2017 educational book - M.MOAM.INFO (2024)

Addressing the Survivorship Care Needs of Patients Receiving Extended Cancer Treatment ..... Julio Garcia-Aguilar, MD, PhD ...... cer-related informationâranging from direct-to-consumer ... to relevant news such as the U.S. Food and Drug Administra- ..... nostic tool in clinical practice,18 and this strategy correlates.

Takeda Oncology

2017 ASCO EDUCATIONAL BOOK

This publication is supported by an educational donation provided by:

AMERICAN SOCIETY OF CLINICAL ONCOLOGY

2017 EDUCATIONAL BOOK Support for this program is funded through

“Making a Difference in Cancer Care WITH YOU” A PEER-REVIEWED, INDEXED PUBLICATION 53rd Annual Meeting | June 2–6, 2017 | Chicago, Illinois | Volume 37

Vol. 37 A special thanks to our Annual Meeting and Program supporters.

American Society of Clinical Oncology Educational Book The 2017 ASCO Educational Book (Print ISSN: 1548-8748 37; Electronic ISSN: 1548-8756) is published by American Society of Clinical Oncology, Inc. (“ASCO”). Requests for permission to reprint all or part of any article published in this title should be directed to Permissions, American Society of Clinical Oncology, Inc., 2318 Mill Road, Suite 800, Alexandria, VA 22314. Tel: (571) 483-1300; fax: (571) 366-9550; or email: [emailprotected]. All other questions should be addressed to ASCO Educational Book Managing Editor, American Society of Clinical Oncology, Inc., 2318 Mill Road, Suite 800, Alexandria, VA 22314. Tel: (571) 483-1300; fax: (571) 366-9550; or email: [emailprotected]. Single issues, both current and back, exist in limited quantities and are offered for sale subject to availability. For further information, email [emailprotected] or call (888) 273-3508. Copyright © 2017 American Society of Clinical Oncology, Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from ASCO. Copies of articles in this publication may be made for personal use. This consent is given on the condition, however, that the copier pay the stated per-copy fee through the Copyright Clearance Center, Inc. (222 Rosewood Drive, Danvers, MA 01923) for any copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. ASCO assumes no responsibility for errors or omissions in this publication. The reader is advised to check the appropriate medical literature and the product information currently provided by the manufacturer of each drug to be administered to verify, among other matters, the dosage, the method and duration of administration, or contraindications. It is the responsibility of the treating physician or other health care professional, relying on the independent experience and knowledge of the patient, to determine drug dosages and the best treatment for the patient. The ideas and opinions expressed in this publication do not necessarily reflect those of ASCO. The mention of any company, product, service, or therapy mentioned does not constitute an endorsem*nt of any kind by ASCO. ASCO assumes no responsibility for any injury or damage to persons or property arising out of or related to any use of the material contained in this publication.

American Society of Clinical Oncology Educational Book Editor in Chief: Don S. Dizon, MD Associate Editor: Nathan Pennell, MD, PhD Guest Editor: Hope S. Rugo, MD Managing Editor: Lindsay F. Pickell, MFA Editorial Coordinator: Christine Melchione Production Manager: Donna Dottellis

Contents 2017 ASCO Annual Meeting Disclosure

2016–2017 Cancer Education Committee

2017 ASCO Educational Book Expert Panel

xiii

Letter From the Editor in Chief

INVITED ARTICLES Can Cancer Truths Be Told? Challenges for Medical Journalism Elaine Schattner

Future Genetic/Genomic Biomarker Testing in Non–Small Cell Lung Cancer David Planchard, Jordi Remon, Fr´ed´erique Nowak, and Jean-Charles Soria

Making the Case for Improving Oncology Workforce Diversity Karen M. Winkfield, Christopher R. Flowers, and Edith P. Mitchell

Minimizing Minimally Invasive Surgery for Endometrial Carcinoma Melissa K. Frey, Stephanie V. Blank, and John P. Curtin

The Road to Addressing Noncommunicable Diseases and Cancer in Global Health Policy Heath Catoe, Jordan Jarvis, Sudeep Gupta, Ophira Ginsburg, and Gilberto de Lima Lopes Jr.

POINTS OF VIEW How Should We Intervene on the Financial Toxicity of Cancer Care? One Shot, Four Perspectives S. Yousuf Zafar, Lee N. Newcomer, Justin McCarthy, Shelley Fuld Nasso, and Leonard B. Saltz

Practice Model for Advanced Practice Providers in Oncology Jamie Cairo, Mary Ann Muzi, Deanna Ficke, Shaunta Ford-Pierce, Katrina Goetzke, Diane Stumvoll, Laurie Williams, and Federico A. Sanchez

BREAST CANCER Breast Cancer in the Central Nervous System: Multidisciplinary Considerations and Management Nancy U. Lin, Laurie E. Gaspar, and Riccardo Soffietti

Lifestyle Interventions to Improve Cardiorespiratory Fitness and Reduce Breast Cancer Recurrence Mark J. Haykowsky, Jessica M. Scott, Kathryn Hudson, and Neelima Denduluri

Novel Targeted Agents and Immunotherapy in Breast Cancer Ingrid A. Mayer, Rebecca Dent, Tira Tan, Peter Savas, and Sherene Loi

Optimal Management of Early and Advanced HER2 Breast Cancer Sara A. Hurvitz, Karen A. Gelmon, and Sara M. Tolaney

The 2017 ASCO Educational Book is published online at asco.org/edbook. Articles that are only available online are denoted with an “e” ahead of the page number.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK

iii

Optimizing Breast Cancer Adjuvant Radiation and Integration of Breast and Reconstructive Surgery Henry M. Kuerer, Peter G. Cordeiro, and Robert W. Mutter

Standard and Genomic Tools for Decision Support in Breast Cancer Treatment N. Lynn Henry, Philippe L. Bedard, and Angela DeMichele

106

Therapeutic Bone-Modifying Agents in the Nonmetastatic Breast Cancer Setting: The Controversy and a Value Assessment Michael Gnant, Catherine Van Poznak, and Lowell Schnipper

116

CANCER PREVENTION, HEREDITARY GENETICS, AND EPIDEMIOLOGY European/U.S. Comparison and Contrasts in Ovarian Cancer Screening and Prevention in a High-Risk Population Marian J. Mourits and G. H. de Bock

124

Social Media and Mobile Technology for Cancer Prevention and Treatment Judith J. Prochaska, Steven S. Coughlin, and Elizabeth J. Lyons

128

CARE DELIVERY AND PRACTICE MANAGEMENT Challenges in Opening and Enrolling Patients in Clinical Trials Julie M. Vose, Meredith K. Chuk, and Francis Giles

139

mHealth: Mobile Technologies to Virtually Bring the Patient Into an Oncology Practice Nathan A. Pennell, Adam P. Dicker, Christine Tran, Heather S. L. Jim, David L. Schwartz, and Edward J. Stepanski

144

Perspectives on the Use of Clinical Pathways in Oncology Care Anne C. Chiang, Peter Ellis, and Robin Zon

155

Precision Oncology: Who, How, What, When, and When Not? Lee Schwartzberg, Edward S. Kim, David Liu, and Deborah Schrag

160

CENTRAL NERVOUS SYSTEM TUMORS American Society for Radiation Oncology 2016 Annual Meeting: Central Nervous System Abstracts Samuel Chao

171

Beyond Alkylating Agents for Gliomas: Quo Vadimus? Vinay K. Puduvalli, Rekha Chaudhary, Samuel G. McClugage, and James Markert

175

Practice-Changing Abstracts From the 2016 Society for Neuro-Oncology Annual Scientific Meeting Marta Penas-Prado

187

DEVELOPMENTAL THERAPEUTICS AND TRANSLATIONAL RESEARCH

Adoptive T-Cell Therapy for Solid Tumors Oladapo Yeku, Xinghuo Li, and Renier J. Brentjens

193

Biomarkers for Checkpoint Inhibition Jeffrey S. Weber

205

Pharmaco*kinetic/Pharmacodynamic Modeling for Drug Development in Oncology Elena Garralda, Rodrigo Dienstmann, and Josep Tabernero

210

2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Strategies to Maximize Patient Participation in Clinical Trials Eric H. Rubin, Mary J. Scroggins, Kirsten B. Goldberg, and Julia A. Beaver

216

Tissue-Agnostic Drug Development Keith T. Flaherty, Dung T. Le, and Steven Lemery

222

GASTROINTESTINAL (COLORECTAL) CANCER Personalizing Adjuvant Therapy for Stage II/III Colorectal Cancer Nadine Jackson McCleary, Al B. Benson III, and Rodrigo Dienstmann

232

Systemic Therapy for Metastatic Colorectal Cancer: From Current Standards to Future Molecular Targeted Approaches Chloe E. Atreya, Rona Yaeger, and Edward Chu

246

GASTROINTESTINAL (NONCOLORECTAL) CANCER Best Practices and Practical Nuances in the Treatment of Gastric Cancer in High-Risk Global Areas Federico A. Sanchez

258

Gastric Cancer in Southern Europe: High-Risk Disease Ramon Andrade De Mello

261

Deploying Immunotherapy in Pancreatic Cancer: Defining Mechanisms of Response and Resistance Gregory L. Beatty, Shabnam Eghbali, and Rebecca Kim

267

Gastric Cancer in Asia: Unique Features and Management Tomoyuki Irino, Hiroya Takeuchi, Masanori Terashima, Toshifumi Wakai, and Yuko Kitagawa

279

Immunotherapy for Esophageal and Gastric Cancer Ronan J. Kelly

292

Pancreatic Adenocarcinoma: Improving Prevention and Survivorship Davendra P. S. Sohal, Field F. Willingham, Massimo Falconi, Kara L. Raphael, and Stefano Crippa

301

The Promise of Immunotherapy in the Treatment of Hepatocellular Carcinoma Anthony El-Khoueiry

311

GENITOURINARY (NONPROSTATE) CANCER Evolving Treatment Paradigm in Metastatic Renal Cell Carcinoma David M. Gill, Neeraj Agarwal, and Ulka Vaishampayan

319

New Developments and Challenges in Rare Genitourinary Tumors: Non-Urothelial Bladder Cancers and Squamous Cell Cancers of the Penis Jeanny B. Aragon-Ching and Lance C. Pagliaro

330

Systemic Therapy for Non–Clear Cell Renal Cell Carcinoma Tian Zhang, Jun Gong, Manuel Caitano Maia, and Sumanta K. Pal

337

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK

GENITOURINARY (PROSTATE) CANCER Diagnosis and Treatment of Prostate Cancer: What Americans Can Learn From International Oncologists Nicholas James, John Graham, Tobias Maurer, Matthias Eiber, and J¨urgen E. Gschwend

344

Personalizing Therapy for Metastatic Prostate Cancer: The Role of Solid and Liquid Tumor Biopsies Terence W. Friedlander, Colin C. Pritchard, and Himisha Beltran

358

Screening and Treating Prostate Cancer in the Older Patient: Decision Making Across the Clinical Spectrum Alicia K. Morgans, William Dale, and Alberto Briganti

370

GERIATRIC ONCOLOGY Improving Quality and Value of Cancer Care for Older Adults Erika E. Ramsdale, Valerie Csik, Andrew E. Chapman, Arash Naeim, and Beverly Canin

383

GLOBAL HEALTH Global Health Initiatives of the International Oncology Community Sana Al-Sukhun, Gilberto de Lima Lopes Jr., Mary Gospodarowicz, Ophira Ginsburg, and Peter Paul Yu

395

Biomarker Testing for Personalized Therapy in Lung Cancer in Low- and Middle-Income Countries Fred R. Hirsch, Bojan Zaric, Ahmed Rabea, Sumitra Thongprasert, Nirush Lertprasertsuke, Mercedes Liliana Dalurzo, and Marileila Varella-Garcia

403

Cancer Care and Control as a Human Right: Recognizing Global Oncology as an Academic Field Alexandru E. Eniu, Yehoda M. Martei, Edward L. Trimble, and Lawrence N. Shulman

409

Thinking Differently in Global Health in Oncology Using a Diagonal Approach: Harnessing Similarities, Improving Education, and Empowering an Alternative Oncology Workforce Natalia M. Rodriguez, Jeannine M. Brant, Dinesh Pendharkar, Hector Arreola-Ornelas, Afsan Bhadelia, Gilberto de Lima Lopes Jr., and Felicia M. Knaul

416

Wedge Resection Versus Anatomic Resection: Extent of Surgical Resection for Stage I and II Lung Cancer Hisao Asamura, Keiju Aokage, and Masaya Yotsukura

426

GYNECOLOGIC CANCER Endometrial Cancer: Is This a New Disease? Kathleen Moore and Molly A. Brewer

435

Whence High-Grade Serous Ovarian Cancer Elise C. Kohn and S. Percy Ivy

443

HEALTH SERVICES RESEARCH, CLINICAL INFORMATICS, AND QUALITY OF CARE More Medicine, Fewer Clicks: How Informatics Can Actually Help Your Practice Debra A. Patt, Elmer V. Bernstam, Joshua C. Mandel, David A. Kreda, and Jeremy L. Warner

2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

450

The Oncology Care Model: Perspectives From the Centers for Medicare & Medicaid Services and Participating Oncology Practices in Academia and the Community Ron Kline, Kerin Adelson, Jeffrey J. Kirshner, Larissa M. Strawbridge, Marsha Devita, Naralys Sinanis, Patrick H. Conway, and Ethan Basch

460

HEMATOLOGIC MALIGNANCIES—LEUKEMIA, MYELODYSPLASTIC SYNDROMES, AND ALLOTRANSPLANT Chronic Myeloid Leukemia: What Every Practitioner Needs to Know in 2017 Hanna Jean Khoury, Loretta A. Williams, Ehab Atallah, and R¨udiger Hehlmann

468

New Insight Into the Biology, Risk Stratification, and Targeted Treatment of Myelodysplastic Syndromes Mintallah Haider, Eric J. Duncavage, Khalid F. Afaneh, Rafael Bejar, and Alan F. List

480

Novel Therapeutics in Acute Myeloid Leukemia Courtney D. DiNardo, Richard M. Stone, and Bruno C. Medeiros

495

HEMATOLOGIC MALIGNANCIES—LYMPHOMA AND CHRONIC LYMPHOCYTIC LEUKEMIA Age and Sex in Non-Hodgkin Lymphoma Therapy: It’s Not All Created Equal, or Is It? Michael Pfreundschuh

505

Current Approaches to Mantle Cell Lymphoma: Diagnosis, Prognosis, and Therapies Jonathon B. Cohen, Jasmine M. Zain, and Brad S. Kahl

512

Health Disparities and the Global Landscape of Lymphoma Care Today Adrienne A. Phillips and Dominic A. Smith

526

Understanding the New WHO Classification of Lymphoid Malignancies: Why It’s Important and How It Will Affect Practice Elaine S. Jaffe, Paul M. Barr, and Sonali M. Smith

535

HEMATOLOGIC MALIGNANCIES—PLASMA CELL DYSCRASIA Established and Novel Prognostic Biomarkers in Multiple Myeloma Mark Bustoros, Tarek H. Mouhieddine, Alexandre Detappe, and Irene M. Ghobrial

548

Hematologic Malignancies: Plasma Cell Disorders Madhav V. Dhodapkar, Ivan Borrello, Adam D. Cohen, and Edward A. Stadtmauer

561

Integration of Genomics Into Treatment: Are We There Yet? Gareth J. Morgan and John R. Jones

569

Myeloma in Elderly Patients: When Less Is More and More Is More Ashley Rosko, Sergio Giralt, Maria-Victoria Mateos, and Angela Dispenzieri

575

LUNG CANCER Caring for the Older Population With Advanced Lung Cancer Carolyn J. Presley, Craig H. Reynolds, and Corey J. Langer

587

Clinical Pathways and the Patient Perspective in the Pursuit of Value-Based Oncology Care Jennifer L. Ersek, Eric Nadler, Janet Freeman-Daily, Samir Mazharuddin, and Edward S. Kim

597

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK

vii

Managing Resistance to EFGR- and ALK-Targeted Therapies Christine M. Lovly, Puneeth Iyengar, and Justin F. Gainor

607

Pathology Issues in Thoracic Oncology: Histologic Characterization and Tissue/Plasma Genotyping May Resolve Diagnostic Dilemmas Ibiayi Dagogo-Jack, Andreas Saltos, Alice T. Shaw, and Jhanelle E. Gray

619

Role of Chemotherapy and Targeted Therapy in Early-Stage Non–Small Cell Lung Cancer Shirish M. Gadgeel

630

MELANOMA/SKIN CANCERS Advances in the Treatment of Advanced Extracutaneous Melanomas and Nonmelanoma Skin Cancers Kimberly M. Komatsubara, Joanne Jeter, Richard D. Carvajal, Kim Margolin, Dirk Schadendorf, and Axel Hauschild

641

Operable Melanoma: Screening, Prognostication, and Adjuvant and Neoadjuvant Therapy Ahmad A. Tarhini, Paul Lorigan, and Sancy Leachman

651

Systemic Therapy Options for Patients With Unresectable Melanoma Melinda Yushak, Paul Chapman, Caroline Robert, and Ragini Kudchadkar

661

PATIENT AND SURVIVOR CARE Addressing the Survivorship Care Needs of Patients Receiving Extended Cancer Treatment Paul B. Jacobsen, Ryan D. Nipp, and Patricia A. Ganz

674

Bench-to-Bedside Approaches for Personalized Exercise Therapy in Cancer Lee W. Jones, Neil D. Eves, and Jessica M. Scott

684

Improving Cancer Care Through the Patient Experience: How to Use Patient-Reported Outcomes in Clinical Practice Kathi Mooney, Donna L. Berry, Meagan Whisenant, and Daniel Sjoberg

695

Pain and Opioids in Cancer Care: Benefits, Risks, and Alternatives Mike Bennett, Judith A. Paice, and Mark Wallace

705

Using the New ASCO Clinical Practice Guideline for Palliative Care Concurrent With Oncology Care Using the TEAM Approach Cardinale B. Smith, Tanyanika Phillips, and Thomas J. Smith

714

PEDIATRIC ONCOLOGY

viii

Advances in the Treatment of Pediatric Bone Sarcomas Patrick J. Grohar, Katherine A. Janeway, Luke D. Mase, and Joshua D. Schiffman

725

Breast Cancer After Childhood, Adolescent, and Young Adult Cancer: It’s Not Just About Chest Radiation David Hodgson, Flora van Leeuwen, Andrea Ng, Lindsay Morton, and Tara O. Henderson

736

Data Commons to Support Pediatric Cancer Research Samuel L. Volchenboum, Suzanne M. Cox, Allison Heath, Adam Resnick, Susan L. Cohn, and Robert Grossman

746

2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

New Classification for Central Nervous System Tumors: Implications for Diagnosis and Therapy Christine E. Fuller, David T. W. Jones, and Mark W. Kieran

753

PROFESSIONAL DEVELOPMENT Collaborating With Advanced Practice Providers Impact and Opportunity Heather M. Hylton and G. Lita Smith

For Our Patients, for Ourselves: The Value of Personal Reflection in Oncology Lidia Schapira, Jane Lowe Meisel, and Ranjana Srivastava

765

Mastering Resilience in Oncology: Learn to Thrive in the Face of Burnout Fay J. Hlubocky, Miko Rose, and Ronald M. Epstein

771

Social Media for Networking, Professional Development, and Patient Engagement Merry Jennifer Markham, Danielle Gentile, and David L. Graham

782

The Road of Mentorship Kelly J. Cooke, Debra A. Patt, and Roshan S. Prabhu

788

SARCOMA Bone Sarcoma Pathology: Diagnostic Approach for Optimal Therapy Andrew E. Rosenberg

794

Fertility, Cardiac, and Orthopedic Challenges in Survivors of Adult and Childhood Sarcoma Emma R. Lipshultz, Ginger E. Holt, Ranjith Ramasamy, Raphael Yechieli, and Steven E. Lipshultz

799

The Current Landscape of Early Drug Development for Patients With Sarcoma Breelyn A. Wilky, Robin L. Jones, and Vicki L. Keedy

807

TUMOR BIOLOGY Higher-Level Pathway Objectives of Epigenetic Therapy: A Solution to the p53 Problem in Cancer Vamsidhar Velcheti, Tomas Radivoyevitch, and Yogen Saunthararajah

812

Metabolic Alterations in Cancer and Their Potential as Therapeutic Targets Jamie D. Weyandt, Craig B. Thompson, Amato J. Giaccia, and W. Kimryn Rathmell

825

Value-Based Medicine and Integration of Tumor Biology Gabriel A. Brooks, Linda D. Bosserman, Isa Mambetsariev, and Ravi Salgia

833

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK

2017 ASCO Annual Meeting Disclosure As the continuing education provider for the 2017 Annual Meeting, ASCO is committed to balance, objectivity, and scientific rigor in the management of financial interactions with for-profit health care companies that could create real or perceived conflicts of interest. Participants in the Meeting have disclosed their financial relationships in accordance with ASCO’s Policy for Relationships with Companies; review the policy at asco.org/rwc. ASCO offers a comprehensive disclosure management system, using one disclosure for all ASCO activities. Authors and participants are required to disclose all interactions with companies. Disclosures are kept on file and can be confirmed or updated with each new activity. Disclosures for the ASCO Educational Book are available to readers online, listed by author for each article. Please email [emailprotected] with specific questions or concerns.

2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

2016–2017 Cancer Education Committee The Cancer Education Committee assesses the need for, plans, develops, and initiates the education programs for the Annual Meeting. Michael A. Thompson, MD, PhD—Chair David R. Spigel, MD—Chair-Elect Apar Kishor Ganti, MD, MBBS—Immediate Past Chair

BREAST CANCER Debra A. Patt, MD, MPH, MBA—Track Leader Judy C. Boughey, MD Jennifer A. Brown, MD Rebecca A. Dent, MD Alberto J. Montero, MD Rinaa S. Punglia, MD, MPH Hope S. Rugo, MD Tallal Younis, MBBCh, FRCP

CANCER PREVENTION, HEREDITARY GENETICS, AND EPIDEMIOLOGY Sofia Merajver, MD—Track Leader Margreet Ausems, MD, PhD Jeff Boyd, PhD Monique A. De Bruin, MD Joanne Jeter, MD Prudence Lam, MD Michael Mullane, MD Surendra S. Shastri, MD, MBBS

GENITOURINARY (NONPROSTATE) CANCER Thomas E. Hutson, DO, PharmD, FACP—Track Leader Jeanny Aragon-Ching, MD, FACP Stephen Boyd Riggs, MD Ulka N. Vaishampayan, MD

GENITOURINARY (PROSTATE) CANCER Nicholas J. Vogelzang, MD, FASCO, FACP—Track Leader Himisha Beltran, MD Chris Parker, MD, BM Bchir, FRCR, FRCP Edward M. Schaeffer, MD, PhD

GERIATRIC ONCOLOGY Heidi Klepin, MD, MS—Track Leader Gretchen Kimmick, MD Stuart Lichtman, MD Hyman Muss, MD, FASCO

GLOBAL HEALTH Peter Paul Yu, MD, FASCO—Track Leader Alex Mutombo Baleka, MD Julie Gralow, MD, FASCO Gilberto de Lima Lopes Jr., MD, MBA, FAMS

GYNECOLOGIC CANCER CARE DELIVERY AND PRACTICE MANAGEMENT Lee S. Schwartzberg, MD, FACP—Track Leader Kelly Bugos, MS, RN, NP Moshe C. Chasky, MD Jeffery C. Ward, MD Sue S. Yom, MD, PhD

CENTRAL NERVOUS SYSTEM TUMORS Nicole A. Shonka, MD—Track Leader Manmeet S. Ahluwalia, MD Priscilla Brastianos, MD Eric L. Chang, MD

DEVELOPMENTAL THERAPEUTICS AND TRANSLATIONAL RESEARCH Francisco J. Esteva, MD, PhD—Track Leader Julia A. Beaver, MD Howard A. Burris, MD, FASCO John C. Morris, MD Naoko Takebe, MD, PhD

Linda Van Le, MD—Track Leader Ronald J. Buckanovich, MD Elizabeth Dickson, MD Elise C. Kohn, MD

HEAD AND NECK CANCER Irina Veytsman, MD—Track Leader John A. Ridge, MD, PhD, FACS Joseph K. Salama, MD John Truelson, MD

HEALTH SERVICES RESEARCH, CLINICAL INFORMATICS, AND QUALITY OF CARE Ethan M. Basch, MD—Track Leader Dawn L. Hershman, MD Nicole M. Kuderer, MD Ryan Nipp, MD Jeremy Warner, MD, MS Yousuf Zafar, MD, MHS

GASTROINTESTINAL (COLORECTAL) CANCER

HEMATOLOGIC MALIGNANCIES—LEUKEMIA, MYELODYSPLASTIC SYNDROMES, AND ALLOTRANSPLANT

Henry Q. Xiong, MD—Track Leader Chi Lin, MD, PhD Arden Morris, MD Donald A. Richards, MD, PhD Ashwin Reddy Sama, MD

Steven Devine, MD—Track Leader Jorge E. Cortes, MD Jonathan M. Gerber, MD Hanna Khoury, MD David Leibowitz, MD

GASTROINTESTINAL (NONCOLORECTAL) CANCER

HEMATOLOGIC MALIGNANCIES—LYMPHOMA AND CHRONIC LYMPHOCYTIC LEUKEMIA

Manish A. Shah, MD—Track Leader Ramon De Mello, MD, PhD Jimmy J. Hwang, MD Vincent J. Picozzi, MD Federico A. Sanchez, MD William Small, MD

John M. Pagel, MD, PhD—Track Leader Carla Casulo, MD Ishmael Jaiyesimi, DO, MS, FACP Matthew A. Lunning, DO Barbara Pro, MD

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK

HEMATOLOGIC MALIGNANCIES—PLASMA CELL DYSCRASIA

PROFESSIONAL DEVELOPMENT

Parameswaran Hari, MD—Track Leader Asher A. Chanan-Khan, MD Irene M. Ghobrial, MD Sagar Lonial, MD Jeffrey Matous, MD Ravi Vij, MD

Kelly J. Cooke, DO—Track Leader Kristin Anderson, MD, MPH Jill Gilbert, MD Laura Goff, MD Roberto A. Leon-Ferre, MD Jane Meisel, MD

LUNG CANCER

SARCOMA

Craig H. Reynolds, MD—Track Leader Shirish M. Gadgeel, MD Puneeth Iyengar, MD, PhD Aaron S. Mansfield, MD Nathan A. Pennell, MD, PhD Anne Tsao, MD

Jonathan Trent II, MD, PhD—Track Leader Robin Lewis Jones, MB, MRCP, MD Min S. Park, MD Nicholas P. Webber, MD

MELANOMA/SKIN CANCERS Ahmad A. Tarhini, MD, PhD—Track Leader Sanjiv S. Agarwala, MD Alexander C. J. van Akkooi, MD, PhD Ragini Kudchadkar, MD Gregory Pennock, MD, FACP

PATIENT AND SURVIVOR CARE Kathi Mooney, PhD, RN—Track Leader Deborah Mayer, PhD, RN, AOCN, FAAN Patricia Robinson, MD Maria A. Rodriguez, MD Nagendra Tirumali, MD Louise C. Walter, MD

PEDIATRIC ONCOLOGY Tara O. Henderson, MD, MPH—Track Leader Sung Won Choi, MD, MS Paul D. Harker-Murray, MD, PhD Michael Ortiz, MD

xii

2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

TUMOR BIOLOGY Ravi Salgia, MD, PhD—Track Leader Kimryn Rathmell, MD, PhD Eliezer M. Van Allen, MD Vamsidhar Velcheti, MD

LIAISONS Sana Al-Sukhun, MD, MSc—International Affairs Committee Anne Chiang, MD, PhD—Quality of Care Committee Christopher Flowers, MD, MS—Health Disparities Committee Stephen Gruber, MD, PhD—Cancer Prevention Committee Paul B. Jacobsen, PhD—Cancer Survivorship Committee Jeffrey Kirshner, MD—Clinical Practice Guidelines Committee Heidi Klepin, MD, MS—Cancer Research Committee Debra Patt, MD, MPH, MBA—Clinical Practice Committee Leonard Saltz, MD—Value Task Force Laura L. Tenner, MD—Ethics Committee

2017 ASCO Educational Book Expert Panel The Expert Panel is a group of well-recognized physicians and researchers in oncology and related fields who have served as peer reviewers of the ASCO Educational Book articles.

Ghassan K. Abou-Alfa, MD Memorial Sloan Kettering Cancer Center, Weill Cornell Medical College Donald I. Abrams, MD San Francisco General Hospital Vandana G. Abramson, MD Vanderbilt University Medical Center Melissa K. Accordino, MD New York-Presbyterian Hospital Ranjana H. Advani, MD Stanford Cancer Institute Sanjiv S. Agarwala, MD St. Luke’s Medical Center Manmeet S. Ahluwalia, MD Cleveland Clinic Jaffer A. Ajani, MD The University of Texas MD Anderson Cancer Center Sana Al-Sukhun, MD, MSc The University of Jordan Terrance L. Albrecht, PhD Wayne State University Laleh Amiri-Kordestani, MD U.S. Food and Drug Administration

David M. Baer, MD, FACP Kaiser Permanente

Susan M. Chang, MD University of California, San Francisco

Christina S. Baik, MD, MPH Seattle Cancer Care Alliance

Stephen J. Chanock, MD National Institute of Health

Karla V. Ballman, PhD Weill Cornell Medicine, Meyer Cancer Center

Alice P. Chen, MD National Cancer Institute at the National Institutes of Health

Tracy Batchelor, MD Mayo Clinic

Helen K. Chew, MD UC Davis Medical Center

Brigitta G. Baumert, MD, PhD, MBA University of Bonn Medical Centre

Stephen K. L. Chia, MD BC Cancer Agency

Abbie Begnaud, MD University of Minnesota

E. Gabriela Chiorean, MD Fred Hutchinson Cancer Research Center

Robert S. Benjamin, MD The University of Texas MD Anderson Cancer Center

Laura Q. M. Chow, MD University of Washington

Wendie Berg, MD, PhD University of Pittsburgh

Hak Choy, MD The University of Texas Southwestern Medical Center

Michael F. Berger, PhD Memorial Sloan Kettering Cancer Center Jan H. Beumer, PharmD, PhD University of Pittsburgh Cancer Institute

Quyen Chu, MD, MBA, FACS LSU Health Sciences Center Adam D. Cohen, MD, PhD Fred Hutchinson Cancer Research Center

Andrea Bezjak, MD Princess Margaret Cancer Centre, University Health Network

Harvey J. Cohen, MD Duke University Medical Center

Susan Blaney, MD Texas Children’s Cancer Center, Baylor College of Medicine

Elise D. Cook, MD, MS The University of Texas MD Anderson Cancer Center

John A. Bridgewater, MD University College London Cancer Institute

Jeffrey Crawford, MD Duke University Medical Center

Jennifer R. Brown, MD Dana-Farber Cancer Institute

Carien L. Creutzberg, MD, PhD Leiden University Medical Center

Frederick R. Appelbaum, MD Fred Hutchinson Cancer Research Center

Paul A. Bunn, MD, FASCO University of Colorado Denver

Katherine D. Crew, MD, MS Columbia University Medical Center

Saro Armenian, DO, MPH City of Hope National Medical Center

Harold J. Burstein, MD, PhD, FASCO Harvard University

Suzanne Eleanor Dahlberg, PhD Dana-Farber Cancer Institute

Gregory T. Armstrong, MD, MSCE St. Jude Children’s Research Hospital

Emiliano Calvo, MD, PhD Clara Comprehensive Cancer Center

Sandra P. D’Angelo, MD Memorial Sloan Kettering Cancer Center

Herve Avet-Loiseau, MD National Cancer Centre Singapore

Lisa A. Carey, MD The University of North Carolina

Nancy E. Davidson, MD University of Pittsburgh Cancer Institute

Hatem A. Azim, MD, PhD Institut Jules Bordet

Kenneth R. Carson, MD, PhD Washington University

Laura A. Dawson, MD Princess Margaret Cancer Centre

Fabrice Andre, MD, PhD Institute Gustave Roussy Christina M. Annunziata, MD, PhD National Cancer Institute at the National Institutes of Health Emmanuel S. Antonarakis, MD The Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK

xiii

Daniel J. DeAngelo, MD, PhD Dana-Farber Cancer Institute H. Joachim Deeg, MD Fred Hutchinson Cancer Research Center Mark A. Dickson, MD Memorial Sloan Kettering Cancer Center Melissa S. Dillmon, MD Harbin Clinic LLC Mary L. Disis, MD University of Washington Ethan Dmitrovsky, MD The University of Texas MD Anderson Cancer Center Susan M. Domchek, MD University of Pennsylvania Perelman School of Medicine Martin H. Dreyling, MD, PhD University Hospital Grosshadem Reinhard Dummer, MD University Hospital Zurich Linda R. Duska, MD University of Virginia Health System Grace K. Dy, MD Roswell Park Cancer Institute Hatem M. El Halabi, MD Cancer Treatment Centers of America at Midwestern Regional Medical Center Andrew M. Evens, DO, FACP Tufts Medical Center, Tufts University Michael S. Ewer, MD, MPH, JD The University of Texas MD Anderson Cancer Center Marwan Fakih, MD City of Hope Comprehensive Cancer Center Michelle A. Fanale, MD The University of Texas MD Anderson Cancer Center

Nathan H. Fowler, MD The University of Texas MD Anderson Cancer Center Matt D. Galsky, MD The Tisch Cancer Institute, Icahn School of Medicine at Mount Sinai Julio Garcia-Aguilar, MD, PhD Memorial Sloan Kettering Cancer Center Edward B. Garon, MD David Geffen School of Medicine at UCLA Peter Gibbs, MBBS, FRACP, MD Royal Melbourne Hospital Mark R. Gilbert, MD Center for Cancer Research, National Cancer Institute Silke Gillessen, MD Kantonsspital St. Gallen Timothy D. Gilligan, MD, FASCO Cleveland Clinic Sharon H. Giordano, MD, MPH The University of Texas MD Anderson Cancer Center Robert Glynne-Jones, MD, FRCP Mount Vernon Cancer Centre Valentin Goede, MD University Hospital of Cologne Lia Gore, MD University of Colorado Cancer Center Mary K. Gospodarowicz, MD, MPH Massachusetts General Hospital Bernardo H. L. Goulart, MD, MS Fred Hutchinson Cancer Research Center Ramaswamy Govindan, MD Washington University School of Medicine William J. Gradishar, MD Robert H. Lurie Comprehensive Cancer Center of Northwestern University

Michael J. Hallek, DO, MSc Tufts Medical Center Michael T. Halpern, MD, PhD, MPH Temple University College of Public Health Stanley R. Hamilton, MD The University of Texas MD Anderson Cancer Center Nasser H. Hanna, MD Indiana University Melvin and Bren Simon Cancer Center Michelle Harvie, PhD Nightingale and Genesis Prevention Centre, Wythenshawe Hospital Nooshin Hashemi Sadraei, MD University of Cincinnati David N. Hayes, MD, MPH UNC Lineberger Comprehensive Cancer Center Axel Heidenreich, MD, PhD Cologne University Roy S. Herbst, MD, PhD Yale University School of Medicine Andrew A. Hertler, MD, FACP New Century Health Jean H. Hoffman-Censits, MD The Sidney Kimmel Cancer Center at Thomas Jefferson University Christine Holmberg, DPhil, MPH Charite´ Universit¨atsmedizin Berlin, Berlin School of Public Health Gabriel N. Hortobagyi, MD, FACP, FASCO The University of Texas MD Anderson Cancer Center Shannon L. Huggins-Puhalla, MD University of Pittsburgh David H. Ilson, MD, PhD Memorial Sloan Kettering Cancer Center

Tatyana A. Feldman, MD John Theurer Cancer Center

Alessandro Gronchi, MD Fondazione IRCCS Istituto Nazionale dei Tumori

Syma Iqbal, MD, FACP University of Southern California

Josephine L. Feliciano, MD University of Maryland Greenebaum Cancer Center

Axel Grothey, MD Mayo Clinic

Claudine Isaacs, MD Georgetown Lombardi Comprehensive Cancer Center

Robert L. Ferris, MD, PhD University of Pittsburgh Cancer Institute

Beverly A. Guadagnolo, MD, MPH The University of Texas MD Anderson Cancer Center

Jeffrey A. Jones, MD, MPH The Ohio State University

Robert A. Figlin, MD, FACP Cedars-Sinai Medical Center

Gordon Hafner, MD Inova Medical Group

Joseph G. Jurcic, MD Columbia University Medical Center

Gunnar Folprecht, MD University Hospital Carl Gustav Carus

William C. Hahn, MD, PhD Dana-Farber Cancer Institute

Thomas J. Kaley, MD Memorial Sloan Kettering Cancer Center

xiv

2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Jeffrey Karnes, MD Mayo Clinic

Cynthia X. Ma, MD, PhD Washington University School of Medicine

Lee N. Newcomer, MD United Health Group

Sean Kehoe, MD University of Birmingham

Helen Mackay, MD Princess Margaret Cancer Centre

Kim Nichols, MD St. Jude Children’s Research Hospital

Kara Kelly, MD Roswell Park Cancer Institute

Amy R. MacKenzie, MD Thomas Jefferson University Hospital

Olatoyosi Odenike, MD The University of Chicago

Gretchen Genevieve Kimmick, MD Duke University

Robert G. Maki, MD, PhD Icahn School of Medicine at Mount Sinai

Travis J. Osterman, DO Vanderbilt University School of Medicine

Michael P. Kosty, MD, FACP Scripps Research Institute Maxwell M. Krem, MD University of Washington Rebecca Sophie Kristeleit, BSc, MRCP, PhD University College London Cancer Institute Geoffrey Y. Ku, MD, MBA Memorial Sloan Kettering Cancer Center Allison W. Kurian, MD, MSc Standford University Ann S. LaCasce, MD Dana-Farber Cancer Institute Martha Lacy, MD Abramson Cancer Center of the University of Pennsylvania Marc Ladanyi, MD Memorial Sloan Kettering Cancer Center Jerome C. Landry, MD, MBA Emory University and Clinic Alexandra Leary, MD, PhD Gustave Roussy Cancer Center

Rami Manochakian, MD Case Western Reserve University Miguel Martin, MD, PhD Hospital General Universitario Gregorio Maraon Viraj A. Master, MD, PhD Winship Cancer Institute of Emory University Heather L. McArthur, MD Cedars Sinai Medical Center

Cynthia Owusu, MD, MS Case Western Reserve University Amit M. Oza, MD Princess Margaret Cancer Centre John M. Pagel, MD Mayo Clinic Paul K. Paik, MD Memorial Sloan Kettering Cancer Center Alberto S. Pappo, MD St. Jude Children’s Research Hospital

Amy E. McKee, MD U.S. Food and Drug Administration Robert R. McWilliams, MD Mayo Clinic

Donald W. Parsons, MD, PhD Texas Children’s Cancer Center, Baylor College of Medicine Shreyaskumar Patel, MD The University of Texas MD Anderson Cancer Center

Bhoomi Mehrotra, MD St. Francis Hospital

Vincent J. Picozzi, MD Virginia Mason Medical Center

Minesh P. Mehta, MD Miami Cancer Institute Alexander M. Menzies, BSc(Med), MBBS, FRACP, PhD Melanoma Institute Australia, Royal North Shore Hospital, The University of Sydney

Seth Pollack, MD Fred Hutchinson Cancer Research Center Sandro Porceddu, MD Princess Alexandra Hospital

Ming Lei, PhD National Cancer Institute at the National Institutes of Health

Jeffrey A. Meyerhardt, MD, MPH Dana-Farber Cancer Institute

Michael A. Postow, MD Memorial Sloan Kettering Cancer Center

Mario M. Leitao, MD Memorial Sloan Kettering Cancer Center

Frederick J. Meyers, MD University of California, Davis

Melanie Powell, MD Barts Health NHS Trust

Daniel J. Lenihan, MD Vanderbilt University Medical Center

Paul A. Meyers, MD Memorial Sloan Kettering Cancer Center

Cornelis J. A. Punt, MD, PhD Academic Medical Center, University of Amsterdam

Jennifer A. Ligibel, MD Dana-Farber Cancer Institute

Linda R. Mileshkin, MBBS, MD, MBioeth Peter MacCallum Cancer Centre

S. Vincent Rajkumar, MD City of Hope

Michael Lim, MD Johns Hopkins University School of Medicine

Sandra A. Mitchell, PhD, RN National Cancer Institute at the National Institutes of Health

Jeffrey Razier, MD Northwestern University

Alison Moliterno, MD Johns Hopkins University

Alyssa G. Rieber, MD The University of Texas MD Anderson Cancer Center

Jason J. Luke, MD, FACP University of Chicago Comprehensive Cancer Center

Karen M. Mustian, PhD, MPH University of Rochester Medical Center

Brian I. Rini, MD Cleveland Clinic Taussig Cancer Institute

Gary H. Lyman, MD, MPH, FASCO, FACP, FRCP Fred Hutchinson Cancer Research Center

Peter L. J. Naredi, MD, PhD Sahlgrenska University Hospital, University of Gothenburg

Julia H. Rowland, PhD National Cancer Institute at the National Institutes of Health

Soon Thye Lim, MD Princess Margaret Hospital

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK

Sameek Roychowdhury, MD, PhD The Ohio State University

A. Keith Stewart, MD Mayo Clinic

Vamsidhar Velcheti, MD Cleveland Clinic

Kathryn J. Ruddy, MD, MPH Mayo Clinic

Ryan J. Sullivan, MD Massachusetts General Hospital Cancer Center, Harvard Medical School

Alan P. Venook, MD University of California, San Francisco

Paul Ruff, MBBCH, MMed, FCP(SA) University of Witwatersrand Faculty of Health Sciences

Christopher Sweeney, MBBS Dana-Farber Cancer Institute

David P. Ryan, MD Massachusetts General Hospital Cancer Center

Antoinette R. Tan, MD Levine Cancer Institute, Carolinas Healthcare System

Joseph K. Salama, MD Duke University Medical Center

Lynne P. Taylor, MD Virginia Mason Medical Center

Stephen E. Sallan, MD Dana-Farber Cancer Institute, Harvard Medical School

Joel E. Tepper, MD UNC Lineberger Comprehensive Cancer Center

Alan Sandler, MD Genentech, Inc.

Evangelos Terpos, MD, PhD National and Kapodistrian University of Athens

Howard M. Sandler, MD Cedars-Sinai Medical Center Hanna Kelly Sanoff, MD, MPH The University of North Carolina at Chapel Hill School of Medicine

William P. Tew, MD Memorial Sloan Kettering Cancer Center, Weill Cornell Medical College Charles R. Thomas, MD Oregon Health & Science University

Nita Seibel, MD National Cancer Institute at the National Institutes of Health

Ian M. Thompson, MD The University of Texas Health Science Center at San Antonio

Manish A. Shah, MD Weill Cornell Medicine, New York-Presbyterian Hospital

Michael A. Thompson, MD, PhD Aurora Research Institute, Aurora Health Care

Armin Shahrokni, MD, MPH Memorial Sloan Kettering Cancer Center

Katherine A. Thornton, MD The Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins

David L. Sher, MD, MPH Rush University Medical Center Liran Shlush, MD, PhD Princess Margaret Cancer Centre

Deborah Toppmeyer, MD Rutgers Cancer Institute of New Jersey Joseph M. Unger, PhD, MS Fred Hutchinson Cancer Research Center

Ravi Vij, MBBS, MD Washington University School of Medicine Victor G. Vogel, MD Geisinger Health System Wendy H. Vogel, MSN, FNP, AOCNP Wellmont Cancer Institute Heather A. Wakelee, MD Stanford University Joan L. Walker, MD The University of Oklahoma Health Sciences Center Christine M. Walko, PharmD, BCOP Moffitt Cancer Center Jeffery C. Ward, MD Swedish Cancer Institute Edmonds Padraig R. Warde, MD Princess Margaret Cancer Centre Jeffrey S. Wefel, PhD The University of Texas MD Anderson Cancer Center Michael Weller, MD University Hospital Zurich Howard J. West, MD Swedish Medical Center Tanya M. Wildes, MD, MSCI Washington University School of Medicine in St. Louis William N. William, MD The University of Texas MD Anderson Cancer Center

Marc Shuman, MD USCF Helen Diller Comprehensive Cancer Center

Neha Vapiwala, MD University of Pennsylvania

Ignacio I. Wistuba, MD The University of Texas MD Anderson Cancer Center

Vernon K. Sondak, MD Moffitt Cancer Center

Anna M. Varghese, MD Memorial Sloan Kettering Cancer Center

Hendrik Witt, PhD University of Heidelberg

xvi

2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Letter From the Editor in Chief

n behalf of my Associate Editor, Dr. Nathan Pennell, and Guest Editor, Dr. Hope S. Rugo, I welcome you to the 2017 ASCO Annual Meeting. It is an honor and privilege to present the 37th volume of the NLM-indexed ASCO Educational Book. The theme of this year’s Meeting is “Making a Difference in Cancer Care WITH YOU,” and this theme celebrates the inclusiveness of those that work together to diagnose and care for people with cancer. With his presidential theme, Dr. Daniel F. Hayes celebrates the inclusive nature of the oncology community. Only when the brightest minds in research, education, and care work together as a community are we able to deliver the highest quality of care to meet the needs of all of our patients. To celebrate the theme of the 2017 Annual Meeting, this volume contains articles coauthored not only by those who diagnose and care for patients with cancer, but also ancillary care specialists, such as nurses and advanced practice providers, and physicians in training. Coauthoring a manuscript takes immense planning and collaboration, and I would like to thank all of the authors for their contributions to the 2017 ASCO Educational Book.

We are honored to have Dr. Rugo join us as a Guest Editor for this year’s Invited Articles. The Invited Articles allow us to explore critical topics in oncology that are closely related to this year’s theme. I would like to thank Dr. Rugo for her dedication and her willingness to oversee this important section. Finally, I would also like to recognize the expert panel who selflessly dedicated their time to perform thorough and thoughtful reviews of the submitted articles. The tremendous work of Dr. Rugo, Dr. Pennell, the expert panel, and all of the authors is especially pertinent to this year’s theme of inclusivity and support. It is my honor to invite you to read the exceptional contributions that comprise this volume. For the first time in several years, the print edition contains the full collection of all of the 2017 articles and is available for purchase on the ASCO University Bookstore. All of the 2017 ASCO Educational Book articles, as well as articles from past volumes, are available to view online for free at www.asco.org/edbook. We welcome your feedback and suggestions on how we can improve the content, so please contact us at [emailprotected] with your comments. Sincerely,

Don S. Dizon, MD Editor in Chief

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK

INVITED ARTICLES This year’s invited articles represent the 2017 ASCO Annual Meeting theme, “Making a Difference in Cancer Care WITH YOU.” These important contributions to the 37th volume of the ASCO Educational Book celebrate the inclusiveness of those that work together to diagnose and care for people with cancer. The authors represent a diverse, multidisciplinary set of expertise and backgrounds. Authors were nominated by the ASCO Educational Book Editors and 2017 ASCO Annual Meeting leadership, and authors developed their topics under the guidance of Dr. Hope S. Rugo, Guest Editor of the Invited Articles.

ARTICLES Can Cancer Truths Be Told? Challenges for Medical Journalism Elaine Schattner, MD Future Genetic/Genomic Biomarker Testing in Non–Small Cell Lung Cancer David Planchard, MD, PhD, Jordi Remon, MD, Fr´ed´erique Nowak, PhD, and Jean-Charles Soria, MD, PhD Making the Case for Improving Oncology Workforce Diversity Karen M. Winkfield, MD, PhD, Christopher R. Flowers, MD, MS, and Edith P. Mitchell, MD, FACP Minimizing Minimally Invasive Surgery for Endometrial Carcinoma Melissa K. Frey, MD, Stephanie V. Blank, MD, John P. Curtin, MD The Road to Addressing Noncommunicable Diseases and Cancer in Global Health Policy Heath Catoe, MD, PhD, Jordan Jarvis, MSc, Sudeep Gupta, MD, MBBS, Ophira Ginsburg, MD, FRCPC, Gilberto de Lima Lopes Jr., MD, MBA, FAMS

INVITED ARTICLES

Oncology in the News

Can Cancer Truths Be Told? Challenges for Medical Journalism Elaine Schattner, MD

ournalism is a field undergoing rapid transformation. In 2016, nearly two-thirds of U.S. adults received news by social media. The Pew Research Center reported that the proportion and number of men and women seeking news on Reddit, Facebook, Twitter, Instagram, and YouTube has been climbing since 2013.1 Meanwhile, traditional newspapers contend with falling print circulation, compete for online traffic, and drop staff. The number of U.S. newsroom employees fell steadily after 2006, from 55,000 to fewer than 33,000 jobs.2 Medical news presents a unique set of challenges, both for journalists and consumers. On the production side, reporters and editors aim to translate doctors’ jargon-loaded updates into digestible, truthful, and appealing bits of information—stories—that resonate with a lay audience. The work is no small task, given the complexity and pace of science and clinical research. On the receiving end, patients or caregivers might read, watch, listen, or skim a feed, consciously or unconsciously taking notes. No matter what the source, an article might influence an individual’s thinking about a personal medical decision. On a larger scale, medical journalism can sway policy makers. The quality and accuracy of news has the potential to alter health outcomes. Put simply, the public depends on reliable news to support everyday medical choices and, occasionally, inform major decisions. When journalists get stories right, they help people to make reasoned choices and ask better questions of their physicians. Conversely, when reporters make errors or editors publish misleading headlines, people with medical conditions and other consumers of news, may be harmed. This article will explore the capacity and limits of health journalism to inform the public about developments in oncology. We will focus on two issues: one is a perennial concern— balancing hype and reality, with appropriate skepticism—in what might be an era of true progress; second is the changing place of journalism amid a torrent of ungated medical

information and stories channeled in blogs, social media posts, and celebrity statements. Finally, we will consider the emerging challenge of distilling valuable and relevant medical news amid a surplus.

BALANCING SKEPTICISM AND HYPE

A 2016 CBS Evening News story, “Promising Brain Cancer Trial Given Breakthrough Status by FDA,” offers an instructive example3 of how news can affect patients’ thinking, hopefulness, and care. Scott Pelley, said by CBS on its website to be “one of the most experienced reporters in broadcast journalism,” anchors the show. He stands upright, against a backdrop of laboratory research images; he speaks with a clear and authoritative voice: “We hope one day to lead the broadcast with a cure for cancer, but tonight, we might have the next best thing” (Fig. 1). “The treatment is audacious, using poliovirus to kill glioblastoma, a vicious brain cancer that can kill in a matter of months,” Pelley states. The program cuts to a young woman who, as told, was the first patient with brain cancer to volunteer in the clinical trial of an experimental treatment at Duke University. The newscaster reviews her case quickly, as a doctor might on rounds. In 2011, a 20-year-old nursing student experienced headaches; doctors found a brain tumor “the size of a tennis ball” and removed 98% of it in surgery; in 2012, the patient had recurrent glioblastoma. 60 Minutes first reported on the experimental brain cancer treatment in March 2015.4 “Something unimaginable happened,” Pelley says in the 2016 segment. Her tumor “shrank for 21 months, until it was gone.” He shows the patient’s MRI to television viewers and explains that it no longer reveals a brain tumor. She and at least two other participants in the phase I trial were doing well and in complete remission for over 3 years, Pelley reported. Critiques of the 60 Minutes and CBS Evening News coverage of the polio-derived brain cancer vaccine appeared at

From Weill Cornell Medicine, New York, NY. Disclosures of potential conflicts of interest provided by the author are available with the online article at asco.org/edbook. Corresponding author: Elaine Schattner, MD, Division of Hematology and Oncology, Department of Medicine, Weill Cornell Medicine, 1300 York Ave., New York, NY 10021; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 3

ELAINE SCHATTNER

FIGURE 1. Scott Pelley, CBS Evening News, May 12, 2016

Related video available here: www.cbsnews.com/news/promising-duke-university-polio-braincancer-trial-given-breakthrough-status-60-minutes/. Accessed February 2017.

Forbes.com5,6 (where I am a contributor) and HealthNewsReview.org, a health journalism watchdog site.7 Criticisms included failure of Pelley’s team to spell out alternative standard and experimental treatments for glioblastoma, no mention of costs, a heavy-weighting of views put forth by those involved with the trial at Duke, a lack of emphasis on the toxicity and deaths experienced by most of the research participants, and the use of the term breakthrough. The harshest piece8 appeared in MedPage Today: “Brain Cancer: Did ‘60 Minutes’ Report Raise False Hope?” The original 60 Minutes feature led to a slew of phone calls to the brain cancer team at Duke University, according to the MedPage story. A later piece9 published by The Hill, “‘60 Minutes’: FDA Fast Tracks Cancer Treatment Using Polio Virus,” reflects perception that television coverage accelerated the experimental trial and vaccine treatment. As told in the later CBS Evening News segment, the investigational agent is being developed by a company based at Duke and involves a scientist interviewed on the program. The clip illustrates the challenges of discerning cancer progress from hype and news from advertisem*nts. When I rewatched this episode on my personal computer in February 2017, an advertisem*nt for a cancer treatment center kicked in. Yet I found the story compelling and valuable, overall; I wanted to know more about the brain cancer vaccine being tested at Duke University, a major academic medical center. Moreover, if I knew an otherwise healthy person with recurrent glioblastoma, I would want them to be aware of this promising treatment, in case they were seeking experimental options. A possible downside of this kind of story is that patients might have their hopes raised about entering the trial, only to find out that they are ineligible. On the plus side, for journalists to provide well-researched stories like this about the vaccine for glioblastoma, hearing of progress, however preliminary and carefully worded, might be comforting to people who have lost loved ones to the condition. One reason the CBS story drew some flak may be that it begins with the term promising in the headline. That word—like breakthrough or miracle in a story having to 4 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

do with cancer—causes journalists, and doctors, to bristle. Some teachers of health care journalism advise these hopeful words be avoided,10 for similar reasons that oncologists instruct younger doctors not to tell a cancer patient they have been cured, but rather to say they are in remission. The problem with using such optimistic terms is they fail to prime the reader, or patient, for disappointment. A research letter11 published in JAMA Oncology, “The Use of Superlatives in Cancer Research,” suggests that excessively positive language appears with undue frequency in journalism about cancer. The article, based on a Google search of terms in news published over 5 days in late June 2015, generated a blitz of coverage in late October 2015, when the paper appeared online. The story resonated, at least among journalists. “Half of the cancer drugs journalists called ‘miracles’ and ‘cures’ were not approved by the FDA,” said Vox.com.12 In a syndicated piece Reuters stated that “Glowing terms are often used for new cancer drugs in health news."13 “If A New Cancer Drug Is Hailed As A Breakthrough, Odds Are It's Not,” stated NPR Health.14 “‘Revolutionary.’ ‘Game changer.’ ‘Miracle.’ How much are we hyping unproven cancer drugs?” asked The Washington Post.15 A March 2017 STAT News opinion16 by two oncology physicians, one of whom is the corresponding author of the original piece, reviews the JAMA Oncology report on superlative language. That column, titled “Few People Actually Benefit From ‘Breakthrough’ Cancer Immunotherapy,” refers to “an ocean of hype” and links to another negative report.17 The message is clear: do not believe promising headlines about oncology drugs. But what if scientific and clinical advances have led to considerable gains for people with cancer? If progress against disease is real, as it may or may not be, a question for journalists is whether skepticism might be flipped: perhaps the current truth is not so bleak.

FAILING TO REPORT PROGRESS THAT IS SLOW, INCREMENTAL, AND IMPERFECT

Between 1991 and 2014, the death rate from cancer fell by 25% in the United States. This impressive figure headlined a report by the American Cancer Society (ACS), “Cancer Statistics, 2017,” published on January 5, 2017.18 In compiling this update, epidemiologists and statisticians drew on mortality data gathered by the National Center for Health Statistics. They also reviewed cancer incidence and survival data from the Surveillance, Epidemiology, and End Results (SEER) Program, the National Program of Cancer Registries, and the North American Association of Central Cancer Registries. Yet a casual survey of my acquaintances who are not oncologists confirmed that some educated people, individuals who read print newspapers and listen to National Public Radio, for instance, remained completely unaware of this favorable trend. A recent search of the New York Times website using terms like “cancer deaths” and “American Cancer Society” finds no mention of the January 2017 ACS report on the 25% decline in U.S. cancer mortality. The Wall Street Journal also appears to have passed on covering this story.

CAN CANCER TRUTHS BE TOLD? CHALLENGES FOR MEDICAL JOURNALISM

Several national news outlets did pick up the ACS report. TIME.com published a short piece,19 “Here’s Why the Cancer Death Rate Has Plummeted.” CNN.com ran a story,20 “US Cancer Deaths Down 25% Since 1991, Report States,” but did not produce searchable video or television coverage. Like CNN, NBC News posted an article21 on its website, but does not appear to have supported it with video or television footage. USA TODAY did not cover the analysis directly, but posted a 47-second video,22 “2017 Is Looking Healthier: Cancer Death Rate Drops a Fourth Since '91.” A caption at USAToday.com attributes the video to Newsy NewsLook, a company23 that provides a “premium video solution that increases views and revenue for Publishers and Creators.” In other words, the USA TODAY story on the cancer decline was produced by a commercial video manufacturer.

FIGURE 2. U.S. Cancer Mortality Trends (19912014)

SPINNING STATISTICS

The 25% statistic raises questions (Fig. 2). A critical reader or journalist might ask if the reported reduction in deaths from cancer was observed only in people already known to have cancer (disease-specific mortality) or if the trend was observed in the larger U.S. population. For this ACS analysis, cancer deaths were tallied in the general population. Specifically, deaths from cancer peaked at 215.1 per 100,000 population in 1991 before falling to 161.2, also per 100,000 population, in 2014. That detail means that the lower reported death rate from cancer cannot be attributed to overdiagnosis. Overdiagnosis would affect the number of reported cancer cases (incidence) and might lower the apparent rate of cancer-specific deaths, but it would not affect mortality from cancer in the general population. The figure persuades because the mortality curve falls steadily between 1991 and 2014; it is not a statistical fluke. U.S. cancer death rates fell for both men and women, although in the past decade, the decline was proportionately greater in men. The ACS authors attributed the trend to reduced smoking, which over time led to fewer lung cancers, and to early detection and treatment of several common cancer forms. The biggest declines were observed in mortality from lung, breast, prostate, and colorectal cancers. Yet the drop is imperfect. Cancer remains the second-leading cause of death in U.S. men and women. Disparities in cancer incidence and deaths persist, based on race, insurance access, and geography. The ACS reported that death rates from cancer among U.S. blacks exceed those among whites by 15% overall. A detail worth considering, because it confounds health statistics and public understanding of those, is the expanding and dynamic U.S. population. Although the cancer death rate fell, the Centers for Disease Control and Prevention reported a rising annual number of cancer deaths, from 514,657 deaths in 1991 to 591,699 deaths in 2014.24 There is no discrepancy in these figures; the absolute increase in cancer deaths can be explained by growth of the population. Based on U.S. Census Bureau estimates,25,26 the population expanded from 252,131 million in 1991 to approximately 318,700 million in 2014; these figures demonstrate

over 26% growth in the U.S. population during the relevant 23-year interval. This sort of apparent contradiction, based on two valid representations of U.S. cancer registry and death data, generates confusion and, sometimes, angry public debate. A relevant example comes from the surprisingly controversial subject of breast cancer statistics.27 A contentious issue that crops up periodically in news, op-eds, on Twitter and advocacy group Facebook pages, is whether or not there has been meaningful progress in reducing breast cancer deaths.28 The facts, based on SEER data, are these: in 1991, deaths from breast cancer numbered 32.69 per 100,000 women in the U.S. population; in 2013, those numbered 20.72, per 100,000 U.S. women.29 These numbers demonstrate a 36% decline in the rate of deaths from breast cancer. Yet the absolute number of deaths declined only slightly during those 22 years, by a few thousand. The Centers for Disease Control and Prevention report that in 1991, deaths from breast cancer numbered 43,583; in 2013, 40,860 women and 464 men died of breast cancer.30,31 Although some might deem these statistical quibbles, these figures affect how and if journalists and others represent and perceive progress against breast cancer. These numbers influence the distribution of funding for research, screening, and care. The breast cancer death rate has declined by 36% in the U.S. population, but the annual U.S. toll of deaths from breast cancer still hovers over 40,000, causing consternation and frustration among patients, advocates, and others. Further clouding the picture is that invasive breast cancer is the most common cancer form not declining in the United States; the recent ACS report indicates asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 5

ELAINE SCHATTNER

that among some groups, such as African Americans, the incidence has been rising.

FOCUS ON DISPARITIES (WHERE INFORMATION MIGHT HELP OUTCOMES)

The fact that cancer death rates have been declining, albeit unevenly, surfaced in a January 2017 JAMA analysis.32 That paper confirms the drop in overall U.S. mortality from cancer after 1980 and highlights disparities. Cancer pockets—marked by a high incidence or death rate from malignancy—exist in broad U.S. geographical areas and small communities. Lung cancer disproportionately affects and kills people in Appalachia, for instance. A high death rate from breast cancer occurs along the lower Mississippi and southern belt states. Clusters, or hot spots, of kidney cancer appear in areas along the Mississippi and in North and South Dakota. Stage at diagnosis and cancer death rates vary among counties within states. These regional differences in U.S. cancer outcomes support the need for improved education and journalism about health. In 2015, Newsweek dedicated an entire issue to cancer. In a long-form feature,33 journalist Jessica Wapner reveals the outlook for patients with cancer in central Appalachia. She visited a region of eastern Kentucky where prevalent poverty and low education levels contribute to cancer’s high toll. People in the area suffer health effects from heavy smoking and excess particulate matter in air. “A long history of poverty and disease in the region has led to a sense of resignation, a fatalistic belief about the inevitability of cancer and the death it brings,” Wapner wrote. “Many people who are diagnosed refuse treatment because they don’t see the point of going through the pain.” In the context of recent data about the declining U.S. cancer death rate, the Newsweek story points to the potential role of news to inform people about progress and nudge them toward better outcomes. Careful journalism might disrupt a fatalistic cycle of disbelief in cancer treatment’s value that leads to late presentation of patients with cancer to physicians, lesser outcomes, and more deaths. Consider the plight of a woman or man living in rural Kentucky with a persistent cough and weight loss, symptoms of a possible lung cancer, today. Just knowing that cancer deaths in the U.S. population are down, by as much as 25% in recent years, might prompt some individuals to visit the doctor rather than ignore early signs of disease.

DESPITE PROGRESS, A DIM AND CONFUSING PICTURE

The recent 25% decline in U.S. cancer deaths—what might be deemed as evidence-based news about cancer that is not anecdotal—reflects progress. Yet it got little attention. Of course, with so much ongoing political changes in January 2017, including the possible repeal of the Affordable Care Act and the affect of the travel ban on doctors, the omission of cancer news from headlines might be understood. When newsrooms are strapped for reporters and editors need to choose stories that draw clicks, careful reporting about cancer 6 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

research, drugs, and clinical developments might be put aside (Sidebar 1). Yet reports about cancer appear constantly and often cast a negative slant. Many recent articles focus on the exorbitant costs of cancer medications. Some stories feed on anger over high drug prices, raise or address economic issues, and mix with political news. Although reports on treatment costs might be set apart from reports on effectiveness, the issues often get conflated, in part because economists and health policy experts discuss pricing models that depend on how well the drugs reportedly work. Since January 2017, overlapping articles have emphasized the ineffectiveness of oncology treatments. One example, “Dozens of New Cancer Drugs Do Little to Improve Survival, Frustrating Patients,”34 appeared in Kaiser Health News on February 9, 2017 (Fig. 3). The piece, also available in Spanish (Nuevas drogas contra el cáncer, ¿ayudan a vivir más?) draws on a lecture on “Unintended Consequences of Expensive Cancer Therapeutics” published in 2014 in JAMA Otolaryngology–Head & Neck Surgery35 and a few recent papers finding marginal, if any, benefit of new oncology drugs. The story leads with the picture of a woman whose breast cancer progressed through multiple treatments, causing pain. It refers to high prices, averaging $171,000 per year, and relies heavily on negative quotes offered by critics of precision oncology. The same article ran in USA TODAY with an abbreviated headline,36 “Dozens of New Cancer Drugs Do Little to Improve Survival,” and on CNN.com,37 “Amid Flurry of New Cancer Drugs, How Many Offer Real Benefits?”

SIDEBAR 1. Pressures on the Quality of Medical Journalism 1. Newsrooms reduce staff for reporting and editing. 2. Many outlets lack fact-checkers. 3. To keep their jobs, reporters need turn out stories quickly. 4. Journalists may lack the time or scientific back ground to critically evaluate reports of new technology and drug development. 5. Income is often incentivized by internet traffic (clicks). 6. Even for news outlets that do not acknowledge paying journalists based on traffic, a freelancer is more likely to get repeat assignments after stories “fly”; reporters and columnists lose jobs when clicks are insufficient. 7. Editors favor topics that drive traffic: stories with splashy headlines on financial toxicity, greedy pharmaceutical companies, and bad doctors gain disproportionate coverage. 8. Journalists hesitate before covering “break throughs,” as they should, for not wanting to seem foolish.

CAN CANCER TRUTHS BE TOLD? CHALLENGES FOR MEDICAL JOURNALISM

In addition to dismissing the value of cancer drugs, recent health news casts doubt on the reliability of medical research. In January 2017, the BBC highlighted the reproducibility crisis with the headline “Most Scientists ‘Can't Replicate Studies by Their Peers.’”38 The journal Nature covers the Reproducibility Initiative, a project funded by the Laura and John Arnold Foundation that aims to replicate key findings in basic cancer research.39 In early March 2017, NPR Health ran a related piece,40 “Reports Of Medical Breakthroughs Often Don't Prove Out.” The triple takeaway might be that new cancer drugs rarely work, cost lots, and that reports of progress cannot be trusted. Yet the well-documented pattern of reduced U.S. cancer mortality supports that modern oncologists are doing something right overall. Perhaps the day-to-day medical news, with a focus on narrative and twists, and only occasional details about treatments, fails to capture the big picture about cancer and incremental progress.

LOSS OF INFORMATION GATEKEEPERS

Thirty years ago, when I graduated from medical school, someone wanting to distribute a factual update, opinion, or

analysis generally needed access to a publisher with printing equipment or a company with radio or TV broadcasting equipment. This is no longer the case. Today a doctor, high school student, refugee, patient with cancer, teacher, celebrity—anyone—can write a few lines, take a photo, or film something and post it to the web. Social media posts, along with other sources of cancer-related information—ranging from direct-to-consumer advertisem*nts sponsored by pharmaceutical companies, to articles put forth by cancer centers on fancy websites, to blog posts authored by individual clinicians—contribute to what might be described as an online free-for-all regarding cancer facts, treatments, and opinion. The expanding volume of stories and data about cancer offers patients and doctors an unprecedented amount of material to sort through. Distilling what is true, relevant, and helpful to an individual patient may be more difficult than ever before. The disruptive and potentially helpful impact of Twitter, a social media platform, is hard to gauge in terms of clinical care and cancer outcomes. Although the numbers of oncologists, patients with cancer, advocates, researchers, and communications specialists representing pharmaceutical

FIGURE 3. Kaiser Health News: “Treating Cancer: Hope vs. Hype"

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 7

ELAINE SCHATTNER

companies, hospitals, and pathology laboratories using social media are rising, the consequences of all this activity remain unclear. Preliminary reports about Twitter and health care cannot be generalized, because they draw on data from the platform’s users. Twitter does facilitate rapid transmission of health news. This could be most helpful during a public health emergency. How it might help patients with cancer and their caregivers is by directing those with a condition or interest, like sarcoma, to relevant news such as the U.S. Food and Drug Administration approval of a new drug, clinical trials, conferences, and websites providing vetted information.

POTENTIAL INFLUENCE OF CANCER NEWS ON PUBLIC HEALTH

Although the clinical and public health impact of most health journalism goes unchecked, some well-documented instances demonstrate the alerting function of high-profile stories. Some of the best-studied examples pertain to oncology and cancer screening and date over several decades.41 In September 1974, First Lady Betty Ford had a mastectomy for breast cancer. Ford’s surgery took place a few weeks before Happy Rockefeller, wife of Vice President Nelson Rockefeller, underwent the same.42 News surrounding the pair’s procedures and their malignant diagnoses generated an uptick in mammography. A 1978 study43 in Public Health Reports found a clear pattern, albeit transient, lasting months, of increased participation in the Health Insurance Plan screening program during that “period of high public attention to breast cancer.”43, p. 320 The authors referred to an “alerting function” of high-profile news; they attributed the term to Earl Ubell, the director of NBC TV News. The newscaster had penned a rare 1976 paper44 on “Responsibility of the Mass Media in the Control of Sexually Transmitted Diseases.” A decade later, President Ronald Reagan underwent successful surgery to remove a small cancer from his colon in the early summer of 1985. The National Cancer Institute reported a spike in calls to its Cancer Information Service from people with questions about colon and rectal cancers; SmithKline Diagnostics, a large manufacturer of a screening kit to check for blood in stool, reported that its supply ran out.45 Although Reagan’s episode did not have any measured effect on public health, such as reduced deaths from colon cancer, his well-publicized case did influence attitudes about colon and rectal cancer. The Los Angeles Times discussed the impact of Reagan’s surgery with Irving Rimer of the American Cancer Society: “The taboo against talking about colon and rectal cancer, about the elimination of wastes from the body, and about the bowels in general has been broken,” Rimer said in July 1985.46 Over time, public cancer disclosures became increasingly frequent. In March 2000, Today Show host Katie Couric gained attention by undergoing a live-televised colonoscopy. After her husband’s death at age 42 from colon cancer, Couric explicitly aimed to encourage screening. Time mag8 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

azine called it “Katie’s Crusade.”47 The strategy worked; in 2003, physicians documented a rise in colonoscopies, particularly among women. Doctors dubbed it “the Couric effect.” Those reporting in the Archives of Internal Medicine48 inserted a note of caution: “While celebrity spokespersons have remarkable potential to transmit important medical information, one notable concern is the possibility for well-meaning public figures to use their influence to promote unproven or even dangerous behaviors,”48, p. 1604 they wrote. “For example, Ms. Couric has advocated colo rectal screening at ages younger than recommended by most medical authorities, declaring, ‘But all the doctors I know—and I know a lot of them—say they had or will get a colonoscopy by their 40th birthday. That ought to tell you something,’” the journal authors added. “Considering these results, celebrity spokespersons should be advised to deliver carefully targeted, evidence-based recommendations that will ultimately improve public health.” A more recent example comes from the double revelation of Angelina Jolie. In May 2013, the actress informed the world that she carries a BRCA mutation in a New York Times op-ed.49 The article emphasized her personal decision to undergo prophylactic, bilateral mastectomy in light of her genetic disposition and strong maternal family history of breast cancer. Her story surely contributed to the increase in BRCA genetic evaluations that ensued. In 2015, Jolie followed up with a column about her oophorectomy.50 Yet a 2016 report in the British Medical Journal found no rise in mastectomy after Jolie’s revelation.51 Her case, and the enormous publicity surrounding her decision, generated greater awareness about BRCA and, possibly, other hereditary cancer syndromes; it led to more DNA testing, but it appears not to have affected surgery rates. Two prominent 2016 cases stand out by their possible consequences for men’s health. The actor Ben Stiller disclosed that he was treated for prostate cancer found 2 years earlier, at age 48. He wrote a controversial blog post52 on Medium: “The Prostate Cancer Test That Saved My Life.” Stiller gave interviews on TV and radio about his prostate cancer evaluation, surgery, and treatment.53 However, because the U.S. Preventive Services Task Force and physician groups advise against prostate cancer screening in men, Stiller got into some hot water over his post and remarks on prostate-specific antigen testing.54,55 Also, in 2016, the Black Eyed Peas musician Jaime Gomez, known as Taboo, announced he had surgery and chemotherapy for testicular cancer.56,57 The ACS named him a global ambassador. Gomez aims to increase awareness and lessen stigma about cancer in his Mexican and Native American communities.58 These cases reveal the potential of cancer news to influence public health. Celebrity health disclosures can help or harm, depending on how effectively, and if, their messages convey medical wisdom. However, most of what people hear about oncology has little to do with celebrities’ experiences. The background effect of everyday stories likely has a greater effect on patients’ decisions, but it may be impossible to measure.

CAN CANCER TRUTHS BE TOLD? CHALLENGES FOR MEDICAL JOURNALISM

A PRESCRIPTION FOR CANCER NEWS

One might consider if and why medical journalism matters. Although a physician or patient might enjoy reading a feature on immune therapy or listening to a podcast on ethical or technical aspects of genetic testing, few individuals would make treatment choices based on what they’ve read in the Atlantic or seen on CNN. Yet exposure to news—what journalists are writing in papers and magazines and saying on radio, TV, and social media—can influence a person’s background view, so that when they enter a physician’s office, they know what questions to ask; a patient might be more wary of an intervention or more willing to accept it. Health news affects whether a person enters a doctor’s office in the first place. I would suggest that good-quality health journalism might be more needed than ever before. As the number of competing online sources of information expands, and patients know their personal doctors less well, the potential consequences of news about oncology, and how that news is steered by editors and social media, will affect cancer patients’ decisions, experiences, and outcomes. Despite progress, negative stories constitute much of what people hear about cancer: patients suffer; many die after treatments fail; medications cost too much and can cause bankruptcy; survivors endure long-term side effects and chronic health problems; researchers fail to reproduce findings; bad oncologists carry out fraudulent billing, etc. News of treatment toxicity, such as recent reports about cardiac effects of oncology drugs,59-61 might scare a patient, so that they decline treatment that is likely to help. Reports on chemobrain, recently substantiated,62 could dissuade anyone from taking the medicines an oncologist recommends. Of course, it is every person’s right to have this kind of information: the good, the bad, and the mixed results. News, presented in a balanced way, could help guide patients and caregivers about the risks and benefits of treatment options. Journalism can inform patients’ decisions about whether to try medicines, whether and when to accept consultation from a palliative care specialist, or choose hospice care. Although premature reports of groundbreaking findings in mice or breakthroughs in the laboratory can mislead, the public deserves to know about advances. The truth includes progress. Producing balanced stories that convey information about progress against cancer, without hype, tasks journalists. Transparency will serve them and their audience: physicians and scientists need be upfront about conflicts of interest and funding; recognize and indicate the limits of conclusions from any study; and be open to correction. What journalists can do, although it is not easy, is to seek varied perspectives. Incorporating viewpoints of scientists, physicians, patients, and others, including some who are not directly involved in a story, should add depth and generally improve coverage (Sidebar 2). As newsrooms shrink and reporters work at a quicker pace, the challenge of producing balanced stories that convey information about real progress against cancer, without hype, may push editors away from the subject. Unless the

SIDEBAR 2. Possible Solutions for Health Journalism 1. Be transparent at all levels of reporting; reveal funding and other conflicts of interest that may influence physicians and scientists in academics and elsewhere, patients who may have organizational or industry ties, journalists, and publishers of news. 2. Reporters should seek input from varied sources. 3. Educate journalists about math, molecular biology, and statistics. 4. Independent health foundations might support in-depth coverage of advances. 5. Patient-driven “advocacy journalism” could help distribute critical information. 6. Increase public access to real-time and anonymized clinical data; this would enable constant reevaluation of published work in context of new information. 7. Consider innovative systems of weighted com mentary, culled from social media and other sources, to develop post-publication peer review of medical reports. data for a treatment or experimental results are so extraordinary that they astonish seasoned oncologists, so much that they use terms like breakthrough or possible cure regarding previously hopeless tumors, journalists may think it wise to play it safe, skip coverage, and report on something else. The reality of incremental progress is unfortunately dull, except, of course, to the patients who experience these advances, their loved ones, and providers of care, including physicians who see them do well. Some journalists might consider, as I have, that detailed information about cancer treatments belongs in doctors’ offices and journals and not in the news. However, if people remain uninformed of trends, they may remain ignorant of the big picture. Even doctors who are not specialists may not be aware of some advances that have occurred in the past decade against metastatic lung cancer, melanoma, and a growing list of previously hopeless tumors. There is a need for good-quality news about cancer, especially in communities in which cancer mortality remains disproportionately high. There is no easy prescription for distilling truth in oncology news. But I am hopeful, as journalists grapple with a changing pace of work and staff and as doctors contend with a changing pace of practice, that better education—of physicians and the public—about cancer, and in basic math, statistics, biology, and other fields relevant to oncology, will prove helpful. For journalists and for doctors, knowing how to interpret, convey, and interpret fluid information, is crucial for public health.

ACKNOWLEDGMENT

I would like to thank Carol DeSantis, of the American Cancer Society, for creating a figure for this article, and the ASCO Edbook staff, for their assistance in preparing this manuscript. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 9

ELAINE SCHATTNER

References 1. Gottfried J, Shearer E. “News Use Across Social Media Platforms 2016.” www.journalism.org/2016/05/26/news-use-across-social-mediaplatforms-2016/. Accessed March 9, 2017.

17. Begley S. “Beware the hype: Top scientists cautious about fighting cancer with immunotherapy.” www.statnews.com/2016/09/25/cancerimmunotherapy-caution/. Accessed March 9, 2017.

2. Barthel M. “Newspapers: Fact Sheet.” www.journalism.org/2016/06/15/ newspapers-fact-sheet/. Accessed March 9, 2017.

18. Siegel RL, Miller KD, Jemal A. Cancer Statistics, 2017. CA Cancer J Clin. 2017;67:7-30.

3. CBS News. “Promising brain cancer trial given breakthrough status by FDA.” www.cbsnews.com/news/promising-duke-university-poliobrain-cancer-trial-given-breakthrough-status-60-minutes/ (video: www.cbsnews.com/videos/bold-cancer-treatment-grantedbreakthrough-status). Accessed March 9, 2017.

19. Oaklander M. “Here’s Why the Cancer Death Rate Has Plummeted.” http://time.com/4622645/cancer-death-rates/. Accessed March 9, 2017.

4. Pelley S. “Killing Cancer.” www.cbsnews.com/news/polio-cancertreatment-duke-university-60-minutes-scott-pelley/. Accessed March 9, 2017.

21. Fox M. Cancer Deaths Fell 25 Percent Since 1991. www.nbcnews. com/health/cancer/cancer-deaths-fell-25-percent-1991-n703651. Accessed March 9, 2017.

5. Kroll D. “What ‘60 Minutes’ Got Right and Wrong On Duke’s Polio Virus Trial Against Glioblastoma.” www.forbes.com/sites/ davidkroll/2015/03/30/60-minutes-covers-dukes-polio-virus-clinicaltrial-against-glioblastoma/. Accessed March 9, 2017.

22. NewsLook. “2017 is looking healthier: Cancer death rate drops a fourth since '91.” www.usatoday.com/videos/news/2017/01/06/2017looking-healthier-cancer-death-rate-drops-fourth-since-'91/ 96238732/. Accessed March 9, 2017.

Weintraub A. “Here's What ‘60 Minutes’ Didn't Tell You About The 6. ‘Miracle’ Glioblastoma Treatment.” https://www.forbes.com/sites/ arleneweintraub/2015/03/30/heres-what-60-minutes-didnt-tell-youabout-the-miracle-glioblastoma-treatment/. Accessed March 9, 2017.

NewsLook. www.newslook.com. Accessed March 9, 2017. 23.

Watkins T. “More than half of 60 Minutes devoted to ‘Killing Cancer’ 7. but still no independent perspective.” www.healthnewsreview. org/2015/03/more-than-half-of-60-minutes-devoted-to-killingcancer/. Accessed March 9, 2017. Yurkiewicz S. “Brain Cancer: Did ‘60 Minutes’ Report Raise False Hope?” 8. www.medpagetoday.com/hematologyoncology/braincancer/50728. Accessed March 9, 2017. 9. Rupert E. “‘60 Minutes’: FDA fast tracks cancer treatment using polio virus.” http://thehill.com/blogs/blog-briefing-room/news/27998160-minutes-fda-fast-tracks-cancer-treatment-using-polio-virus. Accessed March 9, 2017.

20. Scutti S. “US cancer deaths down 25% since 1991, report says.” www.cnn. com/2017/01/06/health/cancer-death-stats-2017/. Accessed March 9, 2017.

Heron M, Anderson RN. Changes in the Leading Cause of Death: 24. Recent Patterns in Heart Disease and Cancer Mortality. National Center for Health Statistics. Data Brief No. 254. www.cdc.gov/nchs/ data/databriefs/db254_table.pdf#1. Accessed March 9, 2017. Byerly E, Deardorff K. National and State Population Estimates: 1990 to 25. 1994. www.census.gov/prod/1/pop/p25-1127.pdf. Accessed March 9, 2017. U.S. Census Bureau, Population Division Annual Estimates of the 26. Resident Population: April 1, 2010 to July 1, 2016 https://factfinder. census.gov/faces/tableservices/jsf/pages/productview.xhtml?. Accessed March 9, 2017. Mulcahy N. “The Mystery of a Common Breast Cancer Statistic.” www. 27. medscape.com/viewarticle/849644. Accessed March 9, 2017.

Schwitzer G. “7 Words (and more) You Shouldn’t Use in Medical 10. News.” www.healthnewsreview.org/toolkit/tips-for-understandingstudies/7-words-and-more-you-shouldnt-use-in-medical-news/. Accessed March 9, 2017.

Schattner E. “Raising The Survival Bar, And Access To Information, On 28. Metastatic Cancer.” www.forbes.com/sites/elaineschattner/2016/05/13/ raising-the-survival-bar-and-access-to-information-on-metastaticcancer/. Accessed March 9, 2017.

11. Abola MV, Prasad V. The Use of Superlatives in Cancer Research. JAMA Oncol. 2016;2:139-141.

29. Howlader N, Noone AM, Krapcho M, et al (eds). SEER Cancer Statistics Review, 1975-2013. http://seer.cancer.gov/csr/1975_2013/. Accessed March 9, 2017.

12. Belluz J. “Half of the cancer drugs journalists called ‘miracles’ and ‘cures’ were not approved by the FDA.” www.vox.com/2015/10/29/9637062/ media-hype-cancer-drugs. Accessed March 9, 2017. 13. Seaman AM. “Glowing terms often used for new cancer drugs in health news.” www.reuters.com/article/us-health-news-terms-canceridUSKCN0SN2OT20151029. Accessed March 9, 2017. 14. Bichell RE. “If A New Cancer Drug Is Hailed As A Breakthrough, Odds Are It’s Not.” http://www.npr.org/sections/health-shots/2015/ 10/29/452610534/if-a-new-cancer-drugs-hailed-as-a-breakthroughodds-are-its-not. Accessed March 9, 2017. 15. Dennis B. “‘Revolutionary.’ ‘Game changer.’ ‘Miracle.’ How much are we hyping unproven cancer drugs?” www.washingtonpost.com/news/toyour-health/wp/2015/10/29/revolutionary-game-changer-miracle-howmuch-are-we-hyping-unproven-cancer-drugs/. Accessed March 9, 2017. 16. Gay N, Prasad V. “Few people actually benefit from ‘breakthrough’ cancer immunotherapy.” www.statnews.com/2017/03/08/immunotherapycancer-breakthrough/. Accessed March 9, 2017.

10 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

30. Centers for Disease Control and Prevention. Deaths from Breast Cancer–United States, 1991. MMWR Weekly. 1994;43:273,279281. 31. Centers for Disease Control and Prevention. Breast Cancer Statistics. www.cdc.gov/cancer/breast/statistics/index.htm. Accessed March 9, 2017. 32. Mokdad AH, Dwyer-Lindgren L, Fitzmaurice C, et al. Trends and Patterns of Disparities in Cancer Mortality Among US Counties, 19802014. JAMA. 2017;317:388-406. 33. Wapner J. “The Cancer Epidemic in Central Appalachia.” www. newsweek.com/2015/07/31/cancer-epidemic-central-appalachia354857.html. Accessed March 9, 2017. 34. Szabo L. “Dozens of New Cancer Drugs Do Little To Improve Survival, Frustrating Patients.” http://khn.org/news/dozens-of-new-cancerdrugs-do-little-to-improve-survival-frustrating-patients/. Accessed March 9, 2017.

CAN CANCER TRUTHS BE TOLD? CHALLENGES FOR MEDICAL JOURNALISM

35. Fojo T, Mailankody S, Lo A. Unintended consequences of expensive cancer therapeutics—the pursuit of marginal indications and a metoo mentality that stifles innovation and creativity: the John Conley Lecture. JAMA Otolaryngol Head Neck Surg. 2014;140:1225-1236. 36. Szabo L. “Dozens of new cancer drugs do little to improve survival.” http://www.usatoday.com/story/news/nation/2017/02/09/newcancer-drugs-do-little-improve-survival/97712858/. Accessed March 9, 2017. 37. Szabo L. “Amid flurry of new cancer drugs, how many offer real benefits?” http://www.cnn.com/2017/02/09/health/hope-vs-hypecancer-drugs-partner/. Accessed March 9, 2017. 38. Feilden T. “Most scientists ‘can’t replicate studies by their peers’.” http://www.bbc.com/news/science-environment-39054778. Accessed March 9, 2017. 39. Baker M, Dolgin E. “Cancer reproducibility project releases first results.” http://www.nature.com/news/cancer-reproducibility-projectreleases-first-results-1.21304. Accessed March 9, 2017. 40. Harris R. “Reports Of Medical Breakthroughs Often Don’t Prove Out.” http://www.npr.org/sections/health-shots/2017/03/06/518802242/ reports-of-medical-breakthroughs-often-dont-prove-out. Accessed March 9, 2017. Lerner Barron H. When Illness Goes Public: Celebrity Patients and How 41. We Look at Medicine. Baltimore, Maryland: Johns Hopkins University Press; 2006. 352. Klemesrud J. “After Breast Cancer Operations A Difficult Emotional 42. Adjustment.” http://www.nytimes.com/1974/10/01/archives/ after-breast-cancer-operations-a-difficult-emotional-adjustment. html. Accessed March 9, 2017. 43. Fink R, Roeser R, Venet W, et al. Effects of news events on response to a breast cancer screening program. Public Health Rep. 1978;93:318327. 44. Ubell E. The responsibility of the mass media in the control of sexually transmitted diseases: a hammer without a nail. Bull N Y Acad Med. 1976;52:1019-1036. 45. Brown ML, Potosky AL. The presidential effect: the public health response to media coverage about Ronald Reagan’s colon cancer episode. Public Opin Q. 1990;54:317-329. Maugh TH II. “Reagan’s Surgery for Colon Cancer Breaks a Taboo, 46. Brings a Floodtide of Calls.” http://articles.latimes.com/1985-07-27/ local/me-6335_1_colon-cancer. Accessed March 9, 2017. 47. Gorman C. “Katie’s Crusade.” http://content.time.com/time/ magazine/article/0,9171,996315,00.html. Accessed March 9, 2017. 48. Cram P, Fendrick AM, Inadomi J, et al. The impact of a celebrity promotional campaign on the use of colon cancer screening: the Katie Couric effect. Arch Intern Med. 2003;163:1601-1605.

49. Jolie A. “My Medical Choice.” www.nytimes.com/2013/05/14/ opinion/my-medical-choice.html. Accessed March 9, 2017. 50. Jolie-Pitt A. “Angelina Jolie Pitt: Diary of a Surgery.” www.nytimes. com/2015/03/24/opinion/angelina-jolie-pitt-diary-of-a-surgery. html. Accessed March 9, 2017. 51. Desai S, Jena AB. Do celebrity endorsem*nts matter? Observational study of BRCA gene testing and mastectomy rates after Angelina Jolie’s New York Times editorial.BMJ2016;355:i6357. 52. Stiller B. “The Prostate Cancer Test That Saved My Life.” https:// medium.com/cancer-moonshot/the-prostate-cancer-test-thatsaved-my-life-613feb3f7c00#. Accessed March 9, 2017. Schattner E. “Ben Stiller and Howard Stern Talk About Prostate 53. Cancer Screening. Great!” www.forbes.com/sites/elaineschattner/ 2016/10/06/ben-stiller-and-howard-stern-talk-about-prostatecancer-screening-great/. Accessed March 9, 2017. 54. Middlebrook H. “Ben Stiller: Prostate cancer test ‘saved my life.’” www.cnn.com/2016/10/04/health/ben-stiller-prostate-cancer/. Lomangino K. “Ben Stiller’s misguided prostate cancer recommendations 55. aren’t based on evidence.” www.healthnewsreview.org/2016/10/benstiller-prostate-cancer/. Accessed March 9, 2017. 56. “Black Eyed Peas Member Taboo Reveals Cancer Secret.” www. thedoctorstv.com/articles/3584-drs-exclusive-black-eyed-peasmember-taboo-reveals-cancer-secret. Accessed March 9, 2017. 57. Gomez P. “Inside The Black Eyed Peas Star Taboo’s Private Health Crisis: How He Beat Cancer Out of the Public Eye.” http://people.com/ music/black-eyed-peas-taboo-cancer-exclusive/. Accessed March 9, 2017. 58. Schattner E. “Taboo Of The Black Eyed Peas Has A Message For Cancer Patients.” www.forbes.com/sites/elaineschattner/2016/11/17/taboothe-black-eyed-peas-musician-has-a-message-for-cancer-patients/. Accessed March 9, 2017. 59. Schattner E. “Heart Health After Cancer: A Growing Concern.” www. forbes.com/sites/elaineschattner/2015/02/23/heart-health-aftercancer-a-growing-need-for-care-and-research-in-cardio-oncology/. Accessed March 9, 2017. 60. Grady D. “Lifesaving Cancer Drugs May in Rare Cases Threaten the Heart.” www.nytimes.com/2016/11/03/health/cancer-drugs-heartrisks.html. Accessed March 9, 2017. 61. Marchione M. “New Cancer Drugs May Damage the Heart.” www. nbcnews.com/health/cancer/new-cancer-drugs-may-damageheart-n677071. Accessed March 9, 2017. 62. Janelsins MC, Heckler CE, Peppone LJ, et al. Cognitive complaints in survivors of breast cancer after chemotherapy compared with age-matched controls: an analysis from a nationwide, multicenter, prospective longitudinal study. J Clin Oncol. 2016;35:506-514.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 11

INVITED ARTICLES

Genomics

Future Genetic/Genomic Biomarker Testing in Non–Small Cell Lung Cancer David Planchard, MD, PhD, Jordi Remon, MD, Frédérique Nowak, PhD, and Jean-Charles Soria, MD, PhD

he magnitude of the challenge that cancer poses is increasingly alarming, with a 60% projected increase in the annual incidence, from 12.7 million new cases in 2008 to 22.2 million new cases projected for 2030.1 However, cancer prognosis has also changed significantly over the past few decades. In the United States, cancer mortality decreased by 32% and 22% in men and women, respectively, between 1990 and 2015,2 reflecting dramatic improvements in cancer care in the last 25 years. Advances in genomic sequencing and molecular marker identification during the last decade have unequivocally demonstrated that cancer is a heterogeneous disease.3 Globally, genotype-directed targeted therapies are revolutionizing cancer care, and genetic alterations in genes such as EGFR, ALK, ROS1, HER2, KIT, and BRAF have been validated as powerful predictive biomarkers in the management of non–small cell lung cancer (NSCLC),4 along with gastric cancer,5 gastrointestinal stromal tumors,6 and melanoma.7 In France, eliminating inequalities in access to molecular profiling tests and consequent treatment is a priority.8 To this end, the French National Cancer Institute (INCa) and the French Ministry of Health established a national network of 28 molecular genetics centers that perform molecular tests for all patients in their region, irrespective of the institution where they are being treated. This program was updated in 2013 introducing next-generation sequencing (NGS) for switching from a gene-by-gene approach to a multiplexed strategy. The economic impact of this strategy has also been evaluated.9

CURRENT TECHNOLOGIC APPROACHES

Implementation of personalized medicine requires widely accessible tumor molecular profiling in routine practice, along with molecular centers for performing high-quality tests. Various methods exist for molecular profiling, including conventional Sanger sequencing, amplification refractory mutations systems, restriction fragment length

polymorphisms, and, more recently, targeted NGS panels.10 As it is now standard to test for a high number of mutations to personalize treatment decisions, use of NGS panels that can evaluate tumor biopsies for a wide range of potentially targetable mutations is increasing. Rapid and low-cost sequencing is providing physicians with the necessary tools to translate genomic information into clinically actionable results. Although the use of NGS is attractive as less DNA is required compared with multiple individual assays, these advancements are not without limitations, and there are substantial improvements to be made in sequencing technologies, data analysis bioinformatics pipelines, and computer resources. Reporting the limitations of an NGS assay along with the result is critical for clinical interpretation, especially in the context of the NGS detection of uncommon molecular alterations for which clinical significance assessment constitutes a real challenge.10 These elements are increasingly being discussed in molecular tumor boards, which are becoming widespread within the clinical sector. For example, in NSCLC, molecular tumor boards are feasible in daily practice allowing treatment recommendations in a majority of these patients (up to 70%), enrichment of their inclusion in clinical trials (57%) or expanded access programs (23%), and limitation of off-label drug use (9%).11 More extensive analysis such as RNA sequencing (RNAseq), whole-exome sequencing, and whole-genome sequencing are also starting to be used in routine practice (Table 1).10,12 Compared with the targeted NGS approach, they have the ability to detect rare and novel mutations that occur outside of specific, predefined regions as well as other types of molecular abnormalities such as gene translocations. Whole-exome sequencing and whole-genome sequencing allow detection of germline events involved in cancer susceptibility.12 However, robust bioinformatics algorithms are needed not only to analyze large volumes of high-throughput data being generated for each patient, but

From the Department of Oncology Medicine, Gustave Roussy, Université Paris-Saclay, Villejuif, France; Department of Oncology Medicine, Hospital de la Vall d’Hebron, Barcelona, Spain; Institut National du Cancer, Boulogne-Billancourt, France; University Paris-Sud and Gustave Roussy Cancer Campus, Villejuif, France. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Jean-Charles Soria, MD, PhD, University Paris-Sud and Gustave Roussy Cancer Campus, 114 Rue Edouard Vaillant, 94805 Villejuif, France; email: [emailprotected]. © 2017 American Society of Clinical Oncology

12 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

FUTURE GENOMIC BIOMARKER TESTING IN NSCLC

TABLE 1. DNA/RNA Sequencing Technique

Advantages

Disadvantages

Exome sequencing

Detection of genetics variations in all protein-coding regions of the genome

No detection of genetic variations in non–protein coding regions, including gene expression regulatory regions

Detection of nucleotide variations and small insertions and deletions

Not feasible if limited material samples

Discovered missense mutations, gene-disrupting mutations, and copy number variants

Slow turnaround time

Exome size relatively small Genome sequencing

Detection of all genetic variations, including protein-coding and regulatory regions

High volume and complex data analysis because of the large size of the human genome

Detection of nucleotide variations and genome reorganizations such as deletions, duplications, or translocations

High-level bioinformatics and computer resources required Greater amounts of DNA needed High cost for clinical use

RNA sequencing

Detection of genetic variations in protein-coding regions

Analysis restricted to genes expressed in the tissue or cell analyzed

RNA expression levels

Genetic variations in untranscribed regions not detected

Detections of RNA slicing variants and fusion transcripts

Adequate tumor tissue

Size of the transcriptome smaller than the genome

also to make predictions on the functional impact for each alteration, to classify drivers and passengers, and to prioritize different targets. As much as the molecular analyses do themselves, the preanalytical steps of NGS tumor genotyping in routine practice also present practical challenges including sample quality, adequate biopsy specimens, and the need for repeat biopsies after development of drug resistance, emphasizing the importance of quality sample collection and proper processing techniques. It is thus incontestable that there is an unmet need for noninvasive assays that can broadly detect actionable genomic alterations.13

CLINICAL UTILITY OF MOLECULAR TESTING

Lung cancer remains the most common cancer at a global scale, both in terms of new cases (1.8 million cases, 12.9% of all total cancer cases) and deaths (1.6 million deaths, representing 19.4% of total cancer deaths).14 It is also among the cancers with the highest mutation rates.15 One of the most important therapeutic advances has been the identification of distinct molecular subsets amenable to targeted therapies, especially among adenocarcinomas, as well as the early success of immune checkpoint inhibitors.16,17 In this indication, tumor genotyping is an essential routine diagnostic tool in clinical practice,18 and this strategy correlates with survival improvement for those patients treated with personalized therapies.4,19 In lung cancer, the availability of several profiling platforms worldwide has seen impressive progress in molecular testing, breaking through the barrier of unselected treatment in NSCLC and pushing out survival limitations. In 2015, molecular screening (EGFR, HER2, KRAS, BRAF, and PIK3CA mutations, as well as ALK rearrangements) was performed

in France at 28 certified centers for about 26,000 patients with NSCLC (Fig. 1). A clinical correlative work with the French Cooperative Thoracic Intergroup highlighted that a genetic driver alteration is recorded in about 50% of the analyses.4 In addition to the organizational framework set up by the French INCa and the Ministry of Health, other examples of molecular profiling programs at a national level include initiatives from the Network Genomic Medicine in Germany,20 the LC-SCRUM in Japan,21 and the Lung Cancer Mutation Consortium in the United States.19 Druggable molecular alterations occur in 20% to 25% of adenocarcinomas.4 The two most common are EGFR mutations, which occur in 12% of the Caucasian population (up to

FIGURE 1. Patients With Non–Small Cell Lung Cancer Screened for a Molecular Alteration in 2015

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 13

SORIA ET AL

50% in Asian population),22 and ALK rearrangements, which are seen in 5% of the population,4 independently of the race. These alterations confer sensitivity to specific EGFR tyrosine kinase inhibitors (TKI) such as erlotinib, gefitinib, afatinib, and icotinib (only available in China), and to ALKs such as crizotinib or ceritinib. Up-front personalized treatment with a TKI confers statistically significant and clinically meaningful improvement in progression-free survival and response rate compared with platinum-based chemotherapy in patients with advanced EGFR-mutated23 or ALK-rearranged NSCLC.24,25 However, almost all patients acquire resistance to these therapies, and identifying the mechanisms of resistance becomes critical for the implementation of personalized treatment at progression, especially among EGFR-mutant tumors. The most frequent pathway of resistance to EGFR TKI is the Thr790Met mutation in exon 20,26 which confers sensitivity to third-generation EGFR TKIs such as osimertinib, improving the outcome at progression compared with standard chemotherapy.27 Other mechanisms of resistance in EGFR-mutant tumors include MET amplification, PIK3CA mutations, EGFR amplification, and transformation to a small-cell phenotype.26 The mechanisms of resistance seen following ALK inhibitor therapy again reflect tumor evolution with secondary ALK mutations, ALK copy number gain, secondary driver mutations in other genes, and bypass pathways.28 A number of structurally distinct and more potent second- and third-generation ALK inhibitors are under evaluation for overcoming crizotinib resistance in the case of secondary ALK mutations. In this context, personalized treatment according to a molecular profile at progression is clearly the optimal strategy. Putative oncogenic drivers in squamous cell carcinoma are rare, but several novel molecular abnormalities are being investigated as potentially actionable targets, specifically FGFR1 amplification and PIK3CA and DDR2 mutations.29 Screening patients for solitary biomarker-driven studies requires both substantial time investment and adequate tumor tissue, resulting in low rates of enrollment. Because of this, new initiatives for lung treatment are in progress with trials integrating molecular screening and both targeted therapy and immunotherapy arms, such as the Lung Cancer Master Protocol (Lung-MAP; NCT02154490) for squamous cell carcinoma,30 and the phase II umbrella trials National Lung Matrix Trial (NCT02664935)31 and SAFIR02 Lung trial (NCT02117167) for squamous and nonsquamous NSCLC (Table 2). In SAFIR02, high throughput molecular analyses (e.g., CGH array, NGS) are used to evaluate the effect of treatment with targeted agents on progression-free survival compared with standard maintenance therapy in patients with metastatic NSCLC.

CHALLENGES AND SOLUTIONS

Beyond the more common nonsquamous NSCLC EGFR and KRAS mutations and ALK rearrangements, an important, albeit smaller, group of patients are found to harbor tumoral HER2, BRAF,4 and MET32 mutations or ROS1,33 NTRK1,34 14 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

and RET21 gene fusions. The development of TKIs for other oncogene-driven NSCLCs may expand the portfolio of precision therapies as we enter a new paradigm of molecular therapies in oncology. As our understanding of tumor taxonomy and genotypes advances, it seems inevitable that some form of NGS platform will become the clinical standard for gene fusion detection instead of running multiple fluorescence in situ hybridization analyses. Both targeted and extensive RNAseq or whole-exome/whole-genome sequencing have the potential to detect ALK and other rearrangements.10 To be widely used in clinical practice, RNAseq approaches have to be optimized for the clinical grade analysis of formalin-fixed, paraffin-embedded tumor samples. In addition to comprehensively identifying mutations in genes that encode members of RTK-induced signaling cascades, genomics efforts have uncovered mutations in genes involved in other important cellular processes (such as KEAP1-CUL3NFE2L2, TP63, and SOX2, involved in the oxidative stress response and differentiation pathways).29 Chromatin-modifying genes are recurrently mutated in lung adenocarcinoma and lung squamous cell carcinoma and also represent potential therapeutic targets in these diseases.29,35 The advent of immunotherapy presents additional challenges for molecular testing in NSCLC. To date, a number of potential biomarkers have been identified, but their relevance to clinical practice is still unclear and requires elucidation in prospective studies. Given that the marketing authorization of some checkpoint inhibitors has been restricted to patients with PD-L1–positive disease,36 immunohistochemical evaluation of PD-L1 expression must be performed in routine practice. Nevertheless, some patients with PD-L1–low or PD-L1–negative tumors respond to these treatments. On a methodologic level, it is essential to harmonize the different detection and scoring methods before the routine use of PD-L1 expression as a predictive marker.37 Recent whole-exome sequencing studies have shown a significant correlation between the total tumor mutational load and the predicted neoantigen load and clinical benefit with immune checkpoint inhibitors.38,39 Characterization of neoantigens as a potential biomarker requires sufficient tumor DNA for whole-exome sequencing and carries major expense. But given the cost of these therapies, this initial outlay would be justified if the assay was sufficiently reliably predictive.40 Obtaining adequate tissue for diagnosis, tissue subtyping, molecular profiling, and treatment planning are critical for optimal patient management. Added to this, at the time of disease progression, a key challenge is also obtaining a recent sampling of the progressive tissue to determine the selection of second-line therapy. However, lack of tissue biopsy at progression is not uncommon, and furthermore, single site biopsies may not provide a representative profile of the overall predominant resistance mechanisms for a given patient.41 Liquid biopsies based on circulating cell-free tumor DNA (ctDNA) analysis have been described as surrogate samples for molecular analysis, replacing solid tumor biopsies.42 This approach offers the potential of real-time

FUTURE GENOMIC BIOMARKER TESTING IN NSCLC

TABLE 2. Molecularly Stratified Ongoing Umbrella Studies in Advanced Stage NSCLC Study

Phase

LungMAP30 U.S.

Multisub-study, randomized phase II/III trial

No. of Patients

Primary Endpoint

Line

Screening Tests

Molecular Subgroups and Treatment

10,000

PFS, ORR, OS

≥ second-line only SCC

NGS

GDC-0032 (PI3K inhibitor)

IHC

MEDI4736 (anti-PD-L1) Palbociclib (CDK 4/6 inhibitor) AZD4547 (FGGFR 1-3 inhibitor) Docetaxel

National Lung Matrix Trial31 U.K.

Multiarm, nonrandomized, noncomparative phase II

620

ORR, PFS

≥ second line

NGS

AZD4547 (FGFR inhibitor) AZD2014 (MTORC1/2 inhibitor) Palbociclib (CDK4/6 inhibitor) Crizotinib (ALK/MET/ROS1 inhibitor) Selumetinib (MEK inhibitor) Docetaxel AZD5363 (AKT inhibitor) AZD9291 (EGFR and T790M inhibitor) MEDI4736 (anti-PDL1)

SAFIR02 Lung Study France

Open-label, multicentric, randomized phase II trial

650

PFS personalized vs. standard maintenance

First-line maintenance after four cycles of CT

NGS

AZD2014 (m-TOR inhibitor)

CGH

AZD4547 (FGFR inhibitor) AZD5363 (AKT inhibitor) AZD8931 (HER2 and EGFR inhibitor) Selumetinib (MEK inhibitor) Vandetanib (VEGF, EGFR inhibitor) Pemetrexed MEDI4736 (anti PD-L1)

Abbreviations: NSCLC, non–small cell lung cancer; PFS, progression-free survival; ORR, objective response rate; OS, overall survival; SCC, squamous cell carcinoma; CT, chemotherapy; NGS, next-generation sequencing; IHC, immunohistochemistry; CGH, comparative genomic hybridization.

sampling of multifocal clonal evolution,43 as well as potential dynamic markers for monitoring the efficacy of treatment44,45 and early detection of resistance mutations.46 Early detection has been reported in patients with EGFR TKIs often prior to radiographic progression46 and allows therapy to be adapted accordingly.47 Further studies confirming clinical implications of monitoring the emergence of resistance mutations in plasma are warranted to guide therapeutic strategies. The phase II APPLE trial (NCT02856893) is one example of a strategic trial that is expected to provide some answers in the near future (Fig. 2). Discrepancies between the tumor biopsy and ctDNA genotyping may result from technologic differences or sampling of different tumor cell populations in a heterogeneous setting.48 As sensitivity and specificity of ctDNA varies across different technology platforms,49 the establishment of robust and standardized protocols for blood sampling, processing, storage, DNA extraction, and analysis will support the role of liquid biopsies as standard tests in the near future for tumor genotyping and predictive biomarkers.49 These recent techniques will ensure that molecular genomic analysis and personalized treatments are soon available to more patients. Finally, circulating tumor cells isolated from the peripheral blood offer a complementary circulating biomarker to ctDNA. Circulating tumor cells permit

further immunohistochemistry/fluorescence in situ hybridization characterization, while single-cell DNA sequencing or RNAseq is also possible, as well as the generation of tumor xenografts to assess drug response.50 However, at present, the technologic complexity of circulating tumor cell isolation and the need to process samples quickly for functional/ genomic studies results in greater expense compared with ctDNA analysis.

CONCLUSION

Molecular genotyping in NSCLC is common in the clinic. In the near future, noninvasive biopsies and standardization of NGS with detection of gene fusion alterations will become the new standard in daily clinical practice. With the development of NGS allowing for the detection of multiple genomic alterations, the need to prioritize these gene alterations to drug response is pressing. The immediate challenges in routine practice includes the cost of molecular profiling, limiting social inequalities that affect access to these tests, the widespread molecular tumor boards, and access to clinical trials for patients with uncommon mutations. Molecular profiling for other thoracic malignancies is another important area currently being addressed. SPECTAlung (NCT02214134) is a pan-European program aimed at screening patients with thoracic tumors (i.e., lung cancer, asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 15

SORIA ET AL

FIGURE 2. APPLE Trial Design

Abbreviations: BM, brain metastases; PFS, progression-free survival; PD, progressive disease; PS, performance status.

malignant pleural mesothelioma, thymoma or thymic carcinoma at any stage) to identify the molecular characteristics of their disease to ensure efficient clinical trial access and personalized treatments in the case of specific mutations. Crossanalysis of mutational data with multiomics data, functional

data, and clinicopathologic data in a larger number of samples is an integral part of this. Thus, international collaborative efforts as well as increased integration of technologic aspects of molecular characterization with clinical data are needed to further advance the treatment of patients with NSCLC.

References 1. Bray F, Jemal A, Grey N, et al. Global cancer transitions according to the Human Development Index (2008-2030): a population-based study. Lancet Oncol. 2012;13:790-801. 2. Byers T, Wender RC, Jemal A, et al. The American Cancer Society challenge goal to reduce US cancer mortality by 50% between 1990 and 2015: results and reflections. CA Cancer J Clin. 2016;66:359-369. 3. Marusyk A, Almendro V, Polyak K. Intra-tumour heterogeneity: a looking glass for cancer? Nat Rev Cancer. 2012;12:323-334. 4. Barlesi F, Mazieres J, Merlio J-P, et al; Biomarkers France contributors. Routine molecular profiling of patients with advanced non-small-cell lung cancer: results of a 1-year nationwide programme of the French Cooperative Thoracic Intergroup (IFCT). Lancet. 2016;387:1415-1426. 5. Bang Y-J, Van Cutsem E, Feyereislova A, et al; ToGA Trial Investigators. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastrooesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet. 2010;376:687-697. 6. Demetri GD, von Mehren M, Blanke CD, et al. Efficacy and safety of imatinib mesylate in advanced gastrointestinal stromal tumors. N Engl J Med. 2002;347:472-480. 7. Hauschild A, Grob J-J, Demidov LV, et al. Dabrafenib in BRAF-mutated metastatic melanoma: a multicentre, open-label, phase 3 randomised controlled trial. Lancet. 2012;380:358-365.

16 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

8. Carbonnaux M, Souquet P-J, Meert A-P, et al. Inequalities in lung cancer: a world of EGFR. Eur Respir J. 2016;47:1502-1509. 9. Institut National du Cancer. Cancer Plan 2014-2019. http://www. e-cancer.fr/Plan-cancer/Plan-cancer-2014-2019-priorites-et-objectifs/ Les-17-objectifs-du-Plan. Accessed February 26, 2017. 10. Goodwin S, McPherson JD, McCombie WR. Coming of age: ten years of next-generation sequencing technologies. Nat Rev Genet. 2016;17:333-351. 11. Planchard D, Faivre L, Sullivan I, et al. 3081 Molecular Tumor Board (MTB) in non-small cell lung cancers (NSCLC) to optimize targeted therapies: 4 years’ experience at Gustave Roussy. Eur J Cancer. 2015;51:S624. 12. Meyerson M, Gabriel S, Getz G. Advances in understanding cancer genomes through second-generation sequencing. Nat Rev Genet. 2010;11:685-696. 13. Crowley E, Di Nicolantonio F, Loupakis F, et al. Liquid biopsy: monitoring cancer-genetics in the blood. Nat Rev Clin Oncol. 2013;10:472-484. 14. Ferlay J, Soerjomataram I, Diksh*t R, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer. 2015;136:E359-E386. 15. Alexandrov LB, Nik-Zainal S, Wedge DC, et al; Australian Pancreatic Cancer Genome Initiative; ICGC Breast Cancer Consortium; ICGC MMML-Seq Consortium; ICGC PedBrain. Signatures of mutational processes in human cancer. Nature. 2013;500:415-421.

FUTURE GENOMIC BIOMARKER TESTING IN NSCLC

16. Tan W-L, Jain A, Takano A, et al. Novel therapeutic targets on the horizon for lung cancer. Lancet Oncol. 2016;17:e347-e362.

33. Shaw AT, Ou S-HI, Bang Y-J, et al. Crizotinib in ROS1-rearranged nonsmall-cell lung cancer. N Engl J Med. 2014;371:1963-1971.

17. Swanton C, Govindan R. Clinical implications of genomic discoveries in lung cancer. N Engl J Med. 2016;374:1864-1873.

34. Suh JH, Johnson A, Albacker L, et al. comprehensive genomic profiling facilitates implementation of the National Comprehensive Cancer Network Guidelines for lung cancer biomarker testing and identifies patients who may benefit from enrollment in mechanism-driven clinical trials. Oncologist. 2016;21:684-691.

18. Lindeman NI, Cagle PT, Beasley MB, et al; College of American Pathologists International Association for the Study of Lung Cancer and Association for Molecular Pathology. Molecular testing guideline for selection of lung cancer patients for EGFR and ALK tyrosine kinase inhibitors: guideline from the College of American Pathologists, International Association for the Study of Lung Cancer, and Association for Molecular Pathology. J Mol Diagn. 2013;15:415-453. 19. Kris MG, Johnson BE, Berry LD, et al. Using multiplexed assays of oncogenic drivers in lung cancers to select targeted drugs. JAMA. 2014;311:1998-2006. 20. Heydt C, Kostenko A, Merkelbach-Bruse S, et al. ALK evaluation in the world of multiplex testing: Network Genomic Medicine (NGM): the Cologne model for implementing personalised oncology. Ann Oncol. 2016;27(Suppl 3):iii25-iii34. 21. Yoh K, Seto T, Satouchi M, et al. Vandetanib in patients with previously treated RET-rearranged advanced non-small-cell lung cancer (LURET): an open-label, multicentre phase 2 trial. Lancet Respir Med. 2017;5:42-50. 22. Midha A, Dearden S, McCormack R. EGFR mutation incidence in nonsmall-cell lung cancer of adenocarcinoma histology: a systematic review and global map by ethnicity (mutMapII). Am J Cancer Res. 2015;5:2892-2911.

35. Collisson EA, Campbell JD, Brooks AN, et al; Cancer Genome Atlas Research Network. Comprehensive molecular profiling of lung adenocarcinoma. Nature. 2014;511:543-550. 36. Reck M, Rodríguez-Abreu D, Robinson AG, et al; KEYNOTE-024 Investigators. Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung cancer. N Engl J Med. 2016;375:1823-1833. 37. Hirsch FR, McElhinny A, Stanforth D, et al. PD-L1 immunohistochemistry assays for lung cancer: results from phase 1 of the Blueprint PD-L1 IHC assay comparison project. J Thorac Oncol. 2017;12:208-222. 38. Rizvi NA, Hellmann MD, Snyder A, et al. Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in nonsmall cell lung cancer. Science. 2015;348:124-128. 39. McGranahan N, Furness AJS, Rosenthal R, et al. Clonal neoantigens elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade. Science. 2016;351:1463-1469. 40. Hiley CT, Le Quesne J, Santis G, et al. Challenges in molecular testing in non-small-cell lung cancer patients with advanced disease. Lancet. 2016;388:1002-1011.

23. Reguart N, Remon J. Common EGFR-mutated subgroups (Del19/L858R) in advanced non-small-cell lung cancer: chasing better outcomes with tyrosine kinase inhibitors. Future Oncol. 2015;11:1245-1257.

41. Piotrowska Z, Nierdest MJ, Mino-Kenudson M. Variation in mechanisms of acquired resistance among EGFR-mutant NSCLC patients with more than 1 postresistance biopsy: metastatic non-small cell lung cancer. Int J Radiat Oncol Biol Phys. 2014;90:S6-S7.

24. Solomon BJ, Mok T, Kim D-W, et al; PROFILE 1014 Investigators. Firstline crizotinib versus chemotherapy in ALK-positive lung cancer. N Engl J Med. 2014;371:2167-2177.

42. Jovelet C, Ileana E, Le Deley M-C, et al. Circulating cell-free tumor DNA analysis of 50 genes by next-generation sequencing in the prospective MOSCATO trial. Clin Cancer Res. 2016;22:2960-2968.

25. Soria JC, Tan DS, Chiari R, et al. First-line ceritinib versus platinumbased chemotherapy in advanced ALK-rearranged non-small-cell lung cancer (ASCEND-4): a randomised, open-label, phase 3 study. Lancet. Epub 2017 Jan 23.

43. Murtaza M, Dawson S-J, Pogrebniak K, et al. Multifocal clonal evolution characterized using circulating tumour DNA in a case of metastatic breast cancer. Nat Commun. 2015;6:8760.

26. Sequist LV, Waltman BA, Dias-Santagata D, et al. Genotypic and histological evolution of lung cancers acquiring resistance to EGFR inhibitors. Sci Transl Med. 2011;3:75ra26. 27. Mok TS, Wu Y-L, Ahn M-J, et al. Osimertinib or platinum-pemetrexed in EGFR T790M-positive lung cancer. N Engl J Med. 2017;376:629640. 28. Gainor JF, Dardaei L, Yoda S, et al. Molecular mechanisms of resistance to first- and second-generation ALK inhibitors in ALK-rearranged lung cancer. Cancer Discov. 2016;6:1118-1133. 29. Cancer Genome Atlas Research Network. Comprehensive genomic charac terization of squamous cell lung cancers. Nature. 2012;489:519-525. 30. Herbst RS, Gandara DR, Hirsch FR, et al. Lung Master Protocol (LungMAP)-a biomarker-driven protocol for accelerating development of therapies for squamous cell lung cancer: SWOG S1400. Clin Cancer Res. 2015;21:1514-1524. 31. Middleton G, Crack LR, Popat S, et al. The National Lung Matrix Trial: translating the biology of stratification in advanced non-small-cell lung cancer. Ann Oncol. 2015;26:2464-2469. 32. Paik PK, Drilon A, Fan P-D, et al. Response to MET inhibitors in patients with stage IV lung adenocarcinomas harboring MET mutations causing exon 14 skipping. Cancer Discov. 2015;5:842-849.

44. Marchetti A, Palma JF, Felicioni L, et al. Early prediction of response to tyrosine kinase inhibitors by quantification of EGFR mutations in plasma of NSCLC patients. J Thorac Oncol. 2015;10:1437-1443. 45. Mok T, Wu Y-L, Lee JS, et al. Detection and dynamic changes of EGFR mutations from circulating tumor DNA as a predictor of survival outcomes in NSCLC patients treated with first-line intercalated erlotinib and chemotherapy. Clin Cancer Res. 2015;21:3196-3203. 46. Sorensen BS, Wu L, Wei W, et al. Monitoring of epidermal growth factor receptor tyrosine kinase inhibitor-sensitizing and resistance mutations in the plasma DNA of patients with advanced non-small cell lung cancer during treatment with erlotinib. Cancer. 2014;120:3896-3901. 47. Remon J, Caramella C, Jovelet C, et al. Osimertinib benefit in EGFRmutant NSCLC patients with T790M-mutation detected by circulating tumour DNA. Ann Oncol. Epub 2017 Jan 18. 48. Sundaresan TK, Sequist LV, Heymach JV, et al. Detection of T790M, the acquired resistance EGFR mutation, by tumor biopsy versus noninvasive blood-based analyses. Clin Cancer Res. 2016;22:1103-1110. 49. Thress KS, Brant R, Carr TH, et al. EGFR mutation detection in ctDNA from NSCLC patient plasma: A cross-platform comparison of leading technologies to support the clinical development of AZD9291. Lung Cancer. 2015;90:509-515. 50. Alix-Panabières C, Pantel K. Clinical applications of circulating tumor cells and circulating tumor DNA as liquid biopsy. Cancer Discov. 2016;6:479-491.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 17

INVITED ARTICLES

Workforce Diversity

Making the Case for Improving Oncology Workforce Diversity Karen M. Winkfield, MD, PhD, Christopher R. Flowers, MD, MS, and Edith P. Mitchell, MD, FACP

ince its inception in 2008, ASCO’s Diversity in Oncology Initiative has engaged in programming designed to support and promote diversity in the oncology workforce.1 The programs, developed by the ASCO Health Disparities Committee and funded through the Conquer Cancer Foundation, include mentoring and award opportunities for medical students and residents from backgrounds that are traditionally underrepresented in medicine (URM). During a recent evaluation of the Initiative, it was recognized that a more comprehensive plan was needed to ensure it successfully met its intended goal. In response to the organization’s goal of increasing diversity and inclusion in oncology professions, the Health Disparities Committee convened a task force in 2015 that ultimately formulated a strategic plan for racial/ ethnic workforce diversity, consonant with ASCO’s overall Workforce Strategic Plan.2 In December 2016, the ASCO Board approved the strategic plan, affirming that diversity is central to its mission and strengthens the organization. Although some still question the rationale for efforts around diversity and inclusion, the current state of our nation and our nation’s health care system clearly speak to the need for a concerted effort to diversify the oncology workforce.

CANCER DISPARITIES

Cancer is a major health care problem worldwide and the second cause of death in the United States. An estimated 600,920 individuals will succumb to cancer in 2017.3 The most common cancer-specific causes of death are cancers of the lung and bronchus, colon and rectum, and prostate in men, and the lung and bronchus, colon and rectum, and breast in women. National attention has increasingly shifted to the issue of health disparities based on race/ethnicity and its impact on the health of the nation. African American, American Indian, and Alaskan Native populations have the poorest health status of all racial/ethnic groups in the United States.4 These communities are plagued by increased incidence of obesity, HIV/AIDS, heart disease, and myriad other conditions. However, African Americans experience the worst cancer outcomes of all races/ethnicities (Fig. 1).5

Several metrics have been used to delineate disparate cancer outcomes experienced by medically underserved communities. Disparities in cancer screening, treatment, and outcomes are well documented by socioeconomic status, health care access, insurance status, and race.4,6,7 For selected cancers, significant differences in incidence and mortality between racial/ethnic groups have been demonstrated. For example, African Americans are more likely to have more advanced-stage disease at the time of cancer diagnosis and to experience lower stage-specific survival rates compared with the white population. Presentation at late stage may be related to access to screening because of lower health insurance coverage rates in African Americans and other minorities compared with non-Hispanic whites. The impact of the Patient Protection and Affordable Care Act and the Health Care and Education Reconciliation Act of 2010 (Affordable Care Act or ACA) on rates of insured cannot be overstated. In 2015, the uninsured rates in the black population dropped to 11% from 21% in 2010. Similarly, for non-Hispanic whites, the uninsured rate declined from 12% in 2010 to 7% in 2015.3 Although all racial/ethnic groups have seen a steady decline in cancer mortality, the overall cancer death rate in 2014 was 15% higher for African Americans than the white population. The persistent gap in mortality suggests that insurance alone is not enough to level the playing field; more is required to improve access to care along the cancer continuum. Cancer disparities arise as a result of complex interactions between biologic, clinical, social, and environmental factors at the patient and the population levels (Fig. 2).8,9 Establishing a comprehensive strategy to meet these needs of patients with cancer requires detailed population-level data on incidence rates and survival disparities that address these factors; an infrastructure to support clinical and basic research to understand how these factors that influence prognosis and survivorship can facilitate strategies and interventions that prevent cancer and improve outcomes; and a well-trained workforce that can use this infrastructure and other resources to examine and address the needs of underserved patient populations.

From the Department of Radiation Oncology, Wake Forest Baptist Health, Winston-Salem, NC; Department of Hematology and Oncology, Winship Cancer Institute at Emory University, Atlanta, GA; Department of Medical Oncology, The Sidney Kimmel Cancer Center at Jefferson University Hospitals, Philadelphia, PA. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Karen M. Winkfield, MD, PhD, Department of Radiation Oncology, Wake Forest Baptist Health, 1 Medical Center Blvd., Winston-Salem, NC 27157; email: [emailprotected]. © 2017 American Society of Clinical Oncology

18 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

IMPROVING ONCOLOGY WORKFORCE DIVERSITY

FIGURE 1. Incidence and Cause-Specific Mortality of the Top Four Cancer Sites in the United States by Race/Ethnicity

From the Surveillance, Epidemiology, and End Results data from 18 areas (San Francisco, Connecticut, Detroit, Hawaii, Iowa, New Mexico, Seattle, Utah, Atlanta, San Jose-Monterey, Los Angeles, Alaska Native Registry, rural Georgia, California excluding San Francisco/San Jose-Monterey/Los Angeles, Kentucky, Louisiana, New Jersey, and Georgia excluding Atlanta/rural Georgia) and U.S. Mortality Files, National Center for Health Statistics, Centers for Disease Control and Prevention. Rates are per 100,000 and age-adjusted to the 2000 U.S. Standard Population (19 age groups; Census P25-1103). (a) Rates for American Indian/Alaska Natives are based on the Contract Health Service Delivery Area counties. (b) Hispanic is not mutually exclusive from the white or black populations, Asian/Pacific Islanders, or American Indians/Alaska Natives. Incidence data for Hispanic numbers are based on North American Association of Central Cancer Registries Hispanic Identification Algorithm and exclude cases from the Alaska Native Registry.

Disparities in health care based on race/ethnicity represent a mutable factor that costs the U.S. government billions of dollars annually. A recent report estimates that from 2009 through 2018, the total cost of these disparities will be approximately $337 billion, with the annual cost estimated to more than double to $50 billion by 2050, attributed in part to the aging black and Hispanic populations.10 Studies have consistently demonstrated the propensity of URM physicians to provide improved access to health care for underserved populations.11-14

ONCOLOGY WORKFORCE

Even the ACA recognized the importance of workforce diversity in stemming health inequities by establishing or

FIGURE 2. Causes of Health Disparities

The complex interplay between socioeconomic status, culture, and biology on cancer disparities. In addition to impacting access to care across the entire cancer continuum from prevention through survivorship, health disparities may influence the genetic environment as well.

renewing funding for several pipeline programs that may impact the racial composition of the workforce.15,16 However, the U.S. Congress has yet to fund the National Health Care Workforce Commission,17 an important component of the ACA that could make recommendations to compel institutions to consider workforce diversity in the design and implementation of their clinical practices. Therefore other strategies must be used to address the increasing challenges that will be faced as the diversity of patients with cancer and survivors continues to increase. According to the U.S. Census Bureau, by 2060, minorities (including African Americans, Hispanics, American Indians, and Alaska Natives) will make up 56% of the U.S. population,18 a major increase from 38% in 2014. Although the lack of workforce diversity has been shown to negatively impact quality and health outcomes for minority patients, a diverse workforce can increase patients’ comfort and trust with their providers, thereby improving patient access to and satisfaction with their health care.12 Lack of diversity is compounded by the growing need for oncologists in general. According to the Association of American Medical Colleges (AAMC), in 1971, there were approximately 3 million cancer survivors in the United States. By 2001, the number of survivors increased to nearly 10 million.19 Given that the number of cancer survivors likely will continue to increase, there will be a corresponding increase in demand for well-trained clinical oncologists. This is particularly relevant as the current oncology workforce continues to age without a corresponding increase in the number of fellowship positions. As noted above, ASCO had previously articulated a strategic plan focused on general workforce development and has now stepped up to address the need for increased racial/ethnic diversity. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 19

WINKFIELD, FLOWERS, AND MITCHELL

BARRIERS TO ADDRESS

Unfortunately, the field of oncology and its subspecialties reflect and magnify the lack of racial and gender diversity present in the general medical field. In 2007, the AAMC reported that 59% of the oncology workforce from a sample of 4,000 physicians was white.19 Black, Hispanic, and female physicians are statistically less represented in oncology than white males, perpetuated by a lack of minority and female clinicians entering the oncology workforce and reinforced by a lack of minority and women role models in leadership positions in academic medicine.20 In 2015, only 3.7% of oncology fellows were black, and 5.3% of fellows were Hispanic.21 These trends are even more worrisome with respect to the lack of diversity in cancer leadership positions. In 2013, minorities held only 4% of National Institutes of Health Research Project grants despite making up 29% of the U.S. population.22 It has been long established that faculty diversity not only benefits medical students and trainees but also provides faculty from all backgrounds with an opportunity to enhance their ability to care for an increasingly diverse population.14 If the oncology workforce continues to lack diversity, the health of the increasingly diverse patient pool will be at stake. Not only is there a need for in-depth training programs to address these disparities, but dissemination of the skills and courses taught to selected individuals who are focused on cultural humility also appears to aid in the dissemination of knowledge to cancer researchers at all levels, including those not in the training program.23 To combat these trends, inclusion and diversity have risen to the forefront as desired characteristics of successful organizations that are essential to competition.24,25 For inclusion to be normalized, it must be integrated into multiple aspects of the entity, including hiring, promotion, and encouraging minority and women leaders to pave the way for those just starting their careers.24 One evidence-based strategy for encouraging more minorities and women to pursue leadership roles in medicine is accessing programs that provide guidance and support for trainees to pursue clinical research. Dedicated mentorship from and collaboration with women and minorities who have already achieved success in the field is an important component of this strategy.26 Supporting trainees at multiple levels within the pipeline can provide a career development pathway that is essential to ensuring a well-trained cohort of leaders and care providers in the future. Moreover, facilitating opportunities for minority mentor and minority mentee relationships among seasoned specialists/researchers and trainees has been demonstrated to be important for the career development of minority professionals.19

PROGRAMS TO IMPROVE PHYSICIAN WORKFORCE DIVERSITY

Increasing minority participation in science, technology, engi neering, and mathematics education at all levels has been recommended as a national priority.27 Although black and Hispanic individuals make up 11.7% and 14.6% of the U.S. population, respectively, they account for 7.2% and 7.7% of college degree holders and 4.8% and 6.1% of individuals in 20 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

science and engineering occupations.27 Having an experienced professional as a mentor can inspire younger mentees to pursue careers in science and clinical care, can provide encouragement for mentees to seek out leadership roles,19 and is associated with greater career satisfaction.28 Studies have shown that fostering such relationships benefits the mentors, influencing academic productivity and career advancement.29,30 Even when underrepresented minorities enter medical school, several hurdles can restrict the successful recruitment and retention of minorities to careers in cancer clinical care and research. A key barrier is lack of exposure to programs that foster understanding and appreciation of the opportunities available in oncology practice and biomedical research. Prominent societies involved in cancer care, including the American Society of Hematology,31 the American Society for Radiation Oncology,32 and the American Association for Cancer Research,33 among others, have established initiatives designed to provide exposure to research and clinical careers at varying stages of career development. Yet, the fact that some oncologists, even those at prominent academic centers with training programs, do not understand what the acronym “URM” stands for is a bit disconcerting. In 2003, the Association of American Medical Colleges shifted the expansion of “URM” from “underrepresented minority” (black population, Mexican Americans, Native Americans, which includes American Indians, Alaska Natives, and Native Hawaiians) to “underrepresented in medicine.”34 URM has grown to reflect the evolving demographics of the nation, but the sentiment of the acronym has remained consistent: “those racial and ethnic populations that are underrepresented in the medical profession relative to their numbers in the general population.” It is imperative that all oncologists—but particularly those who are in a position to influence the character and makeup of training/fellowship programs— acknowledge the lack of diversity in the workforce and understand the importance of reaching out to URM medical students early in their curriculum. For the past 9 years, ASCO’s Diversity in Oncology Initiative program has awarded opportunities for medical students and residents who self-identify as URM. The ASCO Medical Student Rotation Award supports clinical oncology or cancer clinical research rotations and pairs URM medical students with a clinical oncologist who provides ongoing academic and career development advice. The ASCO Resident Travel Award supports residents to attend the ASCO Annual Meeting. Since 2008, the ASCO Diversity in Oncology Initiative has provided over $1 million to fund 137 individuals, and 81 recipients have become ASCO members, providing an early indication of the success of this program amid increasing oncology workforce diversity. However, to ensure sustainability and support the changing needs of the diverse population we serve, ASCO’s recently adopted strategic plan for increasing racial and ethnic diversity in the oncology workforce provides a blueprint for a comprehensive approach to diversity and inclusion.

PROGRAM PARTICIPATION IS KEY

Although ASCO plans to set an example by demonstrating what diversity and inclusion looks like within the organization

IMPROVING ONCOLOGY WORKFORCE DIVERSITY

FIGURE 3. Percentage of URM Residents/Fellows in Oncology Training Programs

Abbreviations: URM, underrepresented in medicine; AA, African American; His, Hispanic. Adapted from: "The State of Cancer Care in America, 2016: A Report by the American Society of Clinical Oncology."38 Note: Data represent the total number of fellows (MDs and DOs) in hematology, hematology/oncology, and clinical oncology graduate medical education programs accredited by the Accreditation Council for Graduation Medical Education.

and by continuing to provide opportunities to URM trainees, their efforts alone are not sufficient. Training programs, academic institutions, and individual practices, whether private or hospital-based, must be willing not only to diversify but to create a truly inclusive environment. Medicine has trailed behind in the recognition that diversity breeds innovation and even improves the financial bottom line. Bringing in new talent fromdiverse backgrounds will require an investment of time, thought, and energy. Leadership must set goals and priorities and articulate a vision for inclusivity that others within the organization will value and accept. Recent reports have demonstrated that while the percentage of Hispanic trainees in oncology is improving, the number of black students graduating from medical school and entering oncology specialties has remained stagnant over the past few decades (Fig. 3).35,36 Even with the slow increase in the number of Hispanic trainees, a workforce that sufficiently reflects the diversity of the United States will not happen over night. Until the number of URM oncologists increases to adequately address our nation's growing needs, thoughtful strategies related to employment of advanced practice providers from diverse backgrounds, such as nurse practitioners, physician assistants, and clinical nurse specialists, among others,37 may provide a critical bridge. However, it is just as important for non-URM oncologists and staff to

understand their cultural biases and learn the importance of cultural sensitivity and cultural humility. This is particularly important for minority-serving institutions that provide cancer care to a greater percentage of patients from diverse backgrounds. ASCO has thoughtfully created several programs related to cultural competency that can be accessed through ASCO University. There is even a membership category for advanced practice providers to help provide the support needed for these valued members of the oncology care team to develop the knowledge and skills required to fully engage in a diverse oncology practice. The effective creation of a workforce that is reflective of the patients it serves first requires an institution to embrace diversity and inclusion as its core values. Change will not come quickly or easily, and therefore, the only way to succeed is for organizational leadership to be intentional in the design and implementation of programming.25 By approving the strategic plan for improving racial/ethnic workforce diversity and sharing it with the world,39 ASCO has raised the bar by openly setting specific goals and inviting others to hold them accountable. Hopefully, oncology training programs and practices around the country will soon follow the example and work toward creating a more diverse and inclusive environment that will enable us to provide better care for all of our patients.

References 1. ASCO. Diversity in Oncology Initiative. https://www.asco.org/practiceguidelines/cancer-care-initiatives/diversity-oncology-initiative. Accessed March 1, 2017. 2. Future supply of and demand for oncologists. J Oncol Pract. 2008; 4:300-302.

3. Siegel RL, Miller KD, Jemal A. Cancer Statistics, 2017. CA Cancer J Clin. 2017;67:7-30. 4. Williams DR, Mohammed SA, Leavell J, et al. Race, socioeconomic status, and health: complexities, ongoing challenges, and research opportunities. Ann N Y Acad Sci. 2010;1186:69-101.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 21

WINKFIELD, FLOWERS, AND MITCHELL

5. National Cancer Institute. Surveillance, Epidemiology, and End Results Program. https://seer.cancer.gov. Accessed March 1, 2017. 6. Keegan TH, DeRouen MC, Parsons HM, et al. Impact of treatment and insurance on socioeconomic disparities in survival after adolescent and young adult Hodgkin lymphoma: a population-based study. Cancer Epidemiol Biomarkers Prev. 2016;25:264-273. 7. Tao L, Foran JM, Clarke CA, et al. Socioeconomic disparities in mortality after diffuse large B-cell lymphoma in the modern treatment era. Blood. 2014;123:3553-3562. 8. Flowers CR, Nastoupil LJ. Socioeconomic disparities in lymphoma. Blood. 2014;123:3530-3531. 9. Warnecke RB, Oh A, Breen N, et al. Approaching health disparities from a population perspective: the National Institutes of Health Centers for Population Health and Health Disparities. Am J Public Health. 2008;98:1608-1615. 10. Waidman T. Estimating the Cost of Racial and Ethnic Health Disparities. The Urban Institute. http://www.urban.org/research/publication/ estimating-cost-racial-and-ethnic-health-disparities. Accessed March 11, 2017. Nivet MA, Taylor VS, Butts GC, et al. Diversity in academic medicine 11. no. 1 case for minority faculty development today. Mt Sinai J Med. 2008;75:491-498. The Sullivan Commission. Missing Persons: Minorities in the Health 12. Professions, A Report of the Sullivan Commission on Diversity in the Healthcare Workforce. http://www.aacn.nche.edu/media-relations/ SullivanReport.pdf. Accessed March 11, 2017. Saha S, Shipman SA. Race-neutral versus race-conscious workforce policy 13. to improve access to care. Health Aff (Millwood). 2008;27:234-245. Saha S, Guiton G, Wimmers PF, et al. Student body racial and ethnic 14. composition and diversity-related outcomes in US medical schools. JAMA. 2008;300:1135-1145. CCBC Library. The Affordable Care Act. http://libraryguides.ccbcmd. 15. edu/ACA. Accessed June 1, 2016. Moy B, Polite BN, Halpern MT, et al. American Society of Clinical 16. Oncology policy statement: opportunities in the patient protection and affordable care act to reduce cancer care disparities. J Clin Oncol. 2011;29:3816-3824. Buerhaus PI, Retchin SM. The dormant National Health Care Workforce 17. Commission needs congressional funding to fulfill its promise. Health Aff (Millwood). 2013;32:2021-2024. 18. Colby SL, Ortman JM. Projections of the Size and Composition of the U.S. Population: 2014 to 2060. US Census Bureau. https://www. census.gov/content/dam/Census/library/publications/2015/demo/ p25-1143.pdf. Accessed March 11, 2017. 19. Association of American Medical Colleges. Forecasting the Supply of and Demand for Oncologists: A Report to the American Society of Clinical Oncology (ASCO) from the AAMC Center for Workforce Studies. http://www.asco.org/sites/new-www.asco.org/files/contentfiles/research-and-progress/documents/Forecasting-the-Supply-ofand-Demand-for-Oncologists.pdf. Accessed March 11, 2017. 20. Deville C, Hwang WT, Burgos R, et al. Diversity in graduate medical education in the United States by race, ethnicity, and sex, 2012. JAMA Intern Med. 2015;175:1706-1708.

23. National Institutes of Health. Clinical Research Training at the NIH Clinical Center. https://report.nih.gov/nihfactsheets/Pdfs/ ClinicalResearchTrainingattheNIHClinicalCenter(CC).pdf. Accessed March 11, 2017. 24. Bersin J. Forbes: Why Diversity and Inclusion Will be a Top Priority for 2016. https://www.forbes.com/sites/joshbersin/2015/12/06/whydiversity-and-inclusion-will-be-a-top-priority-for-2016/#350d91b52ed5. Accessed March 11, 2017. 25. Lightfoote JB, Deville C, Ma LD, et al. Diversity, inclusion, and representation: it is time to act. J Am Coll Radiol. 2016;13: 1421-1425. Edmunds LD, Ovseiko PV, Shepperd S, et al. Why do women choose or 26. reject careers in academic medicine? A narrative review of empirical evidence. Lancet. 2016;388:2948-2958. 27. National Science Board. Science & Engineering Indicators 2016. https:// www.nsf.gov/statistics/2016/nsb20161/#/report. Accessed March 11, 2017. 28. DeCastro R, Griffith KA, Ubel PA, et al. Mentoring and the career satisfaction of male and female academic medical faculty. Acad Med. 2014;89:301-311. 29. Feldman MD, Steinauer JE, Khalili M, et al. A mentor development program for clinical translational science faculty leads to sustained, improved confidence in mentoring skills. Clin Transl Sci. 2012;5:362367. 30. Pfund C, House S, Spencer K, et al. A research mentor training curriculum for clinical and translational researchers. Clin Transl Sci. 2013;6:26-33. American Society for Hematology. ASH Research Programs and 31. Awards. http://www.hematology.org/Research/Programs.aspx. Accessed March 1, 2017. 32. American Society for Radiation Oncology. ASTRO Minority Summer Fellowship Awards. https://www.astro.org/Patient-Care/Research/ Funding-Opportunities/ASTRO-Minority-Summer-Fellowship-Award/. Accessed March 1, 2017. 33. American Association for Cancer Research. Minorities in Cancer Research. http://www.aacr.org/Membership/Pages/Constituency%20Groups/ minorities-in-cancer-research___1C81B8.aspx#.WMROUhLytPc. Accessed March 1, 2017. 34. Association of American Medical Colleges. Underrepresented in Medicine Definition. https://www.aamc.org/initiatives/urm. Accessed March 1, 2017. Deville C, Chapman CH, Burgos R, et al. Diversity by race, 35. Hispanic ethnicity, and sex of the United States medical oncology physician workforce over the past quarter century. J Oncol Pract. 2014;10:e328-e334. 36. Chapman CH, Hwang WT, Deville C. Diversity based on race, ethnicity, and sex, of the US radiation oncology physician workforce. Int J Radiat Oncol Biol Phys. 2013;85:912-918. 37. Kurtin SE, Peterson M, Goforth P, et al. The advanced practitioner and collaborative practice in oncology. J Adv Pract Oncol. 2015;6: 515-527.

21. Brotherton SE, Etzel SI. Graduate medical education, 2014-2015. JAMA. 2015;314:2436-2454.

38. American Society of Clinical Oncology. The state of cancer care in America, 2016: a report by the American Society of Clinical Oncology. J Oncol Pract. 2016;12:339-383.

22. Drew SR. Promoting Diversity in Research: Championing an Inclusive Scientific Workforce. https://www.asbmb.org/asbmbtoday/ asbmbtoday_article.aspx?id=5018. Accessed March 11, 2017.

39. Winkfield KM, Flowers CR, Patel JD, et al. The American Society of Clinical Oncology strategic plan for increasing racial and ethnic diversity in the oncology workforce J Clin Oncol. In press.

22 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

INVITED ARTICLES

Gynecologic Cancer

Minimizing Minimally Invasive Surgery for Endometrial Carcinoma Melissa K. Frey, MD, Stephanie V. Blank, MD, and John P. Curtin, MD

ndometrial cancer is the most common gynecologic malignancy, and, in contrast to many other cancer types, the incidence and mortality of endometrial cancer continue to grow. In the United States, there were approximately 40,000 cases of endometrial cancer in 2006; however, in 2017, there will be an estimated 61,380 new cases and 10,920 deaths.1 The growing obesity epidemic is a considerable contributor to this trend, as more than half of endometrial cancers are attributable to obesity.2,3 Furthermore, as obesity rates continue to rise, the incidence of endometrial cancer is expected to increase. Models predict an incidence of 42.13 cases per 100,000 women by the year 2030, representing a 55% increase over 2010.4,5 Given the substantial increase in the incidence of endometrial cancer, close association with obesity, and the increased prevalence among premenopausal women, management approaches that limit extensive surgical staging will become increasingly important. The Gynecologic Oncology Group LAP2 trial established the oncologic safety of minimally invasive surgery for the treatment of endometrial cancer. This study also demonstrated a reduction in postoperative adverse events and improved quality of life with a minimally invasive approach.6 The LAP2 results culminated in the American College of Obstetricians and Gynecologists (ACOG) and Society of Gynecologic Oncology (SGO) practice bulletin stating that minimally invasive surgery should be embraced as the standard surgical approach for comprehensive surgical staging in women with endometrial cancer.7 Minimally invasive surgery is especially important for obese patients, as obesity has been independently associated with increased surgical complications, and surgical morbidity is most profound in open surgery.8,9 In the LAP2 study, there was a direct relationship between patient body mass index and conversion from laparoscopic approach to laparotomy. In part, this was due to the protocol mandate that all patients have pelvic and para-aortic lymph node sampling performed.

In this review article, we will review two methods that can further minimize minimally invasive surgery for endometrial cancer: (1) assessment of lymph nodes with sentinel lymph node (SLN) mapping, and (2) ovarian preservation at the time of endometrial cancer surgery. We purport that surgical approaches that reduce minimally invasive surgery are essential to the development of safe and cost-effective treatments for patients with endometrial cancer. Additionally, because the majority of women will survive and surpass their endometrial cancer, it is increasingly important to consider the long-term health implications of their treatments and optimize survivorship.

SLN MAPPING

Pelvic and para-aortic lymphadenectomy has been included in the surgical staging criteria for endometrial cancer since 1988.10 Lymph node status is the most important predictor of survival and provides risk assessment that guides postoperative treatment planning.11 The SEPAL study suggested a therapeutic effect of lymphadenectomy, with significantly longer overall survival among patients who had pelvic and para-aortic lymphadenectomy in this retrospective analysis.12 However, two randomized controlled trials have failed to show a survival benefit with pelvic and selective para-aortic lymphadenectomy.13,14 Lymphadenectomy has been associated with prolonged operating time, additional cost, and increased morbidity including lymphedema, lymphocysts, and neuralgia.15 Although the therapeutic benefit of lymphadenectomy remains controversial, most agree that lymph node status can help determine which patients should undergo adjuvant therapy and which patients can avoid additional cancer-directed treatment and the associated morbidity. A Surveillance, Epidemiology, and End Results (SEER) database study demonstrated that patients who undergo lymphadenectomy are less likely to receive pelvic radiation.11 The SLN is the first node to receive drainage from a primary tumor. This lymph node, therefore, is most likely to

From the New York-Presbyterian Hospital, Weill Cornell Medicine, New York, NY; Icahn School of Medicine at Mount Sinai, New York, NY; NYU Langone Medical Center, New York, NY. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Melissa K. Frey, MD, New York-Presbyterian Hospital, Weill Cornell Medicine, 525 E. 68th St., Suite J-130, New York, NY 10065; email: mkf2002@med. cornell.edu. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 23

FREY, BLANK, AND CURTIN

harbor cancer cells for those cancers that spread via the lymphatic system. SLN mapping and ultrastaging of SLNs have been proposed as a surgical method to reduce the morbidity of surgical staging while maintaining the prognostic information of lymph node status assessment. SLN mapping and ultrastaging are currently considered standard of care for the surgical staging of breast cancer, melanoma, and vulvar cancer.16,17

SLN Mapping Technique

Most of the early studies of endometrial cancer SLN mapping used a combination of radioactive tracer with lymphoscintigraphy or single-photon emission CT (SPECT-CT) and colored dye (patent blue, isosulfan blue, and methylene blue) to visualize nodes.18 Drawbacks of radioactive tracers include difficulty detecting SLN close to the cervix as the gamma-probe detects high activity from the cervical injection site, patient inconvenience of having to undergo preoperative injection and imaging, and costly resources and equipment that are not available to all surgeons.19 With the current widespread availability of the robotic platform and near-infrared imaging, many surgeons have replaced the dual injections with indocyanine green (ICG) and immunofluorescence detection (Fig. 1). ICG injection seems to negate the higher rates of failed SLN mapping observed in obese patients, possibly because of differences in the molecular weight of the isosulfan blue versus ICG and that ICG is more prominently visualized in the setting of visceral and retroperitoneal fat.20 Of note, although widely used for endometrial cancer SLN mapping, ICG is not approved by the U.S. Food and Drug Administration for this indication. In other cancers in which SLN mapping is the standard of care, like breast cancer and melanoma, direct peritumoral injection is straightforward; however, this is not the case for endometrial cancer. Several injection locations have

FIGURE 1. SLN Visualized With ICG and Immunofluorescence Detection

Abbreviations: SLN, sentinel lymph node; ICG, indocyanine green.

24 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

been evaluated, including intracervical, uterine subserosal, fundal, and even peritumoral.21-23 The SGO Clinical Practice Statement supports cervical injection, stating that it is a reproducible technique that adequately maps the pelvic lymph nodes and occasionally lower aortic nodes; however, they also have noted that there are insufficient data to suggest that the upper aortic lymph nodes (above the inferior mesenteric artery) can be reliably mapped using current cervical injection techniques.16 López-De la Manzanara Cano et al24 found that a deeper injection (3 cm) had improved detection of para-aortic SLNs. Many propose restricting the cervical injection to low/intermediate-risk tumors that have a low likelihood of para-aortic involvement,25 a restriction that would include the majority of endometrial cancers. The SGO Clinical Practice Statement additionally suggests that decisions regarding para-aortic lymphadenectomy should be determined by tumor histology, intraoperative findings, and status of pelvic lymph nodes at surgery.16 The Memorial Sloan Kettering SLN algorithm mandates that failure to map a SLN results in complete lymphadenectomy on the respective hemipelvis, as well as removal of any suspicious nodes and peritoneal lesions, and meticulous ultrastaging of SLNs. Barlin et al26 found that applying this mapping algorithm significantly reduced the false-negative rate from 14.9% to 1.9%. A recently published modeling analysis proposed an approach termed SLN-restrictive frozen section strategy, in which patients who did not map SLNs would have intraoperative frozen-section evaluation, and the decision to perform a lymph node dissection would be determined by the identification of high-risk uterine features on frozen-section diagnosis.27 Whether the ideal management of failed SLN mapping involves either of these approaches remains unknown, but both present reasonable options. Pathologic ultrastaging of SLN varies among institutions, and clear guidelines have not been established for gynecologic pathologists. The process generally involves examination of multiple deeper level sections of the lymph node using routine staining and keratin immunohistochemical staining. Lymph node metastases are classified according to their size in accordance with the nomenclature used for breast cancer metastases.28 1. Macrometastasis: tumor clusters larger than 2 mm. 2. Micrometastasis: tumor clusters between 0.2 to 2 mm in size. 3. Isolated tumor cells: single tumor cells or tumor clusters that are 0.2 mm or smaller in size. 4. Isolated cytokeratin-positive cells. Kim et al29 found that SLN mapping detected additional low-volume metastases in 4.5% of patients relative to routine lymph node evaluation. Although most groups consider macrometastasis and micrometastasis to be positive SLN, the prognostic value of isolated tumor cells and isolated cytokeratin-positive cells remains uncertain. Furthermore, appropriate treatment of patients with low-volume metastatic disease is not yet known and varies by institution.

MINIMIZING MINIMALLY INVASIVE SURGERY FOR ENDOMETRIAL CARCINOMA

SLN Mapping Efficacy

Despite more than a decade of studies of SLNs in endometrial cancer, it has yet to be established as a standard of care for patients with this disease, and results in the literature vary. This is likely because of a myriad of currently used SLN mapping techniques combined with the complexity and bilaterality of the nodal basins that drain the uterus.30-33 The initial results for SLN mapping were promising, including the SENTI-ENDO trial, which found 100% negative predictive value and 100% sensitivity of SLN when considering the hemipelvis as the unit of analysis and 97% negative predictive value and 84% sensitivity when considering the patient as the unit of analysis.18 However, a meta-analysis of 26 studies found a detection rate of 78% and sensitivity of 93% and cautioned that the demonstrated good diagnostic performance of SLN mapping in endometrial cancer should be interpreted with caution because of the notable small-study effect.34 A more recent meta-analysis identified a higher pooled detection rate (81%) and sensitivity of 96% for detecting lymphatic metastases, rates that approach those observed in breast cancer and melanoma.32 The authors suggest that these improvements may reflect gynecologic surgeons’ growing experience with SLN mapping and increased use of more innovative dye and detection techniques. To account for the learning curve, the SGO Clinical Practice Statement suggests that surgeons should train by performing SLN dissection and then lymphadenectomy on 20 patients prior to adopting SLN as their standard surgical method.

Benefits of SLN Mapping

The most important advantages of SLN mapping include improved detection of metastatic disease through ultrastaging of lymph nodes and reduction in morbidity by eliminating the complete lymph node dissection. Although finding metastases that would have otherwise been missed seems valuable, some might argue that altering therapy based on this information results in overtreatment, and the use of SLN in endometrial cancer has not been shown to improve oncologic outcomes. The inclusion of lymph node dissection in endometrial cancer staging procedures has been shown to increase operative room time, surgical blood loss, length of hospital stay, and morbidity, including permanent lymphedema. Dowdy et al15 found that complications in the first 30 days following surgery occurred in 19.3% versus 37.5% of patients in the non–lymph node dissection versus lymph node dissection group. The 30-day cost-of-care was also found to be significantly higher in the lymph node dissection group, correlating directly with increasing severity of adverse events among these patients. There are yet to be prospective evaluations of the morbidity of SLN mapping in endometrial cancer; however, most would agree that these patients should have an experience that more closely emulates the patients without complete lymphadenectomy.

OVARIAN PRESERVATION

Whereas many debate the necessity of nodal assessment as part of the treatment of patients with endometrial

cancer, hysterectomy and bilateral salpingo-oophorectomy are standard. The rationale for ovarian removal includes detection and removal of occult metastatic disease as well as synchronous ovarian cancers and diminishment of estrogen production. With an amplified incidence of endometrial cancer along with an increasing proportion of diagnoses occurring in younger women, the number of premenopausal women losing their ovaries to endometrial cancer will grow. Almost one-fourth of U.S. women with endometrial cancer are premenopausal at diagnosis,35 and other reported incidences are even higher, such as in Korea, with 45% of cases occurring in premenopausal women, 10% in women under 40.36,37 Removing the ovaries in premenopausal women subjects them to surgical menopause and its attendant symptoms of estrogen deprivation, along with increased risk of cardiovascular disease, osteoporotic fractures, cognitive impairment, and possibly diminished survival, although the most quoted study demonstrating survival benefit to ovarian retention did not include women with cancer.35,38,39 Citing these concerns, several groups have considered the safety of ovarian preservation among young women with endometrial cancer. A query of SEER data found that ovarian preservation in women under 45 with low-grade early-stage endometrial cancer had no effect on either cancer-specific or overall survival.40 A population-based analysis using the National Cancer Database compared the 7% of women with stage I endometrial cancer under the age of 50 with retained ovaries to those who underwent bilateral oophorectomy and also found that ovarian conservation did not adversely affect oncologic outcomes.41 A nationwide study from tumor registries in Korea evaluated 175 women with endometrial cancer who retained their ovaries. With median follow-up of 55 months, 4% recurrence risk was noted and none in patients with stage I endometrioid tumors. All patients who recurred had risk factors including a nonendometrioid histology, contralateral adnexal involvement, or deep myometrial or cervical stromal invasion. The authors concluded that in selected patients (Table 1), oophorectomy need not be a mandatory component of standard surgical therapy for endometrial cancer.42 More recently, a follow-up SEER study of ovarian conservation in young women with early-stage low-grade endometrial cancer found the same cause-specific survival for retaining and removing ovaries but improved overall survival with ovarian retention as well as a lower cumulative risk of death from cardiovascular disease.43 This finding is not necessarily surprising. Several SEER studies have shown that among women with favorable endometrial cancers, cardiovascular disease is a more probable cause of death than is cancer,44,45 in part because of the high likelihood of curative cancer treatment46-52 and the prevalence of cardiovascular disease, especially among patients with endometrial cancer whose risk factors for endometrial cancer are also risk factors for cardiovascular disease. Additionally, when the indication for hysterectomy is not cancer, the case for ovarian retention is compelling. The Nurses’ Health Study showed that all-cause mortality, coronary heart disease mortality, and deaths from all cancers asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 25

FREY, BLANK, AND CURTIN

TABLE 1. Proposed Indications for Ovarian Preservation in Patients With Endometrial Cancer42 Proposed Indicators Patients who want to retain ovarian function. No gross intraoperative extra-uterine tumor spread. No gross abnormality in bilateral ovaries. Negative results of frozen biopsy for lymph nodes suspicious for metastasis. Endometrioid-type histology in preoperative biopsy. Patients who have no inherited predisposition to breast or ovarian cancer.

were diminished when ovaries were retained at the time of hysterectomy for benign disease versus when ovaries were removed.39 Several smaller cohort studies similarly confirm this association.53-56 More recently, a nationwide study in the United Kingdom compared ovarian removal to conservation and found conservation to be associated with lower all-cause mortality as well as lower death rates from heart disease and cancer, causing the authors to conclude that removing ovaries to prevent ovarian cancer comes at the cost of an increased risk of cardiovascular disease and other more prevalent cancers and higher overall mortality.57

CONCLUSION

We have explored two methods to additionally minimize minimally invasive surgery for endometrial cancer—SLN

mapping and ovarian preservation. The greatest obstacle in adopting SLN mapping as standard of care for endometrial cancer is the lack of large prospective studies that perform complete systematic pelvic and para-aortic lymph node dissection as a true control arm. However, according to a survey of SGO members, 28.6% of respondents performed SLN mapping, 16.7% with the exclusion of pelvic lymphadenectomy, and 54% of institutions performed pathologic ultrastaging of SLN.16 Furthermore, the National Comprehensive Cancer Network guidelines for endometrial carcinoma now include a SLN algorithm as an option for surgical management of endometrial cancer. The uncertainty surrounding the value of lymph node assessment in endometrial cancer is not likely to be resolved in the near future. However, SLN mapping is emerging as an effective surgical technique to allow tailored adjuvant therapy for high-risk patients while minimizing the risk of harm that occurs with a complete lymphadenectomy. A compelling case also can be made for ovarian retention in women with early-stage, early-grade endometrioid endometrial cancer, particularly in premenopausal women, but potentially in older women as well. Ovarian preservation has been associated with improved overall survival and lower risk of cardiovascular disease in many studies. This is not to say that ovarian preservation should be standard of care in these cases but that ovarian extirpation should not be automatic for all women with endometrial cancer, as a select group may benefit from keeping their ovaries.

References 1. American Cancer Society. Cancer Facts and Figures 2017. https:// www.cancer.org/research/cancer-facts-statistics/all-cancer-factsfigures/cancer-facts-figures-2017.html. Accessed February 15, 2017.

Gunderson CC, Java J, Moore KN, et al. The impact of obesity on 9. surgical staging, complications, and survival with uterine cancer: a Gynecologic Oncology Group LAP2 ancillary data study. Gynecol Oncol. 2014;133:23-27.

2. Calle EE, Kaaks R. Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms. Nat Rev Cancer. 2004;4: 579-591.

10. Shepherd JH. Revised FIGO staging for gynaecological cancer. Br J Obstet Gynaecol. 1989;96:889-892.

3. Renehan AG, Tyson M, Egger M, et al. Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet. 2008;371:569-578.

11. Sharma C, Deutsch I, Lewin SN, et al. Lymphadenectomy influences the utilization of adjuvant radiation treatment for endometrial cancer. Am J Obstet Gynecol. 2011;205:562.e1-562.e9.

4. Sheikh MA, Althouse AD, Freese KE, et al. USA endometrial cancer projections to 2030: should we be concerned? Future Oncol. 2014;10:2561-2568. 5. Onstad MA, Schmandt RE, Lu KH. Addressing the role of obesity in endometrial cancer risk, prevention, and treatment. J Clin Oncol. 2016;34:4225-4230. 6. Walker JL, Piedmonte MR, Spirtos NM, et al. Laparoscopy compared with laparotomy for comprehensive surgical staging of uterine cancer: Gynecologic Oncology Group Study LAP2. J Clin Oncol. 2009;27: 5331-5336.

12. Todo Y, Kato H, Kaneuchi M, et al. Survival effect of para-aortic lymphadenectomy in endometrial cancer (SEPAL study): a retrospective cohort analysis. Lancet. 2010;375:1165-1172. 13. ASTEC study group; Kitchener H, Swart AM, Qian Q, et al. Efficacy of systematic pelvic lymphadenectomy in endometrial cancer (MRC ASTEC trial): a randomised study. Lancet. 2009;373:125-136. 14. Benedetti Panici P, Basile S, Maneschi F, et al. Systematic pelvic lymphadenectomy vs. no lymphadenectomy in early-stage endometrial carcinoma: randomized clinical trial. J Natl Cancer Inst. 2008;100: 1707-1716.

7. Practice Bulletin No. 149: Endometrial cancer. Obstet Gynecol. 2015;125:1006-1026.

15. Dowdy SC, Borah BJ, Bakkum-Gamez JN, et al. Prospective assessment of survival, morbidity, and cost associated with lymphadenectomy in low-risk endometrial cancer. Gynecol Oncol. 2012;127:5-10.

8. Bouwman F, Smits A, Lopes A, et al. The impact of BMI on surgical complications and outcomes in endometrial cancer surgery--an institutional study and systematic review of the literature. Gynecol Oncol. 2015;139:369-376.

16. Society of Gynecologic Oncology. SGO Clinical Practice Statement: The Role of Sentinel Lymph Node Mapping in Endometrial Cancer. https:// www.sgo.org/clinical-practice/guidelines/the-role-of-sentinel-lymphnode-mapping-in-endometrial-cancer/. Accessed March 7, 2017.

26 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

MINIMIZING MINIMALLY INVASIVE SURGERY FOR ENDOMETRIAL CARCINOMA

17. Abu-Rustum NR. Sentinel lymph node mapping for endometrial cancer: a modern approach to surgical staging. J Natl Compr Canc Netw. 2014;12:288-297. 18. Ballester M, Dubernard G, Lécuru F, et al. Detection rate and diagnostic accuracy of sentinel-node biopsy in early stage endometrial cancer: a prospective multicentre study (SENTI-ENDO). Lancet Oncol. 2011;12:469-476. 19. Cormier B, Rozenholc AT, Gotlieb W, et al; Communities of Practice (CoP) Group of Society of Gynecologic Oncology of Canada (GOC). Sentinel lymph node procedure in endometrial cancer: a systematic review and proposal for standardization of future research. Gynecol Oncol. 2015;138:478-485. 20. Sinno AK, Fader AN, Roche KL, et al. A comparison of colorimetric versus fluorometric sentinel lymph node mapping during robotic surgery for endometrial cancer. Gynecol Oncol. 2014;134:281-286. 21. Abu-Rustum NR, Khoury-Collado F, Gemignani ML. Techniques of sentinel lymph node identification for early-stage cervical and uterine cancer. Gynecol Oncol. 2008;111(Suppl):S44-S50. 22. Khoury-Collado F, Abu-Rustum NR. Lymphatic mapping in endometrial cancer: a literature review of current techniques and results. Int J Gynecol Cancer. 2008;18:1163-1168. Abu-Rustum NR. Update on sentinel node mapping in uterine cancer: 23. 10-year experience at Memorial Sloan-Kettering Cancer Center. J Obstet Gynaecol Res. 2014;40:327-334. 24. López-De la Manzanara Cano C, Cordero García JM, Martín-Francisco C, et al. Sentinel lymph node detection using 99mTc combined with methylene blue cervical injection for endometrial cancer surgical management: a prospective study. Int J Gynecol Cancer. 2014;24:1048-1053. Abu-Rustum NR, Gomez JD, Alektiar KM, et al. The incidence of 25. isolated paraaortic nodal metastasis in surgically staged endometrial cancer patients with negative pelvic lymph nodes. Gynecol Oncol. 2009;115:236-238. 26. Barlin JN, Khoury-Collado F, Kim CH, et al. The importance of applying a sentinel lymph node mapping algorithm in endometrial cancer staging: beyond removal of blue nodes. Gynecol Oncol. 2012;125:531-535. Sinno AK, Peijnenburg E, Fader AN, et al. Reducing overtreatment: 27. a comparison of lymph node assessment strategies for endometrial cancer. Gynecol Oncol. 2016;143:281-286. 28. Schwartz GF, Giuliano AE, Veronesi U; Consensus Conference Committee. Proceedings of the consensus conference on the role of sentinel lymph node biopsy in carcinoma of the breast April 19 to 22, 2001, Philadelphia, Pennsylvania. Hum Pathol. 2002;33:579-589. 29. Kim CH, Soslow RA, Park KJ, et al. Pathologic ultrastaging improves micrometastasis detection in sentinel lymph nodes during endometrial cancer staging. Int J Gynecol Cancer. 2013;23:964-970. Mansel RE, Fallowfield L, Kissin M, et al. Randomized multicenter trial 30. of sentinel node biopsy versus standard axillary treatment in operable breast cancer: the ALMANAC Trial. J Natl Cancer Inst. 2006;98:599-609.

INternational Study on Sentinel nodes in Vulvar cancer (GROINSS-V) I. Gynecol Oncol. 2016;140:8-14. Kang S, Yoo HJ, Hwang JH, et al. Sentinel lymph node biopsy in 34. endometrial cancer: meta-analysis of 26 studies. Gynecol Oncol. 2011; 123:522-527. 35. Rocca WA, Bower JH, Maraganore DM, et al. Increased risk of parkinsonism in women who underwent oophorectomy before menopause. Neurology. 2008;70:200-209. 36. SOG Gynecologic Oncology Committee. Annual report of gynecologic cancer registry program in Korea for 2004. Korean J Obstet Gynecol. 2007;50:28-78. 37. Lee SE, Kim JW, Park NH, et al. Contemporary trends of endometrial cancer in Korean women. Korean J Gynecol Oncol. 2005;16:215-220. Shuster LT, Gostout BS, Grossardt BR, et al. Prophylactic oophorectomy 38. in premenopausal women and long-term health. Menopause Int. 2008;14:111-116. 39. Parker WH, Broder MS, Chang E, et al. Ovarian conservation at the time of hysterectomy and long-term health outcomes in the nurses’ health study. Obstet Gynecol. 2009;113:1027-1037. 40. Wright JD, Buck AM, Shah M, et al. Safety of ovarian preservation in premenopausal women with endometrial cancer. J Clin Oncol. 2009;27:1214-1219. 41. Matsuo K, Machida H, Shoupe D, et al. Ovarian conservation and overall survival in young women with early-stage low-grade endometrial cancer. Obstet Gynecol. 2016;128:761-770. 42. Lee TS, Kim JW, Kim TJ, et al; Korean Gynecologic Oncology Group. Ovarian preservation during the surgical treatment of early stage endometrial cancer: a nation-wide study conducted by the Korean Gynecologic Oncology Group. Gynecol Oncol. 2009;115:26-31. 43. Wright JD, Jorge S, Tergas AI, et al. Utilization and outcomes of ovarian conservation in premenopausal women with endometrial cancer. Obstet Gynecol. 2016;127:101-108. 44. Ward KK, Shah NR, Saenz CC, et al. Cardiovascular disease is the leading cause of death among endometrial cancer patients. Gynecol Oncol. 2012;126:176-179. 45. Felix AS, Bower JK, Pfeiffer RM, et al. High cardiovascular disease mortality after endometrial cancer diagnosis: results from the Surveillance, Epidemiology, and End Results (SEER) Database. Int J Cancer. 2017;140:555-564. Lee NK, Cheung MK, Shin JY, et al. Prognostic factors for uterine cancer 46. in reproductive-aged women. Obstet Gynecol. 2007;109:655-662. 47. Gitsch G, Hanzal E, Jensen D, et al. Endometrial cancer in premenopausal women 45 years and younger. Obstet Gynecol. 1995;85:504-508. 48. Duska LR, Garrett A, Rueda BR, et al. Endometrial cancer in women 40 years old or younger. Gynecol Oncol. 2001;83:388-393. 49. Crissman JD, Azoury RS, Barnes AE, et al. Endometrial carcinoma in women 40 years of age or younger. Obstet Gynecol. 1981;57:699-704.

31. Niebling MG, Pleijhuis RG, Bastiaannet E, et al. A systematic review and meta-analyses of sentinel lymph node identification in breast cancer and melanoma, a plea for tracer mapping. Eur J Surg Oncol. 2016;42:466-473.

50. Evans-Metcalf ER, Brooks SE, Reale FR, et al. Profile of women 45 years of age and younger with endometrial cancer. Obstet Gynecol. 1998;91:349-354.

32. Bodurtha Smith AJ, Fader AN, Tanner EJ. Sentinel lymph node assessment in endometrial cancer: a systematic review and metaanalysis. Am J Obstet Gynecol. Epub 2016 Nov 18.

51. Gallup DG, Stock RJ. Adenocarcinoma of the endometrium in women 40 years of age or younger. Obstet Gynecol. 1984;64:417-420.

33. Te Grootenhuis NC, van der Zee AG, van Doorn HC, et al. Sentinel nodes in vulvar cancer: long-term follow-up of the GROningen

52. Tran BN, Connell PP, Waggoner S, et al. Characteristics and outcome of endometrial carcinoma patients age 45 years and younger. Am J Clin Oncol. 2000;23:476-480.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 27

FREY, BLANK, AND CURTIN

53. McCarthy AM, Menke A, Ouyang P, et al. Bilateral oophorectomy, body mass index, and mortality in U.S. women aged 40 years and older. Cancer Prev Res (Phila). 2012;5:847-854.

56. Rocca WA, Grossardt BR, de Andrade M, et al. Survival patterns after oophorectomy in premenopausal women: a population-based cohort study. Lancet Oncol. 2006;7:821-828.

54. Rivera CM, Grossardt BR, Rhodes DJ, et al. Increased cardiovascular mortality after early bilateral oophorectomy. Menopause. 2009;16:15-23.

57. Mytton J, Evison F, Chilton PJ, et al. Removal of all ovarian tissue versus

55. Rivera CM, Grossardt BR, Rhodes DJ, et al. Increased mortality for neurological and mental diseases following early bilateral oophorectomy. Neuroepidemiology. 2009;33:32-40.

patients with benign disease: study using routine data and data

28 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

conserving ovarian tissue at time of hysterectomy in premenopausal linkage. BMJ. 2017;356:j372.

INVITED ARTICLES

Global Health

The Road to Addressing Noncommunicable Diseases and Cancer in Global Health Policy Heath Catoe, MD, PhD, Jordan Jarvis, MSc, Sudeep Gupta, MD, MBBS, Ophira Ginsburg, MD, FRCPC, and Gilberto de Lima Lopes Jr., MD, MBA, FAMS

remature death and disability from cancer and other noncommunicable diseases (NCDs)—such as diabetes, heart disease, chronic respiratory disease, and others—are on a rapid rise in low- and middle-income countries. Whereas in 1990, 57% of global deaths were attributed to NCDs, they accounted for 70% (38.3 million) in 2013, with 80% of premature deaths reported in low- and middle-income countries.1,2 Historically viewed as conditions largely affecting rich countries and elderly populations, global NCDs were long neglected as a development and even health priority in resource-limited settings. Now, these countries are experiencing an epidemiologic transition in which more patients are afflicted by NCDs, with longer suffering and death at younger ages than in high-income countries.3 Rising global rates of NCDs have enormous economic implications, estimated at a cumulative loss of 47 trillion between 2011 and 2030.4 Premature NCD death and disability, defined by the World Health Organization (WHO) as those younger than age 70, result in people working fewer years, with lower productivity, and result in higher costs to both the health system and individuals.5,6 As NCDs are both a cause and a consequence of poverty, they are a threat to sustainable human development on a global scale. Social, economic, and environmental factors, such as globalization, international trade, urbanization, education, labor practices, household income, and food production, all serve as risk factors for NCDs. This previously under-recognized crisis underlies the importance of global coordinated action to increasingly recognize NCDs as a political issue. In many cases, we have solutions in the form of scientific and technical progress, but these are insufficiently implemented due to a lack of political will. Thus, a movement of stakeholders from across nongovernmental organizations, patient groups, academia, intergovernmental organizations, private sector, and governments have been working to advance global, regional,

and national policy and time-bound measurable commitments to reduce the global burden of NCDs. The inception of global policy to address the growing NCD crisis effectively dates back to the June 1992 United Nations (UN) Conference on Environment and Development in Rio de Janeiro, Brazil. Several key meetings under the auspices of the UN ensued, with a growing consensus and emerging multisector partnerships that would support the path to a set of NCD goals. The 2002 Johannesburg Declaration on Sustainable Development formally addressed the issue of NCDs, and the 2009 Economic and Social Council Ministerial Declaration recognized the burden NCDs placed on countries. These meetings helped lay out frameworks and goals that would eventually lead to the UN High-level Meeting in 2011, a high mark in the global effort to address NCDs.6

THE ROLE OF CIVIL SOCIETY

Whereas the HIV/AIDS movement had strong grassroots advocacy with the voices of patients featuring prominently, technical and policy discussions have been more dominant in driving the early stages of the NCD movement. However, an early catalyst for the NCD movement was a group of patients with diabetes in the Caribbean who drew attention to their preventable foot amputations and lack of prioritization of chronic diseases as a human rights concern.7 They advocated to their governments, leading to the first-ever summit on NCDs involving heads of state in Trinidad and Tobago in September 2007 in which the Port-ofSpain Declaration “Uniting to Stop the Epidemic of Chronic Non-Communicable Diseases” was issued.8 Leaders of CARICOM, a group of 20 countries within the UN system, raised their concerns about diabetes and other NCDs at the UN, setting the stage for the UN High-level Meeting to take place in 2011. Strong global-level civil society mobilization on NCDs began in 2009, with the formation of the NCD Alliance

From the Global Oncology Program, Sylvester Comprehensive Cancer Center at the University of Miami, Miami, FL; Young Professionals Chronic Disease Network, Boston, MA; Department of Medical Oncology, Tata Memorial Centre, Mumbai, India; Laura and Isaac Perlmutter Cancer Center at NYU Langone and Department of Population Health, NYU School of Medicine, New York, NY; Global Oncology Program, Sylvester Comprehensive Cancer Center at the University of Miami, Miami, FL. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Gilberto de Lima Lopes Jr., MD, MBA, FAMS, 1120 NW 14th St., Suite 610N, Miami, FL, 33136; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 29

CATOE ET AL

(https://ncdalliance.org) and other civil society groups, such as the Young Professionals Chronic Disease Network (www. ncdaction.org). As a coalition of disease federations, including the International Diabetes Federation, the World Heart Federation, the Union for International Cancer Control, and the International Union Against Tuberculosis and Lung Disease, the NCD Alliance was established to consolidate funding and influence for policy changes on NCDs as a group and avoid disease silos in global policy for health. Civil society, including academics, pushed for international support and clarification of goals in the form of a political declaration to address NCDs, leading up to the UN High-level Meeting on the Prevention and Control of Non-communicable Diseases in 2011,9,10 at which international heads of state assembled for the second-ever UN High-level Meeting on a health issue.

UNITED NATIONS HIGH-LEVEL MEETING 2011

As part of its wide-ranging mandate, the UN General Assembly convenes high-level meetings to increase awareness while pressing for common ground and policies on issues of global importance.11 Participation in the September 2011 meeting included 82 member states, including representation from 35 heads of state, civil society, private sector, and UN agencies. The political declaration on NCDs emerged after controversial negotiations, largely between groups of developed versus developing nations, on access to medicines, food and beverage policies, tobacco control, and financing commitments.12 Most notably, tensions arose with the addition of intellectual property provisions for the World Trade Organization’s Trade-Related Aspects of Intellectual Property Rights (TRIPS) agreement. These TRIPS “flexibilities” give countries the right, if criteria are met,13 to circumvent patents to issue a compulsory license for medicines considered essential for the public health good at an affordable price.14 Although the HIV/AIDS movement emphasized the right to treatment, this discourse was largely muted at the NCD High-level Meeting. To frame NCDs as a social justice issue and call for action and commitments, students, AIDS activists, and people living with NCDs held the first rally on NCDs outside the UN in New York.7 The UN political declaration on the Prevention and Control of Non-communicable Diseases passed, outlining policies to address the four major NCDs defined by WHO, which are cancer, cardiovascular disease, chronic respiratory diseases, and diabetes. Policies responding to current disease treatment revolved around improving health care infrastructure and systems including increased technical resources.9 The impetus for such shift in policies relates to the goals of sustainable global development based on three pillars defined by economic growth, social equity, and environmental protection.6 Some of the specific policies implemented soon after the 2011 meeting include those from Gabon with free screening for cancers, Niger aiming to ensure access and affordability of medicines, Spain rethinking health strategies on cancer, Mexico funding evidence-based clinical interventions and new technologies, Trinidad and Tobago starting a 30 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Chronic Disease Assistance Program to provide all people with medications, UNASUR (Union of South American Nations) aimed to ensure universal access to medicines, using TRIPS, and India further developing its National Program for Prevention and Control of Cancer to screen for disease.15 To provide member states with a clear road map to address NCDs, WHO was tasked to develop a set of targets, which now guide the global NCD response in the form of the NCD global monitoring framework and the NCD Global Action Plan 2013–2020.16 The targets included advocating and raising awareness, disseminating knowledge and information, encouraging innovation and identifying barriers, advancing multisector action, and advocating for the mobilization of resources. An initial set of nine voluntary targets and 25 indicators for 2025 goals that provided an assessment of NCD mortality and morbidity, risk factors, and national systems response was developed. These targets include: 25% reduction in premature mortality from NCDs, 10% reduction in harmful use of alcohol, 10% reduction in physical inactivity, 30% reduction in tobacco use, 80% coverage of essential NCD medicines and technologies, and 50% coverage of drug therapy and counseling.17

ASSESSMENT THUS FAR

The formal process for monitoring progress on the goals is conducted under WHO auspices via country surveys to assess national capacity for the prevention and control of NCD from the 194 member states. These surveys had been ongoing for several years prior, but the questions varied significantly from prior versions, making comparisons problematic, and validation of the data proved challenging.18 The first survey following the meeting was in 2013 and showed overall progress and improvements, specifically operational national NCD policy with a budget for implementation increased from 32% of countries in 2010 to 50% of countries in 2013, while highlighting the challenges faced by nearly all member states. The WHO Global Survey was comparable in 2010, 2013, and 2015. Questions include public health infrastructure, partnerships and multisector collaboration for NCDs, the existence of NCD-relevant policies, strategies and action plans, capacity for surveillance to address NCDs and their risk factors at the national level, and capacity for NCD prevention, early detection, treatment, and care within the health system. Comparison of 2010 to 2013 showed increases in designating a unit, branch, or department with responsibility for NCDs from 89% to 94%, with a slight decline to 93% by 2015. Increased funding for NCD prevention and health promotions was shown from 81% to 88% from 2010 to 2015. During this period, 11% versus 6% reported absence of funding for NCDs with the variety in funding sources increasing, the most prevalent being general government revenues. In 2015, the first assessed prevalence of palliative care funding was found in 64% of countries.18 Although policies were prevalent, operational policies took time to increase from 33% in 2010 to 63% in 2015. Operational plans addressing cancer specifically increased from 50% in 2010 to 71% in

THE ROAD TO GLOBAL HEALTH IN ONCOLOGY

2015. Cancer registries slightly increased from 80% to 81% to 84%, whereas national, population-based cancer registries changed from 39% to 59%.18 Monitoring risk factors with surveys increased with each assessment, especially tobacco use surveys. Primary prevention and health promotion increased as well as risk factor detection during this period. Although some form of guidelines for cancer management existed for 73% in 2013, only about half of those had fully implemented them. However, in 2015, evidence-based guidelines were assessed, revealing 60% of countries had them, with approximately 55% having some form of implementation of those guidelines. Prevalence of tests and procedures available increased, such as breast cancer mammogram from 81% to 84% as well as cervical cancer from 65% to 74% from 2010 to 2013. Availability of many essential medicines increased, but, for example, oral morphine went from 48% to 56% to 43% from 2010 to 2015. In 2015, 67% of countries had cancer centers or cancer departments at the tertiary level. However, availability of cancer surgery (69%) and subsidized chemotherapy (63%) were distinctly influenced by country income group.18,19 One important point to draw from the WHO global surveys is cancer treatment availability is correlated with income of country. With the 2011 high-level meeting, there may have been hopes to replicate the effect of the high-level meeting for HIV/AIDS in 2001 on funding and donations, which surged afterward, alleviating the treatment-access crisis at that time.9 However, there remains a dearth of domestic funding information on NCD programs.10 As was seen in the 2013 survey, there was policy in place for NCD work but less actually working plans in place, suggesting the difficulty of implementation. Funding is an important consideration with enacting domestic policies for prevention and treatment; however, there is a complex network of factors affecting individual countries in implementing domestic policies. A total of 77% to 87% of low- and middle-income countries in 2015 had a major source of funding for NCD programs come from international donors.19 However, analysis of this funding by the WHO Working Group indicated that donor assistance for health goes mainly to other areas besides NCDs, even though NCDs are more of a health burden.20 Current funding for NCD policies in low- and middle-income countries could be improved by better allocation of international assistance.

FACTORS IN TREATMENT IMPLEMENTATION

Although policy and goals can be set for prevention and treatment of cancer, implementation of such policies is complex, with barriers that will require creative solutions. The high costs of cancer medications can often be an insurmountable barrier to treatment in poor countries. This is in part due to new medications covered under patent laws that are priced for high-income countries.14 International patent law is guided by different conventions, partnerships, and agreements, including obligations under the World Trade Organization and TRIPS agreement.13,21,22 To obtain access to cancer medications, countries can use compulsory

licensing to lower the cost of medication through lower domestic production costs and patent fees. Although this is an ongoing point of contention in international patent law, compulsory licensing has had success before with HIV/AIDS medications.23 Creation of biosimilars, tiered price schemes, public-private partnerships, patent pools, and tax incentives are additional ways in which costs of cancer treatment can be curbed.24 Many governments are now considering how to prioritize medicines for cancer care with limited budgets. Guidance on which medications countries can consider as essential while adapting to their needs and public health priorities exists via the WHO Model List of Essential Medicines. In 2015, 16 cancer drugs were added to the existing WHO Model List of Essential Medicines, providing a suite of medications covering basic oncologic diseases, essentially prioritizing access to basic cancer treatments for 26 cancer types.25 In 2011, the WHO published its Core Medical Equipment guide to further information on essential equipment. Currently the WHO is working on its list of priority medical devices specifically for cancer to guide cost-effective procurement. These guides, expected to be published in summer 2017, will help the lower income, poor resource, and poor infrastructure countries purchase priority medical products with a costeffective logic in mind.

A COUNTRY EXAMPLE: INDIA IN THE SPOTLIGHT

To more thoroughly understand the whole of challenges faced by countries implementing strategies for cancer prevention and treatment, we will look at India’s current state of affairs. India’s National Cancer Control Program is a federally coordinated program that was launched in 1975 with the main aims of creating infrastructure for primary prevention, early detection, and treatment of cancers. The main lynchpin of strategy for providing therapeutic care was the creation of so-called Regional Cancer Centres that are so located as to bridge geographic gaps in the availability of public-sector cancer treatment facilities. There are 27 Regional Cancer Centres in India at present. However, infrastructure and human resource for cancer treatment remain inadequate for a country of India’s physical size and population. For example, there are less than 1,000 trained medical oncologists and only about 340 radiation therapy centers in India.26 A considerable fraction of health care capacity, including cancer care, in India is provided by private-sector industry for which services are beyond the reach of a majority of patients. The added complexity is with grossly uneven distribution of health care services in India in which the majority of infrastructure and human resources are located in urban regions, but the majority of the population resides in rural areas. There have been recent proposals to initiate basic cancer care services, including surgery and chemotherapy for common cancers, at the level of district hospitals, of which there are about 600 in India.26 Several Indian states have initiated programs to deliver cancer care in rural and semi-urban regions along these lines.27 Recently, asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 31

CATOE ET AL

there has been improvement in human resource availability in underserved areas. Several strategies, some incentivized, have been initiated by the federal government to improve rural medical service.28 There has also been a substantial increase in number of postgraduate training positions in oncology in the past few years. An important recent initiative to link more than 80 public- and private-sector cancer centers in India, called the National Cancer Grid, is also noteworthy in this context.29 Its main aim is to provide uniform, evidence-based cancer care across different geographic areas in India. One important aspect of India’s cancer-control scenario is a relatively good quality national cancer registration program that has existed under the aegis of the Indian Council of Medical Research since 1981 and currently comprises 23 population-based cancer registries that collect data on cancer incidence from defined geographic regions. Most of these registries are located in urban areas, and very few collect data on mortality. It is estimated that there are about 1.0 to 1.1 million new cancer cases every year with about 0.5 million deaths for a mortality-to-incidence ratio of about 50%, which is much higher than that seen in developed countries.30 The main contributors to high cancer mortality are advanced stage at diagnosis and inadequate health care infrastructure, especially in rural and semi-urban areas. For example, with a much lower cancer incidence in rural compared with urban areas, the mortality of cancers in males and females is almost equal in rural (95.6–96.6/100,000) and urban (91.2–102.4/100,000) regions of India.31 The main strength of Indian cancer-control scenario is widespread availability of inexpensive medications, including anticancer drug generics, biosimilars, and copies,32 except for a few newer targeted and immunotherapy drugs. The main reason for this has been the development of a robust Indian pharmaceutical industry that has developed considerable expertise in past few decades in making generic versions of most drugs. The generic industry has been helped by legal interpretation of intellectual property rights in India that have generally been against the practice of evergreening of patents.33 India has also used, somewhat sparingly, the route of compulsory licenses for on-patent drugs

in recent years, with mixed results.14,34 The only case of compulsory licensing by India in the domain of oncology was for sorafenib. This generated strong opposition from the patent holder, Bayer. The Government of India also exercises control over pricing of drugs through periodic Drug Price Control Orders under the Essential Commodities Act enacted in 1955. Again, the use of this legislation has had mixed results, for although it has kept essential drug prices under check, scarcity has been created for some medications that have become unviable to manufacture.35 Finally, provision of palliative care for advanced-stage patients has been inadequate, with India faring poorly on the Quality of Death index among various nations.36 There needs to be urgent multisectoral action to improve availability of opioid analgesics, training of health care professionals, creation and adoption of locally relevant palliative care standards, and development of community models for providing home-based palliative care in India.

CONCLUSION

The ideals for treating NCD as a global health problem have a rich historical route that led to the high-level UN meeting in which the world’s countries united around policies enacted to address prevention and treatment. The complexity of implementing those policies, specifically related to cancer, has hindered quickly achieving the WHO global NCD targets. Information on the prevalence and type of cancer in the region, income, and infrastructure is important in making judicious allocations of resources in medication and medical equipment procurement. While considering domestic policies and implementation of cancer prevention and control, one must acknowledge international limitations placed on individual countries through policies and country income. India is a prime example of the interplay between domestic and international factors that must be considered to have an effective system for prevention and treatment of cancer, especially in a low- to middle-income nation.

ACKNOWLEDGMENT

H. Catoe and J. Jarvis contributed equally to this article as first authors. O. Ginsburg and G. Lopes contributed equally to this article as senior authors.

References 1. GBD 2013 Mortality and Causes of Death Collaborators. Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990-2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2015;385:117171. 2. World Health Organization. Noncommunicable diseases. www.who. int/mediacentre/factsheets/fs355/en/. Accessed February 26, 2017. 3. Santosa A, Byass P. Diverse empirical evidence on epidemiological transition in low- and middle-income countries: populationbased findings from INDEPTH Network Data. PLoS One. 2016;11: e0155753.

32 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

4. Bloom DE, Cafiero ET, Jáne-Lopis E, et al. The Global Economic Burden of Noncommunicable Diseases. Geneva, Switzerland: World Economic Forum; 2011. 5. Murray CJ, Barber RM, Foreman KJ, et al; GBD 2013 DALYs and HALE Collaborators. Global, regional, and national disability-adjusted life years (DALYs) for 306 diseases and injuries and healthy life expectancy (HALE) for 188 countries, 1990-2013: quantifying the epidemiological transition. Lancet. 2015;386:2145-2191. 6. The NCD Alliance. Tackling non-communicable diseases to enhance sustainable development. Presented at: World Health Summit. Berlin, Germany; October 22, 2012.

THE ROAD TO GLOBAL HEALTH IN ONCOLOGY

7. Kishore SP, Reddy KS. Non-communicable diseases: equity, action and targets. In Farrar J, Hotez P, Junghanss T, et al. (eds). Manson’s Tropical Diseases, 23rd edition. Philadelphia, PA: Elsevier; 2014;848-853.

20. Nugent R. Policy brief: Bilateral and multilateral financing for NCDs. www.who.int/nmh/ncd-coordination-mechanism/Policybrief5.2docx. pdf. Accessed March 1, 2017.

8. CARICOM. Declaration of Port-of Spain: Uniting to Stop the Epidemic of Chronic NCDs. http://caricom.org/media-center/communications/ statements-from-caricom-meetings/declaration-of-port-of-spainuniting-to-stop-the-epidemic-of-chronic-ncds#. Accessed March 7, 2017.

21. Kapczynski A. The Trans-Pacific Partnership--Is it bad for your health? N Engl J Med. 2015;373:201-203.

9. United Nations. Political declaration of the High-level Meeting of the General Assembly on the Prevention and Control of Noncommunicable Diseases. Presented at: 66th Session of UN General Assembly. New York, NY;2011. www.who.int/nmh/events/un_ncd_ summit2011/political_declaration_en.pdf. 10. Ploeg M, Paton G. NCD Civil Society Campaign Report. www.fctc. org/images/stories/NCD_civil_society_campaign_report.pdf. Accessed March 1, 2017. 11. Pan American Health Organization. FAQS on the United Nations General Assembly High-Level Meeting on the Prevention and Control of NonCommunicable Diseases. www2.paho.org/hq/index.php?option=com_ content&view=article&id=5600:2011-faqs-united-nations-generalassembly-high-level-meeting-prevention-control-ncds&catid=3697:ge neral&Itemid=4015&lang=fr. Accessed March 1, 2017. 12. Cohen D. The final declaration for the UN summit on NCDs. http:// blogs.bmj.com/bmj/2011/09/09/deborah-cohen-the-finaldeclaration-for-the-un-summit-on-ncds-2/. Accessed March 7, 2017. 13. World Trade Organization. Agreement on Trade-Related Aspects of Intellectual Property Rights. www.wto.org/english/docs_e/legal_e/ 27-trips.pdf. Accessed March 1, 2017. 14. Bognar CLFB, Bychkovsky BL, Lopes GL. Compulsory licenses for cancer drugs: does circumventing patent rights improve access to oncology medications? J Glob Oncol. 2016;2:292-301. 15. International Diabetes Federation Summary of Specific Country Commitments Made at the HLM With Key Quotes. https://ncdalliance. org/sites/default/files/rfiles/NCDA Summary of Commitments at and Post NCD Summit.pdf.2011. 16. United Nations. Outcome document of the high-level meeting of the General Assembly on the comprehensive review and assessment of the progress achieved in the prevention and control of noncommunicable diseases. Presented at: 68th Session of UN General Assembly. New York, NY;2014. 17. World Health Assembly. Follow-up to the Political Declaration of the High-level Meeting of the General Assembly on the Prevention and Control of Non-communicable Diseases. Presented at: 66th World Health Assembly Resolutions and Decisions Annexes. Geneva, Switzerland;2013. 18. Riley L, Cowan M, Guthold R. Assessing national capacity for the prevention and control of noncommunicable diseases, 2013. Report of the Americas region. www2.paho.org/hq/index.php?option=com_docman&task=doc_ view&gid=24870&Itemid=270. Accessed March 1, 2017. 19. Riley L, Cowan M. Assessing national capacity for the prevention and control of noncommunicable diseases: global survey. http://apps. who.int/iris/bitstream/10665/246223/1/9789241565363-eng.pdf. Accessed March 1, 2017.

22. Westerhaus M, Castro A. How do intellectual property law and international trade agreements affect access to antiretroviral therapy? PLoS Med. 2006;3:e332. 23. Bollyky TJ. Access to drugs for treatment of noncommunicable diseases. PLoS Med. 2013;10:e1001485. 24. World Health Organization. Global action plan for the prevention and control of noncommunicable diseases 2013-2020. http://apps. who.int/iris/bitstream/10665/94384/1/9789241506236_eng.pdf. Accessed March 1, 2017. 25. Shulman LN, Wagner CM, Barr R, et al. Proposing essential medicines to treat cancer: methodologies, processes, and outcomes. J Clin Oncol. 2016;34:69-75. 26. Gulia S, Sengar M, Badwe R, Gupta S. National Cancer Control Programme in India: proposal for organization of chemotherapy and systemic therapy services. J Global Oncol. Epub 28 Oct 2016. 27. Express News Service. Maharashtra joins hands with Tata Memorial, to train rural doctors in cancer treatment. http://indianexpress.com/ article/india/india-news-india/maharashtra-joins-hands-with-tatamemorial-to-train-rural-doctors-in-cancer-treatment-2857772/. Accessed February 26, 2017. 28. National Health Mission. NHM Components: Health Systems Strengthening: Human Resource. http://nrhm.gov.in/nrhmcomponents/health-systems-strengthening/human-resource. Accessed February 26, 2017. 29. Raghunadharao D, Kannan R, Hingnekar C, et al. Institutional external peer review: a unique National Cancer Grid initiative. Indian J Med Paediatr Oncol. 2015;36:186-188. 30. Mallath MK, Taylor DG, Badwe RA, et al. The growing burden of cancer in India: epidemiology and social context. Lancet Oncol. 2014;15:e205-e212. 31. Diksh*t R, Gupta PC, Ramasundarahettige C, et al; Million Death Study Collaborators. Cancer mortality in India: a nationally representative survey. Lancet. 2012;379:1807-1816. 32. Malhotra H. Biosimilars and non-innovator biotherapeutics in India: an overview of the current situation. Biologicals. 2011;39:321-324. 33. Kapczynski A. Engineered in India--patent law 2.0. N Engl J Med. 2013;369:497-499. 34. 't Hoen E. Access to cancer treatment: a study of medicine pricing issues with recommendations for improving access to cancer medication. http://apps.who.int/medicinedocs/documents/s21758en/s21758en. pdf. Accessed March 1, 2017. 35. Ahmad A, Khan MU, Patel I. Drug pricing policies in one of the largest drug manufacturing nations in the world: are affordability and access a cause for concern? J Res Pharm Pract. 2015;4:1-3. 36. Kar SS, Subitha L, Iswarya S. Palliative care in India: Situation assessment and future scope. Indian J Cancer. 2015;52:99-101.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 33

POINTS OF VIEW The section contains articles describing emerging, highly debated, or controversial topics in cancer research, treatment, and care to benefit patients and the field of oncology.

ARTICLES How Should We Intervene on the Financial Toxicity of Cancer Care? S. Yousuf Zafar, MD, MHS, Lee N. Newcomer, MD, Justin McCarthy, JD, Shelley Fuld Nasso, Leonard B. Saltz, MD Practice Model for Advanced Practice Providers in Oncology Jamie Cairo, DNP, APRN-BC, Mary Ann Muzi, APRN-BC, APCNP, Deanna Ficke, PA-C, Shaunta Ford-Pierce, MSN, NP, FNP-BC, Katrina Goetzke, PA-C, Diane Stumvoll, APGCNP-BC, MSN, Laurie Williams, APNP, MSN, Federico A. Sanchez, MD

FINANCIAL TOXICITY OF CANCER CARE

How Should We Intervene on the Financial Toxicity of Cancer Care? One Shot, Four Perspectives S. Yousuf Zafar, MD, MHS, Lee N. Newcomer, MD, Justin McCarthy, JD, Shelley Fuld Nasso, and Leonard B. Saltz, MD OVERVIEW The median price of a month of chemotherapy has increased by an order of magnitude during the past 20 years, far exceeding inflation over the same period. Along with rising prices, increases in cost sharing have forced patients to directly shoulder a greater portion of those costs, resulting in undue financial burden and, in some cases, cost-related nonadherence to treatment. What can we do to intervene on treatment-related financial toxicity of patients? No one party can single-handedly solve the problem, and the solution must be multifaceted and creative. A productive discussion of the problem must avoid casting blame and, instead, must look inward for concrete starting points toward improvement in the affordability and value of cancer care. With these points in mind, the authors—representatives from the pharmaceutical industry, insurance providers, oncologists, and patient advocacy—have each been asked to respond with a practical answer to the provocative hypothetical question, “If you could propose one thing, and one thing only, in terms of an action or change by the constituency you represent in this discussion, what would that be?”

ancer is one of the most expensive diseases to treat in the United States. The median price of a month of chemotherapy has increased by an order of magnitude during the past 20 years, far exceeding inflation over the same period. Some would maintain that prescribing patterns further contribute to higher costs. In the most common models of cancer care delivery, oncologists have little incentive to contain treatment costs when they prescribe chemotherapy, and only recently have considerations of cost and affordability begun to be openly incorporated into guideline development. Because of increasing deductibles, increasing premiums, cost sharing, coinsurance, and frequent copayments, patients are directly shouldering a greater portion of those costs.1 One in three American families face health care bills they cannot afford, and 50% of elderly Americans with cancer pay at least 10% of their income toward out-of-pocket treatment-related expenses.2,3 A growing body of literature has described the treatment-related financial strain experienced by patients with cancer, often called the financial toxicity of cancer treatment. These studies have described how an increasing portion of patients with cancer are at risk for cutting back on groceries, selling their homes, being nonadherent to their prescribed treatment, or—in the most extreme cases—declaring personal bankruptcy to pay for their cancer treatments.4,5

What can we do to intervene on treatment-related financial toxicity of patients? Without question, any meaningful steps toward lower costs will involve collaboration among the pharmaceutical industry, insurance providers (government and otherwise), oncologists, and patients; no one party can single-handedly solve the problem. The solution must be multifaceted and creative; prosaic appeals to simply lower drug prices surely will fail. A productive discussion of the problem must avoid casting blame and, instead, must look for common ground, and must look inward for concrete starting points toward improvement in the affordability and value of cancer care. With these points in mind, the authors of this article— representatives from the pharmaceutical industry, insurance providers, oncologists, and patient advocates—have each been asked to respond with a practical answer to the provocative hypothetical question, “If you could propose one thing, and one thing only, in terms of an action or change by the constituency you represent in this discussion, what would that be?” Note that this exercise is focused on the question of what changes we would make as opposed to the more comfortable and more often answered question of what changes others should make. We undertake this exercise with the full realization of the artificial nature of a limit to one simple answer of what is necessarily a complex, nuanced, and multifaceted problem and the

From the Duke Cancer Institute, Durham, NC; UnitedHealth Group, Minneapolis, MN; Pfizer, New York, NY; National Coalition for Cancer Survivorship, Washington, DC; Memorial Sloan Kettering Cancer Center; New York, NY. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: S. Yousuf Zafar, MD, MHS, DUMC 2715, 2424 Erwin Rd., Suite 602, Duke Cancer Institute, Durham, NC 27705; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 35

ZAFAR ET AL

awareness that that there is no singular answer that any of us could offer that will be fully inclusive and satisfactory to all voices within each of our stakeholder groups. Rather, we aim here to provide ideas to serve as starting points, both for introspection and to promote discussion.

THE PHARMACEUTICAL INDUSTRY PERSPECTIVE (J. McCarthy)

Invest Additional Resources to Identify Patient Populations Most Likely to Benefit From Therapy

The idea of paying for value when it comes to pharmaceuticals is a widely accepted goal. However, there is still no consensus on what this means. Although this concept is still evolving, it relies at its core on biopharmaceutical companies to demonstrate that the products we develop provide meaningful benefits to patient populations, coupled with a reimbursem*nt system that lowers barriers to high-value products. Biopharmaceutical companies should do their part to invest more resources to ensure that the right product is available to the right patient at the right time. Cancer care is in the midst of an incredible transformation. Many cancers, previously intractable, now can be treated with targeted therapies that greatly boost the chances of better outcomes for patients. Some patients experience long-term benefits from immunotherapies. However, we still know too little about which types of patients are likely to respond best to a particular therapy. The pharmaceutical industry should invest additional resources in studies of a new drug after it has been approved to better understand its utility, whether to identify use at earlier stage of the disease, in a different tumor type, or for an even narrower patient population to avoid patient exposure when the risks are more likely to outweigh the benefits. For our health care system to truly pay for value, stakeholders also must be willing to develop creative reimbursem*nt mechanisms that incentivize high-value care. Payment reform demonstrations are underway across the health care

KEY POINTS • Not only are drug prices rising, but patients also face high cost sharing forcing patients to shoulder a greater portion of costs. • Current reimbursem*nt models provide little incentive to contain costs. • A solution to the problem of financial toxicity must be collaborative and multifaceted. • As a starting point to the discussion, the authors have provided four ideas from each of their stakeholder perspectives as options available to reduce cancer treatment–related financial toxicity, including: highvalue reimbursem*nt models; removing coverage mandates for all FDA-approved cancer therapy; ensuring oncologists consider price in treatment decision making; and encouraging patient engagement in decision making to assure treatment truly matches patients’ values and preferences. 36 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

sector to explore better ways to pay for inpatient and outpatient care. To date, however, little has been done in the prescription drug space. Biopharmaceutical companies and payers—public and private—should collaborate to explore new, high-value reimbursem*nt methods. These methods could include the following: • Outcomes-based contracts. Biopharmaceutical companies and private payers have experimented recently with contracts in which payment for products is tied to achievement of certain therapeutic goals (e.g., avoiding increased hospitalizations or increasing progression-free survival). These two sectors should work together to advocate removal of regulatory and legal barriers. Doing so would allow robust use of these promising tools to expressly tie payment to value. • Value-based insurance design. Another nascent concept is value-based insurance design. Although this idea has been used in the context of outpatient and inpatient services, it has not been applied broadly to drugs. For example, an insurer could dramatically reduce cost sharing for high-value drugs or lower cost sharing after a patient experiences disease failure with a lower-cost medication.6 These types of arrangements are still in their infancy but could help incentivize patients and providers to use and prescribe high-value products. Transformation of our health care system to one that pays on the basis of value, not volume, will require coordination and cooperation across the sector. Biopharmaceutical companies should do their part to help demonstrate the value of our products to patients, payers, and providers.

THE PAYER’S PERSPECTIVE (L. Newcomer)

Remove Coverage Mandates From State and Federal Insurance Law

Insurance regulation forces payers to pay for any U.S. Food and Drug Administration (FDA)–approved cancer therapy in 42 states; Medicare has a similar provision. Such mandatory coverage eliminates any consideration of value. A therapy with mandatory coverage could be curative or could simply add one additional day of life, but the price cannot be negotiated if that therapy has an FDA-approved indication. The laws were well intended originally. As expensive therapies emerged, legislators were concerned that insurers would simply refuse to pay. The unintended consequence of coverage mandates becomes apparent when multiple therapies are available; payers cannot make decisions on the basis of the value of therapy and substitute one therapy for another when it is clinically appropriate. Removal of this legislated requirement would open the marketplace, and pharmaceutical manufacturers would compete on price and outcomes. Payers would compete in the marketplace by offering the best values for therapy within a competitive premium. This competition requires that a transparent and understandable set of criteria for determination of value, partial value, or no value is presented. The market could function normally. The lung cancer therapy necitumumab is an excellent example of why mandates force prices beyond reason.7 This

FINANCIAL TOXICITY OF CANCER CARE

drug was added to cisplatin and gemcitabine and compared with cisplatin and gemcitabine alone in patients with stage IV squamous cell lung cancer. Three percent of the patients who received necitumumab suffered a cardiac arrest. The difference in median progression-free survival was 0.2 months (5.7 vs. 5.5 months), but the overall survival favored the necitumumab group by 1.6 months (11.5 vs. 9.9 months). These results are so meager that the National Comprehensive Cancer Network assigned a category-3 recommendation to the drug—an endorsem*nt that most insurers, including Medicare, do not cover. However, the mandates require coverage at any price, because the drug has an FDA approval. The manufacturer priced this drug at $11,430 per month. The competing regimen, cisplatin and gemcitabine, cost less than $1,000 per month. It is difficult to believe that anyone except the manufacturer would consider this to be a value, but it does not matter. The law mandates coverage; therefore, price is not negotiable. Necitumumab is not an isolated example. Salas-Vega et al8 reviewed 62 new cancer molecules approved between 2003 and 2013 in the United States and Europe. The review showed no evidence to suggest that 16 of those drugs (30%) increased overall survival compared with best alternative treatments. If manufacturers knew that these products would not be reimbursed in the market, they would focus their attention on different molecules that offer better results. Other mandates are emerging. Several states are now considering laws to prohibit step therapy for cancer. Step therapy requires treatment with a preferred regimen before the patient is eligible for a second therapy. This strategy is useful for drugs that have similar clinical response rates, because a payer can obtain competitive bids and then give preference to the lowest-cost regimen. There have been so many drug discoveries in the past decade that many cancer types now have multiple effective agents. Step therapy allows patients to obtain treatment at a lower cost. Prohibition of step therapy eliminates competition, raises costs, and hurts everyone except the pharmaceutical manufacturer. A free market determines prices on the basis of merit, and mandates prevent free market actions. Removal of mandates presents a win-win proposition for patients and payers.

THE ONCOLOGIST’S PERSPECTIVE (L. Saltz)

Know the Price

Any one change that doctors could make would only be a first step toward the ultimate goal of provision of lower-cost, higher-value medicines for our patients. To me, that first step would be physician acceptance, practice, and promotion of transparency in price. In simple terms, that means knowing the prices of the drugs prescribed, considering those prices as one of the many factors in decision making, and discussing the prices of the prescribed drugs as openly as we discuss other risks, toxicities, and benefits. I make a sharp distinction here between the words price and value. A true, constructive consideration of the value

of a particular medicine cannot realistically occur unless we know the amount of money that we are being asked to pay for it. That is the definition of price, or cost, that I refer to for this discussion: the amount of money that will be paid for the drug. I am not, for this exercise, getting into who is paying for it, what other costs are or are not involved, what alternatives are available, or any other of a number of important, arguably relevant considerations, and I am not making a judgment about whether the price is too high, too low, or just right. I am simply saying that we must stop putting our heads in the sand and pretending that we do not need to know, think about, or talk freely about, what the price is. Consider for a moment the inherent ambiguity in the word value. Value can be used as a noun or a verb. The way to define it in a constructive discussion aimed at definitions of high- and low-value care is as a noun; in that respect, the value of a drug would be defined by a ratio of objective positives and negatives of that drug, with price as one of the negatives. Note that price and value move in opposite directions. For any drug with a fixed degree of benefits and adverse effects, the higher the price, the lower the value, and the lower the price, the higher the value. So, we cannot begin to meaningfully determine the value until we know the price, just as we could not meaningfully determine value without knowing the other positive and negative aspects of the drug. Price is not the defining factor in value, but it is one of the components without which the value cannot be defined. Too often, our consideration of price can be distracted by the shift of the discussion to value and its definition as a verb; we value a response, a defined amount of extended life, or relief from a symptom. The verb definition of value necessarily takes us into subjective, as opposed to objective, criteria, and the very nature of these resist correlation with a price. In fact, such a focus prevents delineation of value and distracts us from a meaningful and constructive discussion of what is high-value care and, just as important, what is not. Doctors do not have the ability to unilaterally lower the prices of drugs. Doctors do have the ability to be aware of the prices of the drugs, tests, treatments, and recommendations we offer, both directly in terms of out-of-pocket expenses to our patients and more indirectly to society as a whole. Some have argued that it is only the immediate out-of-pocket expense of the individual that should be considered by doctors and that societal costs are not relevant to a patient-physician relationship. I respectfully disagree. Societal costs ultimately are distributed across the population, and all insured patients eventually bear these costs in terms of increased insurance premiums. The price of insurance and the percentage of paychecks that go toward health care costs have been increasing at a substantially more rapid rate than the increases in average worker wages or inflation. The term financial toxicity has gained increasing traction in our understanding of what these costs are doing to our patients on a regular basis. Even when the initial out-of-pocket asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 37

ZAFAR ET AL

expense may appear small, one can realistically expect that these costs will appear in the insurance premiums for all in the years to come. Even if one were to take the position that the physician focus should be on immediate out-of-pocket expenses of an individual patient (a short-sighted view, as I outlined in the previous paragraph), this would imply a responsibility to understand the coverage and actual out-of-pocket exposure of each patient, as well as, arguably, the ability of each patient to manage those expenses. This often may be beyond provider abilities. It is quite reasonable, however, to assume that vulnerability a patient may have toward the potential cost of even a small part of that therapy increases with more expensive therapy. Physicians see a decrease in simple copayments with fixed nominal costs and an increase in coinsurance charges, whereby the patient will pay a fixed percentage of the price of the drug. In this context, the more expensive drugs create greater out-of-pocket expenses at the same time that they contribute to the aggregate cost of health care and the necessarily compensatory increases in insurance premiums going forward. Physicians frequently talk to patients about intimate and personal details of their lives. Physicians routinely ask about bowel function, bladder function, sexual function, anxiety, depression, alcohol and illicit drug use, and other intimate and personal details that would be far outside normal social discourse. Within this context, there is a startling inconsistency with any conversational taboo regarding costs. Yet, discussion of the prices of treatments has been a taboo in our doctor-patient relationships, and that requires re-evaluation. Bringing the realistic costs to bear in discussions would make the most involved members of society appropriately informed of the magnitude of the challenges faced in paying for drugs at the current prices. It would also facilitate rational discussions of efforts to use more cost-efficient regimens, use less expensive alternatives, or perhaps forego extremely expensive and toxic options that have little chance to provide meaningful benefit. There are very few things in life that people buy without an awareness of the purchase price. Such an awareness helps people make informed decisions about what goods or services they do or do not wish to purchase and can encourage people to make informed decisions about the consideration of alternatives. From an academic perspective, discussion of price is warranted both in clinical trial design and in publications. When a trial is designed that increases the length of treatment or increases the dose of a drug to higher than the standard dose, physicians must know and consider what the costs of those changes will be. When a report is published about a regimen for which the prices of the drugs are known, those prices constitute a nontrivial toxicity to which patients will be exposed. The purpose of academic paper about therapeutic options, and the purpose of open and complete discussions between patients and providers, is to maximize the awareness of the true risks, benefits, and alternatives of the treatment strategy under consideration. It would be wrong to exclude 38 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

consideration of physical toxicities. It is equally counterproductive to exclude consideration of price, or financial toxicity. The inability to provide full awareness of either of these likely will increase, rather than decrease, the prevalence of and the harm done by these toxicities.

THE PATIENT ADVOCATE’S PERSPECTIVE (S. Fuld Nasso)

Engage in Treatment Planning to Better Reflect Patient Values

Patient engagement in treatment decision making can reduce financial toxicity for patients by ensuring that treatments truly match the needs, values, and preferences of patients. A consideration of all clinically meaningful treatment options and their benefits, risks, and out-of-pocket costs should frame the patient decision-making process. At an individual level, patients can play a role by being active participants in decisions about their care, researching their insurance coverage, initiating discussions about the cost of care with their care team, advocating for coverage of the care that they need, and seeking financial assistance from foundations and company-sponsored assistance programs. Empowered patients and family members know that they must advocate on their own behalf, or on behalf of their loved ones, in all aspects of their care, including financial considerations. Of course, not all patients are prepared and knowledgeable about health insurance, and many patients feel overwhelmed by the amount of information they must process about their diagnosis and treatment options, not to mention the question of how they will pay for their care. Patients need assistance with health insurance literacy; there is evidence that patients do not have a thorough understanding of key insurance constructs, like deductibles, copayments, and coinsurance.9 Ideally, patients and caregivers will raise the topic, ask the questions, and seek assistance from their care team and/or a financial counselor. However, a huge barrier for patients is embarrassment about discussions of financial considerations with their care team. It is essential that providers create a welcoming and open environment for patients to express their concerns. Providers should recognize how difficult it is for patients to raise the topic and should open the door to the conversation by asking a question as simple as “Do you have concerns about the cost of your treatment?” At a practice and policy level, the comprehensive treatment planning process that has been defined by the Institute of Medicine10—a definition arrived at with substantial input from oncologists and patients—should be the standard for doctor-patient communication about cancer care. It is important to note that this planning is not about checking a box that a piece of paper was handed to a patient; it is about truly engaging patients in decision making about their care. This plan should include information related to diagnosis, prognosis, treatment goals, expected response to treatment, treatment benefits and harms, out-of-pocket cost of care, and a plan for meeting psychosocial needs. The care planning process also should include consideration of

FINANCIAL TOXICITY OF CANCER CARE

advance care planning and advance directives and should lead to the development of a survivorship plan after treatment. This cancer care planning process should produce a patient-specific care plan that will guide treatment decisions and facilitate care coordination, including effective symptom management to reduce the burden and cost of adverse effects. An important component of the planning process is a discussion of the out-of-pocket costs to a patient. We know that some patients do not wish to discuss costs; they might worry about the perception that the oncologist has of them, they might want the best treatment regardless of cost, or they might fear that a discussion of cost will result in inferior treatments.11 Yet, most patients do want to have this conversation, even if they are reluctant to bring it up. Research shows that having the discussion, even without a change in treatment, can reduce costs for patients.11 Patients are concerned about their total financial responsibility across the life of their treatment, not just the cost of one aspect of treatment. Obviously, that is difficult for one provider to share, given the multidimensional aspects of treatment. To the degree that it is possible, knowledge about the total costs will help patients plan and understand the entire picture, not just the cost of a specific drug. Although some of the value frameworks, including those by ASCO, have considered the price of a drug, the out-of-pocket cost is what is most important to an individual to make decisions. In most cases, that distinction will require an

understanding of the out-of-pocket maximum. It also is important for patients to understand whether any out-ofnetwork services, which do not contribute to the out-ofpocket maximum, will be required.

CONCLUSION

As a starting point to answer the question (How can we reduce patients’ financial toxicity?), we propose four potential solutions from the perspectives of the pharmaceutical industry, payers, physicians, and patients, which we feel are helpful. Of course, we are not the first to propose solutions to the growing financial burden of cancer treatment. The intent of this exercise was to consider solutions from within our own stakeholder groups rather than to pass the responsibility down the road. As interventions to reduce financial toxicity and improve value are considered, all participants should consider and discuss many long- and short-term interventions. Policy interventions, such as facilitation of value-based contracting or removal of the coverage mandate, all warrant consideration and may be helpful to the long-term process but are unlikely to be realized overnight. In the meantime, shortterm interventions, like price awareness and inclusion of cost in treatment and goals of care discussions, are necessary. The discussion cannot stop here. If anything, this exercise demonstrates that all stakeholders—the pharmaceutical industry, payers, providers, and patients—must continue the discussion to ensure the delivery of high-value care.

References 1. Kaiser Family Foundation. Employer health benefits survey, 2016. http://kff.org/health-costs/report/2016-employer-health-benefitssurvey/. Accessed October 16, 2016. 2. Davidoff AJ, Erten M, Shaffer T, et al. Out-of-pocket health care expenditure burden for Medicare beneficiaries with cancer. Cancer. 2013;119:1257-1265. 3. Cohen RA, Gindi RM, Kirzinger WK. Financial Burden of Medical Care: Early Release of Estimates From the National Health Interview Survey, January–June 2011. https://www.cdc.gov/nchs/data/nhis/health_ insurance/financial_burden_of_medical_care_032012.pdf. Accessed March 26, 2017. 4. Zafar SY, Peppercorn JM, Schrag D, et al. The financial toxicity of cancer treatment: a pilot study assessing out-of-pocket expenses and the insured cancer patient’s experience. Oncologist. 2013;18:381-390. 5. Ramsey S, Blough D, Kirchhoff A, et al. Washington state cancer patients found to be at greater risk for bankruptcy than people without a cancer diagnosis. Health Aff (Millwood). 2013;32:1143-1152. 6. Fendrick AM, Buxbaum J, Westrich K. Supporting consumer access to specialty medications through value-based insurance design, 2014.

http://vbidcenter.org/wp-content/uploads/2014/10/vbid-specialtymedications-npc2014-final-web.pdf. Accessed March 8, 2017. 7. Thatcher N, Hirsch FR, Luft AV, et al; SQUIRE Investigators. Necitumumab plus gemcitabine and cisplatin versus gemcitabine and cisplatin alone as first-line therapy in patients with stage IV squamous non-small-cell lung cancer (SQUIRE): an open-label, randomised, controlled phase 3 trial. Lancet Oncol. 2015;16:763-774. 8. Salas-Vega S, Iliopoulos O, Mossialos E. Assessment of overall survival, quality of life, and safety benefits associated with new cancer medicines. JAMA Oncol. 2017;3:382-390. 9. Zafar SY, Tulsky JA, Abernethy AP. It’s time to have ‘the talk’: cost communication and patient-centered care. Oncology (Williston Park). 2014;28:479-480. 10. Institute of Medicine. Delivering high-quality cancer care: charting a new course for a system in crisis, 2013. http://www.nap.edu/catalog. php?record_id=18359. Accessed January 9, 2014. 11. Zafar SY, Chino F, Ubel PA, et al. The utility of cost discussions between patients with cancer and oncologists. Am J Manag Care. 2015;21:607615.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 39

CAIRO ET AL

Practice Model for Advanced Practice Providers in Oncology Jamie Cairo, DNP, APRN-BC, Mary Ann Muzi, APRN-BC, APCNP, Deanna Ficke, PA-C, Shaunta Ford-Pierce, MSN, NP, FNP-BC, Katrina Goetzke, PA-C, Diane Stumvoll, APGCNP-BC, MSN, Laurie Williams, APNP, MSN, and Federico A. Sanchez, MD OVERVIEW According to ASCO, the number of practicing oncologists has remained stable despite growth demands, leading to an overall shortage in many areas of the country. Nurse practitioners and physician assistants are advanced practice providers (APPs) who can assist in the provision of support and care to patients with cancer, but the role of the APP in the oncology setting has not been well defined. There exists a variety of different practice patterns for APPs who work in oncology, and the lack of role definition and absence of an established practice model are considered leading causes of APP attrition. According to the American Academy of Nurse Practitioners, it has been well demonstrated that, when nurse practitioners are allowed to work to the full scope of their education and preparation, there are notable cost reductions and quality improvements in patient care. The focus of APP education and training is on health promotion, disease prevention, and primary care medical management, but most APPs have limited exposure to management of cancer in patients. With this in mind, Aurora Cancer Care developed a practice model for APPs who work in oncology. The goal of the model is to enhance the quality of care delivered to patients and provide a stimulating work environment that fosters excellent collaborative relationships with oncologist colleagues, supports professional growth, and allows APPs to practice to the full extent of their licensure.

urora Health Care is a not-for-profit, large, integrated health care system that provides cancer services in 10 counties, 16 hospitals, and 22 clinics throughout eastern Wisconsin and northern Illinois. A subcommittee of seven APPs who practice in medical oncology was formed with the goals of defining current practice and identifying areas considered problematic or not well defined. The group used the NP Model of Care, designed by Kutzleb et al,1 to identify five major areas of practice and then developed an action plan and rationale for each of these areas with supporting evidence found in the literature. Cancer care is becoming increasingly complex, and health care systems are looking for ways to meet patient needs and address issues of cost, quality, and access including shortages of practicing oncologists in many areas of the country.1,2 APPs can provide high-quality care to patients that not only is cost effective but also can improve outcomes for patients.3 According to the American Academy of Nurse Practitioners, the best and safest outcomes for patients are produced when health care is provided in coordinated networks that recognize and encourage the unique knowledge base and skills of all practitioners.4 Emergency visits and hospital admissions have negative implications for health care systems, payers, and patients. Patients with cancer frequently experience urgent problems related to their cancer and/or its treatment. The majority

of hospital admissions are due to uncontrolled symptoms, such as shortness of breath, pain, fever, and nausea, and vomiting, and these admissions tend to correlate with longer hospital stays and result in a higher rate of mortality.5,6 Patients with cancer who present to the emergency department often experience long wait times and receive costly and fragmented care that does not always meet their needs. APPs can decrease readmission rates and increase the quality of patient care by meeting regularly with patients to manage their treatment plan and treatment-related issues. A model that can lead to improved and more cost-efficient care uses APPs to see patients in urgent care settings. In this model, the APP would perform assessment, triage, and treatment. This approach benefits not only the patient but also the physician, because it allows the physician to keep on track with scheduled appointments without interruptions.7 There are many documented examples of the successful implementation of this model. For patients with oropharyngeal cancer, a weekly nurse practitioner–led symptom management clinic reduced the rates of both acute care hospitalization and chemotherapy dose deviations in patients receiving intensive chemotherapy and radiation.8 The American Cancer Society reports that more than 15.5 million cancer survivors are alive in the United States today, and that number will grow to more than 20 million

From Aurora Cancer Care, Aurora Health Care, Milwaukee, WI. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Federico A. Sanchez, MD, Aurora Health Care, 750 W. Virginia St., P.O. Box 341880, Milwaukee, WI 53204; email: [emailprotected]. © 2017 American Society of Clinical Oncology

40 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

ADVANCED PRACTICE PROVIDER MODEL IN ONCOLOGY

by 2026. Cancer survivors have unique physical and psychosocial needs that require specialized care, and APPs have the expertise to provide that care.9,10 Given appropriate training, APPs also are able to perform select procedures, including bone marrow biopsies and intrathecal chemotherapy administration, independently.11 Nationally, there is a demand for clinicians with palliative care knowledge. APPs are well suited to integration of palliative care into practice during care of chronically and terminally ill patients.12 Use of a collaborative practice model to integrate APPs into oncology practice has been proposed as an ideal solution to the challenge of complex cancer care across multiple settings.4,13 When APPs work to the full extent of their training and licensure, there are improvements in patient and provider satisfaction as well as an overall positive impact on productivity and revenue.14 As health care reform continues to be a topic of national conversation, APPs must be at the table and willing to take an active role in designing innovative models of care as members and leaders of interprofessional teams.14,15 APPs also must take ownership of teaching and mentoring new nurse practitioners and physician assistants. A mentoring program for APPs that is supported and led by APPs can help those new to the field assimilate to their roles. This relationship can provide benefits for the mentee, the mentor, and the organization.16 A meta-analysis found that job satisfaction and commitment, as well as career outcomes of compensation and promotions, were higher among those who had been mentored.17

ROLE OF THE APP IN ONCOLOGY PRACTICE MODEL

Deliver Direct Care and Coordinate the Interdisciplinary Plan of Care for Patients

It is important to introduce the role of the APP to patients as a vital part of the cancer care team. Best practice involves alternating visits between the APP and physician, which allows the APP to have set appointment schedules and an established role in the active care of patients who

KEY POINTS • APPs enrich the delivery of a comprehensive continuum of care for patients with cancer. • Optimal use of APPs increases opportunities for the physician to focus on appropriately complex and more highly reimbursed patient scenarios, increases appointment opportunities for physicians to see new patients, and decreases wait times for patients. APP visits increase billable services and lead to shorter wait times for patients, which leads to improved clinic workflow. • Optimal APP practice leads to a higher levels of job satisfaction, allows for professional growth and development, and decreases APP attrition. • A collaborative practice of APPs with physician colleagues leads to best practice value-based care.

receive treatment. The following are among the many actions the APP will perform: • Formulation of diagnosis and treatment plan in collaboration with the oncologist • Management of chemotherapy in collaboration with the oncologist • Management of symptoms • Survivorship care • Palliative care • Psychosocial intervention • Procedures (e.g., bone marrow biopsies, intrathecal chemotherapy) • Patient education

Serve As a Consultant to Improve Care According to Expertise in Area of Specialization

Examples of the areas in oncology in which APPs can develop an area of expertise that would help an established oncology practice include the following: • Establish a survivorship clinic: Survivorship visits allow for dedicated time to thoroughly discuss post-treatment concerns and guidelines for wellness promotion. Follow-up visits with the APP also offer opportunities to monitor chronic post-treatment side effects/signs of recurrence. • Coordinated hospital consults and daily inpatient rounds: Consultations and rounds by APPs are convenient for physicians and also improves coordination and communication, especially with discharge planning. • Management of oral chemotherapy: Oral chemotherapy use is increasing. Patients need the same level of monitoring, adherence tracking, and symptom management as those who receive other forms of chemotherapy. • Specialized visits to focus on palliative and end-of-life care: APPs can provide palliative care visits that focus on goals of care, symptom management, patient and family education and counseling, coordination, and continuity of care.

Identify Learning Needs of Various Populations and Contribute to the Development of Educational Programs and Resources

APPs can provide and coordinate educational resources as outlined below: • Staff education: APPs can be used as key advisors when educational opportunities are developed for caregivers. APPs hold graduate degrees; on the basis of their experiences and areas of expertise, they are in key positions to provide support and education to other providers within the organization and to serve as role models and provide the leadership needed to implement and expand evidence-based practice. • Mentoring and training of APPs new to practice and/or oncology: APPs can lead the mentoring and training of new colleagues. • Leading research: Nurse practitioners and physician assistants have graduate degrees and are well prepared to not only participate in clinical research but also design and lead such activities. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 41

CAIRO ET AL

• Community speaking: APPs amass a wealth of knowledge and clinical expertise and are an excellent resource for educating the community at public events. They also spread knowledge by participating in professional associations and speaking/presenting at nursing or other advanced practice conferences and meetings. • Working with APP colleagues in primary care and other service lines: Serve as a resource to the organization, providing expertise and support to the existing and emerging practices of other service line advance practice providers.

Evaluate the Impact of Changes in Clinical Practice and Formulate Recommendations About Appropriateness and Cost Effectiveness

As a crucial part of an oncology team, APPs can play an important role in the creation and maintenance of workflow processes that make any oncology practice successful. Examples of roles are the following: • Serve as participant and leader on quality improvement committees • Serve as consultant in design and implementation of new clinical programs • Become involved in leadership on the national level with professional organizations and accreditation programs • Publish articles and make presentations at regional, national, and international conferences

Identify and Build Collaborative Relationships With Physician Care Teams

To guarantee success in the process, it was felt that the APPs and the physicians needed to create outlined relationships. Examples of some relationships are as follows: • The physician serves as a resource for the APP as part of collaborative relationship. A collaborative practice implies an effective working relationship, in which the APP and physician colleague(s) communicate with one another to provide best practice patient care. • Physicians support orientation and training. The oncologist functions as an expert resource and support for the APP. Physician-supported orientation can nurture the collaborative relationship and affect the confidence and autonomy of the APP.

• Physicians market collaborative efforts. Physicians can introduce APPs to referring physicians. The networking effect can help facilitate awareness of the APP as a member of the health care team. • APPs attend local tumor board/multidisciplinary conferences to represent their practice and meet referring surgeons and primary care providers. Attending multidisciplinary conferences and, when possible participating in clinics, can affect APP training and highlight APP knowledge and skills. Participation also offers opportunities for recognition of the APP as a member of the oncology health care team to other health care specialty providers.

FUTURE IMPLEMENTATION PLANS

After the model was formalized by the subcommittee, it was presented to the larger group of APPs who work in medical oncology for discussion and revision. The seasoned and established APPs were encouraged to take the model back to their individual practice sites and evaluate how their current practices fit within the model. There were practicerelated issues in some locations, and the new practice model has served as a guide to redefine the scope of APP practice in those areas. In addition, the model has been used successfully as an orientation resource and as a guide for APPs who are new to a practice. Before development of the model, new caregivers often had questions about their defined roles in the clinics. The practice model has helped answer many of those questions and has led to smoother transitions to new practices. It also has guided both new and established APPs to enact performance goals and clarify the scope of professional development. The model also has benefited physician colleagues. Although many physicians in medical oncology had long worked with APPs and understood the APP scope and practice, others had not and had questions and concerns when a new APP was introduced to their practice areas. This practice model has served as a guide for them and for nursing and clinic supervisors, and it has opened the door for meaningful discussions about collaboration. The model was presented and accepted by the cancer care executive team and now serves as a document to define the role of the APP and the scope of APP practice in the care of patients at the medical oncology practice at Aurora Health Care.

References 1. Kutzleb J, Rigolosi R, Fruhschien A, et al. Nurse practitioner care model: Meeting the health care challenges with a collaborative team. Nurs Econ. 2015;33:297-304, quiz 305.

4. Jensen P, Counts M; American Association of Nurse Practitioners, et al. AANP comments on the IOM report the future of nursing: leading change, advancing health. https://www.aanp.org/images/documents/practice/ AANPIOMResponse92Date8_4_11.pdf. Accessed March 29, 2016.

2. Vose J; American Society of Clinical Oncology. The State of Cancer Care in America, 2016. http://www.asco.org/research-progress/reportsstudies/cancer-care-america-2016#/message-ascos-president. Accessed March 29, 2016.

5. Numico G, Cristofano A, Mozzicafreddo A, et al. Hospital admission of cancer patients: avoidable practice or necessary care? PLoS One. 2015;10:e0120827.

3. Vogel WH. Oncology advanced practitioners bring advanced community oncology care. Am Soc Clin Oncol Educ Book. 2016;35: e97-e100.

6. Sadik M, Ozlem K, Huseyin M, et al. Attributes of cancer patients admitted to the emergency department in one year. World J Emerg Med. 2014;5:85-90.

42 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

ADVANCED PRACTICE PROVIDER MODEL IN ONCOLOGY

7. Shulman L. Efficient and effective models for integrating advanced practice professionals into oncology practice. Am Soc Clin Oncol Educ Book. 2013:33;e377-e379.

12. Dahlin C, Coyne PJ, Cassel JB. The advanced practice registered nurses palliative care externship: a model for primary palliative care education. J Palliat Med. 2016;19:753-759.

8. Mason H, DeRubeis MB, Foster JC, et al. Outcomes evaluation of a weekly nurse practitioner-managed symptom management clinic for patients with head and neck cancer treated with chemoradiotherapy. Oncol Nurs Forum. 2013;40:581-586.

13. Kurtin SE, Peterson M, Goforth P, et al. The advanced practitioner and collaborative practice in oncology. J Adv Pract Oncol. 2015;6:515-527.

9. Corcoran S, Dunne M, McCabe MS. The role of advanced practice nurses in cancer survivorship care. Semin Oncol Nurs. 2015;31: 338-347. 10. Simon S; American Cancer Society. ACS Report: Number of US Cancer Survivors Expected to Exceed 20 Million by 2026. https://www.cancer. org/latest-news/report-number-of-cancer-survivors-continues-togrow.html. Accessed February 11, 2017. 11. Jackson K, Guinigundo A, Waterhouse D. Bone marrow aspiration and biopsy: a guideline for procedural training and competency assessment. J Adv Pract Oncol. 2012;3:260-265.

14. Hain D, Fleck L. Barriers to NP Practice That Impact Healthcare Redesign. http://www.nursingworld.org/MainMenuCategories/ANAMarketplace/ ANAPeriodicals/OJIN/TableofContents/Vol-19-2014/No2-May-2014/ Barriers-to-NP-Practice.html. Accessed March 10, 2016. 15. Consensus Group Institute of Medicine. The Future of Nursing: Leading Change, Advancing Health. http://books.nap.edu/openbook. php?record_id=12956&page=R1. Accessed March 2, 2016. 16. Harrington S. Mentoring new nurse practitioners to accelerate their development as primary care providers: a literature review. J Am Acad Nurse Pract. 2011;23:168-174. 17. Allen TD, Eby LT, Poteet ML, et al. Career benefits associated with mentoring for protégeé: a meta-analysis. J Appl Psychol. 2004;89:127-136.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 43

BREAST CANCER

MANAGEMENT OF BREAST CANCER BRAIN METASTASES

Breast Cancer in the Central Nervous System: Multidisciplinary Considerations and Management Nancy U. Lin, MD, Laurie E. Gaspar, MD, FASTRO, FACR, MBA, and Riccardo Soffietti, MD OVERVIEW Breast cancer is the second most common primary tumor associated with central nervous system (CNS) metastases. Patients with metastatic HER2-positive or triple-negative (estrogen receptor (ER)–negative, progesterone receptor (PR)–negative, HER2-negative) breast cancer are at the highest risk of developing parenchymal brain metastases. Leptomeningeal disease is less frequent but is distributed across breast cancer subtypes, including lobular breast cancer. Initial treatment strategies can include surgery, radiation, intravenous or intrathecal chemotherapy, and/or targeted approaches. In this article, we review the epidemiology of breast cancer brain metastases, differences in clinical behavior and natural history by tumor subtype, and important considerations in the multidisciplinary treatment of these patients. We will highlight new findings that impact current standards of care, clinical controversies, and notable investigational approaches in clinical testing.

pread of cancer to the CNS, either in the form of parenchymal brain metastases or leptomeningeal disease continues to confer a poor prognosis and high symptom burden in many patients, though survival does appear to be improving over time in some patient subsets. Although the area of breast cancer brain metastases has historically been a relatively understudied area, several seminal clinical trials have altered the standard of care over the past few years, and other smaller studies have provided a variety of new treatment options for patients. Furthermore, multiple innovative investigational strategies are being tested in the clinic. For these reasons, more than ever, the management of brain metastases requires a thoughtful, multidisciplinary approach that integrates the anatomic and symptomatic burden of disease, the status of a patient’s extracranial disease and systemic therapy needs, prior therapies, and life expectancy.

RISK FACTORS FOR THE DEVELOPMENT OF CNS METASTASES

Breast cancer is the second most common cancer associated with brain metastases in the United States following lung cancer.1 As patients with advanced breast cancer live longer, the incidence of brain metastases appears to be increasing. In a subset of women, progression in the CNS has become a major life-limiting problem. The incidence of brain metastases in patients presenting with stage I/II invasive breast cancer, according to subtype, is

as follows2: luminal A, 0.1%; luminal B, 3.3%; luminal-HER2, 3.2%; HER2, 3.7%; and triple-negative, 7.4%. Although these numbers are somewhat low, of those patients with distant metastases, approximately 30% to 50% will eventually develop brain metastases.2-5 Factors associated with an increased likelihood of brain metastases include young age, lymph node positivity, higher grade, hormone receptor negativity and HER2 positivity, and time from diagnosis to first metastasis.6 The time from the initial diagnosis of primary breast cancer to the development of brain metastases is also influenced by subtype, with the shortest interval observed for patients with triple-negative disease (27 months), and the longest interval observed for those with ER-positive, HER2-positive disease (54 months).7

PROGNOSTIC AND PREDICTIVE FACTORS OF SURVIVAL

The predictive factors and the prognosis of patients with brain metastases are now considered to be disease specific (Table 1). One tool that can be used is the Disease-Specific Graded Prognostic Assessment (DS-GPA).8 The prognostic factors within the breast-specific GPA are presented in Table 2. Note that the time from primary diagnosis to brain metastases was not an independent significant prognostic factor in the breast GPA and is therefore not a part of the index.7 The DS-GPA was based on the observed outcome of patients referred for a radiation therapy opinion, and

From the Breast Oncology Center, Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA; Department of Radiation Oncology, University of Colorado School of Medicine, Aurora, CO; Department of Neuro-Oncology, University of Turin and City of Health and Science Hospital, Turin, Italy. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Nancy U. Lin, MD, Department of Medical Oncology, Dana-Farber Cancer Institute, 450 Brookline Ave., Boston, MA 02215; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 45

LIN, GASPAR, AND SOFFIETTI

TABLE 1. Median Survival Time (Months) by DS-GPA8 Survival Median (Months)

GPA 1 (0–1)

GPA 2 (1.5–2.0)

GPA 3 (2.5–3.0)

GPA 4 (3.5–4)

NSCLC

3.02

5.49

9.43

14.78

SCLC

4.9

2.79

4.90

7.67

17.05

Melanoma

6.74

3.38

4.70

8.77

13.23

RCC

9.63

3.27

7.29

11.27

14.77

Breast

13.8

3.35

7.70

15.07

25.3

Abbreviations: DS-GPA, Disease-Specific Graded Prognostic Assessment; NSCLC, non–small cell lung cancer; SCLC, small cell lung cancer; RCC, renal cell carcinoma.

patients underwent whole-brain radiotherapy (WBRT), stereotactic radiosurgery (SRS), surgery, or a combination of these treatments. Only 6% of patients had a GPA score of 1, with the remaining patients fairly equally distributed between GPA scores of 2, 3, and 4. This retrospective analysis cannot be used to predict the outcomes according to different treatments. Its utility lies in its use as a stratification tool for clinical trials and the comparison of results between clinical trials. It can also aid the oncologist in determining whether the patient might be best served by hospice. National Comprehensive Cancer Network guidelines as of January 2017 state that CNS imaging of patients with asymptomatic breast cancer is not indicated, based on lack of available evidence of benefit. However, prospective studies to evaluate the risks and benefits of CNS imaging are scant. Given the incidence and relatively short interval to presen-

KEY POINTS • Risk factors for the development of brain metastases in breast cancer include tumor subtype (HER2-positive, triple-negative, estrogen receptor–negative), young age, higher disease grade, and shorter disease-free interval. • Level I evidence, generated in mixed populations of patients with a variety of solid tumors, supports the role of surgical resection in patients with a single brain metastasis who have good performance status and controlled extracranial disease. • For patients presenting with a limited number of brain metastases, the addition of WBRT to SRS improves intracranial control but does not improve survival and can be associated with neurocognitive deficits. Thus, SRS only is a reasonable approach in such patients. A caveat to these data is that they were generated in all-comers with solid tumors, and patients with breast cancer made up only a minority of patients enrolled. • The use of memantine during and after WBRT is associated with delayed time to cognitive decline, reductions in the rates of decline in memory, executive function, and processing speed. • To date, no systemic therapies have gained regulatory approval for the treatment of breast cancer brain metastases; however, several regimens have demonstrated activity in prospective studies, and multiple new approaches are being tested in ongoing clinical trials. 46 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

tation of brain metastases in patients with triple-negative disease, and the high incidence of CNS metastases in patients with HER2-positive breast cancer, further investigation of this issue is highly warranted.

LOCAL THERAPY

The Role of Surgery in Patients With Brain Metastases

Among local treatment options, surgery has a clear role in some subgroups of patients. Three phase III trials have compared surgical resection followed by WBRT with WBRT alone for a single brain metastasis (Fig. 1).9-11 The first two studies have shown a survival benefit for patients receiving the combined treatment (median survival 10 vs. 4–6 months). In the Patchell study, patients who received surgery displayed a lower rate of local relapse (20% vs. 52%) and longer period of functional independence. The third study, which included more patients with active systemic disease (80% vs. 30%–40%) and a low Karnofsky performance status, did not show any benefit with the addition of surgery to WBRT. Therefore, class I evidence shows that the survival benefit of surgical resection in addition to WBRT is limited to the subgroup of patients with controlled systemic disease and good performance status.12 In properly selected patients with two or three brain metastases, who are in good neurologic condition and have controlled systemic disease, complete surgical resection yields results that are comparable to those obtained in single lesions.13 One caveat to these and much of the local therapy literature is that patients of multiple primary histologies were included in the trials, with a relatively small fraction of patients with breast cancer (typically 10%– 20%), and thus, the recommendations are to some extent extrapolations based on a study population with primarily non–small cell lung cancer. For the majority of patients, surgical resection allows an immediate relief of symptoms of intracranial hypertension, a reduction of focal neurologic deficits and seizures, and a rapid steroid taper. Gross total resection of a brain metastasis can be achieved with lower morbidity using contemporary image-guided systems, such as preoperative functional MRI, intraoperative neuronavigation, and cortical mapping.14 An early postoperative MRI can detect residual tumor in up to 20% of patients, and this is associated with an increased risk of local recurrence.15 The impact of surgical methodology on the complication rate and functional outcome, as well as on local relapse in

MANAGEMENT OF BREAST CANCER BRAIN METASTASES

TABLE 2. Prognostic Factors and Assigned Score in Breast Cancer GPA8 Prognostic Factor

0.5

1.5

2.0

KPS

≤ 50

70–80

90–100

n/a

Subtype

Basal

n/a

LumA

HER2

LumB

Age, years

≥ 60

< 60

n/a

Abbreviations: GPA, Graded Prognostic Assessment; KPS, Karnofksy Performance Status; LumA, luminal A; LumB, luminal B.

patients with a single brain metastasis, has been recently analyzed. Overall, the study suggests that postoperative complication rates are not increased by en bloc resection, as compared with piecemeal resection, for lesions in eloquent brain regions or large tumors.16 Leptomeningeal dissemination can be a complication, especially for patients with posterior fossa metastases undergoing a piecemeal resection (13.8%) compared with en bloc resection (5%–6%).17 Last but not least, surgery is important in providing tissue for molecular analysis to define the molecular profile of the brain metastasis, which can be different from that of the primary cancer. This is critical in the near future for tailoring targeted therapies to the molecular profile of the brain metastases.

WBRT Compared With SRS Following Surgical Resection

Despite the randomized study by Patchell et al,18 in which it was found that patients with a single brain metastases who underwent surgical resection and postoperative WBRT had fewer recurrences of cancer in the brain and were less likely to die of neurologic causes as compared with patients treated with surgical resection alone, there has been controversy regarding the role of WBRT in this setting. This led to the N107C/CEC.3 cooperative group study randomly assigning patients with a resected brain metastasis to receive either WBRT or SRS to the cavity. SRS to unresected brain metastases was allowed in both groups. Patients were stratified between primary lung cancer, radio-resistant histologies, or other histologies. This study was presented at the Plenary Session during the 2016 American Society for Radiation Oncology Annual Meeting but is not yet published.19 Approximately 30% of enrolled patients fell into the other

category, although breast cancer is not separated out otherwise. There was no reported difference in overall survival between the two treatment groups (11–12 months), with no difference seen according to age, extracranial disease status, number of brain metastases, histology, or size of resection cavity. However, there was a small but statistically significant difference in the cognitive deterioration–free survival favoring the SRS arm (2.8 months WBRT arm vs. 3.3 months SRS arm; p < .0001). Only 5.4% of patients in the WBRT arm were free of cognitive deterioration at 6 months as opposed to 22.9% in the SRS arm.

SRS With or Without WBRT

Several randomized studies have examined the outcome of SRS with or without WBRT.20-23 One of these was a randomized controlled trial published in 2006 by Aoyama et al20 (JROSG 99-1), which randomly assigned 132 patients with up to four brain metastases amenable to SRS. The primary endpoint was overall survival, but secondary outcomes included local recurrence, rate of salvage brain treatment, functional preservation, toxic effects, and cause of death. The study was closed earlier than the planned accrual when an interim analysis determined that more than 800 patients would be required to detect a significant difference in the primary endpoint. Breast cancer made up only 7% of enrolled patients, the majority being non–small cell lung cancer. In the SRS-only group, the median survival time and the 1-year actuarial survival rate were not significantly different between the two groups. However, the group receiving SRS and WBRT had a lower intracranial recurrence rate at 1 year (47% vs. 77%; p < .001), and required less frequent salvage treatment as opposed to the SRS-only group.

FIGURE 1. Management Algorithm for the Initial Treatment of Patients With Breast Cancer Brain Metastases

This figure provides a broad overview. For details and discussion of nuances of the recommendations, please refer to the text. Treatment recommendations will also depend on performance status, prior therapies, status of extracranial disease, comorbidities, and life expectancy. In most cases, outside of a clinical trial, surgery and/or radiation therapy will be given as initial therapy, and the systemic therapy will be determined by the status of a patient’s extracranial disease (i.e., continue prior systemic therapy if systemic disease is stable, and switch if systemic disease is progressive).

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 47

LIN, GASPAR, AND SOFFIETTI

Another larger study of 359 patients, of which 12% had breast cancer, randomly assigned patients with up to four brain metastases to receive local therapy (surgery or SRS) with or without WBRT.23 Overall survival was similar between the two treatment arms (p = .89), although the local control (surgery vs. surgery WBRT: 59% to 27%, p < .001; SRS vs. SRS + WBRT: 31% vs. 19%, p = .04) and need for further salvage therapy (51% vs. 16%, p value not reported) were improved in the WBRT arm. Lastly, a 2015 meta-analysis by Sahgal et al24 of these randomized studies found that patients age 50 or younger had a significant survival benefit (p = .04) when SRS alone was used. This analysis found that these results were similar between patients with lung cancer and those with breast cancer, although the authors acknowledged the problems with small sample sizes. The authors concluded that SRS alone is the recommended initial therapy of patients age 50 or younger with one to four brain metastases. The above findings showing that WBRT is not associated with improved survival, when combined with the data regarding the neurocognitive effects of WBRT, have led to many guidelines recommending SRS only for patients with one to four brain metastases (American Association of Neurological Surgeons, unpublished data, 2017).22,25

Quality of Life and Cognitive Dysfunction Following WBRT

Cognitive dysfunction following WBRT represents a topic of increasing importance. Historically, radiation-induced dementia with ataxia and urinary incontinence was described in up to 30% of patients by year 1 who were receiving unconventional, large-size fractions of WBRT (6–8.5 Gy), which are no longer used.26 The picture on CT/MRI was that of a leukoencephalopathy (diffuse hyperintensity of the periventricular white matter on T2-weighted and fluid attenuation inversion recovery images) with associated hydrocephalus, for which placement of a ventriculoperitoneal shunt could be of some clinical value. When using more conventional size fractions (up to 3 or 4 Gy per fraction), the risk is that of mild cognitive dysfunction, consisting mainly in learning and memory impairment with a variable degree of damage of the white matter and cortical atrophy on MRI. In recent years, several randomized trials have shed light on the short- and long-term effects of WBRT on neurocognitive function and quality of life. Aoyama et al compared the neurocognitive function of patients who underwent SRS alone or SRS plus WBRT.27 Similar proportions of patients in both arms (p = .85) achieved a three point or more improvement in their Mini Mental State Examination score shortly after therapy (2–3 months). However, subsequent deterioration of neurocognitive function in long-term survivors (up to 36 months) after WBRT was observed. In a small randomized trial, Chang et al have shown that patients treated with SRS plus WBRT were at greater risk of a decline in learning and memory function at 4 months after treatment compared with those receiving SRS alone.21 A randomized phase III trial (Alliance trial) has compared SRS alone with SRS plus 48 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

WBRT in patients with one to three brain metastases using a primary neurocognitive endpoint, defined as decline from baseline in any seven cognitive tests at three months.28 Neurocognitive decline was significantly more frequent after SRS plus WBRT compared with SRS alone (91.7% vs. 63.5%, p < .001). On individual tests, there was more cognitive deterioration in immediate memory (30.4% vs. 8.8.2%, p = .004), delayed memory (51.1% vs. 19.7%, p < .001), and verbal fluency (18.6% vs. 1.9%, p = .01) in the SRS plusWBRT arm. Finally, a quality-of-life analysis of the EORTC 2295226001 trial has shown over 1 year of follow-up no significant differences in the global health-related quality of life, but patients undergoing adjuvant WBRT instead of observation had lower transient cognitive functioning, physical functioning, and more fatigue.22 Patients with arterial hypertension, diabetes, or other vascular diseases are at a higher risk of developing cognitive dysfunction. The pathogenesis of this radiation damage could consist of an injury of the endothelium of small vessels that leads to an accelerated atherosclerosis and ultimately to a chronic ischemia, resulting in a picture similar to that of the small vessel disease of vascular dementia. For this reason, there is interest in investigating vascular dementia treatments to prevent or reduce radiation-induced cognitive decline. One of these approaches is using memantine in combination with WBRT. Memantine is a noncompetitive, low affinity antagonist of the N-methyl-D-aspartate (NMDA) receptor, which is one of the receptors activated by glutamate, the principal excitatory neurotransmitter. Memantine has the potential to block the excessive NMDA stimulation following ischemia, which ordinarily could lead to excitotoxic damage of the normal brain. In a recently published randomized, double-blind, placebo-controlled phase II trial (RTOG 0614), the use of memantine during and after WBRT resulted in a mild improvement of cognitive function over time, specifically delaying time to cognitive decline and reducing the rates of decline in memory, executive function, and processing speed.29 The use of another neurotransmitter regulator, such as donepezil, has shown only modest improvements in cognitive function in a controlled trial, especially among patients with greater pretreatment impairment.30 Radiation-induced cognitive deficits may result, at least in part, from a radiation injury to the neuronal stem cells in the subgranular zone of the hippocampus.31 These stem cells are responsible for maintaining neurogenesis, which is critical for preserving memory function, especially in terms of encoding new episodic memories. Low-dose irradiation in rodents results in a blockade of hippocampal neurogenesis and damage of the neurogenic microenvironment, leading to significant short-term memory impairment. Thereby, it has been hypothesized that sparing the hippocampus during WBRT (hippocampal-avoidance WBRT, HA-WBRT) could prevent the damage of the neuronal progenitor cells and better preserve memory functions.32 The recent singlearm, phase II RTOG 0933 trial has suggested that the conformal avoidance of the hippocampus during WBRT is

MANAGEMENT OF BREAST CANCER BRAIN METASTASES

associated with some sparing of memory and quality of life; specifically, performance on standardized memory tests declined 7% from baseline to 4 months in patients treated with hippocampal-avoidance WBRT, as compared with 30% in an historical control group.33 Importantly, 4.5% of patients developing intracranial progression had involvement of the hippocampal-avoidance area by metastatic disease. In this regard, building on results of RTOG 0933 and RTOG 0614, NRG-CC001 is a phase III trial evaluating the potential combined neuroprotective effects of hippocampal avoidance in addition to memantine during WBRT for brain metastases.34

Clinical Challenges: Tumor Progression, Radionecrosis, and Pseudoprogression

A critical issue is the distinction between post-treatment effects and true tumor progression in some particular scenarios. Following SRS, changes such as an increase in contrast enhancement, necrosis, edema, and mass effect on MRI are difficult to interpret: in this regard, PET with 18F-fluorodeoxyglucose, amino acids or 18F-fluorodeoxythymidine, MRI perfusion, and magnetic resonance spectroscopy may provide additional information, though are rarely diagnostic.35-38 In general, careful monitoring with MRI, sometimes for many months, is needed. Radiation necrosis is commonly treated with steroids. Hyperbaric oxygen and/or the antiVEGF agent bevacizumab, which may allow stabilization/ normalization of the vascular permeability, can be useful in patients not responding to steroids,39 while surgical resection is needed in some patients. In patients receiving immunotherapy-based treatments, an initial increase in the number and size of metastases can

be followed by radiographic stabilization or regression. This pattern might be related to the mechanism of action of immunotherapy, including immune infiltrates and the time to mount an effective immune response. If immune response– related radiographic changes are suspected, the advice is to not interrupt immunotherapy treatment until a short interval scan is obtained.40,41

SYSTEMIC THERAPY

Evidence for Efficacy of Available Endocrine Therapies and Cytotoxic Chemotherapies

To date, no systemic agents have gained regulatory approval for the treatment of breast cancer brain metastases. Nevertheless, as summarized in Table 3, CNS activity in case series or in small, prospective studies has been reported across a range of cytotoxic drugs. For example, Rivera and colleagues reported their experience in a phase I trial testing the combination of capecitabine and temozolomide in 24 patients with breast cancer brain metastases (14 newly diagnosed; 10 with progressive brain metastases after prior local therapy). The observed CNS objective response rate was 18%, with median time to progression of 12 weeks.42 Given the general lack of activity of temozolomide in breast cancer, it would be reasonable to assume that the majority of the activity in this trial can be attributed to the capecitabine. Data from a small series of seven patients treated at Memorial Sloan Kettering Cancer Center corroborate the observation of capecitabine activity in the CNS.43 Anthracyclines are also associated with CNS responses in breast cancer, with response rates ranging widely from 17% to 62%, in small experiences. Single-agent temozolomide

TABLE 3. Summary of Case Reports, Case Series, and Prospective Studies Testing Cytotoxic Chemotherapy in Patients With Breast Cancer Brain Metastases

Agent

Details of Regimen

Type of Study

No. of Patients With Breast Cancer Treated With Specified Regimen

Capecitabine

Capecitabine + temozolomide

Phase I

18%

Rivera et al42

Capecitabine

Case series

43%

Ekenel at al43

Doxorubicin, cyclophosphamide

Case series

17%

Rosner et al44

Pegylated liposomal doxorubicin

Phase II

62%

Caraglia et al45

Liposomal doxorubicin + cyclophosphamide

Retrospective

41%

Linot et al46

Cisplatin + etoposide

Prospective

38%

Franciosi et al47

Cisplatin + etoposide

Case series

55%

Cocconi et al48

Cisplatin + temozolomide

Phase II

40%

Christodoulou et al49

Irinotecan

Irinotecan + iniparib

Phase II

12%

Anders et al50

Temozolomide

Phase II

Trudeau et al51

Temozolomide

Phase II

Christodoulou et al52

Temozolomide

Phase II

Abrey et al53

Temozolomide

Phase II

Siena et al54

Temozolomide + vinorelbine

Phase II

Iwamoto et al55

Anthracycline

Platinum

CNS ORR in Breast Cancer Subset

Reference

Abbreviations: CNS, central nervous system; ORR, objective response rate.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 49

LIN, GASPAR, AND SOFFIETTI

FIGURE 2. Options for Systemic Treatment of Breast Cancer Brain Metastases

This figure provides a broad overview. For details and discussion of nuances of the recommendations, please refer to the text. Note that very few prospective trials have been conducted to evaluate the role of systemic therapy for breast cancer brain metastases, and, in many cases, the recommendations above are therefore based on case reports or case series, or extrapolation from systemic therapy trials. See text for details. Of note, there have not been randomized trials directly comparing a local (i.e., surgery, radiation therapy) versus systemic approach for the treatment of breast cancer brain metastases. Treatment recommendations will depend on performance status, prior therapies, status of extracranial disease, comorbidities, and life expectancy.

appears to have minimal activity in breast cancer and does not clearly add to other agents when given in combination.51,53 CNS activity for platinum salts has been reported in older case series, including response rates of 38% to 55%, albeit in a patient population less heavily pretreated than a typical patient seen today.47,48 In this context, recent efforts (not limited to the brain metastasis space) to identify predictive markers of platinum benefit are relevant. The TNT trial evaluated the efficacy of taxanes versus platinums in the firstline treatment of metastatic triple-negative breast cancer. Of note, patients with active brain metastases were excluded. The study demonstrated differential activity according to BRCA1/2 germline status, with a substantially higher rate of response in extracranial sites to carboplatin compared with docetaxel in BRCA1/2 carriers.56 This hypothesis has not been formally tested in the CNS; however, anecdotally, Jennifer Ligibel, MD, and Judy Garber, MD, MPH, have observed durable CNS responses (including one in excess of three years) in some BRCA1/2 carriers treated with platinum salts (personal communication, January 2017). Case reports of CNS responses to a variety of endocrine agents, including tamoxifen and aromatase inhibitors, are also present in the literature; however, patients with estrogen receptor–positive tumors typically present with brain metastases late in their disease course, when their disease has become hormone refractory.57-60 In general, if considering off-label use of systemic therapy for the treatment of breast cancer brain metastases, the choice of therapy should be in accordance with a patient’s tumor subtype, prior therapies, performance status, and comorbidities, in keeping with national and international guidelines for management of metastatic breast cancer 50 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

(Fig. 2). As with the practice guidelines for use of cytotoxic chemotherapy in patients with extracranial metastases, sequential single-agent chemotherapy is generally preferable to combination chemotherapy.

Investigational Cytotoxic Approaches

In terms of cytotoxic chemotherapy, the trend for the development of compounds focused on the treatment of brain metastases has been to engineer compounds that more effectively penetrate the blood-brain barrier (Table 4). Two of the compounds furthest in development are etirinotecan pegol (NKTR-102) and ANG1005. NKTR-102 is a long-acting topoisomerase-I inhibitor that prolongs exposure to SN38, the active metabolite of irinotecan. CNS activity in breast cancer has been previously observed in the clinic with the parent compound irinotecan.50 In a mouse model of breast cancer brain metastases (MDA-MB-231Br), NKTR-102 prolonged survival compared with conventional irinotecan.61 A phase III trial for patients with heavily pretreated breast cancer (either without brain metastases or with stable brain metastases on study entry) was recently reported.62 Though a negative study overall, a potential signal was observed in the stable brain metastasis subset and a confirmatory study is currently under way. ANG1005 is a novel taxane derivative that is able to penetrate the blood-brain barrier via the low-density lipoprotein receptor–related protein.63 CNS responses have been observed in early-phase studies, and additional studies are ongoing.64,65

HER2-Targeted Therapies

The development of HER2-targeted therapies has dramatically improved overall outcomes for patients with

MANAGEMENT OF BREAST CANCER BRAIN METASTASES

HER2-positive breast cancer, both in the early and advanced stages. Despite these improvements, up to half of patients with advanced HER2-positive breast cancer will relapse in the CNS. The small molecule tyrosine kinase inhibitor (TKI) lapatinib has been studied as a single agent and in combination with chemotherapy in multiple prospective clinical trials. As a single agent in pretreated patients, its activity is modest at best, with CNS responses observed in only 6% of patients.66 Greater activity has been observed in combination with capecitabine, with CNS response rates ranging from 18% to 38% in pretreated patients, and a CNS response rate of 66% in the newly diagnosed setting.66-70 Responses have been durable in many cases, and overall, the results support the concept of evaluating HER2-targeted TKIs for the treatment of brain metastases. HER2-targeted TKIs in clinical development include neratinib, afatinib, and tucatinib, among others (Table 4). Neratinib is an irreversible inhibitor of EGFR and HER2 currently in late-stage clinical testing (NALA, neratinib plus capecitabine versus lapatinib plus capecitabine in patients with HER2+ metastatic breast cancer who have received two or more prior HER2-directed regimens in the metastatic setting, NCT01808573). However, the ongoing phase III study excludes patients with active brain metastases. The Translational Breast Cancer Research Consortium is conducting a phase II study evaluating neratinib in patients with progressive brain metastases. Results of the monotherapy neratinib cohort have been published, with an observed CNS objective response rate of 8%.71 Results of the neratinib/capecitabine

combination cohort are anticipated in mid- to late 2017. Like neratinib, afatinib inhibits both EGFR and HER. Although it has gained regulatory approval in lung cancer, clinical development in breast cancer has been terminated based on negative results of the LUX-Breast 1 and LUX-Breast 3 randomized trials.72,73 In particular, in the LUX-Breast 3 trial, the combination of vinorelbine and afatinib did not afford any additional benefits to patients with HER2-positive breast cancer brain metastases compared with treatment of provider choice.72 In contrast to either neratinib or afatinib, tucatinib (ONT380; ARRY-380) selectively targets HER2 and has minimal activity against EGFR, leading to a more favorable toxicity profile, with less diarrhea and rash. The active metabolite appears to cross the blood-brain barrier, and improvements in survival have been reported in preclinical models of breast cancer brain metastases. In a phase I study of tucatinib with trastuzumab, CNS responses were observed in 7% of patients; approximately one-third of patients achieved stable disease of 16 weeks or longer.74 The triplet of trastuzumabcapecitabine-tucatinib has been studied in a phase IB study among patients with highly refractory, HER2-positive metastatic breast cancer. Objective responses were observed in 61% of patients, including in 42% of patients who had measurable CNS disease at baseline.75 The approach is now being tested in an ongoing randomized trial that includes patients with and without brain metastases. Historically, monoclonal antibodies such as trastuzumab, pertuzumab, or trastuzumab-emtansine were thought to be too large and bulky to penetrate the blood-brain barrier.

TABLE 4. Ongoing Trials of Systemic Therapy for Breast Cancer Brain Metastases Regimen

Phase

Patients With LMD Included?

Breast Cancer Subtypes

ClinicalTrials.gov ID

Neratinib + capecitabine

Yes, if also measurable parenchymal metastasis

HER2+

NCT01494662

Trastuzumab/capecitabine +/− tucatinib (ONT-380)

HER2+

NCT02614794

“High-dose” trastuzumab + pertuzumab

HER2+

NCT02536339

Intrathecal trastuzumab + pertuzumab

Not specified

HER2+

NCT02598427

“High-dose” lapatinib + capecitabine

Yes

HER2+

NCT02650752

Trastuzumab + vinorelbine + everolimus

Diffuse LMD excluded

HER2+

NCT01305941

Abemaciclib

Yes, separate cohort

ER+

NCT02308020

Palbociclib

HER2+ or TNBC

NCT02774681

Palbociclib

Any, but requires evidence of specific pathway alterations in brain metastasis tissue

NCT02896335

Cabozantinib

Yes, if also measurable parenchymal metastasis

ER+ or HER2+

NCT02260531

Pembrolizumab

Yes, separate cohort

Any

NCT02886585

Durvalumab

Any

NCT02669914

Etirinotecan pegol

Any

NCT02312622

Etirinotecan pegol

III

Any

NCT02915744

Cabazitaxel

Any

NCT02166658

Abbreviations: LMD, leptomeningeal disease; ER, estrogen receptor; TNBC, triple-negative breast cancer.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 51

LIN, GASPAR, AND SOFFIETTI

However, studies utilizing 89Zr-labeled trastuzumab as a PET tracer have suggested there is some penetration through a disrupted blood-tumor barrier.76 Penetration across the blood-tumor barrier is supported by emerging case reports and case series of the CNS activity of trastuzumab-emtansine (TDM1), with response rates qualitatively similar to that reported against extracranial disease.77-79 A prospective U.S. study is being planned. Another approach under active investigation in the ongoing PATRICIA study is the use of highdose trastuzumab (6 mg/kg IV weekly) in combination with standard pertuzumab every 3 weeks to drive up concentrations of trastuzumab in brain metastases. Accrual to the study is ongoing.

Clinical Challenges: Is There a Role for Chemoprevention of Brain Metastases?

There is great interest in preventing the emergence of brain metastases (primary prevention) and prolonging the time to subsequent CNS progression in patients who receive initial local therapy (secondary prevention). A common clinical question is whether systemic therapy should be modified following SRS to include a “CNS-active” regimen. At present, there is no direct evidence to support this approach, though the existing data have been sparse and do not perfectly address this question. In the EMILIA trial, a subset analysis was performed among patients who entered the study with stable brain metastases to determine whether the benefit of trastuzumab-emtansine seen in the overall study (compared with lapatinib-capecitabine) held up in the brain metastasis subset.80 Patients in the brain metastasis subset appeared to derive similar relative benefits in terms of overall survival prolongation with trastuzumab-emtansine, and there was no obvious signal in favor of lapatinib-capecitabine with respect to CNS progression. In addition, there were no obvious differences in the incidence of new CNS metastases among patients who entered the study without brain metastases at baseline (2% with trastuzumab-emtansine; 0.7% with lapatinib-capecitabine; p = not significant). The analysis had several limitations including that CNS scans were not mandated per protocol and emerging reports supporting CNS activity of trastuzumab-emtansine, making this potentially a comparison between two “CNS-active” regimens. In general, the consensus approach for treatment of such patients is to continue the prior systemic therapy after SRS, if the systemic disease remains well controlled, but this is an area ripe for clinical trials.81

Other Targeted Approaches Under Clinical Investigation

A number of other targets are being explored in the treatment of breast cancer brain metastases, including CDK4/6 inhibitors, PARP inhibitors, and immunomodulatory therapies (Table 4). When combined with endocrine therapy, the CDK4/6 inhibitors palbociclib and ribociclib prolong progression-free survival compared with endocrine therapy alone.82-84 Among 52 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

the CDK4/6 inhibitors, abemaciclib appears to have the best CNS penetration in preclinical models and has been demonstrated to reach therapeutic levels in human “window of opportunity” studies when given prior to planned resection.85,86 Phase II studies of both palbociclib and abemaciclib for brain metastases are ongoing. Given the frequency of brain metastases in patients with triple-negative breast cancer, there is interest in developing novel targeted approaches in this patient population. PARP inhibitors have clear activity against extracranial metastases in BRCA1/2 carriers; however, single-agent activity in sporadic triple-negative breast cancer has been disappointing.87 Three large randomized phase III trials comparing PARP inhibitors with standard chemotherapy in BRCA1/2 carriers with metastatic breast cancer are ongoing, but all three studies exclude patients with active brain metastases. There is also ongoing interest in combination approaches to sensitize BRCA1/2 wild-type breast cancer to PARP inhibitors, including combinations with platinum salts. The U.S. cooperative groups are collaborating on a planned randomized study (S1416) that will examine cisplatin with or without the PARP inhibitor veliparib (ABT-888). Given that veliparib crosses the blood-brain barrier in preclinical models, a CNS-specific cohort will be concurrently enrolled, with a primary endpoint of progression-free survival.88 There are accumulating preclinical and clinical evidence suggesting that the immune system is critical for disease outcome in breast cancer, particularly in the triple-negative and HER2-positive subtypes.89,90 Moreover, PD-L1 expression appears to be common in breast cancer brain metastases.91 Data in patients with melanoma and lung cancer support potential efficacy of immune checkpoint inhibitors in the CNS.92 Beyond CTLA-4 and PD-1/PD-L1 inhibitors, there is a wealth of novel immunomodulatory compounds, including STING and GITR agonists, and inhibitors of IDO, TIM3, and LAG3. Unfortunately, the vast majority of ongoing trials of immunotherapy in breast cancer specifically exclude patients with active brain metastases, though there are some studies in this population that have recently opened or are in development (Table 4).

MANAGEMENT OF LEPTOMENINGEAL DISEASE

Estimates of the incidence of leptomeningeal metastases vary widely, ranging from 2% to 40%, either alone or associated with parenchymal brain metastases. In a case series of patients with leptomeningeal disease (1998 to 2013) from Memorial Sloan Kettering Cancer Center, both HER2-positive (26% of cases) and triple-negative (25% of cases) breast cancer subtypes were overrepresented, suggesting they are associated with a propensity toward dissemination in the leptomeninges.93 Invasive lobular histology also appears to be associated with leptomeningeal spread.94 The prognosis is poor with median survival of 3.5 to 6 months and 20% survival rate at 1 year.93,95 Since patients can experience very poor survival, it is critical to consider prognostic factors early in weighing management options, including consideration

MANAGEMENT OF BREAST CANCER BRAIN METASTASES

of a more palliative/hospice-oriented course, as appropriate. Favorable prognostic factors include HER2-positive subtype, preserved performance status, and CNS-only involvement. Unfavorable prognostic factors include poor performance status, progressive/treatment-refractory extracranial disease, and major neurological deficits. The most typical management approach is radiation to sites of bulk disease followed by consideration of intra– cerebrospinal fluid (CSF) and/or systemic therapy. Radiation-based approaches, including WBRT, have the potential to provide rapid relief of symptoms, and should be strongly considered, particularly for patients presenting with neurological deficits. Intra-CSF chemotherapy has a role for palliation of neurologic symptoms and should be considered for patients with a large tumor cell load in the CSF.96 Methotrexate, liposomal cytarabine, and thiotepa are the drugs of choice. At the time of the placement of an Ommaya catheter, and prior to injecting drugs into the CSF, flow studies are recommended to rule out the existence of subarachnoid blocks, as these could preclude optimal distribution of drug and increase the risk of leukoencephalopathy.97-100 Systemic chemotherapy has been used off-label to treat patients with leptomeningeal disease based on observed efficacy in case reports and small case series. Regimens with reported efficacy (with caveats given the very limited data) include tamoxifen, aromatase inhibitors, high-dose intravenous methotrexate, capecitabine, lapatinib/capecitabine, and platinum salts. From an investigational standpoint, leptomeningeal disease has frequently been excluded from clinical trials. However, a number of ongoing trials are exploring new therapeutic options (Table 4). Of note, a phase I/II study of intrathecal trastuzumab has recently completed accrual. In this study, trastuzumab was reconstituted in preservative-free sterile

water, USP or preservative-free 0.9% sodium chloride, with an induction phase of more frequent administration, followed by tapering of the frequency of administration. The recommended phase II dose has been identified, and efficacy results from the phase II portion are expected later this year.101 In general, we have not incorporated use of intrathecal trastuzumab into routine clinical practice, pending efficacy results of this study.

CONCLUSION

Increasingly, the management of breast cancer with brain metastases (parenchymal or leptomeningeal disease) requires close multidisciplinary collaboration, balancing the patient’s disease burden in the CNS and extracranially, prior therapies, performance status, comorbidities, life expectancy, and preferences, with available treatment options. Surgical resection should be strongly considered in patients presenting with a single brain metastasis, or a large, symptomatic mass, particularly if they have good performance status and controlled extracranial disease. For patients with expected longer survival, the use of up-front SRS and avoidance of WBRT in the setting of a limited number of brain metastases is preferred. Although there are still no systemic therapies approved for the treatment of breast cancer brain metastases, a number of regimens have demonstrated clear activity in prospective experiences and can be considered in the clinic. At present, systemic therapy is an option for patients whose CNS disease has progressed through standard local therapy, and it can even be considered in patients with newly diagnosed disease in lieu of local approaches in some circ*mstances (e.g., asymptomatic or minimally symptomatic patients). Moving forward, many novel, promising approaches are being tested in the clinic, and results are eagerly awaited.

References 1. Barnholtz-Sloan JS, Sloan AE, Davis FG, et al. Incidence proportions of brain metastases in patients diagnosed (1973 to 2001) in the Metropoli tan Detroit Cancer Surveillance System. J Clin Oncol. 2004;22:2865-2872.

7. Sperduto PW, Kased N, Roberge D, et al. The effect of tumor subtype on the time from primary diagnosis to development of brain metastases and survival in patients with breast cancer. J Neurooncol. 2013;112:467-472.

2. Arvold ND, Oh KS, Niemierko A, et al. Brain metastases after breastconserving therapy and systemic therapy: incidence and characteristics by biologic subtype. Breast Cancer Res Treat. 2012;136:153-160.

8. Sperduto PW, Kased N, Roberge D, et al. Summary report on the graded prognostic assessment: an accurate and facile diagnosisspecific tool to estimate survival for patients with brain metastases. J Clin Oncol. 2012;30:419-425.

3. Lin NU, Claus E, Sohl J, et al. Sites of distant recurrence and clinical outcomes in patients with metastatic triple-negative breast cancer: high incidence of central nervous system metastases. Cancer. 2008;113:2638-2645. 4. Pestalozzi BC, Holmes E, de Azambuja E, et al. CNS relapses in patients with HER2-positive early breast cancer who have and have not received adjuvant trastuzumab: a retrospective substudy of the HERA trial (BIG 1-01). Lancet Oncol. 2013;14:244-248.

9. Patchell RA, Tibbs PA, Walsh JW, et al. A randomized trial of surgery in the treatment of single metastases to the brain. N Engl J Med. 1990;322:494-500. 10. Vecht CJ, Haaxma-Reiche H, Noordijk EM, et al. Treatment of single brain metastasis: radiotherapy alone or combined with neurosurgery? Ann Neurol. 1993;33:583-590.

5. Kennecke H, Yerushalmi R, Woods R, et al. Metastatic behavior of breast cancer subtypes. J Clin Oncol. 2010;28:3271-3277.

11. Mintz AH, Kestle J, Rathbone MP, et al. A randomized trial to assess the efficacy of surgery in addition to radiotherapy in patients with a single cerebral metastasis. Cancer. 1996;78:1470-1476.

6. Graesslin O, Abdulkarim BS, Coutant C, et al. Nomogram to predict subsequent brain metastasis in patients with metastatic breast cancer. J Clin Oncol. 2010;28:2032-2037.

12. Soffietti R, Abacioglu U, Baumert B. Diagnosis and treatment of brain metastases from solid tumors: guidelines from the European Association of Neuro-Oncology (EANO). Neuro-oncol. 2017. In press.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 53

LIN, GASPAR, AND SOFFIETTI

13. Pollock BE, Brown PD, Foote RL, et al. Properly selected patients with multiple brain metastases may benefit from aggressive treatment of their intracranial disease. J Neurooncol. 2003;61:73-80. 14. Vogelbaum MA, Suh JH. Resectable brain metastases. J Clin Oncol. 2006;24:1289-1294. 15. Kamp MA, Rapp M, Bühner J, et al. Early postoperative magnet resonance tomography after resection of cerebral metastases. Acta Neurochir (Wien). 2015;157:1573-1580. 16. Patel AJ, Suki D, Hatiboglu MA, et al. Impact of surgical methodology on the complication rate and functional outcome of patients with a single brain metastasis. J Neurosurg. 2015;122:1132-1143. 17. Suki D, Abouassi H, Patel AJ, et al. Comparative risk of leptomeningeal disease after resection or stereotactic radiosurgery for solid tumor metastasis to the posterior fossa. J Neurosurg. 2008;108:248-257. 18. Patchell RA, Tibbs PA, Regine WF, et al. Postoperative radiotherapy in the treatment of single metastases to the brain: a randomized trial. JAMA. 1998;280:1485-1489. 19. Brown PD, Ballman KV, Cerhan J, et al. N107C/CEC.3: A phase III trial of post-operative stereotactic radiosurgery (SRS) compared with whole brain radiotherapy (WBRT) for resected metastatic brain disease (LBA1). IntJRadiat Oncol Biol Phys. 2016;96:937. 20. Aoyama H, Shirato H, Tago M, et al. Stereotactic radiosurgery plus whole-brain radiation therapy vs stereotactic radiosurgery alone for treatment of brain metastases: a randomized controlled trial. JAMA. 2006;295:2483-2491.

29. Brown PD, Pugh S, Laack NN, et al; Radiation Therapy Oncology Group (RTOG). Memantine for the prevention of cognitive dysfunction in patients receiving whole-brain radiotherapy: a randomized, doubleblind, placebo-controlled trial. Neuro-oncol. 2013;15:1429-1437. 30. Rapp SR, Case LD, Peiffer A, et al. Donepezil for irradiated brain tumor survivors: a phase iii randomized placebo-controlled clinical trial. J Clin Oncol. 2015;33:1653-1659. 31. Gibson E, Monje M. Effect of cancer therapy on neural stem cells: implications for cognitive function. Curr Opin Oncol. 2012;24:672-678. 32. Gondi V, Tomé WA, Mehta MP. Why avoid the hippocampus? A comprehensive review. Radiother Oncol. 2010;97:370-376. 33. Gondi V, Pugh SL, Tome WA, et al. Preservation of memory with conformal avoidance of the hippocampal neural stem-cell compartment during whole-brain radiotherapy for brain metastases (RTOG 0933): a phase II multi-institutional trial. J Clin Oncol. 2014;32:3810-3816. 34. Suh JH. Hippocampal-avoidance whole-brain radiation therapy: a new standard for patients with brain metastases? J Clin Oncol. 2014;32:3789-3791. 35. Chao ST, Ahluwalia MS, Barnett GH, et al. Challenges with the diagnosis and treatment of cerebral radiation necrosis. Int J Radiat Oncol Biol Phys. 2013;87:449-457. 36. Chuang MT, Liu YS, Tsai YS, et al. Differentiating radiation-induced necrosis from recurrent brain tumor using mr perfusion and spectroscopy: a meta-analysis. PLoS One. 2016;11:e0141438.

21. Chang EL, Wefel JS, Hess KR, et al. Neurocognition in patients with brain metastases treated with radiosurgery or radiosurgery plus whole-brain irradiation: a randomised controlled trial. Lancet Oncol. 2009;10:1037-1044.

37. Hatzoglou V, Yang TJ, Omuro A, et al. A prospective trial of dynamic contrast-enhanced MRI perfusion and fluorine-18 FDG PET-CT in differentiating brain tumor progression from radiation injury after cranial irradiation. Neuro-oncol. 2016;18:873-880.

22. Soffietti R, Kocher M, Abacioglu UM, et al. A European Organisation for Research and Treatment of Cancer phase III trial of adjuvant whole-brain radiotherapy versus observation in patients with one to three brain metastases from solid tumors after surgical resection or radiosurgery: quality-of-life results. J Clin Oncol. 2013;31:65-72.

38. Wagner S, Lanfermann H, Eichner G, et al. Radiation injury versus malignancy after stereotactic radiosurgery for brain metastases: impact of time-dependent changes in lesion morphology on MRI. Neuro-oncol. Epub 2016 Sept 15.

23. Kocher M, Soffietti R, Abacioglu U, et al. Adjuvant whole-brain radiotherapy versus observation after radiosurgery or surgical resection of one to three cerebral metastases: results of the EORTC 22952-26001 study. J Clin Oncol. 2011;29:134-141. 24. Sahgal A, Aoyama H, Kocher M, et al. Phase 3 trials of stereotactic radiosurgery with or without whole-brain radiation therapy for 1 to 4 brain metastases: individual patient data meta-analysis. Int J Radiat Oncol Biol Phys. 2015;91:710-717. 25. Tsao MN, Rades D, Wirth A, et al. Radiotherapeutic and surgical management for newly diagnosed brain metastasis(es): An American Society for Radiation Oncology evidence-based guideline. Pract Radiat Oncol. 2012;2:210-225. 26. DeAngelis LM, Delattre JY, Posner JB. Radiation-induced dementia in patients cured of brain metastases. Neurology. 1989;39:789-796. 27. Aoyama H, Tago M, Kato N, et al. Neurocognitive function of patients with brain metastasis who received either whole brain radiotherapy plus stereotactic radiosurgery or radiosurgery alone. Int J Radiat Oncol Biol Phys. 2007;68:1388-1395. 28. Brown PD, Jaeckle K, Ballman KV, et al. Effect of radiosurgery alone vs radiosurgery with whole brain radiation therapy on cognitive function in patients with 1 to 3 brain metastases: a randomized clinical trial. JAMA. 2016;316:401-409.

54 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

39. Boothe D, Young R, Yamada Y, et al. Bevacizumab as a treatment for radiation necrosis of brain metastases post stereotactic radiosurgery. Neuro-oncol. 2013;15:1257-1263. 40. Lin NU, Lee EQ, Aoyama H, et al; Response Assessment in NeuroOncology (RANO) group. Response assessment criteria for brain metastases: proposal from the RANO group. Lancet Oncol. 2015;16:e270-e278. 41. Okada H, Weller M, Huang R, et al. Immunotherapy response assessment in neuro-oncology: a report of the RANO working group. Lancet Oncol. 2015;16:e534-e542. 42. Rivera E, Meyers C, Groves M, et al. Phase I study of capecitabine in combination with temozolomide in the treatment of patients with brain metastases from breast carcinoma. Cancer. 2006;107:13481354. 43. Ekenel M, Hormigo AM, Peak S, et al. Capecitabine therapy of central nervous system metastases from breast cancer. J Neurooncol. 2007;85:223-227. 44. Rosner D, Nemoto T, Lane WW. Chemotherapy induces regression of brain metastases in breast carcinoma. Cancer. 1986;58:832-839. 45. Caraglia M, Addeo R, Costanzo R, et al. Phase II study of temozolomide plus pegylated liposomal doxorubicin in the treatment of brain metastases from solid tumours. Cancer Chemother Pharmacol. 2006;57:34-39.

MANAGEMENT OF BREAST CANCER BRAIN METASTASES

46. Linot B, Campone M, Augereau P, et al. Use of liposomal doxorubicincyclophosphamide combination in breast cancer patients with brain metastases: a monocentric retrospective study. J Neurooncol. 2014;117:253-259. 47. Franciosi V, Cocconi G, Michiara M, et al. Front-line chemotherapy with cisplatin and etoposide for patients with brain metastases from breast carcinoma, nonsmall cell lung carcinoma, or malignant melanoma: a prospective study. Cancer. 1999;85:1599-1605. 48. Cocconi G, Lottici R, Bisagni G, et al. Combination therapy with platinum and etoposide of brain metastases from breast carcinoma. Cancer Invest. 1990;8:327-334. 49. Christodoulou C, Bafaloukos D, Linardou H, et al; Hellenic Cooperative Oncology Group. Temozolomide (TMZ) combined with cisplatin (CDDP) in patients with brain metastases from solid tumors: a Hellenic Cooperative Oncology Group (HeCOG) Phase II study. J Neurooncol. 2005;71:61-65. 50. Anders C, Deal AM, Abramson V, et al. TBCRC 018: phase II study of iniparib in combination with irinotecan to treat progressive triple negative breast cancer brain metastases. Breast Cancer Res Treat. 2014;146:557-566. 51. Trudeau ME, Crump M, Charpentier D, et al. Temozolomide in metastatic breast cancer (MBC): a phase II trial of the National Cancer Institute of Canada - Clinical Trials Group (NCIC-CTG). Ann Oncol. 2006;17:952-956.

62. Perez EA, Awada A, O’Shaughnessy J, et al. Etirinotecan pegol (NKTR102) versus treatment of physician’s choice in women with advanced breast cancer previously treated with an anthracycline, a taxane, and capecitabine (BEACON): a randomised, open-label, multicentre, phase 3 trial. Lancet Oncol. 2015;16:1556-1568. 63. Thomas FC, Taskar K, Rudraraju V, et al. Uptake of ANG1005, a novel pacl*taxel derivative, through the blood-brain barrier into brain and experimental brain metastases of breast cancer. Pharm Res. 2009;26:2486-2494. 64. O’Sullivan CC, Lindenberg M, Bryla C, et al. ANG1005 for breast cancer brain metastases: correlation between (18)F-FLT-PET after first cycle and MRI in response assessment. Breast Cancer Res Treat. 2016;160:51-59. 65. Kumthekar P, Tang S-C, Brenner AJ, et al. ANG1005, a novel brainpenetrant taxane derivative, for the treatment of recurrent brain metastases and leptomeningeal carcinomatosis from breast cancer. J Clin Oncol. 2016;34 (suppl; abstr 2004). 66. Lin NU, Diéras V, Paul D, et al. Multicenter phase II study of lapatinib in patients with brain metastases from HER2-positive breast cancer. Clin Cancer Res. 2009;15:1452-1459. 67. Lin NU, Eierman W, Greil R, et al. Randomized phase II study of lapatinib plus capecitabine or lapatinib plus topotecan for patients with HER2-positive breast cancer brain metastases. J Neurooncol. 2011;105:613-620.

52. Christodoulou C, Bafaloukos D, Kosmidis P, et al; Hellenic Cooperative Oncology Group. Phase II study of temozolomide in heavily pretreated cancer patients with brain metastases. Ann Oncol. 2001;12:249-254.

68. Sutherland S, Ashley S, Miles D, et al. Treatment of HER2-positive metastatic breast cancer with lapatinib and capecitabine in the lapatinib expanded access programme, including efficacy in brain metastases--the UK experience. Br J Cancer. 2010;102:995-1002.

53. Abrey LE, Olson JD, Raizer JJ, et al. A phase II trial of temozolomide for patients with recurrent or progressive brain metastases. J Neurooncol. 2001;53:259-265.

69. Metro G, Foglietta J, Russillo M, et al. Clinical outcome of patients with brain metastases from HER2-positive breast cancer treated with lapatinib and capecitabine. Ann Oncol. 2011;22:625-630.

54. Siena S, Crinò L, Danova M, et al. Dose-dense temozolomide regimen for the treatment of brain metastases from melanoma, breast cancer, or lung cancer not amenable to surgery or radiosurgery: a multicenter phase II study. Ann Oncol. 2010;21:655-661.

70. Bachelot T, Romieu G, Campone M, et al. Lapatinib plus capecitabine in patients with previously untreated brain metastases from HER2positive metastatic breast cancer (LANDSCAPE): a single-group phase 2 study. Lancet Oncol. 2013;14:64-71.

55. Iwamoto FM, Omuro AM, Raizer JJ, et al. A phase II trial of vinorelbine and intensive temozolomide for patients with recurrent or progressive brain metastases. J Neurooncol. 2008;87:85-90.

71. Freedman RA, Gelman RS, Wefel JS, et al. Translational Breast Cancer Research Consortium (TBCRC) 022: a phase ii trial of neratinib for patients with human epidermal growth factor receptor 2-positive breast cancer and brain metastases. J Clin Oncol. 2016;34:945-952.

56. Tutt A, Ellis P, Kilburn L, et al. The TNT trial: a randomized phase III trial of carboplatin (C) compared with docetaxel (D) for patients with metastatic or recurrent locally advanced triple negative or BRCA 1/2 breast cancer (CRUK/07/012). Presented at: San Antonio Breast Cancer Symposium. San Antonio, TX; 2014. Abstract S3-01. 57. Lien EA, Wester K, Lønning PE, et al. Distribution of tamoxifen and metabolites into brain tissue and brain metastases in breast cancer patients. Br J Cancer. 1991;63:641-645. 58. Pors H, von Eyben FE, Sørensen OS, et al. Longterm remission of multiple brain metastases with tamoxifen. J Neurooncol. 1991;10:173177. 59. Madhup R, Kirti S, Bhatt ML, et al. Letrozole for brain and scalp metas tases from breast cancer—a case report. Breast. 2006;15:440-442. 60. Ito K, Ito T, Okada T, et al. A case of brain metastases from breast cancer that responded to anastrozole monotherapy. Breast J. 2009;15:435437. 61. Adkins CE, Nounou MI, Hye T, et al. NKTR-102 Efficacy versus irinotecan in a mouse model of brain metastases of breast cancer. BMC Cancer. 2015;15:685.

72. Cortés J, Dieras V, Ro J, et al. Afatinib alone or afatinib plus vinorelbine versus investigator’s choice of treatment for HER2-positive breast cancer with progressive brain metastases after trastuzumab, lapatinib, or both (LUX-Breast 3): a randomised, open-label, multicentre, phase 2 trial. Lancet Oncol. 2015;16:1700-1710. 73. Harbeck N, Huang CS, Hurvitz S, et al; LUX-Breast 1 study group. Afatinib plus vinorelbine versus trastuzumab plus vinorelbine in patients with HER2-overexpressing metastatic breast cancer who had progressed on one previous trastuzumab treatment (LUX-Breast 1): an open-label, randomised, phase 3 trial. Lancet Oncol. 2016;17:357366. 74. Metzger O, Barry W, Guo H, et al. Phase I dose-escalation trial of ONT380 in combination with trastuzumab in patients (pts) with HER2+ breast cancer brain metastases. J Clin Oncol. 2016;32:5s (suppl; abstr TPS660). 75. Hamilton E, Borges VF, Conlin A, et al. Efficacy results of a phase 1b study of tucatinib (ONT-380), an oral HER2-specific inhibitor, in combination with capecitabine and trastuzumab in HER2+ metastatic breast cancer, including patients with brain metastases. Presented

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 55

LIN, GASPAR, AND SOFFIETTI

at: San Antonio Breast Cancer Symposium. San Antonio, TX; 2016. Abstract P4-21-01. 76. Dijkers EC, Oude Munnink TH, Kosterink JG, et al. Biodistribution of 89Zr-trastuzumab and PET imaging of HER2-positive lesions in patients with metastatic breast cancer. Clin Pharmacol Ther. 2010;87: 586-592. 77. Bartsch R, Berghoff AS, Preusser M. Breast cancer brain metastases responding to primary systemic therapy with T-DM1. J Neurooncol. 2014;116:205-206. 78. Jacot W, Pons E, Frenel JS, et al. Efficacy and safety of trastuzumab emtansine (T-DM1) in patients with HER2-positive breast cancer with brain metastases. Breast Cancer Res Treat. 2016;157:307-318. 79. Keith KC, Lee Y, Ewend MG, et al. Activity of trastuzumab-emtansine (TDM1) in Her2-positive breast cancer brain metastases: a case series. Cancer Treat Commun. 2016;7:43-46. 80. Krop IE, Lin NU, Blackwell K, et al. Trastuzumab emtansine (T-DM1) versus lapatinib plus capecitabine in patients with HER2-positive metas tatic breast cancer and central nervous system metastases: a retro spective, exploratory analysis in EMILIA. Ann Oncol. 2015;26:113-119. 81. Ramakrishna N, Temin S, Chandarlapaty S, et al. Recommendations on disease management for patients with advanced human epidermal growth factor receptor 2-positive breast cancer and brain metastases: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol. 2014;32:2100-2108. 82. Hortobagyi GN, Stemmer SM, Burris HA, et al. Ribociclib as firstline therapy for HR-positive, advanced breast cancer. N Engl J Med. 2016;375:1738-1748. 83. Turner NC, Ro J, André F, et al; PALOMA3 Study Group. Palbociclib in hormone-receptor-positive advanced breast cancer. N Engl J Med. 2015;373:209-219. 84. Finn RS, Martin M, Rugo HS, et al. Palbociclib and letrozole in advanced breast cancer. N Engl J Med. 2016;375:1925-1936. 85. Barroso-Sousa R, Shapiro GI, Tolaney SM. Clinical development of the CDK4/6 inhibitors ribociclib and abemaciclib in breast cancer. Breast Care (Basel). 2016;11:167-173. 86. Sahebjam S, Le Rhun E, Kulanthaivel P, et al. Assessment of concentrations of abemaciclib and its major active metabolites in plasma, CSF, and brain tumor tissue in patients with brain metastases secondary to hormone receptor positive (HR+) breast cancer. J Clin Oncol. 2016;34 (suppl; abstract 526). 87. Tutt A, Robson M, Garber JE, et al. Oral poly(ADP-ribose) polymerase inhibitor olaparib in patients with BRCA1 or BRCA2 mutations and advanced breast cancer: a proof-of-concept trial. Lancet. 2010;376:235-244. 88. Donawho CK, Luo Y, Luo Y, et al. ABT-888, an orally active poly(ADPribose) polymerase inhibitor that potentiates DNA-damaging agents in preclinical tumor models. Clin Cancer Res. 2007;13:2728-2737.

56 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

89. Kroemer G, Senovilla L, Galluzzi L, et al. Natural and therapyinduced immunosurveillance in breast cancer. Nat Med. 2015;21: 1128-1138. 90. Luen SJ, Salgado R, Fox S, et al. Tumour-infiltrating lymphocytes in advanced HER2-positive breast cancer treated with pertuzumab or placebo in addition to trastuzumab and docetaxel: a retrospec tive analysis of the CLEOPATRA study. Lancet Oncol. 2017;18: 52-62. 91. Duchnowska R, Pęksa R, Radecka B, et al; Polish Brain Metastasis Consortium. Immune response in breast cancer brain metastases and their microenvironment: the role of the PD-1/PD-L axis. Breast Cancer Res. 2016;18:43. 92. Goldberg SB, Gettinger SN, Mahajan A, et al. Pembrolizumab for patients with melanoma or non-small-cell lung cancer and untreated brain metastases: early analysis of a non-randomised, open-label, phase 2 trial. Lancet Oncol. 2016;17:976-983. 93. Morikawa A, Jordan L, Rozner R, et al. Characteristics and outcomes of patients with breast cancer with leptomeningeal metastasis. Clin Breast Cancer. 2017;17:23-28. 94. Le Rhun E, Taillibert S, Zairi F, et al. Clinicopathological features of breast cancers predict the development of leptomeningeal metas tases: a case-control study. J Neurooncol. 2011;105:309-315. 95. Chamberlain M, Soffietti R, Raizer J, et al. Leptomeningeal metastasis: a Response Assessment in Neuro-Oncology critical review of endpoints and response criteria of published randomized clinical trials. Neurooncol. 2014;16:1176-1185. 96. Chamberlain M, Junck L, Brandsma D, et al. Leptomeningeal metastases: a RANO proposal for response criteria. Neuro-oncol. Epub 2016 Dec 29. 97. Jaeckle KA. Neoplastic meningitis from systemic malignancies: diag nosis, prognosis and treatment. Semin Oncol. 2006;33:312-323. 98. Jaeckle KA, Phuphanich S, Bent MJ, et al. Intrathecal treatment of neoplastic meningitis due to breast cancer with a slow-release formulation of cytarabine. Br J Cancer. 2001;84:157-163. 99. Mason WP, Yeh SD, DeAngelis LM. 111Indium-diethylenetriamine pentaacetic acid cerebrospinal fluid flow studies predict distribution of intrathecally administered chemotherapy and outcome in patients with leptomeningeal metastases. Neurology. 1998;50: 438-444. 100. Glantz MJ, Jaeckle KA, Chamberlain MC, et al. A randomized controlled trial comparing intrathecal sustained-release cytarabine (DepoCyt) to intrathecal methotrexate in patients with neoplastic meningitis from solid tumors. Clin Cancer Res. 1999;5:3394-3402. 101. Raizer J, Pentsova E, Omuro A, et al. Phase I trial of intrathecal trastzuumab in HER2 positive leptomeningeal metastases. Presented at: 19th Annual Scientific Meeting and Education Day of the Society for Neuro-Oncology. Miami, FL; 2014. Abstract AT-47.

CARDIORESPIRATORY FITNESS IN BREAST CANCER RECURRENCE

Lifestyle Interventions to Improve Cardiorespiratory Fitness and Reduce Breast Cancer Recurrence Mark J. Haykowsky, PhD, Jessica M. Scott, PhD, Kathryn Hudson, MD, and Neelima Denduluri, MD OVERVIEW As patients are living longer after a cancer diagnosis, survivorship is becoming increasingly important in cancer care. The sequelae of multimodality therapies include weight gain and decreased cardiorespiratory fitness, which increase cardiovascular risk. Evidence suggests that physical activity reduces the risk of breast cancer recurrence and death. Avoidance of weight gain after therapy also improves outcomes after a diagnosis of breast cancer. Prospective randomized trials must be performed to determine the benefits of specific physical activity and dietary habits for survivors of breast cancer. This review outlines the important physiologic changes that occur with antineoplastic therapy and the important role of exercise and diet.

reast cancer is the most frequently diagnosed cancer among women and the second leading cause of cancer death in the United States.1 Breast cancer mortality has decreased by nearly 40% during the last 3 decades as a result of advances in prevention, early detection, and treatment.2 As a result of improved survival and population aging, breast cancer is evolving into a disease of older survivors who face an important new set of health care challenges. Nearly one-third of breast cancer survivors have a peak oxygen uptake (peak VO2)—the gold standard measure of cardiorespiratory fitness—that is below the threshold level required for full and independent living.3 A consequence of reduced fitness is decreased survival in healthy populations.4 In accordance with the multiple-hit hypothesis, unfavorable lifestyle factors (e.g., sedentary lifestyle, sarcopenic obesity) coupled with the adverse effects of anticancer therapy result in reduced physiologic and functional reserve capacity. Interventions that improve cardiovascular health and body composition outcomes (e.g., increased muscle mass, decreased visceral adiposity) may play an important role in improving cardiorespiratory fitness, reduce breast cancer recurrence, and improve mortality. The aim of this chapter is to briefly review the mechanisms responsible for reduced peak VO2 in survivors of breast cancer and the role of exercise training to improve peak VO2 and the role of diet, weight reduction, and exercise in the moderation of cardiovascular disease sequelae, reduction of breast cancer recurrence, and improvement in mortality.

CARDIORESPIRATORY FITNESS AMONG BREAST CANCER SURVIVORS

Breast cancer survivors with normal resting left ventricular (LV) systolic function have a peak VO2 that is 19% (5.5 mL/kg/min) lower than healthy age-matched noncancer controls.5-10 The magnitude of the decline in peak VO2 is greatest during the short-term period after completing adjuvant therapy.3,11 Lower levels of cardiorespiratory fitness, as measured by peak VO2, may also have important prognostic implications and result in shorter survival in women with metastatic disease.3 Thus, an important goal is to maintain an optimal level of cardiorespiratory fitness across the breast cancer survivorship continuum.

DETERMINANTS OF PEAK VO2: ROLE OF IMPAIRED CARDIOVASCULAR FUNCTION

Given that VO2 is equal to the product of cardiac output and arterial-venous oxygen content difference, the reduced peak VO2 among breast cancer survivors may be due to central (cardiac) or peripheral (skeletal muscle and its microvasculature) factors that result in decreased oxygen delivery to and/or extraction by the active muscles.5,12 To date, only one study has examined the acute hemodynamic cardiopulmonary response to maximal aerobic exercise in 47 survivors of breast cancer with normal resting LV systolic function (mean age, 59; mean LV ejection fraction, 64%) and 11 age-matched healthy controls.6 As shown in Fig. 1, the decreased peak VO2 in survivors of breast cancer was primarily due to a lower stroke volume and cardiac output, as heart rate and arterial-venous oxygen difference

From the College of Nursing and Health Innovation, The University of Texas at Arlington, Arlington, TX; Memorial Sloan Kettering Cancer Center, New York, NY; US Oncology Network, Texas Oncology, Austin, TX; US Oncology Network, Virginia Cancer Specialists, Arlington, VA. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Mark J. Haykowsky, PhD, College of Nursing and Health Innovation, 411 S. Nedderman Dr., Arlington, TX 76010; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 57

HAYKOWSKY ET AL

FIGURE 1. Fick Determinants of Impaired Peak VO2 in Breast Cancer Survivors

Abbreviations: Peak VO2, peak oxygen uptake; CO, cardiac output; a-vO2Diff, arterial-venous oxygen difference.

during maximal cycle exercise were not significantly different between groups.6 The mechanism responsible for the reduced maximal exercise stroke volume was not examined; however, it may be the result of increased LV afterload as maximal systemic vascular resistance was 11% higher in breast cancer survivors compared with controls.6 Oxygen extraction is directly related to muscle oxygen diffusive conductance (e.g., transport of O2 from hemoglobin to muscle mitochondria) and inversely related to muscle blood flow.12 Accordingly, our finding that maximal arterial-venous oxygen difference was not different between survivors of breast cancer and healthy controls despite a lower maximal cardiac output6 (muscle blood flow) suggests that abnormalities in skeletal muscle microvascular and/or mitochondrial function may play an important role in limiting breast cancer survivors exercise performance.5 To attenuate the decline (during adjuvant therapy) or increase peak VO2 (post-adjuvant therapy), therapies should focus on improving cardiovascular and skeletal muscle function.

KEY POINTS • Breast cancer survivors have reduced cardiorespiratory fitness secondary to impaired cardiovascular reserve. • Exercise training is an effective intervention to improve cardiorespiratory fitness; however, the physiologic mechanisms underpinning this favorable adaptation are unknown. • Concomitant diet and exercise interventions may abrogate breast cancer therapy–induced accelerated CVD sequelae, particularly in overweight/obese women. • Epidemiologic evidence supports the participation in exercise before and after breast cancer diagnosis because it is a contributing factor in decreasing breast cancer recurrence, breast cancer–related mortality, and overall mortality. 58 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

IMPROVEMENT IN PEAK VO2 WITH EXERCISE TRAINING Exercise training is an effective intervention to improve peak VO2, physical functioning, and quality of life, and reduces symptoms of fatigue in breast cancer survivors.13 The magnitude of the change in peak VO2 with training appears to be related to the volume of exercise performed, and the threshold workload required to obtain a clinically significant large increase in peak VO2 (effect size > 1) was 600 intensity-minutes in a 10-week supervised exercise training program performed for 90 minutes per week at 70% peak VO2.14 Although the mechanisms underpinning the exercisetraining mediated improvement in peak VO2 have not been studied, they may be due to favorable changes in cardiac, peripheral vascular, or skeletal muscle function.5

EFFICACY OF DIET AND EXERCISE INTERVENTIONS TO MODULATE THERAPYINDUCED CARDIOVASCULAR DISEASE

Breast cancer survivors who are obese have a significantly higher overall, and breast cancer–related, mortality compared with their counterparts who are at a baseline healthy weight within 12 months after diagnosis.15 There is increasing evidence demonstrating that breast cancer survivors are at a higher risk of morbidity and mortality. Indeed, compared with sex- and age-matched counterparts, patients with breast cancer have an increased incidence of risk factors for cardiovascular disease (CVD; e.g., obesity, hypertension, diabetes, dyslipidemia, exercise intolerance)16 and CVD-specific morbidity (e.g., coronary artery disease and heart failure). Moreover, in survivors older than age 65, CVD is the leading cause of mortality.17,18 As prolonged administration of anticancer therapies becomes increasingly common19 and novel targeted therapies with potentially adverse cardiovascular safety profiles are included in treatment strategies,20 the incidence of CVD morbidity and mortality among breast cancer survivors will likely continue to rise.

CARDIORESPIRATORY FITNESS IN BREAST CANCER RECURRENCE

Thus, defining the feasibility and efficacy of innovative interventions that can attenuate CVD sequelae are of primary research and clinical importance. Exercise, a pleiotropic stimulus leading to physiologic adaptation across multiple organ systems,21 improves insulin sensitivity, decreases lipids, and lowers blood pressure with concomitant improvements in peak VO2 in noncancer settings.22-26 Although comparatively less information is available in oncology settings, recent observational data of patients with breast cancer indicate that adherence to national exercise guidelines for adult patients with cancer (i.e., ≥ 9 MET hours/week) was associated with an adjusted 23% reduction in the risk of CVD events compared with not meeting the guidelines (< 9 MET hours/week; p = .0002).27 The association with exercise did not differ according to age, most CVD risk factors, menopausal status, or anticancer treatment,27 suggesting that, for many patients with breast cancer, exercise is a potent intervention that can modulate CVD sequelae. However, the protective effects of exercise did not extend to women with a body mass index (BMI) of 35 kg/m2 or greater.27 As a result, additional interventions may be required for patients with excess CVD risk associated with obesity. Based on promising data indicating that weight control, physical activity, and/or diet quality reduce cancer recurrence and improve cancer-specific overall survival, the American Cancer Society issued Guidelines on Nutrition and Physical Activity for Cancer Survivors, which call for maintenance of a healthy body weight, regular physical activity regardless of BMI, and modest weight loss for cancer survivors who are overweight or obese.28 Importantly, concomitant diet and exercise interventions in nononcology overweight/obese settings have been shown to improve LV function, exercise capacity, glucose, lipid, and blood pressure control, inflammation markers, body composition, and skeletal muscle function.29 Thus, the synergistic benefits of multicomponent interventions could represent an optimal approach to offset CVD in patients with breast cancer.

CVD SEQUELAE AMONG PATIENTS WITH BREAST CANCER

The incidence of common risk factors for both CVD and cancer such as hypertension (up to 55%),30 diabetes (up to 10%),31 hyperlipidemia (up to 20%),32 obesity (up to 62%),33 and low exercise tolerance (up to 37%)3 likely increase the risk of CVD morbidity and mortality34 For example, Hooning et al examined the long-term causes of mortality among 7,425 women treated for early-stage breast cancer and found that after a median of 13.8 years, survivors diagnosed with one CVD risk factor at any time during the study follow-up had a 1.4- to 3.1-fold higher risk of CVD-related mortality relative to age-matched women among the general population.16,17 Moreover, Playdon et al reported that a weight gain of more than 5% from diagnosis to post-treatment was associated with a 12% increase in the risk of all-cause mortality compared with weight maintenance in a meta-analysis involving 23,832 patients with early-stage breast cancer.33

Similarly, among 3,993 women with early-stage disease (5.8 years postdiagnosis), those with a BMI greater than 30 kg/m2 (classified as obese) had a CVD mortality rate 1.65 times that of women with a normal BMI (18.5–24.9 kg/m2), and each 5 kg weight gain was associated with a 19% increase in CVD mortality.35 These findings highlight the importance of identifying women at the greatest risk of accelerated CVD sequalae so targeted interventions can be initiated.

EVIDENCE OF EFFICACY OF COMBINED DIET AND EXERCISE INTERVENTIONS TO MODULATE CVD SEQUELAE

Evidence from nononcology trials indicate that multicomponent interventions may be critical for improving outcomes such as body composition, peak VO2, and biomarkers linked to CVD outcomes. For example, in 107 obese adults (BMI > 30 mg/kg2) were randomly assigned to one of four groups for 52 weeks: (1) 27 patients in the control group, (2) 26 patients in the diet group, (3) 26 patients in the exercise group, and (4) 28 patients in the diet and exercise group.36 Peak VO2 improved more in the diet and exercise group than in the diet or exercise alone groups (increases of 17% vs. 10% vs. 8%, respectively; p < .001), whereas body weight decreased by 10% in the diet alone group and by 9% in the diet-exercise group, but did not decrease in the exercise group or the control group (p < .001). Similarly, in 439 postmenopausal women who were overweight/obese and randomly assigned to: (1) a reduced calorie, weight loss diet (diet; 118 patients); (2) moderate-to-vigorous intensity aerobic exercise (exercise; 117 patients); (3) a combination of a reduced calorie, weight loss diet and moderate-tovigorous intensity aerobic exercise (diet and exercise; 117 patients); or (4) control (87 patients),37 leptin concentrations, a key regulator of energy homeostasis, metabolism, and adiposity, decreased in all of the intervention groups, but the greatest reduction occurred with diet and exercise (-40%). Taken together, these findings suggest that a combination of weight loss and an exercise program could provide greater improvement in multiple outcomes compared with either intervention alone in overweight/obese populations. To date, the potential cardioprotective properties of multimodal interventions in patients with breast cancer have received limited attention; however, preliminary observational data indicate that adherence to diet and exercise guidelines improves patient morbidity and mortality. For example, among 938 breast cancer survivors in the Iowa Women’s Health Study (mean age, 79; 8.6 years postdiagnosis),38 adherence to the World Cancer Research Fund/ American Institute for Cancer Research (WCRF/AICR) prevention guidelines for weight control, physical activity, and diet was associated with lower all-cause mortality (hazard ratio [HR] 0.67; 95% CI, 0.50–0.94), and a 40% reduction in CVD-specific mortality. The majority of trials of patients with breast cancer have examined the efficacy of either exercise alone to improve functional outcomes (e.g., VO2peak) or diet alone to direct weight management or weight loss. As a result, only approximately six trials have investigated asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 59

HAYKOWSKY ET AL

the effect of a combined diet and exercise intervention among patients with breast cancer. For example, 90 postmenopausal patients with breast cancer receiving adjuvant chemotherapy were randomly assigned to (1) a calcium-rich diet intervention (attention control), (2) calcium-rich diet and exercise, or (3) calcium-rich diet with high fruit and vegetable, low-fat diet and exercise. Demark-Wahnefried et al39 reported that the high fruit and vegetable arm substantially attenuated therapy-induced increases in appendicular body fat. Similarly, Morey et al40 reported that among 641 older, overweight, long-term survivors of breast (289 patients), prostate (261 patients), and colorectal (91 patients) cancer randomly assigned to either a 12-month home-based program of telephone counseling promoting exercise and diet, or wait-list control, weight loss was significantly greater in counseling groups compared with the wait-list control group (2.06 vs. 0.92 kg, respectively; p < .001). The Life After Cancer Epidemiology (LACE) study examined the impact of dietary adherence in 1,901 women diagnosed with early-stage breast cancer. A prudent dietary pattern was defined by a high intake of fruits, vegetables, whole grains, and poultry, whereas a Western diet was defined by high intake of red and processed meats and refined grains. The prudent diet was associated with a significant decrease in risk of overall death (trend p = .02; HR for highest quartile 0.57; 95% CI, 0.36–0.90) and death from non–breast cancer causes (trend p = .003; HR for highest quartile 0.35; 95% CI, 0.17–0.73) independent of physical activity, body habitus, or tobacco use. In contrast, a Western diet was related to an increasing risk of overall death (trend p = .05) and death from non–breast cancer causes (p = .02). Interestingly, both dietary patterns were not associated with reduced risk of breast cancer recurrence or breast cancer– specific mortality.41 The Women's Health Initiative Dietary Modification primary breast cancer prevention trial randomly assigned 48,835 postmenopausal women with no prior history of breast cancer and normal mammograms to undergo dietary intervention or to a control group. The dietary intervention reduced dietary fat intake to 20% of calories, increased fruit and vegetable intake (five servings a day), and increased grains to six servings a day. During year 1, intervention group members participated in 18 group sessions and then quarterly maintenance meetings. The control group participants received dietary guidelines. Although the incidence of breast cancer was not significantly decreased among women in the low-fat diet group, women in the intervention group had improved overall survival at 8.5 years (HR 0.65; 95% CI, 0.45–0.94; p = .02. At the 16-year mark, breast cancer-specific mortality was still lower than those in the control group (234 vs. 443 deaths, respectively; HR 0.82; 95% CI, 0.70–0.96; p = .01). Women with baseline waist circumference of 88 cm or greater and higher baseline levels of dietary fat intake had a stronger interaction in terms of deaths after breast cancer.42 The randomized phase III WINS trial evaluated whether dietary fat reduction affected relapse-free survival among 60 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

2,437 patients with early-stage breast cancer receiving standard-of-care treatment. Women with a dietary fat intake greater than 20% of calories were randomly assigned to a dietary intervention group or a control group. Women were given a fat-gram goal by centrally trained, registered dietitians implementing a low-fat eating plan. Women in the intervention arm underwent 8 biweekly individual counseling sessions, were subsequently contacted every 3 months, and self-monitored their fat-gram intake using a “keeping score” book. Fat intake was externally monitored by unannounced 24-hour telephone recalls performed annually for 5 years. Women enrolled in the intervention group consumed 9.2% calories from fat and lost nearly 6 pounds. Although there was no survival benefit at 19.4 years, an exploratory subgroup analysis of the group with estrogen negative tumors showed higher median survival of 13.6 years in the intervention group compared with 11.7 years in the control arm (HR 0.46; p = .006).42 Although these findings are promising for patients with breast cancer with, or at high risk of obesity, the long-term implications of acute multimodal interventions on CVD morbidity and mortality are unknown. To this end, the Look AHEAD (Action for Health in Diabetes) study examined the incidence of a composite cardiovascular outcome (cardiovascular death, nonfatal myocardial infarction, nonfatal stroke, or hospitalized angina) over 9.6 years in 5,145 overweight or obese individuals with type 2 diabetes randomly assigned to a diet and exercise intervention or control.43 Although improvements in weight loss, fitness, and CVD risk factors were greater in the intervention group, there was no significant difference between the intervention and control group in CVD morbidity and mortality (403 vs. 418 events, respectively; 1.83/100 person-years vs. 1.92/100 person-years, respectively; HR 0.95; 95% CI, 0.83–1.09; p = .505).43 Thus, the effect of multimodal interventions in patients with breast cancer on outcomes other than weight loss, or in patients with concomitant comorbidities such as hypertension, dyslipidemia, diabetes, or exercise intolerance, are important areas for further research. In summary, given that patients with breast cancer now live long enough to be at risk for therapy-induced CVD morbidity and mortality, a research agenda that addresses the nature and magnitude of therapy-related CVD could define important targets for interventions. To this end, observational data indicating that exercise-induced modification of CVD events is attenuated in obese breast cancer survivors27 suggest that, for a subset of patients with breast cancer who are overweight or obese, multimodal interventions may be critical to abrogating accelerated CVD. Prospective trials are needed to define the role of diet and exercise in the management of CVD sequelae in overweight and obese breast cancer survivors.

ROLE OF EXERCISE IN RISK REDUCTION OF BREAST CANCER

The role of exercise in secondary prevention and breast cancer–related mortality is not well-defined. However, the

CARDIORESPIRATORY FITNESS IN BREAST CANCER RECURRENCE

known physiologic effects of exercise suggest that it may have an important role. Exercise may reduce adverse outcomes associated with fat accumulation such as an altered hormonal environment and adipokine production, which may act to promote tumor development and growth.44 Randomized, controlled trials of physical activity among postmenopausal women who are overweight demonstrated declines in serum levels of androgen, estrogen, and leptin— hormones important for tumorigenesis.45-47 For these reasons, exercise may be a modifiable behavior that has the potential to change important breast cancer outcomes. In the next section, we review the salient data on the impact of exercise on breast cancer recurrence and mortality.

Exercise and Risk of Recurrence and Effect on Breast Cancer Mortality: Epidemiologic Data

The majority of studies examining this question have been observational cohort studies based on patient self-reporting, an important limitation. The landmark study is the Nurses’ Health Study, a prospective observational study of 2,987 female registered nurses with stage I to III breast cancer between 1983 and 1998.48 The women were followed every 2 years until 2002 or time of death, and answered questions about their physical activity over the prior year. After adjusting for factors predictive of survival after breast cancer, compared with women who engaged in less than 3 MET-hours/week of physical activity, the relative risk (RR) of death from breast cancer was 0.80 (95% CI, 0.60–1.06) for 3 to 8.9 MET-hours/week; 0.50 (95% CI, 0.31–0.82) for 9 to 14.9 MET-hours/week; 0.56 (95% CI, 0.38–0.84) for 15 to 23.9 MET-hours/week; and 0.60 (95% CI, 0.40–0.89) for 24 or more MET-hours/week (trend p = .004). After multivariable adjustment, the RR of breast cancer recurrence was 0.83 (95% CI, 0.64–1.08) for 3 to 8.9 MET-hours/week; 0.57 (95% CI, 0.38–0.85) 9 to 14.9 MET-hours/week; 0.66 (95% CI, 0.47–0.93) for 15-23.9 MET-hours/week; and 0.74 (95% CI, 0.53–1.04) for 24 or more MET-hours/week as compared with women who engaged in less than 3 MET-hours/week of physical activity (trend p = .05). Interestingly, the RR for each adverse outcome was lowest for the intermediate level of activity, equivalent to walking 3 to 5 hours per week at an average pace. The protective benefit was similar among nonobese and obese women, but the benefit was particularly apparent in women with hormone-responsive tumors. The RR of breast cancer death for women with hormoneresponsive tumors who engaged in 9 or more MET-hours/ week of activity compared with women with hormoneresponsive tumors who engaged in less than 9 MET-hours/ week was 0.50 (95% CI, 0.34–0.74). An important limitation of this study is that the participants were mostly non-Hispanic whites and occupationally hom*ogenous.48 The Collaborative Women’s Longevity Study (CWLS) and the Life After Cancer Epidemiology (LACE) study used the same measure of physical activity as the Nurse’s Health Study, but found different results.49,50 In CWLS, a total of 4,482 eligible women age 20 to 79 diagnosed with invasive breast cancer stages I to III between 1988 and 2001

completed the questionnaire of physical activity a median of 5.6 years after diagnosis. After adjusting for relevant factors, women who engaged in greater levels of activity had a significantly lower risk of dying from breast cancer (HR 0.65; 95% CI, 0.39–1.08 for 2.8 to 7.9 MET hours/week; HR 0.59; 95% CI, 0.35–1.01 for 8.0 to 20.9 MET hours/week; and HR 0.51, 95% CI, 0.29–0.89 for > 21 MET hours/week; trend p = .05). Results were similar for overall survival (HR 0.44; 95% CI, 0.32–0.60 for > 21.0 vs. < 2.8 MET hours/week; trend p < .001) and were similar regardless of a woman’s age (although the majority of women were age > 50), stage of disease, and BMI. Similar to the Nurse’s Health Study, no benefit was demonstrated for vigorous-intensity activity and most of the participants were Caucasian. 50 In the LACE study, 1,970 women age 18 to 79 with stage I to III breast cancer from 1997 to 2000 completed the physical activity questionnaire the prior 6 months only. Age-adjusted results suggested that higher levels of physical activity were associated with reduced risk of recurrence and breast cancer mortality (trend p = .05 and .07, respectively, for highest versus lowest level of hours per week of moderate physical activity), but were not significant after adjusting for prognostic factors and other variables. Of note, in multivariable analyses, there remained a significant protective association between physical activity and allcause mortality (HR 0.66; 95% CI, 0.42–1.03; trend p = .04). A strength and unique characteristic of this study is that it contained 20% minorities. It is hypothesized that the lack of power and healthier nature of the participants led to the null results of this study.49 Several studies have looked at exercise in breast cancer at several time points, an advantage over the studies that evaluate a single point in time. The Health, Eating, Activity, and Lifestyle (HEAL) study was a prospective, observational study of 933 women age 18 and older diagnosed with stage I to III cancer between 1995 and 1998 that measured activity levels the year prior to diagnosis and 2 years after diagnosis. Compared with women who were inactive both before and after diagnosis, women who increased physical activity after diagnosis had a 45% lower risk of death (HR 0.55; 95% CI, 0.22–1.38), and women who decreased physical activity after diagnosis had a fourfold greater risk of death (HR 3.95; 95% CI, 1.45–10.50). Although the risk reductions were observed for total deaths, the majority of deaths were from breast cancer.51 Similarly, the Women’s Health Initiative (WHI) study measured activity at diagnosis and 3 or 6 years post diagnosis in 4,643 postmenopausal women. Women participating in at least 9 MET hours/week (approximately 3 hours per week of brisk walking) of physical activity after diagnosis had lower breast cancer mortality (HR 0.61; 95% CI, 0.35–0.99; p = .049) and lower all-cause mortality (HR 0.54; 95% CI, 0.38–0.79); p < .01). Even in women who were inactive prior to diagnosis, those who increased or maintained physical activity of at least 9 MET hours/week after diagnosis had lower all-cause mortality (HR 0.67; 95% CI, 0.46–0.96).52 The Women’s Healthy Eating and Living Study (WHEL) similarly measured activity at baseline and 1 year in 2,361 asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 61

HAYKOWSKY ET AL

women age 18 and older with stage I to III breast cancer. Adherence to activity guidelines was associated with a 35% lower mortality risk (HR 0.65; 95% CI, 0.47–0.91; p < .01). There was no effect seen on breast cancer events, although deaths were mostly secondary to breast cancer. Unlike the WHI study, in WHEL the change in activity during 1 year was not associated with improved outcomes, potentially secondary to shorter interval between reports.53 The Shanghai Breast Cancer Survival Study (SBCSS) assessed exercise at three time points (6, 18, and 36 months post-diagnosis) in 4,826 Chinese women age 20 to 70 between 2002 and 2006. After adjusting for several covariates, exercise during the first 36 months post-diagnosis was inversely associated with total mortality and recurrence/disease-specific mortality with HRs of 0.70 (95% CI, 0.56–0.88) and 0.60 (95% CI, 0.47–0.76), respectively, regardless of stage and BMI. They observed a dose-response relationship between mortality rates and exercise duration and MET scores, and the mortality association was only among estrogen- and progesterone receptor–negative patients.54 The clear strength of this study is that it evaluated more than two time points; however it is uncertain if this study can be applied to a Western breast cancer population.

EFFECT OF EXERCISE ON BREAST CANCER RECURRENCE AND MORTALITY: CLINICAL TRIALS

Data from randomized controlled trials are limited. The Supervised Trial of Aerobic versus Resistance Training (START) trial randomly selected 242 patients with breast cancer between 2003 and 2005 to usual care, supervised aerobic, or resistance exercise during chemotherapy.55 The trial was originally designed to examine the independent effects of aerobic and resistance exercise on quality of life, healthrelated fitness, and other patient-reported outcomes. As an

exploratory analysis, overall survival and disease-free survival was estimated. Eight-year disease-free survival was 82.7% for the exercise groups compared with 75.6% for the control group (HR 0.68; 95% CI, 0.37–1.24; logrank p = .21). In exploratory subgroup analyses, the strongest effects were among women who were overweight or obese, had stage II/III cancers, estrogen receptor–positive tumors, HER2-positive tumors, and received taxane-based chemotherapies and optimal chemotherapy dosing.56 This study is limited by its exploratory nature but is certainly hypothesis-generating. Phase III studies comparing exercise to usual care in breast cancer survivors are warranted.

EXERCISE SUMMARY AND RECOMMENDATIONS

Supportive data from randomized controlled trials are lacking, but epidemiologic evidence supports the participation in exercise before and after breast cancer diagnosis to decrease breast cancer recurrence, breast cancer–related mortality, and overall mortality. The American Cancer Society encourages cancer survivors to engage in 150 minutes per week of moderate or 75 minutes per week of vigorous aerobic exercise.28

CONCLUSION

Decreased cardiorespiratory fitness, obesity, and a sedentary lifestyle negatively impact the outcomes of breast cancer survivors. Lifestyle interventions to improve survival are of much interest. However, there is a paucity of prospective, randomized clinical trial data to suggest specific exercise and dietary recommendations to improve cardiovascular fitness and reduce breast cancer–specific mortality in this population. Prospective, randomized studies that address interventions are necessary to guide breast cancer survivors.

References 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J Clin. 2016;66:7-30. 2. American Cancer Society. Cancer Facts & Figures. Atlanta: American Cancer Society; 2016. 3. Jones LW, Courneya KS, Mackey JR, et al. Cardiopulmonary function and age-related decline across the breast cancer survivorship continuum. J Clin Oncol. 2012;30:2530-2537. 4. Gulati M, Pandey DK, Arnsdorf MF, et al. Exercise capacity and the risk of death in women: the St James Women Take Heart Project. Circulation. 2003;108:1554-1559. 5. Haykowsky MJ, Beaudry R, Brothers RM, et al. Pathophysiology of exercise intolerance in breast cancer survivors with preserved left ventricular ejection fraction. Clin Sci (Lond). 2016;130: 2239-2244. 6. Jones LW, Haykowsky M, Pituskin EN, et al. Cardiovascular reserve and risk profile of postmenopausal women after chemoendocrine therapy for hormone receptor--positive operable breast cancer. Oncologist. 2007;12:1156-1164.

62 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

7. Jones LW, Haykowsky M, Peddle CJ, et al. Cardiovascular risk profile of patients with HER2/neu-positive breast cancer treated with anthracycline-taxane-containing adjuvant chemotherapy and/or tras tuzumab. Cancer Epidemiol Biomarkers Prev. 2007;16:1026-1031. 8. Burnett D, Kluding P, Porter C, et al. Cardiorespiratory fitness in breast cancer survivors. Springerplus. 2013;2:68. 9. Khouri MG, Hornsby WE, Risum N, et al. Utility of 3-dimensional echocardiography, global longitudinal strain, and exercise stress echocardiography to detect cardiac dysfunction in breast cancer patients treated with doxorubicin-containing adjuvant therapy. Breast Cancer Res Treat. 2014;143:531-539. 10. Koelwyn GJ, Lewis NC, Ellard SL, et al. Ventricular-arterial coupling in breast cancer patients after treatment with anthracycline-containing adjuvant chemotherapy. Oncologist. 2016;21:141-149. 11. Scharhag-Rosenberger F, Kuehl R, Klassen O, et al. Exercise training intensity prescription in breast cancer survivors: validity of current practice and specific recommendations. J Cancer Surviv. 2015;9: 612-619.

CARDIORESPIRATORY FITNESS IN BREAST CANCER RECURRENCE

12. Haykowsky MJ, Tomczak CR, Scott JM, et al. Determinants of exercise intolerance in patients with heart failure and reduced or preserved ejection fraction. J Appl Physiol (1985). 2015;119:739-744.

life in obese older patients with heart failure with preserved ejection fraction: a randomized clinical trial. JAMA. 2016;315:36-46.

13. McNeely ML, Campbell KL, Rowe BH, et al. Effects of exercise on breast cancer patients and survivors: a systematic review and meta-analysis. CMAJ. 2006;175:34-41.

30. Klepin HD, Pitcher BN, Ballman KV, et al. Comorbidity, chemotherapy toxicity, and outcomes among older women receiving adjuvant chemotherapy for breast cancer on a clinical trial: CALGB 49907 and CALGB 361004 (alliance). J Oncol Pract. 2014;10:e285-e292.

14. Beaudry R, Kruger C, Liang Y, et al. Effect of supervised exercise on aerobic capacity in cancer survivors: adherence and workload predict variance in effect. World J Meta-Anal. 2015;3:43-53.

31. Nechuta S, Lu W, Zheng Y, et al. Comorbidities and breast cancer survival: a report from the Shanghai Breast Cancer Survival Study. Breast Cancer Res Treat. 2013;139:227-235.

15. Chan DS, Vieira AR, Aune D, et al. Body mass index and survival in women with breast cancer-systematic literature review and metaanalysis of 82 follow-up studies. Ann Oncol. 2014;25:1901-1914.

32. Jones LW, Haykowsky MJ, Swartz JJ, et al. Early breast cancer therapy and cardiovascular injury. J Am Coll Cardiol. 2007;50:1435-1441.

16. Hooning MJ, Botma A, Aleman BM, et al. Long-term risk of cardiovascular disease in 10-year survivors of breast cancer. J Natl Cancer Inst. 2007;99:365-375. 17. Hooning MJ, Aleman BM, van Rosmalen AJ, et al. Cause-specific mortality in long-term survivors of breast cancer: a 25-year follow-up study. Int J Radiat Oncol Biol Phys. 2006;64:1081-1091. 18. Bradshaw PT, Stevens J, Khankari N, et al. Cardiovascular disease mortality among breast cancer survivors. Epidemiology. 2016;27:6-13. 19. Koelwyn GJ, Khouri M, Mackey JR, et al. Running on empty: cardiovascular reserve capacity and late effects of therapy in cancer survivorship. J Clin Oncol. 2012;30:4458-4461. 20. Moslehi JJ. Cardiovascular toxic effects of targeted cancer therapies. N Engl J Med. 2016;375:1457-1467. 21. Gielen S, Schuler G, Adams V. Cardiovascular effects of exercise training: molecular mechanisms. Circulation. 2010;122:1221-1238. 22. Jones LW, Eves ND, Haykowsky M, et al. Exercise intolerance in cancer and the role of exercise therapy to reverse dysfunction. Lancet Oncol. 2009;10:598-605. 23. Flynn KE, Piña IL, Whellan DJ, et al; HF-ACTION Investigators. Effects of exercise training on health status in patients with chronic heart failure: HF-ACTION randomized controlled trial. JAMA. 2009;301:1451-1459. 24. Erbs S, Höllriegel R, Linke A, et al. Exercise training in patients with advanced chronic heart failure (NYHA IIIb) promotes restoration of peripheral vasomotor function, induction of endogenous regeneration, and improvement of left ventricular function. Circ Heart Fail. 2010;3:486-494. 25. Eisele JC, Schaefer IM, Randel Nyengaard J, et al. Effect of voluntary exercise on number and volume of cardiomyocytes and their mitochondria in the mouse left ventricle. Basic Res Cardiol. 2008;103:12-21. 26. Kavazis AN, McClung JM, Hood DA, et al. Exercise induces a cardiac mitochondrial phenotype that resists apoptotic stimuli. Am J Physiol Heart Circ Physiol. 2008;294:H928-H935. 27. Jones LW, Habel LA, Weltzien E, et al. Exercise and risk of cardiovascular events in women with nonmetastatic breast cancer. J Clin Oncol. 2016;34:2743-2749. 28. Kushi LH, Doyle C, McCullough M, et al; American Cancer Society 2010 Nutrition and Physical Activity Guidelines Advisory Committee. American Cancer Society Guidelines on nutrition and physical activity for cancer prevention: reducing the risk of cancer with healthy food choices and physical activity. CA Cancer J Clin. 2012;62:30-67. 29. Kitzman DW, Brubaker P, Morgan T, et al. Effect of caloric restriction or aerobic exercise training on peak oxygen consumption and quality of

33. Playdon MC, Bracken MB, Sanft TB, et al. Weight gain after breast cancer diagnosis and all-cause mortality: systematic review and metaanalysis. J Natl Cancer Inst. 2015;107:djv275. 34. Scott JM, Koelwyn GJ, Hornsby WE, et al. Exercise therapy as treatment for cardiovascular and oncologic disease after a diagnosis of earlystage cancer. Semin Oncol. 2013;40:218-228. 35. Nichols HB, Trentham-Dietz A, Egan KM, et al. Body mass index before and after breast cancer diagnosis: associations with all-cause, breast cancer, and cardiovascular disease mortality. Cancer Epidemiol Biomarkers Prev. 2009;18:1403-1409. 36. Villareal DT, Chode S, Parimi N, et al. Weight loss, exercise, or both and physical function in obese older adults. N Engl J Med. 2011;364: 1218-1229. 37. Abbenhardt C, McTiernan A, Alfano CM, et al. Effects of individual and combined dietary weight loss and exercise interventions in postmenopausal women on adiponectin and leptin levels. J Intern Med. 2013;274:163-175. 38. Inoue-Choi M, Robien K, Lazovich D. Adherence to the WCRF/AICR guidelines for cancer prevention is associated with lower mortality among older female cancer survivors. Cancer Epidemiol Biomarkers Prev. 2013;22:792-802. 39. Demark-Wahnefried W, Case LD, Blackwell K, et al. Results of a diet/ exercise feasibility trial to prevent adverse body composition change in breast cancer patients on adjuvant chemotherapy. Clin Breast Cancer. 2008;8:70-79. 40. Morey MC, Snyder DC, Sloane R, et al. Effects of home-based diet and exercise on functional outcomes among older, overweight longterm cancer survivors: RENEW: a randomized controlled trial. JAMA. 2009;301:1883-1891. 41. Kwan ML, Weltzien E, Kushi LH, et al. Dietary patterns and breast cancer recurrence and survival among women with early-stage breast cancer. J Clin Oncol. 2009;27:919-926. 42. Tabung FK, Steck SE, Liese AD, et al. Association between dietary inflammatory potential and breast cancer incidence and death: results from the Women’s Health Initiative. Br J Cancer. 2016;114: 1277-1285. 43. Wing RR, Bolin P, Brancati FL, et al; Look AHEAD Research Group. Cardiovascular effects of intensive lifestyle intervention in type 2 diabetes. N Engl J Med. 2013;369:145-154. 44. Cohen DH, LeRoith D. Obesity, type 2 diabetes, and cancer: the insulin and IGF connection. Endocr Relat Cancer. 2012;19:F27-F45. 45. McTiernan A, Tworoger SS, Rajan KB, et al. Effect of exercise on serum androgens in postmenopausal women: a 12-month randomized clinical trial. Cancer Epidemiol Biomarkers Prev. 2004;13:1099-1105.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 63

HAYKOWSKY ET AL

46. McTiernan A, Tworoger SS, Ulrich CM, et al. Effect of exercise on serum estrogens in postmenopausal women: a 12-month randomized clinical trial. Cancer Res. 2004;64:2923-2928. 47. Rogers LQ, Fogleman A, Trammell R, et al. Effects of a physical activity behavior change intervention on inflammation and related health outcomes in breast cancer survivors: pilot randomized trial. Integr Cancer Ther. 2013;12:323-335. 48. Holmes MD, Chen WY, Feskanich D, et al. Physical activity and survival after breast cancer diagnosis. JAMA. 2005;293:2479-2486. 49. Sternfeld B, Weltzien E, Quesenberry CP Jr, et al. Physical activity and risk of recurrence and mortality in breast cancer survivors: findings from the LACE study. Cancer Epidemiol Biomarkers Prev. 2009;18: 87-95. 50. Holick CN, Newcomb PA, Trentham-Dietz A, et al. Physical activity and survival after diagnosis of invasive breast cancer. Cancer Epidemiol Biomarkers Prev. 2008;17:379-386. 51. Irwin ML, Smith AW, McTiernan A, et al. Influence of pre- and postdiagnosis physical activity on mortality in breast cancer survivors:

64 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

the health, eating, activity, and lifestyle study. J Clin Oncol. 2008;26: 3958-3964. 52. Irwin ML, McTiernan A, Manson JE, et al. Physical activity and survival in postmenopausal women with breast cancer: results from the women’s health initiative. Cancer Prev Res (Phila). 2011;4:522-529. 53. Bertram LA, Stefanick ML, Saquib N, et al. Physical activity, additional breast cancer events, and mortality among early-stage breast cancer survivors: findings from the WHEL Study. Cancer Causes Control. 2011;22:427-435. 54. Chen X, Lu W, Zheng W, et al. Exercise after diagnosis of breast cancer in association with survival. Cancer Prev Res (Phila). 2011;4:1409-1418. 55. Courneya KS, Segal RJ, Mackey JR, et al. Effects of aerobic and resistance exercise in breast cancer patients receiving adjuvant chemotherapy: a multicenter randomized controlled trial. J Clin Oncol. 2007;25:4396-4404. 56. Courneya KS, Segal RJ, McKenzie DC, et al. Effects of exercise during adjuvant chemotherapy on breast cancer outcomes. Med Sci Sports Exerc. 2014;46:1744-1751.

NOVEL TARGETED AGENTS AND IMMUNOTHERAPY IN BREAST CANCER

Novel Targeted Agents and Immunotherapy in Breast Cancer Ingrid A. Mayer, MD, MSCI, Rebecca Dent, MSc, MD, FRCP, Tira Tan, MBBS, MRCP, Peter Savas, MBBS, FRACP, and Sherene Loi, MBBS, FRACP, PhD OVERVIEW The treatment of breast cancer is generally determined according to breast cancer subtype: hormone receptor–positive (luminal), triple-negative (basal-like), and HER2-overexpressing breast cancer. Recent years have seen the development of exciting novel and potent therapeutics based on molecular pathways, immune modulation, and antibody conjugates. In this article, we cover new and emerging therapeutic areas and ongoing clinical trials that may result in further improvements in breast cancer outcomes.

reast cancer clinicians have been fortunate in the past to have proven efficacious treatment options to offer their patients. Novel therapeutics continue to expand treatment options for patients with early-stage and advanced breast cancer. For estrogen receptor– positive (ER+), HER2-negative, and HER2-amplified disease, novel agents can combine with existing effective therapies to reverse or delay treatment resistance. Determining the optimal combination and best treatment sequence remains a difficult challenge, however. In contrast, triple-negative breast cancer (TNBC) is a heterogeneous disease that has devastating consequences on relapse with limited treatment options, and improvements in outcomes will rely on novel therapies and identifying the subgroups of patients most likely to benefit. We discuss potential further therapeutic directions for the three main breast cancer subtypes in this article.

MECHANISMS OF ENDOCRINE THERAPY RESISTANCE IN ER+ BREAST CANCER

Acquired resistance (defined as recurrence at least 6–12 months after completion of adjuvant therapy or disease progression more than 6 months after endocrine therapy initiated in the metastatic setting) and occasionally primary resistance (recurrence either within adjuvant therapy or within 6–12 months of completion of adjuvant therapy or disease progression more than 6 months after treatment in the metastatic setting) to antiestrogen therapy is inevitable in patients with ER+ metastatic breast cancer (MBC).1 A variety of mechanisms have been implicated in primary and acquired resistance to endocrine agents (Sidebar 1). Below we review some of the strategies to overcome endocrine therapy resistance.

Cyclin-Dependent Kinases 4 and 6 Inhibitors

Inhibitors of the cyclin-dependent kinases 4 and 6 (CDK4/6) have demonstrated impressive activity in patients with ER+/HER2-negative MBC.2 Palbociclib is an orally active pyridopyrimidine first-in-class compound that is a potent and highly selective reversible inhibitor of CDK4/6.3 By inhibiting CDK4/6, palbociclib prevents tumor cell entry into S phase.4 Consistent with its CDK4/6 specificity, treatment with palbociclib reduces expression of the proliferation marker Ki67 and is completely inactive in Rb-deficient tumor cells.5 Preclinical data have shown that endocrineresistant ER+ breast cancer cells are highly sensitive to palbociclib with and without antihormonal therapy.6 In previously untreated metastatic ER+/HER2-negative MBC, the phase I/II PALOMA-1 trial found an impressive improvement in progression-free survival (PFS) with palbociclib plus letrozole over letrozole alone.7 The confirmatory phase III PALOMA-2 study randomized a total of 666 postmenopausal patients with ER+ MBC and no prior systemic therapy to receive letrozole with palbociclib or letrozole with placebo. Median PFS (the primary endpoint) was 24.8 months versus 14.5 months in favor of the palbociclib arm (hazard ratio [HR], 0.58; 95% CI, 0.46–0.72; p < .000001).6 Response rate was also improved in the palbociclib arm (42.1% vs. 34.7%, p = .031), and clinical benefit rate was 84.9% versus 70.3% (p < .0001). Similar evidence of efficacy was seen in the phase III PALOMA-3 trial with the combination of fulvestrant plus palbociclib, in which the PFS was 9.2 months versus 3.8 months with fulvestrant plus placebo (HR, 0.42; p < .000001) in patients with disease progression after at least one line of hormonal therapy and at most one line of chemotherapy but naive to CDK4/6 inhibitors.2,8 In both phase III trials, the most common grade 3 or 4 adverse

From the Vanderbilt University Medical Center, Nashville, TN; Division of Medical Oncology, National Cancer Centre Singapore, Singapore; Duke-NUS Medical School, Singapore, Singapore; Peter MacCallum Cancer Centre, Melbourne, Australia. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Sherene Loi, Peter MacCallum Cancer Centre, 305 Grattan St., Melbourne, VIC 3000, Australia; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 65

MAYER ET AL

mTOR Inhibitors SIDEBAR 1. Mechanisms of Resistance to Endocrine Agents Primary Resistance

• Receptor tyrosine kinase/growth factor signaling pathway • FGFR amplification • EGFR/ERBB2 mutations • Cell cycle control signaling pathway • Cyclin D1 amplification or expression • MYC amplification and overexpression • Hormone signaling pathway • Loss of ERα • Post-translational modification of ERα • Expression of ER coactivation/corepression factors

Acquired Resistance

• PI3K/AKT1/MTOR signaling pathway • PI3K/AKT/mTOR pathway activation • Mitogen-activated protein (MAP) kinase pathway • MAPK/ERK pathway activation • Hormone signaling pathway • ESR1 mutations • Changes in the tumor microenvironment

event in the palbociclib arms was neutropenia (incidence 62%–65%), but treatment was otherwise well tolerated. Both palbociclib with letrozole for first-line treatment and palbociclib with fulvestrant for second line treatment of patients with ER+/HER2-negative MBC are approved by the U.S. Food and Drug Administration (FDA).

KEY POINTS • Breast cancer was the first solid tumor type leading the way with targeted therapy: tamoxifen and, subsequently, trastuzumab. • ER+, HER2-negative disease has lately seen the emergence of highly effective CDK4/6 inhibitors, which likely will represent a major advance for this breast cancer subtype. • HER2+ breast cancer represents the poster child for oncogene addiction and targeted therapy: the advent of trastuzumab has resulted in early-stage HER2+ disease becoming highly curable, and we now understand that these tumors remain addicted to HER2-signaling. Newer HER2-directed therapies, probably in combination with immunotherapies, will result in patients with advanced disease enjoying long periods of time with excellent quality of life. • Triple negative breast cancer remains the major challenge for breast cancer clinicians. It is hoped that newer DNA damage repair inhibitors, immunotherapies, and antibody-drug conjugates will result in improved survival. 66 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

The addition of the mTOR inhibitor everolimus to endocrine therapy can reverse endocrine resistance in ER+ MBC, as shown in the BOLERO-29 and TAMRAD10 randomized phase III trials. In both studies, all patients were previously exposed to aromatase inhibitors (AIs), and most of them developed progression after initial response (acquired resistance). This led to the approval of everolimus by the FDA and the European Medicines Agency in combination with endocrine therapy after failure of AIs in 2012. In contrast to these positive results, the phase III HORIZON trial found the addition of the mTOR inhibitor temsirolimus to letrozole did not improve PFS over letrozole alone.11 This trial was conducted in the first-line setting, and most patients were AI naive, suggesting that mTOR signaling may have a specific role in acquired resistance to endocrine therapy. There are several ongoing trials that will better define the role of everolimus in advanced disease: BOLERO-6 (NCT01783444), a phase II trial comparing exemestane/everolimus to capecitabine in ER+/HER2-negative disease refractory to AI, and BOLERO-4 (NCT01698918), a phase II single-arm study evaluating the role of everolimus as first-line treatment. Everolimus is also being evaluated in the adjuvant setting with two different studies using two different approaches: (1) SWOG1207 (NCT01674140) will randomly assign high-risk premenopausal and postmenopausal patients to add everolimus or placebo to their standard adjuvant endocrine therapy, and (2) NCT01805271 will evaluate the addition of everolimus to adjuvant endocrine therapy in high-risk ER+/HER2-negative patients with breast cancer who remain disease free after at least 1 year of treatment.

PI3K Inhibitors

PI3K inhibitors consist of pan-PI3K targeting all class I isoforms, isoform-specific PI3K inhibitors, and dual PI3K/mTOR inhibitors. Compounds may also display differential activity for wild-type and mutant PI3K proteins. Response rates with single-agent PI3K inhibitors are far below those of other kinase inhibitors in other cancer types (such as EGFR, ALK, or BRAF inhibitors). Frequent coexisting genetic alterations, compensatory feedback loops, and toxicity that precludes maintaining adequate dose intensity are possible explanations for this diminished efficacy. Buparlisib (BKM120) is a pan-PI3K inhibitor with potent activity against mutant PI3Kα.12 Early phase trials of buparlisib plus endocrine therapy reported activity and a manageable safety profile characterized by transaminitis, hyperglycemia, diarrhea, and mood disorders (anxiety, depression, irritability).12,13 The randomized phase III BELLE-2 trial studied fulvestrant 500 mg plus buparlisib 100 mg daily or placebo in postmenopausal MBC progressing on AIs.14 Buparlisib increased the median PFS by 1.9 months (6.9 months vs. 5.0 months, p < .001). For patients with PI3K/ AKT pathway activation (defined as PIK3CA mutation or PTEN loss, assayed for the majority in the archival primary tumor) there was no difference in the benefit of buparlisib. However, in the subset of patients in whom PIK3CA mutation

NOVEL TARGETED AGENTS AND IMMUNOTHERAPY IN BREAST CANCER

was assessed by circulating tumor DNA at trial entry, buparlisib plus fulvestrant increased PFS in PIK3CA mutant cases compared with fulvestrant alone (7 months vs. 3.2 months; HR, 0.56; p < .001). Using the same treatment arms as BELLE-2, the phase III BELLE-3 trial enrolled AI-experienced patients with disease progression in the past 30 days on an mTOR inhibitor plus endocrine therapy.15 Median PFS for patients in the buparlisib arm was 3.9 months versus 1.8 months for fulvestrant/ placebo, and 6-month PFS rates were 30.6% and 20.1%, respectively. Of 349 patients for whom PIK3CA mutation status from circulating tumor DNA was available, 147 had mutations in the gene. Among those with PIK3CA mutations, PFS was 4.7 months in the buparlisib arm versus 1.6 months in the placebo arm. A similar result was seen with PIK3CA status in tumor tissue. PI3Kα is the isoform predominantly mutated in cancer, and studies have shown that selective inactivation of this isoform is enough to block PI3K/AKT signaling in response to different growth factor stimuli.16-18 Early results with panPI3K inhibition suffered from substantial toxicity and reduced efficacy,19 and inhibiting PI3Kα selectively is designed to achieve a better therapeutic index by targeting the driving isoform in a specific cancer. Two PI3Kα inhibitors are in clinical development for breast cancer in combination with endocrine therapy: alpelisib (BYL719) and taselisib (GDC0032). Single-agent alpelisib showed preferential activity in solid tumors harboring PIK3CA mutations.20,21 Alpelisib plus fulvestrant is being studied in a phase III trial for metastatic ER+ breast cancer progressing on AIs (SOLAR1, NCT02437318) and in a neoadjuvant phase II trial in combination with letrozole (NEO-ORB, NCT01923168). Taselisib is a is a potent inhibitor of p110α, p110δ, and p110γ but with 30-fold less inhibition of p110β relative to p110α and greater selectivity against PIK3CA mutant isoforms than wild-type.22 Taselisib and fulvestrant is being tested in a randomized phase III study in the metastatic setting for women with previous exposure to AIs and enrichment for PIK3CA mutation (SANDPIPER, NCT02340221) and a neoadjuvant phase II trial in combination with letrozole (LORELEI, NCT02273973).

ESR1 Mutations

Mutations in the ligand binding domain of the ER gene ESR1 result in estrogen-independent ER signaling and resistance to antiestrogen therapy.23-26 ESR1 mutations are uncommon in primary breast cancers at the time of diagnosis, but they have been identified in up to 55% of ER+ MBC previously treated with antiestrogen therapy.27 ESR1 mutations may have a role in determining optimal endocrine therapy. In the SoFEA phase III trial,28 ESR1 mutations were found in 39% of patients (63 of 161) with rates of mutation detection unaffected by delays in processing of archival plasma. Patients with ESR1 mutations had improved PFS after taking fulvestrant (45 patients) compared with exemestane (18 patients; HR 0.52; 95% CI, 0.30–0.92; p = .02), whereas patients with wild-type ESR1 had similar PFS after receiving either treatment (HR 1.07; 95% CI, 0.68–1.67; p = .77).29 In PALOMA3,

ESR1 mutations were found in the plasma of 25.3% of patients (91 of 360), with mutations associated with acquired resistance to prior AIs. Fulvestrant plus palbociclib improved PFS compared with fulvestrant plus placebo in both ESR1 mutant (HR 0.43; 95% CI, 0.25–0.74; p = .002) and ESR1 wild-type patients (HR 0.49; 95% CI, 0.35–0.70; p < .001).29 ESR1 mutations are often “polyclonal,” with multiple different mutations detectable in the same patient.29 There is considerable interest in developing antiestrogen therapies that are effective in the presence of ESR1 mutations.

NEW APPROACHES IN TNBC: PARP INHIBITORS AND BEYOND

TNBC, which lacks all three predictive and prognostic immunohistochemical biomarkers, ER, progesterone receptor, and HER2, has few therapeutic options beyond chemotherapy which to date remains the standard of care. Clinically, TNBC has an aggressive tumor biology with the worst disease-specific outcome compared with other subtypes, representing an important challenge and unmet clinical need.30,31 Survival data from clinical trials indicate that the median overall survival for patients with metastatic TNBC (mTNBC) is approximately 11 to 14 months and is indeed much shorter than among patients with other MBC subtypes.32,33 Subtypes of TNBC have been described on the basis of histopathologic features and gene expression profiling, highlighting the heterogeneity and complexity of these tumors.34-37 Four distinct breast cancer subtypes (luminal A, luminal B, HER2-enriched, and basal-like) of prognostic and predictive significance were first described by Perou et al38 in 2000 using microarray analysis. Of the four subtypes, basal-like tumors are typically of triple-negative phenotype, and the vast majority (approximately 80%) of TNBCs are of the basal-like subtype.36,39 In analyzing gene expression profiles of TNBC, Lehmann et al35 identified six distinct molecular subtypes (basal-like 1, basal-like 2, immunomodulatory, mesenchymal, mesenchymal stem-like, and luminal androgen receptor). This was refined into four tumor-specific subtypes (basal-like 1, basal-like 2, mesenchymal, and luminal androgen receptor) following histopathology and lasercapture microdissection, which identified infiltrating lymphocytes and tumor-associated stromal cells contributing to the immunomodulatory and mesenchymal stem-like subtypes, respectively.36 In addition to microarray-based studies, the genomic landscape of this disease has also been extensively interrogated, identifying alterations adding to our burgeoning knowledge of TNBC.27,40 The features and alterations unique to these various subtypes have been incorporated into many ongoing, rationally designed trials to refine treatment strategies. In this article, we discuss notable novel approaches in the treatment of TNBC.

Cytotoxic Chemotherapy

The triple-negative paradox describes a higher responsiveness of TNBC to chemotherapy despite the overall unfavorable prognosis.41 Currently, chemotherapy is the mainstay asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 67

MAYER ET AL

to treating TNBC. Taxane and anthracycline-containing regimens remain the preferred chemotherapeutic options, and recent results have clarified the role of nab-pacl*taxel and carboplatin. The Triple-Negative Albumin-Bound Pacl*taxel Combination International Treatment Study (tnppAcity) is a phase II trial evaluating nab-pacl*taxel in combination with either gemcitabine or carboplatin versus gemcitabine plus carboplatin in previously untreated mTNBC.42 It found a significantly longer PFS (median PFS, 7.4 months vs. 5.4 months, p = .03, and 7.4 months vs. 6 months, p = .02) and overall response rate (ORR; 72% vs. 39% and 44%) in favor of nab-pacl*taxel plus carboplatin compared with pacl*taxel plus gemcitabine or gemcitabine plus carboplatin.42 In the Triple Negative Breast Cancer Trial (TNT), 376 patients with previously untreated mTNBC were randomly assigned to receive either carboplatin or docetaxel monotherapy.43 Although there was no difference in the ORR or PFS between the two arms in the overall population, the key finding was patients who harbored deleterious germline BRCA mutations fared better when treated with carboplatin, with greater ORR (68% vs. 33.3%, p = .03) and PFS (6.8 months vs. 3.1 months, p = .03) compared with docetaxel.43 Thus, both platinum and nab-pacl*taxel should be included in our armamentarium of treatment of TNBC as well as other standard chemotherapies used for other subtypes of breast cancer.

Targeting Defective DNA Repair

A significant proportion of BRCA-mutated breast cancers are TNBC44 or have gene expression profiles similar to basal-like TNBC.45 The BRCA gene complex plays an important role in maintenance of genomic stability via hom*ologous recombination, one of the coordinated pathways that act to identify DNA aberrations and restore genomic stability. Loss of function of BRCA confers a defective hom*ologous recombination phenotype. This affords an opportunity for achieving “synthetic lethality” by using PARP inhibitors.46 Concurrent tumor intrinsic BRCA loss of function and pharmacologic inhibition of PARP results in tumor cell death with a high therapeutic index.47 As a proof of concept, clinical trials in the initial development of PARP inhibitors have focused largely on BRCAmutated tumors.48-50 In a phase II trial of olaparib monotherapy in two sequential cohorts of BRCA-mutated advanced breast cancer, an ORR of 11 of 27 (41%) was seen in the cohort treated at 400 mg twice daily and 6 of 27 (22%) in the cohort treated at 100 mg twice daily. A significant proportion of both cohorts had TNBC; 7 of 13 triple-negative cases (54%) responded to 400 mg twice daily, while 4 of 16 triple-negative cases (25%) responded to 100 mg twice daily.49 Ongoing phase III trials evaluating PARP inhibition in BRCA-mutant MBC include olaparib in OlympiAD (NCT02000622), which will report at the 2017 ASCO Annual Meeting with a press release noting that the trial had reached its primary endpoint, niraparib in BRAVO (NCT01905592), and talazoparib in EMBRACA (NCT01945775). Ongoing efforts are focused on molecular diagnostics beyond BRCA testing to predict benefit from PARP inhibition as well as applying PARP 68 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

inhibitors in a broader population through combination strategies.

Immunotherapy: Checkpoint Inhibition

Approximately 20% of TNBCs are classified as the immunomodulatory subtype and are characterized by genes involved in the immune system.36 Early-stage TNBC has high levels of tumor-infiltrating lymphocytes, and increasing levels of tumor-infiltrating lymphocytes predict a better prognosis.51 These tumor-infiltrating lymphocyte contribute significantly to the gene expression profiles and express immune checkpoint genes such as programmed cell death–1 (PD-1) and programmed cell death ligand–1 (PD-L1).36,52 T-cell checkpoint inhibitors, which relieve the immunosuppressive tumor microenvironment and promote antitumor immune responses, have generated much excitement by demonstrating lasting efficacy in some patients across a broad array of tumor types, including TNBC. In a multicohort phase IB study of monotherapy with the anti-PD1 monoclonal antibody pembrolizumab, the ORR was 18.5% in 28 evaluable patients with TNBC displaying PD-L1 expression (positive staining in stroma or on at least 1% of tumor cells by immunohistochemistry).53 The median duration of response was not reached, and three responders remained on study for at least 1 year. These promising results led to the initiation of KEYNOTE-086 (NCT02447003), a larger single-arm phase II study to evaluate the role of pembrolizumab in advanced TNBC and identify biomarkers of efficacy. The preliminary results of this study will be reported at the 2017 ASCO Annual Meeting. In addition, KEYNOTE-119 (NCT02555657), a randomized phase III study of pembrolizumab versus physician’s choice single-agent chemotherapy in pretreated advanced TNBC, is estimated to complete recruitment in late 2017. Finally, atezolizumab has also shown efficacy as a single agent in a phase IA trial in PD-L1–positive tumors where a cohort of 12 patients with mTNBC were treated, with an ORR of 33%.54 Combinations of immunotherapy and chemotherapy may be more efficacious in TNBC. It has been postulated that chemotherapy could promote an immune response to cancer and hence be synergistic with immune therapy.55 Several trials are investigating combination strategies enhancing the efficacy of immunotherapy and expanding its reach to a broader population of patients. In a phase IB trial of atezolizumab in combination with nab-pacl*taxel in mTNBC, the ORR (including unconfirmed responses) in all patients was an impressive 71%, with a range of 43% in those patients treated in the third line and beyond to 89% in those previously untreated.56 Importantly, the regimen had a tolerable safety profile, and responses were seen in both PD-L1–expressing and PD-L1– nonexpressing tumors. IMpassion130 (NCT02425891), a phase III study of nab-pacl*taxel with or without atezolizumab in previously untreated mTNBC, is ongoing and expected to enroll 900 patients across 270 sites globally. KEYNOTE-355 (NCT02819518) is a two-part phase III study evaluating the safety and efficacy of pembrolizumab in combination with three different chemotherapies, in the first-line setting.

NOVEL TARGETED AGENTS AND IMMUNOTHERAPY IN BREAST CANCER

Androgen Receptor Blockade

Gene expression microarray-based studies have identified the luminal androgen receptor subtype, which highly expresses androgen receptor (AR) messenger RNA in addition to downstream AR targets and coactivators.7,35,57 Thus, it is postulated that AR inhibition would have antitumor activity in a well-defined subgroup of TNBC. A phase II trial of abiraterone acetate, an inhibitor of 17-α-hydroxylase/17,20-lyase (CYP17) in a cohort of heavily pretreated AR-positive (at least 10% by immunohistochemistry) TNBC demonstrated a 6-month clinical benefit rate of 20% (95% CI, 7.7%–38.6%) and PFS of 2.8 months.58 Similarly, in a phase II trial of enzalutamide, a potent AR inhibitor, the 24-week clinical benefit rate was 29% (95% CI, 20%–41%), and median PFS of 14 weeks (95% CI, 8–19 weeks) was seen in the 57 evaluable patients.59 In this study, an androgen-driven diagnostic gene signature was associated with greater clinical benefit, and the phase III ENDEAR trial of pacl*taxel plus enzalutamide/ placebo and enzalutamide monotherapy has been initiated in diagnostic signature positive TNBC (NCT02929576).60

Antibody-Drug Conjugates

Antibody-drug conjugates (ADCs) are a novel class of cancer therapeutics, which amalgamate the selectivity of a targeted treatment and cytotoxicity of chemotherapy, resulting in an improved therapeutic index. Sacituzumab govitecan (IMMU-132) is an anti-Trop-2 ADC consisting of humanized IgG antibody against Trop-2 linked to SN-38, an active metabolite of irinotecan. The Trop-2 protein is an epithelial cancer antigen found to be highly expressed in a majority of TNBC compared with normal tissues and is associated with a poor prognosis and aggressive disease.61 In the first-in-human phase I trial, sacituzumab govitecan had an acceptable safety profile and evidence of efficacy including one confirmed response and two minor responses seen in three of four patients with TNBC.62 In the ongoing multicenter phase II trial, promising PFS of 5.6 months (95% CI, 3.6–7.1 months), overall survival of 14.3 months (95% CI, 10.5–18.8 months), and a response rate of 29% were seen in a heavily pretreated (median of five prior therapies) population of TNBC.63 Sacituzumab govitecan has been given breakthrough therapy and fast-track designation from the FDA, and a phase III international multicenter randomized trial versus treatment of physician’s choice in refractory mTNBC is planned for initiation in 2017 (NCT02574455). Glembatumumab vedotin (CDX-011) is a fully human IgG2 monoclonal antibody with high affinity for extracellular domain of glycoprotein nonmetastatic B linked to the microtubule inhibitor monomethyl auristatin E (MMAE). Glycoprotein nonmetastatic B is highly expressed in TNBC in relation to normal tissue, predicts breast cancer recurrence, and is associated with reduced overall survival.64 Early activity was seen in mTNBC and high-gpNMB-expressing tumors in the phase II EMERGE study.65 The METRIC trial, a randomized phase III study evaluating glembatumumab vedotin versus capecitabine, is ongoing in gpNMB overexpressing TNBC (NCT01997333).

Targeting the PI3K/AKT/mTOR Pathway

Numerous studies have shown that the PI3K/AKT/mTOR pathway is activated in TBNC either through loss of pTEN or INPP4B or mutations in PIK3CA or AKT.66-68 A neoadjuvant study of weekly pacl*taxel/doxorubicin/cyclophosphamide in combination with the AKT inhibitor MK-2206 versus chemotherapy alone showed a pathologic complete response rate of 40% in the combination arm compared with 22% in the chemotherapy alone arm.69 In the metastatic setting, the LOTUS trial (NCT02162719), a phase III study of pacl*taxel alone or in combination with the AKT inhibitor ipatasertib, has completed recruitment and results will be available in 2017. Finally, the luminal AR subtype is known to be enriched with PIK3CA mutations, and the combination of antiandrogens with PI3K inhibitors is currently being evaluated.70

NOVEL THERAPEUTIC DIRECTIONS FOR HER2AMPLIFIED BREAST CANCER

HER2-overexpressing breast cancer exemplifies the concept of oncogene addiction as a highly rewarding treatment target. The success of trastuzumab and subsequent HER2 therapies (pertuzumab, T-DM1, lapatinib, neratinib) highlights that these breastcancers remain highly dependent on the HER2 pathway as treatment resistance develops. Hence the search for more potent agents against the HER2 pathway remains highly active. With the success of the trastuzumab and pertuzumab combination, de-escalation of chemotherapy in the early-stage setting will become a tractable goal. Most patients with advanced disease, however, present with de novo metastatic disease, and here are required therapies that can achieve long term disease control with a favorable toxicity profile and are effective in preventing and managing central nervous system (CNS) metastases. We speculate the treatment in advanced HER2+ disease will become focused on ER+ versus ER-negative status presence or absence and the stability of CNS disease as well as determining the presence of preexisting host immunity as major determinants for deciding ongoing therapy.

Novel HER2 Tyrosine Kinase Inhibitors

The role of HER2 tyrosine kinase inhibitors (TKIs) in the management of early- and late-stage HER2+ breast cancer is evolving. The toxicity of lapatinib and neratinib has hampered their widespread use, and it remains unclear how to best use these drugs in managing CNS disease, and in combination with trastuzumab, pertuzumab or T-DM1. Next-generation HER2 TKIs may ameliorate these issues. Tucatinib (formerly ONT-380, ARRY-380), is a highly selective HER2 tyrosine kinase small-molecule inhibitor.71 This selectivity is expected to ameliorate the diarrhea and skin toxicity, which are the most troublesome side effects of existing HER2 TKIs such as lapatinib. Additionally, studies with an intracranial HER2 xenograft mouse model demonstrated superior survival with tucatinib treatment over lapatinib or neratinib.72 The preclinical rationale for reduced toxicity was confirmed in a phase I study of tucatinib in solid tumors with asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 69

MAYER ET AL

HER2 overexpression.73 The study included an expansion cohort of 17 patients with metastatic HER2+ breast cancer. The dose-limiting toxicity was elevated liver transaminases. At the maximum tolerated dose of 600 mg twice daily, grade 1 or 2 diarrhea occurred in 26% of patients, with no cases of grade 3 or 4 diarrhea. Grade 1 and 2 nausea occurred in 33% of patients. Among the patients with HER2+ breast cancer treated with the maximum tolerated dose or higher, three of 22 had a partial response (14%). A phase IB trial tested tucatinib 300 mg twice daily in combination with capecitabine (100 mg/m2) and/or trastuzumab (6 mg/kg thrice weekly) in patients with prior exposure to trastuzumab, taxane, and T-DM1 (NCT02025192).74 The ORR was 83% for capecitabine plus tucatinib (6 patients), 40% for trastuzumab plus tucatinib (15 patients), and 61% for capecitabine plus trastuzumab plus tucatinib (23 patients). The median duration of response in the triplet arm was 10 months. In the capecitabine-containing arms, grade 1 or 2 diarrhea occurred in 68% of patients. The rate of grade 3 diarrhea was 9% (three of 34), similar to treatment with capecitabine alone. A follow-up phase IB trial tested tucatinib plus T-DM1 in a similar patient population that had not experienced T-DM1 (NCT01983501).75 The recommended phase II dose of tucatinib 300 mg twice daily was used. The ORR was 47% in 34 patients with measurable disease, with similar benefit in those having received one or more prior HER2 agents. The most common grade 3 and 4 toxicities were thrombocytopenia (28%), increased alanine transaminase (16%), and fatigue (12%). Diarrhea incidence was 56% grade 1 or 2 and 4% grade 3. CNS disease remains a challenging problem in the management of advanced HER2+ metastatic disease. Recently, neratinib and afatinib monotherapy have been tested in the setting of progressive CNS metastases, with relatively poor results. The LUX-Breast 3 trial found no benefit and increased toxicity with afatanib alone or afatinib and vinorelbine compared with physician’s choice therapy.76 The Translational Breast Cancer Research Consortium (TBCRC) 022 single-arm phase II trial tested neratinib monotherapy in patients with progressive CNS metastases after prior CNSdirected therapy. The partial response rate was 8%, and 21% of patients experienced grade 3 or worse diarrhea despite loperamide prophylaxis.77 Encouraging results in the setting of CNS metastases have been seen with tucatinib in a combined analysis of patients with CNS metastases across the two phase IB trials described above. The partial response rate was 38% in asymptomatic untreated CNS metastases and 33% in progressive CNS metastases after prior radiotherapy or surgery.78 Considering these results, the phase III HER2CLIMB study is comparing tucatinib versus placebo in combination with capecitabine and trastuzumab (NCT02614794) in patients who have received prior taxane, trastuzumab, pertuzumab, and T-DM1. Asymptomatic CNS metastases and previously treated CNS metastases are permitted. The study enrollment has recently expanded to 480 patients.79 70 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

CDK4/6 Inhibitors

Preclinical data in mouse models of HER2+ breast cancer have shown that CDK4/6 inhibitors can restore sensitivity to anti-HER2 therapy in resistant tumors.80 The randomized phase II monarcHER trial compares abemaciclib plus trastuzumab plus fulvestrant versus abemaciclib plus trastuzumab versus physician’s choice, in ER+/HER2+ MBC with at least two prior anti-HER2 therapies (NCT02675231).81 The phase II PATRICIA trial is delivering palbociclib and trastuzumab with or without letrozole in postmenopausal patients with advanced HER2+ breast cancer and includes ER+ and ERnegative groups (NCT02448420). A phase IB trial of palbociclib and T-DM1 (NCT01976169) and a phase I/II trial of ribociclib and trastuzumab or T-DM1 (NCT02657343) are also under way. A randomized phase III study of maintanence palbociclib with endocrine therapy, pertuzumab, and trastuzumab after chemotherapy in advanced ER+/HER2+ disease is also about to commence (PATINA: NCT02947685).

PI3K/mTOR Inhibitors

Activation of the PI3K pathway, including PIK3CA mutations and PTEN loss, has been noted preclinically to confer resistance to trastuzumab.82 The BOLERO1 and BOLERO2 3 trials investigated the addition of the mTOR inhibitor everolimus to trastuzumab and pacl*taxel in first-line therapy and trastuzumab and vinorelbine following trastuzumab resistance, respectively.83,84 A combined analysis of these studies found that tumors lacking PIK3CA mutations, PTEN loss, and PI3K pathway activation did not benefit from everolimus.85 A phase I study of buparlisib with trastuzumab in trastuzumab resistant advanced HER2+ breast cancer showed good tolerance and an ORR of 17% in 17 patients. Several early phase trials are in progress combining alpelisib (NCT02038010), taselisib (NCT02390427), or pictilisib (NCT00960960) with various combinations of trastuzumab, pertuzumab, and T-DM1.

Novel Antibody and Antibody Conjugates

Germline polymorphisms in CD16A, which encodes the activating Fc receptor FcγRIIIA, have been shown to affect clinical outcomes with trastuzumab.86 Margetuximab is a novel antibody that targets the same epitope as trastuzumab but has a modified Fc portion designed for enhanced affinity for FcγRIIIA receptor and reduced engagement with the inhibitory FcγR receptor.87 In a first-in-human phase I study of margetuximab monotherapy in HER2+ solid tumors, 19 evaluable patients with breast cancer who had received at least one prior anti-HER2 therapy showed an objective response rate of 26%.88 PFS for this group was 5.5 months. Toxicity was favorable, with the most common grade 3 and 4 adverse events being lymphopenia (17%), elevated lipase (8%), and anemia (3%). The currently recruiting phase III SOPHIA trial is comparing margetuximab plus physician’s choice chemotherapy to trastuzumab plus chemotherapy after previous treatment with pertuzumab, trastuzumab and T-DM1 (NCT02492711).89 Patritumab (U3-1287) is a fully human anti-HER3 monoclonal antibody. In a phase IB study of patritumab, trastuzumab,

NOVEL TARGETED AGENTS AND IMMUNOTHERAPY IN BREAST CANCER

and pacl*taxel in patients with metastatic HER2+ breast cancer previously treated with trastuzumab, the objective response rate was 38.9% (18 patients), with median PFS of 274 days.90 No dose-limiting toxicity was reached. The I-SPY-2 phase II neoadjuvant trial added a patritumab and trastuzumab arm in October 2016 (NCT01042379). Following the efficacy and favorable toxicity of T-DM1, a number of ADCs entered early-phase trials. These nextgeneration ADCs incorporate novel linker chemistry and aim to have larger amounts of payload drug attached to each antibody. DS-8201a consists of trastuzumab conjugated with a novel topoisomerase I inhibitor. It has shown efficacy in a T-DM1 resistant PDX, as well as PDXs with low HER2 expression, and is capable of substantial bystander cytotoxicity.91,92 In a phase I trial, no dose-limiting toxicities were seen, and the ORR in 12 patients with breast cancer previously treated with T-DM1 was 42%.93 The XMT-1522 ADC targets an HER2 epitope distinct from the trastuzumab epitope and is conjugated with the cytotoxic agent auristatin.94 It has the possible benefit of avoiding interference with trastuzumab and pertuzumab activity. In preclinical studies, it displayed similar levels of HER2 inhibition as trastuzumab and was effective in T-DM1-resistant and low-HER2-expressing models. It also showed synergistic activity when combined with trastuzumab and pertuzumab.94 A phase IB study is being conducted in both HER2 1–3+ and HER2-amplified advanced breast cancers, as well as other HER2-expressing tumor types (NCT02952729). MM-302 is an ADC of liposomal doxorubicin and a HER2 targeted antibody. Despite an efficacy signal in a phase I trial, the phase II HERMIONE trial of MM-302 plus trastuzumab versus physician’s choice chemotherapy plus trastuzumab was terminated early in December 2016 because of futility (NCT02213744).

Immunotherapy

Multiple lines of evidence support the importance of the immune microenvironment in HER2+ breast cancer and trastuzumab therapy.95-97 The phase IB/II PANACEA trial is evaluating the combination of the anti–PD-1 antibody pembrolizumab with trastuzumab in patients with metastatic HER2+ breast cancer that has progressed after at least one line of therapy (NCT02129556). The trial is including patients with both PD-L1–negative and PD-L1–positive tumors by immunohistochemistry. The randomized phase II KATE2 study is comparing T-DM1 plus the anti–PD-L1 antibody atezolizumab to T-DM1 plus placebo in patients with prior trastuzumab and taxane treatment (NCT02924883). A novel approach to skin metastases involves the application of the topical Toll-like receptor 7 agonist imiquimod directly to skin lesions. A single-arm phase II study of imiquimod and nab-pacl*taxel in patients with treatment refractory breast cancer chest wall metastases showed an ORR of 72% in 14 patients, some of whom were HER2+.98 T-cell cellular therapies are a promising therapeutic intervention for refractory malignancies. In a phase I study,

priming peripheral blood T-cells ex vivo with a HER2 vaccine before expansion in culture and reinfusion produced a response in 43% of patients (seven with HER2+ breast cancer, one with ovarian cancer).99 Chimeric antigen receptor (CAR) T cells have had dramatic success in hematologic malignancies, where they target cell surface receptors such as CD19. This strategy is attractive for HER2+ advanced breast cancer, as HER2 overexpression represents an accessible target for the CAR. Early use of HER2-targeted CAR T cells was associated with substantial toxicity,100 but more recently, improved CAR technology and optimized infusion protocols have been well tolerated and efficacious, as was seen in a phase I/II study of HER2 CAR T cells in HER2-expressing sarcomas.101 CAR T-cell therapy could be an option in advanced refractory HER2+ breast cancers. Vaccination with HER2-derived peptides has repeatedly shown that antigen-specific T-cell immunity can be induced. However, a strong efficacy signal has been lacking, and the benefit of vaccination against HER2 may be low in HER2 amplified tumors. The E75 HER2-derived peptide together with a GM-CSF as an adjuvant was tested in a phase I/II trial of 195 patients with early-stage breast cancer with a range of HER2+ expression. The nature of the vaccine requires that patients possess the HLA-A2 or A3 allele. In the optimally dosed group, 5-year disease-free survival was 94.6% compared with 80.2% in the control group (p = .05, log-rank test). With the E75 vaccine, the benefit of vaccination seems highest in tumors that are HER2 1+ or 2+.102 A similar result was seen with the AE37 vaccine, consisting of a hybrid HER2 peptide designed to enable direct loading onto HLA class 2 molecules.103 The E75 vaccine is being tested in the phase III PRESENT study, restricted to patients with HER2 1+ or 2+ expression. In the advanced disease setting, robust immune responses were noted with the combination of an anti-HER2 vaccine and trastuzumab.104 The combination of T-cell checkpoint inhibitors and vaccines may also be a viable strategy in the advanced disease setting.105

CONCLUSION AND FUTURE DIRECTIONS

In the past year, ER+ disease has seen the debut of the highly active and well-tolerated CDK4/6 inhibitors. CDK4/6 inhibitors are now being evaluated in the adjuvant setting. In TNBC, immunotherapy holds much promise, particularly in combination with other therapies, and the subgroups that define patients most likely to benefit from specific therapies such as PARP inhibition or AR inhibition are becoming increasingly well defined. For HER2+ disease, next-generation HER2 TKIs with improved toxicity and better CNS activity will be a key development if preliminary data are confirmed, and the first results of the immunotherapy studies in combination with HER2-targeted agents are expected soon for advanced disease. As a platform, ADCs are expected to continue to bear fruit in HER2+ disease, and further evidence of efficacy in TNBC is eagerly awaited. The lengthening list of efficacious novel therapies is cause for cautious optimism about the future of breast cancer treatment, and results of ongoing clinical trials are eagerly awaited. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 71

MAYER ET AL

References 1. Dowsett M, Nicholson RI, Pietras RJ. Biological characteristics of the pure antiestrogen fulvestrant: overcoming endocrine resistance. Breast Cancer Res Treat. 2005;93 (Suppl 1):S11-S18. 2. Turner NC, Ro J, André F, et al; PALOMA3 Study Group. Palbociclib in hormone-receptor-positive advanced breast cancer. N Engl J Med. 2015;373:209-219. 3. Fry DW, Harvey PJ, Keller PR, et al. Specific inhibition of cyclindependent kinase 4/6 by PD 0332991 and associated antitumor activity in human tumor xenografts. Mol Cancer Ther. 2004;3:1427-1438. 4. Schwartz GK, LoRusso PM, Dickson MA, et al. Phase I study of PD 0332991, a cyclin-dependent kinase inhibitor, administered in 3-week cycles (schedule 2/1). Br J Cancer. 2011;104:1862-1868. 5. Dean JL, Thangavel C, McClendon AK, et al. Therapeutic CDK4/6 inhibition in breast cancer: key mechanisms of response and failure. Oncogene. 2010;29:4018-4032. 6. Finn RS, Dering J, Conklin D, et al. PD 0332991, a selective cyclin D kinase 4/6 inhibitor, preferentially inhibits proliferation of luminal estrogen receptor-positive human breast cancer cell lines in vitro. Breast Cancer Res. 2009;11:R77. 7. Farmer P, Bonnefoi H, Becette V, et al. Identification of molecular apocrine breast tumours by microarray analysis. Oncogene. 2005;24:4660-4671. 8. Cristofanilli M, Turner NC, Bondarenko I, et al. Fulvestrant plus palbociclib versus fulvestrant plus placebo for treatment of hormonereceptor-positive, HER2-negative metastatic breast cancer that progressed on previous endocrine therapy (PALOMA-3): final analysis of the multicentre, double-blind, phase 3 randomised controlled trial. Lancet Oncol. 2016;17:425-439. 9. Baselga J, Campone M, Piccart M, et al. Everolimus in postmenopausal hormone-receptor-positive advanced breast cancer. N Engl J Med. 2012;366:520-529. 10. Bachelot T, Bourgier C, Cropet C, et al. Randomized phase II trial of everolimus in combination with tamoxifen in patients with hormone receptor-positive, human epidermal growth factor receptor 2-negative metastatic breast cancer with prior exposure to aromatase inhibitors: a GINECO study. J Clin Oncol. 2012;30:2718-2724. 11. Wolff AC, Lazar AA, Bondarenko I, et al. Randomized phase III placebo-controlled trial of letrozole plus oral temsirolimus as first-line endocrine therapy in postmenopausal women with locally advanced or metastatic breast cancer. J Clin Oncol. 2013;31:195-202. 12. Ma CX, Luo J, Naughton M, et al. A phase I trial of BKM120 (buparlisib) in combination with fulvestrant in postmenopausal women with estrogen receptor-positive metastatic breast cancer. Clin Cancer Res. 2016;22:1583-1591. 13. Mayer IA, Abramson VG, Isakoff SJ, et al. Stand up to cancer phase Ib study of pan-phosphoinositide-3-kinase inhibitor buparlisib with letrozole in estrogen receptor-positive/human epidermal growth factor receptor 2-negative metastatic breast cancer. J Clin Oncol. 2014;32:1202-1209. 14. Baselga J, Im S-A, Iwata H, et al. PIK3CA status in circulating tumor DNA (ctDNA) predicts efficacy of buparlisib (BUP) plus fulvestrant (FULV) in postmenopausal women with endocrine-resistant HR+/HER2– advanced breast cancer (BC): first results from the randomized, phase III BELLE-2 trial. Presented at: San Antonio Breast Cancer Symposium. San Antonio, TX; 2015. Abstract S6-S1.

72 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

15. Di Leo A, Seok Lee K, Ciruelos E, et al. BELLE-3: a phase III study of buparlisib + fulvestrant in postmenopausal women with HR+, HER2–, aromatase inhibitor-treated, locally advanced or metastatic breast cancer, who progressed on or after mTOR inhibitor-based treatment. Presented at: San Antonio Breast Cancer Symposium. San Antonio, TX; 2016. Abstract S4-S7. 16. Utermark T, Rao T, Cheng H, et al. The p110α and p110β isoforms of PI3K play divergent roles in mammary gland development and tumorigenesis. Genes Dev. 2012;26:1573-1586. 17. Zhao JJ, Cheng H, Jia S, et al. The p110alpha isoform of PI3K is essential for proper growth factor signaling and oncogenic transformation. Proc Natl Acad Sci U S A. 2006;103:16296-16300. 18. Foukas LC, Claret M, Pearce W, et al. Critical role for the p110α phosphoinositide-3-OH kinase in growth and metabolic regulation. Nature. 2006;441:366-370. 19. Krop IE, Mayer IA, Ganju V, et al. Pictilisib for oestrogen receptorpositive, aromatase inhibitor-resistant, advanced or metastatic breast cancer (FERGI): a randomised, double-blind, placebo-controlled, phase 2 trial. Lancet Oncol. 2016;17:811-821. 20. Furet P, Guagnano V, Fairhurst RA, et al. Discovery of NVP-BYL719 a potent and selective phosphatidylinositol-3 kinase alpha inhibitor selected for clinical evaluation. Bioorg Med Chem Lett. 2013;23:3741-3748. 21. Fritsch C, Huang A, Chatenay-Rivauday C, et al. Characterization of the novel and specific PI3Kα inhibitor NVP-BYL719 and development of the patient stratification strategy for clinical trials. Mol Cancer Ther. 2014;13:1117-1129. 22. Olivero AG, Heffron TP, Baumgardner M, et al. Abstract DDT02-01: discovery of GDC-0032: a beta-sparing PI3K inhibitor active against PIK3CA mutant tumors. Cancer Res. 2013; 73 (8, Supplement) DDT02-DDT01. 23. Jeselsohn R, Yelensky R, Buchwalter G, et al. Emergence of constitutively active estrogen receptor-α mutations in pretreated advanced estrogen receptor-positive breast cancer. Clin Cancer Res. 2014;20:1757-1767. 24. Merenbakh-Lamin K, Ben-Baruch N, Yeheskel A, et al. D538G mutation in estrogen receptor-α: A novel mechanism for acquired endocrine resistance in breast cancer. Cancer Res. 2013;73:6856-6864. 25. Toy W, Shen Y, Won H, et al. ESR1 ligand-binding domain mutations in hormone-resistant breast cancer. Nat Genet. 2013;45:1439-1445. 26. Robinson DR, Wu Y-M, Vats P, et al. Activating ESR1 mutations in hormone-resistant metastatic breast cancer. Nat Genet. 2013;45:14461451. 27. Cancer Genome Atlas Network. Comprehensive molecular portraits of human breast tumours. Nature. 2012;490:61-70. 28. Johnston SR, Kilburn LS, Ellis P, et al; SoFEA Investigators. Fulvestrant plus anastrozole or placebo versus exemestane alone after progression on non-steroidal aromatase inhibitors in postmenopausal patients with hormone-receptor-positive locally advanced or metastatic breast cancer (SoFEA): a composite, multicentre, phase 3 randomised trial. Lancet Oncol. 2013;14:989-998. 29. Fribbens C, O’Leary B, Kilburn L, et al. Plasma ESR1 mutations and the treatment of estrogen receptor-positive advanced breast cancer. J Clin Oncol. 2016;34:2961-2968. 30. Dent R, Trudeau M, Pritchard KI, et al. Triple-negative breast cancer: clinical features and patterns of recurrence. Clin Cancer Res. 2007;13:4429-4434.

NOVEL TARGETED AGENTS AND IMMUNOTHERAPY IN BREAST CANCER

31. Kast K, Link T, Friedrich K, et al. Impact of breast cancer subtypes and patterns of metastasis on outcome. Breast Cancer Res Treat. 2015;150:621-629.

48. Fong PC, Boss DS, Yap TA, et al. Inhibition of poly(ADP-ribose) polymerase in tumors from BRCA mutation carriers. N Engl J Med. 2009;361:123-134.

32. O’Shaughnessy J, Schwartzberg L, Danso MA, et al. Phase III study of iniparib plus gemcitabine and carboplatin versus gemcitabine and carboplatin in patients with metastatic triple-negative breast cancer. J Clin Oncol. 2014;32:3840-3847.

49. Tutt A, Robson M, Garber JE, et al. Oral poly(ADP-ribose) polymerase inhibitor olaparib in patients with BRCA1 or BRCA2 mutations and advanced breast cancer: a proof-of-concept trial. Lancet. 2010;376:235-244.

33. Twelves C, Cortes J, Vahdat L, et al. Efficacy of eribulin in women with metastatic breast cancer: a pooled analysis of two phase 3 studies. Breast Cancer Res Treat. 2014;148:553-561.

50. Kaufman B, Shapira-Frommer R, Schmutzler RK, et al. Olaparib monotherapy in patients with advanced cancer and a germline BRCA1/2 mutation. J Clin Oncol. 2015;33:244-250.

34. Penault-Llorca F, Viale G. Pathological and molecular diagnosis of triple-negative breast cancer: a clinical perspective. Ann Oncol. 2012;23 (Suppl 6):vi19-vi22.

51. Loi S, Sirtaine N, Piette F, et al. Prognostic and predictive value of tumorinfiltrating lymphocytes in a phase III randomized adjuvant breast cancer trial in node-positive breast cancer comparing the addition of docetaxel to doxorubicin with doxorubicin-based chemotherapy: BIG 02-98. J Clin Oncol. 2013;31:860-867.

35. Lehmann BD, Bauer JA, Chen X, et al. Identification of human triplenegative breast cancer subtypes and preclinical models for selection of targeted therapies. J Clin Invest. 2011;121:2750-2767. 36. Lehmann BD, Jovanović B, Chen X, et al. Refinement of triple-negative breast cancer molecular subtypes: implications for neoadjuvant chemotherapy selection. PLoS One. 2016;11:e0157368. 37. Burstein MD, Tsimelzon A, Poage GM, et al. Comprehensive genomic analysis identifies novel subtypes and targets of triple-negative breast cancer. Clin Cancer Res. 2015;21:1688-1698. 38. Perou CM, Sørlie T, Eisen MB, et al. Molecular portraits of human breast tumours. Nature. 2000;406:747-752. 39. Carey LA, Perou CM, Livasy CA, et al. Race, breast cancer subtypes, and survival in the Carolina Breast Cancer Study. JAMA. 2006;295:24922502.

52. Mittendorf EA, Philips AV, Meric-Bernstam F, et al. PD-L1 expression in triple-negative breast cancer. Cancer Immunol Res. 2014;2:361-370. 53. Nanda R, Chow LQM, Dees EC, et al. Pembrolizumab in patients with advanced triple-negative breast cancer: phase Ib KEYNOTE-012 study. J Clin Oncol. 2016;34:2460-2467. 54. Emens LA, Braiteh FS, Cassier P, et al. Abstract 2859: inhibition of PDL1 by MPDL3280A leads to clinical activity in patients with metastatic triple-negative breast cancer (TNBC). Cancer Res. 2015;75 (suppl; abstr 2859). 55. Zitvogel L, Apetoh L, Ghiringhelli F, et al. Immunological aspects of cancer chemotherapy. Nat Rev Immunol. 2008;8:59-73.

40. Nik-Zainal S, Davies H, Staaf J, et al. Landscape of somatic mutations in 560 breast cancer whole-genome sequences. Nature. 2016;534:4754.

56. Adams S, Diamond JR, Hamilton EP, et al Phase Ib trial of atezolizumab in combination with nab-pacl*taxel in patients with metastatic triplenegative breast cancer (mTNBC). J Clin Oncol. 2016;34 (suppl; abstr 1009).

41. Carey LA, Dees EC, Sawyer L, et al. The triple negative paradox: primary tumor chemosensitivity of breast cancer subtypes. Clin Cancer Res. 2007;13:2329-2334.

57. Doane AS, Danso M, Lal P, et al. An estrogen receptor-negative breast cancer subset characterized by a hormonally regulated transcriptional program and response to androgen. Oncogene. 2006;25:3994-4008.

42. Yardley D, Coleman R, Conte P, et al. nab-pacl*taxel + carboplatin or gemcitabine vs gemcitabine/carboplatin as first-line treatment for patients with triple-negative metastatic breast cancer: Results from the randomized phase 2 portion of the tnAcity trial. Presented at: San Antonio Breast Cancer Symposium. San Antonio, TX; 2016. Abstract P5-15-03.

58. Bonnefoi H, Grellety T, Tredan O, et al. A phase II trial of abiraterone acetate plus prednisone in patients with triple-negative androgen receptor positive locally advanced or metastatic breast cancer (UCBG 12-1). Ann Oncol. 2016;27:812-818.

43. Tutt A, Ellis P, Kilburn L, et al. The TNT trial: a randomized phase III trial of carboplatin (C) compared with docetaxel (D) for patients with metastatic or recurrent locally advanced triple negative or BRCA1/2 breast cancer (CRUK/07/012). Presented at: San Antonio Breast Cancer Symposium. San Antonio, TX; 2014. Abstract S3-S1.

59. Traina TA, Miller K, Yardley D, et al Results from a phase 2 study of enzalutamide (ENZA), an androgen receptor (AR) inhibitor, in advanced AR+ triple-negative breast cancer (TNBC). J Clin Oncol. 2015;33 (suppl; abstr 1003).

44. Sonnenblick A, de Azambuja E, Azim HA Jr, et al. An update on PARP inhibitors—moving to the adjuvant setting. Nat Rev Clin Oncol. 2015;12:27-41.

60. Dent R, Schmid P, Cortes J, et al. ENDEAR: A randomized international phase 3 study comparing the efficacy and safety of enzalutamide in combination with pacl*taxel chemotherapy or as monotherapy vs placebo with pacl*taxel in patients with advanced diagnostic-positive triple-negative breast cancer. Presented at: San Antonio Breast Cancer Symposium. San Antonio, TX; 2016. Abstract OT3-02-02.

45. Sorlie T, Tibshirani R, Parker J, et al. Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci U S A. 2003;100:8418-8423.

61. Goldenberg DM, Cardillo TM, Govindan SV, et al. Trop-2 is a novel target for solid cancer therapy with sacituzumab govitecan (IMMU-132), an antibody-drug conjugate (ADC). Oncotarget. 2015;6:22496-22512.

46. Antonia S, Goldberg SB, Balmanoukian A, et al. Safety and antitumour activity of durvalumab plus tremelimumab in non-small cell lung cancer: a multicentre, phase 1b study. Lancet Oncol. 2016;17:299308.

62. Starodub AN, Ocean AJ, Shah MA, et al. First-in-human trial of a novel anti-Trop-2 antibody-SN-38 conjugate, sacituzumab govitecan, for the treatment of diverse metastatic solid tumors. Clin Cancer Res. 2015;21:3870-3878.

47. Kaelin WG Jr. The concept of synthetic lethality in the context of anticancer therapy. Nat Rev Cancer. 2005;5:689-698.

63. Bardia A, Diamond J, Mayer I, et al. Abstract PD3-06: safety and efficacy of anti-Trop-2 antibody drug conjugate, sacituzumab govitecan

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 73

MAYER ET AL

(IMMU-132), in heavily pretreated patients with TNBC. Cancer Res. 2016;76 (4, Supplement):PD3-06. 64. Rose AAN, Grosset A-A, Dong Z, et al. Glycoprotein nonmetastatic B is an independent prognostic indicator of recurrence and a novel therapeutic target in breast cancer. Clin Cancer Res. 2010;16:21472156.

78. Murthy RK, Hamilton E, Borges VF, et al. ONT-380 in the treatment of HER2+ breast cancer central nervous system (CNS) metastases (mets). Presented at: San Antonio Breast Cancer Symposium. San Antonio, TX; 2015. Abstract P4-14-9.

65. Yardley DA, Weaver R, Melisko ME, et al. EMERGE: a randomized phase II study of the antibody-drug conjugate glembatumumab vedotin in advanced glycoprotein NMB-expressing breast cancer. J Clin Oncol. 2015;33:1609-1619.

79. Hamilton EP, Borges VF, Murthy RK, et al. A phase 2 randomized, double-blinded, controlled study of ONT-380 vs. placebo in combination with capecitabine (C) and trastuzumab (T) in patients with pretreated HER2+ unresectable locally advanced or metastatic breast carcinoma (MBC) (HER2CLIMB). Presented at: San Antonio Breast Cancer Symposium. San Antonio, TX; 2016. Abstract OT1-02-09.

66. Marty B, Maire V, Gravier E, et al. Frequent PTEN genomic alterations and activated phosphatidylinositol 3-kinase pathway in basal-like breast cancer cells. Breast Cancer Res. 2008;10:R101.

80. Goel S, Wang Q, Watt AC, et al. Overcoming therapeutic resistance in HER2-positive breast cancers with CDK4/6 inhibitors. Cancer Cell. 2016;29:255-269.

67. Banerji S, Cibulskis K, Rangel-Escareno C, et al. Sequence analysis of mutations and translocations across breast cancer subtypes. Nature. 2012;486:405-409.

81. Tolaney SM, Bourayou N, Goel S, et al. monarcHER: a phase 2 randomized open-label study of abemaciclib plus trastuzumab (T) with or without fulvestrant (F) compared to standard-of-care chemotherapy of physician’s choice plus T in women with HR+, HER2+ advanced breast cancer. Ann Oncol. 2016;27 (suppl_6).

68. Montero JC, Esparís-Ogando A, Re-Louhau MF, et al. Active kinase profiling, genetic and pharmacological data define mTOR as an important common target in triple-negative breast cancer. Oncogene. 2014;33:148-156. 69. Tripathy D, Chien AJ, Hylton NM, et al. Adaptively randomized trial of neoadjuvant chemotherapy with or without the Akt inhibitor MK2206: graduation results from the I-SPY 2 trial. J Clin Oncol. 2015;33 (suppl; abstr 524). 70. Lehmann BD, Pietenpol JA. Identification and use of biomarkers in treatment strategies for triple-negative breast cancer subtypes. J Pathol. 2014;232:142-150. 71. Pheneger T, Bouhana K, Anderson D, et al. In vitro and in vivo activity of ARRY-380: a potent, small molecule inhibitor of ErbB2. Cancer Res. 2014;69 (9 Supplement):1795. 72. Dinkel V, Anderson D, Winski S, et al. Abstract 852: ARRY-380, a potent, small molecule inhibitor of ErbB2, increases survival in intracranial ErbB2+ xenograft models in mice. Cancer Res. 2012;72(8, Supplement)852. 73. Moulder-Thompson S, Borges VF, Baetz TD, et al. Phase 1 study of ONT-380, a HER2 inhibitor, in patients with HER2+ advanced solid tumors, with an expansion cohort in HER2+ metastatic breast cancer (MBC). Clin Cancer Res. Epub 4 Jan 2017. 74. Hamilton E, Borges V, Conlin A, et al. Efficacy results of a phase 1b study of ONT-380, an oral HER2-specific inhibitor, in combination with capecitabine (C) and trastuzumab (T) in HER2+ metastatic breast cancer (MBC), including patients (pts) with brain metastases (mets). Presented at: San Antonio Breast Cancer Symposium. San Antonio, TX; 2016. Abstract P4-21-01. 75. Borges VF, Ferrario C, Aucoin N, et al. Efficacy results of a phase 1b study of ont-380, a CNS-penetrant TKI, in combination with T-DM1 in HER2+ metastatic breast cancer (MBC), including patients (pts) with brain metastases. J Clin Oncol. 2016;34 (suppl; abstr 513). 76. Cortés J, Dieras V, Ro J, et al. Afatinib alone or afatinib plus vinorelbine versus investigator’s choice of treatment for HER2-positive breast cancer with progressive brain metastases after trastuzumab, lapatinib, or both (LUX-Breast 3): a randomised, open-label, multicentre, phase 2 trial. Lancet Oncol. 2015;16:1700-1710. 77. Freedman RA, Gelman RS, Wefel JS, et al. Translational Breast Cancer Research Consortium (TBCRC) 022: a phase II trial of neratinib for patients with human epidermal growth factor receptor 2-positive breast cancer and brain metastases. J Clin Oncol. 2016;34:945-952.

74 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

82. Berns K, Horlings HM, Hennessy BT, et al. A functional genetic approach identifies the PI3K pathway as a major determinant of trastuzumab resistance in breast cancer. Cancer Cell. 2007;12:395-402. 83. Hurvitz SA, Andre F, Jiang Z, et al. Combination of everolimus with trastuzumab plus pacl*taxel as first-line treatment for patients with HER2-positive advanced breast cancer (BOLERO-1): a phase 3, randomised, double-blind, multicentre trial. Lancet Oncol. 2015;16:816-829. 84. André F, O’Regan R, Ozguroglu M, et al. Everolimus for women with trastuzumab-resistant, HER2-positive, advanced breast cancer (BOLERO-3): a randomised, double-blind, placebo-controlled phase 3 trial. Lancet Oncol. 2014;15:580-591. 85. André F, Hurvitz S, Fasolo A, et al. Molecular alterations and everolimus efficacy in human epidermal growth factor receptor 2-overexpressing metastatic breast cancers: combined exploratory biomarker analysis from BOLERO-1 and BOLERO-3. J Clin Oncol. 2016;34:2115-2124. 86. Musolino A, Naldi N, Bortesi B, et al. Immunoglobulin G fragment C receptor polymorphisms and clinical efficacy of trastuzumab-based therapy in patients with HER-2/neu-positive metastatic breast cancer. J Clin Oncol. 2008;26:1789-1796. 87. Nordstrom JL, Gorlatov S, Zhang W, et al. Anti-tumor activity and toxico*kinetics analysis of MGAH22, an anti-HER2 monoclonal antibody with enhanced Fcγ receptor binding properties. Breast Cancer Res. 2011;13:R123. 88. Burris H, Giaccone G, Im S-A, et al. Updated findings of a first-inhuman, phase I study of margetuximab (M), an Fc-optimized chimeric monoclonal antibody (MAb), in patients (pts) with HER2-positive advanced solid tumors. J Clin Oncol. 2015;33 (suppl; abstr 523). 89. Rugo HS, Pegram MD, Gradishar WJ, et al SOPHIA: a phase 3, randomized study of margetuximab (M) plus chemotherapy (CTX) vs trastuzumab (T) plus CTX in the treatment of patients with HER2+ metastatic breast cancer (MBC). J Clin Oncol. 2016;34 (suppl; abstr TPS630). 90. Mukai H, Saeki T, Aogi K, et al. Patritumab plus trastuzumab and pacl*taxel in human epidermal growth factor receptor 2-overexpressing metastatic breast cancer. Cancer Sci. 2016;107:1465-1470. 91. Ogitani Y, Aida T, Hagihara K, et al. DS-8201a, a novel HER2-targeting ADC with a novel DNA topoisomerase I inhibitor, demonstrates a

NOVEL TARGETED AGENTS AND IMMUNOTHERAPY IN BREAST CANCER

promising antitumor efficacy with differentiation from T-DM1. Clin Cancer Res. 2016;22:5097-5108. 92. Ogitani Y, Hagihara K, Oitate M, et al. Bystander killing effect of DS8201a, a novel anti-human epidermal growth factor receptor 2 antibody-drug conjugate, in tumors with human epidermal growth factor receptor 2 heterogeneity. Cancer Sci. 2016;107:1039-1046. 93. Tamura K, sh*tara K, Naito Y, et al Single agent activity of DS-8201a, a HER2-targeting antibody-drug conjugate, in breast cancer patients previously treated with T-DM1: phase 1 dose escalation. Ann Oncol. 2016;27 (suppl_6):LBA17. 94. Bergstrom DA, Bodyak N, Yurkovetskiy A, et al. Abstract LB-231: a novel, highly potent HER2-targeted antibody-drug conjugate (ADC) for the treatment of low HER2-expressing tumors and combination with trastuzumab-based regimens in HER2-driven tumors. Cancer Res. 2015;75 (suppl; abstr LB-231). 95. Stagg J, Loi S, Divisekera U, et al. Anti-ErbB-2 mAb therapy requires type I and II interferons and synergizes with anti-PD-1 or anti-CD137 mAb therapy. Proc Natl Acad Sci U S A. 2011;108:7142-7147. 96. Loi S, Michiels S, Salgado R, et al. Tumor infiltrating lymphocytes are prognostic in triple negative breast cancer and predictive for trastuzumab benefit in early breast cancer: results from the FinHER trial. Ann Oncol. 2014;25:1544-1550. 97. Salgado R, Denkert C, Campbell C, et al. Tumor-infiltrating lymphocytes and associations with pathological complete response and eventfree survival in HER2-positive early-stage breast cancer treated with lapatinib and trastuzumab: a secondary analysis of the NeoALTTO trial. JAMA Oncol. 2015;1:448-454.

98. Salazar LG, Lu H, Reichow JL, et al. Topical imiquimod plus nabpacl*taxel for breast cancer cutaneous metastases: a phase 2 clinical trial. JAMA Oncol. Epub 19 Jan 2017. 99. Disis ML, Dang Y, Coveler AL, et al. HER-2/neu vaccine-primed autologous T-cell infusions for the treatment of advanced stage HER-2/ neu expressing cancers. Cancer Immunol Immunother. 2014;63:101-109. 100. Morgan RA, Yang JC, Kitano M, et al. Case report of a serious adverse event following the administration of T cells transduced with a chimeric antigen receptor recognizing ERBB2. Mol Ther. 2010;18:843-851. 101. Ahmed N, Brawley VS, Hegde M, et al. Human epidermal growth factor receptor 2 (HER2)-specific chimeric antigen receptor-modified T Cells for the immunotherapy of HER2-positive sarcoma. J Clin Oncol. 2015;33:1688-1696. 102. Benavides LC, Gates JD, Carmichael MG, et al. The impact of HER2/neu expression level on response to the E75 vaccine: from U.S. Military Cancer Institute Clinical Trials Group Study I-01 and I-02. Clin Cancer Res. 2009;15:2895-2904. 103. Mittendorf EA, Ardavanis A, Symanowski J, et al. Primary analysis of a prospective, randomized, single-blinded phase II trial evaluating the HER2 peptide AE37 vaccine in breast cancer patients to prevent recurrence. Ann Oncol. 2016;27:1241-1248. 104. Disis ML, Wallace DR, Gooley TA, et al. Concurrent trastuzumab and HER2/neu-specific vaccination in patients with metastatic breast cancer. J Clin Oncol. 2009;27:4685-4692. 105. Moynihan KD, Opel CF, Szeto GL, et al. Eradication of large established tumors in mice by combination immunotherapy that engages innate and adaptive immune responses. Nat Med. 2016;22:1402-1410.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 75

HURVITZ, GELMON, AND TOLANEY

Optimal Management of Early and Advanced HER2 Breast Cancer Sara A. Hurvitz, MD, FACP, Karen A. Gelmon, MD, FRCPC, and Sara M. Tolaney, MD, MPH OVERVIEW Approximately 15%–20% of breast cancer is HER2 positive, and patients with this subtype of disease historically had worse outcomes than patients with HER2-negative disease. However, the introduction of HER2-directed therapies has dramatically altered outcomes for these patients, especially for persons with early disease. However, despite these achievements, metastatic disease is still not curable. This review summarizes the current treatment approach for patients in the preoperative and adjuvant setting, including data regarding selecting the optimal chemotherapy partner as well as determining the duration and type of anti-HER–directed therapy. This article also reviews how to approach patients with advanced HER2-positive disease and discusses promising new therapies that are in development.

he initial studies of HER2-positive breast cancer focused on advanced disease, in which a novel monoclonal antibody directed against HER2 (which soon came to be known as trastuzumab) was shown to have single-agent activity in tumors that overexpressed HER2.1,2 After the initial discovery of activity, the pivotal study by Slamon et al3 showed benefit in terms of progression-free survival (PFS) and overall survival (OS) for both the combination of doxorubicin and cyclophosphamide with the antibody as well as the couplet of pacl*taxel and trastuzumab. Although the anthracycline combination was more active, the cardiotoxicity that was reported led to the approval in 1998 of trastuzumab and pacl*taxel as standard therapy for HER2-positive advanced breast cancer.4

WHERE ARE WE WITH THE TREATMENT OF ADVANCED HER2-POSITIVE BREAST CANCER?

The years after the 1998 approval of trastuzumab were dominated by studies examining the combination of trastuzumab with almost every known cytotoxic, exploring the prolonged administration of the antibody past progression and examining different dosing schedules (including the three-weekly scheduling that has become widely used).5-7 In addition, many studies examined the optimal way to define HER2 in the laboratory, with the recognition that the best results were seen in those tumors that overexpressed HER2.8 Although many individuals with true HER2-positive advanced disease respond to treatment, the majority of patients develop a resistance and disease progression, which has led to the pursuit of additional anti-HER2 agents

or combinations to improve outcomes. Studies have tried to exploit other pathways in addition to HER2 to improve outcomes and to develop novel agents. This section briefly summarizes the work since trastuzumab became the standard of care, concentrating on those strategies that have influenced guidelines.

Targeting HER2 With Other Agents

Tyrosine kinase inhibitors. There were theoretical reasons to look at small molecules to target HER2, with the idea that they may have both mechanistic and practical advantages over a large antibody. Many of the small-molecule tyrosine kinase inhibitors (TKIs) had less specificity than trastuzumab, which was potentially of value in a tumor with heterogeneity. In addition, small molecules could potentially cross the blood-brain barrier, be given on a continuous schedule, be orally available, and possibly have less cardiac toxicity than the approved drug, trastuzumab. The first widely tested agent was lapatinib, which reversibly binds to and inhibits the intracellular domain of HER1 and HER2 and was shown to have both single-agent activity and be able to be combined with cytotoxics, including capecitabine and pacl*taxel. The combination of capecitabine and lapatinib was compared with capecitabine alone, showing an improved time to progression with a hazard ratio (HR) of 0.49 (95% CI, 0.34–0.71; p < .001) but no OS benefit; this led to its approval in the second-line setting.9,10 In addition, combined antiHER2 therapy with trastuzumab and lapatinib showed activity in multiple pretreated patients with advanced breast cancer compared with lapatinib alone, with improved PFS

From the David Geffen School of Medicine, University of California, Los Angeles, CA; BC Cancer Agency, University of British Columbia, Vancouver, BC, Canada; Dana-Farber Cancer Institute, Boston, MA. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Karen A. Gelmon, MD, FRCPC, BC Cancer Agency, University of British Columbia, 600 West 10th Ave., Vancouver, BC V5Z4E6, Canada; email: kgelmon@ bccancer.bc.ca. © 2017 American Society of Clinical Oncology

76 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

OPTIMAL MANAGEMENT OF HER2 BREAST CANCER

(from 8 weeks to 11 weeks with the combination; HR, 0.74; 95% CI, 0.58–0.94; p = .011) and OS (from 10 months to 14 months; HR, 0.74; 95% CI, 0.57–0.97; p = .026).11 In a head-tohead comparison in the first-line setting in combination with taxanes (MA31), lapatinib was shown to be inferior to trastuzumab. This trial demonstrated that trastuzumab in combination with a taxane had significantly longer intention-to-treat PFS of 11.3 months compared with lapatinib combined with a taxane of 9.0 months (HR, 1.37; 95% CI, 1.13–1.65; p < .001) and more toxicity, in terms of diarrhea and rash, was observed with lapatinib compared with trastuzumab combined with taxane (p < .001).12 Although initial data suggested a benefit in brain metastases, this has not been clearly shown with specific studies such as CEREBEL or compared with other agents as in the EMILIA trial.13-15 The toxicity of the drug with diarrhea (up to 60% any grade) and rash (up to 27% any grade), as well as the efficacy of newer agents such as trastuzumab emtansine (T-DM1), led to the changing role of lapatinib from the second-line setting to later lines. Other small TKIs may have additional activity. Initial studies of neratinib, an irreversible pan-HER TKI of HER1, HER2, and HER4, suggest that it may be more active than lapatinib with single-agent activity. Burstein et al16 reported a median PFS of 22.3 and 39.6 weeks, respectively, among patients previously treated with trastuzumab (66 patients) and those that were treatment naive (70 patients), with objective response rates of 24% for patients who received prior trastuzumab treatment and 56% for the trastuzumab-naive cohort. Diarrhea, which is the major toxicity with neratinib administration, must be controlled early to derive the benefit from this drug and maintain dosing. The NEfERT trial was an open-label randomized study in first-line metastatic disease comparing neratinib and pacl*taxel to trastuzumab plus pacl*taxel. Median PFS was 12.9 months with neratinib/pacl*taxel and 12.9 months with trastuzumab/pacl*taxel (HR, 1.02; 95% CI, 0.81–1.27; p = .89).17 With neratinib/

KEY POINTS • The advent of anti-HER2 therapies have transformed the prognosis of HER2-overexpressing breast cancer. • Dual targeting of HER2 with the antibodies pertuzumab and trastuzumab has been a successful strategy in both early and metastatic breast cancer. • In the the metastatic setting, TDM-1, a novel antibodydrug conjugate, has become standard second-line therapy. • In low-risk, early-stage breast cancer, modified protocols with limited chemotherapy are effective, although 12 months of anti-HER2 treatment remains the standard. • Current neoadjuvant therapy results in higher pCR rates but also highlights the differences between estrogen receptor–positive, HER2-positive breast cancers and those that are HER2-positive but estrogen receptor negative.

pacl*taxel, the incidence of central nervous system recurrences was lower (relative risk, 0.48; 95% CI, 0.29–0.79; p = .002) and time to central nervous system metastases was delayed (HR, 0.45; 95% CI, 0.26–0.78; p = .004). Common grade 3/4 adverse events were diarrhea (30.4% with neratinib/pacl*taxel, 3.8% with trastuzumab/pacl*taxel), neutropenia (12.9% vs. 14.5%), and leukopenia (7.9% vs. 10.7%); no grade 4 diarrhea was observed. The NALA phase III study is comparing neratinib in combination with capecitabine to capecitabine plus lapatinib for patients with HER2-positive metastatic breast cancer who have received two or more prior HER2-directed regimens (NCT01808573). ONT-380 is a reversible TKI with selective HER2 inhibition with antitumor activity, and it is suggested to improve activity in brain metastases.18,19 This agent is being combined with trastuzumab T-DM1 as well as capecitabine or both; further data are pending to further delineate this agent’s potential benefit. Antibodies. Although the initial development was rather slow, pertuzumab has now established its role in the firstline advanced setting. Activity was seen in phase II studies after preclinical studies showed the impact of blocking HER2 and HER3, which led to the CLEOPATRA study. CLEOPATRA randomly assigned patients with newly diagnosed advanced HER2-positive breast cancer to docetaxel and trastuzumab with or without the addition of intravenous pertuzumab.20-22 This international phase III study showed a very powerful PFS of 18.5 months compared with 12.4 months in the trastuzumab arm and an OS benefit of 56.5 months in the trastuzumab/pertuzumab arm compared with 40.8 months in the trastuzumab cohort. Although the majority of patients in this study had de novo metastatic disease, the small number with prior exposure to trastuzumab in the adjuvant setting also responded. Studies have also shown benefit for pertuzumab with pacl*taxel and vinorelbine, providing additional options for combination treatment and leading to its widespread use. This drug is well tolerated, causing only minimal increases in diarrhea for many patients and no notable additional cardiotoxicity. To date, there are no data to suggest a continued benefit of pertuzumab past progression. T-DM1 is an antibody-drug conjugate linking trastuzumab with the powerful agent emtansine, a derivative of maytansine, an older microtubule cytotoxic initially evaluated in the 1970s.23,24 A large phase III study, EMILIA, was launched in the second-line metastatic setting comparing single-agent T-DM1 to capecitabine and lapatinib for patients with prior trastuzumab and taxane exposure.15 This study showed a statistically and clinically relevant benefit in PFS, response, duration of response, and OS for T-DM1, with a median 3.2-month PFS benefit (HR, 0.65; 95% CI, 0.55–0.77; p < .001) and a median 5.8-month OS benefit (HR, 0.68; 95% CI, 0.55–0.85; p < .001) compared with capecitabine plus lapatinib. The drug is very well tolerated, with the major side effects being thrombocytopenia (seen for 12.9% of patients) and occasional increases in liver enzymes (occurring for 2.9% and 4.3% of patients, for aspartate aminotransferase asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 77

HURVITZ, GELMON, AND TOLANEY

and alanine aminotransferase, respectively). An analysis of the EMILIA study showed that dose decreases may be associated with decreased efficacy, suggesting a rather narrow therapeutic window.25 Of interest is that despite being a large antibody complex, there is activity against brain metastases seen in the EMILIA study and a number of other reports. The TH3RESA study reported efficacy of T-DM1 for a heavily pretreated population, confirming the role of T-DM1 in advanced HER2-positive cancers. This randomized phase III study compared T-DM1 to physician’s choice therapy and demonstrated an improved PFS of 6.2 months for T-DM1 compared with 3.3 months for physician’s choice therapy (p < .001).26 In the first-line setting, a phase II study comparing T-DM1 to trastuzumab and docetaxel showed improvements in PFS of 14.2 months for T-DM1 compared with 9.2 months for trastuzumab, with an HR of 0.59 (95% CI, 0.36–0.97; p = .035).27 MARIANNE was a three-arm randomized phase III study comparing trastuzumab plus a taxane (either pacl*taxel or docetaxel) to single-agent T-DM1 or to the doublet T-DM1 and pertuzumab; it was hoped that this would to lead to the first-line approval for this well-tolerated drug.28 Although expectations favored the experimental arms, there was disappointment for many when the addition of pertuzumab to T-DM1 did not improve the PFS of the monotherapy T-DM1. In addition, there was a noninferior PFS outcome for the two T-DM1 arms compared with the trastuzumab plus taxane arm. Fewer notable adverse events and better quality-of-life outcomes were seen for the two experimental T-DM1–containing arms. There was not a trastuzumab/ pertuzumab arm similar to the CLEOPATRA cohort in the MARIANNE study. This trastuzumab/pertuzumab triplet remains the standard first-line treatment, except for selected patients who may not tolerate a taxane. MM-302 is a new experimental agent comprising a HER2-targeted nanoparticle containing doxorubicin, a cytotoxic with well-known anti-HER activity.28 MM-302 has been studied alone, in combination with trastuzumab, and in combination with cyclophosphamide and has demonstrated safety and preliminary efficacy. A phase II trial (HERMIONE) in anthracycline-naive advanced disease with prior progression on pertuzumab and T-DM1 that randomly assigned patients to receive MM-302 plus trastuzumab compared with physician’s choice chemotherapy plus trastuzumab was recently closed early because an unfavorable futility analysis (NCT02213744).29 Other new agents. Ertumaxomab, a bispecific antibody targeting HER2 and cluster of differentiation-3 with selective binding to activatory Fcγ-type I/III receptors, has been shown to elicit an immune response and antitumor activity and is now in an open-label dose-escalating study of patients with HER2-expressing advanced solid tumors (NCT01569412).30,31 In addition, there are a number of studies of trastuzumab biosimilars being reported. These agents may have the role of providing choice for patients if these drugs are truly more advantageous economically. 78 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Targeting of two pathways. Laboratory studies suggested that angiogenic pathways were active in HER2-positive cancers, leading to the idea of targeting both VEGF and HER2.32 Many studies were initiated; in the advanced setting, the phase III trial of bevacizumab and trastuzumab in addition to docetaxel (AVEREL) did not show a statistically notable benefit for dual targeting of 424 patients who were randomly assigned to treatment, leading to a general abandonment of this combination strategy.33 Other early strategies included combining heat shock protein agents with trastuzumab.34 Although responses were seen, these have not led to large phase III studies or changes in guidelines. With approximately half of HER2-overexpressing cancers also being estrogen receptor (ER)–positive, endocrine and anti-HER2 agents are an obvious combination in the advanced setting. An initial phase II study was done with trastuzumab and letrozole, with a 3.3-month PFS for the endocrine therapy alone and a 14.1-month PFS for the anti-HER2 therapy and endocrine therapy.35 The phase III randomized TANDEM study of anastrozole and trastuzumab showed improvements in PFS with the addition of anti-HER2 therapy (from 2.4 months for anastrozole vs. 4.8 months for the combined anastrozole and trastuzumab).36 The combination of lapatinib plus letrozole was studied in another large phase III trial. In the confirmed HER2-positive population, the median PFS rose from 3 months for the letrozole arm compared with 8.2 months for the lapatinib and letrozole arm, which led to the regulatory approval of this combination.37 However, the results of both of these studies were inferior to those of trastuzumab with chemotherapy for this patient population, leading to a limited uptake of this strategy as upfront treatment of most patients with ER-positive, HER2-positive advanced breast cancer. This is an option for patients with low-burden disease or comorbidities. More recently, the phase II PERTAIN trial (NCT01491737) enrolled 258 postmenopausal women with HER2-positive, hormone receptor–positive, metastatic, or locally advanced breast cancer to receive first-line pertuzumab plus trastuzumab and an aromatase inhibitor (anastrozole or letrozole), or trastuzumab plus an aromatase inhibitor. Preliminary results showed that adding pertuzumab to trastuzumab and an aromatase inhibitor significantly reduced the risk of progression or death by 35% versus treatment with trastuzumab and an aromatase inhibitor alone (HR, 0.65; 95% CI, 0.48–0.89; p = .0070). The median duration of response was 27.1 months and 15.1 months in the pertuzumab arm and the trastuzumab-only arm, respectively (HR, 0.57; 95% CI, 0.36–0.91; p = .02). Knowing that the phosphoinositide 3 kinase/protein kinase B (AKT)/mechanistic target of rapamycin (mTOR) pathway has been implicated in resistance and that activating mutations in PI3KCA or PTEN (phosphatase and tensin hom*olog) are seen in a large number of HER2-positive metastatic tumors, initial phase II studies combined everolimus with trastuzumab and either vinorelbine or pacl*taxel to target mTOR concurrently with HER2.38,39 There are now reports from two phase III studies that randomly assigned

OPTIMAL MANAGEMENT OF HER2 BREAST CANCER

patients with advanced HER2-positive breast cancer to receive either pacl*taxel plus trastuzumab (BOLERO I) or vinorelbine plus trastuzumab (BOLERO III), each with an experimental arm adding oral everolimus. In BOLERO I, which was in the first-line setting, there was no improvement in PFS, although an analysis of the HER2-positive/ER-negative cohort suggested an improved PFS from 13.1 months to 20.3 months.40 In BOLERO III, which was done in the later setting for patients who had progressed with prior trastuzumab, there was a statistically significant improvement in PFS from 5.78 months to 7.0 months (HR, 0.78; 95% CI, 0.65–0.95; p = .0067).41 There was considerable toxicity seen with the addition of everolimus and no improvement in OS was reported at this time. This lack of clinically relevant benefit coupled with toxicity has not led to changes in standard clinical practice although some individual patients may benefit. Correlative work suggested that patients with low PTEN concentrations may preferentially respond, but this needs further evaluation prior to being considered a predictive marker. Newer and more specific phosphoinositide 3 kinase inhibitors are in clinical trials in the advanced HER2-positive setting, including alpelisib, taselisib, and pictilisib. A phase II study of alpelisib, an alpha-specific inhibitor, has shown tolerability and responses among patients who had previously progressed during or following treatment with T-DM1.42 More recently, a number of studies have been initiated with checkpoint inhibitors and either trastuzumab or T-DM1 in advanced HER2-positive breast cancer. Preclinical laboratory work, as well as the demonstration of tumor-infiltrating lymphocytes (TILs) in HER2-positive breast cancer, has led to excitement in this area and a number of studies. These include the phase Ib/II PANACEA trial, which is combining the anti–PD-1 inhibitor pembrolizumab (MK-3475) with trastuzumab (NCT02129556). The Canadian Clinical Trials Group is combining durvalumab with trastuzumab in multiply pretreated patients with HER2-positive metastatic breast cancer to assess toxicity and biologic activity (NCT0264968). In addition, defining which cancers may respond to these costly therapies will be important. Vaccines are also being studied and initial reports have shown immune responses. Finally, an area of new interest is the combination of CDK4/6 inhibitors and anti-HER2 agents, with the preclinical evidence of activity in this subtype suggesting in transgenic mouse models that resistance may be overcome by this strategy and the tumors become resensitized to EGFR/ HER2 blockade.43 Currently, studies of palbociclib and T-DM1 (NCT01976169) and abemaciclib plus trastuzumab are being done after preliminary activity has been reported (NCT02675231). The PATRICIA trial is comparing palbociclib plus trastuzumab with or without letrozole for patients with triple-positive advanced cancers who have had prior trastuzumab treatment (NCT0244840).

Ongoing Issues

Although there has been major progress in the treatment of HER2-positive advanced breast cancer, there are a number of ongoing issues that have not been resolved. First, as new

agents are introduced into the neoadjuvant and adjuvant settings, including potentially pertuzumab, the selection of drugs and the optimal sequence in the advanced setting for previously treated patients may need to be revised. There are currently no data on how long to continue treatment for patients who appear to have a complete response, leading to a rather ad hoc approach to these rare cases. Most patients continue with anti-HER2 therapy indefinitely, because there is concern about tumor recurrence. We continue to struggle with tumor resistance, because this is the major cause of treatment failure. How much does tumor heterogeneity impact resistance and will this limit some of our more specific and HER2-directed treatments? What are the mechanisms of resistance? Although there have been clues, we are still a long way from defining them among most individual patients. Serial sampling and cell-free DNA studies may help our understanding. A notable number of patients still develop central nervous system disease, which continues to be difficult to treat despite studies of new agents and intrathecal drugs, including intrathecal trastuzumab. The development of new drugs is not easy with the small numbers of patients with advanced HER2-positive disease eligible for studies in some centers, generally requiring multicenter trials, which adds to the complexity and expense. This is a statement of our success but does extend the time to develop potentially active agents, particularly in later lines of therapy. Finally, will patients be able to afford new agents in the future or possibly continued treatment in the present?

OPTIMIZING NEOADJUVANT/ADJUVANT TREATMENT OPTIONS FOR THE PATIENT WITH HER2-AMPLIFIED BREAST CANCER

It has now been 12 years since trastuzumab was first reported to significantly improve disease-free survival (DFS) for HER2-positive breast cancer in the curative setting.44 The positive impact of trastuzumab cannot be overstated, because its use has been shown to alter the natural course of this disease, transforming it from an aggressive subtype with poor outcomes to one that may be expected to have a prognosis as favorable as HER2-normal disease (Table 1).45-48 In the past decade, findings from multiple studies have become available to inform the optimal treatment of this form of breast cancer. These trials have addressed fundamental issues, including optimal duration of trastuzumab, timing with chemotherapy, chemotherapy backbone, and relative risks and benefits of adding other novel biologic therapies, including the use of dual HER2 targeting. Importantly, ongoing studies are now addressing the use of less toxic regimens with novel therapeutics and are evaluating prognostic or predictive biomarkers for long-term outcome. These data will no doubt be invaluable in maximizing the therapeutic index for patients diagnosed with early-stage HER2-driven disease. This section reviews how results from large adjuvant studies have significantly influenced management of this disease and describes major findings from neoadjuvant clinical trials that have provided essential clinical and molecular information in an efficient and cost-effective manner. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 79

HURVITZ, GELMON, AND TOLANEY

Four Trials, One Enormous Breakthrough

In February 2000, the first phase III clinical trial to evaluate the use of adjuvant trastuzumab (NSABP B-31) was launched; within less than 2 years, three additional large randomized adjuvant trials dedicated to evaluating trastuzumab in HER2-overexpressing or amplified breast cancer were activated, including the NCCTG N9831, BCIRG-006, and BIG 01-01 HERceptin Adjuvant (HERA) studies44,51,52 (Table 2). Updated results for each of these trials, with 8 years53 to 10 years50,54 of median follow-up, demonstrated persistent substantial DFS and OS benefits associated with the addition of trastuzumab to standard chemotherapy. The designs of each of these clinical trials, as well as the design and results of subsequent studies, have provided important data to guide optimal treatment of early-stage disease. These will each be considered below.

Optimal Timing of Trastuzumab With Chemotherapy

Although several studies used concurrent chemotherapy and trastuzumab,44,51,56 two were designed to give the trastuzumab after chemotherapy.52,57 In HERA, 5,102 patients who had completed adjuvant chemotherapy were randomly assigned to either observation or 1- or 2-year treatment with trastuzumab.52 With 10 years of follow-up, DFS (HR, 0.77; 95% CI, 0.68–0.86; p < .0001) and OS (HR, 0.74; 95% CI, 0.64–0.86; p < .0001) remained statistically significantly better with a year of trastuzumab compared with observation.54 In a similarly designed study (FNCLCC-PACS-04),57 528 patients with HER2-positive breast cancer were randomly assigned after chemotherapy to receive trastuzumab for a year or to undergo observation, but they did not demonstrate a notable improvement in DFS with trastuzumab; the small size of the study, coupled with the fact that only 84% of the patients received at least 6 months of trastuzumab, may explain these discordant results. To date, the N9831 trial is the only study that has prospectively compared the sequential or concurrent approaches.58 With a median follow-up of 6 years, patients treated with concurrent pacl*taxel chemotherapy and trastuzumab had

a 5-year DFS of 84.4% compared with 80.1% for patients treated with pacl*taxel followed by trastuzumab (HR, 0.77; 99.9% CI, 0.53–1.11; p = .0216). Although this did not cross the prespecified boundary for significance (p = .00116), this trend toward an increase in DFS with the concurrent administration led the authors to conclude that trastuzumab should be given concurrently with taxane chemotherapy.

Optimal Length of HER2-Targeted Therapy

Although the original decision to give trastuzumab for 1 year was relatively arbitrary, we now have the benefit of data from several studies that have addressed the ideal length of trastuzumab treatment. In addition, several other studies addressing the optimal length of trastuzumab treatment have enrolled patients but results are yet to be reported. The HERA trial was the first to address length of therapy by including not only an observation and 1-year trastuzumab arm but also a 2-year trastuzumab arm.52 At a median 8 years of follow-up, there was no difference in DFS for patients treated for 1 or 2 years (HR, 0.99; 95% CI, 0.85–1.14; p = .86).55 Importantly, the rates of grade 3/4 adverse events were higher for patients in the 2-year group (20.4%) compared with the 1-year group (16.3%). This included a higher rate of cardiac toxicity (4.1% and 7.2% for the 1-year and 2-year groups, respectively). To date, two trials that evaluated whether a shorter course of trastuzumab yields similar outcomes to 1 year have been reported. FinHer was a 1,010-patient trial, in which patients were randomly assigned to receive three cycles of docetaxel or vinorelbine followed three cycles of 5′ fluorouracil/epirubicin/cyclophosphamide.56 The 232 women with HER2-positive breast cancer were randomly assigned to receive a 9-week course of trastuzumab concurrently with the vinorelbine or docetaxel. Three-year recurrence-free survival was significantly improved for trastuzumab-treated patients (HR, 0.42; 95% CI, 0.21–0.83; p = .01) with a trend toward improved OS (HR, 0.41; 95% CI, 0.16–1.08; p = .07).56 With longer follow-up, however,48 the distant DFS benefit was no longer statistically significant but still tended to be in favor

TABLE 1. Overall Survival for HER2-Positive, Trastuzumab-Treated Early Disease Similar to or Better Than HER2-Normal Disease No. of Patients With HER2-Positive Disease (%)

Study

Reference

Yes Trastuzumab

No Trastuzumab

No. of Patients With HER2-Negative Disease (%)

BCIRG-005 and BCIRG-006

Mackey et al49 and Slamon et al50

1,841/2,149 (86)*

870/1,073 (81)

2,647/3,298 (80)*

NOAH

Gianni et al46

87/117 (74)

74/118 (63)

75/99 (76)

Italian Registry

Musolino et al

4.1

52/53 (98)

140/161 (87)

1,108/1,186 (93)*

GeparQuattro

von Minckwitz et al

5.4

392/446 (88)

FinHer6

Joensuu et al48

12/115 (90)

45 47

Median Follow-Up (Years)

889/1,049 (85)*

21/116 (82)

61/778 (92)

*Data from these studies indicate that, in general, HER2-positive trastuzumab-treated disease has a similar or better outcome than HER2-negative disease. Studies that included patients with both HER2-positive (trastuzumab treated and trastuzumab nontreated) and HER2-negative disease are included (NOAH, Italian Registry, GeparQuattro, FinHer). In addition, BCIRG-005 (HER2 negative) and BCIRG-006 (HER2 positive) are included, because these studies were conducted at many of the same sites during similar time periods. Patients with HER2-positive disease were referred to BCIRG-006 and those with HER2-negative disease were referred to BCIRG-005.

80 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

OPTIMAL MANAGEMENT OF HER2 BREAST CANCER

TABLE 2. Four Large Initial Adjuvant Trastuzumab Trials

DFS

Median FollowUp (Years)

3,351

10-year DFS for groups C/2 vs. A/1: 73%, AC/PH; and 62%, AC/P

10-year OS for groups C/2 vs. A/1: 84%, AC/PH; and 75%, AC/P

0.63

0.60

Hazard Ratio* Trial

Reference

Arms

NCCTG N9831 and NSABP B-31

Perez et al53

NCCTG N9831**: AC → wP (group A), AC → wP → wH (group B),† and AC → wPwH (group C)

No. of Patients

Crossover (%)

NSABP B-31‡: AC → P (group 1) and AC → PwH (group 2) HERA

Goldhirsch et al55

Standard chemotherapy and then observation vs. H for 1 years vs. H for 2 years

5,090

72%, H for 1 year; 66%, observation

84%, H for 1 year; 79%, observation

0.76

BCIRG-006

Slamon et al50

AC → T, AC → TH, and TCH

3,222

10-year: 75%, AC/ TH; 73%, TCH; and 68%, AC/T

10-year: 86%, AC/ TH; 83%, TCH; and 79%, AC/T

0.63, AC/ TH; 0.76, TCH

0.72, AC/ TH; and 0.77, TCH

10.3

3.1

*Statistically significant. **NCCTG N9831: AC → P, doxorubicin/cyclophosphamide for four cycles followed by weekly pacl*taxel 12×; AC→ wP → wH, doxorubicin/cyclophosphamide for four cycles followed by weekly pacl*taxel 12× followed by weekly trastuzumab for 1 year; AC → wPwH, doxorubicin/cyclophosphamide for four cycles followed by weekly pacl*taxel plus weekly trastuzumab 12× followed by weekly trastuzumab to complete a year. †Not included in the joint analysis. ‡NSABP B-31: AC → P, doxorubicin/cyclophosphamide for four cycles followed by pacl*taxel (weekly 12× or every 3 weeks 4×); AC → PwH, doxorubicin/cyclophosphamide for four cycles followed by pacl*taxel (weekly 12× or every 3 weeks 4×) plus weekly trastuzumab followed by weekly trastuzumab to complete a year. HERA: trastuzumab every 3 weeks for 1 year or 2 years. BCIRG-006: AC/TH, doxorubicin/cyclophosphamide for four cycles followed by docetaxel/trastuzumab for four cycles, followed by maintenance with trastuzumab every 3 weeks to complete a year; AC/T, doxorubicin/cyclophosphamide for four cycles followed by docetaxel for four cycles; TCH, docetaxel/carboplatin/trastuzumab for six cycles followed by maintenance with trastuzumab every 3 weeks to complete a year. Abbreviations: AC, doxorubicin; DFS, disease-free survival; H, trastuzumab; OS, overall survival; P, pertuzumab; PH, pertuzumab and trastuzumab; T, docetaxel; TCH, docetaxel/carboplatin/trastuzumab; wH, weekly trastuzumab; wP, weekly pacl*taxel; wPwH, weekly pacl*taxel plus weekly trastuzumab.

of trastuzumab-based therapy (HR, 0.65; 95% CI, 0.38–1.12; p = .12). Thus, although the safety and financial aspects of 9 weeks of trastuzumab treatment are attractive, the longterm benefits are not certain. The second study reported to date to evaluate a shorter duration of trastuzumab is PHARE, a phase III noninferiority study aimed to evaluate 6 versus 12 months of trastuzumab for patients who had completed chemotherapy, surgery, and up to 6 months of trastuzumab treatment.59 The prespecified noninferiority HR margin was set at 1.15. With a median follow-up of 42.5 months, the HR was 1.28 (95% CI, 1.05–1.56; p = .29); thus, noninferiority of 6 months of trastuzumab compared with 12 months was not demonstrated. An ongoing phase III study with a noninferiority DFS endpoint (PERSEPHONE) is also addressing 6 versus 12 months of trastuzumab. This study completed accrual of more than 4,000 patients in 2015. Cardiac safety data from the first 2,500 patients enrolled were reported in 2016, demonstrating a substantial reduction in cardiac events associated with 6 months of therapy compared with 12 months.60 Several studies addressing the optimal length of trastuzumab therapy are ongoing (Table 3). On the basis of currently available data, the standard of care remains 1 year of trastuzumab treatment. It should be noted that one phase III randomized study, EXTENET (described below), is

evaluating a year of HER2-targeted therapy with neratinib for patients who already completed a full year of trastuzumab for early-stage disease.61 Thus, in addition to evaluating a novel HER2-targeted therapy in the adjuvant setting, this study is addressing whether 2 years of HER2-targeted therapy improves outcomes compared with 1 year. A prespecified early analysis at the 2-year mark demonstrated that invasive DFS was significantly improved for patients who received neratinib, especially those with hormone receptor–coexpressing cancer. These intriguing data are in contrast with the 8-year HERA results, in which 2 years of trastuzumab did not provide additional benefit compared with 1 year, regardless of hormone receptor expression.55 One theory to explain this differential outcome is that in contrast with trastuzumab, neratinib may more effectively interfere with receptor crosstalk between human EGFRs and ERs, leading to particular benefit in hormone receptor–positive tumors. Pending longer follow-up of this trial as well as data to guide management of the gastrointestinal toxicity associated with neratinib, 1 year of trastuzumab in the adjuvant setting remains the standard of care.

Optimizing the Cardiac Risk

Although the DFS and OS benefits of trastuzumab clearly support its use in the curative setting, the risk of cardiac asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 81

HURVITZ, GELMON, AND TOLANEY

TABLE 3. Ongoing Studies Evaluating Duration of Trastuzumab Study

No. of Patients (Enrollment Status)

ClinicalTrials.gov Identifier

PERSEPHONE

4,089 (closed)

SHORT-HER

Treatment Arms

Endpoint

NCT00712140

Chemotherapy concurrent or sequential with trastuzumab 12 vs. 6 months

DFS

2,500 (closed)

NCT00629278

AC or EC 4× → TH (H for 1 year) vs. TH 3× → FEC 3×

DFS

SOLD

2,168 closed)

NCT00593697

TH 3× → FEC 3× vs. TH 3× → FEC 3× → H for 1 year

DFS

BOLD-1

1,366 (open)

NCT02625441

THP 3×

iDFS

Hellenic Oncology Research Group

489 (closed)

NCT00615602

TH 3× → H for 1 year FEC 4× → TH 4× → H for 6 months

DFS

FEC 4× → TH 4× → H for 12 months Abbreviations: AC, doxorubicin/cyclophosphamide; DFS, disease-free survival; EC, epirubicin and cyclophosphamide; FEC, fluorouracil/epirubicin/cyclophosphamide; H, trastuzumab; iDFS, invasive disease– free survival; TH, trastuzumab; THP, trastuzumab/pertuzumab.

toxicity should be carefully considered, especially in the early-stage setting, in which a relatively substantial proportion of patients may be cured with local measures alone. A meta-analysis that included eight randomized controlled trials of trastuzumab (11,991 patients) reported a 5.11 times higher risk of congestive heart failure (2.5% vs. 0.4%) and a 1.83 times higher risk of left ventricular ejection fraction (LVEF) decline (11.2% vs. 5.6%) for trastuzumab-treated patients compared with control-treated patients.62 Although these rates of cardiac dysfunction are relatively low compared with the improvements in DFS and OS, it is concerning to note that 7%–10% of patients who began anthracycline-based chemotherapy in the B-3163 and N983164,65 trials were unable to ever receive trastuzumab-based therapy because of unacceptably low cardiac function. This raises an important point relating to the chemotherapy backbone: the majority of adjuvant trastuzumab trials used an anthracycline, making it difficult to distinguish the relative impact on cardiac outcome contributed by trastuzumab and anthracycline. Seven-year follow-up of the B-31 trial reported cardiac events for 4.0% of patients treated with trastuzumab compared with 1.3% of patients in the control arm.63 The prevalence of clinically occult cardiac damage is difficult to gauge, however, because this study was designed to measure LVEF data for asymptomatic patients for up to only 18 months. Of 947 trastuzumab-treated patients in the B-31 trial, 12% stopped taking trastuzumab because of an asymptomatic decline in LVEF; altogether, 15.5% stopped trastuzumab prematurely, owing to cardiac-related issues.63 The N9831 study also reported the rates of asymptomatic LVEF decline observed in the 18–21 months postrandomization. Of 1,944 patients who began post-AC (doxorubicin/ cyclophosphamide) treatment, LVEF declined 10% or more for 26% of patients treated with AC/T (arm A), 35% of those treated with sequential pacl*taxel and trastuzumab AC/T/H (arm B), and 40% of those treated with concurrent pacl*taxel trastuzumab AC/trastuzumab (arm C).65 A small proportion of patients (33%) consented to have another LVEF measurement at the 6-year time point. Data from these 651 patients 82 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

showed that a substantial proportion in each treatment arm had a decrease in LVEF of at least 10% (arm A, 21%; arm B, 20%; and arm C, 23%), a decrease in LVEF of at least 15% (arm A, 9%; arm B, 7%; and arm C, 9%), and a decrease to below the lower limits of normal (arm A, 6%; arm B, 5%; and arm C, 5%). The percentage of patients with LVEF decline was similar among the treatment arms, and the median LVEF change from baseline to year 6 also appeared to be similar among the three treatment arms (−3.0%, −2.5%, and −3.0% in arms A, B, and C, respectively), leading the authors to speculate that one explanation for long-term LVEF dysfunction may be related to anthracycline exposure as opposed to trastuzumab exposure.65 To date, the only adjuvant study comparing a nonanthracycline regimen to an anthracycline regimen is BCIRG-006.51 This study was prospectively designed to not only evaluate the relative efficacy of the two trastuzumab-containing arms (docetaxel/carboplatin/trastuzumab [TCH] and doxorubicin/cyclophosphamide followed by AC/trastuzumab) to the AC/T control arm but also aimed to prospectively follow cardiac function out to 5 years. Data from 48 months of follow-up demonstrated congestive heart failure for 2.0% of patients in the AC/trastuzumab arm, 0.7% in the AC/T arm, and 0.4% in the TCH arm. Moreover, a decline in mean LVEF of greater than 10 points was reported for 18.6% of patients in the AC/trastuzumab arm, 9.4% in the TCH arm (AC/trastuzumab vs. TCH: p < .001), and 11.2% in the AC/T arm.51 With over 5 years of follow-up, the decline in mean LVEF did not appear to persist over time in the TCH arm.50 However, persistence in this decline was observed among anthracycline-treated patients. In terms of efficacy, with a median follow-up of 10.3 years, both trastuzumab-containing arms demonstrated significant improvements in both DFS (AC/ trastuzumab vs. AC/T: HR, 0.72; p < .0001; TCH vs. AC/T: HR, 0.77; p = .0011) and OS (AC/trastuzumab vs. AC/T: HR, 0.63; p < .0001; TCH vs. AC/T: HR, 0.76; p < .0075).50 Although the study was not powered to test equivalence of the two trastuzumab-based arms, it is notable that 10-year DFS was quite similar in the two trastuzumab arms for higher-risk

OPTIMAL MANAGEMENT OF HER2 BREAST CANCER

patients with lymph node–positive (69.6%, AC/trastuzumab; 68.4%, TCH) or 4 or greater lymph node–positive disease (62.8%, AC/trastuzumab; 62.9%, TCH). To date, more than 5,000 patients have been treated with TCH-based therapy in clinical trials.51,66-70

Optimal Study Design to Evaluate Novel Therapies: Neoadjuvant Versus Adjuvant Settings

Around the same time that the large adjuvant trastuzumab studies were enrolling patients, two studies were started to evaluate the use of trastuzumab in primary breast tumors.71-73 Both showed that trastuzumab more than doubled pathologic complete response (pCR) rates and also improved relapse-free/event-free survival.46,71,73 Subsequently, several other trials were conducted to evaluate neoadjuvant trastuzumab.74-76 Collectively, data from these studies support the routine clinical use of neoadjuvant trastuzumab, especially for larger tumors. Traditionally, clinical trials of new agents have been conducted in the adjuvant setting. However, there are several potential advantages to the use of a neoadjuvant study design. First, the pCR rate appears to be a reliable surrogate marker of long-term outcome, especially for HER2-positive breast cancer.77 This enables a relatively rapid readout of primary endpoints as well as smaller sample sizes. Neoadjuvant studies have been conducted to compare the activity and safety of trastuzumab to lapatinib78-80; to evaluate dual HER2 targeting with trastuzumab plus lapatinib,68,81-85 trastuzumab plus pertuzumab,67,86 and T-DM1 plus pertuzumab87;

and to gauge activity of combining HER2- and hormonally targeted approaches (Table 4).70,88,89 In addition, the neoadjuvant setting allows for serial biopsies to be performed, thus enabling in vivo molecular analyses to be conducted to assess for novel markers of response or resistance to therapy. This will be critical as we aim to personalize treatment regimens to provide an individual the highest therapeutic benefit with the least amount of toxicity. That said, although the design allows for more cost-effective trials to be done in an efficient manner, their small size makes it unlikely these studies will be powered to evaluate long-term outcomes. Thus, adjuvant studies are still needed to validate promising findings from the neoadjuvant setting. It is hoped that utilization of the neoadjuvant study design for the testing of novel targeted therapeutics will allow for weeding out of the less effective or more toxic agents, thus sparing the high expense, long follow-up, and large patient numbers required in the adjuvant setting.

Optimal HER2-Targeted Agents

Tyrosine kinase inhibitors. Evidence supporting activity of the oral TKI lapatinib in the preclinical and metastatic settings provided strong rationale for the evaluation of lapatinib alone and in combination with trastuzumab in the adjuvant setting.9,11,90-92 In the TEACH trial, 3,147 patients with HER2-positive stage I–IIIC breast cancer who had completed adjuvant chemotherapy were randomly assigned to lapatinib (1,500 mg daily) for 12 months or placebo.93 With a median follow-up of approximately 4 years, lapatinib-treated

TABLE 4. Ongoing Adjuvant/Neoadjuvant Studies Study

No. of Patients (Enrollment Status)

ClinicalTrials.gov Identifier

KATHERINE

1,487 (closed)

APHINITY

4,806 (closed)

Treatment Arms

Endpoint

NCT01772472

Adjuvant T-DM1 vs. trastuzumab (patients with residual disease after neoadjuvant treatment)

iDFS

NTC01358877

Chemotherapy/trastuzumab vs. chemotherapy/trastuzumab/pertuzumab

iDFS Cardiac safety

KAITLIN

1,846 (closed)

NCT01966471

AC or FEC → T-DM1/pertuzumab

iDFS

BOLD-1

1,366 (open)

NCT02625441

Taxane/trastuzumab/pertuzumab 3× → FEC 3×

ATEMPT

500 (open)

NCT01853748

T-DM1 for 1 year vs. pacl*taxel/trastuzumab for 12 weeks → trastuzumab for 1 year (stage I disease)

DFS

NeoPhoebe

50 (closed)

NCT01816594

Trastuzumab/pacl*taxel/buparlisib vs. trastuzumab/pacl*taxel/placebo

pCR

GeparOcto

950 (open)

NCT02125344

PMCb vs. ETC

pCR

AC or FEC → taxane/trastuzumab/pertuzumab iDFS

Taxane/trastuzumab 3× → FEC 3× → trastuzumab for 1 year

If HER2+, also pertuzumab/trastuzumab Predix-HER2

200 (open)

NCT02568839

Docetaxel/sq trastuzumab/pertuzumab vs. T-DM1

pCR

Therapy arms switched if no response after cycle 2 TEAL

30 (open)

NCT02073487

T-DM1/lapatinib → nanoparticle albumin–bound pacl*taxel vs. trastuzumab/pertuzumab/pacl*taxel

pCR

Abbreviations: DFS, disease-free survival; ETC, epirubicin/pacl*taxel/cyclophosphamide; FEC, fluorouracil/epirubicin/cyclophosphamide; iDFS, invasive disease–free survival; pCR, pathologic complete response; PMCb, pacl*taxel/nonpegylated liposomal doxorubicin/carboplatin; sq, subcutaneous.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 83

HURVITZ, GELMON, AND TOLANEY

patients had an 87% DFS compared with 83% for placebo-treated patients (HR, 0.83; 95% CI, 0.70–1.00; p = .053). It now appears, based on results from other comparative trials (discussed above and below), that lapatinib is less clinically active than trastuzumab. However, it is possible that the sequential administration of lapatinib after chemotherapy lessened its benefit. In addition, it is notable that a large proportion of patients enrolled when they were more than a year from their diagnosis of cancer (29% of lapatinib-treated patients were more than 4 years from their diagnosis) and, importantly, on central review, 21% of patients were found to have HER2-normal cancer. DFS analysis of the patients with centrally confirmed HER2-positive disease did suggest a significant reduction in risk (HR, 0.82; p = .04). That said, the DFS benefits were borderline at best, leading most to envision this as a negative study. Another large study that evaluated lapatinib in the curative setting was ALTTO.66,94 This study was unique, in that it not only compared a year of trastuzumab treatment (T) to a year of lapatinib, but it also evaluated a sequential arm (12 weeks of trastuzumab followed by 34 weeks of lapatinib) and an arm that used dual HER2 targeting. In 2011, the lapatinib arm was closed after an interim analysis determined futility to show noninferiority compared with trastuzumab/ lapatinib. With a median follow-up of 4.5 years and a prespecified level of significance of .025 for each of the comparisons, lapatinib/trastuzumab was not shown to significantly improve DFS compared with trastuzumab (HR, 0.84; 95% CI, 0.70–1.02; p = .048), nor was trastuzumab followed by lapatinib shown to be different from trastuzumab (HR, 0.96; 95% CI, 0.80–1.15; p = .61). Moreover, compared to trastuzumab, lapatinib was associated with lower rates of completion of HER2-targeted therapy, owing to its notable toxicity.94 Lapatinib has also been evaluated in combination with chemotherapy in at least seven neoadjuvant clinical trials.68,79-81,83-85 pCR rates with lapatinib were significantly inferior to trastuzumab in two trials making the head-to-head comparison.79,80 The effects of single-agent lapatinib, trastuzumab, or their combination have been assessed in several of these trials.68,81,83-85 Although all of these studies demonstrated numeric improvements in pCR with dual HER2 blockade, only two of these studies demonstrated a statistically notable improvement in pCR.81,85 The toxicity associated with lapatinib resulted in lower rates of completion of HER2-targeted therapy in several of these trials.81,83,85 Moreover, the pCR benefits noted in one of the studies (NeoALTTO) was not shown to translate into event-free survival benefits.82 Given its unfavorable safety profile and lack of demonstrated notable benefit in two large adjuvant studies and multiple smaller neoadjuvant studies, lapatinib is not considered appropriate therapy in the early-stage setting. That said, another TKI, neratinib, is showing promise in the adjuvant setting. As mentioned above, EXTENET is a phase III placebo-controlled study in which 2,840 patients who had completed trastuzumab were randomly assigned to 1 year of neratinib or placebo.61 At 2-year follow-up, invasive DFS was 93.9% for neratinib-treated patients compared with 91.6% 84 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

for the control arm (stratified HR, 0.67; 95% CI, 0.50–0.91; p = .0091). This benefit came at the expense of relatively severe gastrointestinal toxicity, with 40% of patients reporting grade 3 diarrhea. An ongoing study is being conducted to evaluate whether diarrhea can be mitigated with primary prophylaxis with loperamide (NCT02400476). At this point, neratinib is not available outside of a clinical trial. Pertuzumab. In September 2013, based on the results of two relatively small phase II trials,67,86 the U.S. Food and Drug Administration approved three neoadjuvant regimens for HER2-positive tumors greater than 2 cm in size. The three approved regimens all used dual HER2 targeting with pertuzumab and trastuzumab given concurrently with chemotherapy. One of the regimens approved (trastuzumab/ pertuzumab 4×, followed postoperatively by fluorouracil/ epirubicin/cyclophosphamide 3×) was based on the results of NeoSphere,86 a four-arm randomized phase II trial that compared docetaxel plus either trastuzumab (TH), pertuzumab (TP), or both trastuzumab and pertuzumab. A fourth arm that used a chemotherapy-free regimen comprised of pertuzumab and trastuzumab was also included (HP). The pCR rates were significantly higher in the trastuzumab/pertuzumab arm and, importantly, the combination was shown to be relatively safe. The study was not powered to demonstrate event-free survival benefit but an exploratory analysis at 5-year follow-up showed a numerical trend in favor of the trastuzumab/pertuzumab arm compared with TH.95 The other two regimens approved in 2013, docetaxel, carboplatin, trastuzumab and pertuzumab (TCHP) and fluorouracil/ epirubicin/cyclophosphamide 3× followed by trastuzumab/ pertuzumab 3×) were based on TRYPHAENA, a 225-patient, three-arm study primarily aimed to evaluate the cardiac safety of three pertuzumab/trastuzumab-based regimens.67 pCR rates (in breast and lymph nodes) were similarly high among the three treatment arms (notably, 64% for TCHP and 55% for fluorouracil/epirubicin/cyclophosphamide followed by trastuzumab/pertuzumab). No long-term data from this study are currently available, but the tested regimens appeared to be safe from a cardiac perspective. The largest neoadjuvant study reported to use pertuzumab and trastuzumab in combination with chemotherapy is GeparSepto.96 This phase III trial was aimed to test noninferiority of nanoparticle albumin–bound pacl*taxel– based chemotherapy to solvent-based pacl*taxel. All patients with HER2-positive disease in this study (n = 396) received both pertuzumab and trastuzumab; thus, relative benefits imparted by dual HER2-targeted therapy compared with trastuzumab could not be assessed. pCR rates for the HER2-positive subset were 62% with nanoparticle albumin– bound pacl*taxel and 54% with solvent pacl*taxel (p = 0.13), providing further evidence of the activity of pertuzumab/ trastuzumab-based therapy. Although these data and the regulatory approval support the clinical use of trastuzumab and pertuzumab in the neoadjuvant setting, long-term safety and efficacy data from larger confirmatory studies are awaited (Table 3) before routine use of dual HER2 targeting in the adjuvant setting.

OPTIMAL MANAGEMENT OF HER2 BREAST CANCER

Optimizing Therapy for Hormone Receptor– Coexpressing Disease

At least half of HER2-positive breast cancer coexpresses one or both hormone receptors, and this coexpression may serve as a pathway for resistance to HER2-targeted therapy. This does not mean that HER2-targeted therapy is inactive in hormone receptor–positive breast cancer. In fact, analyses from the AC/trastuzumab and AC/T arms of the BCIRG-00651 and B-3153 trials show that the HRs for DFS are very similar for hormone receptor–positive (HR, 0.65 and 0.61 for BCIRG-006 and B-31, respectively) and hormone receptor–negative (HR, 0.64 and 0.62 for BCIRG-006 and B-31, respectively) disease. This also holds true for OS. Similarly, subset analysis of the HERA study at 10 years of follow-up also demonstrates long-term trastuzumab benefit for all patients regardless of HR status.54 Although trastuzumab imparts DFS and OS benefit regardless of hormone receptor status, the presence of ER may indicate more indolent, luminal-like tumor behavior. For example, Kaplan–Meier curves from HERA indicate that although the long-term risk of recurrence is similar in hormone receptor–positive and hormone receptor–negative subtypes, patients with hormone receptor–negative disease have earlier recurrences, in keeping with more aggressive disease biology. Further evidence supporting the notion that disease behavior differs based on hormone receptor expression comes from neoadjuvant clinical trials, which have consistently shown that pCR rates are lower for hormone receptor–positive, HER2-positive breast cancer than for hormone receptor–negative disease.67,77,81,83 That said, the longer follow-up of the NeoSphere trial95 indicates that patients with hormone receptor coexpression have numerically higher PFS compared with those with tumors lacking hormone receptors (5-year PFS for patients who achieved pCR: 90% if hormone receptor positive, 84% if hormone receptor negative; 5-year PFS for patients who did not achieve pCR: 80% if hormone receptor positive, 72% if hormone receptor negative). Thus, patients with hormone receptor–positive tumors may do better in the long run. Intriguing biomarker analyses from HERA suggest that although ER-positive tumors with a high level of HER2 amplification (by FISH ratio) derive clear benefit from trastuzumab, those with a low level of HER2 amplification may not receive benefit from trastuzumab-based therapy.97 Several clinical trials aimed to evaluate cotargeting hormone receptor and HER2 have been conducted. The first of these, TBCRC-006, evaluated 12 weeks of neoadjuvant lapatinib plus trastuzumab (with letrozole for ER-positive tumors).88 pCR (breast) in HER2-positive/ hormone receptor–positive tumors was 21% in this proofof-concept study, indicating that a relatively well-tolerated chemotherapy-free regimen might be highly effective for patients if accurate biomarkers for selection could be identified. Trastuzumab emtansine has also been evaluated in the neoadjuvant and adjuvant settings. The WGS-ADAPT

study compared four cycles of T-DM1, either alone or in combination with endocrine therapy, to trastuzumab plus endocrine therapy for patients with hormone receptor– positive, HER2-positive disease.89 This relatively short course of T-DM1 was associated with an impressive pCR rate (breast and lymph nodes) of 41%, which was considerably higher than that achieved with trastuzumab plus endocrine therapy. Although neither of these relatively small studies has changed the standard of care, the intriguing results should encourage the investigation of whether less toxic regimens like these might be beneficial for selected patient populations. In December 2016, the results of the NSABP B-52 trial were presented. This study was designed to evaluate whether the addition of an aromatase inhibitor to standard chemotherapy plus HER2-targeted therapy (TCHP) would improve pCR rates for hormone receptor–positive/ HER2-positive breast cancer, and to also test whether endocrine therapy would be antagonistic in combination with chemotherapy.70 Although the addition of endocrine therapy to TCHP did not lead to a statistically notable improvement in pCR (41% for TCHP vs. 46% for TCHP plus endocrine therapy), it did not appear to be antagonistic, leaving room for future studies to test less toxic chemotherapy regimens concurrently with hormone therapy approaches. In summary, in just over a decade, the management of early-stage HER2-positive breast cancer has changed drastically because of the development of highly effective biologically targeted therapy. The therapeutic options available to the patient in both the neoadjuvant and adjuvant settings are now nearly countless, making the choice of optimal therapy somewhat difficult at times. Our pursuit to provide patients with the safest and most effective therapies for their particular disease requires us to design carefully selected clinical trials with attention toward the discovery of molecular drivers of disease biology and markers of response to therapy.

DE-ESCALATING TREATMENT IN THE ADJUVANT SETTING IN HER2-POSITIVE DISEASE

Although there has been much work done to improve outcomes for patients with HER2-positive breast cancer by adding additional HER2-targeted agents to a standard chemotherapy backbone, it is important to consider that there may be some patients for whom we may be able to de-escalate treatment. One way to achieve this would be to use biomarkers that predict which patients are likely to benefit from less therapy. Although there are several biomarkers being explored, there is not yet one identified that can help us preselect patients for less therapy. Another approach would be to consider using clinical features to help determine which patients may be able to achieve good outcomes with less toxic regimens. Three clinical features we could consider to help us select patients asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 85

HURVITZ, GELMON, AND TOLANEY

include tumor size, patient age, and response to preoperative therapy.

Using Biomarkers to De-escalate Therapy

Much work has been done to try to identify biomarkers that may help us identify patients that are likely to benefit to anti-HER2 therapy. A meta-analysis performed by Loibl et al98 suggests that the presence of a PIK3CA mutation is associated with a significantly lower rate of pCR to anti-HER2 therapy; however, this difference in pCR was not associated with a difference in DFS. These data suggest that although those patients without a PIK3CA mutation may achieve better pCR rates than those with a mutation, the mutation is not predictive of long-term outcomes and thus cannot help us select patients that may be able to receive less therapy. In a recent meta-analysis presented by Denkert et al,99 high levels of TILs correlated with pCR rates among patients receiving anti-HER2 therapy and were associated with improved PFS. These data are in contrast with data from the N9831 study, which suggested that the presence of stromal TILs was associated with an improvement in recurrence-free survival of patients treated with chemotherapy alone but not of patients treated with chemotherapy and trastuzumab. In addition, high levels of stromal TILs were associated with lack of trastuzumab benefit.100 Further work must be done to assess whether those patients with high TILs may achieve similar outcomes with less chemotherapy, and whether replacing chemotherapy with immunotherapy may be beneficial for these patients.

De-escalating Therapy Based on Clinical Parameters: Tumor Size

Systemic therapy for small (stage I) HER2-positive breast cancers has been a challenge for clinicians. This is largely attributable to the fact that the pivotal adjuvant trastuzumab trials included very few patients with stage I disease and even fewer with tumors smaller than 1 cm. In addition, as mammographic screening has become more widespread, the number of women diagnosed with T1 tumors has increased significantly. For example, among middle-aged women in the U.S. Surveillance, Epidemiology, and End Results registry, the diagnosis of T1 tumors increased from 143.5 per 100,000 to 163.5 per 100,000 women between 1990 and 1998.101 Thus, providers face management of small, node-negative HER2-positive tumors with increasing frequency. Moreover, data from several retrospective studies looking at outcomes for patients with untreated tumors suggest that even the smallest node-negative HER2positive tumors may have a substantial risk of recurrence (Table 5). Some of the most informative prognostic data for stage I HER2-positive breast cancers comes from a population-based cohort from British Columbia; this study demonstrated 10-year relapse-free survival of 71.6% for patients with stage I HER2-positive disease.102 Similarly, a Finnish cohort of patients with pT1N0 disease had 72% 9-year distant DFS.103 No patients in either cohort received trastuzumab therapy. 86 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Studies looking at outcomes for T1abN0 tumors also suggest that even these tumors have a notable risk of recurrence. An MD Anderson Cancer Center study demonstrated that 5-year recurrence-free survival was 77.1% among 98 patients with untreated HER2-positive tumors that were less than or equal to 1 cm.108 In a study examining a slightly larger population of patients within the National Comprehensive Cancer Network with similar disease characteristics, Vaz-Luis et al110 demonstrated 5-year distant recurrence-free survival of 94%–96%. In a Kaiser Permanente study in which a minority of patients received chemotherapy and/or trastuzumab, 5-year distant invasive recurrence-free interval was 96.5%.107,110 Although definitions for recurrence vary across retrospective studies and some trials included patients who received systemic therapy, the rates of recurrence across trials range from approximately 5% to 30%, indicating that these patients are at more than just minimal risk of recurrence. Because these patients were excluded from the large adjuvant trials, the Adjuvant Pacl*taxel and Trastuzumab (APT) trial was designed to prospectively address treatment of patients with small, node-negative HER2-positive breast cancer.111 Eligible patients in this single-arm trial had a primary tumor size of less than or equal to 3 cm and had node-negative or N1mic disease (nodal disease greater than 0.2 mm but not more than 2 mm). Patients were treated with weekly pacl*taxel and trastuzumab for 12 weeks, followed by completion of 1 year of trastuzumab. The study enrolled 406 patients, of which 67% had hormone receptor–positive tumors and 49.5% of tumors were less than or equal to 1 cm; 8.9% of patients had tumors greater than 2 cm and 1.5% of patients had N1mic disease (the remainder had N0 disease). Survival free from invasive disease, the primary endpoint of the trial, was 98.7% (95% CI, 97.7%–99.8%) at 3 years. Toxicity in the APT trial was minimal; 3.2% of patients experienced an asymptomatic but notable decline in cardiac ejection fraction, 0.5% of patients (2 patients) developed symptomatic heart failure. The majority of cardiotoxicity events were reversible after trastuzumab was held.111 In a substudy of chemotherapy-related amenorrhea following the adjuvant trastuzumab regimen, which included 64 APT trial participants who were premenopausal at the time of APT trial enrollment, 28% of women were amenorrheic at a median of 51 months from study enrollment.112 This compares favorably to the approximately 50% rate of chemotherapy-associated amenorrhea seen for premenopausal recipients of the standard AC/trastuzumab adjuvant regimen113 and suggests that the trastuzumab regimen may have the added benefit of decreased fertility concerns among young, appropriately selected women with HER2-positive cancers. Another prospective, single-arm phase II trial that looked at treatment of patients with early-stage HER2positive breast cancer administered four cycles of docetaxel with cyclophosphamide and trastuzumab, followed by every-3-week trastuzumab to complete a year of therapy.

OPTIMAL MANAGEMENT OF HER2 BREAST CANCER

trastuzumab, and those with notable comorbidity appear to be less likely to complete adjuvant trastuzumab therapy. Given evidence that shorter-duration trastuzumab adversely affects outcomes,59 adherence to trastuzumab may be an important factor to consider when selecting an initial regimen. Previously conducted clinical trials of adjuvant trastuzumab-based regimens included few older patients, with women older than age 60 comprising approximately 15% of all participants. In addition, patients older than age 70, as well as those with cardiac conditions, were either excluded or poorly represented. Because trial eligibility is selective with regard to comorbidities, the landmark trials of trastuzumab provide limited data regarding tolerability and effectiveness of therapy among older patients. Multiple studies do, however, suggest that older women with more cardiac comorbidity may be at higher risk of cardiac toxicity with anthracycline-based therapy.105,116,117 When making decisions about these patients, it is important to factor in the potential benefits and toxicities of therapy, particularly in disease settings like HER2-positive cancer, in which recurrences can occur earlier rather than later. Older patients without notable comorbidities should be considered for standard adjuvant regimens; however, individuals with multiple medical problems may be those

Of those patients enrolled, 284 (57.6%) had stage I HER2-positive disease. Three-year DFS for patients with node-negative disease was 97.8% (95% CI, 95.6–98.9).114 Because this DFS is very similar to outcomes in the APT trial, there is likely little role for the addition of cyclophosphamide in the management of the lowest-risk HER2positive tumors. Work is ongoing to determine whether even less toxic regimens may be effective in this population. The ATEMPT trial (NCT01853748) recently completed accrual and randomly assigned patients in a 3:1 fashion to T-DM1 or to the pacl*taxel plus trastuzumab regimen used in the APT trial. This study was designed to compare clinically relevant toxicities between the two arms and to also examine DFS among those patients receiving T-DM1.

De-escalating Therapy Based on Patient Age

Elderly individuals comprise another group of patients for whom we should consider de-escalation of therapy. Data from Freedman et al115 looking at the incidence of earlystage HER2-positive disease by age within the National Comprehensive Cancer Network showed that 26% of cases arise for patients older than age 60. These data also demonstrate that older patients were less likely to receive adjuvant

TABLE 5. Observational Cohort Studies of Small HER2-Positive Breast Cancer

Reference

Type of Cohort

Tumors Included (Subgroups)

Chia et al102

British Columbia

206

65.9 (10)

71.2 (10)

pT1abcN0

71.6 (10)

77.5 (10)

Tovey et al104 Joensuu et al

No. HER2+

Chemotherapy Treated (%)

Trastuzumab Treated (%)

Percent of DFS/ RFS (Years)

Percent of DDFS/ DRFS (Years)

pT1abN0

United Kingdom

N0, grade 1–2

NR 72 (9)

Finland

pT1abcN0

Rom et al106

Germany

pT1abcN0

97.1 (1)

98.5 (1)

Fehrenbacher et al107

Kaiser Permanente

pT1abN0

234

25.6

8.1

94.1 (5)

96.5 (5)*

Gonzalez-Angulo et al108

MD Anderson

pT1abN0

77.1 (5)

86.4 (5)

Curigliano et al109

European Institute of Oncology

pT1abN0HR+

25.4

92 (5)

pT1abN0HR−

43.7

91 (5)

France

pT1abN0

73 (10)

80 (10)**

NCCN database

pT1bN0HR+

94 (5)

pT1bN0HR−

94 (5)

pT1aN0HR+

102

96 (5)

pT1aN0HR−

94 (5)

105

Rouanet et al

Vaz-Luis et al110

*Recurrence-free interval (invasive disease only). **Metastasis-free survival. Abbreviations: DDFS, distant disease-free survival; DFS, disease-free survival; DRFS, distant recurrence-free survival; NCCN, National Comprehensive Cancer Network; NR, not recorded; OS, overall survival; RFS, recurrence-free survival. Data reported are from original publications, as referenced. Point estimates for outcomes are included; original publications include confidence intervals. Point estimates must be interpreted in the context of confidence intervals.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 87

HURVITZ, GELMON, AND TOLANEY

that should be considered for less toxic treatments. The ATOP trial (NCT02414646) is currently assessing a less toxic regimen for older patients; in this trial, patients older than age 60 with stage I–III HER2-positive breast cancer, for whom standard regimens are not felt to be appropriate, are treated with 1 year of adjuvant T-DM1. Another study looking at de-escalation of therapy for older patients is the RESPECT trial (NCT01104935) conducted in Japan, which is randomly assigning women age 69–81 with stage I–IIIA HER2-positive breast cancer to treatment with trastuzumab alone versus trastuzumab plus chemotherapy. This study will have the potential to address the question of whether trastuzumab monotherapy has a place in treated elderly patients with HER2-positive disease.

De-escalation of Therapy Based on Response to Preoperative Therapy

Consideration of de-escalation of therapy is also warranted for the group of patients who achieve a pCR to preoperative therapy. Data suggest that patients with HER2-positive breast cancer who achieve a pCR have better long-term outcomes, with improved DFS and OS.77 In a pooled analysis

of 12 international trials (11,955 patients), a notable association was found between pCR and event-free survival. There is, however, no association of treatment effects on long-term outcomes, suggesting that randomized trials with long-term follow-up are needed to understand outcomes for specific therapies. It is therefore critical that we begin designing clinical trials to assess outcomes for patients who achieve a pCR to a highly active regimen, rather than just administering further therapies with associated toxicities that may not be providing additional benefit. We should also consider using escalation of biologic therapy as a mechanism to de-escalate chemotherapy. If adding pertuzumab to trastuzumab-based chemotherapy is found to improve long-term outcomes in the APHINITY trial, perhaps patients may achieve similar outcomes with less chemotherapy and highly effective biologic therapy. One strategy would be to consider doing a prospective randomized trial of doxorubicin/cyclophosphamide/taxane with trastuzumab and pertuzumab compared with a taxane plus trastuzumab and pertuzumab regimen and see whether patients receiving less chemotherapy can do just as well.

References 1. Slamon DJ, Clark GM, Wong SG, et al. Human breast cancer: correlation of relapse and survival with amplification of the HER-2/neu oncogene. Science. 1987;235:177-182.

9. Geyer CE, Forster J, Lindquist D, et al. Lapatinib plus capecitabine for HER2-positive advanced breast cancer. N Engl J Med. 2006;355:27332743.

2. Carter P, Presta L, Gorman CM, et al. Humanization of an antip185HER2 antibody for human cancer therapy. Proc Natl Acad Sci USA. 1992;89:4285-4289.

10. Cameron D, Casey M, Oliva C, et al. Lapatinib plus capecitabine in women with HER-2-positive advanced breast cancer: final survival analysis of a phase III randomized trial. Oncologist. 2010;15: 924-934.

3. Slamon DJ, Leyland-Jones B, Shak S, et al. Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2. N Engl J Med. 2001;344:783-792. 4. Giordano SH, Temin S, Kirshner JJ, et al; American Society of Clinical Oncology. Systemic therapy for patients with advanced human epidermal growth factor receptor 2-positive breast cancer: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol. 2014;32:2078-2099. 5. Marty M, Cognetti F, Maraninchi D, et al. Randomized phase II trial of the efficacy and safety of trastuzumab combined with docetaxel in patients with human epidermal growth factor receptor 2-positive metastatic breast cancer administered as first-line treatment: the M77001 study group. J Clin Oncol. 2005;23:4265-4274. 6. von Minckwitz G, du Bois A, Schmidt M, et al. Trastuzumab beyond progression in human epidermal growth factor receptor 2-positive advanced breast cancer: a German Breast Group 26/Breast International Group 03-05 study. J Clin Oncol. 2009;27:1999-2006. 7. Leyland-Jones B, Gelmon K, Ayoub JP, et al. Pharmaco*kinetics, safety, and efficacy of trastuzumab administered every three weeks in combination with pacl*taxel. J Clin Oncol. 2003;21:3965-3971. 8. Wolff AC, Hammond ME, Hicks DG, et al; American Society of Clinical Oncology; College of American Pathologists. Recommendations for human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists clinical practice guideline update. J Clin Oncol. 2013;31:3997-4013.

88 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

11. Blackwell KL, Burstein HJ, Storniolo AM, et al. Overall survival benefit with lapatinib in combination with trastuzumab for patients with human epidermal growth factor receptor 2-positive metastatic breast cancer: final results from the EGF104900 study. J Clin Oncol. 2012;30:2585-2592. 12. Gelmon KA, Boyle FM, Kaufman B, et al. Lapatinib or trastuzumab plus taxane therapy for human epidermal growth factor receptor 2-positive advanced breast cancer: final results of NCIC CTG MA.31. J Clin Oncol. 2015;33:1574-1583. 13. Bachelot T, Romieu G, Campone M, et al. Lapatinib plus capecitabine in patients with previously untreated brain metastases from HER2positive metastatic breast cancer (LANDSCAPE): a single-group phase 2 study. Lancet Oncol. 2013;14:64-71. 14. Pivot X, Manikhas A, Żurawski B, et al. CEREBEL (EGF111438): a phase III, randomized, open-label study of lapatinib plus capecitabine versus trastuzumab plus capecitabine in patients with human epidermal growth factor receptor 2-positive metastatic breast cancer. J Clin Oncol. 2015;33:1564-1573. 15. Verma S, Miles D, Gianni L, et al; EMILIA Study Group. Trastuzumab emtansine for HER2-positive advanced breast cancer. N Engl J Med. 2012;367:1783-1791. 16. Burstein HJ, Sun Y, Dirix LY, et al. Neratinib, an irreversible ErbB receptor tyrosine kinase inhibitor, in patients with advanced ErbB2positive breast cancer. J Clin Oncol. 2010;28:1301-1307.

OPTIMAL MANAGEMENT OF HER2 BREAST CANCER

17. Awada A, Colomer R, Inoue K, et al. Neratinib plus pacl*taxel vs trastuzumab plus pacl*taxel in previously untreated metastatic ERBB2positive breast cancer: the NEfERT-T randomized clinical trial. JAMA Oncol. 2016;2:1557-1564. 18. Moulder-Thompson S, Borges VF, Baetz TD, et al. Phase 1 study of ONT-380, a HER2 inhibitor, in patients with HER2+ advanced solid tumors, with an expansion cohort in HER2+ metastatic breast cancer (MBC). Clin Cancer Res. Epub 2017 Jan 4. 19. Borges VF, Ferrario C, Aucoin N, et al. Efficacy results of a phase 1b study of ONT 380, a CNS penetrant TKI in combination with T-DM1 in HER2 + metastatic breast cancer including patients with brain metastases. J Clin Oncol. 2016;34 (suppl; abstr 513). 20. Baselga J, Gelmon KA, Verma S, et al. Phase II trial of pertuzumab and trastuzumab in patients with human epidermal growth factor receptor 2-positive metastatic breast cancer that progressed during prior trastuzumab therapy. J Clin Oncol. 2010;28:11381144. 21. Swain SM, Baselga J, Kim SB, et al; CLEOPATRA Study Group. Pertuzumab, trastuzumab, and docetaxel in HER2-positive metastatic breast cancer. N Engl J Med. 2015;372:724-734. 22. Baselga J, Cortés J, Im SA, et al. Biomarker analyses in CLEOPATRA: a phase III, placebo-controlled study of pertuzumab in human epidermal growth factor receptor 2-positive, first-line metastatic breast cancer. J Clin Oncol. 2014;32:3753-3761. 23. Krop IE, Beeram M, Modi S, et al. Phase I study of trastuzumab-DM1, an HER2 antibody-drug conjugate, given every 3 weeks to patients with HER2-positive metastatic breast cancer. J Clin Oncol. 2010;28:26982704. 24. Burris HA 3rd, Rugo HS, Vukelja SJ, et al. Phase II study of the antibody drug conjugate trastuzumab-DM1 for the treatment of human epidermal growth factor receptor 2 (HER2)-positive breast cancer after prior HER2-directed therapy. J Clin Oncol. 2011;29:398-405. 25. Wang J, Song P, Schrieber S, et al. Exposure-response relationship of T-DM1: insight into dose optimization for patients with HER2-positive metastatic breast cancer. Clin Pharmacol Ther. 2014;95:558-564. 26. Krop IE, Kim SB, González-Martín A, et al; TH3RESA study collaborators. Trastuzumab emtansine versus treatment of physician’s choice for pretreated HER2-positive advanced breast cancer (TH3RESA): a randomised, open-label, phase 3 trial. Lancet Oncol. 2014;15:689699. 27. Hurvitz SA, Dirix L, Kocsis J, et al. Phase II randomized study of trastuzumab emtansine versus trastuzumab plus docetaxel in patients with human epidermal growth factor receptor 2-positive metastatic breast cancer. J Clin Oncol. 2013;31:1157-1163. 28. Perez EA, Barrios C, Eiermann W, et al. Trastuzumab emtansine with or without pertuzumab versus trastuzumab plus taxane for human epidermal growth factor receptor 2-positive, advanced breast cancer: primary results from the phase III MARIANNE study. J Clin Oncol. 2017;35:141-148. 29. Miller K, Cortes J, Hurvitz SA, et al. HERMIONE: a randomized phase 2 trial of MM-302 plus trastuzumab versus chemotherapy of physician’s choice plus trastuzumab in patients with previously treated, anthracycline-naïve, HER2-positive, locally advanced/metastatic breast cancer. BMC Cancer. 2016;16:352. 30. Kiewe P, Hasmüller S, Kahlert S, et al. Phase I trial of the trifunctional anti-HER2 x anti-CD3 antibody ertumaxomab in metastatic breast cancer. Clin Cancer Res. 2006;12:3085-3091.

31. Haense N, Atmaca A, Pauligk C, et al. A phase I trial of the trifunctional anti Her2 × anti CD3 antibody ertumaxomab in patients with advanced solid tumors. BMC Cancer. 2016;16:420. 32. Konecny GE, Meng YG, Untch M, et al. Association between HER2/neu and vascular endothelial growth factor expression predicts clinical outcome in primary breast cancer patients. Clin Cancer Res. 2004;10:1706-1716. 33. Gianni L, Romieu GH, Lichinitser M, et al. AVEREL: a randomized phase III trial evaluating bevacizumab in combination with docetaxel and trastuzumab as first-line therapy for HER2-positive locally recurrent/ metastatic breast cancer. J Clin Oncol. 2013;31:1719-1725. 34. Modi S, Stopeck A, Linden H, et al. HSP90 inhibition is effective in breast cancer: a phase II trial of tanespimycin (17-AAG) plus trastuzumab in patients with HER2-positive metastatic breast cancer progressing on trastuzumab. Clin Cancer Res. 2011;17:5132-5139. 35. Huober J, Fasching PA, Barsoum M, et al. Higher efficacy of letrozole in combination with trastuzumab compared to letrozole monotherapy as first-line treatment in patients with HER2-positive, hormone-receptorpositive metastatic breast cancer - results of the eLEcTRA trial. Breast. 2012;21:27-33. 36. Kaufman B, Mackey JR, Clemens MR, et al. Trastuzumab plus anastrozole versus anastrozole alone for the treatment of postmenopausal women with human epidermal growth factor receptor 2-positive, hormone receptor-positive metastatic breast cancer: results from the randomized phase III TAnDEM study. J Clin Oncol. 2009;27:5529-5537. 37. Johnston S, Pippen J Jr, Pivot X, et al. Lapatinib combined with letrozole versus letrozole and placebo as first-line therapy for postmenopausal hormone receptor-positive metastatic breast cancer. J Clin Oncol. 2009;27:5538-5546. 38. Esteva FJ, Guo H, Zhang S, et al. PTEN, PIK3CA, p-AKT, and p-p70S6K status: association with trastuzumab response and survival in patients with HER2-positive metastatic breast cancer. Am J Pathol. 2010;177:1647-1656. 39. Hurvitz SA, Dalenc F, Campone M, et al. A phase 2 study of everolimus combined with trastuzumab and pacl*taxel in patients with HER2overexpressing advanced breast cancer that progressed during prior trastuzumab and taxane therapy. Breast Cancer Res Treat. 2013;141:437-446. 40. Hurvitz SA, Andre F, Jiang Z, et al. Combination of everolimus with trastuzumab plus pacl*taxel as first-line treatment for patients with HER2-positive advanced breast cancer (BOLERO-1): a phase 3, randomised, double-blind, multicentre trial. Lancet Oncol. 2015;16:816-829. 41. André F, O’Regan R, Ozguroglu M, et al. Everolimus for women with trastuzumab-resistant, HER2-positive, advanced breast cancer (BOLERO-3): a randomised, double-blind, placebo-controlled phase 3 trial. Lancet Oncol. 2014;15:580-591. 42. Jain S, Nye LE, Santa-Maria CA, et al. Phase I study of alpelisib and T-DM1 in trastuzumab refractory HER2 positive metastatic breast cancer. J Clin Oncol. 2016;34 (suppl; abstr 588). 43. Goel S, Wang Q, Watt AC, et al. Overcoming therapeutic resistance in HER2-positive breast cancers with CDK4/6 inhibitors. Cancer Cell. 2016;29:255-269. 44. Romond EH, Perez EA, Bryant J, et al. Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med. 2005;353:1673-1684.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 89

HURVITZ, GELMON, AND TOLANEY

45. Musolino A, Ciccolallo L, Panebianco M, et al. Multifactorial central nervous system recurrence susceptibility in patients with HER2positive breast cancer: epidemiological and clinical data from a population-based cancer registry study. Cancer. 2011;117:1837-1846. 46. Gianni L, Eiermann W, Semiglazov V, et al. Neoadjuvant and adjuvant trastuzumab in patients with HER2-positive locally advanced breast cancer (NOAH): follow-up of a randomised controlled superiority trial with a parallel HER2-negative cohort. Lancet Oncol. 2014;15:640-647. 47. von Minckwitz G, Rezai M, Fasching PA, et al. Survival after adding capecitabine and trastuzumab to neoadjuvant anthracyclinetaxane-based chemotherapy for primary breast cancer (GBG 40– GeparQuattro). Ann Oncol. 2014;25:81-89. 48. Joensuu H, Bono P, Kataja V, et al. Fluorouracil, epirubicin, and cyclophosphamide with either docetaxel or vinorelbine, with or without trastuzumab, as adjuvant treatments of breast cancer: final results of the FinHer Trial. J Clin Oncol. 2009;27:5685-5692. 49. Mackey JR, Pieńkowski T, Crown J, et al. Long-term outcomes after adjuvant treatment of sequential versus combination docetaxel with doxorubicin and cyclophosphamide in node-positive breast cancer: BCIRG-005 randomized trial. Ann Oncol. 2016;27:1041-1047. 50. Slamon DJ, Eiermann W, Robert NJ, et al. Ten year follow-up of BCIRG-006 comparing doxorubicin plus cyclophosphamide followed by docetaxel (AC→T) with doxorubicin plus cyclophosphamide followed by docetaxel and trastuzumab (AC→TH) with docetaxel, carboplatin and trastuzumab (TCH) in HER2+ early breast cancer. Cancer Res. 2015;76 (suppl; abstr S5-04). 51. Slamon D, Eiermann W, Robert N, et al; Breast Cancer International Research Group. Adjuvant trastuzumab in HER2-positive breast cancer. N Engl J Med. 2011;365:1273-1283. 52. Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al; Herceptin Adjuvant (HERA) Trial Study Team. Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med. 2005;353:1659-1672. 53. Perez EA, Romond EH, Suman VJ, et al. Trastuzumab plus adjuvant chemotherapy for human epidermal growth factor receptor 2-positive breast cancer: planned joint analysis of overall survival from NSABP B-31 and NCCTG N9831. J Clin Oncol. 2014;32:3744-3752. 54. Jackisch C, Piccart MJ, Gelber RD, et al. HERA TRIAL: 10-year follow up of trastuzumab after adjuvant chemotherapy in HER2 positive breast cancer — final analysis. Cancer Res. 2015;76 (suppl; abstr PD5-01). 55. Goldhirsch A, Gelber RD, Piccart-Gebhart MJ, et al; Herceptin Adjuvant (HERA) Trial Study Team. 2 years versus 1 year of adjuvant trastuzumab for HER2-positive breast cancer (HERA): an open-label, randomised controlled trial. Lancet. 2013;382:1021-1028. 56. Joensuu H, Kellokumpu-Lehtinen PL, Bono P, et al; FinHer Study Investigators. Adjuvant docetaxel or vinorelbine with or without trastuzumab for breast cancer. N Engl J Med. 2006;354:809-820. 57. Spielmann M, Roché H, Delozier T, et al. Trastuzumab for patients with axillary-node-positive breast cancer: results of the FNCLCC-PACS 04 trial. J Clin Oncol. 2009;27:6129-6134. 58. Perez EA, Suman VJ, Davidson NE, et al. Sequential versus concurrent trastuzumab in adjuvant chemotherapy for breast cancer. J Clin Oncol. 2011;29:4491-4497. 59. Pivot X, Romieu G, Debled M, et al; PHARE trial investigators. 6 months versus 12 months of adjuvant trastuzumab for patients with HER2positive early breast cancer (PHARE): a randomised phase 3 trial. Lancet Oncol. 2013;14:741-748.

90 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

60. Earl HM, Vallier AL, Dunn J, et al. Trastuzumab-associated cardiac events in the Persephone trial. Br J Cancer. 2016;115:1462-1470. 61. Chan A, Delaloge S, Holmes FA, et al; ExteNET Study Group. Neratinib after trastuzumab-based adjuvant therapy in patients with HER2positive breast cancer (ExteNET): a multicentre, randomised, doubleblind, placebo-controlled, phase 3 trial. Lancet Oncol. 2016;17:367377. 62. Moja L, Tagliabue L, Balduzzi S, et al. Trastuzumab containing regimens for early breast cancer. Cochrane Database Syst Rev. 2012;4:CD006243. 63. Romond EH, Jeong JH, Rastogi P, et al. Seven-year follow-up assessment of cardiac function in NSABP B-31, a randomized trial comparing doxorubicin and cyclophosphamide followed by pacl*taxel (ACP) with ACP plus trastuzumab as adjuvant therapy for patients with node-positive, human epidermal growth factor receptor 2-positive breast cancer. J Clin Oncol. 2012;30:3792-3799. 64. Perez EA, Suman VJ, Davidson NE, et al. Cardiac safety analysis of doxorubicin and cyclophosphamide followed by pacl*taxel with or without trastuzumab in the North Central Cancer Treatment Group N9831 adjuvant breast cancer trial. J Clin Oncol. 2008;26:12311238. 65. Advani PP, Ballman KV, Dockter TJ, et al. Long-term cardiac safety analysis of NCCTG N9831 (Alliance) adjuvant trastuzumab trial. J Clin Oncol. 2016;34:581-587. 66. Piccart-Gebhart MJ, Holmes AP, Baselga J, et al. First results from the phase III ALTTO trial (BIG 2-06; NCCTG [Alliance] N063D) comparing one year of anti-HER2 therapy with lapatinib alone (L), trastuzumab alone (T), their sequence (T→L), or their combination (T+L) in the adjuvant treatment of HER2-positive early breast cancer (EBC). J Clin Oncol. 2014;32 (suppl; abstr LBA4). 67. Schneeweiss A, Chia S, Hickish T, et al. Pertuzumab plus trastuzumab in combination with standard neoadjuvant anthracycline-containing and anthracycline-free chemotherapy regimens in patients with HER2-positive early breast cancer: a randomized phase II cardiac safety study (TRYPHAENA). Ann Oncol. 2013;24:2278-2284. 68. Hurvitz SA, Miller JM, Dichmann R, et al. Final analysis of a phase II 3-arm randomized trial of neoadjuvant trastuzumab or lapatinib or the combination of trastuzumab and lapatinib, followed by six cycles of docetaxel and carboplatin with trastuzumab and/or lapatinib in patients with HER2+ breast cancer (TRIO-US B07). Cancer Res. 2013;73 (suppl; abstr S1-02). 69. Slamon DJ, Swain SM, Buyse M, et al. Primary results from BETH, a phase 3 controlled study of adjuvant chemotherapy and trastuzumab ± bevacizumab in patients with HER2-positive, node-positive or high risk node-negative breast cancer. Cancer Res. 2013;73 (suppl; abstr S1-03). 70. Rimawi MF, Cecchini RS, Rastogi P, et al. A phase III trial evaluating pCR in patients with HR+, HER2-positive breast cancer treated with neoadjuvant docetaxel, carboplatin, trastuzumab, and pertuzumab (TCHP) +/- estrogen deprivation: NRG Oncology/NSABP B-52. Cancer Res. 2016;77 (suppl; abstr S3-06). 71. Gianni L, Eiermann W, Semiglazov V, et al. Neoadjuvant chemotherapy with trastuzumab followed by adjuvant trastuzumab versus neoadjuvant chemotherapy alone, in patients with HER2-positive locally advanced breast cancer (the NOAH trial): a randomised controlled superiority trial with a parallel HER2-negative cohort. Lancet. 2010;375:377-384. 72. Buzdar AU, Ibrahim NK, Francis D, et al. Significantly higher pathologic complete remission rate after neoadjuvant therapy with trastuzumab,

OPTIMAL MANAGEMENT OF HER2 BREAST CANCER

pacl*taxel, and epirubicin chemotherapy: results of a randomized trial in human epidermal growth factor receptor 2-positive operable breast cancer. J Clin Oncol. 2005;23:3676-3685. 73. Buzdar AU, Valero V, Ibrahim NK, et al. Neoadjuvant therapy with pacl*taxel followed by 5-fluorouracil, epirubicin, and cyclophosphamide chemotherapy and concurrent trastuzumab in human epidermal growth factor receptor 2-positive operable breast cancer: an update of the initial randomized study population and data of additional patients treated with the same regimen. Clin Cancer Res. 2007;13:228-233. 74. Coudert BP, Largillier R, Arnould L, et al. Multicenter phase II trial of neoadjuvant therapy with trastuzumab, docetaxel, and carboplatin for human epidermal growth factor receptor-2-overexpressing stage II or III breast cancer: results of the GETN(A)-1 trial. J Clin Oncol. 2007;25:2678-2684. 75. Untch M, Rezai M, Loibl S, et al. Neoadjuvant treatment with trastuzumab in HER2-positive breast cancer: results from the GeparQuattro study. J Clin Oncol. 2010;28:2024-2031. 76. Untch M, Fasching PA, Konecny GE, et al. Pathologic complete response after neoadjuvant chemotherapy plus trastuzumab predicts favorable survival in human epidermal growth factor receptor 2-overexpressing breast cancer: results from the TECHNO trial of the AGO and GBG study groups. J Clin Oncol. 2011;29:3351-3357. 77. Cortazar P, Zhang L, Untch M, et al. Pathological complete response and long-term clinical benefit in breast cancer: the CTNeoBC pooled analysis. Lancet. 2014;384:164-172. 78. Valachis A, Nearchou A, Lind P, et al. Lapatinib, trastuzumab or the combination added to preoperative chemotherapy for breast cancer: a meta-analysis of randomized evidence. Breast Cancer Res Treat. 2012;135:655-662. 79. Untch M, Loibl S, Bischoff J, et al; German Breast Group (GBG); Arbeitsgemeinschaft Gynäkologische Onkologie-Breast (AGO-B) Study Group. Lapatinib versus trastuzumab in combination with neoadjuvant anthracycline-taxane-based chemotherapy (GeparQuinto, GBG 44): a randomised phase 3 trial. Lancet Oncol. 2012;13:135-144. 80. Alba E, Albanell J, de la Haba J, et al. Trastuzumab or lapatinib with standard chemotherapy for HER2-positive breast cancer: results from the GEICAM/2006-14 trial. Br J Cancer. 2014;110:1139-1147.

85. Guarneri V, Frassoldati A, Bottini A, et al. Preoperative chemotherapy plus trastuzumab, lapatinib, or both in human epidermal growth factor receptor 2-positive operable breast cancer: results of the randomized phase II CHER-LOB study. J Clin Oncol. 2012;30:1989-1995. 86. Gianni L, Pienkowski T, Im YH, et al. Efficacy and safety of neoadjuvant pertuzumab and trastuzumab in women with locally advanced, inflammatory, or early HER2-positive breast cancer (NeoSphere): a randomised multicentre, open-label, phase 2 trial. Lancet Oncol. 2012;13:25-32. 87. Hurvitz SA, Martin M, Symmans WF, et al. Pathologic complete response (pCR) rates after neoadjuvant trastuzumab emtansine (T-DM1 [K]) + pertuzumab (P) vs docetaxel + carboplatin + trastuzumab + P (TCHP) treatment in patients with HER2-positive (HER2+) early breast cancer (EBC) (KRISTINE). J Clin Oncol. 2016;34 (suppl; abstr 500). 88. Rimawi MF, Mayer IA, Forero A, et al. Multicenter phase II study of neoadjuvant lapatinib and trastuzumab with hormonal therapy and without chemotherapy in patients with human epidermal growth factor receptor 2-overexpressing breast cancer: TBCRC 006. J Clin Oncol. 2013;31:1726-1731. 89. Harbeck N, Gluz O, Christgen M, et al. Efficacy of 12-weeks of neoadjuvant TDM1 with or without endocrine therapy in HER2positive hormone-receptor-positive early breast cancer: WSG-ADAPT HER2+/HR+ phase II trial. J Clin Oncol. 2015;33 (suppl; abstr 506). 90. Konecny GE, Pegram MD, Venkatesan N, et al. Activity of the dual kinase inhibitor lapatinib (GW572016) against HER-2-overexpressing and trastuzumab-treated breast cancer cells. Cancer Res. 2006;66:1630-1639. 91. Gomez HL, Doval DC, Chavez MA, et al. Efficacy and safety of lapatinib as first-line therapy for ErbB2-amplified locally advanced or metastatic breast cancer. J Clin Oncol. 2008;26:2999-3005. 92. Scaltriti M, Verma C, Guzman M, et al. Lapatinib, a HER2 tyrosine kinase inhibitor, induces stabilization and accumulation of HER2 and potentiates trastuzumab-dependent cell cytotoxicity. Oncogene. 2009;28:803-814. 93. Goss PE, Smith IE, O’Shaughnessy J, et al; TEACH investigators. Adjuvant lapatinib for women with early-stage HER2-positive breast cancer: a randomised, controlled, phase 3 trial. Lancet Oncol. 2013;14:88-96.

81. Baselga J, Bradbury I, Eidtmann H, et al; NeoALTTO Study Team. Lapatinib with trastuzumab for HER2-positive early breast cancer (NeoALTTO): a randomised, open-label, multicentre, phase 3 trial. Lancet. 2012;379:633-640.

94. Piccart-Gebhart M, Holmes E, Baselga J, et al. adjuvant lapatinib and trastuzumab for early human epidermal growth factor receptor 2-positive breast cancer: results from the randomized phase III adjuvant lapatinib and/or trastuzumab treatment optimization trial. J Clin Oncol. 2016;34:1034-1042.

82. de Azambuja E, Holmes AP, Piccart-Gebhart M, et al. Lapatinib with trastuzumab for HER2-positive early breast cancer (NeoALTTO): survival outcomes of a randomised, open-label, multicentre, phase 3 trial and their association with pathological complete response. Lancet Oncol. 2014;15:1137-1146.

95. Gianni L, Pienkowski T, Im YH, et al. 5-year analysis of neoadjuvant pertuzumab and trastuzumab in patients with locally advanced, inflammatory, or early-stage HER2-positive breast cancer (NeoSphere): a multicentre, open-label, phase 2 randomised trial. Lancet Oncol. 2016;17:791-800.

83. Robidoux A, Tang G, Rastogi P, et al. Lapatinib as a component of neoadjuvant therapy for HER2-positive operable breast cancer (NSABP protocol B-41): an open-label, randomised phase 3 trial. Lancet Oncol. 2013;14:1183-1192.

96. Untch M, Jackisch C, Schneeweiss A, et al; German Breast Group (GBG); Arbeitsgemeinschaft Gynäkologische Onkologie—Breast (AGO-B) Investigators. Nab-pacl*taxel versus solvent-based pacl*taxel in neoadjuvant chemotherapy for early breast cancer (GeparSeptoGBG 69): a randomised, phase 3 trial. Lancet Oncol. 2016;17:345-356.

84. Carey LA, Berry DA, Cirrincione CT, et al. Molecular heterogeneity and response to neoadjuvant human epidermal growth factor receptor 2 targeting in CALGB 40601, a randomized phase III trial of pacl*taxel plus trastuzumab with or without lapatinib. J Clin Oncol. 2016;34:542549.

97. Loi S, Dafni U, Karlis D, et al. Effects of estrogen receptor and human epidermal growth factor receptor-2 levels on the efficacy of trastuzumab: a secondary analysis of the HERA trial. JAMA Oncol. 2016;2:1040-1047.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 91

HURVITZ, GELMON, AND TOLANEY

98. Loibl S, Majewski I, Guarneri V, et al. PIK3CA mutations are associated with reduced pathological complete response rates in primary HER2positive breast cancer: pooled analysis of 967 patients from five prospective trials investigating lapatinib and trastuzumab. Ann Oncol. 2016;27:1519-1525. 99. Denkert C, von Minckwitz G, Darb-Esfahani S, et al. Evaluation of tumor-infiltrating lymphocytes (TILs) as predictive and prognostic biomarkers in different subtypes of breast cancer treated with neoadjuvant therapy—a meta-analysis of 3771 patients. Cancer Res. 2017;77 (suppl; abstr S1-09). 100. Perez EA, Ballman KV, Tenner KS, et al. Association of stromal tumorinfiltrating lymphocytes with recurrence-free survival in the n9831 adjuvant trial in patients with early-stage HER2-positive breast cancer. JAMA Oncol. 2016;2:56-64. 101. Hanrahan EO, Gonzalez-Angulo AM, Giordano SH, et al. Overall survival and cause-specific mortality of patients with stage T1a,bN0M0 breast carcinoma. J Clin Oncol. 2007;25:4952-4960. 102. Chia S, Norris B, Speers C, et al. Human epidermal growth factor receptor 2 overexpression as a prognostic factor in a large tissue microarray series of node-negative breast cancers. J Clin Oncol. 2008;26:5697-5704. 103. Joensuu H, Isola J, Lundin M, et al. Amplification of erbB2 and erbB2 expression are superior to estrogen receptor status as risk factors for distant recurrence in pT1N0M0 breast cancer: a nationwide population-based study. Clin Cancer Res. 2003;9:923-930. 104. Tovey SM, Brown S, Doughty JC, et al. Poor survival outcomes in HER2-positive breast cancer patients with low-grade, node-negative tumours. Br J Cancer. 2009;100:680-683. 105. Chen J, Long JB, Hurria A, et al. Incidence of heart failure or cardiomyopathy after adjuvant trastuzumab therapy for breast cancer. J Am Coll Cardiol. 2012;60:2504-2512. 106. Rom J, Schumacher C, Gluz O, et al. Association of HER2 Overexpression and Prognosis in Small (T1N0) Primary Breast Cancers. Breast Care (Basel). 2013;8:208-214. 107. Fehrenbacher L, Capra AM, Quesenberry CP Jr, et al. Distant invasive breast cancer recurrence risk in human epidermal growth factor receptor 2-positive T1a and T1b node-negative localized breast cancer diagnosed from 2000 to 2006: a cohort from an integrated health care delivery system. J Clin Oncol. 2014;32:2151-2158.

92 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

108. Gonzalez-Angulo AM, Litton JK, Broglio KR, et al. High risk of recurrence for patients with breast cancer who have human epidermal growth factor receptor 2-positive, node-negative tumors 1 cm or smaller. J Clin Oncol. 2009;27:5700-5706. 109. Curigliano G, Viale G, Bagnardi V, et al. Clinical relevance of HER2 overexpression/amplification in patients with small tumor size and node-negative breast cancer. J Clin Oncol. 2009;27:5693-5699. 110. Vaz-Luis I, Ottesen RA, Hughes ME, et al. Outcomes by tumor subtype and treatment pattern in women with small, node-negative breast cancer: a multi-institutional study. J Clin Oncol. 2014;32:2142-2150. 111. Tolaney SM, Barry WT, Dang CT, et al. Adjuvant pacl*taxel and trastuzumab for node-negative, HER2-positive breast cancer. N Engl J Med. 2015;372:134-141. 112. Ruddy KJ, Guo H, Barry W, et al. Chemotherapy-related amenorrhea after adjuvant pacl*taxel-trastuzumab (APT trial). Breast Cancer Res Treat. 2015;151:589-596. 113. Abusief ME, Missmer SA, Ginsburg ES, et al. The effects of pacl*taxel, dose density, and trastuzumab on treatment-related amenorrhea in premenopausal women with breast cancer. Cancer. 2010;116:791798. 114. Jones SE, Collea R, Paul D, et al. Adjuvant docetaxel and cyclophosphamide plus trastuzumab in patients with HER2-amplified early stage breast cancer: a single-group, open-label, phase 2 study. Lancet Oncol. 2013;14:1121-1128. 115. Freedman RA, Hughes ME, Ottesen RA, et al. Use of adjuvant trastuzumab in women with human epidermal growth factor receptor 2 (HER2)-positive breast cancer by race/ethnicity and education within the National Comprehensive Cancer Network. Cancer. 2013;119:839846. 116. Chavez-MacGregor M, Zhang N, Buchholz TA, et al. Trastuzumabrelated cardiotoxicity among older patients with breast cancer. J Clin Oncol. 2013;31:4222-4228. 117. Arpino G, Ferrero JM, de la Haba-Rodriguez J, et al. Primary analysis of PERTAIN: a randomized, two-arm, open-label, multicenter phase II trial assessing the efficacy and safety of pertuzumab given in combination with trastuzumab plus an aromatase inhibitor in first-line patients with HER2-positive and hormone receptor-positive metastatic or locally advanced breast cancer. Presented at: San Antonio Breast Cancer Symposium. San Antonio, TX; 2016. Abstract S3-04.

OPTIMIZING BREAST CANCER RADIOTHERAPY AND SURGICAL OUTCOMES

Optimizing Breast Cancer Adjuvant Radiation and Integration of Breast and Reconstructive Surgery Henry M. Kuerer, MD, PhD, Peter G. Cordeiro, MD, and Robert W. Mutter, MD OVERVIEW Postmastectomy radiotherapy (PMRT) reduces the risk of locoregional and distant recurrence and improves overall survival in women with lymph node–positive breast cancer. Because of stage migration and improvements in systemic therapy and other aspects of breast cancer care, the absolute benefit of PMRT and regional nodal irradiation may be small in some favorable subsets of patients with very low nodal burden, and newer consensus guidelines do not mandate PMRT in all node-positive cases. The use and need for PMRT may considerably complicate breast reconstruction after mastectomy and therefore mandates multidisciplinary input that takes into account patient choice given potential risk of acute and long-term toxicities, benefits, life expectancy, the biology of the tumor, plans for systemic therapy, and actual tumor burden. Management of axillary lymph node metastases is changing with selective use of axillary lymph node dissection for advanced disease, sentinel lymph node biopsy alone for clinically and pathologic node-negative cases receiving mastectomy, and targeted axillary dissection alone among patients with eradication of initial biopsy-proven nodal metastases with neoadjuvant systemic therapy use. In general, when the need for PMRT is anticipated, autologous reconstruction should be delayed. This comprehensive article reviews the current indications and implications regarding integration of breast cancer surgery and timing of reconstruction with optimum radiation delivery to achieve the best possible patient outcomes.

n contemporary practice, there is uniform consensus that PMRT is indicated for patients at high risk of local regional failure, such as those with stage III breast cancers. However, locoregional and distant recurrence rates are lower than in past decades when randomized controlled trials demonstrated overall survival benefits in all patients with lymph node–positive disease.1-3 Therefore, there is much controversy on the role of PMRT in patients with earlier stage breast cancer, particularly among those with low-volume nodal metastases identified with sentinel lymph node (SLN) dissection. Added to this treatment conundrum, recent evidence suggests that regional nodal irradiation (RNI) in itself may provide a survival benefit in patients with early-stage breast cancer despite only modest reductions in locoregional recurrence.4-6 These RNI studies highlight the importance of long-term follow-up of prospective studies to fully assess the impact of locoregional therapies on breast cancer–specific and all-cause mortality. At the same time, stage migration from increased screening and improved diagnostic imaging, in addition to advances in systemic therapy, surgical techniques, pathologic evaluation, and radiotherapy delivery, all must be taken into consideration when applying the results from past locoregional studies to patients assessed in the clinic today.7 These advances add complexity to counseling

regarding the absolute risks and benefits of PMRT for each individual patient. In this context, new consensus guidelines related to the use of PMRT were recently released by the ASCO, the American Society for Radiation Oncology (ASTRO), and the Society of Surgical Oncology (SSO) to provide additional guidance on some continued areas of controversy, including the role of PMRT in patients with one to three positive lymph nodes, the role of PMRT in the setting of preoperative chemotherapy, as well as technical aspects of PMRT such as indications for internal mammary node irradiation (IMNI).2 Coupled with great institutional variability in applying PMRT and RNI guidelines is the complexity of integrating breast and nodal surgery together with plastic reconstructive surgery. Breast cancer nodal metastases plays a considerable role in determining radiotherapy indications and treatment targeting, yet recently there have also been marked changes in axillary management, starting with SLN biopsy for clinically node–negative disease and newer techniques to stage the axilla after neoadjuvant systemic therapy in node-positive breast cancer and new trial results related to use of adjuvant radiotherapy without completion dissection for SLN-positive breast cancer. Finally, patient decisions regarding whether to pursue immediate reconstruction

From the Department of Breast Surgical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX; Plastic and Reconstructive Surgery Service, Memorial Sloan Kettering Cancer Center, New York, NY; Department of Radiation Oncology, Mayo Clinic, Rochester, MN. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Henry M. Kuerer, MD, PhD, The University of Texas MD Anderson Cancer Center, 1400 Pressler St., Unit 1434, Houston, TX 77030; email: hkuerer@ mdanderson.org. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 93

KUERER, CORDEIRO, AND MUTTER

may be impacted by a recommendation for PMRT, as PMRT may increase the risk of complications and adverse cosmetic outcome, and immediate reconstruction has been reported to pose challenges to PMRT delivery.8-10 All of these factors underscore the imperative of close communication by multidisciplinary teams to best prospectively coordinate and deliver patient-centered breast cancer care. This special ASCO educational article will address the current indications and implications regarding integration of breast cancer surgery with optimum radiation delivery to achieve the best possible patient outcomes.

INDICATIONS AND IMPLICATIONS FOR OPTIMAL CONTEMPORARY ADJUVANT RADIATION THERAPY DELIVERY

The PMRT Randomized Controlled Trials

In the period from 1978 to 1990, three seminal randomized controlled trials (Table 1) were conducted (one in British Columbia, Canada, and two in Denmark) that evaluated the role of PMRT in patients receiving systemic therapy.11-13 Eligibility criteria included one or more pathologically involved axillary lymph node. In addition, approximately 8% and 10% of patients in the premenopausal Danish Breast Cancer Cooperative Group (DBCG) 82b study and the postmenopausal DBCG 82c study, respectively, were node negative and enrolled on the basis of a primary tumor more than 5 cm or

KEY POINTS • Prospective multidisciplinary review, communication, and coordination is necessary to optimize breast cancer local-regional control, survival, cosmetic outcome, and define a unified clear patient-centered path forward. • PMRT reduces the risk of locoregional and distant recurrence and improves overall survival in women with lymph node–positive breast cancer. Because of stage migration and improvements in systemic therapy and other aspects of breast cancer care, the absolute benefit of PMRT may be small in some favorable subsets of patients with very low nodal burden. • PMRT use may complicate breast reconstruction and requires multidisciplinary input that takes into account patient choice given potential risk of acute and longterm toxicities, benefits, life expectancy, the biology of the tumor, plans for systemic therapy, and actual tumor burden. • Management of axillary lymph node metastases is changing with selective use of axillary lymph node dissection for advanced disease, sentinel lymph node biopsy alone, and targeted axillary dissection alone among patients with eradication of initial biopsy-proven nodal metastases with neoadjuvant systemic therapy. • Immediate two-stage implant-based reconstruction is usually preferable in the majority of patients with breast cancer facing PMRT due to its preservation of autologous tissue and often acceptable outcomes, and PMRT can be administered either to the tissue expander or the final implant. 94 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

invasion of the skin or pectoral fascia. Each study reported similar findings, a 9% to 10% absolute improvement in 10year overall survival with the addition of PMRT. However, there has remained heterogeneity in the uptake of PMRT, particularly in the subset of women with one to three positive nodes. Some physicians have favored offering PMRT in the majority of women meeting eligibility criteria (i.e., one or more positive lymph nodes) based on the high level of evidence provided by these prospective randomized controlled trials.14 Others highlighted that inadequate axillary surgery and inadequate systemic therapy regimens without taxanes and anti-HER2 therapy have limited applicability to patients with low nodal burden in modern practice who have much lower recurrence rates.15,16 In the DBCG 82b and 82c studies, a median of just seven lymph nodes were identified in pathologic specimens from axillary dissections. In the British Columbia study, the median number was 11. To address these concerns, the DBCG performed an analysis of patients with one to three positive lymph nodes from the DBCG 82b and 82c studies but in which they excluded patients with fewer than eight nodes removed. That analysis demonstrated a statistically significant overall survival benefit with the addition of PMRT (57% vs. 48%; p = .03).14 In 2014, the Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) published an individual patient data meta-analysis on the effects of PMRT. In it, they specifically assessed the role of PMRT in patients who underwent axillary lymph node dissection. For the 1,133 women with one to three positive nodes who underwent axillary lymph node dissection and received systemic therapy, PMRT reduced the 20-year breast cancer mortality rate from 49.4% to 41.5% (relative risk 0.78; p = .01).17 These findings lead to unanimous agreement by an ASCO/ASTRO/SSO expert panel in 2016 that PMRT reduces recurrence and breast cancer mortality for patients with T1-T2 breast cancer with one to three positive axillary lymph nodes.18 However, there was also recognition by the ASCO/ASTRO/SSO panel that patients with T1-T2 breast cancer and one to three positive nodes are a heterogeneous group with varying prognoses. They highlighted that subsets of patients in this population are likely to have such a low recurrence risk that the benefit of PMRT may be outweighed by the potential risks.18 Indeed, improvements in systemic therapy have significantly reduced both locoregional and distant recurrence risk because the three landmark randomized PMRT studies were conducted. It is noteworthy that patients included in the aforementioned EBCTCG analysis received cyclophosphamide, methotrexate, and fluorouracil (CMF), or tamoxifen, and the duration of tamoxifen for most patients was just 1 year.17 These agents and schedules have since been replaced by more effective strategies, including anthracyclines, taxanes, HER2-targeted therapies, and prolonged endocrine therapy, frequently with aromatase inhibitors. In addition, the introduction of SLN dissection has resulted in the identification of smaller volume axillary macro- and micrometastases.19 Therefore, improvements in both multidisciplinary management and stage migration have reduced

1982–1989

1982–1990

DBCG 82b

DBCG 82c

1996–2004

1979–1986

British-Columbia

EORTC 22922/10925

Dates

Study

4,004

1,375

1,708

318

No. of Patients

Stage I, II, or III with a centrally or medially located primary tumor, irrespective of axillary involvement, or pN+ and externally located

Postmenopausal, age < 70, pN+, primary tumor > 5 cm, invasion of skin or pectoral fascia

Premenopausal, pN+, or primary tumor > 5 cm, invasion of skin or pectoral fascia

Premenopausal, pN+

Inclusion Criteria

According to institutional practice; of patients enrolled, 25% received chemotherapy, 30% received hormonal therapy, and 30% received both.

Tamoxifen for 1 yr

CMF eight to nine cycles

CMF for 6–12 months

Systemic Therapy

TABLE 1. Select PMRT and RNI Randomized Controlled Trials

Breast or with/ without chest wall vs. breast or with/without chest wall and RNI (axillary, SC, and IMNs)

Chest wall, axillary, SC and IMNs

Chest wall, axillary, SC, and bilateral IMNs

Radiotherapy Target

Locoregional events 8 (RNI) vs. 10% (no RNI)

Isolated LRR alone as first event 4 (RT) vs. 29% (no RT)

Isolated LRR alone as first event 5 (RT) vs. 26% (no RT)

20-yr ILRFS, 90 (RT) vs. 74% (no RT); p = .002. For 1–3 positive nodes, 91 vs. 79%

Locoregional Recurrence

10-yr DDFS 78 (RNI) vs. 75% (no RNI); p = .02

Distant metastases alone as first event 25 (RT) vs. 39% (no RT)

Distant metastases alone as first event 34 (RT) vs. 26% (no RT)

20-yr SBCFS, 48 (RT) vs. 31% (no RT), p = .004. For 1–3 positive nodes, 58 vs. 44%

Distant Recurrence

10-yr DFS 72 (RNI) vs. 69% (no RNI); p = .004

10-year DFS, 36 (RT) vs. 24% (no RT); p < .001. For 1 3 positive nodes 44% vs. 31%

10-year DFS, 48 (RT) vs. 34% (no RT); p < .001. For 1–3 positive nodes, 54 vs. 39%

20-yr BCFS, 48 (RT) vs. 30% (no RT); p = .001. For 1–3 positive nodes, 57 vs. 41%

Disease-Free Survival

10-yr breast cancer mortality 12 (RNI) vs. 14% (no RNI); p = .02

Not reported

20-yr BCSS, 53 (RT) vs. 38% (no RT); p = .008. For 1–3 positive nodes 64 vs. 53%

Breast Cancer– Specific Survival

Continued

10-yr OS, 82 (RNI) vs. 81% (no RNI); p = .06

10-yr OS, 45 (RT) vs. 35% (no RT); p = .03. For 1–3 positive nodes, 55 vs. 44%

10-yr OS, 54 (RT) vs. 45% (no RT); p < .001. For 1–3 positive nodes, 62 vs. 54%

20-yr OS, 47 (RT) vs. 37% (no RT); p = .03. For 1–3 positive nodes, 57 vs. 50%

Overall Survival

OPTIMIZING BREAST CANCER RADIOTHERAPY AND SURGICAL OUTCOMES

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 95

96 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

2000–2007

Dates

1,832

No. of Patients pN+ or primary tumor ±5 cm or primary tumor ±2 cm, < 10 axillary nodes removed, and one of grade 3, ER negativity, or lymphovascular invasion

Inclusion Criteria According to institutional practice; of patients enrolled. 86% received an anthracycline; 25% received both an anthracycline and taxane; 57% received an aromatase inhibitor, and 19% received tamoxifen. After June 2005, trastuzumab was recommended for patients with HER2-positive disease.

Systemic Therapy

Locoregional Recurrence 10-yr ILRFS, 95 (RNI) vs. 92% (no RNI); p = .009

Radiotherapy Target Breast vs. breast plus RNI (axillary, SC, and IMNs)

10-yr DDFS, 86 (RNI) vs. 82% (no RNI); p = .03

Distant Recurrence 10-yr DFS. 82, (RNI) vs. 77% (no RNI); p = .01

Disease-Free Survival 10-yr breast cancer mortality, 10 (RNI) vs. 12% (no RNI); p = .11

Breast Cancer– Specific Survival

10-yr OS, 83 (RNI) vs. 82% (no RNI); p = .38

Overall Survival

Abbreviations: BCFS, breast cancer–free survival; BCSS, breast cancer–specific survival; CMF, cyclophosphamide, methotrexate, and fluorouracil; DDFS, distant disease-free survival; DFS, disease-free survival; ER, estrogen receptor; ILRFS, isolated local recurrence-free survival; IMNS, internal mammary node; LRR, locoregional recurrence; OS, overall survival; pN+, pathologically node positive; RNI, regional nodal irradiation; RT, radiotherapy; SBCFS, systemic breast cancer–free survival; SC, supraclavicular; yr, year.

NCIC MA.20

Study

TABLE 1. Select PMRT and RNI Randomized Controlled Trials (Cont'd)

KUERER, CORDEIRO, AND MUTTER

OPTIMIZING BREAST CANCER RADIOTHERAPY AND SURGICAL OUTCOMES

the absolute risk of recurrence in patients with one to three positive lymph nodes considered for PMRT compared with years past and with it the absolute risk reduction that can likely be expected for some subsets with the addition of PMRT.20-22 Yet more recently published prospective studies still demonstrated improvements in the therapeutic ratio with the addition of RNI (i.e., high axillary, supraclavicular, and internal mammary irradiation) in more contemporary treated populations with low axillary nodal burden.

IMNI, RNI, and the Relationship Between Locoregional and Distant Relapse

The National Cancer Institute of Canada MA.20 clinical trial assessed the role of the addition of RNI to whole breast irradiation (WBI) in women after breast-conserving surgery.4 Fifty percent of the study population consisted of women with just one positive axillary lymph node, and 85% had one to three positive nodes. An additional 10% of women had highrisk node-negative disease. Eighty-six percent of patients received anthracycline-based chemotherapy, and 26% also received a taxane. Both tamoxifen and aromatase inhibitors were administered according to institutional practice, with 57% receiving an aromatase inhibitor alone or after a period of tamoxifen and 19% receiving tamoxifen alone. HER2-directed therapy was only recommended in the final 20 months that the study was open. MA.20 did not meet its primary endpoint of a 5% improvement in 5-year survival. Moreover, the 2% absolute improvement in 10-year rate of breast cancer mortality with the addition of RNI to WBI did not reach significance (hazard ratio 0.80; p = .11). However, RNI significantly improved the 10-year rate of disease-free survival from 77.0% to 82.0% (hazard ratio 0.76; p = .01). Interestingly, the absolute improvement in the rate of 10-year distant disease-free survival with the addition of RNI to WBI was 4.0%, greater than the 3.0% improvement in 10-year isolated locoregional disease-free survival. In the same journal issue, the European Organization for Research and Treatment of Cancer (EORTC) published results of EORTC 22922/10925, which also evaluated the role of RNI in patients with early-stage breast cancer.5 The study population was slightly different than MA.20, with 44% of the population being node negative and 24% having undergone mastectomy. In this larger study, the 2% absolute improvement in breast cancer mortality with the addition of RNI reached statistical significance (hazard ratio 0.82; p = .02). Similar to MA.20, a provocative finding was that RNI prevented more distant events than locoregional events. MA.20 and EORTC 22922/10925 evaluated the role of comprehensive RNI, including axillary, supraclavicular, and IMNI. Despite IMNI being a component of PMRT in 20 of the 22 trials in the EBCTCG analysis, the need for IMNI has been questioned, as the risk of isolated nodal failures in the IMNs has historically been reported to be 1% or less.5,15 In addition, targeting the IMNs increases the dose to the heart and lungs, raising concern that IMNI could increase the late effects of treatment.23 DBCG-IMN was a prospective population-based cohort study that specifically evaluated

the effect of IMNI in patients with node-positive early-stage breast cancer.6 All patients with right-sided disease were allocated to IMNI, whereas patients with left-sided disease received no IMNI to minimize cardiac exposure. The majority of patients enrolled had one to three positive nodes. With a median follow-up of 8.9 years, right-sided patients who received IMNI had an 8-year overall survival rate of 75.9% versus 72.2% for left-sided patients treated without IMNI (hazard ratio 0.82; p = .005). The results of these three studies evaluating nodal irradiation contradict a highly cited principle put forth in an earlier EBCTCG analysis of trials initiated between 1951 and 1991 that differences in radiotherapy and extent of surgery that result in a less than 10% difference in 5-year local recurrence risk are unlikely to impact breast cancer mortality.24 Given the relatively high number of distant events prevented, the findings suggest that clinically substantial residual locoregional disease may go undetected or be only detected after a distant relapse has occurred. Therefore, caution should be exercised if de-escalating radiotherapy based on retrospective locoregional patterns of failure data alone.16,25,26 It is worth noting that most clinically detected locoregional recurrences after mastectomy occur on the chest wall, not in the regional lymphatics.15,27 The data from MA.20 and EORTC 22922/10925 is not directly applicable to PMRT because the majority of patients in these studies underwent breast conservation therapy. However, because PMRT typically includes both chest wall and RNI, it is reasonable to infer that if patients with similar disease features were treated with mastectomy, the proportional and absolute reduction in recurrence with PMRT would have been at least as great. At the same time, favorable long-term breast cancer event rates have been reported in single-institution retrospective analyses of well-staged and carefully selected patients (a majority with T1, estrogen receptor–positive, and a single involved axillary micro- or macrometastases) treated with mastectomy and systemic therapy without PMRT.1,28 In women with one to three positive lymph nodes being considered for PMRT, radiation oncologists must carefully consider individual patient and clinical features that not only influence the risk of locoregional relapse, but also distant relapse. Factors such as patient age, tumor size, number, size, and percentage of sampled nodes involved, grade, subtype, proximity of margins, and molecular profiling, if available, may all assist in estimating a patient’s risk of recurrence and assist in identifying those with low nodal burden who are most likely to benefit from PMRT.4,5,16,25,27,29-33

PMRT in the Setting of Neoadjuvant Chemotherapy

Increasingly, patients are being treated with preoperative systemic therapy, and for most breast cancer subtypes, tumor response is the most important prognostic factor for recurrence in that setting.34 Patients with residual disease in the lymph nodes after preoperative chemotherapy are at elevated risk of recurrence, and it is generally agreed that these patients should be treated with PMRT.35,36 Patients who have a complete response in the breast but residual asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 97

KUERER, CORDEIRO, AND MUTTER

disease in the axillary nodes after preoperative chemotherapy have a similar risk of recurrence as those with residual disease in both the breast and nodes and should also receive PMRT.37 Whether such an approach is also appropriate for low-grade, slowly proliferating estrogen receptor–positive tumors (i.e., luminal A tumors) in whom preoperative endocrine therapy approaches are increasingly considered is uncertain and warrants further investigation.38 Although retrospective analyses suggest locoregional recurrence risk is low, there is a dearth of prospective data on the benefits of administration or the safety of omitting PMRT in patients with lymph node–positive clinical stage II breast cancer who are converted to node negative after preoperative systemic therapy or who achieve a pathologic complete response in both the breast and axillary nodes.39,40 The NRG Oncology Group 9353 trial randomizes patients with biopsy-proven axillary node involvement before preoperative chemotherapy who become pathologically node negative at the time of mastectomy to PMRT or no irradiation. In patients who undergo lumpectomy, the randomization is to WBI versus whole breast plus RNI (NCT01872975). Eligible patients are best treated as part of this clinical trial.

Toxicity of Radiotherapy in Modern Practice

Finally, the potential benefits in disease control with the administration of radiotherapy must be weighed with the risks of toxicity in each individual patient and take into consideration their values on minimizing treatment morbidity versus avoiding recurrence. For example, complications of PMRT are higher in women pursuing implant-based reconstruction (discussed below), as well as those who have previously undergone axillary lymph node dissection, relative to SLN biopsy.41-43 Although toxicity outcomes of patients treated in past eras provide valuable lessons on the importance of optimizing radiotherapy delivery, the multidisciplinary team must be familiar with the expected acute and late toxicity of PMRT using the techniques and technology of today, not of decades past.44,45 For example, a greater appreciation for the potential risks of low-dose cardiac irradiation and improved radiotherapy planning and delivery have led to much lower cardiac exposure in women undergoing radiotherapy today.45 Proton therapy is being investigated for breast cancer and routinely enables heart, lung, and other nontarget normal tissue doses to be significantly lower than optimized photon and/or electron techniques and can further improve targeting of the IMNs in patients with challenging anatomy, such as those with bilateral reconstruction (Fig. 1).46-50 Therefore, there is strong justification to support clinical trials and prospective registries investigating whether these newer techniques can further improve long-term outcomes.

PMRT Targeting and Techniques

PMRT should be delivered to both the chest wall and regional lymphatics, including the IMNs, provided that appropriate normal tissue constraints can be met.18 At the Mayo Clinic in Rochester, MN, 50 Gy in 25 fractions is prescribed. We do not routinely boost the chest wall given concern about the 98 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

potential additional risk to the reconstructed breast mound.42 For photon PMRT, in cases of bilateral reconstruction, we routinely deflate the contralateral expander before CT simulation to avoid exposure from partially wide tangents that are generally used to target the IMNs. An added advantage of proton therapy is that contralateral tissue expander deflation is not necessary because protons are administered from anteriorly directed beams.46 For both photon and proton PMRT, the ipsilateral expander is overinflated before simulation to facilitate the second stage of reconstruction and is maintained in the same state during treatment to ensure reproducibility of the radiotherapy plan. Evidence suggests that proton therapy further improves targeting in the setting of immediate reconstruction.46,48,50 In patients undergoing WBI, hypofractionation resulted in less acute and late toxicity than conventional fraction.51 Whether hypofractionated schedules can further improve the therapeutic ratio in patients undergoing PMRT, including those with reconstruction, is an important area of future investigation. In summary, the absolute benefit of PMRT in some subsets of women with node-positive breast cancer is likely smaller in today’s practice because of a lower baseline risk of recurrence. However, the relative benefit of PMRT may be greater because of improved systemic therapy, resulting in less risk of early systemic dissemination, better targeting of areas at risk with modern treatment planning, and reductions in dose to nontarget normal tissues. Thus, careful multidisciplinary consideration of individual risk factors for recurrence, toxicity, and patient values is crucial to optimize patient counseling. Investment by all stakeholders in locoregional therapy randomized studies will be required to address many of the controversies that persist.

MANAGEMENT OF THE AXILLA IN PATIENTS WHO UNDERGO MASTECTOMY AND BREAST CONSERVATION

There have been many changes regarding the surgical management of node-positive breast cancer with respect to axillary surgery over the last 5 to 10 years. However, every day, all clinicians initially evaluating patients with breast cancer, and specifically the breast surgical oncologist, evaluate routine complexities that are discussed, including the need for or choice of mastectomy or choice of breast conservation. Further confounding these discussions is the potential for neoadjuvant systemic therapy in decisions regarding ultimate surgery and implications for PMRT. It becomes obvious and intuitive that these complex decisions are best made prospectively by a multidisciplinary team including the breast surgical oncologist, the radiation oncologist, the plastic surgeon, and the medical oncologists in conjunction, of course, with the patient.

SLN Biopsy for Clinically Node-Negative Patients

Simply stated, intraoperative lymphatic mapping and SLN biopsy for breast cancer is the accepted international standard for evaluating patients with a clinical node-negative axilla whether the patient receives breast-conserving therapy or mastectomy with or without immediate breast reconstruction.52

OPTIMIZING BREAST CANCER RADIOTHERAPY AND SURGICAL OUTCOMES

FIGURE 1. Axial (A), Coronal (B), and Sagittal (C) Colorwash Images of the Intensity-Modulated Proton Therapy Plan of a Patient With Bilateral Tissue Expander Reconstruction Undergoing PMRT A

Chest wall (magenta) and internal mammary (blue) clinical target volume is well covered while sparing the heart (red) and contralateral reconstructed breast mound.

Before recently, it was more or less dogmatic that a patient with node-positive breast cancer and receiving mastectomy would be a candidate for PMRT. This concept is changing based on new consensus guidelines taking into account tumor biology as well as tumor burden such that any given patient with one or two positive SLNs may not be recommended PMRT.2 However, despite these new guidelines, most patients who are younger with node-positive breast cancer and those patients with larger primary tumors and other risk factors will in fact be recommended for PMRT.3 Currently, ultrasound with biopsy of suspicious axillary nodes is an excellent methodology to identify patients with axillary nodal metastases and can be quite valuable in

making decisions regarding the use of neoadjuvant systemic therapy and planning for PMRT and reconstruction. Despite the utility of ultrasound, most imaging is neither sensitive nor specific for definitively identifying breast cancer nodal metastases.53 A negative nodal ultrasound does not rule out metastatic carcinoma, and this has led some clinicians to perform a separate SLN procedure if the status of the lymph node would change systemic therapy sequencing and/or the type of reconstruction based on the potential need for PMRT. This has not been the MD Anderson approach— patients with a benign nodal ultrasound but with large primary tumors would either undergo preoperative systemic therapy if indicated or primary surgery and SLN biopsy, asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 99

KUERER, CORDEIRO, AND MUTTER

usually with a tissue implant that could then easily be deflated for delivery of the PMRT if indicated and then be expanded for final reconstruction.54

Unsuspected Nodal Micrometastases Identified in SLNs

Patients receiving breast conservation and in whom unsuspected nodal metastases is identified in one or two lymph nodes do not have formal axillary lymph node dissection based on American College of Surgeons Oncology Group Z0011 randomized trial of axillary dissection versus observation for SLN-positive disease.55 For those with more nodal metastases identified receiving breast conservation, a multidisciplinary decision is made for recommendation of axillary lymph node dissection as the current standard, although some patients decline dissection, and groups of clinicians do offer nodal radiotherapy for this group of patients if axillary dissection is not performed. For unsuspected nodal micrometastases in patients with mastectomy, the standard has been recommendation for completion axillary lymph node dissection, although many patients decline this surgery and opt for observation, particularly when systemic therapy is given with or without PMRT.56 In this regard, although the National Comprehensive Cancer Network guidelines suggest that axillary radiotherapy could be used in this scenario instead of completion axillary dissection, the recent PMRT consensus guidelines specifically state that if completion axillary dissection is not performed after a positive sentinel node, patients should receive PMRT only if there is already sufficient information to justify its use.2 The International Breast Cancer Study Group (IBCSG) 23-01 noninferiority randomized clinically T1/2, N0 patients with micrometastases identified in SLNs to dissection versus observation.57 Twenty-two percent of patients on that study who underwent breast-conserving surgery received no adjuvant radiotherapy or received intraoperative partial breast irradiation without any axillary treatment. An additional 9% underwent mastectomy without adjuvant radiotherapy. The 5-year local regional recurrence rates in both arms of the study were less than 3%, and no notable disease-free or overall survival difference was identified. Based on this additional study taken together with the Z0011 study, it has become standard practice to avoid axillary dissection among similar patients. It is been well known for several decades that radiotherapy can be used to treat undissected breast cancer nodal disease. The EORTC 10981-22023 AMAROS (After Mapping of the Axilla, Radiotherapy or Surgery?) trial enrolled clinically node-negative patients with positive SLNs and a breast cancer less than 3 cm in diameter (approximately 17% of each group underwent mastectomy, and about 5% of cases in each arm had three or more positive SLNs).41 The study reported a 5-year axillary recurrence rate of 0.43% after dissection and 1.19% in the axillary radiotherapy group with no overall or disease-free survival difference yet significantly lower rates of clinical lymphedema in the radiotherapy arm compared with dissection group (23% compared with 11%; 100 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

p < .0001).41 Thus, axillary radiotherapy seems to control regional residual microscopic disease among patients with positive SLNs, although, taken together with the results from the Z00011 and IBCSG study, may potentially overtreat some patients.

Management of Biopsy-Proven Axillary Nodal Metastases: SLN Biopsy and Targeted Axillary Dissection

Neoadjuvant systemic therapy can eradicate documented breast cancer axillary nodal metastases in 40%–74% of patients.53,58 Therefore, it was of interest to see if the SLN procedure could accurately identify patients without residual disease to avoid axillary lymph node dissection and potential complications associated with that type of surgery. Retrospective and prospective clinical trials demonstrated that the false-negative rates of SLN biopsy alone were often higher among patients with documented nodal metastases treated with neoadjuvant systemic therapy compared with patients receiving surgery first.59-62 The false-negative rates markedly decreased after placing a marker within the lymph node with documented carcinoma, such that it could be localized and tested after neoadjuvant systemic therapy as an accurate reflection of the remaining lymph nodes.41,63 The concept intuitively makes sense, as the lymph node to test would be an actual lymph node that had carcinoma before the introduction of systemic therapy. The new technique, called targeted axillary dissection (TAD), in which a radioactive seed is placed within the previously biopsy-proven positive clipped node after preoperative chemotherapy and removed along with any other SLNs, reduced the falsenegative rates to about 2% in The University of Texas MD Anderson Cancer Center clinical trial.64 Ensuring removal of the clipped lymph node seems to improve the accuracy of sentinel lymphadenectomy after preoperative chemotherapy, because approximately 25% of the time, the clipped node with documented carcinoma is not retrieved as a standard SLN, probably because of fibrosis in the lymphatics and nodes secondary to treated carcinoma.64 Patients at MD Anderson who do not have residual nodal metastases found on TAD do not go on to full completion axillary lymph node dissection. Patients with residual nodal metastases with TAD have an approximately 50% chance of harboring additional nodal disease and undergo completion dissection or are enrolled in the Alliance for Clinical Trials in Oncology 11202 (NCT01901094), a trial with the primary aim to determine whether axillary radiation alone is not inferior to axillary lymph node dissection with radiotherapy among patients with initial documented nodal disease both before and after receipt of neoadjuvant systemic therapy.

RECONSTRUCTION IMPLICATIONS OF PMRT

General Approach to Breast Reconstruction in the Patient Needing Radiation

PMRT is used with increasing frequency in the treatment of patients with advanced breast cancer. The plastic surgeon is faced with the challenge of reconstructing a breast before

OPTIMIZING BREAST CANCER RADIOTHERAPY AND SURGICAL OUTCOMES

radiotherapy or sometimes in a previously radiated field. In these patients, the timing and sequence of mastectomy, reconstruction, chemotherapy, and radiation has not been clearly established. A reasonable algorithm for management of these complex patients is outlined in Fig. 2. Generally, if radiation is anticipated, immediate two-stage implant-based reconstruction is recommended. This allows for reconstruction of a usually acceptable breast mound and leaves autologous tissue as a potential salvage in the case of failure of the implant-based reconstruction. This section will outline why it is generally preferable not to do immediate autologous reconstruction and the outcomes and timing for immediate twostage reconstruction with implants in the setting of PMRT.

AUTOLOGOUS RECONSTRUCTION AND RADIOTHERAPY

Immediate Autologous Flap Reconstruction

Immediate breast reconstruction with a flap followed by PMRT remains controversial and is usually not recommended.65,66 This is principally the result of multiple studies that have demonstrated higher complication rates, flap fibrosis, fat necrosis, and poor aesthetic outcomes in flap reconstructions that receive radiation.65,67,68 Although free flap reconstruction can result in high flap-survival rates, these patients frequently require additional procedures and often a second flap for salvage of the reconstruction. Although there are some proponents of immediate autologous reconstruction followed by PMRT who believe that acceptable aesthetic results are feasible,69 this approach remains controversial, and most guidelines do not routinely recommend autologous reconstruction in patients who will definitely need PMRT.70

DELAYED AND DELAYED-IMMEDIATE AUTOLOGOUS FLAP RECONSTRUCTION

The most acceptable approach for the patient choosing autologous reconstruction is to perform reconstruction after the completion of mastectomy, chemotherapy, and radiation. Alternatively, a two-stage reconstruction can be performed by first placing a tissue expander underneath the mastectomy flaps and then radiating the tissue expander after early, rapid expansion. Once radiated, a flap reconstruction can be performed after a delay. This approach has been described as “delayed-immediate reconstruction” and in concept preserves more native mastectomy skin for the delayed reconstruction, potentially to the benefit of the aesthetic outcome. However, this approach does involve an additional surgical procedure. Further, it is not clear whether performing the flap reconstruction within a very short period of time after radiation therapy may actually increase the chances of perioperative complications as compared with traditional delayed autologous breast reconstruction.

ALLOPLASTIC RECONSTRUCTION AND RADIOTHERAPY

Two-stage implant-based reconstruction is the most common approach to breast reconstruction, and it is well-established that the long-term reconstructive failure rate of prosthetic reconstruction is significantly lower in nonradiated versus radiated reconstruction. The largest prospective series of immediate two-stage breast reconstruction has demonstrated that patients who undergo PMRT after implant reconstruction will lose the implant reconstruction in 9.1% of cases as compared with only 0.5% in the nonirradiated group. In addition, the high-grade capsular contracture rate is greater

FIGURE 2. Breast Reconstruction Management Algorithm When Postmastectomy Radiation Is Anticipated

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 101

KUERER, CORDEIRO, AND MUTTER

FIGURE 3. Timing and Sequence of Immediate Two-Stage Prosthetic Breast Reconstruction With Mastectomy, Chemotherapy, and Radiation

(6.9% in radiated versus 0.5% nonradiated), and the aesthetic outcomes are generally inferior in patients with radiated reconstruction.71 It is an open question whether immediate alloplastic reconstruction is still acceptable given these inferior outcomes and more importantly whether this procedure can be timed and sequenced appropriately within the oncologic treatment scheme.

TIMING AND SEQUENCE OF IMMEDIATE RECONSTRUCTION WITH MASTECTOMY, CHEMOTHERAPY, AND RADIATION

How can the reconstructive surgeon best collaborate with the oncologic surgeon, oncologist, and radiation oncologist to provide the patient with a successfully reconstructed breast mound in the face of PMRT? Figure 3 outlines the two principal approaches with regard to the sequencing of surgery and PMRT in patients who undergo two-stage prosthetic breast reconstruction: either the expander can be exchanged for permanent implant before radiation or the tissue expander is first radiated and then exchanged for the permanent implant. There have been numerous proponents for both of these approaches, although the literature primarily consists of small, retrospective series, often without control subjects.42 The protocol at Memorial Sloan Kettering Cancer Center (Fig. 2) was derived primarily for patients who underwent mastectomy, adjuvant chemotherapy, and then PMRT. These patients underwent immediate reconstruction with a tissue expander at time of mastectomy, expansion during adjuvant chemotherapy, and exchange for permanent implant 102 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

4 weeks postchemotherapy. The final implant reconstruction was then radiated 4 weeks later. This timing and sequence of oncologic treatment and reconstruction is not possible in patients undergoing neoadjuvant chemotherapy followed by mastectomy followed by radiation. In patients for whom treatment follows the neoadjuvant chemotherapy to mastectomy to PMRT sequence, exchange before radiation is not feasible because of the resulting 4- to 5-month gap between chemotherapy and PMRT. Thus, the approach to these patients involves rapid expansion within 6 weeks and radiation to the tissue expander within 8 weeks postsurgery. The exchange to permanent implant is performed 6 months after completion of PMRT. The largest prospective series that compares long-term outcomes of patients receiving prosthetic breast reconstructions with PMRT to the expander versus those receiving radiation to the permanent implant demonstrates that reconstructions undergoing radiation therapy to the tissue expander before exchange to the permanent implant have doubled the 6-year predicted failure rate (32%) of those receiving radiation after placement of the permanent implant (16%).8 Therefore, in principle, the final implant should be radiated to minimize reconstructive failure. The data on long-term aesthetic outcomes of the two approaches are not entirely clear. Nava et al72 found that subjective evaluations of shape and symmetry assessed by several surgeons and the patient’s opinion of the final reconstruction favored radiation of the final implant. However, the data from Cordeiro et al8 suggest that patients with radiation to the tissue expander might have slightly better aesthetic outcomes

OPTIMIZING BREAST CANCER RADIOTHERAPY AND SURGICAL OUTCOMES

and slightly lower capsular contracture grades than patients with radiation to the permanent implant. With either approach, one must keep in mind that it is still possible to have good-to-excellent outcomes in approximately 50% of patients. What then is the ideal timing of reconstruction and PMRT? What is the timing of radiation to the tissue expander or to the permanent implant? For patients who have already undergone neoadjuvant chemotherapy, the decision is a priori dictated by the oncologic treatment because these patients cannot delay radiotherapy and should receive radiation to the tissue expander. In the case of patients who undergo mastectomy followed by adjuvant chemotherapy, the reconstructive surgeon is faced with the dilemma of choosing to recommend radiation therapy to the tissue expander, accepting that approach’s higher rate of reconstructive failure as worthwhile given its potentially superior aesthetic result? Or should long-term viability of the reconstruction be more important than a lesser aesthetic result? The pros and cons of these options should be discussed with the patient. Perhaps the best approach is to provide the patient with the data, review her goals and expectations, and then let the patient make the final decision. One could argue that because the overall implant survival rate, aesthetic outcomes, and capsular contracture rates are significantly worse in patients who undergo immediate prosthetic breast reconstruction in the face of PMRT, these patients should not be reconstructed and should instead undergo delayed breast reconstruction. However, patients who undergo mastectomy and PMRT alone would likely

never undergo delayed two-stage prosthetic reconstruction because they cannot be expanded successfully. These patients would then be relegated to delayed reconstruction with autologous tissue. Many may not be candidates for autologous reconstruction because of age, comorbidity, or inadequate tissue at donor site; may not survive long enough to ever undergo delayed reconstruction; or may simply not be interested in further extensive, complicated operations given what they have already experienced in the course of cancer treatment. One could also argue that by providing a very simple reconstructive approach consisting of small operations and quick recoveries, most of these patients are still extremely happy to have some form of reconstructed breast and are accepting of the aesthetic tradeoff. Patient-reported outcomes data demonstrate lower satisfaction levels in those patients with radiated implant reconstructions as compared with those that are not radiated.71 However, overall satisfaction in many patients remains high enough that it is certainly worthwhile. We therefore still strongly advocate immediate two-stage implant-based breast reconstruction in any patient who might be interested, despite the potential need for PMRT.

ACKNOWLEDGMENT

This work was supported by a Cancer Center Support Grant from the National Institutes of Health (CA16672) and the P.H. and Fay Etta Robinson Distinguished Professorship in Cancer Research (H.M.K.), Mayo Clinic Breast Cancer SPORE (P50-CA116201), and the American Society for Radiation Oncology (R.W.M.).

References 1. McBride A, Allen P, Woodward W, et al. Locoregional recurrence risk for patients with T1,2 breast cancer with 1-3 positive lymph nodes treated with mastectomy and systemic treatment. Int J Radiat Oncol Biol Phys. 2014;89:392-398. 2. Recht A, Comen EA, Fine RE, et al. Postmastectomy radiotherapy: an American Society of Clinical Oncology, American Society for Radiation Oncology, and Society of Surgical Oncology focused guideline update. J Clin Oncol. 2016;34:4431-4442. 3. Sharma R, Bedrosian I, Lucci A, et al. Present-day locoregional control in patients with t1 or t2 breast cancer with 0 and 1 to 3 positive lymph nodes after mastectomy without radiotherapy. Ann Surg Oncol. 2010;17:2899-2908. 4. Whelan TJ, Olivotto IA, Parulekar WR, et al; MA.20 Study Investigators. Regional nodal irradiation in early-stage breast cancer. N Engl J Med. 2015;373:307-316. 5. Poortmans PM, Collette S, Kirkove C, et al; EORTC Radiation Oncology and Breast Cancer Groups. Internal mammary and medial supraclavicular irradiation in breast cancer. N Engl J Med. 2015;373:317-327.

misleading statistics for survival in cancer. N Engl J Med. 1985;312: 1604-1608. 8. Cordeiro PG, Albornoz CR, McCormick B, et al. What is the optimum timing of postmastectomy radiotherapy in two-stage prosthetic reconstruction: radiation to the tissue expander or permanent implant? Plast Reconstr Surg. 2015;135:1509-1517. 9. Jagsi R, Jiang J, Momoh AO, et al. Complications after mastectomy and immediate breast reconstruction for breast cancer: a claims-based analysis. Ann Surg. 2016;263:219-227. 10. Kronowitz SJ, Lam C, Terefe W, et al. A multidisciplinary protocol for planned skin-preserving delayed breast reconstruction for patients with locally advanced breast cancer requiring postmastectomy radiation therapy: 3-year follow-up. Plast Reconstr Surg. 2011;127: 2154-2166. 11. Ragaz J, Jackson SM, Le N, et al. Adjuvant radiotherapy and chemotherapy in node-positive premenopausal women with breast cancer. N Engl J Med. 1997;337:956-962.

6. Thorsen LB, Offersen BV, Danø H, et al. DBCG-IMN: a population-based cohort study on the effect of internal mammary node irradiation in early node-positive breast cancer. J Clin Oncol. 2016;34:314-320.

12. Overgaard M, Hansen PS, Overgaard J, et al. Postoperative radiotherapy in high-risk premenopausal women with breast cancer who receive adjuvant chemotherapy. Danish Breast Cancer Cooperative Group 82b Trial. N Engl J Med. 1997;337:949-955.

7. Feinstein AR, Sosin DM, Wells CK. The Will Rogers phenomenon. Stage migration and new diagnostic techniques as a source of

13. Overgaard M, Jensen MB, Overgaard J, et al. Postoperative radiotherapy in high-risk postmenopausal breast-cancer patients given adjuvant

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 103

KUERER, CORDEIRO, AND MUTTER

tamoxifen: Danish Breast Cancer Cooperative Group DBCG 82c randomised trial. Lancet. 1999;353:1641-1648. 14. Overgaard M, Nielsen HM, Overgaard J. Is the benefit of postmastectomy irradiation limited to patients with four or more positive nodes, as recommended in international consensus reports? A subgroup analysis of the DBCG 82 b&c randomized trials. Radiother Oncol. 2007;82: 247-253. 15. Recht A, Gray R, Davidson NE, et al. Locoregional failure 10 years after mastectomy and adjuvant chemotherapy with or without tamoxifen without irradiation: experience of the Eastern Cooperative Oncology Group. J Clin Oncol. 1999;17:1689-1700. 16. Taghian A, Jeong JH, Mamounas E, et al. Patterns of locoregional failure in patients with operable breast cancer treated by mastectomy and adjuvant chemotherapy with or without tamoxifen and without radiotherapy: results from five National Surgical Adjuvant Breast and Bowel Project randomized clinical trials. J Clin Oncol. 2004;22: 4247-4254. 17. EBCTCG (Early Breast Cancer Trialists’ Collaborative Group); McGale P, Taylor C, Correa C, et al. Effect of radiotherapy after mastectomy and axillary surgery on 10-year recurrence and 20-year breast cancer mortality: meta-analysis of individual patient data for 8135 women in 22 randomised trials. Lancet. 2014;383:2127-2135. 18. Recht A, Comen EA, Fine RE, et al. Postmastectomy radiotherapy: an American Society of Clinical Oncology, American Society for Radiation Oncology, and Society of Surgical Oncology focused guideline update. Pract Radiat Oncol. 2016;6:e219-e234. 19. Tvedskov TF, Jensen MB, Balslev E, et al. Stage migration after introduction of sentinel lymph node dissection in breast cancer treatment in Denmark: a nationwide study. Eur J Cancer. 2011;47:872-878. 20. van Laar C, van der Sangen MJ, Poortmans PM, et al. Local recurrence following breast-conserving treatment in women aged 40 years or younger: trends in risk and the impact on prognosis in a populationbased cohort of 1143 patients. Eur J Cancer. 2013;49:3093-3101. 21. Slamon D, Eiermann W, Robert N, et al; Breast Cancer International Research Group. Adjuvant trastuzumab in HER2-positive breast cancer. N Engl J Med. 2011;365:1273-1283. 22. Cao L, Cai G, Xu F, et al. Trastuzumab improves locoregional control in HER2-positive breast cancer patients following adjuvant radiotherapy. Medicine (Baltimore). 2016;95:e4230. 23. Darby SC, Ewertz M, McGale P, et al. Risk of ischemic heart disease in women after radiotherapy for breast cancer. N Engl J Med. 2013; 368:987-998. 24. Clarke M, Collins R, Darby S, et al; Early Breast Cancer Trialists’ Collaborative Group (EBCTCG). Effects of radiotherapy and of differences in the extent of surgery for early breast cancer on local recurrence and 15-year survival: an overview of the randomised trials. Lancet. 2005;366:2087-2106. 25. Truong PT, Olivotto IA, Kader HA, et al. Selecting breast cancer patients with T1-T2 tumors and one to three positive axillary nodes at high postmastectomy locoregional recurrence risk for adjuvant radiotherapy. Int J Radiat Oncol Biol Phys. 2005;61:1337-1347. 26. Taghian AG, Jeong JH, Mamounas EP, et al. Low locoregional recurrence rate among node-negative breast cancer patients with tumors 5 cm or larger treated by mastectomy, with or without adjuvant systemic therapy and without radiotherapy: results from five national surgical adjuvant breast and bowel project randomized clinical trials. J Clin Oncol. 2006;24:3927-3932.

104 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

27. Katz A, Strom EA, Buchholz TA, et al. Locoregional recurrence patterns after mastectomy and doxorubicin-based chemotherapy: implications for postoperative irradiation. J Clin Oncol. 2000;18:2817-2827. 28. Moo TA, McMillan R, Lee M, et al. Selection criteria for postmastectomy radiotherapy in t1-t2 tumors with 1 to 3 positive lymph nodes. Ann Surg Oncol. 2013;20:3169-3174. 29. Wallgren A, Bonetti M, Gelber RD, et al; International Breast Cancer Study Group Trials I through VII. Risk factors for locoregional recurrence among breast cancer patients: results from International Breast Cancer Study Group Trials I through VII. J Clin Oncol. 2003;21:1205-1213. 30. Cheng JC, Chen CM, Liu MC, et al. Locoregional failure of postmastectomy patients with 1-3 positive axillary lymph nodes without adjuvant radiotherapy. Int J Radiat Oncol Biol Phys. 2002;52:980-988. 31. Katz A, Strom EA, Buchholz TA, et al. The influence of pathologic tumor characteristics on locoregional recurrence rates following mastectomy. Int J Radiat Oncol Biol Phys. 2001;50:735-742. 32. Mamounas EP, Tang G, Fisher B, et al. Association between the 21gene recurrence score assay and risk of locoregional recurrence in node-negative, estrogen receptor-positive breast cancer: results from NSABP B-14 and NSABP B-20. J Clin Oncol. 2010;28:1677-1683. 33. Jegadeesh NK, Kim S, Prabhu RS, et al. The 21-gene recurrence score and locoregional recurrence in breast cancer patients. Ann Surg Oncol. 2015;22:1088-1094. 34. Cortazar P, Zhang L, Untch M, et al. Pathological complete response and long-term clinical benefit in breast cancer: the CTNeoBC pooled analysis. Lancet. 2014;384:164-172. 35. Mamounas EP, Anderson SJ, Dignam JJ, et al. Predictors of locoregional recurrence after neoadjuvant chemotherapy: results from combined analysis of National Surgical Adjuvant Breast and Bowel Project B-18 and B-27. J Clin Oncol. 2012;30:3960-3966. 36. Recht A, Somerfield MR, Edge SB. Postmastectomy radiotherapy: An American Society of Clinical Oncology, American Society for Radiation Oncology, and Society of Surgical Oncology focused guideline update summary. J Oncol Pract. 2016;12:1258-1261. 37. von Minckwitz G, Untch M, Blohmer JU, et al. Definition and impact of pathologic complete response on prognosis after neoadjuvant chemotherapy in various intrinsic breast cancer subtypes. J Clin Oncol. 2012;30:1796-1804. 38. Ellis MJ, Suman VJ, Hoog J, et al. Randomized phase II neoadjuvant comparison between letrozole, anastrozole, and exemestane for postmenopausal women with estrogen receptor-rich stage 2 to 3 breast cancer: clinical and biomarker outcomes and predictive value of the baseline PAM50-based intrinsic subtype--ACOSOG Z1031. J Clin Oncol. 2011;29:2342-2349. 39. Marks LB, Prosnitz LR. Reducing local therapy in patients responding to preoperative systemic therapy: are we outsmarting ourselves? J Clin Oncol. 2014;32:491-493. 40. White J, Mamounas E. Locoregional radiotherapy in patients with breast cancer responding to neoadjuvant chemotherapy: a paradigm for treatment individualization. J Clin Oncol. 2014;32:494-495. 41. Donker M, van Tienhoven G, Straver ME, et al. Radiotherapy or surgery of the axilla after a positive sentinel node in breast cancer (EORTC 10981-22023 AMAROS): a randomised, multicentre, openlabel, phase 3 non-inferiority trial. Lancet Oncol. 2014;15:1303-1310. 42. Momoh AO, Ahmed R, Kelley BP, et al. A systematic review of complications of implant-based breast reconstruction with prereconstruction

OPTIMIZING BREAST CANCER RADIOTHERAPY AND SURGICAL OUTCOMES

and postreconstruction radiotherapy. Ann Surg Oncol. 2014;21: 118-124. 43. Warren LE, Miller CL, Horick N, et al. The impact of radiation therapy on the risk of lymphedema after treatment for breast cancer: a prospective cohort study. Int J Radiat Oncol Biol Phys. 2014;88: 565-571. 44. Henson KE, McGale P, Taylor C, et al. Radiation-related mortality from heart disease and lung cancer more than 20 years after radiotherapy for breast cancer. Br J Cancer. 2013;108:179-182. 45. Beck RE, Kim L, Yue NJ, et al. Treatment techniques to reduce cardiac irradiation for breast cancer patients treated with breast-conserving surgery and radiation therapy: a review. Front Oncol. 2014;4:327. 46. Jimenez RB, Goma C, Nyamwanda J, et al. Intensity modulated proton therapy for postmastectomy radiation of bilateral implant reconstructed breasts: a treatment planning study. Radiother Oncol. 2013;107:213-217. 47. MacDonald SM, Jimenez R, Paetzold P, et al. Proton radiotherapy for chest wall and regional lymphatic radiation; dose comparisons and treatment delivery. Radiat Oncol. 2013;8:71. 48. MacDonald SM, Patel SA, Hickey S, et al. Proton therapy for breast cancer after mastectomy: early outcomes of a prospective clinical trial. Int J Radiat Oncol Biol Phys. 2013;86:484-490. 49. Cuaron JJ, Chon B, Tsai H, et al. Early toxicity in patients treated with postoperative proton therapy for locally advanced breast cancer. Int J Radiat Oncol Biol Phys. 2015;92:284-291. 50. Bradley JA, Dagan R, Ho MW, et al. Initial report of a prospective dosimetric and clinical feasibility trial demonstrates the potential of protons to increase the therapeutic ratio in breast cancer compared with photons. Int J Radiat Oncol Biol Phys. 2016;95:411-421. 51. Haviland JS, Owen JR, Dewar JA, et al; START Trialists’ Group. The UK Standardisation of Breast Radiotherapy (START) trials of radiotherapy hypofractionation for treatment of early breast cancer: 10-year follow-up results of two randomised controlled trials. Lancet Oncol. 2013;14:1086-1094. 52. Lyman GH, Somerfield MR, Bosserman LD, et al. Sentinel lymph node biopsy for patients with early-stage breast cancer: American Society of Clinical Oncology clinical practice guideline update. J Clin Oncol. 2016 Dec 12. 53. van la Parra RF, Kuerer HM. Selective elimination of breast cancer surgery in exceptional responders: historical perspective and current trials. Breast Cancer Res. 2016;18:28. 54. Kronowitz SJ, Hunt KK, Kuerer HM, et al. Delayed-immediate breast reconstruction. Plast Reconstr Surg. 2004;113:1617-1628. 55. Giuliano AE, Ballman K, McCall L, et al. Locoregional recurrence after sentinel lymph node dissection with or without axillary dissection in patients with sentinel lymph node metastases: long-term follow-up from the American College of Surgeons Oncology Group (Alliance) ACOSOG Z0011 randomized trial. Ann Surg. 2016;264:413-420. 56. FitzSullivan E, Bassett RL, Kuerer HM, et al. Outcomes of sentinel lymph node-positive breast cancer patients treated with mastectomy without axillary therapy. Ann Surg Oncol. 2017;24:652-659. 57. Galimberti V, Cole BF, Zurrida S, et al; International Breast Cancer Study Group Trial 23-01 investigators. Axillary dissection versus no

axillary dissection in patients with sentinel-node micrometastases (IBCSG 23-01): a phase 3 randomised controlled trial. Lancet Oncol. 2013;14:297-305. 58. Dominici LS, Negron Gonzalez VM, Buzdar AU, et al. Cytologically proven axillary lymph node metastases are eradicated in patients receiving preoperative chemotherapy with concurrent trastuzumab for HER2-positive breast cancer. Cancer. 2010;116:2884-2889. 59. Boileau JF, Poirier B, Basik M, et al. Sentinel node biopsy after neoadjuvant chemotherapy in biopsy-proven node-positive breast cancer: the SN FNAC study. J Clin Oncol. 2015;33:258-264. 60. Boughey JC, Suman VJ, Mittendorf EA, et al; Alliance for Clinical Trials in Oncology. Sentinel lymph node surgery after neoadjuvant chemotherapy in patients with node-positive breast cancer: the ACOSOG Z1071 (Alliance) clinical trial. JAMA. 2013;310:1455-1461. 61. Caudle AS, Kuerer HM. Targeting and limiting surgery for patients with node-positive breast cancer. BMC Med. 2015;13:149. 62. Kuehn T, Bauerfeind I, Fehm T, et al. Sentinel-lymph-node biopsy in patients with breast cancer before and after neoadjuvant chemotherapy (SENTINA): a prospective, multicentre cohort study. Lancet Oncol. 2013;14:609-618. 63. Caudle AS, Yang WT, Mittendorf EA, et al. Selective surgical localization of axillary lymph nodes containing metastases in patients with breast cancer: a prospective feasibility trial. JAMA Surg. 2015;150:137-143. 64. Caudle AS, Yang WT, Krishnamurthy S, et al. Improved axillary evaluation following neoadjuvant therapy for patients with node-positive breast cancer using selective evaluation of clipped nodes: implementation of targeted axillary dissection. J Clin Oncol. 2016;34:1072-1078. 65. Tran NV, Chang DW, Gupta A, et al. Comparison of immediate and delayed free TRAM flap breast reconstruction in patients receiving postmastectomy radiation therapy. Plast Reconstr Surg. 2001;108:78-82. 66. Spear SL, Ducic I, Low M, et al. The effect of radiation on pedicled TRAM flap breast reconstruction: outcomes and implications. Plast Reconstr Surg. 2005;115:84-95. 67. Rogers NE, Allen RJ. Radiation effects on breast reconstruction with the deep inferior epigastric perforator flap. Plast Reconstr Surg. 2002;109:1919-1924; discussion 1925-1926. 68. Mirzabeigi MN, Smartt JM, Nelson JA, et al. An assessment of the risks and benefits of immediate autologous breast reconstruction in patients undergoing postmastectomy radiation therapy. Ann Plast Surg. 2013;71:149-155. 69. Carlson GW, Page AL, Peters K, et al. Effects of radiation therapy on pedicled transverse rectus abdominis myocutaneous flap breast reconstruction. Ann Plast Surg. 2008;60:568-572. 70. Gradishar WJ, Anderson BO, Balassanian R, et al. Breast Cancer Version 2.2015. J Natl Compr Canc Netw. 2015;13:448-475. 71. Cordeiro PG, Albornoz CR, McCormick B, et al. The impact of postmastectomy radiotherapy on two-stage implant breast reconstruction: an analysis of long-term surgical outcomes, aesthetic results, and satisfaction over 13 years. Plast Reconstr Surg. 2014;134: 588-595. 72. Nava MB, Pennati AE, Lozza L, et al. Outcome of different timings of radiotherapy in implant-based breast reconstructions. Plast Reconstr Surg. 2011;128:353-359.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 105

HENRY, BEDARD, AND DEMICHELE

Standard and Genomic Tools for Decision Support in Breast Cancer Treatment N. Lynn Henry, MD, PhD, Philippe L. Bedard, MD, and Angela DeMichele, MD, MSCE OVERVIEW Over the past few decades, comprehensive characterization of the cancer genome has elucidated pathways that drive cancer and mechanisms of resistance to therapy and provided important insights for development of new therapies. These advances have resulted in the development of prognostic and predictive tools for use in clinical settings, which can assist clinicians and patients in making informed decisions about the benefits of established therapies. In early-stage breast cancer, multiparameter genomic assays are now available for decision making about the duration of adjuvant endocrine therapy and the use of adjuvant chemotherapy. Similarly, in metastatic disease, there are multiple commercially available next-generation sequencing options for identifying genetic alterations in tumors that may be targeted with a drug. Although these tools hold great promise for providing precision medicine, it can be difficult for the treating physician to evaluate their clinical utility and appropriately select tools for individual clinical situations. This review summarizes the currently available genomic tools in breast cancer, the data underlying their clinical validity and utility, and how they can be used in conjunction with standard clinicopathologic data for making adjuvant and metastatic treatment decisions.

he knowledge generated by The Cancer Genome Atlas on the genetic profile of breast and other cancers along with the development and widespread availability of sophisticated technologies for commercial testing of the cancer genome in clinical settings has led to a desire to use these tools to improve patient care and outcomes. This has catalyzed the development of a variety of prognostic and predictive tools designed to assist clinicians and patients in making informed decisions about the benefits of established therapies, as well as potentially identifying new treatment options. Although these tools hold promise for personalized care and rational molecularly based treatments, it can be difficult for the treating physician to evaluate their quality and clinical utility and select tools that are appropriate for an individual patient and/or clinical situation. This review summarizes the framework for evaluating new genomic tools in breast cancer and the data underlying their utility for making adjuvant and metastatic treatment decisions.

QUALITY OF BIOMARKERS

All biomarkers, including genomic and molecular tools, require rigorous development and evaluation prior to incorporation into clinical care.1 According to the Evaluation of Genomic Applications in Practice and Prevention framework, biomarkers must have analytic validity, clinical validity, and clinical utility.2 The assay for the biomarker must

be accurate and reproducible. Once an assay is analytically valid, it must be shown in multiple independent cohorts to have the ability to divide a population of interest into separate groups, and it must be demonstrated that use of the biomarker adds to current clinical care in a meaningful way without introducing substantial risk to the patient. Although it would be ideal to evaluate all biomarkers in prospective trials, similar to what is done for medications, this is not practical, so the Tumor Marker Utility Grading System was developed to assess the level of evidence supporting the clinical utility of individual biomarkers,3 which allows all potential new biomarkers to be evaluated in a standardized way.

USING MOLECULAR TOOLS TO MAKE ADJUVANT ENDOCRINE THERAPY DECISIONS

Prognostication Without Treatment

Adjuvant endocrine therapy reduces the risk of breast cancer recurrence and improves survival4 for hormone receptor–positive (HR+) breast cancer irrespective of age, menopausal status, involvement of axillary lymph nodes, or tumor size.5 A variety of molecular tools, including the 21gene recurrence score assay (Oncotype DX),6-8 the 70-gene signature (MammaPrint),9-11 the PAM50 risk-of-recurrence assay (Prosigna),12 the Breast Cancer Index,13,14 and EndoPredict,15,16 can identify subsets of patients who are HR+/

From the University of Utah, Salt Lake City, UT; Department of Medicine, Division of Medical Oncology & Hematology, Princess Margaret Cancer Centre, University Health Network, University of Toronto, Toronto, Ontario, Canada; Abramson Cancer Center, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Angela DeMichele, MD, MSCE, University of Pennsylvania School of Medicine, Perelman Center for Advanced Medicine, 10 South, 3400 Civic Center Blvd., Philadelphia, PA 19104; email: [emailprotected]. © 2017 American Society of Clinical Oncology

106 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

GENOMIC TOOLS FOR BREAST CANCER DECISION SUPPORT

TABLE 1. Summary of Available Multiparameter Genomic Assays for Decision Making in Early Breast Cancer Name

Description

Results

References

Breast Cancer Index

HoxB13/IL17BR plus molecular-grade index

Low vs. high risk for both prognosis and prediction

13,14,17

EndoPredict

11-gene signature

Low vs. high risk

15,18,19

uPA/PAI-1 (Femtelle)

Urokinase plasminogen activator plus plasminogen activator inhibitor type 1

Low vs. high risk

70-gene breast cancer recurrence assay (MammaPrint)

70-gene signature

Low vs. high risk

Mammostrat

5-gene signature

Low vs. moderate vs. high risk

21-gene recurrence score assay (Oncotype DX)

21-gene signature

Low vs. intermediate vs. high risk

6,7,21

PAM50 risk of recurrence score (Prosigna)

46-gene signature plus 18-gene proliferation score plus tumor size

Low vs. intermediate vs. high risk

12,22,23

HER2− and have a low risk of distant recurrence (< 5% at 5 years and/or < 10% at 10 years; Table 1). However, the vast majority of patients included in these series were treated with adjuvant endocrine therapy (and some also received chemotherapy). There are less data regarding prognostic performance for patients who were untreated. MammaPrint (70-gene signature) was initially developed from a cohort of 78 patients with lymph node–negative breast cancer who were younger than age 55 by using distant metastasis-free survival at 5 years from diagnosis to derive the “good prognosis” from “poor prognosis” 70-gene signature.9 All patients who were distant metastasis-free at 5 years, as well as 29 of 34 patients who developed metastases, had not received adjuvant endocrine therapy or chemotherapy. A subsequent independent validation by

KEY POINTS • Numerous multiparameter genomic assays have been developed that provide prognostic information for HR+, HER2−, node-negative breast cancer, and a subset are also predictive of the benefit from adjuvant chemotherapy and endocrine therapy. • Studies are underway examining the clinical utility of multiparameter genomic assays for determining benefit from extended adjuvant endocrine therapy. • Comprehensive next-generation sequencing approaches are being developed to identify targetable lesions, and clinical tests focused on actionable mutations are currently available. • Liquid biopsies, which can be used to identify circulating tumor cells and circulating tumor DNA, are less invasive, may better reflect tumor heterogeneity, and can be used to identify tumor mutations that are complementary to those found in tumor biopsies. • Prospective trials using genomic tumor testing for treatment selection in the metastatic setting have not demonstrated clinical utility in improving patient outcomes. Thus, this testing is not recommended by ASCO for this purpose, although it may be useful for eligibility in investigational trials.

the TRANSBIG consortium included 302 patients from six European institutions with node-negative breast cancer diagnosed between 1980 and 1998 who had not received adjuvant endocrine therapy or chemotherapy.24 The 10-year distant metastasis-free survival in the “low-risk” 70-gene profile group approached 90%. Similarly, a subset of 198 patients with available RNA from this cohort were analyzed for a 76-gene prognostic profile,25 independently developed by investigators at the Erasmus Cancer Center in Rotterdam, the Netherlands, with minimal gene overlap with the 70gene profile.26 The 10-year distant metastasis-free survival in the “good risk” 76-gene profile group was 94% in the absence of systemic therapy.25 The Oncotype DX assay, a 21-gene signature, was initially validated in a cohort of 668 patients with HR+, node-negative breast cancer treated with adjuvant tamoxifen and no chemotherapy in NSABP B-14.6 This trial randomly selected patients to receive 5 years of tamoxifen versus placebo. For patients in the placebo group, a low-risk recurrence score lower than 18 was associated with a 10-year distant disease-free survival of 85.9% compared with 93.1% in patients who received tamoxifen.27 Recent data from prospective clinical trials demonstrate excellent outcomes for patients with HR+ breast cancer with “low-risk” gene-expression profiles treated with endocrine therapy alone. In the TAILORx trial, patients with HR+, HER2−, node-negative breast cancer with an Oncotype DX recurrence score of 10 or lower were treated with tamoxifen or an aromatase inhibitor, and their rate of 5-year freedom-from-distant recurrence was 99.3%.7 Similarly, in the West German Study Group Plan B, trial patients with HR+, HER2−, breast cancer with recurrence scores of 11 or lower (including patients with 1–3 involved lymph nodes) were treated with endocrine therapy alone without chemotherapy, and their 3-year disease-free survival was 98%.28 In the MINDACT trial, patients with HR+, HER2− breast cancer with up to three positive lymph nodes who were low risk based on clinical (based upon Adjuvant! Online) and genomic (based upon MammaPrint) criteria had a 5-year distant disease-free survival of 97%.11 asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 107

HENRY, BEDARD, AND DEMICHELE

These excellent outcomes raise the question of whether these patients might also fare well without endocrine treatment. Outside of well-controlled clinical trials, up to 40% of patients with early-stage HR+ breast cancer are nonadherent with tamoxifen or aromatase inhibitor therapy.29,30 Persistence with endocrine treatment decreases over time, due in part to side effects, including vasomotor and mood symptoms, arthralgias, weight gain, and sexual dysfunction. Intriguingly, a retrospective analysis of the randomized Stockholm Tamoxifen trial (STO-3) reported a 20-year disease-specific survival rate of 94% in node-negative women who received no adjuvant systemic therapy and who had a predefined “ultra-low-risk” MammaPrint score; in the tamoxifen-treated group, the 20-year disease-specific survival was 97%.31 Although only 15% of patients in this analysis had an ultra-low-risk MammaPrint score, molecular tools may identify a higher proportion of patients with indolent biology in the modern era of mammographic breast cancer screening.32 However, considering the long natural history of recurrence risk for HR+ breast cancer, the generally favorable safety profile of tamoxifen and aromatase inhibitors, and the continued reduction in recurrence risk beyond the duration of endocrine treatment,4 the current evidence is insufficient to withhold endocrine therapy based upon the results of molecular testing.

Prediction of Benefit From Endocrine Treatment

Data for prediction of endocrine treatment benefit with molecular tools are even more limited. In the NSABP B-14 trial, the greatest benefit with adjuvant tamoxifen versus placebo was observed in patients with low (< 18) and intermediate (18–30) Oncotype Dx scores compared with high (≥ 31) scores (for low scores: 10-year distant relapse-free survival [DRFS], 93.1% vs. 85.9%; p = .039; for intermediate: 10-year DRFS, 79.5% vs. 62.2%; p = .02; for high: 10-year DRFS, 70.3% vs. 68.7%; p = .82).27 The National Cancer Institute of Canada Clinical Trials Group MA.12 trial randomly selected premenopausal women with stage I–III breast cancer of any hormonal status to receive tamoxifen versus placebo following adjuvant chemotherapy.33 A retrospective analysis using the PAM50 assay showed a benefit in disease-free survival for tamoxifen (hazard ratio 0.52; 95% CI, 0.32–0.86) in the luminal (A + B) subtypes compared with nonluminal subtypes (hazard ratio 0.80; 95% CI, 0.50–1.29), although the interaction test was not statistically significant (p = .24).34 There are no randomized data available for prediction of endocrine treatment benefit with other molecular tools. These data are insufficient to inform endocrine treatment decisions for patients with HR+ breast cancer identified using standardized immunohistochemistry testing methods.35 Data to evaluate the differential predictive benefit of one endocrine treatment approach versus another are also limited. In the TransATAC study, the prognostic value of recurrence score was compared for postmenopausal patients who were HR+ and randomly selected to receive 5 years of anastrozole versus tamoxifen using a multivariate model adjusted for tumor size, tumor grade, nodal status, and age.8 108 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

The hazard ratios for distant recurrence per 50-point change in recurrece scores were similar in each treatment group, and tests for recurrence score × treatment interactions were not noteworthy. The TEAM trial compared 5 years of upfront exemestane versus a switch regimen of tamoxifen (2 years to 3 years) followed by exemestane (3 years to 2 years). No differential effect in DRFS prognostication was observed in patients treated with 5-year exemestane or the switch regimen according to Mammostrat score, which stratifies patients based on five immunohistochemical markers.36

Prognostication of Late Relapse

HR+ breast cancer has a unique natural history, with a relatively constant ongoing annual hazard of distant relapse following an initial peak within the first 5 years of diagnosis.37,38 Recent studies demonstrate a reduction in recurrence risk with extended adjuvant endocrine therapy with tamoxifen39,40 or an aromatase inhibitor41-43 following 5 years of upfront tamoxifen. Data regarding the benefit of extended adjuvant endocrine therapy after 5 years of upfront aromatase inhibitor or a switch regimen of 2 year to 3 years of tamoxifen followed by 2 year to 3 years of aromatase inhibitor are conflicting.44-47 Extended adjuvant endocrine treatment is associated with bothersome side effects and potentially serious risks, such as endometrial cancer and venous thromboembolism with tamoxifen and bone fracture and cardiovascular events with aromatase inhibitors. A variety of molecular tools may identify patients at very low risk of late relapse more than 5 years from diagnosis who might be spared extended adjuvant endocrine treatment. In NSABP B-14, node-negative patients treated with 5 years of tamoxifen who had high ESR1 mRNA expression and recurrence scores lower than 18 had a low risk of distant recurrence in years 5–15 (6.8%) compared with scores of 18–30 (11.2%) and scores 31 or higher (16.4%; p = .01).48 The risk-of-recurrence score model integrates the expression profile of a subset of 46 genes from the PAM50 intrinsic subtype classifier with an 18-gene proliferation score and tumor size. In TransATAC, recurrence score and risk of recurrence were associated with relapse risk during years 5–10 from diagnosis in multivariate analysis,49 although risk-of-recurrence scoring was a stronger prognosticator of late recurrence than recurrence scoring. A combined analysis of TransATAC and the ABCSG-8 trial found that the risks of distant relapse in years 5–10 for women with low, intermediate, and high risk-of-recurrence scores were 2.4%, 8.3%, and 16.8%, respectively.50 For women with node-positive disease, the 5–10-year distant recurrence risk with a low risk-of-recurrence score was only 3.3%. Likewise, a combined analysis of the ABCSG-6 and ABCSG-8 trials found that EPclin, which combines the EndoPredict score with tumor size and nodal status, was prognostic for late distant recurrence (> 5 years) in patients with HR+, HER2− breast cancer.51 At 10 years of follow-up, the rate of distant metastasis was 1.8% for EPclin-low patients compared with 17.1% for EPclin-high patients. BCI combines the two-gene HOXB13/IL17BR ratio with the five proliferation

GENOMIC TOOLS FOR BREAST CANCER DECISION SUPPORT

gene Molecular Grade Index. A cubic risk model was developed using the tamoxifen arm of the STO-3 and was found to be prognostic of late recurrence in an independent validation cohort of HR+, node-negative patients.17 Risks of distant recurrence in years 5–10 were 2.5%, 16.9%, and 15.0% for the low-, intermediate-, and high-risk cohorts, respectively. A linear risk model was also shown to prognosticate for late recurrence in TransATAC13 and restratify low and intermediate reucrrence score groups into distant recurrence risk subsets.52 A recently published ASCO guideline using data available through August 2014 did not recommend any molecular tools to inform extended adjuvant endocrine therapy.53 The review panel concluded that clinical validity for late recurrence was not shown for any individual assay in more than one study, and clinical utility had not been demonstrated. Although there are limitations of the available evidence, the consistency of highly favorable outcomes for late recurrence risk reported for “low-risk” patients across studies suggests that these molecular tools that are driven by quantification of proliferation may play an increasingly prevalent role in extended adjuvant treatment decisions.

Tools for Chemotherapy Decision Making

Clinicopathologic tools. Breast cancer is heterogeneous. At diagnosis, patients can present with breast tumors that differ by stage, histology, and pathologic characteristics. When deciding whom to treat with cytotoxic chemotherapy, providers synthesize the data for individual patients to develop a personalized treatment recommendation. In particular, treatment decisions are driven by clinical factors such as patient age and comorbidities, as well as pathologic factors such as disease stage, tumor grade, and receptor status, including estrogen receptor and progesterone receptor overexpression as well as overexpression or amplification of HER2. A number of organizations have developed guidelines for chemotherapy use in patients with stage I–III breast cancer based on these standard clinicopathologic characteristics, including the National Comprehensive Cancer Network, ASCO, and the St. Gallen International Expert Consensus Panel.54-57 Quantitative decision aids have been developed for use by providers making treatment recommendations. One of the first to be widely used was Adjuvant! Online (www.adjuvantonline.com), in which providers input details including age, tumor size, nodal involvement, grade, and estrogen and progesterone receptor status and receive estimates of likelihood of recurrence or mortality within 10 years based on different treatment options. The data used to develop the tool were derived from Surveillance, Epidemiology, and End Results data for estimates of prognosis as well as data from the Early Breast Cancer Trialists’ Collaborative Group meta-analysis for estimates of response to endocrine and cytotoxic chemotherapy.58 Although useful, the tool has limitations, including lack of incorporation of HER2 status into the estimates. In addition, the site is currently offline for updating.

A second decision aide has been developed that addresses some of the limitations of Adjuvant! Online. This model, called PREDICT (www.predict.nhs.uk), is derived from a database of patients treated in the United Kingdom between 1999 and 2003 and provides survival estimates for patients based on clinicopathologic factors, including HER2, Ki67, and mode of detection of breast cancer.59 It has been validated using a number of independent data sets.60,61 In addition to providing 5- and 10-year survival data, the site also provides information about likelihood of benefit from secondand third-generation chemotherapy regimens as well as endocrine therapy. Although these decision aides can be used for treatment decision making for patients with estrogen receptor–negative, progesterone receptor–negative, and HER2− breast cancer (termed "triple-negative" breast cancer), decisions regarding use of chemotherapy are primarily driven by tumor size and lymph node involvement.54,55,57,62 Similarly for patients with HER2+ disease, use of chemotherapy plus anti-HER2 therapy is recommended for most patients with tumor size greater than 5 mm and/or lymph node involvement because of the reduction in risk of recurrence with chemotherapy and the additional 40% improvement in disease-free survival from the addition of trastuzumab.54,55 Therefore, although these decision aids can estimate prognosis for an individual patient, they are often less useful for making decisions about treatment. In contrast, for patients with HR+ disease, the Adjuvant! Online and PREDICT decision aids may be more useful for informing decision making regarding treatment with chemotherapy in addition to endocrine therapy. Multiparameter genomic assays. More recently, multiparameter genomic assays have been developed that complement standard clinicopathologic data for treatment selection (Table 1). A subset of these has been recommended for use in the most recent ASCO biomarker guidelines53 because of demonstrated clinical utility, including Oncotype DX,6,7,63 EndoPredict,15,18,21 PAM50/Prosigna,12,19,22 Breast Cancer Index,13,14,17 and the combination of urokinase plasminogen activator and plasminogen activator inhibitor type 1.23 Importantly, current ASCO guidelines recommend these assays for use in patients with HR+, HER2−, node-negative breast cancer.53 The guidelines do not recommend use of these assays for guiding adjuvant systemic therapy decisions in patients with HR−, HER2+, or node-positive disease.53 Selected assays will be discussed in more detail below; details about the other assays can be found in the provided references. The first assay to be incorporated into routine clinical use in the United States was Oncotype DX.6 The assay was originally developed using samples from patients with HR+, node-negative breast cancer treated with tamoxifen on a randomized clinical trial. The assay analyzes the expression of 16 tumor-related genes and five housekeeping genes from formalin-fixed, paraffin-embedded tumor specimens to generate a recurrence score that corresponds with the 10-year risk of distant disease recurrence assuming 5 years of treatment with tamoxifen. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 109

HENRY, BEDARD, AND DEMICHELE

Subsequent studies using samples derived from a separate trial demonstrated that the recurrence score is also predictive of response to chemotherapy. Patients with low scores (0–17) were shown to have no benefit from adjuvant chemotherapy in addition to endocrine therapy, whereas those with high scores (31–100) were shown to benefit from chemotherapy followed by endocrine therapy.63 For those with intermediate scores (18–30), the optimal treatment approach remains uncertain and is currently being investigated in the TailoRX clinical trial (NCT00310180). Observational studies have demonstrated that incorporation of the assay into routine clinical care has resulted in decreased use of chemotherapy for women with node-negative breast cancer.20 In addition, data from prospective-retrospective trials have demonstrated there is similar benefit from use of Oncotype DX in patients with node-positive disease8,64; the prospective SWOG 1007 RxPONDER clinical trial (NCT01272037) is ongoing to provide more definitive results. The use of a different multiparameter genomic assay, MammaPrint, for chemotherapy decision making has also been studied in a large prospective randomized trial, MINDACT.11 MammaPrint separates patients into two categories, either good or poor prognosis. In MINDACT, the investigators evaluated patient prognosis based on both standard clinicopathologic factors using Adjuvant! Online and genomic factors using MammaPrint. Those patients who were discordant, with high clinical risk but low genomic risk, were randomly selected to receive chemotherapy or not, in additional to endocrine therapy. The 5-year rate of survival without distant metastases in those patients who did not receive chemotherapy was 94.7% (CI, 92.5% to 96.2%), which met the criteria for success in this trial. They concluded that those patients at low genomic risk of recurrence might not require chemotherapy, despite being at high risk based on standard clinicopathologic factors. Of note, the results of this trial were published after the development of the most recent ASCO biomarker guidelines.53,65 In summary, multiparameter gene expression assays complement standard pathologic factors and provide additional information to support treatment decision making. The primary use of multiparameter assays, therefore, is to inform the decision about adjuvant systemic therapy in situations in which chemotherapy in being considered. For patients who definitely will or will not be receiving chemotherapy based on standard clinicopathologic factors, including comorbidities and patient preferences, testing is unnecessary because it will not alter the planned treatment. Additional high levels of evidence data are forthcoming regarding use of these assays in other patient populations, including those with node-positive disease.

THE ROLE OF GENOMIC TESTING IN THE METASTATIC SETTING

Breast cancer is the second leading cause of cancer death in women.66 Approximately 30% of women diagnosed with the disease will ultimately recur, and over 40,000 women die 110 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

annually of metastatic breast cancer.66 Standard treatment is guided by expression of HR or HER2, with sequential endocrine therapies initially in most HR+ disease, chronic antiHER2 therapy (with or without chemotherapy) in HER2+ disease, and sequential chemotherapy in triple-negative and endocrine-resistant disease.67 Development of resistance is universal, and patients are in a continual state of alternating disease control and progression. Receptor expression and genetic changes can differ between a breast primary and metastases,68 as tumors continue to evolve both stochastically and in response to treatment. Several meta-analyses69 have documented that pooled estimates for the absolute frequency in changes from positive to negative ranged from 5.7% to 9.5% for estrogen receptor status and 17% to 24% for progesterone receptor status, whereas ranges for changes from negative to positive ranged from 3% to 8.8% and 6.9% to 7.3% for estrogen receptor and progesterone receptor, respectively. The overall rate of absolute change in HER2 status (in either direction) was approximately 6%, and some studies have demonstrated that discordance is associated with shorter survival.70-72 Next-generation sequencing (NGS) technologies have led to the development of numerous commercial assays that can detect genomic variability both within tumors and in the circulation through the identification of intact circulating tumor cells (CTCs) and shed tumor DNA (patient tumor DNA [ptDNA]). A number of studies have reported on the spectrum of mutations identified by massively parallel sequencing in primary and metastatic breast cancer,73-76 and similar recurrent genomic alterations have been identified both in tumor and blood.77-81 Although these data have the potential to improve prognostication, expand therapeutic targets,82-85 or enable tracking of therapeutic response,86 evidence supporting the clinical utility of either tumor- or blood-based genomic assays for these purposes is scant.

Available Tools for Genomic Testing of Metastatic Disease

Approaches to evaluating the spectrum of genetic mutations or other alterations, such as copy number changes, generally use NGS approaches to enable simultaneous evaluation of many genes, with commercial and proprietary panels in widespread use. This requires tumor biopsy, which can be difficult or risky depending on the location of disease, limits the amount of tissue that can be obtained, and may not be representative of the entire tumor because of tumor heterogeneity. Technologies underlying these panels use whole-exome sequencing on paraffin-embedded tumor specimens and bioinformatics approaches to identify driver mutations, mutational hot spots, and those that are actionable—that is, for which there are potentially effective targeted therapies. Those approved for use in making treatment decisions must be done in a Clinical Laboratory Improvement Amendments–certified laboratory, meeting requirements for clinical and analytic validity. Platforms can vary substantially in sensitivity, specificity, and the spectrum of mutations that can be detected, and there are biologic

GENOMIC TOOLS FOR BREAST CANCER DECISION SUPPORT

challenges to identifying true drivers in this context of tremendous biologic heterogeneity.87 In addition to identifying mutations, amplifications in many genes are also biologically important, and thus, copy-number changes are typically included in these tests. Concurrently, technologies have developed to measure circulating markers of tumor burden, including CTCs and tumor-derived DNA. These technologies are appealing as a way to provide a liquid biopsy from the blood, obviating the difficulties in procuring surgical tumor specimens. However, these technologies also differ in sensitivity and specificity, require specific blood processing and storage protocols, and, in the case of CTCs, require immediate processing. CTCs can measure tumor burden and be profiles for surface receptors, including estrogen receptor, progesterone receptor, and HER2. ptDNA can detect specific mutations but with the same limitations as tumor NGS. Prior studies of concordance between tumor and ptDNA in breast cancer have been mainly restricted to analysis of pathogenic mutations, demonstrating concordance rates above 70%88-90 when the same platform is used. In addition, as shown in a prospective study of 32 patients comparing NGS data sets from three distinct patient-matched samples types (formalin-fixed, paraffin-embedded and CTC-DNA ptDNA) using common amplicon-based resequencing panels, CTC-DNA and ptDNA evaluation yielded complementary molecular information from the same blood sample.85,91 However, in clinical practice, the need for rapid turnaround times for clinical decision making and differences in commercial platforms can lead to lack of concordance between tumor and blood due to purely technical reasons, such as differences in test coverage of genes, laboratory variant reporting practices, variant classification, and allele frequency thresholds for detection based on total sequencing depth. Consideration of these limitations is are extremely important for practicing oncologists when ordering and interpreting the results of such tests to avoid erroneous conclusions about potential therapeutic targets or the gain or loss of specific mutations or overall changes in mutational burden under the pressure of therapy.

Current Knowledge of the Unique Biology of Metastatic Disease

The development of massively parallel sequencing technologies such as NGS has led to a proliferation of studies that have characterized primary tumors and enable comparison of metastatic breast tumors to matched primaries. The Cancer Genome Atlas characterized the genomic landscape of early breast cancer, demonstrating that approximately onethird of tumors have TP53 or PIK3CA mutations, and up to 20% have amplifications in ERBB2, FGFR1, and CCND1.73 Studies taking a broad approach with NGS have demonstrated that whole-exome or whole-genome sequencing uncovers discordant, novel mutations in both primaries and matched metastases.79,92-94 Studies using comparative genomic hybridization to detect copy number changes between matched primary and distant metastases have been

contradictory, with some but not all studies finding increased copy number changes in metastases compared with primaries.95-97 The largest metastatic profiling study to date, examining 216 metastatic breast tumors/blood pairs compared with 712 TCGA primary tumors as reference, found that 12 genes (TP53, PIK3CA, GATA3, ESR1, MAP3K1, CDH1, AKT1, MAP2K4, RB1, PTEN, CBFB, and CDKN2A) were identified as significantly mutated in metastatic breast cancer (false discovery rate < 0.1). Eight genes (ESR1, FSIP2, FRAS1, OSBPL3, EDC4, PALB2, IGFN1, and AGRN) were more frequently mutated in metastatic breast cancer as compared with early-stage breast cancer (false discovery rate < 0.01). ESR1 was identified both as a driver and as a metastatic gene (n = 22; odds ratio 29; 95% CI, 9–155; p = 1.2e-12) and also presented with focal amplification (n = 9) for a total of 31 metastatic breast cancers with either ESR1 mutation or amplification, including 27 HR+ and HER2− metastatic breast cancers (19%). HR+/HER2− metastatic breast cancers presented a high prevalence of mutations in genes located in the mTOR pathway (TSC1 and TSC2) as compared with HR+/ HER2− early-stage breast cancer (6% and 0.7%, respectively; p = .0004). Other actionable genes were more frequently mutated in HR+ metastatic breast cancer, including ERBB4 (n = 8), NOTCH3 (n = 7), and ALK (n = 7). Analysis of mutational signatures revealed a notable increase in APOBEC-mediated mutagenesis in HR+/HER2− metastatic tumors as compared with primary TCGA samples (p < 2e-16). These data and others that are emerging paint a picture of the enormous genetic heterogeneity of metastatic breast cancer.

Clinical Utility of Genomic Testing for Patient and Treatment Selection in Metastatic Trials and Practice

Despite advances in technology and our understanding of metastatic tumor biology as well as the proliferation of clinical and commercial tools to perform genomic assessment of tumor genomics and ptDNA, the clinical utility of these tests has not yet been established in metastatic breast cancer. Clinical utility of a genomic test in this context is defined as the identification of targetable alterations (those causing perturbations in proteins, pathways, or both that can be specifically intercepted pharmacologically98) with the use of such agents leading to a favorable outcome over standard of care. As stated earlier, high-level evidence supporting the use of such tools in clinical practice relies on prospective studies designed and powered for clinically meaningful outcomes. Only one trial to date, the SAFIR trial,77 has prospectively profiled metastatic breast tumors and assessed treatment responses based upon genomically guided decision making. This trial, conducted in 18 centers in France, enrolled 423 patients with biopsy-accessible tumors. Biopsy samples from 407 patients underwent comparative genomic hybridization, genome-wide single nucleotide polymorphism array, and Sanger sequencing of PIK3CA (exons 10 and 21) and AKT1 (exon 4). The primary outcome was the proportion of patients for whom genomic analysis identified a targeted therapy, with a goal of achieving a 30% or higher success asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 111

HENRY, BEDARD, AND DEMICHELE

rate. Overall, 46% of patients were found to have actionable mutations, and therapy could be personalized in 13%. Of the 43 assessable patients, 9% had partial response to therapy, and 21% had stable disease for at least 16 weeks. Although these results are promising and demonstrate feasibility, they do not provide evidence that using this approach is superior to use of estrogen receptor, progesterone receptor, or HER2 overexpression for clinical decision making. Similar multidisease trials are either completed or underway. To date, results are mixed, and it is too early to define a role for testing specifically in breast cancer. However, there are hints that such an approach could have traction. In the MOSCATO trial,99 patients were used as their own control subjects. The progression-free survival (PFS) from the most recent therapy on which the patient had just experienced progression before enrollment in MOSCATO was compared with the PFS observed under the targeted therapy selected within the MOSCATO trial based upon molecular genomic testing of the patient’s tumor, which was selected from over 60 phase I trials at the study center. A total of 33% of patients treated within the MOSCATO trial had an improved outcome, defined as at least a 30% increase in their PFS with the targeted therapy as compared with their baseline reference PFS. Moreover, 62% of the patients had an objective response or stable disease. The SHIVA trial100 took a slightly different approach. This was an open-label, randomized, controlled phase II trial that included adult patients with any metastatic solid tumor refractory to standard of care, provided they had good performance status, disease that was accessible for a biopsy or resection of a metastatic site, and at least one measurable lesion. The molecular profile of each patient’s tumor was established with large-scale genomic testing, and the trial enrolled only patients for whom a molecular alteration was identified within one of three molecular pathways (HR, phosphoinositide 3-kinase/ AKT/mTOR, and RAF/mitogen-activated protein kinase kinase), which could be matched to 1 of 10 regimens, including 11 available molecularly targeted agents. Patients were randomly assigned (1:1) to receive a matched molecularly targeted agent (experimental group) or treatment of the physician’s choice (control group). This trial was not limited

to breast cancer, although patients with breast cancer constituted 20% of the study population, and patients with tumor alterations that matched a standard-of-care therapy (such as tamoxifen for HR+ disease) were not included in the analytical group. With 11.3 months of follow-up, there was no considerable difference in PFS between those who received a matched therapy and those who received physician’s choice (hazard ratio 0.88; 95% CI, 0.65–1.19; p = .41) nor were the findings noteworthy within each of the specific types of alterations. The small size of the subset of patients with breast cancer precluded analysis for that group specifically. Similar trials, including MATCH,101 SAFIR02, and TAPUR, are ongoing and enrolling patients with breast cancer. Given the current state of evidence on the biology of breast cancer metastases and the lack of definitive utility of genomic testing tools for treatment selection, ASCO guidelines102 currently recommend using ER, PR, and HER2 status of the metastatic tumor for treatment selection and support biopsy of metastatic sites for this purpose. The panel considered any use of NGS testing to be investigational and does not recommend the use of this testing to initiate systemic therapy or direct selection of new therapy outside a research setting.

FUTURE DIRECTIONS

In the past decade, there have been considerable advances in the development of genomic assays and incorporation of these tools into routine clinical care, primarily for chemotherapy decision-making. Increasingly, there are also data evaluating use of these same assays for making decisions about extended adjuvant endocrine therapy, although use of these tools has not yet been incorporated into treatment guidelines. In the arena of metastatic breast cancer, comprehensive genomic analysis of metastatic lesions is being intensively studied to determine if the identified changes have sufficient clinical utility to guide treatment. Continued technological advances will lead to more comprehensive findings from both tumors and liquid biopsies, at lower cost. Results of studies examining the impact of this knowledge on disease outcomes, and the clinical utility of these results for guiding patient care, are eagerly awaited.

References 1. Henry NL, Hayes DF. Cancer biomarkers. Mol Oncol. 2012;6:140-146. 2. Teutsch SM, Bradley LA, Palomaki GE, et al; EGAPP Working Group. The Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Initiative: methods of the EGAPP Working Group. Genet Med. 2009;11:3-14. 3. Hayes DF, Bast RC, Desch CE, et al. Tumor marker utility grading system: a framework to evaluate clinical utility of tumor markers. J Natl Cancer Inst. 1996;88:1456-1466. 4. Early Breast Cancer Trialists’ Collaborative Group (EBCTCG); Davies C, Godwin J, Gray R, et al. Relevance of breast cancer hormone receptors and other factors to the efficacy of adjuvant tamoxifen: patient-level meta-analysis of randomised trials. Lancet. 2011;378:771-784.

112 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

5. Eifel P, Axelson JA, Costa J, et al. National Institutes of Health Consensus Development Conference statement: adjuvant therapy for breast cancer, November 1-3, 2000. J Natl Cancer Inst. 2001;93:979989. 6. Paik S, Shak S, Tang G, et al. A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. N Engl J Med. 2004;351:2817-2826. 7. Sparano JA, Gray RJ, Makower DF, et al. Prospective validation of a 21gene expression assay in breast cancer. N Engl J Med. 2015;373:20052014. 8. Dowsett M, Cuzick J, Wale C, et al. Prediction of risk of distant recurrence using the 21-gene recurrence score in node-negative and

GENOMIC TOOLS FOR BREAST CANCER DECISION SUPPORT

node-positive postmenopausal patients with breast cancer treated with anastrozole or tamoxifen: a TransATAC study. J Clin Oncol. 2010;28:1829-1834. 9. van de Vijver MJ, He YD, van’t Veer LJ, et al. A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med. 2002;347:1999-2009. 10. Mook S, Schmidt MK, Viale G, et al; TRANSBIG Consortium. The 70gene prognosis-signature predicts disease outcome in breast cancer patients with 1-3 positive lymph nodes in an independent validation study. Breast Cancer Res Treat. 2009;116:295-302. 11. Cardoso F, van’t Veer LJ, Bogaerts J, et al; MINDACT Investigators. 70gene signature as an aid to treatment decisions in early-stage breast cancer. N Engl J Med. 2016;375:717-729. 12. Dowsett M, Sestak I, Lopez-Knowles E, et al. Comparison of PAM50 risk of recurrence score with oncotype DX and IHC4 for predicting risk of distant recurrence after endocrine therapy. J Clin Oncol. 2013;31:2783-2790. 13. Sgroi DC, Sestak I, Cuzick J, et al. Prediction of late distant recurrence in patients with oestrogen-receptor-positive breast cancer: a prospective comparison of the breast-cancer index (BCI) assay, 21gene recurrence score, and IHC4 in the TransATAC study population. Lancet Oncol. 2013;14:1067-1076.

the ABCSG-8 trial treated with adjuvant endocrine therapy alone. Ann Oncol. 2014;25:339-345. 23. Harbeck N, Schmitt M, Meisner C, et al; Chemo-N 0 Study Group. Ten-year analysis of the prospective multicentre Chemo-N0 trial validates American Society of Clinical Oncology (ASCO)-recommended biomarkers uPA and PAI-1 for therapy decision making in nodenegative breast cancer patients. Eur J Cancer. 2013;49:1825-1835. 24. Buyse M, Loi S, van’t Veer L, et al; TRANSBIG Consortium. Validation and clinical utility of a 70-gene prognostic signature for women with node-negative breast cancer. J Natl Cancer Inst. 2006;98:1183-1192. 25. Desmedt C, Piette F, Loi S, et al; TRANSBIG Consortium. Strong time dependence of the 76-gene prognostic signature for node-negative breast cancer patients in the TRANSBIG multicenter independent validation series. Clin Cancer Res. 2007;13:3207-3214. 26. Wang Y, Klijn JG, Zhang Y, et al. Gene-expression profiles to predict distant metastasis of lymph-node-negative primary breast cancer. Lancet. 2005;365:671-679. 27. Paik S, Shak S, Tang G, et al. Expression of the 21 genes in the Recurrence Score assay and tamoxifen clinical benefit in the NSABP study B-14 of node negative, estrogen receptor positive breast cancer. J Clin Oncol. 2005;23:510.

14. Jerevall PL, Ma XJ, Li H, et al. Prognostic utility of HOXB13:IL17BR and molecular grade index in early-stage breast cancer patients from the Stockholm trial. Br J Cancer. 2011;104:1762-1769.

28. Gluz O, Nitz UA, Christgen M, et al. West German Study Group Phase III PlanB Trial: First prospective outcome data for the 21-gene recurrence score assay and concordance of prognostic markers by central and local pathology assessment. J Clin Oncol. 2016;34:2341-2349.

15. Filipits M, Rudas M, Jakesz R, et al; EP Investigators. A new molecular predictor of distant recurrence in ER-positive, HER2-negative breast cancer adds independent information to conventional clinical risk factors. Clin Cancer Res. 2011;17:6012-6020.

29. Partridge AH, LaFountain A, Mayer E, et al. Adherence to initial adjuvant anastrozole therapy among women with early-stage breast cancer. J Clin Oncol. 2008;26:556-562.

16. Martin M, Brase JC, Calvo L, et al. Clinical validation of the EndoPredict test in node-positive, chemotherapy-treated ER+/HER2- breast cancer patients: results from the GEICAM 9906 trial. Breast Cancer Res. 2014;16:R38. 17. Zhang Y, Schnabel CA, Schroeder BE, et al. Breast cancer index identifies early-stage estrogen receptor-positive breast cancer patients at risk for early- and late-distant recurrence. Clin Cancer Res. 2013;19:41964205. 18. Buus R, Sestak I, Kronenwett R, et al. Comparison of EndoPredict and EPclin with Oncotype DX recurrence score for prediction of risk of distant recurrence after endocrine therapy. J Natl Cancer Inst. 2016;108:djw149. 19. Filipits M, Nielsen TO, Rudas M, et al; Austrian Breast and Colorectal Cancer Study Group. The PAM50 risk-of-recurrence score predicts risk for late distant recurrence after endocrine therapy in postmenopausal women with endocrine-responsive early breast cancer. Clin Cancer Res. 2014;20:1298-1305. 20. Henry NL, Braun TM, Ali HY, et al. Associations between use of the 21gene recurrence score assay and chemotherapy regimen selection in a statewide registry. Cancer. 2017;123:948-956. 21. Dubsky P, Filipits M, Jakesz R, et al; Austrian Breast and Colorectal Cancer Study Group (ABCSG). EndoPredict improves the prognostic classification derived from common clinical guidelines in ER-positive, HER2-negative early breast cancer. Ann Oncol. 2013;24:640-647. 22. Gnant M, Filipits M, Greil R, et al; Austrian Breast and Colorectal Cancer Study Group. Predicting distant recurrence in receptor-positive breast cancer patients with limited clinicopathological risk: using the PAM50 Risk of Recurrence score in 1478 postmenopausal patients of

30. Henry NL, Azzouz F, Desta Z, et al. Predictors of aromatase inhibitor discontinuation as a result of treatment-emergent symptoms in earlystage breast cancer. J Clin Oncol. 2012;30:936-942. 31. Esserman LJ, Thompson CK, Yau C, et al. Identification of tumors with an indolent disease course: MammaPrint ultralow signature validation in a retrospective analysis of a Swedish randomized tamoxifen trial. Cancer Res. 2016;76 (suppl; abstr P6-09-01). 32. Welch HG, Prorok PC, O’Malley AJ, et al. Breast-cancer tumor size, overdiagnosis, and mammography screening effectiveness. N Engl J Med. 2016;375:1438-1447. 33. Bramwell VH, Pritchard KI, Tu D, et al. A randomized placebo-controlled study of tamoxifen after adjuvant chemotherapy in premenopausal women with early breast cancer (National Cancer Institute of Canada— Clinical Trials Group Trial, MA.12). Ann Oncol. 2010;21:283-290. 34. Chia SK, Bramwell VH, Tu D, et al. A 50-gene intrinsic subtype classifier for prognosis and prediction of benefit from adjuvant tamoxifen. Clin Cancer Res. 2012;18:4465-4472. 35. Hammond ME, Hayes DF, Dowsett M, et al. American Society of Clinical Oncology/College Of American Pathologists guideline recom mendations for immunohistochemical testing of estrogen and pro gesterone receptors in breast cancer. J Clin Oncol. 2010;28:2784-2795. 36. Bartlett JM, Bloom KJ, Piper T, et al. Mammostrat as an immuno histochemical multigene assay for prediction of early relapse risk in the tamoxifen versus exemestane adjuvant multicenter trial pathology study. J Clin Oncol. 2012;30:4477-4484. 37. Cossetti RJ, Tyldesley SK, Speers CH, et al. Comparison of breast cancer recurrence and outcome patterns between patients treated from 1986 to 1992 and from 2004 to 2008. J Clin Oncol. 2015;33:65-73.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 113

HENRY, BEDARD, AND DEMICHELE

38. Saphner T, Tormey DC, Gray R. Annual hazard rates of recurrence for breast cancer after primary therapy. J Clin Oncol. 1996;14:2738-2746.

prognostic information on late distant metastases in ER+/HER2- breast cancer patients. Br J Cancer. 2013;109:2959-2964.

39. Davies C, Pan H, Godwin J, et al; Adjuvant Tamoxifen: Longer Against Shorter (ATLAS) Collaborative Group. Long-term effects of continuing adjuvant tamoxifen to 10 years versus stopping at 5 years after diagnosis of oestrogen receptor-positive breast cancer: ATLAS, a randomised trial. Lancet. 2013;381:805-816.

52. Sestak I, Zhang Y, Schroeder BE, et al. Cross-stratification and differential risk by breast cancer index and recurrence score in women with hormone receptor-positive lymph node-negative early-stage breast cancer. Clin Cancer Res. 2016;22:5043-5048.

40. Gray RG, Rea D, Handley K, et al. ATTom: long-term effects of continuing adjuvant tamoxifen to 10 years versus stopping at 5 years in 6,953 women with early breast cancer. J Clin Oncol. 2013;31 (suppl; abstr 5). 41. Goss PE, Ingle JN, Martino S, et al. A randomized trial of letrozole in postmenopausal women after five years of tamoxifen therapy for early-stage breast cancer. N Engl J Med. 2003;349:1793-1802. 42. Mamounas EP, Jeong JH, Wickerham DL, et al. Benefit from exemestane as extended adjuvant therapy after 5 years of adjuvant tamoxifen: intention-to-treat analysis of the National Surgical Adjuvant Breast And Bowel Project B-33 trial. J Clin Oncol. 2008;26:1965-1971. 43. Jakesz R, Greil R, Gnant M, et al; Austrian Breast and Colorectal Cancer Study Group. Extended adjuvant therapy with anastrozole among postmenopausal breast cancer patients: results from the randomized Austrian Breast and Colorectal Cancer Study Group Trial 6a. J Natl Cancer Inst. 2007;99:1845-1853. 44. Goss PE, Ingle JN, Pritchard KI, et al. Extending aromatase-inhibitor adjuvant therapy to 10 years. N Engl J Med. 2016;375:209-219. 45. Mamounas EP, Bandos H, Lembersky BC, et al. A randomized, doubleblinded, placebo-controlled clinical trial of extended adjuvant endocrine therapy (tx) with letrozole (L) in postmenopausal women with hormone-receptor (+) breast cancer (BC) who have completed previous adjuvant tx with an aromatase inhibitor (AI): results from NRG Oncology/NSABP B-42. Presented at: San Antonio Breast Cancer Symposium. December 2016; San Antonio, TX. 46. Blok EJ, van de Velde CJH, Meershoek-Klein Kranenbarg EM, et al. Optimal duration of extended letrozole treatment after 5 years of adjuvant endocrine therapy; results of the randomized phase III IDEAL trial (BOOG 2006-05). Presented at: San Antonio Breast Cancer Symposium. December 2016; San Antonio, TX. 47. Tjan-Heijnen VC, Van Hellemond IE, Peer PG, et al. First results from the multicenter phase III DATA study comparing 3 versus 6 years of anastrozole after 2-3 years of tamoxifen in postmenopausal women with hormone receptor-positive early breast cancer. Presented at: San Antonio Breast Cancer Symposium. December 2016; San Antonio,TX. 48. Wolmark N, Mamounas EP, Baehner FL, et al. Prognostic impact of the combination of recurrence score and quantitative estrogen receptor expression (ESR1) on predicting late distant recurrence risk in estrogen receptor-positive breast cancer after 5 years of tamoxifen: results from NRG Oncology/National Surgical Adjuvant Breast and Bowel Project B-28 and B-14. J Clin Oncol. 2016;34:2350-2358. 49. Sestak I, Dowsett M, Zabaglo L, et al. Factors predicting late recurrence for estrogen receptor-positive breast cancer. J Natl Cancer Inst. 2013;105:1504-1511. 50. Sestak I, Cuzick J, Dowsett M, et al. Prediction of late distant recurrence after 5 years of endocrine treatment: a combined analysis of patients from the Austrian breast and colorectal cancer study group 8 and arimidex, tamoxifen alone or in combination randomized trials using the PAM50 risk of recurrence score. J Clin Oncol. 2015;33:916-922. 51. Dubsky P, Brase JC, Jakesz R, et al; Austrian Breast and Colorectal Cancer Study Group (ABCSG). The EndoPredict score provides

114 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

53. Harris LN, Ismaila N, McShane LM, et al; American Society of Clinical Oncology. Use of biomarkers to guide decisions on adjuvant systemic therapy for women with early-stage invasive breast cancer: American Society of Clinical Oncology Clinical Practice Guideline. J Clin Oncol. 2016;34:1134-1150. 54. Denduluri N, Somerfield MR, Eisen A, et al. Selection of optimal adjuvant chemotherapy regimens for human epidermal growth factor receptor 2 (HER2)-negative and adjuvant targeted therapy for HER2positive breast cancers: an American Society of Clinical Oncology guideline adaptation of the Cancer Care Ontario Clinical Practice Guideline. J Clin Oncol. 2016;34:2416-2427. 55. Gradishar WJ, Anderson BO, Balassanian R, et al. NCCN Guidelines insights breast cancer, version 1.2016. J Natl Compr Canc Netw. 2015;13:1475-1485. 56. Coates AS, Winer EP, Goldhirsch A, et al; Panel Members. Tailoring therapies--improving the management of early breast cancer: St Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2015. Ann Oncol. 2015;26:1533-1546. 57. Henry NL, Somerfield MR, Abramson VG, et al. Role of patient and disease factors in adjuvant systemic therapy decision making for earlystage, operable breast cancer: American Society of Clinical Oncology Eendorsem*nt of Cancer Care Ontario Guideline recommendations. J Clin Oncol. 2016;34:2303-2311. 58. Ravdin PM, Siminoff LA, Davis GJ, et al. Computer program to assist in making decisions about adjuvant therapy for women with early breast cancer. J Clin Oncol. 2001;19:980-991. 59. Wishart GC, Bajdik CD, Dicks E, et al. PREDICT Plus: development and validation of a prognostic model for early breast cancer that includes HER2. Br J Cancer. 2012;107:800-807. 60. Maishman T, Copson E, Stanton L, et al; POSH Steering Group. An evaluation of the prognostic model PREDICT using the POSH cohort of women aged ≤40 years at breast cancer diagnosis. Br J Cancer. 2015;112:983-991. 61. Wishart GC, Bajdik CD, Azzato EM, et al. A population-based validation of the prognostic model PREDICT for early breast cancer. Eur J Surg Oncol. 2011;37:411-417. 62. Berry DA, Cirrincione C, Henderson IC, et al. Estrogen-receptor status and outcomes of modern chemotherapy for patients with nodepositive breast cancer. JAMA. 2006;295:1658-1667. 63. Paik S, Tang G, Shak S, et al. Gene expression and benefit of chemotherapy in women with node-negative, estrogen receptorpositive breast cancer. J Clin Oncol. 2006;24:3726-3734. 64. Albain KS, Barlow WE, Shak S, et al; Breast Cancer Intergroup of North America. Prognostic and predictive value of the 21-gene recurrence score assay in postmenopausal women with nodepositive, oestrogen-receptor-positive breast cancer on chemotherapy: a retrospective analysis of a randomised trial. Lancet Oncol. 2010;11: 55-65. 65. Harris LN, Ismaila N, McShane LM, et al. Reply to D.C. Sgroi et al, T. Sanft et al, M.S. Copur et al, and M.P. Goetz et al. J Clin Oncol. 2016;34:3946-3948.

GENOMIC TOOLS FOR BREAST CANCER DECISION SUPPORT

66. Howlader N, Noone AM, Krapcho M, et al. SEER Cancer Statistics Review, 1975–2013. https://seer.cancer.gov/csr/1975_2013/. Accessed March 5, 2017. 67. Gradishar WJ, Anderson BO, Balassanian R, et al. Breast cancer version 2.2015. J Natl Compr Canc Netw. 2015;13:448-475. 68. Zardavas D, Irrthum A, Swanton C, et al. Clinical management of breast cancer heterogeneity. Nat Rev Clin Oncol. 2015;12:381-394. 69. Aurilio G, Disalvatore D, Pruneri G, et al. A meta-analysis of oestrogen receptor, progesterone receptor and human epidermal growth factor receptor 2 discordance between primary breast cancer and metastases. Eur J Cancer. 2014;50:277-289. 70. Dieci MV, Barbieri E, Piacentini F, et al. Discordance in receptor status between primary and recurrent breast cancer has a prognostic impact: a single-institution analysis. Ann Oncol. 2013;24:101-108. 71. Hoefnagel LD, Moelans CB, Meijer SL, et al. Prognostic value of estrogen receptor α and progesterone receptor conversion in distant breast cancer metastases. Cancer. 2012;118:4929-4935. 72. Lindström LS, Karlsson E, Wilking UM, et al. Clinically used breast cancer markers such as estrogen receptor, progesterone receptor, and human epidermal growth factor receptor 2 are unstable throughout tumor progression. J Clin Oncol. 2012;30:2601-2608.

85. Parsons HA, Beaver JA, Cimino-Mathews A, et al. Individualized molecular analyses guide efforts (IMAGE): a prospective study of molecular profiling of tissue and blood in metastatic triple-negative breast cancer. Clin Cancer Res. 2017;23:379-386. 86. Esposito A, Bardelli A, Criscitiello C, et al. Monitoring tumor-derived cell-free DNA in patients with solid tumors: clinical perspectives and research opportunities. Cancer Treat Rev. 2014;40:648-655. 87. Lawrence MS, Stojanov P, Polak P, et al. Mutational heterogeneity in cancer and the search for new cancer-associated genes. Nature. 2013;499:214-218. 88. Higgins MJ, Jelovac D, Barnathan E, et al. Detection of tumor PIK3CA status in metastatic breast cancer using peripheral blood. Clin Cancer Res. 2012;18:3462-3469. 89. Madic J, Kiialainen A, Bidard FC, et al. Circulating tumor DNA and circulating tumor cells in metastatic triple negative breast cancer patients. Int J Cancer. 2015;136:2158-2165. 90. Rothé F, Laes JF, Lambrechts D, et al. Plasma circulating tumor DNA as an alternative to metastatic biopsies for mutational analysis in breast cancer. Ann Oncol. 2014;25:1959-1965. 91. Strauss WM, Carter C, Simmons J, et al. Analysis of tumor template from multiple compartments in a blood sample provides complementary access to peripheral tumor biomarkers. Oncotarget. 2016;7:26724-26738.

73. Cancer Genome Atlas Network. Comprehensive molecular portraits of human breast tumours. Nature. 2012;490:61-70.

92. Ding L, Ellis MJ, Li S, et al. Genome remodelling in a basal-like breast cancer metastasis and xenograft. Nature. 2010;464:999-1005.

74. Morganella S, Alexandrov LB, Glodzik D, et al. The topography of mutational processes in breast cancer genomes. Nat Commun. 2016;7:11383.

93. Krøigård AB, Larsen MJ, Lænkholm AV, et al. Clonal expansion and linear genome evolution through breast cancer progression from pre-invasive stages to asynchronous metastasis. Oncotarget. 2015;6:5634-5649.

75. Nik-Zainal S, Alexandrov LB, Wedge DC, et al; Breast Cancer Working Group of the International Cancer Genome Consortium. Mutational processes molding the genomes of 21 breast cancers. Cell. 2012;149:979-993. 76. Nik-Zainal S, Van Loo P, Wedge DC, et al; Breast Cancer Working Group of the International Cancer Genome Consortium. The life history of 21 breast cancers. Cell. 2012;149:994-1007. 77. André F, Bachelot T, Commo F, et al. Comparative genomic hybridisation array and DNA sequencing to direct treatment of metastatic breast cancer: a multicentre, prospective trial (SAFIR01/UNICANCER). Lancet Oncol. 2014;15:267-274. 78. Craig DW, O’Shaughnessy JA, Kiefer JA, et al. Genome and transcriptome sequencing in prospective metastatic triple-negative breast cancer uncovers therapeutic vulnerabilities. Mol Cancer Ther. 2013;12:104-116. 79. Manso L, Mourón S, Tress M, et al. Analysis of paired primary-metastatic hormone-receptor positive breast tumors (HRPBC) uncovers potential novel drivers of hormonal resistance. PLoS One. 2016;11:e0155840. 80. Roy-Chowdhuri S, de Melo Gagliato D, Routbort MJ, et al. Multigene clinical mutational profiling of breast carcinoma using next-generation sequencing. Am J Clin Pathol. 2015;144:713-721. 81. Vasan N, Yelensky R, Wang K, et al. A targeted next-generation sequencing assay detects a high frequency of therapeutically targetable alterations in primary and metastatic breast cancers: implications for clinical practice. Oncologist. 2014;19:453-458. 82. Arnedos M, Vicier C, Loi S, et al. Precision medicine for metastatic breast cancer--limitations and solutions. Nat Rev Clin Oncol. 2015;12:693-704. 83. Meric-Bernstam F, Brusco L, Shaw K, et al. Feasibility of large-scale genomic testing to facilitate enrollment onto genomically matched clinical trials. J Clin Oncol. 2015;33:2753-2762. 84. Zardavas D, Baselga J, Piccart M. Emerging targeted agents in metastatic breast cancer. Nat Rev Clin Oncol. 2013;10:191-210.

94. Shah SP, Morin RD, Khattra J, et al. Mutational evolution in a lobular breast tumour profiled at single nucleotide resolution. Nature. 2009;461:809-813. 95. Friedrich K, Weber T, Scheithauer J, et al. Chromosomal genotype in breast cancer progression: comparison of primary and secondary manifestations. Cell Oncol. 2008;30:39-50. 96. Kuukasjärvi T, Karhu R, Tanner M, et al. Genetic heterogeneity and clonal evolution underlying development of asynchronous metastasis in human breast cancer. Cancer Res. 1997;57:1597-1604. 97. Nishizaki T, DeVries S, Chew K, et al. Genetic alterations in primary breast cancers and their metastases: direct comparison using modified comparative genomic hybridization. Genes Chromosomes Cancer. 1997;19:267-272. 98. Wagle N, Berger MF, Davis MJ, et al. High-throughput detection of actionable genomic alterations in clinical tumor samples by targeted, massively parallel sequencing. Cancer Discov. 2012;2:82-93. 99. PRNewswire. The Prospective MOSCATO 01 Trial demonstrates that molecular “portraits” improve outcome of patients with metastatic cancer. Presented at: The MAP Meeting; September 2016; London, U.K. 100. Le Tourneau C, Delord JP, Gonçalves A, et al; SHIVA investigators. Molecularly targeted therapy based on tumour molecular profiling versus conventional therapy for advanced cancer (SHIVA): a multicentre, open-label, proof-of-concept, randomised, controlled phase 2 trial. Lancet Oncol. 2015;16:1324-1334. 101. Mullard A. NCI-MATCH trial pushes cancer umbrella trial paradigm. Nat Rev Drug Discov. 2015;14:513-515. 102. Van Poznak C, Somerfield MR, Bast RC, et al. Use of biomarkers to guide decisions on systemic therapy for women with metastatic breast cancer: American Society of Clinical Oncology Clinical Practice Guideline. J Clin Oncol. 2015;33:2695-2704.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 115

GNANT, VAN POZNAK, AND SCHNIPPER

Therapeutic Bone-Modifying Agents in the Nonmetastatic Breast Cancer Setting: The Controversy and a Value Assessment Michael Gnant, MD, FACS, Catherine Van Poznak, MD, and Lowell Schnipper, MD OVERVIEW Clinical trials and meta-analyses investigating bisphosphonates as an adjuvant breast cancer therapy have shown a consistent trend, with postmenopausal women and women receiving ovarian suppression with gonadotropin-releasing hormone therapy gaining improved breast cancer outcomes with the use of adjuvant bisphosphonate therapy. The interpretation of these data is controversial, because the primary endpoints of the majority of adjuvant bisphosphonate studies have been negative. Pros and cons as well as the value of adjuvant bisphosphonate therapy are discussed here.

espite notable recent advances in therapy, breast cancer remains one of the leading causes of cancer deaths among women worldwide. Adjuvant endocrine therapy is the state-of-the-art adjuvant treatment for all patients with estrogen receptor–positive early-stage breast cancer. For postmenopausal patients, aromatase inhibitors have been identified as one standard of care for this endocrine treatment, owing to their superior efficacy compared with tamoxifen, as demonstrated in several large clinical trials.1-8 The main side effect of aromatase inhibitors is their ability to compromise bone health,9-11 based on the (oncologically intended) reduction of estradiol levels.12 Thus, the concomitant use of bone-targeted agents has been extensively investigated to protect patients from these side effects and to prevent treatment-induced bone loss and fractures.13 Several trials using antiresorptive agents such as bisphosphonates have demonstrated that treatment-induced bone loss can be successfully prevented.13-21 It remains controversial whether these successful interventions actually lead to a notable reduction in the incidence of fractures. More recently, the anti–receptor activator of nuclear factor kappa-B ligand denosumab was shown to dramatically reduce fractures in a pivotal phase III trial. As a result, most clinical practice guidelines recommend monitoring of bone mineral density, as well as treatment with bisphosphonates or denosumab.22 In addition to their ability to protect and restore bone health for patients with early breast cancer, antiresorptive drugs have also been investigated for their oncologic benefits as adjuvant therapy. This was based on their antineo-

plastic potential, which was well described in preclinical in vitro and in vivo studies, as well as on putative indirect effects of antiresorptive therapies on the bone microenvironment.23 Clinical trials and meta-analyses investigating bisphosphonates as an adjuvant breast cancer therapy have shown a consistent trend, with postmenopausal women and women receiving ovarian suppression with gonadotropin-releasing hormone therapy gaining improved breast cancer outcomes with the use of adjuvant bisphosphonate therapy. The interpretation of these data is controversial, because the primary endpoints of adjuvant bisphosphonate studies have been negative. Pros and cons as well as the value of adjuvant bisphosphonate therapy are discussed here.

THE CASE FOR USING BONE-MODIFYING AGENTS AS ROUTINE ADJUVANT THERAPY

Early Studies

More than 2 decades ago, several trials investigated the adjuvant effects of the first-generation bisphosphonate clodronate. In a pivotal German trial,24 302 patients were selected because of the presence of disseminated tumor cells (DTCs) in their bone marrow and they were randomly assigned to receive oral clodronate or not. Early results from this trial demonstrated significant improvements in disease-free survival (DFS; 87% vs. 71% at 3 years; p < .0001) and even overall survival (96% vs. 85%; p = .0001); however, later updates did not confirm these findings. In addition, the Royal Marsden trial examining 2 years of clodronate treatment versus placebo yielded notable outcome improvements (HR at 5

From the Department of Surgery, Comprehensive Cancer Center, Medical University of Vienna, Waehringer Guertel, Austria; University of Michigan, Ann Arbor, MI; Hematology/ Oncology Division, Harvard Medical School, Beth Israel Deaconess Medical Center, Boston, MA. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Lowell Schnipper, MD, Hematology/Oncology Division, Harvard Medical School, Beth Israel Deaconess Medical Center, Rabb 430, 330 Brookline Ave., Boston, MA 02215; email: [emailprotected]. © 2017 American Society of Clinical Oncology

116 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

ADJUVANT THERAPEUTIC BONE-MODIFYING AGENTS IN BREAST CANCER

years, 0.69).25 A third adjuvant clodronate trial not only failed to confirm these promising results, but it even reported a detrimental effect of adjuvant clodronate on outcomes (i.e., increasing nonbone metastases).26 As a result of these controversial trial results, oral bisphosphonates never became accepted as a standard of care.

More Recent Studies With Oral Bisphosphonates

Oral bisphosphonates were also studied in more recent larger trials. NSABP B-34 is a randomized, double-blind, placebocontrolled study among more than 3,300 patients with breast cancer. Patients were stratified by age, axillary nodal status, and hormone receptor status and were randomly assigned to either 1,600 mg of oral clodronate per day for 3 years or placebo. After a median follow-up of almost 8 years, overall DFS did not differ between the groups (hazard ratio [HR], 0.91); however, a beneficial effect was observed for postmenopausal patients (see below).27 This was also seen in the GAIN study, a multicenter, open-label, randomized controlled phase III trial that recruited more than 3,000 patients to investigate the adjuvant effect of ibandronate in node-positive early breast cancer. Patients were randomized in a 2:1 ratio to either 50 mg of ibandronate per day or placebo for 2 years. Again, the overall results of this trial were negative: oral ibandronate did not improve outcomes of patients but there was a positive trend for DFS in the postmenopausal subgroup,28 fueling the discussion about differential effects of adjuvant bisphosphonates according to menopausal status.

Adjuvant Studies With Intravenous Aminobisphosphonates

Most adjuvant bisphosphonate studies have used zoledronic acid, the most potent bisphosphonate. With respect to DTCs, smaller studies among women with high-risk early breast cancer have reported that monthly zoledronic acid

KEY POINTS • Adjuvant breast cancer systemic therapy is given with curative intent. • Data are evolving to suggest that the adjuvant use of a bone-modifying agent (bisphosphonate or denosumab) may have anti–breast cancer effects for postmenopausal women (or women receiving ovarian-suppressing therapy). • The primary endpoints of the majority of adjuvant bisphosphonate studies to date have been negative, with the positive anticancer findings identified through subset analyses. • Controversy exists over whether the existing data are sufficient to influence adjuvant breast cancer treatment recommendations. • The value of adjuvant bone-modifying agents for adjuvant breast cancer care will be discussed, taking into account the clinical benefit and toxicity of these agents and their cost.

in addition to treatment with cytotoxic anticancer therapy can effectively increase DTC clearance and can reduce DTC numbers and persistence in bone marrow compared with standard therapy alone.29-31 These bisphosphonate-mediated decreases in DTC have been suggested as a potential mechanism underlying the observed clinical benefits in the large adjuvant studies. In the ABCSG-12 trial, anticancer effects with zoledronic acid were seen both in and outside bone. When patients who received zoledronic acid were compared with those who did not, improved DFS and fewer locoregional, visceral, and nonvisceral recurrences were observed.32 After longerterm follow-up (median 76 months), a persistent benefit in DFS more than 3 years after completion of treatment suggests a sustained “carryover” benefit from adding zoledronic acid to endocrine therapy.33 In addition, zoledronic acid also produced a significantly improved overall survival (HR, 0.59; p = .042). Further analyses on the mature dataset from ABCSG-12 revealed a significant difference in zoledronic acid treatment effects based on patient age at enrollment33: although no significant decrease was observed in women age 40 or younger, zoledronic acid produced a 42% reduction in the risk of DFS events among premenopausal women older than age 40 at study entry (HR, 0.58; p = .003). In addition, zoledronic acid was associated with a strong trend toward a 43% reduction in the risk of death for this older subset of patients (HR, 0.57; p = .057). These results were supported by two adjuvant zoledronic acid trials (ZO-FAST and Z-FAST) among postmenopausal women, in which DFS was a secondary end point. In ZO-FAST,34 the zoledronic acid group showed a significant DFS improvement of 41% (HR, 0.588; log-rank p = .0314) after a median follow-up of 3 years. The sibling study, Z-FAST,35 showed similar results, in which zoledronic acid yielded reduced disease recurrence at 61 months of follow-up. In the AZURE trial, however, the addition of zoledronic acid to standard adjuvant breast cancer therapy did not significantly increase DFS compared with the overall population.36 Notably, a differential effect of the bisphosphonate was observed with respect to menopausal status of trial patients. There was no difference in DFS with zoledronic acid among pre- or perimenopausal patients; however, among patients who were postmenopausal for at least 5 years before study entry, zoledronic acid significantly reduced the risk of DFS events by 24% (p = .02) and the risk of death by 29% (p = .017).37 Thus, the indirect metastasis-preventing effect of bisphosphonates is confined to postmenopausal women and to premenopausal women who receive ovarian function suppression, but not younger patients without ovarian function suppression. This suggests that estrogen effects on the bone microenvironment play a substantial role in determining who may benefit most from adjuvant bisphosphonate therapy.38 Eventually, the large Early Breast Cancer Trialists Collaborative Group meta-analysis confirmed this by assembling patient-level data on the majority of patients included in any asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 117

GNANT, VAN POZNAK, AND SCHNIPPER

adjuvant clinical bisphosphonate trial. Based on data from almost 19,000 patients, the results of this meta-analysis were clearly positive: bisphosphonates were demonstrated to have a positive effect on the recurrence of bone metastasis and overall survival for postmenopausal patients with breast cancer.39 It is important to note that outcome benefits appear to be confined to patients who are postmenopausal (either naturally or therapy induced) at diagnosis; clinically important benefits were seen for these women, with improvements in overall breast cancer recurrence, distant recurrence at any site, bone recurrence, and breast cancer-specific mortality (relative risk of 0.86, 0.82, 0.72, and 0.82, respectively).

DFS Results of Adjuvant Denosumab

In an early and premature analysis of the ABCSG-18 data,40 adjuvant denosumab appears to yield DFS benefits that are similar to what was observed in the bisphosphonate metaanalysis. Further follow-up of this trial as well as the results of the large D-CARE trial are expected for 2018 and will clarify the outcome effects of adjuvant denosumab.

RESERVATIONS ON THE USE OF BONEMODIFYING AGENTS AS ANTICANCER THERAPY IN THE NONMETASTATIC SETTING

Adjuvant breast cancer studies investigating bone-modifying agents with a bisphosphonate or denosumab have been reported, and additional studies are ongoing.41 The majority of reported studies have identified a positive effect of the bone-modifying agent in secondary or exploratory analyses. The anticancer benefits are particular to postmenopausal women. For the purposes of this discussion, the term “postmenopausal” applies to women who are clinically not pre- or perimenopausal and includes women receiving ovarian suppression with gonadotropin-releasing hormone therapy. The fundamentals of bisphosphonate pharmacology are known, including absorption, distribution, elimination, pharmaco*kinetics, and pharmacodynamics and the impact that structural alterations to the bisphosphonate chemical structure have on potency.42 Bisphosphonates enter the osteoclast by endocytosis. Nitrogen-containing bisphosphonates, such as zoledronic acid and ibandronate, inhibit farnesyl pyrophosphate synthase and prevent the prenylation of small guanosine 5′-triphosphatase proteins essential for the function and survival of osteoclasts. The non–nitrogencontaining bisphosphonates, such as clodronate, are incorporated into adenosine 5′-triphosphate analogs in the osteoclast and promote apoptosis.42 Bisphosphonates may have direct or indirect antitumor effects within the bone and may impact tissues outside of the skeleton. There are data suggesting that bisphosphonates alter tumor behavior; affect host or tumor vasculature, the tumor microenvironment, and its associated immune cells, fibroblasts, stromal cells, and macrophages; and effect circulating factors.43,44 It is not known whether any, or which, of these properties are related to the potential anticancer 118 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

effects seen with adjuvant bisphosphonate therapy for postmenopausal women with breast cancer. Incorporating an adjuvant bone-modifying agent into the care plan of postmenopausal women may reduce the risk of osteoporosis, osteoporotic fractures, and possibly the risk of breast cancer outcomes. The risk of serious toxicities affecting the gastrointestinal tract, renal system, osteonecrosis of the jaw, and atypical fractures appears to be lower than the potential benefits in anticancer outcomes. Yet the use of adjuvant bisphosphonates for postmenopausal women with early-stage breast cancer does not appear to be uniformly embraced by clinicians, patients, or health care payers. Indeed, the Canadian Bone and the Oncologist New Updates meeting debated “are adjuvant bisphosphonates now standard of care in early stage breast cancer” and the majority of the meeting attendees voted “no.”45 Ten potential reasons to hesitate before adopting adjuvant bisphosphonates in the care of postmenopausal women with breast cancer are outlined here. 1. The adjuvant bisphosphonate clinical trials were not designed to test the hypothesis that there would be an effect in postmenopausal women that differs from pre- and perimenopausal women. 2. Meta-analyses do not substitute for well-designed, randomized clinical trials testing an a priori hypothesis. 3. Confidence in the data and the durability of the anticancer findings may be questioned. First, ZO-FAST and Z-FAST are “twin” studies in which zoledronic acid was used up front or in a delayed manner for postmenopausal women receiving an adjuvant aromatase inhibitor.46 These parallel studies do not report the same cancer outcomes at 5 years. ZO-FAST reports improved DFS with the use of immediate use of zoledronic acid (HR, 0.66; 95% CI, 0.44–0.97; p = .0375), and Z-FAST does not have statistical significance for disease recurrence or death between the arms.34,47 Second, the long-term follow-up of ABCSG-12 does not show retained statistical improvement in DFS.48 The long-term follow-up of AZURE does not show maintained statistical significance for improved overall survival with use of zoledronic acid for participants who were more than 5 years since menopause.37 4. No proven mechanism to explain the different cancer outcomes among pre-, peri-, and postmenopausal women has been identified. 5. For postmenopausal women, adjuvant bisphospho nates appear to affect the risk of bone metastases. Estrogen receptor–positive breast cancers are more likely to metastasize to bone than estrogen receptor–negative, progesterone receptor–negative, and HER2-negative (triple-negative) breast cancers.49 Yet tumor characteristics (i.e., estrogen receptor status) did not correlate with benefit from adjuvant bisphosphonate use. 6. Patient selection factors that may be used in the decision to treat with adjuvant bone-modifying therapy are not refined. The use of adjuvant bisphosphonate

ADJUVANT THERAPEUTIC BONE-MODIFYING AGENTS IN BREAST CANCER

therapy for all postmenopausal women with a risk of breast cancer recurrence seems indiscriminate. 7. If the decision is made to use an adjuvant bonemodifying agent, the data do not provide clarity on the optimal time to start an adjuvant bone-modifying agent or on the drug selection, dose, dosing interval, or duration of therapy. 8. If there are challenges in tolerating adjuvant therapy, it is not known whether the bone-modifying agent should be discontinued or perhaps changed to an alternative drug and/or dosing schedule. 9. Polypharmacy can negatively influence compliance with, adherence to, and persistence of medication use. These factors must be investigated as related to adjuvant bone-modifying therapy. 10. The value of using an adjuvant bone-modifying agent and the financial toxicities to the patient and the health care system have not been prospectively defined. In summary, confidence in the data is undermined by the limited hypothesis testing and the absence of a proven biologic mechanism to account for the reported anticancer outcomes among postmenopausal women treated with adjuvant bisphosphonate therapy. This is complicated by challenges in identifying which postmenopausal women to treat and an optimal treatment regimen. Guiding principles in medical care address having an understanding of the drug to be used, its mode of action, the risk-benefit profile for the patient to be treated, the dosage to be prescribed, the dosing interval to be used, the duration of therapy, and the toxicity profile (which may include financial toxicities) prior to prescribing a medication. Today, there remains uncertainty on many of these key factors. Additional data are needed to optimally understand the utility of adjuvant bone-modifying agents and to identify those patients most likely to benefit from therapy.

BISPHOSPHONATES AS ADJUVANT THERAPY FOR EARLY BREAST CANCER: A VALUE ASSESSMENT

The value associated with the use of bisphosphonates as adjunctive therapy for early breast cancer is controversial. Administration of bisphosphonates has two plausible goals: one is to abrogate bone loss (particularly when using aromatase inhibitors) and the other relates to potential beneficial effects on disease recurrence, as well as mortality, for postmenopausal women. Multiple studies have addressed this question, yet controversy remains. The following remarks attempt to view the question of the utility of these agents in the context of the value framework developed by ASCO. ASCO’s Value in Cancer Care Task Force has been committed to identifying approaches that support the highest quality of cancer care, while bending the cost curve downward. The basic assumption is that provision of high-quality cancer prevention and treatment practices at the lowest reasonable cost is the embodiment of high-value care. Examples of the initiatives the ASCO task force has undertaken thus

far include participation in the “Choosing Wisely” campaign of the American Board of Internal Medicine Foundation and Consumer Reports.50,51 Ten commonly used practices in medical oncology were identified on the basis of there being no evidence that they add clinical value but do have high associated costs. These included unnecessary imaging, unnecessary staging in early cases of breast and prostate cancer, proper use of cytokines and high-quality end-of-life care that shifts from cancer-directed to symptom-directed treatment,50,51 and the development of a model system with which to assess the value of antineoplastic therapies.52,53 ASCO’s model framework enables assessment of the incremental benefit associated with a new drug or regimen when compared in a prospective randomized clinical trial. The original intent of the framework was to provide physicians and patients with a rapidly accessible, easily comprehensible way of viewing comparative options for management of a specific cancer, as well as to demonstrate the cost associated with administering the therapies both from the perspective of societal and out-of-pocket costs to the patient. The latter have been a steadily growing burden, frequently resulting in emotional stress, physical symptoms, personal bankruptcy, and possibly earlier mortality.52,53 The use of tools such as this and other frameworks is another area that is receiving substantial attention from the payer community. The ASCO Value Assessment Framework54,55 measures the incremental clinical impact (termed the net health benefit [NHB]) of a new therapy compared with a control treatment when these have been evaluated in a prospective randomized trial. The framework includes several parameters, each of which is assigned a maximum point score that reflects the difference between the arms in a trial. Two distinct frameworks have been developed, one for advanced disease and the other for the potentially curative setting. The elements within the framework are as follows: 1. Clinical benefit: the maximum score is 100 when comparing magnitude of difference of the test regimen and its control. 2. Toxicity: a maximum of 20 points can be added or subtracted from the clinical benefit score depending on the relative toxicity or lack thereof of the new therapy compared with the comparator. 3. Bonus points: these are awarded for specific mile stones reached such as improved symptom control in advanced disease. 4. NHB: the NHB represents the sum of clinical benefit and bonus scores from which is subtracted (or added) the toxicity score. 5. Cost for adjuvant regimens or the total cost (based on average sale price of the control and test regimens): the cost of the new regimen is not factored into a final value assessment but is represented in parallel with the clinical benefit. The goal is to promote discussion between physician and patient about the clinical impact of a therapy and whether the extent of benefit is justified by the financial cost to the patient and family unit. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 119

GNANT, VAN POZNAK, AND SCHNIPPER

The Early Breast Cancer Trialists Collaborative Group performed a meta-analysis of all prospective randomized trials and concluded that “some years of adjuvant bisphosphonate treatment can reduce breast cancer recurrence rates in bone and improve breast cancer survival, but have provided clear evidence of benefit only in women who are postmenopausal (natural or induced) at the time bisphosphonates are started.”39 The European Society of Medical Oncology has evaluated the aggregated data from a large number of clinical trials and has published guidelines that support the use of adjuvant bisphosphonates as part of the systemic treatment of early-stage breast cancer for postmenopausal women, whether they are postmenopausal through a natural menopause or as a result of ovarian suppression.41 The value assessment of adjuvant bisphosphonate is based on a subset of 11,767 postmenopausal women included in the meta-analysis of trials comparing bisphosphonate use versus no bisphosphonate use in early breast cancer. The included studies used durations of bisphosphonate treatment varying between 2 and 5 years, with no conclusion as to the optimal period of treatment. The meta-analysis demonstrated an 18% reduction in the risk of breast cancer mortality at 10 years among post-

menopausal women who took a bisphosphonate compared with those treated with placebo. In absolute terms, the advantage is quite small (3.3%). When using the ASCO value framework, this difference in outcome yields a clinical benefit score of 18 points (of a possible 100). Most studies of bisphosphonates registered small increases in toxicity. Were there no toxicity, the ASCO framework would yield no added or subtracted points. Because the toxicities were generally few and mild, for the purposes of this analysis, the clinical benefit will be reduced by 5%, thereby subtracting 1 from the clinical benefit score to yield an NHB of 17. To place this in perspective, in the study comparing ibrutinib to chlorambucil as first-line therapy for chronic lymphocytic leukemia, the NHB score was 77 based upon a reduction in risk of death of 84%. In a comparison between chemotherapy (doxorubicin, cyclophosphamide, and pacl*taxel vs. the same agents plus trastuzumab), the NHB score in favor of the trastuzumab-containing regimen was 47. The cost of the adjuvant endocrine therapy plus bisphosphonate will be presented in parallel with the NHB calculations. Comparisons with clinical trials in which the lest agent yielded a larger NHB will be shared to demonstrate the range of possible outcomes in the value analysis.

References 1. Burstein HJ, Temin S, Anderson H, et al. Adjuvant endocrine therapy for women with hormone receptor-positive breast cancer: American Society of Clinical Oncology clinical practice guideline focused update. J Clin Oncol. 2014;32:2255-2269.

8. Thürlimann B, Keshaviah A, Coates AS, et al; Breast International Group (BIG) 1-98 Collaborative Group. A comparison of letrozole and tamoxifen in postmenopausal women with early breast cancer. N Engl J Med. 2005;353:2747-2757.

2. Dubsky PC, Jakesz R, Mlineritsch B, et al. Tamoxifen and anastrozole as a sequencing strategy: a randomized controlled trial in postmenopausal patients with endocrine-responsive early breast cancer from the Austrian Breast and Colorectal Cancer Study Group. J Clin Oncol. 2012;30:722-728.

9. Becker T, Lipscombe L, Narod S, et al. Systematic review of bone health in older women treated with aromatase inhibitors for early-stage breast cancer. J Am Geriatr Soc. 2012;60:1761-1767.

3. Dowsett M, Forbes JF, Bradley R, et al; Early Breast Cancer Trialists' Collaborative Group. Aromatase inhibitors versus tamoxifen in early breast cancer: patient-level meta-analysis of the randomised trials. Lancet. 2015;386:1341-1352. 4. Forbes JF, Cuzick J, Buzdar A, et al; Arimidex, Tamoxifen, Alone or in Combination (ATAC) Trialists’ Group. Effect of anastrozole and tamoxifen as adjuvant treatment for early-stage breast cancer: 100-month analysis of the ATAC trial. Lancet Oncol. 2008;9:45-53. 5. Goss PE, Ingle JN, Martino S, et al. Randomized trial of letrozole following tamoxifen as extended adjuvant therapy in receptor-positive breast cancer: updated findings from NCIC CTG MA.17. J Natl Cancer Inst. 2005;97:1262-1271. 6. Howell A, Cuzick J, Baum M, et al; ATAC Trialists’ Group. Results of the ATAC (Arimidex, Tamoxifen, Alone or in Combination) trial after completion of 5 years’ adjuvant treatment for breast cancer. Lancet. 2005;365:60-62. 7. Jakesz R, Greil R, Gnant M, et al; Austrian Breast and Colorectal Cancer Study Group. Extended adjuvant therapy with anastrozole among postmenopausal breast cancer patients: results from the randomized Austrian Breast and Colorectal Cancer Study Group Trial 6a. J Natl Cancer Inst. 2007;99:1845-1853.

120 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

10. Body JJ. Increased fracture rate in women with breast cancer: a review of the hidden risk. BMC Cancer. 2011;11:384. 11. Bouvard B, Soulié P, Hoppé E, et al. Fracture incidence after 3 years of aromatase inhibitor therapy. Ann Oncol. 2014;25:843-847. 12. Cummings SR, Browner WS, Bauer D, et al; Study of Osteoporotic Fractures Research Group. Endogenous hormones and the risk of hip and vertebral fractures among older women. N Engl J Med. 1998;339:733-738. 13. Gnant M. Role of bisphosphonates in postmenopausal women with breast cancer. Cancer Treat Rev. 2014;40:476-484. 14. Gnant M, Mlineritsch B, Luschin-Ebengreuth G, et al; Austrian Breast and Colorectal Cancer Study Group (ABCSG). Adjuvant endocrine therapy plus zoledronic acid in premenopausal women with earlystage breast cancer: 5-year follow-up of the ABCSG-12 bone-mineral density substudy. Lancet Oncol. 2008;9:840-849. 15. Brufsky A, Bundred N, Coleman R, et al; Z-FAST and ZO-FAST Study Groups. Integrated analysis of zoledronic acid for prevention of aromatase inhibitor-associated bone loss in postmenopausal women with early breast cancer receiving adjuvant letrozole. Oncologist. 2008;13:503-514. 16. Bundred NJ, Campbell ID, Davidson N, et al. Effective inhibition of aromatase inhibitor-associated bone loss by zoledronic acid in

ADJUVANT THERAPEUTIC BONE-MODIFYING AGENTS IN BREAST CANCER

postmenopausal women with early breast cancer receiving adjuvant letrozole: ZO-FAST study results. Cancer. 2008;112:1001-1010. 17. Body JJ. Aromatase inhibitors-induced bone loss in early breast cancer. Bonekey Rep. 2012;1:201. 18. Delmas PD, Balena R, Confravreux E, et al. Bisphosphonate risedronate prevents bone loss in women with artificial menopause due to chemotherapy of breast cancer: a double-blind, placebo-controlled study. J Clin Oncol. 1997;15:955-962. 19. Powles TJ, McCloskey E, Paterson AH, et al. Oral clodronate and reduction in loss of bone mineral density in women with operable primary breast cancer. J Natl Cancer Inst. 1998;90:704-708. 20. Saarto T, Blomqvist C, Välimäki M, et al. Chemical castration induced by adjuvant cyclophosphamide, methotrexate, and fluorouracil chemotherapy causes rapid bone loss that is reduced by clodronate: a randomized study in premenopausal breast cancer patients. J Clin Oncol. 1997;15:1341-1347. 21. Shapiro CL, Halabi S, Hars V, et al. Zoledronic acid preserves bone mineral density in premenopausal women who develop ovarian failure due to adjuvant chemotherapy: final results from CALGB trial 79809. Eur J Cancer. 2011;47:683-689. 22. Gnant M, Pfeiler G, Dubsky PC, et al; Austrian Breast and Colorectal Cancer Study Group. Adjuvant denosumab in breast cancer (ABCSG-18): a multicentre, randomised, double-blind, placebocontrolled trial. Lancet. 2015;386:433-443. 23. Gnant M, Clézardin P. Direct and indirect anticancer activity of bisphosphonates: a brief review of published literature. Cancer Treat Rev. 2012;38:407-415. 24. Diel IJ, Jaschke A, Solomayer EF, et al. Adjuvant oral clodronate improves the overall survival of primary breast cancer patients with micrometastases to the bone marrow: a long-term follow-up. Ann Oncol. 2008;19:2007-2011. 25. Powles T, Paterson A, McCloskey E, et al. Reduction in bone relapse and improved survival with oral clodronate for adjuvant treatment of operable breast cancer [ISRCTN83688026]. Breast Cancer Res. 2006;8:R13. 26. Saarto T, Vehmanen L, Virkkunen P, et al. Ten-year follow-up of a randomized controlled trial of adjuvant clodronate treatment in nodepositive breast cancer patients. Acta Oncol. 2004;43:650-656. 27. Paterson AH, Anderson SJ, Lembersky BC, et al. Oral clodronate for adjuvant treatment of operable breast cancer (National Surgical Adjuvant Breast and Bowel Project protocol B-34): a multicentre, placebo-controlled, randomised trial. Lancet Oncol. 2012;13:734-742. 28. von Minckwitz G, Möbus V, Schneeweiss A, et al. German adjuvant intergroup node-positive study: a phase III trial to compare oral ibandronate versus observation in patients with high-risk early breast cancer. J Clin Oncol. 2013;31:3531-3539. 29. Aft R, Naughton M, Trinkaus K, et al. Effect of zoledronic acid on disseminated tumour cells in women with locally advanced breast cancer: an open label, randomised, phase 2 trial. Lancet Oncol. 2010;11:421-428. 30. Banys M, Solomayer EF, Gebauer G, et al. Influence of zoledronic acid on disseminated tumor cells in bone marrow and survival: results of a prospective clinical trial. BMC Cancer. 2013;13:480. 31. Rack B, Jückstock J, Genss EM, et al. Effect of zoledronate on persisting isolated tumour cells in patients with early breast cancer. Anticancer Res. 2010;30:1807-1813.

32. Gnant M, Mlineritsch B, Schippinger W, et al; ABCSG-12 Trial Investigators. Endocrine therapy plus zoledronic acid in premenopausal breast cancer. N Engl J Med. 2009;360:679-691. 33. Gnant M, Mlineritsch B, Stoeger H, et al; Austrian Breast and Colorectal Cancer Study Group, Vienna, Austria. Adjuvant endocrine therapy plus zoledronic acid in premenopausal women with early-stage breast cancer: 62-month follow-up from the ABCSG-12 randomised trial. Lancet Oncol. 2011;12:631-641. 34. Coleman R, de Boer R, Eidtmann H, et al. Zoledronic acid (zoledronate) for postmenopausal women with early breast cancer receiving adjuvant letrozole (ZO-FAST study): final 60-month results. Ann Oncol. 2013;24:398-405. 35. Brufsky A, Harker WG, Beck JT, et al. Zoledronic acid inhibits adjuvant letrozole-induced bone loss in postmenopausal women with early breast cancer. J Clin Oncol. 2007;25:829-836. 36. Coleman RE, Marshall H, Cameron D, et al; AZURE Investigators. Breast-cancer adjuvant therapy with zoledronic acid. N Engl J Med. 2011;365:1396-1405. 37. Coleman R, Cameron D, Dodwell D, et al; AZURE investigators. Adjuvant zoledronic acid in patients with early breast cancer: final efficacy analysis of the AZURE (BIG 01/04) randomised open-label phase 3 trial. Lancet Oncol. 2014;15:997-1006. 38. Ottewell PD, Wang N, Brown HK, et al. Zoledronic acid has differential antitumor activity in the pre- and postmenopausal bone microenvironment in vivo. Clin Cancer Res. 2014;20:2922-2932. 39. Coleman R, Powles T, Paterson A, et al; Early Breast Cancer Trialists’ Collaborative Group (EBCTCG). Adjuvant bisphosphonate treatment in early breast cancer: meta-analyses of individual patient data from randomised trials. Lancet. 2015;386:1353-1361. 40. Gnant M, Pfeiler G, Dubsky PC, et al. The impact of adjuvant denosumab on disease-free survival: results from 3,425 postmenopausal patients of the ABCSG-18 trial. Presented at: San Antonio Breast Cancer Symposium. San Antonio, TX; 2015. Abstract S2-02. 41. Hadji P, Coleman RE, Wilson C, et al. Adjuvant bisphosphonates in early breast cancer: consensus guidance for clinical practice from a European Panel. Ann Oncol. 2016;27:379-390. 42. Russell RG, Xia Z, Dunford JE, et al. Bisphosphonates: an update on mechanisms of action and how these relate to clinical efficacy. Ann N Y Acad Sci. 2007;1117:209-257. 43. Holen I, Coleman RE. Anti-tumour activity of bisphosphonates in preclinical models of breast cancer. Breast Cancer Res. 2010;12: 214. 44. Santini D, Stumbo L, Spoto C, et al. Bisphosphonates as anticancer agents in early breast cancer: preclinical and clinical evidence. Breast Cancer Res. 2015;17:121. 45. Jacobs C, Amir E, Paterson A, et al. Are adjuvant bisphosphonates now standard of care of women with early stage breast cancer? A debate from the Canadian Bone and the Oncologist New Updates meeting. J Bone Oncol. 2015;4:54-58. 46. Aapro M. Improving bone health in patients with early breast cancer by adding bisphosphonates to letrozole: the Z-ZO-E-ZO-FAST program. Breast. 2006;15 (Suppl 1):30-40. 47. Brufsky AM, Harker WG, Beck JT, et al. Final 5-year results of Z-FAST trial: adjuvant zoledronic acid maintains bone mass in postmenopausal breast cancer patients receiving letrozole. Cancer. 2012;118:11921201.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 121

GNANT, VAN POZNAK, AND SCHNIPPER

48. Gnant M, Mlineritsch B, Stoeger H, et al; Austrian Breast and Colorectal Cancer Study Group, Vienna, Austria. Zoledronic acid combined with adjuvant endocrine therapy of tamoxifen versus anastrozol plus ovarian function suppression in premenopausal early breast cancer: final analysis of the Austrian Breast and Colorectal Cancer Study Group Trial 12. Ann Oncol. 2015;26:313-320. 49. Kennecke H, Yerushalmi R, Woods R, et al. Metastatic behavior of breast cancer subtypes. J Clin Oncol. 2010;28:3271-3277.

52. Zafar SY, Peppercorn JM, Schrag D, et al. The financial toxicity of cancer treatment: a pilot study assessing out-of-pocket expenses and the insured cancer patient’s experience. Oncologist. 2013;18: 381-390. 53. Institute of Medicine. Delivering Affordable Cancer Care in the 21st Century: Workshop Summary. Washington, DC: National Academies Press; 2013.

50. Schnipper LE, Smith TJ, Raghavan D, et al. American Society of Clinical Oncology identifies five key opportunities to improve care and reduce costs: the top five list for oncology. J Clin Oncol. 2012;30:1715-1724.

54. Schnipper LE, Davidson NE, Wollins DS, et al; American Society of Clinical Oncology. American Society of Clinical Oncology statement: a conceptual framework to assess the value of cancer treatment options. J Clin Oncol. 2015;33:2563-2577.

51. Schnipper LE, Lyman GH, Blayney DW, et al. American Society of Clinical Oncology 2013 top five list in oncology. J Clin Oncol. 2013;31:43624370.

55. Schnipper LE, Davidson NE, Wollins DS, et al. Updating the American Society of Clinical Oncology Value Framework: revisions and reflections in response to comments received. J Clin Oncol. 2016;4:2925-2934.

122 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

CANCER PREVENTION, HEREDITARY GENETICS, AND EPIDEMIOLOGY

MOURITS AND DE BOCK

European/U.S. Comparison and Contrasts in Ovarian Cancer Screening and Prevention in a High-Risk Population Marian J. Mourits, MD, PhD, and G. H. de Bock, PhD OVERVIEW The history of screening and prevention of ovarian cancer among high-risk women in the United States and Europe is one of mutual inspiration, with researchers learning from each others’ findings and insights and collaborating with investigators from both sides of the Atlantic ocean. Examples of simultaneous and joint development of knowledge and scientific points of view include the paradigm shift from ovarian to fallopian tube high-grade serous cancer and the cessation of simultaneous adoption of ovarian cancer screening by clinicians in both the United States and Europe. Examples of joint efforts with fruitful results include international collaboration in large population-based, genome-wide association studies and in epidemiologic database studies. Research in the field of hereditary ovarian cancer is a great example of mutual inspiration and joint efforts for the purpose of improving knowledge and health care for women with hereditary ovarian cancer.

varian cancer is the most lethal of all gynecologic cancers. The poor prognosis of ovarian cancer is largely attributable to the fact that patients with the disease present late. Although the symptom index for ovarian cancer may help to identify women with the disease, symptoms are not early signs, and most women are diagnosed at an advanced stage.1,2 Once the malignancy is detected, usually when classified at International Federation of Gynecology and Obstetrics stages III to IV, standard treatment consists of a combination of debulking surgery and chemotherapy, and survival rates have shown little improvement.3 Over the last decade, it became clear that ovarian cancer is not a single disease; different histologic subtypes of epithelial ovarian cancer with different molecular pathogeneses and prognoses can be identified.4 This knowledge will guide future research initiatives to improve early detection of and prognosis for epithelial ovarian cancer.5

OVARIAN CANCER SCREENING

Although ovarian cancer screening cannot prevent cancer, it was long hoped that screening might permit detection at an early stage when a cure is possible. Data from general population screening were disappointing; therefore, a large, more sophisticated screening study began in the 1990s in the United Kingdom.6 Postmenopausal women age 45 or older were randomly assigned to a screening or control group. Women randomly assigned to screening were offered three annual screens that included: cancer anti-

gen 125 (CA125) measurements; pelvic ultrasonographies if the CA125 measurement was greater than 30 U/mL; and referrals for gynecologic counseling if the ovarian volume reached 8.8 mL or greater. The development of epithelial ovarian cancer was the study endpoint. The median survival of women with index cancers was longer for the screened group than for the control group (72.9 vs. 41.8 months; p = .01), however, the number of deaths attributable to ovarian cancer did not differ.6 To further improve screening results, a new ovarian cancer risk algorithm was designed using pelvic ultrasonography and trends in serum CA125. This algorithm was developed by the U.K. Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) group, another large screening study.7 Outcomes of the UKCTOCS study showed a favorable stage distribution using the risk of ovarian cancer algorithm, however, there was no notable survival benefit in the screened group compared with the control group.8 In the United States, the Prostate, Lung, Colorectal, and Ovarian (PLCO) cancer screening trial investigators randomly assigned women between age 55–74 to an annual screening group and a control group. The screened group underwent an annual pelvic ultrasound and serum CA125 measurement. Increased morbidity was reported owing to high false-positive results (8%) in the screening group, which resulted in women undergoing surgery, however, no reduction in ovarian cancer mortality by screening was found.9 Because the positive and negative predictive value of screening depends on the incidence of the disease, screening was

From the Departments of Gynecologic Oncology and Epidemiology, University of Groningen, University Medical Center Groningen, Groningen, Netherlands. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Marian J. Mourits, MD, PhD, Department of Gynecologic Oncology, University of Groningen, University Medical Center Groningen, Hanzeplein 1, 9700RB Groningen, Netherlands; email: [emailprotected]. © 2017 American Society of Clinical Oncology

124 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

U.S./E.U. COMPARISON AND CONTRASTS IN HIGH-RISK OVARIAN CANCER CARE

expected to be more effective for a high-risk population. The U.K. Familial Ovarian Screening Study (UKFOCSS) was developed as a prospective cohort study to assess the value of screening in a high-risk population specifically. The UKFOCSS recruited more than 5,000 high-risk women between 2002 and 2009, and screening was performed with four monthly CA125 measurements analyzed by the risk of ovarian cancer algorithm. Although the final UKFOCSS results are not yet available, screening is not expected to improve ovarian cancer–specific survival nor to be cost-effective.

HEREDITARY OVARIAN CANCER

Since the BRCA genes were discovered in 1994 and 1995,10,11 clinicians worldwide have begun developing guidelines for a systematic ovarian cancer screening program for women with BRCA1/2 mutations, consisting mostly of an annual pelvic ultrasound and serum CA125 measurement.12-16 Lynch syndrome (LS) is another hereditary syndrome with an increased ovarian cancer risk. LS is an autosomal dominant predisposition characterized by germline mutations in one of four DNA mismatch repair genes: MLH1, MSH2, MSH6, and PMS2.17 For female carriers with LS, endometrial cancer is, after colon cancer, the most common tumor type with a cumulative lifetime risk of 21%–71%; the risk of ovarian cancer is between 6% and 12%.18 Because of these high cancer risks, women with LS are regularly surveyed. Endometrial cancer surveillance seems to be effective in early detection of endometrial cancer19-21; however, the value of surveillance for ovarian cancer has not yet been proven.19,22 In a recent review on ovarian cancer in LS, the mean age of women with LS and ovarian cancer was 45.3 years and patients had a wide age range of onset (between age 19–82).23 For these patients, ovarian cancer was mostly diagnosed at an early stage (International Federation of Gynecology and Obstetrics stage I–II), exhibited a variety of histopathological subtypes (frequently endometrioid or clear cell), and had a survival rate of 86%.23 Data on the role of surveillance in the detection of ovarian cancer in women with LS were scarce, and the early stage could not be attributed to screening.23

KEY POINTS • Ovarian cancer screening is not effective in early detection of the disease. • Most, if not all, high-grade serous ovarian cancers arise in the fallopian tube. • All women with epithelial ovarian cancer should be offered genetic counseling and testing to reduce morbidity and mortality for patients and their relatives. • The only effective strategy to prevent high-risk women from dying of the disease is to remove the ovaries and fallopian tubes before the cancer incidence rises. • Research in the field of hereditary ovarian cancer is an example of a joint effort and fruitful collaboration between researchers on both sides of the Atlantic ocean.

TIME TO STOP OVARIAN CANCER SCREENING After 2 decades of ovarian cancer screening, and despite major efforts in large prospective trials, no evidence of a survival benefit of screening has been reported. Clinicians, almost simultaneously in the United States and European Union, began to omit gynecologic screening and instead adopted risk-reducing salpingo-oophorectomy (RRSO) and reported on their results.24-30 In 2009, a meta-analysis on risk-reduction estimates showed that RRSO, performed at ages 35–40 for BRCA1 and 40–45 for BRCA2 mutation carriers (i.e., before the cancer incidence rises31), is effective in the detection of more than 96% of BRCA-associated ovarian cancers (hazard ratio, 0.21; 95% CI, 0.12–0.39).32

NEW PARADIGM OF OVARIAN CANCER IN BRCA1/2 MUTATION CARRIERS

Since the adoption of RRSO for BRCA1/2 mutation carriers, increasing percentages of fallopian tube (pre)malignancies have been found. In 1998, Dubeau33 was the first to propose that the various ovarian cancers (serous, endometrioid, mucinous, and clear cell) resemble the epithelium of the fallopian tube, endometrium, endocervix, and gastrointestinal tract, respectively. In 2001, a group of Dutch researchers published a small series on the fallopian tubes of high-risk women and found preneoplastic lesions in benign fallopian tube tissue, not in controls.34 One patient, a BRCA1 mutation carrier, showed loss of the wild-type BRCA1 allele in a severely dysplastic lesion of the distal fallopian tube.34 The publication by Piek et al34 opened the eyes of many pathologists around the world, including Crum and colleagues35 in Boston, Massachusetts, who were the most successful in further elaborating the new paradigm. They were the first to describe the phenomenon of tubal intraepithelial carcinomas, later designated serous tubal intraepithelial carcinomas. From that point, fallopian tubes were more carefully examined, which resulted in an increasing incidence of premalignant and early stages of high-grade serous cancer in prophylactically removed fallopian tubes.36-39 Many research projects have since been initiated and are still ongoing to find definitive evidence that the fallopian tube is the tissue of origin of pelvic high-grade serous cancer.40,41

IDENTIFICATION OF MUTATION CARRIERS

Since the isolation of BRCA1/2, the National Comprehensive Cancer Network, which is an alliance of leading U.S. cancer centers, and various family cancer clinics in Europe have developed guidelines for surveillance and prophylactic surgery.12,14,16 More recently, with the introduction of nextgeneration sequencing and the availability of gene panels, genetic testing for patients with ovarian cancer and family members of mutation carriers is within reach of many women. Year by year, the costs for genetic testing have dropped dramatically, and genetic counseling and testing was recently incorporated in practice guidelines in the United States and Europe.42,43 However, referral for genetic counseling and testing is not implemented among all patients with ovarian cancer in the United States and Europe, and asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 125

MOURITS AND DE BOCK

accessibility differs among patient groups. A recent study on adherence to National Comprehensive Cancer Network guidelines showed differences in genetic testing for patients with ovarian cancer in the United States: women were more frequently tested if they were younger at diagnosis, had a lower stage of ovarian cancer, were white, had private/ managed care insurance, and had a family history of cancer.44 Adherence and access to genetic counseling guidelines in different European countries has not yet been studied. Because genetic counseling and testing of all patients with ovarian cancer can reduce morbidity and mortality from ovarian (and breast) cancer among their relatives, and because prophylactic surgery is cost-effective, referral of all women with epithelial ovarian cancer should be encouraged, regardless of age, histologic type, and family history.42,43

COMPARISON AND CONTRASTS BETWEEN THE UNITED STATES AND EUROPEAN UNION

The knowledge and understanding of inherited ovarian cancer has expanded greatly since the discovery of BRCA1/2. Scientific expertise has developed on both sides of the Atlantic ocean, and researchers all over the world are sharing new findings and insights on the implications of the hereditary cancer syndrome and the ovarian cancer paradigm shift. Collaboration between geneticists and epidemiologists from Western countries resulted in large cohorts of

BRCA1/2 mutation carriers (e.g., CIMBA, IBCCS, kConFab, BCFR, GEO-HEBON, EMBRACE, GENEPSO), resulting in numerous important studies on risk estimates, genetic modifiers and correlation of cancer incidence, and lifestyle and reproductive factors. There are no scientific controversies on the pathogenesis and extraovarian origin of high-grade serous ovarian cancer, and the paradigm shift concerning the cell of origin arising from the fallopian tube is an excellent example of mutual inspiration and collaboration.34,45 Regarding clinical implications, no controversies exist regarding the cessation of ovarian cancer screening, which was adopted almost simultaneously on both sides of the ocean.42,43 If there are contrasts between countries and continents, they exist mostly in the field of access to genetic counseling and testing for all patients with ovarian cancer and in models of care for women at increased risk.46,47 One contrast across the ocean seems to be the extent of risk-reducing surgery, with or without hysterectomy. Although the overall risk of uterine cancer after RRSO is not increased, more clinicians in the United States than in Europe are inclined to offer a hysterectomy with RRSO.47 In conclusion, research and guidelines on hereditary ovarian cancer is a great example of mutual inspiration and joint efforts of researchers from all over the world, for the purpose of improving knowledge and health care for women with hereditary ovarian cancer.

References 1. Goff BA, Mandel LS, Drescher CW, et al. Development of an ovarian cancer symptom index: possibilities for earlier detection. Cancer. 2007;109:221-227. 2. Mourits MJ, de Bock GH. Symptoms are not early signs of ovarian cancer. BMJ. 2009;339:b3955. 3. Vergote I, Amant F, Kristensen G, et al. Primary surgery or neoadjuvant chemotherapy followed by interval debulking surgery in advanced ovarian cancer. Eur J Cancer. 2011;47 (Suppl 3):S88-S92. 4. Banerjee S, Kaye SB. New strategies in the treatment of ovarian cancer: current clinical perspectives and future potential. Clin Cancer Res. 2013;19:961-968. 5. McGee J, Bookman M, Harter P, et al; behalf of the participants of the 5th Ovarian Cancer Consensus on Conference. 5th Ovarian Cancer Consensus Conference: individualized therapy and patient factors. Ann Oncol. Epub 2017 Jan 24. 6. Jacobs IJ, Skates SJ, MacDonald N, et al. Screening for ovarian cancer: a pilot randomised controlled trial. Lancet. 1999;353:1207-1210. 7. Menon U, Gentry-Maharaj A, Hallett R, et al. Sensitivity and specificity of multimodal and ultrasound screening for ovarian cancer, and stage distribution of detected cancers: results of the prevalence screen of the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS). Lancet Oncol. 2009;10:327-340. 8. Jacobs IJ, Menon U, Ryan A, et al. Ovarian cancer screening and mortality in the UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS): a randomised controlled trial. Lancet. 2016;387: 945-956.

126 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

9. Buys SS, Partridge E, Black A, et al; PLCO Project Team. Effect of screening on ovarian cancer mortality: the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Randomized Controlled Trial. JAMA. 2011;305:2295-2303. 10. Miki Y, Swensen J, Shattuck-Eidens D, et al. A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science. 1994;266:66-71. 11. Wooster R, Bignell G, Lancaster J, et al. Identification of the breast cancer susceptibility gene BRCA2. Nature. 1995;378:789-792. 12. Vasen HF, Haites NE, Evans DG, et al; European Familial Breast Cancer Collaborative Group. Current policies for surveillance and management in women at risk of breast and ovarian cancer: a survey among 16 European family cancer clinics. Eur J Cancer. 1998;34:19221926. 13. American College of Obstetricians and Gynecologists; ACOG Committee on Practice Bulletins--Gynecology; ACOG Committee on Genetics; Society of Gynecologic Oncologists. ACOG Practice Bulletin No. 103: hereditary breast and ovarian cancer syndrome. Obstet Gynecol. 2009;113:957-966. 14. U.S. Preventive Services Task Force. Screening for ovarian cancer: recommendation statement. Ann Fam Med. 2004;2:260-262. 15. Walsh CS, Blum A, Walts A, et al. Lynch syndrome among gynecologic oncology patients meeting Bethesda guidelines for screening. Gynecol Oncol. 2010;116:516-521. 16. Lancaster JM, Powell CB, Kauff ND, et al; Society of Gynecologic Oncologists Education Committee. Society of Gynecologic Oncologists

U.S./E.U. COMPARISON AND CONTRASTS IN HIGH-RISK OVARIAN CANCER CARE

17. Boyd J. Molecular genetics of hereditary ovarian cancer. Oncology (Williston Park). 1998;12:399-406; discussion 409-410, 413.

Rebbeck TR, Kauff ND, Domchek SM. Meta-analysis of risk reduction 32. estimates associated with risk-reducing salpingo-oophorectomy in BRCA1 or BRCA2 mutation carriers. J Natl Cancer Inst. 2009;101:8087.

18. Koornstra JJ, Mourits MJ, Sijmons RH, et al. Management of extracolonic tumours in patients with Lynch syndrome. Lancet Oncol. 2009;10:400-408.

33. Dubeau L. The cell of origin of ovarian epithelial tumors and the ovarian surface epithelium dogma: does the emperor have no clothes? Gynecol Oncol. 1999;72:437-442.

19. Renkonen-Sinisalo L, Bützow R, Leminen A, et al. Surveillance for endometrial cancer in hereditary nonpolyposis colorectal cancer syndrome. Int J Cancer. 2007;120:821-824.

34. Piek JM, van Diest PJ, Zweemer RP, et al. Dysplastic changes in prophylactically removed Fallopian tubes of women predisposed to developing ovarian cancer. J Pathol. 2001;195:451-456.

20. Gerritzen LHM, Hoogerbrugge N, Oei ALM, et al. Improvement of endometrial biopsy over transvagin*l ultrasound alone for endometrial surveillance in women with Lynch syndrome. Fam Cancer. 2009;8:391397.

35. Kindelberger DW, Lee Y, Miron A, et al. Intraepithelial carcinoma of the fimbria and pelvic serous carcinoma: evidence for a causal relationship. Am J Surg Pathol. 2007;31:161-169.

Education Committee statement on risk assessment for inherited gynecologic cancer predispositions. Gynecol Oncol. 2007;107:159-162.

21. Helder-Woolderink JM, De Bock GH, Sijmons RH, et al. The additional value of endometrial sampling in the early detection of endometrial cancer in women with Lynch syndrome. Gynecol Oncol. 2013;131:304308. Lu KH, Daniels M. Endometrial and ovarian cancer in women with 22. Lynch syndrome: update in screening and prevention. Fam Cancer. 2013;12:273-277. Helder-Woolderink JM, Blok EA, Vasen HF, et al. Ovarian cancer in 23. Lynch syndrome; a systematic review. Eur J Cancer. 2016;55:65-73. Kauff ND, Satagopan JM, Robson ME, et al. Risk-reducing salpingo24. oophorectomy in women with a BRCA1 or BRCA2 mutation. N Engl J Med. 2002;346:1609-1615. Rebbeck TR, Lynch HT, Neuhausen SL, et al; Prevention and Observation 25. of Surgical End Points Study Group. Prophylactic oophorectomy in carriers of BRCA1 or BRCA2 mutations. N Engl J Med. 2002;346:1616-1622. Rutter JL, Wacholder S, Chetrit A, et al. Gynecologic surgeries and 26. risk of ovarian cancer in women with BRCA1 and BRCA2 Ashkenazi founder mutations: an Israeli population-based case-control study. J Natl Cancer Inst. 2003;95:1072-1078. Domchek SM, Friebel TM, Neuhausen SL, et al. Mortality after bilateral 27. salpingo-oophorectomy in BRCA1 and BRCA2 mutation carriers: a prospective cohort study. Lancet Oncol. 2006;7:223-229. Finch A, Beiner M, Lubinski J, et al; Hereditary Ovarian Cancer 28. Clinical Study Group. Salpingo-oophorectomy and the risk of ovarian, fallopian tube, and peritoneal cancers in women with a BRCA1 or BRCA2 Mutation. JAMA. 2006;296:185-192.

36. Leeper K, Garcia R, Swisher E, et al. Pathologic findings in prophylactic oophorectomy specimens in high-risk women. Gynecol Oncol. 2002;87:52-56. 37. Lamb JD, Garcia RL, Goff BA, et al. Predictors of occult neoplasia in women undergoing risk-reducing salpingo-oophorectomy. Am J Obstet Gynecol. 2006;194:1702-1709. Callahan MJ, Crum CP, Medeiros F, et al. Primary fallopian tube 38. malignancies in BRCA-positive women undergoing surgery for ovarian cancer risk reduction. J Clin Oncol. 2007;25:3985-3990. Reitsma W, Mourits MJ, de Bock GH, et al. Endometrium is not the 39. primary site of origin of pelvic high-grade serous carcinoma in BRCA1 or BRCA2 mutation carriers. Mod Pathol. 2013;26:572-578. Kim J, Coffey DM, Creighton CJ, et al. High-grade serous ovarian cancer 40. arises from fallopian tube in a mouse model. Proc Natl Acad Sci USA. 2012;109:3921-3926. Kroeger PT Jr, Drapkin R. Pathogenesis and heterogeneity of ovarian 41. cancer. Curr Opin Obstet Gynecol. 2017;29:26-34. National Institute for Health and Care Excellence. Familial breast 42. cancer: classification, care and managing breast cancer and related risks in people with a family history of breast cancer. www.nice.org. uk/guidance/cg164. Accessed February 1, 2017. National Comprehensive Cancer Network. Genetics screening. www. 43. nccn.org/professionals/physician_gls/pdf/genetics_screening.pdf. Accessed February 1, 2017. 44. Febbraro T, Robison K, Wilbur JS, et al. Adherence patterns to National Comprehensive Cancer Network (NCCN) guidelines for referral to cancer genetic professionals. Gynecol Oncol. 2015;138:109-114.

29. Hermsen BB, Olivier RI, Verheijen RH, et al. No efficacy of annual gynaecological screening in BRCA1/2 mutation carriers; an observational follow-up study. Br J Cancer. 2007;96:1335-1342.

45. Crum CP, Drapkin R, Miron A, et al. The distal fallopian tube: a new model for pelvic serous carcinogenesis. Curr Opin Obstet Gynecol. 2007;19:3-9.

30. van der Velde NM, Mourits MJ, Arts HJ, et al. Time to stop ovarian cancer screening in BRCA1/2 mutation carriers? Int J Cancer. 2009; 124:919-923.

46. Mourits MJ, de Bock GH. Managing hereditary ovarian cancer. Maturitas. 2009;64:172-176.

31. Chen S, Parmigiani G. Meta-analysis of BRCA1 and BRCA2 penetrance. J Clin Oncol. 2007;25:1329-1333.

47. Walker JL, Powell CB, Chen LM, et al. Society of Gynecologic Oncology recommendations for the prevention of ovarian cancer. Cancer. 2015;121:2108-2120.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 127

PROCHASKA, COUGHLIN, AND LYONS

Social Media and Mobile Technology for Cancer Prevention and Treatment Judith J. Prochaska, PhD, MPH, Steven S. Coughlin, PhD, and Elizabeth J. Lyons, PhD, MPH OVERVIEW Given the number of lives affected by cancer and the great potential for optimizing well-being via lifestyle changes, patients, providers, health care systems, advocacy groups, and entrepreneurs are looking to digital solutions to enhance patient care and broaden prevention efforts. Thousands of health-oriented mobile websites and apps have been developed, with a majority focused upon lifestyle behaviors (e.g., exercise, diet, smoking). In this review, we consider the use and potential of social media and mHealth technologies for cancer prevention, cancer treatment, and survivorship. We identify key principles in research and practice, summarize prior reviews, and highlight notable case studies and patient resources. Further, with the potential for scaled delivery and broad reach, we consider application of social media and mHealth technologies in low-resource settings. With clear advantages for reach, social media and mHealth technologies offer the ability to scale and engage entire populations at low cost, develop supportive social networks, connect patients and providers, encourage adherence with cancer care, and collect vast quantities of data for advancing cancer research. Development efforts have been rapid and numerous, yet evaluation of intervention effects on behavior change and health outcomes are sorely needed, and regulation around data security issues is notably lacking. Attention to broader audiences is also needed, with targeted development for culturally diverse groups and non-English speakers. Further investment in research to build the evidence base and identify best practices will help delineate and actualize the potential of social media and mHealth technologies for cancer prevention and treatment.

ew cancer cases in the United States number nearly 1.7 million annually. With earlier detection and improved treatments, the 5-year cancer survival rate increased from 49% during 1975 to 1977 to 69% during 2005 to 2011. Yet, cancer remains the second leading cause of death in the United States, with a substantial proportion of cancers preventable. Tobacco use alone is estimated to cause 29% of all cancer deaths,1 and more than one in five cancer diagnoses are related to lifestyle factors of obesity, physical inactivity, alcohol consumption, dietary factors, sexual health, and sun exposure.2 Vaccinations and regular cancer screening also are important for cancer prevention and early intervention. Among cancer survivors, quitting smoking and maintaining a healthy body weight through physical activity and healthy nutrition reduces the risk of disease recurrence or progression. Given the number of lives affected by cancer and the great potential for optimizing well-being via lifestyle changes, patients, providers, health care systems, advocacy groups, and entrepreneurs are looking to digital solutions to enhance patient care and broaden prevention efforts. In this review, we consider the use and potential of social media

and mHealth technologies for cancer prevention and cancer care. Social media are websites and applications (apps) that allow users to create, share, and participate via virtual communities and networks. Social media can provide fellowship with others, because of sharing common attitudes, interests, goals, or experiences, person-to-person, in real time, at low or no cost. mHealth, more broadly, refers to the delivery, facilitation, and communication of health-related information via mobile telecommunication and multimedia technologies (e.g., handheld devices, smartphones, tablets). The boom in mHealth has been made possible by the high penetration of internet access and increased use of smartphones. An estimated 89% of United States adults are now online, with smartphone ownership at 72%.3 As such, social media and mHealth technologies offer the ability to scale and engage entire populations, develop supportive social networks, connect patients and providers, encourage adherence with cancer care, and collect vast quantities of data for advancing cancer research. Our review attends to the use of social media and mHealth technologies in cancer prevention, cancer treatment, and survivorship. The field is broad and emerging rapidly with

From the Department of Medicine, Stanford Prevention Research Center, Stanford University, Stanford, CA; Department of Clinical and Digital Health Sciences, College of Allied Health Sciences, Augusta University, Augusta, GA; Department of Nutrition and Metabolism, The University of Texas Medical Branch, Galveston, TX. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Judith Prochaska, PhD, MPH, Stanford University, Medical School Office Building, X316, 1265 Welch Rd., Stanford, CA 94305; email: [emailprotected]. © 2017 American Society of Clinical Oncology

128 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

SOCIAL MEDIA AND MOBILE TECHNOLOGY FOR CANCER PREVENTION AND TREATMENT

the need for determination of evidence base and identification of best practices for patient care and data security. Given the breadth of our interest, a comprehensive review is not feasible. Instead, we identify key principles in research and practice, summarize prior reviews, and highlight notable case studies and patient resources. Further, with the potential for scaled delivery and broad reach, we consider application of social media and mHealth technologies in low-resource settings and best practices for dissemination.

SOCIAL MEDIA APPLICATIONS TO CANCER PREVENTION AND CANCER CARE

Social media come in several forms with differing audiences and emphases (Table 1). Among United States adults online, 79% use Facebook, 32% Instagram, 31% Pinterest, 29% LinkedIn, and 24% Twitter.4 Further, social media use in the United States has become routine, with daily use reported by 76% of Facebook users, 51% of Instagram users, and 42% of Twitter users. Most health-oriented research has been performed on general social media outlets such as Facebook and Twitter, with relatively little information available on smaller or specialized networks such as Snapchat. Yet, the emphases of specialized networks may make some platforms more optimally suited for specific intervention components. For example, video on YouTube or photos on Instagram may be effective for instruction and role modeling. Smaller and more private social networks may be preferred when discussing sensitive topics such as weight, tobacco, heavy alcohol use, or sexual activity. If using a larger and more general social medium, it may be prudent to consider private invitation-only groups, such as the example presented on use of Twitter to deliver private, peer-to-peer, quit-smoking groups (Sidebar 1). Closed quit-smoking groups targeting young adults also have been tested on Facebook5,6 and WhatsApp7 with encouraging short-term effects.

KEY POINTS • Innovations in mHealth and social media applications are occurring across the cancer spectrum, from primary prevention to screening, early diagnosis, treatment, survivorship, and end-of-life care. • Thousands of health-oriented mobile websites and apps have been developed, with most focused upon lifestyle behaviors (e.g., exercise, diet, stress, smoking). • Advantages of social media and mHealth technologies include low- or no-cost, high scalability, self-tracking and tailored feedback functionalities, use of images and video for enhanced health literacy, broad reach, and data sharing for large-scale analytics. • Although development efforts have been rapid and numerous, evaluation of intervention effects on behavior change and health outcomes are sorely needed, and regulation around data security issues is notably lacking. • Targeted development is also needed for culturally diverse groups and non-English speakers.

TABLE 1. Categories of Existing Social Media and Popular Examples Category

Examples

Major general-purpose social media outlets

Facebook; Twitter

Social media with a chronic illness focus

Smartpatients; CaringBridge; PatientsLikeMe

Photo-emphasizing social media

Instagram; Snapchat

Video-emphasizing social media

YouTube; Periscope

Blogs and message board–style networks

Tumblr; Reddit; Medium

Social video game or simulation networks

Xbox Live; Apple GameCenter; Second Life

Social media can provide varying degrees of anonymity, which may be attractive for stigmatized behaviors or medical conditions. When faced with the unknowns of a new diagnosis and a menu of treatment options, each with particular risks and benefits, social media can provide a unique connection with others who have direct personal experience. For example, with a focus on empowering patients, PatientsLikeMe is a free website, organized by medical conditions, where people can share health data, track their progress, connect with others, and contribute to big data analytics. PatientsLikeMe reports nearly 450,000 registered users and offers communities on nine cancer types. With a specific focus on cancer survivors, Springboard Beyond Cancer addresses more than 20 symptoms and health behaviors. The site promotes skills training and use of strategies for active self-management among cancer survivors with the aim of lessening the impact of disease and treatment side effects and improving quality of life.9 The mobile-optimized website draws existing information from Cancer.org, Cancer.gov, and literature related to survivorship and health behavior interventions. With social media sites that are largely uncurated or expert moderated, patients should be forewarned that negative or inaccurate health information might be posted. For example, user communities may encourage excessive dieting, vaccine avoidance, or use of nonevidence-based treatments (e.g., laser or herbs for quitting smoking). Harassment also can be a problem on more open networks such as Twitter and Reddit. Review of online content on breast cancer identified difficulty finding accurate information because of the lack of regulated sites.10 Although social media has become an important channel for disseminating findings from medical studies, the problem of fake news, including fake health news, is real, with growing recognition of the need for countermeasures.11,12

KEY PRINCIPLES OF SOCIAL MEDIA TO ENHANCE CANCER PREVENTION AND TREATMENT

At the foundation of social media applications for cancer prevention and control are techniques related to social support, health communication, self-regulation, and motivation enhancement. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 129

PROCHASKA, COUGHLIN, AND LYONS

SIDEBAR 1. Tweet2Quit Smoking-Cessation Intervention

Description

In private and by invitation-only 20-person groups, Tweet2Quit fostered peer-to-peer support and accountability for maintaining commitment to quit smoking. The Twitter-based intervention encouraged engagement via two scheduled automessages a day: (1) discussion questions based on tobacco treatment clinical practice guidelines and (2) individualized autofeedback based on past-day participation. A customized computer program automatically downloaded the group’s tweets daily, analyzed those who tweeted versus not, and sent prewritten and varied messages that praised tweeters for participating and encouraged nontweeters to do so. The groups lasted 100 days.

Study Design

In a two-group randomized controlled trial with 160 tobacco smokers, Tweet2Quit was combined with a web guide (smokefree.gov) and nicotine patch. The comparison group received the web guide and nicotine patches without the Twitter support group. Tobacco abstinence was reported at 60 days follow-up.

Examples of Group Tweeting

M1: I've smoked.:-( but I hide when I do bc I'm ashamed.:-p M2: Who you hiding from? YOU are the one that wants to quit…start over and try again! M3: Its ok to trip u just need to get back on track it sounds like u want to quit maybe u need more patches M1: I am going to get more and start fresh. Ty!!! M4: It's ok to stumble. just keep getting back up. you can do it! M1: When I saw myself failing I stopped tweeting so much. Didnt want to bring the rest of you down.:-/ M2: You need to keep tweeting! Maybe WE can bring you back UP! M3: Know we r all here to help anytime day or night u want to smoke txt us we r here for u

Study Findings

Tweet2Quit participants reported significantly greater sustained tobacco abstinence compared with control subjects: 40% vs. 20%; p = .012. Engagement was high, with participants averaging 57 tweets over an average of 47 days. More tweeting was associated with quitting (p = .003).8

Study Limitations

The sample was largely non-Hispanic white (88%), and outcomes were self-reported and short term (60 days). A larger randomized controlled trial is underway with an ethnically diverse sample and 6-month bioconfirmed outcomes of tobacco abstinence.

Future Applications

Social media may be leveraged to create support groups to attend to other cancer-related behaviors such as diet, physical inactivity, and excessive alcohol use.

Social Support, Influence, and Norms

Online social networking for fostering social support has a long research history, from online mailing lists and message boards to more modern iterations such as Instagram. Social support is important for behavior change broadly,13-15 and ample evidence indicates that existing social media groups can provide informational and emotional support to cancer survivors and caregivers.16-18 Online communities have been linked to increased empowerment19 and retention20; engagement with the communities has been linked to behavior change success for weight loss, smoking cessation, and other cancer-related behaviors,8,21,22 although some effects are small.23-25 Additionally, structured short message service and text messages to generate forum discussions, provide reminders, or offer tips and strategies have been effective build-ins.26 Ideally, social support is bidirectional, and 130 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

attention should be paid to facilitate receipt as well as provision of social support. A recent intervention study found that expressing social support was associated with perceived bonding within the social media group and positive coping strategies, whereas receipt alone of supportive messages was not.27 With a focus on influencing perceived social norms, social media interventions have demonstrated preliminary efficacy for reducing problematic alcohol consumption.28,29 Yet, of concern, the literature also finds social networking associated with negative outcomes related to social comparison, such as poor body image and depression.13,30 When designing interventions for cancer prevention and survivorship, it is important to consider potential unintended negative consequences and attempt to avoid or ameliorate them. For example, implementing weight-related programming

SOCIAL MEDIA AND MOBILE TECHNOLOGY FOR CANCER PREVENTION AND TREATMENT

in photo-sharing media may require private groups, stricter rules, or additional intervention to reduce negative social comparisons to participants with lower weights, “thinspiration” accounts, or slim celebrities.

Health Communication

Communication campaigns using social media such as Twitter and Facebook are increasingly popular. Both large-scale national and international campaigns as well as smaller campaigns by local organizations and clinics have demonstrated engagement with their target audiences using social media.14,15 Role model narratives are effective methods of persuasion with demonstrated positive impacts on cancer prevention behaviors19-21 and can easily be delivered using video and photo tools in most popular social media systems. Evaluation of a breast cancer awareness campaign launched on Facebook by the Centers for Disease Control and Prevention found greatest engagement for posts with photos rather than status/links or videos; posts released in the early morning and afternoon (2:00 PM to 6:00 PM) versus other time periods; and posts shared earlier (2014) than later (2016) in the campaign.31 Social media also can provide opportunities for truly interactive intervention methods. For example, a study found that participation in cocreating antismoking campaign content on Facebook produced greater information searching and intention to quit than simply viewing the content online.22

Self-Regulation

Self-regulation techniques, such as goal setting and feedback, are the foundation of many interventions that seek to change health behaviors, both for cancer prevention and adherence with cancer treatment regimens. Social networks are incorporated into some health-related apps and websites to promote self-regulatory skill-building,26 and many general social networks include large subcommunities related to these topics. Some forms of these media may be particularly well suited to promoting self-regulation. For example, video-sharing services can provide highly detailed instruction and rich feedback from peers as well as experts.32

Motivation Enhancement

Social media shows promise for delivery of general and social rewards. In fact, several scholars have suggested that virtual rewards such as badges may be more effective when implemented within some form of social network, to emphasize personal status, group affiliation, and reputation.33,34 Recommendations for gamification emphasize the importance of social engagement, personal reflection, and nurturing game elements for producing long-term motivation,35,36 all of which can be facilitated via social media.

Engagement

Inadequate engagement can be a major limitation to cancer-related social media interventions.37 Research consistently has found that posting photos results in a greater amount of engagement than other post types.31,38,39 A study

of scientific communication with the public across social media platforms by the European Organization for Nuclear Research found that “wow” photographs (i.e., awe-inspiring photographs) produced the most engagement, especially when posted on the photo-emphasizing platform Instagram.40 Another recurring finding is that users may prefer different social media platforms, making formative research and/or use of multiple channels an important consideration.7,41

MHEALTH APPS AND WEARABLE DEVICES FOR CANCER PREVENTION AND CANCER CARE

A full range of mHealth apps are available for download from digital marketplaces (e.g., iTunes, Google Play) for use on smartphones, tablets, and other handheld devices. Thousands of health-oriented apps have been developed, with most focused upon lifestyle behaviors (e.g., exercise, diet, stress, smoking).42 Yet, a mere 36 comprise half of the downloads. The focused use is attributed to the very limited functionality of most mHealth apps: just 10% can connect to a device or sensor, only 2% sync with providers' systems, and few incorporate social networking functions.43 Table 2 presents categories and examples of mHealth apps relevant to cancer prevention and cancer care. Several reviews have been published on mHealth apps. With attention to the prevention, detection, and management of cancer, one review identified 295 mHealth apps available in 2012.44 Most common were apps on breast cancer (47%) or cancer in general (29%), apps aimed at raising cancer awareness (32%), providing cancer education (26%), supporting fundraising (13%), assisting in early detection (12%), or promoting a charitable organization (10%). Far fewer were apps designed to support disease management (4%), cancer prevention (2%), or social support (1%). The authors conducted a companion systematic review of the

TABLE 2. Categories of mHealth Apps With a Cancer Focus and Examples Category

Examples

General health apps

Find a Health Center; Medscape

Health risk assessment apps

BRisk; BCSC; Rotterdam Prostate Cancer Risk Calculator

Quit-smoking apps for patients/providers

ASPIRE; QuitStart; QuitGuide; QuitMedKit

Diet and fitness apps

SuperTracker; SWORKIT; Endomondo

Self-regulation apps with social networking

Fitbit; Lose It!; My Fitness Pal; QuitNet

Symptom navigator apps

My PearlPoint Cancer Side Effects Helper

Patient portals

OhMD

Health condition trackers

My Breast Cancer Journey

Screening exam apps

ePrognosis Cancer Screening

Environmental exposure apps

Detox Me, Healthy Living Mobile App

Cancer treatment and survivorship apps

Cancer.Net (ASCO), iCancerHealth; National Comprehensive Cancer Network

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 131

PROCHASKA, COUGHLIN, AND LYONS

health literature (1990–2012) and could not identify a single empirical evaluation of a cancer-focused mHealth app. With a focus on breast health, a search of breast symptoms and diseases in major app stores identified 185 mHealth apps, of which 139 (75%) focused on breast cancer. Most of the apps (51%) were educational, 16% were self-assessment tools, only 14% were deemed evidence-based, and a mere 13% involved medical professionals in their development. Potential patient safety concerns were identified in 29 (16%) of the apps. Needed are mHealth cancer prevention apps informed by behavior change theory that attend to multiple risk factors and are appropriate for patients with low health and e-health literacy. As an illustrative example, the purposeful design of a breast cancer prevention app is summarized in Sidebar 2.45 A recent study conducted with 54 women at elevated risk for breast cancer evaluated, in a randomized controlled design, the combination of a wearable technology to monitor physical activity (Fitbit One) with a smartphone app to monitor diet (My Fitness Pal), and coaching calls from trained counselors. The goal was weight loss. Women randomized to the wearable plus mHealth app plus coaching achieved significantly greater weight loss (4.4 vs. 0.08 kg; p = .004) than women randomized to usual care.46

With a focus on managing symptoms following breast cancer treatment, The-Optimal-Lymph-Flow health IT system is an mHealth site with an electronic assessment and education on self-care strategies for lymphedema symptom management.47 Evaluated over 12 weeks with 355 survivors of breast cancer, 97% reported high satisfaction with ease of use, and participants reported less pain, less soreness, less aching, less tenderness, fewer lymphedema symptoms, and improved symptom distress (all p values < .05). In the area of tobacco control for cancer prevention, a number of apps have been developed with good interest. A 2014 search identified 546 smoking-cessation apps in the Apple Store and Google Play, which were downloaded an estimated 3.2 million times in the United States and 20 million times worldwide.48 A review specifically of Android apps for quitting smoking identified 225 apps available between 2013 and 2014.49 Most provided simplistic tools (e.g., calculators, trackers). Use of tailoring was limited, though positively related to app popularity and user ratings of quality. The numbers are anticipated to rise as interest in mHealth apps and wearable health devices continues to grow. The past 2 years (2014–2016) saw a doubling in consumer use. One in three adults now report using an mHealth app and

SIDEBAR 2. Development of the Physical Activity and Your Nutrition for Cancer (PYNC) Prevention App

Objective

To promote healthy diet, nutrition, physical activity, and weight loss among women at risk of breast cancer who have varying levels of health literacy and e-health literacy.

Methods

An eight-step process is being followed to ensure that the intervention materials are appropriate for the intended audience. Development to date has included literature reviews, conceptual design, drafting informational and motivational content, acceptability review with community members, and scientific review by the research team. Remaining steps include prototyping materials, assessment of health literacy level, usability testing with community members, and final modifications.

Framework

The app uses Leventhal’s Common Sense Model of Health Behavior, which describes how thoughts and beliefs about health and disease risk influence behavior.

Components

The app draws upon commercially available technology for monitoring physical activity, caloric intake, diet, and nutrition (Fitbit, LoseIt!, and USDA’s ChooseMyPlate) while providing evidence-based information about breast cancer and ways that women can reduce their risk of the disease.

Prototype Feedback

Recommendations included use of “more relaxed language” and presentation of information “in a more visual way.” Other suggestions included ideas for easy-to-prepare healthy foods, instruction on how to read food labels, and information on environmental contaminants and chemicals that may influence cancer risk, such as cleaning and beauty products.

Future Directions

Next steps are testing the efficacy of the mHealth intervention in increasing physical activity, improving diet and nutrition, and managing weight through a randomized controlled trial. 132 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

SOCIAL MEDIA AND MOBILE TECHNOLOGY FOR CANCER PREVENTION AND TREATMENT

21% a wearable device, with use greatest among adults age 18–34. The most popular mHealth app segments are fitness (59%) and diet/nutrition (52%), followed by symptom navigators (36%), patient portals (28%), health condition trackers (25%), medication trackers (12%), and disease-management apps (10%). Most consumers (77%) and doctors (85%) view health wearables as helping to engage patients in their health, and over a third of physicians have recommended mHealth apps to their patients.43,50 In the area of cancer care, novel wearable technology concepts include balance sensors for patients with chemotherapyinduced peripheral neuropathy51 and Google glasses with a fluorescence imaging system for complete resection of tumors in surgical oncology.52 The demonstrated evidence, however, for mHealth apps in promoting and sustaining behavior change is still limited. A 2016 review of 38 articles of mobile phone applications for behavior change, four specific to cancer, was unable to identify a single best practice approach to evaluate mHealth apps, which the authors noted was further complicated by a general lack of regulation.53 Similarly, a systematic review of randomized controlled trials testing the efficacy of mHealth apps for cancer prevention identified only four trials for smoking cessation and two for sun safety and concluded a meta-analysis was premature in this area.54 Health apps also have been developed to help consumers reduce exposures to known or suspected carcinogens and other toxicants in work or home environments. App functions include education, scanning of product bar codes at point-of-purchase, and self-tracking. With the same limitations acknowledged above, to date, the environmental health apps have not been tested for acceptability, feasibility, or effectiveness in randomized controlled trials.55

PRIVACY AND CONFIDENTIALITY CONCERNS WITH SOCIAL MEDIA AND MHEALTH TECHNOLOGIES

Although technologies such as smartphone mHealth apps and other remote monitoring devices have the potential to transform oncology care,56 they also raise new considerations with regard to patient privacy and confidentiality. Apps may support a patient’s self-report of symptoms or passively record location and other information using global positioning systems, accelerometers, and physiologic sensors. The ability to collect large amounts of personal data over long periods of time provides clinicians and researchers with insights into disease treatment and progression and also raises unique ethical issues.57,58 We consider in this study the privacy and confidentiality concerns of social media and consumer-oriented mHealth technologies; patient safety, data security, and confidentiality of mHealth technologies; and regulatory developments. With direct application to practice, we also consider clinician-patient discussion points regarding the risks and benefits of using mHealth technologies. Patients who purchase consumer-facing smartphone apps and other mHealth technologies (e.g., apps for weight loss

and wearable devices for monitoring steps, heart rate, and sleep) may not be well informed of privacy practices. Systematic reviews of health and wellness apps available from generic app stores have identified deficiencies in the extent to which data uses are documented and appropriate security measures are implemented.59,60 Among the most commonly used apps available for iOS and Android, only 183 of 600 (31%) had privacy policies, and 66% of the privacy policies did not specifically address the app.59 Consumers may be unaware that smartphone apps may share sensitive information such as sensor data on location with third parties such as advertisers. Many apps sold direct to consumers send unencrypted data to third party sites for advertising or analytics.61 The main security risk is unauthorized access to data during collection, transmission, or storage. Unencrypted data (e.g., global positioning system coordinates, telephone numbers, email addresses, health information) transmitted over the internet can be intercepted. Efforts have been made to create secure devices and apps, but many contain serious flaws.62 Security threats also exist for provider-facing mHealth technologies. Ethical and regulatory issues related to mHealth technologies used by providers for patient care relate to patient safety and the security and confidentiality of patient data transmitted and stored in mobile medical apps.63 Hackers and malware pose an increasing threat to the security of mobile medical apps.

REGULATION AND CERTIFICATION OF MEDICAL APPS AND MHEALTH TECHNOLOGIES

In some countries, government agencies have begun to regulate or curate medical apps.63-65 In 2013, the U.S. Food and Drug Administration (FDA) released guidance for mobile medical apps that draws a distinction between unregulated apps and mobile medical apps that are subject to overt FDA regulation.66 Apps that convert a mobile platform such as a smartphone or tablet computers into a medical device are regulated by the FDA.63 The FDA regulates mobile apps that pose a greater risk to patients if they do not function as intended (e.g., apps that perform clinical tests such as blood or urine analysis, apps that display diagnostic images from x-rays and MRI, and apps that remotely display data from bedside monitors). The FDA focuses on technical issues related to patient safety and the security and integrity of information but not patient privacy.62 Consumer-oriented apps for general health education are mostly unregulated.66 In Europe, an Irish app (ONCOassist) for the iPhone and iPad that contains prognostic tools and calculators for oncologists at the point-of-care, has received Conformite Europeenne certification indicating that it complies with relevant European Union legislation.67 The European Medical Device Directive MDD 93/42/EEC mentions software in its definition of a medical device. In the United States, the Health Insurance Portability and Accountability Act (HIPAA) contains the primary set of regulations that guide the privacy and security of health asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 133

PROCHASKA, COUGHLIN, AND LYONS

information.68 HIPAA regulations require covered entities and their business associates (e.g., physicians, hospitals, health plans) to protect health information that identifies an individual and that relates to an individual’s physical or mental health or health care services provided to the individual.69 Developers of mobile apps and sensors must consider whether the software and information technology will be used by a covered entity and whether it will include any protected health information. For example, an app that assists a health care provider with following up patients must be designed to allow the provider to comply with HIPAA.69 HIPAA requires that identifiable health information be encrypted so that only those authorized to read it can do so.68 In the United Kingdom, the National Health Service established a Health Apps Library that endorses apps considered to be relevant to people in Great Britain and that provide trustworthy information, comply with data storage regulations, and do not pose potential risks if used improperly.70 A recent assessment of 79 apps certified as clinically safe and trustworthy by the Health Apps Library found systematic gaps in compliance with data protection principles.70 None of the 79 apps encrypted personal information stored locally, 66% (23 of 35) of apps sending identifying information over the internet did not use encryption, and 20% (7 of 35) did not have a privacy policy.70 The authors noted that app users cannot see into the inner workings of apps or the services to which they connect; hence, they must trust developers to comply with privacy regulations and security best practices.70 Medical information stored on apps or transmitted via the internet or Bluetooth should be secured using encryption.71

WHAT SHOULD CLINICIANS TELL THEIR PATIENTS ABOUT PRIVACY AND CONFIDENTIALITY?

Clinicians can only provide limited guarantees about privacy protection. Data collected on mobile phones can be subpoenaed as part of legal proceeds in civil or criminal cases.57 Because of the potential for hacking of personal data from mHealth apps, the security of data collected via mobile phones cannot be guaranteed.57 As stated, many mHealth apps do not use encryption when transferring data.72 A further issue is that telecommunication companies record metadata and data transferred over their networks and sell them to third parties. Patients’ trust in their clinicians contributes to treatment adherence and continuity of care and, in turn, plays an important role in the adoption of mHealth technologies.68 Clinicians should discuss the risks and benefits of using mHealth technologies as part of patient-centered care.68 Providers should be aware of their institutions’ privacy and security policies as part of their ethical obligation to ensure patient-physician confidentiality. Before using mHealth technologies, clinicians should obtain informed consent from patients so that they understand the benefits, risks, and potential harms. The rapid pace of development, early efforts at regulation, and the complex nature of the 134 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

risks posed by using mHealth technologies raise challenges in communicating risks to patients.57 Discussion of the potential risks (e.g., data harvesting, data breaches), benefits (e.g., self-awareness/self-management, attention to adherence and lifestyle behaviors, patient-provider communications), and unknowns (e.g., optimal balance of tech to touch) is warranted.

USING SOCIAL MEDIA AND MHEALTH APPS IN LOW-RESOURCE SETTINGS

Globally, by 2030, the burden of cancer is predicted to worsen significantly in low-income (82% increase in incidence) and lower-middle income (70% increase) countries.73 The rise in mobile phone access worldwide74 affords opportunity for delivering social media and mHealth technologies to improve cancer awareness, encourage timely screening, and secure follow-up care.75 In the United States, mobile technologies have bridged the digital divide.76 By ethnicity, African Americans and English-speaking Hispanics are just as likely as whites to own a mobile phone and use it for a wider range of activities.76 In a survey of female public housing residents in Boston, nearly all reported mobile phone use for calls (97%) and texts (84%); recent use (past day) of the internet was 65%, social media 59%, and email 28%; 70% had a Facebook account and 12% a Twitter account.77 Social media users were more likely to be Hispanic and Spanish speaking. Broad reach, low or no cost, and high scalability make social media and mHealth apps particularly well suited for application in resource-poor settings. Social media can be used across platforms (i.e., Android, iOS, and personal computers) and can connect individuals over long distances, which can be valuable to individuals in rural areas with rare cancers who do not have peers or role models readily available otherwise. Even for those with more common cancers, online social media allows social interaction without the burden of travel to clinics or support group locations. Research indicates barriers to engaging in care among some low-income groups, such as residents in public housing.78 Social media and mHealth technologies may aid outreach efforts with appropriate messaging and support for cancer prevention efforts. Needed and worthy of evaluation is the extent to which people with lower levels of health literacy or numeracy find cancer-related use of social media and mHealth apps to be helpful or practical and whether apps are effective in helping culturally diverse groups to reduce their risk of cancer. Emphasized is the thoughtful development and use of mHealth applications to solve health disparities, not widen them. To inform development of a social media smoking-cessation intervention, focus groups were conducted with 33 Hispanic, Spanish-speaking, current and former smokers in the San Francisco Bay area.79 Most participants owned a smartphone (84%), and the majority of cell phone owners reported daily texting (81%) and Facebook use (69%). The participants valued the communal aspect of social media

SOCIAL MEDIA AND MOBILE TECHNOLOGY FOR CANCER PREVENTION AND TREATMENT

and suggested strategically tailoring groups based on key features (e.g., age, gender, language preference). Participants reported preferring visual, educational, and motivational messages connected with existing services. Development of social media and mHealth programs for diverse settings and communities can be achieved with limited investment by drawing upon existing resources. Content analyses of various social media groups (e.g., Facebook groups, individuals using the same Twitter hashtag) have identified several types of social support provided,16,80 and numerous interventions have shown that behavior change techniques can be effectively delivered via existing social media tools.7,23,81 Hence, expending resources to create new cancer-focused mobile apps or websites may not be necessary to deliver effective prevention and treatment interventions. Even if the long-term goal is to create an entirely new system, existing tools can provide a method for prototype testing. For example, combinations of personal emails and group sessions via social media can be used to test out the potential effects of face-to-face or app-based delivery of these techniques. An example of effective low-cost leveraging of mobile technologies comes from work in Ambanja, Madagascar, where smartphones were used to take and transmit high-definition images for the detection of cervical intraepithelial neoplasia of grade 2 or worse as an adjunct to standard on-site examination.82

CONCLUSION

Exciting innovations in mHealth and social media applications are occurring across the cancer spectrum, from primary prevention to screening, early diagnosis, treatment, survivorship, and end-of-life care. These new platforms and technologies avail social engagement and support as well as personalized data points for patients and providers to inform care decisions. Cancer-prevention applications include

attention to tobacco use, diet, physical activity, and sleep; there are screening apps and cancer risk calculators to raise awareness; and links to patient communities or providers for symptom management. Advantages of social media and mHealth technologies include low or no cost, high scalability, self-tracking and tailored feedback functionalities, use of images and video for enhanced health literacy, broad reach, and data sharing for large-scale analytics. Although development efforts have been rapid and numerous, frameworks and investigations of efficacy for achieving and sustaining behavioral change and positive health outcomes are sorely needed, and regulation concerning data security issues is notably lacking. Targeted development is also needed for culturally diverse groups and for non-English speakers. Further investment in research to build the evidence base and identify best practices will help delineate and actualize the potential of social media and mHealth technologies for cancer prevention and treatment.

ACKNOWLEDGMENT

J. J. Prochaska’s research is funded by the National Cancer Institute (R01-CA-204356), the National Heart, Lung and Blood Institute (R01-HL-117736), the State of California’s Tobacco-Related Disease Research Program (24RT-0035 and 25IR-0032), and an intramural grant from the Stanford Cancer Institute. J. J. Prochaska is on the advisory board for Carrot Sense, a digital health company. S. S. Coughlin’s research is funded by the Office of the Assistant Secretary of Defense for Health Affairs under award no. W81XWH-16-1-0774 and by intramural support provided by the Augusta University College of Allied Health Sciences. E. J. Lyons is supported by a Mentored Research Scholar Grant in Applied and Clinical Research (MRSG-14-165-01-CPPB) from the American Cancer Society and the Claude D. Pepper Older Americans Independence Center (P30-AG-024832).

References 1. World Cancer Research Fund International. Cancer preventability estimates for diet, nutrition, body fatness, and physical activity. http:// www.wcrf.org/int/cancer-facts-figures/preventability-estimates/ cancer-preventability-estimates-diet-nutrition. Accessed September 29, 2015. 2. Colditz GA, Wei EK. Preventability of cancer: the relative contributions of biologic and social and physical environmental determinants of cancer mortality. Annu Rev Public Health. 2012;33: 137-156. 3. Poushter J. “Smartphone ownership and internet usage continues to climb in emerging economies.” Pew Research Center’s Global Attitudes Project, February 22, 2016. http://www.pewglobal. org/2016/02/22/smartphone-ownership-and-internet-usagecontinues-to-climb-in-emerging-economies/. Accessed January 5, 2017. 4. Greenwood S, Perrin A, Duggan M. “Social media update 2016.” Pew Research Center: Internet, Science & Tech, November 11, 2016. http:// www.pewinternet.org/2016/11/11/social-media-update-2016/. Accessed January 5, 2017.

5. Ramo DE, Thrul J, Chavez K, et al. Feasibility and quit rates of the tobacco status project: a Facebook smoking cessation intervention for young adults. J Med Internet Res. 2015;17:e291. 6. Baskerville NB, Azagba S, Norman C, et al. Effect of a digital social media campaign on young adult smoking cessation. Nicotine Tob Res. 2016;18:351-360. 7. Cheung YTD, Chan CHH, Lai C-KJ, et al. Using WhatsApp and Facebook online social groups for smoking relapse prevention for recent quitters: a pilot pragmatic cluster randomized controlled trial. J Med Internet Res. 2015;17:e238. 8. Pechmann C, Delucchi K, Lakon CM, et al. Randomised controlled trial evaluation of Tweet2Quit: a social network quit-smoking intervention. Tob Control. Epub 2016 Feb 29. 9. Springboard Beyond Cancer. https://smokefree.gov/springboard/. Accessed January 6, 2017. 10. Quinn EM, Corrigan MA, McHugh SM, et al. Breast cancer information on the internet: analysis of accessibility and accuracy. Breast. 2012;21:514-517.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 135

PROCHASKA, COUGHLIN, AND LYONS

11. Hitlin P. “Health issues topped the list of scientific studies reaching wide audiences in 2016.” Pew Research Center. http://www. pewresearch.org/fact-tank/2016/12/28/health-issues-topped-thelist-of-scientific-studies-reaching-wide-audiences-in-2016/. Accessed January 6, 2017. 12. Romano A. “The year social media changed everything.” Vox, December 31, 2016. http://www.vox.com/2016/12/31/13869676/ social-media-influence-alt-right. Accessed January 5, 2017. 13. Lup K, Trub L, Rosenthal L. Instagram #instasad?: exploring associations among instagram use, depressive symptoms, negative social comparison, and strangers followed. Cyberpsychol Behav Soc Netw. 2015;18:247-252. 14. Park H, Reber BH, Chon M-G. Tweeting as health communication: health organizations’ use of Twitter for health promotion and public engagement. J Health Commun. 2016;21:188-198. 15. Borgmann H, Loeb S, Salem J, et al. Activity, content, contributors, and influencers of the twitter discussion on urologic oncology. Urol Oncol. 2016;34:377-383. 16. Gage-Bouchard EA, LaValley S, Mollica M, et al. Communication and exchange of specialized health-related support among people with experiential similarity on Facebook. Health Commun. 2016;2:1-8.

28. Boyle SC, Earle AM, LaBrie JW, et al. PNF 2.0? Initial evidence that gamification can increase the efficacy of brief, web-based personalized normative feedback alcohol interventions. Addict Behav. 2017;67: 8-17. 29. Flaudias V, de Chazeron I, Zerhouni O, et al. Preventing alcohol abuse through social networking sites: a first assessment of a two-year ecological approach. J Med Internet Res. 2015;17:e278. 30. Holland G, Tiggemann M. A systematic review of the impact of the use of social networking sites on body image and disordered eating outcomes. Body Image. 2016;17:100-110. 31. Theiss SK, Burke RM, Cory JL, et al. Getting beyond impressions: an evaluation of engagement with breast cancer-related Facebook content. mHealth. 2016;2:41. 32. Basch CH, Menafro A, Mongiovi J, et al. A content analysis of YouTube videos related to prostate cancer. Am J Mens Health. Epub 2016 Sep 29. 33. Antin J, Churchill EF. Badges in social media: a social psychological perspective. Paper presented at: ACM CHI Conference on Human Factors in Computing Systems; May 2011; Vancouver, BC, Canada.

17. Bender JL, Jimenez-Marroquin MC, Ferris LE, et al. Online communities for breast cancer survivors: a review and analysis of their characteristics and levels of use. Support Care Cancer. 2013;21:1253-1263.

34. Laschke M, Hassenzahl M. Being a “mayor” or a “patron”? The difference between owning badges and telling stories. Paper presented at: ACM CHI Conference on Human Factors in Computing Systems; May 2011; Vancouver, BC, Canada.

18. Gage-Bouchard EA, LaValley S, Mollica M, et al. Cancer communication on social media: examining how cancer caregivers use Facebook for cancer-related communication. Cancer Nurs. Epub 2016 Jul 20.

35. Nicholson S. A user-centered theoretical framework for meaningful gamification. Paper presented at: Games+Learning+Society 8.0; June 2012; Madison, WI.

19. Beach WA, Dozier DM, Buller MK, et al. The Conversations About Cancer (CAC) Project-Phase II: national findings from viewing When Cancer Calls…and implications for Entertainment-Education (E-E). Patient Educ Couns. 2016;99:393-399.

36. Lucero A, Karapanos E, Arrasvuori J, et al. Playful or gameful?: Creating delightful user experiences. Interaction. 2014;21:34-39.

20. Green MC. Narratives and cancer communication. J Commun. 2006;56:S163-S183. 21. Murphy ST, Frank LB, Chatterjee JS, et al. Comparing the relative efficacy of narrative vs nonnarrative health messages in reducing health disparities using a randomized trial. Am J Public Health. 2015;105:2117-2123. 22. Namkoong K, Nah S, Record RA, et al. Communication, reasoning, and planned behaviors: unveiling the effect of interactive communication in an anti-smoking social media campaign. Health Commun. 2017;32:41-50.

37. Cavallo DN, Chou W-YS, McQueen A, et al. Cancer prevention and control interventions using social media: user-generated approaches. Cancer Epidemiol Biomarkers Prev. 2014;23:1953-1956. 38. Rus HM, Cameron LD. Health communication in social media: message features predicting user engagement on diabetes-related Facebook pages. Ann Behav Med. 2016;50:678-689. 39. Strekalova YA, Krieger JL. A picture really is worth a thousand words: public engagement with the National Cancer Institute on social media. J Cancer Educ. 2017;32;155-157. 40. Kahle K, Sharon AJ, Baram-Tsabari A. Footprints of fascination: digital traces of public engagement with particle physics on CERN’s social media platforms. PLoS One. 2016;11:e0156409.

23. Maher CA, Lewis LK, Ferrar K, et al. Are health behavior change interventions that use online social networks effective? A systematic review. J Med Internet Res. 2014;16:e40.

41. Owen JE, Curran M, Bantum EO, et al. Characterizing social networks and communication channels in a Web-based peer support intervention. Cyberpsychol Behav Soc Netw. 2016;19:388-396.

24. Ashrafian H, Toma T, Harling L, et al. Social networking strategies that aim to reduce obesity have achieved significant although modest results. Health Aff (Millwood). 2014;33:1641-1647.

42. Misra S. “New reports find more than 165,000 mobile health apps now available.” iMedicalApps, September 17, 2015. http://www. imedicalapps.com/2015/09/ims-health-apps-report/. Accessed January 5, 2017.

25. Williams G, Hamm MP, Shulhan J, et al. Social media interventions for diet and exercise behaviours: a systematic review and meta-analysis of randomised controlled trials. BMJ Open. 2014;4:e003926. 26. Lyons EJ, Lewis ZH, Mayrsohn BG, et al. Behavior change techniques implemented in electronic lifestyle activity monitors: a systematic content analysis. J Med Internet Res. 2014;16:e192. 27. Namkoong K, McLaughlin B, Yoo W, et al. The effects of expression: how providing emotional support online improves cancer patients’ coping strategies. J Natl Cancer Inst Monogr. 2013;2013:169-174.

136 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

43. McCarthy J. “How many health apps actually matter?” Healthcare IT News. September 23, 2015. http://www.healthcareitnews.com/ news/how-many-health-apps-actually-matter. Accessed January 5, 2017. 44. Bender JL, Yue RYK, To MJ, et al. A lot of action, but not in the right direction: systematic review and content analysis of smartphone applications for the prevention, detection, and management of cancer. J Med Internet Res. 2013;15:e287.

SOCIAL MEDIA AND MOBILE TECHNOLOGY FOR CANCER PREVENTION AND TREATMENT

45. Coughlin S, Besenyi G, Bowen D, et al. Development of the PYNC smartphone app for preventing breast cancer in women. mHealth. 2017;3:5. 46. Hartman SJ, Nelson SH, Cadmus-Bertram LA, et al. Technology- and phone-based weight loss intervention: pilot RCT in women at elevated breast cancer risk. Am J Prev Med. 2016;51:714-721. 47. Fu MR, Axelrod D, Guth AA, et al. mHealth self-care interventions: manag ing symptoms following breast cancer treatment. mHealth. 2016;2:28. 48. Bricker JB, Mull KE, Kientz JA, et al. Randomized, controlled pilot trial of a smartphone app for smoking cessation using acceptance and commitment therapy. Drug Alcohol Depend. 2014;143:87-94. 49. Hoeppner BB, Hoeppner SS, Seaboyer L, et al. How smart are smartphone apps for smoking cessation? A content analysis. Nicotine Tob Res. 2016;18:1025-1031. 50. Safavi K, Webb K, MacCracken L, et al. Patients want a heavy dose of digital. https://acnprod.accenture.com/t20160629T045303__w__/ us-en/_acnmedia/PDF-6/Accenture-Patients-Want-A-Heavy-Dose-ofDigital-Infographic.pdf#zoom=50. Accessed January 17, 2017. 51. Schwenk M, Grewal GS, Holloway D, et al. Interactive sensor-based balance training in older cancer patients with chemotherapy-induced peripheral neuropathy: a randomized controlled trial. Gerontology. 2016;62:553-563. 52. Shao P, Ding H, Wang J, et al. Designing a wearable navigation system for image-guided cancer resection surgery. Ann Biomed Eng. 2014;42:2228-2237.

64. Yetisen AK, Martinez-Hurtado JL, da Cruz Vasconcellos F, et al. The regulation of mobile medical applications. Lab Chip. 2014;14:833-840. 65. Cortez NG, Cohen IG, Kesselheim AS. FDA regulation of mobile health technologies. N Engl J Med. 2014;371:372-379. 66. U.S. Department of Health and Human Services, Food and Drug Administration. “Mobile medical applications: guidance for industry and Food and Drug Administration staff.” February 9, 2015. http://www. fda.gov/downloads/MedicalDevices/DeviceRegulationandGuidance/ GuidanceDocuments/UCM263366.pdf. Accessed May 12, 2016. 67. Boulos MNK, Brewer AC, Karimkhani C, et al. Mobile medical and health apps: state of the art, concerns, regulatory control and certification. Online J Public Health Inform. 2014;5:229. 68. Hale TM, Kvedar JC. Privacy and security concerns in telehealth. Virtual Mentor. 2014;16:981-985. 69. Greene AH. “When HIPAA applies to mobile applications.” http:// www.mobihealthnews.com/11261/when-hipaa-applies-to-mobileapplications. Accessed January 5, 2017. 70. Huckvale K, Prieto JT, Tilney M, et al. Unaddressed privacy risks in accredited health and wellness apps: a cross-sectional systematic assessment. BMC Med. 2015;13:214. 71. Center for Democracy Technology. Best practices for mobile app developers. https://www.cdt.org/files/pdfs/Best-Practices-MobileApp-Developers.pdf. Accessed May 10, 2016. 72. He D, Naveed M, Gunter CA, et al. Security concerns in Android mHealth apps. AMIA Annu Symp Proc AMIA Symp. 2014;2014:645–54.

53. McKay FH, Cheng C, Wright A, et al. Evaluating mobile phone applications for health behaviour change: a systematic review. J Telemed Telecare. Epub 2016 Oct 18.

73. World Health Organization. Global status report on noncommunicable diseases 2010. World Health Organization; Report No.: ISBN 978-92-4156422-9, 2011.

54. Coughlin S, Thind H, Liu B, et al. Mobile phone apps for preventing cancer through educational and behavioral interventions: state of the art and remaining challenges. JMIR Mhealth Uhealth. 2016;4:e69.

74. International Telecommunications Union. The world in 2014: ICT facts and figures. https://www.itu.int/en/ITU-D/Statistics/Documents/ facts/ICTFactsFigures2014-e.pdf. Accessed January 3, 2017.

55. Coughlin SS, Jacobs M, Thind H. On the need for research-tested smartphone applications for reducing exposure to known or suspected breast carcinogens in work and home environments. J Environ Health Sci. 2015;1(4): 10.15436/2378-6841.15.e004.

75. Eskandar H, Land M-A, Arnold V, et al. Mobile technology in cancer control for emerging health systems: digital divide or digital provide?. Cancer Contr. 2015;65-70.

56. Kessel KA, Vogel MM, Schmidt-Graf F, et al. Mobile apps in oncology: a survey on health care professionals’ attitude toward telemedicine, mHealth, and oncological apps. J Med Internet Res. 2016;18:e312.

76. Monica A. “Racial and ethnic differences in how people use mobile technology.” Pew Research Center, April 30, 2015. http://www. pewresearch.org/fact-tank/2015/04/30/racial-and-ethnic-differencesin-how-people-use-mobile-technology/. Accessed January 3, 2017.

57. Carter A, Liddle J, Hall W, et al. Mobile phones in research and treatment: ethical guidelines and future directions. JMIR Mhealth Uhealth. 2015;3:e95.

77. Quintiliani LM, Reddy S, Goodman R, et al. Information and communication technology use by female residents of public housing. mHealth. 2016;2:2.

58. Martínez-Pérez B, de la Torre-Díez I, López-Coronado M. Privacy and security in mobile health apps: a review and recommendations. J Med Syst. 2015;39:181.

78. Bowen DJ, Battaglia TA, Murrell SS, et al. What do public housing residents say about their health? Prog Community Health Partnersh. 2013;7:39-47.

59. Sunyaev A, Dehling T, Taylor PL, et al. Availability and quality of mobile health app privacy policies. J Am Med Inform Assoc. 2015;22:e28-e33.

79. Anguiano B, Brown-Johnson C, Rosas LG, et al. Latino adults’ perspectives on treating tobacco use via social media. JMIR Mhealth Uhealth. 2017;5:e12.

60. Dehling T, Gao F, Schneider S, et al. Exploring the far side of mobile health: information security and privacy of mobile health apps on iOS and Android. JMIR Mhealth Uhealth. 2015;3:e8.

80. Turner-McGrievy GM, Tate DF. Weight loss social support in 140 characters or less: use of an online social network in a remotely delivered weight loss intervention. Transl Behav Med. 2013;3:287-294.

61. Mense A, Steger S, Sulek M, et al. Analyzing privacy risks of mHealth applications. Stud Health Technol Inform. 2016;221:41-45.

81. Laranjo L, Arguel A, Neves AL, et al. The influence of social networking sites on health behavior change: a systematic review and metaanalysis. J Am Med Inform Assoc. 2015;22:243-256.

62. Hall JL, McGraw D. For telehealth to succeed, privacy and security risks must be identified and addressed. Health Aff (Millwood). 2014;33:216-221. 63. Barton AJ. The regulation of mobile health applications. BMC Med. 2012;10:46.

82. Ricard-Gauthier D, Wisniak A, Catarino R, et al. Use of smartphones as adjuvant tools for cervical cancer screening in low-resource settings. J Low Genit Tract Dis. 2015;19:295-300.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 137

CARE DELIVERY AND PRACTICE MANAGEMENT

CLINICAL TRIALS ADMINISTRATIVE BURDEN

Challenges in Opening and Enrolling Patients in Clinical Trials Julie M. Vose, MD, MBA, FASCO, Meredith K. Chuk, MD, and Francis Giles, MB, MD OVERVIEW Clinical trials are key elements of the processes that account for many of the recent advances in cancer care, including decreased mortality rates and increased survivorship; better supportive care; and improved understanding of cancer risk, prevention, and screening. This research also has led to the validation of numerous exciting new types of cancer treatments, such as molecularly targeted therapies and immunotherapies. Clinical trials, however, are becoming more and more challenging to conduct. Research programs must comply with legal and regulatory requirements that can be inefficient and costly to implement and often are variably interpreted by institutions and sponsors and sponsors’ representatives, including contract research organizations. Some of these requirements are essential to protect the safety of trial participants, to promote the scientific integrity of research, or to ensure that trial conduct is efficient and adequately resourced. Such requirements are important to preserve. However, some requirements do not fulfill any of these goals and, in fact, hinder research and slow patient access to safe and effective treatments. This article discusses some of the identified issues that are slowing the process of cancer clinical trials, such as conservatively interpreted guidelines by pharmaceutical companies and contract research organizations; overprotective language for contracts; and patient protections by health systems and universities. The article also discusses possible solutions to these problems that are slowing down the cancer therapies that patients need.

linical trials are key elements of the processes that account for many of the recent advances in cancer care, including decreased mortality rates and increased survivorship; better supportive care; and improved understanding of cancer risk, prevention, and screening. This research also has led to the validation of numerous new types of cancer treatments, such as molecularly targeted therapies and immunotherapies. Clinical trials, however, are becoming more and more challenging to conduct. Research programs must comply with legal and regulatory requirements that can be inefficient and costly to implement and often are variably interpreted by institutions and sponsors and sponsors representatives, including contract research organizations. Some of these requirements are essential to protect the safety of trial participants, to promote the scientific integrity of research, or to ensure that trial conduct is efficient and adequately resourced. Such requirements are important to preserve. However, some requirements do not fulfill any of these goals and, in fact, hinder research and slow patient access to safe and effective treatments. To address the problem of administrative and regulatory burden on cancer clinical trials, the American Society of Clinical Oncology (ASCO) partnered with the Association of American Cancer Institutes on the Best Practices in Cancer

Clinical Trials Initiative (the Initiative). The purpose of the Initiative is to promote practical solutions to meeting existing regulatory and administrative requirements on research. Both ASCO and the Association of American Cancer Institutes have previously explored various strategies to streamline the conduct of clinical trials, such as the development of supportive tools and templates, networking sessions, and the development of common guidelines and standards. This Initiative was an opportunity to expand on the current work in this area. The Initiative was overseen by a multidisciplinary working group, including hematologists and oncologists, research nurses, administrators, managers, and industry representatives. Officials from the U.S. Food and Drug Administration (FDA) and the National Cancer Institute, contract research organization staff, and patient advocates attended. The main elements of the project included (1) a stakeholder survey to identify the most pressing issues in clinical trials that could be addressed by the Initiative and to gather data on use of existing tools and resources, (2) an invitational workshop, which convened many leading oncology professionals and policy makers to identify potential solutions for improving the efficiency and conduct of cancer clinical trials, and (3) dissemination of the recommendations from the workshop through publication; the ASCO Annual Meeting; and the

From the Division of Hematology/Oncology, University of Nebraska Medical Center, Omaha, NE; Office of Hematology and Oncology Products, U.S. Food and Drug Administration, Rockville, MD; Division of Hematology/Oncology, Northwestern University Feinberg School of Medicine; Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, IL. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Julie M. Vose, MD, MBA, FASCO, 987680 Nebraska Medical Center, Omaha, NE 68198; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 139

VOSE, CHUK, AND GILES

development of practical resources, toolkits, and follow-up meetings with relevant organizations and individuals. This article provides a summary of the stakeholder survey and of the workshop.1

ADVERSE EVENT REPORTING: FDA GUIDANCE

As stated in the Code of Federal Regulations (CFR), the primary objective of the FDA in reviewing an investigational new drug application (IND) is “to assure the safety and rights of subjects” (21 CFR §312). One of the principle ways that this is accomplished is by monitoring adverse events that occur during the course of clinical trials. Sponsors of clinical trials conducted under an IND are required under 21 CFR §312.32 to report any suspected adverse reaction that is both serious and unexpected as an IND safety report to the FDA and to all participating investigators. In September 2010, the FDA published a final rule that amended the safety reporting requirements for INDs.2 This rule became effective on March 28, 2011. This rule was adopted to clarify the requirements for safety reporting to improve the quality of reporting, reduce the number of uninformative reports, expedite the FDA review of important safety information, and improve the ability to detect valid safety signals. Sponsors often submitted IND safety reports of individual events that were a result of the underlying disease—events that occurred often in the population evaluated, or events that were study endpoints. This resulted in submission of a large number of uninterpretable and uninformative safety reports that strained the limited resources of the FDA, investigators, and institutional review boards and did not contribute meaningfully to the development of a drug safety profile. The FDA published two guidance documents to help sponsors and investigators comply with the requirements of the 2010 final rule: Safety Reporting Requirements for INDs and

KEY POINTS • Some steps that have been shown to dramatically improve the efficiency of the clinical research process include establishing clear timelines; developing transparent metrics; and defining the appropriate role for center clinical investigators. • Another key step is the development of standardized study budgets, based on fair market value as applicable to the center’s level and location, with dynamic benchmarking with comparable centers. • The ratio of and relative priority given to commercially sponsored studies, investigator-initiated studies/trials, the National Clinical Trials Network, or federally funded trials should be addressed. • The FDA guidance documents on adverse event reporting should be followed by all sponsors, contract research organizations, and investigators for cancer clinical trials. • Health care institutions should work with the investigators and the sponsors on reasonable terms for master contracts that should be universally accepted. 140 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

BA/BE Studies, in December 2012,3 and Safety Assessment for IND Safety Reporting, in December 2015 (Draft Guidance).4 A brief review of the these guidance documents with a focus on reporting events from clinical trials is below, followed by a review of results of an internal FDA audit about the quality of safety reporting in oncology and efforts to improve the problem of ongoing uninformative safety reporting.

Review of FDA Guidance on IND Safety Reporting

The final rule clarified the following definitions to be used for reporting purposes: • Adverse event: any untoward medical occurrence associated with the use of a drug, whether it is considered drug related or not • Suspected adverse reaction: any adverse event for which there is a reasonable possibility that the drug caused the adverse event (i.e., evidence to suggest a causal relationship between the investigational drug and the adverse event) • Life-threatening adverse event or life-threatening suspected adverse reaction: an event that, in the view of either the investigator or sponsor, places the patient at immediate risk of death • Serious adverse event or serious suspected adverse reaction: an adverse event that, in the view of either the investigator or sponsor, results in any of the following outcomes: death, a life-threatening adverse event, inpatient hospitalization or prolongation of existing hospitalization, a persistent or significant incapacity or substantial disruption of the ability to conduct normal life functions, or a congenital anomaly/birth defect. Important medical events that do not meet the above criteria may be considered serious when, based upon appropriate medical judgment, they may jeopardize the patient and may require medical or surgical intervention to prevent one of the outcomes listed in this definition. • Unexpected adverse event or unexpected suspected adverse reaction: An adverse event is considered unexpected if it is not listed as occurring with the particular drug in the investigator brochure or other risk information, or is not listed at the specificity or severity as the current event. Sponsors of clinical trials conducted under an IND application must notify the FDA and all participating investigators in an IND safety report (a 7- or 15-day report, depending on the type of event) of potential serious risks from clinical trials or other sources. In the IND safety report, the sponsor must identify all reports of similar events previously submitted to FDA, and the sponsor should take these events and any other relevant information into consideration for the assessment of causality and significance of the suspected adverse reaction being reported. Sponsors must report suspected adverse reactions that are both serious and unexpected. Reports that do not satisfy all three criteria should not be submitted to the FDA as IND safety reports.

CLINICAL TRIALS ADMINISTRATIVE BURDEN

The issue of causality is often the most difficult to assess, but it is the most critical to avoid uninformative reporting. The regulations state that there must be a reasonable possibility that the adverse event was caused by the drug. This should be interpreted as the presence of enough evidence to suggest a causal relationship. Per FDA regulations, the determination of causality for the purposes of reporting rests with the sponsor, not the investigator. The sponsor has access to the most up-to-date and comprehensive information available about the drug and is best able to make informed and consistent decisions about causality. This is a difference between the FDA regulations and the ICH E2A guideline,5 which allows the determination of causality to be made by the investigator or the sponsor. Sponsors should consider the following when assessing causality: • Single adverse events usually are uninterpretable and would not meet the criteria for expedited reporting except in cases of events that are uncommon and known to be associated with use of a drug, such as Stevens-Johnson syndrome or angioedema. • Multiple occurrences of events not commonly associated with drug exposure but otherwise not common in the population (e.g., tendon rupture), may be informative. Single events with strong evidence of causation, (i.e., a strong temporal relationship or recurrence on rechallenge) may constitute sufficient evidence for an expedited report, but generally more than one similar event is needed to suspect a causal relationship. • Adverse events that are likely to occur (i.e., are anticipated) in the population under evaluation, whether as a result of the age of the patient, nature of the disease, or concomitant therapy (e.g., cardiac disease in older patients with risk factors or fever and neutropenia in patients receiving cytotoxic chemotherapy) should not be reported as single events, because there is inadequate information to determine a reasonable possibility of causality. These events require aggregate analysis across the development program to determine if they truly occur more often in patients exposed to the drug. These aggregate analyses require that the sponsor have a system in place for ongoing review and analysis of safety data throughout the development of the drug. If these analyses reveal there is an imbalance between patients who did and those who did not receive the investigational drug, this information should be reported in an IND safety report. An event is considered unexpected if it is not listed in or if it occurs at a severity or frequency that is unusual from that listed in the investigator brochure or other risk information. The investigator brochure should contain a list of adverse events that have been observed with use of the drug and for which a causal relationship is suspected or confirmed. Clinical judgment is required to establish and then maintain this list after periodic review of safety information from ongoing clinical investigations. Adverse events that qualify for IND safety reporting must be submitted to the FDA and participating investigators as

soon as possible, but no later than 15 calendar days after the sponsor determines that the information qualifies for reporting. Unexpected fatal or life-threatening suspected adverse reactions must be reported as soon as possible, but no later than 7 days after the sponsor receives the information. Follow-up reports are only required for relevant information that is necessary to evaluate the suspected adverse reaction. For clinical trials that are blinded, the blind generally should be broken for IND safety reports submitted to FDA and investigators, because information about drug exposure is necessary to interpret the event, treat the patient, and institute any changes in trial conduct, such as increased monitoring or changes to the informed consent document. This unblinding should not affect the integrity of the trial, because it should be infrequent for single events. A data monitoring committee or independent safety team should review safety data to determine if aggregate reporting of any particular adverse event is appropriate. Sponsors also are required to submit safety information from other clinical studies, epidemiologic studies, and pooled analyses of multiple studies; findings from in vitro studies that suggest a notable risk in humans; and any increased rate of serious suspected adverse reactions other than that listed in the protocol or IB. Clinical judgment is required to determine what is a clinically meaningful increase on the basis of the trial population(s), nature and severity of the adverse event, and the magnitude of increase. The submission of an IND safety report for the findings listed above should be enough to require a change to the protocol (e.g., monitoring or eligibility criteria) or to the informed consent document.

Internal Audit of Expedited Safety Reports in the Office of Hematology and Oncology

From the years 2006 to 2014, the Office of Hematology and Oncology Products received an average of 17,686 expedited safety reports per year. Additional analysis of the number of reports per IND per year that were submitted before and after the implementation of the 2010 final rule on IND safety reporting showed that, not only was there no change since the implementation of the final rule, there was actually a slight increase.5 In 2015, medical officers in the Office of Hematology and Oncology Products who were responsible for evaluating IND safety reports conducted a review of 160 initial safety reports submitted to commercial INDs and concluded that only 14% met criteria for reporting.6 The remaining 86% of the reports were determined to be uninformative for a variety of reasons: 54% of the reports were for adverse events that were expected on the basis of information in the investigator brochure or product labeling, in 50% of the reports, the sponsor did not conclude that the adverse event was related to the drug; and, of the reports that met all three criteria of serious, suspected, and unexpected events, 42% of the events were determined to be anticipated on the basis of the FDA review (e.g., febrile neutropenia in a patient who received a backbone of cytotoxic chemotherapy). asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 141

VOSE, CHUK, AND GILES

Steps Toward Process Improvement

The FDA has been encouraging sponsors to develop mechanisms to reduce uninformative reporting. Several sponsors have been successful at dramatically decreasing the number of initial and follow-up safety reports through a variety of measures, including establishing dedicated teams of physicians to review safety reports, implementing consistent thresholds for determination of causality, and identifying and reporting only clinically relevant follow-up information that directly contributes to the assessment of the suspected adverse reaction.7 In addition, 21 CFR § 312.32 (c)1(v) allows for submission of IND safety reports in an “electronic format that FDA can process, review, and archive.” The FDA is exploring the digital submission of expedited safety reports based on ICH E2B guidelines—Technical Requirements for Registration of Pharmaceuticals for Human Use—for postmarket safety report submissions5 on the basis of a successful pilot study conducted by the Office of Hematology and Oncology Products and the Office of Surveillance and Epidemiology.7 This move to standardize reporting and submit safety information as data sets uses mechanisms already in place for safety reporting in the postmarket setting. This method of submission will allow for a more consistent and streamlined way to receive, process, and analyze safety information, the ability to better detect safety signals and ensure the protection of patients, and the identification of relevant events at time of reporting. Efficient and timely submission and review of relevant safety data are imperative to ensure patient safety in clinical trials. Unfortunately, revisions to the IND safety reporting requirements instituted in the final rule in 2010 did not result in the desired decrease in the number of uninformative safety reports, which continue to be burdensome for the FDA, investigators, and investigational review boards to process and review; these reports also make the detection of genuine safety signals more difficult. Perceived barriers by sponsors to implementation of the provisions of the 2010 final rule include a lack of international harmonization on all elements of reporting as well as concerns related to unblinding during the course of clinical trials and to thresholds for reporting serious and unexpected adverse reactions.8 Efforts to improve and streamline the reporting process are ongoing, but successful implementation will require all sponsors to identify barriers to and institute mechanisms for decreasing uninformative IND safety reporting; efforts also will require regulators to continue to engage with all stakeholders to optimize the process for IND safety reporting.

BREAKING DOWN THE BARRIERS: THE PATH FORWARD

The barriers to patient participation in cancer clinical studies are numerous; less than 5% of patients participate in a clinical trial. Conducting relevant public education campaigns; addressing financial and other study access barriers; and increasing physician advocacy for, and conduct of, clinical trials are important ways to address this major challenge. 142 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

In cancer centers where the conduct of clinical studies is an important core activity, there is increasing concern about the escalation of attendant financial and personnel costs of study conduct. Efforts to standardize and streamline the process of opening, conducting, and closing studies at cancer centers are ongoing. Some key steps that have been shown to dramatically improve the efficiency of this process include: 1. Map the institutional processes used to open, conduct, close, and report a study. The involvement of all relevant stakeholders and of professional process improvement colleagues is critical to the success of this step. It is important to leverage an underlying proven, data-driven, process-improvement methodology, such as Define, Measure, Analyze, Improve, Control (i.e., DMAIC), that can be scaled according to the scope and depth of current and desired clinical study activity. 2. Eliminate all unnecessary duplicative steps in the process, and pay particular attention to steps that are necessary for full compliance with applicable mandatory standards (e.g., National Cancer Institute– designated cancer centers). Establish clear timelines for—and definitions of roles, responsibilities, and deliverables of those involved in—the remaining essential steps. Develop an ongoing dynamic feedback system to monitor the efficiency of these key steps. 3. Develop transparent metrics for expected productivity and outcomes of key administrative, financial, and research staff. 4. Define the appropriate role for center clinical investigators in study budget development and study institutional resource allocation, with appropriate consideration of conflict-of-interest issues, academic freedom, and operational efficiency. Develop stan dardized study budgets that are based on fair market value as applicable to the level and location of the center. Ensure that such fair-market-value budgets are dynamically benchmarked with comparable centers and are offered in a transparent manner to all sponsors for all studies, regardless of whether the study is offered directly from a sponsor or via intermediary entities. 5. Develop policies, in conjunction with comparable clinical research sites, on institutional responses to site evaluation/screening questionnaires; on responses to sponsor/contract research organization requests for site evaluation/qualification visits; on minimal qualifications/experience levels of external staff who conduct or monitor the study; on the nature and frequency of remote and on-site study monitoring activities; and on standards for authorship/acknowledgment expectations in studyrelated publications. 6. A specific issue that merits the development of institutional policies is the ratio of, and relative pri ority given to, studies or trials that are commercially

CLINICAL TRIALS ADMINISTRATIVE BURDEN

sponsored and investigator initiated, that are National Clinical Trials Network studies, or that are federally funded. Additional specific issues include the role of central institutional review boards, participation in multicenter groups with common consensus administrative and budget policies, and the development of specific alliances with sponsors to improve the investigator-initiated studies/trials process.

Major opportunities for standardization of approaches to clinical research conduct between cancer centers exist. Key steps will involve central registers/repositories of commonly requested study conduct documents and central records of investigator and institutional research capabilities, interests, infrastructure/resources and productivity. Increasing the role of central key organizations, such as ASCO, in developing, monitoring, and refining policies and procedures to optimize clinical research conduct also will be crucial.

References 1. Vose JM, Levit LA, Hurley P, et al. Addressing administrative and regulatory burden in cancer clinical trials: summary of a stakeholder survey and workshop hosted by the American Society of Clinical Oncology and the Association of American Cancer Institutes. J Clin Oncol. 2016;34:3796-3802. 2. U.S. Food and Drug Administration. Investigational new drug safety reporting requirements for human drug and biological products and safety reporting requirements for bioavailability and bioequivalence studies in humans. Fed Regist. 2010;75: 59935-59963. 3. U.S. Food and Drug Administration. Guidance for Industry and Investigators: Safety Reporting Requirements for INDs and BA/BE Studies. http://www.fda.gov/downloads/Drugs/GuidanceComplia nceRegulatoryInformation/Guidances/UCM227351.pdf. Accessed February 1, 2017. 4. U.S. Food and Drug Administration. Safety Assessment for IND Safety Reporting: Guidance for Industry Draft Guidance. http://www.fda.

gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/ Guidances/UCM477584.pdf. Accessed February 1, 2017. 5. International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use. ICH Harmonised Tripartite Guideline: Clinical Safety Data Management Definitions and Standards for Expedited Reporting E2A. http://www.ich.org/ fileadmin/Public_Web_Site/ICH_Products/Guidelines/Efficacy/E2A/ Step4/E2A_Guideline.pdf. Accessed February 1, 2017. 6. Jarow JP, Casak S, Chuk M, et al. The majority of expedited investigational new drug safety reports are uninformative. Clin Cancer Res. 2016;22:2111-2113. 7. Khozin S, Chuk M, Kim T, et al. Regulatory watch: evaluating the potential for digital submission of expedited premarket safety reports to the FDA. Nat Rev Drug Discov. 2016;15:670-671. 8. Archdeacon P, Grandinetti C. Vega JM, et al. Optimizing expedited safety reporting for drugs and biologics subject to an investigational new drug application. Ther Innov Regul Sci. 2014;48:200-207.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 143

PENNELL ET AL

mHealth: Mobile Technologies to Virtually Bring the Patient Into an Oncology Practice Nathan A. Pennell, MD, PhD, Adam P. Dicker, MD, PhD, Christine Tran, MS, Heather S. L. Jim, PhD, David L. Schwartz, MD, FACR, and Edward J. Stepanski, PhD OVERVIEW Accompanied by the change in the traditional medical landscape, advances in wireless technology have led to the development of telehealth or mobile health (mHealth), which offers an unparalleled opportunity for health care providers to continually deliver high-quality care. This revolutionary shift makes the patient the consumer of health care and empowers patients to be the driving force of management of their own health through mobile devices and wearable technology. This article presents an overview of technology as it pertains to clinical practice considerations. Telemedicine is changing the way clinical care is delivered without regard for proximity to the patient, whereas nonclinical telehealth applications affect distance education for consumers or clinicians, meetings, research, continuing medical education, and health care management. Technology has the potential to reduce administrative burdens and improve both efficiency and quality of care delivery in the clinic. Finally, the potential for telehealth approaches as cost-effective ways to improve adherence to treatment is explored. As telehealth advances, health care providers must understand the fundamental framework for applying telehealth strategies to incorporate into successful clinical practice.

elehealth encompasses a broad variety of technologies with clinical applications to deliver virtual health care services. Because there is no universal definition, the terms telehealth, telemedicine, eHealth, digital health, or mobile health (mHealth) often are used interchangeably. However, the U.S. Department of Health & Human Services defines telehealth as the use of electronic information and telecommunication technologies to support and promote long-distance clinical health care, patient and professional health-related education, public health, and health administration.1 Although this broad definition includes both clinical and nonclinical applications, the term telemedicine is confined to clinical services in remote locations and is defined as allowing health care professionals to remotely evaluate, diagnose, and treat patients using telecommunications technology.2 These clinical applications encompass services that support remote electronic clinical consultation, such as diagnosis, patient communication, disease management, remote monitoring, and clinician support. Meanwhile, nonclinical applications can include distance education for consumers or clinicians, administrative meetings, research, continuing medical education, or health care management.3 Telehealth innovations enable the delivery of care irrespective of geographic location, bringing about a fundamental

shift in U.S. health care by bringing health care to the patient. Moreover, the need to improve quality, access, equity, and affordability of health care supports the utilization of telehealth across several medical disciplines. The potential shortage of oncology services is pointed out in ASCO’s report, The State of Cancer Care in America: 20164; evidence-based health research supports the use of telehealth in the oncology setting and its ability to increase access to patients with cancer.5-7 For example, in a systematic review of experiences for patients with cancer who have participated in telehealth interventions, telehealth was noted to be an advantageous approach to reduce treatment burden and disruption to patient lives.8 Health care professionals who use telehealth to export their clinical expertise enable patients to experience decreased travel time, immediate access to care, early detection of health issues, increased patient autonomy, reduced caregiver burden, and increased patient satisfaction with health care.

TELEHEALTH TECHNOLOGY

The most commonly used telehealth technology employs video conferencing to connect a patient to a health care provider.9 Video conferencing integrates telecommunications

From the Department of Hematology and Medical Oncology, Cleveland Clinic Taussig Cancer Institute, Cleveland, OH; Department of Radiation Oncology, The Sidney Kimmel Cancer Center at Thomas Jefferson University, Philadelphia, PA; Department of Health Outcomes and Behavior, Moffitt Cancer Center, Tampa, FL; Department of Radiation Oncology, University of Tennessee Health Sciences Center West Cancer Clinic; University of Tennessee Health Sciences Center, Vector Oncology, Memphis, TN. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Nathan A. Pennell, MD, PhD, Department of Hematology and Medical Oncology, Cleveland Clinic Taussig Cancer Institute, 9500 Euclid Ave, R-35, Cleveland, OH 44195; email: [emailprotected]. © 2017 American Society of Clinical Oncology

144 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

USE OF MHEALTH TECHNOLOGIES TO IMPROVE PATIENT CARE IN ONCOLOGY

technology to allow patients and providers to “electronically collaborate face to face, in real time, and share all types of information, including data, documents, sound, and picture.”9 This type of interactive video conferencing environment allows for patient-provider consultation, discussion, education, and patient monitoring. The use of telehealth technology offers great promise and currently is being used in health care in a number of ways. The clinical applications of telehealth range from drug formulary apps to reference programs, educational apps, medical tools (patient documentation apps, patient monitoring apps, nursing apps, imaging apps, and clinical apps), payer tools, decision support tools, and patient support tools.10 The technological advances of telehealth include wearable sensors (pedometer/ accelerometer, or sensors of sleep, weight, blood pressure, heart rate, temperature, environment exposure, blood levels, falls, and geolocation), data entry technologies (exercise testing, diet, mood/stress levels, symptoms, health-related quality of life, functional status, social support, medication, tobacco use, pillbox sensors, and alcohol use), ingestible/ implantable sensors, biometric sticker sensors, and the ability of smartphones to be used as otoscopes, ophthalmoscopes, and microscopes. This technology can be used to remotely collect and send data for interpretation by a health care provider.11 Telehealth interventions also have been expanded to social media sites such as Twitter to foster healthy lifestyles through the use of wearables for self-monitoring and social media to facilitate support for behavioral changes.12 The U.S. Food and Drug Administration also has approved imaging apps, which allow radiologists to interpret images or ophthalmologists to use color vision plates for clinical evaluation when a more traditional outlook is not available. Digital images also are a type of store-andforward technology, which permits the electronic transmission of medical files to be used at the convenience of providers to then make diagnoses, recommendations, and treatment plans. Whether the device exists as a standalone item, such as a smartphone, wearable, or hybrid (e.g., smartwatch), the information can be used by remotely monitoring health,

KEY POINTS • Oncology health care is ripe for digital health disruption with the convergence of mobile technology, platforms, networks, and the introduction of machine learning. • Digital platforms that include telemedicine, internet of things, and wearables are scalable. • mHealth technology, including virtual scribes, real-time location systems, and peer-to-peer messaging apps, has the potential to improve the efficiency and quality of clinical cancer care. • Treatment nonadherence in oncology occurs at a high rate and is associated with worse outcomes. • Innovative, collaborative research will be pivotal to transform mHealth into a standard part of modern cancer care relevant to the 21st century health care marketplace.

medical behavior (e.g., compliance, movement, symptoms, vital signs, diet) or a person’s location.11 Moreover, the demand to satisfy uniform quality of telehealth services has been met recently through the American Telemedicine Association. These practice guidelines and technical standards include practice guidelines for videoconferencing-based telemental health, evidence-based practices for telemental health, core standards for telemedicine operations, practice guidelines for teledermatology, telehealth practice recommendations for diabetic retinopathy, home telehealth clinical guidelines, and clinical guidelines for telepathology.13 The standardization of telehealth guidelines may help reduce the cost of equipment and increase adoption by making telecommunication independent of hardware used.

HEALTH CARE CONSUMER AND PROVIDER PERSPECTIVES

The goal of telehealth is equal efficiency with in-person care, and physician-patient encounters via telehealth recently have reported consistent performances compared with standard face-to-face care.14 In a randomized, controlled trial for patients with prostate cancer that used telehealth after radical prostatectomy to assess the efficiency, satisfaction, and cost of remote virtual visits versus traditional office visits, telehealth was equivalent in patient and provider satisfaction and time allocated to care.15 In another study to evaluate the opinion on the use of telehealth in oncology, a majority of responders cited advantages of oncologic apps that included better documentation, improved and continual care for patients, enhancement of communication between provider and patient, improved patient compliance, possible use of data for scientific evaluation, and potential for patient-independent information.16 Overall, 84.3% supported the use of oncologic apps complementary to traditional treatment.16 Critics of telehealth cited issues related to legal uncertainty, data privacy, and insecure data transfer and storage.16 Moreover, in a group of surveyed health care professionals, the most common medical app functions included drug-referencing tools, clinical decision-support tools, communication, electronic health record (EHR) access, and medical education materials.17 The amount of scientific material that clinicians must memorize is large, so reference programs and educational apps help enable clinicians to choose clinically appropriate and cost-effective drugs, quickly search and access information/textbooks, perform calculations, log experiences, communicate, and input specific patient information for diagnosis.10 The adoption of telehealth technology relies on patient participation and the motivation of patients to become partners in their health care. With a consumer-based foundation, telehealth shifts medicine to more participatory care and an improved health care system composed of patient empowerment. This paradigm shift in responsibility allows patients to manage their health, health network, and heath information, and it leverages emerging technologies for a patient-centered ecosystem. In a survey to assess patient asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 145

PENNELL ET AL

attitudes toward telehealth, patients had a positive overall attitude and cited an opportunity for improved self-efficacy and improved provider-driven medical management.18 Moreover, respondents mentioned comfortableness in being remotely monitored with confidence in privacy protection. Findings about telehealth from the experience of a cancer survivor illustrated analytic themes that included how telehealth limited the disruption to people’s lives, how telehealth could enable close and personalized relationships between cancer survivors and service providers, and how survivors felt that they had immediate access to professional advice, which acted as a safety net for possible issues in treatment.8 Nevertheless, individual differences in digital literacy (i.e., the competency and technical skills to operate digital devices and conceptually understand their functionality) have the potential to widen health disparities and must be addressed as telehealth becomes more widespread.19,20

TELEHEALTH CHALLENGES FOR CLINICAL PRACTICE

Despite its potential, telehealth issues of privacy and security remain ongoing concerns for health care professionals and patients alike. For telehealth to complement traditional approaches in the delivery of health care, it must be delivered to both clinicians and patients with confidence that the privacy, confidentiality, and security of their data will be safeguarded within compliance of the Health Insurance Portability and Accountability Act (HIPAA). In an emerging field, the means for securing data includes understanding the roles of cybersecurity and developing a mobile technology policy to ensure that protected health information data are safe. Moreover, patient portals tethered to EHRs include advanced technology as part of their system to provide scheduling, billing, and clinical support, but there is no policy for telehealth applications to be fully integrated into health information systems in hospitals or provider organizations.21 The variation in telehealth data and a patient’s EHR displays the difficulty of management for telehealth and need for integration. In the progression of telehealth, health care institutions must establish a method for health care providers to access the EHR at the time of a telehealth encounter and establish a foundation of interoperable standards. Multiple factors on both the individual and organizational levels are crucial to clinician acceptance and adoption of telehealth technology. Clinician acceptance of telehealth technology depends on a full integration into the workflow, added value to patient care, administrative convenience, and facilitated communication among multidisciplinary teams.22 Although usefulness and ease of use were cited as important factors to the adoption of telehealth, the argument of whether it is an affordable option is still in discussion among health care professionals, who have referenced cost issues as limiting the adoption of telehealth tools.23,24 Elements related to costs (e.g., the question of how to bill for telehealth) act as barriers to its adoption. Reimbursem*nt regulations for medical services were planned before telehealth technology, which thus gives each state the option 146 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

of whether to cover telehealth. These variations in reimbursem*nt relate to service coverage, payment methodology, distance requirements, eligible patient populations, authorized technology, and patient consent.25 Moreover, traditional concepts of liability and malpractice still apply to telehealth practitioners, who are more vulnerable to legal issues and who may face an additional fee for malpractice insurance.26 Despite technological advances, legal and regulatory challenges concerning provider licensure, credentialing, and privileging processes remain an obstacle for all allied health professionals. Mutual recognition models such as the multistate Nurse Licensure Compact or the Interstate Medical Licensure Compact are just beginning to develop to help facilitate telehealth interactions across state boundaries and into the mainstream. Additionally, the mandate for credentialing and privileging in multiple, separate health care facilities offer similar challenged for health care providers to deliver telemedicine.

FUTURE DIRECTIONS FOR TELEHEALTH

Telehealth is the future to improved access to specialized medicine, preventive care, monitoring of chronic conditions, and improved patient outcomes and satisfaction. It has the potential to reduce fragmentation of care and allow access to care despite the distance from major medical centers. In a 2014 study, telehealth industry growth and its potential to decrease care costs within the health care system were demonstrated; the study outlined $5 billion in savings on the basis of an estimated 100 million telemedicine visits across the world.27 Demands for improved access to care in rural areas or to underserved populations that have been a challenge historically because of a shortage of clinicians or because of financial or geographic barriers also create the potential for a new telehealth ecosystem and novel health care model. Telehealth can overcome many of these barriers; it already has increased the quality of care and reduced costs by reducing the readmissions and emergency visits in rural communities.28 Telehealth effectiveness also has been demonstrated through research in rural and remote areas, where telehealth satisfaction reached 94%.29 These findings suggest a general acceptance of therapies delivered via telehealth, which advocates for its unparalleled opportunity. Growing interest in tele-oncology also shows the potential to increase access from a comprehensive cancer center to patients in rural areas by offering consultations, supervision of chemotherapy administration, oral medication adherence, or symptom management.30

THE POTENTIAL OF MHEALTH TECHNOLOGY TO IMPROVE EFFICIENCY AND CLINICAL CARE

mHealth technology has a tremendous potential to improve clinical care; its uses range from telemedicine patient encounters to the collection of patient-reported outcomes and improved adherence to therapies with apps and mobile devices. However, there is a lack of research about what patients will benefit the most, what the efficiency of telehealth

USE OF MHEALTH TECHNOLOGIES TO IMPROVE PATIENT CARE IN ONCOLOGY

is at saving costs or time, and whether its contribution to a greater provider burden significantly hinders the advancement of telehealth. Apps for electronic patient-reported outcomes are available now from the Apple and Android app stores. One example is the Strength Through Insight app (Fig. 1).31 The Strength Through Insight study aims to assess the feasibility of collecting survey data from patients through digital technologies and hand-held devices.31 Practitioners may worry about the impact these technologies have on their day-to-day workflows and how demands for increasing technological innovation may interfere with their primary job of caring for patients. To what extent are these changes taking into account improvements in the efficiency of patient care? Efficiency has not been a major consideration in the design of much of health care technology, but there are a number of areas in which mHealth tools can be used not just to improve compliance or billing but also to benefit day-to-day practices.

Virtual scribes, connected by audio and video to the patient and practitioner through a wireless connection such as Google Glass,38,39 could provide the same advantages as an in-person scribe but without the space issues or intrusiveness of an additional person in the room. There could also be cost advantages, such as reduced expense in hiring and training scribes in a HIPAA-compliant location that can link out to clinics around the world, even, potentially, in countries where highly educated individuals are available at reduced cost. Patients would still have to consent to this service, and there are important issues related to protection of protected health information and data security that must be addressed, but hospital systems around the country already are adopting this model with some success.39 A pilot study to investigate the impact of virtual scribes on documentation time and on patient and physician satisfaction is planned (unpublished observation).

REDUCING THE BURDEN OF DOCUMENTATION IN THE EHR WITH VIRTUAL SCRIBES

mHealth technology does not always have to connect to the outside world. Real-time location system (RTLS) technology is emerging as a useful tool to help improve patient flow within clinics and hospitals by allowing real-time localization of patients and practitioners.40,41 In general, patients or practitioners wear a badge that allows them to be tracked in real time by a variety of possible means (e.g., wireless local area networking (Wi-Fi), radio-frequency identification (RFID), or global positioning system (GPS), and the patterns of movement and time spent in a particular location can be recorded. This can help with clinic flow and treatment chair management, and it can decrease room turnover time.40,42 RTLS also can allow rapid localization (which can be a tedious process) of individual practitioners to sign orders, for example. Some RTLS systems allow hands-free verbal communication through the badges.43 Although little data exist specifically in the oncology field about the use of RTLS to improve efficiency, data in other health care settings supports RTLS as a viable option, and a number of prominent institutions, including a cancer center affiliation of one author (N. A. P.), has adopted this technology.44,45 In an example of how RTLS can be used, a timer starts when patients are roomed; if no practitioner enters the room within 15 minutes, a nurse is alerted to find the practitioner and to reassure the patient. As a result of the positive effect on clinic flow as well as the possible impact of the Hawthorne effect (i.e., that watching someone tends to influence their behavior), studies have shown the patient wait times can be lowered and satisfaction scores can be improved by RTLS.42

The primary components of health care technology that practitioners interact with on a day-to-day basis are the EHR, clinical decision support tools, and clinical physician order entry.32 The primary intent behind adoption of these tools has been the reduction in preventable medical errors, as outlined in the Institute of Medicine report “To Err is Human; Building a Safer Health Care System,”33 and their use is encouraged through the Health Information Technology for Economic and Clinical Health (HITECH) Act.34 Although much time and money have been spent on their adoption, little time has been spent making these systems user friendly or efficient. Additional requirements specific to oncology, such as meeting criteria for participation in the Oncology Care Model,35 only worsen the bureaucratic burden. There is a growing realization that documentation in the EHR places a substantial time burden on practitioners and is drastically reducing the amount of time physicians can spend face to face in direct patient interaction. This has consequences in reduced patient and physician satisfaction as well as in reduced clinical productivity and income.36 The need for documentation in an EHR is not going away anytime soon, so a workaround, the medical scribe, has allowed practitioners to spend more time with patients. Scribes, who usually are unlicensed professionals hired to retrieve from and transcribe data into the EHR, have been shown in various clinical settings to decrease time spent in documentation and to improve both the quality of documentation and patient satisfaction.37 Scribes introduce challenges too, including space issues in the exam room, patient discomfort with a stranger in the room during sensitive conversations, and—of course—expense and availability of trained scribes issues. However, the capacity of telemedicine for instantaneous, real-time communication anywhere in the world now means that the scribe does not have to be in the same room or even in the same country as the practitioner.

REAL-TIME LOCATION SYSTEMS

USE OF MOBILE TECHNOLOGY FOR PHYSICIAN-TO-PHYSICIAN COMMUNICATION

Communication between health care practitioners is critically important to high-quality health care, especially in a field as multidisciplinary as oncology. This is true whether it occurs between nurse and physician, between resident and supervising physicians on a health care team, or between asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 147

PENNELL ET AL

FIGURE 1. The Strength Through Insight App

The app allows patients and their caregivers to build a partnership for communication throughout their cancer treatment. The survey uses standard questions that can be answered digitally via an app at a set schedule.

148 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

USE OF MHEALTH TECHNOLOGIES TO IMPROVE PATIENT CARE IN ONCOLOGY

consulting services. Although there are a great many ways that practitioners communicate, ranging from alphanumeric pagers to email, the ubiquity of mobile phones and texting/ instant-messaging apps opens up a whole new arena of opportunity for communication. Mobile phones have been tested in medical settings and compared with pagers in terms of speed of communication and reduction in medical errors; results generally are in favor of mobile phones.46 However, the speed and ease with which practitioners can be reached with mobile phones has drawbacks. Although replacement of pagers with mobile phones has been shown to improve efficiency and decrease the time needed to reach physicians, it may not improve nursing satisfaction with communication. In fact, in one study, use of mobile phones reduced face-to-face communication of nurses with doctors. Instead communication primarily occurred by texting or phone calls, which were considered less meaningful.47 Studies also have suggested that the ease of mobile phone–based communication significantly increases the number of messages, which can be disruptive to workflow.48 Finally, questions about the security of protected health information depend on the specific application used for texting; institutions that choose texting as a preferred communication method must provide a properly secure environment.49 Despite the risks of increased interruptions and security, mobile phones seem likely to replace pagers and other types of physician-to-physician communication, given their prominence in all other aspects of our lives. In middle-income countries, mobile phones may represent the best available means of communication. An example of the use of mobile technology in this manner is the widespread use of the web-based messaging app, WhatsApp (WhatsApp, Inc., Mountain View, CA), which has approximately one billion users worldwide and has been tested in a number of health care settings.50 WhatsApp has advantages compared with short message service texting in that closed groups can be created, and all communications can be viewed securely by all group members, which allows supervision of team communications. Notifications can be sent when a message has been read, and the app is fairly inexpensive, because practitioners can use their own phones and communicate with the wireless network of the institution. In some countries, such as Israel, WhatsApp is used by up to 96% of physicians, and up to 71% use it for communication of patient information and for consultations.50 Several studies have shown WhatsApp to be a viable method of communicating patient information, asking questions of supervising physicians, and getting feedback among members of a health care team.51,52 Given the importance of teamwork and multidisciplinary care to oncology,53 the availability of a secure and rapid method for team communication would have tremendous potential to aid patient care. However, there are concerns about the security of WhatsApp for communicating protected health information,54 because the app security is endto-end encrypted only if all members of a communication

group have the most up-to-date version of the software. WhatsApp represents an intriguing illustration of the potential for web-based messaging for clinical communication. However, before adoption of a specific app for professional communication, the policy of the institution about the use of such technology must be clear. Many oncologists and other practitioners view health information technology as a burden that decreases their face-to-face time with patients and contributes to burnout, but it is important to point out that technology also has the potential to improve efficiency and reduce time spent on low-value tasks. Although this is by no means a comprehensive list, the examples of virtual scribes to reduce time spent typing on the EHR, RTLS to reduce time spent moving patients or searching for providers, and use of mobile apps for better team communication should illustrate how technology may reduce burdens in caring for patients with cancer. As these technologies advance, however, it will be critically important to study their effects on patient care and practitioner well-being and to make sure that the rapid pace of technological development does not conflict with laws in place to protect patient confidentiality.

MHEALTH APPROACHES TO IMPROVING TREATMENT ADHERENCE

Decreased adherence to treatment is well documented for many chronic diseases. Adherence rates vary across diseases and patient factors, with an overall nonadherence rate of 24.8%.55 This is an important topic, given that decreased treatment efficacy is a consequence of nonadherence.56 The empiric data specific to adherence in oncology is more limited. Innovative use of mobile technology is well suited to support strategies to improve adherence in oncology, although study of these approaches is still limited. Given that use of telehealth approaches may provide a cost-effective way to improve outcomes for patients with cancer, validated approaches to use of this technology are highly desirable. Although there are decades of experience in measuring and improving adherence in many chronic diseases, this topic has not received the same attention in oncology. Given the historical dominance of infused therapies in oncology, the concept of adherence as understood in other therapeutic areas was not relevant—antineoplastic treatment was delivered in direct view of the health care team. Empiric work conducted to understand adherence to cancer treatment is more limited, and strategies to optimize patient adherence have not been incorporated into usual care. A preponderance of the empiric work on treatment adherence in oncology has focused on imatinib for chronic myeloid leukemia, or on hormonal therapy for breast cancer (e.g., tamoxifen, aromatase inhibitors), because these were among the first widely used oral medications to require long-term administration in oncology.57 Adherence rates in these indications are similar to those documented in other therapeutic areas; adherence to oral chemotherapy has ranged in empiric studies from 50% to 89%, depending on the definition of adherence and the study methodology.57-59 asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 149

PENNELL ET AL

An important methodologic consideration in adherence research is the measure of adherence used. Measures can be described as direct measures (e.g., blood levels, provider observation) or indirect measures that are further subdivided into objective (e.g., prescription fills) or subjective (e.g., patient self-report) measures. Comparison of objective and subjective measures shows that patients systematically over-report adherence behavior.60 Treatment adherence is a critical issue for oncology, because studies consistently have shown that nonadherence leads to worse outcomes, including decreased survival.61 Data from studies of infused therapies are instructive to define the risk of missed doses. A study of relative dose intensity of adjuvant chemotherapy delivered to patients with breast cancer found that the cohort of patients who received a relative dose intensity of less than 65% achieved an overall survival equivalent to a control group who received no adjuvant chemotherapy.62 Increased mortality underscores another difference between the consequences of poor adherence in oncology and those in other diseases, in which the risk is limited to increased morbidity. Nonadherence also leads to increased health care utilization and increased cost.63 Predictors of nonadherence include patient factors (e.g., age, gender, amount of social support), treatment factors (e.g., frequency and severity of adverse effects), and health care team factors (e.g., education from physician about disease linked to treatment information).58 Cost of treatment also has been linked to decreased adherence rates.64,65 Given the emerging focus on improving outcomes while containing health care costs, the need to implement cost-effective strategies to improve treatment adherence is paramount. In addition, the ability to connect to patients outside clinical settings is a compelling approach, given the importance of patient engagement and symptom management in promoting adherence for patients with cancer. For both of these reasons, research into mHealth approaches to manage treatment adherence is desirable. Given that smartphones are becoming ubiquitous, interventions to improve medication adherence through smartphone applications are broadly available.66 Greater than 90% of American adults owned a cell phone in 2015, an increase from 65% in 2005.67 There is a growing body of evidence that even simple interventions, such as text message reminders, improve adherence in a variety of chronic diseases.68 Greater than 80% of cell phone users report sending or receiving text messages.69 Essentially all (99%) texts are opened, and 90% are read within 3 minutes.70 Short message service– and multimedia message service–based texting programs and smartphone applications are being introduced into the health care setting.71-73 Prospective research remains limited, but early studies indicate that digital mHealth interventions can improve patient engagement and adherence to treatment.69,74 Real-time mobile links between patients and providers can relieve logistic burdens of facility-based care, improve symptom tracking, enhance patient compliance, and shift symptom control to the at-home setting.75 However, 150 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

it is also true that the features present in basic smartphone apps vary enormously; therefore, not all apps can be expected to have the same usability or outcomes.76 An evaluation of the features of 272 mobile phone apps purported to promote medication adherence, and readily available in an app store, found that only six of these apps had even half of the desirable features.76 For example, flexible scheduling was the most common feature found across the 272 apps but was still available in only 56.3% of the apps. Password protection, as desired to optimize data privacy, was available for only 13.2% of these apps. Standard approaches to evaluate the quality of apps as required to meet their stated goals are needed to facilitate decision making by patients and providers. In contrast to standard treatments for many chronic diseases, treatments in oncology are often associated with risks for toxicity. This cause of nonadherence may require a different approach than those effective in other therapeutic indications. That is, a text message reminder may not increase treatment adherence in a patient with severe diarrhea who is electing not to take his oral chemotherapy in response to treatment-related symptoms. Instead, approaches that include symptom monitoring of emerging toxicity that prompt the health care team to conduct proactive management would be expected to provide value to improve adherence in oncology. Basch et al77 used an electronic system to routinely collect patient reported outcomes on common symptoms as part of usual care. Patients in this experimental condition continued to receive therapy for an average of 2 months longer, and they experienced increased 1-year survival, compared with a control group treated with usual care. These data provide evidence that symptom assessment and management may help improve outcomes in the context of oncology care. A recent pilot study combined symptom monitoring and adherence assessment in patients with early-stage breast cancer who initiated treatment with aromatase inhibitors (Fig. 2).78 Patients were stratified to receive text and/or email alerts reminding them to complete surveys or to a group that logged onto a website to complete surveys on an ad lib schedule. The group receiving text/email alerts completed 74% of surveys compared with 38% in the ad lib group. Post-study interviews found a high level of acceptance for the mobile surveys; patients stated that they felt that weekly surveys better captured their symptoms compared with waiting for their in-clinic appointment. Additionally, the alert group had nominally better quality of life than the ad lib group. Beyond systemic therapy, patient-facing technology also may improve broader patient acceptance of their complex care journey, including locoregional treatment. Adherence to immediate postoperative care is ripe for mHealth engagement. Surgical recovery is a traumatic part of the overall cancer care continuum and is punctuated by discomfort, disability, and anxiety. For example, the emotional burden of cancer surgery in the head and neck region is heightened by disfigurement and debilitation. Surgeons and allied

USE OF MHEALTH TECHNOLOGIES TO IMPROVE PATIENT CARE IN ONCOLOGY

FIGURE 2. Screen Shots From the Patient Care Monitor (Vector Oncology, Memphis, TN) Used for Mobile Health Adherence Monitoring

providers field drop-in visits to manage minor problems, but these visits distract from their urgent duties. Real-time or asynchronous mobile communication could empower appropriate patient self-care, preempt needless anxiety, and decompress clinic schedules. A pilot study that involved a surgical specialty team was conducted at an academic referral center and used a commercial automated text-based intervention to address the immediate postoperative care engagement needs of patients with head and neck cancer.79 Thirty-two patients were approached, and 23 patients (72%) enrolled. All enrolled patients texted their providers, although frequency (median, seven texts; range, two to 44 texts) varied. Socially isolated patients and those who faced surgical complications used the platform more frequently. Patient satisfaction with the platform was high (mean,3.8 on a four-point Likert scale). Radiation treatment is complex, lengthy (often 30 to 35 daily treatments during 6 to 7 weeks), costly, and toxic. It has been shown that gaps in treatment yield poorer outcomes for patients as a result of accelerated tumor regrowth during breaks in treatment. Compliance is crucial to ensure the best chance for local control and cure; unfortunately, adherence to radiation is a challenge. A review of 564 patients with head and neck cancer who received radiotherapy at a tertiary academic center was conducted to quantify the extent of this problem in a modern patient population covered by a spectrum of private insurance and public

indigent care.80 Three-hundred sixteen patients (56% of all enrolled) suffered a treatment break; 114 missed a single session, 202 missed multiple treatments. Seventy percent of uninsured patients had treatment delays compared with 47% of privately insured patients (p ≤ .0001). Uninsured patients most often missed treatment because of nonmedical/ logistic reasons. Delay was predictive for local recurrence (p = .0002) and overall survival (p < .0001). Among noncompliant patients, there was a higher likelihood for local recurrence in indigent patients. Our results highlight cancer control needs specific to disadvantaged communities at risk for poor radiotherapy adherence. A complex mix of social and human elements—including patient trust in providers, effectiveness of toxicity management, and quality of patient support—create a constellation of determinate factors. Emerging research has shown that mHealth informatics platforms can positively affect health care delivery in indigent cancer populations.74,75 Interestingly, a pilot study published by Percac-Lima et al81 found that telephone navigation directed to at-risk patients significantly improved cancer clinic visit adherence. In summary, early patient-centered studies leveraging mHealth applications to engage patients with cancer about their treatments confirm an exciting beginning. However, these are just the first steps in a longer journey, on which all hype will wear thin quickly. Careful work is needed to refine personalized telehealth tools/utility measures, propel asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 151

PENNELL ET AL

stakeholder enthusiasm, and secure sustainable reimbursem*nt models. Momentum toward mobile consumer self-fulfillment in our modern economy is undeniable. Our patients soon will demand dependable, useful, and thoughtfully designed mobile tools to optimize the acute, recuperative, and long-term survivorship phases of their cancer care. Inclusive multidisciplinary teams will be necessary to keep the cancer care experience relevant to the 21st-century patient, and the changes will require the buy-in and expertise of clinicians, social scientists, computer/data scientists,

product designers, health systems experts, and health care policy makers among others. Additional work must focus on best practices to improve outcomes and balance patient and provider burden. Comparing approaches will be challenging because of the variability in features of apps and tools grouped under the mHealth label. Attention to scientific methodology will be especially important to ensure that potentially cosmetic improvements in patient satisfaction and adherence that we engender with technology actually lead to meaningful downstream clinical outcome improvements.

References 1. Health Resources and Services Administration (HRSA). Telehealth Programs. https://www.hrsa.gov/ruralhealth/telehealth/. Accessed January 21, 2017. 2. AMD Telemedicine. Telemedicine Defined. http://www.amdtelemedicine. com/telemedicine-resources/telemedicine-defined.html. Accessed January 21, 2017. 3. HealthIT.gov. What Types of Telehealth Services Can I Offer? https:// www.healthit.gov/providers-professionals/faqs/what-typestelehealth-services-can-i-offer. Accessed January 24, 2017. 4. American Society of Clinical Oncology. The State of Cancer Care in America: 2016. http://www.asco.org/research-progress/reportsstudies/cancer-care-america-2016#/message-ascos-president. Accessed February 7, 2017. 5. Melton L, Brewer B, Kolva E, et al. Increasing access to care for young adults with cancer: results of a quality-improvement project using a novel telemedicine approach to supportive group psychotherapy. Palliat Support Care. 2016:1-5. 6. Kinney AY, Boonyasiriwat W, Walters ST, et al. Telehealth personalized cancer risk communication to motivate colonoscopy in relatives of patients with colorectal cancer: the family CARE randomized controlled trial. J Clin Oncol. 2014;32:654-662. 7. Schwartz MD, Valdimarsdottir HB, Peshkin BN, et al. Randomized noninferiority trial of telephone versus in-person genetic counseling for hereditary breast and ovarian cancer. J Clin Oncol. 2014;32: 618-626. 8. Cox A, Lucas G, Marcu A, et al. Cancer survivors’ experience with telehealth: a systematic review and thematic synthesis. J Med Internet Res. 2017;19:e11. 9. Nelson R, Staggers N. Health Informatics: An Interprofessional Approach. St. Louis, MO: Elsevier; 2017. 10. Waegemann CP. mHealth: History, Analysis, and Implementation. In Moumtzoglou A. M-Health Innovations for Patient-Centered Care. Hershey, PA: IGI Global, 2016;1-19. 11. Wood WA, Bennett AV, Basch E. Emerging uses of patient generated health data in clinical research. Mol Oncol. 2015;9:1018-1024. 12. Chung AE, Skinner AC, Hasty SE, et al. Tweeting to health: a novel mHealth intervention using Fitbits and Twitter to foster healthy lifestyles. Clin Pediatr (Phila). 2016;56:26-32. 13. The American Telemedicine Association. Practice Guidelines and Resources. http://thesource.americantelemed.org/resources/ telemedicine-practice-guidelines. Accessed February 3, 2017.

152 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

14. Gros DF, Lancaster CL, López CM, et al. Treatment satisfaction of home-based telehealth versus in-person delivery of prolonged exposure for combat-related PTSD in veterans. J Telemed Telecare. 2016;1357633X16671096. Epub 2016 Sep 26. 15. Viers BR, Lightner DJ, Rivera ME, et al. Efficiency, satisfaction, and costs for remote video visits following radical prostatectomy: a randomized controlled trial. Eur Urol. 2015;68:729-735. 16. Kessel KA, Vogel MM, Schmidt-Graf F, et al. Mobile apps in oncology: a survey on health care professionals’ attitude toward telemedicine, mHealth, and oncological apps. J Med Internet Res. 2016;18:e312. 17. Boulos MN, Brewer AC, Karimkhani C, et al. Mobile medical and health apps: state of the art, concerns, regulatory control and certification. Online J Public Health Inform. 2014;5:229. 18. McGillicuddy JW, Weiland AK, Frenzel RM, et al. Patient attitudes toward mobile phone-based health monitoring: questionnaire study among kidney transplant recipients. J Med Internet Res. 2013;15:e6. 19. Nelson R, Joos I, Wolf DM. Social Media for Nurses: Educating Practitioners and Patients in a Networked World. New York, NY: Springer Publishing Company; 2013. 20. Sclafani J, Tirrell TF, Franko OI. Mobile tablet use among academic physicians and trainees. J Med Syst. 2013;37:9903. 21. Ozdalga E, Ozdalga A, Ahuja N. The smartphone in medicine: a review of current and potential use among physicians and students. J Med Internet Res. 2012;14:e128. 22. Yu P, Wu MX, Yu H, et al. The challenges for the adoption of mHealth. Paper presented at: IEEE International Conference on Service Operations and Logistics, and Informatics; June 2006; Shanghai, China. 23. Gagnon MP, Ngangue P, Payne-Gagnon J, et al. m-Health adoption by healthcare professionals: a systematic review. J Am Med Inform Assoc. 2016;23:212-220. 24. Patel M, Dine J, Asch D. Resident use of smartphones while providing patient care. J Gen Intern Med. 2011;26:S103-S104. 25. Thomas L, Capistrant G. State Telemedicine Gaps Analysis: Coverage and Reimbursem*nt. http://www.americantelemed.org/main/policypage/state-telemedicine-gaps-reports. Accessed January 29, 2017. 26. Chee J. Tele-Medical Malpractice: Negligence in the Practice of Telemedicine and Related Issues. http://www.ctel.org/research/ TeleMedical Malpractice Negligence in the Practice of Telemedicine and Related Issues.pdf. Accessed January 29, 2017. 27. Lee P, Stewart D, Calugar-Pop C. Technology, Media, and Telecommunications Predictions 2014. https://www2.deloitte.com/

USE OF MHEALTH TECHNOLOGIES TO IMPROVE PATIENT CARE IN ONCOLOGY

us/en/pages/technology-media-and-telecommunications/articles/ tmt-predictions.html. Accessed January 29, 2017. 28. Institute of Medicine (IOM). The Role of Telehealth in an Evolving Health Care Environment: Workshop Summary. https://www.nap. edu/download/13466. Accessed January 29, 2017. 29. Wood J, Mulrennan S, Hill K, et al. Telehealth clinics increase access to care for adults with cystic fibrosis living in rural and remote Western Australia. J Telemed Telecare. 2016;1357633X16660646. Epub 2016 Jul 20. 30. Sabesan S. Medical models of teleoncology: current status and future directions. Asia Pac J Clin Oncol. 2014;10:200-204.

47. Wu RC, Morra D, Quan S, et al. The use of smartphones for clinical communication on internal medicine wards. J Hosp Med. 2010;5:553559. 48. Quan SD, Wu RC, Rossos PG, et al. It’s not about pager replacement: an in-depth look at the interprofessional nature of communication in healthcare. J Hosp Med. 2013;8:137-143. 49. Nguyen C, McElroy LM, Abecassis MM, et al. The use of technology for urgent clinician to clinician communications: a systematic review of the literature. Int J Med Inform. 2015;84:101-110.

31. Thomas Jefferson University. Strength Through Insight. http://www. jefferson.edu/strength-through-insight.html. Accessed February 7, 2017.

50. Siegal G, Dagan E, Wolf M, et al. Medical information exchange: pattern of global mobile messenger usage among otolaryngologists. Otolaryngol Head Neck Surg. 2016;155:753-757.

32. Fasola G, Macerelli M, Follador A, et al. Health information technology in oncology practice: a literature review. Cancer Inform. 2014;13:131139.

51. Khanna V, Sambandam SN, Gul A, et al. “WhatsApp”ening in orthopedic care: a concise report from a 300-bedded tertiary care teaching center. Eur J Orthop Surg Traumatol. 2015;25:821-826.

33. Homsted L. Institute of Medicine report: to err is human—building a safer health care system. Fla Nurse. 2000;48:6.

52. Johnston MJ, King D, Arora S, et al. Smartphones let surgeons know WhatsApp: an analysis of communication in emergency surgical teams. Am J Surg. 2015;209:45-51.

34. Mennemeyer ST, Menachemi N, Rahurkar S, et al. Impact of the HITECH Act on physicians’ adoption of electronic health records. J Am Med Inform Assoc. 2016;23:375-379. 35. Thomas CA, Ward JC. The oncology care model: a critique. Am Soc Clin Oncol Educ Book. 2016;35:e109-e114. 36. Gellert GA, Ramirez R, Webster SL. The rise of the medical scribe industry: implications for the advancement of electronic health records. JAMA. 2015;313:1315-1316. 37. Campbell LL, Case D, Crocker JE, et al. Using medical scribes in a physician practice. J AHIMA. 2012;83:64-69. 38. Brady K, Shariff A. Virtual medical scribes: making electronic medical records work for you. J Med Pract Manage. 2013;29:133-136. 39. Hein I. Electronic Record Keeping With Google Glass and Helpers. http://www.medscape.com/viewarticle/874206. Accessed February 4, 2017. 40. Stübig T, Zeckey C, Min W, et al. Effects of a WLAN-based real time location system on outpatient contentment in a level I trauma center. Int J Med Inform. 2014;83:19-26. 41. Kamel Boulos MN, Berry G. Real-time locating systems (RTLS) in healthcare: a condensed primer. Int J Health Geogr. 2012;11:25. 42. Dobson I, Doan Q, Hung G. A systematic review of patient tracking systems for use in the pediatric emergency department. J Emerg Med. 2013;44:242-248. 43. Schulmeister L. Technology and the transformation of oncology care. Semin Oncol Nurs. 2016;32:99-109. 44. Versus. Versus Technology, Inc, Announces Collaboration with Cleveland Clinic to Create Clinical Patient Flow Model. http://www. versustech.com/rtls-news/press-releases/versus-technology-incannounces-collaboration-with-cleveland-clinic-to-create-clinicalpatient-flow-model/. Accessed February 4, 2017. 45. Versus. Pediatric Clinics Expedite Visits with RTLS Technology. http:// www.versustech.com/rtls-news/press-releases/pediatric-clinicspatient-flow-rtls/. Accessed February 4, 2017. 46. Soto RG, Chu LF, Goldman JM, et al. Communication in critical care environments: mobile telephones improve patient care. Anesth Analg. 2006;102:535-541.

53. Osarogiagbon RU, Rodriguez HP, Hicks D, et al. Deploying team science principles to optimize interdisciplinary lung cancer care delivery: avoiding the long and winding road to optimal care. J Oncol Pract. 2016;12:983-991. 54. Watson L, Pathiraja F, Depala A, et al. Ensuring safe communication in health care: a response to Johnston et al on their paper “Smartphones let surgeons know WhatsApp: an analysis of communication in emergency surgical teams.” Am J Surg. 2016;211:302-303. 55. DiMatteo MR. Variations in patients’ adherence to medical recommendations: a quantitative review of 50 years of research. Med Care. 2004;42:200-209. 56. Osterberg L, Blaschke T. Adherence to medication. N Engl J Med. 2005;353:487-497. 57. Hohneker J, Shah-Mehta S, Brandt PS. Perspectives on adherence and persistence with oral medications for cancer treatment. J Oncol Pract. 2011;7:65-67. 58. Gater A, Heron L, Abetz-Webb L, et al. Adherence to oral tyrosine kinase inhibitor therapies in chronic myeloid leukemia. Leuk Res. 2012;36:817-825. 59. Makubate B, Donnan PT, Dewar JA, et al. Cohort study of adherence to adjuvant endocrine therapy, breast cancer recurrence and mortality. Br J Cancer. 2013;108:1515-1524. 60. Kribbs NB, Pack AI, Kline LR, et al. Objective measurement of patterns of nasal CPAP use by patients with obstructive sleep apnea. Am Rev Respir Dis. 1993;147:887-895. 61. Ganesan P, Sagar TG, Dubashi B, et al. Nonadherence to imatinib adversely affects event free survival in chronic phase chronic myeloid leukemia. Am J Hematol. 2011;86:471-474. 62. Bonadonna G, Valagussa P, Moliterni A, et al. Adjuvant cyclophosphamide, methotrexate, and fluorouracil in node-positive breast cancer: the results of 20 years of follow-up. N Engl J Med. 1995;332:901-906. 63. Wu EQ, Johnson S, Beaulieu N, et al. Healthcare resource utilization and costs associated with non-adherence to imatinib treatment in chronic myeloid leukemia patients. Curr Med Res Opin. 2010;26: 61-69.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 153

PENNELL ET AL

64. Streeter SB, Schwartzberg L, Husain N, et al. Patient and plan characteristics affecting abandonment of oral oncolytic prescriptions. J Oncol Pract. 2011; 7:46s-51s.

73. Stinson JN, Jibb LA, Nguyen C, et al. Development and testing of a multidimensional iPhone pain assessment application for adolescents with cancer. J Med Internet Res. 2013;15:e51.

65. Farias AJ, Du XL. Association between out-of-pocket costs, race/ ethnicity, and adjuvant endocrine therapy adherence among Medicare patients with breast cancer. J Clin Oncol. 2017;35:86-95.

74. Kannisto KA, Koivunen MH, Välimäki MA. Use of mobile phone text message reminders in health care services: a narrative literature review. J Med Internet Res. 2014;16:e222.

66. Bailey SC, Belter LT, Pandit AU, et al. The availability, functionality, and quality of mobile applications supporting medication selfmanagement. J Am Med Inform Assoc. 2014;21:542-546.

75. Jibb LA, Stevens BJ, Nathan PC, et al. A smartphone-based pain management app for adolescents with cancer: establishing system requirements and a pain care algorithm based on literature review, interviews, and consensus. JMIR Res Protoc. 2014;3:e15.

67. Anderson M. Technology Device Ownership: 2015. http://www. pewinternet.org/2015/10/29/technology-device-ownership-2015. Accessed January 25, 2017. 68. Thakkar J, Kurup R, Laba TL, et al. Mobile telephone text messaging for medication adherence in chronic disease: a meta-analysis. JAMA Intern Med. 2016;176:340-349. 69. Duggan M. Cell Phone Activities 2013. http://www.pewinternet. org/2013/09/19/cell-phone-activities-2013/. Accessed January 25, 2017. 70. Johnson D. SMS Open Rates Exceed 99%. http://www.tatango.com/ blog/sms-open-rates-exceed-99/. Accessed January 25, 2017. 71. Hall AK, Cole-Lewis H, Bernhardt JM. Mobile text messaging for health: a systematic review of reviews. Annu Rev Public Health. 2015;36: 393-415. 72. Semple JL, Sharpe S, Murnaghan ML, et al. Using a mobile app for monitoring post-operative quality of recovery of patients at home: a feasibility study. JMIR Mhealth Uhealth. 2015;3:e18.

154 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

76. Santo K, Richtering SS, Chalmers J, et al. Mobile phone apps to improve medication adherence: a systematic stepwise process to identify highquality apps. JMIR Mhealth Uhealth. 2016;4:e132. 77. Basch E, Deal AM, Kris MG, et al. Symptom monitoring with patientreported outcomes during routine cancer treatment: a randomized controlled trial. J Clin Oncol. 2016;34:557-565. 78. Graetz I, McKillop CM, Stepanski E, et al. Use of a web-based app to improve breast cancer symptom management and aromatase inhibitor adherence. J Clin Oncol. 2017;35 (suppl 5S; abstr 89). 79. Sosa A, Heineman N, Thomas K, et al. Improving patient health engagement with mobile texting: a pilot study in the H&N postoperative setting. Head Neck. In press. 80. Thomas K, Martin T, Gao A, et al. Interruptions of head & neck radiotherapy across insured and indigent patient populations. J Oncol Pract. In press. 81. Percac-Lima S, López L, Ashburner JM, et al. The longitudinal impact of patient navigation on equity in colorectal cancer screening in a large primary care network. Cancer. 2014;120:2025-2031.

CLINICAL PATHWAYS IN ONCOLOGY CARE

Perspectives on the Use of Clinical Pathways in Oncology Care Anne C. Chiang, MD, PhD, Peter Ellis, MD, and Robin Zon, MD OVERVIEW Pathways and guidelines are valuable tools to provide evidence-based care in oncology. Pathways may be more restrictive than guidelines because they attempt (where possible) to reduce cost, add efficiency, and remove unwarranted variability. Pathways offer an opportunity to measure, report, and improve quality of care; they can drive to evidence-based targeted therapy where appropriate; they can enhance efficiency through standardization; and, finally, they can be a vehicle to enhance participation in clinical trials. Pathway implementation requires understanding and commitment on the part of the physician and leadership as they may initially disrupt workflow, but ultimately have the ability to enhance patient care. ASCO criteria have been published for the development and implementation of high-quality oncology pathway programs. Future challenges for pathways include incorporation of molecular testing and appropriate targeted care in a real-time precision oncology approach.

athways have been used for more than a decade in the oncology space, with the aims of improving patient care and communication while focusing on outcomes and maximizing resource utilization by reducing unwarranted variation and promoting the use of higher value therapy. However, the magnitude and growing importance of pathways is evident in the role of reimbursem*nt, already implemented by some payers, and pathway utilization expansion, such as in the Oncology Care Model. Additionally, with the statutory implementation of the Medicare Access and CHIP Reauthorization Act of 2015 (MACRA), the U.S. health care system of reimbursem*nt is transitioning away from a volumeincentivized, provider-centric model to a value-based, patientcentered model. Well-designed pathways have the potential to help us adapt to this transformation by serving as a foundation for comprehensive patient care, while promoting efficient, higher quality care. This in turn can potentially control costs and better position practices to assume financial risk for our patient populations, while assuring best care for the patient. Pathways could possibly serve as a central component of oncology practice and as a cornerstone of future payment methodologies.

DEFINITION: GUIDELINES VERSUS PATHWAYS

The terms “guidelines” and “pathways” are used frequently in discussions regarding quality and cost-effective oncology care. A guideline is a listing of all treatments that are considered (by a panel of experts) to be within a “standard of

care” for a given presentation of disease. Practice guidelines assist practitioner and patient decisions about appropriate care, defined by empirical evidence, addressing specific clinical circ*mstances and aligning practice with state of the art oncology. Guidelines do not formally address cost and resource utilization. The primary aim for guidelines is not standardization, but rather to ensure that care delivered has been demonstrated to be effective by evidential review. In a Venn diagram, pathway treatment recommendations would be included within the scope of the guidelines, but with a much smaller numerical set. An exception to this might be when a pathway is updated more frequently than a guideline, incorporating new evidence. Pathways are also known as care pathways, critical pathways, care maps, or integrated care pathways. A pathway references the same literature sources, but attempts to choose a single therapy from the acceptable options that is best for a given presentation of disease. This process involves committees of physician peers with disease expertise and uses consensus to determine the best option based on a platform of efficacy, toxicity, and cost. Characteristics of a pathway program include the pathway serving as a multidisciplinary management tool, based on high-level evidence applicable to a specific group of people, wherein interventions are defined, optimized, and sequenced. Generally, pathways will support the Triple Aim of health care, which is better perception of care by the patient, improved professional effectiveness and coordination of care, while

From the Yale University School of Medicine, New Haven, CT; University of Pittsburgh School of Medicine, Pittsburgh, PA; Michiana Hematology Oncology, South Bend, IN. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Anne Chiang, MD, PhD, Yale University School of Medicine, Smilow Cancer Hospital, 20 York St., NP-304, New Haven, CT 06520; email: anne.chiang@yale. edu. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 155

CHIANG, ELLIS, AND ZON

controlling/reducing costs. Ideally, pathways should be available at the point of care for a patient for real-time clinical decision support. The pathway system should also be able to document the decision making of the clinician for future study, reporting, and compliance.

RATIONALE FOR USE OF PATHWAY PROGRAMS

Whereas access to the ASCO and the National Comprehensive Cancer Network guidelines is open, pathway systems are currently only commercially available and thus require substantial commitments of time and resources for practice implementation. However, there are multiple compelling reasons to support pathway use. First, the status quo of relying upon the individual expertise of any one physician to decide appropriate care is no longer acceptable in our current age of accountability. There is now a need to prove the quality of care to stakeholders (e.g., patients, referring providers, and payers) rather than the “trust me” mantra of old. Measuring care is an indispensable part of care delivery, but, unfortunately, our current electronic platforms for documenting care are not up to the task as they cannot have the innate flexibility needed to react to the rapidly changing science of oncology. Pathway systems can provide such data demanded by new legislation, such as MACRA and Merit-based Incentive Payment System. Second, the costs of all medical care are rising faster than inflation with oncology care contributing significantly to this increase. By standardizing care—where feasible and appropriate—to best evidenced-based care, pathways programs offer the possibility of driving down costs, increasing quality, and improving outcomes. Standardization also offers the potential for improved patient flow, error reduction, and cost savings without compromising patient access or care. Limiting “unwarranted variability” is in everyone’s best interest. Finally, as oncology care becomes more personalized, the options for care more numerous, and the volume of infor-

KEY POINTS • Pathways are a subset of guidelines designed to standardize care with best evidence and reduce unwarranted variability and cost. • Pathways offer the hope of improved quality by standardization around best evidence, enhancing clinical trial availability, offering decision support at the point of care, and providing measurable outcomes. • Implementation of pathways requires significant understanding and commitment of the provider institution, but has the promise of being an important quality and cost tool. • Pathways may be influenced by the viewpoint of the developer. ASCO members are often most comfortable with the implementation of provider-based pathways in clinical care. • Pathways may enhance the use of appropriate targeted (personalized) care. 156 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

mation that informs decision making ever more complex, it becomes increasingly difficult for a single provider to remember the nuanced details of treatment of every state and stage of disease presentation. At the same time, restrictions in provider availability will likely tighten as the baby boomers continue to age and patients live longer and better lives with their cancer diagnoses. The need for a point-ofcare decision support tool is increasingly becoming self-evident. For institutions committed to advancing the knowledge-base of cancer treatment through research, a pathway can simplify and potentially enhance accrual of patients by being able to incorporate open clinical trials into the treatment algorithm.

THE ROAD MAP TO SUCCESSFUL PATHWAY IMPLEMENTATION

After the decision has been made to proceed with the adoption of clinical pathways, there are several key aspects to a successful launch. The initial “big picture” buy-in of all of the clinicians in the institution is critical; they must understand and agree that a provider-based solution is preferable to a payer- or government-based solution both for patient care and provider satisfaction. The big picture will necessarily look different to an academic physician compared with a clinical physician, but can be equally compelling. Once buy-in is obtained, the system chosen for use should be integrated into the normal physician workflow as seamlessly as is possible. Clinician expectations have to be managed, stressing that all efforts are being made to minimize disruption in the clinic, but that it is impossible to add a variable without some alteration in flow. Adjustments to workflow or the product should not be based upon a perception of what it will be like, but rather upon the experience of hands-on usage and subsequent thoughtful adjustment to obtain the goal of quality patient care and good workflow. This iterative process is possibly the single most important aspect of implementation. Each clinic may have its own uniqueness and character demanding slight variations in rollout that are often easily accommodated. After initial implementation, giving feedback to the physician and monitoring that feedback are vital to success. Clinician feedback regarding the advantages of a decision support tool and data collection are helpful in maintaining compliant use of the system. For example, the ability to provide data on eligible patients throughout a network to academic physicians for screening or clinical trials development is an immediate positive byproduct. Facilitating workflow by offering all the information necessary for patient care in one location is appreciated by busy clinicians trying to get through a busy clinic. Saving time to look up side effects or dose reductions allows for more time spent with each patient. Changes in workflow are always difficult to manage, especially when physicians already have their own established care preferences. Attempting change in a stepwise fashion is preferable to allow time for adoption. The content of a pathway may differ based upon the principles and priorities of the pathway developer. A transparent

CLINICAL PATHWAYS IN ONCOLOGY CARE

SIDEBAR 1. ASCO Recommendations to Improve the Development and Use of Clinical Pathways in Oncology 1. P ursue a collaborative, national approach to reduce the unsustainable administrative burdens associated with the unmanaged proliferation of oncology pathways. 2. Adopt a process for development of oncology pathways that is consistent and transparent to all stakeholders. 3. E nsure that pathways address the full spectrum of cancer care, from diagnostic evaluation through medical, surgical, and radiation treatments, and include imaging, laboratory testing, survivorship, and end-of-life care. 4. U pdate pathways continuously to reflect new scientific knowledge and insights gained from clinical experience and patient outcomes, to promote the best possible evidence-based care. 5. R ecognize patient variability and autonomy and allow physicians to easily diverge from pathways when evidence and patient needs dictate. 6. Implement oncology pathways in ways that promote administrative efficiencies for both oncology providers and payers. 7. Promote education, research, and access to clinical trials in oncology clinical pathways. 8. D evelop robust criteria to support certification of oncology pathway programs; pathway programs should be required to qualify based on these criteria, and payers should accept all oncology pathway programs that achieve certification through such a process. 9. Support research to understand the effect of pathways on care and outcomes. and evidence-based development process of pathways is vital to assure the confidence of those affected by the pathway: the provider, patient, and payer. Many practices have found that implementation of a well-designed pathway program is both doable and advantageous in the delivery and measurement of quality oncology care.

THE ASCO PERSPECTIVE OF CLINICAL PATHWAYS

On January 12, 2016, the ASCO Policy Statement on Clinical Pathways on Oncology1 was released with recommendations to ensure that clinical pathways in oncology enhance—and not diminish—patient care (Sidebar 1). The intent of the statement was to elevate awareness about clinical pathways in oncology and to convey a cautionary note that no current mechanism exists to ensure the integrity, efficient implementation, and outcome assessments for these treatment management tools. The release came within 1 year of the establishment of the Task Force on Clinical Pathways, which was charged with better understanding the concerns and barriers to providing high-quality, evidence based care, as articulated by ASCO members and other stakeholders. Specifically, ASCO’s State Affiliate Council and Clinical Practice Committee cited a number of concerns, including: lack of transparency in disclosing conflict of interest and disclosure describing the methodology used in

development, a focus on cost savings with efficacy and safety as secondary considerations, a cumbersome appeals process and lack of reimbursem*nt for off-pathway treatment, lack of pathways for rare and in-patient–treated cancers, implementation concerns, as well as lack of publicized analytics supporting pathway utilization. Members especially highlighted the associated unsustainable administrative burdens related to the multitude of oncology pathways some providers are required to track and manage, as well as the continued requirement of pre-authorization when pathway compliant. The Clinical Pathways Task Force has continued its efforts to ensure that pathways are consistently developed and transparent to all stakeholders, and used in the way they are intended—to guarantee quality care while helping to reduce unwanted variations in care and controlling costs. Recently, the Task Force developed and published ASCO criteria for the development and implementation of high-quality oncology pathway programs while actively seeking and respecting input—through direct stakeholder meetings and interviews—with patient advocates, payers, vendors, and providers (Table 1).2 The criteria focus on three key areas: development, implementation/use, and analysis and are intended for use by multiple stakeholders to evaluate clinical pathway programs and guide their future development.

TABLE 1. ASCO Criteria for High-Value Pathways Development

Implementation and Use

Analytics

Expert driven and reflects stakeholder input

Clear and achievable expected outcomes

Efficient and public reporting of performance metrics

Transparent, evidence-based, patient-focused, clinically driven, and up-to-date

Integrated, cost-effective technology, and decision support

Outcomes-driven incentives

Comprehensive and promotes participation in clinical trials

Efficient processes for communication and adjudication

Promotion of research in value and effect on pathways and care transformation

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 157

CHIANG, ELLIS, AND ZON

BENEFITS OF USING ASCO CLINICAL PATHWAY CRITERIA FOR PRACTICING ONCOLOGISTS

Regardless of the site of service, ASCO criteria are intended to be used in a manner that enhances the ability of the provider to evaluate pathway programs for their practice. Because providers are becoming increasingly focused on optimizing efficiencies, including reducing costs while preserving high quality care for the patient, the criteria may help in assessing programs that may best attain practice management goals. Furthermore, as various stakeholders collaborate to improve the delivery of care, pathway utilization may intensify. Cancer care is generally a multidisciplinary effort, and pathways, if comprehensively developed as proposed in the criteria, can be widely used by the caregiver team to optimally coordinate patient care. Reducing redundancy and unnecessary testing while assuring the necessary evaluations and treatments are delivered would be paramount to the success of a pathway program in this collaboration. On a broader scale, there are differences between payerand provider-facing pathways. Collaboration between payers and providers regarding pathway utilization and criteria compliance may offer opportunity to help minimize some of the administrative issues for both stakeholder groups, be leveraged as a measure to control costs, and inform reimbursem*nt discussions.

EVOLUTION OF CLINICAL PATHWAYS: CHALLENGES OF PRECISION ONCOLOGY

The ASCO criteria promote a much needed benefit of continually updated comprehensive pathways to help with the management of rapidly developing clinical advances. All stakeholder groups, including payers and patients, are quick to point out variation of care and resource utilization between providers. As pathways integrate scientific advances, including precision medicine and rapid learning system–validated evidence, there should be opportunity to maximize resource alignment and promote value. In addition, informed pathways will assist the provider in delivering appropriate, equitable care for all patients. To serve as example, pathway programs can potentially assist providers in ensuring the growing complexity of molecular testing is used optimally so patients can receive targeted and other personalized care to achieve best outcomes. Currently, the utilization of U.S. Food and Drug Administration–approved and National Comprehensive Cancer Network guideline-approved tests, such as EGFR testing in patients with newly diagnosed advanced lung cancer, is not known. One study estimates 18% of patients with newly diagnosed lung cancer undergo testing within 6 months of diagnosis, with 37% patients with presumed nonsquamous histology (receiving bevacizumab or pemetrexed) undergoing EFGR testing.3 Certainly, pathway use can help to ensure and track the use of appropriate molecular tests. Precision oncology now refers to the use of molecular testing to identify mutation-based treatment options such as vemurafenib and dabrafenib for patients with BRAFV600 158 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

mutations or erlotinib for EGFR mutations. With the increasing uptake of genomic testing, in part enabled by reduced cost, many clinicians and patients are faced with trying to understand the implications of molecular testing results on treatment decisions. For example, should patients with various disease types who have BRAFV600 mutations receive targeted therapy, even if the drugs have not yet been tested or approved for those specific disease-types? Two meta-analyses have shown that trials utilizing molecularlybased treatments led to better outcomes compared with those with nontargeted therapies.4,5 However, the prospectively randomized French SHIVA trial did not show benefit in progression-free survival for patients who were treated with molecularly targeted agents versus investigator’s choice. In this trial, patients with any metastatic solid tumor that had progressed on standard treatments underwent mandatory tumor biopsy to obtain tissue for genomic testing. Patients who had molecular alterations that mapped to the PI3K/ AKT/mTOR, or RAF/MEK, or hormone receptor pathways were randomly selected to receive standard of care versus molecularly targeted agents including erlotinib, lapatanib plus trastuzumab, sorafenib, imatinib, dasatinib, vemurafenib, everolimus, abiraterone, letrozole and tamoxifen. Of 741 patients screened, 293 (40%) had at least one molecular mutation identified; 92 patients were randomly selected to receive the control standard-of-care arm and 99 patients to the experimental molecularly targeted arm. The median progression-free survival was 2.0 months compared to 2.3 months (HR 0.88; p = 0.41) in the control versus experimental arms, respectively.6 There are several reasons for the lack of benefit seen in patients receiving targeted therapy in the SHIVA trial. Mutations in driver genes identified by sequencing may be silent passenger mutations or mutations that are resistant to therapies, such as EGFR exon 20 insertions that are not associated with response to EGFR inhibitors. Alternatively, some cancers may require multiple hits in different signaling pathways or involve epigenetic or post-transcriptional control. Finally, discrepancies between commercially available tests or discordance between primary, metastatic, or just heterogeneous tumors may complicate the use of molecular alterations for treatment. Although there is no doubt that some molecularly based treatments based on single mutations (e.g., erlotinib and crizotinib in lung cancer) can improve patient outcomes and should be incorporated into clinical pathways, much work remains to be done before large-scale genomic testing and subsequent targeted therapies becomes standard of care.

VALUE-BASED CARE AND PATHWAYS: FUTURE CHALLENGES

Several current initiatives aim to improve quality of care and care delivery, including the development of learning health care systems and value-based frameworks. These initiatives, along with pathways, are influenced by the different perspectives and needs of the stakeholders. For example, the definition of value may differ between health systems,

CLINICAL PATHWAYS IN ONCOLOGY CARE

payers, patients, oncologists, and manufacturers. Similarly, the perspectives of payer- and provider-facing pathways may differ. Because there is limited published data regarding pathway performance, escalation of widespread pathway program analysis as it pertains to patient outcomes and the financial aspects surrounding cost and value, is needed. Additionally, learning health care systems have challenges pertaining to data sharing, interoperability of electronic

health care systems, and patient-reported outcomes. These initiatives, although currently being developed independently, have the potential to enhance patient-centered care as an integrated strategy while still achieving the goals of quality, value, and cost control. As pathway programs continue to evolve, improvements in patient care will be achieved as value and learning health system learnings are integrated as essential components.

References 1. Zon RT, Frame JN, Neuss MN, et al. American Society of Clinical Oncology policy statement on clinical pathways in oncology. J Oncol Pract. 2016;12:261-266.

4. Schwaederle M, Zhao M, Lee JJ, et al. Impact of precision medicine in diverse cancers: a meta-analysis of phase II clinical trials. J Clin Oncol. 2015;33:3817-3825.

2. Zon RT, Edge SB, Page RD, et al. American Society of Clinical Oncology criteria for high-quality clinical pathways in oncology. J Oncol Pract. Epub 2017 Feb 7.

5. Jardim DL, Schwaederle M, Wei C, et al. Impact of a biomarker-based strategy on oncology drug development: a meta-analysis of clinical trials leading to FDA approval. J Natl Cancer Inst. 2015;107:djv253. (Erratum in: J Natl Cancer Inst. 2016;108:djv423).

3. Shen C, Kehl KL, Zhao B, et al. Utilization patterns and trends in epidermal growth factor receptor (EGFR) mutation testing among patients with newly diagnosed metastatic lung cancer. Clin Lung Cancer. Epub 2016 Nov 10.

6. Le Tourneau C, Delord JP, Gonçalves A, et al; SHIVA investigators. Molecularly targeted therapy based on tumour molecular profiling versus conventional therapy for advanced cancer (SHIVA): a multicentre, open-label, proof-of-concept, randomised, controlled phase 2 trial. Lancet Oncol. 2015;16:1324-1334.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 159

SCHWARTZBERG ET AL

Precision Oncology: Who, How, What, When, and When Not? Lee Schwartzberg, MD, Edward S. Kim, MD, David Liu, MD, MPH, and Deborah Schrag, MD, FASCO OVERVIEW Precision oncology, defined as molecular profiling of tumors to identify targetable alterations, is rapidly developing and has entered the mainstream of clinical practice. Genomic testing involves many stakeholders working in a coordinated fashion to deliver high-quality tissue samples to high-quality laboratories, where appropriate next-generation sequencing (NGS) molecular analysis leads to actionable results. Clinicians should be familiar with the types of genomic variants reported by the laboratory and the technology used to determine the results, including limitations of current testing methodologies and reports. Interpretation of genomic results is best undertaken with multidisciplinary input to reduce uncertainty in clinical recommendations relating to a documented variant. Non–small cell lung cancer has emerged as a prototype disease where genomic data from at least several well-documented alterations with approved targeted agents are essential for optimal treatment from diagnosis of advanced disease. Due to the development of resistance to targeted therapies, resampling and retesting of tumors, including using liquid biopsy technology after clinical progression, may be important in making treatment decisions. The value of molecular profiling depends on avoiding both underutilization for well-documented variant target-drug pairs and overutilization of variant-drug therapy without proven benefit. As techniques evolve and become more cost effective, the use of molecular testing may prove to add more specificity and improve outcomes for a larger number of patients.

he goal of precision medicine is simply to deliver the right cancer treatment to the right patient at the right dose and the right time. Several lines of investigation came together nearly simultaneously to usher in the beginning of the precision oncology era. In 1998, the BCR-ABL rearrangement in chronic myeloid leukemia was successfully targeted by the small molecule imatinib, leading to dramatic clinical remissions and U.S. Food and Drug Administration approval in 2001. The first draft sequence of the human genome was accomplished the same year,1 followed by the first cancer genome.2 Rapid discovery of multiple, nonoverlapping driver mutations and tyrosine kinase inhibitors with clinically effective inhibitory properties in non–small cell lung cancer and melanoma led to assays of alterations performed by polymerase chain reaction (PCR) quickly and inexpensively. Use of these biomarkers to drive treatment decisions in solid tumors raised expectations and interest in molecular profiling. Sequencing technology and costs improved rapidly during the early 2000s, particularly with the advent of NGS on formalin-fixed, paraffin-embedded tissue whereby massive parallel sequencing allows determination of alterations in a large number of genes through a timely, cost-effective process. Underpinning precision oncology is the concept of somatic mutations as the foundation of cancer development.3

Mutations in oncogenes rendering them constitutively active are considered driver mutations and are central control points for progression of malignancies. Conversely, tumor suppressor genes, involved naturally in controlling tumor pathogenesis, can cause cancer progression when inactivated through mutation or allele loss. Multiple processes result in dysregulation of the genetic machinery in DNA RNA or protein, leading to altered expression of the protein coded for by the gene. To capture the entire spectrum of potential alterations, multiple technologies, termed a multi- or pan-omic approach, are best considered. The vast number of choices of technologies, commercial entities offering testing, and sometimes conflicting results have overwhelmed clinicians looking to obtain molecular information that will result in clinical utility for their patients. Even in academic centers, oncologists report varying confidence in their ability to use the genomic findings appropriately.4 At its most fundamental level, a genomic test with clinical utility should be predictive of a treatment response from a targeted agent. An early example in solid tumor oncology was the ability to test for HER2 positivity as defined as fluorescent in situ hybridization–based gene amplification or immunohistochemistry to demonstrate overexpression of the protein. Positive results predicted response to

From the University of Tennessee Health Science Center, Memphis, TN; Levine Cancer Institute, Charlotte, SC; Dana-Farber Cancer Institute, Boston, MA. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Lee Schwartzberg, MD, West Cancer Center, 7945 Wolf River Blvd., Germantown, TN 38138; email: [emailprotected]. © 2017 American Society of Clinical Oncology

160 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PRECISION ONCOLOGY: WHO, HOW, WHAT, WHEN, AND WHEN NOT

trastuzumab-based therapies, whereas HER2-negative tumors did not derive benefit from this approach. As we have moved into multiplex testing of many genes or other biologic species, including messenger RNA and proteins, the same criteria should apply—is the variant alteration sufficiently predictive of response to a paired agent? To date, success in using precision approaches to treatment have been mixed. A prospective phase II study of molecular profiling to assign matched therapy did not show superior outcomes for the matched group but suffered from serious methodologic design issues.5 Large retrospective series have documented that 80%–90% of patients tested will have potentially actionable genomic alterations, although the definition of actionable can vary substantially.6-9 However, only a minority of patients to date actually receive genomically directed therapy, usually on a clinical trial.

TECHNICAL ASPECTS OF NGS TESTING FOR THE CLINICIAN

Types of Alterations Detected

A range of genomic somatic variants can be ascertained with NGS, including single nucleotide variants (SNVs), also known as point mutations, and small insertions or deletions of bases (indels), which can lead to a nonfunctional or absent protein. Additionally, copy number variants, which reflect amplifications and deletions of genes and/or larger portions of a chromosome, gene rearrangements and fusion genes can be detected.

Read Depth and Coverage

This criterion refers to the number of times a particular base position in the DNA is read during the NGS analysis. The greater the coverage of a particular alteration, the more likely it is to be detected, which is especially important in tumor samples with low tumor content. By covering the same area of the gene fragment multiple times, the likelihood of picking up a variation of low allelic frequency is enhanced.

KEY POINTS • The goal of precision oncology has begun to be realized through multiplex molecular testing including NGS. • Oncologists should be familiar with technical aspects of NGS to facilitate selecting the most appropriate and costeffective testing platform. • Considerations for molecular testing include which tissue type to utilize, timing of profiling in the disease course, extent of panel to order, and degree of clinical annotation reported. • Actionable biomarkers of non–small cell lung cancer make this disease a paradigm for precision oncology at diagnosis of advanced disease, during therapy, and at time of progression. • Interpretation of molecular data to facilitate best practice remains a challenge; clinical trial participation and sharing of linked molecular/clinical data sets are strongly encouraged.

For hot spot testing, coverage of at least 100–300X is recommended.

Breadth and Scope of Testing

How many genes are included and what areas of the gene are analyzed. The most frequent NGS offerings today are hot spot testing, where alterations in exons or intron/exon junction areas of a preselected panel of cancer genes, including known activating oncogenes and tumor suppressor genes, are analyzed. Targeted hot spot panels focus on the best-annotated cancer genes, typically 35 to 350 genes, and provide high depth of coverage. The greater depth allows for assessing lower allele frequency and can account for intratumoral heterogeneity and low allele frequency of the alteration. NGS panels are not ideal for large-scale rearrangements and/or deletions and certain fusion genes. Addition of RNA sequencing can help identify these alterations. Recently, whole-exome sequencing and whole-transcriptome sequencing have become available at academic and some commercial laboratories. At the moment, the value of whole-exome sequencing information is largely confined to the translational research space, where it offers enormous potential to produce novel variant-pathogenic associations leading to clinical trials investigating new agents. Lengthy turnaround time and lack of clinical associations for the large majority of genomic alterations preclude current effective clinical use.

Variant Calling

The bioinformatics approach to lining up the vast amount of information obtained in an NGS sequence, and accurately calling variants, is important to achieve quality results. Variant quality scores are generated for each test within a laboratory. Technical validity results can be provided to the practitioner upon request and may be useful in determining which assay to use due to interlaboratory differences. Federal guidelines for technical validity do not currently exist for NGS tests, which are classified as laboratory developed tests.

Variant Allele Frequency

This reflects the percentage of reads identifying a variant divided by the overall coverage of that locus. If tumor cells represent 100% of the sample DNA analyzed, heterozygous loci such as seen in germline mutations should be near 50% variant allele frequency, hom*ozygous loci should be near 100% and reference loci should be near zero. In actual practice, the contamination from normal cells, local copy number alterations and tumor heterogeneity often yield unpredictable variant allele frequency.

Variant Meaning

Particularly for single nucleotide variants, it is not always easy to determine if a mutation is pathogenic or not. Publicly available databases, such as the Catalogue of Somatic Mutations in Cancer (COSMIC),10 and the laboratory’s internal databases, are reviewed by the evaluating pathologist, and a asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 161

SCHWARTZBERG ET AL

determination is made whether the alteration is pathogenic, probably pathogenic, probably benign (meaning that it is likely a single nucleotide polymorphism without functional significance), benign representing a known single nucleotide polymorphism or a variant of unknown significance. As panel testing grows larger, the reporting of a variant of unknown significance has grown dramatically. It is hoped that in the future, sharing of genomic data will settle the issue for the growing number of alterations without a clinical correlate to call pathogenic or not. At the moment, utilizing a laboratory with deep molecular expertise, including molecular pathologists and geneticists on staff to help make the call, is extremely important. Alternatively, third-party organizations like N-of-1 use extensive resource capabilities to perform this function for laboratories or health care systems. Their role is to generate content relevant to the spectrum of variants so that clinically appropriate decisions can be made.

Tumor Only Versus Tumor Normal

When a tumor alone is tested, variants are compared with databases such as COSMIC and ClinVar11 to determine whether the variant is a known pathogenic variant or a known single nucleotide polymorphism. Simultaneously sequencing tumor and normal tissue allows more precise calling of somatic mutations. Moreover, germline cancer predisposition genes can be clearly distinguished from somatic mutations in the same genes. As bioinformatics improves, value from the additional cost and complexity of sequencing both tumor and normal tissue routinely appears to be diminishing.12 Though advances in bioinformatic techniques and reference germline databases are improving the accuracy of tumor-only sequencing, matched-tumor and normal-tissue sequencing is still the gold standard for somatic mutation detection.

IMPLEMENTING PRECISION ONCOLOGY TESTING AND INTERPRETATION IN PRACTICE

Doing precision oncology optimally depends on getting operational issues of testing right. Many considerations factor into selecting the right molecular test (Sidebar 1). Communication between medical oncologists and local pathologists becomes more critical than ever, particularly when the material will be sent to an outside facility. Local pathologists control the tissue, and the rationale for testing and the technical needs of the outside laboratory must be clearly stated. Standard operating procedures for molecular testing are useful to facilitate the process and improve the likelihood of timely, successful and accurate molecular result reporting. Importantly, tissue blocks must be assessed for adequate tumor tissue so that the results are interpretable and infrequent mutations can be characterized. Formalin-fixed, paraffin-embedded samples, including fine-needle aspirates and cytology samples with sufficient cellularity, can be used for NGS. The amount of DNA needed, expressed either in nanograms or the number of slides necessary to do testing, should be considered upfront to avoid a quantity-not-sufficient 162 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

SIDEBAR 1. Diagnostic Considerations in Molecular Testing • Choice of assay and design • Cost • Tissue quality • Turnaround time • Clinical Laboratory Improvement Amendments and/or College of American Pathologists certification • Bioinformatics analysis • Clinical interpretation result. NGS technology requires at least 10 to 20 slides for a complete analysis, so the pathologist may have to evaluate multiple blocks to pick the sample with the most tumor tissue likely to yield an interpretable result. Many patients undergo fine-needle aspiration or core biopsies for histologic diagnosis of malignancy, so remnant tissue may be sparse and careful decision making weighing the risks and benefits of biopsy for the express purpose of genomic testing is essential. For patients likely to require molecular testing at some point in their course, it is helpful to plan the initial biopsy of metastatic disease with this need in mind, so that tissue will be available later. Decisions to rebiopsy are complex and include morbidity and cost associated with the procedure versus the value of assessing the current tumor biology, particularly after exposure to genomic-altering agents. Typically, specific informed consent for testing in the context of clinical decision making is not required for molecular profiling. An oncology clinic’s general consent form for testing and treatment should cover molecular testing under the scope of medical practice. If patient results will be used in a prospective registry maintained by the practice, the institution, the testing laboratory or an academic consortium, informed consent based on a collection and analysis protocol is advisable. Should molecular alterations render a patient eligible for a clinical trial, the patient will be required to provide consent again to use this information for the study.

Who and When?

Many patients with metastatic disease may be good candidates for genomic testing at varying times in their clinical course. Patients with disease with fewer or no standard treatment options are candidates for early molecular profiling in the hope that they will be a candidate for a clinical trial evaluating a particular alteration. Such trials are called basket studies and typically are agnostic to the tissue of origin as long as specific variants are identified. When no trial is available or the patient is not eligible, using an approved agent for a specific alteration in another disease state (e.g., BRAF V600E mutation) might be appropriate after failure of standard therapy. Often there is no definitive trial data to base decisions on appropriateness of molecular targeted therapy in noninvestigational settings, and a balance of risks and benefits of an unknown approach should be carefully

PRECISION ONCOLOGY: WHO, HOW, WHAT, WHEN, AND WHEN NOT

TABLE 1. Online Knowledge Bases to Aid Clinical Decision Making Resource

Website

My Cancer Genome

www.mycancergenome.org

JAX Clinical Knowledgebase

https://ckb.jax.org

Clinical Interpretation of Variants in Cancer

https://civic.genome.wustl. edu

Oncology Knowledge Base

https://onco*kb.org

Clinical Genome

https://clinicalgenome.org

weighted. One potential hierarchy for decision making is presented in Table 1. Emerging evidence suggests overall tumor mutational load, analyzed in either large target gene panels of 300 to 600 genes or utilizing whole-exome sequencing, can be predictive of response to immune checkpoint inhibitors.13,14 Additionally, mismatch repair deficiency assessment through genomic analysis is another valuable molecular assay with applicability to immune-oncology therapies.15 In general, early-stage patients undergoing definitive treatment do not typically require somatic gene panels. They will not have actionable alterations provided by NGS beyond what can be ascertained from standard histologic evaluation (e.g., estrogen receptor, progesterone receptor, and HER2 in the case of early-stage breast cancer). Broader molecular information will be of research use only. For certain diseases where a first-line decision depends on multiple molecular markers, such as advanced non–small cell lung cancer, the use of a multiplex NGS panel at diagnosis becomes increasingly attractive given the growing number of targetable genes, the ability to simultaneously obtain the information from one sample and the ever lower cost of multiplex testing. Conversely, when other disease states exhaust evidence-based lines of standard therapy, panel testing is often appropriate if the patient remains a good candidate for further treatment. A decision must be made whether to send a new biopsy or use archival tissue. Discordant genomic results may be seen between primary tumors and metastases, although this is highly disease site specific. For instance, RAS mutations in colorectal cancer are an early event in tumorigenesis, and reliable actionable information regarding use of anti-EGFR antibodies in the metastatic setting can obtained from the primary tumor.16 Conversely, estrogen receptor 1 mutations conferring resistance to aromatase inhibitors in breast cancer appear to occur as a consequence of exposure to aromatase inhibitors and are unlikely to be present in the primary tumor.17 Molecular evolution of the tumor has been documented in many cancers under the selective pressure of prior therapy. This is most apparent in patients receiving targeted therapy whereby one of the mechanisms of resistance is secondary mutations in the gene of interest or along the relevant pathway.18 In these circ*mstances, repeat biopsies may be very informative. Currently, the degree of heterogeneity exhib-

ited by metastatic lesions in various sites is unclear. When heterogeneity is suspected, liquid biopsies for circulating tumor DNA or circulating tumor cells may reflect the clinical situation better, presumably integrating tumor status from a variety of sites and reflecting a composite mutational landscape.19 This concept is attractive but far from established and requires further study. However, minimal invasive tissue sampling using blood samples to yield components such as circulating tumor DNA and circulating tumor cells is likely to become standard as an alternative to biopsy in clinically risky situations and in monitoring progressive alterations over tumor during exposure to targeted agents.

Molecular Tumor Board

As precision oncology expands, there is an increased need to include multiple domain experts in decision making to best harness the massive amount of information wisely. In the community setting where generalist oncologists now have to add management of genomic data to the clinical information, having access to additional expertise is enormously valuable. Either virtual or real-time molecular tumor boards can be accomplished in a practice setting. Ideally, members would include medical, surgical, and radiation oncologists as would be found in any conventional tumor board complimented by pathologists, genetic counselors, and research staff. Additional expertise is often available from commercial laboratories in the form of molecular pathologists and molecular geneticists. In more robust practice settings, access to biostatisticians, bioinformaticists, epidemiologists, and translational scientists may be available to participate. Multiple databases are publicly available for searching during molecular tumor boards to help in making variant-therapy associations. A number of free, frequently updated, and deeply curated websites offer information on a large number of variants and can be very useful to the practicing clinician in helping ascertain whether a particular therapy is right for a patient (Table 2). Archived, online molecular tumor boards such as those provided by ASCO are a good reference source.20

Data Integration

Integration of molecular data into electronic health records remains in the infancy of development. Typically, genomic results are sent to the oncologist in Portable Document Format and are therefore not available in structured fields that are searchable, filterable, and linked to clinical data. Integration of clinical and genomic data are a necessary goal to aid electronic matching of patients to molecular-based trials and to aggregate multiple N-of-1 experiments on individual patients to develop real-world evidence of benefit. Custom interfaces can be developed through relationships with third-party genomic laboratories and information technology companies or as a standalone in larger institutions that possess deep bioinformatics and information technology resources. Several laboratories now offer resources, such as the Caris Life Sciences Molecular Intelligence portal21 and the Foundation Medicine Interactive Cancer Explorer asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 163

SCHWARTZBERG ET AL

TABLE 2. U.S. Food and Drug Administration–Approved Drugs and Companion or Complementary Diagnostics for Non–Small Cell Lung Cancer Targeted Agents

Tumor

Blood

Erlotinib

cobas EGFR Mutation Test v2

Gefitinib

Therascreen EGFR RGQ PCR Kit

Afatinib

Therascreen EGFR RGQ PCR Kit

Osimertinib

cobas EGFR Mutation Test v2

EGFR

cobas EGFR Mutation Test v2

ALK/ROS1 Crizotinib

ALK IHC (D5F3, Ventana) ALK Break Apart FISH Probe Kit (Vysis)

Ceritinib Alectinib PD-L1 Pembrolizumab

PD-L1 IHC (22c3, Dako)

Nivolumab

PD-L1 IHC (28-8, Dako)

Atezolizumab

PD-L1 IHC (SP142, Ventana)

Abbreviations: FISH, fluorescence in situ hybridization; IHC, immunohistochemistry; PCR, polymerase chain reaction, RGQ, Rotor-Gene Q.

portal,22 which allow result searches with some data-basing capability, and provide documentation of available preclinical and clinical research pertaining to observed variants and therapies. The clinical interpretation of molecular alterations is at the heart of providing the value of precision oncology.

NON–SMALL CELL LUNG CANCER AS THE PARADIGM OF PRECISION ONCOLOGY

Recently, lung cancer, after several decades of choosing platinum-based doublets for every patient, has undergone a transformation integrating precision medicine. There are now numerous biomarkers needed for treatment assessment in patients with lung cancer (Table 3), and this number will continue to increase as new molecularly defined subsets are identified. When diagnosing a patient, measuring EGFR mutation, and ALK or ROS1 fusions, will help determine whether a tyrosine kinase inhibitor (TKI) should be used in lieu of cytotoxic chemotherapy.23 Recently, PDL1 expression (tumor proportion score ≥ 50%) has proven to be effective in enriching patients with lung cancer who may benefit from immunotherapy (pembrolizumab) instead of chemotherapy.24 When considering patients who are diagnosed with non–small cell lung cancer, these biomarkers may alter treatment decisions in approximately 50% of patients to biologic agents instead of cytotoxic chemotherapy. Increasing utilization of targeted therapies also brings to the forefront the growing clinical challenge of acquired drug resistance, currently a very active area of research. This is best exemplified by the emergence of the EGFR T790M mutation, which occurs in 50% of patients previously treated with an EGFR-TKI.25 These gatekeeper mutations, which directly interfere with drug-target interactions, are a recurring theme across many kinase-driven tumors treated with kinase 164 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

inhibitors.26 Propelling their significance is the accompanying development of (1) therapies designed to target these mutations and overcome resistance (e.g., T790M/osimertinib), and (2) noninvasive assays that can monitor status of the resistance mutations (e.g., cobas EGFR Mutation Test v2), sequentially and in real time. The complexity associated with acquired resistance is compounded by intra- and intertumor heterogeneity27 and adaptive tumor biology that is facilitated by genetic instability.25 The selective pressure of kinase inhibition can lead to the disappearance of drug resistance mutations,28 emergence of varying resistance mechanisms at different metastatic sites,29 histology transformation to small cell lung cancer,30 or emergence of new resistant clones (e.g., C797S, Leu792).31 Each of these situations presents unique treatment approaches. For example, there are reported responses to specific small cell lung cancer treatments for tumors that have transformed into small cell lung cancer.30 Rechallenge with first-generation EGFR TKIs in patients where the T790M clone disappearshas also been successful.32 The utilization of the evolving genetic landscape of tumors to inform treatment decisions will be made possible through sequential and real-time monitoring of the patient. Rebiopsy with traditional biopsy techniques at time of progression, which may occur at multiple time points through the treatment course, is not safe for the patient, nor feasible from a practical perspective. Furthermore, patients may have multiple tumors. Therefore, the challenge is also identifying the tumor that would yield the best-quality biopsy; however, that approach still does not address the potential for intertumor heterogeneity. Much progress has been made to address these complex and evolving issues through the development of noninvasive plasma-based assays for the detection of emerging

PRECISION ONCOLOGY: WHO, HOW, WHAT, WHEN, AND WHEN NOT

resistance mutations. The U.S. Food and Drug Administration approved the first liquid biopsy–based companion diagnostic to detect the T790M resistance mutation in patients whose disease is progressing on erlotinib, gefitinib or afatinib, for consideration of osimertinib. Furthermore, the search for assays, utilizing PCR- or NGS-based detection methods, that have high sensitivity and specificity, are cost-effective, and have high concordance with tumor biopsies is intensifying.33-35 Some studies note, however, that liquid biopsy is still not ready for replacement of tumor biopsies but, in some instances, such as monitoring response or progression, may be prioritized.36,37 Therefore, in cases in which a liquid biopsy test is negative for a resistance mutation, guidelines recommend a tissue biopsy.23 Collectively, these research efforts are converging to create a new paradigm in precision medicine in oncology. The discovery of resistance mutations, designing new drugs that target these resistance mechanisms, and development of noninvasive techniques to monitor emergence of resistance are all integral components advancing the field forward. The following case highlights the precision medicine revolution occurring in lung cancer.

Case Example

Mr. G, a 55-year-old man, has felt fatigued during the last couple months. A persistent cough led to a doctor’s appointment. He did not have a history of smoking, although his parents did smoke cigarettes. His performance status was good, and he did not have any other chronic medical conditions. Radiographic imaging with CT identified several lesions in the lungs bilaterally. A CT–guided biopsy revealed a well-differentiated adenocarcinoma. The rest of the staging scans revealed stage IV disease in the bilateral lungs. Molecular testing was performed and revealed an exon 19 deletion. The patient started treatment with afatinib 40 mg daily. He had grade 1 acne and grade 1 diarrhea. Follow-up CT after 2 months of treatment revealed significant response of the tumors. He continued treatment, eventually dose reducing to 30 mg after several months of treatment. The patient continued taking afatinib for 20 months when a restaging scan revealed new disease bilaterally. The disease was peripherally located, with the largest lesion approximately 1 cm. Performing a tissue biopsy would be challenging but

feasible. A serum circulating tumor DNA test revealed a T790M mutation. The patient then started treatment with osimertinib 80 mg daily. Restaging after 2 months revealed shrinkage of the tumors. He is tolerating the therapy well and continues at this time.

PRECISION ONCOLOGY IN THE ERA OF VALUE-BASED MEDICINE

The goal of precision testing is to identify the optimal therapy for the patient that will maximize their survival and quality of life. In some cases, there are well-validated biomarkers in a specific tumor context with high-quality clinical evidence (often approved by the U.S. Food and Drug Administration) of improved efficacy using a specific targeted agent or class of agents vis-à-vis an unselective therapy. For example, every new diagnosis of metastatic non–small cell lung cancer should undergo molecular testing for EGFR mutation, ALK rearrangement, ROS1 rearrangement and PD-L1 expression, all of which have demonstrated improved benefit with targeted agents for tumors positive for these biomarkers compared with chemotherapy in the first-line setting. In contrast, mutated RAS is a contraindication for the addition of anti-EGFR therapy for metastatic colorectal cancer due to well-demonstrated lack of efficacy in this setting. Testing for these biomarkers in the appropriate clinical context is the standard of care and is covered by insurance. With proven benefits, clear indications and financial coverage, the major challenge in this subset is underutilization of testing and dissemination and implementation of timely adoption to maximize benefits for all patients. The National Comprehensive Cancer Network, ASCO, and other tumor-specific societies provide regularly updated guidelines and are good references for evidence-based testing. Examples of these validated context-biomarker-drug combinations are listed in Table 3. However, there is a much longer list of context-biomarkerdrug combinations without sufficient evidence to make standard of care (Table 4). The cost of NGS has decreased by orders of magnitude in recent years. Cancer centers and other academic medical centers often have their own “home-grown” panels of cancer genes, anda number of companies offer gene panel sequencing for several thousand dollars within a few weeks. Given the reasonable cost, rapid turnaround time, and the potential for discovery of

TABLE 3. Examples of U.S Food and Drug Administration–Approved Biomarkers/Drug Pairs for Specific Tumors Biomarker

Drug

Tumor Context

HER2/neu (ERBB2) expression

Trastuzumab, pertuzumab

Metastatic breast cancer

EGFR L858R

Erlotinib

Metastatic NSCLC

BCR-ABL1 fusion

Imatinib

Chronic myeloid leukemia

17p deletion

Venetoclax

Chronic lymphocytic leukemia

KIT expression

Imatinib

Gastrointestinal stromal tumor

BRAF V600E

BRAF and MEK inhibitors

Metastatic melanoma

Abbreviation: NSCLC, non–small cell lung cancer.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 165

SCHWARTZBERG ET AL

TABLE 4. Examples of Precision Tests Without Established Clinical Utility Biomarker

Drug

Tumor Context

EGFR L858R

EGFR TKI

Non-NSCLC tumor

BRAF V600E mutation

BRAF and MEK inhibitors

Nonmelanoma

BRAF L597 mutations

BRAF and MEK inhibitors

Any tumor

ATM mutation

PARP inhibitor + alkylator

Any tumor

Abbreviations: NSCLC, non–small cell lung cancer; TKI, tyrosine kinase inhibitor.

new biomarkers that are targetable, the use of panel testing has proliferated in academic medical centers and the community. Standard biomarkers are tested in these gene panels, but in addition, alterations in genes without sufficient evidence of corresponding efficacious therapy are also routinely presented. Interpretation and communication of this data to patients and translation into therapeutic interventions is a daunting challenge for clinicians. For example, BRAF V600E–mutated melanomas respond exceptionally well to BRAF and MEK inhibitors, but the response in colorectal cancers to these drugs as monotherapies has been disappointing.38,39 Closely curated databases of genomic alterations such as Onco*kB40 and MyCancerGenome41 have developed frameworks to assist prioritization of therapies for genomic alterations (Fig. 1). However, use of these biomarkers to select therapy is still largely experimental and should be done in the setting of a clinical study at an experienced center, where structured

support is available for the patient and data can be appropriately collected and aggregated to answer clinical and research questions. It should be emphasized that in communication with patients, clinicians should make clear that beyond the limited set of validated tests with corresponding validated therapies, selection of therapy based on tumor genomic profiles is experimental and with no clearly established benefit. Whenever possible, patients should be encouraged to participate in clinical studies. Molecular stratification of patient tumors increases the challenge of accruing sufficient patients to power detection of benefit (or lack thereof) in these tumor subsets. A variety of clinical trial designs have been developed to validate predictive biomarkers, including random assignment of patients stratified by biomarkers (IPASS,42 MARVEL43), enrichment studies with assignment to study arms by biomarker status (BATTLE,44 I-SPY 245) and adaptive trial designs.46 NCI-MATCH (NCT02465060) is a National Cancer Institute–sponsored clinical trial designed to

FIGURE 1. Example of Hierarchy of Evidence of Genomic Alterations

Used with permission.40

166 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PRECISION ONCOLOGY: WHO, HOW, WHAT, WHEN, AND WHEN NOT

handle the problem of low accrual by combining multiple tumor types as a basket trial based on molecular alteration rather than tumor type, as well as to maximize the number of participating sites to optimize enrollment.47 ASCO has initiated the Targeted Agent and Profiling Utilization Registry Study (TAPUR), which will collect real-world data on use of approved agents to treat molecular targeted variants across disease types.48 Umbrella trials, such as the National Cancer Institute–sponsored Lung MAP trial (NCT02154490), recruit patients of a given tumor type (recurrent metastatic squamous cell carcinoma) and place them into arms based on biomarkers (e.g., PI3KCA mutation) with targeted therapies (e.g., taselisib).49 The vast majority of patients do not participate in biomarker-driven clinical studies, and their genomic and clinical data would be a huge boon for research if able to be collected, aggregated and structured appropriately. This has prompted initiatives to share and pool data between multiple institutions, such as the American Association for Cancer Research–sponsored GENIE project50 and ORIEN.51 Further, direct patient collaborations such as the Metastatic Breast Cancer Project,52 in which individual patients directly give permission for clinical data and tumor tissue to be collected from disparate medical centers and centrally analyzed, could provide important data in low-frequency disease. ASCO is further developing the CancerLinQ53 program to create a data platform in which clinical (and genomic, where available) data from the much larger group of patients treated in a broader range of settings can be collected and analyzed both for clinical and research benefit. The Cancer Moonshot Initiative54 identified data aggregation and a common data ecosystem as key components of accelerating the pace of cancer research. Beyond the current set of existing tests, new promising technologies are being developed. Cell-free tumor DNA found in blood plasma has been detected in multiple metastatic tumor settings19; a liquid biopsy avoids the morbidity of traditional biopsies and allows more frequent monitoring, enabling earlier detection of response or development of resistance. Further, tumor genomic heterogeneity has been demonstrated between primary and metastatic lesions and 55 different metastatic lesions56,57 and even in different regions58 of the same lesion. A liquid biopsy may thus present an integrated profile of the tumor. Further, novel techniques using bio-informatic approaches to infer deficiencies in DNA repair pathways from genomic data59 may predict response to DNA-damaging therapies. Single-cell RNA sequencing,60 or deconvolution of bulk RNA sequencing61 to identify specific immune cell subsets in the tumor microenvironment,62

may assist in predicting which tumors are likely to respond to immune therapy. It is likely that in the future mulitple omic approaches including genomic DNA alterations, epigenetic modifications, transcriptome-based expression of mRNA, proteomic expression, and alterations in regulatory molecules such as microRNA and immune factors will provide a more integrated portrait of the tumor and microenvironment. In a value-based reimbursem*nt world, where quality is defined by outcome/cost, it is essential for oncology practices to maintain up-to-date lists of biomarker/target–driven pairs for which there is compelling evidence that biomarker testing identifies an important therapeutic opportunity (e.g., crizotinib for ALK-rearranged lung cancer) or allows for avoidance of a toxic therapy (e.g., cetuximab in KRAS mutant colorectal cancer). Quality metrics will increasingly focus on avoiding underutilization of these established tests, as clinical evidence has already established biomarker selected therapy as a superior strategy. In contrast, use of broad panel tests is more complex. Payers may have prior authorizations built in to limit use of panel testing in certain clinical circ*mstances. Panel tests may be valuable if they are used to identify a batch of validated biomarker/target–driven pairs as well as to support investigation. However, overuse of panel tests should also be avoided. It is critical that when these tests are obtained, oncologists have access to the necessary support for interpretation. Finally, it is anticipated that linking these genomic reports with detailed treatment histories and clinical outcomes such as response, duration of response and survival will accelerate discovery of efficacious therapeutic strategies.

CONCLUSION

Precision oncology has clinical utility in the here and now, but the promise for the future is much greater. With rapid improvements in technology, enhanced ability to probe beyond single DNA alterations to other molecular components that influence tumor behavior and represent targets for new therapeutics is clearly in sight. Responsible use of this remarkable technology will depend on generating evidence through new clinical trial designs, aggregation of molecular and clinical data in real-world databases and careful analysis to determine relevant target-agent associations. Ultimately, the approach must prove value across specific patient populations. The practicing oncologist should make an effort to understand the power and limitations of the current testing and treatment landscape to help patients make the best informed decisions.

References 1. Venter JC, Adams MD, Myers EW, et al. The sequence of the human genome. Science. 2001;291:1304-1351.

3. Stratton MR, Campbell PJ, Futreal PA. The cancer genome. Nature. 2009;458:719-724.

2. Sjöblom T, Jones S, Wood LD, et al. The consensus coding sequences of human breast and colorectal cancers. Science. 2006;314:268-274.

4. Gray SW, Hicks-Courant K, Cronin A, et al. Physicians’ attitudes about multiplex tumor genomic testing. J Clin Oncol. 2014;32:1317-1323.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 167

SCHWARTZBERG ET AL

5. Le Tourneau C, Delord JP, Gonçalves A, et al; SHIVA investigators. Molecularly targeted therapy based on tumour molecular profiling versus conventional therapy for advanced cancer (SHIVA): a multicentre, open-label, proof-of-concept, randomised, controlled phase 2 trial. Lancet Oncol. 2015;16:1324-1334. 6. Sholl LM, Do K, Shivdasani P, et al. Institutional implementation of clinical tumor profiling on an unselected cancer population. JCI Insight. 2016;1:e87062. 7. Meric-Bernstam F, Brusco L, Shaw K, et al. Feasibility of large-scale genomic testing to facilitate enrollment onto genomically matched clinical trials. J Clin Oncol. 2015;33:2753-2762. 8. Johnson DB, Dahlman KH, Knol J, et al. Enabling a genetically informed approach to cancer medicine: a retrospective evaluation of the impact of comprehensive tumor profiling using a targeted next-generation sequencing panel. Oncologist. 2014;19:616-622. 9. Schwaederle M, Daniels GA, Piccioni DE, et al. On the road to precision cancer medicine: analysis of genomic biomarker actionability in 439 patients. Mol Cancer Ther. 2015;14:1488-1494. 10. COSMIC. http://cancer.sanger.ac.uk/cosmic/contact. Accessed February 10, 2017. 11. National Center for Biotechnology Information. Clinvar. www.ncbi. nlm.nih.gov/clinvar. Accessed February 10, 2017. 12. Garofalo A, Sholl L, Reardon B, et al. The impact of tumor profiling approaches and genomic data strategies for cancer precision medicine. Genome Med. 2016;8:79. 13. Rizvi NA, Hellmann MD, Snyder A, et al. Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in nonsmall cell lung cancer. Science. 2015;348:124-128.

professionals/physician_gls/f_guidelines.asp. Accessed February 10, 2017. 24. Herbst RS, Baas P, Kim DW, et al. Pembrolizumab versus docetaxel for previously treated, PD-L1-positive, advanced non-small-cell lung cancer (KEYNOTE-010): a randomised controlled trial. Lancet. 2016;387:1540-1550. 25. Kosaka T, Yatabe Y, Endoh H, et al. Analysis of epidermal growth factor receptor gene mutation in patients with non-small cell lung cancer and acquired resistance to gefitinib. Clin Cancer Res. 2006;12:57645769. 26. Barouch-Bentov R, Sauer K. Mechanisms of drug resistance in kinases. Expert Opin Investig Drugs. 2011;20:153-208. 27. Fisher R, Pusztai L, Swanton C. Cancer heterogeneity: implications for targeted therapeutics. Br J Cancer. 2013;108:479-485. 28. Piotrowska Z, Niederst MJ, Karlovich CA, et al. Heterogeneity underlies the emergence of EGFRT790 wild-type clones following treatment of T790M-positive cancers with a third-generation EGFR inhibitor. Cancer Discov. 2015;5:713-722. 29. Engelman JA, Zejnullahu K, Mitsudomi T, et al. MET amplification leads to gefitinib resistance in lung cancer by activating ERBB3 signaling. Science. 2007;316:1039-1043. 30. Niederst MJ, Sequist LV, Poirier JT, et al. RB loss in resistant EGFR mutant lung adenocarcinomas that transform to small-cell lung cancer. Nat Commun. 2015;6:6377. 31. Thress KS, Paweletz CP, Felip E, et al. Acquired EGFR C797S mutation mediates resistance to AZD9291 in non-small cell lung cancer harboring EGFR T790M. Nat Med. 2015;21:560-562.

14. Snyder A, Makarov V, Merghoub T, et al. Genetic basis for clinical response to CTLA-4 blockade in melanoma. N Engl J Med. 2014;371:2189-2199.

32. Hata A, Katakami N, Yoshioka H, et al. Spatiotemporal T790M heterogeneity in individual patients with EGFR-mutant non-small-cell lung cancer after acquired resistance to EGFR-TKI. J Thorac Oncol. 2015;10:1553-1559.

15. Sacher AG, Gandhi L. Biomarkers for the clinical use of PD-1/PDL1 inhibitors in non-small-cell lung cancer: a review. JAMA Oncol. 2016;2:1217-1222.

33. Wu YL, Sequist LV, Hu CP, et al. EGFR mutation detection in circulating cell-free DNA of lung adenocarcinoma patients: analysis of LUX-Lung 3 and 6. Br J Cancer. 2017;116:175-185.

16. Han CB, Li F, Ma JT, et al. Concordant KRAS mutations in primary and metastatic colorectal cancer tissue specimens: a meta-analysis and systematic review. Cancer Invest. 2012;30:741-747. 17. Jeselsohn R, Yelensky R, Buchwalter G, et al. Emergence of constitutively active estrogen receptor-α mutations in pretreated advanced estrogen receptor-positive breast cancer. Clin Cancer Res. 2014;20:1757-1767. 18. Russo M, Siravegna G, Blaszkowsky LS, et al. Tumor heterogeneity and lesion-specific response to targeted therapy in colorectal cancer. Cancer Discov. 2016;6:147-153. 19. Alix-Panabieres C, Pantel K. Clinical applications of circulating tumor cells and circulating tumor DNA as liquid biopsy. Cancer Discov. 2016;5:479-491. 20. American Society of Clinical Oncology. Molecular Oncology Tumor Boards. https://university.asco.org/motb. Accessed February 10, 2017. 21. Caris Life Sciences. MI Profile Report. www.carislifesciences.com/ platforms/cmi-overview/mi-profile. Accessed February 10, 2017.

34. Weber B, Meldgaard P, Hager H, et al. Detection of EGFR mutations in plasma and biopsies from non-small cell lung cancer patients by allelespecific PCR assays. BMC Cancer. 2014;14:294. 35. Sundaresan TK, Sequist LV, Heymach JV, et al. Detection of T790M, the acquired resistance EGFR mutation, by tumor biopsy versus noninvasive blood-based analyses. Clin Cancer Res. 2016;22:1103-1110. 36. Oxnard GR, Thress KS, Alden RS, et al. 135O_PR: Plasma genotyping for predicting benefit from osimertinib in patients (pts) with advanced NSCLC. J Thorac Oncol. 2016; 11 (4, Suppl)S154. 37. Jenkins S, Yang J, Ramalingam S, et al. 134O_PR: Plasma ctDNA analysis for detection of EGFR T790M mutation in patients (pts) with EGFR mutation-positive advanced non-small cell lung cancer (aNSCLC). J Thorac Oncol. 2016; 11(4, Suppl)S153-S154. 38. Kopetz S, Desai J, Chan E, et al. Phase II pilot study of vemurafenib in patients with metastatic BRAF-mutated colorectal cancer. J Clin Oncol. 2015;33:4032-4038.

22. FoundationICE. https://foundationice.com. Accessed February 10, 2017.

39. Corcoran RB, Ebi H, Turke AB, et al. EGFR-mediated re-activation of MAPK signaling contributes to insensitivity of BRAF mutant colorectal cancers to RAF inhibition with vemurafenib. Cancer Discov. 2012;2:227-235.

23. National Comprehensive Cancer Network. Clinical Practice Guidelines in Oncology: Non-Small Cell Lung Cancer. www.nccn.org/

40. Memorial Sloan Kettering Cancer Center. Onco*kB. www.onco*kb.org. Accessed February 10, 2017.

168 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PRECISION ONCOLOGY: WHO, HOW, WHAT, WHEN, AND WHEN NOT

41. Vanderbilt-Ingram Cancer Center. My Cancer Genome. www. mycancergenome.org. Accessed February 10, 2017. 42. Mok TS, Wu Y-L, Thongprasert S, et al. Gefitinib or carboplatinpacl*taxel in pulmonary adenocarcinoma. N Engl J Med. 2009;361:947957. 43. Mandrekar SJ, Sargent DJ. Clinical trial designs for predictive biomarker validation: theoretical considerations and practical challenges. J Clin Oncol. 2009;27:4027-4034. 44. Kim ES, Herbst RS, Wistuba II, et al. The BATTLE trial: personalizing therapy for lung cancer. Cancer Discov. 2011;1:44-53. 45. Barker AD, Sigman CC, Kelloff GJ, et al. I-SPY 2: an adaptive breast cancer trial design in the setting of neoadjuvant chemotherapy. Clin Pharmacol Ther. 2009;86:97-100. 46. Kelloff GJ, Sigman CC. Cancer biomarkers: selecting the right drug for the right patient. Nat Rev Drug Discov. 2012;11:201-214. 47. McNeil C. NCI-MATCH launch highlights new trial design in precisionmedicine era. J Natl Cancer Inst. 2015;107:133. 48. American Society of Clinical Oncology. TAPUR. www.tapur.org. Accessed February 10, 2017. 49. LUNG-MAP. LUNG-MAP clinical trial. www.lung-map.org. Accessed February 10, 2017. 50. Memorial Sloan Kettering Cancer Center. cBioPortal for Cancer Genomics. www.cbioportal.org. Accessed February 10, 2017.

53. Shah A, Stewart AK, Kolacevski A, et al. Building a rapid learning health care system for oncology: why CancerLinQ collects identifiable health information to achieve its vision. J Clin Oncol. 2016;34:756-763. 54. National Cancer Institute. Cancer Moonshot. www.cancer.gov/research/ key-initiatives/moonshot-cancer-initiative. Accessed February 10, 2017. 55. Brastianos PK, Carter SL, Santagata S, et al. Genomic characterization of brain metastases reveals branched evolution and potential therapeutic targets. Cancer Discov. 2015;5:1164-1177. 56. Faltas BM, Prandi D, Tagawa ST, et al. Clonal evolution of chemotherapyresistant urothelial carcinoma. Nat Genet. 2016;48:1490-1499. 57. Juric D, Castel P, Griffith M, et al. Convergent loss of PTEN leads to clinical resistance to a PI(3)Kα inhibitor. Nature. 2015;518:240-244. 58. Gerlinger M, Rowan AJ, Horswell S, et al. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. N Engl J Med. 2012;366:883-892. 59. Alexandrov LB, Nik-Zainal S, Wedge DC, et al; Australian Pancreatic Cancer Genome Initiative; ICGC Breast Cancer Consortium; ICGC MMML-Seq Consortium; ICGC PedBrain. Signatures of mutational processes in human cancer. Nature. 2013;500:415-421. 60. Tirosh I, Izar B, Prakadan SM, et al. Dissecting the multicellular ecosystem of metastatic melanoma by single-cell RNA-seq. Science. 2016;352:189-196.

http://

61. Gentles AJ, Newman AM, Liu CL, et al. The prognostic landscape of genes and infiltrating immune cells across human cancers. Nat Med. 2015;21:938-945.

52. Broad Institute of MIT and Harvard. Metastatic Breast Cancer Project. www.mbcproject.org. Accessed February 10, 2017.

62. Newman AM, Liu CL, Green MR, et al. Robust enumeration of cell subsets from tissue expression profiles. Nat Methods. 2015;12:453-457.

51. Oncology Research Information Exchange oriencancer.org. Accessed February 10, 2017.

Network.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 169

CENTRAL NERVOUS SYSTEM TUMORS

CNS ABSTRACTS FROM ASTRO 2016 ANNUAL MEETING

American Society for Radiation Oncology 2016 Annual Meeting: Central Nervous System Abstracts Samuel Chao, MD OVERVIEW The American Society for Radiation Oncology's (ASTRO) 2016 scientific program presented a number of excellent abstracts focusing on brain and spine tumors. Selected abstracts will be reviewed in this article.

here were a number of excellent abstracts regarding central nervous system tumors presented at ASTRO’s 2016 Annual Meeting. This review will highlight some of these abstracts, but is not meant to be comprehensive. The exclusion of some abstracts does not reflect on the quality of those abstracts, as all abstracts presented were superb.

GLIOBLASTOMA/GLIOMA

EORTC 26981-22981 defined the current standard of care for glioblastoma, showing that the addition of temozolomide resulted in a survival advantage.1 In a secondary analysis, Hegi et al2 demonstrated that the methylation of O-6-methylguanine-DNA methyltransferase resulted in better survival and response to temozolomide. However, there has not been an updated nomogram to help predict survival that incorporates O-6-methylguanine-DNA methyltransferase methylation. Gittleman et al3 used data from the Radiation Therapy Oncology Group (RTOG) 0525 and 0825 studies to develop this nomogram. They found that older age at diagnosis, male gender, lower Karnofsky performance status (KPS), lack of a gross total resection, and unmethylated O-6-methylguanine-DNA methyltransferase predicted for worse survival. A nomogram was developed and can be accessed through this website: http://cancer4. case.edu/rCalculator/rCalculator.html. Treatment options for recurrent glioblastoma are limited. A phase I trial was conducted at Yale looking at mibefradil and reported by Lester-Coll et al.4 This drug is a calcium channel blocker, but was found to have activity in gliomas as a radiation sensitizer. In this study, mibefradil was given 5 days prior to resection, and the tissue was analyzed for the presence of drug. Radiation was given to a total dose of 30 Gy in five fractions. Drug dose was escalated in a 3 + 3 design. The study was able to escalate to a final dose level of 200 mg per day. Median progression-free survival was

5.25 months, and median overall survival (OS) was 12.75 months. One patient had a complete response. Drug levels in the tumor correlated to what was required for tumor cell sensitization. The researchers are currently doing a phase I trial in upfront glioblastoma. Other studies included one that looked at salicylic acid (Zhang et al5), which appears to affect growth inhibition and apoptosis related to c-Jun-N-terminal kinase and AMPK (AMP-activated protein kinase) activation and arrest in the G2/M phase in vitro. This may have promise for a future study. A couple of studies specifically looked at institutional data regarding low-grade gliomas. The Mayo Clinic looked at their series of patients with low-grade glioma. Kreofsky et al6 found that the median OS was 11.8, 8.1, 14.2, and 14 years for observation, radiation therapy (RT) alone, chemotherapy alone, and RT and chemotherapy, respectively (p = .02). They recommended that observation is reasonable for appropriately selected patients with low-grade glioma, in particular those younger than age 40 and with a gross total resection. Wahl et al7 looked at patients treated at the University of California, San Francisco, and found that patients with 1p19q codeletion and pretreatment tumor volume of less than 68 cc had 0% risk of progression during treatment. Median progression-free survival was 4.9 years, and median OS was not reached. This suggests that there may be a subgroup of patients who can avoid radiation.

BRAIN METASTASES

Studies to date looking at stereotactic radiosurgery (SRS) alone versus SRS with whole-brain RT (WBRT) have failed to demonstrate a strong role for WBRT upfront given its effect on neurocognitive function and quality of life.8,9 It is recognized, however, that there might be patients who may benefit, as a subset analysis of a prospective trial

From the Department of Radiation Oncology, Rose Ella Burkhardt Brain Tumor and Neuro-oncology Center, Cleveland Clinic, Cleveland, OH. Disclosures of potential conflicts of interest provided by the author are available with the online article at asco.org/edbook. Corresponding author: Samuel Chao, MD, Department of Radiation Oncology, Rose Ella Burkhardt Brain Tumor and Neuro-oncology Center, Cleveland Clinic, 9500 Euclid Ave., T28, Cleveland, OH 44195. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 171

SAMUEL CHAO

demonstrated a potential survival benefit to WBRT.10 Churilla et al11 performed a secondary analysis of EORTC 22952-26001 that randomly selected patients who received SRS or surgical resection to observation versus WBRT. They asked the question whether WBRT translates into a survival benefit for patients with a limited competing risk from their extracranial disease. They found no interaction from the effect of WBRT and time to extracranial progression. There was no difference in OS between patients with favorable versus unfavorable graded prognostic assessment, so the patients with the best survival did not seem to benefit from WBRT. The study authors concluded that WBRT could be omitted in patients undergoing SRS or surgical resection. It is clear from these studies, however, that withholding WBRT does increase the risk for intracranial recurrence.8,10,12-14 Thus, patients who do not receive WBRT have a higher rate of needing salvage therapies, including more SRS. Given the costs of SRS, there is concern that withholding WBRT will increase the overall costs of brain metastases management in a patient. Miller et al15 used single-institutional data comparing various costs including cancer-specific costs, brain metastases–specific cost, and cumulative total costs of health care between those receiving SRS alone and those receiving SRS and WBRT upfront. The study authors showed that there was no difference in the various costs between the two cohorts. Additional sensitivity analyses and modeling were recommended by the authors to identify if there is a subset of patients for which SRS with WBRT is most cost-effective. Brown et al16 reported as a late-breaking abstract the results of NCCTG N107C. This was a phase III trial comparing WBRT and SRS to the resection cavity for patients with resected brain metastases. In total, 194 patients were enrolled in this study. Median follow-up was 15.6 months. There was shorter cognitive deterioration-free survival

KEY POINTS • For gliomas, and in particular glioblastoma, there is a need to better determine prognosis for treatment selection. • Systemic therapies need further study. This includes a better understanding of patients who may avoid radiation, particularly in low-grade gliomas. • More studies seem to prove stereotactic radiosurgery alone is the best approach for patients with a limited number of brain metastases. In this era, an understanding of the costs of stereotactic radiosurgery is necessary. • Resection cavity radiosurgery may be an alternative to whole-brain radiation for patients with resected brain metastases. Systemic agents need further study in the management of brain metastases. • Spine stereotactic radiosurgery has become an established technique in treating spine metastases. The presented abstracts focused on understanding the toxicities and survival rates following spine radiosurgery. 172 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

of 2.8 months in those receiving WBRT compared with 3.2 months for those that received SRS to the resection cavity (p < .0001). As expected, overall intracranial tumor control at 12 months was 78.6% with WBRT and was better compared with SRS alone, which was 54.7% (p = .0001). However, quality of life was better in the SRS arm. One interesting result was that surgical bed relapse at 12 months was rather high at 44.4% compared with SRS at 21.8% for WBRT. The final manuscript is eagerly awaited to see if these factors contributed to a higher than expected surgical bed relapse rate for the SRS-only group. Mahajan et al17 reported the results of an MD Anderson Cancer Center prospective study comparing observation and SRS to the resection cavity for completely resected brain metastases. In this phase III study, 132 patients with 140 resected brain metastases were randomly selected to receive SRS to cavity versus observation. Median follow-up was 12.6 months. The authors found that local control was better in the SRS arm at 72% at 12 months compared with 45% for observation. There was no difference in distant brain metastasis control. OS was the same in both arms. Those with a preoperative tumor size of more than 3 cm in diameter benefitted the most. Adjuvant SRS is recommended postresection for local control over no additional therapy. There has been interest in using systemic agents to manage brain metastases upfront. Magnuson et al18 performed a multi-institutional analysis comparing upfront epidermal growth factor receptor tyrosine kinase inhibitor therapy compared with upfront RT. From four institutions, 162 patients were pooled. Median OS was longer in the upfront RT group at 29.4 months compared with upfront epidermal growth factor receptor tyrosine kinase inhibitor therapy at 20.5 months (p = .0015). Upfront SRS had significantly improved survival, but upfront WBRT did not. Median intracranial progression-free survival was improved in patients receiving upfront RT compared with upfront epidermal growth factor receptor tyrosine kinase inhibitor therapy (21.1 vs. 13.4 months; p = .003). The authors stressed the importance of a prospective study given these results.

SPINE STEREOTACTIC RADIOSURGERY

The role of spine stereotactic radiosurgery is being studied by RTOG 0631, a phase II/III study with the phase III component comparing single-fraction conventional radiation with spine SRS.19 One toxicity that has been noted by those treating with SRS is vertebral compression fractures (VCF). Prior studies have shown that a dose of 20 Gy and more per fraction increases the risk of VCF.20 Thibault et al21 used CT-based segmentation to assess the volume of lytic vertebral body metastatic disease and predict the risk of VCF for spine radiosurgery. In 55 patients, 100 spine segments were analyzed. Of these, 56% had lytic disease. Median dose was 24 Gy in two fractions. Those who developed fractures had pre-existing osteolytic disease. The threshold for VCF was a lytic tumor burden of 11.6% or more. One may consider prophylactic stabilization or vertebral augmentation in this group of patients.

CNS ABSTRACTS FROM ASTRO 2016 ANNUAL MEETING

Using the patients treated at MD Anderson Cancer Center, Deegan et al22 looked at long-term toxicities from spine SRS. This includes grade 4 myelopathy in 1.7%, sensory radiculopathy in 8.5%, and vertebral body collapse in 13.6%, with 10.2% requiring vertebroplasty or surgery. Most of these occur within 2 years of treatment. The only toxicity that occurred after 4 years was a VCF. There is no good way to predict how patients receiving spine SRS will do with regard to survival. This is important, as spine SRS is a relatively expensive treatment but is felt to improve local control with good pain relief compared with conventional RT. Spine SRS may be offered to patients with better survival, but those who have short anticipated survival may be best treated with conventional RT. Balagamwala et al23 used a database comprising 444 patients and looked at factors that contribute to survival. On univariate analysis, patients with a KPS higher than 70, controlled systemic disease, single-level spine disease, absence of visceral metastases, and longer time from diagnosis of primary had improved survival. The authors did recursive partitioning

analysis. Class I patients were defined as KPS higher than 70 and controlled systemic disease and had a median OS of 26.7 months. Class II patients were defined as KPS higher than 70 and uncontrolled systemic disease or KPS 70 or lower, age 54 or older, and no visceral metastases. These patients had a median OS of 13.4 months. Recursive partitioning analysis class III patients had KPS 70 or lower, age 54 or older, and visceral metastases or KPS 70 or less and age younger than 54. The median OS was only 4.5 months in this group. The authors felt that spine SRS as upfront treatment is best reserved for patients who are recursive partitioning analysis class I and II. For class III patients, conventional RT may be favored, and spine SRS can be reserved as salvage treatment.

CONCLUSION

There were numerous excellent abstracts presented at ASTRO's 2016 Annual Meeting. Only a few of these were reviewed in this summary. With continuing research like this, better treatments with lower toxicities and improved survival may be discovered.

References 1. Stupp R, Mason WP, van den Bent MJ, et al; European Organisation for Research and Treatment of Cancer Brain Tumor and Radiotherapy Groups; National Cancer Institute of Canada Clinical Trials Group. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med. 2005;352:987-996. 2. Hegi ME, Diserens AC, Gorlia T, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med. 2005;352:997-1003. 3. Gittleman HR, Lim D, Kattan M, et al. An independently validated nomogram for individualized estimation of survival among patients with newly diagnosed glioblastoma: NRG Oncology/RTOG 0525 and 0825. Presented at: American Society for Radiation Oncology 2016 Annual Meeting. Boston, MA; September 25–28, 2016. Abstract 208.

9. Soffietti R, Kocher M, Abacioglu UM, et al. A European Organisation for Research and Treatment of Cancer phase III trial of adjuvant whole-brain radiotherapy versus observation in patients with one to three brain metastases from solid tumors after surgical resection or radiosurgery: quality-of-life results. J Clin Oncol. 2013;31: 65-72. 10. Aoyama H, Tago M, Shirato H; Japanese Radiation Oncology Study Group 99-1 (JROSG 99-1) Investigators. Stereotactic radiosurgery with or without whole-brain radiotherapy for brain metastases: secondary analysis of the JROSG 99-1 randomized clinical trial. JAMA Oncol. 2015;1:457-464.

4. Lester-Coll NH, Kluytenaar J, Pavlik KF, et al. Mibefradil dihydrochloride with hypofractionated radiation for recurrent glioblastoma: preliminary results of a phase 1 dose expansion trial. Presented at: American Society for Radiation Oncology 2016 Annual Meeting. Boston, MA; September 25–28, 2016. Abstract 210.

11. Churilla TM, Handorf E, Soffietti R, et al. Does whole-brain radiation therapy for oligometastatic brain metastases translate into a survival benefit for patients with a limited competing risk from extracranial disease? A secondary analysis of EORTC 22952-26001. Presented at: American Society for Radiation Oncology 2016 Annual Meeting. Boston, MA; September 25–28, 2016. Abstract 126.

5. Zhang J, Zhang Q, Xu M, et al. Radiosensitizing effects of salicylic acid in glioma and the mechanisms. Presented at: American Society for Radiation Oncology 2016 Annual Meeting. Boston, MA; September 25–28, 2016. Abstract 63.

12. Chang EL, Wefel JS, Hess KR, et al. Neurocognition in patients with brain metastases treated with radiosurgery or radiosurgery plus whole-brain irradiation: a randomised controlled trial. Lancet Oncol. 2009;10:1037-1044.

6. Kreofsky CR, Youland RS, Buckner JC, et al. Single-institution experience of low-grade glioma patient outcomes eligible for RTOG 9802. Presented at: American Society for Radiation Oncology 2016 Annual Meeting. Boston, MA; September 25–28, 2016. Abstract 203.

13. Sahgal A, Aoyama H, Kocher M, et al. Phase 3 trials of stereotactic radiosurgery with or without whole-brain radiation therapy for 1 to 4 brain metastases: individual patient data meta-analysis. Int J Radiat Oncol Biol Phys. 2015;91:710-717.

7. Wahl M, Phillips J, Molinaro A, et al. Omission of radiation therapy for low-grade gliomas: molecular and radiographic correlates of treatment response and disease progression on a phase 2 clinical trial of adjuvant temozolomide. Presented at: American Society for Radiation Oncology 2016 Annual Meeting. Boston, MA; September 25–28, 2016. Abstract 204.

14. Brown PD, Jaeckle K, Ballman KV, et al. Effect of radiosurgery alone vs radiosurgery with whole brain radiation therapy on cognitive function in patients with 1 to 3 brain metastases: a randomized clinical trial. JAMA. 2016;316:401-409.

8. Kocher M, Soffietti R, Abacioglu U, et al. Adjuvant whole-brain radiotherapy versus observation after radiosurgery or surgical resection of one to three cerebral metastases: results of the EORTC 22952-26001 study. J Clin Oncol. 2011;29:134-141.

15. Miller JA, Kotecha R, Mohammadi AM, et al. The economic implication of upfront whole-brain radiation therapy for patients with limited brain metastasis. Presented at: American Society for Radiation Oncology 2016 Annual Meeting. Boston, MA; September 25–28, 2016. Abstract 367.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 173

SAMUEL CHAO

16. Brown PD, Ballman KV, Cerhan J, et al. N107C/CEC.3: a phase III trial of postoperative stereotactic radiosurgery (SRS) compared with whole brain radiotherapy (WBRT) for resected metastatic brain disease. Presented at: American Society for Radiation Oncology 2016 Annual Meeting. Boston, MA; September 25–28, 2016. Abstract LBA1. 17. Mahajan A, Ahmed S, Li J, et al. Postoperative stereotactic radiosurgery versus observation for completely resected brain metastases: results of a prospective randomized study. Presented at: American Society for Radiation Oncology 2016 Annual Meeting. Boston, MA; September 25–28, 2016. Abstract 3. 18. Magnuson WJ, Amini A, Patil T, et al. Deferring radiation therapy for brain metastases in patients with EGFR-mutant non-small cell lung cancer: a multi-institutional analysis. Presented at: American Society for Radiation Oncology 2016 Annual Meeting. Boston, MA; September 25–28, 2016. Abstract 128. 19. Ryu S, Pugh SL, Gerszten PC, et al. RTOG 0631 phase 2/3 study of image guided stereotactic radiosurgery for localized (1-3) spine metastases: phase 2 results. Pract Radiat Oncol. 2014;4:76-81.

174 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

20. Sahgal A, Atenafu EG, Chao S, et al. Vertebral compression fracture after spine stereotactic body radiotherapy: a multi-institutional analysis with a focus on radiation dose and the spinal instability neoplastic score. J Clin Oncol. 2013;31:3426-3431. 21. Thibault I, Zhou S, Campbell M, et al. Volume of lytic vertebral body metastatic disease quantified using computed tomography (CT)-based image segmentation predicts fracture risk following spine stereotactic body radiation therapy (SBRT). Presented at: American Society for Radiation Oncology 2016 Annual Meeting. Boston, MA; September 25–28, 2016. Abstract 133. 22. Deegan BJ, Ho JC, Allen PK, et al. Toxicity profile of spine stereotactic radiosurgery among long-term survivors. Presented at: American Society for Radiation Oncology 2016 Annual Meeting. Boston, MA; September 25–28, 2016. Abstract 62. 23. Balagamwala EH, Miller JA, Reddy CA, et al. Recursive partitioning analysis is predictive of overall survival for patients undergoing spine radiosurgery for spine metastasis. Presented at: American Society for Radiation Oncology 2016 Annual Meeting. Boston, MA; September 25–28, 2016. Abstract 333.

BEYOND ALKYLATING AGENTS FOR GLIOMAS

Beyond Alkylating Agents for Gliomas: Quo Vadimus? Vinay K. Puduvalli, MD, Rekha Chaudhary, MD, Samuel G. McClugage, MD, and James Markert, MD OVERVIEW Recent advances in therapies have yielded notable success in terms of improved survival in several cancers. However, such treatments have failed to improve outcome in patients with gliomas for whom surgery followed by radiation therapy and chemotherapy with alkylating agents remain the standard of care. Genetic and epigenetic studies have helped identify several alterations specific to gliomas. Attempts to target these altered pathways have been unsuccessful due to various factors, including tumor heterogeneity, adaptive resistance of tumor cells, and limitations of access across the blood-brain barrier. Novel therapies that circumvent such limitations have been the focus of intense study and include approaches such as immunotherapy, targeting of signaling hubs and metabolic pathways, and use of biologic agents. Immunotherapeutic approaches including tumor-targeted vaccines, immune checkpoint blockade, antibody-drug conjugates, and chimeric antigen receptor–expressing cell therapies are in various stages of clinical trials. Similarly, identification of key metabolic pathways or converging hubs of signaling pathways that are tumor specific have yielded novel targets for therapy of gliomas. In addition, the failure of conventional therapies against gliomas has led to a growing interest among patients in the use of alternative therapies, which in turn has necessitated developing evidence-based approaches to the application of such therapies in clinical studies. The development of these novel approaches bears potential for providing breakthroughs in treatment of more meaningful and improved outcomes for patients with gliomas.

he past decade has seen important breakthroughs in the treatment of newly diagnosed gliomas with the publication of mature and practice-changing results of several clinical trials. However, nearly all these trials have been based on the combination of radiation therapy and alkylating agents. The promise of targeted therapies, which has resulted in notable successes in several other cancers, has not been realized in patients with gliomas despite numerous trials of agents targeting the most common signaling pathways altered in these tumors.1 Although ongoing studies are actively examining the mechanisms for the failure of targeted therapies in gliomas, alternative approaches that seek to attack tumor cells in ways that circumvent tumor resistance and heterogeneity are being increasingly explored. One of the more exciting of therapeutic strategies that has emerged in recent years involves immunotherapy involving a variety of methods including cell-free and cell-based vaccines, antibody-drug conjugates, and checkpoint blockade, which exploit the expression of tumor-specific antigens and neutralize tumor-mediated immunosuppression.2 Another emerging area is the identification and targeting of tumor-specific metabolic and protein-processing pathways that act as hubs for converging cellular processes vital for tumor cell survival.3 Targeting such hubs has the potential to disable the complex signaling networks that tumor cells

depend on for survival and resistance to therapy. However, the slow progress in developing effective therapies against gliomas has also resulted in patients seeking alternative and often untested therapies that are used concurrent with or as alternatives to standard therapy4; the rigorous assessment of such treatments through systematic studies is emerging as an equally important aspect of cancer care. The following sections examine the current state of these varied approaches and their impact on the treatment of patients with gliomas.

NEW APPROACHES TO GLIOMA THERAPY: TARGETED THERAPIES AND BEYOND

Current Standards of Care for Gliomas

Recent studies have established new standards of care for patients with gliomas. For adults with World Health Organization (WHO) grade II glioma after maximum safe resection, radiation therapy (RT) followed by chemotherapy using a combination of procarbazine, lomustine, and vincristine resulted in improvement of survival compared with RT alone, particularly for patients with low-grade oligodendroglioma.5 The same regimen also resulted in improved overall survival (OS) in patients with WHO grade III (anaplastic) oligodendrogliomas that had codeletions of chromosome 1p and 19q.6,7 Further characterization of this benefit is being

From The Ohio State University Comprehensive Cancer Center, Columbus, OH; University of Cincinnati, Cincinnati, OH; University of Alabama at Birmingham, Birmingham, AL. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Vinay K. Puduvalli, MD, Division of Neuro-oncology, 320 W. 10th Ave., Suite M410, Starling Loving Hall, The Ohio State Wexner Medical Center, Columbus, OH 43210; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 175

PUDUVALLI ET AL

explored in a randomized CODEL trial that seeks to compare the benefits of RT with procarbazine, lomustine, and vincristine with that of RT with temozolomide (TMZ) against 1p/19q codeleted anaplastic gliomas.8 The optimal standard of care for patients with anaplastic gliomas without 1p/19q codeletion is currently being explored in a multicenter CATNON trial that randomly assigned patients to four different treatment arms to assess the benefit of adding TMZ as adjuvant or concurrent therapy with RT. Recently reported interim results of this study indicated that the two arms with adjuvant TMZ had a better outcome compared with the two without.9 Based on these data, the trial has been modified to eliminate the arms without adjuvant TMZ and now continues with two arms (RT followed by TMZ vs. RT with TMZ followed by TMZ), the results of which are awaited. In the setting of recurrent grade II or grade III gliomas, there are no clear new standards of care; currently used treatments include reirradiation, alkylating agents, and treatment of secondary glioblastoma (GBM) with bevacizumab. Lastly, the current standard of care for adults up to age 70 with newly diagnosed GBM after maximum safe resection consists of chemoradiation therapy (6 weeks) with concurrent daily TMZ followed by up to six monthly cycles of adjuvant TMZ, which improved survival particularly in the subgroup of patients in whom tumors have promoter methylation of methyl-guanyl methyltransferase (MGMT).10,11 Efforts to intensify adjuvant TMZ dosing or to add bevacizumab to this regimen have failed to improve survival in this setting. However, recent data showed that the addition of low-intensity alternating electrical fields to the standard of care therapy along with adjuvant TMZ, using transducer arrays applied to the scalp for more than 18 hours a day (tumor-treatment fields; Optune) improved OS in adults with newly diagnosed GBM independent of the MGMT promoter status.12 In elderly patients, a shortened course of chemoradiotherapy (3 weeks) followed by up to 12 months of adjuvant TMZ was

KEY POINTS • Alkylating therapies remain the cornerstone of standardof-care therapy for gliomas despite advances. • Identification and targeting of specific signaling pathways most commonly altered in gliomas have failed to yield improvement in outcomes in patients with these tumors. • There is urgent need for a broader targeting of the diverse pathways that mediate adaptive resistance to treatment and facilitate tumor recurrence. • Novel strategies include immunotherapeutic approaches, targeting of metabolic pathways, and harnessing of newer insights into the biology of gliomas. • Additionally, the lack of curative therapies for gliomas has also increasingly encouraged patients to use alternative therapies with little scientific support. Critical assessment and systematic study of such treatment options is essential for providing the best care for patients with gliomas. 176 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

both tolerated and improved OS compared with RT alone.13 Treatments for recurrent GBM includes bevacizumab, nitrosoureas, tumor-treatment fields, and several other chemotherapeutic agents that are currently less frequently used given their limited benefit. None of these treatments have provided a survival benefit, although bevacizumab received regulatory approval based on response rate and improved progression-free survival (PFS).14,15

Targeted Therapies Against Gliomas: Rationale and Limitations

Extensive genetic, epigenetic, and molecular studies have been conducted to delineate the key signaling pathways and alterations in gliomas.16-20 Early studies had identified key roles for three major pathways in gliomagenesis and progression including (1) the receptor tyrosine kinase/ phosphoinositide 3-kinase (PI3K)/Akt pathway, including alterations in EGFR, Her2, PDGFR, FGFR, and cMET (approximately 90%), (2) the p53 pathway including alterations in TP53, MDM2, and MDM4 (approximately 85%), and (3) the Rb and cell cycle–related pathways including defects in RB, CDKN, CDK, and cyclins (approximately 80%).16 More indepth analysis has revealed the striking complexity of genetic and epigenetic changes in both low- and high-grade gliomas.19,20 These data provided strong rationale for several clinical trials targeting key GBM relevant pathways, especially those of receptor tyrosine kinase inhibitors initially in the setting of recurrent GBM and subsequently in the newly diagnosed setting; however, it was soon evident that single-agent trials of these agents were largely ineffective in providing benefit in either PFS or OS for this patient population.21-23 Focusing on the strategy of overcoming bypass pathways that the tumor cells deploy to establish resistance to the action of single agents, combination strategies targeting multiple pathways were subsequently tried but were again strikingly unsuccessful in providing an improvement in outcome.24 These results point to limitations in our understanding of the complexity of survival and resistance mechanisms adopted by glioma cells and the need for more concerted effort to delineate the adaptive mechanisms of these cells that can be new targets for therapy.

Recent Advances in Understanding the Biology of Gliomas

In-depth genetic and epigenetic analyses of large numbers of low- and high-grade gliomas have recently yielded novel insights into the complexity of the alterations in these tumors.25 From a clinical perspective, this has also provided evidence that histologic diagnosis may not accurately correlate with outcome; this has led to a revision of the WHO classification of brain tumors with modifications that allow incorporation of such molecular markers into conventional diagnostic approaches to align better with clinical outcome.26 Although some of these markers have shown potential to be predictive markers that can help selection of treatment options, most are prognostic in nature and inform largely of the intrinsic behavior of the tumors without

BEYOND ALKYLATING AGENTS FOR GLIOMAS

specificity to the treatments used.6,7,11,27-29 From the perspective of the basic biology of the tumors, the studies have shown that tumor heterogeneity is one of the most critical factors that dictates tumor behavior. Such heterogeneity is seen to be not only spatial (with different regions of the same glioma evolving along distinct pathways) but also temporal (with emergence of new mutations and hence biologic behavior with tumor treatment and progression).30-32 It is also becoming clear that treatments can induce mutations that can contribute to clonal evolution of gliomas.33 Such heterogeneity has provided an explanation for why therapies targeting single or even multiple pathways in gliomas fail to uniformly affect the majority of tumor cells or eliminate emergent clones during the course of tumor growth and therapy.

Emergent Targets for Antiglioma Therapies

Novel therapies for gliomas must overcome tumor heterogeneity and disable resistance mechanisms to treatment to be effective in improving outcome. The search for such strategies has resulted in the identification of novel targets that promise to change conventional approaches and are in advanced clinical studies or in early stages of investigations. Immunotherapy has emerged as one of the most promising strategies against gliomas currently in clinical trials and is aimed at either disabling immunosuppression induced by tumor cells or enable tumor targeting of immune cells by identification of overexpressed proteins or neoantigens. Several other emerging areas that bear promise to change therapeutic outcome of patients include targeting signaling hubs, use of biologic agents, disabling tumor-specific metabolic pathways, and activating death pathways that can eliminate tumor cells independent of internal tumor heterogeneity. The following sections briefly review these approaches and outline ongoing or potential clinical trial strategies related to these approaches. The high frequency (approximately 80%) of mutations in isocitrate dehydrogenase 1 (IDH1) in WHO grade II and III gliomas as well as secondary GBM34 associated with increased levels of 2-hydroxyglutarate (2HG), a putative oncometabolite that is believed to drive gliomagenesis,35 has provided a strong rationale for targeting mutant IDH1 as a therapy against gliomas36; this has led to development of pharmacologic inhibitors of IDH1 that are currently in early clinical trials. A phase II trial of IDH-305, a selective R132H-IDH1 inhibitor against progressive WHO grade II and III gliomas, is due to open shortly (NCT02977689). Another trial aimed at low-grade gliomas that have high 2HG as measured by magnetic resonance spectroscopy proposes to assess changes in 2HG in tumor tissue and clinical outcome in terms of tumor response (NCT02987010). Another agent, AGI-5198, was shown to inhibit the ability of mutant IDH1 to produce 2HG in glioma cells, suggesting a potential for clinical activity in these tumors.37 Tumor cells including glioma cells, unlike normal cells, when subject to hypoxic environments, preferentially continue to use the anaerobic tricarboxylic acid cycle even after

normoxic conditions are restored, the so-called Warburg effect.3 Hypoxic regions are a key feature of GBMs and associated with areas of pseudopallisading necrosis, which also show increased expression of hypoxia-inducible factor α (HIF1α), a key player in inducing the Warburg effect.38 HIF1α is stabilized in the setting of hypoxia and acts as a transcription factor that triggers a number of changes in gene expression and protein signaling aimed at increasing levels of tumor cell defense mechanisms include resistance pathways, accelerated metabolism, and angiogenic factors.39 Therapeutic agents that target upstream effectors that stabilize HIF1α such as PI3K and mTOR have failed to yield substantial responses or improved outcome in early trials, likely because of activation of bypass pathways. Agents that directly target HIF1α have been tested in early trials, but data regarding their efficacy have not been promising.39 More recently, the identification of phosphokinase M2 (PKM2) as being highly expressed in cancers, being transcriptionally upregulated by HIF1α, and promoting the Warburg effect,40,41 has triggered efforts to develop PKM2 inhibitors as anticancer agents.42,43 Depletion of PKM2 in cancer cells reverses the Warburg effect and inhibits tumor formation, providing a strong rationale for targeting PKM2 to inhibit cancer metabolism and tumor growth. Novel inhibitors to inhibit PKM2 are currently under development and may provide a novel therapeutic option against GBM (Fig. 1).42,43 Targeting basic cellular processes common to oncogenic pathways could potentially disable resistance mechanisms deployed by GBM cells and overcome the effect of heterogeneity. Heat shock response is one such highly conserved process that protects cells against adverse environmental stresses (e.g., oxidative stress, acidosis, or metabolic stress).44 Heat shock response is mediated by the heat shock protein (HSP) family, which directs protein folding, oligomerization, and secretion, enabling the cell to generate potent resistance and survival mechanisms. Hsp90, a key member of this family, regulates folding and stabilization of several oncoproteins and is overexpressed in cancer cells; a highaffinity form of the protein is specifically expressed by cancer cells, allowing them to rapidly process proteins unlike normal cells.45 Targeting Hsp90 can destabilize client oncoproteins, leading to their proteosomal degradation disabling crucial defense mechanisms used by GBM and sensitizing them to treatment.46 Several Hsp90 inhibitors that block ATPase activity in a tumor-selective manner are in clinical trials.47 Their use against GBM has been limited due to their inability to cross the blood-brain barrier, short duration of action, or unacceptable toxicity profile.48 Second-generation Hsp90 inhibitors such as AUY922,49 Onalespib,50,51 and Debio0932,52 some of which cross the blood-brain barrier, are currently in clinical trials against cancer and being considered for clinical trials against gliomas. Another central regulator of protein processing involves the unfolded protein response (UPR), which is an evolutionally conserved central defense mechanism activated when protein that protects allows cells to adapt to endoplasmic reticulum (ER) stress.53 ER stress results in the incorrect asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 177

PUDUVALLI ET AL

folding and improper glycosylation of newly synthesized proteins. UPR allows cells to re-establish homeostasis by inducing a cell cycle arrest and blocking of protein translation, which prevents new protein formation during the period of ER stress.53 Cancer cells including glioma cells are frequently subject to hypoxia and nutrient deprivation triggering tumor-specific ER stress, as a result of which they become highly reliant on the UPR for survival, making it an ideal target for therapeutic targeting.54 Given that the UPR has also emerged as a mechanism for resistance to conventional therapies in solid tumors,55 inhibitors of the UPR may serve to overcome resistance to standard antiglioma treatments and enhance their antitumor efficacy. The UPR is mediated by the dissociation of GRP78 from its inhibitory association with three transducer proteins—protein kinase R–like ER kinase, inositol-requiring enzyme 1, or activating transcription factor 6—allowing it to bind to and chaperone unfolded proteins.55 Once released, protein kinase R–like ER kinase, inositol-requiring enzyme 1, or activating transcription factor 6 initiate downstream signals that cause transcriptional arrest and assist in alleviating ER stress. Inhibitors of GRP78 can disrupt signaling through the UPR, facilitate the reversal of ER, and consequently sensitize tumor cells to conventional treatments. AR-12, a novel orally bioavailable agent, has been reported to downregulate GRP78 and affect the UPR.56 Inhibitors of the mitogen-activated protein kinase kinase/extracellular signal–regulated kinase pathways also decrease levels of GRP78 and can potentially inhibit the UPR. Novel inhibitors of GRP78 pathway are in continued development, and these agents are expected to enter clinical trials.57 Several novel and unconventional agents are currently entering clinical trials that have shown promising preclinical data against gliomas. BXQ-950 is a first-in-class agent that is a lipid-protein complex composed of Saposin C (SapC), a lysosomal protein, and a phospholipid (dioleoylphosphatidylserine [DOPS]) assembled into nanovesicles (SapC-DOPS), which selectively kill tumor cells through targeting of phosphatidylserine on the cancer cell surface, activating the ceramide cell death pathway as demonstrated in recent preclinical studies in gliomas.58 The agent is currently in a first-in-human phase I trial including in patients with recurrent GBM (NCT02859857). G-202 (mipsagargin) is another novel prodrug that is activated by prostate-specific membrane antigen, which is expressed by GBM and tumor-associated vasculature but not in normal tissue, and is currently in phase II trials against recurrent GBM (NCT02876003 and NCT02067156). Glioma stem cells have also been assessed as novel targets in GBM; BBI608 (napabucasin), an orally bioavailable STAT3 inhibitor that targets cancer stem cells,59 is currently in a phase I/II trial against recurrent GBM in combination with TMZ (NCT02315534).

IMMUNOTHERAPY FOR GLIOMA

Immunotherapy has emerged as one of the most exciting of therapeutic strategies against gliomas with a variety of approaches that harness the recent insights gained into cell based and humoral immune responses against cancer. 178 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

The following section outlines several key strategies and associated ongoing clinical trials including a summary or early results available to date (Table 1).

Checkpoint Inhibitors

Checkpoint inhibitors (CPIs) work by interacting with molecules involved in the normal immune-inhibitory pathways of the body, tasked with limiting immune responses and avoidance of autoimmune reactivity.60 A phase III trial (Checkmate 143; NCT02017717) is underway, looking at treatment of patients with recurrent GBM with nivolumab, a PD-1 inhibitor, and ipilimumab, a CTLA-4 inhibitor.61 Preliminary data for this trial showed 90% of patients receiving a combination of nivolumab and ipilimumab having grade 3 or 4 adverse events in response to treatment. Preliminary efficacy results showed a 12-month OS of 40% for nivolumab (3 mg/kg), 30% for nivolumab (1 mg/kg) plus ipilimumab (3 mg/kg), and 25% for nivolumab (3 mg/kg) plus ipilimumab (1 mg/kg). Another phase III trial (Checkmate 498; NCT02617589) is investigating treatment with nivolumab combined with RT in adults with newly diagnosed MGMT promoter unmethylated GBM compared with standard therapy. KEYNOTE-028 is a phase I trial looking at the use of pembrolizumab (PD-1 inhibitor) for solid tumors and includes a GBM cohort of 26 patients.62 A total of 73.1% of patients experienced treatment-related adverse events, with 15.4% experiencing grade 3 or 4 adverse events. One patient exhibited a partial response to treatment, whereas 12 patients exhibited stable disease. Results regarding PFS and OS did not show noteworthy improvements from current standard therapy. Additional studies are underway. The relatively modest effect of CPIs in trials to date may be because of the fact that GBM is not generally primed for immune response. Future studies using CPIs in combination therapies that may increase the degree of immune activity against GBM cells may provide better outcomes

T-Cell Therapies

Adoptive T-cell therapy is a novel approach to glioma immunotherapy that allows the targeting of treatment to a patient’s specific tumor-associated antigen (TAA) profile, thus limiting the off-target effect to surrounding nonmalignant cells. Autologous lymphocytes are harvested and grown ex vivo, allowing for modification and recognition of specific TAAs prior to implantation in the patient.64,65 Another novel approach uses chimeric antigen receptor (CAR) T cells, which uses autologous T-cell extraction and transduction to express modified tumor antigen T-cell receptors on the cell surface, allowing for chimeric T-cell activation independent of surface major histocompatibility complexes.66 Brown et al67 recently published a case report of one patient involved in a phase I trial (NCT02208362), treating patients with recurrent GBM using CAR T-cells targeting the TAA interleukin-13 receptor alpha 2 (IL-13Rα2). The patient exhibited tumor regression from 70%–100% in all lesions, an effect that was maintained for 7.5 months. Furthermore, this group

BEYOND ALKYLATING AGENTS FOR GLIOMAS

TABLE 1. Active Clinical Trials for Immunotherapy in Glioma Phase

Population

Design

Estimated Primary Completion Date

NCT Identification

IMA950 (multiple tumor antigens)

I/II

Newly diagnosed GBM

Open-label, single-group assignment

March 2016

NCT02343406

PEPIDH1M (IDH1)

Recurrent grade II glioma

Open-label, single-group assignment

May 2017

NCT02193347

IDH1 peptide vaccine (IDH1R132H)

Grade III/IV glioma

Open-label, single-group assignment

August 2018

NCT02454634

HSPPC-96 (heat shock protein)

Recurrent GBM

Randomized, open-label, parallel assignment

July 2017

NCT01814813

ICT-107 (allogenic TAAs)

III

Newly diagnosed GBM

Randomized, double-blind, parallel assignment

December 2019

NCT02546102

DC vaccine (allogenic tumor lysate)

Newly diagnosed or recurrent GBM

Nonrandomized, open-label, parallel assignment

October 2018

NCT02010606

DC vaccine (autologous tumor lysate)

Pilot

Newly diagnosed GBM

Open-label, single-group assignment

November 2016

NCT01957956

DC vaccine (tumor lysate)

Recurrent GBM

Nonrandomized, open-label, parallel assignment

July 2018

NCT01808820

Human CMV pp65-LAMP mRNA-pulsed DC vaccine

Newly diagnosed GBM

Randomized, double-blind, parallel assignment

March 2019

NCT02366728

ICT-121 (CD 133)

Recurrent GBM

Open-label, single-group assignment

November 2017

NCT02049489

DCVax-L (autologous tumor lysate)

III

Newly diagnosed GBM

Randomized, double-blind, parallel assignment

November 2016

NCT00045968

Pembrolizumab (PD-1 inhibitor)

Solid tumors (Recurrent GBM)

Open-label, single-group assignment

August 2017

NCT02054806

Nivolumab (PD-1 inhibitor) with/without ipilimumab (CTLA-4 inhibitor)

III

Recurrent GBM

Randomized, open-label, parallel assignment

February 2017

NCT02017717

Durvalumab (PD-1 Inhibitor)

Newly diagnosed or recurrent GBM

Nonrandomized, open-label, parallel assignment

July 2017

NCT02336165

Pembrolizumab

Recurrent GBM

Randomized, open-label, parallel assignment

August 2016

NCT02337491

Nivolumab

III

Newly diagnosed GBM

Randomized, open-label, parallel assignment

March 2019

NCT02617589

Indoximod (IDO inhibitor)

I/II

Recurrent GBM

Nonrandomized, open-label, parallel assignment

December 2016

NCT02052648

Genetically modified T cells (IL-13Rα2)

Recurrent GBM

Nonrandomized, open-label, parallel assignment

December 2018

NCT02208362

CAR T cells (EGFRvIII)

I/II

Recurrent GBM

Nonrandomized, single-group

December 2018

NCT01454596

CAR T cells (CMV, HER2)

Recurrent GBM

Open-label, single-group assignment

June 2014

NCT01109095

CAR T cells (EphA2)

I/II

Newly diagnosed or recurrent GBM

Randomized, open-label, parallel assignment

September 2016

NCT02575261

MV-CEA (measles virus)

Recurrent GBM

Nonrandomized, open-label, parallel assignment

June 2017

NCT00390299

PVSRIPO (poliovirus)

Recurrent GBM

Open-label, single-group assignment

January 2017

NCT01491893

DNX2401 (adenovirus)

Recurrent GBM

Open-label, single-group assignment

December 2015

NCT01956734

Toca 511 (retrovirus) + Toca FC

II/III

Recurrent GBM

Randomized, open-label, parallel assignment

November 2017

NCT02414165

Intervention (Target/Origin) Peptide Vaccines

DC Therapies

CPIs

T-Cell Therapies

Viral Therapy

Continued

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 179

PUDUVALLI ET AL

TABLE 1. Active Clinical Trials for Immunotherapy in Glioma (Cont'd) Phase

Population

Design

Estimated Primary Completion Date

NCT Identification

DNX2401 + pembrolizumab

Recurrent GBM

Open-label, single-group assignment

December 2019

NCT02798406

DC vaccine + nivolumab

Recurrent GBM

Randomized, open-label, parallel assignment

May 2018

NCT02529072

Nivolumab + galunisertib (TGF-βR1K inhibitor)

I/II

Refractory solid tumors (recurrent GBM)

Nonrandomized, single group, open label

April 2018

NCT02423343

Pembrolizumab + HSPPC-96

Newly diagnosed GBM

Randomized, parallel assignment

May 2018

NCT03018288

Intervention (Target/Origin) Combination Therapy

Abbreviations: NCT, National Clinical Trial; GBM, glioblastoma; DC, dendritic cell; TAAs, tumor-associated antigens; CPIs, checkpoint inhibitors; IDO, indoleamine 2,3-dioxygenase; CAR, chimeric antigen receptor; CEA, carcinoembryonic antigen; TGF, transforming growth factor. Data derived from https://clinicaltrials.gov.

reported 10 patients currently undergoing treatment, with minimal side effects, and CAR T cells detected in cerebrospinal fluid or tumor cyst fluid for more than 7 days.68 These findings suggest that CAR T cells may be well tolerated by patients and are capable of producing a relevant treatment response in vivo. Other early-stage CAR T-cell trials include NCT01454596 (National Cancer Institute), treating patients in whom GBM expresses EGFR type III (EGFRvIII), a known tumor-specific antigen, and NCT01109095, testing the safety of CAR T cells targeting HER2, a tumor-specific antigen expressed on 87% of GBM cells.69,70 For this second study, the gene expressing the HER2 antibody was transduced into T cells selected for their reactivity to cytomegalovirus (CMV), postulating that these cells would be more reactive as they would respond to both tumor cells and CMV, a viral antigen found in many patients with GBM.

Peptide Vaccine Therapies

Vaccination strategies for GBM are aimed at creating vaccines targeting specific tumor antigens, such as EGFRvIII and isocitrate dehydrogenase 1 (IDH1).60,66,71 The ACT III trial was a phase II trial of newly diagnosed patients with GBM treated with rindopepimut (CDX-110), a vaccine targeting EGFRvIII. Early results showed a survival benefit, with patients exhibiting a PFS at 5.5 months of 66% and median OS of 21.8 months.72 However, the phase III trial (ACT IV) was terminated early when it failed to show a survival benefit.73 It has been speculated that the results may have been influenced by the patients in the control arm faring better than would be expected of typical control subjects with GBM.66,73 Reardon et al74 reported the results of a phase I/II trial of SL-107, a peptide vaccine targeting three tumor antigens— IL-13Rα2, Ephrin A2 (EphA2), and Survivin—in patients with recurrent GBM. Early results showed a partial response in one patient (more than 33 weeks in duration) and stable disease in 15 patients (median duration 8 weeks). Migliorini and Dutoit75 reported results of a trial using IMA950, a peptide vaccine composed of 11 tumor-specific peptides.66 For the six patients treated under the initial protocol, median OS was 17.5 months (range 11–21 months). Another target of interest under investigation for potential vaccination therapy is HSP.76,77 A phase II trial (NCT01814813) is evaluat180 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

ing heat shock protein–peptide complex 96 (HSPPC96) with bevacizumab in patients with recurrent glioma. Similarly, IDH1 is a novel target for vaccine therapy, with one phase I trial in patients with grade III/IV glioma testing an IDH1 vaccine (NOA-16; NCT02454634).

Dendritic Cell Therapies

Dendritic cell (DC) therapies function by harvesting DCs from the patient and exposing them to the tumor-specific peptides or tumor lysate ex vivo, prior to being injected back into the patient.5,18 Previous clinical trials using DC therapy have shown encouraging results, with improvement in OS and 2-year survival compared with the current therapy.60 Santos et al78 published data from a phase II (NCT01280552) trial using ICT-107, a DC vaccine encompassing autologous DCs incubated ex vivo with six tumor-specific peptides. The data showed a relationship between HLA-A2–positive patients and an immune response to treatment, associated with both OS and PFS. The Mayo Clinic group reported a trial (MC1272; NCT01957956) of DC therapy in patients with newly diagnosed GBM.79 Autologous DCs were pulsed with allogenic tumor lysate from two human GBM cell lines. Mean follow-up was roughly 1 year (range 0.19–1.77), with 80% OS. The University of California, Los Angeles group reported recently on a phase IIa trial (NCT01635283) of DC therapy in patients with grade II glioma treated with autologous DCs pulsed with autologous tumor lysate.80 No difference was noted in time to progression between study patients and a matched cohort. One phase III trial (NCT00045968) using a tumor-lysate pulsed DC vaccine (DCVax-L) was recently completed in December 2016 with results not yet reported.

Oncolytic Viral Therapies

Oncolytic virus therapy uses genetically modified viral vectors that have the ability to both directly attack malignant cells as well as produce a durable host immune response to them.81,82 NCT00390299 is a phase I trial for patients with recurrent GBM, assessing the safety and efficacy of measles virus transfected with human carcinoembryonic antigen as a marker for replication in vivo.66 The PVSRIPO trial (NCT01491893) is evaluating a recombinant poliovirus (PSVRIPO) and has shown promising early results.83,84 Patel

BEYOND ALKYLATING AGENTS FOR GLIOMAS

FIGURE 1. Metabolic Pathways Active in GBM Involving Enzymes of Glycolysis, the Pentose Phosphate Pathway, Fatty Acid and Glutamine Metabolism, and Their Regulation by Known Oncogenes and Tumor Suppressor Genes in Proliferating Cells

Growth factor/PI3K/AKT signaling stimulates glucose uptake and flux through the early part of glycolysis. Tyrosine kinase signaling negatively regulates flux through at PKM2, making glycolytic intermediates available for macromolecular synthesis. Myc has been found to promote glutamine metabolism and inhibit oxidative metabolism by activating pyruvate dehydrogenase kinase (PDK). p53 decreases metabolic flux through glycolysis in response to cell stress. Abbreviations: Acetyl-CoA, acetyl coenzyme A; ACL, ATP citrate lyase; AMPK, 5' adenosine monophosphate–activated protein kinase; DCA, dichloroacetate; FBP, fructose 1,6-bisphosphate; NADP, nicotinamide adenine dinucleotide phosphate; NADPH, reduced form of nicotinamide adenine dinucleotide phosphate; PDH, pyruvate dehydrogenase; PEP, phosphoenolpyruvate. Reproduced with modifications from Wolf et al63 under the Creative Common Attribution License.

et al85 recently reported on a new clinical trial using a recombinant oncolytic herpes simplex virus that functions through direct oncolytic action against malignant cells and transfection of a viral payload causing malignant cells to secrete IL-12. Tejada et al86 reported preliminary results from a phase I trial (NCT01956734) using an oncolytic adenovirus

(DNX-2401) and TMZ with noteworthy results: one patient alive 30 months after treatment with no evidence of progression and two further patients alive at 23 months. Aghi et al87 also reported the preliminary results of three phase I trials using replicating retrovirus Toca 511 in patients with recurrent GBM, delivered via three distinct methods. Toca asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 181

PUDUVALLI ET AL

511 is a recombinant retrovirus that transfects yeast cytosine deaminase into malignant cells, allowing for subsequent treatments with Toca FC (a 5-fluorocytosine derivative) to convert to 5-fluorouracil within the malignant cells.87,88 Median OS for all three trial groups was 12.1–13.6 months.

Combination Immunotherapy Approaches

Several trials are investigating the use of combinations of immunologic agents to elicit the synergistic effects of such therapies. These include the CAPTIVE trial (NCT02798406), a phase 2 trial combining pembrolizumab with DNX-2401 (oncolytic adenovirus), the AVERT trial (NCT02529072), a phase I trial testing a combination of nivolumab and a DC vaccine (pp65) against recurrent glioma, a phase I/II trial (NCT02423343) for several solid malignancies, including recurrent GBM, testing nivolumab in conjunction with galunisertib, a transforming growth factor-beta receptor I kinase inhibitor, and a trial (NCT03018288) using pembrolizumab in conjunction with HSPPC96, an HSP peptide vaccine. Further research is likely needed to understand the critical interplay of immunosuppressive mechanisms within GBM, but combination trials such as these will hopefully provide us with useful information regarding concurrent immunotherapy treatments.

INTEGRATIVE MEDICINE AND ALTERNATIVE THERAPIES AGAINST GLIOMAS

Scope and Definition of Integrative Medicine

Integrative medicine, a new name for an ancient field of medicine, was previously known as alternative medicine, but because the term “alternative” implies in lieu of traditional medical therapy, the name has been recently been changed to the more inclusive name. Practitioners in oncology often encounter patients seeking advice about integrative oncology practices but whom are often ill equipped to address such questions, which may result in minimizing or avoidance of discussion regarding these queries. This common practice could lead to the patients becoming reluctant to discuss such therapies with their clinical team and potentially withholding information about the integrative techniques that they may be using. It is noted that up to 65% of cancer survivors report using integrative medicine practices at some time during their clinical course,89 making it a highly relevant issue in the management of oncology patients. It has hence become important for practitioners in oncology to address this issue, encourage a discussion with the patients related to integrative therapy, and provide evidence-based information to help them make appropriate choices. Although reducing the potential for unreported use of integrative medicines by patients and encouraging open dialogue is a major reason for practitioners to educate themselves about integrative medicine, there are also evidence-based practices that demonstrate not only improvement in quality of life measures but also an increase in OS and PFS in patients with cancer. 182 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Role of Stress and Mitigation of Stress

Stress and stress reduction is one of the most difficult fields to study in medicine because of often-subjective measures but is also the most interesting of all integrative medicine fields. Thaker et al90 used a validated stress model to test the hypothesis that stress can induce tumorigenesis through stimulation of the B-adrenergic receptor on tumor cells. Nude mice were placed in a stress-inducing restraint system, and human ovarian carcinoma cells were inoculated into the peritoneal cavity. As compared with the control nonstressed mice, the ovarian cancer cells grew by up to 275% in the stressed-out mice. The pathophysiology of this response is poorly understood but is thought to be secondary to stimulation of the B-adrenergic receptor on tumor cells and transcriptional upregulation of VEGF. A legitimate question hence is whether decreasing stress can cause a decrease in tumor burden and improve survival. In this context, a randomized trial showed a considerable survival benefit of stress reduction in patients with metastatic breast cancer.91 Patients undergoing adjuvant breast cancer therapy were randomized to an intervention arm that taught strategies to “reduce stress, improve mood, and alter health behaviors” for 1 year (26 sessions). At 11 years of follow-up, there was a striking improvement in median survival in the intervention arm (6.1 years) versus the assessment-only arm (4.8 years). Multivariate analyses confirmed that patients randomized to the intervention arm had a significantly lower risk of death because of breast cancer (hazard ratio 0.44; p = .016). Although there is a paucity of such trials, the importance of mental health in patients with cancer, often minimized in routine care, has been shown to improve quality of life measures and now even survival in a statistically meaningful way. Incorporating stress reduction techniques such meditation, psychologic therapy, and psychiatric intervention to address issues such as anxiety and depression into routine oncological clinical practice hence must be an integral part of the management of the patient with cancer in a modern era of medicine.

Exercise and Its Impact on Cancer

All areas of medicine accept that regular exercise is important for maintaining mental and physical health, but the question of whether exercise can actually affect cancer growth has only recently been systematically addressed. In a laboratory study, mice were randomized to a cage with a wheel for running or into a cage without a wheel, and mice in both groups were inoculated with B16F10 melanoma cells. Strikingly, after 4 weeks of running, the exercise mice had a 61% (p < .01) reduction in tumor as compared with the mice that were not exercising. Exercised mice had increased natural killer cell mobilization that was believed to be the pathophysiology due to which tumor progression was decreased.92 In an era of immunotherapy, these data suggest that exercise may have an important part in therapy. Primary brain tumors such as high-grade gliomas are highly aggressive and treatment resistant. A very interesting trial examined the role of exercise in patients with

BEYOND ALKYLATING AGENTS FOR GLIOMAS

recurrent grade III and grade IV malignant gliomas in which 243 patients with a Karnofsky performance status 70 or higher were prospectively given a self-administered questionnaire that assessed exercise behavior and performed a 6-minute walk test to assess functional capacity.93 Exercise was an independent predictor of survival (p = .0081), although, interestingly, functional capacity was not. Exercise was also a better predictor of survival than Karnofsky performance status, age, sex, grade, and number of prior progressions. The adjusted hazard ratio mortality was 0.64 (95% CI, 0.46–0.91) for patients reporting strenuous exercise. That strenuous exercise was an independent predictor of survival is a striking finding in an extremely treatmentresistant tumor given that to date, no chemotherapy, radiation therapy, nor surgical therapy have been shown to improve survival in a meaningful way in recurrent highgrade gliomas. Patients often stop exercising after the diagnosis of cancer because of several factors, including fatigue and concerns regarding the impact of physical exercise on their fragile health status, an attitude that may be inadvertently encouraged by clinicians. The data presented above and in many animal models point to the contrary and suggest that patients should continue to exercise as vigorously as possible not only to maintain performance status but also for its potential effect on controlling tumor growth and improving survival.

Diet and Nutrition

One of the most controversial areas of integrative oncology is related to diet and nutrition and its effect on cancer. There are often concerns about weight loss and poor protein intake in patients with a hypermetabolic state. The prevalent belief among clinicians that diet has no notable effect on tumors has been challenged by emerging data in animals as well as humans. This is highlighted by the intriguing results of a randomized, double-blinded, placebo-controlled study assessing the role of flax seed in postmenopausal patients with breast cancer conducted by Thompson et al.94 After initial biopsy, patients were randomized to a diet taking 25 grams of flaxseed daily in a muffin versus one with no flaxseed and with no other changes in their regular diets. At the time of lumpectomy, the flaxseed arm was found to have a considerable decrease (median 34.2%) in Ki-67 labeling index and considerable increase in apoptotic index (30.7%). No quantifiable changes on these indices were noted in the placebo group. Other similar studies have shown that even small changes in diet for a short period of time can induce changes at a cellular level. However, the question of whether diet can change tumorigenesis in a clinically meaningful way has been challenging to answer because of the difficulties in evaluating the effect of diet in an evidenced-based manner. Randomized controlled trials are virtually impossible, and so clinicians have to often rely on retrospective trial data, which are fraught with statistical bias. One large randomized control trial, the PREDIMED study, conducted in Spain on 4,282 women age 60 to 80 at high cardiovascular risk, showed compelling results in this context. These subjects

were randomized to a Mediterranean diet with olive oil, a Mediterranean diet with nuts, or a low-fat control diet and monitored for development of breast cancer. There were 35 cases of breast cancer in a median follow-up of 4.8 years; intriguingly, the multivariable-adjusted hazard ratios for the Mediterranean diet plus olive oil group versus the control group was 0.32 (95% CI, 0.13–0.79).95 The impact of these results can be considered in the context that a chemotherapy agent that showed similar results in preventing breast cancer would have been considered highly successful. Other specialties in medicine such as cardiology have willingly embraced the impact of stress reduction, exercise, and diet in the prevention and treatment of disease. In oncology, such acceptance has lagged behind despite cumulative data that support the direct beneficial effects of such modalities on quality of life and survival. The reasons for the resistance in accepting the value of such measures in the therapeutic strategies against cancer are unclear. It can be speculated that this may be due to the lack of a direct and verifiable logical link between cancer and, for instance, exercise. It is also possible that the lack of conventionally accepted evidence based on robust clinical trials raise skepticism about the results of small uncontrolled studies or anecdotal experience. Further, in the era of highly specific and targeted therapies, it is possible that the role of a broader and less specific intervention such as diet or exercise may have lesser acceptance as a legitimate anticancer therapy. Perhaps clinicians also have concerns that patients may choose integrative practices in lieu of traditional medical therapy with potential medical consequences. Equally likely, however, is that clinicians trained in oncology receive little or no training in aspects of integrative medicine given the traditional stigma and biases associated with this field in traditional training programs as being a nonspecific science. However, it is highly encouraging that Integrative Oncology as a field is moving forward and that there is a growing recognition of the obligation in clinicians to familiarize themselves with the fundamentals of this field and to critically examine the evidence in the field as well as generate carefully designed studies, including through new study designs and clinical trial approaches to provide evidence-based results on which practice can be based. This is imperative for the benefit of our patients who are currently bombarded with nonevidence-based recommendations from a variety of nonmedical sources, including the internet and social media, every day.

CONCLUSION AND FUTURE DIRECTIONS

Recent advances in treatment have yielded incremental improvement in the outcomes of patients with gliomas; however, paradigm-shifting therapies that provide considerable prolongation of survival and improvement in quality of life for these patients have been elusive. The need for unconventional approaches to targeting gliomas has led investigators to explore targets and strategies that go beyond traditional chemotherapeutic agents, including the ones outlined in the sections above. A variety of other strategies, asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 183

PUDUVALLI ET AL

including gene editing, noncoding RNAs, biologic therapies including viral and nonviral vectors, in addition to cell-based therapies, heat-based therapies, novel surgical techniques, and local delivery using blood-brain barrier disruption and convention-enhanced deliveries, as well as novel approaches with radiation therapy techniques, are under development.

An exhaustive coverage of these techniques is outside the scope of this review, but it is recognized that these equally cutting-edge approaches are under active study. These novel strategies bear great promise in providing the long overdue improvement in quality of life and survival for patients with gliomas.

References 1. Wang H, Xu T, Jiang Y, et al. The challenges and the promise of molecular targeted therapy in malignant gliomas. Neoplasia. 2015;17:239-255. 2. Platten M, Bunse L, Wick W, et al. Concepts in glioma immunotherapy. Cancer Immunol Immunother. 2016;65:1269-1275. 3. Agnihotri S, Zadeh G. Metabolic reprogramming in glioblastoma: the influence of cancer metabolism on epigenetics and unanswered questions. Neuro-oncol. 2016;18:160-172. 4. Horneber M, Bueschel G, Dennert G, et al. How many cancer patients use complementary and alternative medicine: a systematic review and metaanalysis. Integr Cancer Ther. 2012;11:187-203. Buckner JC, Shaw EG, Pugh SL, et al. Radiation plus procarbazine, CCNU, 5. and vincristine in low-grade glioma. N Engl J Med. 2016;374:13441355. Cairncross G, Wang M, Shaw E, et al. Phase III trial of chemoradiotherapy 6. for anaplastic oligodendroglioma: long-term results of RTOG 9402. J Clin Oncol. 2013;31:337-343. van den Bent MJ, Brandes AA, Taphoorn MJ, et al. Adjuvant 7. procarbazine, lomustine, and vincristine chemotherapy in newly diagnosed anaplastic oligodendroglioma: long-term follow-up of EORTC brain tumor group study 26951. J Clin Oncol. 2013;31:344-350. Jaeckle K, Vogelbaum M, Ballman K, et al. CODEL (Alliance-N0577; 8. EORTC-26081/22086; NRG-1071; NCIC-CEC-2): phase III randomized study of RT vs. RT+TMZ vs. TMZ for newly diagnosed 1p/19q-codeleted anaplastic oligodendroglial tumors. Analysis of patients treated on the original protocol design (PL02.005). Neurology. 2016; (suppl; abstr 2001). Van Den Bent MJ, Vogelbaum MA, Nowak AK, et al. Results of the 9. interim analysis of the EORTC randomized phase III CATNON trial on concurrent and adjuvant temozolomide in anaplastic glioma without 1p/19q co-deletion: an Intergroup trial. J Clin Oncol. 2016;34 (suppl; abstr LBA2000). 10. Stupp R, Mason WP, van den Bent MJ, et al; European Organisation for Research and Treatment of Cancer Brain Tumor and Radiotherapy Groups; National Cancer Institute of Canada Clinical Trials Group. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med. 2005;352:987-996. 11. Hegi ME, Diserens AC, Gorlia T, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med. 2005;352:9971003. 12. Stupp R, Taillibert S, Kanner AA, et al. Maintenance therapy with tumor-treating fields plus temozolomide vs temozolomide alone for glioblastoma: a randomized clinical trial. JAMA. 2015;314:2535-2543. 13. Perry J, Laperriere, N, O'Callaghan, CJ, et al. A phase III randomized controlled trial of short-course radiotherapy with or without concomitant and adjuvant temozolomide in elderly patients with glioblastoma. J Clin Oncol. 2016;34: (suppl; abstr LBA2).

184 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Friedman HS, Prados MD, Wen PY, et al. Bevacizumab alone and in 14. combination with irinotecan in recurrent glioblastoma. J Clin Oncol. 2009;27:4733-4740. 15. Kreisl TN, Kim L, Moore K, et al. Phase II trial of single-agent bevacizumab followed by bevacizumab plus irinotecan at tumor progression in recurrent glioblastoma. J Clin Oncol. 2009;27:740-745. 16. Cancer Genome Atlas Research Network. Comprehensive genomic characterization defines human glioblastoma genes and core pathways. Nature. 2008;455:1061-1068. 17. Verhaak RG, Hoadley KA, Purdom E, et al; Cancer Genome Atlas Research Network. Integrated genomic analysis identifies clinically relevant subtypes of glioblastoma characterized by abnormalities in PDGFRA, IDH1, EGFR, and NF1. Cancer Cell. 2010;17:98-110. Noushmehr H, Weisenberger DJ, Diefes K, et al; Cancer Genome 18. Atlas Research Network. Identification of a CpG island methylator phenotype that defines a distinct subgroup of glioma. Cancer Cell. 2010;17:510-522. 19. Brennan CW, Verhaak RG, McKenna A, et al; TCGA Research Network. The somatic genomic landscape of glioblastoma. Cell. 2013;155: 462-477. 20. Brat DJ, Verhaak RG, Aldape KD, et al; Cancer Genome Atlas Research Network. Comprehensive, integrative genomic analysis of diffuse lower-grade gliomas. N Engl J Med. 2015;372:2481-2498. 21. Lassman AB, Rossi MR, Raizer JJ, et al. Molecular study of malignant gliomas treated with epidermal growth factor receptor inhibitors: tissue analysis from North American Brain Tumor Consortium Trials 01-03 and 00-01. Clin Cancer Res. 2005;11:7841-7850. 22. Rich JN, Reardon DA, Peery T, et al. Phase II trial of gefitinib in recurrent glioblastoma. J Clin Oncol. 2004;22:133-142. 23. De Witt Hamer PC. Small molecule kinase inhibitors in glioblastoma: a systematic review of clinical studies. Neuro-oncol. 2010;12:304-316. 24. Sathornsumetee S, Reardon DA. Targeting multiple kinases in glioblastoma multiforme. Expert Opin Investig Drugs. 2009;18:277292. 25. Kim H, Zheng S, Amini SS, et al. Whole-genome and multisector exome sequencing of primary and post-treatment glioblastoma reveals patterns of tumor evolution. Genome Res. 2015;25:316-327. 26. Louis DN, Perry A, Reifenberger G, et al. The 2016 World Health Organization classification of tumors of the central nervous system: a summary. Acta Neuropathol. 2016;131:803-820. 27. Bell EH, Pugh SL, McElroy JP, et al. Molecular-based recursive partitioning analysis model for glioblastoma in the temozolomide era: a correlative analysis based on NRG Oncology RTOG 0525. JAMA Oncol. Epub 2017 Jan 12. 28. Weller M, Tabatabai G, Kästner B, et al; DIRECTOR Study Group. MGMT promoter methylation is a strong prognostic biomarker for benefit

BEYOND ALKYLATING AGENTS FOR GLIOMAS

from dose-intensified temozolomide rechallenge in progressive glioblastoma: the DIRECTOR Trial. Clin Cancer Res. 2015;21:20572064.

49. Sessa C, Shapiro GI, Bhalla KN, et al. First-in-human phase I doseescalation study of the HSP90 inhibitor AUY922 in patients with advanced solid tumors. Clin Cancer Res. 2013;19:3671-3680.

29. Huse JT, Aldape KD. The evolving role of molecular markers in the diagnosis and management of diffuse glioma. Clin Cancer Res. 2014; 20:5601-5611.

50. Do K, Speranza G, Chang LC, et al. Phase I study of the heat shock protein 90 (Hsp90) inhibitor onalespib (AT13387) administered on a daily for 2 consecutive days per week dosing schedule in patients with advanced solid tumors. Invest New Drugs. 2015;33:921-930.

30. Patel AP, Tirosh I, Trombetta JJ, et al. Single-cell RNA-seq highlights intratumoral heterogeneity in primary glioblastoma. Science. 2014; 344:1396-1401. 31. Meyer M, Reimand J, Lan X, et al. Single cell-derived clonal analysis of human glioblastoma links functional and genomic heterogeneity. Proc Natl Acad Sci USA. 2015;112:851-856. 32. Kim J, Lee IH, Cho HJ, et al. Spatiotemporal evolution of the primary glioblastoma genome. Cancer Cell. 2015;28:318-328. 33. Wang J, Cazzato E, Ladewig E, et al. Clonal evolution of glioblastoma under therapy. Nat Genet. 2016;48:768-776. 34. Yan H, Parsons DW, Jin G, et al. IDH1 and IDH2 mutations in gliomas. N Engl J Med. 2009;360:765-773. Losman JA, Kaelin WG Jr. What a difference a hydroxyl makes: 35. mutant IDH, (R)-2-hydroxyglutarate, and cancer. Genes Dev. 2013; 27:836-852. Cohen AL, Holmen SL, Colman H. IDH1 and IDH2 mutations in gliomas. 36. Curr Neurol Neurosci Rep. 2013;13:345. Rohle D, Popovici-Muller J, Palaskas N, et al. An inhibitor of mutant 37. IDH1 delays growth and promotes differentiation of glioma cells. Science. 2013;340:626-630. Dang CV, Kim JW, Gao P, et al. The interplay between MYC and HIF in 38. cancer. Nat Rev Cancer. 2008;8:51-56. Burroughs SK, Kaluz S, Wang D, et al. Hypoxia inducible factor pathway 39. inhibitors as anticancer therapeutics. Future Med Chem. 2013;5: 553-572. Yang W, Lu Z. Nuclear PKM2 regulates the Warburg effect. Cell Cycle. 40. 2013;12:3154-3158. Yang W, Zheng Y, Xia Y, et al. ERK1/2-dependent phosphorylation and 41. nuclear translocation of PKM2 promotes the Warburg effect. Nat Cell Biol. 2012;14:1295-1304. Iqbal MA, Gupta V, Gopinath P, et al. Pyruvate kinase M2 and cancer: 42. an updated assessment. FEBS Lett. 2014;588:2685-2692. 43. Li N, Feng L, Liu H, et al. PARP inhibition suppresses growth of EGFR-mMutant cancers by targeting nuclear PKM2. Cell Rep. Epub 2016 Apr 13. 44. Jolly C, Morimoto RI. Role of the heat shock response and molecular chaperones in oncogenesis and cell death. J Natl Cancer Inst. 2000;92:1564-1572. 45. Whitesell L, Lindquist SL. HSP90 and the chaperoning of cancer. Nat Rev Cancer. 2005;5:761-772. 46. Miyata Y, Nakamoto H, Neckers L. The therapeutic target Hsp90 and cancer hallmarks. Curr Pharm Des. 2013;19:347-365. 47. Neckers L, Workman P. Hsp90 molecular chaperone inhibitors: are we there yet? Clin Cancer Res. 2012;18:64-76. 48. Waza M, Adachi H, Katsuno M, et al. Alleviating neurodegeneration by an anticancer agent: an Hsp90 inhibitor (17-AAG). Ann N Y Acad Sci. 2006;1086:21-34.

51. Shapiro GI, Kwak E, Dezube BJ, et al. First-in-human phase I dose escalation study of a second-generation non-ansamycin HSP90 inhibitor, AT13387, in patients with advanced solid tumors. Clin Cancer Res. 2015;21:87-97. Isambert N, Delord JP, Soria JC, et al. Debio0932, a second-generation 52. oral heat shock protein (HSP) inhibitor, in patients with advanced cancer-results of a first-in-man dose-escalation study with a fixeddose extension phase. Ann Oncol. 2015;26:1005-1011. 53. Nagelkerke A, Bussink J, Sweep FC, et al. The unfolded protein response as a target for cancer therapy. Biochim Biophys Acta. 2014;1846:277-284. Prabhu A, Sarcar B, Kahali S, et al. Targeting the unfolded protein 54. response in glioblastoma cells with the fusion protein EGF-SubA. PLoS One. 2012;7:e52265. 55. Ma Y, Hendershot LM. The role of the unfolded protein response in tumour development: friend or foe? Nat Rev Cancer. 2004;4:966-977. 56. Booth L, Roberts JL, Cruickshanks N, et al. Regulation of OSU-03012 toxicity by ER stress proteins and ER stress-inducing drugs. Mol Cancer Ther. 2014;13:2384-2398. Wang M, Kaufman RJ. The impact of the endoplasmic reticulum 57. protein-folding environment on cancer development. Nat Rev Cancer. 2014;14:581-597. 58. Wojton J, Meisen WH, Jacob NK, et al. SapC-DOPS-induced lysosomal cell death synergizes with TMZ in glioblastoma. Oncotarget. 2014;5:9703-9709. 59. Li Y, Rogoff HA, Keates S, et al. Suppression of cancer relapse and metastasis by inhibiting cancer stemness. Proc Natl Acad Sci USA. 2015;112:1839-1844. 60. Tivnan A, Heilinger T, Lavelle EC, et al. Advances in immunotherapy for the treatment of glioblastoma. J Neurooncol. 2017;131:1-9. 61. Sampson J, Omuro A, Vlahovic G, et al. IMCT-03: Safety and activity of nivolumab monotherapy and nivolumab in combination with ipilimumab in recurrent glioblastoma: updated results from CHECKMATE-143. Neuro-oncol. 2015;17:v107. 62. Reardon DA, Kim T-M, Frenel J-S, et al. ATIM-35: Results of the phase IB KEYNOTE-028 multi-cohort trial of pembrolizumab monotherapy in patients with recurrent PD-L1-positive glioblastoma multiforme (GBM). Neuro-oncol. 2016;18:vi25-vi26. 63. Wolf A, Agnihotri S, Guha A. Targeting metabolic remodeling in glioblastoma multiforme. Oncotarget. 2010;1:552-577. Dhodapkar KM, Cirignano B, Chamian F, et al. Invariant natural killer 64. T cells are preserved in patients with glioma and exhibit antitumor lytic activity following dendritic cell-mediated expansion. Int J Cancer. 2004;109:893-899. 65. Sawamura Y, Hosokawa M, Kuppner MC, et al. Antitumor activity and surface phenotypes of human glioma-infiltrating lymphocytes after in vitro expansion in the presence of interleukin 2. Cancer Res. 1989;49:1843-1849.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 185

PUDUVALLI ET AL

66. Dunn-Pirio AM, Vlahovic G. Immunotherapy approaches in the treatment of malignant brain tumors. Cancer. 2017;123:734-750.

81. Kaufman HL, Kohlhapp FJ, Zloza A. Oncolytic viruses: a new class of immunotherapy drugs. Nat Rev Drug Discov. 2015;14:642-662.

67. Brown CE, Alizadeh D, Starr R, et al. Regression of glioblastoma after chimeric antigen receptor T-cell therapy. N Engl J Med. 2016;375: 2561-2569.

82. Saha D, Ahmed SS, Rabkin SD. Exploring the antitumor effect of virus in malignant glioma. Drugs Future. 2015;40:739-749.

68. Brown C, Alizadeh D, Starr R, et al. ATIM-13: phase I study of chimeric antigen receptor-engineered T cells targeting IL13Rα2 for the treatment of glioblastoma. Neuro-oncol. 2016;18:vi20. 69. Hegde M, Corder A, Chow KKH, et al. Combinational targeting offsets antigen escape and enhances effector functions of adoptively transferred T cells in glioblastoma. Mol Ther. 2013;21:2087-2101. 70. Humphrey PA, Wong AJ, Vogelstein B, et al. Anti-synthetic peptide antibody reacting at the fusion junction of deletion-mutant epidermal growth factor receptors in human glioblastoma. Proc Natl Acad Sci USA. 1990;87:4207-4211. 71. Swartz AM, Batich KA, Fecci PE, et al. Peptide vaccines for the treatment of glioblastoma. J Neurooncol. 2015;123:433-440. 72. Schuster J, Lai RK, Recht LD, et al. A phase II, multicenter trial of rindopepimut (CDX-110) in newly diagnosed glioblastoma: the ACT III study. Neuro-oncol. 2015;17:854-861. Weller M, Butowski N, Tran D, et al. ATIM-03: ACT IV: an international, 73. double-blind, phase 3 trial of rindopepimut in newly diagnosed, EGFRvIII-expressing glioblastoma. Neuro-oncol. 2016;18:vi17-vi18. Reardon D, Peereboom D, Nabors B, et al. ATIM-11: phase 2 trial 74. OF SL-701, a novel immunotherapy comprised of synthetic short peptides against GBM targets IL-13Rα2, EphA2, and SURVIVIN, in adults with second-line recurrent GBM: interim results. Neuro-oncol. 2016;18:vi20. Migliorini D, Dutoit V. ATIM-21: IMA950 peptide-based vaccine 75. adjuvanted with poly-ICLC in combination with standard therapy in newly diagnosed HLA-A2 glioblastoma patients: preliminary results. Neuro-oncol. 2016;18:vi22. Beaman GM, Dennison SR, Chatfield LK, et al. Reliability of HSP70 76. (HSPA) expression as a prognostic marker in glioma. Mol Cell Biochem. 2014;393:301-307. Berwin B, Hart JP, Pizzo SV, et al. Cutting edge: CD91-independent 77. cross-presentation of GRP94(gp96)-associated peptides. J Immunol. 2002;168:4282-4286. 78. Santos R, Pinilla C, Swanson SJ, et al. ATIM-19: categorizing immune responders with fusion metrics and simulation for association to survival and progression free survival with immune response in HLA-A2+ patients with GBM from a phase 2 trial of dendritic cell (DC) immunotherapy (ICT-107). Neuro-oncol. 2016;18:vi22. 79. Parney IF, Gustafson MP. ATIM-31: allogeneic tumor lysate/autologus dendritic cell vaccines in newly diagnosed glioblastoma: clinical trial MC1272. Neuro-oncol. 2016;18:vi24-vi25. 80. Moughon D, Everson R, Odesa S, et al. ATIM-32: phase IIA clinical trial evaluating dendritic cell vaccine for the treatment of low-grade gliomas. Neuro-oncol. 2016;18:vi25.

186 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

83. Desjardins A, Sampson JH, Peters KB, et al. Oncolytic polio/rhinovirus recombinant (PVSRIPO) against recurrent glioblastoma (GBM): optimal dose determination. J Clin Oncol. 2015;33 (suppl; abstr 2068). 84. Desjardins A, Sampson JH, Peters KB, et al. Patient survival on the dose escalation phase of the Oncolytic Polio/Rhinovirus Recombinant (PVSRIPO) against WHO grade IV malignant glioma (MG) clinical trial compared to historical controls. J Clin Oncol. 2016;34 (suppl; abstr 2061). 85. Patel DM, Foreman PM, Nabors LB, et al. Design of a phase I clinical trial to evaluate M032, a genetically engineered HSV-1 expressing IL-12, in patients with recurrent/progressive glioblastoma multiforme, anaplastic astrocytoma, or gliosarcoma. Hum Gene Ther Clin Dev. 2016;27:69-78. 86. Tejada S, Valle RD, Gallego J, et al. ACTR-15: a phase I study of the oncolytic virus DNX-2401 and a short course temozolomide for glioblastoma at first recurrence. Neuro-oncol. 2016;18:vi4. 87. Aghi M, Vogelbaum M, Kalkanis S, et al. ATIM-05: complementary clinical and ancillary data from 123 patients with recurrent high grade glioma from three phase 1 trials of TOCA 511 and TOCA FC: update and justification for a phase 2/3 trial. Neuro-oncol. 2016;18:vi18. Ostertag D, Amundson KK, Lopez Espinoza F, et al. Brain tumor 88. eradication and prolonged survival from intratumoral conversion of 5-fluorocytosine to 5-fluorouracil using a nonlytic retroviral replicating vector. Neuro-oncol. 2012;14:145-159. Mao JJ, Palmer CS, Healy KE, et al. Complementary and alternative 89. medicine use among cancer survivors: a population-based study. J Cancer Surviv. 2011;5:8-17. Thaker PH, Han LY, Kamat AA, et al. Chronic stress promotes tumor 90. growth and angiogenesis in a mouse model of ovarian carcinoma. Nat Med. 2006;12:939-944. Andersen BL, Yang HC, Farrar WB, et al. Psychologic intervention 91. improves survival for breast cancer patients: a randomized clinical trial. Cancer. 2008;113:3450-3458. 92. Pedersen L, Idorn M, Olofsson GH, et al. Voluntary running suppresses tumor growth through epinephrine- and IL-6-dependent NK cell mobilization and redistribution. Cell Metab. 2016;23:554-562. Ruden E, Reardon DA, Coan AD, et al. Exercise behavior, functional 93. capacity, and survival in adults with malignant recurrent glioma. J Clin Oncol. 2011;29:2918-2923. Thompson LU, Chen JM, Li T, et al. Dietary flaxseed alters tumor 94. biological markers in postmenopausal breast cancer. Clin Cancer Res. 2005;11:3828-3835. Toledo E, Salas-Salvadó J, Donat-Vargas C, et al. Mediterranean diet 95. and invasive breast cancer risk among women at high cardiovascular risk in the PREDIMED Trial: a randomized clinical trial. JAMA Intern Med. 2015;175:1752-1760.

ABSTRACTS FROM SNO 2016

Practice-Changing Abstracts From the 2016 Society for Neuro-Oncology Annual Scientific Meeting Marta Penas-Prado, MD OVERVIEW The most relevant practice-changing presentations at the 2016 Society for Neuro-Oncology (SNO) Annual Scientific Meeting revolved around the topic of the new 2016 World Health Organization (WHO) classification of central nervous system (CNS) tumors. The most notable change in this new classification is the introduction of molecular markers into the morphologic classification of diffuse gliomas (isocitrate dehydrogenase [IDH] mutation, 1p19q codeletion, and H3K27M mutation), ependymomas (RELA fusion), medulloblastomas (WNT- and sonic hedgehog–activated), and other embryonal tumors (C19MC amplification), thus allowing for more precise diagnosis of these entities compared with the use of morphologic features alone. Among the clinical trials presented, only one phase III trial evaluating a device therapy for treatment of newly diagnosed glioblastoma (EF14; tumor-treating fields) met prespecified statistical criteria for success, showing a modest benefit in progression-free survival and overall survival in patients without progression after radiation and concurrent temozolomide. Other topics of interest included the spatial and temporal heterogeneity of primary brain tumors and the prevalence of burnout among neuro-oncologists.

he new 2016 WHO classification of CNS tumors and its practice-changing implications were discussed in detail in several presentations during the 2016 SNO Annual Scientific Meeting.1-6 Historically, gliomas and other primary brain tumors have been classified based on morphology, with glioma grading depending on cellularity, nuclear pleomorphism, and presence of microvascular proliferation and necrosis. It is widely accepted that tumor grade is the primary predictor of biologic behavior and prognosis. However, recent data demonstrate that the molecular marker IDH mutation provides better prognostication than grade in malignant gliomas, and anaplastic gliomas without IDH mutations (IDH wild-type) have a similarly dismal prognosis as glioblastoma. Additionally, the presence of 1p19q codeletion is linked to the oligodendroglioma lineage and also associated with better prognosis and response to therapy than tumors of astrocytoma lineage. This current 2016 update of the WHO classification of tumors of the CNS is the result of the collaboration of 117 contributors from 20 countries and discussion of the most controversial issues by a working group of 35 neuropathologists, neuro-oncological clinical advisors, and scientists at a 3-day consensus conference. The changes introduced in this 2016 update to the 2007 edition are detailed by Louis et al.7

The 2016 WHO CNS tumor classification moves beyond the classic histology description toward an integrated diagnosis, adding molecular-genetic markers to define many CNS tumors, although the grade determination is still being defined primarily on histologic criteria. As a result, the classification now includes a major restructuring of diffuse gliomas, with incorporation of new molecularly defined entities and deletion of some variants lacking diagnostic and/or biologic relevance (Table 1). The WHO grade II diffuse astrocytomas, WHO grade III anaplastic astrocytomas, and WHO grade IV astrocytomas (glioblastoma), are now each divided into IDH-mutant, IDH wild-type, and not otherwise specified (NOS) categories. The diagnosis of WHO grade II oligodendroglioma and WHO grade III anaplastic oligodendroglioma requires the demonstration of both an IDH mutation and combined whole-arm losses of 1p and 19q (1p/19q codeletion). In the absence of testing or the setting of inconclusive genetic results, an otherwise histologically typical oligodendroglial tumor should be diagnosed as NOS. In instances in which histology and molecular genetic features are discordant (e.g., a diffuse glioma with morphologic features of astrocytoma, but harboring an IDH mutation and 1p/19q codeletion, or a tumor with appearance of oligodendroglioma, but with IDH, ATRX, and TP53 mutations), the genotype features determine the final diagnosis. Importantly, a molecularly defined

From the Department of Neuro-Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX. Disclosures of potential conflicts of interest provided by the author are available with the online article at asco.org/edbook. Corresponding author: Marta Penas-Prado, MD, Department of Neuro-Oncology, The University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd., Unit 431, Houston, TX 77030; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 187

MARTA PENAS-PRADO

TABLE 1. 2016 WHO Changes to the Classification of Diffuse Gliomas New Molecularly Defined Entities

Entities or Variants Deleted

Variants Added

Diffuse astrocytoma, IDH-mutant

Gliomatosis cerebri

Epithelioid glioblastoma

Diffuse astrocytoma, IDH wild-type

Protoplasmic and fibrillary astrocytoma variants

Diffuse astrocytoma, NOS Anaplastic astrocytoma, IDH-mutant Anaplastic astrocytoma, IDH wild-type Anaplastic astrocytoma, NOS Glioblastoma, IDH wild-type Glioblastoma, IDH-mutant Glioblastoma, NOS Diffuse midline glioma, H3K27M-mutant Oligodendroglioma, IDH-mutant and 1p/19q-codeleted Oligodendroglioma, NOS Anaplastic oligodendroglioma, IDH-mutant and 1p/19q-codeleted Anaplastic oligodendroglioma, NOS Oligoastrocytoma, NOS* Anaplastic oligoastrocytoma, NOS* Abbreviations: WHO, World Health Organization; IDH, isocitrate dehydrogenase; NOS, not otherwise specified; CNS, central nervous system. *The diagnosis of mixed gliomas is strongly discouraged in the 2016 WHO classification of CNS tumors. Diagnostic molecular testing should be used for further classification as a “molecularly” astrocytic or oligodendroglial tumor, whenever possible. NOS is to be used in the absence of diagnostic molecular testing or when results are inconclusive; for oligoastrocytomas, NOS category should also be used in the very rare instance of a dual genotype oligoastrocytoma.

group of tumors, diffuse midline gliomas, H3K27M-mutant, has been introduced in the new 2016 WHO classification. These are tumors primarily seen in children (but also in adults), characterized by K27M mutations in the histone H3 gene H3F3A (or less commonly in the HIST1H3B gene) and

KEY POINTS • The new 2016 WHO classification of CNS tumors introduced molecular markers into the morphological classification of diffuse gliomas, ependymomas, medulloblastomas, and other embryonal tumors, allowing for more precise diagnosis of these entities. • The 2016 CNS WHO classification includes a major restructuring of diffuse gliomas; WHO grade II diffuse astrocytomas, grade III anaplastic astrocytomas, and grade IV astrocytomas (glioblastoma) are now each divided into IDH-mutant, IDH wild-type, and NOS categories. • Only one randomized phase III non–placebo-controlled trial was presented that met prespecified statistical criteria for success, evaluating the role of tumor-treating fields for treatment of newly diagnosed glioblastoma in combination with maintenance temozolomide. • Recognition of spatial and temporal hetereogeneity of gliomas is key for the development of the new generation of clinical trials. • A survey among SNO members revealed a high prevalence of symptoms of burnout and high levels of stress, and showed that patient care was the most satisfying career aspect. 188 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

a diffuse growth pattern, midline location (thalamus, brain stem, or spinal cord), and poor prognosis. Even though the changes in the classification of diffuse gliomas are the most relevant due to the frequency of diffuse gliomas in clinical practice, other important changes have been introduced in this 2016 classification. New glioma variants have been included, such as epithelioid glioblastoma and anaplastic pleomorphic xanthoastrocytoma, as well as a new mixed neuronal-glial tumor entity, diffuse leptomeningeal glioneuronal tumor. A genetically defined ependymoma variant has also been included that is called ependymoma, RELA fusion-positive, which accounts for the majority of supratentorial ependymomas in children. Medulloblastomas and other embryonal tumors are also being reclassified based on molecular subtypes when possible, and the general term PNET, or primitive neuroectodermal tumor, has been removed and substituted by the wastebasket diagnosis of CNS embryonal tumor, NOS, to be used when further histologic and/or molecular characterization is not possible (Table 2). Resembling the new classification of diffuse gliomas, medulloblastomas should also be classified following an integrated diagnosis model, with incorporation of both molecular group and histologic phenotype. The classification of embryonal tumors is expected to continue to evolve when molecular markers allow more precise cataloguing of these tumors and their subtypes in the near future. Other noteworthy changes to the 2016 WHO classification of CNS tumors include the introduction of brain invasion as a criterion for atypical meningioma and a soft tissue-type grading system

ABSTRACTS FROM SNO 2016

for the new combined entity of solitary fibrous tumor/ hemangiopericytoma. The WHO CNS tumor classification remains a work in progress. Of note, only markers with widely recognized diagnostic value and standardized, widely available methods of detection have been incorporated. Prognostic markers such as MGMT promoter methylation in glioblastoma have therefore not been incorporated, because of the lack of diagnostic value and lack of consensus on methods for detection. One of the major shortcomings is the difficulty to rapidly incorporate diagnostically relevant new molecular findings into standard WHO updates. For this reason, a new initiative will soon commence that will facilitate input and consensus review of novel molecular data, perhaps facilitating incorporation into future CNS tumor classifications. This initiative has been named cIMPACT-NOW (the Consortium to Inform Molecular and Practical Approaches to CNS Tumor Taxonomy) and is sponsored by the International Society of Neuropathology.8

FINAL ANALYSIS OF TUMOR-TREATING FIELDS PHASE III CLINICAL TRIAL FOR NEWLY DIAGNOSED GLIOBLASTOMA

The final results of NovoCure EF-14, an industry-sponsored, multicenter, non–placebo-controlled, randomized phase III trial in newly diagnosed glioblastoma testing the efficacy of tumor-treating fields (TTFields) in combination with maintenance temozolomide after initial treatment with chemoradiation were presented at this 2016 SNO meeting.9 TTFields are low-intensity (1–3 V/cm), intermediatefrequency (200 kHz) alternating electric fields applied to the shaved scalp via transducer arrays, connected to a portable device set to generate the electric fields (Optune; Novocure Ltd.). In preclinical models, TTFields have been shown to cause mitotic arrest and apoptosis by disrupting mitotic spindle formation during metaphase.10 Notably, a prior randomized phase 3 trial in 237 patients with recurrent glioblastoma comparing treatment with TTFields to physician’s choice standard chemotherapy failed to demonstrate an improvement in overall survival (OS) or progression-free survival (PFS), although adverse effects were generally minor.11

The current study in newly diagnosed glioblastoma aimed to evaluate the efficacy and safety of TTFields used in combination with maintenance temozolomide after completing therapy with radiation and concurrent temozolomide. The interim results were previously reported at SNO 2014 and published in JAMA in 2015.10 Randomization was 2:1 favoring the experimental arm. The primary endpoint of the study was PFS in the intent-to-treat population, assessed by a blinded central review panel, with OS in an as-treated population as a powered secondary endpoint. OS was to be tested statistically only if the primary was met to avoid an increase in the risk of a false-positive result. The as-treated population was defined as patients who were able to receive at least one adjuvant cycle of temozolomide (i.e., started cycle 2) and also excluded patients in the control arm who received TTFields therapy outside the protocol (a total of 35 patients were removed for analysis of OS at the time of interim analysis). Blinded central review was not performed in real time to dictate treatment decisions, and some discrepancies in interpretation between local and central review were identified in both the experimental and standard arm.10 An unusual feature of this trial was that, in the TTFields arm, patients were allowed to continue TTFields in combination with second-line chemotherapy until the second radiologic progression or clinical deterioration, for a maximum of 24 months. At the published prespecified interim analysis, which included data from 315 patients,10 the difference in PFS was 4.0 versus 7.1. months (i.e., an increase in 3.1 months favoring the TTFields arm, with a hazard ratio of 0.62), whereas the final analysis with 695 patients9 found the difference was 4.0 versus 6.7 months with a hazard ratio of 0.63 (i.e., an increase of 2.7 months favoring the TTFields arm). In terms of OS in an as-treated population (powered secondary endpoint), there was a difference of 5 months both at interim and final analysis with a hazard ratio of 0.65, although the difference in intent-to-treat population at interim analysis was smaller (3 months). The respective 2-, 3-, and 4-year OS rates (secondary objective) were 43% (CI, 38%–47%) versus 30% (CI, 24–37), 24% (CI, 19–29) versus 16% (CI, 11–23), and 17% (CI, 13–23) versus 10% (CI, 6–18), respectively

TABLE 2. 2016 WHO Changes to the Classification of Medulloblastomas and Other Embryonal Tumor New Molecularly Defined Entities

Entities or Variants Deleted

Entities Added

Medulloblastoma, WNT-activated

Primitive neuroectodermal tumor

CNS embryonal tumor, NOS

Medulloblastoma, SHH-activated and TP53-mutant Medulloblastoma, SHH-activated and TP53 wild-type Medulloblastoma, non-WNT/non-SHH Medulloblastoma, group 3 Medulloblastoma, group 4 Medulloblastoma, NOS Embryonal tumor with multilayered rosettes, C19MC-altered Embryonal tumor with multilayered rosettes, NOS Abbreviations: WHO, World Health Organization; CNS, central nervous system; NOS, not otherwise specified; SHH, sonic hedgehog.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 189

MARTA PENAS-PRADO

TABLE 3. Comparative Outcome Data From RTOG 0525, RTOG 0825, EF-14 Interim Analysis, and EF-14 Final Analysis EF-14† RTOG 0525 (2013)*

RTOG 0825 (2014)**

Interim (JAMA 2015)

Final (SNO 2016)

Median PFS (Months) Standard/ Experimental

5.5/6.7

7.3/10.7

4.0/7.1

4.0/6.7

Median OS (Months) Standard/ Experimental

16.6/14.9

16.1/15.7

15.6/20.5‡

16/21

N (Standard/Experimental)

411/422

309/312

105/210

695

Abbreviations: JAMA, Journal of the American Medical Association; SNO, Society for Neuro-Oncology; PFS, progression-free survival; OS, overall survival. *Randomization after chemoradiation. **Randomization by day 10 of chemoradiation. †Randomization after chemoradiation. ‡Defined as per protocol population (prespecified analysis). Median OS in intent-to-treat population also reported: 16.6 vs. 19.6 months.

(p < .05 for all time points). Therefore, this phase III trial met prespecified statistical criteria for success, with a modest improvement in both PFS and OS. This is in striking contrast with the negative results of the previous phase III trial in recurrent glioblastoma.11 Table 3 summarizes EF-14 outcome data (median PFS and median OS) and provides the results of two other recent phase III trials, RTOG 0525 (standard vs. dose-dense temozolomide)12 and RTOG 0825 (bevacizumab vs. placebo in addition to chemoradiation),13 both large, multicenter randomized trials in newly diagnosed glioblastoma. Although direct comparison of patient outcomes among different trials is not statistically appropriate, RTOG 0525 and RTOG 0825 provide a reference of expected outcome of patients with newly diagnosed glioblastoma treated in clinical trials, and it is worth discussing in some detail to highlight potential limitations of this trial. In RTOG 0525 and EF-14, patients were randomly assigned after chemoradiation, whereas in RTOG 0825, random assignment occurred at the beginning of chemoradiation (patients had to be randomly assigned by day 10 of chemoradiation). In all three trials, PFS and OS were measured from random assignment. When comparing outcome data, notably, the temozolomide control arm in EF-14 trial showed worse PFS than in RTOG 0525 and RTOG 0825, whereas OS seemed equivalent. In both RTOG 0525 and RTOG 0825, patients with suspected treatment effect rather than true tumor progression (pseudoprogression) were allowed to continue therapy, unless there was a new lesion or clinical decline. In the EF-14 trial, patients with suspected progression after chemoradiation were excluded from participation. One potential factor that can account for differences in PFS in a given trial or among trials is the misclassification of true early tumor progression and pseudoprogression. Unfortunately, both traditional imaging criteria and clinical criteria are often misleading when trying to differentiate true progression from early treatment effect, thus being challenging to correctly include or exclude patients from trial participation and to estimate PFS. It is known that the peak of pseudoprogression happens within 3 to 4 months of completion of chemoradiation. Unfortunately, a blinded central radiologic review does not eliminate the problem of 190 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

having inadequate tools to determine true versus false tumor progression. If, inadvertently, more patients with true progression right after chemoradiation were included in the temozolomide-only control arm of EF-14 (as suggested by worse PFS than in previous RTOG trials), this may have introduced an involuntary source of bias into the randomized study, favoring the experimental arm. In summary, the positive results of EF-14 bring one additional therapy (TTFields) into the treatment arsenal against newly diagnosed glioblastoma. However, in view of the previous negative phase III trial in recurrent glioblastoma and the modest differences found in PFS and OS, it is still unclear which patients truly derive most benefit from this therapy, a question particularly relevant in the context of globally increasing health care costs. Unfortunately, patients with the most aggressive tumors (i.e., showing rapid tumor progression early during their course) are still lacking valid treatment options. In many cases, these tumors are also likely to have high MGMT activity as well as other unfavorable molecular markers. Additional research to investigate novel therapies for newly diagnosed and recurrent glioblastoma must continue to achieve much better improvement in survival, in the order of years rather than several months, as is the reality today. We must see the progress in OS for high-grade gliomas that has occurred in other malignancies like chronic myeloid leukemia or her2+ metastatic breast cancer.14,15

OTHER TOPICS OF INTEREST

Spatial and Temporal Heterogeneity of Tumors

It is well known that gliomas represent a group of highly heterogeneous tumors, not only spatially (with different areas of the same tumor demonstrating different morphology, grade, or even molecular findings), but also temporarily, with an evolving molecular makeup during tumor treatment and evolution.16 In this regard, Costello17 presented intriguing data regarding IDH mutations during glioma evolution. Despite being an early oncogenic molecular event, IDH status can change during subclonal evolution and tumor recurrence via epigenetic mechanisms. It is becoming increasingly clear that the failure of molecularly targeted therapies for recurrent tumors may be due, at least in part, to dynamic changes in the molecular alterations

ABSTRACTS FROM SNO 2016

driving the growth of the recurrence as compared with the original tumor. In consequence, taking therapy decisions based on molecular analysis on original tumor tissue is likely inappropriate in most cases. Recognition of this phenomenon should lead to changes in research practice, obtaining new tumor tissue for analysis whenever feasible before enrollment to clinical trials. In addition, it is also apparent that we do not have targeted drug therapies today that are sufficiently brain penetrant to access infiltrating cells, able to inhibit cellular targets long enough to exert sufficient antitumor activity, or specific enough in their action to be safely combined with other targeted therapies,18 and a summary of the 2nd CNS Anticancer Drug Discovery and Development Conference will be published in CNS Oncology in the next few months.18

Investigating Burnout and Career Satisfaction Among Neuro-Oncologists

Professional burnout is common among U.S. physicians, but prevalence, root causes, and consequences of burnout

syndrome in the neuro-oncology community had not been studied until now. Barbara O’Brien, MD, Shlomit Yust-Katz, MD, and Alvina Acquaye, MS, spoke on Education Day at the SNO meeting about their ongoing study on burnout in neuro-oncology and its preliminary findings.19 Among 324 SNO members from the U.S. and Canada who completed an anonymous online survey, 30% reported current symptoms of burnout and 45% reported experiencing burnout in the past. More than 70% of the participants reported working more than 50 hours per week and administrative burden was high. Nearly half of participants reported significant stress and did not meet exercise and sleep recommendations for a healthy lifestyle. Interestingly, despite the unique challenges of caring for patients with brain or spinal cord tumors, patient care was reported as the most satisfying career aspect. Data collection and analysis are still ongoing, but these preliminary results reflect a high prevalence of burnout and stress in the neuro-oncology community and point toward the need for interventions to reduce undue administrative burden.

References 1. Perry A. WHO Overview. Presented at: 21st Annual Scientific Meeting and Education Day of the Society for Neuro-Oncology. Scottsdale, AZ; November 2016.

10. Stupp R, Taillibert S, Kanner AA, et al. Maintenance therapy with tumor-treating fields plus temozolomide vs temozolomide alone for glioblastoma: a randomized clinical trial. JAMA. 2015;314:2535-2543.

2. Brat D. IDH-omas. Presented at: 21st Annual Scientific Meeting and Education Day of the Society for Neuro-Oncology. Scottsdale, AZ; November 2016.

11. Stupp R, Wong ET, Kanner AA, et al. NovoTTF-100A versus physician’s choice chemotherapy in recurrent glioblastoma: a randomised phase III trial of a novel treatment modality. Eur J Cancer. 2012;48:21922202.

3. Ligon K, Eberhart C, Phillips J, et al. Mock integrative diagnostic tumor board. Presented at: 21st Annual Scientific Meeting and Education Day of the Society for Neuro-Oncology. Scottsdale, AZ; November 2016. 4. Ahluwalia M, Buckner J, van den Bent M, et al. Mock molecular treatment tumor board. Presented at: 21st Annual Scientific Meeting and Education Day of the Society for Neuro-Oncology. Scottsdale, AZ; November 2016. 5. Louis D. Keynote Presentation: The 2016 WHO classification of CNS tumors: an overview and a review of diffuse gliomas in adults. Presented at: 21st Annual Scientific Meeting and Education Day of the Society for Neuro-Oncology. Scottsdale, AZ; November 2016. 6. Ellison D. Keynote Presentation: The 2016 WHO classification of CNS tumors – What’s new for pediatrics? Presented at: 21st Annual Scientific Meeting and Education Day of the Society for NeuroOncology. Scottsdale, AZ; November 2016.

12. Gilbert MR, Wang M, Aldape KD, et al. Dose-dense temozolomide for newly diagnosed glioblastoma: a randomized phase III clinical trial. J Clin Oncol. 2013;31:4085-4091. 13. Gilbert MR, Dignam JJ, Armstrong TS, et al. A randomized trial of bevacizumab for newly diagnosed glioblastoma. N Engl J Med. 2014;370:699-708. 14. Kantarjian H, O’Brien S, Jabbour E, et al. Improved survival in chronic myeloid leukemia since the introduction of imatinib therapy: a singleinstitution historical experience. Blood. 2012;119:1981-1987. 15. Mendes D, Alves C, Afonso N, et al. The benefit of HER2-targeted therapies on overall survival of patients with metastatic HER2-positive breast cancer--a systematic review. Breast Cancer Res. 2015;17:140. 16. Johnson BE, Mazor T, Hong C, et al. Mutational analysis reveals the origin and therapy-driven evolution of recurrent glioma. Science. 2014;343:189-193.

7. Louis DN, Perry A, Reifenberger G, et al. The 2016 World Health Organization Classification of Tumors of the Central Nervous System: a summary. Acta Neuropathol. 2016;131:803-820.

17. Costello J. Keynote Presentation: The parallel lives of mutations and epimutations during tumor evolution. Presented at: 21st Annual Scientific Meeting and Education Day of the Society for NeuroOncology. Scottsdale, AZ; November 2016.

8. Louis DN, Aldape K, Brat DJ, et al. Announcing cIMPACT-NOW: the Consortium to Inform Molecular and Practical Approaches to CNS Tumor Taxonomy. Acta Neuropathol. 2017;133:1-3.

18. Levin VA, Tonge PJ, Gallo JM, et al. CNS Anticancer Drug Discovery and Development Conference white paper. Neuro-oncol. 2015;17: vi1-vi26.

9. Stupp R, Idbaih A, Steinberg DM, et al. Prospective, multi-center phase III trial of tumor treating fields together with temozolomide compared to temozolomide alone in patients with newly diagnosed glioblastoma. Neuro-oncol. 2016;18:i1-i150.

19. O’Brien B, Yust-Katz S, Acquaye A. Neuro-oncology burnout and career satisfaction: overview and preliminary results. Presented at: 21st Annual Scientific Meeting and Education Day of the Society for NeuroOncology. Scottsdale, AZ; November 2016.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 191

DEVELOPMENTAL THERAPEUTICS AND TRANSLATIONAL RESEARCH

ADOPTIVE T-CELL THERAPY FOR SOLID TUMORS

Adoptive T-Cell Therapy for Solid Tumors Oladapo Yeku, MD, PhD, Xinghuo Li, and Renier J. Brentjens, MD, PhD OVERVIEW Chimeric antigen receptor (CAR) T-cell therapy is an innovative form of immunotherapy wherein autologous T cells are genetically modified to express chimeric receptors encoding an antigen-specific single-chain variable fragment and various costimulatory molecules. Upon administration, these modified T cells traffic to, and recognize, cancer cells in an HLA-independent manner. CAR T-cell therapy has shown remarkable success in the treatment of CD-19–expressing B-cell acute lymphocytic leukemia. However, clinical gains to the same magnitude have not been reported in solid tumors. Several known obstacles to CAR T-cell therapy for solid tumors include target antigen identification, effective trafficking to the tumor, robust activation, proliferation, and in vivo cytotoxicity. Beyond these T-cell intrinsic properties, a complex and dynamic immunosuppressive tumor microenvironment in solid tumors hinders T-cell efficacy. Notable advancements in CAR design to include multiple costimulatory molecules, ligands, and soluble cytokines have shown promise in preclinical models, and some of these are currently in early-phase clinical trials. In this review, we discuss selected solid tumor malignancies and relevant preclinical data and highlight clinical trial results that are available. Furthermore, we outline some obstacles to CAR T-cell therapy for each tumor and propose strategies to overcome some of these limitations.

AR T-cell therapy for solid tumor malignancies is an exciting frontier in cancer immunotherapy. The general architecture of a CAR consists of a single-chain variable fragment (scFv) derived against a predetermined tumor-associated antigen (TAA) followed by a CD3ζ domain required for provision of signal 1 and T-cell activation upon antigen recognition.1 Upon transfection into autologous T cells, first-generation CAR T cells targeting HER2/Neu-expressing breast and ovarian cancer cell lines showed increased interleukin-2 (IL-2) production and cytotoxicity.2 However, it was subsequently realized that sustained activity and proliferation after receptor engagement required a secondary signal, or signal 2.1 Additional genetic modifications to include costimulatory molecules, such as CD283 and 4-1BB,4 to the CD3ζ signaling domain led to second-generation CARs (28ζ and 4-1BBζ, respectively). Acting in concert, provision of both signal 1 and signal 2 mitigated the anergy and activation-induced cell death observed with first-generation CAR T cells.5 Direct comparison of first- and second-generation CARs directed against CD19, a TAA expressed on malignant B cells, revealed superior expansion, tumor infiltration, and persistence in favor of the second-generation CAR design.6 Additional genetic modifications have yielded third-generation CARs composed of two distinct costimulatory domains, such as CD28/4-1BB/CD3ζ or CD28/OX-40/CD3ζ, all with varying degrees of efficacy.7-9 More recently, other approaches to optimize CAR T-cell efficacy via engineered

expression of tethered or soluble ligands, cytokines, or scFvs10,11 also have been reported. However, despite ongoing success in the management of CD19+ B-cell hematologic malignancies, progress in the solid tumor landscape has been met with many obstacles. One is the identification of suitable neoantigens or TAAs to serve as targets for CAR T-cell therapy. The biologic heterogeneity of solid tumor malignancies does not lend to an approach of one antigen fits all. This difficulty is compounded by the frequent expression of putative target antigens on normal tissues that leads to on-target, off-tumor toxicity.12 Despite this, acceptable antigens, such as EGFR variant III (EGFRIII),13 GD2,14 mucin 1 (MUC-1),9 mucin 16 (MUC-16),15 carcinoembryonic antigen,16 mesothelin,17 CA-IX,18 and prostate-specific membrane antigen (PSMA)19 have been characterized and are in various stages of clinical development (Table 1). Besides identification of a suitable TAA, trafficking of administered CAR T cells to the tumor is another challenge to effective therapy. Consequently, experimental models to improve innate CAR T-cell trafficking via coexpression of chemokine receptors20 and compartmental/intercavitary administration of CAR T cells are being investigated.21 Perhaps the most notable limitation lies in the dynamic, complex, and often inhibitory tumor microenvironment present in many solid tumor malignancies. For instance, myeloid-derived suppressor cells and tumor-associated macrophages (TAMs) decrease local tryptophan levels in

From the Department of Medicine, Memorial Sloan Kettering Cancer Center, New York, NY; Weill Cornell Medicine, New York, NY; Center for Cell Engineering, and Molecular Pharmacology and Chemistry Program, Memorial Sloan Kettering Cancer Center, New York, NY. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Renier J. Brentjens, MD, PhD, Memorial Sloan Kettering Cancer Center, 1275 York Ave., Box 242, New York, NY 10065; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 193

YEKU, LI, AND BRENTJENS

the tumor microenvironment,22 depriving CAR T cells of an essential amino acid necessary for optimal function. In addition, regulatory T cells, myeloid-derived suppressor cells, and TAMs elaborate inhibitory cytokines such as IL-4, IL-10, leukemia inhibitor factor, and transforming growth factor β—all of which further repress T-cell function.23-25 Strategies aimed at overcoming these limitations are currently areas of intense investigation.

GLIOBLASTOMA

IL-13 receptor α2 (IL-13Rα2) and EGFRIII are two major targets that have been investigated for CAR T-cell therapy against glioblastoma. IL-13Rα2 is overexpressed in more than 50% of glioblastomas, but limited expression on normal brain tissue is retained.34 Importantly, IL-13Rα2 expression has been reported on both stem-like and more differentiated malignant cells, making it a favorable target with the potential to eliminate tumor-initiating cells and prevent tumor recurrence. Kahlon et al35 generated a first-generation IL-13Rα2–specific CAR that redirected human CD8+ cytotoxic T lymphocytes to eradicate established glioblastoma tumor in an orthotopic xenograft model. In a separate study, IL-13Rα2–specific CAR T cells targeted glioma stem–like cancer-initiating cells and abrogated their tumor-initiating activity in mice.36 A phase Ι trial was conducted in three patients with recurrent glioblastoma who received repetitive intracranial infusions of first-generation IL-13Rα2–specific CAR T cells without nonmyeloablative preconditioning.26 Only transient antiglioma responses were observed in two patients. The unsatisfactory response may be explained by poor expansion and persistence of CAR T cells in vivo, because the trial used first-generation CAR T cells. As previously mentioned, first-generation CAR T cells show diminished expansion upon repeated antigen stimulation.37 In a recent case report, a patient showed tumor regression after multiple intracranial infusions of second-generation

KEY POINTS • CAR T-cell therapy has emerged as a promising immunotherapeutic approach for solid tumor malignancies and several promising candidates are in early-phase clinical trials. • Despite tumor and antigen heterogeneity, several TAAs such as MUC-16, GD2, EGFRIII, mesothelin and PSMA have been identified as targets for CAR T-cell therapy. • Clinical responses have been reported in a small subset of solid tumor malignancies; however, increased response rates and responses across a broader range of tumor types are required. • CAR T-cell efficacy is limited by various intrinsic and extrinsic factors, including poor trafficking to tumor site and an immunosuppressive tumor microenvironment. • Further genetic engineering to optimize CAR design (armored CAR T cells) or combinatorial approaches with cytotoxic, targeted therapy, and immunomodulatory agents are currently under investigation. 194 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

IL-13Rα2–specific CAR T cells.38 Interestingly, CAR T cells with intracavitary administration prevented only local tumor recurrence but failed to control tumor progression at distant sites. In contrast, intraventricular infusions resulted in tumor regression in all intracranial and spinal tumors. EGFRIII is a tumor-specific, mutated form of wild-type EGFR and is commonly expressed in glioblastoma. Because of an absence in normal tissues, EGFRIII is ideally suited to minimize on-target, off-tumor toxicity. Multiple preclinical studies demonstrate that EGFRIII-specific CAR T cells recognize and eliminate antigen-positive glioblastoma tumors in vitro and in vivo without cross-reacting with wild-type receptors present on normal tissues.13,39-41

NEUROBLASTOMA

In contrast to glioblastoma, neuroblastoma originates from immature neurons and mostly occurs in infants and young children. Multiple targets, including GD2 and CD171, have been identified and tested for development of CAR T-cell therapy. GD2 is expressed on tumors of neuroectodermal origin, including neuroblastoma and melanoma.42 In a preclinical study, GD2-specific CAR T cells exhibited potent cytotoxicity and cytokine production in response to antigen stimulation.43 A phase I clinical trial by Louis et al27 reported a complete remission rate of 27% (three of 11 patients) in patients treated with first-generation GD2-specifc CAR T cells without lymphodepletion. Furthermore, CAR T-cell persistence was observed for up to 192 weeks in this study.27 CD171 is a surface antigen expressed on many types of cancer, including neuroblastoma. Functionally, CD171 has been reported to enhance tumor cell activity.44 The first CD171-specifc CAR was developed by Gonzalez et al,45 and the engineered T cells displayed robust antitumor activity in vitro. However, subsequent treatment with first-generation GD2-targeting CD8+ lymphocytes in clinical trials failed to control disease progression, and CAR T-cell persistence was inversely correlated with disease burden.28 The authors speculated that the minimal antitumor response was due in part to the lack of coadministration of IL-2, which is especially critical to support the function of first-generation CARs. It is also worthwhile to note that absence of a CD4+ subset in transferred T cells may have compromised function and persistence; emerging data indicate that optimal CAR T-cell efficacy requires both CD4+ and CD8+ compartments.46

Prospects

Efficient CAR T-cell trafficking and localization to the tumor site are prerequisites for optimal antitumor efficacy. This is especially challenging for neuro-oncological malignancies such as glioblastoma because of limited T-cell infiltration in brain. CAR T cells modified to express chemokine receptors, such as chemokine receptor 2, have shown improved trafficking and tissue homing in a neuroblastoma model.47 An alternative strategy is to target the tumor vasculature. Local delivery of tumor necrosis factor α (TNF-α) has been reported to upregulate the expression of adhesion molecules, such

ADOPTIVE T-CELL THERAPY FOR SOLID TUMORS

TABLE 1. Selected Clinical Trials of CAR T-Cell Therapy for Solid Tumors Best Response Data

Trial

Tumor

Target

CAR Design

Phase

Brown et al26

Glioblastoma

IL-13Rα2

CD3ζ

Louis et al27

Neuroblastoma

GD2

CD3ζ

CR, 27%; 19 patients**

Park et al28

Neuroblastoma

CD171

CD3ζ

Feng et al

Non–small cell lung cancer

EGFR

4-1BB/CD3ζ

PR, 18%; SD, 45%; 11 patients

Beatty et al17

Mesothelioma/pancreatic cancer

Mesothelin

4-1BB/CD3ζ

PR, 50%; 2 patients

Junghans et al30

Prostate cancer

PSA

CD3ζ

PR, 40%; 5 patients

Lamers et al

Renal cell carcinoma

CAIX

FcRγ

Kershaw et al32

Ovarian cancer

Folate receptor α

FcRγ

Ahmed et al

Sarcoma

HER2

CD28/CD3ζ

I/II

SD, 24%; 17 patients

NCT02209376

Glioblastoma

EGFRIII

4-1BB/CD3ζ

N/A

NCT01454596

Malignant glioma

EGFRIII

CD28/CD3ζ

I/II

N/A

18,31

Glioblastoma Brain cancer NCT02664363

Glioblastoma

EGFRIII

†

N/A

NCT02208362

Glioblastoma

IL-13Rα2

4-1BB/CD3ζ

N/A

NCT02311621

Neuroblastoma

CD171

4-1BB/CD3ζ

N/A

Ganglioneuroblastoma

CD28/41BB/CD3ζ

NCT01822652

Neuroblastoma

GD2

CD28/OX40/CD3ζ

N/A

NCT01818323

Head and neck cancer

ErbB

CD28/CD3ζ

N/A

NCT02547961

Breast cancer

HER2

CD28/CD3ζ

I/II

N/A

NCT02349724

Lung cancer

CEA

†

N/A

Mesothelin

CD28/CD3ζ

N/A

Mesothelin

4-1BB/CD3ζ

N/A

Mesothelin

†

I/II

N/A

Colorectal cancer Gastric cancer Breast cancer Pancreatic cancer NCT02414269

Malignant pleural disease Mesothelioma metastases Lung cancer Breast cancer

NCT02159716

Pancreatic cancer Ovarian cancer Mesothelioma

NCT01583686

Cervical cancer Pancreatic cancer Ovarian cancer Mesothelioma Lung cancer

NCT01140373

Prostate cancer

PSMA

CD28/CD3ζ

N/A

NCT02498912

Ovarian cancer

Muc-16

CD28/CD3ζ

N/A

NCT00902044

Sarcoma

HER2

CD28/CD3ζ

N/A Continued

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 195

YEKU, LI, AND BRENTJENS

TABLE 1. Selected Clinical Trials of CAR T-cell Therapy for Solid Tumors (Cont'd) Trial

Tumor

Target

CAR Design

Phase

Best Response Data

NCT02107963

Sarcoma

GD2

OX40/CD28/CD3ζ

N/A

Osteosarcoma Neuroblastoma Melanoma *Patients underwent craniotomy before CAR therapy. **Patients with NED before CAR therapy were not included in denominator of responders. †Not listed on clinicaltrials.gov. Abbreviations: CAR, chimeric antigen receptor; CEA, carcinoembryonic antigen; CR, complete response; EGFRIII, EGFR variant III; FcR, fragment crystallizable receptor; GD2, disialoganglioside GD2; IL-13Rα2, interleukin-13 receptor α2; MUC-16, mucin 16; N/A, not applicable; NED, no evidence of disease; PD, progressive disease; PR, partial response; PSA, prostate-specific antigen; PSMA, prostate-specific membrane antigen; SD, stable disease.

as vascular cell adhesion protein 1 and intracellularadhesion molecule 2 on endothelial cells, and to enhance T-cell infiltration.48 Therefore, genetically modifying CAR T cells to secrete TNF-α is one potential approach to overcome this limitation and improve CAR T-cell efficacy. Combining CAR T cells with lenalidomide has been reported to enhance the formation of immune synapses and improve persistency of CAR T cells in vivo,49 providing a rationale for combinatorial approaches for CAR T-cell therapy.

HEAD AND NECK CANCER

A target of particular interest is the ErbB receptor family, which contains four members, designated EGFR (or ErbB-1), ErbB-2 (HER2 or neu), ErbB-3, and ErbB-4.50 ErbB receptors are transmembrane tyrosine kinase proteins that promote cell growth and inhibit apoptosis. Overexpression of these receptors, especially ErbB1 and ErbB2, have been observed in many malignancies, such as head and neck, breast, and lung cancers.51-53 ErbB receptors can exist either in hom*odimeric or heterodimeric configurations,54 and it has recently been appreciated that the transforming potential of the heterodimeric configuration is superior.55 In addition, targeting individual ErbB receptors often results in acquired resistance because of enhanced activity of nontargeted receptors. In light of this, Davies et al56 developed a second-generation CAR that incorporates a chimeric polypeptide, T1E, designed to achieve broad specificity for the ErbB network. ErbB-specific CAR T cells recognized and lysed several ErbB-positive tumor cell lines in vitro. These cell lines showed expression of a broad range of receptor combinations. In SCID-beige mice, CAR T-cell administration led to the eradication of established xenografts derived from ErbB1/2-overexpressing and ErbB2/3-overexpressing tumors. All four ErbB receptors are widely expressed in normal tissues, albeit at lower levels, which could lead to on-target, off-tumor toxicity. Van der Stegen et al57 examined treatment toxicity in SCIDbeige mice after delivery of the ErbB-specific CAR T cells via different routes. Compared with the intraperitoneal route, intratumoral delivery promoted tumor regression without eliciting any cytokine release syndrome. Consideration of intratumoral delivery has been proposed in clinical trials.58 196 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Prospects Multiple mechanisms have been exploited by cells in head and neck squamous cell carcinoma to escape immune surveillance. Data suggest that 55% to 65% of head and neck squamous cell carcinomas express PD-L1, which binds to its cognate receptor PD-1 on T cells, and suppress immune responses.59 The presence of infiltrating regulatory T cells also contributes to the immunosuppressive tumor microenvironment via secretion of IL-10 and transforming growth factor β and via direct inhibition of T cells.60 Therefore, strategies to optimize T-cell efficacy for head and neck squamous cell carcinoma could involve rational combinations of anti-PD-1/PD-L1 antibody with CAR T cells or armored CAR T cells modified to secrete blocking PD-1/PD-L1 scFvs.

BREAST CANCER

HER2 and mesothelin are two TAAs currently under investigation. Amplification of HER2 oncogene leads to uncontrolled cell proliferation and occurs in approximately 20% of breast cancers.61 Globerson-Levin et al62 generated a HER2-specific, second-generation CAR containing CD28 and fragment crystallizable receptor (FcγR) signaling domains and tested its efficacy in a syngeneic mouse mammary tumor model. Transduced T cells exhibited potent cytotoxic capacity and cytokine secretion upon antigen recognition.62 In addition, repeated injections of HER2-directed CAR T cells eliminated spontaneous HER2-positive tumors and enhanced survival in transgenic mice. Mesothelin is a glycoprotein expressed on a broad range of solid tumors, with limited expression on normal tissues.63 Mesothelin expression has been shown to be enriched in triple-negative breast cancer and is associated with poor outcomes.64 Patients with triple-negative breast cancer are not suitable for targeted therapy or hormone therapy, so adoptive transfer of mesothelin-specific CAR T cells offers an alternative option. Tchou et al65 engineered mesothelin-specific CAR T cells and reported a cytolytic capacity against primary breast tumor cells in vitro. However, in vivo antitumor activity was not evaluated in this study.

ADOPTIVE T-CELL THERAPY FOR SOLID TUMORS

Prospects

A major therapeutic challenge to therapy in breast cancer is acquired resistance that results from antigen escape. For instance, under selective pressure, HER2 can undergo proteolysis to cleave the extracellular domain without compromising kinase activity. One approach to overcome this limitation is to use a dual-targeting CAR system, in which engineered T cells coexpress two CARs that recognize two distinct antigens. Redirected T cells can be activated in the presence of either antigen, in essence creating an or-switch, to mitigate antigen-loss escape.66 Alternatively, CAR T cells can be modified to secrete inflammatory cytokines, such as IL-12, or costimulatory ligands, such as 4-1BB ligand, to stimulate an endogenous immune response against tumor cells via epitope spreading.67,68

NON–SMALL CELL LUNG CANCER

Overexpression of EGFR is commonly seen in patients with non–small cell lung cancer, and small molecules inhibiting EGFR kinase activity have shown therapeutic benefits. Feng et al29 reported efficacy of second-generation EGFR-specific CAR T cells that incorporate CD137 and CD3ζ signaling domains. In vitro antitumor efficacy was demonstrated via potent cytotoxicity and by interferon γ (IFN-γ) and IL-2 secretion in response to EGFR-positive lung carcinoma cells. In a phase I clinical study, two of 11 patients with refractory non–small cell lung cancer experienced a partial response after treatment with second-generation EGFR-specific CAR T cells after lymphodepletion. CAR T cells were detected in the peripheral blood of treated patients along with detection of CAR T cells at tumor sites, and eradication of EGFR-positive tumor cells was noted in post-treatment biopsies.29 Mesothelin and carcinoembryonic antigen are also two attractive targets because of their elevated expressions in non–small cell lung cancer.69,70 Multiple preclinical studies have reported antitumor efficacy of mesothelin- and carcinoembryonic antigen–specific CAR T cells against antigen-positive tumors, such as ovarian and liver cancers. However, direct evidence of antitumor efficacy against primary tumor samples or lung cancer cell lines has not been evaluated.71-74

MESOTHELIOMA

In addition to breast and lung cancer, mesothelin is overexpressed on the majority of mesotheliomas. Carpenito et al71 engineered several mesothelin-specific CARs that used different combinations of costimulatory domains and compared their antitumor efficacy. Despite equivalent cytotoxicity in vitro, third-generation CARs, which contained CD137 and CD28 costimulatory domains in tandem, showed marginally superior tumor rejection in a subcutaneous mesothelioma tumor model compared with second-generation CARs that had either costimulatory domain alone. In a separate study, a fully humanized second-generation anti-mesothelin CAR mediated tumor elimination in vitro and in vivo.72 Importantly, CAR T-cell activation was not subverted by soluble tumor-secreted or recombinant mesothelin. This

mitigates the concern that CAR T cells could be blocked or preoccupied by the soluble portion of mesothelin detected in some patients. In addition to CAR development, identifying an optimal route of administration has been explored. Using an orthotopic mesothelioma xenograft model, Adusumilli et al73 showed that intrapleural delivery of second-generation mesothelin-directed CAR T cells vastly outperformed intravenous delivery, requiring 30-fold fewer CAR T cells to induce tumor eradication. In a phase I clinical trial, four patients with advanced mesothelioma or pancreatic cancer were treated with repetitive intravenous infusions of second-generation mesothelin-specific CAR T cells. Moderate antitumor responses were observed, and CAR T cell persistence and trafficking to the tumor site were detected. Interestingly, this study also reported induction of an antitumor humoral immune response after CAR T-cell therapy, evidenced by an elevated antibody response to a variety of tumor-associated proteins. This observation highlights the potential of CAR T-cell therapy to elicit a systemic immune response targeted to a broader range of antigens mediated via epitope spreading.17 One patient experienced anaphylaxis and cardiac arrest after the third infusion on this trial, and this adverse event was believed to be associated with the development of antibodies against the murine-derived scFv.75

Prospects

Like many other solid tumors, lung cancer and mesothelioma possess an immunosuppressive microenvironment. Overexpression of inhibitory molecules, such as PD-L1 and indoleamine 2,3-dioxygenase (IDO) by tumor cells and myeloid-derived suppressor cells have been reported in patients with non–small cell lung cancer or mesothelioma.76-78 Multiple strategies, including additional modification of CAR T cells and combinatorial approaches, can be adopted to overcome these obstacles and enhance CAR T-cell efficacy. For instance, CAR T cells can be engineered to express dominant negative PD-1 receptors79 or anti–PD-1/PD-L1 agents to promote resistance to such inhibition.11 In addition, rational combinations with PD-1/PD-L1 blockade antibody or IDO inhibitors may restore CAR T-cell activity.

OVARIAN CANCER

Several antigens have been exploited as targets for CAR T-cell therapy in ovarian cancer. Barber et al80 engineered a first-generation NKG2D receptor CAR that recognizes the cognate NKG2D ligand expressed on ovarian cancer cell lines and patient-derived primary ovarian cancer samples. In both cell lines and primary samples, these CAR T cells were activated, secreted proinflammatory cytokines, and lysed tumor cells in an NKG2D-dependent fashion. In vitro efficacy and repression of flank-implanted ovarian cancer cells in a xenogeneic model using HER2/neu-directed second-generation CAR T cells also have been reported.81 The Lewis-Y (LeY+) antigen is a carbohydrate molecule that has been shown to be overexpressed on 70% of epithelialderived tumors.82-84 Westwood et al85 designed a CD28ζ asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 197

YEKU, LI, AND BRENTJENS

second-generation CAR directed against LeY+ tumors, one of which included ovarian cancer in an OVCAR-3 tumor model. These CAR T cells showed significantly enhanced IFN-γ production, proliferation, and cytotoxicity when exposed to LeY+ OVCAR-3 cells.85 Furthermore, treatment with LeY+-specific CARs inhibited growth of flank-implanted OVCAR-3 in immunodeficient NOD-SCID mice. Another TAA under development is MUC-16. MUC-16 is a membrane-associated molecule that belongs to the mucin family of glycoproteins.86 The extracellular domain of MUC-16 is cleaved into a soluble antigen (cancer antigen 125 [CA-125]), leaving a retained portion (MUC-16-CD) that can be targeted by adoptively transferred engineered T cells.15 Chekmasova et al15 engineered a second-generation (CD28ζ) MUC-16CD–directed CAR that showed efficacy against OVCAR-3 and patient-derived tumor samples. Armored CAR T-cells which have been engineered to secrete IL-12 directed against MUC-16-CD have been shown to be superior in vitro and in vivo to second-generation MUC-16-CD–directed CARs.87 Similarly, mesothelin, a glycoprotein molecule expressed on pleural, pericardial, and peritoneal cells88 has been explored as a TAA in ovarian cancer. Carpenito et al71 reported notable in vitro cytotoxicity using mesothelin-directed third-generation (CD28/4-1BBζ) CAR T cells. Folate receptor α(FRα) is a cell surface–anchored glycosylphosphatidylinositol molecule89 that is highly expressed on ovarian cancer cells,90 and it has been shown to be predictive of negative outcomes in patients with ovarian cancer.91 On the basis of the preclinical efficacy of folate receptor–directed CAR T cells,92 Kershaw et al32 conducted a phase I clinical trial using first-generation FR-positive–specific CAR T cells with or without exogenous IL-2 in patients with relapsed/ refractory epithelial ovarian cancer. All 14 patients treated in this study had progressive disease. There was no reported decline in CA-125 or antitumor response.32 In one of the cohorts in this study, the adoptively transferred cells were labeled with indium-111 to facilitate in vivo imaging. After intravenous administration, most of the labeled T cells persisted in the lungs, without any evidence of specific localization to the tumor sites. This finding partially explained the decreased systemic persistence and lack of efficacy in this trial.

Prospects

The inhibitory tumor microenvironment in ovarian cancer, including the highly suppressive ascitic microenvironment,93 is an important obstacle that needs to be addressed for CAR T cells to be successful in this disease. One approach is to armor the CAR T cells with soluble cytokines, such as IL-12,21 a proinflammatory cytokine that has been shown to enhance the cytotoxic capability of effector T cells94 and to reprogram dendritic cells and myeloid-derived suppressor cells.95 Potential combinations of checkpoint blockade with second-generation or armored CAR T cells also could be explored as a means to augment CAR T-cell efficacy via recruitment of endogenous effector T cells.96,97 198 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PROSTATE CANCER

Prostate stem-cell antigen and PSMA are two of the most commonly used target antigens for CAR T-cell therapy for prostate cancer. Predominantly found on prostate tissue, prostate stem-cell antigen is a glycosylphosphatidylinositol-anchored antigen located on the cell surface.98 In contrast, PSMA is a type II transmembrane protein that reportedly is present at low levels on the cytosolic/apical surface of normal prostate tissue.99 However, during malignant transformation to prostate adenocarcinoma, it translocates to the extracellular/luminal side of the epithelium.100 Zhong et al8 generated a PSMA-directed third-generation CAR by engineering the 4-1BB receptor costimulatory molecule in tandem with CD28 and CD3ζ (named P28BBζ) and tested its efficacy against a human prostate cancer cell line in an SCID/ beige mouse model. These CAR T cells showed robust proliferation and cytotoxicity in vitro. In tumor-bearing mice, treatment with P28BBζ greatly enhanced survival compared with control mice. Mechanistically, these T cells showed increased intracellular signaling and enhanced production of granzyme, IFN-γ, TNF-α, and granulocyte-macrophage colony-stimulating factor. Hillerdal et al101 also have reported efficacy of a prostate stem-cell antigen–directed third-generation CAR that uses CD28 and OX-40 costimulatory molecules. In addition to robust proliferation, cytokine production, degranulation, and cytotoxicity upon recognition of prostate stem-cell antigen–expressing cells, these CAR T cells also were able to significantly delay subcutaneous tumor growth and prolong survival in nude mice. A phase I clinical trial by Junghans et al102 reported a response rate, by prostate-specific antigen level, of 40% (two of five patients) with a first-generation PSMA-directed CAR after nonmyeloablative preconditioning and concurrent IL-2 administration. In another phase I report, Slovin et al30 reported tolerability and systemic persistence of up to 2 weeks with second-generation PSMA-directed CAR T cells.

Prospects

TAMs have been implicated in prostate cancer.103 Specifically, TAMs are recruited to and infiltrate the tumor stroma in a colony stimulating factor-1 (CSF-1)/CSF-1 receptor (CSF-1R) –dependent fashion,104 where it has been shown to promote tumor and vascular growth105 and to mediate resistance to hormonal therapy.106 In experimental models, clodronate-mediated depletion of TAMs led to notable inhibition of tumor growth.105 One approach to optimize CAR T-cell therapy for prostate cancer might involve preconditioning therapy with either pharmacologic (AZD6495) or antibody-mediated (anti–CSF-1R) depletion of TAMs before CAR T-cell administration. Alternatively, second-generation CAR T cells can be armored via additional genetic modifications to secrete soluble CSF-1R inhibitors.

RENAL CELL CARCINOMA

Carboxy-anhydrase-IX (CA-IX) expression in metastatic renal cell carcinoma has been exploited as a target for adoptive transfer of engineered T cells.18 CA-IX is a metalloprotease

ADOPTIVE T-CELL THERAPY FOR SOLID TUMORS

that reversibly catalyzes the hydration of carbon dioxide.107 Although it is useful as a TAA in renal cell carcinoma, it also is expressed on several normal tissues, such as the gastric mucosa epithelium, small intestine epithelium, duodenum, and biliary tree.108 In addition, expression of CA-IX is inducible in many other tissues under hypoxic conditions.109 In preclinical studies, Weijtens et al110 showed robust cytokine production and cytotoxic activity of first-generation CA-IX–directed engineered T cells against renal carcinoma cells. Lamers at al31 initially treated three patients with CA-IX–positive metastatic clear cell renal cell carcinoma with first-generation CA-IX–specific CAR T cells and exogenous IL-2 administration without nonmyeloablative preconditioning. Two of these patients developed grade 2 to 4 liver enzyme toxicity, and liver biopsies showed cholangitis that involved T-cell infiltration around bile ducts and confirmation of CA-IX expression on the biliary ductal epithelium. Furthermore, all three patients developed antibodies against the murine-derived scFv. To abrogate any more toxicity, the investigators pre-administered unmodified antibody from which the scFv was derived (cG250) to saturate and protect the liver before CAR T cell administration. With this amended approach, Lamers et al18 successfully eliminated treatment-associated hepatoxicity in all four patients who received antibody pretreatment. Curiously, they were unable to detect any human anti-mouse antibodies against the cellular product in patients who underwent antibody pretreatment, which suggests that perhaps the nonspecific inflammation caused by the cholangitis contributed to the generation of human anti-mouse antibodies. Despite CAR T-cell persistence of 3 to 5 weeks, there were no clinical responses.18

Prospects

Myeloid-derived suppressor cells111,112 have been shown to facilitate T-cell suppression via arginase-mediated downregulation of the T-cell receptor ζ chain.113 Increased levels of circulating regulatory T cells also have been reported in patients with renal cell carcinoma114 and are inversely correlated with survival.115 Sunitinib is a U.S. Food and Drug Administration–approved multikinase inhibitor for the treatment of metastatic renal cell carcinoma, and it has been shown to decrease myeloid-derived suppressor cells,116 enhance type-I IFN responses, and decrease regulatory T cells function in patients with renal cell carcinoma.117 Could sunitinib be used as preconditioning and maintenance therapy after CAR T-cell administration? This hypothesis could readily be subject to testing with a second-generation or armored CARs in a syngeneic model of metastatic renal cell carcinoma.118

SARCOMA

Although sarcomas represent a heterogeneous group of mesenchymal-derived neoplasms, there has been some success in identifying TAAs that are expressed across different sarcoma subtypes. Ahmed et al119 exploited the expression of HER2 on osteosarcomas by engineering a second-generation HER2-directed CAR construct. These HER2-specific

T cells showed robust cytokine production, proliferation, and cytotoxicity in vivo. Adoptive transfer of these genetically modified T cells effectively treated both localized and metastatic osteosarcoma in SCID mice. Second-generation (CD28ζ) NKG2D ligand-directed CAR T cells also have shown efficacy in preclinical in vitro models of Ewing sarcoma.120 Another approach reported by Huang et al121 involved generation of an anti–IL-11 receptor α chain (IL-11Rα) second-generation CAR. IL-11Rα expression has been reported on multiple tumor types, including osteosarcoma,122 prostate cancer,123 and breast cancer.124 Signaling via the IL-11/ IL-11Rα pathway has been shown, among many other things, to promote osteoclastogenesis.125,126 IL-11Rα–specific CAR T cells were effective against both primary tumors and pulmonary metastasis in a nude mouse model of osteosarcoma.121 In a phase I/II trial by Ahmed et al,33 19 patients with HER2-positive sarcoma were treated with second-generation HER2-specific CAR T cells without nonmyeloablative preconditioning. Adoptively transferred cells were detectable for up to 9 months in a fraction of treated patients. Furthermore, in patients who underwent metastatectomy 9 to 15 weeks after CAR T-cell therapy, HER2-specific CAR T cells were detected in the tumor samples by qualitative polymerase chain reaction.33 Of the 17 evaluable patients, four had stable disease for as long as 12 weeks to 14 months. Three patients who underwent metastatectomy after CAR T-cell therapy remained in remission for up to 16 months.

Prospects

The importance of angiogenesis and vascular invasion in sarcoma has been well described.127 In addition, the presence of M2-polarized TAMs has been reported, and these cells also could contribute to pathologic vasculogenesis via VEGF production.128 Could CAR T cells be additionally modified to secrete soluble VEGF inhibitors? Perhaps they could be used in combination with anti-VEGF antibodies or multikinase inhibitors like pazopanib or sunitinib? Preconditioning or combination with immune-modifying agents, such as trabectedin129 or mifamuritide, which act against monocytes/ macrophages, could be explored as a means to optimize CAR T-cell efficacy for this disease.

CONCLUSION

Despite enthusiasm for adoptive immunotherapy, many obstacles must be addressed before CAR T-cell therapy joins the armamentarium for management of solid tumors. In tumor types that have more than one TAA, there is the question of which is the optimal target to minimize tumor escape via antigen loss/downregulation. When more than one TAA is expressed, could scFvs against both antigens be engineered in an or-activation or and-activation configuration to combat tumor heterogenicity or to improve safety, respectively? The prerequisite for nonmyeloablative preconditioning also must be rigorously assessed in syngeneic solid tumor models and clinical trials. There might be a hypothetical benefit to remodeling the endogenous lymphoid populations in anticipation of activation/recruitment asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 199

YEKU, LI, AND BRENTJENS

by specifically armored CAR T cells, but this remains to be tested. Appropriate preclinical models and mechanisms of efficacy and resistance to CAR T-cell therapy also should be explored, ideally before clinical development. Driven mostly by the importance of demonstrating antitumor efficacy against human cancer cell lines, the clear majority of preclinical CAR T cell validation experiments have been in the context of SCID/beige or other immunodeficient tumor models. These models potentially could underestimate the immunomodulatory effect of the endogenous immune systems of the hosts and the effects of the immunosuppressive tumor microenvironment on adoptively transferred T cells. Consequently, more effort is being directed at understanding the interaction of the tumor microenvironment and the endogenous immune system in immunocompetent mouse models in addition to the prerequisite xenogeneic research. The route of CAR T-cell administration also could be tailored to each solid tumor malignancy according to what is known about each tumor’s biology. For example, clinical trials of intrapleural and intraperitoneal administration of CAR T cells for mesothelioma and ovarian cancer, respectively, are in progress. Lingering issues with toxicities in the form of

cytokine release syndrome, neurotoxicity, and off-tumor cytotoxicity also are being investigated. Ultimately, knowledge of how best to mitigate these toxicities, coupled with rational combinations of chemotherapy, surgery, radiotherapy, or immunomodulators, will pave the way for the next breakthroughs in CAR T-cell therapy for solid tumor malignancies.

Acknowledgment

Oladapo Yeku and Xinghuo Li contributed equally to this article. This work was funded in part by the following: National Institutes of Health Grants No. R01CA138738-05, PO1CA059350, PO1CA190174-01; the Ovarian Cancer Research Fund Grant No. 327501; Memorial Sloan Kettering T32 Investigational Therapeutics Training Program Grant No. T32-CA009207; the annual Terry Fox Run for Cancer Research in New York, NY, Grant No. 29410; Kate’s Team; Carson Family Charitable Trust Grant No. 10171; William Lawrence and Blanche Hughes Foundation Grant No. 10251; Emerald Foundation Grant No. 11625; and the Experimental Therapeutics Center of Memorial Sloan Kettering Cancer Center Grant No. 13072.

References 1. Sadelain M, Brentjens R, Rivière I. The basic principles of chimeric antigen receptor design. Cancer Discov. 2013;3:388-398. 2. Stancovski I, Schindler DG, Waks T, et al. Targeting of T lymphocytes to Neu/HER2-expressing cells using chimeric single chain Fv receptors. J Immunol. 1993;151:6577-6582. 3. Brentjens RJ, Santos E, Nikhamin Y, et al. Genetically targeted T cells eradicate systemic acute lymphoblastic leukemia xenografts. Clin Cancer Res. 2007;13:5426-5435. 4. Kalos M, Levine BL, Porter DL, et al. T cells with chimeric antigen receptors have potent antitumor effects and can establish memory in patients with advanced leukemia. Sci Transl Med. 2011;3:95ra73. 5. Chambers CA, Allison JP. Co-stimulation in T cell responses. Curr Opin Immunol. 1997;9:396-404.

11. Suarez ER, Chang K, Sun J, et al. Chimeric antigen receptor T cells secreting anti–PD-L1 antibodies more effectively regress renal cell carcinoma in a humanized mouse model. Oncotarget. 2016;7:3434134355. 12. Morgan RA, Yang JC, Kitano M, et al. Case report of a serious adverse event following the administration of T cells transduced with a chimeric antigen receptor recognizing ERBB2. Mol Ther. 2010;18:843-851. 13. Johnson LA, Scholler J, Ohkuri T, et al. Rational development and characterization of humanized anti-EGFR variant III chimeric antigen receptor T cells for glioblastoma. Sci Transl Med. 2015;7:275ra22. 14. Prapa M, Caldrer S, Spano C, et al. A novel anti-GD2/4-1BB chimeric antigen receptor triggers neuroblastoma cell killing. Oncotarget. 2015;6:24884-24894.

6. Savoldo B, Ramos CA, Liu E, et al. CD28 costimulation improves expansion and persistence of chimeric antigen receptor-modified T cells in lymphoma patients. J Clin Invest. 2011;121:1822-1826.

15. Chekmasova AA, Rao TD, Nikhamin Y, et al. Successful eradication of established peritoneal ovarian tumors in SCID-beige mice following adoptive transfer of T cells genetically targeted to the MUC16 antigen. Clin Cancer Res. 2010;16:3594-3606.

7. Zhao Y, Wang QJ, Yang S, et al. A herceptin-based chimeric antigen receptor with modified signaling domains leads to enhanced survival of transduced T lymphocytes and antitumor activity. J Immunol. 2009;183:5563-5574.

16. Chmielewski M, Hahn O, Rappl G, et al. T cells that target carcinoembryonic antigen eradicate orthotopic pancreatic carcinomas without inducing autoimmune colitis in mice. Gastroenterology. 2012;143:1095-1107.e2.

8. Zhong XS, Matsush*ta M, Plotkin J, et al. Chimeric antigen receptors combining 4-1BB and CD28 signaling domains augment PI3kinase/ AKT/Bcl-XL activation and CD8+ T cell-mediated tumor eradication. Mol Ther. 2010;18:413-420.

17. Beatty GL, Haas AR, Maus MV, et al. Mesothelin-specific chimeric antigen receptor mRNA-engineered T cells induce anti-tumor activity in solid malignancies. Cancer Immunol Res. 2014;2:112-120.

9. Wilkie S, Picco G, Foster J, et al. Retargeting of human T cells to tumorassociated MUC1: the evolution of a chimeric antigen receptor. J Immunol. 2008;180:4901-4909.

18. Lamers CH, Sleijfer S, van Steenbergen S, et al. Treatment of metastatic renal cell carcinoma with CAIX CAR-engineered T cells: clinical evaluation and management of on-target toxicity. Mol Ther. 2013;21:904-912.

10. Yeku OO, Brentjens RJ. Armored CAR T-cells: utilizing cytokines and pro-inflammatory ligands to enhance CAR T-cell anti-tumour efficacy. Biochem Soc Trans. 2016;44:412-418.

19. Zuccolotto G, Fracasso G, Merlo A, et al. PSMA-specific CAR-engineered T cells eradicate disseminated prostate cancer in preclinical models. PLoS One. 2014;9:e109427.

200 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

ADOPTIVE T-CELL THERAPY FOR SOLID TUMORS

20. Moon EK, Carpenito C, Sun J, et al. Expression of a functional CCR2 receptor enhances tumor localization and tumor eradication by retargeted human T cells expressing a mesothelin-specific chimeric antibody receptor. Clin Cancer Res. 2011;17:4719-4730.

36. Brown CE, Starr R, Aguilar B, et al. Stem-like tumor-initiating cells isolated from IL13Rα2 expressing gliomas are targeted and killed by IL13-zetakine-redirected T Cells. Clin Cancer Res. 2012;18: 2199-2209.

21. Koneru M, O’Cearbhaill R, Pendharkar S, et al. A phase I clinical trial of adoptive T cell therapy using IL-12 secreting MUC-16(ecto) directed chimeric antigen receptors for recurrent ovarian cancer. J Transl Med. 2015;13:102.

37. Hombach A, Wieczarkowiecz A, Marquardt T, et al. Tumor-specific T cell activation by recombinant immunoreceptors: CD3 zeta signaling and CD28 costimulation are simultaneously required for efficient IL-2 secretion and can be integrated into one combined CD28/CD3 zeta signaling receptor molecule. J Immunol. 2001;167:6123-6131.

22. Ninomiya S, Narala N, Huye L, et al. Tumor indoleamine 2,3-dioxygenase (IDO) inhibits CD19-CAR T cells and is downregulated by lymphodepleting drugs. Blood. 2015;125:3905-3916. 23. Li X, Ye F, Chen H, et al. Human ovarian carcinoma cells generate CD4(+)CD25(+) regulatory T cells from peripheral CD4(+)CD25(−) T cells through secreting TGF-beta. Cancer Lett. 2007;253:144-153. 24. Chen LL, Ye F, Lü WG, et al. Evaluation of immune inhibitory cytokine profiles in epithelial ovarian carcinoma. J Obstet Gynaecol Res. 2009;35:212-218. 25. Zhao X, Ye F, Chen L, et al. Human epithelial ovarian carcinoma cellderived cytokines cooperatively induce activated CD4+CD25−CD45RA+ naïve T cells to express forkhead box protein 3 and exhibit suppressive ability in vitro. Cancer Sci. 2009;100:2143-2151. 26. Brown CE, Badie B, Barish ME, et al. Bioactivity and safety of IL13Rα2redirected chimeric antigen receptor CD8+ T cells in patients with recurrent glioblastoma. Clin Cancer Res. 2015;21:4062-4072. 27. Louis CU, Savoldo B, Dotti G, et al. Antitumor activity and long-term fate of chimeric antigen receptor-positive T cells in patients with neuroblastoma. Blood. 2011;118:6050-6056. 28. Park JR, Digiusto DL, Slovak M, et al. Adoptive transfer of chimeric antigen receptor re-directed cytolytic T lymphocyte clones in patients with neuroblastoma. Mol Ther. 2007;15:825-833. 29. Feng K, Guo Y, Dai H, et al. Chimeric antigen receptor-modified T cells for the immunotherapy of patients with EGFR-expressing advanced relapsed/refractory non-small cell lung cancer. Sci China Life Sci. 2016;59:468-479. 30. Slovin SF, Wang X, Hullings M, et al Chimeric antigen receptor (CAR+) modified T cells targeting prostate specific membrane antigen (PSMA) in patients (pts) with castrate metastatic prostate cancer (CMPC). J Clin Oncol. 2013;31 (suppl; abstr TPS3115). 31. Lamers CH, Sleijfer S, Vulto AG, et al. Treatment of metastatic renal cell carcinoma with autologous T-lymphocytes genetically retargeted against carbonic anhydrase IX: first clinical experience. J Clin Oncol. 2006;24:e20-e22. 32. Kershaw MH, Westwood JA, Parker LL, et al. A phase I study on adoptive immunotherapy using gene-modified T cells for ovarian cancer. Clin Cancer Res. 2006;12:6106-6115. 33. Ahmed N, Brawley VS, Hegde M, et al. Human epidermal growth factor receptor 2 (HER2) –specific chimeric antigen receptor-modified T cells for the immunotherapy of HER2-positive sarcoma. J Clin Oncol. 2015;33:1688-1696. 34. Brown CE, Warden CD, Starr R, et al. Glioma IL13Rα2 is associated with mesenchymal signature gene expression and poor patient prognosis. PLoS One. 2013;8:e77769. 35. Kahlon KS, Brown C, Cooper LJ, et al. Specific recognition and killing of glioblastoma multiforme by interleukin 13-zetakine redirected cytolytic T cells. Cancer Res. 2004;64:9160-9166.

38. Brown CE, Alizadeh D, Starr R, et al. Regression of glioblastoma after chimeric antigen receptor T-cell therapy. N Engl J Med. 2016;375:25612569. 39. Morgan RA, Johnson LA, Davis JL, et al. Recognition of glioma stem cells by genetically modified T cells targeting EGFRvIII and development of adoptive cell therapy for glioma. Hum Gene Ther. 2012;23:1043-1053. 40. Ohno M, Ohkuri T, Kosaka A, et al. Expression of miR-17-92 enhances anti-tumor activity of T-cells transduced with the anti-EGFRvIII chimeric antigen receptor in mice bearing human GBM xenografts. J Immunother Cancer. 2013;1:21. 41. Choi BD, Suryadevara CM, Gedeon PC, et al. Intracerebral delivery of a third generation EGFRvIII-specific chimeric antigen receptor is efficacious against human glioma. J Clin Neurosci. 2014;21:189-190. 42. Schulz G, Cheresh DA, Varki NM, et al. Detection of ganglioside GD2 in tumor tissues and sera of neuroblastoma patients. Cancer Res. 1984;44:5914-5920. 43. Rossig C, Bollard CM, Nuchtern JG, et al. Epstein-Barr virus-specific human T lymphocytes expressing antitumor chimeric T-cell receptors: potential for improved immunotherapy. Blood. 2002;99:2009-2016. 44. Gavert N, Conacci-Sorrell M, Gast D, et al. L1, a novel target of betacatenin signaling, transforms cells and is expressed at the invasive front of colon cancers. J Cell Biol. 2005;168:633-642. 45. Gonzalez S, Naranjo A, Serrano LM, et al. Genetic engineering of cytolytic T lymphocytes for adoptive T-cell therapy of neuroblastoma. J Gene Med. 2004;6:704-711. 46. Turtle CJ, Hanafi LA, Berger C, et al. CD19 CAR-T cells of defined CD4+:CD8+ composition in adult B cell ALL patients. J Clin Invest. 2016;126:2123-2138. 47. Craddock JA, Lu A, Bear A, et al. Enhanced tumor trafficking of GD2 chimeric antigen receptor T cells by expression of the chemokine receptor CCR2b. J Immunother. 2010;33:780-788. 48. Calcinotto A, Grioni M, Jachetti E, et al. Targeting TNF-α to neoangiogenic vessels enhances lymphocyte infiltration in tumors and increases the therapeutic potential of immunotherapy. J Immunol. 2012;188:2687-2694. 49. Kuramitsu S, Ohno M, Ohka F, et al. Lenalidomide enhances the function of chimeric antigen receptor T cells against the epidermal growth factor receptor variant III by enhancing immune synapses. Cancer Gene Ther. 2015;22:487-495. 50. Hynes NE, MacDonald G. ErbB receptors and signaling pathways in cancer. Curr Opin Cell Biol. 2009;21:177-184. 51. Rogers SJ, Harrington KJ, Rhys-Evans P, et al. Biological significance of c-erbB family oncogenes in head and neck cancer. Cancer Metastasis Rev. 2005;24:47-69. 52. Foley J, Nickerson NK, Nam S, et al. EGFR signaling in breast cancer: bad to the bone. Semin Cell Dev Biol. 2010;21:951-960.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 201

YEKU, LI, AND BRENTJENS

53. Hirsch FR, Varella-Garcia M, Cappuzzo F. Predictive value of EGFR and HER2 overexpression in advanced non–small cell lung cancer. Oncogene. 2009;28 (Suppl 1):S32-S37. 54. Holbro T, Civenni G, Hynes NE. The ErbB receptors and their role in cancer progression. Exp Cell Res. 2003;284:99-110. 55. Olayioye MA, Neve RM, Lane HA, et al. The ErbB signaling network: receptor heterodimerization in development and cancer. EMBO J. 2000;19:3159-3167. 56. Davies DM, Foster J, Van Der Stegen SJ, et al. Flexible targeting of ErbB dimers that drive tumorigenesis by using genetically engineered T cells. Mol Med. 2012;18:565-576. 57. van der Stegen SJ, Davies DM, Wilkie S, et al. Preclinical in vivo modeling of cytokine release syndrome induced by ErbB-retargeted human T cells: identifying a window of therapeutic opportunity? J Immunol. 2013;191:4589-4598. 58. van Schalkwyk MC, Papa SE, Jeannon JP, et al. Design of a phase I clinical trial to evaluate intratumoral delivery of ErbB-targeted chimeric antigen receptor T-cells in locally advanced or recurrent head and neck cancer. Hum Gene Ther Clin Dev. 2013;24: 134-142. 59. Ferris RL. Immunology and immunotherapy of head and neck cancer. J Clin Oncol. 2015;33:3293-3304. 60. Albers AE, Ferris RL, Kim GG, et al. Immune responses to p53 in patients with cancer: enrichment in tetramer-positive p53 peptidespecific T cells and regulatory T cells at tumor sites. Cancer Immunol Immunother. 2005;54:1072-1081. 61. Witton CJ, Reeves JR, Going JJ, et al. Expression of the HER1-4 family of receptor tyrosine kinases in breast cancer. J Pathol. 2003;200:290297. 62. Globerson-Levin A, Waks T, Eshhar Z. Elimination of progressive mammary cancer by repeated administrations of chimeric antigen receptor-modified T cells. Mol Ther. 2014;22:1029-1038. 63. Morello A, Sadelain M, Adusumilli PS. Mesothelin-targeted CARs: driving T cells to solid tumors. Cancer Discov. 2016;6:133-146. 64. Li YR, Xian RR, Ziober A, et al. Mesothelin expression is associated with poor outcomes in breast cancer. Breast Cancer Res Treat. 2014;147:675-684. 65. Tchou J, Wang LC, Selven B, et al. Mesothelin, a novel immunotherapy target for triple negative breast cancer. Breast Cancer Res Treat. 2012;133:799-804. 66. Ruella M, Barrett DM, Kenderian SS, et al. Dual CD19 and CD123 targeting prevents antigen-loss relapses after CD19-directed immunotherapies. J Clin Invest. 2016;126:3814-3826. 67. Pegram HJ, Lee JC, Hayman EG, et al. Tumor-targeted T cells modified to secrete IL-12 eradicate systemic tumors without need for prior conditioning. Blood. 2012;119:4133-4141. 68. Zhao Z, Condomines M, van der Stegen SJ, et al. Structural design of engineered costimulation determines tumor rejection kinetics and persistence of CAR T cells. Cancer Cell. 2015;28:415-428. 69. Kachala SS, Bograd AJ, Villena-Vargas J, et al. Mesothelin overexpression is a marker of tumor aggressiveness and is associated with reduced recurrence-free and overall survival in early-stage lung adenocarcinoma. Clin Cancer Res. 2014;20:1020-1028. 70. Grunnet M, Sorensen JB. Carcinoembryonic antigen (CEA) as tumor marker in lung cancer. Lung Cancer. 2012;76:138-143.

202 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

71. Carpenito C, Milone MC, Hassan R, et al. Control of large, established tumor xenografts with genetically retargeted human T cells containing CD28 and CD137 domains. Proc Natl Acad Sci USA. 2009;106:33603365. 72. Lanitis E, Poussin M, Hagemann IS, et al. Redirected antitumor activity of primary human lymphocytes transduced with a fully human antimesothelin chimeric receptor. Mol Ther. 2012;20:633-643. 73. Adusumilli PS, Cherkassky L, Villena-Vargas J, et al. Regional delivery of mesothelin-targeted CAR T cell therapy generates potent and long-lasting CD4-dependent tumor immunity. Sci Transl Med. 2014;6:261ra151. 74. Emtage PC, Lo AS, Gomes EM, et al. Second-generation anticarcinoembryonic antigen designer T cells resist activation-induced cell death, proliferate on tumor contact, secrete cytokines, and exhibit superior antitumor activity in vivo: a preclinical evaluation. Clin Cancer Res. 2008;14:8112-8122. 75. Maus MV, Haas AR, Beatty GL, et al. T cells expressing chimeric antigen receptors can cause anaphylaxis in humans. Cancer Immunol Res. 2013;1:26-31. 76. Konishi J, Yamazaki K, Azuma M, et al. B7-H1 expression on non– small cell lung cancer cells and its relationship with tumor-infiltrating lymphocytes and their PD-1 expression. Clin Cancer Res. 2004;10:50945100. 77. Mansfield AS, Roden AC, Peikert T, et al. B7-H1 expression in malignant pleural mesothelioma is associated with sarcomatoid histology and poor prognosis. J Thorac Oncol. 2014;9:1036-1040. 78. Suzuki Y, Suda T, Furuhashi K, et al. Increased serum kynurenine/ tryptophan ratio correlates with disease progression in lung cancer. Lung Cancer. 2010;67:361-365. 79. Cherkassky L, Morello A, Villena-Vargas J, et al. Human CAR T cells with cell-intrinsic PD-1 checkpoint blockade resist tumor-mediated inhibition. J Clin Invest. 2016;126:3130-3144. 80. Barber A, Zhang T, DeMars LR, et al. Chimeric NKG2D receptorbearing T cells as immunotherapy for ovarian cancer. Cancer Res. 2007;67:5003-5008. 81. Yoon SH, Lee JM, Cho HI, et al. Adoptive immunotherapy using human peripheral blood lymphocytes transferred with RNA encoding Her-2/ neu-specific chimeric immune receptor in ovarian cancer xenograft model. Cancer Gene Ther. 2009;16:489-497. 82. Zhang S, Zhang HS, Cordon-Cardo C, et al. Selection of tumor antigens as targets for immune attack using immunohistochemistry: II. Blood group-related antigens. Int J Cancer. 1997;73:50-56. 83. Miyake M, Taki T, Hitomi S, et al. Correlation of expression of H/Le(y)/ Le(b) antigens with survival in patients with carcinoma of the lung. N Engl J Med. 1992;327:14-18. 84. Sakamoto J, Furukawa K, Cordon-Cardo C, et al. Expression of Lewis-a, Lewis-b, X, and Y blood group antigens in human colonic tumors and normal tissue and in human tumor-derived cell lines. Cancer Res. 1986;46:1553-1561. 85. Westwood JA, Smyth MJ, Teng MW, et al. Adoptive transfer of T cells modified with a humanized chimeric receptor gene inhibits growth of Lewis-Y-expressing tumors in mice. Proc Natl Acad Sci USA. 2005;102:19051-19056. 86. Dharma Rao T, Park KJ, Smith-Jones P, et al. Novel monoclonal antibodies against the proximal (carboxy-terminal) portions of MUC16. Appl Immunohistochem Mol Morphol. 2010;18:462-472.

ADOPTIVE T-CELL THERAPY FOR SOLID TUMORS

87. Koneru M, Purdon TJ, Spriggs D, et al. IL-12 secreting tumor-targeted chimeric antigen receptor T cells eradicate ovarian tumors in vivo. OncoImmunology. 2015;4:e994446.

106. Zhu P, Baek SH, Bourk EM, et al. Macrophage/cancer cell interactions mediate hormone resistance by a nuclear receptor derepression pathway. Cell. 2006;124:615-629.

88. Chang K, Pastan I. Molecular cloning of mesothelin, a differentiation antigen present on mesothelium, mesotheliomas, and ovarian cancers. Proc Natl Acad Sci USA. 1996;93:136-140.

107. Tafreshi NK, Lloyd MC, Bui MM, et al. Carbonic anhydrase IX as an imaging and therapeutic target for tumors and metastases. Subcell Biochem. 2014;75:221-254.

89. Elnakat H, Ratnam M. Distribution, functionality and gene regulation of folate receptor isoforms: implications in targeted therapy. Adv Drug Deliv Rev. 2004;56:1067-1084.

108. Pastoreková S, Parkkila S, Parkkila AK, et al. Carbonic anhydrase IX, MN/CA IX: analysis of stomach complementary DNA sequence and expression in human and rat alimentary tracts. Gastroenterology. 1997;112:398-408.

90. Toffoli G, Cernigoi C, Russo A, et al. Overexpression of folate binding protein in ovarian cancers. Int J Cancer. 1997;74:193-198. 91. Toffoli G, Russo A, Gallo A, et al. Expression of folate binding protein as a prognostic factor for response to platinum-containing chemotherapy and survival in human ovarian cancer. Int J Cancer. 1998;79:121-126. 92. Hwu P, Yang JC, Cowherd R, et al. In vivo antitumor activity of T cells redirected with chimeric antibody/T-cell receptor genes. Cancer Res. 1995;55:3369-3373. 93. Kim S, Kim B, Song YS. Ascites modulates cancer cell behavior, contributing to tumor heterogeneity in ovarian cancer. Cancer Sci. 2016;107:1173-1178. 94. Zhao J, Zhao J, Perlman S. Differential effects of IL-12 on Tregs and non-Treg T cells: roles of IFN-γ, IL-2 and IL-2R. PLoS One. 2012;7:e46241. 95. Kerkar SP, Goldszmid RS, Muranski P, et al. IL-12 triggers a programmatic change in dysfunctional myeloid-derived cells within mouse tumors. J Clin Invest. 2011;121:4746-4757.

109. Ivanov S, Liao SY, Ivanova A, et al. Expression of hypoxia-inducible cellsurface transmembrane carbonic anhydrases in human cancer. Am J Pathol. 2001;158:905-919. 110. Weijtens ME, Willemsen RA, Valerio D, et al. Single chain Ig/gamma gene-redirected human T lymphocytes produce cytokines, specifically lyse tumor cells, and recycle lytic capacity. J Immunol. 1996;157:836843. 111. Walter S, Weinschenk T, Stenzl A, et al. Multipeptide immune response to cancer vaccine IMA901 after single-dose cyclophosphamide associates with longer patient survival. Nat Med. 2012;18:1254-1261. 112. Finke JH, Rayman PA, Ko JS, et al. Modification of the tumor microenvironment as a novel target of renal cell carcinoma therapeutics. Cancer J. 2013;19:353-364. 113. Rodriguez PC, Zea AH, Culotta KS, et al. Regulation of T cell receptor CD3zeta chain expression by L-arginine. J Biol Chem. 2002;277:2112321129.

96. Oelkrug C, Ramage JM. Enhancement of T cell recruitment and infiltration into tumours. Clin Exp Immunol. 2014;178:1-8.

114. Cesana GC, DeRaffele G, Cohen S, et al. Characterization of CD4+CD25+ regulatory T cells in patients treated with high-dose interleukin-2 for metastatic melanoma or renal cell carcinoma. J Clin Oncol. 2006;24:1169-1177.

97. Esposito A, Criscitiello C, Curigliano G. Immune checkpoint inhibitors with radiotherapy and locoregional treatment: synergism and potential clinical implications. Curr Opin Oncol. 2015;27:445-451.

115. Siddiqui SA, Frigola X, Bonne-Annee S, et al. Tumor-infiltrating Foxp3CD4+CD25+ T cells predict poor survival in renal cell carcinoma. Clin Cancer Res. 2007;13:2075-2081.

98. Tricoli JV, Schoenfeldt M, Conley BA. Detection of prostate cancer and predicting progression: current and future diagnostic markers. Clin Cancer Res. 2004;10:3943-3953.

116. Ko JS, Zea AH, Rini BI, et al. Sunitinib mediates reversal of myeloidderived suppressor cell accumulation in renal cell carcinoma patients. Clin Cancer Res. 2009;15:2148-2157.

99. Leek J, Lench N, Maraj B, et al. Prostate-specific membrane antigen: evidence for the existence of a second related human gene. Br J Cancer. 1995;72:583-588.

117. Finke JH, Rini B, Ireland J, et al. Sunitinib reverses type-1 immune suppression and decreases T-regulatory cells in renal cell carcinoma patients. Clin Cancer Res. 2008;14:6674-6682.

100. DeMarzo AM, Nelson WG, Isaacs WB, et al. Pathological and molecular aspects of prostate cancer. Lancet. 2003;361:955-964.

118. Tracz A, Mastri M, Lee CR, et al. Modeling spontaneous metastatic renal cell carcinoma (mRCC) in mice following nephrectomy. J Vis Exp. 2014;(86): 51485.

101. Hillerdal V, Ramachandran M, Leja J, et al. Systemic treatment with CAR-engineered T cells against PSCA delays subcutaneous tumor growth and prolongs survival of mice. BMC Cancer. 2014;14:30. 102. Junghans RP, Ma Q, Rathore R, et al. Phase I trial of anti-PSMA designer CAR T Cells in prostate cancer: possible role for interacting interleukin 2 T cell pharmacodynamics as a determinant of clinical response. Prostate. 2016;76:1257-1270. 103. Qian BZ, Pollard JW. Macrophage diversity enhances tumor progression and metastasis. Cell. 2010;141:39-51. 104. Ostrand-Rosenberg S, Sinha P. Myeloid-derived suppressor cells: linking inflammation and cancer. J Immunol. 2009;182:4499-4506. 105. Halin S, Rudolfsson SH, Van Rooijen N, et al. Extratumoral macrophages promote tumor and vascular growth in an orthotopic rat prostate tumor model. Neoplasia. 2009;11:177-186.

119. Ahmed N, Salsman VS, Yvon E, et al. Immunotherapy for osteosarcoma: genetic modification of T cells overcomes low levels of tumor antigen expression. Mol Ther. 2009;17:1779-1787. 120. Lehner M, Götz G, Proff J, et al. Redirecting T cells to Ewing’s sarcoma family of tumors by a chimeric NKG2D receptor expressed by lentiviral transduction or mRNA transfection. PLoS One. 2012;7:e31210. 121. Huang G, Yu L, Cooper LJ, et al. Genetically modified T cells targeting interleukin-11 receptor α-chain kill human osteosarcoma cells and induce the regression of established osteosarcoma lung metastases. Cancer Res. 2012;72:271-281. 122. Lewis VO, Ozawa MG, Deavers MT, et al. The interleukin-11 receptor alpha as a candidate ligand-directed target in osteosarcoma: consistent data from cell lines, orthotopic models, and human tumor samples. Cancer Res. 2009;69:1995-1999.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 203

YEKU, LI, AND BRENTJENS

123. Campbell CL, Jiang Z, Savarese DM, et al. Increased expression of the interleukin-11 receptor and evidence of STAT3 activation in prostate carcinoma. Am J Pathol. 2001;158:25-32. 124. Hanavadi S, Martin TA, Watkins G, et al. Expression of interleukin 11 and its receptor and their prognostic value in human breast cancer. Ann Surg Oncol. 2006;13:802-808.

127. Engellau J, Bendahl PO, Persson A, et al. Improved prognostication in soft tissue sarcoma: independent information from vascular invasion, necrosis, growth pattern, and immunostaining using whole-tumor sections and tissue microarrays. Hum Pathol. 2005;36:994-1002.

125. Schwertschlag US, Trepicchio WL, Dykstra KH, et al. Hematopoietic, immunomodulatory and epithelial effects of interleukin-11. Leukemia. 1999;13:1307-1315.

128. Castelli C, Rivoltini L, Rodolfo M, et al. Modulation of the myeloid compartment of the immune system by angiogenic- and kinase inhibitor-targeted anti-cancer therapies. Cancer Immunol Immunother. 2015;64:83-89.

126. Teramura M, Kobayashi S, Yoshinaga K, et al. Effect of interleukin 11 on normal and pathological thrombopoiesis. Cancer Chemother Pharmacol. 1996;38 (Suppl):S99-S102.

129. Germano G, Frapolli R, Belgiovine C, et al. Role of macrophage targeting in the antitumor activity of trabectedin. Cancer Cell. 2013;23:249-262.

204 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

BIOMARKERS FOR CHECKPOINT INHIBITION

Biomarkers for Checkpoint Inhibition Jeffrey S. Weber, MD, PhD OVERVIEW The identification of predictive biomarkers for the benefit of cancer immunotherapy is the holy grail of the burgeoning immunotherapy field. Recent work has shown that there are a core of concepts that establish the presence of an immune cell–infiltrate, an inflammatory signature of the tumor microenvironment, and the availability of target antigens defined by mutated neoantigens, as critical for the success of the checkpoint blockade. Genetic analyses have shown that resistance to PD-1 blockade, either innate or adaptive, may be due to existing or de novo mutations in signaling pathways critical for T-cell function in a modest proportion of cases. Major hurdles in the field that remain to be overcome are the difficulty of obtaining tumor biopsies for biomarker assessment, the heterogeneity of biomarker expression within tumors and within different tumors from the same patient, and the inducibility of some biomarkers by disease-related processes. Although assessment of peripheral blood or serum biomarkers would be ideal, few data suggest that they would reliably predict outcome with checkpoint blockade. Ultimately, some amalgamated biomarker that includes tumor and host factors will be required to predict which patients are likely to benefit from, or be resistant to, the effects of checkpoint inhibition.

he field of cancer immunotherapy has expanded enormously over the last decade, with many new trials and multiple new approvals of checkpoint inhibitors for solid tumors in 2016. Ipilimumab, nivolumab, and pembrolizumab have become mainstays of treatment for metastatic melanoma,1-6 lung cancer,7-10 and other solid tumors, and numerous combination trials are underway in efforts to optimize the use of checkpoint inhibition. Recently, atezolizumab, the first anti–PD-L1 antibody to be approved by the U.S. Food and Drug Administration for the treatment of platinum-resistant bladder cancer.11 Lessons learned from evaluating biomarkers of toxicity and outcome in patients with melanoma will undoubtedly help accelerate the development of checkpoint inhibition for other cancers, and may suggest new strategies for overcoming innate and adaptive resistance to checkpoint inhibition. The most critical question in the field of cancer immunotherapy is whether biomarkers can be defined that predict benefit from the use of these drugs and allow oncologists to choose patients most likely to respond to them. In melanoma and non–small cell lung cancer (NSCLC), a variety of studies have suggested that tumors have three potential immune profiles: (1) those that are infiltrated with T cells and express an “inflammatory” signature of genes, which could be amenable to checkpoint inhibition; (2) tumors that are devoid of any T-cell or inflammatory infiltrate on histologic examination and have a noninflamed,

or “cold” gene expression profile and could be amenable to adoptive cell therapy; (3) and tumors that have T cells and other immune cells present, but only at the periphery or within the stromal tissue and not within the tumor itself and might be amenable to antiangiogenic therapy.12 The “hot” tumors are most likely to respond to PD-1/PDL1 blockade and have been associated with a previously primed immune response, but have been infiltrated with T cells with high levels of PD-1. Cold tumors that lack a T-cell infiltrate may be good candidates for an adoptive cell therapy strategy, and tumors that have immune cells that fail to infiltrate the tumor tissue may be appropriate for strategies employing antiangiogenesis agents or other drugs that promote T-cell migration.

PD-L1

PD-L1 is the critical ligand for the checkpoint molecule PD-1 on T cells. Its overexpression on tumor cells is a form of adaptive resistance to the presence of T cells that are infiltrating tumors.13,14 A number of studies have evaluated the association of PD-L1 expressed on tumor cells and/or immune cell expression assessed by immunohistochemical staining and its clinical effect on the efficacy of PD-1/PDL1 blockade.15 Although most studies are in agreement that the higher the level of tumor cell membrane PD-L1 expression, the better the outcome with PD-1/PD-L1 blockade, it is clear that patients whose tumors stain negatively for PD-L1 may still gain benefit from checkpoint inhibition.16

From the Laura and Isaac Perlmutter Cancer Center, NYU Langone Medical Center, New York, NY. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Jeffrey S. Weber, MD, PhD, Laura and Isaac Perlmutter Cancer Center, NYU Langone Medical Center, 522 First Ave., Room 1310, Smilow Building, New York, NY 10016; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 205

JEFFREY S. WEBER

This compromises the utility of PD-L1 as a biomarker to choose patients for therapy, because the investigators are unable to define patients who should not receive immunotherapy. In current studies, no less than three different antibodies are routinely used for PD-L1 staining assays, with three different scoring systems. The Ventana SP263, Dako 22C3, and Dako 28-8 antibodies have been most commonly used, and when evaluated for the pathologists’ concordance in scoring using NSCLC specimens, a 90% rate was achieved.17 Nonetheless, some trials include tumor staining, others allow staining of tumor and immune cells, and yet others include staining of the immune-infiltrating cells only within the scoring system. Nonetheless, PD-L1 is an important biomarker for some tumors, and it has been used as a companion biomarker for the approval of pembrolizumab in patients with NSCLC.9,10 In NSCLC and melanoma, patients with the highest levels of PD-L1 tumor staining have an excellent chance of achieving a response to PD-1 blockade. PD-L1 expression, a cytolytic or gamma interferon–related gene expression signature, CD8 density, and mutational load have been evaluated for their utility as biomarkers for the efficacy of PD-1 blockade, and mutational load appeared to be independent of the expression of T-cell and PD-L1 markers (Weber et al, unpublished data, 2017).

T-CELL INFILTRATION

The number of CD8 T-cells infiltrating the tumor microenvironment and expressing PD-1 and/or CTLA-4 appears to be a key indicator of success with checkpoint inhibition, and both PD-1 and CTLA-4 blockade may increase the proportion of infiltrating T cells. The best parameter associated with response to PD-1 blockade was a high density of CD8+ T cells at the invasive tumor margin.18 Assessment of CD8+ cells in the tumor itself were less useful, as was tumor and invasive margin PD-1 expression or the expression of PD-L1 on the tumor cells and invasive margin cells. The number of intratumoral CD8+ T cells that were PD-1+ may also be associated with response to PD-1 blockade. In a different study, the number of double positive PD-1+/CTLA-4+ CD8 T cells within the tumor-infiltrating population was most strongly associated with outcome.19 In patients receiving sequential PD-1 then CTLA-4 blockade with a planned switch, the CD8 T-cell infiltrate detected by immunohistochemistry was strongly associated

Key Points • PD-L1 staining of tumors is associated with response and survival after treatment with PD-1 blockade. • The magnitude of the CD8 T-cell infiltrate is an important correlate of benefit from anti–PD-1 therapy. • An inflammatory tumor signature is also associated with benefit from anti–PD-1 therapy. • T-cell receptor diversity is associated with benefit from CTLA-4 blockade; in contrast, T-cell receptor clonality is associated with benefit from PD-1 blockade. • Mutational and neoantigen load have a modest association with benefit from PD-1 blockade. 206 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

with response to treatment in the cohort that received PD-1 antibody first (Weber et al, unpublished data, 2017). Immune T-cell receptor (TCR) DNA sequencing utilizes a multiplex polymerase chain reaction (PCR) assay with forward primers specific for each V-gene segment and reverse primers specific for each J-gene segment. The TCR repertoire from circulating peripheral blood T cells was determined prior to and after anti–CTLA-4 antibody therapy. There was a substantial increase in the diversity (the number of unique TCR V-beta sequences) of the peripheral blood T cells. This increase demonstrated that no specific clone or subgroup of clones was expanded preferentially. These data suggest that a number of important clones have been disinhibited and allowed to proliferate by CTLA-4 blockade.20,21 Interestingly, the immune-related toxicity associated with anti–CTLA-4 antibodies was also associated with increases in the TCR diversity, suggesting that some of the clones that were disinhibited and allowed to proliferate generated proinflammatory or autoimmune hyper-responsiveness. In another study, tumor biopsies from patients with metastatic melanoma were analyzed by TCR V-beta chain immune sequencing before anti–PD-1 antibody therapy.18 Patients whose tumors had more clonal T-cell repertoire were most likely to respond to PD-1 blockade. When a metric of T cell number and clonality of the TCR was calculated, those with progressive disease had the lowest values. Analysis of tumors obtained after starting anti–PD-1 therapy showed that patients whose tumors exhibited high expansion of preexisting T-cell clones were most likely to respond to therapy. A “focused” TCR repertoire, defined by DNA sequencing of the rearranged beta chain variable regions of the TCR within tumors, was associated with a good outcome with PD-1/ PD-L1 blockade,18 whereas a more “diverse” repertoire infiltrating tumors was associated with benefit from anti–CTLA4 antibodies, as indicated above.20-22 The focused repertoire would seem likely to encode TCRs specific for neoantigens, and many of the TCR sequences associated with a good outcome with PD-1 blockade can be found in the periphery in patients with melanoma. TCR diversity is also independent of the T-cell infiltrate or PD-L1 staining, both of which are dependent variables as a marker of outcome.

MUTATION AND NEOANTIGEN LOAD

The T-cell repertoire reflects the host immune response to cancer, but the tumor itself is a key determinant of success with checkpoint inhibition because there is a relationship between increased nonsynonymous variants or somatic mutations in tumors and outcome with checkpoint inhibition.23 Patients who had tumors, which, like melanoma, possessed a high frequency of somatic mutations,were more likely to respond to checkpoint inhibition with anti–CTLA4 and anti–PD-1 antibodies. In lung cancer, the number of smoking-related mutations,24 and in gastrointestinal cancers,25 the level of mutations dictated by the presence of mismatch repair deficiency were associated with the benefit of PD-1 blockade. Mismatch repair deficiency occurs in a small proportion of colorectal cancers as well as cancers

BIOMARKERS FOR CHECKPOINT INHIBITION

of the uterus, stomach, biliary tract, pancreas, ovary, prostate, and small intestine. Tumors that possess defects in the mismatch repair pathway have thousands of somatic mutations in regions of repeated DNA, known as microsatellites. Many different mismatch repair deficient tumors possess a prominent immune infiltrate and a cytokine-rich tumor microenvironment in which the tumors express PD-L1 and the effectors express different immune checkpoints including PD-1, CTLA-4, and LAG-3, which is consistent with a primed immune response.26 The number of neoantigens—mutated proteins that are expressed and could be recognized in the context of MHC class I or II molecules as a processed peptide antigen by T cells, which is related to the total mutational burden—is critically associated with outcome for checkpoint inhibitors. However, the correlation between the burden of neoantigens and clinical benefit was less clear-cut when increasing rigorous thresholds for the binding affinity of peptides were applied and the neoantigens thus defined did not possess any shared sequences or features that were preferentially observed in patients who were responding.27 These data suggest that the clinical relevance of neoantigens depends on the proper antigen processing and affinity of the neo-epitope peptide and HLA expression by the tumor, which is frequently aberrant. An additional issue is that of clonality, the likelihood that the majority of tumor cells express the neo-epitope in question, as opposed to a “branched” mutation that might be expressed by a small proportion of tumor cells and not clinically relevant. Better algorithms might also be needed to assess the immunogenicity of mutation-derived neo-epitopes.28,29 Interestingly, BRCA2 mutations, which are associated with increased rates of DNA damage and a higher mutational load, are also associated with response to PD-1 blockade.30

TUMOR GENE EXPRESSION PROFILE

The nature of the tumor microenvironment also plays an important role in resistance or susceptibility to checkpoint inhibition. A tumor gene expression signature that reflects a series of gamma interferon–inducible genes may define a “hot,” inflamed tumor, and is associated in several studies with a good outcome with checkpoint inhibition; its loss is associated with resistance to ipilimumab therapy.30,31 Melanomas that are class II MHC–positive respond to PD-1/ PD-L1 blockade and may share the interferon responsive signature.31 In a recent study that included whole-exome sequencing of tumors from 16 patients with melanoma, multiple copy-number alterations resulted in the loss of interferon gamma pathway genes in 12 patients whose disease did not respond to ipilimumab.32 Mice bearing melanoma tumors that lacked one of these genes, IFNGR1, also had an impaired response to anti–CTLA-4 therapy and substantially reduced overall survival compared with their counterparts whose tumors had wild-type IFNGR1. Tumor samples were collected from patients with melanoma treated with CTLA-4 blockade followed by PD-1 blockade at progression at multiple time points during therapy. Tumor biopsies during CTLA4 blockade demonstrated higher density of CD8+ T cells in

responders compared with nonresponders, suggesting a pharmacodynamic effect of the treatment that was associated with benefit.33 When tumor gene expression profiling for patients exposed to either PD-1 or CTLA-4 blockade was performed, there was only a modest overlap in the genes that were increased at baseline or during early therapy and associated with outcome with either therapy, indicating a very different mode of action of the two antibodies. Response to PD-1 blockade was associated with pathways of cytolytic activity, antigen processing, and interferon gamma signaling. Expression of VEGFA was decreased in responders and increased with therapy in nonresponders, suggesting a mechanism of therapeutic resistance, as observed by others, and a potential target for therapy. In contrast, resistant tumors displayed a transcriptional signature (called the innate anti–PD-1 resistance, or IPRES), which was associated with increased expression of genes involved in the regulation of the epithelial-mesenchymal transition, cell adhesion, extracellular matrix remodeling, angiogenesis, and wound healing.30 In addition to mesenchymal transition genes, immunosuppressive genes including IL10, VEGFA, VEGFC, and monocyte and macrophage chemotactic genes such as CCL2, CCL7, CCL8, and CCL13 were associated with a poor outcome with PD-1 blockade. Those signatures of mesenchymal-invasive transition, angiogenesis, and wound-healing signatures have been detected in the resistant melanomas from patients receiving BRAF-inhibitor therapy, suggesting that induction of these signatures may negatively impact responsiveness to combinatorial anti–PD-1/PD-L1 therapy.30,34 The IPRES signature was found to be increased in metastases compared with primary melanomas and was also detected in most different types of malignancy. Deletion of the PTEN gene, commonly found in melanoma, has a deleterious effect on antitumor immunity with checkpoint inhibition, and leads to a “cold” tumor with high levels of immune suppressive cytokines with sparse and inactive T cells.35 The PTEN-deleted population had increased Akt signaling, and consistent with that finding, high levels of p-Akt expression in pretreatment tumor cells. In addition, CTLA-4 expression on the tumors and infiltrating T cells of patients with melanoma was associated with poor response rate and overall survival.36 There is also an association between tumor activation of the WNT/β-catenin signaling pathway and absence of a T-cell gene expression signature, which leads to deficiencies of infiltrating dendritic cells and a “cold” tumor microenvironment.37 This might be overcome with the use of a STING agonist, which can augment expression of interferon gamma pathway genes.38 Clinical examination of host biomarkers from large clinical trials of PD-1 blockade has shown that neutrophil-to-lymphocyte ratios, baseline lactate dehydrogenase, and eosinophil numbers are associated with outcome to checkpoint blockade, although none of these markers can reliably identify a patient who will not benefit from treatment.39-41 When tumors become resistant to PD-1 blockade after an initial response, exhibiting adaptive resistance to therapy, the induction of tumor JAK1 and JAK2 mutations, or deletion of beta 2-microglobulin may be responsible, leading to asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 207

JEFFREY S. WEBER

impaired T-cell immunity and inability to detect tumor antigens.42 In patients with innate resistance to anti–PD-1 antibodies that never respond to treatment, inactivating JAK1/ JAK2 mutations are not common, but are associated with low PD-L1 expression and lack of antitumor response.43

PERIPHERAL BLOOD AND MICROBIOME

The definition of serum biomarkers associated with the benefit of PD-1 blockade is still immature. In one series of patients receiving either nivolumab or pembrolizumab, a mass spectrometry–defined signature of proteins included those associated with acute phase reactant, complement, and wound-healing pathways.44 The complement pathway has not been clearly shown in the past to play a role in T-cell activation, but recent work suggests that, in murine models, T cells have complement receptors, and that C5a and C3a can inhibit T-cell proliferation and activation.45,46 High pretreatment serum levels of angiopoetin-2 was found to be associated with reduced overall survival in patients who were treated with anti–CTLA-4 or anti–PD-1 antibodies.47 CTLA-4 and PD-1 blockade increased serum angiopoietin early after treatment initiation in a cohort of patients, whereas the addition of bevacizumab to ipilimumab resulted in decreased serum concentrations of angiopoietin. Increased angiopoietin levels were associated with reduced response to checkpoint inhibition. Although immune populations detected in the peripheral circulation may not reflect events in the tumor microenvironment, a recent study demonstrated that baseline frequencies

of myeloid-derived suppressor cells, CD4+/DF25+/FOXp3+ T regulatory cells, and high levels of eosinophils were associated with clinical benefit in patients with melanoma treated with ipilimumab.40 High baseline frequencies of circulating CD4+/CD25-high/FOXp3+ T regulatory cells were associated with improved overall survival in this cohort. There is a long history of work suggesting that the composition of the host microbiota in mice is associated with a favorable outcome with immunotherapy and checkpoint blockade, and recent data suggest that both clinical outcome and the immune-related adverse events often seen with checkpoint blockade may be associated with specific microbial taxa.48-51

CONCLUSION

In conclusion, there is no clear-cut and clinically useful single biomarker associated with the benefit of checkpoint blockade, or which could be used to select patients that would not benefit from this treatment. Developing the tools to define pathways of benefit, and that have the negative predictive value to predict innate and adaptive resistance to PD-1/PD-L1 and CTLA-4 inhibition, will undoubtedly require an amalgamated biomarker that combines tumor cell–intrinsic and host T-cell specific determinants. Current efforts in which peripheral blood cells, tumor and serum, as well as microbiome specimens are routinely collected in patients before and after treatment with checkpoint inhibition will be critical to research in defining biomarkers of response and resistance to immunotherapies.

References 1. Hodi FS, O’Day SJ, McDermott DF, et al. Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med. 2010;363:711-723.

9. Garon EB, Rizvi NA, Hui R, et al; KEYNOTE-001 Investigators. Pembrolizumab for the treatment of non-small-cell lung cancer. N Engl J Med. 2015;372:2018-2028.

2. Robert C, Thomas L, Bondarenko I, et al. Ipilimumab plus dacarbazine for previously untreated metastatic melanoma. N Engl J Med. 2011;364:2517-2526.

10. Reck M, Rodríguez-Abreu D, Robinson AG, et al; KEYNOTE-024 Investigators. Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung cancer. N Engl J Med. 2016;375:1823-1833.

3. Ribas A, Puzanov I, Dummer R, et al. Pembrolizumab versus investigator-choice chemotherapy for ipilimumab-refractory melanoma (KEYNOTE-002): a randomised, controlled, phase 2 trial. Lancet Oncol. 2015;16:908-918.

11. Powles T, Eder JP, Fine GD, et al. MPDL3280A (anti-PD-L1) treatment leads to clinical activity in metastatic bladder cancer. Nature. 2014;515:558-562.

4. Robert C, Long GV, Brady B, et al. Nivolumab in previously untreated melanoma without BRAF mutation. N Engl J Med. 2015;372:320-330. 5. Larkin J, Chiarion-Sileni V, Gonzalez R, et al. Combined nivolumab and ipilimumab or monotherapy in untreated melanoma. N Engl J Med. 2015;373:23-34. 6. Postow MA, Chesney J, Pavlick AC, et al. Nivolumab and ipilimumab versus ipilimumab in untreated melanoma. N Engl J Med. 2015;372:2006-2017.

12. Teng MW, Ngiow SF, Ribas A, et al. Classifying cancers based on T-cell infiltration and PD-L1. Cancer Res. 2015;75:2139-2145. 13. Topalian SL, Hodi FS, Brahmer JR, et al. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N Engl J Med. 2012;366:2443-2454. 14. Taube JM, Klein A, Brahmer JR, et al. Association of PD-1, PD-1 ligands, and other features of the tumor immune microenvironment with response to anti-PD-1 therapy. Clin Cancer Res. 2014;20:5064-5074.

7. Borghaei H, Paz-Ares L, Horn L, et al. Nivolumab versus docetaxel in advanced nonsquamous non-small-cell lung cancer. N Engl J Med. 2015;373:1627-1639.

15. Danilova L, Wang H, Sunshine J, et al. Association of PD-1/PD-L axis expression with cytolytic activity, mutational load, and prognosis in melanoma and other solid tumors. Proc Natl Acad Sci USA. 2016;113:E7769-E7777.

8. Brahmer J, Reckamp KL, Baas P, et al. Nivolumab versus docetaxel in advanced squamous-cell non-small-cell lung cancer. N Engl J Med. 2015;373:123-135.

16. Daud AI, Wolchok JD, Robert C, et al. Programmed death-ligand 1 expression and response to the anti-programmed death 1 antibody pembrolizumab in melanoma. J Clin Oncol. 2016;34:4102-4109.

208 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

BIOMARKERS FOR CHECKPOINT INHIBITION

17. Ratcliffe MJ, Sharpe A, Midha A, et al. Agreement between programmed cell death ligand-1 diagnostic assays across multiple protein expression cut-offs in non-small cell lung cancer. Clin Cancer Res. Epub 2017 Jan 10. 18. Tumeh PC, Harview CL, Yearley JH, et al. PD-1 blockade induces responses by inhibiting adaptive immune resistance. Nature. 2014;515:568-571. 19. Daud AI, Loo K, Pauli ML, et al. Tumor immune profiling predicts response to anti-PD-1 therapy in human melanoma. J Clin Invest. 2016;126:3447-3452. 20. Robert L, Tsoi J, Wang X, et al. CTLA4 blockade broadens the peripheral T-cell receptor repertoire. Clin Cancer Res. 2014;20:2424-2432. 21. Cha E, Klinger M, Hou Y, et al. Improved survival with T cell clonotype stability after anti-CTLA-4 treatment in cancer patients. Sci Transl Med. 2014;6:238ra70. 22. Kvistborg P, Philips D, Kelderman S, et al. Anti-CTLA-4 therapy broadens the melanoma-reactive CD8+ T cell response. Sci Transl Med. 2014;6:254ra128. 23. Snyder A, Makarov V, Merghoub T, et al. Genetic basis for clinical response to CTLA-4 blockade in melanoma. N Engl J Med. 2014;371:2189-2199. 24. Rizvi NA, Hellmann MD, Snyder A, et al. Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in nonsmall cell lung cancer. Science. 2015;348:124-128. 25. Le DT, Uram JN, Wang H, et al. PD-1 blockade in tumors with mismatchrepair deficiency. N Engl J Med. 2015;372:2509-2520. 26. Llosa NJ, Cruise M, Tam A, et al. The vigorous immune micro environment of microsatellite instable colon cancer is balanced by multiple counter-inhibitory checkpoints. Cancer Discov. 2015;5: 43-51.

35. Peng W, Chen JQ, Liu C, et al. Loss of PTEN promotes resistance to T cell-mediated immunotherapy. Cancer Discov. 2016;6:202-216. 36. Chakravarti N, Ivan D, Trinh VA, et al. High cytotoxic T-lymphocyteassociated antigen 4 and phospho-Akt expression in tumor samples predicts poor clinical outcomes in ipilimumab-treated melanoma patients. Melanoma Res. 2017;27:24-31. 37. Spranger S, Bao R, Gajewski TF. Melanoma-intrinsic β-catenin signalling prevents anti-tumour immunity. Nature. 2015;523:231-235. 38. Woo SR, Fuertes MB, Corrales L, et al. STING-dependent cytosolic DNA sensing mediates innate immune recognition of immunogenic tumors. Immunity. 2014;41:830-842. 39. Weide B, Martens A, Hassel JC, et al. Baseline biomarkers for outcome of melanoma patients treated with pembrolizumab. Clin Cancer Res. 2016;22:5487-5496. 40. Martens A, Wistuba-Hamprecht K, Geukes Foppen M, et al. Baseline peripheral blood biomarkers associated with clinical outcome of advanced melanoma patients treated with ipilimumab. Clin Cancer Res. 2016;22:2908-2918. 41. Kelderman S, Heemskerk B, van Tinteren H, et al. Lactate dehydrogenase as a selection criterion for ipilimumab treatment in metastatic melanoma. Cancer Immunol Immunother. 2014;63:449458. 42. Shin DS, Zaretsky JM, Escuin-Ordinas H, et al. Primary resistance to PD-1 blockade mediated by JAK1/2 mutations. Cancer Discov. 2017;7:188-201. 43. Zaretsky JM, Garcia-Diaz A, Shin DS, et al. Mutations associated with acquired resistance to PD-1 blockade in melanoma. N Engl J Med. 2016;375:819-829.

27. Van Allen EM, Miao D, Schilling B, et al. Genomic correlates of response to CTLA-4 blockade in metastatic melanoma. Science. 2015;350:207211.

44. Weber J, Sznol M, Kluger H, et al. A test identifying advanced melanoma patients with long survival outcomes on nivolumab shows potential for selection for benefit from combination checkpoint blockade. Paper presented at: 31st Society for Immunotherapy of Cancer Annual Meeting; November 2016; National Harbor, MD.

28. McGranahan N, Furness AJ, Rosenthal R, et al. Clonal neoantigens elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade. Science. 2016;351:1463-1469.

45. Wang Y, Sun SN, Liu Q, et al. Autocrine complement inhibits IL10dependent T-cell-mediated antitumor immunity to promote tumor progression. Cancer Discov. 2016;6:1022-1035.

29. Roszik J, Haydu LE, Hess KR, et al. Novel algorithmic approach predicts tumor mutation load and correlates with immunotherapy clinical outcomes using a defined gene mutation set. BMC Med. 2016;14:168177.

46. Nabizadeh JA, Manthey HD, Steyn FJ, et al. The complement C3a receptor contributes to melanoma tumorigenesis by inhibiting neutrophil and CD4+ T cell responses. J Immunol. 2016;196:47834792.

30. Hugo W, Zaretsky JM, Sun L, et al. Genomic and transcriptomic features of response to anti-pd-1 therapy in metastatic melanoma. Cell. 2016;165:35-44.

47. Wu X, Giobbie-Hurder A, Liao X, et al. Angiopoietin-2 as a biomarker and target for immune checkpoint therapy. Cancer Immunol Res. 2017;5:17-28.

31. Johnson DB, Estrada MV, Salgado R, et al. Melanoma-specific MHC-II expression represents a tumour-autonomous phenotype and predicts response to anti-PD-1/PD-L1 therapy. Nat Commun. 2016;7:10582.

48. Vétizou M, Pitt JM, Daillère R, et al. Anticancer immunotherapy by CTLA-4 blockade relies on the gut microbiota. Science. 2015;350:10791084.

32. Gao J, Shi LZ, Zhao H, et al. Loss of IFN-γ pathway genes in tumor cells as a mechanism of resistance to anti-CTLA-4 therapy. Cell. 2016;167:397-404.e9.

49. Sivan A, Corrales L, Hubert N, et al. Commensal Bifidobacterium promotes antitumor immunity and facilitates anti-PD-L1 efficacy. Science. 2015;350:1084-1089.

33. Chen PL, Roh W, Reuben A, et al. Analysis of immune signatures in longitudinal tumor samples yields insight into biomarkers of response and mechanisms of resistance to immune checkpoint blockade. Cancer Discov. 2016;6:827-837.

50. Pitt JM, Vétizou M, Daillère R, et al. Resistance mechanisms to immune-checkpoint blockade in cancer: tumor-intrinsic and -extrinsic factors. Immunity. 2016;44:1255-1269.

34. Hugo W, Shi H, Sun L, et al. Non-genomic and immune evolution of melanoma acquiring MAPKi resistance. Cell. 2015;162:1271-1285.

51. Dubin K, Callahan MK, Ren B, et al. Intestinal microbiome analyses identify melanoma patients at risk for checkpoint-blockade-induced colitis. Nat Commun. 2016;7:10391.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 209

GARRALDA, DIENSTMANN, AND TABERNERO

Pharmaco*kinetic/Pharmacodynamic Modeling for Drug Development in Oncology Elena Garralda, MD, Rodrigo Dienstmann, MD, and Josep Tabernero, MD, PhD OVERVIEW High drug attrition rates remain a critical issue in oncology drug development. A series of steps during drug development must be addressed to better understand the pharmaco*kinetic (PK) and pharmacodynamic (PD) properties of novel agents and, thus, increase their probability of success. As available data continues to expand in both volume and complexity, comprehensive integration of PK and PD information into a robust mathematical model represents a very useful tool throughout all stages of drug development. During the discovery phase, PK/PD models can be used to identify and select the best drug candidates, which helps characterize the mechanism of action and disease behavior of a given drug, to predict clinical response in humans, and to facilitate a better understanding about the potential clinical relevance of preclinical efficacy data. During early drug development, PK/PD modeling can optimize the design of clinical trials, guide the dose and regimen that should be tested further, help evaluate proof of mechanism in humans, anticipate the effect in certain subpopulations, and better predict drug-drug interactions; all of these effects could lead to a more efficient drug development process. Because of certain peculiarities of immunotherapies, such as PK and PD characteristics, PK/PD modeling could be particularly relevant and thus have an important impact on decision making during the development of these agents.

igh drug attrition rates in oncology have been a major concern during recent years.1 Only 5% of agents that show anticancer activity in preclinical development ultimately are approved upon demonstration of efficacy in a phase III clinical trial. Although the advent of molecular targets has led to a reduction in attrition rates—to 55% in the case of kinase inhibitors—rates remain unacceptably high,2 which has led to an unsustainable economic model of drug discovery for the pharmaceutical industry and to spiraling costs of drugs that finally receive approval.2,3 These high attrition rates compared with those of other therapeutic areas can be explained in part by particular characteristics of oncology drugs,4 including a narrow therapeutic index, complex pharmacology, the lack of data from healthy patients, a sparser PK sampling, high interindividual variability, and frequent use of therapies in combination to achieve maximum efficacy. Also, major differences in cancer targets and mechanisms of action of anticancer drugs (from conventional cytotoxic chemotherapies to small-molecule targeted agents to immune checkpoint targeted therapies) reject a one-size-fits-all model for PK and PD analyses. Different solutions, including adaptive trial design5; a more extensive use of biomarkers from the early stages6-8; and novel tools, such as PK/PD modeling and simulation to aid the different steps of drug development, have been discussed.9,10

PHARMACOLOGICAL AUDIT TRAIL

The Pharmacological Audit Trail (PhAT) is a conceptual framework developed by Banerji and Workman11 to facilitate rational decision making during drug development. By integrating PK and PD data, this tool allows for the codification of a series of biomarker-driven questions that should be raised in a sequential way at the appropriate stages of drug development. When these relevant issues or benchmarks are addressed, the likelihood of failure of a drug would decrease. The PhAT allows us to address critical aspects though out all the process, from the identification of the population most likely to respond and thus to define the target population, to the development of biomarkers of response, to understand the mechanisms of resistance once the treatment fails and finally to establish potential mechanisms to overcome such resistance (such as defining a new combination regimen or the identification of a potential new target).

Pharmaco*kinetics

PK is the study of the drug concentrations in the body during a period of time, and it includes the processes by which the drug is absorbed, distributed, metabolized, and excreted (also known in colloquial terms as what the body does to the drug).

From the Early Drug Development Unit, Vall d’Hebron University Hospital and Vall d´Hebron Institute of Oncology, CIBERONC, Universitat Autònoma de Barcelona, Barcelona, Spain. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Josep Tabernero, MD, PhD, Early Drug Development Unit, Vall d’Hebron University Hospital and Vall d´Hebron Institute of Oncology, CIBERONC, Universitat Autònoma de Barcelona, Barcelona, Spain; email: [emailprotected]. © 2017 American Society of Clinical Oncology

210 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PHARMAco*kINETIC/PHARMACODYNAMIC MODELING FOR DRUG DEVELOPMENT

Both the importance and utility of PK studies in early drug development have long been recognized.4,12 Different questions must be addressed. Are we reaching an appropriate PK exposure in humans? What is the correct schedule of administration? What is the correlation of the PK values with the toxicities observed during the trial? The answer to this specific question, for example, could help manage different toxicity profiles that could correlate with the maximum drug concentration and the area under the curve, which could help determine a different schedule of administration.7 Finally, is there a relevant food effect or drug-drug interaction?

Pharmacodynamics

PD is the study of the relationship between drug concentration and its biologic effects (also known in colloquial terms as what the drug does to the body). Overall, there are two main types of biomarkers in this field13: (1) predictive biomarkers that constitute any measurement associated to response/lack of response or toxicity and (2) mechanism-of-action biomarkers that reveal insights into the PD effects of a drug. Selection of the correct biomarker remains an important challenge, and the bottom-line questions surely must be these: are we modulating the intended target, and does this modulation translate into clinical benefit? Just one example to illustrate this point is the phase I trial of the protein tyrosine kinase Src inhibitor saracatinib.14 The maximum-tolerated dose was achieved, and the recommended phase II dose was determined. Importantly, different schemes of treatments were tested, and PK analysis confirmed the proposed dose as optimal for PD effects in the tumor, with substantial reduction in Src activity. Another example is a PD study performed to evaluate the effects of different doses and schedules of cetuximab.15 Results showed that every-other-week dosages of cetuximab had the same functional PD effect as weekly administration in patients with metastatic colorectal cancer; these results confirmed that the every-other-week dosage could be the appropriate one to treat this patient population.

KEY POINTS • There is a clear need to improve the speed and efficiency of clinical drug development in oncology. • A deeper, more systematic knowledge of the PK and PD properties of a particular drug is required to better guide rational decision making during drug development. • The use of computational models and simulation can help quantify and understand the relationship between exposure (i.e., PK) and response (i.e., PD). • PK/PD modeling is a useful tool throughout all stages of drug development, and applications differ during the preclinical and clinical stage. • Given the particular characteristics of immune therapies, PK/PD modeling could be particularly relevant during the immune therapy development.

Despite the fact that PK/PD biomarkers remain crucial for phase I trials, incorporation of the use of predictive biomarkers from the outset clearly could be crucial for acceleration of drug development.16 This is especially true with molecular targeted therapies, for which the cancer patient subpopulations that are most likely to respond to treatment can be identified to increase the probability of success. The development of anaplastic lymphoma kinase inhibitors in patients with ALK rearrangements,17 or BRAF inhibitors in patients with mutant V600E BRAF melanoma,18 is a clear example of this molecular enrichment approach. Complete validation of the biomarker probably will be carried out in subsequent phase II and III trials; incorporation of the biomarkers earlier, though, will potentiate more informative study designs about the biology of the tumor and treatment resistance mechanisms.

Pharmaco*kinetics/Pharmacodynamics Modeling

Model-based drug development employs mathematical and statistical models to describe disease progression, PK, and PD; to improve study design; and to better enable decision making.19 It works as a tool to respond more certainly to the questions raised in the PhAT at a lower cost. The term PK/PD modeling refers to a PK- and PD-driven exploratory analysis, which is based on a mathematical model.20 These models can help to better understand the relationship between exposure (PK) and response (PD) as well as the change between these relationships as a function of drug intake. The normal assumption for a study design is the linear relationship between exposure to a medication and its activity; however, this relationship is not always so simple. For example, many monoclonal antibodies exhibit nonlinear PK behavior. PK/PD modeling can be used throughout all the stages of clinical development.21 Because models work with data in an iterative manner, the PK/PD model will be more reliable and valuable when more data are available. For the tool to reach its full potential, the tool should be developed from the preclinical stages to incorporate new data as the drug moves forward, which thus refines and improves the model. The updated model will help with the next steps of development. At the end of this building process, a well-defined model will include different submodels, which will be able to predict trial outcomes through the use of data of hundreds of multiple individual patients and multiple trial designs (Fig. 1). PK/PD modeling allows us to address a number of key questions at the various stages of the drug discovery and development process (i.e., PhAT). Pharmaco*kinetics/pharmacodynamics modeling in preclinical stages. The main objective of early drug development is to select promising compounds that will be screened for efficacy and safety. The most promising agents will be studied more, and those with an acceptable efficacy/safety profile will enter the clinic. However, the translation of the efficacy and safety results from a preclinical level to the patient population remains a major challenge. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 211

GARRALDA, DIENSTMANN, AND TABERNERO

One of the key aims of translational, mainly mechanistic, PK/PD modeling is to generate a priori simulations that help support predictions about efficacy between different species (e.g., in vitro-in vivo predictions and xenograft-to-clinical correlations); thus, the importance of this tool lies at the preclinical-to-clinical interface.21,22 The biggest cost-saving potential of drug modeling would be a determination of which compounds should move forward and which should be dropped. The use of these models in the preclinical setting has several potential advantages.23,24 It can improve lead optimization and the selection of the optimal compound, predict clinical potency estimates (e.g., effective concentration of a drug that gives half-maximal response), and predict the drug exposure needed. It also can provide guidance about the initial tested dose in clinical trials, the dose range, the suggested administration scheme, and even the optimal sampling required in the trial (Fig. 2).25 Other advantages include prediction of oral bioavailability and assessment of the potential for drug-drug interactions. Other types of modeling, such as synergy-response surface modeling, can help predict the result of drug combinations and can better define a whole development strategy. Pharmaco*kinetics/pharmacodynamics modeling in the clinical stage. There are several applications of PK/PD modeling in clinical development.24,26 This include effective establishment of the relationship between exposure and biomarker, exposure and response, or biomarker and response

FIGURE 1. The Model-Building Process

212 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

relationships; earlier decision making about go or no-go plans on the basis of PK and PD characteristics of the drug; reduced numbers of phase II trials needed to obtain enough data for a phase III trial or for registration; and evaluation of different schemes of drug administration and study designs. One important aspect of modeling is that it can increase prediction of drug-drug interactions. In the field of oncology, these interactions are a substantial problem, because many agents have a narrow therapeutic index and because most of the anticancer agents will be metabolized by cytochrome P450. The use of modeling can unmask these interactions and the importance of these interactions, and such data can be sent to the health authorities. With this approach, additional drug-drug interaction studies were avoided during the development of ceritinib, for example.27 Population PK/PD models also are becoming more important; in addition to the characterization of PK and PD, the models include relationships between covariates such as patient characteristics (e.g., age, body weight, renal function). This enables the assessment and the quantification of potential sources of variability in exposure and response within a specific target population, even under sparse sampling conditions. This approach is extremely relevant for assessments of special populations (e.g., pediatric population, frail patients, renal or liver impairment).28 A comprehensive example that illustrates the use of PK/ PD modeling in the clinic is the development of everolimus. Everolimus blocks the mammalian target of rapamycin, or

PHARMAco*kINETIC/PHARMACODYNAMIC MODELING FOR DRUG DEVELOPMENT

mTOR, pathway and inhibits the downstream S6 kinase 1. PK/PD models predicted that the inhibition of S6 kinase 1 in peripheral blood mononuclear cells was related to tumor effect and suggested that, although a weekly dose of 20 to 30 mg of everolimus already was associated with an antitumor effect, daily administration could cause a greater effect with higher dose intensity.29,30 This supported the incorporation of S6 kinase 1 as a PD biomarker of mTOR signaling and guided the selection of the doses explored during the phase I studies. The phase I trial showed that everolimus administered as 10 mg daily or as 50 mg weekly would be the recommended phase II dosage, although the PD effect was more sustained with the daily dosage.30,31 Subsequently, a phase III trial showed an increased overall survival in patients with renal cancer who received 10 mg of everolimus daily; approval for this indication was obtained.32 Another phase III trial showed that 10 mg of everolimus daily increased progression-free survival in patients with pancreatic neuroendocrine tumors.33 One area in which PK/PD modeling is especially reliable is that of biologics (drugs such as monoclonal antibodies that are produced by using biologic organisms or purified from a natural source) because of their ability to translate across species.34 In general, the determination of the firstin-human dose is based mainly on toxicology properties and

consideration of the no-observed-adverse-effect level, which is determined in preclinical safety studies and then reduced by an appropriate safety factor. However, with monoclonal antibodies, particularly agonist therapies, an additional approach is recommended—an approach that uses the minimal anticipated biologic effect level (MABEL) approach. The MABEL approach considers the pharmacological properties of the drug and the anticipated dose level that leads to a minimal biologic effect level in humans. MABEL is calculated mainly by integrating all of the available in vitro and in vivo information by PK/PD modeling.

DRUG MODELING WITH IMMUNOTHERAPIES

The advent of immune therapy represents a groundbreaking milestone for the treatment of patients with cancer, and the number of immuno-oncology agents entering drug development has continued to increase. However, because of the particular characteristics of these agents, their development has different challenges that must be considered. Until recently, because most agents have had a direct dose-response curve, determination of the maximum-tolerated dose (MTD) has been the principal parameter for definition of the recommended phase II dose. However, the safety profile of most immunotherapies is different from those of targeted therapies or cytotoxic agents, and

FIGURE 2. Simulation of Exposure and Efficacy Data of Hypothetical Dosing Regimens in a Preclinical Mouse Model

A mouse xenograft model was treated with a drug with two different regimens: 30 mg/kg once daily or 1 mg/kg twice daily. Observed and predicted plasma levels were plotted with the simulated PD responses. This strategy can help determine what dose should be further tested. Abbreviations: PD, pharmacodynamic; PK, pharmaco*kinetic. Modified from Tuntland et al.5

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 213

GARRALDA, DIENSTMANN, AND TABERNERO

the MTD generally is not reached; this provides an optimal setting for PK/PD-based dose recommendations. Moreover, the relationships between dose, response and toxicity are more complex with immunotherapies than with other drugs, so the old paradigms of the higher the dose the better, and the higher the dose the higher the toxicity profile is completely obsolete.35 To date, most of these immunotherapeutic agents are antibodies and have much more complex PK profiles than small molecules that have specific absorption, distribution, and metabolism characteristics. The size of an immunotherapy agent is much bigger, they are mainly administered intravenously, their distribution is more limited, and their elimination mainly depends on proteolytic degradation rather than biliary and renal excretion. The PD characteristics also are different. For example, although the toxicity of small molecules can be as a result of an on-target or off-target mechanism, the toxicity that has been observed with immunotherapies is mainly because of an activation of the immunogenicity, with a delayed onset of most of the immune adverse events. This will be especially relevant when combination therapies are considered.36 Another potential advantage of immuno-oncology therapies is that, given their mechanisms of action, repeated PD measurements can be tracked for dynamic biomarker assessment and immunologic monitoring (e.g., CD4+ and CD8+ cells or the levels of different cytokines), which potentially could add value as predictive biomarkers or surrogates of tumor response. Despite these considerations, most phase I studies of immune agents lack PK and PD data. Additional knowledge of these parameters clearly would enrich PK/PD models and optimize the development of these treatments.35 A clear example of the importance of modeling in drug development is the case of pembrolizumab. The large phase I KEYNOTE 001 trial started with a standard 3 + 3 design, dose-escalation cohort to explore the MTD of pembrolizumab. Several dosages, from 2 to 10 mg/kg every 2 and 3 weeks, were evaluated.37 Despite the safety of the higher dosage studied (10 mg/kg every 2 weeks), the mechanism-based translational model, which focused mainly on intratumor exposure prediction, suggested that robust clinical activity would be observed from a dosage of 2 mg/kg every 3 weeks.38 This dosage, therefore,was recommended to be tested for clinical efficacy in additional clinical trials. Model-based characterization of the PK of pembrolizumab also was performed,39 and it indicated an absence of covariate effects and supported the pembrolizumab dosage

of 2 mg/kg every 3 weeks. Pembrolizumab at a dosage of 2 mg/kg every 3 weeks now is approved for non–small cell lung cancer tumors that express programmed death-ligand 1 and for metastatic melanoma. This clearly illustrates how drug modeling can transform early PK and PD results into a robust clinical trial design and can increase knowledge about the pharmacological properties of a drug. Given the costs of these agents and the challenge of reimbursem*nt for health authorities and insurers, determination of the most appropriate dose is paramount. Notably, pembrolizumab was approved just 4 years after the phase I clinical trial started, through breakthrough designation by the FDA. This timeframe clearly contrasts with the 10 or greater years that former drugs traditionally took to be approved.40

CONCLUSION

As we progress in drug development, the importance of strategic thinking and rational decision making aimed at improvements of results remains clear. The PhAT represents a stepwise approach that allows for critical decision making that is based on biomarker and clinical endpoints, and it should be adopted and embraced more widely in clinical research. Implementation of PK/PD modeling from early drug development promises a substantial impact on general efficiency as an excellent tool to help address critical questions and to evaluate different scenarios. For optimal modeling, the groundwork must begin early in preclinical development and the model must be finely tuned as results are obtained and sequentially analyzed. Because of the common practice of using an MTD measurement, model-based drug development generally has not been considered during development decisions for anticancer therapies. This will probably change with novel drugs such as immunotherapeutics, because the MTD often is not reached. For PK/PD modeling to deliver on its promise the entire drug development community will need to learn and understand this approach in order to trust the models and be reassured of their utility. As drug development evolves from determinations of MTDs to determinations of the optimal biologic (or immunologic) dose, the need for validated biomarkers will be of critical importance. Newer trial designs, coupled with new response and efficacy assessments, also will be required to optimize and expedite the development of novel agents, immunotherapies in particular.

References 1. Hutchinson L, Kirk R. High drug attrition rates: where are we going wrong? Nat Rev Clin Oncol. 2011;8:189-190. 2. Walker I, Newell H. Do molecularly targeted agents in oncology have reduced attrition rates? Nat Rev Drug Discov. 2009;8:15-16. 3. Kola I, Landis J. Can the pharmaceutical industry reduce attrition rates? Nat Rev Drug Discov. 2004;3:711-715.

214 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

4. Tuntland T, Ethell B, Kosaka T, et al. Implementation of pharmaco*kinetic and pharmacodynamic strategies in early research phases of drug discovery and development at Novartis Institute of Biomedical Research. Front Pharmacol. 2014;5:174. 5. Chang M, Chow SC, Pong A. Adaptive design in clinical research: issues, opportunities, and recommendations. J Biopharm Stat. 2006;16: 299-309.

PHARMAco*kINETIC/PHARMACODYNAMIC MODELING FOR DRUG DEVELOPMENT

6. Colburn WA, Lee JW. Biomarkers, validation and pharmaco*kineticpharmacodynamic modelling. Clin Pharmaco*kinet. 2003;42:997-1022.

24. Rajman I. PK/PD modelling and simulations: utility in drug development. Drug Discov Today. 2008;13:341-346.

7. Tan DS, Thomas GV, Garrett MD, et al. Biomarker-driven early clinical trials in oncology: a paradigm shift in drug development. Cancer J. 2009;15:406-420.

25. Aarons L, Karlsson MO, Mentré F, et al; COST B15 Experts. Role of modelling and simulation in Phase I drug development. Eur J Pharm Sci. 2001;13:115-122.

8. Yap TA, Sandhu SK, Workman P, et al. Envisioning the future of early anticancer drug development. Nat Rev Cancer. 2010;10:514-523.

26. Gibbs JP. Prediction of exposure-response relationships to support first-in-human study design. AAPS J. 2010;12:750-758.

9. Burman CF, Wiklund SJ. Modelling and simulation in the pharmaceutical industry: some reflections. Pharm Stat. 2011;10:508-516.

27. Mould DR, Walz AC, Lave T, et al. Developing exposure/response models for anticancer drug treatment: special considerations. CPT Pharmacometrics Syst Pharmacol. 2015;4:e00016.

10. Gieschke R, Steimer JL. Pharmacometrics: modelling and simulation tools to improve decision making in clinical drug development. Eur J Drug Metab Pharmaco*kinet. 2000;25:49-58. 11. Banerji U, Workman P. Critical parameters in targeted drug develop ment: the pharmacological audit trail. Semin Oncol. 2016;43: 436-445. 12. Workman P. How much gets there and what does it do? The need for better pharmaco*kinetic and pharmacodynamic endpoints in contemporary drug discovery and development. Curr Pharm Des. 2003;9:891-902. 13. Biomarkers Definitions Working Group. Biomarkers and surrogate endpoints in clinical trials: proposed definitions and conceptual framework. Clin Pharmacol Ther. 2001;69:89-95. 14. Baselga J, Cervantes A, Martinelli E, et al. Phase I safety, pharmaco*kinetics, and inhibition of SRC activity study of saracatinib in patients with solid tumors. Clin Cancer Res. 2010;16:4876-4883. 15. Tabernero J, Cervantes A, Rivera F, et al. Pharmacogenomic and pharmacoproteomic studies of cetuximab in metastatic colorectal cancer: biomarker analysis of a phase I dose-escalation study. J Clin Oncol. 2010;28:1181-1189. 16. Carden CP, Sarker D, Postel-Vinay S, et al. Can molecular biomarkerbased patient selection in phase I trials accelerate anticancer drug development? Drug Discov Today. 2010;15:88-97. 17. Balis FM, Thompson PA, Mosse YP, et al. First-dose and steady-state pharmaco*kinetics of orally administered crizotinib in children with solid tumors: a report on ADVL0912 from the Children’s Oncology Group phase 1/pilot consortium. Cancer Chemother Pharmacol. 2017;79:181-187. 18. Chapman PB, Hauschild A, Robert C, et al; BRIM-3 Study Group. Improved survival with vemurafenib in melanoma with BRAF V600E mutation. N Engl J Med. 2011;364:2507-2516. 19. Gobburu JV, Marroum PJ. Utilisation of pharmaco*kineticpharmacodynamic modelling and simulation in regulatory decisionmaking. Clin Pharmaco*kinet. 2001;40:883-892. 20. Chaikin P, Rhodes GR, Bruno R, et al. Pharmaco*kinetics/ pharmacodynamics in drug development: an industrial perspective. J Clin Pharmacol. 2000;40:1428-1438. 21. Barrett JS, Gupta M, Mondick JT. Model-based drug development applied to oncology. Expert Opin Drug Discov. 2007;2:185-209. 22. Sinha VK, Snoeys J, Osselaer NV, et al. From preclinical to human: prediction of oral absorption and drug-drug interaction potential using physiologically based pharmaco*kinetic (PBPK) modeling approach in an industrial setting—a workflow by using case example. Biopharm Drug Dispos. 2012;33:111-121. 23. Thiel C, Schneckener S, Krauss M, et al. A systematic evaluation of the use of physiologically based pharmaco*kinetic modeling for crossspecies extrapolation. J Pharm Sci. 2015;104:191-206.

28. Thai HT, Mazuir F, Cartot-Cotton S, et al. Optimizing pharmaco*kinetic bridging studies in pediatric oncology using physiologically-based pharmaco*kinetic modelling: application to docetaxel. Br J Clin Pharmacol. 2015;80:534-547. 29. O’Donnell A, Faivre S, Burris HA III, et al. Phase I pharmaco*kinetic and pharmacodynamic study of the oral mammalian target of rapamycin inhibitor everolimus in patients with advanced solid tumors. J Clin Oncol. 2008;26:1588-1595. 30. Tabernero J, Rojo F, Calvo E, et al. Dose- and schedule-dependent inhibition of the mammalian target of rapamycin pathway with everolimus: a phase I tumor pharmacodynamic study in patients with advanced solid tumors. J Clin Oncol. 2008;26:1603-1610. 31. Tanaka C, O’Reilly T, Kovarik JM, et al. Identifying optimal biologic doses of everolimus (RAD001) in patients with cancer based on the modeling of preclinical and clinical pharmaco*kinetic and pharmacodynamic data. J Clin Oncol. 2008;26:1596-1602. 32. Motzer RJ, Escudier B, Oudard S, et al; RECORD-1 Study Group. Efficacy of everolimus in advanced renal cell carcinoma: a double-blind, randomised, placebo-controlled phase III trial. Lancet. 2008;372:449456. 33. Yao JC, Shah MH, Ito T, et al; RAD001 in Advanced Neuroendocrine Tumors, Third Trial (RADIANT-3) Study Group. Everolimus for advanced pancreatic neuroendocrine tumors. N Engl J Med. 2011;364:514-523. 34. Zhao L, Ren TH, Wang DD. Clinical pharmacology considerations in biologics development. Acta Pharmacol Sin. 2012;33:1339-1347. 35. Postel-Vinay S, Aspeslagh S, Lanoy E, et al. Challenges of phase 1 clinical trials evaluating immune checkpoint-targeted antibodies. Ann Oncol. 2016;27:214-224. 36. Agoram BM, Martin SW, van der Graaf PH. The role of mechanismbased pharmaco*kinetic-pharmacodynamic (PK-PD) modelling in translational research of biologics. Drug Discov Today. 2007;12:10181024. 37. Patnaik A, Kang SP, Rasco D, et al. Phase I study of pembrolizumab (MK-3475; anti-PD-1 monoclonal antibody) in patients with advanced solid tumors. Clin Cancer Res. 2015;21:4286-4293. 38. Elassaiss-Schaap J, Rossenu S, Lindauer A, et al. Using modelbased “learn and confirm” to reveal the pharmaco*kineticspharmacodynamics relationship of pembrolizumab in the KEYNOTE-001 trial. CPT Pharmacometrics Syst Pharmacol. 20176:2128. 39. Ahamadi M, Freshwater T, Prohn M, et al. Model-based characterization of the pharmaco*kinetics of pembrolizumab: a humanized anti-pd-1 monoclonal antibody in advanced solid tumors. CPT Pharmacometrics Syst Pharmacol. 2017;6:49-57. 40. DiMasi JA, Grabowski HG. Economics of new oncology drug development. J Clin Oncol. 2007;25:209-216.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 215

RUBIN ET AL

Strategies to Maximize Patient Participation in Clinical Trials Eric H. Rubin, MD, Mary J. Scroggins, MA, Kirsten B. Goldberg, MA, and Julia A. Beaver, MD OVERVIEW Despite considerable interest and success in oncology drug development, the minority of patients with cancer diagnoses enroll in clinical trials. Multiple obstacles account for this low enrollment rate. An improvement in patient participation in clinical trials could increase patient access to novel and potentially promising agents, provide faster trial results, and, with implementation of rational eligibility criteria, allow for a better understanding of the drug’s safety and efficacy in a heterogeneous population. We present barriers and potential solutions to maximize patient participation, including a review of the ASCO and Friends of Cancer Research (FoCR) Modernizing Eligibility Criteria Project, U.S. Food and Drug Administration (FDA) regulatory considerations, an industry perspective, and a patient perspective.

he goal of oncology clinical trials is to understand the risks and benefits of a therapy and to facilitate and expedite the development of safe and effective drugs to treat patients with cancer. Clinical trials also provide patients with access to investigational agents; however, U.S. oncology clinical trials only enroll approximately 3% of patients diagnosed with a new cancer.1 Multiple barriers can contribute to this low rate of enrollment, including those at the patient, physician, institutional, and protocol levels.2,3 Patient-level barriers to enrollment in clinical trials result from the fear that clinical trials will delay initiation of antineoplastic drugs (particularly if biopsy and genomic sequencing are required), fear of undergoing additional testing and procedures, or concerns about enrolling in randomized trials that might include a placebo or perceived inferior investigational or control arm. Other patient barriers include socioeconomic issues, such as concerns over travel costs with increased frequency of follow-up visits in a clinical trial. The lack of clinical trial access and/ or a decline in functional status are additional commonly reported reasons that patients are not enrolled in clinical trials.4 Physician-level barriers such as lack of knowledge about new agents and available clinical trials may also present obstacles to enrollment. Institutional-level barriers are reflected by the number of available protocols at one institution, and the fact that for many community practices, knowledge about the potential for referral to clinical trials may be limited. Overly restrictive eligibility criteria are a major protocol-level barrier. Given these obstacles, strategies are needed to maximize patient participation in clinical trials.

MAXIMIZING CLINICAL TRIAL PARTICIPATION AND IMPLEMENTATION OF MODERN ELIGIBILITY: INDUSTRY PERSPECTIVE From an industry perspective, key parameters in clinical trial implementation and participation are as follows: (1) speed and efficiency in evaluating the safety and efficacy of an experimental oncology agent, (2) investigator and site experience with investigational drug trials (including obtaining patient informed consent and assessing adverse events), (3) speed of trial initiation at sites, (4) site accrual rates, (5) site data quality, and (6) investigator experience with the pathway targeted by the experimental agent. Protocols for industry-sponsored clinical trials undergo a rigorous internal review process, typically involving multiple review committees with members possessing expertise in trial design, statistics, and regulatory, safety, data management, and operational aspects of clinical trials. During protocol development, industry sponsors typically obtain investigator input and ensure that the patient perspective is understood in order to confirm feasibility and maximize accrual rates. Single-arm, personalized treatments (e.g., trials that select among various treatments based on a biomarker evaluation of a tumor biopsy) are often attractive to patients; thus, these trials tend to accrue rapidly. Single-arm trials may be sufficient for regulatory approval in some cases, such as rare cancers and/or where initial data suggest a remarkable improvement relative to existing treatment options. However, in most cases, randomized controlled trials will be necessary to demonstrate clinical benefit. In these trials, accrual rates can vary widely depending on patient perspectives on the potential benefits of the experimental and control arms.

From the Merck Research Laboratories, North Wales, PA; Pinkie Hugs, LLC, Washington, DC; U.S. Food and Drug Administration, Silver Spring, MD. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Julia A. Beaver, U.S. Food and Drug Administration, 10903 New Hampshire Ave., WO 22, Room 2359, Silver Spring, MD 20993; email: julia.beaver@fda. hhs.gov. © 2017 American Society of Clinical Oncology

216 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

MAXIMIZING PATIENT CLINICAL TRIAL PARTICIPATION

Clinical Trial Designs to Improve Enrollment

Depending on the molecular target for an investigational agent, it may be preferable to use clinical trials that use enrichment designs (i.e., eligibility is dependent on having a “positive” result from a biomarker test performed on a tumor or blood specimen from the patient), particularly if the target is a mutated gene that has a low prevalence among a given tumor type. Using adaptive trials (e.g., trials that include prespecified changes based on accumulating data) may improve efficiency relative to use of multiple sequential trials, each requiring several internal and external committee reviews as well as separate site contracting and trial initiation efforts.5-7 However, there may be tension in terms of the degree of flexibility desired by an industry sponsor versus prespecification of sample sizes based on statistical evaluations of efficacy and safety. In addition, adaptive trials increase protocol complexity, particularly if multiple amendments are required even if the adaptive changes are prespecified. In some types of enrichment trials, it may be expected that that the biomarker test may identify responsive patients regardless of tumor type. In this case, “basket” trials have become increasing popular, in which eligibility is determined by the biomarker test rather than the tumor type.8 However, this approach may create challenges in terms of selection of site principal investigators, because cancer centers are not organized by biomarkers but by tumor type. Thus, it may be difficult to identify an optimal principal investigator in terms of patient accrual to the trial.

KEY POINTS • Despite considerable interest and success in oncology drug development, the minority of patients with cancer diagnoses enroll in clinical trials, owing to barriers at the patient, physician, institutional, and protocol levels. • An ASCO and FoCR Modernizing Eligibility Criteria Project, in collaboration with the FDA and other stakeholders, is working toward evaluating clinical trial entrance criteria that may unnecessarily restrict clinical trial access and providing recommendations for a more rational approach to determining inclusion and exclusion criteria. • Working groups with representatives from ASCO, FoCR, FDA, patient advocacy programs, industry, and others have prioritized assessments on criteria for patients with brain metastases, organ dysfunction, history of prior malignancy, and HIV and for those younger than age 18 to come up with recommendations for a more nuanced approach to determining eligibility. • Creative clinical trial designs such as enrichment designs, master protocols, and “basket” trials in the right clinical scenario can maximize patient participation and efficiency. • Clinical trials hold the promise of providing benefit to patients/survivors and patient-level barriers to enrollment can be addressed through reasoned interventions.

Master protocols are another trial design that has become increasingly popular as a means of improving efficiency of oncology drug development for a particular tumor type.9 When testing multiple drugs with non-overlapping eligibility criteria (e.g., drugs targeting non-overlapping gene mutations), these designs have advantages over use of multiple two-arm randomized registration trials. First, grouping these trials under a single protocol, with a common control arm, reduces the overall screen failure rate. For example, assuming a prevalence of 20% for biomarkers A, B, C, and D in a given histologic cancer type (with no overlap among each subpopulation) and a need for 200 biomarker-positive patients each on an experimental arm and in the treatment-control arm, 8,000 patients would need to be screened in the case of four separate randomized studies, whereas only 2,163 would need to be screened in the case of a single five-arm study with four experimental arms and one control treatment arm. Second, process and operational efficiencies are improved through the ability to amend a single master protocol as needed as drugs enter and exit the trial. For example, after implementation, sponsors enrolling new drugs would benefit from the presence of a preexisting infrastructure. Although master protocols can improve the efficiency of drug development, they may not be fully endorsed by industry, particularly if trial arms include similar agents.

Implementation of Modern Eligibility

Although there are already examples of industry sponsors embracing changes in traditional eligibility criteria that have been suggested recently by various stakeholders (see section on modernizing eligibility criteria), challenges remain. Because industry typically desires to register new drugs in many different countries, registration trials usually involve sites from many different countries, and in certain countries, the view on reducing eligibility restrictions may differ from that of the United States, particularly as related to age. Furthermore, although mitigating factors have been described, allowing patients with impaired organ function or per formance status may bias evaluation of safety and efficacy, potentially leading to premature discontinuation of the development of a particular agent. In some trials, accrual may be dominated by non-U.S. sites as a result of limited access of patients in some countries to investigational drugs or newly approved drugs. This can result in U.S. filing applications consisting of patient data from predominantly nonU.S. sites.

Strategies to Address Low Enrollment

Some studies enroll poorly, which may relate to several possible issues. These reasons may include lack of interest in the investigational agent or the control arm treatment, overly restrictive eligibility criteria, complex requirements (e.g., prolonged inpatient stays, uncomfortable procedures, or multiple invasive biopsies), or, in the case of enrichment trials, a low prevalence of biomarker positivity. Another possible reason for slow enrollment is the existence of similar studies competing for the same patient population, asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 217

RUBIN ET AL

particularly if the eligible population is uncommon (e.g., a biomarker-selected population in which the prevalence of biomarker positivity is low). Although many sites attempt to limit the number of trials competing for the same population, this does not address competition for patients between sites. Sponsors may evaluate several options in cases of slow enrollment. A commonly used option is to simply add additional sites. If there is a high rate of screen failures, eligibility criteria may be re-evaluated, particularly if one criterion is a predominant reason for screen failures. For example, the number and types of previous treatment allowed may be too restrictive. In addition, for randomized trials, patients and investigators may not view the control arm as an attractive treatment option. In this case, adding additional treatment options to the control arm (e.g., allowing investigators to choose among a list of treatments), or allowing crossover to the investigation arm upon disease progression, may improve accrual rates.

MODERNIZING ELIGIBILITY CRITERIA FOR CLINICAL TRIALS

Historically, eligibility criteria were appropriately put in place because of concerns over safety in selected populations but, in many cases, clinical trial protocols are copied forward between and within drug development portfolios and are not always based on a rational analysis. It is critically important for developers of clinical trials to take a more thoughtful approach to the selection of eligibility criteria, not only to provide improved access to clinical trials for patients with cancer but also to understand a drug’s safety and efficacy in a more representative population. Despite years of recognition of this issue, inclusion and exclusion criteria remain prohibitive for many patients.10-13 Certain populations in particular are frequently excluded from oncology clinical trials, including patients with HIV, brain metastases, history of prior malignancies, poor performance status, and comorbidities and those younger than age 18.13 Of approximately 300 commercial Investigational New Drug Applications submitted to the FDA's Office of Hematology and Oncology Products in 2015, only 3.7% included pediatric patients; 60% required Eastern Cooperative Oncology Group performance status of 0–1; 77% excluded known, active, or symptomatic central nervous system or brain metastases; 47% allowed treated or stable brain metastases; and 84.2% excluded patients with known or active HIV (with only 1.7% allowing patients to enroll with adequate CD4 counts).14 Multiple stakeholders realize that taking a more rational approach to eligibility criteria will result in improved patient benefit.

ASCO/Friends of Cancer Research/FDA Modernizing Eligibility Criteria Project and Working Groups

The ASCO and FoCR Modernizing Eligibility Criteria Project, in collaboration with the FDA and other stakeholders, is working toward evaluating clinical trial entrance criteria that may unnecessarily restrict clinical trial access and 218 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

providing recommendations for a more rational approach to determining inclusion and exclusion criteria.13,15 Working groups were formed with representatives from ASCO, FoCR, FDA, patient advocacy programs, industry, the National Cancer Institute, biostatisticians, pharmacologists, and clinical investigators to come up with recommendations for a more nuanced approach to determining eligibility. These groups have prioritized assessments on criteria for patients with brain metastases, organ dysfunction, history of prior malignancy, and HIV and those younger than age 18.

Working Group Preliminary Recommendations

Publications from these working groups are pending; however, preliminary recommendations presented at the FoCR Annual Meeting in November 2016 detail current working group thinking.16 The brain metastases working group endorsed the routine inclusion of patients with treated or stable brain metastases in all phases of clinical trials unless there is a compelling rationale for exclusion.16 In certain instances, patients with new, active, or progressing brain metastases may also be included, taking the history of the patient’s disease, trial phase and design, drug mechanism, and potential for central nervous system interaction into account. Patients with leptomeningeal disease may also have specific situations that warrant an eligible cohort in earlyphase trials. The minimum age working group proposed that pediatric-specific cohorts be included in dose-finding studies in which strong scientific rationale is present. This rationale could be based on preclinical data or an understanding of the mechanism of the disease. In later stages of drug development, the group proposed that trials in diseases that span adult and pediatric populations should enroll pediatric patients, particularly patients age 12 and older. Others, including the FDA, have also suggested the inclusion of patients age 12–17 in appropriate adult disease-specific trials.17 The working group also proposed that HIV-related eligibility criteria be rationally developed and focus on current and past CD4 and T-cell counts, a history of AIDSdefining conditions, and status of HIV treatment.16 The working group advised that HIV should be considered a comorbidity and antiretroviral therapy should be considered a concomitant medication. The organ dysfunction working group proposed that eligibility regarding renal function should be based on creatinine clearance rather than serum creatinine levels, and knowledge of a drug’s excretion by a specific organ system could inform exclusion criteria cutoffs. In addition, it was advised that exclusions based on prior malignancy be liberalized.

Regulatory Considerations of Eligibility

Regulatory incentives that might encourage a thoughtful approach to eligibility are also possible. For example, an expanded marketing claim could be granted if an adequately studied patient cohort included a previously excluded population such as patients with brain metastases. In addition, postmarketing requirements or commitments, such as the study of organ impairment, might be unnecessary if these

MAXIMIZING PATIENT CLINICAL TRIAL PARTICIPATION

populations were included previously. Pediatric incentives include the ability to address requirements from the Pediatric Research Equity Act to study the effects of drugs on children. So that efficacy is not compromised in trials intended to support registration, a broader clinical trial population could include a prespecified, more narrowly defined population for the primary efficacy evaluation.

ASCO/FoCR/FDA Eligibility Criteria Future Directions

Appropriate eligibility criteria define a patient population that will result in patient protection, but strict eligibility criteria can negatively affect patient participation and result in failure to understand a drug’s safety and efficacy in a representative population. Future endeavors of the ASCO/FoCR/ FDA project will focus on approaches to appropriately defining drug washout periods, exclusion of concomitant medications, and inclusion of elderly patients.16,18 The Modernizing Eligibility Criteria Project advocates for a culture shift in the approach to inclusion and exclusion criteria and will continue to pursue a broad implementation of this rational approach to eligibility.

IMPROVING ACCESS TO CLINICAL TRIALS: A PATIENT PERSPECTIVE

or unequal recruitment (including physician conscious and unconscious bias), (5) patient-physician communication (or the lack thereof), (6) access and logistical considerations (including financial, geographic, and educational considerations), and (7) the fear factor (e.g., of being randomized or of randomization not being so random and of being the object of unfettered experimentation). These barriers and others combine to create limited and unequal participation in clinical trials, with differing effects on various populations, partially depending on their experience with the health care system. For example, “history” is often cited as a barrier to participation by populations described as vulnerable or “special” (a misnomer); however, history can be a barrier for any population and any individual who understands the medical misconduct and infamous studies that litter the research landscape. When known and understood, events such as the Nazi experiments (1940s), the Willowbrook studies (1956–1972), the Jewish Chronic Disease Hospital studies (1963), the AIDS trials (1980s), and the Tuskegee syphilis study (1932–1972), which is perhaps the most often cited example, prompt or exacerbate distrust in the system and increase reluctance to participate even with the potential of benefit to the participant or to others.

Although it is important to acknowledge that there is no single patient or patient advocate perspective or consensus on clinical trials. The views and suggestions offered here are those of an individual advocate informed and enriched by more than 20 years of active engagement across numerous advocacy groups.

Distinguishing Myth From Reality

A Goal of Patient Benefit

Trial Participation in the Age of Personalized Medicine

Simply put, clinical trials hold the promise—the enormously hopeful potential—of providing benefit to patients/ survivors. Patient advocates, some as research advocates, support the clinical trial enterprise in numerous ways and participate in clinical trials for many reasons—one being the hope of personal benefit and another being the desire to contribute to the likelihood of benefit to future generations. Low trial enrollment decreases the speed and likelihood of trial progress and thus ultimate patient benefit (practice changing or incremental), wastes precious human and funding resources, and compromises confidence in the entire enterprise (thus becoming an additional barrier to enrollment).

Overview of Patient Barriers

Barriers to participation in clinical trials vary widely across institutions, professions, and populations. Low trial enrollment, relative to the available pool of participants, is a recurring topic or theme at nearly all conferences, meetings, and other gatherings of individuals involved in the clinical trial enterprise, seeking access to trials, or hoping to benefit from them. The long list of often-overlapping barriers includes (1) a history of unethical trials, (2) lack of understanding of clinical trials or the availability of specific trials, (3) lack of trust in the medical system, (4) uneven

Important overarching concerns related to clinical trial participation include limited understanding of clinical trial terminology, standards, and protections. These concerns support the rise and maintenance of myths, such as those described in Table 1.

Clinical trial participation is the primary route through which biospecimens are obtained and banked, thus serving as a gateway for individual access to personalized medicine and health care. As such, it is increasingly important for all populations to be represented in the clinical trial enterprise. Because they are not equally represented, it is not surprising that banked biospecimens do not represent the diversity of the general population and that the findings derived from these biospecimens are not widely generalizable to segments of the population. Without trial participation across populations, underrepresented populations will have little or no “skin in the game.” An unintended consequence is likely to be an increase in cancer health disparities. Therefore, with a substantial share of research efforts and research dollars focused on personalized medicine and health care, representative trial participation is a must.

Potential Strategies and Interventions

Like the barriers to clinical trial participation and low enrollment, potential solutions are frequently offered, if not implemented. The following potential solutions are offered as doable, feasible, and measurable: 1. Acting on what we know and have researched (including using best practices), asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 219

RUBIN ET AL

TABLE 1. Myths and Realities of Clinical Trial Participation Myth

Reality

The Tuskegee syphilis study is the reason for low enrollment of black patients and perhaps other vulnerable populations.

The Tuskegee syphilis study or what it represents is a reason (i.e., one among many others), not the reason. In fact, it can become an excuse for the opportunity to participate not being offered.

Black and Hispanic patients are less likely to participate in clinical trials than white patients.

Although both groups are less likely to be invited to participate in trials than white patients, when asked, they are slightly more likely to enroll.

Trial participants are guinea pigs and may be experimented on in ways beyond their consent.

Numerous regulations and safeguards, including 45CFR46, ensure that human participants are protected and that research is ethical. All U.S. research involving human participants is reviewed and monitored by an institutional review board, whose focus is to protect participants.

Clinical trials are or should be an option only when potential participants have no other treatment options.

Clinical trial participation can and should be an option for any patient who meets specific trial eligibility requirements and wants to participate. Not all patients will choose to participate, but all should be given the option to do so when appropriate.

Some clinical trial participants will receive treatment inferior to standard of care, perhaps through randomization or placebo use.

All clinical trial participants will receive at least standard of care. Randomization and placebo strike fear in the hearts of potential participants. This fear can be educated away.

2. Focusing on recruitment of—as opposed to continuously studying—vulnerable and special populations (as a largely untapped resource and as a matter of good conscience and good science), 3. Funding and implementing bidirectional clinical trial education and awareness (to include a sustained public awareness campaign, communications skills, and cultural sensitivity training for the public, patients and survivors, health care providers, and researchers), 4. Developing trial-specific educational material (as well as trial-general material with substantial meaningful patient advocate involvement), 5. Instituting accountability relative to uneven/unequal recruitment and unmet goals (to include the requirement for rigorously reviewed populationspecific recruitment goals and the implementation of consequences where warranted), 6. Reviewing and modernizing eligibility requirements (understanding, for example, that exclusion of potential participants with comorbidities—unless scientifically warranted—has a profound effect on eligibility by population and that trial participants should more closely align with the general population that might benefit from the trial), 7. Formalizing engagement of patient advocates beyond recruitment and throughout and beyond protocol development and review, and 8. Requiring an informed consent process as well as a signed informed consent document.

This may seem a smorgasbord or data dump of possibilities. It is not. Instead, it is a listing of potential solutions that can and should often be combined and fashioned into interventions that move from discussing low enrollment and the concomitant barriers to overcoming them.

In Summary—A Patient Advocate’s Perspective

This cancer survivor and patient advocate’s perspective focuses on the implementation of interventions old and new, alone and in combination. The breadth and depth of research on clinical trial participation have been extensive and the conversation is ongoing; however, to effect measurable change, we must move from conversation to research-based and best practice–informed action, that is, reasoned interventions.

CONCLUSION

Although restrictions on clinical trial entry for the protection of patients are appropriate and supported by all stakeholders, an examination of more nuanced eligibility is appropriate in many cases. This rational approach to defining eligibility will benefit patients by providing clinical trial access and ultimately resulting in a greater knowledge of a drug upon approval. Additional barriers can also be approached with similar efforts to expand and maximize patient participation in clinical trials. Through collaborative efforts across academia, government, industry, and advocacy, there is great promise and potential for maximizing patient participation in oncology clinical trials.

References 1. Lara PN Jr, Higdon R, Lim N, et al. Prospective evaluation of cancer clinical trial accrual patterns: identifying potential barriers to enrollment. J Clin Oncol. 2001;19:1728-1733.

3. Bennette CS, Ramsey SD, McDermott CL, et al. Predicting low accrual in the National Cancer Institute’s Cooperative Group Clinical Trials. J Natl Cancer Inst. 2015;108:djv324.

2. Baquet CR, Henderson K, Commiskey P, et al. Clinical trials: the art of enrollment. Semin Oncol Nurs. 2008;24:262-269.

4. Sohal DP, Rini BI, Khorana AA, et al. Prospective clinical study of precision oncology in solid tumors. J Natl Cancer Inst. 2015;108:djv332.

220 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

MAXIMIZING PATIENT CLINICAL TRIAL PARTICIPATION

5. Prowell TM, Theoret MR, Pazdur R. Seamless oncology-drug development. N Engl J Med. 2016;374:2001-2003. 6. Papadimitrakopoulou V, Lee JJ, Wistuba II, et al. The BATTLE-2 study: a biomarker-integrated targeted therapy study in previously treated patients with advanced non-small cell lung cancer. J Clin Oncol. 2016;34:3638-3647. 7. Berry DA. Adaptive clinical trials in oncology. Nat Rev Clin Oncol. 2011;9:199-207. 8. Beckman RA, Antonijevic Z, Kalamegham R, et al. Adaptive design for a confirmatory basket trial in multiple tumor types based on a putative predictive biomarker. Clin Pharmacol Ther. 2016;100: 617-625. 9. Herbst RS, Gandara DR, Hirsch FR, et al. Lung Master Protocol (LungMAP)-a biomarker-driven protocol for accelerating development of therapies for squamous cell lung cancer: SWOG S1400. Clin Cancer Res. 2015;21:1514-1524.

12. f*cks A, Weijer C, Freedman B, et al; National Surgical Adjuvant Breast and Bowel Program. Pediatric Oncology Group. A study in contrasts: eligibility criteria in a twenty-year sample of NSABP and POG clinical trials. J Clin Epidemiol. 1998;51:69-79. 13. Kim ES, Bernstein D, Hilsenbeck SG, et al. Modernizing eligibility criteria for molecularly driven trials. J Clin Oncol. 2015;33:2815-2820. 14. Sridhara R. Expansion of eligibility criteria: trial design considerations. Presented at: Friends of Cancer Research Annual Meeting; November 2016; Washington, DC. 15. Kim ES, Atlas J, Ison G, et al. Transforming clinical trial eligibility criteria to reflect practical clinical application. Am Soc Clin Oncol Educ Book. 2016;36:83-90. 16. Kim ES. ASCO-Friends of Cancer Research Modernizing Eligibility Criteria Project. Presented at: Friends of Cancer Research Annual Meeting; November 2016; Washington, DC.

10. Yusuf S, Collins R, Peto R. Why do we need some large, simple randomized trials? Stat Med. 1984;3:409-420.

17. Chuk MK, Mulugeta Y, Roth-Cline M, et al. Enrolling adolescents in disease/target-appropriate adult oncology clinical trials of investigational agents. Clin Cancer Res. 2017;23:9-12.

11. George SL. Reducing patient eligibility criteria in cancer clinical trials. J Clin Oncol. 1996;14:1364-1370.

18. Singh H, Beaver JA, Kim G, et al. Enrollment of older adults on oncology trials: an FDA perspective. J Geriatr Oncol. Epub 2016 Dec 22.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 221

FLAHERTY, LE, AND LEMERY

Tissue-Agnostic Drug Development Keith T. Flaherty, MD, Dung T. Le, MD, and Steven Lemery, MD, MHS OVERVIEW The U.S. Food and Drug Administration (FDA) has approved drugs to treat patients with tumor types based on a single anatomic site, such as renal cell carcinoma or melanoma, rather than on a biomarker alone. This standard approach is based on a number of factors, including heterogeneity of drug effects in different biomarker-positive tumor types. Additionally, drug development for some drugs was primarily directed toward a specific genomic abnormality in a specific tumor type (e.g., drugs for anaplastic lymphoma kinase [ALK] fusion-positive non–small cell lung cancer). In such cases, differences in biology, differences in natural histories of different cancers, differences in mutation frequencies among cancers, or differences in concomitant therapies may have necessitated diverse development considerations. As described in U.S. regulations [21 CFR 201, CFR 201.57(c)(2)], the indications and usage section of drug labeling “must state that a drug is indicated for the treatment, prevention, mitigation, cure, or diagnosis of a recognized disease or condition or of a manifestation of a recognized disease or condition, or for the relief of symptoms associated with a recognized disease or condition.” Such regulations, however, do not require that disease be defined solely as a specific tumor type. This manuscript will highlight scientific/biologic issues, clinical trial designs, and regulatory issues pertaining to the development of drugs agnostic of tumor type. Although the manuscript will discuss regulatory considerations as understood by the authors regarding tissue-agnostic drug development, it should not be considered formal or binding FDA guidance or policy.

ome of the earliest successes in developing targeted therapy for genetically defined subsets of cancer occurred by targeting genetic alterations that, in retrospect, were nearly restricted to one or two cancer types. This was the case for abl kinase targeting with imatinib, EGFR inhibitors for EGFR activating mutations, and ALK inhibitors for ALK fusions. As the tools to unravel the molecular biology of cancer have enabled complete characterization of the hundreds of cases of all common cancers and many uncommon ones, it is clear that cancers arise from common somatic genetic building blocks.

SCIENTIFIC/BIOLOGIC CONSIDERATIONS AND TARGET SELECTION FOR TRIALS TO IDENTIFY PATIENTS FOR TREATMENT AGNOSTIC OF TUMOR TYPE

The Cancer Genome Atlas project and other publicly funded studies rediscovered common genetic alterations that were variably represented across cancer types defined by site of origin (e.g., PIKC3A, RAS, BRAF, Her-2, TP53, PTEN, CDKN2A, etc.).1 New insights were also gleaned into the commonness of genetic changes in components of complex, multicomponent molecular machinery (such as the SWI-SNF and spliceosome complexes). In the case of Her-2 amplified and BRAF mutant tumors, it has become clear that the spectrum

of efficacy observed in one tumor type can vary substantially when comparing various cancer types harboring a specific genetic alteration.2-5 Perhaps most striking is the case of BRAF, where BRAF inhibitor monotherapy has profound efficacy in melanoma that is not yet equaled in colorectal cancer, even with triple-drug regimens targeting BRAF, MEK, and EGFR.6 This precedent established the principle that one should assume heterogeneity, not hom*ogeneity when investigating novel targeted therapy strategies. More recently, even immunotherapy has been subject to similar considerations. For the field-changing class of PD-1/PD-L1 antibodies, it has been established that higher mutation burden, infiltration of CD8+ T cells, and expression of PD-L1 on tumor and/or infiltrating immune cells can predict response.7 But, the predictive accuracy varies across cancer types.8 As new immunotherapies are being developed, the question arises as to whether their development would be accelerated by understanding whether new single agents or combinations building on a PD-1/PD-L1 antibody backbone might confer benefit similarly or differently in various cancer types that are profiled at the level of these analytes. Preclinical models that might aid in predicting the most-responsive or most-resistant tumor types for a given therapy are poorly developed for many cancer types. Models that reflect the full genetic complexity of human cancer,

From the Massachusetts General Hospital Cancer Center, Boston, MA; The Sidney Kimmel Cancer Center at Johns Hopkins, Baltimore, MD; Office of Hematology and Oncology Products, Center for Drug Evaluation and Research, U.S. Food and Drug Administration, Silver Spring, MD. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Steven Lemery, MD, MHS, U.S. Food and Drug Administration, 10903 New Hampshire Ave., Silver Spring, MD 20993; email: [emailprotected]. © 2017 American Society of Clinical Oncology

222 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

TISSUE-AGNOSTIC DRUG DEVELOPMENT

in a representative microenvironment, and in the setting of an intact immune system, are a particular unmet need in cancer therapeutic development. Thus, new therapies are brought into clinical development with little ability to predict which cancer types would be most impacted.

Principles of Variable Response to Oncogene/Signal Transduction Targeted Therapy

In cases of single-agent targeted therapy development, variable response within a single tumor type is the rule. The full spectrum of durable responses and de novo refractory disease has been well cataloged in the case of BRAF in melanoma, EGFR in EGFR mutant non–small cell lung cancer,9 and ALK in non–small cell lung cancers harboring ALK fusions.10 The addition of Her-2 antibodies to conventional cytotoxic chemotherapy in Her-2 amplified breast cancer follows the same pattern.2 Extensive investigation into the causes of de novo resistance and susceptibility have highlighted the contribution of cells in the tumor microenvironment providing growth factor-mediated survival signals, compensatory signaling as a consequence of dysregulated feedback mechanisms, and concomitant somatic genetic alterations present at baseline that mediate compensatory signaling. Even the contribution of tumor/immune interactions has been implicated in responsiveness of BRAF mutant melanomas to BRAF inhibitors.11 It is logical to hypothesize that heterogeneity in one or more of these features would account for variable response to a new therapy being prospectively investigated. Taking the case of BRAF, sensitivity was equally demonstrated in melanoma and colorectal cancer (CRC) cell lines harboring V600 BRAF mutations.12 But, only after observing widespread unresponsiveness in the colorectal population treated as one of two expansion cohorts in the phase I trial of vemurafenib were preclinical investigations launched that uncovered the ability of EGFR receptor signaling to rescue MAP kinase signaling and maintain PI3K pathway signaling in colon cancer models.13,14 Interestingly, upregulation

KEY POINTS • Most somatic genetic alterations in cancer are distributed across several to many cancer types. • Preclinical model systems poorly predict relative sensitivity or resistance to therapies targeting these alterations. • Early-phase clinical trials increasingly explore comparative efficacy across the spectrum of biomarkerdefined tumor types. • To date, durable objective responses have been observed in patients across different microsatellite instability– high/mismatch repair–deficient tumor types when treated with checkpoint inhibitors. • If scientifically and clinically appropriate, investigating the effects of a drug agnostic of tumor type may be one pathway for drug development; however, every drug presents unique circ*mstances in regard to the population of patients who might benefit from it.

of EGFR has been implicated as a component of receptor tyrosine kinase–mediated resistance in melanoma, but as a component of acquired, not de novo, resistance.15 Whereas the presence or absence of concomitant genetic alterations beyond the index alteration that is being therapeutically targeted can be assessed relatively easily in the era of clinical next-generation sequencing, predictive adaptive resistance mechanisms that do not have a hard-wired somatic genetic basis are not readily diagnosed a priori.

Somatic Genetics Features

Concomitant genetic alterations that might mediate resistance to a targeted therapy (such as PTEN or CDKN2A loss in concert with a V600 BRAF mutation in melanoma) can be considered as effect modifiers. To mediate de novo resistance, such alterations must be present in the vast majority of tumor cell clones, presuming that cells that lack such alterations would be sensitive to therapy. Subclonal events that are present in a small minority of cells would be expected to account for resistance and disease progression following an initial period of disease control or response. Extreme examples of this are the presumed presence of gatekeeper mutations that impair drug binding to EGFR and ALK inhibitors that co-occur with an activating mutation that activates the kinase domain of EGFR or a translocation that drives ALK overexpression.16,17 Such gatekeeper mutations are rarely found in the untreated state but commonly emerge under selective pressure of targeted therapy over the course of months. As such, these types of acquired resistance mechanisms seem to have little to do with variable initial response to treatment. Another possibility has recently emerged in the targeted therapy development landscape: subclonal presence of the targeted oncogene itself. This appears to be the case in certain PIK3CA mutant cancers.18,19 For years, PIK3CA mutant cancers were minimally impacted by the first generation of nonspecific PI3K inhibitors. With the emergence of inhibitors that are relatively selective for the alpha isoform of PI3K, which is the isoform that is activated by the mutation, muted responses have been observed across cancer types that harbor these common mutations (including breast cancers, head and neck cancers, and endometrial cancers). Genetic characterization of tumor specimens procured at baseline and at the time of disease progression has revealed that PIK3CA mutations are commonly subclonal and that PIK3CA wild-type tumors cells become over-represented in progression samples. In cases where mutant PI3KCA is clonal, concomitant genetic loss of PTEN can mediate resistance, providing a mechanism analogous to the other oncogene targeted therapy precedents described above. The PIK3CA case provides another dimension that may account for variable response and resistance across cancers that harbor genetic features: truncal versus subclonal target gene alteration.

Lineage-Specific Resistance

Beyond the somatic genetic makeup of cancers, the large remainder of resistance mechanisms to oncogene/signal asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 223

FLAHERTY, LE, AND LEMERY

transduction targeted therapy lies in the epigenetic domain. On the surface, this refers to any de novo or adaptive resistance mechanism that reflects altered gene expression or signaling without requirement of an upstream genetic alteration that activates or inactivates. For the case of variable response to BRAF inhibitors in melanoma versus colorectal cancer, an epigenetic mechanism appears to underlie the utilization of EGFR signaling to maintain CRAF-MEK-ERK signaling in the face of BRAF inhibition.14,20 Interestingly, this program does not appear to be as readily accessible to melanoma, raising the hypothesis that cell of origin might be an important factor for lineage-specific therapeutic resistance. Extensive research has highlighted the baseline transcriptional state of BRAF mutant melanomas and how that state is altered upon exposure to BRAF inhibitors. It appears that coordinated alteration in the melanocyte-specific transcription factor MITF and other TFE-3 family transcription factors can, but does not always, facilitate utilization of the beta-catenin signaling pathway and altered cellular metabolism.21,22 These downstream consequences of BRAF inhibitor therapy might have their analog in other cancers that harbor BRAF mutations, but are almost certainly mediated by other transcription factors that play a role in the tissue-specific identity of those cells. The relative contribution of concomitant genetic alterations and epigenetic phenomena is far from completely understood in any of the cases of successfully developed oncogene/signal transduction targeted therapy. In each case, it is only after a therapeutic effect has been observed that analysis of patients’ tumor samples and parallel investigation in preclinical models have shed light on variable response within and across tumor types. It is this unmet need that, in part, informs the rise of functional diagnostics as an approach for judging response in minimally manipulated tumor samples ex vivo or, using novel microdevices for local delivery of drugs, in vivo.23,24 These technologies are in their infancy with regard to use in cancer drug development. As such, it remains challenging to anticipate how the currently standard preclinical methods alone will enable prediction of therapeutic effects of a new class of therapies being brought forward.

The Clinical Trial Paradigm

In light of the current state of preclinical prediction, the standard paradigm for early-phase clinical trials is to perform dose escalation in unselected or biomarker-selected patients across tumor types followed by an exploration of efficacy in expansion cohorts. In the case of the vemurafenib phase I trial, unselected patients were enrolled in dose escalation (in part because diagnostic methods for BRAF mutation testing in real time were lacking), and two expansion cohorts were explored in which patients with melanoma and CRC were required to have a V600 BRAF mutation.4,5 The target population was initially 20 patients. As opposed to formal phase II trials with statistical power to rule out a meaningful response rate, dose expansion cohorts of this size serve roughly the same purpose as the first stage of 224 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

accrual in a Simon two-stage design. Depending on the clinical context with regards to unmet medical need, different threshold levels of response might be sought. The 20-patient CRC cohort was sufficient to declare vemurafenib as unworthy of further investigation as a single agent, whereas the response rate of greater than 60% in melanoma provided a clear signal that further single-agent development was warranted. Additionally, other cancer types in which BRAF mutations are less commonly found were investigated systematically in a dedicated phase II trial in which patients with any type of solid tumor could be enrolled.25 As in the vemurafenib example, a new oncogene/signal transduction targeted therapy is most commonly evaluated in the most-prevalent tumor-defined populations and/or the ones with the greatest unmet clinical need. With a path forward clearly established in melanoma, a parallel phase II investigation in other cancers harboring V600 BRAF mutations could be conducted while the confirmatory phase II trial in melanoma was performed for the purposes of regulatory approval. This approach effectively created a staggered drug development strategy, one that was driven by prevalence as well as observed efficacy. Out of an intention to streamline the process of drug development, pharmaceutical companies now typically embrace dynamic decision making and adaptation of phase I/II trials to enable broad exploration across cancer types and acceleration within cancer types in the setting of variable efficacy.26 Of course, if heterogeneity is not seen, then the possibility exists of maintaining a tumor type–agnostic approach to development through phase II. This topic will be discussed in more detail later in the article. One additional point in regards to drug development for oncogene/signal transduction targeted therapies is the difficulty in finding sparsely distributed, genetically defined populations. In the case of a genetic alteration that is present in 1% of a certain cancer type, it is both inefficient and costly to perform tests seeking alterations in such a single gene. Next-generation sequencing platforms for routine clinical use can solve this problem by simultaneously sequencing many cancer genes at one time, but have only recently been introduced and are used systematically in only a small number of major academic medical centers. Diagnostic companies have developed these platforms for centralized testing so that individual pathology laboratories need not develop their own capacity. However, in both scenarios, medical insurance payers are reluctant to cover the cost of these tests. Absent more widespread availability of next-generation sequencing tests, it becomes prohibitive to screen 100 patients to find one. The ongoing NCI-MATCH trial performs a sequencing analysis that identifies alterations in more than 150 genes for the purpose of empowering dozens of parallel phase II trials. But, with a total sample of only 6,000 patients, this trial simply demonstrates the efficiency of such an approach. As originally conceived, NCI-MATCH placed priority on determining the genetic make-up of tumors at the time of study entry, rather than relying on archival tumor material

TISSUE-AGNOSTIC DRUG DEVELOPMENT

that might have been obtained at the time of primary tumor resection and potentially confounded by selective pressure of intervening therapy. To execute this, fresh biopsies were required at the time of study entry. This design decision reflects optimization of diagnostic accuracy, while introducing cost, risk associated with invasive procedures for research purposes, and a time delay while next generation sequencing is performed and analyzed. While this approach would have little impact on the representation of truncal mutations in a given tumor sample, it ensured that subclonal genetic alterations that may arise through tumor evolution and therapeutic resistance would be captured. During conduct of the trial, a modified approach was incorporated to allow submission of biopsies procured within the preceding 6 months prior to study entry provided that an intervening response to therapy had not occurred.

APPROACH TO THE DEVELOPMENT OF DRUGS FOR PATIENTS WITH MSI-H TUMORS

The promise of developing mismatch repair (MMR) deficiency as the first predictive biomarker across multiple tumor types for response to a novel therapeutic is supported by strong biologic rationale, availability of commercially used diagnostics for patient identification, and the urgent, unmet medical needs of patients with refractory cancers. The accumulation of evidence that PD-1 inhibition can provide durable benefit in patients with MMR deficiency, coupled with the explosion of technologies to identify these patients, leaves traditional approval pathways that require substantial evidence of effectiveness for each tumor type inadequate to help those in desperate need of therapy today. MMR deficiency refers to the deficiency in proteins responsible in DNA repair when a mismatch occurs in the replication process. Specifically, these proteins are MLH1, PMS2, MSH2, and MSH6. Tumors deficient in these proteins accumulate many mutations because they lack the capacity to repair these mistakes. A tumor may acquire these deficiencies as part of an inherited disorder in one of these genes known as Lynch syndrome, a double somatic mutation in the tumor, or by hypermethylation of the MLH1 gene.27-29 Immunohistochemistry testing for the absence of the MMR proteins can be used to identify these tumors.30 Regardless of the mechanism for deficiency, this leads to the phenomenon of microsatellite instability. Microsatellites are repetitive DNA sequences that are prone to accumulation of mutations when tumors are MMR deficient. Microsatellite instability-high (MSI-H) refers to a tumor that, by polymerase chain reaction–based testing, has been shown to have shifts in more than 30% of specific microsatellite loci. As the shifts are compared with normal DNA, polymerase chain reaction–based testing does require normal tissue. The recognition that MMR-deficient tumors were potentially immunogenic predated the current era of immune checkpoint blockade therapeutics. One of the pathologic characteristics of MSI-H CRC is the presence of immune infiltrating cells.31-34 In particular, the presence of cytotoxic T cells in the tumor microenvironment suggests recognition

of tumor antigen by these cells. The presence and progression of the tumor despite immune cell infiltration, however, support immune evasion by the cancer. The mechanism of immune recognition is not completely understood; however, these tumors are considered “hypermutated” due to accumulation of 10- to 100-fold the number of mutations as their microsatellite-stable counterparts.35,36 These mutations can lead to the presentation of neoantigens to the immune system, making the prospect of immune recognition more likely.

Biologic Commonalities Among MMR-Deficient Tumors

The frequency of MSI varies across tumor types and stages within a tumor type but can be found in diverse histologies, ranging from those with higher frequencies such as colon, gastric, and endometrial cancers to those less commonly associated with MSI such as prostate cancer.35,37 However, among those tumor types that have been studied histologically and by sequencing, the characteristics that may be predictive of response to immunotherapy are shared among MSI tumors: T-cell infiltration, PD-L1 expression, and high mutation burden. These features are found in both Lynch-associated and sporadic tumors.

CRC

MSI is present in 15% of CRC, with a majority of cases sporadic in etiology and 3% due to Lynch syndrome.29 In advanced-stage disease, the frequency is 3% to 5%. Testing is already part of the National Comprehensive Cancer Network guidelines for CRC (version 1.2017).38 Testing is recommended for the identification of patients with possible Lynch syndrome, a negative predictive marker for adjuvant therapy for stage II colon cancer, and now as a positive predictive biomarker for PD-1 inhibition. Better prognosis in early-stage disease is possibly due to immune recognition of these tumors. However, this is not true in metastatic disease, and MSI may portend a worse prognosis.39,40 Pathologists can often identify these tumors without molecular diagnostics, as they tend to originate from right-sided primaries and have medullary differentiation, signet cell features, mucin production, poor differentiation status, and dense T-cell infiltration.31,32,34 PD-L1 expression is also more frequent in MSI cancers than microsatellite-stable cancers.41 In the Cancer Genome Atlas analysis of CRCs, 77% of the 30 hypermutated tumors with a complete data set were MSI-H.42 Nineteen of these were due to MLH1 hypermethylation.

Endometrial Cancer

Universal testing of endometrial tumors for MMR deficiency is now recommended for the identification of patients with Lynch syndrome (National Comprehensive Cancer Network guidelines, version 1.2017). Data correlating MSI status to patient outcomes has been mixed. In a Japanese study, 40% of 191 cases of surgically resected endometrial cancer were found to be MMR deficient by IHC and this correlated with asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 225

FLAHERTY, LE, AND LEMERY

better OS.43 However, in a study of patients younger than age 40, MMR deficiency was associated with worse outcomes.44 Similar to CRC, MSI-H endometrial cancers have higher CD3+ and CD8+ T-cell infiltrates than their microsatellite-stable counterparts.45,46 Furthermore, PD-L1 expression is also mostly seen on immune-infiltrating cells.45 Integrated genomic analysis of 373 endometrial cancers including MSI testing found MSI in 40% of endometrioid tumors and 2% of serous tumors.47 MSI tumors fell into the hypermutated group, with most of them falling into the MLH1 hypermethylated group.

Gastric Cancer

Although MSI testing is not routinely performed in gastric cancer, some studies report an MSI-H frequency as high as 30%.37 In a post hoc analysis of a large trial of perioperative chemotherapy versus surgery alone, MMR deficiency was a positive prognostic factor in those undergoing surgery alone but negative in those treated with chemotherapy.48 This is reminiscent of the adjuvant story in colon cancer. Pathologically similar to other MSI-H cancers, they are associated with tumor-infiltrating lymphocytes and mucin phenotypes.49,50 These tumors are predominantly intestinal in type. As part of the Cancer Genome Atlas analysis, 295 gastric cancers were analyzed. Four subtypes of gastric cancer were proposed: Epstein-Barr virus (EBV)–associated tumors that are associated with PD-L1 and PD-L2 amplification; microsatellite-unstable tumors; genomically stable tumors; and tumors with chromosomal instability.51 The MSI-H cancers were again characterized to have high mutation rates. Likewise, PD-L1 expression on tumor and tumor immune infiltrates are more commonly associated with EBV+ or MSI-H gastric cancers.50,52,53

Other MSI-H Cancers

Microsatellite instability can be found across multiple tumor types at varying frequency.35,37 These variations may be due to the baseline characteristics such as geographic region, stage of disease, and family history. Furthermore, the assay for detection of MMR or MSI varied as well. MSI can be detected in 2% of pancreatic adenocarcinomas, 10% of ampullary cancers, and 10% of ovarian cancers. The availability of data regarding the prognostic implications of MSI status varies among different histologies; however, once metastatic, the cancers are uniformly fatal. It is hypothesized that regardless of histology, the common features of immune cell infiltration, PD-L1 expression, and high mutation frequency will make MSI an important predictive marker for susceptibility to immune checkpoint blockade. Furthermore, routinely used assays are already being used to identify these patients, and MSI/MMR status is now also being reported on molecular profiling and next-generation sequencing panels.

Hints of PD-1 Antibody Activity in MSI-H Cancers

Intriguingly, PD-1 inhibition has shown some level of activity in these specific histologies known to have a subset of can226 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

cer with MSI. In CRC, single-agent PD-1 inhibition has only been active in MSI-H tumors. The single response in the first nivolumab study and the response in the PD-L1–positive selected study of pembrolizumab in CRC were found to be MSI-H.54,55 PD-L1 was not a good predictive biomarker of response in the latter study. The overall response rate (ORR) was 13% in PD-L1–selected endometrial cancer.56 MSI status has not been reported on these patients. However, in a study of pembrolizumab in PD-L1–positive gastric cancer, a 22% ORR was observed, with all eight responders achieving partial responses. Twenty-four tumors were tested for MSI and 17% (four tumors) were determined to be MSI-H. Of these, two MSI-H tumors were noted to have responded to treatment, accounting for at least 25% of the responses.57,58 In an analysis of 319 esophagogastric cancers by investigators at Memorial Sloan Kettering Cancer Center, they identified 12 patients with MSI-H tumors, of which three were treated with anti–PD-1 with best response of complete response, partial response, and stable disease. One patient with an EBV+ cancer had a complete response to combination PD-1 and cytotoxic T lymphocyte antigen-4 inhibition.59

MMR Deficiency as a Predictive Biomarker for PD-1 Inhibition

In a prospective study, MMR deficiency was explored as a predictive biomarker of response to pembrolizumab.60 Patients with CRC were identified as MMR deficient based on immunohistochemistry testing for the MMR proteins or by polymerase chain reaction–based testing for MSI. An additional cohort of MSI-H non-CRC patients was also treated. The original article reported on 11, 21, and nine patients with MMR-deficient CRC, MMR-proficient CRC, and MMR-deficient non-CRC with ORR of 40%, 0%, and 71%, respectively. Importantly, in the MMR-deficient cohorts of patients who were all previously treated with standard therapies, the median duration of response and overall survival were not reached. Median follow-ups were 36 and 21 weeks for the MMR-deficient CRC and non-CRC cohorts, respectively. As expected, whole-exome sequencing resulted in a mean of 1,782 somatic mutations per tumor in MMR–deficient tumors and 73 in MMR-proficient tumors. Updated data reported at the 2016 ASCO Annual Meeting 61,62 showed the ORR was 57% with 11% complete response rate in the MMR-deficient CRC cohort and 53% with a 30% complete response rate in the MMR-deficient non-CRC cohort. The disease control rates were 89% and 73%, respectively. Responses were seen in CRC and endometrial, gastric, pancreaticobiliary, small bowel, and prostate cancers. At the same meeting, preliminary data with nivolumab was reported that showed an ORR of 26% with an additional 30% stable disease in MSI-H CRC.63 Based on a clinically significant response rate with longerthan-expected response duration in traditionally incurable advanced cancers, physicians and patients are optimistic that the evidence is accumulating that will allow access to these agents for patients outside of clinical trials and access programs.

TISSUE-AGNOSTIC DRUG DEVELOPMENT

REGULATORY CONSIDERATIONS REGARDING TISSUE-AGNOSTIC DEVELOPMENT

General Considerations

Federal regulations governing drug development do not require disease to be defined based on a single tumor type. Food-labeling regulations contain the following definition of disease: “… damage to an organ, part, structure, or system of the body such that it does not function properly (e.g., cardiovascular disease), or a state of health leading to such dysfunctioning (e.g., hypertension); except that diseases resulting from essential nutrient deficiencies (e.g., scurvy, pellagra) are not included in this definition.”64 Nevertheless, prior to determining whether a drug can be developed based on a molecular pathway, investigators or drug developers should determine whether the approach is scientifically and clinically appropriate. In the BRAF example cited above, tissue-agnostic development may not have been appropriate. Importantly, the mechanistic explanation for lack of response in CRC has been elucidated, and this has led to clinical trials that are investigating the effects of BRAF inhibitors in combination with other inhibitors of the MEK or EGFR pathways.6,14,65 Ultimately, determining whether a sponsor should develop a drug irrespective of histology will depend on several preclinical and clinical factors, including data supporting the scientific rationale (e.g., as discussed above) and the context of treatment of patients with different tumor types. For example, if a drug is most effective in combination regimens, tissue-agnostic development may not be appropriate given the differences in standard therapies administered to patients across tumor types. It may be more appropriate to consider tissue-agnostic development in situations where a drug-target combination appears to demonstrate very high activity (e.g., breakthrough-like) across multiple tumor types where the clinical effect can easily be demonstrated. Other developmental considerations may include differences in natural history across diverse cancers and how investigators or sponsors will propose to generate data in a sufficient number of patients with various tumor types.

use of a corresponding therapeutic product.67 An analytically and clinically validated device reduces the risk of withholding appropriate therapy from a patient who is mutation/ biomarker positive who receives a false negative test result or administering inappropriate therapy in the case of a false positive result.67 From a practical perspective, for rare mutation-tumor combinations, it may be preferable to develop an IVD test as part of a larger panel of tests (e.g., as part of a next-generation sequencing panel). Factors unique to IVD development for tissue-agnostic use may include differences in amount of tumor collected at biopsy among different tumor types, differences in tumor heterogeneity among tumor types, and differences in stromal tissue surrounding tumors. The FDA recently published a perspective regarding both the potential benefits and challenges of complex IVD signatures.68 Sponsors are encouraged to meet with the FDA early to facilitate companion diagnostic development. FDA can provide advice to sponsors to determine whether an investigational device exemption is necessary to use an IVD in the trial and to provide guidance in regards to what data would be necessary to approve a companion IVD if the device is necessary for the safe use of the drug.

Residual Uncertainty

An indication that is truly tissue agnostic would allow for the treatment of both adults and children. Such a tissue-agnostic approach, if appropriate, could therefore benefit children by bringing drugs to treat children with cancer more expeditiously. Sponsors who are assessing the effects of a drug across various tumor sites based on a biomarker should consider how they will address the needs of children who have a tumor that possesses that biomarker. For example, the FDA has recommended the inclusion of adolescents (age 12–17) in disease- and target-appropriate adult oncology trials.66 Sponsors should determine whether additional formulations of a drug would be necessary to address the needs of younger children so that younger children can enroll in clinical trials and potentially benefit from that therapy.

Uncertainty may arise in a development program about a drug’s effectiveness in all tumor types with a specific fusion, mutation, or biomarker, particularly because some tumor-biomarker combinations may be exceedingly rare. The rarity of certain tumor-biomarker combinations may also make it impossible to conduct randomized trials, especially in settings where equipoise would not exist based on unprecedented antitumor activity observed in single-arm trials. For example, the FDA granted regular approval of crizotinib for ROS-1–positive non–small cell lung cancer based on the benefit of an objective response rate of 66% (95% CI, 51–79) by independent review and a median duration of response of 18.3 months.69 Depending on the strength of evidence across tumor types, multiple regulatory mechanisms exist to address this residual uncertainty. These range from requiring additional data in the premarket setting to requiring postmarketing data in the setting of accelerated approval. If data are adequately collected, “real-world” evidence also may provide supportive data regarding tumor response or lack thereof across rare tumor types.70 Postmarketing requirements for drugs granted accelerated approval would not necessarily need to include randomized trials. Alternatively, if there is a histology-biomarker combination that is more common (e.g., ALK in lung cancer), a randomized trial (if necessary) in that dominant tumor type could be conducted to provide supportive evidence of safety and clinical benefit of a high (durable) response rate observed in other tumor types.

Companion Diagnostic

CONCLUSION

Pediatric Development

A companion or complementary in vitro diagnostic (IVD) device provides essential information for the safe and effective

Ultimately, the goal of drug development is to bring effective drugs that benefit patients as quickly as possible. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 227

FLAHERTY, LE, AND LEMERY

Investigating the effects of a drug agnostic of tumor type may be one pathway for drug development; however, every drug presents unique circ*mstances in regard to the population of patients who might benefit from it. Furthermore, development agnostic of tumor type could actually slow drug development if there are differential effects across tumor types by diverting resources from enrolling patients in a predominant population or in the tumor type

most likely to respond. Therefore, input from all stakeholders is recommended prior to embarking on such an approach.

ACKNOWLEDGMENT

The authors thank Kirsten Goldberg for her editorial review and Amy McKee and Richard Pazdur for scientific review of the regulatory considerations section.

References 1. Hoadley KA, Yau C, Wolf DM, et al; Cancer Genome Atlas Research Network. Multiplatform analysis of 12 cancer types reveals molecular classification within and across tissues of origin. Cell. 2014;158:929-944.

15. Müller J, Krijgsman O, Tsoi J, et al. Low MITF/AXL ratio predicts early resistance to multiple targeted drugs in melanoma. Nat Commun. 2014;5:5712.

2. Esteva FJ, Valero V, Booser D, et al. Phase II study of weekly docetaxel and trastuzumab for patients with HER-2-overexpressing metastatic breast cancer. J Clin Oncol. 2002;20:1800-1808.

16. Pao W, Miller VA, Politi KA, et al. Acquired resistance of lung adenocarcinomas to gefitinib or erlotinib is associated with a second mutation in the EGFR kinase domain. PLoS Med. 2005;2:e73.

3. Bang YJ, Van Cutsem E, Feyereislova A, et al; ToGA Trial Investigators. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet. 2010;376:687-697.

17. Ignatius Ou SH, Azada M, Hsiang DJ, et al. Next-generation sequencing reveals a novel NSCLC ALK F1174V mutation and confirms ALK G1202R mutation confers high-level resistance to alectinib (CH5424802/ RO5424802) in ALK-rearranged NSCLC patients who progressed on crizotinib. J Thorac Oncol. 2014;9:549-553.

Flaherty KT, Puzanov I, Kim KB, et al. Inhibition of mutated, activated 4. BRAF in metastatic melanoma. N Engl J Med. 2010;363:809-819.

18. Mayer IA, Abramson VG, Formisano L, et al. A phase Ib study of alpelisib (BYL719), a PI3Kα-specific inhibitor, with letrozole in ER+/ HER2- metastatic breast cancer. Clin Cancer Res. 2017;23:26-34.

5. Kopetz S, Desai J, Chan E, et al. Phase II pilot study of vemurafenib in patients with metastatic BRAF-mutated colorectal cancer. J Clin Oncol. 2015;33:4032-4038. Atreya CE, Van Cutsem E, Bendell JC, et al. Updated efficacy of the 6. MEK inhibitor trametinib, BRAF inhibitor dabrafenib, and antiEGFR antibody panitumumab in patients with BRAF V600E mutated metastatic colorectal cancer. J Clin Oncol. 2015;33 (suppl; abstr 103). 7. Tumeh PC, Harview CL, Yearley JH, et al. PD-1 blockade induces responses by inhibiting adaptive immune resistance. Nature. 2014;515:568-571. 8. Reck M, Rodríguez-Abreu D, Robinson AG, et al; KEYNOTE-024 Investigators. Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung cancer. N Engl J Med. 2016;375:1823-1833. Sequist LV, Martins RG, Spigel D, et al. First-line gefitinib in patients 9. with advanced non-small-cell lung cancer harboring somatic EGFR mutations. J Clin Oncol. 2008;26:2442-2449.

19. Juric D, Castel P, Griffith M, et al. Convergent loss of PTEN leads to clinical resistance to a PI(3)Kα inhibitor. Nature. 2015;518:240-244. 20. Sun C, Wang L, Huang S, et al. Reversible and adaptive resistance to BRAF(V600E) inhibition in melanoma. Nature. 2014;508:118-122. 21. Haq R, Shoag J, Andreu-Perez P, et al. Oncogenic BRAF regulates oxidative metabolism via PGC1α and MITF. Cancer Cell. 2013;23:302315. 22. O’Connell MP, Marchbank K, Webster MR, et al. Hypoxia induces phenotypic plasticity and therapy resistance in melanoma via the tyrosine kinase receptors ROR1 and ROR2. Cancer Discov. 2013;3:1378-1393. 23. Montero J, Sarosiek KA, DeAngelo JD, et al. Drug-induced death signaling strategy rapidly predicts cancer response to chemotherapy. Cell. 2015;160:977-989.

10. Kwak EL, Bang YJ, Camidge DR, et al. Anaplastic lymphoma kinase inhibition in non-small-cell lung cancer. N Engl J Med. 2010;363:1693-1703.

24. Jonas O, Landry HM, Fuller JE, et al. An implantable microdevice to perform high-throughput in vivo drug sensitivity testing in tumors. Sci Transl Med. 2015;7:284ra57.

11. Cooper ZA, Juneja VR, Sage PT, et al. Response to BRAF inhibition in melanoma is enhanced when combined with immune checkpoint blockade. Cancer Immunol Res. 2014;2:643-654.

25. Hyman DM, Puzanov I, Subbiah V, et al. Vemurafenib in multiple nonmelanoma cancers with BRAF V600 mutations. N Engl J Med. 2015;373:726-736.

12. Bollag G, Hirth P, Tsai J, et al. Clinical efficacy of a RAF inhibitor needs broad target blockade in BRAF-mutant melanoma. Nature. 2010;467:596-599.

26. Theoret MR, Pai-Scherf LH, Chuk MK, et al. Expansion cohorts in firstin-human solid tumor oncology trials. Clin Cancer Res. 2015;21:45454551.

13. Corcoran RB, Ebi H, Turke AB, et al. EGFR-mediated re-activation of MAPK signaling contributes to insensitivity of BRAF mutant colorectal cancers to RAF inhibition with vemurafenib. Cancer Discov. 2012;2:227-235.

27. Boland CR, Lynch HT. The history of Lynch syndrome. Fam Cancer. 2013;12:145-157.

14. Prahallad A, Sun C, Huang S, et al. Unresponsiveness of colon cancer to BRAF(V600E) inhibition through feedback activation of EGFR. Nature. 2012;483:100-103.

228 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

28. Lynch HT, Lynch PM, Lanspa SJ, et al. Review of the Lynch syndrome: history, molecular genetics, screening, differential diagnosis, and medicolegal ramifications. Clin Genet. 2009;76:1-18. 29. Boland CR, Goel A. Microsatellite instability in colorectal cancer. Gastroenterology. 2010;138:2073-2087.e3.

TISSUE-AGNOSTIC DRUG DEVELOPMENT

30. Shia J. Immunohistochemistry versus microsatellite instability testing for screening colorectal cancer patients at risk for hereditary nonpolyposis colorectal cancer syndrome. Part I. The utility of immunohistochemistry. J Mol Diagn. 2008;10:293-300.

48. Smyth EC, Wotherspoon A, Peckitt C, et al. Mismatch repair deficiency, microsatellite instability, and survival: an exploratory analysis of the medical research council adjuvant gastric infusional chemotherapy (MAGIC) trial. JAMA Oncol. Epub 2017 Feb 23.

31. Shia J, Ellis NA, Paty PB, et al. Value of histopathology in predicting microsatellite instability in hereditary nonpolyposis colorectal cancer and sporadic colorectal cancer. Am J Surg Pathol. 2003;27:1407-1417.

49. Kim JY, Shin NR, Kim A, et al. Microsatellite instability status in gastric cancer: a reappraisal of its clinical significance and relationship with mucin phenotypes. Korean J Pathol. 2013;47:28-35.

32. Alexander J, Watanabe T, Wu TT, et al. Histopathological identification of colon cancer with microsatellite instability. Am J Pathol. 2001;158:527-535.

50. Kawazoe A, Kuwata T, Kuboki Y, et al. Clinicopathological features of programmed death ligand 1 expression with tumor-infiltrating lymphocyte, mismatch repair, and Epstein-Barr virus status in a large cohort of gastric cancer patients. Gastric Cancer. Epub 2016 Sep 14.

33. Dolcetti R, Viel A, Doglioni C, et al. High prevalence of activated intraepithelial cytotoxic T lymphocytes and increased neoplastic cell apoptosis in colorectal carcinomas with microsatellite instability. Am J Pathol. 1999;154:1805-1813. 34. Kim H, Jen J, Vogelstein B, et al. Clinical and pathological characteristics of sporadic colorectal carcinomas with DNA replication errors in microsatellite sequences. Am J Pathol. 1994;145:148-156. 35. Dudley JC, Lin MT, Le DT, et al. Microsatellite instability as a biomarker for PD-1 blockade. Clin Cancer Res. 2016;22:813-820. Lin EI, Tseng LH, Gocke CD, et al. Mutational profiling of colorectal cancers 36. with microsatellite instability. Oncotarget. 2015;6:42334-42344. 37. Lee V, Murphy A, Le DT, et al. Mismatch repair deficiency and response to immune checkpoint blockade. Oncologist. 2016;21:1200-1211. National Comprehensive Cancer Network. National Comprehensive 38. Cancer Network Guidelines. Version 1.2017. https://www.nccn.org/ professionals/physician_gls/pdf/colon.pdf. Accessed February 8, 2017. Goldstein J, Tran B, Ensor J, et al. Multicenter retrospective analysis 39. of metastatic colorectal cancer (CRC) with high-level microsatellite instability (MSI-H). Ann Oncol. 2014;25:1032-1038. 40. Venderbosch S, Nagtegaal ID, Maughan TS, et al. Mismatch repair status and BRAF mutation status in metastatic colorectal cancer patients: a pooled analysis of the CAIRO, CAIRO2, COIN, and FOCUS studies. Clin Cancer Res. 2014;20:5322-5330. 41. Llosa NJ, Cruise M, Tam A, et al. The vigorous immune microenvironment of microsatellite instable colon cancer is balanced by multiple counterinhibitory checkpoints. Cancer Discov. 2015;5:43-51. 42. Cancer Genome Atlas Network. Comprehensive molecular characterization of human colon and rectal cancer. Nature. 2012;487:330-337.

51. Cancer Genome Atlas Research Network. Comprehensive molecular characterization of gastric adenocarcinoma. Nature. 2014;513:202-209. 52. Derks S, Liao X, Chiaravalli AM, et al. Abundant PD-L1 expression in Epstein-Barr Virus-infected gastric cancers. Oncotarget. 2016;7:3292532932. 53. Ma C, Patel K, Singhi AD, et al. Programmed Death-Ligand 1 expression is common in gastric cancer associated with Epstein-Barr Virus or microsatellite instability. Am J Surg Pathol. 2016;40:1496-1506. 54. Lipson EJ, Sharfman WH, Drake CG, et al. Durable cancer regression off-treatment and effective reinduction therapy with an anti-PD-1 antibody. Clin Cancer Res. 2013;19:462-468. 55. Sehdev A, Cramer HM, Ibrahim AA, et al. Pathological complete response with anti-PD-1 therapy in a patient with microsatellite instable high, BRAF mutant metastatic colon cancer: a case report and review of literature. Discov Med. 2016;21:341-347. 56. Ott PA, Bang YJ, Berton-Rigaud D, et al. Pembrolizumab in advanced endometrial cancer: preliminary results from the phase Ib KEYNOTE-028 study. J Clin Oncol. 2016;34 (suppl; abstr 5581). 57. Muro K, Chung HC, Shankaran V, et al. Pembrolizumab for patients with PD-L1-positive advanced gastric cancer (KEYNOTE-012): a multicentre, open-label, phase 1b trial. Lancet Oncol. 2016;17:717-726. 58. Ilson DH. “Anti–PD-1 Treatment With Pembrolizumab in Gastric/ Gastroesophageal Junction Cancers: Who Is Likely to Respond?” The ASCO Post, June 25, 2016. www.ascopost.com/issues/ june-25-2016/anti-pd-1-treatment-with-pembrolizumab-ingastricgastroesophageal-junction-cancers-who-is-likely-to-respond/. 59. Janjigian YY, Sanchez-Vega F, Jonsson P, et al. Clinical next generation sequencing (NGS) of esophagogastic (EG) adenocarcinomas identifies distinct molecular signatures of response. Ann Oncol. 2016;27:207242.

43. Kato M, Takano M, Miyamoto M, et al. DNA mismatch repair-related protein loss as a prognostic factor in endometrial cancers. J Gynecol Oncol. 2015;26:40-45.

60. Le DT, Uram JN, Wang H, et al. PD-1 blockade in tumors with mismatchrepair deficiency. N Engl J Med. 2015;372:2509-2520.

44. Shih KK, Garg K, Levine DA, et al. Clinicopathologic significance of DNA mismatch repair protein defects and endometrial cancer in women 40 years of age and younger. 2011;123:88-94.

61. Le DT, Uram JN, Wang H, et al. Programmed death-1 blockade in mismatch repair deficient colorectal cancer. J Clin Oncol. 2016; (suppl; abstr 103).

45. Howitt BE, Shukla SA, Sholl LM, et al. Association of polymerase e-mutated and microsatellite-instable endometrial cancers with neoantigen load, number of tumor-infiltrating lymphocytes, and expression of PD-1 and PD-L1. JAMA Oncol. 2015;1:1319-1323.

62. Diaz LA, Uram JN, Wang H, et al. Programmed death-1 blockade in mismatch repair deficient cancer independent of tumor histology. J Clin Oncol. 2016; (suppl; abstr 3003).

46. Kanopienė D, Smailytė G, Vidugirienė J, et al. Impact of microsatellite instability on survival of endometrial cancer patients. Medicina (Kaunas). 2014;50:216-221.

63. Overman MJ, Kopetz S, McDermott RS, et al. Nivolumab ± ipilimumab in treatment (tx) of patients (pts) with metastatic colorectal cancer (mCRC) with and without high microsatellite instability (MSI-H): CheckMate-142 interim results. J Clin Oncol. 2016;34 (suppl; abstr 3501).

47. Kandoth C, Schultz N, Cherniack AD, et al; Cancer Genome Atlas Research Network. Integrated genomic characterization of endometrial carcinoma. Nature. 2013;497:67-73.

U.S. Food and Drug Administration. Code of Federal Regulations 64. Title 21. https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/ CFRSearch.cfm?fr=101.93. Accessed February 8, 2017.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 229

FLAHERTY, LE, AND LEMERY

65. Tabernero J, van Geel R, Guren T, et al. Combination of encorafenib and cetuximab with or without alpelisib in patients with advanced BRAF-mutant colorectal cancer (BRAFm CRC): phase 2 results. Ann Oncol. 2016;27:127-128. 66. Chuk MK, Mulugeta Y, Roth-Cline M, et al. Enrolling adolescents in disease/target-appropriate adult oncology clinical trials of investigational agents. Clin Cancer Res. 2017;23:9-12. 67. U.S. Food and Drug Administration. In Vitro Companion Diagnostic Devices Guidance for Industry and Food and Drug Administration Staff. https://www.fda.gov/downloads/MedicalDevices/

230 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

DeviceRegulationandGuidance/GuidanceDocuments/UCM262327. pdf. Accessed February 8, 2017. 68. Beaver JA, Tzou A, Blumenthal GM, et al. An FDA perspective on the regulatory implications of complex signatures to predict response to targeted therapies. Clin Cancer Res. Epub 2016 Dec 19. 69. Kazandjian D, Blumenthal GM, Luo L, et al. Benefit-risk summary of crizotinib for the treatment of patients with ROS1 alteration-positive, metastatic non-small cell lung cancer. Oncologist. 2016;21:974-980. 70. Sherman RE, Anderson SA, Dal Pan GJ, et al. Real-world evidence: what is it and what can it tell us? N Engl J Med. 2016;375:2293-2297.

GASTROINTESTINAL (COLORECTAL) CANCER

MCCLEARY, BENSON, AND DIENSTMANN

Personalizing Adjuvant Therapy for Stage II/III Colorectal Cancer Nadine Jackson McCleary, MD, MPH, Al B. Benson III, MD, FACP, FASCO, and Rodrigo Dienstmann, MD OVERVIEW This review focuses on three areas of interest with respect to the treatment of stage II and III colon and rectal cancer, including (1) tailoring adjuvant therapy for the geriatric population, (2) the controversy as to the optimal adjuvant therapy strategy for patients with locoregional rectal cancer and for patients with colorectal resectable metastatic disease, and (3) discussion of the microenvironment, molecular profiling, and the future of adjuvant therapy. It has become evident that age is the strongest predictive factor for receipt of adjuvant chemotherapy, duration of treatment, and risk of treatment-related toxicity. Although incorporating adjuvant chemotherapy for patients who have received neoadjuvant chemoradiation and surgery would appear to be a reasonable strategy to improve survivorship as an extrapolation from stage III colon cancer adjuvant trials, attempts at defining the optimal rectal cancer population that would benefit from adjuvant therapy remain elusive. Similarly, the role of adjuvant chemotherapy for patients after resection of metastatic colorectal cancer has not been clearly defined because of very limited data to provide guidance. An understanding of the biologic hallmarks and drivers of metastatic spread as well as the micrometastatic environment is expected to translate into therapeutic strategies tailored to select patients. The identification of actionable targets in mesenchymal tumors is of major interest.

lthough there is no uniform number at which physiologic aging occurs, there is little known about optimal treatment of colon cancer involving lymph nodes following surgical resection for adults age 75 or older.1,2 A substantial number of patients with colorectal cancer (CRC; 40%) are adults age 75 or older.3 Standards for adjuvant chemotherapy following resection of colon cancer were established based on results of three large randomized clinical trials: MOSAIC (Multicenter International Study of Oxaliplatin/5-Fluorouracil/Leucovorin in the Adjuvant Treatment of Colon Cancer), NSABP C-07 (National Surgical Adjuvant Breast and Bowel Project), and XELOXA NO16968 (XELOX in Adjuvant Colon Cancer Treatment). Yet with less than 1% (MOSAIC) and 5% (NSABP C-07), respectively, of those trials including older adults, it proves difficult to extrapolate standards of adjuvant chemotherapy to older adults in the real-world setting (the proportion age 75 or older is not reported in XELOXA NO16968). Several pooled analyses show potential for survival benefit among some older adults; however, nearly two-thirds do not receive adjuvant treatment.3,4 Nonreceipt of systemic chemotherapy is particularly prevalent among those older adults diagnosed with colon cancer who also have geriatric syndromes (e.g., delirium, frailty) or active comorbid medical conditions.5 Expertise in delivering care to this growing subset of patients is predom-

inantly driven by provider experience and possible bias, given the limited clinical trial data available to guide use of adjuvant chemotherapy in the older adult population. Here, we review the available data and recommendations for adjuvant treatment recommendations for adults age 75 or older diagnosed with stage III colon cancer.

TAILORING ADJUVANT THERAPY FOR THE GERIATRIC POPULATION

Adjuvant Chemotherapy for Stage III Colon Cancer

General population. The benefit of adjuvant chemotherapy has been clearly established in the adjuvant setting for node-positive colon cancer. Standard treatment options include fluorouracil (FU) or capecitabine with or without oxaliplatin (Table 1). The addition of FU to surgical resection led to 17% improvement in disease-free survival and 13% improvement in overall survival among patients with node-positive colon cancer.6 The addition of capecitabine led to similar improvements in disease-free (hazard ratio [HR], 0.87; 95% CI, 0.75–1.00) and overall survival (HR, 0.84; 95% CI, 0.69–1.01) compared with bolus FU/leucovorin (p for equivalence < .001, with median follow-up of 3.8 years).7 The addition of oxaliplatin to FU further leads to an absolute improvement in disease-free and overall survival at 10 years by an additional 8%.8 The addition of oxaliplatin

From the Dana-Farber Cancer Institute, Boston, MA; Division of Hematology/Oncology, Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, IL; Oncology Data Science Group, Vall d’Hebron Institute of Oncology, Barcelona, Spain; Sage Bionetworks, Fred Hutchinson Cancer Research Center, Seattle, WA. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Al B. Benson III, MD, FACP, FASCO, Division of Hematology/Oncology, Robert H. Lurie Comprehensive Cancer Center, Northwestern University, 676 N. St Clair St., Suite 850, Chicago, IL 60611; email: [emailprotected]. © 2017 American Society of Clinical Oncology

232 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PERSONALIZING ADJUVANT THERAPY FOR STAGE II/III COLORECTAL CANCER

to capecitabine results in a similar improvement, with reductions of 20% and 17% in the relative risk of recurrence or death (95% CI, 0.69–0.93; p = .004) and risk of death (95% CI, 0.70–0.99; p = .04), respectively.9 The extent to which older adults derive benefit from adjuvant chemotherapy was not established in these trials, given that most trials limited participation to those younger than age 757,10 or limited the number of adults age 75 or older.3,4 Older adults. A number of pooled and subpopulation analyses have been conducted to fill the gap in knowledge regarding survival benefit of older adults receiving adjuvant chemotherapy for stage III colon cancer (refer to Table 123,24). In a pooled analysis of seven randomized clinical trials of adjuvant chemotherapy included in the ACCENT (Adjuvant Colon Cancer End Points) study, older adults did not experience substantial benefit from adjuvant fluoropyrimidine or combination chemotherapy regimens regarding disease-free survival (HR, 1.05; 95% CI, 0.94–1.19), overall survival (HR, 1.08; 95% CI, 0.95–1.23), or time to recurrence survival (HR, 1.06; 95% CI, 0.93–1.22).22 Older adults seemed to have a reduced overall survival benefit from oxaliplatin-based chemotherapy with a similar disease-free survival benefit compared with younger adults receiving oxaliplatin-based chemotherapy. There was no difference in rates of death within experimental or control arms, suggesting that it is unlikely that the substantial interaction noted between treatment and age would be explained by early deaths attributable to treatment-related toxicity.22 In contrast, comorbidity and age did not appear to affect disease-free or overall survival among older adults enrolled in four randomized clinical trials evaluating adjuvant fluoropyrimidine with or without oxaliplatin including comorbidity

KEY POINTS • Age is the strongest predictive factor for receipt of adjuvant chemotherapy, duration of treatment, and risk of treatment-related toxicity. • Available data support disease-free and overall survival benefit after adjuvant therapy among older adults age 70–74 years with colon cancer, but variable outcomes for those age 75 years or older. • Attempts at defining the optimal rectal cancer population that would benefit from adjuvant therapy remain elusive. • In stage II disease, microsatellite instability and/or high “immunoscores” associate with very good prognosis and support a no-adjuvant-treatment approach. On the other hand, empirical evidence for the addition of supervised gene expression classifiers to the clinical decision-making paradigm is scarce. • Irrespective of tumor stage, activation of a gene expression signature of epithelial-mesenchymal transition correlates with an invasive-inflamed microenvironment infiltrated with stromal and immunosuppressive cells, which confers poor prognosis and limited benefit with standard adjuvant chemotherapies.

data defined by the Charlson Comorbidity Index or the National Cancer Institute Combined Index.15 The ACCORE study evaluated 191 patients age 70 or older and 338 patients younger than 70 receiving adjuvant 5-fluorouracil (5-FU) or capecitabine with or without oxaliplatin for CRC in Denmark from 2001 to 2012.25 Older adults experienced similar 10-year CRC-specific overall survival compared with younger patients but did experience higher rates of mortality owing to other causes after controlling for performance status and presence of comorbid medical conditions. Older adults received equivalent doses of capecitabine but fewer doses of oxaliplatin and 5-FU compared with younger patients. Disease-free survival and CRC-specific mortality were not affected by reductions in chemotherapy dose intensity. This and other pooled analyses from clinical trials are limited by the relatively small number of older adults enrolled. Despite this, rates of use of oxaliplatin found in the Surveillance, Epidemiology and End Results (SEER) database increased rapidly in adults age 65 or older diagnosed with stage III colon cancer (from 52% in 2004 to 73% in 2007, albeit at reduced rates among individuals older than 85 and those with comorbid medical conditions).26 In the general population of 5,489 adults 75 or older, 2,395 (44%) received chemotherapy within 120 days of surgery and 3,096 (56%) did not.3 Rates of chemotherapy administration were higher in academic centers (75% in a National Comprehensive Cancer Network assessment [61% oxaliplatin]) versus nonacademic and community sites (42% in a SEER-Medicare analysis [42% oxaliplatin], 45% in a New York State Cancer Registry-Medicare study [28% oxaliplatin], and 52% in the CanCORS study). Oxaliplatin receipt decreased with increasing age, from 46% of adults age 75–79 to 7% among those age 85 or older. However, the benefit of adjuvant chemotherapy in this heterogeneous older population was comparable to that observed in pooled analyses of selected fit older adults participating in clinical trials, suggesting retention of survival benefit among subsets of older adults amenable to and receiving adjuvant chemotherapy within 120 days of surgical resection. Older adults appear to receive a benefit from adjuvant chemotherapy in some, but not all, studies. Survival seems to differ across age categories, with decreasing survival benefit with increasing age. A review of the National Cancer Institute SEER database linked to the Medicare database (SEER-Medicare) noted a predicted increased 5-year survival benefit of 14% among patients age 70–74 compared with 8% among those age 80–84.17 Survival benefit persists in older adults age 80–89, despite only 43% of the 8,141 octogenarians included in the National Cancer Database from 2006 to 2011.4 Regardless of potential benefit for some older adults, older age remains the strongest determinant of initiation, duration, and completion of adjuvant chemotherapy.3,27-29 Older adults are also more likely to have delays in initiation of adjuvant chemotherapy and are less likely to complete the full 6 months of adjuvant therapy, factors that also increase mortality risk.25,30,31 asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 233

MCCLEARY, BENSON, AND DIENSTMANN

TABLE 1. Survival Outcomes for Adjuvant Chemotherapy Among Older Adults

Treatment Arms

Older Adult Cohort, Age (No. of Patients)

NSABP C-06 phase III (200611)

UFT/LV vs. 5-FU/LV

≥ 60 (939)

National Cancer Database (200512), retrospective study

Adjuvant chemotherapy

All (1990– 1991: 12,413; 2001– 2002: 14,187)

Receipt of chemotherapy (%):

N/A

70–79 (1990– 1991: 4,103; 2001– 2002: 4,086)

N/A

≥ 80 (1990– 1991): 2,593; 2001– 2002: 3,305)

N/A

Trial (Year), Study Design

Regimen-Specific Data

RelapseFree Survival

Time to Recurrence

1.41 (1.18– 1.69); p = .002 (referent = age < 60)

N/A

1.40 (1.12–1.74); p = .03 (referent = age < 60)

Similar OS regardless of age

Disease-Free Survival

Overall Survival

NO16968 phase III (20159)

XELOX vs. bolus 5-FU/LV

≥ 70 (409)

0.86 (0.64– 1.16); p = NR

N/A

0.91 (0.66–1.26); p = NR

7 randomized clinical trials (20046), pooled analysis

5-FU/LV vs. surgery alone

≥ 60 (1,864)

63% vs. 55%; p = .001 at 5 years

N/A

69% vs. 62%; p = .0005 at 5 years

SEER-Medicare (200213), retrospective study

5-FU/LV vs. surgery alone

≥ 65 (4,768)

N/A

0.66 (0.60–0.73)

ACCENT (200114)

5-FU/LV vs. surgery alone

> 70 (506)

Overall 0.68 (0.60– 0.76); p < .001

N/A

Overall 0.76 (0.68–0.85); p < .001

XELOXA, AVANT, X-ACT, NSABP C-08 (201515), pooled analysis

5-FU or capecitabine vs. XELOX or FOLFOX

5-FU: ≥ 70 (424); XELOX or FOLFOX: ≥ 70 (480)

0.77 (0.62– 0.95); p = .014

N/A

0.78 (0.61–0.99); p = .045)

Phase II dose escalation study of capecitabine (201216)

Capecitabine

≥ 70 (82)

50% completed planned therapy at 80% relative dose intensity; stable QoL during treatment; 26% grade 3 handfoot syndrome

N/A

SEER-Medicare, NYSCR, CanCORS, NCCN (20123), retrospective study

Fluoropyrimidine (5FU, capecitabine) ± oxaliplatin

≥ 75 (5,489)

SEER-Medicare

N/A

Adjuvant chemotherapy: 0.60 (0.53–0.68); p = NR

Similar DFS and OS regardless of age

Oxaliplatin: 0.84 (0.69–1.04); p = NR Continued

234 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PERSONALIZING ADJUVANT THERAPY FOR STAGE II/III COLORECTAL CANCER

TABLE 1. Survival Outcomes for Adjuvant Chemotherapy Among Older Adults (Cont'd)

Trial (Year), Study Design

Treatment Arms

Older Adult Cohort, Age (No. of Patients)

Regimen-Specific Data

Disease-Free Survival

RelapseFree Survival

Time to Recurrence

NYSCR

N/A

Overall Survival Adjuvant chemotherapy: 0.76 (0.58–1.01); p = NR Oxaliplatin: 0.82 (0.51–1.33, p = NR)

CanCORS

N/A

NCCN

N/A

Adjuvant chemotherapy: 0.48 (0.19–1.21), p = NR Oxaliplatin: N/A Adjuvant chemotherapy: 0.42 (0.17–1.03); p = NR Oxaliplatin: 1.84 (0.48–7.05); p = NR

SEER-Medicare (200917), retrospective study

MOSAIC phase III (201218)

Adjuvant chemotherapy

FU/LV vs. FOLFOX4

≥ 66 (7,182)

Receipt of chemotherapy (%):

N/A

Overall survival by age:

66–69

N/A

0.47 (0.33– 0.65); p < .001

70–74

N/A

0.32 (0.25– 0.40); p < .001

75–79

N/A

0.41 (0.34– 0.50); p < .001

80–84

16.5

N/A

0.59 (0.49– 0.72); p < .001

≥ 85

N/A

0.54 (0.41– 0.71); p < .001

69.1% (61.3%– 75.8%) vs. 65.8% (57.8%– 72.7%)

N/A

78.8% (71.2%– 84.6%) vs. 69.9% (61.9%– 76.5%);

75.8% (0.73–1.65) vs. 76.1% (68.6–82.1)

0.72 (0.47–1.11); p = .14 at 5 years (referent = age < 70)

1.10 (0.73–1.65); p = .661 at 6 years (referent = age < 70)

70–75 (315)

0.93 (0.64– 1.35); p = .710 at 5 years (referent = age < 70) NO16968 phase III (201119)

FU/FA vs. XELOX

< 65 vs. ≥ 65 (1,886 overall)

No change in DFS/ OS by age reported

NSABP C-07 phase III (201120)

FU/LV vs. FLOX

≥ 70 (396)

62% vs. 62.8%;1.17 (0.94– 1.46); p = .16 (referent = age < 70)

No change in DFS/OS by age reported N/A

N/A

76.3% vs. 71.6% 1.32 (1.03– 1.70), p = .30 (referent = age < 70)

Continued

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 235

MCCLEARY, BENSON, AND DIENSTMANN

TABLE 1. Survival Outcomes for Adjuvant Chemotherapy Among Older Adults (Cont'd)

Treatment Arms

Older Adult Cohort, Age (No. of Patients)

NSABP C-06 phase III (200611)

FU/LV vs. UFT/LV

≥ 60 (939)

1.41 (1.18– 1.69); p = .002 at 5 years (referent = age < 60)

N/A

1.40 (1.12–1.74), p = .03 at 5 years (referent = age < 60)

X-ACT phase III (201221)

Bolus FU/LV vs. capecitabine

≥ 70 (397)

58.1% vs. 55.8%; 0.97 (0.72–1.31) at 5 years

N/A

68.8% vs. 65.0%; 0.91 (0.65–1.26) at 5 years

ACCENT (201322), pooled analysis of 7 adjuvant studies

Oral/IV FU ± irinotecan or oxaliplatin

≥ 70 (2,575)

Oxaliplatin-based regimens (1,119)

0.94 (0.78– 1.13); p = .09

N/A

0.86 (0.69–1.06); p = .36

1.04 (0.85–1.27); p = .05

Oral fluoropyrimidine (757)

0.92 (0.92– 1.41); p = .13

N/A

1.20 (0.93–1.54); p = .09

1.13 (0.90–1.41); p = .16

Trial (Year), Study Design

Regimen-Specific Data

Disease-Free Survival

RelapseFree Survival

Time to Recurrence

Overall Survival

Abbreviations: 5-FU, fluorouracil; CanCORS, Cancer Care Outcomes Research and Surveillance Consortium; DFS, disease-free survival; FA, folinic acid; FLOX, bolus fluorouracil/leucovorin/oxaliplatin; FOLFOX, fluorouracil/leucovorin/oxaliplatin; FU, fluorouracil; IV, intravenous; LV, leucovorin; MOSAIC, Multicenter International Study of Oxaliplatin/5-Fluorouracil/Leucovorin in the Adjuvant Treatment of Colon Cancer; N/A, not available; NCCN, National Comprehensive Cancer Network; NR, not reached; NSABP, National Surgical Adjuvant Breast and Bowel Project; NYSCR, New York State Cancer Registry; OS, overall survival; QoL, quality of life; SEER, Surveillance, Epidemiology and End Results; UFT, uracil tegafur; XELOX, capecitabine/oxaliplatin. Data are presented as hazard ratios (95% CIs) unless otherwise indicated.

Molecular Profile of CRC Among Older Adults

Could a molecular profile determine those older adults unlikely to benefit from adjuvant chemotherapy? We sought to identify a subset of molecular markers unique to older adults diagnosed with colon or rectal cancer. We examined the presence of the CpG island methylator phenotype; microsatellite instability (MSI); KRAS, BRAF, and PIK3CA mutations; and nuclear CTNNβ1 expression status by age at CRC diagnosis within a large prospective cohort study. Tumor nuclear CTNNβ1 appeared to be associated with higher mortality among older adults diagnosed with CRC.32 However, subsequent examination of the impact of nuclear CTNNβ1 and a host of additional molecular factors on prognosis for older adults diagnosed with colon or rectal cancer did not confirm a particular molecular phenotype among older adults diagnosed with colon or rectal cancer (N. J. McCleary, MD, MPH, and A. J. Bass, manuscript in preparation, 2017). Additional study is underway to examine whether a particular molecular phenotype predicts survival among a cohort of older adults receiving chemotherapy for colon or rectal cancer.

Modifying Risk, Enhancing Benefit of Adjuvant Chemotherapy for Older Adults

Potential benefit from adjuvant chemotherapy in older adults must be balanced by the potential for risk attributable to increased toxicity, reduced organ function, sarcopenia, limited social support, or unanticipated decline in physical function.33 Although prospective clinical trials cannot delineate patients most at risk for poor clinical or physical outcomes from specific adjuvant chemotherapy regimens, doses, or duration of treatment, we can glean 236 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

recommendations for treatment decisions from a few notable studies. First, we can predict treatment-related toxicity across multiple cancer types for older adults. Moving beyond the limitations of the Eastern Cooperative Oncology Group or Karnofsky performance status, the comprehensive geriatric assessment has been shown to predict those older adults at risk for toxicity across a number of cancer types and stages, including colon cancer.34,35 The comprehensive geriatric assessment is a feasible, validated instrument that allows both patient and provider evaluation of functional status, medications, social support, cognition, nutrition, psychologic state, and comorbidity to better assess overall fitness. This assessment may identify issues affecting treatment decision making for both patient and provider. The geriatric assessment can predict overall morbidity and mortality but, more specifically, it can anticipate chemotherapy-related toxicity.36-39 A cancer-specific comprehensive geriatric assessment has shown benefit in treatment selection for older adults diagnosed with lung cancer in the ambulatory setting40 and for hospitalized older adults.41 It is now being embedded within a prospective multicenter treatment clinical trial to assess its ability to risk-stratify patients (A. Hurria, personal communication, ALLIANCE meeting, Chicago, IL, November 2016). This and other indices of frailty,42 or risk of increased morbidity and mortality associated with chemotherapy, will not only provide parameters for discussion with older adults regarding the additive risks versus benefits of adjuvant chemotherapy, but they will also potentially inform provider decision making regarding initiation and dosing of treatment.43-45

PERSONALIZING ADJUVANT THERAPY FOR STAGE II/III COLORECTAL CANCER

Second, we can consider the potential impact of particular adjuvant chemotherapy regimens on organ function and physical function absent from any specific comorbid medical condition. Common measures of performance status underestimate the physiologic changes in organ function occurring with aging.24 Bone marrow reserves decrease with increasing age. Chemotherapy treatment can lead to depletion of the bone marrow, thereby increasing the risk of cytopenias and subsequent risks of bleeding or infection. Aging is also associated with decreases in renal and hepatic function, bone and muscle mass, and risks of altered cognition, potentially increasing the risk of treatment-related toxicity.24 Exercise is recommended for secondary cancer prevention following resection of colon cancer and may serve as a useful adjunct during the postoperative treatment course.46,47 However, older adults receiving chemotherapy are susceptible to a decline in physical function, potentially limiting their ability to exercise. Oxaliplatin-induced neuropathy further affects this physical decline and increases the potential risk of falls and limits independence.24,48-50 Third, we can discuss the relative benefit of adjuvant chemotherapy among those older adults with active, unmanaged comorbid medical conditions and competing risk of death or disability. Comorbid medical conditions appear to have a greater effect on older adults diagnosed with advanced CRC.51 Comorbid medical conditions may impact drug absorption and clearance. The presence of comorbid medical conditions predict for concomitant medications and risk of drug interactions.52 Regular careful review of patient medications, as promoted by geriatric assessment, can limit the potential risk of drug-drug interactions.53 In the adjuvant setting, renal excretion of both capecitabine and oxaliplatin requires dose adjustment for creatinine clearance below 50 mL/min.24 Capecitabine also requires dose adjustment for patients taking warfarin. Cognitive impairment increases the risk of nonadherence to capecitabine. Finally, it is incumbent on us as oncology providers to understand the full impact of adjuvant chemotherapy on older adults beyond disease-free and overall survival.54 Disease-free and overall survival are the primary outcomes used to determine the standards for adjuvant chemotherapy regardless of age at diagnosis. However, other clinical and quality outcomes may be of interest to patients. Although outcomes of interest have not yet been specifically identified for older adults, few clinical trials evaluate outcomes beyond traditional outcomes of disease-free and overall survival to include outcomes potentially pertinent to older adults, such as the impact of adjuvant chemotherapy on “quality of survival and functional independence.”34 Given this, we can consider the traditional outcomes as a measure of treatment response, but we cannot fully comment on other benefits that older adults may experience as a result of adjuvant chemotherapy. Adults age 65 or older reported greater decline in physical and mental health within the first 6 months of diagnosis of CRC compared with age-matched controls as part of the Medicare Health Outcomes Survey, particularly among patients diagnosed with

stage III or IV CRC.55 How do we best define “functional independence” and “quality of survival” over the course of adjuvant chemotherapy administration and afterward? What is an acceptable threshold for additional outcomes beyond which treatment should not be recommended regardless of potential disease-free or overall survival benefit? We must begin exploring those additional outcomes of importance to older adults to determine the full impact of adjuvant chemotherapy on older adults and develop strategies to improve outcomes globally.

CONTROVERSIES IN THE ADJUVANT SETTING

Adjuvant Chemotherapy for Rectal Cancer

For many years, the standard of care for patients with locally advanced clinical stage II to III rectal cancer included surgery, often resulting in a permanent ostomy, followed by adjuvant chemotherapy and chemoradiation.11-14,16,18-21 This strategy improved both overall survival and the risk of locoregional failure. An example of the outcome benefits of combined adjuvant chemoradiation, published more than a decade ago, include the U.S. Intergroup 0144 trial, which evaluated the so-called sandwich approach of chemotherapy followed by chemoradiation followed by additional chemotherapy and compared bolus versus infusional 5-FU regimens for patients with T3-4N0M0 or T1-4N1,2M0 disease.56 The locoregional failure rate for those who received low anterior resection was between 3% and 5%. Three-year overall survival was between 81% and 83%. A pooled analysis of North American phase III combined modality adjuvant trials identified three different risk groups defined by TN stage, including T1-T2N1 and T3N0 (intermediate); T1-2N2, T3N1, and T4N0 (moderately high); and T3N2, T4N1, and T4N2 (high), which correlated with survival and disease control.57,58 Fiveyear overall survival rates for the intermediate group were 78%–85% compared with 25%–57% for those with high-risk lesions. Different treatment strategies depending upon risk were therefore implied. Subsequently there was a profound shift in the treatment approach for clinical stage II to III rectal cancer as data emerged supporting the use of neoadjuvant chemoradiation; however, this therapeutic evolution generated considerable controversy as to the role of adjuvant chemotherapy, a controversy that has persisted. Neoadjuvant chemoradiation has become the preferred treatment of locally advanced rectal cancer because of evidence demonstrating improved outcomes, better tolerability, and, in many cases, considerable downstaging resulting in sphincter-preserving surgery and thus avoiding a permanent ostomy. A hallmark study from the Working Group of Surgical Oncology/Working Group of Radiation Oncology/ Working Group of Medical Oncology of the Germany Cancer Society (CAO/ARO/AIO-94) compared preoperative chemoradiotherapy with postoperative chemoradiotherapy for locally advanced rectal cancer, demonstrating significant improvement in 5-year cumulative incidence of local relapse favoring a preoperative approach (6% vs. 13%; p = .006).59 There were considerably less acute and long-term toxic effects in the preoperative group, although 5-year overall survival asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 237

MCCLEARY, BENSON, AND DIENSTMANN

rates were similar (76% vs. 74%). Patients also received four cycles of postoperative 5-FU. Long-term follow-up data showed improved outcomes for the preoperative patients who achieved complete and intermediate tumor regressions and the overall 10-year cumulative incidence of local relapse continued to favor the patients treated preoperatively (7.1% vs. 10.1%; p = .048); there was no change in overall survivorship (59.6% vs. 59.9%).60,61 A recent meta-analysis of more than 10,000 patients who participated in randomized controlled trials confirmed the improved rate of local control with neoadjuvant chemoradiation, including after total mesorectal excision, although there was no improvement in long-term survival.62 Although incorporating adjuvant chemotherapy for patients who have received neoadjuvant chemoradiation and surgery would appear to be a reasonable strategy to improve survivorship as an extrapolation from stage III colon cancer adjuvant trials, attempts at defining the optimal rectal cancer population that would benefit from adjuvant therapy remain elusive. This paucity of consistent evidence has resulted in variability in practice patterns. For example, a National Comprehensive Cancer Network CRC database assessment of nearly 2,000 patients with stage II/III rectal cancer who received neoadjuvant chemoradiation showed that a sizable minority of patients did not receive adjuvant chemotherapy.63 A SEER-Medicare database analysis noted that one in three patients did not receive adjuvant therapy after neoadjuvant chemoradiation and resection.64 Some investigations have attempted to select patients who may not require adjuvant therapy after neoadjuvant chemoradiation and surgery. For example, a study of 176 patients reported that those who achieved a complete response (15.3% of patients staged as ypT0M0) had 5-year disease-free and overall survival rates of 96% and 100%, respectively, suggesting that adjuvant therapy would provide no further meaningful benefit for these individuals.65 In a retrospective study of 851 patients, 330 received preoperative short-course radiation (2,500 cGy administered in five fractions without chemotherapy) and 123 received adjuvant chemotherapy.66 A subgroup analysis showed that adjuvant therapy improved disease-specific survival and overall survival only for those patients who had at least two high-risk features such as pT4 tumor, inadequate lymph node sampling, lymphovascular invasion, perineural invasion, poor differentiation, obstruction, or perforation. EORTC 22921 was a randomized trial of 1,011 patients evaluating FU-based adjuvant chemotherapy after preoperative chemoradiation for patients with clinical stage T3 or T4 resectable rectal cancer.67 Patients were assigned to one of four treatment arms including preoperative radiotherapy with or without chemotherapy and preoperative radiotherapy with or without chemotherapy followed by adjuvant chemotherapy. There was relatively poor adherence to adjuvant chemotherapy, because only 43% of patients received the planned dose. At a median follow-up of 10.4 years, there was no substantial difference in overall survival among the four treatment groups (48.4%–51.9%), nor were 238 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

there differences in disease-free survival rates and cumulative incidence of distant metastases. Most recurrences were noted within 5 years. A recently reported Italian study of 634 evaluable patients concluded that adjuvant 5-FU did not improve 5-year overall or disease-free survival, including among those who obtained a complete pathologic response and overall downstaging rates; 28% of patients, however, never received the assigned adjuvant chemotherapy.68 A Dutch study of 437 eligible patients closed prematurely for accrual reasons; however, there was no difference in 5-year cumulative incidence for either local regional recurrence or in 5-year distance recurrences after postoperative fluoropyrimidine monotherapy.69 Systematic reviews and meta-analyses also were recently reported to address the role of adjuvant chemotherapy after neoadjuvant therapy and surgery. An analysis of four phase III clinical trials of nearly 1,200 patients with ypTNM stage II and III rectal cancers and a R0 resection found no difference in overall survival comparing those who received adjuvant chemotherapy versus observation.70 There were patients with tumors located at 10 to 15 cm from the anal verge who had improved disease-free survival and fewer distant metastases when treated with adjuvant chemotherapy. Another analysis of randomized controlled trials in retrospective studies of nearly 5,500 patients reported improvement in both 5-year overall survival and disease-free survival for those patients treated with neoadjuvant chemoradiation, surgery, and adjuvant chemotherapy.71 The improvement in 5-year overall survival was largest among patients who were downstaged and in the retrospective series. A third systematic review and meta-analysis of five randomized trials including 2,398 patients did not show an advantage for those who received adjuvant chemotherapy although there was a substantial adjuvant chemotherapy effect for patients who were randomized after surgery (753 patients).72 In two trials, there was a difference in disease-free survival for those who received FU and oxaliplatin compared with single-agent 5-FU; however, in two other trials, FU and oxaliplatin did not show a substantial difference. Overall, the authors concluded that adjuvant chemotherapy provided no “strong scientific evidence” to support its use for those who received preoperative chemoradiation. A number of treatment strategies have been the subject of recent clinical trials and have informed current or planned global clinical trial portfolios73,74 (Table 2). For example, there is interest in the “wait and watch” approach for patients who have obtained a complete response after chemoradiation and in strategies to encompass neoadjuvant chemotherapy while reserving radiation for those with suboptimal response.73,74 The overall goal is to avoid more extensive intervention with the associated risk of toxicity and long-term sequelae for patients who may not need such an approach and to intensify therapy for those who are at highest risk for recurrence. Current National Comprehensive Cancer Network guidelines recommend a number of options for patients with clinical stage II and III rectal cancer, including (1) neoadjuvant therapy comprising long-course chemoradiation with either

PERSONALIZING ADJUVANT THERAPY FOR STAGE II/III COLORECTAL CANCER

capecitabine or infusional 5-FU, short-course radiation, or a preferred chemotherapy regimen with oxaliplatin and a fluoropyrimidine followed by chemoradiation; or (2) adjuvant therapy is recommended after surgery for those who have received neoadjuvant oxaliplatin and a fluoropyrimidine followed by chemoradiation surveillance, whereas adjuvant chemotherapy with oxaliplatin and a fluoropyrimidine is recommended as the preferred regimen for those treated with chemoradiation or short-course radiation.75

Adjuvant Therapy for Resectable Metastatic Disease

It has long been known that there is a subgroup of patients with colon or rectal cancer who have potentially resectable

metastatic disease and can enjoy long-term survival after surgery. The introduction of combination chemotherapy for metastatic CRC has resulted in improvement in response, progression-free survival, and overall survival. In addition, there is a perceived benefit of combination chemotherapy for patients with resectable metastatic disease or those who obtain a substantial response to therapy rendering them with resectable disease. An advantage of preoperative chemotherapy for patients with resectable or potentially resectable metastatic disease is to determine “chemosensitivity” and also to identify those individuals who may be resistant to therapy and develop more rapid disease progression. For patients with rectal cancer and potentially resectable

TABLE 2. Select Current and Planned Rectal Cancer Trials Trial

Treatment

NO148 PROSPECT phase III

FOLFOX × 6 → response ≥ 20% → TME → FOLFOX × 6 FOLFOX × 6 → response < 20% → 5-FU/capecitabine RT → TME → FOLFOX × 2 versus 5-FU/capecitabine RT → TME → FOLFOX x 8

NRG G1-002 (TNT) phase II High risk

FOLFOX × 8 → RT + capecitabine → surgery FOLFOX × 8 → RT + capecitabine + veliparib → surgery Additional arms planned

CCTG CO28 NeoTEMS phase II

cT1-3N0 FOLFOX/CAPOX → TEMS/TAMIS → ypTO/T1Good → surveillance cT1-3N0 FOLFOX/CAPOX → TEMS/TAMIS → ypT1Bad or higher → TMEa

OPRA (13-213) phase II

ChemoRT + chemotherapy → TME ChemoRT + chemotherapy → surveillance

RIA trial phase II High risk

FOLFOX/aflibercept × 6 → capecitabine RT → surgery FOLFOX × 6 → capecitabine RT → surgery

PIER phase II Intermediate risk

FOLFOX/panitumumab → no progression → TME FOLFOX/panitumumab → progression → capecitabine RT

ARISTOTLE NCRI phase III

Capecitabine RT → surgery Irinotecan + capecitabine RT → surgery

AIO/ARO/A10-04

ChemoRT → surgery → FOLFOX → 5-FU

AIO/ARO/A10-12 phase II

Chemotherapy → chemoRT → surgery ChemoRT → chemotherapy → Surgery

AIO/ARO/A10-16 Low risk

Surgery → FOLFOX (pN+) versus Capecitabine RT → surgery → capecitabine

High risk

Capecitabine RT → surgery → capecitabine versus 5-FU/oxaliplatin RT → FOLFOX → surgery

RENO

ChemoRT → cCR → watch and wait

RAPIDO phase III

ChemoRT → MRI/CT → response → TME → CAPOX versus RT → CAPOX → MRI/CT → response → TME

aypT1Bad, RI, high grade. Abbreviations: 5-FU, fluorouracil; CAPOX, capecitabine/oxaliplatin; cCR, complete clinical response; chemoRT, chemoradiotherapy; FOLFOX, fluorouracil/leucovorin/oxaliplatin; pN+, pathologically node positive; RI, microscopic positive margin; RT, radiotherapy; TAMIS, transanal minimally invasice surgery; TEMS, transanal endoscopic micro-surgery; TME, total mesorectal excision.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 239

MCCLEARY, BENSON, AND DIENSTMANN

metastatic disease, preoperative chemoradiotherapy is often considered to reduce the risk of local regional recurrence, particularly when the goal of surgery is curative intent. A perioperative approach for those with resectable metastatic disease incorporates a total chemotherapy treatment period of approximately 6 months including preoperative therapy followed by surgery and adjuvant chemotherapy. The role of adjuvant chemotherapy, however, has not been clearly defined because of very limited data to provide guidance. A systematic review of 642 evaluable patients with liver metastases evaluated surgery versus surgery and chemotherapy, demonstrating improvement in disease-free and progression-free survival favoring chemotherapy without a survival advantage.76 A meta-analysis of 10 studies including nearly 1,900 patients showed no survival benefit for patients who received perioperative chemotherapy for resectable liver metastases compared with surgery alone; however, a disease-free survival benefit was noted.77 Similar results were observed in additional analyses.78,79 EORTC 40983 evaluated six cycles of fluorouracil/leucovorin/oxaliplatin before and after surgical resection of liver metastases compared with surgery alone, demonstrating a 40% response to preoperative fluorouracil/leucovorin/oxaliplatin and improvement in progression-free survival for eligible patients who were resected with no overall survival benefit.80 There is consensus in the National Comprehensive Cancer Network guidelines that adjuvant chemotherapy after resection of metastatic disease remains an option of care.75

MOLECULAR PROFILES AND THE FUTURE OF ADJUVANT THERAPY: MICROENVIRONMENT MATTERS

Retrospective biomarker analyses of multiple clinical trials in the adjuvant setting strongly support the feasibility of refining prognostic stratification in CRC by factoring in molecular features with pathologic tumor staging.81 However, validated predictive markers of adjuvant therapy benefit for stage II or III CRCs are still lacking.81 To date, the only molecular marker with proven clinical utility in early-stage CRC is MSI, which associates with very good prognosis in stage II disease irrespective of adjuvant chemotherapy, supporting a no-adjuvanttreatment approach.82 On the other hand, patients with MSI stage III CRC derive benefit from adjuvant chemotherapy, with no differential benefit compared with the microsatellite stability (MSS) population in clinical trials assessing 5-FU or oxaliplatinbased regimens.83 Interestingly, there is a possible interaction between MSI status and primary tumor location in stage III treated disease, with a better prognosis limited to right-sided tumors.84 This association reinforces the known intrinsic biologic differences between proximal and distal CRC.85 Mounting evidence indicates that an enhanced lymphocytic reaction in CRC is a critical determinant of the risk of dissemination to distant metastasis.86 A clinical translation of this finding was the establishment of a scoring system, called the “immunoscore,” based on the abundance of two distinct lymphocyte populations (CD8+ cytotoxic T cells and CD3+ T memory cells) at the tumor center and at its invasive 240 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

margin. In a large validation study, time to recurrence and overall survival were significantly longer for patients with stage II and III colon cancer with immunoscore high tumors, independent of clinicopathological factors.87 MSI cancers characteristically exhibit strong infiltration of the tumor microenvironment with immune cells, which relates to hypermutation rates and higher neoantigen loads.88 However, a subset of MSS tumors also have increased intratumoral adaptive immune gene expression and high immunoscores. These “immune-activated” tumors, irrespective of stage, have improved survival outcomes and the immunoscore was shown to be superior to MSI in predicting patients’ disease-specific recurrence and survival in multivariable models.89 These data strengthen the concept that reduced immune cytotoxicity is a major factor driving metastases in CRC. However, most patients with early-stage CRC have MSS and/or a medium/low immunoscore, which associate with an intermediate to poor prognosis and do not help prioritize adjuvant chemotherapy in stage II or III disease. The same is true for tumors harboring BRAF V600E mutations, which are an independent prognostic factor of reduced overall survival in multiple studies, particularly in MSS left-sided disease, but not a marker of chemosensitivity/resistance to 5-FU or oxaliplatin-based regimens in the adjuvant setting.81 In addition to microsatellite status and gene mutations, which did not demonstrate predictive value for standard chemotherapy benefit in early-stage CRC, different groups explored the potential clinical utility of gene expression signatures in this context. The transcriptomic profile of a tumor, encompassing cancer cell, immune, and stromal signals, is intimately linked to its phenotype and clinical behavior. Gene expression profiling has been used extensively to identify biologically hom*ogeneous subtypes of the disease through unsupervised clustering. An international effort dedicated to large-scale data sharing and coordinated analytics cross-compared independent transcriptomic-based CRC subtyping systems and resulted in a consensus molecular classification that allows the categorization of most CRC tumors into one of four robust intrinsic subtypes.90 The consensus molecular subtype (CMS) features are summarized in Table 3. There are striking differences in prognosis with this unsupervised gene expression signature, confirming that the biologic processes implicated in each subtype are clinically relevant.90 The CMS4 mesenchymal group is associated with a significantly higher risk of distant relapse and death for patients diagnosed with early-stage CRC, irrespective of validated clinicopathological features, MSI status, and BRAF V600E mutations.81 These tumors exert a proangiogenic and stromagenic influence on the microenvironment, which is highly infiltrated with endothelial cells and cancer-associated fibroblasts. In addition, CMS4 mesenchymal tumors are enriched with immunosuppressive cells, such as regulatory T cells, B cells, and myeloid-derived suppressor cells, which are negative regulators of cytotoxic T cells.91-93 This effect is explained in part by high expression of transforming growth factor-β and chemokines attracting myeloid cells (C-C motif chemokine ligand CCL2) and related cytokines

PERSONALIZING ADJUVANT THERAPY FOR STAGE II/III COLORECTAL CANCER

(interleukin-23 and interleukin-17).92,93 The proangiogenic/ stromagenic/immunosuppressive phenotype of CMS4 mesenchymal tumors, with their invasive-inflamed microenvironment, is intimately linked to higher chances of metastatic spread and resistance to therapy.94,95 Indeed, retrospective biomarker analysis of the NSABP C-07 randomized clinical trial showed poor prognosis and no benefit from adjuvant oxaliplatin-based chemotherapy in the subset of patients with stage III CRC whose tumors displayed a mesenchymal phenotype.96 However, the clinical utility of using intrinsic CRC subtyping to identify patients for oxaliplatin treatment requires validation in independent clinical trial cohorts. Similarly, the value of supervised gene expression classifiers for adjuvant chemotherapy selection remains to be proven. Different prognostic signatures, such as Oncotype DX Colon Cancer, ColoPrint, Veridex, and GeneFx Colon, have been widely evaluated retrospectively in clinical cohorts.81 Irrespective of assay, gene panel size, and tissue source (fresh, frozen, formalin fixed, paraffin embedded), analysis of the various transcriptomes in CRC can effectively classify patients into subgroups at low and high risk of disease relapse. The original hypothesis was that patients whose tumors are categorized as high risk have increased benefit from adjuvant chemotherapy. In theory, the prognostic information provided by these signatures could have the greatest clinical utility when used as a complement to T stage and MSI status, specifically for patients who have pT3pN0 MSS disease.97 However, the relative chemotherapy benefit for Oncotype DX Colon Cancer was shown to be similar across risk groups.98,99 Despite the fact that gene expression–based risk scores seem to add little to risk models with known prognostic factors,100 incorporation of the signature results into clinical practice was associated with changes in treatment recommendation for nearly 50% of patients with pT3pN0 MSS CRC compared with traditional clinicopathological assessment variables alone.101 Prospective validation of these signatures has not yet been presented, and

currently only one trial (PARSC [Prospective Study for the Assessment of Recurrence Risk in Stage II Colon Cancer Patients]) is comparing risk assessment using the ColoPrint profile versus a clinical risk assessment based on the investigator’s judgment and American Society of Clinical Oncology recommendations for high-risk disease. Furthermore, economic studies assessing the cost-effectiveness of using gene expression signatures to select patients with CRC who have a high risk of relapse (and to base adjuvant chemotherapy decision making on this criterion) are not yet available. Given the fact that high risk scores in supervised signatures have substantial overlap with a mesenchymal phenotype,102 it is understandable that these prognostic classifiers have limited predictive value for adjuvant chemotherapy selection. This finding is in stark contrast with prognostic gene expression classifiers in early-stage breast cancer, in which high risk scores associate with high proliferation rates and increased benefit from more aggressive adjuvant chemotherapy.103 In summary, pathways that coordinate the creation of an immunosuppressive microenvironment and stromal invasiveness are the key drivers of a prometastatic state in CRC.86 These processes are strongly enriched in the CMS4 mesenchymal CRC population,90 which is poorly responsive to standard chemotherapies.95,96 The following investigations should be pursued by the scientific community: (1) correlating response patterns of targeted agents and immunotherapies with the CMS classification in existing clinical trials; (2) adapting the design of future trials, such as adding stratification factors or increasing their power to allow these retrospective correlative analyses to be performed; and (3) designing prospective clinical trials in CRC that incorporate new biomarkers with drug repositioning and/or novel matched targeted agents and immunotherapies.91 Different academic groups are working on a practical and robust CMS classifier that works on formalin-fixed, paraffin-embedded primary CRC tissues (either gene expression or immunohistochemistry based).104 Molecular classifiers based on

TABLE 3. Clinical and Molecular Features of Intrinsic Gene Expression–Based CRC Subtypes Feature

CMS1 (MSI Immune)

CMS2 (Canonical)

CMS3 (Metabolic)

CMS4 (Mesenchymal)

Prevalence in earlystage CRC, approximate %

Primary tumor site

Enriched right side of the colon

Enriched left side of the colon and rectum

Enriched right side of the colon

Enriched left side of the colon and rectum

Cancer cell features

MSI, hypermutated, hypermethylated, enriched for BRAF mutations

MSS, chromosomal instability, EGFR, and ERBB2 upregulation

Mixed MSI/MSS status, chromosomal instability, metabolic deregulation, enriched for KRAS mutations

MSS, chromosomal instability, epithelial-mesenchymal transition and stemness

Microenvironment features

Infiltrated with cytoxic T, helper T, and natural killer cells

Limited immune cell or stromal cell infiltration

Highly infiltrated with stromal cells, regulatory T cells, B cells, and myeloid-derived suppressor cells

Prognosis

Better relapse-free survival and worse survival after relapse

Better relapse-free and overall survival

Worse relapse-free and overall survival

Abbreviations: CMS, consensus molecular subtype; CRC, colorectal cancer; MSI, microsatellite instability; MSS, microsatellite stability.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 241

MCCLEARY, BENSON, AND DIENSTMANN

intrinsic tumor phenotypes are already being investigated in prospective clinical trials in the metastatic setting, such as in the MoTriColor project, a large pan-European effort pioneering novel molecularly guided trials in metastatic CRC, but the biologic differences between micro- and macrometastatic disease must be taken into account when translating data garnered from advanced-stage CRC into early-stage disease regarding treatment decisions.91 The recent failures with cetuximab and bevacizumab in adju-

vant trials in stage III CRC exposed this challenge and call into question our traditional paradigm of drug development (namely, considering agents for testing in the curative setting only after they are found to be beneficial in the treatment of patients with metastatic disease). We believe that this new biologic understanding is expected to guide drug selection in future adjuvant clinical trials and is hoped to increase cure rates and survival in CRC.

References 1. National Cancer Institute Surveillance, Epidemiology, and End Results Program (SEER). Cancer Stat Facts: Colon and Rectum Cancer, 2017. https://seer.cancer.gov/statfacts/html/colorect.html. Accessed January 25, 2017.

13. Sundararajan V, Mitra N, Jacobson JS, et al. Survival associated with 5-fluorouracil-based adjuvant chemotherapy among elderly patients with node-positive colon cancer. Ann Intern Med. 2002;136: 349-357.

2. Sorrentino JA, Sanoff HK, Sharpless NE. Defining the toxicology of aging. Trends Mol Med. 2014;20:375-384.

14. Sargent DJ, Goldberg RM, Jacobson SD, et al. A pooled analysis of adjuvant chemotherapy for resected colon cancer in elderly patients. N Engl J Med. 2001;345:1091-1097.

3. Sanoff HK, Carpenter WR, Stürmer T, et al. Effect of adjuvant chemotherapy on survival of patients with stage III colon cancer diagnosed after age 75 years. J Clin Oncol. 2012;30:2624-2634. 4. Bergquist JR, Thiels CA, Spindler BA, et al. Benefit of postresection adjuvant chemotherapy for stage III colon cancer in octogenarians: analysis of the National Cancer Database. Dis Colon Rectum. 2016;59:1142-1149. 5. Koroukian SM, Xu F, Bakaki PM, et al. Comorbidities, functional limitations, and geriatric syndromes in relation to treatment and survival patterns among elders with colorectal cancer. J Gerontol A Biol Sci Med Sci. 2010;65A:322-329. 6. Gill S, Loprinzi CL, Sargent DJ, et al. Pooled analysis of fluorouracilbased adjuvant therapy for stage II and III colon cancer: who benefits and by how much? J Clin Oncol. 2004;22:1797-1806. 7. Twelves C, Wong A, Nowacki MP, et al. Capecitabine as adjuvant treatment for stage III colon cancer. N Engl J Med. 2005;352:26962704. 8. André T, de Gramont A, Vernerey D, et al. Adjuvant fluorouracil, leucovorin, and oxaliplatin in stage II to III colon cancer: updated 10year survival and outcomes according to BRAF mutation and mismatch repair status of the MOSAIC Study. J Clin Oncol. 2015;33:4176-4187. 9. Schmoll HJ, Tabernero J, Maroun J, et al. Capecitabine plus oxaliplatin compared with fluorouracil/folinic acid as adjuvant therapy for stage III colon cancer: final results of the NO16968 randomized controlled phase III trial. J Clin Oncol. 2015;33:3733-3740. 10. André T, Boni C, Mounedji-Boudiaf L, et al; Multicenter International Study of Oxaliplatin/5-Fluorouracil/Leucovorin in the Adjuvant Treatment of Colon Cancer (MOSAIC) Investigators. Oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment for colon cancer. N Engl J Med. 2004;350:2343-2351. 11. Lembersky BC, Wieand HS, Petrelli NJ, et al. Oral uracil and tegafur plus leucovorin compared with intravenous fluorouracil and leucovorin in stage II and III carcinoma of the colon: results from National Surgical Adjuvant Breast and Bowel Project Protocol C-06. J Clin Oncol. 2006;24:2059-2064. 12. Jessup JM, Stewart A, Greene FL, et al. Adjuvant chemotherapy for stage III colon cancer: implications of race/ethnicity, age, and differentiation. JAMA. 2005;294:2703-2711.

242 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

15. Haller DG, O’Connell MJ, Cartwright TH, et al. Impact of age and medical comorbidity on adjuvant treatment outcomes for stage III colon cancer: a pooled analysis of individual patient data from four randomized, controlled trials. Ann Oncol. 2015;26:715-724. 16. Chang HJ, Lee KW, Kim JH, et al. Adjuvant capecitabine chemotherapy using a tailored-dose strategy in elderly patients with colon cancer. Ann Oncol. 2012;23:911-918. 17. Zuckerman IH, Rapp T, Onukwugha E, et al. Effect of age on survival benefit of adjuvant chemotherapy in elderly patients with stage III colon cancer. J Am Geriatr Soc. 2009;57:1403-1410. 18. Tournigand C, André T, Bonnetain F, et al. Adjuvant therapy with fluorouracil and oxaliplatin in stage II and elderly patients (between ages 70 and 75 years) with colon cancer: subgroup analyses of the Multicenter International Study of Oxaliplatin, Fluorouracil, and Leucovorin in the Adjuvant Treatment of Colon Cancer trial. J Clin Oncol. 2012;30:3353-3360. 19. Haller DG, Tabernero J, Maroun J, et al. Capecitabine plus oxaliplatin compared with fluorouracil and folinic acid as adjuvant therapy for stage III colon cancer. J Clin Oncol. 2011;29:1465-1471. 20. Yothers G, O’Connell MJ, Allegra CJ, et al. Oxaliplatin as adjuvant therapy for colon cancer: updated results of NSABP C-07 trial, including survival and subset analyses. J Clin Oncol. 2011;29:3768-3774. 21. Twelves C, Scheithauer W, McKendrick J, et al. Capecitabine versus 5-fluorouracil/folinic acid as adjuvant therapy for stage III colon cancer: final results from the X-ACT trial with analysis by age and preliminary evidence of a pharmacodynamic marker of efficacy. Ann Oncol. 2012;23:1190-1197. 22. McCleary NJ, Meyerhardt JA, Green E, et al. Impact of age on the efficacy of newer adjuvant therapies in patients with stage II/III colon cancer: findings from the ACCENT database. J Clin Oncol. 2013;31:2600-2606. 23. Moth EB, Vardy J, Blinman P. Decision-making in geriatric oncology: systemic treatment considerations for older adults with colon cancer. Expert Rev Gastroenterol Hepatol. 2016;10:1321-1340. 24. McCleary NJ, Dotan E, Browner I. Refining the chemotherapy approach for older patients with colon cancer. J Clin Oncol. 2014;32: 2570-2580.

PERSONALIZING ADJUVANT THERAPY FOR STAGE II/III COLORECTAL CANCER

25. Lund CM, Nielsen D, Dehlendorff C, et al. Efficacy and toxicity of adjuvant chemotherapy in elderly patients with colorectal cancer: the ACCORE study. ESMO Open. 2016;1:e000087.

44. Wildiers H, Heeren P, Puts M, et al. International Society of Geriatric Oncology consensus on geriatric assessment in older patients with cancer. J Clin Oncol. 2014;32:2595-2603.

26. Lund JL, Stürmer T, Sanoff HK, et al. Determinants of adjuvant oxaliplatin receipt among older stage II and III colorectal cancer patients. Cancer. 2013;119:2038-2047.

45. Decoster L, Van Puyvelde K, Mohile S, et al. Screening tools for multidimensional health problems warranting a geriatric assessment in older cancer patients: an update on SIOG recommendations. Ann Oncol. 2015;26:288-300.

27. Dobie SA, Baldwin LM, Dominitz JA, et al. Completion of therapy by Medicare patients with stage III colon cancer. J Natl Cancer Inst. 2006;98:610-619. 28. Bradley CJ, Given CW, Dahman B, et al. Adjuvant chemotherapy after resection in elderly Medicare and Medicaid patients with colon cancer. Arch Intern Med. 2008;168:521-529. 29. Schrag D, Cramer LD, Bach PB, et al. Age and adjuvant chemotherapy use after surgery for stage III colon cancer. J Natl Cancer Inst. 2001;93:850-857. 30. Dobie SA, Warren JL, Matthews B, et al. Survival benefits and trends in use of adjuvant therapy among elderly stage II and III rectal cancer patients in the general population. Cancer. 2008;112:789-799. 31. Neugut AI, Matasar M, Wang X, et al. Duration of adjuvant chemotherapy for colon cancer and survival among the elderly. J Clin Oncol. 2006;24:2368-2375. 32. McCleary NJ, Sato K, Nishihara R, et al. prognostic utility of molecular factors by age at diagnosis of colorectal cancer. Clin Cancer Res. 2016;22:1489-1498. 33. Broughman JR, Williams GR, Deal AM, et al. Prevalence of sarcopenia in older patients with colorectal cancer. J Geriatr Oncol. 2015;6:442-445. 34. Extermann M, Hurria A. Comprehensive geriatric assessment for older patients with cancer. J Clin Oncol. 2007;25:1824-1831. 35. Kelly CM, Shahrokni A. Moving beyond Karnofsky and ECOG performance status assessments with new technologies. J Oncol. 2016;2016:6186543. 36. de Glas NA, Bastiaannet E, Engels CC, et al. Validity of the online PREDICT tool in older patients with breast cancer: a population-based study. Br J Cancer. 2016;114:395-400. 37. Hurria A, Togawa K, Mohile SG, et al. Predicting chemotherapy toxicity in older adults with cancer: a prospective multicenter study. J Clin Oncol. 2011;29:3457-3465. 38. Hurria A, Akiba C, Kim J, et al. Reliability, validity, and feasibility of a computer-based geriatric assessment for older adults with cancer. J Oncol Pract. 2016;12:e1025-e1034. 39. McCleary NJ, Wigler D, Berry D, et al. Feasibility of computer-based self-administered cancer-specific geriatric assessment in older patients with gastrointestinal malignancy. Oncologist. 2013;18:64-72. 40. Gajra A, Loh KP, Hurria A, et al. Comprehensive geriatric assessmentguided therapy does improve outcomes of older patients with advanced lung cancer. J Clin Oncol. 2016;34:4047-4048. 41. Baitar A, Kenis C, Moor R, et al. Implementation of geriatric assessment-based recommendations in older patients with cancer: a multicentre prospective study. J Geriatr Oncol. 2015;6:401-410. 42. Ferrat E, Paillaud E, Caillet P, et al. Performance of four frailty classifications in older patients with cancer: prospective elderly cancer patients cohort study. J Clin Oncol. Epub 2017 Jan 17. 43. Pal SK, Katheria V, Hurria A. Evaluating the older patient with cancer: understanding frailty and the geriatric assessment. CA Cancer J Clin. 2010;60:120-132.

46. Meyerhardt JA, Giovannucci EL, Holmes MD, et al. Physical activity and survival after colorectal cancer diagnosis. J Clin Oncol. 2006;24:35273534. 47. Meyerhardt JA, Heseltine D, Niedzwiecki D, et al. Impact of physical activity on cancer recurrence and survival in patients with stage III colon cancer: findings from CALGB 89803. J Clin Oncol. 2006;24:35353541. 48. van Erning FN, Janssen-Heijnen ML, Wegdam JA, et al. The course of neuropathic symptoms in relation to adjuvant chemotherapy among elderly patients with stage III colon cancer: alongitudinal study. Clin Colorectal Cancer. Epub 2016 Sep 17. 49. Mols F, Beijers T, Lemmens V, et al. Chemotherapy-induced neuropathy and its association with quality of life among 2- to 11-year colorectal cancer survivors: results from the population-based PROFILES registry. J Clin Oncol. 2013;31:2699-2707. 50. Tofthagen C, Overcash J, Kip K. Falls in persons with chemotherapyinduced peripheral neuropathy. Support Care Cancer. 2012;20:583589. 51. Meyerhardt JA, McCleary NJ, Niedzwiecki D, et al. Impact of age and comorbidities on treatment effect, tolerance, and toxicity in metastatic colorectal cancer (mCRC) patients treated on CALGB 80203. J Clin Oncol. 2009;27:15s (suppl; abstr 4038). 52. Stepney R, Lichtman SM, Danesi R. Drug-drug interactions in older patients with cancer: a report from the 15th Conference of the International Society of Geriatric Oncology, Prague, Czech Republic, November 2015. Ecancermedicalscience. 2016;10:611. 53. Turner JP, Shakib S, Bell JS. Is my older cancer patient on too many medications? J Geriatr Oncol. Epub 2016 Nov 11. 54. Sanoff HK, Goldberg RM, Pignone MP. A systematic review of the use of quality of life measures in colorectal cancer research with attention to outcomes in elderly patients. Clin Colorectal Cancer. 2007;6:700-709. 55. Quach C, Sanoff HK, Williams GR, et al. Impact of colorectal cancer diagnosis and treatment on health-related quality of life among older Americans: a population-based, case-control study. Cancer. 2015;121:943-950. 56. Smalley SR, Benedetti JK, Williamson SK, et al. Phase III trial of fluorouracil-based chemotherapy regimens plus radiotherapy in postoperative adjuvant rectal cancer: GI INT 0144. J Clin Oncol. 2006;24:3542-3547. 57. Gunderson LL, Sargent DJ, Tepper JE, et al. Impact of T and N stage and treatment on survival and relapse in adjuvant rectal cancer: a pooled analysis. J Clin Oncol. 2004;22:1785-1796. 58. Gunderson LL, Callister M, Marschke R, et al. Stratification of rectal cancer stage for selection of postoperative chemoradiotherapy: current status. Gastrointest Cancer Res. 2008;2:25-33. 59. Sauer R, Becker H, Hohenberger W, et al; German Rectal Cancer Study Group. Preoperative versus postoperative chemoradiotherapy for rectal cancer. N Engl J Med. 2004;351:1731-1740.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 243

MCCLEARY, BENSON, AND DIENSTMANN

60. Sauer R, Liersch T, Merkel S, et al. Preoperative versus postoperative chemoradiotherapy for locally advanced rectal cancer: results of the German CAO/ARO/AIO-94 randomized phase III trial after a median follow-up of 11 years. J Clin Oncol. 2012;30:1926-1933. 61. Fokas E, Liersch T, Fietkau R, et al. Tumor regression grading after preoperative chemoradiotherapy for locally advanced rectal carcinoma revisited: updated results of the CAO/ARO/AIO-94 trial. J Clin Oncol. 2014;32:1554-1562. 62. Rahbari NN, Elbers H, Askoxylakis V, et al. Neoadjuvant radiotherapy for rectal cancer: meta-analysis of randomized controlled trials. Ann Surg Oncol. 2013;20:4169-4182. 63. Khrizman P, Niland JC, ter Veer A, et al. Postoperative adjuvant chemotherapy use in patients with stage II/III rectal cancer treated with neoadjuvant therapy: a national comprehensive cancer network analysis. J Clin Oncol. 2013;31:30-38. 64. Haynes AB, You YN, Hu CY, et al. Postoperative chemotherapy use after neoadjuvant chemoradiotherapy for rectal cancer: analysis of Surveillance, Epidemiology, and End Results-Medicare data, 19982007. Cancer. 2014;120:1162-1170. 65. García-Albéniz X, Gallego R, Hofheinz RD, et al. Adjuvant therapy sparing in rectal cancer achieving complete response after chemoradiation. World J Gastroenterol. 2014;20:15820-15829. 66. Loree JM, Kennecke HF, Renouf DJ, et al. Effect of adjuvant chemotherapy on stage II rectal cancer outcomes after preoperative short-course radiotherapy. Clin Colorectal Cancer. 2016;15:352.e1-359.e1. 67. Bosset JF, Calais G, Mineur L, et al; EORTC Radiation Oncology Group. Fluorouracil-based adjuvant chemotherapy after preoperative chemoradiotherapy in rectal cancer: long-term results of the EORTC 22921 randomised study. Lancet Oncol. 2014;15:184-190. 68. Sainato A, Cernusco Luna Nunzia V, Valentini V, et al. No benefit of adjuvant fluorouracil leucovorin chemotherapy after neoadjuvant chemoradiotherapy in locally advanced cancer of the rectum (LARC): long term results of a randomized trial (I-CNR-RT). Radiother Oncol. 2014;113:223-229. 69. Breugom AJ, van Gijn W, Muller EW, et al; Cooperative Investigators of Dutch Colorectal Cancer Group and Nordic Gastrointestinal Tumour Adjuvant Therapy Group. Adjuvant chemotherapy for rectal cancer patients treated with preoperative (chemo)radiotherapy and total mesorectal excision: a Dutch Colorectal Cancer Group (DCCG) randomized phase III trial. Ann Oncol. 2015;26:696-701. 70. Breugom AJ, Swets M, Bosset JF, et al. Adjuvant chemotherapy after preoperative (chemo)radiotherapy and surgery for patients with rectal cancer: a systematic review and meta-analysis of individual patient data. Lancet Oncol. 2015;16:200-207. 71. Petrelli F, Coinu A, Lonati V, et al. A systematic review and metaanalysis of adjuvant chemotherapy after neoadjuvant treatment and surgery for rectal cancer. Int J Colorectal Dis. 2015;30:447-457. 72. Bujko K, Glimelius B, Valentini V, et al. Postoperative chemotherapy in patients with rectal cancer receiving preoperative radio(chemo) therapy: a meta-analysis of randomized trials comparing surgery ± a fluoropyrimidine and surgery + a fluoropyrimidine ± oxaliplatin. Eur J Surg Oncol. 2015;41:713-723. 73. Des Guetz G, Nicolas P, Perret GY, et al. Does delaying adjuvant chemotherapy after curative surgery for colorectal cancer impair survival? A meta-analysis. Eur J Cancer. 2010;46:1049-1055. 74. Deng Y, Chi P, Lan P, et al. Modified FOLFOX6 with or without radiation versus fluorouracil and leucovorin with radiation in neoadjuvant

244 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

treatment of locally advanced rectal cancer: initial results of the Chinese FOWARC multicenter, open-label, randomized three-arm phase III trial. J Clin Oncol. 2016;34:3300-3307. 75. Benson AB 3rd, Venook AP, Bekaii-Saab T, et al. Rectal cancer, version 2.2015. J Natl Compr Canc Netw. 2015;13:719-728, quiz 728. 76. Ciliberto D, Prati U, Roveda L, et al. Role of systemic chemotherapy in the management of resected or resectable colorectal liver metastases: a systematic review and meta-analysis of randomized controlled trials. Oncol Rep. 2012;27:1849-1856. 77. Wang ZM, Chen YY, Chen FF, et al. Peri-operative chemotherapy for patients with resectable colorectal hepatic metastasis: a metaanalysis. Eur J Surg Oncol. 2015;41:1197-1203. 78. Araujo R, Gonen M, Allen P, et al. Comparison between perioperative and postoperative chemotherapy after potentially curative hepatic resection for metastatic colorectal cancer. Ann Surg Oncol. 2013;20:4312-4321. 79. Khoo E, O’Neill S, Brown E, et al. Systematic review of systemic adjuvant, neoadjuvant and perioperative chemotherapy for resectable colorectal-liver metastases. HPB (Oxford). 2016;18:485-493. 80. Nordlinger B, Sorbye H, Glimelius B, et al; EORTC Gastro-Intestinal Tract Cancer Group; Cancer Research UK; Arbeitsgruppe Lebermetastasen und–tumoren in der Chirurgischen Arbeitsgemeinschaft Onkologie (ALMCAO); Australasian Gastro-Intestinal Trials Group (AGITG); Fédération Francophone de Cancérologie Digestive (FFCD). Perioperative FOLFOX4 chemotherapy and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC 40983): long-term results of a randomised, controlled, phase 3 trial. Lancet Oncol. 2013;14:1208-1215. 81. Dienstmann R, Salazar R, Tabernero J. Personalizing colon cancer adjuvant therapy: selecting optimal treatments for individual patients. J Clin Oncol. 2015;33:1787-1796. 82. Sargent DJ, Marsoni S, Monges G, et al. Defective mismatch repair as a predictive marker for lack of efficacy of fluorouracil-based adjuvant therapy in colon cancer. J Clin Oncol. 2010;28:3219-3226. 83. Gavin PG, Colangelo LH, Fumagalli D, et al. Mutation profiling and microsatellite instability in stage II and III colon cancer: an assessment of their prognostic and oxaliplatin predictive value. Clin Cancer Res. 2012;18:6531-6541. 84. Sinicrope FA, Mahoney MR, Smyrk TC, et al. Prognostic impact of deficient DNA mismatch repair in patients with stage III colon cancer from a randomized trial of FOLFOX-based adjuvant chemotherapy. J Clin Oncol. 2013;31:3664-3672. 85. Missiaglia E, Jacobs B, D'Ario G, et al. Distal and proximal colon cancers differ in terms of molecular, pathological, and clinical features. Ann Oncol. 2014;25:1995-2001. 86. Mlecnik B, Bindea G, Kirilovsky A, et al. The tumor microenvironment and Immunoscore are critical determinants of dissemination to distant metastasis. Sci Transl Med. 2016;8:327ra26. 87. Galon J, Mlecnik B, Marliot F, et al. Validation of the Immunoscore (IM) as a prognostic marker in stage I/II/III colon cancer: results of a worldwide consortium-based analysis of 1,336 patients. J Clin Oncol. 2016;34 (suppl; abstr 3500). 88. Giannakis M, Mu XJ, Shukla SA, et al. Genomic correlates of immunecell infiltrates in colorectal carcinoma. Cell Rep. 2016;15:857-865. 89. Mlecnik B, Bindea G, Angell HK, et al. Integrative analyses of colorectal cancer show immunoscore is a stronger predictor of patient survival than microsatellite instability. Immunity. 2016;44:698-711.

PERSONALIZING ADJUVANT THERAPY FOR STAGE II/III COLORECTAL CANCER

90. Guinney J, Dienstmann R, Wang X, et al. The consensus molecular subtypes of colorectal cancer. Nat Med. 2015;21:1350-1356. 91. Dienstmann R, Vermeulen L, Guinney J, et al. Consensus molecular subtypes and the evolution of precision medicine in colorectal cancer. Nat Rev Cancer. 2017;17:79-92. 92. Angelova M, Charoentong P, Hackl H, et al. Characterization of the immunophenotypes and antigenomes of colorectal cancers reveals distinct tumor escape mechanisms and novel targets for immunotherapy. Genome Biol. 2015;16:64. 93. Becht E, de Reynies A, Giraldo NA, et al. Immune and stromal classification of colorectal cancer is associated with molecular subtypes and relevant for precision immunotherapy. Clin Cancer Res. 2016;22:4057-4066. 94. Straussman R, Morikawa T, Shee K, et al. Tumour micro-environment elicits innate resistance to RAF inhibitors through HGF secretion. Nature. 2012;487:500-504. 95. Roepman P, Schlicker A, Tabernero J, et al. Colorectal cancer intrinsic subtypes predict chemotherapy benefit, deficient mismatch repair and epithelial-to-mesenchymal transition. Int J Cancer. 2014;134:552562.

98. Gray RG, Quirke P, Handley K, et al. Validation study of a quantitative multigene reverse transcriptase-polymerase chain reaction assay for assessment of recurrence risk in patients with stage II colon cancer. J Clin Oncol. 2011;29:4611-4619. 99. Yothers G, O’Connell MJ, Lee M, et al. Validation of the 12-gene colon cancer recurrence score in NSABP C-07 as a predictor of recurrence in patients with stage II and III colon cancer treated with fluorouracil and leucovorin (FU/LV) and FU/LV plus oxaliplatin. J Clin Oncol. 2013;31:4512-4519. 100. Di Narzo AF, Tejpar S, Rossi S, et al. Test of four colon cancer risk-scores in formalin fixed paraffin embedded microarray gene expression data. J Natl Cancer Inst. 2014;106:dju247. 101. Srivastava G, Renfro LA, Behrens RJ, et al. Prospective multicenter study of the impact of oncotype DX colon cancer assay results on treatment recommendations in stage II colon cancer patients. Oncologist. 2014;19:492-497. 102. De Sousa E Melo F, Wang X, Jansen M, et al. Poor-prognosis colon cancer is defined by a molecularly distinct subtype and develops from serrated precursor lesions. Nat Med. 2013;19:614-618.

96. Song N, Pogue-Geile KL, Gavin PG, et al. Clinical outcome from oxaliplatin treatment in stage II/III colon cancer according to intrinsic subtypes: secondary analysis of NSABP C-07/NRG Oncology Randomized Clinical Trial. JAMA Oncol. 2016;2:1162-1169.

103. Albain KS, Barlow WE, Shak S, et al; Breast Cancer Intergroup of North America. Prognostic and predictive value of the 21-gene recurrence score assay in postmenopausal women with node-positive, oestrogenreceptor-positive breast cancer on chemotherapy: a retrospective analysis of a randomised trial. Lancet Oncol. 2010;11:55-65.

97. Kopetz S, Tabernero J, Rosenberg R, et al. Genomic classifier ColoPrint predicts recurrence in stage II colorectal cancer patients more accurately than clinical factors. Oncologist. 2015;20:127-133.

104. Trinh A, Trumpi K, De Sousa EMF, et al. Practical and robust identification of molecular subtypes in colorectal cancer by immunohistochemistry. Clin Cancer Res. 2017;23:387-398.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 245

ATREYA, YAEGER, AND CHU

Systemic Therapy for Metastatic Colorectal Cancer: From Current Standards to Future Molecular Targeted Approaches Chloe E. Atreya, MD, PhD, Rona Yaeger, MD, and Edward Chu, MD OVERVIEW Over the past 20 years, substantial advances have been made in the treatment of patients with metastatic colorectal cancer (mCRC). In particular, there is now a wide range of options for the front-line treatment of mCRC. Sophisticated molecular technologies have been developed to identify novel prognostic and predictive biomarkers for CRC. DNA sequencing technology has made remarkable advances in recent years, primarily as a result of the development of next-generation sequencing and whole exome sequencing, which are powerful new tools for the discovery of predictive molecular biomarkers to facilitate the delivery of personalized medicine. In addition to tumor tissue, recent efforts have focused on analyzing circulating tumor DNA in peripheral blood. Herein, we review the evolution of standard chemotherapy and targeted therapy strategies for the treatment of mCRC in the front-line setting, the molecular technologies that are presently being used to facilitate our ability to practice individualized medicine, and the practical aspects of applying molecular biomarkers to everyday clinical practice.

RC remains a major public health problem in the United States and worldwide. In 2017, there will be an estimated 135,000 new cases diagnosed in the United States.1 CRC is the second leading cause of cancer deaths, with an estimated 50,000 deaths each year. Approximately 20% of newly diagnosed CRC is metastatic at the time of initial presentation. Perhaps more importantly, up to 50% of patients who initially present with early-stage CRC will eventually be diagnosed with metastatic disease. Substantial progress has been made in the treatment of mCRC during the last 2 decades, so that median overall survival (OS) is now in the 30-month range. For patients with mCRC, systemic chemotherapy has been the main treatment approach.2,3 For nearly 40 years, from the mid-1950s to 1996, the fluoropyrimidine 5-fluorouracil (5FU) was the only agent approved for the treatment of mCRC. However, since 1996 with the approval of the topoisomerase I inhibitor irinotecan, considerable advances have been made with the approval of several cytotoxic, biologic, and targeted agents by the U.S. Food and Drug Administration. In addition to irinotecan, the cytotoxic agents include oxaliplatin, a third-generation platinum analog, and two oral fluoropyrimidines, capecitabine and TAS-102. Bevacizumab, an anti-VEGF antibody, was approved in 2004, along with cetuximab, an anti-EGFR antibody. In 2006, panitumumab, an anti-EGFR antibody, was approved for use in the disease-refractory setting, and in 2012, the anti-VEGF recombinant fusion protein,

Ziv aflibercept, and regorafenib, a multikinase small molecule inhibitor, were approved.

OXALIPLATIN VERSUS IRINOTECAN

Four randomized clinical trials have directly compared the clinical efficacy of oxaliplatin-based chemotherapy with irinotecan-based chemotherapy in the first-line treatment setting.4-7 The most well-known was the GERCOR C97-3 study conducted by Tournigand et al4 in France, and this was the first large, randomized clinical trial investigating leucovorin plus 5-FU (46-hour infusion) and oxaliplatin (FOLFOX6) compared with leucovorin plus 5-FU and irinotecan (FOLFIRI) for the front-line treatment of mCRC. This study clearly documented the virtually identical clinical efficacy of FOLFOX6 and FOLFIRI chemotherapy with respect to overall response rate (ORR; 56% vs. 54%, respectively), median time to tumor progression (8.5 vs. 8.1 months), and median OS (20.6 vs. 21.5 months). Similar results were subsequently reported by the Gruppo Oncologico Dell’Italia Meridionale in Italy and the U.S. CALGB Cooperative Group (CALGB 80203).5,6 The Hellenic Oncology Group in Greece conducted a clinical study in which the bolus weekly schedule of 5-FU was used instead of an infusional schedule as the backbone fluoropyrimidine regimen in combination with irinotecan or oxaliplatin, and they also showed no difference between irinotecan and oxaliplatin with respect to clinical efficacy.7

From the Gastrointestinal Oncology Program, UCSF Helen Diller Family Comprehensive Cancer Center, San Francisco, CA; Memorial Sloan Kettering Cancer Center, New York, NY; University of Pittsburgh Cancer Institute, Pittsburgh, PA. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Edward Chu, MD, University of Pittsburgh Cancer Institute, 5150 Centre Ave., Fifth Floor, Room 571, Pittsburgh, PA 15232; email: [emailprotected]. © 2017 American Society of Clinical Oncology

246 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

SYSTEMIC THERAPY FOR METASTATIC COLORECTAL CANCER

ORAL VERSUS INTRAVENOUS 5-FU

FOLFOX Versus XELOX

The NO16966 randomized phase III study was initially designed as a two-arm, open-label study to compare the clinical efficacy of leucovorin plus 5-FU (22-hour infusion) and oxaliplatin (FOLFOX4) with oral fluoropyrimidine plus capecitabine and oxaliplatin (XELOX).8 When it was clear that bevacizumab was going to receive approval by the U.S. Food and Drug Administration, this study was subsequently amended to a two-by-two, placebo-controlled design with two of the arms including bevacizumab.9 In the initial cohort of 634 patients who were treated with only FOLFOX4 or XELOX, no differences were observed with respect to ORR, progression-free survival (PFS), and OS. Both combination regimens were relatively well-tolerated. However, FOLFOX4 was associated with more grade 3/4 neutropenia (44% vs. 7%), febrile neutropenia (4.8% vs. 0.9%), and grade 3/4 venous thromboembolic agents (6.3% vs. 3.8%) than XELOX. In contrast, XELOX was associated with an increased incidence of grade 3 diarrhea (19% vs. 11%) and grade 3 hand-foot syndrome (6% vs. 1%) compared with FOLFOX4. Of note, the rates of grade 3/4 neurotoxicity were similar between XELOX and FOLFOX4. Ducreux et al10 conducted a randomized study of XELOX compared with FOLFOX6 in the first-line treatment of mCRC, and the primary endpoint of this study was overall ORR. The secondary endpoints were PFS, OS, quality of life, and pharmacoeconomics. No differences were observed between the XELOX and FOLFOX6 arms in terms of the clinical efficacy endpoints of ORR (42% vs. 46%, respectively), PFS (8.8 vs. 9.3 months, respectively), and OS (19.9 vs. 20.5 months, respectively). However, as had been previously reported in

KEY POINTS • Anti-VEGF and anti-EGFR antibodies can be effectively used in combination with cytotoxic chemotherapy for the first-line treatment of metastatic colorectal cancer. • The sidedness of the primary tumor is an important factor in determining the potential role of anti-VEGF and anti-EGFR antibodies for the first-line treatment of metastatic colorectal cancer. Anti-EGFR antibody therapy should not be administered in patients with a right-sided primary tumor. • Genomic testing should be performed at the time of diagnosis of metastatic disease to evaluate for potential alterations in the KRAS, NRAS, and BRAF genes, which will guide patient selection for anti-EGFR antibody therapy, and to also inform decisions about potential curative-intent resection of metastases. • Induction chemotherapy with FOLFOXIRI, with or without bevacizumab, should be considered in patients with a BRAF mutation and good performance status. • Treatment with the immune checkpoint inhibitors nivolumab or pembrolizumab should be considered in patients with microsatellite unstable/mismatch repair defective tumors, as per the 2017 NCCN guideline (U.S. Food and Drug Administration approval is pending).

the N016966 study, the incidence of grade 3/4 myelosuppression was higher in patients treated with FOLFOX6 (47%) compared with XELOX (5%). Patients treated with XELOX experienced more grade 3/4 thrombocytopenia (12% vs. 5%) and diarrhea (14% vs. 7%) than those treated with FOLFOX6. In contrast to the NO16966 study, FOLFOX6 was associated with a higher incidence of neurotoxicity (26% vs. 11%) compared with XELOX.

FOLFIRI Versus XELIRI

A meta-analysis of six clinical studies was conducted by Guo et al11 to investigate the clinical efficacy of the oral capecitabine plus irinotecan (XELIRI) and FOLFIRI combination regimens in the first-line treatment of mCRC. No significant differences in clinical efficacy, as reflected in ORR, PFS, and OS, between XELIRI and FOLFIRI were identified. In terms of side effects, both treatment regimens were relatively well-tolerated with similar safety profiles. The FNCLCC ACCORD 13/053 study was a randomized phase II clinical trial that investigated the efficacy and safety of XELIRI and FOLFIRI as first-line therapy of mCRC.12 Patients were randomly assigned to receive XELIRI or FOLFIRI, and bevacizumab was included in both treatment arms. The 6-month PFS was 82% in the XELIRI arm and 85% in the FOLFIRI. In general, both XELIRI and FOLFIRI were well-tolerated with a manageable safety profile, and the most frequent toxicities were grade 3/4 neutropenia (18% vs. 26%, respectively) and grade 3 diarrhea (12% vs. 5%, respectively).

Triplet Cytotoxic Chemotherapy

In patients with good performance status and who are believed to be able to tolerate aggressive combination chemotherapy, it is clear that doublet chemotherapy has superior clinical efficacy over single-agent fluoropyrimidine chemotherapy, whether it is infusional 5-FU or oral capecitabine. The next issue to consider is whether triplet chemotherapy with all three of the active cytotoxic agents, 5-FU, oxaliplatin, and irinotecan, could provide improved clinical efficacy in the up-front treatment setting. To directly address this question, the Gruppe Oncologico Nord Ovest (GONO) of Italy conducted the first randomized phase III study to compare leucovorin plus 5-FU, oxaliplatin, and irinotecan (FOLFOXIRI) with FOLFIRI in the front-line setting.13 A total of 244 patients were randomly assigned, and Falcone et al reported improved ORR, PFS, and median OS in patients treated with the FOLFOXIRI triplet combination regimen when compared with FOLFIRI. Moreover, treatment with FOLFOXIRI compared with FOLFIRI resulted in improved R0 surgical resection rate for all patients (15% vs. 6%, respectively) and for those specific patients with liver-limited disease (36% vs. 12%, respectively). The triplet combination regimen was fairly well-tolerated, although there was an increase in grade 2/3 neurotoxicity (19% vs. 0%) and grade 3/4 neutropenia (50% vs. 28%) compared with FOLFIRI. However, the incidence of febrile neutropenia and grade 3/4 diarrhea was not significantly different between the two treatment arms. A final analysis was conducted after a median follow asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 247

ATREYA, YAEGER, AND CHU

up of 5 years that confirmed the superiority of the FOLFOXIRI regimen.14 In addition, a risk-stratified analysis based on the Köhne prognostic model was performed, which showed that FOLFOXIRI, when compared with FOLFIRI, was associated with improved PFS and OS in all risk subgroups. FOLFOXIRI chemotherapy is associated with improvements in PFS and OS; the absolute benefit in OS is 7% at 5 years. In a relatively fit patient population, this triplet combination regimen is feasible and is associated with a manageable safety profile. Moreover, in patients who are able to undergo R0 surgical resection of liver-limited disease, there does not appear to be an increase in liver toxicity. Finally, initial treatment with FOLFOXIRI does not appear to have a negative effect on the outcomes of patients who received subsequent treatment in the secondline setting.

ANTI-VEGF VERSUS ANTI-EGFR THERAPY IN COMBINATION WITH CYTOTOXIC CHEMOTHERAPY

PEAK was a phase II study in which modified FOLFOX6 (mFOLFOX6) was used as the chemotherapy backbone, and patients were randomly assigned to receive the anti-VEGF antibody bevacizumab or the anti-EGFR antibody panitumumab.15 A total of 285 patients were enrolled, and the primary endpoint of the study was PFS, with secondary endpoints of ORR, OS, and safety. Overall, ORR and median PFS were nearly identical between the arms treated with bevacizumab or panitumumab (54% vs. 58%, respectively, and 10.1 vs. 10.9 months, respectively). However, median OS was significantly improved in patients treated with panitumumab when compared with bevacizumab (34.2 vs. 24.3 months; hazard ratio [HR] 0.62; p = .009). When an extended RAS analysis was performed and the clinical data were re-analyzed, the improvement in OS was maintained with panitumumab compared with bevacizumab (41.3 vs. 28.9 months; HR 0.63). Moreover, median PFS in patients treated with panitumumab was found to be significantly improved by 4.5 months compared with bevacizumab (13 vs. 9.5 months; HR 0.65; p = .029). FIRE-3 was a study conducted in Germany in which patients with previously untreated mCRC with wild-type KRAS were treated with the FOLFIRI chemotherapy backbone, and patients were randomly assigned to receive either bevacizumab or cetuximab.16 The primary endpoint of this study was ORR, with secondary endpoints of PFS, OS, R0 surgical resection rate, and safety. In the original analysis of this study, ORR and PFS were virtually identical between the arms treated with bevacizumab or cetuximab (58% vs. 62%, respectively, and 10.3 vs. 10.0 months, respectively). However, a significant 3.7-month improvement in OS was observed in patients treated with cetuximab, which represented a 23% reduction in the risk of death (HR 0.77; p = .017). The safety profiles of the two arms of the study were as expected and manageable. This was the first direct head-to-head comparison of cetuximab and bevacizumab in the front-line treatment setting, and although ORR 248 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

and PFS were identical, cetuximab treatment resulted in a potentially clinically meaningful improvement in OS. At the time of the initial publication and the presentation of this work at the 2013 ASCO Annual Meeting, the potential effect of the improvement of the FOLFIRI plus cetuximab combination remained unclear given the somewhat limited information relating to duration of second-line and subsequent salvage therapies. In the United States, the CALGB/SWOG 80405 phase III randomized study compared the potential benefit of cetuximab and bevacizumab added to cytotoxic chemotherapy.17 In contrast to the FIRE-3 study, the primary endpoint of this study was OS, and patients could receive either FOLFOX or FOLFIRI as their cytotoxic regimen depending on physician preference. Of note, 74% of patients received FOLFOX whereas 26% received FOLFIRI. Overall, no significant differences were observed in PFS (10.4 vs. 10.8 months, respectively) and OS (29.9 vs. 29.0 months, respectively) in patients treated with cetuximab compared with bevacizumab. When the specific chemotherapy regimen was analyzed, no significant differences in OS were identified among patients treated with FOLFOX or FOLFIRI chemotherapy. This was the largest randomized study in the front-line setting conducted to date, and the take-home conclusions were that FOLFOX/ FOLFIRI in combination with either bevacizumab or cetuximab were effective treatment options and that an OS of approximately 30 months established a new benchmark for first-line treatment. Venook et al18 recently investigated the potential effect of primary tumor location on the clinical efficacy of patients treated on CALGB/SWOG 80405, and these findings were reported at the 2016 ASCO Annual Meeting. In a careful chart review, they determined that 68% of the primary tumors came from the left side of the colon or rectum and 27% of the tumors came from the right side. When OS was determined by sidedness of the primary tumor, there was an improvement in OS for patients with left-sided tumors compared with right-sided tumors (33.3 vs. 19.4 months, respectively), which was highly significant (p < .0001). For patients treated with bevacizumab, the improvement in OS was maintained in patients with left-sided tumors compared with right-sided tumors, albeit still higher for leftsided primary tumors (31.4 vs. 24.2 months, respectively). However, in patients treated with cetuximab, there was a striking 19.3-month difference in which OS was 36.0 months for left-sided tumors and only 16.7 months for right-sided tumors. These findings are important as they highlight the importance of sidedness as an important predictive marker and the role of sidedness in determining response to antiEGFR antibody therapy and has now been confirmed in other clinical studies.19,20 Patients with right-sided tumors clearly derive greater benefit from bevacizumab compared with cetuximab and, in fact, derive little benefit from cetuximab. In contrast, patients with left-sided tumors derive benefit from both cetuximab and bevacizumab, although it appears that the median OS is improved by nearly 5 months with cetuximab.

SYSTEMIC THERAPY FOR METASTATIC COLORECTAL CANCER

Triplet Cytotoxic Chemotherapy in Combination With Biologic Agents

TRIBE was a multicenter, phase III study conducted in 34 Italian oncology centers in which patients with mCRC were randomly assigned to receive FOLFOXIRI plus bevacizumab or FOLFIRI plus bevacizumab as first-line treatment.21,22 A total of 508 patients were enrolled in this study. The median PFS was 12.1 months with FOLFOXIRI plus bevacizumab compared with 9.7 months with FOLFIRI plus bevacizumab (HR 0.75; p = .003), and ORR was increased to 65% compared with 53% for patients treated with FOLFIRI plus bevacizumab. The median OS was 31 months in the FOLFOXIRI plus bevacizumab group compared with 25.8 months in patients treated with FOLFIRI plus bevacizumab, and this difference approached significance (HR 0.79; p = .054). Of note, the FOLFOXIRI/bevacizumab combination was equally effective in wild-type and mutant KRAS tumors, and although the numbers were small, this combination appeared to be especially active in the mutant BRAF subgroup. To date, only phase II studies have investigated FOLFOXIRI in combination with the anti-EGFR antibodies cetuximab and panitumumab. Saridaki et al23 conducted a pilot phase II study of FOLFOXIRI plus cetuximab, and they reported an ORR of 70%, with median time to progression of 10.2 months and median OS of 30.3 months. In addition, R0 surgical resection was performed in 37% of patients. Although this combination regimen was relatively well-tolerated, the incidence of grade 3/4 diarrhea was 53%. The GONO group in Italy investigated the combination of FOLFOXIRI plus panitumumab in patients with wild-type RAS and BRAF mCRC, and they reported an 89% ORR and a median PFS of 11.3 months.24 In addition, R0 surgical resection was achieved in 35% of the treated patients. Once the 5-FU infusion dose was reduced from 3,000 mg/m2 to 2,400 mg/m2 after two of the first three patients experienced grade 3/4 diarrhea, this combination regimen was found to be well-tolerated, and the most common grade 3/4 toxicities were neutropenia (48%), diarrhea (35%), asthenia (27%), mucositis (14%), and skin conditions (14%).

CURRENT TREATMENT OPTIONS FOR MCRC

In 2017, there is now a wide range of treatment options for the first-line therapy of mCRC.25 The current standard of care for first-line treatment is combination cytotoxic chemotherapy using the fluoropyrimidine backbone (5-FU or capecitabine) with either oxaliplatin (FOLFOX or XELOX) or irinotecan (FOLFIRI or XELIRI) in combination with the antiVEGF agent bevacizumab or anti-EGFR agents (cetuximab or panitumumab) for patients with wild-type RAS and BRAF. The choice between bevacizumab and anti-EGFR agents for the first-line setting in patients with wild-type RAS depends on clinical presentation and individual patient factors. Recent studies suggest that the sidedness of the primary tumor is an important predictive biomarker and that patients who present with right-sided primary tumors derive little benefit from anti-EGFR therapy. Patients with left-sided

primary tumors derive clinical benefit from either bevacizumab or cetuximab, although it appears that the benefit may be greater with cetuximab.

MOLECULAR PROFILING OF CRC: WHAT, WHEN, AND HOW

Genomic analysis of mCRC provides both prognostic and predictive information. Genomic analysis should be performed in all patients with mCRC, including patients with limited tumor burden who are being evaluated for hepatectomy or other metastasectomy. The National Comprehensive Cancer Network (NCCN) guideline recommends genomic testing at the time of diagnosis of metastatic disease.25 An array of sequencing tests, which differ in the number of genes analyzed, depth of sequencing, and evaluation of mutations and/or copy number alterations, are currently available. For clinical care, genomic analysis should evaluate for the presence of activating mutations in the KRAS, NRAS, and BRAF genes to guide the selection of patients for the anti-EGFR therapies, cetuximab or panitumumab. The KRAS, NRAS, and BRAF are oncogenes that encode proteins involved in the classic mitogen-activated protein kinase (MAPK) pathway that regulates cell proliferation and survival, and mutations in these genes are found in about 45%, 4%, and 8% of mCRC, respectively.26 Activating mutations in these genes occur as an early event in colorectal tumorigenesis27 in defined hotspots, and KRAS-, NRAS-, and BRAF-activating mutations are nearly universally exclusive.28 In mCRC, either the primary tumor or metastasis can be analyzed. Studies of paired tumors suggest near complete concordance for mutations in KRAS, NRAS, and BRAF between colorectal primaries and metastases.29,30 Genomic alterations in the KRAS, NRAS, and BRAF genes provide predictive information for the selection of patients for the anti-EGFR antibodies cetuximab or panitumumab. Activating mutations in these genes result in constitutively activated proteins, whose activation does not require upstream signaling, such as through EGFR, and whose activation leads to negative feedback loops that limit EGFR activation. Multiple clinical trials indicate that colorectal tumors harboring activating KRAS or NRAS mutations do not benefit from anti-EGFR therapies and may actually experience accelerated growth with use of these drugs.26,31,32 Thus, prior to the consideration for treatment with the anti-EGFR antibodies, RAS mutation testing should be performed to analyze for mutations, both at the more common sites in exon 2 (codons 12 and 13) and outside exon 2 (exon 3 at codons 59 and 61, and exon 4 at codons 117 and 146). In colorectal tumors harboring BRAF V600E mutations, increasing evidence through subset analysis of clinical trial and retrospective data also suggest lack of response to EGFR inhibitors, both as single agents and in combination with chemotherapy. In addition, two meta-analyses found no PFS benefit for EGFR inhibitors in BRAF-mutant mCRC.33,34 These data are limited by the overall low frequency of BRAF V600E mutation and the aggressive nature of mCRC with this mutation (see below). asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 249

ATREYA, YAEGER, AND CHU

Genotyping the KRAS, NRAS, and BRAF genes also provides important prognostic information. Several studies have evaluated the effect of KRAS mutations on the outcomes of patients with mCRC. Although early results were inconsistent, as some studies found no effect and others found harm associated with KRAS mutations, recent data increasingly suggest that the presence of a KRAS mutation is associated with worse outcomes. A retrospective analysis of 918 patients with mCRC at Memorial Sloan Kettering Cancer Center identified an HR of 1.6 (95% CI, 1.29–1.90; p < .001) for OS with the presence of a RAS mutation on multivariate analysis adjusting for age at diagnosis of metastatic disease, gender, location of primary tumor, synchronous or metachronous disease, and occurrence of metastasectomy or hepatic arterial infusion treatment treated as a timedependent covariates.35 Few series have looked at outcomes in NRAS-mutant mCRC. An updated analysis from Memorial Sloan Kettering Cancer Center found that, compared with RAS–wild-type mCRC, NRAS-mutant and KRAS-mutant mCRC had an HR of 2.0 (95% CI, 1.3–2.8; p < .01) and 1.5 (95% CI, 1.2–1.8; p < .01) for OS, respectively, on multivariate analysis (Rona Yaegar, unpublished data). The presence of the BRAF V600E mutation is a strong negative prognostic marker in mCRC. Data from the randomized phase III Medical Research Council COIN trial in mCRC, for example, identified an OS of 8.8 months for patients with BRAF-mutant mCRC, 14.4 months for KRAS exon 2–mutant mCRC, and 20.1 months for KRAS exon 2–wild-type mCRC.36 Similarly, many series have reported median OS of less than a year for BRAF-mutant mCRC.26,37 The presence of a BRAF mutation in mCRC has been associated with T4 primary tumors, poor tumor differentiation, and peritoneal metastasis.37-40 The presence of mutations in the RAS and BRAF genes may also influence recurrence risk after metastasectomy, and patients with RAS-mutated or BRAF-mutated tumors should be carefully selected for surgical treatment with curative intent. In patients undergoing hepatectomy for colorectal liver metastases, the presence of a RAS mutation, in comparison with wild-type the wild-type variant, was associated with significantly worse recurrence-free survival (RFS) and OS, with similar liver RFS at 3 years, but lower lung RFS at 3 years.41 In RAS-mutant mCRC, there is a high risk of recurrence as well as shorter survival after hepatectomy in node-positive primary tumors, larger tumors, and after more than 7 cycles of preoperative chemotherapy. The presence of a BRAF mutation is associated with a high risk of recurrence after metastasectomy of liver, lung, or peritoneal disease, and patients with BRAF-mutant mCRC experience shorter survival after metastasectomy compared with patients with BRAF–wild-type disease.42 Patients with BRAF-mutant mCRC should be carefully selected and informed of the increased risk of recurrence before metastasectomy; and, unless metastatic disease is truly limited, systemic clinical trial options (see below) should be considered rather than aggressive surgical debulking. Prior to an appreciation for the need to perform extended RAS genotyping, clinical testing should focus on hotspot 250 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

alterations in exon 2 of KRAS. Genotyping of codons 12 and 13 in exon 2 of KRAS was often done with Sanger sequencing or real-time polymerase chain reaction (PCR).43 Sanger sequencing is limited by low sensitivity with a limit of detection of about 20% and is laborious. Real-time PCR is more sensitive, but requires unique primers for each possible mutation. PCR-based assays have been expanded to cover all hotspots in RAS and can be used for extended RAS testing. However, with the decreased cost of sequencing, many groups have shifted to multiplexed genotyping platforms that include more genes. A mass spectrometry based multiplatform assay can detect alterations with a sensitivity limit of detection of about 5%. For this assay, the target DNA is amplified, a single base extension is performed, and then the small DNA products with unique mass value according to the mutation generated are measured by a mass spectrometer. These assays can analyze several genes in a single sample and genotype multiple hotspots, but require prior knowledge of all potential mutation sites of interest. In recent years, target enrichment by hybrid capture has allowed next-generation sequencing (NGS) of subsets of the genome of clinical interest, as done by FoundationOne, Caris Life Sciences, and Tempus.44 NGS assays are highly sensitive, can analyze a large panel of genes, and detect novel mutations, small insertions and deletions (indels), copy number alterations, and select gene fusions and rearrangements from small amounts of DNA. In addition to providing predictive and prognostic information, when using a multigene NGS panel to perform molecular profiling of mCRC, the number of mutated genes identified can be used to discriminate between somatic mCRC and microsatellite-unstable mCRC.45 Microsatellite instability–high (MSI-H) mCRC results from mutations in the mismatch repair (MMR) genes that cause a malfunctioning gene product or from promoter methylation causing epigenetic silencing of MMR protein expression and is diagnosed with a PCR assay to identify changes in length of dinucleotide/trinucleotide repeats or with immunohistochemical analysis for retained expression of the MMR proteins (MLH1, MSH2, MSH6, PMS2). MSI-H tumors exhibit a higher mutation burden. Using a small number of test cases with known MSI status, a mutation number cutoff to discriminate between microsatellite-stable (MSS) and MSI-H cases can be identified.46 The mutation number cutoff varies by the assay and the number of genes analyzed. Information about MSI status of mCRC now has important clinical implications, and the updated 2017 NCCN guideline recommends that all mCRC be evaluated for MMR deficiency, and the anti-PD1 antibodies, pembrolizumab and nivolumab, have been added as treatment options for patients with unresectable MSI-H or MMR-deficient mCRC.25 Use of a single assay may provide a cost-effective option to both evaluate RAS and BRAF mutation status and screen for microsatellite instability in mCRC. Because activating mutations in the KRAS, NRAS, and BRAF genes are nearly universally exclusive, some groups use a hier archical system of genotyping to improve cost-effectiveness

SYSTEMIC THERAPY FOR METASTATIC COLORECTAL CANCER

of genomic analysis. Genomic alterations are analyzed sequentially—usually first in KRAS exon 2, then extended KRAS testing, BRAF V600E, and finally NRAS testing based on frequency of these alterations—with the analysis halted if a mutation is detected. In tumors that are wild-type at all three genes, it may be worth considering alternative driver alterations that may impact response to EGFR inhibition and may be targetable. ERBB2 amplification and MAP2K1 mutations appear to predict resistance to EGFR inhibition and may be targetable.46-49 NTRK fusions, although very rare in mCRC, are potentially targetable and early clinical trials with novel agents that target NTRK alterations have reported promising initial results.50 Cell-free fragments of DNA are shed into the bloodstream by cells undergoing apoptosis and necrosis, and the circulating free DNA (cfDNA) released by tumor cells can be collected and amplified to look for somatic mutations. Looking to the future, cfDNA may represent a novel method to perform molecular profiling in mCRC, obviating the need to obtain tumor tissue and potentially allowing for a noninvasive method to perform genomic analysis at multiple time points during therapy, evaluating for clonal evolution, as well as molecular mechanisms of response and resistance. Both focused analysis for several genes and multigene NGS assays have been applied to detect somatic alterations within circulating tumor DNA in mCRC.51 Genomic analysis of cfDNA has been done most often in patients with response to targeted therapy who then acquire resistance.52,53 The use of cfDNA to characterize KRAS, NRAS, and BRAF mutation status and guide the use of EGFR inhibitor treatment is not standard in mCRC and may be limited by the lower sensitivity of the assay.

mCRC tumors. In combination, however, the ORR to cobimetinib plus atezolizumab was 20% in 20 KRAS-mutant tumors, with the majority of responses or disease stabilization persisting for more than 6 months. It is now hypothesized that MEK inhibitor–mediated intratumoral T-cell infiltration and MHC I upregulation is a RAS mutation–independent effect. For this reason, the follow-up phase III trial (NCT02788279) allows up to half of patients to have RAS wild-type tumors. If a tumor is found to be RAS wild-type, additional biomarker testing should be considered to try to identify potentially actionable targets. It should also be noted that as per the 2017 NCCN guideline, cetuximab and panitumumab therapy are only recommended for left-sided tumors, further refining the appropriate patient population for anti-EGFR antibodies.25

MSI/MMR

The anti–PD-1 antibodies, pembrolizumab or nivolumab, have been included as treatment options for patients with unresectable MSI-H or MMR-deficient CRC in the recently updated 2017 NCCN guideline25; although, as of this writing, neither agent is U.S. Food and Drug Administration–approved for mCRC. The data to support the NCCN recommendation come from the interim results of KEYNOTE-016, a phase II study of pembrolizumab in MSI-H tumors (NCT01876511), and CheckMate 142, a study of nivolumab or nivolumab combinations in recurrent or mCRC (NCT02060188).56-58 Given the limitations of cross-study comparisons, these results are summarized in Table 1. With immunotherapy, time to response may be long, such that response rates tend to increase as the data matures. Whereas primary progression occurs quickly, responses and stable disease are impressively durable. KEYNOTE-164 is a phase II trial of pembrolizumab as monotherapy in patients with previously-treated MSI-H/ MMR-deficient mCRC, and this study is now closed to accrual after reaching the planned enrollment of 120 patients.60 For newly identified patients with MSI-H/MMRdeficient mCRC, KEYNOTE-177 is an ongoing phase III study of first-line pembrolizumab compared with standard of care chemotherapy with a planned enrollment of 270 patients (NCT02563002).61 In response to those patients with MSS mCRC who ask about immunotherapy, none of the 25 patients with MMR-proficient mCRC who received pembrolizumab during the phase II study achieved objective responses (median

TRANSLATING BIOMARKERS INTO CLINICAL PRACTICE

RAS Mutations

If a mutation is detected in KRAS or NRAS at codons 12, 13, 61, 117, or 146, the recommendation is for cetuximab or panitumumab to not be considered as potential treatment options.54 RAS remains an elusive target, although many early phase trial strategies are aimed at targeting RAS and/or its downstream effectors. One study that has generated a great deal of interest is the phase 1B study of the MEK inhibitor, cobimetinib, combined with the anti-PD-L1 agent, atezolizumab.55 Based on prior clinical data, neither drug along would be expected to produce responses in KRAS-mutated, MSS

TABLE 1. MSI-H/dMMR CRC: Interim Results of Anti–PD-1 Antibody Trials Therapy Pembrolizumab

Nivolumab Nivolumab + ipilimumab

No. of Patients

Confirmed ORR, %

Median DOR (Months)

Median PFS (Months)

Median OS (Months)

9.6

Reference

*Median follow-up of 9.3 months. **At least 12 weeks of follow-up. Abbreviations: MSI-H, microsatellite instability high; CRC, colorectal cancer; ORR, objective response rate; DOR, duration of response; PFS, progression-free survival; OS, overall survival; NR, not reached.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 251

ATREYA, YAEGER, AND CHU

TABLE 2. BRAF-Mutated Metastatic Colorectal Cancer: Phase II Trial Results Therapy

No. of Patients

Confirmed ORR, %

Median DOR (Months)

Median PFS (Months)

Median OS (Months)

Vemurafenib (V)

2.1

7.7

V + cetuximab (C)

3.7

7.1

C + irinotecan (I)

V+C+I

54*

4.4

Dabrafenib (D) + trametinib (T)

3.5

8.7

D + panitumumab (P)

6.9

3.5

13.2

T+P

2.6

8.2

D+T+P

7.6, estimate

4.4

9.1

Encorafenib (E) + C

4.6

4.2

12.4

E + C + alpelisib

9.9

5.4

13.1

Reference

*Patients were randomly assigned to treatment. Abbreviations: ORR; objective response rate; DOR, duration of response; PFS, progression-free survival; OS, overall survival; NA, not available; NR, not relevant.

PFS, 2.4 months); comparable results were observed for the 20 patients with MSS mCRC treated with nivolumab with or without ipilimumab.56,57 Research efforts are underway to identify a subset of patients with immune-infiltrated MSS tumors who may benefit from checkpoint blockade.62 In addition, several combination strategies are being developed to prime MSS tumors for immunotherapy by first inducing intratumoral T-cell accumulation, including the cobimetinib plus atezolizumab trial discussed above.

BRAF Mutations

The presence of a BRAF V600E mutation predicts aggressive disease biology and poor response to standard therapies including anti-EGFR antibodies.26,37,63-66 However, the potential treatment implications of favorable prognosis BRAF mutations at codons 594 and 596 are less clear.67,68 With a BRAF V600 mutation, the ability to achieve disease control with first-line therapy may be a critical determinant of survival outcomes.69 In patients with good performance status, induction therapy with FOLFOXIRI should be considered.70 The TRIBE study showed that the median OS of 53 patients with BRAF-mutated CRC was 19 months when treated with first-line FOLFOXIRI plus bevacizumab compared with 10.7 months with FOLFIRI plus bevacizumab (HR 0.54; 95% CI, 0.24–1.20).22 Additionally, approximately a quarter of BRAF V600E–mutated mCRCs also exhibit MMR/MSI. Responses to checkpoint inhibitor therapy have been observed in the presence of a BRAF mutation and MMR/MSI.58,71 As such, immunotherapy options, such as enrollment on the KEYNOTE-177 trial (NCT02563002) and/or other immunotherapy-based clinical studies should be considered in this setting. Several studies have investigated the role of BRAF inhibitors as single-agents and in various combination strategies, for the treatment of BRAF V600E–mutated mCRC72-79 (Table 2). What we have learned thus far is that BRAF inhibitor monotherapy is ineffective, although inclusion of a BRAF inhibitor in a combination strategy appears to be important, and 252 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

three-drug therapies are somewhat more active than doublets. In the first randomized comparison with a standard of care regimen, the S1406 cooperative group study found that the addition of vemurafenib to irinotecan plus cetuximab significantly improved PFS (HR 0.42; 95% CI, 0.26–0.66; p < .001).76 Some patients clearly benefit from targeted inhibitor strategies, although disease stabilization is more common than radiographic response. Early emergence of drug resistance remains a challenge. For newly identified patients with BRAF V600E-mutated mCRC, an ongoing phase III trial that is expected to enroll 645 patients, BEACON CRC, is testing the combination of binimetinib (MEK inhibitor) with encorafenib and cetuximab, as compared with standard therapy (NCT02928224).

HER2 Amplification/Overexpression

The presence of HER2 amplification or a HER2-activating mutation may predict resistance to anti-EGFR antibodies. However, the current NCCN guideline does not yet discourage the use of cetuximab or panitumumab.25,48,80-82 Several HER2-directed treatment strategies have been, or are being, evaluated in mCRC.49,83-85 HERACLES (HER2 Amplification for Colorectal Cancer Enhanced Stratification) was the first large, phase II clinical trial to evaluate the tyrosine kinase inhibitor lapatinib in combination with the HER2-targeted monoclonal antibody trastuzumab in patients with HER2-amplified mCRC. HER2 positivity was defined as 2+/3+ by immunohistochemistry or fluorescence in situ hybridization–positive. The HERACLES study found that 27 patients with HER2-amplified mCRC refractory to standard therapy, including anti-EGFR antibodies, had a 30% response rate to lapatinib plus trastuzumab, with an 8.9-month median duration of response and a median PFS of 4.9 months.49 A follow-up study, HERACLES-RESCUE, is evaluating T-DM1 after progression during trastuzumab and lapatinib treatment is underway.86 T-DM1 is an antibody-drug conjugate whereby trastuzumab (the T portion) is connected via a stable

SYSTEMIC THERAPY FOR METASTATIC COLORECTAL CANCER

thioether linker to emtansine (the DM1 portion), a potent microtubule chemotherapy agent. Once trastuzumab binds to HER2-expressing cells, the linker is broken down, releasing DM1 intracellularly. Considerations for patients with newly identified HER2 overexpression include ongoing basket trials. NCI-Molecular Analysis for Therapy Choice (NCI-MATCH, NCT02465060) is testing T-DM1 in HER2-amplified cancers and afatinib, an irreversible tyrosine kinase inhibitor that targets HER2, EGFR, and HER4, in HER2-mutated cancers. The MyPathway study is testing the combination of trastuzumab with pertuzumab, a monoclonal antibody that inhibits HER2 hom*odimerization and heterodimerization with other HER family members, in HER2-amplified and mutated tumors (NCT02091141).85 The interim efficacy data from 34 patients enrolled in MyPathway is similar to what has been reported with the HERACLES study: 38% ORR, with a 10.3-month median duration of response, and a median PFS of 4.6 months. Notably, none of the nine patients with mutant KRAS and HER2amplified/overexpressed mCRC responded to trastuzumab plus pertuzumab.85,86

Emerging Biomarkers

The aforementioned basket trials and several smaller studies contain rational therapeutic options for a large number

of genetic aberrations found in CRC, including EGFR, AKT, PIK3CA, and MAP2K1 mutations, as well as MET and FGFR amplification. Additionally, PARP inhibitors may be effective for BRCA1/2-mutated CRCs. Other emerging CRC-relevant DNA damage response targets include ataxia-telangiectasia–mutated (ATM) and ataxia-telangiectasia and Rad3-related (ATR).87 Oncogenic fusions are of particular interest, as their targeting has led to exquisite responses in other cancer types. As proof-of-concept, patients with mCRC with a NTRK1 and an ALK gene rearrangement were treated with entrectinib, a selective pan-TRK, ROS1, and ALK inhibitor.50,88 Both patients experienced partial responses. A basket study of entrectinib for the treatment of patients with NTRK, ROS1, or ALK fusions is ongoing (NCT02568267). For patients with high RSPO3 gene expression, which may arise from translocations of RSPO3 and PTPRK, a trial of the anti-RSPO3 antibody, OMP-131R10, is in progress (NCT02482441). In 2017, KRAS, NRAS, MSI/MMR, and BRAF are the main molecular markers that currently influence standard-ofcare practice. However, enrollment of patients with these and other potentially actionable biomarkers in clinical trials holds promise for making increasingly personalized and effective treatment options available to future patients with mCRC.

References 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2017. CA Cancer J Clin. 2017;67:7-30. 2. Cremolini C, Schirripa M, Antoniotti C, et al. First-line chemotherapy for mCRC—a review and evidence-based algorithm. Nat Rev Clin Oncol. 2015;12:607-619. 3. Weinberg BA, Marshall JL, Hartley M, et al. A paradigm shift from onesize-fits-all to tailor-made therapy for metastatic colorectal cancer. Clin Adv Hematol Oncol. 2016;14:116-128. 4. Tournigand C, André T, Achille E, et al. FOLFIRI followed by FOLFOX6 or the reverse sequence in advanced colorectal cancer: a randomized GERCOR study. J Clin Oncol. 2004;22:229-237. 5. Colucci G, Gebbia V, Paoletti G, et al; Gruppo Oncologico Dell’Italia Meridionale. Phase III randomized trial of FOLFIRI versus FOLFOX4 in the treatment of advanced colorectal cancer: a multicenter study of the Gruppo Oncologico Dell’Italia Meridionale. J Clin Oncol. 2005;23:4866-4875. 6. Venook A, Niedzwiecki D, Hollis D, et al. Phase III study of with untreated metastatic adenocarcinoma of the colon or rectum (MCRC): CALGB 80203 preliminary results. J Clin Oncol. 2006;24:18S (suppl; abstr 3509). 7. Kalofonos HP, Aravantinos G, Kosmidis P, et al. Irinotecan or oxaliplatin combined with leucovorin and 5-fluorouracil as first-line treatment in advanced colorectal cancer: a multicenter, randomized, phase II study. Ann Oncol. 2005;16:869-877. 8. Cassidy J, Clarke S, Díaz-Rubio E, et al. Randomized phase III study of capecitabine plus oxaliplatin compared with fluorouracil/folinic acid plus oxaliplatin as first-line therapy for metastatic colorectal cancer. J Clin Oncol. 2008;26:2006-2012.

9. Saltz LB, Clarke S, Díaz-Rubio E, et al. Bevacizumab in combination with oxaliplatin-based chemotherapy as first-line therapy in metastatic colorectal cancer: a randomized phase III study. J Clin Oncol. 2008;26:2013-2019. 10. Ducreux M, Bennouna J, Hebbar M, et al; GI Group of the French Anti-Cancer Centers. Capecitabine plus oxaliplatin (XELOX) versus 5-fluorouracil/leucovorin plus oxaliplatin (FOLFOX-6) as first-line treatment for metastatic colorectal cancer. Int J Cancer. 2011;128:682690. 11. Guo Y, Shi M, Shen X, et al. Capecitabine plus irinotecan versus 5-FU/ leucovorin plus irinotecan in the treatment of colorectal cancer: a meta-analysis. Clin Colorectal Cancer. 2014;13:110-118. 12. Ducreux M, Adenis A, Pignon JP, et al. Efficacy and safety of bevacizumab-based combination regimens in patients with previously untreated metastatic colorectal cancer: final results from a randomised phase II study of bevacizumab plus 5-fluorouracil, leucovorin plus irinotecan versus bevacizumab plus capecitabine plus irinotecan (FNCLCC ACCORD 13/0503 study). Eur J Cancer. 2013; 49:1236-1245. 13. Falcone A, Ricci S, Brunetti I, et al; Gruppo Oncologico Nord Ovest. Phase III trial of infusional fluorouracil, leucovorin, oxaliplatin, and irinotecan (FOLFOXIRI) compared with infusional fluorouracil, leucovorin, and irinotecan (FOLFIRI) as first-line treatment for metastatic colorectal cancer: the Gruppo Oncologico Nord Ovest. J Clin Oncol. 2007;25:1670-1676. 14. Masi G, Vasile E, Loupakis F, et al. Randomized trial of two induction chemotherapy regimens in metastatic colorectal cancer: an updated analysis. J Natl Cancer Inst. 2011;103:21-30.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 253

ATREYA, YAEGER, AND CHU

15. Schwartzberg LS, Rivera F, Karthaus M, et al. PEAK: a randomized, multicenter phase II study of panitumumab plus modified fluorouracil, leucovorin, and oxaliplatin (mFOLFOX6) or bevacizumab plus mFOLFOX6 in patients with previously untreated, unresectable, wild-type KRAS exon 2 metastatic colorectal cancer. J Clin Oncol. 2014;32:2240-2247. 16. Heinemann V, von Weikersthal LF, Decker T, et al. FOLFIRI plus cetuximab versus FOLFIRI plus bevacizumab as first-line treatment for patients with metastatic colorectal cancer (FIRE-3): a randomised, open-label, phase 3 trial. Lancet Oncol. 2014;15:1065-1075. 17. Venook AP, Niedzwiecki D, Lenz HJ, et al. CALGB/SWOG 80405: phase III trial of irinotecan/5-FU/leucovorin (FOLFIRI) or oxaliplatin/5-FU/ leucovorin (mFOLFOX6) with bevacizumab (BV) or cetuximab (CET) for patients (pts) with KRAS wild-type (wt) untreated metastatic adenocarcinoma of the colon or rectum (MCRC). J Clin Oncol. 2014;32:5s (suppl; abstr LBA3). 18. Venook AP, Niedzwiecki D, Innocenti F, et al. Impact of primary (1°) tumor location on overall survival (OS) and progression-free survival (PFS) in patients (pts) with metastatic colorectal cancer (mCRC): analysis of CALGB/SWOG 80405 (Alliance). J Clin Oncol. 2016;34 (suppl; abstr 3504).

29. Brannon AR, Vakiani E, Sylvester BE, et al. Comparative sequencing analysis reveals high genomic concordance between matched primary and metastatic colorectal cancer lesions. Genome Biol. 2014;15:454. 30. Vakiani E, Janakiraman M, Shen R, et al. Comparative genomic analysis of primary versus metastatic colorectal carcinomas. J Clin Oncol. 2012;30:2956-2962. 31. Amado RG, Wolf M, Peeters M, et al. Wild-type KRAS is required for panitumumab efficacy in patients with metastatic colorectal cancer. J Clin Oncol. 2008;26:1626-1634. 32. Van Cutsem E, Köhne CH, Hitre E, et al. Cetuximab and chemotherapy as initial treatment for metastatic colorectal cancer. N Engl J Med. 2009;360:1408-1417. 33. Pietrantonio F, Petrelli F, Coinu A, et al. Predictive role of BRAF mutations in patients with advanced colorectal cancer receiving cetuximab and panitumumab: a meta-analysis. Eur J Cancer. 2015;51:587-594. 34. Rowland A, Dias MM, Wiese MD, et al. Meta-analysis of BRAF mutation as a predictive biomarker of benefit from anti-EGFR monoclonal antibody therapy for RAS wild-type metastatic colorectal cancer. Br J Cancer. 2015;112:1888-1894.

19. Cremolini C, Antoniotti C, Moretto R, et al. First-line therapy for mCRC - the influence of primary tumour location on the therapeutic algorithm. Nat Rev Clin Oncol. 2017;14:113.

35. Yaeger R, Cowell E, Chou JF, et al. RAS mutations affect pattern of metastatic spread and increase propensity for brain metastasis in colorectal cancer. Cancer. 2015;121:1195-1203.

Holch JW, Ricard I, Stintzing S, et al. The relevance of primary tumour 20. location in patients with metastatic colorectal cancer: A meta-analysis of first-line clinical trials. Eur J Cancer. 2017;70:87-98.

Maughan TS, Adams RA, Smith CG, et al; MRC COIN Trial Investigators. 36. Addition of cetuximab to oxaliplatin-based first-line combination chemotherapy for treatment of advanced colorectal cancer: results of the randomised phase 3 MRC COIN trial. Lancet. 2011;377:2103-2114.

21. Loupakis F, Cremolini C, Masi G, et al. Initial therapy with FOLFOXIRI and bevacizumab for metastatic colorectal cancer. N Engl J Med. 2014;371:1609-1618. Cremolini C, Loupakis F, Antoniotti C, et al. FOLFOXIRI plus 22. bevacizumab versus FOLFIRI plus bevacizumab as first-line treatment of patients with metastatic colorectal cancer: updated overall survival and molecular subgroup analyses of the open-label, phase 3 TRIBE study. Lancet Oncol. 2015;16:1306-1315. Saridaki Z, Androulakis N, Vardakis N, et al. A triplet combination 23. with irinotecan (CPT-11), oxaliplatin (LOHP), continuous infusion 5-fluorouracil and leucovorin (FOLFOXIRI) plus cetuximab as first-line treatment in KRAS wt, metastatic colorectal cancer: a pilot phase II trial. Br J Cancer. 2012;107:1932-1937. 24. Fornaro L, Lonardi S, Masi G, et al. FOLFOXIRI in combination with panitumumab as first-line treatment in quadruple wild-type (KRAS, NRAS, HRAS, BRAF) metastatic colorectal cancer patients: a phase II trial by the Gruppo Oncologico Nord Ovest (GONO). Ann Oncol. 2013;24:2062-2067. 25. Benson AB III, Venook AP, Cederquist L, et al. Clinical Guidelines in Oncology (NCCN Guidelines). Colon Cancer, Version I.2017. National Comprehensive Cancer Network. https://www.nccn.org/ professionals/physician_gls/pdf/colon.pdf. Accessed March 13, 2017. 26. Douillard JY, Oliner KS, Siena S, et al. Panitumumab-FOLFOX4 treatment and RAS mutations in colorectal cancer. N Engl J Med. 2013;369:10231034.

37. Tran B, Kopetz S, Tie J, et al. Impact of BRAF mutation and microsatellite instability on the pattern of metastatic spread and prognosis in metastatic colorectal cancer. Cancer. 2011;117:4623-4632. Atreya CE, Greene C, McWhirter RM, et al. Differential radiographic 38. appearance of BRAF V600E-mutant metastatic colorectal cancer in patients matched by primary tumor location. J Natl Compr Canc Netw. 2016;14:1536-1543. Clancy C, Burke JP, Kalady MF, et al. BRAF mutation is associated 39. with distinct clinicopathological characteristics in colorectal cancer: a systematic review and meta-analysis. Colorectal Dis. 2013;15:e711-e718. 40. Yaeger R, Cercek A, Chou JF, et al. BRAF mutation predicts for poor outcomes after metastasectomy in patients with metastatic colorectal cancer. Cancer. 2014;120:2316-2324. 41. Vauthey JN, Zimmitti G, Kopetz SE, et al. RAS mutation status predicts survival and patterns of recurrence in patients undergoing hepatectomy for colorectal liver metastases. Ann Surg. 2013;258:619626, discussion 626-627. Passot G, Denbo JW, Yamash*ta S, et al. Is hepatectomy justified for 42. patients with RAS mutant colorectal liver metastases? An analysis of 524 patients undergoing curative liver resection. Surgery. 2017;161:332-340.

27. Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell. 1990;61:759-767.

Vnencak-Jones CL, Berger MF, Pao W. Types of molecular tumor 43. testing. My Cancer Genome. https://www.mycancergenome.org/ content/molecular-medicine/types-of-molecular-tumor-testing/. Accessed March 13, 2017.

28. Janakiraman M, Vakiani E, Zeng Z, et al. Genomic and biological characterization of exon 4 KRAS mutations in human cancer. Cancer Res. 2010;70:5901-5911.

44. Frampton GM, Fichtenholtz A, Otto GA, et al. Development and validation of a clinical cancer genomic profiling test based on massively parallel DNA sequencing. Nat Biotechnol. 2013;31:1023-1031.

254 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

SYSTEMIC THERAPY FOR METASTATIC COLORECTAL CANCER

45. Hechtman JF, Zehir A, Yaeger R, et al. Identification of targetable kinase alterations in patients with colorectal carcinoma that are preferentially associated with wild-type RAS/RAF. Mol Cancer Res. 2016;14:296-301. 46. Stadler ZK, Battaglin F, Middha S, et al. Reliable detection of mismatch repair deficiency in colorectal cancers using mutational load in nextgeneration sequencing panels. J Clin Oncol. 2016;34:2141-2147. 47. Martin V, Landi L, Molinari F, et al. HER2 gene copy number status may influence clinical efficacy to anti-EGFR monoclonal antibodies in metastatic colorectal cancer patients. Br J Cancer. 2013;108:668-675. 48. Raghav KPS, Overman MJ, Yu R, et al. HER2 amplification as a negative predictive biomarker for anti-epidermal growth factor receptor antibody therapy in metastatic colorectal cancer. J Clin Oncol. 2016;34 (suppl; abstr 3517). 49. Sartore-Bianchi A, Trusolino L, Martino C, et al. Dual-targeted therapy with trastuzumab and lapatinib in treatment-refractory, KRAS codon 12/13 wild-type, HER2-positive metastatic colorectal cancer (HERACLES): a proof-of-concept, multicentre, open-label, phase 2 trial. Lancet Oncol. 2016;17:738-746. Sartore-Bianchi A, Ardini E, Bosotti R, et al. Sensitivity to entrectinib 50. associated with a novel LMNA-NTRK1 gene fusion in metastatic colorectal cancer. J Natl Cancer Inst. 2015;108:djv306. 51. Diaz LA Jr, Bardelli A. Liquid biopsies: genotyping circulating tumor DNA. J Clin Oncol. 2014;32:579-586. Misale S, Yaeger R, Hobor S, et al. Emergence of KRAS mutations and 52. acquired resistance to anti-EGFR therapy in colorectal cancer. Nature. 2012;486:532-536. Siravegna G, Mussolin B, Buscarino M, et al. Clonal evolution and 53. resistance to EGFR blockade in the blood of colorectal cancer patients. Nat Med. 2015;21:795-801. Atreya CE, Corcoran RB, Kopetz S. Expanded RAS: refining the patient 54. population. J Clin Oncol. 2015;33:682-685. Bendell J, Kim T, Goh B, et al. Clinical activity and safety of cobimetinib 55. (cobi) and atezolizumab in colorectal cancer (CRC). J Clin Oncol. 2016;34 (suppl; abstr 3502). Le DT, Uram JN, Wang H, et al. PD-1 blockade in tumors with mismatch56. repair deficiency. N Engl J Med. 2015;372:2509-2520. 57. Overman M, Kopetz S, McDermott R, et al. Nivolumab ± ipilimumab in treatment (tx) of patients (pts) with metastatic colorectal cancer (mCRC) with and without high microsatellite instability (MSI-H): CheckMate-142 interim results. J Clin Oncol. 2016;34 (suppl; abstr 3501). 58. Overman M, Lonardi S, Leone F, et al. Nivolumab in patients with DNA mismatch repair deficient/microsatellite instability high metastatic colorectal cancer: update from CheckMate 142. J Clin Oncol. 2017;35 (suppl 4S; abstract 519). 59. Le D, Uram J, Wang H, et al. Programmed death-1 blockade in mismatch repair deficient colorectal cancer. J Clin Oncol. 2016;34 (suppl; abstr 103). 60. Le D, Andre T, Kim T, et al. KEYNOTE-164: phase 2 study of pembrolizumab for patients with previously treated, microsatellite instability-high advanced colorectal carcinoma. J Clin Oncol. 2016;34 (suppl; abstr TPS363). 61. Diaz L, Le D, Yoshino T, et al. KEYNOTE-177: first-line, open-label, randomized, phase 3 study of pembrolizumab versus investigatorchoice chemotherapy for mismatch repair-deficient or microsatellite instability-high metastatic colorectal carcinoma. J Clin Oncol. 2016;34 (suppl; abstr TPS3639).

62. Mlecnik B, Bindea G, Angell HK, et al. Integrative analyses of colorectal cancer show immunoscore is a stronger predictor of patient survival than microsatellite instability. Immunity. 2016;44:698-711. 63. Samowitz WS, Sweeney C, Herrick J, et al. Poor survival associated with the BRAF V600E mutation in microsatellite-stable colon cancers. Cancer Res. 2005;65:6063-6069. 64. Yokota T, Ura T, Shibata N, et al. BRAF mutation is a powerful prognostic factor in advanced and recurrent colorectal cancer. Br J Cancer. 2011;104:856-862. 65. Bokemeyer C, Van Cutsem E, Rougier P, et al. Addition of cetuximab to chemotherapy as first-line treatment for KRAS wild-type metastatic colorectal cancer: pooled analysis of the CRYSTAL and OPUS randomised clinical trials. Eur J Cancer. 2012;48: 1466-1475. Morris V, Overman MJ, Jiang ZQ, et al. Progression-free survival 66. remains poor over sequential lines of systemic therapy in patients with BRAF-mutated colorectal cancer. Clin Colorectal Cancer. 2014;13:164171. 67. Cremolini C, Di Bartolomeo M, Amatu A, et al. BRAF codons 594 and 596 mutations identify a new molecular subtype of metastatic colorectal cancer at favorable prognosis. Ann Oncol. 2015;26:20922097. 68. Zheng G, Tseng LH, Chen G, et al. Clinical detection and categorization of uncommon and concomitant mutations involving BRAF. BMC Cancer. 2015;15:779. 69. Seligmann JF, Fisher D, Smith CG, et al. Investigating the poor outcomes of BRAF-mutant advanced colorectal cancer: Analysis from 2530 patients in randomised clinical trials. Ann Oncol. Epub 2016 Dec 19. 70. Van Cutsem E, Cervantes A, Adam R, et al. ESMO consensus guidelines for the management of patients with metastatic colorectal cancer. Ann Oncol. 2016;27:1386-1422. 71. Sehdev A, Cramer HM, Ibrahim AA, et al. Pathological complete response with anti-PD-1 therapy in a patient with microsatellite instable high, BRAF mutant metastatic colon cancer: a case report and review of literature. Discov Med. 2016;21:341-347. 72. Kopetz S, Desai J, Chan E, et al. Phase II pilot study of vemurafenib in patients with metastatic BRAF-mutated colorectal cancer. J Clin Oncol. 2015;33:4032-4038. 73. Hyman DM, Puzanov I, Subbiah V, et al. Vemurafenib in multiple nonmelanoma cancers with BRAF V600 mutations. N Engl J Med. 2015;373:726-736. 74. Yaeger R, Cercek A, O’Reilly EM, et al. Pilot trial of combined BRAF and EGFR inhibition in BRAF-mutant metastatic colorectal cancer patients. Clin Cancer Res. 2015;21:1313-1320. 75. Hong DS, Morris VK, El Osta B, et al. Phase IB study of vemurafenib in combination with irinotecan and cetuximab in patients with metastatic colorectal cancer with BRAFV600E mutation. Cancer Discov. 2016;6:1352-1365. 76. Kopetz S, McDonough S, Morris V, et al. Randomized phase II study of irinotecan and cetuximab with or without vemurafenib in BRAFmutant metastatic colorectal cancer (SWOG1406). J Clin Oncol. 2017;35 (suppl 4S; abstract 520). 77. Corcoran RB, Atreya CE, Falchook GS, et al. Combined BRAF and MEK inhibition with dabrafenib and trametinib in BRAF V600-mutant colorectal cancer. J Clin Oncol. 2015;33:4023-4031.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 255

ATREYA, YAEGER, AND CHU

78. Corcoran R, André T, Yoshino T, et al. Efficacy and circulating tumor DNA (ctDNA) analysis of the BRAF inhibitor dabrafenib (D), MEK inhibitor trametinib (T), and anti-EGFR antibody panitumumab (P) in patients (pts) with BRAF V600E–mutated (BRAFm) metastatic colorectal cancer (mCRC). Ann Oncol. 2016;27 (Supplement 6): vi1-vi14. 79. Tabernero J, Van Geel V, Guren T, et al. Phase 2 results: encorafenib (ENCO) and cetuximab (CETUX) with or without alpelisib (ALP) in patients with advanced BRAF-mutant colorectal cancer (BRAFm CRC). J Clin Oncol. 2016;34 (suppl; abstr 3544). 80. Vlacich G, Coffey RJ. Resistance to EGFR-targeted therapy: a family affair. Cancer Cell. 2011;20:423-425. 81. Bertotti A, Papp E, Jones S, et al. The genomic landscape of response to EGFR blockade in colorectal cancer. Nature. 2015;526:263-267. 82. Yonesaka K, Zejnullahu K, Okamoto I, et al. Activation of ERBB2 signaling causes resistance to the EGFR-directed therapeutic antibody cetuximab. Sci Transl Med. 2011;3:99ra86. 83. Ramanathan RK, Hwang JJ, Zamboni WC, et al. Low overexpression of HER-2/neu in advanced colorectal cancer limits the usefulness of

256 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

trastuzumab (Herceptin) and irinotecan as therapy. A phase II trial. Cancer Invest. 2004;22:858-865. 84. Hurwitz H, Hainsworth J, Swanton C, et al. Targeted therapy for gastrointestinal (GI) tumors based on molecular profiles: Early results from MyPathway, an open-label phase IIa basket study in patients with advanced solid tumors. J Clin Oncol. 2016;34 (suppl 4S; abstr 653). 85. Siena S, Bardelli A, Sartore-Bianchi A, et al. HER2 amplification as a ‘molecular bait’ for trastuzumab-emtansine (T-DM1) precision chemotherapy to overcome anti-HER2 resistance in HER2 positive metastatic colorectal cancer: the HERACLES-RESCUE trial. J Clin Oncol. 2016;34 (suppl 4S; abstr TPS774). 86. Hurwitz H, Raghav K, Burris H, et al. Pertuzumab + trastuzumab for HER2amplified/overexpressed metastatic colorectal cancer (mCRC): interim data from MyPathway. J Clin Oncol. 2017;35 (suppl 4S; abstract 676). 87. Weber AM, Ryan AJ. ATM and ATR as therapeutic targets in cancer. Pharmacol Ther. 2015;149:124-138. 88. Amatu A, Somaschini A, Cerea G, et al. Novel CAD-ALK gene rearrangement is drugable by entrectinib in colorectal cancer. Br J Cancer. 2015;113:1730-1734.

GASTROINTESTINAL (NONCOLORECTAL) CANCER

FEDERICO A. SANCHEZ

Best Practices and Practical Nuances in the Treatment of Gastric Cancer in High-Risk Global Areas Federico A. Sanchez, MD OVERVIEW Gastric cancer is an aggressive disease that is very frequent in Latin America. The reasons for this increased incidence is not clear. Associated with the lack of minimum health care opportunities, lack of accurate statistics and reporting data beyond epidemiologic data, and raw nonreliable data, there is little known of the actual clinical course and treatment of these patients. Understanding epidemiologic data may allow us to encourage the adequate use and distribution of the meager resources that exist.

he incidence of gastric cancer in Latin America and in Central America is quite high. In the neighboring countries of Central America—Guatemala, El Salvador, and Honduras—the incidence of gastric cancer is high enough to be considered the most common malignancy. Actual statistics and information about the number of occurrences and the clinical course of this condition are unknown. In Central America, with the exception of Panama and Costa Rica, a mature, well-developed cancer registry program is sorely lacking.1 Internet sites like www.WorldAtlas.com quote a gastric cancer incidence in Guatemala as high as 23.7 in 100,000. Limited information gathered through National Cancer Institute of Guatemala states that gastric cancer remains the leading cause of cancer-related deaths.1 Regardless, gastric cancer is the second most common cause of cancer mortality worldwide and is the leading infection-associated cancer.2,3 Gastric cancer has clear geographic and ethnic variability.3,4 The highlands of the Pacific coast of Latin America have the highest incidence and mortality rates of the sites studied so far; the areas affected include Mexico, Guatemala, Honduras, Costa Rica, and Colombia.5 As already mentioned, with the exception of Panama and Costa Rica, any type of information or literature about the epidemiology of gastric cancer in Central America is limited.1,3 Gastric cancer development has to be considered a multifactorial process, in which many conditions play a role; historically, the roles of diet, infection, ethnic background, socioeconomic conditions, and the altitude enigma6 have been addressed. The impact of these factors may actually affect the histologic type of cancer (personal observation on the frequency of signet ring type histology) and the anatomic location of the sites of the cancer (proximal vs.

distal). It should be noted that, without much data and on the basis of personal observations, at least in Guatemala, gastric cancers in lower socioeconomic populations tend to present with signet ring morphology and a more virulent behavior; anatomically, they tend to be distal cancers. This article attempts to outline the known facts associated with this condition in Guatemala and northern Central America.

RISK FACTORS

Dietary factors may induce cancer, but dietary habits also may be protective; diet is being studied aggressively. Historically, worldwide studies have documented protection from diets high in vegetables and fruit. Conversely, high intake of processed, red meat or smoked preserved foods seems to increase the risk of gastric cancer. A meta-analysis of Latin American studies done by Bonequi et al4 showed that a trend toward this protection did exist, but the association was considered weak. Results for the association with the intake of red meat, processed meat, and salt did not vary from other global studies and did not indicate an increase in risk for the development of gastric cancer.4 Smoking and drinking clearly showed a notable impact on the development of this malignancy. Smoking increased the incidence of gastric, overall risk increased 60% between smokers and nonsmokers. The dose response meta-analysis for this cancer documented an increase in gastric cancer risk of 12% per exposure of 10-pack years.4 Alcohol use seemed to increase the risk of gastric cancer by 61%, but there was notable variability on the amounts of exposure reported.4 Education and ethnicity may go hand in hand as surrogate indicators of socioeconomic status and income potential in countries where the median income of the population is approximately $1.60 per day.1 Clearly, ethnicity has been

From Aurora Cancer Care, Aurora Health Care, Milwaukee, WI. Disclosures of potential conflicts of interest provided by the author are available with the online article at asco.org/edbook. Corresponding author: Federico A. Sanchez, MD, Aurora Health Care, 750 W. Virginia St., P.O. Box 341880, Milwaukee, WI 53204; email: [emailprotected]. © 2017 American Society of Clinical Oncology

258 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

GASTRIC CANCER IN CENTRAL AMERICA

studied worldwide, but, in Latin America, where 10% of the population is indigenous,7 the lack of information is widespread, and even the more advanced economies of the area lack statistics and data with which to work.1,8 We know that poverty is associated with poor outcomes of cancer care; this has been attributed to poor access to medical care either because of low availability or, probably more important, because of the high cost of care and the lack of accessible, affordable resources in these low-income countries.7 Also, poverty and ethnic background fuel persistent cultural beliefs, which hinder both access and availability to medical care.4,8 The lack of adequate educational and medical resources at the national level in these societies also affects the incidence of gastric cancer. Other risk factors worth mentioning include the association with infections and the site of residence. We have known for some time that the presence of infections, in particular Helicobacter pylori (H. pylori), clearly increases the risk of gastric cancer. This association, identified in the 1990s, has been aggressively studied worldwide.9 In association with presence of infections, the site of residence for this population may play a role. Interesting data have been published on the role of high altitude and the development of gastric cancer. Torres et al6 reviewed published statistics and data that documented clearly higher gastric cancer mortality in the Americas and a higher incidence of disease concentrated in the nations along the Pacific Rim. Observational analysis cited in the article document notable changes in incidence of this disease in short distances, sometimes as small as 150 miles, but the most distinct difference between the communities was the altitude.6 Implications of these observations are important; these observations have led to the development of the altitude enigma concept. Historically, geographic barriers like mountains or large bodies of water allow development of diverse genetic patterns in humans and in bacteria or other organisms. If we include the impact of population changes triggered by immigration, slavery, and historical events, the implications are staggering. Implications and associations studied so far include evaluation of H. pylori genotypes and haplotypes on the basis of altitude. The ancestral origin of the H. pylori strain studied by phylogenetic haplotypes has been documented by de Sablet et al,10 who noted clear

KEY POINTS • In the Americas, the highest incidence of gastric cancer is found the mountain areas of the Pacific Rim. • Gastric cancer has ethnic and geographic variability. • Dietary issues may not play as large a role in the pathogenesis of gastric cancer. • Altitude may be a surrogate for bacterial, dietary, and environmental factors that may predispose to gastric cancer. • Interactions between H. pylori serotype and population genetic makeup could explain some of the epidemiological variability for gastric cancer.

variation between the African ancestries populations that live in the coastal regions of Colombia compared with the mestizo population (European/Amerindian) in the highlands. It was determined that the incidence of H. pylori from European ancestry was more common in the highlands with the mestizo population, whereas the coastal populations had 66% African-originated H. pylori.10 These researchers also observed an increase in more severe gastritis and DNA damage in the populations affected by the European-ancestry H. pylori, but they still could not explain, by this finding alone, the 25-fold difference in incidence of gastric cancer between coastal and mountain regions of Colombia. Clearly, other risk factors do exist. The possibility of coinfections, either chronic helminthiasis or Epstein-Barr virus, has been suggested.11,12 Diet, as previously mentioned, also has been strongly evaluated.

EPIDEMIOLOGY OF GASTRIC CANCER IN CENTRAL AND LATIN AMERICA

A difference in the type of gastric cancers according to socioeconomic status has been noted. As documented in all of the previously mentioned citations, gastric cancers in low-income countries tend to be noncardia, affect the poor more often, and have a male-to-female ratio of 2:1.3 In contrast, the higher socioeconomic, more educated, and more affluent groups tend to suffer from cardia-based gastric cancer, which presents with a ratio of 5:1 in men to women and 2:1 in white to black individuals.13 To explain the different pattern of development for gastric cancer in the cardia in the developed countries, other dietary behavioral issues—in particular, obesity and tobacco use—have been suggested.14 These suggestions carry many concerns. Does the different carcinogenesis pattern develop a different kind of biologic behavior? Does the different pattern of development imply a different pattern of genetic amplification and overexpression that would affect how we treat this condition? Clearly, these are all good questions that, without adequate databases, will be hard to answer. Corral et al3 observed that, as Hispanic people have migrated to the United States, there has been a slight shift proximal of the site of origin of the malignancy; however, malignancies remain noncardia in origin. Cancers of the antrum in Central America occurred in 73.6%; cancer of the corpus was slightly more common, at 54%, in the U.S. Hispanic population. Thus, there is either a genetic predisposition or a residual epidemiologic drift brought on by cultural and social affinities; it is hard to know which is true.

IMPLICATIONS

This brief review outlines the issues created by the inequities in medical care. Cancer is one of the most common medical conditions across all continents and social strata and should be considered a public health issue. The problem of adequate care is compounded by the multiple other social issues surrounding illness and public health cost that exist, in particular in low-income countries, which results in a lack of information worldwide. Few studies outside the asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 259

FEDERICO A. SANCHEZ

high-income countries have been done to provide indicators, to identify the magnitude and scope of this problem, and to plan an approach to tackle it. Treatment patterns and outcomes in Latin American countries, in particular in Central America, usually are underreported; any accurate static or

information is elusive. Cancer control can be achieved only if we develop a system to acquire relevant data that could be analyzed to allow development of a systematic approach to address this problem across all countries and socioeconomic strata.

References 1. Barnoya J. Cancer in Guatemala: first steps against a growing problem. http://www.nypcancerprevention.org/archive/issue/20/cancer_ prevention/feature/guatemala.shtml. Accessed January 17, 2017. 2. World Atlas. Countries With the Highest Prevalence of Stomach Cancer in the World. http://www.worldatlas.com/articles/countries-with-thehighest-prevalence-of-stomach-cancer.html. Accessed January 17, 2017. 3. Corral JE, Delgado Hurtado JJ, Domínguez RL, et al. The descriptive epidemiology of gastric cancer in Central America and comparison with United States Hispanic populations. J Gastrointest Cancer. 2015;46:21-28.

8. Sampieri CL, Mora M. Gastric cancer research in Mexico: a public health priority. World J Gastroenterol. 2014;20:4491-4502. 9. Parsonnet J, Friedman GD, Vandersteen DP, et al. Helicobacter pylori infection and the risk of gastric carcinoma. N Engl J Med. 1991;325:1127-1131. 10. de Sablet T, Piazuelo MB, Shaffer CL, et al. Phylogeographic origin of Helicobacter pylori is a determinant of gastric cancer risk. Gut. 2011;60:1189-1195.

4. Bonequi P, Meneses-González F, Correa P, et al. Risk factors for gastric cancer in Latin America: a meta-analysis. Cancer Causes Control. 2013;24:217-231.

11. Whary MT, Sundina N, Bravo LE, et al. Intestinal helminthiasis in Colombian children promotes a Th2 response to Helicobacter pylori: possible implications for gastric carcinogenesis. Cancer Epidemiol Biomarkers Prev. 2005;14:1464-1469.

5. Dominguez RL, Crockett SD, Lund JL, et al. Gastric cancer incidence estimation in a resource-limited nation: use of endoscopy registry methodology. Cancer Causes Control. 2013;24:233-239.

12. Ryan,JL Morgan,DR Dominguez,RL, et al; High levels of Epstein-Barr virus DNA in latently infected gastric adenocarcinoma. Lab Invest. 2009;89:80-90.

6. Torres J, Correa P, Ferreccio C, et al. Gastric cancer incidence and mortality is associated with altitude in the mountainous regions of Pacific Latin America. Cancer Causes Control. 2013;24:249-256.

13. Crew KD, Neugut AI. Epidemiology of gastric cancer. World J Gastroenterol. 2006;12:354-362.

7. Hall G, Patrinos H. Indigenous People: Poverty and Human Development in Latin America 1994-2004 (vol. 8). Washington, DC: The World Bank; 2006;221-240.

260 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

14. Vaughan TL, Davis S, Kristal A, et al. Obesity, alcohol, and tobacco as risk factors for cancers of the esophagus and gastric cardia: adenocarcinoma versus squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev. 1995;4:85-92.

GASTRIC CANCER AND HIGH-RISK AREAS

Gastric Cancer in Southern Europe: High-Risk Disease Ramon Andrade De Mello, MD, PhD OVERVIEW Gastric cancer is an aggressive disease. Several risk factors are involved in gastric cancer pathogenesis, likely Helicobacter pylori (H. pylori) infection, genetic factors in hereditary syndromes, lifestyle, and diet. However, well-implemented screening strategies are lacking in most countries, including those in Southern Europe. Nevertheless, gastric cancer outcomes are better in some Southern European countries than in others, likely because of the incidence and distribution of different histologic types. Robotic surgery has been gaining favor as a treatment of early-stage disease, and the need for perioperative chemotherapy or adjuvant chemoradiotherapy (CRT) for locally advanced disease has been debated. In the metastatic setting, trastuzumab in combination with chemotherapy has helped to extend survival compared with chemotherapy alone for HER2-positive disease. This article will describe how gastric cancer is assessed and treated in Southern Europe in an attempt to correlate these approaches from a global perspective.

astric cancer is a very aggressive disease worldwide.1 Specifically, it is the third most important cause of cancer-related death, and in Southern Europe, it is the sixth most common malignancy. Therefore, Southern Europe is considered a high-risk area for gastric cancer.1-3 In Europe, the incidence of gastric cancer is not hom*ogeneous, and the risk is low in some areas, such as Central Europe. The mortality rate is 9.7% in men and 4.6% in women.1 Moreover, some familial syndromes, such as hereditary diffuse gastric cancer syndrome, which is related to the CDH1 gene mutation, play an important role in the etiology of gastric cancer.4,5 Furthermore, the incidence of H. pylori and genetic factors are responsible for a variety of related diseases and their presentation. Currently, H. pylori infection is considered a component of gastric cancer development. In addition, factors related to the improvement of the population’s living conditions, a better diet, and improved food preservation may improve the course of the disease.1,6 Although the incidence of gastric cancer continues to decrease in the United States and Europe, gastric cancer remains an important leading cause of cancer death worldwide.7 Geographic differences in the disease burden of gastric cancer (highest in Japan, Korea, and regions of Latin America) suggest that environmental and dietary factors play major roles in gastric cancer risk. Histologically, gastric adenocarcinomas are classified as intestinal or diffuse; diffuse-type gastric cancers make up 15% to 10% of cases and are characterized by the submucosal spread of neoplastic signet ring cells. In this article, we discuss how gastric cancer is assessed and treated in

Southern Europe in an attempt to correlate these approaches from a global perspective.

RISK FACTORS AND HEREDITARY SYNDROMES

Several genetic and hereditary factors have been identified to interact with gastric cancer pathogenesis. Nevertheless, environmental risk factors have been studied on a population basis because they can be modified to reduce the prevalence of this disease. Although Western Europe is considered an overall low-risk area, gastric cancer is a challenge for oncology health care professionals in Southern Europe, which is considered a high-risk area. However, gastric cancer mortality has been decreasing in recent decades (since 1971).8 Although most gastric adenocarcinomas are presumed to be sporadic, approximately 5% to 10% arise in individuals with a family history of gastrointestinal cancer, and 3% to 5% of these cases are estimated to be associated with inherited cancer predisposition syndromes.8 Following the decreases observed since the 1970s in Portugal, further declines in gastric cancer mortality were projected for 2015 and 2020, with an expected number of deaths of approximately 1,400 and 1,300 in men and 900 and 800 in women, respectively, corresponding to crude rates of 28.9/100,000 and 28.2/100,000 in men and 16.9/100,000 and 16.1/100,000 in women, respectively.7-10 Another important issue is the prevalence of H. pylori, which increases with age, from childhood to age 45 to 50 in adults. In addition, the prevalence of H. pylori has not exhibited consistent trends in adults and children since the

From the Department of Biomedical Sciences and Medicine, University of Algarve, Faro, Portugal, and the Department of Medical Oncology, Clatterbridge Cancer Centre, Merseyside, United Kingdom. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Ramon Andrade de Mello, MD, PhD, Department of Biomedical Sciences and Medicine, University of Algarve, Campus de Gambelas, Edifício 2, Piso 2, 8005-139, Faro, Portugal; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 261

RAMON ANDRADE DE MELLO

1990s. Regarding overall survival (OS), some screening programs implemented in certain areas, such as Korea, may improve survival because of early diagnosis. Nevertheless, survival is higher in Portugal than in Europe overall. This difference is likely because of a higher incidence of tumors with a better prognosis, such as noncardia tumors and intestinal histologic type tumors and not because of generalized screening programs.7,9,11 Furthermore, efforts to reduce the prevalence of H. pylori infection are important because such efforts reduce gastric cancer mortality, as reported in previous studies of developed countries such as the United States and Central Europe.

PATHOLOGY AND MOLECULAR PROFILE

Genetic alterations are closely associated with the development of neoplasia, and the body of evidence identifying molecular factors that affect gastric cancer is continuously growing. Specifically, approximately 140 genes have been identified to drive cancer, a subset of which have been implicated in hereditary cancer syndromes. However, the types and patterns of these mutations are highly variable and heterogeneous within a single entity as well as between different tumor categories. For example, the hyperactivation of the phosphatidylinositol-3 kinase (PI3K)/Akt/ mTOR signaling pathway in gastric cancer has frequently been identified to be related to mutations and/or amplifications of the PIK3CA gene and a loss of function of PTEN, which play a crucial role in the regulation of this pathway. The PI3K/Akt/mTOR signaling pathway plays an important role in mediating multiple cellular functions including cell growth, proliferation, metabolism, survival, and angiogenesis. In 2014, researchers from The Cancer Genome Atlas (TCGA) research network studied 295 stomach tumors and found that 80% harbor different grades of mutations in the PIK3CA gene and amplifications of receptor tyrosine kinase genes, such as ERBB3, ERBB2, and EGFR, which increase the activities of these proteins in gastric cancer. Akt acts as a central regulator of cell survival by transcriptionally and post-translationally interacting with antiapoptotic

KEY POINTS • Gastric cancer is a very aggressive disease with a higher incidence in Latin America, Southern Europe, and Asia compared with other areas. • Several genetic and hereditary factors have been identified to interact with gastric cancer pathogenesis. • The eradication of H. pylori reduces the incidence of gastric cancer and peptic ulcers as well as the prevalence and cost of managing dyspepsia. • Minimally invasive surgery has been demonstrated to improve the short-term outcomes in selected patients compared with open surgery, especially the rate of perioperative complications. • Irinotecan in combination with 5-FU (FOLFIRI or IF) is an option for the treatment of patients with chemotherapynaive disease with a poor performance status. 262 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

signals. In addition, Akt phosphorylates Bad, a member of the BCL2 family of antiapoptotic proteins, at SER136 and Caspase9, a protease, and at SER196, which partially inhibits cell death and supports cell survival signals. Akt also regulates antiapoptotic transcriptional functions by translocating to the nucleus and regulating the transcription of the forkhead box O (FoxO) family of transcription factors. The FoxO family of transcription factors regulates cell death signals by expressing various members of both intrinsic and extrinsic modes of apoptosis as well as cyclin-dependent kinase inhibitors. Upon nuclear translocation, Akt represses the transcription of FoxO1, FoxO3, and FoxO4, thereby enhancing cell survival signals. In 2014, researchers from TCGA examined 295 stomach tumors and identified subtypes using complex statistical analyses of molecular data obtained from six molecular analysis platforms. Thereafter, they described a new molecular characterization that defines four major genomic subtypes of gastric cancer: positive for Epstein-Barr virus, microsatellite instability, chromosomally instability, and genomic stability. At least three of these subtypes, which includes 80% of the studied gastric cancer cases (Epstein-Barr virus–positive, microsatellite instability, and chromosomally instability subgroups), house different grades of mutations in the PIK3CA gene and amplifications of receptor tyrosine kinase genes, such as ERBB3, ERBB2, and EGFR. In a recent Chinese study, the authors suggested that a mutation in the GTPase RHOA gene and its oncogenic signaling pathway represent a strong biomarker-driven therapeutic target for Asian gastric cancer. This comprehensive strategy represents a promising approach for the development of hit compounds. RHOA is frequently overexpressed in the gastric cancer tumors of Japanese and Chinese patients, whereas gastric cancer datasets from TCGA depository exhibited RHOA mutations, not mere overexpression, in diffuse-type gastric cancer tumors. More recent evidence suggests that changes in PHOSPHO2-KLHL23 mRNA expression were the most significant in gastric adenocarcinoma. PHOSPHO2 is important for metabolism and the vitamin B6 metabolism pathway, and KLHL23 is implicated in cone-rod dystrophy and the vitamin B6 metabolism pathway. Ribosomal protein L17 (RPL17), also known as RPL23, is a component of the large 60S ribosome subunit and promotes multidrug resistance in gastric cancer cells by suppressing drug-induced apoptosis. In a Korean study, Choi and colleagues12 screened read-through transcription events from stomach adenocarcinoma RNA-seq data and selected three candidates, PHOSPHO2-KLHL23, RPL17-C18orf32, and PRR5-ARHGAP8, to assess their biologic role in gastric cancer. They suggested that PHOSPHO2-KLHL23 was the most significantly upregulated transcript in stomach tumor tissues (p < .0001), and our investigation revealed that the KLHL23 protein is related to this tumorigenic effect. One to three percent of all gastric cancers may be considered hereditary diffuse gastric cancer.13 Furthermore, consistent with the biallelic CDH1 inactivation and consequent E-cadherin loss of function, E-cadherin protein expression,

GASTRIC CANCER AND HIGH-RISK AREAS

as assessed by immunohistochemistry, is almost always abnormal in hereditary diffuse gastric cancer, in contrast to the normal complete membranous expression in adjacent normal (nontumoral) epithelium.3,5,14-16 The E-cadherin gene, CDH1, is located on chromosome 16q22.1, and heterozygous germline CDH1 mutations have been described in 18% to 40% of hereditary diffuse gastric cancer families.4,17 The 120 kDa glycoprotein encoded by CDH1 features a large extracellular domain, a transmembrane segment and a short cytoplasmic domain. E-cadherin is a transmembrane calcium-dependent protein that is mainly expressed at the basolateral membrane of epithelial cells, where it plays important roles in cell-cell adhesion at the adherens junctions to maintain epithelial integrity. Furthermore, several other genes are involved in hereditary diffuse gastric cancer predisposition, including CTNNA1. Like CDH1, CTNNA1 is involved in intercellular cell adhesion, and CTNNA1 encodes the protein α-E-catenin, which functions in a complex with β-catenin, where it binds the cytoplasmic domain of E-cadherin to the cytoskeleton.14,18

CLINICAL APPROACHES AND TREATMENT

Currently, screening strategies are not well implemented throughout European and Latin American countries. Nevertheless, the eradication of H. pylori reduces the incidence of gastric cancer and peptic ulcers as well as the prevalence and cost of managing dyspepsia. Specifically, economic analyses suggest that the eradication of H. pylori is cost-effective in controlling gastric cancer for high-risk populations.19-21 Table 1 summarizes current clinical trials involved in gastric cancer prevention and/or early detection. In terms of molecular approaches, a single molecular alteration that has been universally accepted as an independent prognostic factor in gastric cancer has not been identified. Instead, many gene expression signatures have been used to classify tumors into intrinsic subtypes and predict the survival of patients with gastric cancer,22,23 and carriers of germline mutations in different genes associated with cancer predisposition have an increased risk for various tumor types.24 The identification of germline mutations in families offers the opportunity for early intervention in relatives as yet unaffected by cancer who may be at high risk. Specifically, 25 frequently mutated genes have been identified in gastric adenocarcinoma (e.g., PIK3CA, RHOA, ARID1A, KRAS, MUC6, RNF43, CNGA4, TP53, SMAD4, CDH1, APC, MLH1, MSH2, MSH6, STK11), and these mutations correspond to four tumor subtypes: positive for Epstein-Barr virus, microsatellite stable (hypermutated), genomically stable (predominantly diffuse subtype), and chromosomally instable.25 However, well-implemented strategies and policies to correlate these data with screening tools to avoid familiar gastric cancer cases are lacking. Furthermore, populations at a low risk for gastric cancer may also benefit from screening and treatment because of the effects on nonmalignant upper gastrointestinal diseases. However, public health authorities have been slow to consider the benefits of population-based screening and treatment as a means of

reducing the morbidity and mortality associated with H. pylori infection.21 Nevertheless, novel techniques are revolutionizing approaches for the treatment of gastric cancer. For example, robot-assisted surgery has recently improved conventional minimally invasive surgery. Specifically, Ceccarelli and colleagues26 reported an interesting retrospective Italian review of 363 consecutive patients undergoing robot-assisted surgery at an Italian general surgery unit from September 2012 to June 2016. The entire cohort and subgroups that underwent three of the most-performed surgeries (i.e., gastric resections, right colectomy, and liver resections) were analyzed. This analysis suggested that the benefits of minimally invasive surgery compared with open surgery have improved short-term outcomes in selected patients, as evidenced by lower perioperative complication rates and earlier recovery, resulting in an improved quality of life. These benefits were also suggested in elderly populations; the risk of death or morbidity was not increased among elderly patients compared with younger patients in the three groups examined in their retrospective cohort. Thus, their study showed robot-assisted surgery to be a safe and effective technique for the aging patient population, especially for major abdominal cancer surgery.26

Early Disease

Endoscopic resection may be suitable for well differentiated, early-stage gastric cancer that is smaller than 2 cm, confined to the mucosa, and is not ulcerated. Intestinal Lauren histology and no evidence of lymphovascular invasion also indicate mucosectomy in the following tumors: intramucosal cancers without ulceration, irrespective of tumor size; ulcerated intramucosal cancers less than 3 cm; or cancers with early invasion into the submucosa measuring less than 3 cm. Endoscopic submucosal dissection has proven more effective than endoscopic mucosal resection but requires greater skill and instrumentation and entails a significant risk for complications, including perforation.23,27 For locally advanced disease, complete resection with adequate margins remains the cornerstone of curative treatment. In gastric cancer, the type of resection, subtotal compared with total gastrectomy, depends on the anatomic location of the primary tumor. A total esophagectomy with a partial gastrectomy or an extended gastrectomy is generally performed for esophagogastric junction cancers, but the extent of lymph node dissection remains a subject of controversy. Nevertheless, consensus exists regarding lymphadenectomy: it must include at least 15 lymph nodes, and gastrectomy with D2 lymph node dissection is a recommended procedure.27

Neoadjuvant Chemotherapy

The indication for neoadjuvant chemotherapy is also a subject of great interest. According to the European Organisation for Research and Treatment of Cancer trial, which examined 40,954 inpatients with locally advanced gastric cancer, neoadjuvant chemotherapy did not provide a asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 263

RAMON ANDRADE DE MELLO

survival benefit.28 However, a meta-analysis by Sjoquist and colleagues showed that neoadjuvant chemotherapy provided a significant survival benefit to patients with gastroesophageal cancer (p = .005).29 Thus, the Spanish Society of Medical Oncology recommends neoadjuvant chemotherapy for patients with stage IB disease. The POET30 and CROSS31 studies assessed the role of neoadjuvant CRT for locally advanced gastric and gastroesophageal carcinoma. The pathologic complete response was significantly higher for patients in the CRT group, but OS did not significantly differ between groups.30,31 However, the meta-analysis by Sjoquist and colleagues supports an increase in OS for patients who have undergone CRT.29

current radiotherapy) with chemotherapy alone (XP every 3 weeks for 6 cycles) in patients with at least D2 lymphadenectomy and R0 resection. The long-term follow-up analysis showed no differences in the outcomes (PFS and OS; p = .527). However, a subgroup analysis showed that CRT significantly improved PFS (p = .04) in patients with node-positive, intestinal-type gastric cancer.40 Furthermore, the ARTIST II trial is expected to demonstrate the role of S-1 with or without oxaliplatin and radiotherapy. Currently, CRT has a limit role when surgery is D2 quality; and it is commonly reserved for patients with node-positive disease, insufficient lymphadenectomy, or questionable surgery precedence41.

Perioperative Chemotherapy

Advanced gastric cancer is a challenge for oncologists, especially because the clinical status of some patients is too poor to begin chemotherapy. However, chemotherapy is mandatory for patients with a good performance status to improve OS. Specifically, the combination of cisplatin and fluoropyrimidine is the treatment cornerstone for patients with HER2-negative disease. Both CF (cisplatin and 5-FU) and ECF can be considered standard combinations that extend OS in Southern Europe.16 However, head-to-head studies comparing the efficacy of these treatment modalities are not available. Furthermore, docetaxel, cisplatin, and 5-FU– based chemotherapy is considered a more effective option than CF, but it exhibits a worse toxicity profile.42 Regarding platinum toxicity and efficacy, some studies demonstrated that oxaliplatin can replace cisplatin because of its improved tolerability.3,43 Furthermore, trastuzumab in combination with chemotherapy significantly improved OS according to the ToGA trial (p = .0046), and this drug is also being used in Southern Europe in a metastatic setting for patients with HER2-positive disease.3,44 Conversely, irinotecan combined with 5-FU (FOLFIRI or IF) is an option for patients with chemotherapy-naive disease.23,45,46 However, less than 60% of patients receive second or later lines of therapy for gastric cancer in clinical practice.44,47 Therefore, first-line treatment should be maximized for these patients to attain clinical outcomes and quality of life. Nevertheless, several efforts are being made to develop tolerable drugs for patients with advanced, previously treated gastric cancer, such as antiangiogenic drugs. For example, ramucirumab showed significant efficacy as a monotherapy (REGARD) or in combination with pacl*taxel (RAINBOW). Specifically, the REGARD trial randomly assigned 445 patients to ramucirumab or placebo, and ramucirumab produced a significant OS benefit (5.2 vs. 3.8 months; HR, 0.77) over placebo.48 In addition, the RAINBOW trial randomly assigned 665 patients to ramucirumab plus pacl*taxel or pacl*taxel plus placebo, and the ramucirumab plus pacl*taxel arm showed a significantly superior OS (9.6 vs. 7.3 months; HR, 0.80, p = 0.017) over pacl*taxel monotherapy.49 More recently, apatinib was also shown to be superior to best supportive care in previously treated patients,47 and an Italian study50 showed that the combination of irinotecan and 5-FU is a manageable and acceptable regimen. Moreover, third-line

At many centers in Portugal, Spain, and most European countries, perioperative chemotherapy has been adopted as an interesting option for medically fit patients with resectable locally advanced (cT2 or higher, any N) distal esophageal, esophagogastric junction, or gastric tumors. The British MAGIC trial,32 the French FNLCC/FFCD 9703 study,33 and a meta-analysis34 have shown that perioperative chemotherapy significantly increases R0 rates and survival outcomes without significantly increasing perioperative complications or mortality. Moreover, the approach based on 3-cytotoxic agent combination also exhibits a quite tolerable grade 3 to 4 toxicity rate, however it is not normally used.34

Adjuvant Chemoradiotherapy/Chemotherapy

Based on the evidence of the INT-0116 trial and CALGB 80101 study, adjuvant CRT is indicated in patients with stage IB–IV (M0) resected gastric or gastroesophageal junction adenocarcinoma.35,36 The MacDonald regimen (CRT based on fluorouracil [5-FU]/leucovorin [LV]) improved progression-free survival (PFS; hazard ratio [HR], 1.52; p < .001) and OS (HR, 1.35; p = .005) compared with surgery alone. Furthermore, the CALGB study compared the INT-0116 regimen with epirubicin, cisplatin, and 5-FU (or ECF) before and after 5-FU/radiotherapy in resected gastroesophageal junction or gastric cancer without observing differences in the 3-year OS (52% and 50% for ECF and 5-FU/LV, respectively). The role of adjuvant trastuzumab for patients with HER2-positive disease is still being assessed in the phase II TOXAG trial. Nevertheless, patients with stage II or III gastric cancer submitted to D2-resection significantly benefited from S-1 at 1 year in the ACTS-GC randomized phase III trial (HR, 0.708; 95% CI, 0.510–0.983).37,38 However, the CLASSIC phase III trial39 also demonstrated that XELOX (capecitabine, oxaliplatin) significantly (p = .0015) benefited patients with stage II and III disease: the median 5-year PFS was 68% compared with 53%, and the estimated 5-year OS was 78% compared with 69% in the XELOX and the surgery-only groups, respectively.37-39 Currently, adjuvant approaches (chemotherapy vs. CRT) are controversial for patients with stage II-III D2 disease after resection. The ARTIST trial is an important study that compared CRT (cisplatin and capecitabine, or XP) with con264 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Metastatic Disease

GASTRIC CANCER AND HIGH-RISK AREAS

FOLFIRI has been shown to benefit patients with heavily pretreated disease without excessive toxicity. In particular, almost half of patients experienced disease control in their study.50 Furthermore, immunotherapy acquired an important role in gastric cancer. In January 2017, a randomised phase III study, NCT02267343, indicated that nivolumab (a human monoclonal IgG4 antibody which blocks the human PD-1 receptor) has superior survival (p < .0001) when compared to placebo in pretreated patients with advanced gastric cancer.51 Thus, it also could be considered a promising option for treat these patients in later lines.

FUTURE DIRECTIONS

Gastric cancer is a challenge to health care professionals worldwide, especially in high-risk areas. Moreover, the

landscape of gastric cancer is different in Southern Europe and Mediterranean countries than in other Central and Northern European countries. Environmental and genetic factors constitute an important background to understand the disease trajectory, mainly in the diagnosis and screening phases. However, well-implemented screening programs are lacking in these high-risk countries, and most patients consequently present with late-stage gastric cancer at diagnosis. Moreover, the eradication of H. pylori infection is important to decrease this manageable risk factor. Currently, novel target drugs and immune-checkpoints inhibitors could be promising options to prolong the outcomes of advanced patients with good quality of life. Further epidemiologic studies are warranted to improve the disease outcomes and implement preventive strategies.

References 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2015. CA Cancer J Clin. 2015;65:5-29. 2. Li Q, Zhang N, Jia Z, et al. Critical role and regulation of transcription factor FoxM1 in human gastric cancer angiogenesis and progression. Cancer Res. 2009;69:3501-3509. 3. de Mello RA, Marques AM, Araújo A. HER2 therapies and gastric cancer: a step forward. World J Gastroenterol. 2013;19:6165-6169. 4. Hansford S, Kaurah P, Li-Chang H, et al. Hereditary diffuse gastric cancer syndrome: CDH1 mutations and beyond. JAMA Oncol. 2015;1:23-32. 5. Oliveira C, Sousa S, Pinheiro H, et al. Quantification of epigenetic and genetic 2nd hits in CDH1 during hereditary diffuse gastric cancer syndrome progression. Gastroenterology. 2009;136:2137-2148. 6. de Mello RA, Costa BM, Reis RM, et al. Insights into angiogenesis in non-small cell lung cancer: molecular mechanisms, polymorphic genes, and targeted therapies. Recent Patents Anticancer Drug Discov. 2012;7:118-131. 7. Pimentel-Nunes P, Mourão F, Veloso N, et al. Long-term follow-up after endoscopic resection of gastric superficial neoplastic lesions in Portugal. Endoscopy. 2014;46:933-940. 8. Morais S, Ferro A, Bastos A, et al. Trends in gastric cancer mortality and in the prevalence of Helicobacter pylori infection in Portugal. Eur J Cancer Prev. 2016;25:275-281. 9. Bastos J, Peleteiro B, Barros R, et al. Sociodemographic determinants of prevalence and incidence of Helicobacter pylori infection in Portuguese adults. Helicobacter. 2013;18:413-422. 10. Santos A-C, Barros H. Prevalence and determinants of obesity in an urban sample of Portuguese adults. Public Health. 2003;117:430-437. 11. Lunet N, Pina F, Barros H. Regional trends in Portuguese gastric cancer mortality (1984-1999). Eur J Cancer Prev. 2004;13:271-275. 12. Choi ES, Lee H, Lee CH, et al. Overexpression of KLHL23 protein from read-through transcription of PHOSPHO2-KLHL23 in gastric cancer increases cell proliferation. FEBS Open Bio. 2016;6:1155-1164.

Wright NA (eds). Histopathological, Molecular, and Genetic Profile of Hereditary Diffuse Gastric Cancer: Current Knowledge and Challenges for the Future. Cham, Switzerland: Springer, 2016;371-391. 15. Brooks-Wilson AR, Kaurah P, Suriano G, et al. Germline E-cadherin mutations in hereditary diffuse gastric cancer: assessment of 42 new families and review of genetic screening criteria. J Med Genet. 2004;41:508-517. 16. Luis M, Tavares A, Carvalho LS, et al. Personalizing therapies for gastric cancer: molecular mechanisms and novel targeted therapies. World J Gastroenterol. 2013;19:6383-6397. 17. Suriano G, Yew S, Ferreira P, et al. Characterization of a recurrent germ line mutation of the E-cadherin gene: implications for genetic testing and clinical management. Clin Cancer Res. 2005;11: 5401-5409. 18. Rimm DL, Koslov ER, Kebriaei P, et al. Alpha 1(E)-catenin is an actinbinding and -bundling protein mediating the attachment of F-actin to the membrane adhesion complex. Proc Natl Acad Sci USA. 1995;92:8813-8817. 19. Moayyedi P, Feltbower R, Brown J, et al; Leeds HELP Study Group. Effect of population screening and treatment for Helicobacter pylori on dyspepsia and quality of life in the community: a randomised controlled trial. Lancet. 2000;355:1665-1669. 20. Parsonnet J, Friedman GD, Vandersteen DP, et al. Helicobacter pylori infection and the risk of gastric carcinoma. N Engl J Med. 1991;325:1127-1131. 21. O'Connor A, O'Morain CA, Ford AC. Population screening and treatment of Helicobacter pylori infection. Nat Rev Gastroenterol Hepatol. Epub 2017 Jan 5. 22. Salem A, Hashem S, Mula-Hussain LY, et al. Management strategies for locoregional recurrence in early-stage gastric cancer: retrospective analysis and comprehensive literature review. J Gastrointest Cancer. 2012;43:77-82.

13. Guilford P, Hopkins J, Harraway J, et al. E-cadherin germline mutations in familial gastric cancer. Nature. 1998;392:402-405.

23. Martin-Richard M, Custodio A, García-Girón C, et al. SEOM guidelines for the treatment of gastric cancer 2015. Clin Transl Oncol. 2015;17:996-1004.

14. van der Post RS, Gullo I, Oliveira C, et al. Stem Cells, Pre-neoplasia, and Early Cancer of the Upper Gastrointestinal Tract. In Jansen M and

24. Chun N, Ford JM. Genetic testing by cancer site: stomach. Cancer J. 2012;18:355-363.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 265

RAMON ANDRADE DE MELLO

25. Bass AJ, Thorsson V, Shmulevich I, et al; Cancer Genome Atlas Research Network. Comprehensive molecular characterization of gastric adenocarcinoma. Nature. 2014;513:202-209. 26. Ceccarelli G, Andolfi E, Biancafarina A, et al. Robot-assisted surgery in elderly and very elderly population: our experience in oncologic and general surgery with literature review. Aging Clin Exp Res. Epub 2016 Nov 30. 27. Gotoda T, Iwasaki M, Kusano C, et al. Endoscopic resection of early gastric cancer treated by guideline and expanded National Cancer Centre criteria. Br J Surg. 2010;97:868-871. 28. Schuhmacher C, Gretschel S, Lordick F, et al. Neoadjuvant chemotherapy compared with surgery alone for locally advanced cancer of the stomach and cardia: European Organisation for Research and Treatment of Cancer randomized trial 40954. J Clin Oncol. 2010;28:5210-5218. 29. Sjoquist KM, Burmeister BH, Smithers BM, et al; Australasian GastroIntestinal Trials Group. Survival after neoadjuvant chemotherapy or chemoradiotherapy for resectable oesophageal carcinoma: an updated meta-analysis. Lancet Oncol. 2011;12:681-692. Stahl M, Walz MK, Stuschke M, et al. Phase III comparison of 30. preoperative chemotherapy compared with chemoradiotherapy in patients with locally advanced adenocarcinoma of the esophagogastric junction. J Clin Oncol. 2009;27:851-856. Shapiro J, van Lanschot JJB, Hulshof MC, et al; CROSS study group. 31. Neoadjuvant chemoradiotherapy plus surgery versus surgery alone for oesophageal or junctional cancer (CROSS): long-term results of a randomised controlled trial. Lancet Oncol. 2015;16:1090-1098. Cunningham D, Allum WH, Stenning SP, et al; MAGIC Trial Participants. 32. Perioperative chemotherapy versus surgery alone for resectable gastroesophageal cancer. N Engl J Med. 2006;355:11-20. Ychou M, Boige V, Pignon J-P, et al. Perioperative chemotherapy 33. compared with surgery alone for resectable gastroesophageal adenocarcinoma: an FNCLCC and FFCD multicenter phase III trial. J Clin Oncol. 2011;29:1715-1721. Ronellenfitsch U, Schwarzbach M, Hofheinz R, et al. Preoperative 34. chemo(radio)therapy versus primary surgery for gastroesophageal adenocarcinoma: systematic review with meta-analysis combining individual patient and aggregate data. Eur J Cancer. 2013;49:31493158. 35. Macdonald JS, Smalley SR, Benedetti J, et al. Chemoradiotherapy after surgery compared with surgery alone for adenocarcinoma of the stomach or gastroesophageal junction. N Engl J Med. 2001;345:725730. 36. Smalley SR, Benedetti JK, Haller DG, et al. Updated analysis of SWOG-directed intergroup study 0116: a phase III trial of adjuvant radiochemotherapy versus observation after curative gastric cancer resection. J Clin Oncol. 2012;30:2327-2333. 37. Sakuramoto S, Sasako M, Yamaguchi T, et al; ACTS-GC Group. Adjuvant chemotherapy for gastric cancer with S-1, an oral fluoropyrimidine. N Engl J Med. 2007;357:1810-1820. 38. Sasako M, Sakuramoto S, Katai H, et al. Five-year outcomes of a randomized phase III trial comparing adjuvant chemotherapy with S-1 versus surgery alone in stage II or III gastric cancer. J Clin Oncol. 2011;29:4387-4393.

266 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Noh SH, Park SR, Yang H-K, et al; CLASSIC trial investigators. Adjuvant 39. capecitabine plus oxaliplatin for gastric cancer after D2 gastrectomy (CLASSIC): 5-year follow-up of an open-label, randomised phase 3 trial. Lancet Oncol. 2014;15:1389-1396. 40. Park SH, Sohn TS, Lee J, et al. Phase III trial to compare adjuvant chemotherapy with capecitabine and cisplatin versus concurrent chemoradiotherapy in gastric cancer: final report of the adjuvant chemoradiotherapy in stomach tumors trial, including survival and subset analyses. J Clin Oncol. 2015;33:3130-3136. Ajani JA, D'Amico TA, Almhanna K, et al. Gastric cancer, version 3.2016, 41. NCCN clinical practice guidelines in oncology. J Natl Compr Canc Netw. 2016;14:1286-1312. 42. Van Cutsem E, Moiseyenko VM, Tjulandin S, et al; V325 Study Group. Phase III study of docetaxel and cisplatin plus fluorouracil compared with cisplatin and fluorouracil as first-line therapy for advanced gastric cancer: a report of the V325 Study Group. J Clin Oncol. 2006;24:49914997. 43. Cunningham D, Starling N, Rao S, et al; Upper Gastrointestinal Clinical Studies Group of the National Cancer Research Institute of the United Kingdom. Capecitabine and oxaliplatin for advanced esophagogastric cancer. N Engl J Med. 2008;358:36-46. 44. Bang Y-J, Van Cutsem E, Feyereislova A, et al; ToGA Trial Investi gators. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastro-oesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet. 2010;376:687-697. 45. Dank M, Zaluski J, Barone C, et al. Randomized phase III study comparing irinotecan combined with 5-fluorouracil and folinic acid to cisplatin combined with 5-fluorouracil in chemotherapy naive patients with advanced adenocarcinoma of the stomach or esophagogastric junction. Ann Oncol. 2008;19:1450-1457. 46. Shah MA. Update on metastatic gastric and esophageal cancers. J Clin Oncol. 2015;33:1760-1769. 47. de Mello RA, de Oliveira J, Antoniou G. Angiogenesis and apatinib: a new hope for patients with advanced gastric cancer? Future Medicine. 2016;13:295-298. 48. Fuchs CS, Tomasek J, Yong CJ, et al; REGARD Trial Investigators. Ramucirumab monotherapy for previously treated advanced gastric or gastro-oesophageal junction adenocarcinoma (REGARD): an international, randomised, multicentre, placebo-controlled, phase 3 trial. Lancet. 2014;383:31-39. Wilke H, Muro K, Van Cutsem E, et al; RAINBOW Study Group. 49. Ramucirumab plus pacl*taxel versus placebo plus pacl*taxel in patients with previously treated advanced gastric or gastro-oesophageal junction adenocarcinoma (RAINBOW): a double-blind, randomised phase 3 trial. Lancet Oncol. 2014;15:1224-1235. 50. Pasquini G, Vasile E, Caparello C, et al. Third-line chemotherapy with irinotecan plus 5-fluorouracil in caucasian metastatic gastric cancer patients. Oncology. 2016;91:311-316. 51. Kang YK, Satoh T, Ryu MR, et al. Nivolumab (ONO-4538/BMS-936558) as salvage treatment after second or later-line chemotherapy for advanced gastric or gastro-esophageal junction cancer (AGC): a double-blinded, randomized, phase III trial. J Clin Oncol. 2017;35 (suppl; abstr 2).

DEPLOYING IMMUNOTHERAPY IN PANCREATIC CANCER

Deploying Immunotherapy in Pancreatic Cancer: Defining Mechanisms of Response and Resistance Gregory L. Beatty, MD, PhD, Shabnam Eghbali, and Rebecca Kim OVERVIEW The immune reaction to pancreatic ductal adenocarcinoma (PDAC) is a strong prognostic determinant of clinical outcomes and may be a promising therapeutic target. We use multiplex immunohistochemistry to illustrate distinct patterns of T-cell and myeloid cell infiltration seen in PDAC that have therapeutic implications and discuss the current state of immunotherapy in this disease. Based on collective findings from clinical and preclinical studies, two conceptual models have emerged for applying immunotherapy in PDAC that involve (1) restoring elements of T-cell immunosurveillance and (2) redirecting myeloid cells to condition tumors with increased sensitivity to cytotoxic therapies. Overall, the success of immunotherapy in PDAC will most likely rely on strategic combinations of therapies that are informed by well-designed correlative analyses that consider the spatial heterogeneity of immune responses detected in malignant tissues.

herapeutic resistance to standard cytotoxic therapies (e.g., radiation and chemotherapy) is a hallmark of PDAC, a disease for which the 5-year overall survival rate has remained below 10% for the past two decades despite considerable efforts to improve clinical outcomes.1 The recent advent of immunotherapy has brought new hope to this disease with the possibility of redirecting the immune system to recognize and eliminate malignant cells. However, PDAC is capable of orchestrating several mechanisms of immune escape that can thwart the therapeutic potential of immunotherapy.2 Thus, unlike many solid malignancies for which a sizable subset of patients have demonstrated remarkable responsiveness to immunotherapy, PDAC has shown striking resistance. Nonetheless, lessons learned from preclinical models and several clinical trials investigating immunotherapy in PDAC have provided critical insight into strategies to circumvent immune resistance in this disease. This knowledge is now guiding the next generation of immunotherapy in PDAC with an emphasis on rationally designed and novel treatment combinations.

THE IMMUNE REACTION TO PDAC

The tumor microenvironment is a critical determinant of treatment resistance. In PDAC, this microenvironment is commonly marked by a dense fibrotic reaction with recruitment of fibroblasts and leukocytes, which together can impede the efficacy of therapeutics.3 The leukocyte infiltrate seen in PDAC is complex and heterogeneous. Myeloid cells are the most prominent component of this

infiltrate and are associated with a worse prognosis in patients with surgically resected PDAC.4,5 In contrast, CD3+ T-cell infiltration has been found in some studies to correlate with improved overall survival for at least a subset of patients with resected PDAC.4,6 Other reports, though, have failed to demonstrate a correlation between T-cell density and overall survival, suggesting that for many patients, the mere presence of T cells may have little clinical significance.7,8 This is consistent with the notion that the type and location of T cells within tumors is equally as important as their density.9 The location of T cells in PDAC may inform resistance mechanisms to productive T-cell immunosurveillance. For example, CD3+ T-cell infiltrates are more prominent in regions of chronic pancreatitis than in PDAC.7 In addition, CD3+ T cells have been identified more commonly at the invasive front of PDAC with fewer cells detected in the center, suggesting mechanisms of T-cell exclusion orchestrated by malignant cells.4 Within primary resected PDAC tumors, CD3+ T-cell infiltrates have been found associated with tertiary lymphoid structures which include B cells, dendritic cells, and other immune cell populations.10,11 Tumor-infiltrating CD3+ T cells also cluster adjacent to nests of malignant cells (Fig. 1A) and are commonly found diffusely scattered and trapped within stromal tissue (Fig. 1B). In contrast, direct interaction between CD3+ T cells and malignant cells is infrequent (Fig. 1C and D). Thus, although CD3+ T-cell infiltrates can be found in the majority (approximately 75%) of resected primary PDAC tumors, they appear to be trapped within the stroma.10,12

From the Abramson Cancer Center, University of Pennsylvania, Philadelphia, PA; Department of Medicine, Division of Hematology-Oncology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA; Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Gregory L. Beatty, MD, PhD, Abramson Cancer Center of the University of Pennsylvania, Smilow Center for Translational Research, Room 8-112, 3400 Civic Center Blvd., Building 421, Philadelphia, PA 19104; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 267

BEATTY, EGHBALI, AND KIM

FIGURE 1. Patterns of T-Cell Infiltration in Human PDAC

Shown are representative images of CD3+ T cells (purple) seen clustering (A, red arrows) adjacent to CK19+ tumor cells (shown in brown) and trapped (B, red arrows) in stromal tissue adjacent to CK19+ tumor cells. Insets show higher magnification. CD3+ (C, purple) and CD8+ (D, brown) T cells are seen adjacent to but excluded from interacting with tumor cells. Dotted red line illustrates stromal barrier between T-cell infiltrates and malignant cells. Nuclear staining (light blue) is illustrated by hematoxylin counterstain. Immunohistochemical staining was performed on surgically resected human PDAC specimens using the Ventana Discovery Ultra automated staining system. T, tumor.

The presence of tertiary lymphoid structures in surgically resected PDAC has led to the suggestion that this disease may be more immunogenic than previously appreciated.10

KEY POINTS • CD3+ T-cell infiltrates are associated with favorable clinical outcomes in pancreatic ductal adenocarcinoma (PDAC), whereas myeloid cell recruitment to tumors portends a poor prognosis. • Multiplex immunohistochemistry can be used to define the quality and quantity of distinct patterns of CD3+ T cell infiltration in PDAC including formation of tertiary lymphoid aggregates, clustering adjacent to tumor cell nests, trapping within stromal tissue, and interaction with malignant cells. • Strategies designed to harness the potential of T cells for the treatment of PDAC may need to address a state of functional paralysis associated with tumor-infiltrating CD3+ T cells. • Clinical experience with immunotherapy and preclinical modeling suggests a need for rationally designed combination treatment strategies that consider elements of immunosuppression imparted by the tumor microenvironment. • Two conceptual models for applying immunotherapy to PDAC have emerged: restoring elements of T-cell immunosurveillance and redirecting the myeloid reaction to PDAC for enhancing the efficacy of cytotoxic therapies. 268 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

These lymphoid structures are detected by immunohistochemistry in most samples analyzed via serial tissue sectioning.10 Moreover, characterization of the T-cell receptor repertoire in primary PDAC by deep sequencing analysis has found that the majority of the T-cell infiltrate is represented by only a few T-cell clones.10 However, it is unknown whether these clones are confined to tertiary lymphoid structures or represent true tumor-infiltrating T cells. As T-cell entry into tumors is via the bloodstream and not lymphatics, the antitumor potential of T cells found in tertiary lymphoid structures is uncertain. Indeed, the clonal repertoire of T cells detected in tertiary lymphoid structures, as defined by Vβ T-cell receptor expression, appears to be distinct from T cells detected infiltrating the tumor stroma.10 As a result, analysis of the immune microenvironment of PDAC using techniques (e.g., RNA and flow cytometry) that do not consider the spatial location of T cells within tumor tissue may overestimate the quality and quantity of the lymphocyte infiltrate in PDAC.13-15 Nonetheless, despite their unclear role in regulating PDAC biology, tertiary lymphoid structures are associated with a more favorable prognosis in surgically resected PDAC.16,17 Expansion of tumor-infiltrating CD3+ T cells isolated from surgically resected PDAC tumors has showed that at least a subset of T cells in PDAC has tumor reactivity.10,18 However, the frequency of the expanded T cells displaying tumor reactivity is usually low (less than 1%).10 In addition, within the majority of PDAC specimens (more than 80%), CD3+ T cells show a considerable decrease or loss of CD3zeta chain

DEPLOYING IMMUNOTHERAPY IN PANCREATIC CANCER

expression, which is required for T-cell receptor signaling and activation.12 A notable decrease in CD3ζ expression on T cells has also be detected in nearly half (9 of 19) of peritumoral lymph nodes analyzed and is seen in the peripheral blood of patients with metastatic PDAC compared with healthy control subjects.12,19 Decreased CD3zeta expression correlates with limited T-cell capacity to secrete cytokines, in particular interferon (IFN)-gamma.19 Moreover, consistent with the notion of poor T-cell activation in PDAC, genomic profiling of human PDAC tumors suggests that despite potentially targetable neoantigens being present in nearly all PDAC samples, albeit at much lower levels than seen in other tumors such as melanoma, T cells detected in tumor tissue lack an activation gene signature.13 Thus, strategies designed to harness the potential of T cells for the treatment of PDAC must address a state of functional paralysis associated with tumor-infiltrating and circulating CD3+ T cells in this disease.

CLINICAL EXPERIENCE WITH IMMUNOTHERAPY IN PANCREATIC CANCER

The success of immunotherapy is reliant on activating potent and durable T-cell immunity against PDAC. T-cell immunosurveillance is dependent on elements of the cancer immunity cycle that proposes that tumors harbor unique antigens capable of being recognized by T cells and that these antigens, when appropriately presented by antigen-presenting cells, can prime and activate T cells to infiltrate tumors, where they then recognize and eliminate malignant cells.2,20 Multiple clinical-grade therapeutics are available for bolstering elements of the cancer immunity cycle (Fig. 2). The most extensively evaluated approach in PDAC has involved the use of vaccines (Table 1). Vaccines can induce T-cell responses against PDAC. In a phase I study, vaccination with GVAX, an irradiated allogeneic whole tumor cell vaccine expressing granulocyte-

macrophage colony-stimulating factor (GM-CSF), stimulated delayed-type hypersensitivity responses to autologous tumor cells in a subset of patients and induced the development of mesothelin-specific CD8+ T cells that correlated with improved disease-free survival.21,44 Based on this finding, GVAX combined with low-dose cyclophosphamide (Cy), as a strategy to deplete regulatory T (Treg) cells, was subsequently tested with or without an attenuated Listeria-based mesothelin vaccine (CRS-207) in a prime/boost strategy intended to stimulate and maintain tumor-specific immunity.24 In this phase II study, Cy plus GVAX followed by CRS-207 compared with GVAX alone was associated with improved overall survival (6.1 vs. 3.9 months) in patients with previously treated metastatic PDAC. Moreover, the development of mesothelin-specific CD8+ T cells in response to treatment was associated with longer overall survival. These promising results led to a phase IIB study comparing Cy/GVAX plus CRS-207 with CRS-207 alone or chemotherapy alone in patients with previously treated PDAC. However, the combination of Cy/GVAX plus CRS-207 produced no survival benefit over chemotherapy alone. Similarly, algenpantucel-L, an irradiated allogeneic tumor cell vaccine expressing murine alpha-1,3-galactosyltransferase, was recently found in a phase III study of patients with resected PDAC to not significantly impact overall survival when combined with standard of care versus standard of care only.23 A large phase III study evaluating sequential or simultaneous telomerase peptide vaccination in combination with chemotherapy in patients with locally advanced or metastatic PDAC also demonstrated no notable improvement in overall survival with chemoimmunotherapy with a potential trend toward worse outcomes in the sequentially treated group.35 Limitedefficacy data have also been observed for a range of other tumor-specific peptide-based vaccines despite their capacity to stimulate tumor-specific T-cell immune responses

FIGURE 2. Immunotherapeutic Strategies for the Treatment of PDAC

Multiple therapeutic options with clinical-grade agents capable of restoring distinct elements of the cancer immunity cycle exist for the treatment of PDAC.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 269

BEATTY, EGHBALI, AND KIM

TABLE 1. Completed Phase II/III Clinical Trials of Immunotherapy in PDAC Treatment

Trial Phase

Setting

No. of Patients

ORR, N (%)

Median PFS (Months)

Median OS (Months)

GVAX/CRT

Stage I-II res

17.3

24.8

Algenpantucel-L + CRT following surgery

Stage I-II res

14.1*

Chemo

III

Stage I-II res

722

30.4

Met

303

Chemo + algenpantucel-L Cy/GVAX + CRS-207

27.3 6.1

Cy/GVAX Chemo

Reference

3.9 4.6

CRS-207

5.4

CRS-207 + Cy/GVAX

3.8

Postoperative K-ras vaccine

I/II

Res

27.5

KIF20A/VEGFR1,2/Gem

Stage III-IV

8 (12.1)

4.7–5.2

9.0–10.0

PPV

Stage IV

7.9

KIF20A

I/II

UR, met

8 (25.8)

1.8

4.7

Ras

I/II

Ras

Stage II-III

36+ (PDAC)

47+ (PDAC)

Ras + GM-CSF

I/II

Res or LA

1 (2)

25.6

Gem

II/III

UR or met

159

3.71 (active)

8.36 (active)

3.75 (placebo)

8.54 (placebo)

Antisense oligo TGF-β2

I/II

1 (2.7)

Gem/capecitabine

III

UR LA, met, res

1,062

63 (18)

6.4

7.9

31 (9)

4.5

6.9

Gem + VEGFR2

+ GV1001 sequential + GV1001 concurrent

55 (16)

6.6

8.4

Anti–CTLA-4 (ipilimumab)

LA or met

CIK

Met

2.75

6.65

CapCell + ifosfamide

I/II

UR (stage II-IV)

4 (28.6)

10.25

III

Res

11.5*

28.5

39,40

15.2

26.5

14.2

18.8

FU + Cis + IFN-α-2b + RT Surgery

64 NS

Stage III

Surgery + chemo

21.7

Surgery + chemo + IL-2

Gem

110

I/II

Gem + IMM-101 I(131) KAb201

27.52

31.07

(2.9)

2.4

5.6

(10.7)

4.1

6.7

1 (6)

5.2

Abbreviations: CapCell, encapsulated CYP2B1; chemo, chemotherapy; CIK, cytokine-induced killer cells; cis, cisplatin; CRT, chemoradiation therapy; FU, 5-fluorouracil; gem, gemcitabine; I(131) KAb201, radiolabeled anti-CEA antibody; KIF20A, Rab6-binding kinesin-derived peptide; IFN, interferon; IL, interleukin; LA, locally advanced; met, metastatic; NR, not reported; NS, not specified; ORR, overall response rate; OS, overall survival; PFS, progression-free survival; PPV, personalized peptide vaccine; res, resected; TGF, transforming growth factor; UR, unresectable. *Disease-free survival reported.

(Table 1). Together, these data collectively demonstrate the capacity of vaccines to elicit tumor-specific T-cell responses in patients that are in some cases associated with improved clinical outcomes, but suggest the importance of immune resistance mechanisms that regulate the efficacy of T-cell immunosurveillance in this disease. T-cell effector activity is an exquisitely regulated process that is controlled by a balance of positive and negative signaling pathways. In this regard, immune checkpoint molecules 270 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

(e.g., CTLA-4 and PD-1/PD-L1) provide negative signals to T cells to limit their expansion and effector activity in tissues.45 Disruption of these signals using blocking antibodies has demonstrated remarkable success in stimulating antitumor T-cell immunity in a subset of patients across a wide range of malignancies.46 However, for patients with PDAC, treatment with checkpoint inhibitors against CTLA-4, PD-L1, and Lag-3 has not produced considerable clinical activity.36,47 Interestingly, PD-L1 transcriptional expression in PDAC,

DEPLOYING IMMUNOTHERAPY IN PANCREATIC CANCER

unlike other solid malignancies, is correlated with worse disease outcomes in patients with surgically resected PDAC.48 Nonetheless, emerging evidence suggests that a subset of patients with cancer with mismatch repair (MMR) deficiency may be particularly sensitive to PD-1/PD-L1 blockade.49 However, while this population may represent as much as 15–20% of PDAC patients,50-52 it is noteworthy that the predictive value of MMR deficiency, as seen in colorectal cancer, is correlated with an enhanced quality and density of tumor-infiltrating T cells.49,53 It is currently unclear, though, whether a similar correlation will be observed in PDAC. Thus, ongoing studies are investigating the prospect of PD-1–blocking antibodies for this population. Nonetheless, for the majority of patients with PDAC, exceptional resistance to checkpoint immunotherapy has emerged as a common theme. The ability of vaccines to modulate the immune microenvironment in PDAC by stimulating the formation and activation of tertiary lymphoid aggregates in tumor tissue has suggested that immune checkpoint inhibitors may enhance the efficacy of vaccines.11 Combining CTLA-4–blocking antibodies with chemotherapy and vaccines, though, has been largely ineffective in PDAC.54-56 However, multiple studies are underway that seek to provide further insight into the capacity of vaccines and chemotherapy to combine with immune checkpoint inhibitors as a strategy to convert PDAC from immunoresistant to immunosensitive (Table 2). The ultimate goal is to sequentially restore major elements of the cancer immunity cycle by layering in therapies that can stimulate tumor-specific T-cell expansion, activation, trafficking, and effector activity. An alternative approach to stimulating tumor-specific T-cell activity in patients is to adoptively transfer tumor-reactive T cells. This approach bypasses the need for in vivo T-cell priming and allows for assessment of downstream mechanisms that may regulate T-cell infiltration and effector activity within tumors.57 For example, engineering T cells with a chimeric antigen receptor (CAR) that recognizes mesothelin, which is expressed on the surface of malignant cells, can yield potent major histocompatibility complex (MHC)–independent cytolytic activity in vitro against autologous PDAC tumor cells.58 In PDAC, expression of MHC class I molecules is frequently altered in primary and metastatic lesions with poor infiltration by T cells seen in MHC class I–negative tumors.8 Thus, CARs offer a unique strategy for overcoming elements of immune escape mediated by downregulation or loss of antigen processing and presentation machinery in tumor cells. CARs combine the protein recognition capacity of antibodies with intracellular stimulatory components of the T-cell receptor to redirect T cells against a tumor-specific target protein.57 However, clinical benefit with CAR T cells in PDAC has thus far been limited, despite evidence for potential clinical activity.59 For instance, one patient with metastatic PDAC responded to CAR T-cell therapy with a complete metabolic response in the liver detected on fluorodeoxyglucose-PET/CT imaging, but ultimately succumbed to disease because of

progression of their primary pancreatic lesion. Similar mixed clinical responses have been seen in patients with advanced-stage PDAC treated with intravenous infusions of MUC1-specific lymphocytes in combination with intradermal vaccination using MUC1-expressing dendritic cells.60 The finding of mixed tumor responses seen in these studies implies tumor lesion heterogeneity but also suggests that T cells may be capable of producing, in some cases, remarkable clinical activity. Chemotherapy alone may also alter the immune microenvironment in PDAC. In patients with surgically resectable disease, preoperative therapy with chemoradiotherapy or chemotherapy has shown the capacity to decrease myeloid cell and Treg infiltrates, leading to an increase in the ratio of CD8 T cells to Treg cells.15 Similarly, a separate study showed that neoadjuvant chemoradiotherapy can increase T-cell infiltrates, which is a stronger predictor of long-term outcomes than pathologic response to treatment.61 To enhance the immunostimulatory capacity of chemotherapy, cytokines such as IFN-alpha-2b have been investigated in the adjuvant setting but have not been found to produce noteworthy improvement in clinical outcomes.39 The use of molecularly targeted therapies to disrupt signaling pathways may enhance cancer immunogenicity. For example, an early-phase clinical trial has suggested the capacity of MEK inhibitors to uncover therapeutic benefit with PD-L1 checkpoint blockade in patients with MMR-proficient colorectal cancer.62 The potential immune-enhancing effects of inhibiting the MEK pathway is suggested by preclinical studies showing enhanced expression of cancer differentiation antigens and MHC expression in the setting of MEK inhibition.63,64 Similarly, in preclinical models of PDAC, MAPK has been found to regulate PD-L1 expression on malignant cells which can inhibit CD8 T cell antitumor activity.65 In this model system, combining MAPK inhibition with PD-1 blockade produced T-cell–dependent antitumor immunity. Epigenetic modulation may also have a significant role in defining the immunogenicity of cancer cells.66 To this end, epigenetic modifiers, such as inhibitors of DNA methyltransferase (DNMT) or histone deacetylase (HDAC), have been found to enhance cancer cell immunogenicity through increased MHC expression and in preclinical models, combine with checkpoint immunotherapy to produce increased T cell–dependent antitumor activity.67 Thus, incorporating molecular targeted therapies to improve cancer cell immunogenicity may be a promising therapeutic avenue for shifting PDAC from immune-resistant to immune-sensitive. Overall, findings from clinical trials investigating immunotherapy in PDAC suggest that many patients, although not all, respond to vaccines by eliciting tumor-specific T-cell responses. However, the productivity of vaccine-induced T cells as well as adoptively transferred T cells has been marginal, and based on histologic analyses of tissues showing discreet patterns of infiltration and activation, attention has turned to the tumor microenvironment as a major determinant and barrier to the efficacy of T-cell immunotherapy in PDAC. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 271

BEATTY, EGHBALI, AND KIM

TABLE 2. Immune Targets Under Active Clinical Investigation in Pancreatic Cancer Target Category

Target

Target Agent (Active Clinical Trials)

Checkpoint molecules

CTLA-4

Tremelimumab (NCT02311361, NCT02639026) Ipilimumab (NCT01896869)

PD-1/PD-L1

Pembrolizumab (NCT02713529, NCT02331251, NCT02305186, NCT01174121) MEDI4736 (NCT02311361, NCT02639026, NCT02826486) Nivolumab (NCT02451982, NCT02423954)

Myeloid recruitment

Immune-suppressive molecules

Vaccines

Immune agonists Adoptive cell therapy Cytokines

Other

B7-H3

Enoblituzumab (NCT02475213)

B7-H3 × CD3

MGD009 (NCT02628535)

IDO

Indoximod (NCT02077881)

CSF-1R

AMG820 (NCT02713529)

BTK

Ibrutinib (NCT02562898)

CXCR4

BL-8040 (NCT02826486)

PI3K

INCB050465 (NCT02646748)

FAK

Defactinib (NCT02546531)

JAK

INCB039110 (NCT02646748)

Whole tumor cell vaccine

GVAX (NCT02451982, NCT01896869)

Kras

TG01 (NCT02261714)

p53

P53MVA (NCT02432963)

WT1

DSP-7888 (NCT02498665)

p97

CB-5083 (NCT02243917)

hTERT

INO-1400 (NCT02960594)

MUC16

DMCU4064A (NCT02146313)

VEGFR2

Ramucirumab (NCT02581215)

CA-125

Oregovomab (NCT01959672)

CD40

RO7009789 (NCT02588443)

CD40/4-1BBL oncolytic virus

LOAd703 (NCT02705196)

Anti-PSCA CAR

BPX-601 (NCT02744287)

Anti-mesothelin CAR

NCT01583686

IL-12

INO-9012 (NCT02960594)

IL-15

ALT-803 (NCT02559674)

IL-2

Aldesleukin (NCT01174121)

IL-10

AM0010 (NCT02009449, NCT02923921)

Young TIL

NCT01174121

Abbreviations: hTERT, human catalytic reverse transcription subunit of telomerase; IDO, indoleamine 2,3-dioxygenase; LOAd703, oncolytic adenovirus; P53MVA, modified vaccinia virus Ankara vaccine expressing p53; PI3K, phosphatidylinositide 3-kinase; TIL, tumor-infiltrating lymphocyte.

PRECLINICAL MODELING TO GUIDE THE APPLICATION OF IMMUNOTHERAPY IN PDAC The development of genetic mouse models that recapitulate salient features of human PDAC, including the immune reaction, offer an opportunity to rapidly study the tumor microenvironment, novel therapeutic targets, and combination treatment regimens.68,69 In these genetic mouse models, similar to human disease, CD3+ T-cell infiltrates can be detected in lymphoid aggregates adjacent to tumors (Fig. 3A) and are sometimes found diffusely scattered within tumor tissue (Fig. 3B), but infrequently seen to interact directly with malignant cells (Fig. 3C). In contrast, myeloid cells are a common component of the immune reaction to both 272 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

mouse and human PDAC and can be found in close contact with malignant cells (Fig. 4). Moreover, inducing pancreatic inflammation pharmacologically70-72 or with radiotherapy73 has been shown to drive pancreatic cancer development and accelerate tumor progression, implying that the myeloid reaction to PDAC can have a protumor role. Depletion of myeloid cell subsets in genetic mouse models of PDAC has been shown to alter the immune dynamics in tumors with increased CD3+ T-cell infiltration.74 In addition, pharmacologic inhibition of macrophages has been suggested as a strategy to inhibit metastasis formation.75 Disrupting myeloid cell recruitment to tumors can also alter tumor sensitivity to cytotoxic therapies. For example, inhibiting

DEPLOYING IMMUNOTHERAPY IN PANCREATIC CANCER

FIGURE 3. Patterns of T-Cell Infiltration in Murine PDAC Tumors

Shown are representative immunohistochemical images showing low-power (A) and high-power (B) magnifications of CD3+ T cells (purple) detected in lymphoid aggregates adjacent to CK19+ PDAC cells (brown). (C) CD3+ T cells (purple) are confined to lymphoid structures demarcated by Lyve-1 (dark blue). Shown are low-power (D) and high-power (E) magnification images of CD3+ T cells (purple) seen trapped in stromal tissue adjacent to CK19+ PDAC cells (brown) and rare direct cell-cell interaction between CD3+ T cells (purple) and malignant CK19+ cells (brown; F). Nuclear staining (light blue) is illustrated by hematoxylin counterstain. Immunohistochemical staining was performed on PDAC specimens obtained from KrasG12D/+; Trp53R172H/+; Pdx-1Cre mice using the Ventana Discovery Ultra automated staining system.

myeloid cell infiltration by antagonizing the CCL2/CCR2 pathway has been found to increase the efficacy of chemotherapy76,77 and radiotherapy78 in mouse models of PDAC. Blockade of macrophage colony-stimulating factor using

neutralizing antibodies has also been shown to decrease myeloid accumulation in PDAC and in doing so, enhance the efficacy of radiotherapy.73 CXC chemokines, involved in the recruitment of neutrophils and immature myeloid cells, are

FIGURE 4. Patterns of Myeloid Cell Infiltration in Human PDAC

Low-power (A) and high-power (B) magnification images showing CD15+ granulocytes (brown) surrounding tumor cells (yellow T). (C) CD14+ macrophages (brown) are seen closely interacting with the periphery of a nest of tumor cells (yellow T). Red arrows indicate cellular localization patterns in A–C. (D) CD14+ macrophages (purple) are seen to encompass CK19+ malignant cells (orange). CD14+ macrophage (purple) recruitment to CK19+ tumor cell structures (brown) is heterogeneous within tumor tissues, with some tumor clusters showing rare single cells excluded from interacting with malignant cells and other tumor clusters (E) showing robust macrophage infiltration and close interaction with PDAC cells (F). The black arrows in E and F mark tumor cells, and the red arrows mark macrophages to illustrate the distance separating macrophages and tumors in each image. Nuclear staining (light blue) was detected by hematoxylin counterstain. Immunohistochemical staining was performed on surgically resected human PDAC specimens using the Ventana Discovery Ultra automated staining system.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 273

BEATTY, EGHBALI, AND KIM

also detected at increased levels in both mouse and human PDAC. Chemokine (C-X-C motif) receptor (CXCR) 2–dependent recruitment of neutrophils and myeloid cell progenitors has been found to limit the efficacy of cytotoxic chemotherapy and immune checkpoint inhibition with PD-1–blocking antibodies in mouse models of PDAC.79 Promising early phase clinical results investigating small-molecule inhibitors of CCR2 in patients with both borderline resectable/locally advanced80 and unresectable81 disease support a role for inhibiting myeloid cell recruitment to tumor tissues for improving the efficacy of cytotoxics in PDAC. Although blocking myeloid recruitment to tumors is one strategy for shifting the immune reaction in PDAC from tumor-promoting to tumor-inhibitory, the biology of myeloid cells is inherently pliable such that under the appropriate conditions, they can also acquire antitumor properties.82 This biology creates an opportunity to leverage myeloid cell recruitment to tumors for potential therapeutic benefit. For example, CD40 agonists have been found to impart antitumor and antifibrotic properties on tumor-infiltrating myeloid cells in vivo.83,84 This myeloid-dependent antifibrotic activity induced by CD40 agonists can shift PDAC tumors from resistant to chemotherapy to sensitive to chemotherapy, thereby implying a potential role for immunotherapy in enhancing the efficacy of cytotoxic therapies. CD40 agonists are best appreciated for their capacity to license antigen-presenting cells with T-cell stimulatory properties.85 In mouse models of spontaneously arising PDAC, CD40 agonists can reverse functional T-cell paralysis detected in lymphoid structures adjacent to tumor tissue.83 However, the capacity of CD40 agonists to restore productive T-cell immunosurveillance is exquisitely regulated by macrophages residing outside of the tumor microenvironment.86 This finding implies that strategies to reverse elements of immune suppression imposed by myeloid cells may be critical to the success of T-cell immunotherapy. Inhibition of myeloid cell activation through targeting of Bruton’s tyrosine kinase (BTK) has been found to inhibit tumor progression.87 Moreover, combining BTK inhibition with chemotherapy produces T-cell–dependent tumor regressions in an orthotopic model of PDAC.87 T-cell–dependent antitumor activity can also be induced in orthotopic models when chemotherapy is combined with inhibitors of CCR2, CSF1R, and CXCR2, which all limit myeloid accumulation in tumors.73,77,79 These findings suggest that redirecting or disrupting the myeloid reaction in tumors may be a cardinal feature for the success of T-cell immunotherapy in PDAC. The inflammatory reaction that surrounds PDAC is likely directed, at least in part, by tumor cell-intrinsic signaling pathways. For example, Kras activation can drive the release of factors (e.g., interleukin [IL]-8 and GM-CSF) that induce the recruitment and accumulation of myeloid cells in tumor tissue.88,89 In addition, hyperactivation of focal adhesion kinase (FAK) activity in malignant cells has been found to stimulate the release of chemoattractants involved in myeloid cell recruitment.90 Disruption of FAK signaling genetically or pharmacologically delays tumor progression, reduces 274 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

fibrosis, and decreases myeloid cell accumulation in mouse models of PDAC. In patients with resected PDAC, FAK activity has been associated with decreased CD8+ T-cell infiltration.90 Moreover, FAK inhibition can enhance the efficacy of both chemotherapy and PD-1 antagonists in mice with spontaneously arising PDAC tumors.90 Overall, genetic mouse models have suggested that the tumor microenvironment is a key determinant of T-cell efficacy in PDAC. However, whether the microenvironment acts as a physical barrier to T-cell infiltration is less clear. Adoptive transfer studies in a genetic mouse model of PDAC has shown that tumor-reactive T cells can effectively penetrate the fibrotic matrix that surrounds malignant cells.91 However, the functional capacity of tumor-infiltrating T cells is fleeting and associated with upregulation of multiple negative regulatory molecules.91 In essence, tumor-infiltrating T cells may ultimately become trapped in the stromal compartment, as is also seen in human disease.12 Elements of the stroma, including fibroblasts, have been implicated in this active sequestration of T cells away from malignant cells through secretion of chemokines including CXCL12.92,93 Inhibiting the interaction of CXCL12 with its receptor, CXCR4, which can be found on T cells among many other cell types including myeloid cells, has shown potential to restore T-cell infiltration and, in doing so, uncover therapeutic activity with immune checkpoint inhibitors including CTLA-4 and PD-L1–blocking antibodies.92 Thus, elements of the tumor microenvironment including fibroblasts and myeloid cells possess properties capable of inhibiting the efficacy of immunotherapy in PDAC.

STRATEGICALLY APPLYING IMMUNOTHERAPY TO PDAC

Unlike other malignancies in which monotherapy with immune checkpoint inhibitors can produce extraordinary activity in some patients, PDAC is a disease that has demonstrated remarkable immunologic resistance. Applying immunotherapy to PDAC will undoubtedly require strategically designed combinations of therapies. Genetic mouse models of PDAC have been strongly predictive of immunotherapy outcomes and thus can offer a high-throughput platform for the study of treatment combinations.69 From clinical and preclinical studies, two conceptual models for applying immunotherapy in PDAC have emerged: (1) restoring elements of the cancer immunity cycle to stimulate productive T-cell immunosurveillance and (2) redirecting the immune reaction to PDAC for enhancing the efficacy of cytotoxic therapies. Although strategies capable of invoking T-cell immunity are critical for treatment response durability, leveraging the immune microenvironment in PDAC to improve outcomes with cytotoxic chemotherapy and radiotherapy may be particularly relevant for tumor debulking.85 Effective T-cell immunity in PDAC will likely require a multipronged approach that involves (1) conditioning the tumor microenvironment by reversing elements of immunosuppression and therapeutic resistance, (2) activation of tumor-specific T-cell responses using vaccines, targeted

DEPLOYING IMMUNOTHERAPY IN PANCREATIC CANCER

therapy, chemotherapy, radiotherapy, or transfer of T-cell immunity, and (3) maintenance therapy to sustain T-cell effector activity and counteract negative regulatory signals encountered within tumors. Well-designed correlative analyses applied to early phase clinical trials that incorporate tissue analysis using genomic and proteomic assays (e.g., multiplex immunohistochemistry) that consider the spatial heterogeneity of tumors will be fundamental to rapidly learning from each patient treated and interpreting treatment responses and failures based on expected biologic activity. Given considerable heterogeneity seen in the immune microenvironment in PDAC, it is also likely that immunotherapy must ultimately be personalized. One approach to this is actively being explored and involves applying genetic determinants (e.g., MMR deficiency and BRCA1/2 mutations) to identifying patient subgroups that may be more likely to respond to a particular immunotherapeutic strategy. For example, MMR deficiency has been suggested from clinical studies to be a potential biomarker for clinical activity with PD-1 antagonists.49 Similarly, in mouse models of BRCA mutant PDAC, IL-6–neutralizing antibodies enhance the efficacy of PD-L1 antagonists.94 The application of immunotherapy to PDAC has been associated with mixed clinical responses. This has been

seen with adoptive cell therapy using CAR T cells as well as MUC-1–specific lymphocytes.59,60 In addition, CD40 agonists administered in combination with chemotherapy have produced heterogeneous treatment responses in individual lesions detected within the same patient.95 Together, these findings suggest that tumor heterogeneity could emerge as a major challenge to the success of immunotherapy in PDAC. In conclusion, immunotherapy is a novel treatment approach to PDAC that leverages the specificity and diversity of the immune system for cancer therapy. Multiple clinical trials evaluating immunotherapy in PDAC are ongoing (Table 2). However, PDAC has repeatedly triumphed over novel therapeutic strategies in the past. Thus, for immunotherapy to be different and successful in this disease, it must be applied strategically and guided by rigorous “bedside-tobench and back” research.

ACKNOWLEDGMENT

This work was supported by National Institutes of Health grant R01-CA-197916 (to G. L. Beatty), a 2015 Pancreatic Cancer Action Network-AACR Career Development Award supported by an anonymous foundation through grant 15-20-25-BEAT (to G. L. Beatty), and Doris Duke Charitable Foundation grant 2013107 (to G. L. Beatty).

References 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J Clin. 2016;66:7-30. 2. Beatty GL, Gladney WL. Immune escape mechanisms as a guide for cancer immunotherapy. Clin Cancer Res. 2015;21:687-692.

11. Lutz ER, Wu AA, Bigelow E, et al. Immunotherapy converts nonimmunogenic pancreatic tumors into immunogenic foci of immune regulation. Cancer Immunol Res. 2014;2:616-631.

3. Feig C, Gopinathan A, Neesse A, et al. The pancreas cancer microenvironment. Clin Cancer Res. 2012;18:4266-4276.

12. von Bernstorff W, Voss M, Freichel S, et al. Systemic and local immunosuppression in pancreatic cancer patients. Clin Cancer Res. 2001;7:925s-932s.

4. Ino Y, Yamazaki-Itoh R, Shimada K, et al. Immune cell infiltration as an indicator of the immune microenvironment of pancreatic cancer. Br J Cancer. 2013;108:914-923.

13. Bailey P, Chang DK, Forget MA, et al. Exploiting the neoantigen landscape for immunotherapy of pancreatic ductal adenocarcinoma. Sci Rep. 2016;6:35848.

5. Wang WQ, Liu L, Xu HX, et al. Infiltrating immune cells and gene mutations in pancreatic ductal adenocarcinoma. Br J Surg. 2016;103:1189-1199.

14. Bailey P, Chang DK, Nones K, et al; Australian Pancreatic Cancer Genome Initiative. Genomic analyses identify molecular subtypes of pancreatic cancer. Nature. 2016;531:47-52.

6. Tewari N, Zaitoun AM, Arora A, et al. The presence of tumourassociated lymphocytes confers a good prognosis in pancreatic ductal adenocarcinoma: an immunohistochemical study of tissue microarrays. BMC Cancer. 2013;13:436.

15. Shibuya KC, Goel VK, Xiong W, et al. Pancreatic ductal adenocarcinoma contains an effector and regulatory immune cell infiltrate that is altered by multimodal neoadjuvant treatment. PLoS One. 2014;9:e96565.

7. Helm O, Mennrich R, Petrick D, et al. Comparative characterization of stroma cells and ductal epithelium in chronic pancreatitis and pancreatic ductal adenocarcinoma. PLoS One. 2014;9:e94357. 8. Ryschich E, Nötzel T, Hinz U, et al. Control of T-cell-mediated immune response by HLA class I in human pancreatic carcinoma. Clin Cancer Res. 2005;11:498-504. 9. Galon J, Pagès F, Marincola FM, et al. Cancer classification using the Immunoscore: a worldwide task force. J Transl Med. 2012;10:205. 10. Poschke I, Faryna M, Bergmann F, et al. Identification of a tumorreactive T-cell repertoire in the immune infiltrate of patients with resectable pancreatic ductal adenocarcinoma. Oncoimmunology. 2016;5:e1240859.

16. Castino GF, Cortese N, Capretti G, et al. Spatial distribution of B cells predicts prognosis in human pancreatic adenocarcinoma. OncoImmunology. 2015;5:e1085147. 17. Hiraoka N, Ino Y, Yamazaki-Itoh R, et al. Intratumoral tertiary lymphoid organ is a favourable prognosticator in patients with pancreatic cancer. Br J Cancer. 2015;112:1782-1790. 18. Meng Q, Liu Z, Rangelova E, et al. Expansion of tumor-reactive T cells from patients with pancreatic cancer. J Immunother. 2016;39:81-89. 19. Schmielau J, Nalesnik MA, Finn OJ. Suppressed T-cell receptor zeta chain expression and cytokine production in pancreatic cancer patients. Clin Cancer Res. 2001;7:933s-939s. 20. Chen DS, Mellman I. Oncology meets immunology: the cancerimmunity cycle. Immunity. 2013;39:1-10.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 275

BEATTY, EGHBALI, AND KIM

21. Lutz E, Yeo CJ, Lillemoe KD, et al. A lethally irradiated allogeneic granulocyte-macrophage colony stimulating factor-secreting tumor vaccine for pancreatic adenocarcinoma. A phase II trial of safety, efficacy, and immune activation. Ann Surg. 2011;253:328-335. 22. Hardacre JM, Mulcahy M, Small W, et al Addition of algenpantucel-L immunotherapy to standard adjuvant therapy for pancreatic cancer: a phase 2 study. J Gastrointest Surg. 2013;17:94-100; discussion 101. 23. NewLink Genetics Announces Results from Phase 3 IMPRESS Trial of Algenpantucel-L for Patients with Resected Pancreatic Cancer. http:// investors.linkp.com/releasedetail.cfm?releaseid=969978. Accessed January 29, 2017. 24. Le DT, Wang-Gillam A, Picozzi V, et al. Safety and survival with GVAX pancreas prime and Listeria Monocytogenes-expressing mesothelin (CRS-207) boost vaccines for metastatic pancreatic cancer. J Clin Oncol. 2015;33:1325-1333. 25. Le DT, Ko AH, Wainberg ZA. Results from a phase 2b, randomized, multicenter study of GVAX pancreas and CRS-207 compared to chemotherapy in adults with previously-treated metastatic pancreatic adenocarcinoma (ECLIPSE Study). J Clin Oncol. 2017; 4s (suppl; abstr 345). 26. Wedén S, Klemp M, Gladhaug IP, et al. Long-term follow-up of patients with resected pancreatic cancer following vaccination against mutant K-ras. Int J Cancer. 2011;128:1120-1128. 27. Suzuki N, Hazama S, Iguchi H, et al. Phase II clinical trial of peptide co*cktail therapy for patients with advanced pancreatic cancer: VENUSPC study. Cancer Sci. 2017;108:73-80. 28. Yutani S, Komatsu N, Yosh*tomi M, et al. A phase II study of a personalized peptide vaccination for chemotherapy-resistant advanced pancreatic cancer patients. Oncol Rep. 2013;30:1094-1100. 29. Asahara S, Takeda K, Yamao K, et al. Phase I/II clinical trial using HLAA24-restricted peptide vaccine derived from KIF20A for patients with advanced pancreatic cancer. J Transl Med. 2013;11:291. 30. Gjertsen MK, Bakka A, Breivik J, et al. Ex vivo ras peptide vaccination in patients with advanced pancreatic cancer: results of a phase I/II study. Int J Cancer. 1996;65:450-453. 31. Toubaji A, Achtar M, Provenzano M, et al. Pilot study of mutant ras peptide-based vaccine as an adjuvant treatment in pancreatic and colorectal cancers. Cancer Immunol Immunother. 2008;57:1413-1420. 32. Gjertsen MK, Buanes T, Rosseland AR, et al. Intradermal ras peptide vaccination with granulocyte-macrophage colony-stimulating factor as adjuvant: Clinical and immunological responses in patients with pancreatic adenocarcinoma. Int J Cancer. 2001;92:441-450. 33. Yamaue H, Tsunoda T, Tani M, et al. Randomized phase II/III clinical trial of elpamotide for patients with advanced pancreatic cancer: PEGASUS-PC Study. Cancer Sci. 2015;106:883-890. 34. Oettle H, Seufferlein T, Luger T, et al. Final results of a phase I/II study in patients with pancreatic cancer, malignant melanoma, and colorectal carcinoma with trabedersen. J Clin Oncol. 2012;30 (suppl; abstr 4034). 35. Middleton G, Silco*cks P, Cox T, et al. Gemcitabine and capecitabine with or without telomerase peptide vaccine GV1001 in patients with locally advanced or metastatic pancreatic cancer (TeloVac): an open-label, randomised, phase 3 trial. Lancet Oncol. 2014;15: 829-840. 36. Royal RE, Levy C, Turner K, et al. Phase 2 trial of single agent Ipilimumab (anti-CTLA-4) for locally advanced or metastatic pancreatic adenocarcinoma. J Immunother. 2010;33:828-833.

276 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

37. Chung MJ, Park JY, Bang S, et al. Phase II clinical trial of ex vivoexpanded cytokine-induced killer cells therapy in advanced pancreatic cancer. Cancer Immunol Immunother. 2014;63:939-946. 38. Salmons B, Löhr M, Günzburg WH. Treatment of inoperable pancreatic carcinoma using a cell-based local chemotherapy: results of a phase I/ II clinical trial. J Gastroenterol. 2003;38:78-84. 39. Schmidt J, Abel U, Debus J, et al. Open-label, multicenter, randomized phase III trial of adjuvant chemoradiation plus interferon Alfa2b versus fluorouracil and folinic acid for patients with resected pancreatic adenocarcinoma. J Clin Oncol. 2012;30:4077-4083. 40. Knaebel HP, Märten A, Schmidt J, et al. Phase III trial of postoperative cisplatin, interferon alpha-2b, and 5-FU combined with external radiation treatment versus 5-FU alone for patients with resected pancreatic adenocarcinoma -- CapRI: study protocol [ISRCTN62866759]. BMC Cancer. 2005;5:37. 41. Lygidakis NJ, Sgourakis G, Georgia D, et al. Regional targeting chemoimmunotherapy in patients undergoing pancreatic resection in an advanced stage of their disease: a prospective randomized study. Ann Surg. 2002;236:806-813. 42. Dalgleish AG, Stebbing J, Adamson DJ, et al. Randomised, open-label, phase II study of gemcitabine with and without IMM-101 for advanced pancreatic cancer. Br J Cancer. 2016;115:789-796. 43. Sultana A, Shore S, Raraty MG, et al. Randomised Phase I/ II trial assessing the safety and efficacy of radiolabelled anticarcinoembryonic antigen I(131) KAb201 antibodies given intraarterially or intravenously in patients with unresectable pancreatic adenocarcinoma. BMC Cancer. 2009;9:66. 44. Jaffee EM, Hruban RH, Biedrzycki B, et al. Novel allogeneic granulocytemacrophage colony-stimulating factor-secreting tumor vaccine for pancreatic cancer: a phase I trial of safety and immune activation. J Clin Oncol. 2001;19:145-156. 45. Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer. 2012;12:252-264. 46. Zou W, Wolchok JD, Chen L. PD-L1 (B7-H1) and PD-1 pathway blockade for cancer therapy: Mechanisms, response biomarkers, and combinations. Sci Transl Med. 2016;8:328rv4. 47. Brahmer JR, Tykodi SS, Chow LQ, et al. Safety and activity of antiPD-L1 antibody in patients with advanced cancer. N Engl J Med. 2012;366:2455-2465. 48. Birnbaum DJ, Finetti P, Lopresti A, et al. Prognostic value of PDL1 expression in pancreatic cancer. Oncotarget. 2016;7:71198-71210. 49. Le DT, Uram JN, Wang H, et al. PD-1 blockade in tumors with mismatchrepair deficiency. N Engl J Med. 2015;372:2509-2520. 50. Riazy M, Kalloger SE, Sheffield BS, et al. Mismatch repair status may predict response to adjuvant chemotherapy in resectable pancreatic ductal adenocarcinoma. Mod Pathol. 2015;28:13831389. 51. Nakata B, Wang YQ, Yashiro M, et al. Prognostic value of microsatellite instability in resectable pancreatic cancer. Clin Cancer Res. 2002;8:2536-2540. 52. Eatrides JM, Coppola D, Diffalha SA, et al. Microsatellite instability in pancreatic cancer. J Clin Oncol. 2016;34:e15753. 53. Mlecnik B, Bindea G, Angell HK, et al. Integrative Analyses of Colorectal Cancer Show Immunoscore Is a Stronger Predictor of Patient Survival Than Microsatellite Instability. Immunity. 2016;44: 698-711.

DEPLOYING IMMUNOTHERAPY IN PANCREATIC CANCER

54. Le DT, Lutz E, Uram JN, et al. Evaluation of ipilimumab in combination with allogeneic pancreatic tumor cells transfected with a GM-CSF gene in previously treated pancreatic cancer. J Immunother. 2013;36:382389. 55. Aglietta M, Barone C, Sawyer MB, et al. A phase I dose escalation trial of tremelimumab (CP-675,206) in combination with gemcitabine in chemotherapy-naive patients with metastatic pancreatic cancer. Ann Oncol. 2014;25:1750-1755. 56. Wang-Gillam A, Plambeck-Suess S, Goedegebuure P, et al. A phase I study of IMP321 and gemcitabine as the front-line therapy in patients with advanced pancreatic adenocarcinoma. Invest New Drugs. 2013;31:707-713.

70. Collins MA, Bednar F, Zhang Y, et al. Oncogenic Kras is required for both the initiation and maintenance of pancreatic cancer in mice. J Clin Invest. 2012;122:639-653. 71. Guerra C, Collado M, Navas C, et al. Pancreatitis-induced inflammation contributes to pancreatic cancer by inhibiting oncogene-induced senescence. Cancer Cell. 2011;19:728-739. 72. Guerra C, Schuhmacher AJ, Cañamero M, et al. Chronic pancreatitis is essential for induction of pancreatic ductal adenocarcinoma by K-Ras oncogenes in adult mice. Cancer Cell. 2007;11:291-302. 73. Seifert L, Werba G, Tiwari S, et al. Radiation therapy induces macrophages to suppress T-cell responses against pancreatic tumors in mice. Gastroenterology. 2016;150:1659-1672.e5.

57. Beatty GL, O’Hara M. Chimeric antigen receptor-modified T cells for the treatment of solid tumors: defining the challenges and next steps. Pharmacol Ther. 2016;166:30-39.

74. Stromnes IM, Brockenbrough JS, Izeradjene K, et al. Targeted depletion of an MDSC subset unmasks pancreatic ductal adenocarcinoma to adaptive immunity. Gut. 2014;63:1769-1781.

58. Beatty GL, Haas AR, Maus MV, et al. Mesothelin-specific chimeric antigen receptor mRNA-engineered T cells induce anti-tumor activity in solid malignancies. Cancer Immunol Res. 2014;2:112-120.

75. Griesmann H, Drexel C, Milosevic N, et al. Pharmacological macrophage inhibition decreases metastasis formation in a genetic model of pancreatic cancer. Gut. Epub 2016 Mar 24.

59. Beatty GL, O’Hara MH, Nelson AM, et al. Safety and antitumor activity of chimeric antigen receptor modified T cells in patients with chemotherapy refractory metastatic pancreatic cancer. J Clin Oncol. 2015;33 (suppl; abstr 3007).

76. Sanford DE, Belt BA, Panni RZ, et al. Inflammatory monocyte mobilization decreases patient survival in pancreatic cancer: a role for targeting the CCL2/CCR2 axis. Clin Cancer Res. 2013;19:3404-3415.

60. Kondo H, Hazama S, Kawaoka T, et al. Adoptive immunotherapy for pancreatic cancer using MUC1 peptide-pulsed dendritic cells and activated T lymphocytes. Anticancer Res. 2008;28(1B):379-387.

77. Mitchem JB, Brennan DJ, Knolhoff BL, et al. Targeting tumorinfiltrating macrophages decreases tumor-initiating cells, relieves immunosuppression, and improves chemotherapeutic responses. Cancer Res. 2013;73:1128-1141.

61. Homma Y, Taniguchi K, Murakami T, et al. Immunological impact of neoadjuvant chemoradiotherapy in patients with borderline resectable pancreatic ductal adenocarcinoma. Ann Surg Oncol. 2014;21:670-676.

78. Kalbasi A, Komar C, Tooker GM, et al. Tumor-derived CCL2 mediates resistance to radiotherapy in pancreatic ductal adenocarcinoma. Clin Cancer Res. 2017;23:137-148.

62. Bendell JC, Kim TW, Goh BC, et al. Clinical activity and safety of cobimetinib (cobi) and atezolizumab in colorectal cancer (CRC). J Clin Oncol. 2016;34 (suppl; abstr 3502). 63. Hu-Lieskovan S, Mok S, Homet Moreno B, et al. Improved antitumor activity of immunotherapy with BRAF and MEK inhibitors in BRAF(V600E) melanoma. Sci Transl Med. 2015;7:279ra41. 64. Kono M, Dunn IS, Durda PJ, et al. Role of the mitogen-activated protein kinase signaling pathway in the regulation of human melanocytic antigen expression. Mol Cancer Res. 2006;4:779-792. 65. Zhang Y, Velez-Delgado A, Mathew E, et al. Myeloid cells are required for PD-1/PD-L1 checkpoint activation and the establishment of an immunosuppressive environment in pancreatic cancer. Gut. 2017;66:124-136. 66. Lampen MH, van Hall T. Strategies to counteract MHC-I defects in tumors. Curr Opin Immunol. 2011;23:293-298. 67. Kim K, Skora AD, Li Z, et al. Eradication of metastatic mouse cancers resistant to immune checkpoint blockade by suppression of myeloidderived cells. Proc Natl Acad Sci USA. 2014;111:11774-11779. 68. Hingorani SR, Wang L, Multani AS, et al. Trp53R172H and KrasG12D cooperate to promote chromosomal instability and widely metastatic pancreatic ductal adenocarcinoma in mice. Cancer Cell. 2005;7:469483. 69. Lee JW, Komar CA, Bengsch F, et al. Genetically engineered mouse models of pancreatic cancer: the KPC model (LSL-Kras(G12D/+) ;LSL-Trp53(R172H/+) ;Pdx-1-Cre), its variants, and their application in ommuno-oncology drug discovery. Curr Protoc Pharmacol. 2016;73:14.39.1-14.39.20.

79. Steele CW, Karim SA, Leach JD, et al. CXCR2 inhibition profoundly suppresses metastases and augments immunotherapy in pancreatic ductal adenocarcinoma. Cancer Cell. 2016;29:832-845. 80. Nywening TM, Wang-Gillam A, Sanford DE, et al. Targeting tumourassociated macrophages with CCR2 inhibition in combination with FOLFIRINOX in patients with borderline resectable and locally advanced pancreatic cancer: a single-centre, open-label, dosefinding, non-randomised, phase 1b trial. Lancet Oncol. 2016;17: 651-662. 81. Noel MS, Hezel AF, Linehan D, et al. Orally administered CCR2 selective inhibitor CCX872-b clinical trial in pancreatic cancer. J Clin Oncol. 2017; 35 (suppl; abstr 276). 82. Long KB, Beatty GL. Harnessing the antitumor potential of macrophages for cancer immunotherapy. OncoImmunology. 2013;2:e26860. 83. Beatty GL, Chiorean EG, Fishman MP, et al. CD40 agonists alter tumor stroma and show efficacy against pancreatic carcinoma in mice and humans. Science. 2011;331:1612-1616. 84. Long KB, Gladney WL, Tooker GM, et al. IFNγ and CCL2 cooperate to redirect tumor-Infiltrating monocytes to degrade fibrosis and enhance chemotherapy efficacy in pancreatic carcinoma. Cancer Discov. 2016;6:400-413. 85. Beatty GL, Li Y, Long KB. Cancer immunotherapy: activating innate and adaptive immunity through CD40 agonists. Expert Rev Anticancer Ther. 2017;17:175-186. 86. Beatty GL, Winograd R, Evans RA, et al. Exclusion of T cells from pancreatic carcinomas in mice is regulated by Ly6C(low) F4/80(+) extratumoral macrophages. Gastroenterology. 2015;149: 201-210.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 277

BEATTY, EGHBALI, AND KIM

87. Gunderson AJ, Kaneda MM, Tsujikawa T, et al. Bruton tyrosine kinasedependent immune cell cross-talk drives pancreas cancer. Cancer Discov. 2016;6:270-285. 88. Pylayeva-Gupta Y, Lee KE, Hajdu CH, et al. Oncogenic Kras-induced GMCSF production promotes the development of pancreatic neoplasia. Cancer Cell. 2012;21:836-847. 89. Sparmann A, Bar-Sagi D. Ras-induced interleukin-8 expression plays a critical role in tumor growth and angiogenesis. Cancer Cell. 2004;6:447-458. 90. Jiang H, Hegde S, Knolhoff BL, et al. Targeting focal adhesion kinase renders pancreatic cancers responsive to checkpoint immunotherapy. Nat Med. 2016;22:851-860. 91. Stromnes IM, Schmitt TM, Hulbert A, et al. T cells engineered against a native antigen can surmount immunologic and physical barriers to treat pancreatic ductal adenocarcinoma. Cancer Cell. 2015;28:638-652.

278 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

92. Feig C, Jones JO, Kraman M, et al. Targeting CXCL12 from FAPexpressing carcinoma-associated fibroblasts synergizes with antiPD-L1 immunotherapy in pancreatic cancer. Proc Natl Acad Sci USA. 2013;110:20212-20217. 93. Ene-Obong A, Clear AJ, Watt J, et al. Activated pancreatic stellate cells sequester CD8+ T cells to reduce their infiltration of the juxtatumoral compartment of pancreatic ductal adenocarcinoma. Gastroenterology. 2013;145:1121-1132. 94. Mace TA, Shakya R, Pitarresi JR, et al. IL-6 and PD-L1 antibody blockade combination therapy reduces tumour progression in murine models of pancreatic cancer. Gut. Epub 2016 Oct 21. 95. Beatty GL, Torigian DA, Chiorean EG, et al. A phase I study of an agonist CD40 monoclonal antibody (CP-870,893) in combination with gemcitabine in patients with advanced pancreatic ductal adenocarcinoma. Clin Cancer Res. 2013;19:6286-6295.

GASTRIC CANCER IN ASIA

Gastric Cancer in Asia: Unique Features and Management Tomoyuki Irino, MD, PhD, Hiroya Takeuchi, MD, PhD, Masanori Terashima, MD, PhD, Toshifumi Wakai, MD, PhD, and Yuko Kitagawa, MD, PhD OVERVIEW Gastric cancer (GC) poses a burden to patients across the globe as the third leading cause of cancer deaths worldwide. Incidence of GC is particularly high in Asian countries, which is attributed to the prevalence of Helicobacter pylori (H. pylori) infection and has prompted the establishment of unique treatment strategies. D2 gastrectomy, which was established in the 1950s in Japan, has served as a gold standard for locally advanced GC for over half a century. Since the beginning of the 21st century, endoscopic resection (ER) techniques and minimally invasive laparoscopic surgery have greatly changed the treatment of patients with early GC. S-1, which showed a striking survival benefit in a large randomized trial in Japan, has been used as adjuvant therapy for the last decade. Likewise, S-1–based chemotherapy regimens are currently the standard of care for the treatment of unresectable/metastatic GC in Asia. Along with the development of standardized therapy, novel techniques and new drugs have been rapidly brought into clinical practice. State-of-the-art sentinel node (SN) navigation surgery enables clinicians to perform truly minimally invasive surgery for early GC, and appropriate chemotherapy regimens are now determined by a tumor’s molecular expression. New classifications based on gene signatures are proposed and may replace conventional clinical classifications. Such highly individualized treatment has the potential to alter our clinical practice in GC in the near future. The best practice in each geographic region should be shared and integrated, resulting in the best practice without borders.

C is the fifth most common cancer worldwide with 952,000 new patients diagnosed in 2012, and the third leading cause of cancer deaths in both men and women.1 Geographic distribution of incidence and mortality of GC is known to be disproportionate, namely it is particularly high in Asian countries and low in the Western hemisphere. This uneven distribution has been closely correlated with the prevalence of H. pylori infection, which is undoubtedly the primary cause of noncardia GC (Fig. 1).2-4 GC has therefore been one of the leading causes of cancer mortality in Asia. In Japan, as well as other high-risk areas, the infection rate of H. pylori was very high a hundred years ago with an estimated prevalence of 80% in people born in the 1900s; however, the prevalence has gradually declined and no more than 10% of people born in the 2000s harbor H. pylori.5 This fact is well associated with the trend of incidence and mortality of GC in Japan. At that time, because surgical resection of the stomach was the only treatment of GC, Japanese surgeons struggled to determine how to manage this intractable cancer that presented with extensive lymph node metastasis and systemic dissemination. They finally devised two approaches: one approach was the elucidation of lymphatics around the stomach followed by the estab-

lishment of classification of GC, and the other approach was implementation of a screening program for GC. Meticulous analysis of gastric lymphatics was initially studied by Inoue et al in 1936. Based on the study, Tamaki Kajitani and other Japanese gastric surgeons established systematic lymph node dissection of the stomach in the 1950s, namely today’s D2 lymphadenectomy. The fundamental concept of D2 gastrectomy was resection of the primary tumor with whole dissection of the lymphatic system around the stomach, which was most likely to involve not only macroscopic (visible) but also microscopic (invisible) nodal metastases. Gastrectomy with adequate lymphadenectomy was reasonable in light of cancer surgery and thus has been widely accepted by Japanese surgeons. Careful, organized findings about GC was exhaustively collected and finally published in the first edition of the Japanese Classification of Gastric Carcinoma in 1964. Surgical resection has been the predominant treatment of GC thereafter. In contrast, a major problem was that the cancer was already advanced and metastatic when many patients presented with their symptoms. Thus, it was crucial to find cancer at an early stage to achieve cure. Toshio Kurokawa et al first performed a mass screening program for GC in

From the Division of Gastric Surgery, Shizuoka Cancer Center, Shizuoka, Japan; Department of Surgery, Keio University School of Medicine, Tokyo, Japan; Division of Gastric Surgery, Shizuoka Cancer Center, Shizuoka, Japan; Division of Digestive and General Surgery, Niigata University Graduate School of Medical and Dental Science, Niigata, Japan; Department of Surgery, Keio University School of Medicine, Tokyo, Japan. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Yuko Kitagawa, MD, PhD, FACS, Department of Surgery, Keio University School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo, 160-8582, Japan; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 279

IRINO ET AL

FIGURE 1. Prevalence of Helicobacter pylori Infection by Country Along With the Incidence and Mortality of Gastric Cancer

the 1950s “with the intention of early discovery and prevention of cancer through the diagnosis of people who felt they were perfectly healthy.”6 His tremendous effort also led to the development of diagnostic tools and techniques used today, represented by endoscopy and double-contrast barium radiography. Owing to the screening program, over half of GCs were detected at an early stage in the early 1980s, whereas less than 5% were detected before 1955. Although the question whether mass screening programs for GC can ultimately reduce mortality of GC is still open to debate,7 it

KEY POINTS • As a result of the large numbers of patients with GC in Asia, we have established our own evidence-based best practices. • The sentinel node concept was demonstrated to be applicable in certain early GCs, enabling highly individualized sentinel node navigation surgery. • Conventional D2 gastrectomy followed by adjuvant chemotherapy using S-1 or capecitabin plus oxaliplatin serves as a gold standard for locally advanced GC in Asia. • S-1 is one of the key drugs used in the Asian population and S-1–based chemotherapy regimens currently achieve the best survival in unresectable/metastatic GC. • In the new era of precision medicine, the best treatment strategy is determined by a tumor’s molecular expression and gene signatures including EpsteinBarr virus status, microsatellite and/or chromosomal instability, genomic stability, epithelial mesenchymal transition, p53 activity, cytokine signaling, cell proliferation, and DNA methylation. 280 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

is no longer in doubt that such a program detects a large number of early GCs that are curable. These unique features have greatly influenced the management of GC in Asian countries, particularly in Japan. Nowadays, much evidence comes from Western countries where evidence-based medicine has been widely accepted for many years. Guidelines and clinical practices in Asian countries, for example in breast cancer and colorectal cancer, depend much on external evidence from the West. In contrast, by taking advantage of the large numbers of patients with GC, we have developed best practices not by extrapolating external evidence, but by establishing our own evidence. In this review, we outline how the management of GC has been established, by tracing back the history of GC treatment in Japan and then discuss whether in future we should follow our own path as done so far, or should collaborate with other countries in the West to establish common evidence.

SURGICAL MANAGEMENT OF EARLY GC: TRULY MINIMALLY INVASIVE SURGICAL THERAPY

The high incidence of early-stage GC in Japan has had considerable influence on its surgical management, and has resulted in the development of endoscopic treatment and minimally invasive surgery. ER was first reported by Tsuneoka et al in 1969, which has since become the standard of care for early GCs in which the risk of lymph node involvement is most likely to be zero.8,9 Early cancers that are not eligible for endoscopic therapy are primarily treated by surgical resection of the stomach with D1+/D2 lymphadenectomy, as is performed for

GASTRIC CANCER IN ASIA

patients with locally-advanced GC, because there is a high risk (approximately 15%) of lymph node metastasis.10,11 Open gastrectomy was the standard of care until a decade ago; however, since 1991, when laparoscopic gastrectomy as minimally-invasive surgery for patients with gastric cancer was first performed by Seigo Kitano et al, laparoscopic surgery has become popular worldwide. The Japan Clinical Oncology Group (JCOG) is now conducting a large, randomized phase III trial comparing laparoscopic gastrectomy with conventional open gastrectomy in patients with early GC (JCOG0912), and short-term outcomes showed the safety of the procedure in line with the result from a large phase III trial in Korea (KLASS-01).10,11 Although many randomized, controlled trials of laparoscopic gastrectomy for early GC have demonstrated surgical and oncologic noninferiority, only a limited study provided evidence of its superiority over conventional surgery. This is in part because the procedure in the abdomen is more or less the same between open and laparoscopic surgery (i.e., gastrectomy with lymphadenectomy).12 Therefore, how to avoid gastrectomy without compromising radicality, which is truly minimally invasive, is one of the long-standing issues surgeons have attempted to solve for decades. SN biopsy was first introduced by Cabanas et al in penile carcinoma in 1977, and also dramatically altered the surgical and oncologic management of breast cancer.13 If nodal status can be pathologically confirmed prior to gastrectomy, radical lymphadenectomy would not be necessary for patients without lymph node involvement, which accounts for up to 90% of all early GCs. Hence, if the SN concept can be applied to the gastric lymphatic drainage, which is relatively complicated, it would be possible to mitigate the burden of gastrectomy.14 To evaluate whether the SN theory can be applied to GC, a large, prospective, multicenter phase II study was implemented by the Japan Society of SN navigation surgery, in which a standardized protocol and technique with an endoscopic dual tracer injection method were used.15 A total of 397 patients with cT1/T2 gastric adenocarcinoma smaller than 4 cm were analyzed, and the SN detection rate was 97.5% and the overall accuracy of nodal evaluation for metastasis was 99.0% with only four false-negative cases (1.0%). The study confirmed that the SN theory can be applied to the complicated gastric lymphatic drainage and a large phase III study is currently being conducted in which long-term oncologic safety and quality of life will be evaluated. D2 gastrectomy alone has been the standard of care for early GCs as well as locally advanced cancers. According to the SN study, 96% of metastatic nodes reside within the D2 area, which indicates the need for the radical and complete technique of D2 gastrectomy. However, D2 gastrectomy might be overtreatment for patients who are at very low risk of lymph node metastasis. Gastric surgeons have long confronted this dilemma, that it is hard to please everybody. Hence, SN navigation surgery can help individualize surgical treatment, thereby appropriately providing radical and less-invasive procedures for patients with early GC.

On the basis of the SN theory, a new rendezvous-style surgical procedure using endoscopy and laparoscopy was developed for early GC. The concept of the procedure, which is called laparoscopy-endoscopy cooperative surgery (LECS), was originally developed for submucosal tumors of the gastrointestinal tract and various modified methods have been currently proposed.16 Nonexposed endoscopic wall–inversion surgery is one of the modified procedures, in which a full-thickness gastric wall resection can be achieved in a closed manner.17 By combining nonexposed endoscopic wall–inversion surgery with SN navigation, Goto et al reported the first case of stomach-preserving surgery in a 55-yearold female patient with 2-cm diffuse-type early GC that was not eligible for ER.18 (Fig. 2) Although this minimally invasive technique is just being developed and thus needs to be carefully evaluated, it has great potential to change the role of gastrectomy for early GC, which has served as a gold standard for decades.

TREATMENT STRATEGY FOR LOCALLY ADVANCED GC: PERIOPERATIVE TREATMENT

Since Kajitani et al established D2 gastrectomy alone as the standard of care for locally advanced GC for many years until S-1 (an orally-administered combination drug of tegafur and gimeracil plus oteracil) was introduced as adjuvant therapy in 2006. In the late 1990s, results from two randomized trials from Holland (Dutch trial) and the United Kingdom (MRC trial) comparing D2 with D1 were published, in which D2 gastrectomy yielded no survival benefit and showed more postoperative complications and high mortality.19,20 However, Japanese surgeons did not accept high morbidity and in-hospital death rates in the D2 group as sufficient reason to change their management practices, and D2 gastrectomy remained the standard of care in Japan. Results of a 15year follow-up analysis of the Dutch trial were published in 2010, which demonstrated advantages of D2 gastrectomy in terms of locoregional recurrence (22% in D1 vs. 12% in D2) and GC-related death (48% in D1 vs. 37% in D2; hazard ratio [HR] 0.74; 95% CI, 0.59–0.93; p = .01), although there were no significant differences in disease-free survival (p = .31) or risk of recurrence (p = .10).21 Based on these results, and that it is currently performed safely in Western high-volume centers, D2 gastrectomy is currently recommended for medically-fit patients in the guideline of the National Comprehensive Cancer Network and European Society of Medical Oncology.22,23 Five large, randomized, controlled trials were conducted in Japan by JCOG to evaluate the survival benefit of extended lymphadenectomy and surgical procedure because there had been much discussion among Japanese surgeons about the optimal extent of lymphadenectomy (Table 1).24-27 JCOG9501, the first big surgical randomized trial that evaluated survival benefit of prophylactic paraaortic lymph node dissection, demonstrated a negative result. The results between the two groups were strikingly similar for 5-year overall survival (OS; 69.2% for D2 and 70.3% for extended D2; HR 1.03; 95% CI, 0.77–1.37) and other relevant outcomes.24 The asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 281

IRINO ET AL

FIGURE 2. Sentinel Node Navigation Surgery for Early Gastric Cancer in Combination With Nonexposed Endoscopic Wall-Invasion Surgery

(A) Fluorescence image after indocyanine green was injected around the primary tumor. (B) Visualization of indocyanine green through infrared camera. (C) Circumferential seromuscular incision followed by linear suturing. (D) Endoscopic removal of the primary lesion.

second trial tested noninferiority of spleen-preservation during gastrectomy in patients with locally-advanced cancer that was not located in the greater curvature (JCOG0110).26 Because standard D2 total gastrectomy had included the nodal station 10 (splenic hilar nodes), splenectomy was necessary to complete D2 dissection. The study demonstrated noninferiority of spleen preservation in regards to 5-year OS (75.1% for splenectomy and 76.4% for spleen preservation; HR 1.21; 90.7% CI, 0.67–1.16 with noninferiority margin 1.21; p = .025), having altered the definition of D2 since January 2017. The third study, JCOG1001, started in 2010 and compared bursectomy with nonbursectomy in patients with cT3/T4 GC. As a result, pancreatic fistula was twofold more likely to occur in the bursectomy group, whereas bursectomy failed to provide significant 5-year OS benefit (HR 1.07; 95% CI, 0.81–1.42; one-sided p = .68), and the data monitoring committee recommended early publication after all patients completed recruitment as planned.27 None of the trials showed a benefit of extended surgery, and conventional D2 gastrectomy without splenectomy and bursectomy is currently recognized as the standard procedure for locally-advanced cancer. Therefore, perioperative treatments should have promising potential to improve survival in patients with locally advanced GC. Although surgery was regarded as the most effective treatment, the importance of perioperative therapy was also recognized; however, there was no evidence to support this. The first large, randomized phase III trial by JCOG 282 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

commenced in 1988 in which surgery plus adjuvant mitomycin C (MMC) and 5-fluorouracil (5-FU) followed by tegafur demonstrated no survival benefit compared with surgery alone in patients with serosa-negative GC (Table 2).29 Importantly, T1 cancer accounted for 32.8% of all tumors in this study, suggesting that surgery alone for T1 cancer yielded good survival without need for any adjuvant therapy. Other JCOG studies compared curative surgery alone with MMC, 5-FU, cytosine arabinoside followed by oral 5-FU in serosa-negative GC (JCOG9206-1),30 and intraperitoneal and intravenous cisplatin followed by oral 5-FU in serosa-positive GC (JCOG9206-2).31 However, both trials yielded negative results, failing to show the advantage of adjuvant chemotherapy. In 1984, an oral tegafur/uracil drug named UFT was developed in Japan and has become commercially available. Taking advantage of this “homemade” drug, a national randomized trial comparing surgery alone with surgery plus adjuvant UFT was conducted (N-SAS-GC).32 The patient accrual was slow and only 190 patients (38% of initially-planned) were finally included in the study. The result was favorable for the adjuvant therapy group; however, the 5-year recurrence-free survival rate was relatively poor, calling for a replication of the study to validate the result. Meanwhile, another homemade agent named S-1 was developed and launched in 1999. S-1, commercially available as TS-1 in Asia and Teysuno in Europe, is an orally bioavailable fluoropyrimidine antagonist consisting of tegafur

GASTRIC CANCER IN ASIA

TABLE 1. Practice-Changing Randomized Phase III Trials in Gastric Cancer Surgery in Asia Ref 24

Recruit Time 1995– 2001

Country

Trial

Registry

No. of Patients

Target

Arms

Primary Endpoint

JCOG9501

NCT00149279

523

T2-4

D2 gastrectomy

vs. D2/PAND gastrectomy 25

1995– 2003

JCOG9502

NCT00149266

167

T2-4

Transhiatal approach

2002– 2009

JCOG0110

NCT00147147

505

T2-4/ N0-2

Splenectomy

2010– 2015

JCOG1001

NCT00152243

1,204

T2-4

Nonbursectomy

2008– 2013

JP, KR, SIN

JCOG0705/ KGCA01 (REGATTA)

NCT03001726

175

Stage IV

Chemotherapy alone vs. gastrectomy, chemotherapy

1.03 (0.77– 1.37)

.85

1.36 (0.89– 2.08)

.92

0.88 (0.67– 1.16)*

.025

1.07 (0.81– 1.42)

.68

1.09 (0.78– 1.52)

.70

5-year OS: 52.3%

5-year OS: 75% vs. 76.4%

vs. bursectomy 28

5-year OS: 69.2%

vs. 37.9%

vs. spleen-preservation 27

HR (95% CI)

vs. 70.3%

vs. left transthoracic approach 26

p Value

Result

3-year OS: 86% vs. 83.3%

median OS 16.6 vs. 14.3 months

*Noninferiority margin < 1.21. Abbreviations: HR; hazard ratio; CI, confidence interval; JP, Japan; PAND; para-aortic node dissection; OS, overall survival; KR; Korea; Ref, reference; SIN, Singapore.

with two modulators of 5-FU activity, 5-chloro-2,4-dihydroxypyridine and potassium oxonate in a molar ratio of 1:0.4:1.36 The landmark phase III ACTS-GC trial was initiated in 2002, comparing D2 gastrectomy alone with surgery plus S-1 for 1 year in patients with pathologic stage II/III GC.33 The result was remarkable; adjuvant S-1 yielded a 32% reduction of risk for 3-year OS (80.1% for S-1 vs. 70.1% for surgery alone; HR 0.68; 95% CI, 0.52–0.87; p = .003). These results dramatically changed clinical practice in Japan, where surgery had been believed to be the only effective treatment of locally advanced GC. After these findings, the majority of chemotherapy regimens have been developed in combination with S-1. Another adjuvant study was started in 2006 in three Asian countries, South Korea, China, and Taiwan, which aimed to seek an optimal adjuvant regimen after curative D2 gastrectomy. This study was abbreviated as the CLASSIC trial, and compared surgery alone with capecitabine plus oxaliplatin (CapeOX) as adjuvant therapy.35,37 This study was stopped in accordance with the recommendation by the data monitoring committee for reasons of benefit, unveiling the efficacy of CapeOX with a 3-year disease-free survival HR of 0.56 (95% CI, 0.44–0.72; p < .0001). Efficacy of adjuvant chemoradiation therapy was first demonstrated by the INT0116 study in the United States.38 However, because surgery alone has been deemed to offer good local control, radiation therapy, which also provides local control, has not become common in Japan. Most clinical trials from Japan comprise of chemotherapy alone without radiation. For that reason, adjuvant chemoradiation therapy in an Asian population was first evaluated in South Korea. This randomized phase III trial (ARTIST trial) compared adju-

vant capecitabine and cisplatin (XP) with XP plus concurrent capecitabine radiotherapy (XRT) after curative D2 gastrectomy.34,39 No additional benefit with radiation was observed in both 5-year OS (HR 1.130; 95% CI, 0.775–1.647; p = .527) and disease-free survival (HR 0.710; 95% CI, 0.520–1.050; p = .922), although the XP group showed significantly more local recurrence (XP 13% vs. XRT 7%; p = .033). However, subgroup analyses suggested some benefit for patients with positive lymph node involvement (HR 0.700; 95% CI, 0.493–0.994), motivating them to start another study (ARTIST-2 trial) in which three arms of chemotherapy or chemoradiation regimens will be compared: adjuvant S-1 and oxaliplatin (SOX) plus radiation (45 Gy) and two chemotherapy regimens (S-1 alone for 1 year and SOX for 6 months) for patients with node-positive disease after curative D2 gastrectomy. Adjuvant S-1 has become the standard of care since 2006 when the results of the ACTS-GC trial was unveiled. The final result was published in 2011, showing a 5-year OS of 71.7% with S-1 compared with 61.1% with surgery alone (HR 0.669; 95% CI, 0.540–0.828). In contrast, patients diagnosed with a more advanced stage (stage IIIB) demonstrated unsatisfactory survival outcomes, with no significant difference in 5-year OS (50.2% for S-1 vs. 44.1% for surgery alone; HR 0.791; 95% CI, 0.520–1.205).40 These data have provoked a discussion for another strategy to improve survival among patients with high-risk of relapse, which includes the possibility of neoadjuvant treatment and/or regimens with multiple agents. Several clinical trials are currently being conducted in Asia. A randomized phase III trial by JCOG was initiated in 2016 comparing surgery plus adjuvant S-1 compared with preopasco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 283

IRINO ET AL

erative SOX and surgery plus adjuvant S-1 in patients with cT3-4 node-positive disease (JCOG1509). Another group in Japan is evaluating the superiority of S-1 plus docetaxel compared with S-1 alone for patients with stage III disease

after curative D2 gastrectomy (START-2 trial). Nivolumab, an anti–PD-1 monoclonal antibody, demonstrated a survival benefit in patients with unresectable advanced or recurrent gastric or gastroesophageal junction (GEJ) cancer refractory

TABLE 2. Practice-Changing Randomized Phase III Trials For Neoadjuvant Therapy in Asia Ref 29

Recruit Time 1988– 1992

Country

Trial

Registry

No. of Patients

JCOG8801

Not registered

573

Target

Arms

Primary Endpoint

cT1-3/ any N

Surgery alone

vs. MMC, 5-FU IV, 5-FU PO 30

1993– 1994

JCOG9206-1

Not registered

252

cT2-3/ N1-2

Surgery alone

1993– 1998

JCOG9206-2

NCT00147147

268

cT3-4/ N0-2

Surgery alone

RFS

vs. cisplatin IP, cisplatin IV, 5-FU IV, UFT 32

1997– 2001

N-SAS-GC

NCT00152243

190

cT2/ N1-2

Surgery alone

vs. UFT 33

2001– 2004

ACTS-GC

NCT00152217

1,059

Stage II/ III

Surgery alone

2004– 2008

ARTIST

NCT00323830

458

D2, R0

2006– 2009

CH, KR, TW

CLASSIC

NCT00411229

1,035

D2, R0

Surgery alone

0.738 (0.498– 1.093)

.17

5-year RFS: 83.7

(77.1%– 90.2%)

.13

vs. 88.8%

(83.2%– 94.3%)

5-year OS: 60.9%

(52.6%– 69.2%)

vs. 62.0%

(53.7%– 70.2%)

5-year OS: 73%

0.48 (0.26– 0.89)

.017

0.68 (0.52– 0.87)

.003

0.740 (0.520– 1.050)

.0922

0.58 (0.47– 0.72)

80%) and has a very high number of tribal and socially challenged families (up to 90% in several districts). Three different WhatsApp groups, mentored by an oncology specialist, have been formed. Cloud-based electronic medical records, to store data electronically, have also been initiated, and physicians are being encouraged to capture data. Registration and physical data records with photocopies of the medical documents are being carried by the patient and are also kept in hospital records.

THINKING DIFFERENTLY IN GLOBAL HEALTH IN ONCOLOGY

TABLE 2. Examples of Community Events Event

Description

Pink and blue bingo nights

Separate bingo events for men and women. The evening begins with a circle conversation about breast, cervical, colorectal, and prostate cancer screening. Fecal occult blood kits are distributed to those eligible for screening. Exams can be scheduled during the meeting. Bingo follows the cancer education component.

Cancer awareness poster/photography contests

This event can engage school-aged children and adolescents in cancer awareness. The community can sponsor a poster or photography contest that features healthy behaviors to prevent cancer. Winners are awarded prizes.

Generations of wellness photos

Women of all generations can attend mammography screening together and have a professional photo taken following the mammogram. Children and those not screened can be welcome to attend. The photo is printed on site and framed for the grandmother and her children and grandchildren.

Cup art

While waiting for a mammogram or cervical cancer–screening test, women can artistically paint a mug, which is later fired in a kiln and given to the woman at a later date. An educator is present to provide information about cancer prevention and early detection for both men and women. Women are encouraged to schedule and/or bring their husbands to the clinic for cancer screening.

Dress making/beading

Ceremonial dance is important to American Indians. Women can gather to work on competition dresses and complete beadwork while a guest speaker can discuss cancer screening.

“Tough Enough to Wear Pink”

Rodeo or athletic event that encourages both men and women to wear pink in honor of breast cancer awareness month.

More than 15,000 patients have been registered and have used various services. The number of new patients being registered is increasing regularly and per month varies from district to district in the range of 5 to 30. There are more than 400 outpatient visits per month in the best performing district. The number of inpatients is also increasing. The inpatient services are being used for palliative and supportive care as well. When a patient arrives at a unit, he/she is seen by the physician, and his/her histologic verification is confirmed. The patient is diagnosed through biopsy and appropriately staged. Diagnostic services, if not available in the hospital, are outsourced. After evaluation, the patient is brought to the group tumor board. Once the board reaches a decision, the patient is informed and counseled on the decided course of action. Chemotherapy, if needed, is started locally and offered at no cost. Surgery and radiotherapy services of the nearest cancer centers are used (Fig. 2).

FIGURE 1. Basic Drawing of a Nodal Cancer Unit

All centers have started performing chemotherapy services as per standard guidelines. The number of chemotherapy sessions ranges from 5 to 150 per month in different hospitals. All classes of drugs are being used, including those that are emetogenic or with potential of acute reactions. Initial outcomes suggest standard toxicity levels comparable to any other center. There has been only one case of mortality related to chemotherapy toxicity. Patients in advanced or terminal stages of the disease are receiving proper palliative and end-of-life care in the districts, which has come as a big relief for the families. These units have begun serving as centers of public education on cancer and are routinely involved in various activities like rallies, public speeches, poster exhibitions, leaflet distributions, etc. Media outlets and other government machinery are being used. The units have also become centers of professional education. One of the centers is being used for hands-on training for physicians and has been incorporated in to training modules. The physicians have been designated by the state as nodal cancer officers. Patient data are serving as a local cancer registry and enabling the studying and understanding of many small epidemiologic deviations in the districts, revealing quite striking variations in patterns across different districts. One of the best performing districts has registered 1,564 patients (790 [50.5%] males and 774 [49.5%] females). The distribution of most common cancers is listed in Tables 3 and 4, with head and neck cancer being the most common among men and breast cancer among women. Among men, lung is the second most common, whereas among women, it is head and neck cancer. Prostate cancer is commonly reported. In men, upper gastrointestinal malignancies (esophagus and stomach) are also common. In women, incidence of ovarian cancer is high, higher than cervical cancer. Incidence of hematologic malignancies including leukemia, lymphoma, and myeloma is higher. Multiple satisfaction surveys conducted among patients attending the services have shown complete confidence in asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 421

RODRIGUEZ ET AL

FIGURE 2. Process of Patient Care

the system. Many patients have chosen district hospitals over tertiary centers for comparable services. All countries with a high burden of poor, uneducated, tribal, and socially challenged populations, including India, are facing a huge problem of access to care. The problem is multiplied because of the unavailability of qualified oncology personnel or specialized cancer centers. This health care delivery model tries to find a solution to these issues. Bringing affordable care physically close to the population can only bring a positive change. This can be fit into the diagonal approach to system strengthening. More and more tertiarylevel specialized cancer centers have shown confidence in districts and have started referring back local patients for intermediate care. More and more chemotherapy sessions are being performed. This proves the reproducibility of the chemotherapy facility under guidance of oncologists. The model reinforces the need for as well as the acceptance of decentralized specialized care. The program has now been running for more than 3 years and has proven its sustainability. Acceptance by other state governments, after evaluation, shows its administrative and political acceptability. Financial outcomes must be analyzed, but one thing is certain: as government services are free, patients are significantly saving out-of-pocket costs by being locally treated and not having to travel to seek care. The extension of cancer services via a remote support system improves access to care, especially for those living in rural and underserved areas with complex health problems. With the use of electronic medical record– and

WhatsApp-based interfaces, specialists are able to consult primary care providers like doctors and nurses on the care of patients with cancer, including the administration of complex chemotherapy protocols and management of side effects. Although a toxicity study is currently underway in district hospital settings, more than 3,000 patients have been treated at these centers. Many of these patients had previously been unable to receive chemotherapy because of access barriers. Those who received chemotherapy both at private or tertiary establishments and district hospitals have not reported any major differences in toxicity or clinicians’ ability to manage toxicity and adverse effects. The primary goal of patients accessing treatment at the standard of care was met. The results of this model show that it is an effective way to treat patients with cancer, administer chemotherapy, and provide palliative care in underserved areas. Implementation of this model would allow other states and nations with limited resources to treat greater numbers of patients with cancer than they are currently able to treat. Epidemiologic data being generated appear to be different from those of the national data registry. For example, the burden of cervical cancer appears to be lower, whereas ovarian cancer is much higher. These differences are important and warrant serious thinking over causative mechanisms. If micromanagement of preventive strategies is to be planned, these data could be very helpful. The methodology of counseling camps is turning out to be an excellent tool for the education of local physicians,

TABLE 3. Disease Pattern in Male Population (790 Patients)

TABLE 4. Disease Pattern in Female Population (774 Patients)

Type

No. of Patients

Percent

Type

No. of Patients

Percent

Head and neck

353

44.7

Breast

308

39.8

Lung

110

14.1

Head and neck

10.5

Hematologic malignancy

12.1

Ovary

08.9

Prostate

05.9

Hematologic malignancies

07.6

Upper gastrointestinal

05.8

Cervix

05.3

Other

138

17.4

Other

216

27.9

422 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

THINKING DIFFERENTLY IN GLOBAL HEALTH IN ONCOLOGY

helping in setting local systems and processes, building confidence among the local population, and creating wide public awareness on the issue of cancer. Increasing attendance in districts is showing better participation of patients in the care continuum. The point of contact for cancer vis-avis nodal cancer units created are helping to drive cancer care more effectively. These physicians are now undertaking and leading all of the following activities: communitybased cancer awareness, prevention, education, counseling and appropriate referrals, administering chemotherapy, conducting post-treatment surveillance, and providing palliative care.20 This model offers an alternative solution to managing workforce issues in oncology and establishes a new model of health care delivery in cancer care. The innovative model of empowerment using existing infrastructure and human resources touches on all proposed building blocks of an effective health system as advocated by the World Health Organization and has the potential to expand to other countries with limited resources.21 This model can serve as an important role in expansion to universal health care. The empowerment of an alternative oncology workforce using basic level physicians can help solve many global issues of access to cancer care.

A DIAGONAL RESPONSE TO WOMEN’S CANCERS: EXAMPLES FROM THE MEXICAN HEALTH SYSTEM

Effective health systems must encompass the six overlapping components of the cancer care and control continuum by developing integrated programs for primary prevention, early detection, diagnosis, treatment, survivorship, and long-term follow-up and palliation; in other words, mapping and supporting the cancer- and patient-specific journey.22 A diagonal approach is a strategy in which resources for disease-specific intervention priorities, like cancer, are distributed in ways that strengthen the entire health system by driving improvements in systemic areas including human resource development, financing, service provision, drug supply, and quality assurance.23 This approach overcomes the barriers between vertical (disease-specific) and horizontal (systemic) approaches by making full use of potential synergies between different programs and offers the opportunity to implement person-centered, instead of disease-focused approaches.24 A diagonal approach to cancer care addresses the false dichotomy of prevention versus treatment by strengthening integration of programs along the entire continuum of care. This approach can help to link cancer care and control with many services associated with a broad range of health promotion and treatment activities and reinforce human resources and physical infrastructure in health systems in ways that avoid creation of parallel structures for service delivery.25 A diagonal response also seeks to identify opportunities for optimal use of existing health programs or platforms, including those in other sectors, such as education, to address multiple health priorities and raise public awareness.

In this article, we discuss examples for the case of Mexico and women’s cancers. Although Mexico has seen a steady decline in cervical cancer mortality, which peaked at close to 16 per 100,000 women in the late 1980s and then steadily declined to a rate of less than 8 in 2008, breast cancer mortality rose steadily, reaching over 9 per 100,000 by 1995 and has remained relatively stable since.26,27 One example of a diagonal approach is the inclusion of cancer in national health insurance programs. In 2003, Mexico underwent a remarkable health reform that introduced the System of Social Protection in Health that includes a publicly funded health insurance scheme, the Seguro Popular de Salud (Popular Health Insurance), to cover universal access to an essential package of services with financial protection, especially targeting the poor and informal workers.28 As of 2012, the Seguro Popular had affiliated and covered more than 50 million previously uninsured Mexicans and by 2015 further expanded coverage to reach more than 56 million people. The number of covered diseases and interventions has steadily and considerably increased over time, including a growing list of cancers.24 Both breast and cervical cancer treatment are included in Mexico’s Seguro Popular since 2005 and 2007, respectively. Despite the inclusion of breast cancer in Seguro Popular, access to services for early detection of breast cancer remains limited. The 2012 National Health and Nutrition Survey showed that only one in five Mexican women ages 40 to 69 reported having an annual mammogram or breast clinical exam, with large disparities across the poorest and wealthiest quintiles. The majority of hospital cases are detected at later stages, especially in poorer states and municipalities, and mortality rates are high and increasing despite better access to treatment.24 To address the problem of late detection, a number of innovative education, training, and awareness-building interventions have been put in place. A prominent example of the potential for applying a diagonal approach is to integrate interventions for the prevention, early detection, treatment, survivorship, and palliation of women’s cancer into antipoverty or maternal and child health programs. For example, the Mexican human development and poverty alleviation program, Oportunidades (now called Prospera), is a social welfare scheme created in 1997 that offers conditional cash transfers to more than 90% of poor, urban, and rural families for the purpose of promoting education, health, and nutrition.24 Women are the recipients of the cash transfers, and as part of the program, participate in a variety of information and educational outreach activities.29 Cervical cancer mortality in Mexico has concentrated among the poorest quintiles despite the fact that this is an easily preventable disease through early detection.26 Oportunidades and now Prospera include a broad range of activities around cervical cancer. Education initiatives for cervical cancer prevention and clinic visit incentives for women to receive the Papanicolaou test have shown a positive impact on increasing the numbers of beneficiary women who are tested for cervical cancer as well as the willingness of asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 423

RODRIGUEZ ET AL

indigenous women to take the test and encourage other women in their communities to do so.29 Furthermore, as of 2015, the HPV vaccination is now included in the Prospera package. The inclusion of information on breast cancer has been much more difficult to achieve. As part of an effort to increase access to early detection of breast cancer, information was included in manuals distributed through Oportunidades, and the program was encouraged to include education and awareness-building on women’s cancers in community workshops and educational outreach.26,30 Through Oportunidades, female household heads and community health promoters throughout the country were trained with basic information about breast health and self-examination.24 This work must be evaluated and extended as part of Prospera. Other breast health awareness initiatives in Mexico have explored and developed various educational innovations to provide breast health education for women in their communities and to ensure a properly trained primary health workforce. A multi-institutional group, spearheaded by the civil society organization Tómatelo a Pecho, A.C., and working with the Seguro Popular, the National Institute of Public Health of Mexico, and state governments, was created to train an extensive network of community health workers, nurses, and primary care physicians on early diagnosis and the triaging of high-risk cases with family history.31 The group worked with local organizations to develop and implement a “train-the-trainer” program to improve breast cancer knowledge among community health workers, including professional health promoters who then trained nonprofessional community health promoters. The educational strategy was designed using a competency-based approach with an emphasis on student-centered activities, innovative tools, collaborative work, and hands-on problem-solving. Training materials included manuals for physicians and nurses, educational kits and workshop development guides for health promoters, and various recreational games involving the identification of warning signs, breast self-examination techniques, treatment, and return to daily life.32 Participants were surveyed before and after training and demonstrated improvements in understanding of breast cancer

as a problem, understanding of screening, treatment, and insurance coverage issues, and knowledge of breast cancer risk factors, symptoms, and what constitutes a family history of breast cancer.31 The training modules have since been and are now available online. More extensive training on survivorship, pain management, and palliative care is now underway.24 These innovative interventions to improve training, education, and awareness constitute a diagonal approach and build on overall efforts to strengthen primary care and link to specialized tertiary treatment options, instead of developing parallel systems for early detection of cancers. These examples deserve rigorous evaluation, as they suggest that diagonal strategies for early detection of breast cancer can be implemented through integration into national insurance and social security schemes and that antipoverty, maternal and child health, sexual and reproductive health, and other programs can serve as platforms for addressing early detection and prevention of cancer.24

CONCLUSION

The case studies presented in this article discuss different strategies for improving cancer management along the entire continuum of care. Although cultural diversity and regional idiosyncrasies across the world will always exist, and context-relevant solutions will always be needed, there are key challenges in cancer care that are universal. Improving global cancer survival requires innovative solutions for the education of both health care professionals and the public. Community empowerment through training of community health promoters and alternative workforces can ensure the uptake and sustainability of improvements in early detection and quality of care. Finally, these solutions need not be cancer specific or developed in parallel silos; health systems can be strengthened through diagonal approaches that find synergies across diseases and build off existing programs or platforms.

ACKNOWLEGMENT

G. Lopes and F. M. Knaul contributed equally to this article as senior authors.

References 1. World Health Organization. Cancer Fact Sheet. http://www. who.int/mediacentre/factsheets/fs297/en/. Accessed November 2, 2011. 2. American Cancer Society. Cancer Facts and Figures 2016. https:// www.cancer.org/research/cancer-facts-statistics/all-cancerfacts-figures/cancer-facts-figures-2016.html. Accessed January 10, 2017. 3. LeDuc T. World Life Expectancy. World Health Rankings. http://www. worldlifeexpectancy.com/. Accessed January 25, 2017. 4. Wagner JM. Shifting family structure: Theoretical perspectives of worldwide population and family composition. In Stevenson EL and Herschberger PE (eds). Fertility and Assisted Reproductive Technology

424 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

(ART): Theory, Research, Policy, and Practice for Health Care Practitioners. New York: Springer; 2016;15-28. 5. Vogel O, Cowens-Alvarado R, Eschiti V, et al. Circle of life cancer education: giving voice to American Indian and Alaska Native communities. J Cancer Educ. 2013;28:565-572. 6. Ehsani M, Taleghani F, Hematti S, et al. Perceptions of patients, families, physicians and nurses regarding challenges in cancer disclosure: a descriptive qualitative study. Eur J Oncol Nurs. 2016;25:55-61. 7. Abazari P, Taleghani F, Hematti S, et al. Exploring perceptions and preferences of patients, families, physicians, and nurses regarding cancer disclosure: a descriptive qualitative study. Support Care Cancer. 2016;24:4651-4659.

THINKING DIFFERENTLY IN GLOBAL HEALTH IN ONCOLOGY

8. Haozous EA, Knobf MT, Brant JM. Understanding the cancer pain experience in American Indians of the Northern Plains. Psychooncology. 2011;20:404-410. 9. Silbermann M (ed). Cancer Care in Countries and Societies in Transition. Cham, Switzerland: Springer; 2016. 10. Bassah N, Seymour J, Cox K. A modified systematic review of research evidence about education for pre-registration nurses in palliative care. BMC Palliat Care. 2014;13:56. 11. Eggenberger E, Heimerl K, Bennett MI. Communication skills training in dementia care: a systematic review of effectiveness, training content, and didactic methods in different care settings. Int Psychogeriatr. 2013;25:345-358. 12. Gysels M, Richardson A, Higginson IJ. Communication training for health professionals who care for patients with cancer: a systematic review of effectiveness. Support Care Cancer. 2004;12:692-700. 13. Lipmanowicz H, McCandless K. The Surprising Power of Liberating Structures: Simple Rules to Unleash a Culture of Innovation. Seattle, WA: Liberating Structures Press; 2013. 14. Orians CE, Erb J, Kenyon KL, et al. Public education strategies for delivering breast and cervical cancer screening in American Indian and Alaska Native populations. J Public Health Manag Pract. 2004;10:46-53. 15. Becker SA, Affonso DD, Beard MB. Talking circles: Northern Plains tribes American Indian women’s views of cancer as a health issue. Public Health Nurs. 2006;23:27-36. 16. Matloub J, Creswell PD, Strickland R, et al. Lessons learned from a community-based participatory research project to improve American Indian cancer surveillance. Prog Community Health Partnersh. 2009;3:47-52. 17. Pendharkar D, Agarwal P, Tripathi C. Innovative healthcare delivery model to expand access and outreach of cancer care services. J Cancer Res Ther. 2016;12:2-5. 18. Jhaveri D, Larkins S, Kelly J, et al. Remote chemotherapy supervision model for rural cancer care: perspectives of health professionals. Eur J Cancer Care (Engl). 2016;25:93-98. 19. Arora S, Thornton K, Murata G, et al. Outcomes of treatment for hepatitis C virus infection by primary care providers. N Engl J Med. 2011;364:2199-2207. 20. Miesfeldt S. The inaugural cancer control in primary care course in Bhopal India. https://connection.asco.org/blogs/inaugural-cancercontrol-primary-care-course-bhopal-india. Accessed on March 13, 2016.

21. World Health Organization. Everybody business: strengthening health systems to improve outcomes. WHO’s framework for action. http:// www.who.int/healthsystems/strategy/everybodys_business.pdf. Accessed March 16, 2017. 22. Knaul FM, Alleyne G, Piot P, et al. Health system strengthening and cancer: a diagonal response to the challenge of chronicity. In Knaul FM, Gralow JR, Atun R, Bhadelia A (eds). Closing the Cancer Divide: An Equity Imperative. Cambridge, MA: Harvard Global Equity Initiative; 2012;95-122. 23. Frenk J. Bridging the divide: global lessons from evidence-based health policy in Mexico. Lancet. 2006;368:954-961. 24. Knaul FM, Bhadelia A, Atun R, et al. Achieving effective universal health coverage and diagonal approaches to care for chronic illnesses. Health Aff (Millwood). 2015;34:1514-1522. 25. Farmer P, Frenk J, Knaul FM, et al. Expansion of cancer care and control in countries of low and middle income: a call to action. Lancet. 2010;376:1186-1193. 26. Knaul FM, Bhadelia A, Gralow J, et al. Meeting the emerging challenge of breast and cervical cancer in low- and middle-income countries. Int J Gynaecol Obstet. 2012;119:S85-S88. 27. Knaul FM, Nigenda G, Lozano R, et al. Breast cancer in Mexico: a pressing priority. Reprod Health Matters. 2008;16:113-123. 28. Knaul FM, González-Pier E, Gómez-Dantés O, et al. The quest for universal health coverage: achieving social protection for all in Mexico. Lancet. 2012;380:1259-1279. 29. Sánchez López G. External Evaluation of Oportunidades 2008. 19972007: 10 Years of Intervention in Rural Areas. Volume II. The Challenge of Services Quality: Health and Nutrition Outcomes. http://lanic. utexas.edu/project/etext/oportunidades/2008/sanchez_eng.pdf. Accessed March 16, 2017. 30. Luciani S, Cabanes A, Prieto-Lara E, et al. Cervical and female breast cancers in the Americas: current situation and opportunities for action. Bull World Health Organ. 2013;91:640-649. 31. Keating NL, Kouri EM, Ornelas HA, et al. Evaluation of breast cancer knowledge among health promoters in Mexico before and after focused training. Oncologist. 2014;19:1091-1099. 32. Magaña-Valladares L, González-Robledo MC, Rosas-Magallanes C, et al. Training primary health professionals in breast cancer prevention: evidence and experience from Mexico. J Cancer Educ. Epub 30 Jun 2016.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 425

ASAMURA, AOKAGE, AND YOTSUKURA

Wedge Resection Versus Anatomic Resection: Extent of Surgical Resection for Stage I and II Lung Cancer Hisao Asamura, MD, Keiju Aokage, MD, and Masaya Yotsukura, MD OVERVIEW Currently, surgery for lung cancer with curative intent consists of resection (removal) of the proper extent of lung parenchyma that bears the cancer lesion along with locoregional lymph nodes to assess possible cancer metastasis. Lobectomy, at least, is preferred with regard to the extent of parenchymal resection. The history of lung cancer surgery, which started around 1933 as pneumonectomy (resection of the entire lung on either side), can be characterized as an attempt to minimize the extent of parenchymal resection. In the early 1960s, pneumonectomy was replaced by lobectomy, which has long been respected as the standard surgical mode. However, the transition from lobectomy to a lesser resection, such as segmentectomy or wedge resection, was not recommended because of the results of a randomized trial performed by the North American Lung Cancer Study Group in the 1980s. As of now, the extent of parenchymal resection remains lobectomy, and lesser resection is indicated only for patients who have a compromised pulmonary reserve. Very recently, because of the advent of CT screening programs and improvements in imaging technology, fainter and smaller lung cancers are being discovered. For these smaller and earlier lung cancers, there is some uncertainty about whether lobectomy still should be indicated, as it is for larger tumors with a diameter of 3 cm or more. Therefore, several randomized trials are ongoing to compare lobectomy with lesser resections; endpoints are overall survival and postoperative pulmonary function. Until the results of these trials are available, lung cancer should still be removed by lobectomy rather than by limited resection, such as segmentectomy or wedge resection.

urgery continues to be the mainstay in the treatment of early-stage lung cancer and may ensure the total removal of cancer cells limited to the lung parenchyma and locoregional lymph nodes to achieve a cure. The first successful en bloc left-sided pneumonectomy for lung cancer was performed by Graham and Singer1 in 1933. Pneumonectomy is the removal of one entire lung on either side, which represents the largest amount of lung parenchyma that could be removed safely by surgery. Since then, surgeons have tried to remove tumors by removing smaller amounts of lung parenchyma while considering postoperative pulmonary function. The extent of parenchymal resection gradually was reduced from pneumonectomy to lobectomy in the early 1960s, and additional attempts to minimize the extent of resection even more, from lobectomy to sublobar resection (i.e., wedge resection and segmentectomy), were made in the 1980s. However, a smaller resection inevitably increases the risk of incomplete removal of the tumor and subsequent local tumor recurrence. In this sense, surgeons have been trying to achieve an optimal balance between the radicality of cancer surgery and a safe surgical margin.2

On the basis of the results of the landmark study by Ginsberg et al3 in the 1980s, lobectomy has been considered the optimal mode of pulmonary resection for lung cancer when combined with clearance of the lymphatic route of the pulmonary hilum and mediastinum, a procedure originally described by Cahan4 as radical lobectomy. However, because of the advent of CT screening programs and improvements in imaging technology, fainter and smaller lung cancers are being encountered in daily practice. For these smaller and earlier lung cancers, it is not quite clear whether lobectomy is an optimal surgery. Therefore, very recently, several randomized trials have been undertaken to compare lobectomy and lesser, limited resection.5 This article presents the technical aspects of standard and lesser, limited resections for lung cancer, a historical overview of the evolution of lung cancer surgery since the 1930s, and previous and ongoing clinical studies that aim to identify the optimal extent of the lung parenchyma to be resected in radical, curative surgery for early-stage (stages I and II) lung cancer.

From the Division of Thoracic Surgery, Keio University School of Medicine, Tokyo, Japan; Division of Thoracic Surgery, National Cancer Center Hospital East, Chiba, Japan; Keio University School of Medicine, Tokyo, Japan. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Hisao Asamura, MD, Division of Thoracic Surgery, Keio University School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan; email: [emailprotected]. © 2017 American Society of Clinical Oncology

426 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

EXTENT OF PARENCHYMAL PULMONARY RESECTION FOR LUNG CANCER

FIGURE 1. Schematic Drawing of the Extent of Pulmonary Resections

(A) Wedge resection; (B) segmentectomy; (C) lobectomy; and (D) pneumonectomy.

PARENCHYMAL PULMONARY RESECTION FOR LUNG CANCER: OPERATIVE MODES TO RESECT DIFFERENT AMOUNTS OF THE LUNG PARENCHYMA

The present-day surgery for lung cancer with curative intent consists of complete removal of the primary lesion of the lung and clearance of the locoregional lymphatic drainage

KEY POINTS • The present-day standard operative mode for the resection of lung cancer is at least lobectomy (i.e., lobectomy, pneumonectomy) with mediastinal/hilar lymph node dissection/sampling. • Lesser resections, such as wedge resection and segmental resection, are referred to as limited resection or sublobar resection. • Only one randomized trial to compare limited resection with lobectomy showed a higher incidence of local recurrence and a poor prognosis with limited resection. • Several randomized trials are underway to determine whether limited resection can provide at least an equivalent prognosis and better postoperative pulmonary function compared with lobectomy in earlystage non–small cell lung cancer. • Limited resection is routinely used for the resection of lung cancer in patients with limited pulmonary reserve or high risk of comorbidity.

route. Therefore, it involves resection (removal) of the proper extent of lung parenchyma that bears the cancer lesion together with locoregional lymph nodes to assess possible cancer metastasis.6 For resection of the lung parenchyma, the following surgical modes technically could be selected according to the extent of the disease and its nature (Fig. 1): pneumonectomy (removal of the entire lung on either side), bilobectomy (removal of two adjacent lobes), lobectomy (removal of a single lobe), segmentectomy (removal of a single segment or adjacent segments), and wedge/partial resection (removal of wedge-shaped parenchyma regardless of the bronchovascular anatomy). If the proximal portion of the bronchus is involved by direct extension of the tumor or if there is lymph node metastasis at the hilum and neither lobectomy nor pneumonectomy could ensure that the resected end of the bronchus would be tumor free, a sleeve resection, which entails combined resection of the proximal portion of the bronchus and reconstruction, might be considered in conjunction with lobectomy (sleeve lobectomy) or pneumonectomy (sleeve pneumonectomy) to ensure a safe surgical margin. Sleeve resection enables tumor-free resection without sacrifice of uninvolved lung parenchyma. The techniques used at the pulmonary hilum, though, can be divided into two types: anatomic (pneumonectomy, bilobectomy, lobectomy, and segmentectomy) and nonanatomic resection (wedge resection). In anatomic resection, the extent of resected parenchyma is determined according to the extent of perfusion of pulmonary vessels as well as by the extent of aeration of bronchi, which are divided at the asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 427

ASAMURA, AOKAGE, AND YOTSUKURA

hilum. In nonanatomic resection, the extent of parenchymal resection is determined solely according to the location of the target lesion.

SUBLOBAR RESECTIONS

Among the modes of pulmonary resection, segmentectomy and wedge resection are both referred to as sublobar resection, because the resected parenchyma by these two modes are smaller than lobe. Despite this similar classification, their technical characteristics are quite different, especially at the hilum. In segmental resection, the hilum needs to be dissected for individual division and ligation, exactly as in lobectomy and pneumonectomy. This is why segmentectomy is considered anatomic resection (Fig. 2A). Conversely, in wedge resection, the extent of resection is determined according to what the surgeons think is appropriate, regardless of the bronchovascular anatomy. Thus, wedge resection is considered nonanatomic resection (Fig. 2B). Therefore, although they both are considered sublobar resection, these two operative modes have technical differences.

Evolution of Lung Cancer Surgery: A History of Minimization

The history of lung cancer surgery can be thought of in terms of minimization of the extent of parenchymal resection. Surgeons try to achieve an optimal balance between the radicality of cancer surgery and the preservation of postoperative lung function. The earliest report about pneumonectomy of the right side was by Kummel7 in 1910; the patient, a 40-year-old man, died on the sixth postoperative day. After a series of early postoperative deaths after pneumonectomy in the 1920s, Evarts Graham1 in St. Louis reported the first successful pneumonectomy,

which was performed on a 48-year-old man with lung cancer by using a tourniquet technique, in 1933. After this landmark operation, successful reports of pneumonectomy for lung cancer were documented. In 1940, Overholt8 reviewed 110 pneumonectomies, including 15 of his own patient cases, for benign and malignant lung diseases, and reported a mortality rate of 65% for the malignant group. He also stated that the operability of primary lung cancer was 25%.8 In the 1940s, pneumonectomy was established as the standard mode of pulmonary resection for lung cancer. In 1950, Allison9 performed pneumonectomy with intrapericardial ligation of the pulmonary vessels, and, more importantly, the addition of locoregional lymph node dissection to pneumonectomy was proposed as radical surgery for lung cancer. Cahan10 called this procedure radical pneumonectomy, which indicated the combination of parenchymal resection and lymph node dissection. In the 1950s and 1960s, pneumonectomy gradually was replaced by lobectomy. This era was the transitional phase from pneumonectomy to lobectomy for lung cancer. In 1950, Churchill11 reported a better 5-year survival rate with lobectomy (19%) than with pneumonectomy (12%). In 1960, Cahan4 again defined radical lobectomy as an operation in which one or two lobes of an entire lung are excised in a block dissection along with certain of their regional hilar and mediastinal lymphatics (Fig. 3). The extent of lymph node dissection also was defined according to the primary site of the lung cancer. Cahan4 analyzed the outcome of 48 radical lobectomies for primary and metastatic lung cancers and concluded that survival for 5 years or longer was associated in large part with more extensive lymphatic dissection and radical lobectomy. In the 1970s and 1980s, lobectomy became recognized as the standard mode of resection for primary lung

FIGURE 2. Sublobar Resections

(A) Posterior segmentectomy of the right upper lobe. Note the individual division and ligation of the pulmonary bronchovascular structures before the division of the lung parenchyma. (B) Wide wedge resection. Abbreviations: S1, superior segment; S2, posterior segment; S3, anterior segment of the right upper lobe.

428 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

EXTENT OF PARENCHYMAL PULMONARY RESECTION FOR LUNG CANCER

FIGURE 3. Radical Lobectomy by Cahan (1960)4

The extent of parenchymal resection (lobe) and lymph node dissection is determined according to the location of the primary tumor.

cancer, and pneumonectomy was no longer the standard approach. However, lesser resections (i.e., segmentectomy and wedge resection) for peripheral lung cancer always have been reserved for compromised patients who could not tolerate more extensive procedures, such as lobectomy or pneumonectomy. Churchill and Belsey12 originally introduced segmental resection as segmental pneumonectomy for the treatment of benign lung diseases in 1939. This technique was advocated for use later in patients with cancer who had limited pulmonary reserve and inoperable disease. In 1973, Jensik et al13 suggested that anatomic pulmonary segmentectomy could be applied effectively to small primary lung cancers when the surgical margins were sufficient. These reports stimulated a debate about the optimal resection technique for early-stage non–small cell lung cancer (NSCLC). The optimal technique was addressed in a prospective, randomized trial conducted by the Lung Cancer Study Group in 247 patients with stage IA NSCLC.3 Limited pulmonary resection, including anatomic segmentectomy and nonanatomic wide-wedge resections, was compared with lobectomy to evaluate the postoperative prognosis and pulmonary function. This study showed a 39% increase in local recurrence and a nonsignificant decrease in overall survival after sublobar resection. It included patients with tumors of 3 cm or less and a significant number of nonanatomic wedge resections (i.e., one of three sublobar procedures). In retrospect, both of these parameters may have significantly limited the effectiveness of sublobar resection. In short, this study solidified lobectomy as the procedure of choice for the treatment of this disease on the basis of the inferior postoperative survival and increased locoregional recurrence in the limited-resection group. This still is the only randomized trial to compare limited resection with lobectomy directly;

therefore, the gold standard for lung cancer remains lobectomy with lymph node sampling/dissection. Recently, the results of a prospective, randomized trial (ACOSOG Z0030) to evaluate the prognostic significance of lymph node dissection in lung cancer were published.14,15 This study compared systematic sampling and dissection for N0 or nonhilar N1, T1, or T2 NSCLC (stages I and II). This study did not show that lymph node dissection offered a prognostic advantage compared with sampling. The authors concluded that, if systematic and thorough presection sampling of the mediastinal and hilar lymph nodes is negative, mediastinal lymph node dissection does not improve survival in patients with early-stage NSCLC, but these results are not generalizable to patients whose disease was staged radiographically or those with higher-stage tumors. On the basis of the combination of these results, it is widely accepted that the present-day gold standard should be lobectomy, at least, with lymph node sampling/dissection for stages I and II disease.

A NEW WHO CLASSIFICATION OF LUNG CANCER AND NEW CONCEPT OF EARLY FORMS OF ADENOCARCINOMA

Recently, a new classification of adenocarcinoma of the lung was published to provide uniform terminology and diagnostic criteria, with a particular focus on the classification of earlier forms of adenocarcinoma.16 New concepts, such as adenocarcinoma in situ and minimally invasive adenocarcinoma for small solitary adenocarcinomas with either pure lepidic growth or predominant lepidic growth of less than 5 mm, were introduced to define patients who, if they were to undergo complete resection, could be expected to have 100% or near-100% disease-specific survival, respectively. However, adenocarcinomas are classified according to the predominant pattern (lepidic, acinar, papillary, or asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 429

ASAMURA, AOKAGE, AND YOTSUKURA

solid) after comprehensive histologic subtyping. These earlier forms, such as adenocarcinoma in situ and minimally invasive adenocarcinoma, were recognized only after the advent of high-resolution CT scans and the dissemination of CT screening programs. In a registry study, patients with an earlier disease stage were included in a stage IA group, and the proportion of these patients might be associated with the difference in survival, especially for stage IA disease. The surgical significance of these classifications also has been analyzed.17 Recently, the prognosis of 545 patients with radiographically determined noninvasive adenocarcinomas of the lung of ground-glass opacity was reported; a consolidation-to-tumor ratio of 0.25 or less in cT1a was used as the radiologic criteria of noninvasive cancer, and the lesion was resected by lobectomy.18,19 The reported 5-year survival rates for noninvasive and invasive adenocarcinomas were 96.7% and 88.9%, respectively. This superb surgical outcome supports the possibility of lesser resection, such as segmentectomy and wedge resection, for patients with early lung cancers.

POSSIBILITY OF SUBLOBAR, LIMITED RESECTION FOR EARLY-STAGE LUNG CANCER

Technical and Pathologic Considerations

Several factors must be weighed in terms of the characteristics of sublobar resection, especially segmental resection, when it is considered as an option for radical resection for lung cancer, for which no tumor tissue can be left behind. In sublobar resections, the lung parenchyma must be transected and divided for completion of the procedure, whereas, in lobectomy, the fissure is divided to remove the entire lobe. In relation to the nature of these procedures, some technical limitations in sublobar resection include tumor size, location, histologic type as lung cancer, and nodal involvement. Tumor size and location, in particular, are closely related to the safe surgical margin when performed as radical resection. Tumor size and local recurrence after sublobar resection have been extensively studied. It has been shown repeatedly that tumors larger than 2 cm have a significantly higher local recurrence rate than those smaller than 2 cm.20-22 Another important factor is the location of the tumor in relation to the pleural surface/hilum. A fundamental, geometric understanding of a lung segment is that the segment is fan shaped with the base on the pleural surface and the apex at the pulmonary hilum. Therefore, the distance between a tumor and the resection line inevitably is closer when the tumor is located close to the hilum, even if the tumor is small. Generally, even for tumors of 2 cm or less in diameter, segmentectomy/wedge resection should be performed only when tumors are located in the outer third of the lung parenchyma. Other unfavorable factors for limited resection are an aggressive histology, such as small cell carcinoma, and lymph node involvement. These conditions indicate that there is a greater possibility of tumor spread in the lobe that contains the segment. Very recently, the results of 164 intentional, extended segmentectomies were reported as a comparison with 430 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

lobectomy.23 The incidence of locoregional recurrence after segmentectomy was reported according to the location of the resected site. The incidences of local and locoregional recurrences, respectively, were 21.9% and 21.9% for the right upper lobe, 10.5% and 15.8% for the left upper lobe, 4.2% and 4.2% for the bilateral superior segment of the lower lobes, and 20.8% and 37.5% for the bilateral basal segments. The incidence of local or locoregional recurrence was significantly higher for segmentectomies of the right upper lobe and bilateral basal segments. These results indicated that tumors in the basal segments should be resected by lobectomy rather than by segmentectomy, even if the tumors are small. Segmentectomy could be desirable or undesirable according to on the anatomic location. This issue is related solely to the anatomic structure of the lobe and segment as a functional unit in the lung.

Oncologic Considerations

Currently, limited, sublobar resection for lung cancer could be considered for the following situations: (1) limited resection of T1N0M0 lung cancer in compromised patients who have limited cardiopulmonary reserve regardless of the type of lesion; (2) limited resection for early lung cancer with a predominantly ground-glass opaque appearance (pathologic adenocarcinoma in situ/minimally invasive adenocarcinoma); and (3) limited resection for small but invasive lung cancers located in the periphery of the lung. As discussed in the previous section, considerable interest in sublobar resection began in the 1970s and 1980s, when the feasibility of limited resection for patients with compromised cardiopulmonary reserve was demonstrated. Then, the 5-year survival rate and the recurrence rate were considered inferior to those for lobectomy, and sublobar resection was restricted to patients who had impaired cardiac function or significant comorbidities that would preclude conventional lobectomy. Decreased survival and increased recurrence were demonstrated in 173 patients with stage I NSCLC who underwent sublobar resection or lobectomy, as reported in an early work by Warren and Faber.24 However, recent single-institution, retrospective investigations to evaluate the equivalency of sublobar resection to lobectomy in patients who have limited cardiopulmonary reserve contradicted these earlier results and demonstrated that stage I disease has a survival advantage regardless of the extent of surgical resection or histologic subtype. Campione et al25 found no significant difference in survival between lobectomy and anatomic segmentectomy in a series of 121 patients with stage IA disease. Other studies showed similar results with segmentectomy and lobectomy.26-33 The surgical indication of limited resection for patients with stage IA disease who have limited cardiopulmonary reserve is respected as a reasonable treatment of choice. As discussed in the previous section, adenocarcinoma in situ and minimally invasive adenocarcinoma comprise a novel concept that refers to the noninvasive or minimally invasive nature of adenocarcinoma with a unique, ground-glass opacity. The use of limited resection for patients who have nonsolid (pure)

EXTENT OF PARENCHYMAL PULMONARY RESECTION FOR LUNG CANCER

or part-solid (mixed) ground-glass opaque presentation has been assessed in a variety of retrospective Japanese studies. In each of these studies, patients with adenocarcinoma in situ or minimally invasive adenocarcinoma tumors had prolonged survival and lower recurrence after resection than those with other subtypes of NSCLC. The application of sublobar resection for these early tumors was based upon a clinicopathologic study on the correlation between the degree of invasive growth (stromal invasion) and the prognosis. Sakurai et al34 classified 380 resected adenocarcinomas of 2 cm or less in diameter according to the degree of invasive growth (i.e., structural deformity and location in the adenocarcinoma) and showed that patients who had noninvasive or minimally invasive adenocarcinomas could achieve 100% survival despite the mode of pulmonary resection. On the basis of these clinicopathologic observations, sublobar resection for adenocarcinoma in situ/minimally invasive adenocarcinoma tumors with ground-glass opacity could be considered reasonable according to the location and size of the tumor. The indication of sublobar resection needs to be considered from not only an oncologic but also an anatomic perspective. For occurrences in which the tumor is located deep inside the lung parenchyma, a safe surgical margin cannot be ensured in sublobar resection, because the surgical margin is close to the hilar structures. A lung segment is generally shaped like a sector, with the apex at the hilum, which indicates that there is a shorter distance between the tumor and resected margin in the area close to the hilum. The tumor diameter also affects the distance to the surgical margin. Therefore, sublobar resection should be applied only when the tumor is located in the outer third of the lung parenchyma and preferably when the tumor is 2 cm or less in diameter. For tumors located in the inner two-thirds of

the lung parenchyma or for those larger than 2 cm in diameter, lobectomy still should be selected regardless of the tumor pathology.

Ongoing Trials to Compare Lobectomy and Limited Resection

For histologically invasive lung cancers that are located in the periphery of the lung as a small (2 cm or less: T1a) solitary nodule, the feasibility of limited, sublobar resection needs to be defined from a present-day perspective. This means that the Lung Cancer Study Group study in the late 1980s must be revised.3 Indeed, the present-day routine work-up for patients with resectable lung cancer is much different than the work-up used in the 1980s. A few prospective studies are ongoing. Randomized clinical trials that enrolled patients with peripheral lung cancers of no more than 2 cm in diameter as the target lesions are underway in the United States (Cancer and Leukemia Group B study CALGB140503) and Japan (Japan Clinical Oncology Group and West Japan Oncology Group study JCOG0802/ WJOG4607L).35 As of January 2017, the targeted number of patients in the Japan Clinical Oncology Group study was registered, and data maturation is awaited. The design of the phase III Japan Clinical Oncology Group study is shown in Fig. 4. In this trial, the endpoints are overall survival (primary) and postoperative pulmonary function (secondary), and the targeted accrual is 1,100 patients. If the results show that the prognoses of patients who undergo segmentectomy is not significantly inferior to that of patients who undergo lobectomy and that the postoperative pulmonary function is significantly better for the patients who undergo segmentectomy, then we can conclude definitively that the standard surgical mode for these early tumors should be segmentectomy.

FIGURE 4. Design of JCOG0802/WJOG4607L, a Randomized Trial to Compare the Prognoses and Postoperative Function After Segmentectomy or Lobectomy for Non–Small Cell Lung Cancer ≤ 2 cm Diameter

Abbreviation: OS, overall survival.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 431

ASAMURA, AOKAGE, AND YOTSUKURA

CONCLUSION

Lobectomy with hilar and mediastinal lymph node dissection/sampling is still the present-day gold standard for the resection of lung cancer. Sublobar resection such as segmentectomy and wedge resection could be reasonably used for compromised patients at high risk. The use of sublobar resection might be justified for most early lung cancer with

no or minimal inasive features located in the outer region of the lung parenchyma. The possibility of sublobar resection for lung cancer with overt invasive features is under investigation, with particular focus on tumors 2 cm or less in diameter. The results of several on-going trials are awaited. Lobectomy should be recognized as the standard mode of resection for good-risk patients.

References 1. Graham EA, Singer JJ. Successful removal of an entire lung for carcinoma of the bronchus. CA Cancer J Clin. 1974;24:238-242. 2. Asamura H, Grunenwald D, Pass H, et al (eds). The IASLC Multidisciplinary Approach to Thoracic Oncology. Aurora, CO: IASLC Press; 2014;403409. 3. Ginsberg RJ, Rubinstein LV. Randomized trial of lobectomy versus limited resection for T1 N0 non–small cell lung cancer. Ann Thorac Surg. 1995;60:615-622. 4. Cahan WG. Radical lobectomy. J Thorac Cardiovasc Surg. 1960;39:555572. 5. Aokage K, Yoshida J, Hishida T, et al. Limited resection for early-stage non–small cell lung cancer as function-preserving radical surgery: a review. Jpn J Clin Oncol. 2017;47:7-11. 6. Lee PC, Altoki NK. Extent of resection for stage I lung cancer. In Principles and Practice of Lung Cancer, (ed 4). Philadelphia, PA: Lippencott Williams & Wilkins; 2010;459-465. 7. Kummel H. Karzinom total resektion einer lunge wegen karzinom. Zentralbl Chir. 1911;38:427-428. 8. Overholt RH. Pneumonectomy for malignant and suppurative diseases of the lung. J Thorac Surg. 1940;9:17-61. 9. Allison PR. Intrapericardial approach to the lung root in the treatment of bronchial carcinoma by dissection pneumonectomy. J Thorac Surg. 1946;15:99-117. 10. Cahan WG, Watson WL, Pool JL. Radical pneumonectomy. J Thorac Surg. 1951;22:449-473. 11. Churchill ED, Sweet RH, Soutter L, et al. The surgical management of carcinoma of the lung: a study of the cases treated at the Massachusetts General Hospital from 1930 to 1950. J Thorac Surg. 1950;20:349-365. 12. Churchill ED, Belsey R. Segmental pneumonectomy in bronchiectasis: lingular segment of the left upper lobe. Ann Surg. 1939;109:481-499. 13. Jensik RJ, Faber LP, Milloy FJ, et al. Segmental resection for lung cancer: a fifteen-year experience. J Thorac Cardiovasc Surg. 1973;66:563-572. 14. Allen MS, Darling GE, Pechet TT, et al. Morbidity and mortality of major pulmonary resections in patients with early-stage lung cancer: initial results of the randomized, prospective ACOSOG Z0030 trial. Ann Thorac Surg. 2006;81:1013-1019. 15. Darling GE, Allen MS, Decker PA, et al. Randomized trial of mediastinal lymph node sampling versus complete lymphadenectomy during pulmonary resection in the patient with N0 or N1 (less than hilar) non–small cell carcinoma: results of the American College of Surgery Oncology Group Z0030 Trial. J Thorac Cardiovasc Surg. 2011;141:662670.

432 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

16. Travis WD, Brambilla E, Noguchi M, et al. International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society international multidisciplinary classification of lung adenocarcinoma. J Thorac Oncol. 2011;6:244-285. 17. Van Schil PE, Asamura H, Rusch VW, et al. Surgical implications of the new IASLC/ATS/ERS adenocarcinoma classification. Eur Respir J. 2012;39:478-486. 18. Asamura H, Hishida T, Suzuki K, et al. Radiographically determined noninvasive adenocarcinoma of the lung: survival outcomes of Japan Clinical Oncology Group 0201. J Thorac Cardiovasc Surg. 2013;146:2430. 19. Suzuki K, Koike T, Asakawa T, et al. A prospective radiological study of thin-section computed tomography to predict pathological noninvasiveness in peripheral clinical IA lung cancer (Japan Clinical Oncology Group 0201). J Thorac Oncol. 2011;6:751-756. 20. Bando T, Yamagihara K, Ohtake Y, et al. A new method of segmental resection for primary lung cancer: intermediate results. Eur J Cardiothorac Surg. 2002;21:894-899. 21. Fernando HC, Santos RS, Benfield JR, et al. Lobar and sublobar resection with and without brachytherapy for small stage IA non– small cell lung cancer. J Thorac Cardiovasc Surg. 2005;129:261-267. 22. Okada M, Nishio W, Sakamoto T, et al. Effect of tumor size on prognosis in patients with non–small cell lung cancer: the role of segmentectomy as a type of lesser resection. J Thorac Cardiovasc Surg. 2005;129:8793. 23. Nishio W, Yoshimura M, Maniwa Y, et al. Re-assessment of intentional extended segmentectomy for clinical T1aN0 non–small cell lung cancer. Ann Thorac Surg. 2016;102:1702-1710. 24. Warren WH, Faber LP. Segmentectomy versus lobectomy in patients with stage I pulmonary carcinoma: five-year survival and patterns of intrathoracic recurrence. J Thorac Cardiovasc Surg. 1994;107:10871093. 25. Campione A, Ligabue T, Luzzi L, et al. Comparison between segmentectomy and larger resection of stage IA non–small cell lung carcinoma. J Cardiovasc Surg (Torino). 2004;45:67-70. 26. Tsubota N, Ayabe K, Doi O, et al. Ongoing prospective study of segmentectomy for small lung tumors. Ann Thorac Surg. 1998;66:17871790. 27. Okada M, Yoshikawa K, Hatta T, et al. Is segmentectomy with lymph node assessment an alternative to lobectomy for non–small cell lung cancer of 2 cm or smaller? Ann Thorac Surg. 2001;71:956-960. 28. Martin-Ucar AE, Nakas A, Pilling JE, et al. A case-matched study of anatomical segmentectomy versus lobectomy for stage I lung cancer in high-risk patients. Eur J Cardiothorac Surg. 2005;27:675-679.

EXTENT OF PARENCHYMAL PULMONARY RESECTION FOR LUNG CANCER

29. Okada M, Koike T, Higashiyama M, et al. Radical sublobar resection for small-sized non–small cell lung cancer: a multicenter study. J Thorac Cardiovasc Surg. 2006;132:769-775. 30. El-Sherif A, Gooding WE, Santos R, et al. Outcomes of sublobar resection versus lobectomy for stage I non–small cell lung cancer: a 13-year analysis. Ann Thorac Surg. 2006;82:408-415. 31. Yendamuri S, Sharma R, Demmy M, et al. Temporal trends in outcomes following sublobar and lobar resections for small (≤ 2 cm) non–small cell lung cancers: a Surveillance Epidemiology End Results database analysis. J Surg Res. 2013;183:27-32. 32. Cao C, Chandrakumar D, Gupta S, et al. Could less be more? A systematic review and meta-analysis of sublobar resections versus

lobectomy for non–small cell lung cancer according to patient selection. Lung Cancer. 2015;89:121-132. 33. Altorki NK, Kamel MK, Narula N, et al. Anatomical segmentectomy and wedge resections are associated with comparable outcomes for patients with small cT1N0 non–small cell lung cancer. J Thorac Oncol. 2016;11:1984-1992. 34. Sakurai H, Maeshima A, Watanabe S, et al. Grade of stromal invasion in small adenocarcinoma of the lung: histopathological minimal invasion and prognosis. Am J Surg Pathol. 2004;28:198-206. 35. Nakamura K, Saji H, Nakajima R, et al. A phase III randomized trial of lobec tomy versus limited resection for small-sized peripheral non–small cell lung cancer (JCOG0802/WJOG4607L). Jpn J Clin Oncol. 2010;40:271-274.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 433

GYNECOLOGIC CANCER

ENDOMETRIAL CANCER: IS THIS A NEW DISEASE

Endometrial Cancer: Is This a New Disease? Kathleen Moore, MD, and Molly A. Brewer, DVM, MD, MS OVERVIEW The incidence of endometrial cancer is increasing, and the age of onset is younger than in prior years. Although endometrial cancer still occurs more commonly in older women, for whom the mortality rate is increasing, it also is being diagnosed in younger and younger women. The underlying cause of the increase in incidence is the epidemic of obesity and the resulting hyperinsulinemia. Conservative treatment may be indicated for younger women who wish to retain their fertility. Lifestyle modifications such as diet and exercise can modulate the risk of developing endometrial cancer as well as prevent recurrence and other comorbidities associated with obesity.

n 2012, 527,600 women worldwide were diagnosed with endometrial cancer.1 The mortality rate was 1.7 to 2.4 per 100,000 women. In 2017, there will be an estimated 61,380 cases of endometrial cancer diagnosed in the United States and more than 10,920 deaths.2 Historically, the age of onset was typically in postmenopausal women, with a strong association with obesity. As compared with just 3 years ago, this is almost 10,000 more cases and over 2,000 more deaths as a result of endometrial cancer. However, in the past 10 years, the incidence of endometrial cancer in young women has increased dramatically as a result of earlier-onset obesity. It is the fourth most common cancer for women in the United States.

ENDOMETRIAL CANCER

Endometrial Cancer in the Older Population

The ideal management of endometrial cancer in an older population, typically defined as older than age 65, is largely unknown, despite the fact that this population is the most rapidly growing age group in the United States. Endometrial cancer, which is predominantly a disease of postmenopausal women, is expected to increase in prevalence with an increasingly older population. Whereas 25% of cases will be diagnosed in patients over age 70, 50% of deaths from endometrial cancer will occur in this same older age group. Despite the increased rates of endometrial cancer mortality seen in elderly patients, studies show these patients receive less aggressive therapy than their younger counterparts, which is presumed to be due, in part, to the physician bias that older patients cannot tolerate aggressive therapy. Prior literature supports this, showing that advanced age is an independent risk factor for perioperative morbidity, even when controlling for medical comorbidities.3 A Surveillance,

Epidemiology, and End Results (SEER) program analysis of over 37,000 women with endometrial cancer found age-related changes in management. For example, whereas 95% of women younger than age 70 underwent a hysterectomy, only 67% of women older than age 80 underwent a hysterectomy. Similar findings were reported for lymphadenectomy, with only 25% of women older than age 80 as compared with almost 50% of women age 60 to 69 undergoing full staging.4 A multi-institutional study from France found similar trends, with declining rates of pelvic lymphadenectomy with age (85% vs. 46% in patients younger than age 65 years and older than age 80, respectively).5 When considering the optimal treatment approach for the older patient with endometrial cancer, there are several factors in play. One factor to consider is the mortality risk a particular cancer has for a particular patient. In other words, what type of endometrial cancer do older patients develop? What is their risk for extra-uterine spread? Can we use this information to direct the planned surgical intervention? Although not therapeutic, lymphadenectomy can direct subsequent therapy depending on findings. Identification of those patients at highest risk for extra-uterine spread of disease allows us to potentially triage our surgical approach based on these risks. Another factor to consider is the risk of surgical intervention(s) on the older patient. The intersection of these two factors can help define whether a surgical intervention is offered and, if offered, what type of surgical intervention that might be.

Disease Risk

Increasing age plays a strong role in predicting recurrence in endometrial cancer. Gynecologic Oncology Group (GOG)

From the Stephenson Cancer Center, Section of Gynecologic Oncology, Department of Obstetrics and Gynecology, University of Oklahoma Health Sciences Center, Oklahoma City, OK; Department of Obstetrics and Gynecology, Division of Gynecologic Oncology, Carole and Ray Neag Comprehensive Cancer Center, University of Connecticut Health Center, Farmington, CT. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Molly A. Brewer, DVM, MD, MS, Carole and Ray Neag Comprehensive Cancer Center, University of Connecticut Health Center, 263 Farmington Ave., MC 2947, Farmington, CT 06030; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 435

MOORE AND BREWER

FIGURE 1. Recurrence-Free Survival Stratified by Number of Uterine Risk Factors and Age From Gynecologic Oncology Protocol 2499

Study 99 was designed to evaluate surgery alone, including lymphadenectomy, versus surgery and adjuvant pelvic radiotherapy in patients with intermediate-risk, stage I, and occult stage II endometrioid endometrial adenocarcinoma. A subset analysis identified a group with a 25% risk of recurrence and labeled as high intermediate risk (H-IR). H-IR patients included those older than age 70 plus one uterine factor, those age 50–70 plus two factors, or any age plus three factors. Uterine risk factors included grade 2 or 3, presence of lymph vascular space invasion, and depth of invasion to the outer one-third of the myometrium.6 Other similar studies such as the first Post-Operative Radiotherapy in Endometrial Cancer trial also found age was associated with increased disease recurrence.7 An ancillary data analysis of the GOG LAP2 study, which compared laparoscopic versus open hysterectomy and lymphadenectomy in clinical stage I patients, evaluated the pathologic findings and disease-related outcomes by increasing age. They reported that, of the entire LAP2 population, 37% met H-IR criteria, and 43% of those were older than age 70. Only 23% of the entire LAP2 population was older than age 70, but 55% of those patients met H-IR criteria. When looking at rates of adjuvant treatment across age groups in this H-IR subset, as age increased, significantly less radiotherapy was administered, with 60%

KEY POINTS • Obesity with a BMI greater than 30 is responsible for up to 81% of the endometrial cancer diagnosed. • Older women experience increased mortality rates from both their cancer and the surgery, and risk needs to be closely balanced with benefit. • Younger women also have an increase in endometrial cancer and its precursors and can be successfully treated conservatively to maintain fertility. • Changing one’s lifestyle, although critical, is not often successful, which may be, in part, a result of a lack of counseling from physicians as well as a long history of an unhealthy lifestyle. 436 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

of patients younger than age 50 receiving radiotherapy versus 24%–27% in patients age 50–79 and only 17.3% in patients age 80 or older (p = .011). Despite these differences, receipt of adjuvant radiotherapy did not impact progression-free survival (PFS) or overall survival (OS) in this subset, and the only factors that were significant for OS were grade 3 disease (p = .016; hazard ratio [HR], 1.99), nonwhite race (p = .043; HR, 1.62), age (p < .001; HR, 1.05), and body mass index (BMI; p = .008; HR, 1.03). Although age remained important in models of overall survival, it is interesting to note that of the 370 patients older than age 70 who met H-IR criteria, the majority did so by age and just one uterine risks factor, usually grade 2 or 3 disease. The recurrence risk in this group was only 3.3%, as compared with those with all three uterine factors at 20.9%.7,8 The idea that the GOG H-IR algorithm identifies many lowrisk patients was reinforced by subset analysis of GOG 249. Here, the true discriminator of risk appears to be presence of all three uterine risk factors (Fig. 1). How increasing age modifies this risk is unknown, but these appear to be the highest-risk patients who may require adjuvant therapy to improve outcomes.

Surgical Risk

Options for primary treatment include hysterectomy alone (including vagin*l hysterectomy with regional anesthesia), hysterectomy with lymph node staging, or no hysterectomy at all and treatment with primary radiation. The decision whether to plan a lymphadenectomy is a controversial topic in all endometrial cancer cases, much less in cases with older patients. From a pure morbidity standpoint, we do have data on perioperative and postoperative complications in older patients undergoing a hysterectomy with and without planned lymphadenectomy as compared with either approach in younger patients. Minimally invasive surgical management of many types of cancers is commonly used, and there is a large amount of data showing similar oncologic outcomes and decreased morbidity with minimally invasive techniques versus laparotomy.10,11 Several small

ENDOMETRIAL CANCER: IS THIS A NEW DISEASE

FIGURE 2. Description of a Linear Model With Outcome of Maximum Toxicity Including Explanatory Variables

retrospective studies show decreased morbidity in older patients treated with minimally invasive techniques, but to date, we have no prospective data comparing outcomes in the elderly.12,13 GOG protocol GOG 2222 or LAP2 was a randomized, prospective clinical trial to compare comprehensive surgical staging by laparotomy versus laparoscopy for the treatment of patients with stage I–IIA endometrial cancer (2,616 patients). LAP2 is the largest prospective trial to date looking at minimally invasive surgical approaches in clinically earlystage endometrial cancer. This study includes 1,477 patients age 60 or older, 762 of whom are age 60–69 and 715 of whom are age 70 or older. An ancillary data analysis was performed to look at both the operative risks of hysterectomy and lymphadenectomy in an older population and to assess histologic, tumor-related risk, and disease-specific outcome. There was noted to be a much higher rate of complications in the laparotomy group, and this difference became larger with increasing age starting at age 60. Patients younger than age 50 had the same rates of postoperative complications (laparotomy, 15.9% vs. laparoscopy, 15.7%), whereas patients age 80 or older had very different rates of complications (laparotomy, 38.9% vs. laparoscopy, 19.8%).14 Comparison of complications during and following laparotomy versus laparoscopy by age and controlling for race, BMI, stage, and grade found higher rates of postoperative antibiotics (odds ratio [OR], 1.63) and hospital stay longer than 2 days (OR, 12.77), as compared with patients younger than 60. Patients age 60 or older in the laparotomy group had higher rates of readmission (OR, 1.52), ileus (OR, 2.16), pneumonia (OR, 2.36), deep vein thrombosis (OR, 2.87), and arrhythmia (OR, 3.21).14

Figure 2 describes a linear model with outcome of maximum toxicity, including explanatory variables such as age, race, and BMI. The change in maximum toxicity before approximately age 60 is not significant, but after age 60, the toxicity appears to increase sharply, with interaction between the surgical approach and age having a moderate effect (p = .035). As age increases, the benefit from laparoscopy appears to increase as well, and according to this model, the benefit occurs beginning at age 60,14 demonstrating that for older patients who are fit enough to be considered for hysterectomy and lymphadenectomy, a minimally invasive approach is favored. There has been a shift from laparoscopic to a robotic approach, and with this shift comes potentially important changes for the older population. Robotic surgery requires a steep Trendelenberg position and potentially longer operative times, which can place patients at risk for ischemic optic neuropathy. Risk factors for this include hypertension, diabetes, known cardiovascular disease, and narrow-angle glaucoma all of which are more common with age.15 Despite the potential complications noted above, the data surrounding use of robotic hysterectomy with or without lymphadenectomy are similar to the laparoscopic data with less complications and acceptable surgical outcomes. Krause et al reported on 705 patients, 50 of whom were older than age 75 and underwent robotic gynecologic procedures, and found that, other than increased length of stay longer than 1 day (30.0% vs. 15%; p < .01) and higher incidence of postoperative arrhythmia (8.0% vs. 1.2%; p < .01), there were no other differences in the relatively low rate of intraoperative and postoperative complications. Of note, the arrhythmias did not result in cardioversion for any patient in the elderly group and only one patient in the younger group.16 Similarly, asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 437

MOORE AND BREWER

Backes et al compared open versus robotic hysterectomy with or without lymphadectomy and reinforced the safety of robotic surgery in an older population (here defined as age 70 or older). As compared with laparotomy, they report a higher incidence of pelvic lymphadenectomy completion, decreased blood loss and need for transfusion, shorter hospital stay, and no increase in intraoperative complications.17 No incidence of ischemic optic neuropathy has been reported to date in these series of older patients. Data also show increased risk of complications by age for patients who undergo an exploratory laparotomy for hysterectomy with or without a lymphadenectomy. A SEER analysis looked at over 25,000 women age 65 or older who underwent hysterectomy for endometrial cancer. Compared with women age 65–69, women age 85 or older were more likely to have perioperative complications (12% vs. 17%), postoperative medical complications (24% vs. 34%), a longer hospital stay (3 vs. 5 days), and require more blood transfusions (6% vs. 10%). Perioperative mortality rates were significantly higher in patients age 85 or older compared with those age 65–69 (1.6% vs. 0.4%). These results were the same when controlling for medical comorbidities.3 These data on postoperative complications are important to consider when selecting surgical procedures and optimal approaches in the older patient, as surgical complications can impact overall survival. Certain complications that occur within 30 days of surgery were more important than preoperative patient risk and intraoperative factors in determining survival after major surgery.18

Endometrial Cancer in the Premenopausal Population

The standard treatment of endometrial cancer is a hysterectomy or a bilateral salpingo-oophorectomy (BSO) with or without staging. However, with the increase in incidence in obesity that is starting at younger ages, the incidence of endometrial cancer is increasing in the premenopausal population. The worldwide prevalence of obesity more than doubled between 1980 and 2014. In 2014, more than 1.9 billion adults age 18 or older were overweight. Of these, more than 600 million adults were obese. About 13% of the world’s adult population (11% of men and 15% of women) were obese in 2014, and 39% of adults age 18 or older (38% of men and 40% of women) were overweight.19 Obesity is the single biggest risk factor for endometrial cancer. The major precursor lesion for endometrial cancer is atypical endometrial hyperplasia (AEH). It is estimated that up to 50% of women with an endometrial biopsy with AEH will have a concomitant adenocarcinoma or will develop endometrial cancer. If there are not atypical nuclear changes, the risk of developing endometrial cancer is much lower. Many of the studies have focused on treating precursor lesions, including hyperplasia without atypia, despite a low risk for progression to cancer. Treatment of AEH or endometrial cancer in premenopausal women who wish to retain their fertility has centered around progesterone treatment to offset the unopposed estrogen created by the aromatase 438 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

conversion of androgen to estrogen. Many studies have used oral progestins, either medroxyprogesterone acetate or megestrol acetate, a potent progestin. Recently, there has been an interest in the use of the levonorgestrel intrauterine device (IUD), which secretes a continuous amount of progestin and avoids the side effects of oral progestin such as increased appetite, depression, and increased venous thromboembolism risk. There is additional literature in the last several years of combining the levonorgestrel IUD with a gonadotropin-releasing hormone (GnRh) agonist to induce menopause. In addition, once menopause has been created, the use of aromatase inhibitors (AIs) is being used to offset the effect of obesity on the aromatase system.

Oral Progestin

There are numerous studies that report on the response to oral progestin. Chen et al reported that of 53 patients, 39 (74%) achieved a complete response (CR) after a median period of 6 (3–24) months on either medroxyprogesterone acetate or megestrol acetate. A CR was less frequent among obese than nonobese patients (four of 12 [33%] vs. 35 of 41 [85%]; p = .001), which suggested that obesity may reduce the risk of responding to conservative treatment. Twenty-six percent of the women with a CR recurred. Fifty-two percent of those women wishing to conceive became pregnant.20 Simpson et al described 45 patients with AEH (19) and endometrial cancer (25) treated with oral progestin. Fifty-five percent achieved a CR, but 54% of those that responded recurred within 3 years. Five of the 53 patients achieved a pregnancy, and two delivered a live infant.21 Ramirez et al reported on 81 patients with endometrial cancer treated with oral progestin: 76% responded to treatment, and 24% of those who responded recurred within 19 months. Twenty of the 62 patients who responded became pregnant.22 Wang et al described six patients with endometrial cancer who underwent hysteroscopic resection of local lesions combined with oral administration of megestrol acetate for 3 to 6 months. They all had a CR, and half of them became pregnant by natural means (without assisted reproductive technology) and delivered healthy infants.23

Levonorgestrel IUD

Over the last 10 years, with the development of the levonorgestrel IUD, there have been several studies using the IUD in lieu of oral progestin. Scarselli et al found a 94% regression rate in 34 patients, but only 11% had AEH, and none had endometrial cancer.24 Varma et al evaluated 105 women and found regression in 67% of women with AEH and a 90% overall regression rate.25 Wildemeersch followed 20 patients long term with an IUD. Forty percent had AEH, none had cancer, and all women except one developed normal endometrium. One woman with AEH developed simple hyperplasia.26 Gallos et al, in a meta-analysis, compared the response rate of oral progestin to the levonorgestrol IUD and found that the response rate of the oral progestin was lower than the IUD for complex hyperplasia (69% vs. 90%) and for AEH (60% vs. 90%).27 Given the reasonable response

ENDOMETRIAL CANCER: IS THIS A NEW DISEASE

rate to the IUD, Pronin et al analyzed 70 patients with either AEH or grade 1 endometrial cancer using the IUD with a GnRH agonist for at least 6 months and found a 72% CR in the patients with endometrial cancer and 92% CR in the AEH, suggesting that adding the GnRH agonist significantly improved the relative risk (RR).28 Minig et al also combined the IUD with a GnRH agonist and found a CR in 95% of the patients with AEH and 57% CR in women with endometrial cancer.29 A large U.K. meta-analysis of 408 women with endometrial cancer and 151 with AEH had a live birth rate of 28% in the endometrial cancer group and 26% live birth rate in the AEH group. Treatments included oral progestin and an IUD with or without GnRH agonists.27 One caveat noted was the risk of conservative treatment. In one case report, a borderline ovarian cancer was found in a young woman. Review of the literature suggested that 5% of women with endometrial cancer had a synchronous ovarian cancer, so careful evaluation of the ovaries is important prior to conservative treatment of endometrial cancer.30 There is increasing interest in combining the progesterone releasing IUD with a GnRH agonist and AIs. Studies in patients with breast cancer comparing the use of tamoxifen with AIs found a 48% lower rate of endometrial cancer in the patients taking the AI, suggesting there may be a role for AIs in the conservative treatment of AEH or endometrial cancer, particularly in women that are obese.31 In conclusion, there is emerging evidence that treating young women with AEH or endometrial cancer with the combination of a levonorgestrel IUD and GnRH agonists may produce an excellent regression rate and a reasonable pregnancy rate. Addition of an AI in obese women may also have a positive impact. More evidence is needed before we have a clear picture of the best treatment of endometrial cancer or AEH in young women.

Lifestyle and Endometrial Cancer

Endometrial cancer has a strong correlation with lifestyle. Obesity with a BMI greater than 30 is responsible for up to 81% of the endometrial cancer diagnosed.32 There is emerging literature in breast and colon cancer suggesting that obese/inactive patients have a higher mortality than those who are thinner and more physically active. Obesity may affect tumorigenesis and tumor progression through insulin resistance and hyperinsulinemia, increased bioavailability of steroid hormones, and localized inflammation.33 Other comorbidities, in particular, cardiovascular disease, which is associated with obesity, are the cause of mortality in women with endometrial cancer at 10 years post-diagnosis, not endometrial cancer, suggesting that the obesity is associated with not only an increased risk of death from endometrial cancer, but also other comorbidities.34 A study by Arem et al35 found that patients with BMIs of 25–29 had an HR of death of 1.74, compared with 1.84 and 2.39 for patients with BMIs of 30–35 and 35 and higher, respectively. Regular exercise (more than 7 hours/week) was associated with an HR of death of 0.57. In adjusted models, a woman who did moderate to vigorous physical

activity and who reported more than 7 hours of moderate to vigorous physical activity per week prior to a diagnosis of endometrial cancer had a 43% lower risk of 5-year, allcause mortality when compared with women who never/ rarely exercised. The same study found more than a twofold increased mortality risk among class II obese women (BMI > 35) when compared with women with BMIs lower than 35. BMI, physical activity, and health status all were associated with an increased risk of death from endometrial cancer (HR, 2.28) and were highly correlated.36 Another study by Modesitt et al37 evaluated physical fitness in obese women with and without endometrial cancer and found the patients with endometrial cancer had significantly poorer physical fitness and higher glucose levels when compared with those without endometrial cancer, even though the women had equivalent BMIs. In the EPIC trial, a large cohort study in Europe to investigate the association between lifestyle and cancer risk, Dossus et al conducted a nested case control study of patients with endometrial cancer and evaluated four categories of characteristics to determine risk of endometrial cancer.38 They found that insulin resistance/metabolic syndrome had an OR of 2.5 for developing endometrial cancer, the steroid factor (estradiol, dehydroepiandrosterone, androstenedione, estrone, and testosterone) had an OR of 1.68 of developing endometrial cancer, and the inflammation factor (cytokine; especially IL-6, TNF receptor, and C peptide) had an OR of 1.62 of developing endometrial cancer. Correlations with BMI and waist circumference were strongest for Creactive protein, IL-6, IL-1 receptor antagonist, triglycerides, C peptide, and estradiol. All of these factors were statistically significantly higher in women who developed endometrial cancer compared with the controls. Although there was a significantly higher BMI and waist circumference in the patients with endometrial cancer compared with the controls, the mean BMI in the control group was 26.4 compared with 28.1, which was not enough of a difference for BMI to completely explain the increase in endometrial cancer risk.38 A small study investigated the impact of exercise and diet change on endometrial cancer survivors39 and found that 84% of patients adhered to the intervention and had weight loss of 4.4 kg at 6 months and 4.0 kg at 1 year. The patients on the intervention started with a mean weight of 95.7 kg and had a mean weight of 92.7 kg at 1 year, whereas the control patients had a baseline weight of 94 kg and at 1 year were 95.4 kg. Those patients who did not participate in an intervention gained rather than lost weight. The authors concluded that patients with endometrial cancer are difficult to motivate to change their lifestyle, and only 30% of participants succeeded in the weight reduction goal. A second study by this group showed that patients with endometrial cancer had overall unhealthy lifestyles with poor eating habits and minimal physical activity as well as multiple comorbidities, and 93% had abdominal obesity.40 A meta-analysis evaluating insulin resistance found that both fasting insulin levels and nonfasting C peptide were significantly higher in patients with endometrial cancer.41 A Danish asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 439

MOORE AND BREWER

study evaluated factors associated with mortality from endometrial cancer and found that education status, but not BMI, was associated with an increase in mortality from endometrial cancer. However, the majority of the patients had a BMI of less than 35, which is a different population that the current U.S. population.42 Insulin resistance plays an important role in endometrial cancer. High serum levels of insulin are associated with an increased risk of endometrial cancer, particularly in overweight/obese women, who also have an increase in estrogen activity.43 Type 2 diabetes (noninsulin dependent) results in increased insulin levels for long periods both before and after the disease onset and is associated with an increased risk of atypical hyperplasia and endometrial cancer, independent of obesity.44 Insulin reduces the liver production of sex hormone–binding globulin (SHBG), and chronically high insulin because insulin resistance is associated with high serum levels of testosterone. Insulin stimulates the ovarian and adrenal cortex production of androgens (especially androstenedione and testosterone) through the 17α-hydroxylase and 17,20-lyase activities, which are subsequently metabolized into estrogen from the aromatase enzyme in adipose tissue. Insulin also has direct proliferative effects on the endometrium, working as a growth factor, similar to insulin-like growth factor 1 as well. Epidemiologic data on postmenopausal women suggest an increased risk of endometrial cancer in nondiabetic women with hyperinsulinemia, in diabetic women with insulin resistance, and in women with metabolic syndrome.45 Studies have shown a direct link between estrogen receptors and cell surface receptors such as insulin-like growth factor 1 receptor and EGFR, which cause the activation of kinase cascade pathways, including PI3K/AKT/mTOR and MAPK, which are directly associated with cell proliferation.46 In addition, the loss of tumor suppressor PTEN has been found in 40%–80% of type 1 endometrial cancers.47 Increased insulin-like growth factor 1 levels in addition to loss of PTEN leads to the increased activation of the kinase signaling cascades. Adiponectin and insulin growth factor– binding protein help to regulate glucose levels and insulin sensitivity and therefore serve as protective factors against endometrial cancer development. Reduced levels of these protective molecules have been found in individuals with obesity and hyperinsulinemia. The most important lifestyle change that will help correct the underlying hyperinsulin state and obesity is aerobic exercise. Limiting energy-dense foods such as carbohydrates is also an important aspect of improving the underlying metabolic abnormalities that promote endometrial pathology as well as a host of other diseases. Independent of its influence on BMI,48 physical activity improves glucose uptake by skeletal muscles, which reduces insulin resistance and insulin levels as well as estrogen and androgen levels. The use of metformin can reduce the risk of cancer in women with diabetes or nondiabetic hyperinsulinemia. It can also be used in patients with polycystic ovarian syndrome, for whom it reduces the insulin level as well as the androgen level.49 440 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Metformin also reduces the proliferation of endometrial cells and may be an important aspect of treating and preventing endometrial cancer in addition to lifestyle changes. The metabolic issues associated with obesity and hyperinsulinemia are associated with other causes of mortality. A 2012 study focusing on cardiovascular mortality used SEER registries to show that patients with endometrial cancer were more likely to die of cardiovascular disease (35.9%), followed by other causes, including other malignancies, than they were to die of their endometrial cancer.50 In a study by Clark et al, only 29% of patients reported ever being told of the relationship between obesity and endometrial cancer. Despite this, over 50% of the patients reported attempting to lose weight through lifestyle changes after diagnosis. Those who were most likely to make lifestyle modifications were those who had received adequate counseling by a physician.51 Thus the primary care physician, the obstetrician/gynecologist, the gynecologic oncologist, and the bariatric surgeons are extremely important factors for changing the lifestyle of these obese, hyperinsulinemic patients with such a high risk of endometrial cancer and other diseases.52 Moore et al published a meta-analysis in 2010 concluding that increased physical activity and decreased sedentary time were associated with a decreased risk of endometrial cancer. Women who were inactive and who sat for 9 hours per day had twice the risk of endometrial cancer compared with active women who sat fewer than 3 hours per day (RR 2.14; 95% CI, 1.48–3.10). However, if women exercised and sat for 9 or more hours per day, their reduced risk of endometrial cancer was attenuated, suggesting that prolonged inactivity is an independent risk factor for endometrial cancer.53 There is compelling evidence that lifestyle changes are important for both prevention and treatment of endometrial cancer. Women who have already developed endometrial cancer would benefit from structured exercise interventions, as this would decrease their obesity and insulin resistance and decrease their death from other diseases such as cardiovascular disease. In addition, their mortality from endometrial cancer would be expected to decrease. Given the obesity epidemic and the general decrease in physical activity, interventions are needed prior to the development of endometrial cancer and should be a collaboration between the pediatricians, the primary care physicians, and, once endometrial cancer develops, the gynecologic oncologist. Without these lifestyle interventions, the incidence and mortality from endometrial cancer will continue to increase along with other obesity-associated diseases. Treatment of insulin resistance is an important aspect of preventing endometrial cancer and improving outcomes and consists of exercise, dietary changes, and the use of metformin.

CONCLUSION

The incidence of and mortality from endometrial cancer is increasing primarily because of the increased incidence of obesity and the resulting hyperinsulinemia. Older women have an increase in mortality from both their cancer and the surgery, and risk needs to be closely balanced with benefit.

ENDOMETRIAL CANCER: IS THIS A NEW DISEASE

Minimally invasive surgery carries a lower risk of surgical complications, particularly in the older age group. Younger women are also experiencing an increase in rates of endometrial cancer and its precursors and can be successfully treated conservatively to maintain fertility. Changes in lifestyle are critical to managing this increase in risk and mor-

tality. Weight loss and exercise are key to decreasing the hyperinsulinemia that drives the development of endometrial cancer. Changing one’s lifestyle, although critical, is not often successful, which may be, in part, a result of a lack of counseling from physicians as well as a long history of an unhealthy lifestyle.

References 1. Torre LA, Bray F, Siegel RL, et al. Global cancer statistics, 2012. CA Cancer J Clin. 2015;65:87-108.

colorectal cancer patients. Surg Laparosc Endosc Percutan Tech. 2013;23:532-535.

2. American Cancer Society. Cancer Facts & Figures 2017. https://www. cancer.org/research/cancer-facts-statistics/allcancer-facts-figures/ cancer-facts-figures-2017.html. Accessed March 21, 2017.

14. Bishop EA, Java J, Moore KN, et al. Surgical outcomes among a geriatric population of endometrial cancer patients: An NRG/GOG ancillary data analysis of GOG LAP2. Int J Gynecol Cancer. In press.

3. Wright JD, Lewin SN, Barrena Medel NI, et al. Morbidity and mortality of surgery for endometrial cancer in the oldest old. Am J Obstet Gynecol. 2011;205:66.e1-66.e8.

15. Dunker S, Hsu HY, Sebag J, et al. Perioperative risk factors for posterior ischemic optic neuropathy. J Am Coll Surg. 2002;194:705-710.

4. Wright JD, Lewin SN, Barrena Medel NI, et al. Endometrial cancer in the oldest old: tumor characteristics, patterns of care, and outcome. Gynecol Oncol. 2011;122:69-74. 5. Poupon C, Bendifallah S, Ouldamer L, et al. Management and survival of elderly and very elderly patients with endometrial cancer: an agestratified study of 1228 women from the FRANCOGYN group. Ann Surg Oncol. Epub 2016 Dec 22. 6. Keys HM, Roberts JA, Brunetto VL, et al; Gynecologic Oncology Group. A phase III trial of surgery with or without adjunctive external pelvic radiation therapy in intermediate risk endometrial adenocarcinoma: a Gynecologic Oncology Group study. Gynecol Oncol. 2004;92:744-751. Creutzberg CL, van Putten WL, Koper PC, et al. Surgery and 7. postoperative radiotherapy versus surgery alone for patients with stage-1 endometrial carcinoma: multicentre randomised trial. PORTEC Study Group. Post Operative Radiation Therapy in Endometrial Carcinoma. Lancet. 2000;355:1404-1411. 8. Bishop EA, Java J, Moore KN, et al. Pathologic and treatment outcomes among a geriatric population of endometrial cancer patients: a Gynecologic Oncology Group ancillary data study. Gynecol Oncol. 2014;S24 (suppl; abstr 57). McMeekin DS, Filiaci G, Aghajanian C, et al. GOG 249: a randomized 9. phase III trial of pelvic radiation therapy versus vagin*l cuff brachytherapy followed by pacl*taxel/carboplatin chemotherapy in patients with high-risk, early stage endometrial cancer: a Gynecologic Oncology Group trial. Presented at: Society of Gynecologic Oncology Annual Meeting. Tampa Bay, FL; 2014. 10. Veldkamp R, Kuhry E, Hop WC, et al; Colon cancer Laparoscopic or Open Resection Study Group (COLOR). Laparoscopic surgery versus open surgery for colon cancer: short-term outcomes of a randomised trial. Lancet Oncol. 2005;6:477-484. Smith JA Jr, Chan RC, Chang SS, et al. A comparison of the incidence 11. and location of positive surgical radical prostatectomy. J Urol. 2007;178:2385-2390. Bogani G, Cromi A, Uccella S, et al. Perioperative and long-term 12. outcomes of laparoscopic, open abdominal, and vagin*l surgery for endometrial cancer in patients aged 80 years or older. Int J Gynecol Cancer. 2014;24:894-900. 13. Hatakeyama T, Nakanishi M, Murayama Y, et al. Laparoscopic resection for colorectal cancer improves short-term outcomes in very elderly

16. Krause AK, Muntz HG, McGonigle KF. Robotic-assisted gynecologic surgery and perioperative morbidity in elderly women. J Minim Invasive Gynecol. 2016;23:949-953. 17. Backes FJ, ElNaggar AC, Farrell MR, et al. Perioperative outcomes for laparotomy compared to robotic surgical staging of endometrial cancer in the elderly: a retrospective cohort. Int J Gynecol Cancer. 2016;26:1717-1721. 18. Khuri SF, Henderson WG, DePalma RG, et al; Participants in the VA National Surgical Quality Improvement Program. Determinants of long-term survival after major surgery and the adverse effect of postoperative complications. Ann Surg. 2005;242:326-343. World Health Organization. Obesity and Overweight. www. 19. who.int/mediacentre/factsheets/fs311/en/. Accessed February 8, 2017. 20. Chen M, Jin Y, Li Y, et al. Oncologic and reproductive outcomes after fertility-sparing management with oral progestin for women with complex endometrial hyperplasia and endometrial cancer. Int J Gynaecol Obstet. 2016;132:34-38. 21. Simpson AN, Feigenberg T, Clarke BA, et al. Fertility sparing treatment of complex atypical hyperplasia and low grade endometrial cancer using oral progestin. Gynecol Oncol. 2014;133:229-233. 22. Ramirez PT, Frumovitz M, Bodurka DC, et al. Hormonal therapy for the management of grade 1 endometrial adenocarcinoma: a literature review. Gynecol Oncol. 2004;95:133-138. 23. Wang Q, Guo Q, Gao S, et al. Fertility-conservation combined therapy with hysteroscopic resection and oral progesterone for local early stage endometrial carcinoma in young women. Int J Clin Exp Med. 2015;8:13804-13810. 24. Scarselli G, Bargelli G, Taddei GL, et al. Levonorgestrel-releasing intrauterine system (LNG-IUS) as an effective treatment option for endometrial hyperplasia: a 15-year follow-up study. Fertil Steril. 2011;95:420-422. 25. Varma R, Soneja H, Bhatia K, et al. The effectiveness of a levonorgestrelreleasing intrauterine system (LNG-IUS) in the treatment of endometrial hyperplasia--a long-term follow-up study. Eur J Obstet Gynecol Reprod Biol. 2008;139:169-175. 26. Wildemeersch D, Janssens D, Pylyser K, et al. Management of patients with non-atypical and atypical endometrial hyperplasia with a levonorgestrel-releasing intrauterine system: long-term follow-up. Maturitas. 2007;57:210-213.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 441

MOORE AND BREWER

27. Gallos ID, Yap J, Rajkhowa M, et al. Regression, relapse, and live birth rates with fertility-sparing therapy for endometrial cancer and atypical complex endometrial hyperplasia: a systematic review and metaanalysis. Am J Obstet Gynecol. 2012;207:266.e1-266.12. 28. Pronin SM, Novikova OV, Andreeva JY, et al. Fertility-sparing treatment of early endometrial cancer and complex atypical hyperplasia in young women of childbearing potential. Int J Gynecol Cancer. 2015;25:10101014. 29. Minig L, Franchi D, Boveri S, et al. Progestin intrauterine device and GnRH analogue for uterus-sparing treatment of endometrial precancers and well-differentiated early endometrial carcinoma in young women. Ann Oncol. 2011;22:643-649. 30. Shamshirsaz, AA, Withiam-Leitch, M, Odunsi, K, et al. Young patients with endometrial carcinoma selected for conservative treatment: A need for vigilance for synchronous ovarian carcinomas, case report and literature review. Gynecol Oncol. 2007;104:757-760. 31. Chlebowski RT, Schottinger JE, Shi J, et al. Aromatase inhibitors, tamoxifen, and endometrial cancer in breast cancer survivors. Cancer. 2015;121:2147-2155. 32. Nevadunsky NSVAA, Van Arsdale A, Strickler HD, et al. Obesity and age at diagnosis of endometrial cancer. Obstet Gynecol. 2014;124:300-306. 33. Calle EE, Kaaks R. Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms. Nat Rev Cancer. 2004;4:579-591. 34. Ward KK, Shah NR, Saenz CC, et al. Cardiovascular disease is the leading cause of death among endometrial cancer patients. Gynecol Oncol. 2012;126:176-179. 35. Arem H, Park Y, Pelser C, et al. Prediagnosis body mass index, physical activity, and mortality in endometrial cancer patients. J Natl Cancer Inst. 2013;105:342-349. 36. Arem H, Pfeiffer RM, Moore SC, et al. Body mass index, physical activity, and television time in relation to mortality risk among endometrial cancer survivors in the NIH-AARP Diet and Health Study cohort. Cancer Causes Control. 2016;27:1403-1409.

40. von Gruenigen VE, Waggoner SE, Frasure HE, et al. Lifestyle challenges in endometrial cancer survivorship. Obstet Gynecol. 2011;117:93-100. 41. Hernandez AV, Pasupuleti V, Benites-Zapata VA, et al. Insulin resistance and endometrial cancer risk: a systematic review and meta-analysis. Eur J Cancer. 2015;51:2747-2758. 42. Seidelin UH, Ibfelt E, Andersen I, et al. Does stage of cancer, comorbidity or lifestyle factors explain educational differences in survival after endometrial cancer? A cohort study among Danish women diagnosed 2005-2009. Acta Oncol. 2016;55:680-685. 43. Gunter MJ, Hoover DR, Yu H, et al. A prospective evaluation of insulin and insulin-like growth factor-I as risk factors for endometrial cancer. Cancer Epidemiol Biomarkers Prev. 2008;17:921-929. 44. Barone BB, Yeh HC, Snyder CF, et al. Long-term all-cause mortality in cancer patients with preexisting diabetes mellitus: a systematic review and meta-analysis. JAMA. 2008;300:2754-2764. 45. Kaaks R, Lukanova A, Kurzer MS. Obesity, endogenous hormones, and endometrial cancer risk: a synthetic review. Cancer Epidemiol Biomarkers Prev. 2002;11:1531-1543. Cust AE, Kaaks R, Friedenreich C, et al. Metabolic syndrome, plasma 46. lipid, lipoprotein and glucose levels, and endometrial cancer risk in the European Prospective Investigation into Cancer and Nutrition (EPIC). Endocr Relat Cancer. 2007;14:755-767. Schmandt RE, Iglesias DA, Co NN, et al. Understanding obesity and 47. endometrial cancer risk: opportunities for prevention. Am J Obstet Gynecol. 2011;205:518-525. 48. Hawley JA, Lessard SJ. Exercise training-induced improvements in insulin action. Acta Physiol (Oxf). 2008;192:127-135. 49. Palomba S, Falbo A, Zullo F, et al. Evidence-based and potential benefits of metformin in the polycystic ovary syndrome: a comprehensive review. Endocr Rev. 2009;30:1-50. 50. Papatla K, Huang M, Slomovitz B. The obese endometrial cancer patient: how do we effectively improve morbidity and mortality in this patient population? Ann Oncol. 2016;27:1988-1994.

37. Modesitt SC, Geffel DL, Via J, et al. Morbidly obese women with and without endometrial cancer: are there differences in measured physical fitness, body composition, or hormones? Gynecol Oncol. 2012;124:431-436.

51. Clark LH, Ko EM, Kernodle A, et al. Endometrial cancer survivors’ perceptions of provider obesity counseling and attempted behavior change: are we seizing the moment? Int J Gynecol Cancer. 2016;26:318-324.

38. Dossus L, Lukanova A, Rinaldi S, et al. Hormonal, metabolic, and inflammatory profiles and endometrial cancer risk within the EPIC cohort--a factor analysis. Am J Epidemiol. 2013;177:787-799.

52. Koutoukidis DA, Beeken RJ, Lopes S, et al. Attitudes, challenges and needs about diet and physical activity in endometrial cancer survivors: a qualitative study. Eur J Cancer Care (Engl). Epub 2016 Jun 21.

39. von Gruenigen V, Frasure H, Kavanagh MB, et al. Survivors of uterine cancer empowered by exercise and healthy diet (SUCCEED): a randomized controlled trial. Gynecol Oncol. 2012;125:699-704.

53. Moore SC, Gierach GL, Schatzkin A, et al. Physical activity, sedentary behaviours, and the prevention of endometrial cancer. Br J Cancer. 2010;103:933-938.

442 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

WHENCE HIGH-GRADE SEROUS OVARIAN CANCER

Whence High-Grade Serous Ovarian Cancer Elise C. Kohn, MD, and S. Percy Ivy, MD OVERVIEW Our understanding of epithelial ovarian cancer has blossomed, and we now recognize that it is a collection of varied histologic and molecularly different malignancies, many of which may not derive from a true ovarian anatomic precursor. High-grade serous ovarian cancer (HGSOC) is a unique type of epithelial cancer. It is characterized by nearly universal mutation in and dysfunction of p53, genomic instability rather than driver mutations, advanced stage at onset, and probable fallopian tube epithelium origin, with a serous tubal in situ carcinoma precursor. Germline deleterious mutations in BRCA1 and BRCA2, as well as other less prevalent genes involved in DNA repair, such as PALB2 and RAD51c, are associated with its carcinogenesis and may predict susceptibility to classes of treatment agents, including DNA-damaging agents and DNA repair inhibitors. Loss of function of these genes is associated with hom*ologous recombination dysfunction (HRD). It is now recognized that there may be HGSOC with wild-type BRCA1 and BRCA2 with an identifiable HRD phenotype. Such HRD tumors also may be more susceptible to certain classes of treatments and may be phenotypically detectable with a composite molecular biomarker that has been shown to be predictive for response to PARP inhibitors. Use of this new knowledge of the anatomic and molecular background of HGSOC has led to the rational design of novel combinations of treatment classes to create an HRD-like cellular environment and thus drive treatment benefits.

he past 2 decades have brought about great progress and change in the field of ovarian cancer diagnosis, treatment, and research. “Ovarian cancer” has gone from singular to plural, and our diagnosis, treatment, and research have followed suit. Along with these changes have come new classifications, new drugs, and great opportunities to improve the quality and quantity of life for the women afflicted with this cancer. Keeping up with the frequent changes may have been daunting, although the scientific progress has also brought important answers that open viable directions to rethink screening and prevention and to target therapy more directly.

HIGH-GRADE SEROUS OVARIAN CANCER

What Is High-Grade Serous “Ovarian” Cancer?

The most common histology of malignancies of the ovary is now recognized to be an epithelial cancer emanating most commonly or most likely from the epithelium of the fimbria of the fallopian tube. This group of cancers was previously lumped together as high-grade epithelial ovarian cancer of serous or serous papillary type. An independent tumor of the fallopian tube(s) was not recognized, in part because the two organs are in such close proximity that their distinction was abrogated with tumor progression. The new World Health Organization histologic classification and grading system embraced the two-tiered grading system of low and

high grades in their revision in 2014.1 High-grade serous tumors are generally recognized by their lack of architecture and sheets of malignant cells, often enlarged and dysmorphic nuclei, and with further molecular characterization, nearly 100% TP53 mutation frequency. These can be ascertained with relative confidence by immunohistochemistry demonstrating overexpression of nuclear p53 staining or complete lack of such staining within the tumor, the latter being the loss-of-function p53 mutations. The World Health Organization classification recognizes the likely precursor lesion to be serous tubal in situ carcinoma lesions,2,3 from which progression to invasive carcinoma may be found, albeit generally in small lesions. The outward-facing exposure of the tubal (and ovarian) epithelium supports early shedding and implantation. The lack of an anatomic barrier between the pelvis and the abdomen, coupled with the permissive environment of the omentum, buoys local colonization and further invasion. This is a likely reason why high-grade serous cancers present with advanced stage with abdominal involvement in more than 70% of patients.4 The sine qua non of high-grade serous cancers is the dysregulation of p53 and associated effects on DNA repair, leading to genomic instability and the characteristic of high copy number variability.5 These tumors are also characterized by expression of WT-1, estrogen receptor α, and PAX8.6-8

From the Cancer Therapy Evaluation Program, National Cancer Institute, Rockville, MD. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Elise C. Kohn, MD, Cancer Therapy Evaluation Program, National Cancer Institute, 9609 Medical Center Dr., MSC9737, Rockville, MD 20850-9737; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 443

KOHN AND IVY

High-grade serous cancers are now being evaluated for subset analysis. Gene expression sets were found to segregate high-grade serous cancers into four descriptive groups: proliferative, mesenchymal, immune, and differentiated5,9; these groups have yet to be applied diagnostically or clinically. Further studies are ongoing to characterize the genomic patterns. The most validated prognostic and predictive biomarker within high-grade serous cancers is germline deleterious mutation in either BRCA1 or BRCA2 (gBRCA)4,10 and, with somewhat less support, somatic hom*ozygous loss of function of BRCA1 or BRCA2.11 As true suppressor genes, both copies must be disrupted or lost for the malignancy phenotype. The proteins encoded by BRCA1 and BRCA2 are critical for maintenance of the high-fidelity double-stranded DNA repair pathway, hom*ologous recombination repair.4,12 Loss of function of these genes requires loss normal p53 regulation for cellular viability; this is consistent with the observation that p53 overexpression precedes actual serous tubal in situ carcinoma formation.3 The Cancer Genome Atlas, which analyzed biospecimens from cases of newly diagnosed highgrade serous cancer, described 14% of HGSOC as having gBRCA status.5 Another approximately 6% have somatic hom*ozygous loss. Methylation of BRCA1 promoter has been described as associated with loss of function; however, controversy remains if this consistently yields a hom*ologous recombination dysfunction (HRD) phenotype, as does gBRCA or hom*ozygous somatic loss. More recently, studies have evaluated other proteins and genes within the hom*ologous recombination pathway and validated other genes wherein germline deleterious mutations have been observed. These are found in lower frequency, accounting for about 7% additional germline heritable mutations associated with ovarian cancer.13-16 Altogether, inclusive of BRCA1 methylation, this describes approximately one-third of all serous cancers. gBRCA is prognostic of generally good outcomes, at least up to the first postdiagnosis decade,17 and is predictive of platinum sensitivity and PARP inhibitor sensitivity. Studies are ongoing

KEY POINTS • HGSOC is an independent histologic and molecular set of cancers. • HGSOC is genomically unstable and can be classified by molecular subgroups, the clinical application of which is yet undetermined. • Biomarker tests have been developed that identify an HRD molecular phenotype, approximating BRCA-like drug sensitivity behavior. • Optimal treatment directions may be best identified by focusing on the development of clinical synthetic lethality across high-grade serous cancer molecular types. • Clinical synthetic lethality approaches may incorporate disruption of the tumor microenvironment and the immune milieu. 444 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

to validate prognostic and predictive utility of germline mutations in the other genes associated with familial ovarian cancer. Biomarkers to identify cancers with HRD, those that are gBRCA-like, have been developed,18,19 and one such biomarker has been approved as a companion diagnostic to the PARP inhibitor rucaparib.20 Earlier transcription array studies also led to the identification of a subset of ovarian cancers that overexpressed cyclin E.5,9,21-24 This has now been further supported by genomic studies such as The Cancer Genome Atlas.5 It is estimated that disruption of the G1/S cell-cycle transition by CCNE1 amplification (20% as estimated by The Cancer Genome Atlas), by overexpression or amplification of CCND1 or CCND2 (19%), or loss of regulation of the G1/S checkpoint by loss of function of pRB (10%) will account for nearly one-third to one-half of cases. Disruption of normal G1/S transitions also leads to poor DNA repair, also contributing to the classic genomic instability phenotype of ovarian cancers.25-27

What Are the Clinical Implications of a Diagnosis of HGSOC?

The preponderance of the women in clinical trials and represented in retrospective studies have HGSOC. Thus, much of the data in the literature on susceptibility to treatment, duration of response, and overall survival are driven by the behavior of this most prevalent type of ovarian cancer. Staging is used to categorize cancers for prognostic purposes, to guide therapeutic decisions, and as a classification tool for data analysis. The current 2014 International Federation of Gynecology and Obstetrics staging system,28 the primary system used worldwide, is a four-tiered system with staging based on pathologic evaluation of surgical staging. It is thus biased by the completeness and depth of surgery. However, practically, most trials and therapeutic triage are based on disease being early stage or organ confined (stage I) or advanced disease, which includes local pelvic extension. This is pertinent to high-grade serous cancers, more than 70% of which are advanced disease at presentation. Not included in International Federation of Gynecology and Obstetrics staging but of recognized importance for decades is the role of the extent of residual disease after primary or interval debulking surgery.4 Residual disease affects prognosis and is not specific to ovarian cancer type in its utility. The molecular makeup of high-grade serous cancer may have the greatest implication to patient prognosis and treatment secondary to diagnosis of ovarian cancer type. The aggressive genomic instability, caused by different molecular mechanisms, may lead to selective treatment directions. How this will affect initial therapy for high-grade serous cancers is currently the subject of many clinical trials. However, the molecular makeup has already been used to define access to one class of new anticancer agents approved for use in ovarian cancer. gBRCA-associated ovarian cancers have been shown to be substantially more susceptible to the class of PARP inhibitors, with platinum-sensitive gBRCA patients responding best (range, 35%–50% or more) and the lowest response rate (7%–12%) in women with wild-type

WHENCE HIGH-GRADE SEROUS OVARIAN CANCER

BRCA1 and BRCA2 whose tumors are platinum resistant.20,29 gBRCA status is thus a validated predictive biomarker for use of PARP inhibitors. The drive to identify other patients whose tumors may respond to PARP inhibitors has led to a test that is used to define HRD, where biology argues susceptibility to these DNA repair inhibitors.18-20 Laboratory and translational science has now broadened membership in the class of DNA repair inhibitor agents beyond the PARP inhibitors.12,30 Disruption of hom*ologous recombination can also come from inhibition of other key events in the complex hom*ologous recombination pathway.30 ATR and ATM kinases are critical to this form of DNA repair, and they have been found to have deleterious cancer-associated germline mutations. Inhibitors of these kinases are now in clinical testing.12 Another key element required for adequate DNA repair is either cell-cycle delay or sufficient time in the necessary cell-cycle phase to allow repair to proceed and complete. Block in G1/S or G2/M affects the type and extent of injury or repair, as well as potentially the type of cell death.27,31,32 Inhibitors of cell-cycle regulatory proteins are now recognized as potential targeted agents for cancer treatment and could be included in the DNA repair inhibitor class. Example agents include inhibitors of WEE-1 kinase and CHEK1 kinase.33-37 These kinases represent a yinyang scenario that ultimately affect a G2/M cell-cycle halt to allow DNA repair to proceed. Dysregulation of this cell-cycle checkpoint has been shown to propagate DNA damage because of inability to repair and have been shown to drive cells into apoptosis, autophagy, and mitotic catastrophe.38 Early clinical trials of agents targeting these kinases have had mixed results. AZ1775, a WEE-1 inhibitor, has some single-agent activity in gBRCA ovarian cancer and limited

single-agent activity otherwise. Preclinical and early clinical data suggest that it can synergize with chemotherapy or targeted agents to greatly improve their activity. A secondgeneration CHEK1 inhibitor with some inhibition against CHEK2, a modulator of both G1/S and G2/M, has been reported to have clinical activity in non-gBRCA recurrent highgrade ovarian cancer, and study is being expanded.

GENERATION OF CLINICAL SYNTHETIC LETHALITY

Clinical synthetic lethality may occur when a common underlying event(s) or drug causes a gain- or loss-of-function phenotype that, when combined with a drug targeted to a different pathway, collaborates to augment or create antitumor effects (Fig. 1).30,39 For example, the targeting of PARP and its many downstream functions synergizes with existing loss of hom*ologous repair function in tumors with hom*ozygous loss of function of BRCA1 or BRCA2.12 This results in greater clinical benefit in these patients than is seen in patients with wild-type and hom*ologous recombination-intact HGSOC.29 The latter subgroup of women do respond, albeit in a limited fashion. Investigations into creating clinical synthetic lethality to improve their outcomes to PARP inhibitors build on either contextual or chemical synthetic lethality. Chemical synthetic lethality occurs with the introduction of an additional agent(s) or modification of the microenvironment; contextual synthetic lethality leverages existing endogenous behaviors to greater benefit.30

Clinical Synthetic Lethality Opportunities in HGSOC

Recent reports of targeted drug combinations have introduced opportunities to examine the potential of clinical

FIGURE 1. Generation of Clinical Synthetic Lethality

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 445

KOHN AND IVY

synthetic lethality. For example, the combination of cediranib, a pan-VEGFR 1–3 inhibitor, and the PARP inhibitor olaparib demonstrated an unexpectedly high response rate and progression-free survival in women with HGSOC.40,41 Greater activity was observed in women without gBRCA in an unplanned post hoc subset analysis of the cediranib/olaparib study, 5 versus 16.5 months for single agent versus combination.40 Angiogenesis inhibitors have been shown to cause hypoxia and to alter local blood flow.42-44 Hypoxia has been shown to downregulate expression of critical DNA repair enzymes.45 Hypoxia induction, combined with chemical disruption of DNA repair with a PARP inhibitor, is an example of clinical synthetic lethality. Definitive studies are now ongoing to evaluate the benefits of this combination in platinum-sensitive (NCT02446600) and platinum-resistant (NCT02502266) HGSOC. Our understanding of the local tumor and stromal milieu of HGSOC has opened new directions for therapeutic investigation. It has long been known that microvessel density and angiogenic profusion is more common in advanced and aggressive ovarian cancers and parses out to be more common in the high-grade serous cancers.46,47 Not surprisingly, antiangiogenic therapies have clear benefits in newly diagnosed ovarian cancers48,49 and in recurrent disease, as single agents as well as in combinations.42,50-52 The local tumor microenvironment has immune infiltration. The strong presence of tumor-infiltrating lymphocytes is prognostic of outcome in ovarian cancer.53 This observation has been confirmed to hold fast in HGSOC.47,54 Interestingly, it appears that highly vascularized tumors may have different immune infiltration than those not vascularized and that the combination of the immune infiltration type and vascularity may affect prognosis. Patients with highgrade serous cancers containing high regulatory T-cell infiltration and high vascularity did better than patients with T-cell infiltration without vascularity.47 Characterization is ongoing to understand what types of immune phenotypes are within that milieu to understand how to better use immune-modulating agents. More recently, there is evidence that the same factors that drive angiogenesis are also important in attenuating the immune response.55 VEGF induces the accumulation of myeloid-derived suppressor cells and regulatory T cells and inhibits the migration of T lymphocytes from the vasculature into the tumor.56 A link has been proposed between hypoxia stress and immune suppression through the HIF1α and VEGF pathways through recruitment of regulatory T cells.57 This microenvironmental interaction between the stromal and tumor vasculature and the peritumoral and intratumoral immune responses may help identify reasons that current immune checkpoint inhibitor approaches are not as successful as anticipated in some cancers, including ovarian cancer. It has been hypothesized that there may be additional benefits to combining angiogenesis inhibitors, stromal inhibitors, or DNA repair inhibitors with immune checkpoint modulation; both preclinical and clinical investigations are ongoing. 446 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Propagation of poorly or unrepaired DNA in cells that do not succumb to such injury may result in mutations that, although perhaps not harmful, may create or unmask neoantigens.58-60 Not all such neoantigens may play a role in immune stimulation. It appears that there may be select common epitopes,58 or cancer-testis antigens, such as NYESO-1, that may activate T cell–mediated immunity more globally in patients with HGSOC. Current studies are incorporating measures of neoantigens and selective responsiveness to targeting cancer-testis antigens to test these questions. It remains unclear if these findings will be tumor-type specific, microenvironment (e.g., organ) specific, or generalizable. Clinical approaches to test these hypotheses include combinations of immune checkpoint inhibitors with angiogenesis inhibitors, some of which also incorporate DNA-damaging agents. New trials targeting immune checkpoint inhibitors with angiogenesis inhibitors have been initiated.

HRD PHENOTYPES AND BIOMARKERS

The ability to measure hom*ologous repair defects in a semiquantitative fashion to identify and select patients for treatment with PARP inhibitors is in the early stages of phenotype analysis but appears promising. The measurement of genomic instability cannot be quantitated with a single test; the presence or absence of gBRCA1/2 mutations is insufficient to provide a more global assessment of this highly plastic genome in HGSOC. Recently, three independent DNA-based measures (unweighted sum of scores, higher than 42) of genomic instability on the basis of loss of heterozygosity, telomeric allelic imbalance, and large-scale state transitions have been described as characterizing HRD.18,19 This has been validated prospectively for ovarian cancer in the study of niraparib presented at the 2016 Congress of the European Society for Medical Oncology. It was further investigated retrospectively using biospecimens and data from women with triple-negative breast cancer who received iniparib with cisplatin and gemcitabine. Triple-negative breast cancer tumors, including BRCA1/2 wild-type tumors, were more likely to respond to platinum-containing therapy if they demonstrated HRD as measured by a weighted summed score of loss of heterozygosity, telomeric allelic imbalance, and large-scale state transition.61 Rucaparib treatment was examined in a phase II trial for women with platinum-sensitive HGSOC, ARIEL2. The overall response rate was reported as 70%.20 The Foundation Medicine companion diagnostic HRD test for a BRCAness signature was evaluated in this trial, in which 40% of patients with the signature and 8% without the signature demonstrated response to rucaparib. This signature may prove useful in identifying patients who will benefit from PARP inhibitor therapy. The PARP inhibitor niraparib was examined in a randomized prospective trial of maintenance or placebo for women with high-grade ovarian cancer who have completed platinum-based therapy for recurrent disease. gBRCA patients receiving niraparib versus placebo had significantly longer median progression-free survival, 21 versus 5.5 months. The niraparib compared with placebo outcome was 12.9

WHENCE HIGH-GRADE SEROUS OVARIAN CANCER

versus 3.8 months in the gBRCA wild type cohort with HRD as measured using a composite HRD test. Among patients with platinum-sensitive, recurrent ovarian cancer, the median duration of progression-free survival was significantly longer among those receiving niraparib than among those receiving placebo, regardless of the presence or absence of gBRCA mutations or HRD status. The presence of an HRD phenotype correlated with outcome for the patients in each of the settings described above. These initial steps are critically important in the development of phenotypic biomarkers that can be used to select patients with hom*ologous DNA repair defects for treatment with PARP inhibitors and other inhibitors that interrogate the DNA damage response that is an integral part of cell replication and genomic instability.

CONCLUSION

HGSOC, now incorporating also high-grade endometrioid ovarian cancers, is a collection of relatively similar entities. They appear to originate from fimbrial fallopian tube epithelium and require p53 dysfunction to develop their characteristic genomic instability. Differential degrees of DNA repair dysfunction have been identified in different molecularly characterized subsets of HGSOC that may lead to selected future targeted clinical approaches. Leveraging the endogenous DNA repair dysfunction as identified in gBRCA or HRD patients with exogenously derived DNA repair dysfunction caused by induction or augmentation of local hypoxia is an example of clinical synthetic lethality that may further direct successful treatment combinations.

References 1. Kurman RJ, Carcangiu ML, Herrington CS, et al. WHO Classification of Tumours of the Female Reproductive Organs. Lyon: WHO Press; 2014.

response and survival in ovarian, fallopian tube, and peritoneal carcinomas. Clin Cancer Res. 2014;20:764-775.

2. Jarboe E, Folkins A, Nucci MR, et al. Serous carcinogenesis in the fallopian tube: a descriptive classification. Int J Gynecol Pathol. 2008;27:1-9.

15. Stover EH, Konstantinopoulos PA, Matulonis UA, et al. Biomarkers of response and resistance to DNA repair targeted therapies. Clin Cancer Res. 2016;22:5651-5660.

3. Mehra K, Mehrad M, Ning G, et al. STICS, SCOUTs and p53 signatures; a new language for pelvic serous carcinogenesis. Front Biosci (Elite Ed). 2011;3:625-634. 4. Jayson GC, Kohn EC, Kitchener HC, et al. Ovarian cancer. Lancet. 2014;384:1376-1388. 5. Cancer Genome Atlas Research Network. Integrated genomic analyses of ovarian carcinoma. Nature. 2011;474:609-615. 6. Sieh W, Köbel M, Longacre TA, et al. Hormone-receptor expression and ovarian cancer survival: an Ovarian Tumor Tissue Analysis consortium study. Lancet Oncol. 2013;14:853-862. 7. Rodgers LH, Ó hAinmhire E, Young AN, et al. Loss of PAX8 in highgrade serous ovarian cancer reduces cell survival despite unique modes of action in the fallopian tube and ovarian surface epithelium. Oncotarget. 2016;7:32785-32795. 8. de Cristofaro T, Di Palma T, Soriano AA, et al. Candidate genes and pathways downstream of PAX8 involved in ovarian high-grade serous carcinoma. Oncotarget. 2016;7:41929-41947. 9. Tothill RW, Tinker AV, George J, et al; Australian Ovarian Cancer Study Group. Novel molecular subtypes of serous and endometrioid ovarian cancer linked to clinical outcome. Clin Cancer Res. 2008;14:51985208. 10. Timms KM, Abkevich V, Hughes E, et al. Association of BRCA1/2 defects with genomic scores predictive of DNA damage repair deficiency among breast cancer subtypes. Breast Cancer Res. 2014;16:475. 11. Moschetta M, George A, Kaye SB, et al. BRCA somatic mutations and epigenetic BRCA modifications in serous ovarian cancer. Ann Oncol. 2016;27:1449-1455. 12. O’Connor MJ. Targeting the DNA damage response in cancer. Mol Cell. 2015;60:547-560. 13. Norquist BM, Harrell MI, Brady MF, et al. Inherited mutations in women with ovarian carcinoma. JAMA Oncol. 2016;2:482-490. 14. Pennington KP, Walsh T, Harrell MI, et al. Germline and somatic mutations in hom*ologous recombination genes predict platinum

16. Walsh T, Casadei S, Lee MK, et al. Mutations in 12 genes for inherited ovarian, fallopian tube, and peritoneal carcinoma identified by massively parallel sequencing. Proc Natl Acad Sci USA. 2011;108:18032-18037. 17. Kotsopoulos J, Rosen B, Fan I, et al. Ten-year survival after epithelial ovarian cancer is not associated with BRCA mutation status. Gynecol Oncol. 2016;140:42-47. 18. Abkevich V, Timms KM, Hennessy BT, et al. Patterns of genomic loss of heterozygosity predict hom*ologous recombination repair defects in epithelial ovarian cancer. Br J Cancer. 2012;107:1776-1782. 19. Birkbak NJ, Wang ZC, Kim JY, et al. Telomeric allelic imbalance indicates defective DNA repair and sensitivity to DNA-damaging agents. Cancer Discov. 2012;2:366-375. 20. Swisher EM, Lin KK, Oza AM, et al. Rucaparib in relapsed, platinumsensitive high-grade ovarian carcinoma (ARIEL2 Part 1): an international, multicentre, open-label, phase 2 trial. Lancet Oncol. 2017;18:75-87. 21. Etemadmoghadam D, George J, Cowin PA, et al; Australian Ovarian Cancer Study Group. Amplicon-dependent CCNE1 expression is critical for clonogenic survival after cisplatin treatment and is correlated with 20q11 gain in ovarian cancer. PLoS One. 2010;5:e15498. 22. Etemadmoghadam D, Weir BA, Au-Yeung G, et al; Australian Ovarian Cancer Study Group. Synthetic lethality between CCNE1 amplification and loss of BRCA1. Proc Natl Acad Sci USA. 2013;110:19489-19494. 23. Farley J, Smith LM, Darcy KM, et al; Gynecologic Oncology Group. Cyclin E expression is a significant predictor of survival in advanced, suboptimally debulked ovarian epithelial cancers: a Gynecologic Oncology Group study. Cancer Res. 2003;63:1235-1241. 24. Karst AM, Jones PM, Vena N, et al. Cyclin E1 deregulation occurs early in secretory cell transformation to promote formation of fallopian tube-derived high-grade serous ovarian cancers. Cancer Res. 2014;74:1141-1152. 25. Jabbour-Leung NA, Chen X, Bui T, et al. Sequential combination therapy of CDK inhibition and doxorubicin is synthetically lethal in p53-mutant triple-negative breast cancer. Mol Cancer Ther. 2016;15:593-607.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 447

KOHN AND IVY

26. Johnson SF, Cruz C, Greifenberg AK, et al. CDK12 inhibition reverses de novo and acquired PARP inhibitor resistance in BRCA wild-type and mutated models of triple-negative breast cancer. Cell Reports. 2016;17:2367-2381. 27. Alagpulinsa DA, Ayyadevara S, Yaccoby S, et al. A cyclin-dependent kinase inhibitor, dinaciclib, impairs hom*ologous recombination and sensitizes multiple myeloma cells to PARP inhibition. Mol Cancer Ther. 2016;15:241-250. 28. Prat J; FIGO Committee on Gynecologic Oncology. Staging classification for cancer of the ovary, fallopian tube, and peritoneum. Int J Gynaecol Obstet. 2014;124:1-5. 29. Gelmon KA, Tischkowitz M, Mackay H, et al. Olaparib in patients with recurrent high-grade serous or poorly differentiated ovarian carcinoma or triple-negative breast cancer: a phase 2, multicentre, open-label, non-randomised study. Lancet Oncol. 2011;12:852861. 30. Ivy SP, de Bono J, Kohn EC. The “Pushmi-Pullyu” of DNA repair: clinical synthetic lethality. Trends Cancer. 2017;2:646-656. 31. Dillon MT, Barker HE, Pedersen M, et al. Radiosensitization by the ATR inhibitor AZD6738 through generation of acentric micronuclei. Mol Cancer Ther. 2017;16:25-34. 32. Jirawatnotai S, Sittithumcharee G. Paradoxical roles of cyclin D1 in DNA stability. DNA Repair (Amst). 2016;42:56-62. 33. Jackson SP, Helleday T. DNA repair. Drugging DNA repair. Science. 2016;352:1178-1179. 34. Matheson CJ, Backos DS, Reigan P. Targeting WEE1 kinase in cancer. Trends Pharmacol Sci. 2016;37:872-881. 35. Karnitz LM, Zou L. Molecular pathways: targeting ATR in cancer therapy. Clin Cancer Res. 2015;21:4780-4785. 36. Morgan MA, Parsels LA, Zhao L, et al. Mechanism of radiosensitization by the Chk1/2 inhibitor AZD7762 involves abrogation of the G2 checkpoint and inhibition of hom*ologous recombinational DNA repair. Cancer Res. 2010;70:4972-4981. 37. Bauman JE, Chung CH. CHK it out! Blocking WEE kinase routs TP53 mutant cancer. Clin Cancer Res. 2014;20:4173-4175. 38. Morgan MA, Parsels LA, Maybaum J, et al. Improving the efficacy of chemoradiation with targeted agents. Cancer Discov. 2014;4:280291. 39. McLornan DP, List A, Mufti GJ. Applying synthetic lethality for the selective targeting of cancer. N Engl J Med. 2014;371:1725-1735. 40. Ivy SP, Liu JF, Lee JM, et al. Cediranib, a pan-VEGFR inhibitor, and olaparib, a PARP inhibitor, in combination therapy for high grade serous ovarian cancer. Expert Opin Investig Drugs. 2016;25:597611. 41. Liu JF, Barry WT, Birrer M, et al. Combination cediranib and olaparib versus olaparib alone for women with recurrent platinumsensitive ovarian cancer: a randomised phase 2 study. Lancet Oncol. 2014;15:1207-1214. 42. Azad NS, Posadas EM, Kwitkowski VE, et al. Combination targeted therapy with sorafenib and bevacizumab results in enhanced toxicity and antitumor activity. J Clin Oncol. 2008;26:3709-3714. 43. Lee JM, Peer CJ, Yu M, et al. Sequence-specific pharmaco*kinetic and pharmacodynamic phase I/Ib study of olaparib tablets and carboplatin in women’s cancer. Clin Cancer Res. Epub 2016 Sep 23.

448 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

44. Ng C, Zhang Z, Lee SI, et al. CT perfusion as an early biomarker of treatment efficacy in advanced ovarian cancer: an ACRIN and GOG study. Clin Cancer Res. Epub 2017 Feb 7. 45. Glazer PM, Hegan DC, Lu Y, et al. Hypoxia and DNA repair. Yale J Biol Med. 2013;86:443-451. 46. Hollingsworth HC, Kohn EC, Steinberg SM, et al. Tumor angiogenesis in advanced stage ovarian carcinoma. Am J Pathol. 1995;147:33-41. 47. Townsend KN, Spowart JE, Huwait H, et al. Markers of T cell infiltration and function associate with favorable outcome in vascularized highgrade serous ovarian carcinoma. PLoS One. 2013;8:e82406. 48. Burger RA, Brady MF, Bookman MA, et al; Gynecologic Oncology Group. Incorporation of bevacizumab in the primary treatment of ovarian cancer. N Engl J Med. 2011;365:2473-2483. 49. Perren TJ, Swart AM, Pfisterer J, et al; ICON7 Investigators. A phase 3 trial of bevacizumab in ovarian cancer. N Engl J Med. 2011;365:2484-2496. 50. Aghajanian C, Blank SV, Goff BA, et al. OCEANS: a randomized, doubleblind, placebo-controlled phase III trial of chemotherapy with or without bevacizumab in patients with platinum-sensitive recurrent epithelial ovarian, primary peritoneal, or fallopian tube cancer. J Clin Oncol. 2012;30:2039-2045. 51. Cannistra SA, Matulonis UA, Penson RT, et al. Phase II study of bevacizumab in patients with platinum-resistant ovarian cancer or peritoneal serous cancer. J Clin Oncol. 2007;25:5180-5186. 52. Pujade-Lauraine E, Hilpert F, Weber B, et al. Bevacizumab combined with chemotherapy for platinum-resistant recurrent ovarian cancer: the AURELIA open-label randomized phase III trial. J Clin Oncol. 2014;32:1302-1308. 53. Zhang L, Conejo-Garcia JR, Katsaros D, et al. Intratumoral T cells, recurrence, and survival in epithelial ovarian cancer. N Engl J Med. 2003;348:203-213. 54. Webb JR, Milne K, Watson P, et al. Tumor-infiltrating lymphocytes expressing the tissue resident memory marker CD103 are associated with increased survival in high-grade serous ovarian cancer. Clin Cancer Res. 2014;20:434-444. 55. Voron T, Marcheteau E, Pernot S, et al. Control of the immune response by pro-angiogenic factors. Front Oncol. 2014;4:70. 56. Kandalaft LE, Motz GT, Duraiswamy J, et al. Tumor immune surveillance and ovarian cancer: lessons on immune mediated tumor rejection or tolerance. Cancer Metastasis Rev. 2011;30:141-151. 57. Chouaib S, Messai Y, Couve S, et al. Hypoxia promotes tumor growth in linking angiogenesis to immune escape. Front Immunol. 2012;3:21. 58. Nathanson T, Ahuja A, Rubinsteyn A, et al. Somatic mutations and neoepitope hom*ology in melanomas treated with CTLA-4 blockade. Cancer Immunol Res. 2017;5:84-91. 59. Rizvi NA, Hellmann MD, Snyder A, et al. Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in nonsmall cell lung cancer. Science. 2015;348:124-128. 60. Snyder A, Makarov V, Merghoub T, et al. Genetic basis for clinical response to CTLA-4 blockade in melanoma. N Engl J Med. 2014;371:2189-2199. 61. Telli ML, Timms KM, Reid J, et al. hom*ologous recombination deficiency (HRD) score predicts response to platinum-containing neoadjuvant chemotherapy in patients with triple-negative breast cancer. Clin Cancer Res. 2016;22:3764-3773.

HEALTH SERVICES RESEARCH, CLINICAL INFORMATICS, AND QUALITY OF CARE

PATT ET AL

More Medicine, Fewer Clicks: How Informatics Can Actually Help Your Practice Debra A. Patt, MD, MPH, MBA, Elmer V. Bernstam, MD, MSE, MS, FACP, FACMI, Joshua C. Mandel, MD, David A. Kreda, and Jeremy L. Warner, MD, MS OVERVIEW In the information age, we expect data systems to make us more effective and efficient—not to make our lives more difficult! In this article, we discuss how we are using data systems, such as electronic health records (EHRs), to improve care delivery. We illustrate how US Oncology is beginning to use real-world evidence to facilitate trial accrual by automatic identification of eligible patients and how big data and predictive analytics will transform the field of oncology. Some information systems are already being used at the point of care and are already empowering clinicians to improve the care of their patients in real time. Telehealth platforms are being used to bridge gaps that currently exist in expertise, geography, and technical capability. Optimizing virtual collaboration, such as through virtual tumor boards, is empowering communities that are geographically disparate to coordinate care. Informatics methods can provide solutions to the challenging problems of how to manage the vast amounts of data confronting the practicing oncologist, including information about treatment regimens, side effects, and the influence of genomics on the practice of oncology. We also discuss some of the challenges of clinical documentation in the modern era, and review emerging efforts to engage patients as digital donors of their EHR data.

s cancer specialists, we have grown up in an information age in which we expect data systems to make us more effective and more efficient. There has been an unprecedented improvement in cancer death rates over the past 25 years.1 We expect machine learning and other informatics innovations to help us advance the quality of care even faster, and not simply result in additional boxes to check in an EHR. In 2017, there are growing concerns that these expectations have not been met, to the point at which some clinicians are citing dissatisfaction with health information technology as a major driver of job dissatisfaction.2 Despite this negativity, many tools are in development or operational for use in the clinic today to help make us better at what we do. Recently, there has been much interest in facilitating data systems to become more integrated and interoperable and to deliver care faster. These are recurring themes within the ASCO's participation in the Cancer Moonshot work and in the 21st Century Cures legislation that passed in 2016.3 The 2016 President’s Cancer Panel report, Improving Cancer-Related Outcomes With Connected Health, states, “We live at a most exciting and critical time of technological advances with potential to help individuals manage and improve their own health and support high-quality, patient-centered cancer care.”4

In this article, we explore some of the success that informatics can bring to the practice of oncology. First, we review some of the currently existing informatics capabilities at one author’s large integrated practice, US Oncology. Second, we discuss the topic of incorporating external knowledge into oncology practice and how informatics can provide point-of-care solutions. Third, we discuss the challenges of clinical documentation in the 21st century and how informatics tools can be used to make sense of messy real-world data. Finally, we turn toward patients and discuss how new technologies are emerging to enable digital donations to research.

CREATING AN INFORMATICS-ENABLED ONCOLOGY PRACTICE

Big Data

Over the past decade, we have seen advancements in big data systems that allow us to aggregate data in cancer beyond what has traditionally been collected by the cancer registry system. This is used across systems of care delivery to understand outcomes in various disease states outside of clinical trials in the form of real-world evidence. It remains a limitation in adult oncology that only 2% to 3% of patients enroll in prospective clinical trials, and age- and

From Texas Oncology, Austin, TX, McKesson Specialty Health and the US Oncology Network, The Woodlands, TX; School of Biomedical Informatics, The University of Texas Health Science Center, Houston, TX; Verily (Google Life Sciences), USA Research Faculty, Harvard Medical School, Cambridge, MA; Vanderbilt University, Vanderbilt Cancer Registry, Nashville, TN. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Jeremy L. Warner, MD, MS, Vanderbilt University, 2220 Pierce Ave., 777 PRB, Nashville, TN 37232; email: [emailprotected]. © 2017 American Society of Clinical Oncology

450 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

HOW INFORMATICS CAN HELP YOUR PRACTICE

race-related disparities persist, yet we would like to learn from all patients with cancer.5,6 There are also some diseases and novel molecular mechanisms in cancer that make clinical trials a challenge to open and accrue patients appropriately because of their rarity. Big data systems are a good answer to these challenges as they allow us to look for the needles within the haystacks and learn collectively. Some real-world evidence is also being used in other countries and is now even being considered by the U.S. Food and Drug Administration (FDA) for drug approval, with some caveats.7 The FDA issued draft guidance about the use of real-world evidence to support regulatory decisions around medical devices in 2016.8 Using real-world data systems to screen patients within care delivery systems, patients with rare diseases can be identified for clinical trials with just-intime mechanisms including opening clinical trials at the site of care when these patients are identified. Several big data aggregators are pioneers in this new landscape: (1) CancerLinQ, ASCO’s data aggregation and sharing platform among many EHR sources to enlighten outcomes and report quality in cancer care9,10; (2) the National Cancer Institute’s Genomic Data Commons (https://gdc.cancer.gov); (3) the Oncology Research Information Exchange Network, a collaboration of many prominent North American cancer centers; (4) the American Association for Cancer Research’s Project Genomics Evidence Neoplasia Information Exchange; (5) the Triangle Census Research Network at Duke University, informing data aggregation and dissemination; and (6) Project Data Sphere. Additional private entities that have invested tremendous resources in developing solutions for better use of cancer data include TriNetX, McKesson Specialty Health,

KEY POINTS • As cancer specialists, we have grown up in an information age in which we expect data systems to make us more effective and more efficient; despite recent concerns surrounding health information technology, we are convinced that there is significant potential that is yet to be met on the large scale. • US Oncology is a large integrated practice that has implemented big data, predictive analytics, and telehealth applications at the point of care. • Factual and contextual knowledge, especially regarding the interpretation of genomic sequence variation in cancer, will require external knowledge support that can be integrated into clinician workflow using emerging informatics technologies. • The challenges of clinical documentation in the 21st century are significant because of increasing care complexity and regulatory and billing requirements, but informatics technologies exist to facilitate documentation and its secondary use. • An emerging technology called Sync for Science will be piloted in 2017, with a goal of enabling patients to become digital donors for EHR-based research such as that envisioned by the Precision Medicine Initiative’s All of Us research program.

Flatiron Health, and IBM Watson Health, among many others. Other systems that are internal to organizations are integrating molecular data to identify patients for selection for clinical trials. For example, Syapse is a commercial partner that works within health systems to implement precision medicine programs. US Oncology research has a just-in-time mechanism of clinical trial initiation called the STAR program to accrue patients when identified with n-of1 tumors that would otherwise be hard to accrue in independent systems. Clinical decision support systems (CDSS) have become integrated in cancer care in many ways: facilitating compliance among clinicians in prescribing therapeutic interventions within guidelines of care delivery (a quality enhancement), facilitating screening for research accrual for appropriately selected patients, and using apps and other forms of digital patient engagement to inform patients to act on, alter, or contact their providers regarding their plans of care given their individualized data. Across a large network of US Oncology community practices that vary from urban and suburban to rural and frontier in their locations, a CDSS platform called Clear Value Plus was implemented, providing an interface at the point of service for chemotherapy ordering in a value-based mechanism within nationally accepted guidelines. The implementation of this CDSS significantly improved reportable data, guideline compliance, and exception reporting, making therapy decisions easier for doctors at the point of service, in addition to enhancing guideline compliance for patients and providing the necessary data to practices to proceed with prior authorization with payers, thus enhancing quality and time efficiencies.11 We fully acknowledge the real concerns regarding alert fatigue in the implementation of CDSS, and strongly encourage the efforts being made in the field of human-computer interaction to improve this experience. Predictive analytics platforms are being used to improve outcomes in patients with cancer. Data from EHRs and other data sources have been used to develop models to predict the risk for hospitalization in other diseases. For example, in populations with low-socioeconomic status and heart failure, risk prediction and the interventions based upon it reduced hospitalization risk in a high-risk and difficult-to-treat population.12 Warner et al have previously described an EHR-based predictive model for hospital-acquired complications.13 Using such models to inform the care team about risk and facilitate appropriate interventions may be effective at reducing hospitalizations and readmissions in highrisk groups. Similar models have been developed for high-risk cancers and are being used to inform clinicians about risk and facilitate support systems and care interventions to reduce the risk for hospitalization accordingly.14 Prediction of treatment intolerance, which can lead to nonadherence, especially in novel therapies, is also an evolving area of research, although the prior body of evidence suggests that side effects are a common reason for nonadherence and early discontinuation for traditional therapies as well.15 Given that recent studies suggest that treatment discontinuation asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 451

PATT ET AL

FIGURE 1. Predictive Analytics to Understand Risk

Multiple sources of data are used to create a risk model that can predict likely outcomes for individual patients, such as the risk for hospitalization after the administration of chemotherapy.

because of intolerance is the most common reason for revolutionary drugs such as ibrutinib,16,17 it will be critically important to identify vulnerable populations. Presently, predictive analytic platforms can come across the EHR in the form of CDSS so that they are available to clinicians at the point of care (Fig. 1). Telehealth and virtual collaboration platforms are another way US Oncology uses data systems to enhance care delivery with efficiency. Use of these platforms is growing in scope, scale, and prevalence throughout the United States, and many states are currently considering policies that influence their implementation.18 Innovations in platforms of interaction telehealth and virtual collaboration allow us to bridge existing gaps in geography and expertise. In the US Oncology network of oncology practices, there are sites of service that vary from urban and suburban to rural and frontier, and staffing and subspecialty expertise is also variable. Telehealth platforms have allowed for consultation with subspecialty experts in neuro-oncology and genetic risk assessment that otherwise would have required a drive of several hours, a geographic barrier to care that frequently results in diminished utilization of subspecialty services. This has allowed patients to access subspecialty services and treatments they otherwise would not have access to and it helps us make quality care global. Our present abilities to implement telehealth include multimodal virtual collaboration (between clinicians or between clinicians and patients) and remote review of imaging and pathology, but also have become enhanced in our ability to complement the physical examination with universal serial bus attachments such as sphygmomanometers, stethoscopes, ophthalmoscopes, electrocardiographs, ultrasound probes, and cameras to make the skin examination more sensitive than to the human eye. Teledermatoscopy programs have been implemented to diagnose and follow at-risk skin lesions to detect 452 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

early melanoma.19 There are even multimodal fiber optic probes that can be used at remote sites with a clinician’s assistance to interrogate cutaneous lesions and replace the need for some biopsies in skin cancer.20 How we interact with these systems continues to change over time. These are not only technological advancements but also ways we must think differently about supporting clinical workflow to optimize the patient experience as our technological capabilities grow. Optimizing virtual collaboration with all of these platforms also allows virtual multidisciplinary planning, which can often be a critical quality measure in planning cancer care treatment. Virtual tumor boards are now in existence in many networks today, allowing geographically disparate multidisciplinary planning.21 All of these systems are tools that close physical and mental gaps that limit care delivery today.

INTEGRATING EXTERNAL KNOWLEDGE INTO THE ONCOLOGY WORKFLOW

Generally speaking, there are four types of knowledge pertinent to the day-to-day practice of oncology: procedural, transactional, factual, and contextual. The first category, procedural knowledge, includes those factors pertinent to daily practice and is usually specific to a given location. Examples may include (1) what antibiotics and antiemetics are available in the hospital formulary, (2) the times and days laboratory technicians are available to assist with bone marrow biopsy and aspiration procedures, and (3) standard protocols decided by consensus or disease group leadership. Often, procedural knowledge is kept locally in the form of standard operating procedures. The next category, transactional knowledge, includes those factors pertinent to the business aspects of oncology practice. This includes knowledge about what billing codes (e.g., International Classification of Diseases, 10th Revision,

HOW INFORMATICS CAN HELP YOUR PRACTICE

Clinical Modification) are necessary and sufficient to justify a given level of professional billing, what billing codes will translate into an appropriate diagnosis-related group for a given hospitalization, and details of negotiated contracts with third-party insurers and pharmaceutical companies. As with procedural knowledge, most transactional knowledge is location specific, although some, such as information about International Classification of Diseases codes, may be amenable to external knowledge resources. The last two categories are inter-related. Factual knowledge includes information about a disease, a prognosis, and associated treatments. One is likely to find this type of knowledge in an encyclopedia or a medical textbook. Importantly, factual knowledge is by convention limited to a representative example or a range of commonly expected examples. In other words, factual knowledge is generic and often not applicable to an individual patient. On the other hand, contextual knowledge takes into account features of an individual patient and is necessary (although not sufficient) for the practice of precision or personalized medicine.22 In oncology practice, context includes comorbidities; performance status; treatment history, including prior drug exposures, length and depth of response, and pertinent adverse events; behavioral determinants of health such as substance abuse; psychological distress and psychosocial support systems; and belief systems (e.g., Jehovah’s Witnesses will not accept blood transfusions, which may influence decisions about the intensity of cytotoxic chemotherapy). The crux of the issue of knowledge management in oncology is what proportion of knowledge is internal as opposed to external. Internalized knowledge is that which is available to a practitioner through memory, with or without prompting. Externalized knowledge is that which is available through any type of ancillary resource. Internal knowledge is not to be taken lightly; after all, much of the 4 years of medical school and 5 to 7 years of postgraduate training are focused on the tasks of acquiring and retaining knowledge. Nevertheless, it was observed many decades ago that a practicing clinician could not possibly grasp the totality of medical knowledge.23 We have previously determined with some back-of-the-envelope calculations that it would be necessary to read 272 articles each day, 365 days a year, just to keep up with the cancer literature.24 So how is a practicing clinician to approach and master both facts and context? Historically, the incorporation of external knowledge into the clinical workflow falls under CDSS. CDSS is part of the lingua franca of biomedical informatics and features prominently in recent regulations, including the meaningful-use rules and the FDA’s guidance on the regulation of mobile medical applications.25,26 Interestingly, one of the earliest CDSS, ONCOCIN,27 was an oncology protocol management system. Indeed, the selection of chemotherapy protocols and tracking of their dosing parameters is one of the areas most amenable to external knowledge support. Although certain EHRs provide the means to build and maintain local chemotherapy regimen libraries, these are rarely

comprehensive. In 2011, Dr. Peter Yang founded the site HemOnc.org, with the goal of creating a freely available, comprehensive, and accurate resource for chemotherapy regimen details.28 The site listed more than 1,000 regimens by mid-2013, and, as of February 2017, HemOnc.org listed more than 2,000 disease-specific chemotherapy regimens across 84 distinct solid oncology, benign, and malignant hematology conditions; to our knowledge, it is the largest resource of its kind. Over time, the initial focus on capturing details of dosing and timing of chemotherapeutics has expanded to also include information on comparative efficacy and toxicity for randomized controlled trials and overall response rates for nonrandomized studies (Fig. 2). HemOnc.org and similar resources can offer the practicing oncologist the ready means to bring external knowledge to bear, especially when prescribing obscure or infrequently used regimens. Another solution to this knowledge management problem is pathways. Pathways take into account cost, reimbursem*nt, efficacy, and the likelihood of treatment-related complications to varying degrees. Some, such as the National Comprehensive Cancer Network guidelines, provide a fair amount of latitude in treatment selection; others, including several vendor products, enforce treatment choices with potential penalties for overrides. Zon et al29 recently criticized the pathway approach for lacking clear processes, placing additional administrative burdens on oncology practices and not yet clearly demonstrating an impact on patient health outcomes. The other area most in need of external knowledge support is genomically guided treatment. This is a knowledge space that is simply too large to manage without assistance. Knowledge support in this evolving area is taking several forms: (1) extensively curated somatic panel test reports, (2) molecular tumor boards that convene experts either locally or remotely, and (3) genomic knowledge bases. Although curated reports are critical, they suffer from two major flaws: (1) they are a snapshot from the time when the test was obtained and will not reflect the new genotype-phenotype knowledge that is constantly emerging, and (2) they are subject to considerable variation, as recently demonstrated by Balmaña et al.30 Molecular tumor boards can be both clinically useful and educational but do not necessarily fall within normal clinic workflows. Genomic knowledge bases hold great promise but currently lack uniformity in format and interpretation. Recently, Ritter et al,31 on behalf of the ClinGen Somatic Cancer Working Group, described a consensus for minimum variant level data, which is followed by knowledge bases such as My Cancer Genome32 and ClinGen.33 The FDA has recently issued draft guidance on the use of public human genetic variant databases to support clinical validity of next-generation sequencing panels.34 ASCO, the Association for Molecular Pathology, and the College of American Pathologists recently issued a unified set of standards and guidelines for the interpretation and reporting of sequence variants in cancer. These efforts should eventually improve the uniformity of genomic test results.35 Once genomic data are integrated into the EHR, the capacity for further innovation expands.36 In particular, a new asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 453

PATT ET AL

standard produced by Health Level Seven International, called the Fast Healthcare Interoperability Resources (FHIR) standard, increases the ability to bring external knowledge, including genomics, into the clinical workflow. Warner et al previously demonstrated a FHIR-based app, called SMART Precision Cancer Medicine, that links users to factual genomic knowledge (via Gene Wiki), contextual genomic knowledge (via My Cancer Genome), and contextual chemotherapy regimen knowledge (via HemOnc.org).37 This app was designed to launch from a tablet device with the intent of seamless integration into the workflow of a busy clinic; it and similar apps can also easily be integrated into certain EHR environments to provide a seamless user experience.38 ASCO is currently investigating the possibility of creating an app that will bring the results from multiple genomic knowledge bases to clinician users. It is clear that clinical decision support, especially the invasive variety that disrupts workflow through alerts and reminders, can be perceived negatively. In anticipation of a backlash, Bates et al39 produced the seminal article “Ten Commandments for Effective Clinical Decision Support: Making the Practice of Evidence-Based Medicine a Reality” in 2003. This group and others have also documented the frequent practice of overriding alerts, even when the result may be a fatal drug interaction.40,41 Nevertheless, it is likely that clinical decision support and passive knowledge support will increasingly become available within the clinical workflow, ideally in the form of apps clinicians can select and customize to meet their needs.42 The final issue that must be addressed is the accuracy of knowledge. Although there is no shortage of studies demonstrating the fallibility and malleability of internal knowledge

(the seminal paper by Tversky and Kahneman43 is an excellent primer), the failure of accuracy of external knowledge is often treated more harshly. This is likely an issue of trust more than anything. Failure of internal knowledge banks may be attributable to a variety of factors, but it is the rare practitioner who has a fundamental lack of trust in his or her own knowledge. Conversely, external knowledge that is incorrect and provably false can raise issues of trust pertaining to the entire knowledge base. This phenomenon is well demonstrated by the ongoing skepticism of the Wikipedia resource, despite academic publications showing high levels of accuracy in certain areas of the medical domain.44-47 Various approaches have been used to increase trust in external knowledge bases, especially those that are openly collaborative: (1) clear attribution of content to well-known experts, (2) restriction of content creation to vetted individuals, and (3) stamps of approval from specialty societies or other agencies. It remains to be seen which of these approaches, or a combination thereof, will be most successful in gaining the trust of the user community.

THE CHALLENGES OF CLINICAL DOCUMENTATION IN THE 21ST CENTURY

Clinical documentation has always served multiple purposes, including clinical (record of clinical reasoning, decisions, and clinically relevant events), billing and financial (justifying payment for services rendered), and legal (What happened? Who knew what and when?). Over time, practices and processes evolved that variably addressed all of these purposes. Some of these processes and practices were formal, but some were informal and specific to individual clinicians. With the implementation of EHRs, many of

FIGURE 2. A Portion of the Docetaxel for Non–Small Cell Lung Cancer Regimen on HemOnc.org, Showing Comparative Efficacy for 11 Randomized Controlled Trials

Hyperlinks under "Study" link to the original articles; those under "Comparator" link to other regimens on HemOnc.org. Abbreviations: IV, intravenous; OS, overall survival; PFS, progression-free survival.

454 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

HOW INFORMATICS CAN HELP YOUR PRACTICE

these processes were changed, either consciously or unconsciously. Although there are clear benefits of computerizing health care, there are also a number of challenges, particularly related to documentation. In this section, we discuss three challenges posed by the computerization of clinical documentation and the changing health care environment.

Structuring Inherently Messy Clinical Reality

The first challenge relates to a fundamental mismatch between our tools and what we want to accomplish. Clinical reality is difficult to define precisely. Consider this simple example: Is hypertension a disease of the blood vessels? Heart? Kidneys? Brain? Clearly all are affected and involved. However, putting hypertension into a category with well-defined necessary and sufficient conditions is challenging. Further, clinical concepts are continuously evolving. For example, the definition of a gene has changed dramatically as our understanding of molecular biology has improved. Similarly, genomically informed therapy is constantly evolving. Today’s variant of unknown significance may be an actionable (targetable) variant tomorrow. With paper records, we were not tied to specific categories. We recorded our thoughts using natural language rather than trying to express our thoughts using a predefined set of categories that are often inadequate to represent our intended meaning. Freehand drawings or diagrams could be inserted where appropriate, and shorthand was widely used. Rosenbloom et al48 published an overview of the tensions between structured and unstructured clinical documentation. There is no easy answer to this challenge. However, there are promising developments. First, natural language processing technology can be very useful when 100% accuracy is not required. For example, algorithms can identify specific concepts in the text even when they are not referenced with a particular name (e.g., breast cancer, brst ca, and IDC [invasive ductal carcinoma] can be recognized as synonymous). This can be very useful for a variety of purposes, including identifying cases of a particular malignancy, cancer stage, or treatment outcome in large clinical data sets and for automatically summarizing complex patient histories. A review of natural language processing in oncology was recently published in JAMA Oncology.49 Second, the field of human-computer interaction has developed into an engineering discipline with validated approaches to matching users and tasks (e.g., a clinician who has to write a note documenting an office visit) to specific tools and their characteristics. Usability experts can define existing workflow, identify areas that can be improved, and guide implementation of systems that match user needs. Professional organizations have recognized that improving the usability of clinical systems has the potential to improve clinical outcomes (e.g., by reducing errors) and have published recommendations for incorporating usability into the design of clinical systems.50

Competing Priorities (Business Versus Clinical Needs)

In many important ways, health care is a business. Institutions are reasonably concerned about their financial performance and must comply with an increasing regulatory burden. As a result, the decision to implement a clinical system is often driven more by business concerns rather than clinical needs, for example, the need to document compliance and streamline financial (billing) operations. Clinical reimbursem*nt has traditionally relied on documentation of specific services rendered. Thus, clinical notes now contain specific billing-oriented phrases such as “40 minutes spent at the bedside with greater than 50% of this time spent on counseling.” This, along with the need to document in increasing detail to justify specific service levels, has led to administratively compliant but clinically less useful documentation. Further, computerized physician order entry is a very effective way to track and influence clinical behavior. An undesirable behavior (e.g., daily laboratories) can simply be made inconvenient to order (e.g., by requiring a daily written justification). Thus, health information technology has increased the ability of the business enterprise to monitor and influence the clinical enterprise without assuming direct responsibility for clinical outcomes. This challenge is primarily social, rather than technical. For a variety of reasons, clinicians have been reluctant advocates for clinical priorities. As a result, business priorities may outweigh clinical priorities at times simply because the clinical enterprise lacks effective representation when the relevant decisions are being made. To their credit, organizations increasingly recognize the need for clinical champions in board rooms and are hiring clinician-informaticians to lead clinical information technology efforts (e.g., chief medical information officers).

Ease of Creating Data Versus Useful Information

Current computer technology makes it very easy to generate and replicate data. For example, with a few clicks, one can copy and paste radiology reports, laboratory studies, past notes, and any other data contained in a clinical system. As a result, notes that were previously succinct have become unreadable. In contrast, it requires much more effort to summarize the clinically relevant facts. Thus, ironically, health information technology has decreased our ability to manage information. Patients who enter the hospital with hypercalcemia leave with hypercalcemia, and errors are perpetuated verbatim from note to note. Part of the problem is that trainees are encouraged to document fully to avoid being accused of missing something important and to support billing. However, errors may creep into a long note that is assembled from pieces of other notes. Institutions struggle to develop policies that balance the need to repeatedly document the same information (e.g., often the physical exam does not change from day to day in a hospitalized patient, unless it does) and ensure that important changes are not missed. Currently, there is general agreement that cut and paste or cloning of asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 455

PATT ET AL

clinical notes is undesirable. However, there is not yet a consensus regarding best practice or how to configure clinical systems to support best practice. Clearly, the current model where by some reports clinicians are spending twice as much time documenting as they are with patients, is no longer tenable. Creative solutions such as voice-to-text systems with predictive analytic features that can autocomplete notes, scribes that can be present in a clinical encounter with disrupting workflow or rapport, and structured authoring tools will all need to be exhaustively tested in the field in order to help busy oncologists get through their clinical day.

DIGITAL DONATIONS: PATIENT CONSENT AND ENABLING TECHNOLOGIES IN THE EHR

Under the first stages of the meaningful-use EHR incentive program (2010–2015), adoption of EHR systems increased from 51% to 87% in outpatient practices51 and from 16% to 84% in hospitals.52 Increased availability of clinical data (including problem lists, laboratory results, prescription history, and free-text notes) presents a growing opportunity for researchers.53 For example, combining EHR data with adverse event reporting databases has led to automated detection of previously unknown adverse drug reactions.54 EHRs also present an opportunity for prospective research studies, as an adjunct to (and a cross-check for) data collected directly from research participants through traditional paper-based forms or recent innovations using app-based interactions.55 However, researcher access to EHR data has traditionally been limited to institutional settings, where data from a single clinical system or a small network of collaborating systems are available to researchers within the network. These systems expand to form wider networks with more data available to qualified researchers, as in the network of networks established by the Patient-Centered Outcomes Research Institute’s Clinical Data Research Network awardees.56 Such networks are constructed along the grain of institutional boundaries, with careful legal agreements needing to be established among entities as a prerequisite for data sharing. These institution-based studies can provide relatively easy access to EHR data by creating their own legal frameworks for intramural data sharing. On the other hand, many research studies cut across the grain of institutional boundaries. For example, diseased-focused organizations such as the Multiple Myeloma Research Foundation create community-based registries that identify patients across the country on the basis of disease state and without regard to institutional affiliation.57 We call these participant-based studies. We should highlight that this distinction is not a bright line; studies such as the Precision Medicine Initiative’s All of Us research program pursue a hybrid approach by recruiting from in-network health care systems as well as the general population.58 One model for sharing clinical records with participantbased studies is to engage participants to mediate the transfer. For example, after a participant completes an informed 456 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

consent process, researchers might ask the patient to collect her own clinical records from the hospitals where she has received care. This model confers comprehensive access to clinical records by leveraging a patient’s right under the Health Insurance Portability and Accountability Act to access a designated record set.59 But the barriers for this model are formidable for otherwise willing participants, including driving to multiple hospitals to visit the medical records departments and filling out multiple authorization forms. In addition, data arriving from faxed or photocopied page-formatted documents instead of electronic structured data add sources of error as well as cost and time. These shortcomings can be addressed through existing law and additional technology, in particular an application programming interface (API) that can retrieve and move data from EHRs to researchers. Three key enablers are established by federal laws and regulations: (1) The right under the Health Insurance Portability and Accountability Act for a patient to access his or her own medical records, (2) the meaningful-use stage 3 requirement that patients may access their health information with the applications they choose, and (3) the meaningful-use common clinical data set, which establishes a minimum set of data to be made available to patients through such an API, including patient demographics, allergies, immunization, medications, laboratory results, and vital signs.60 Of note, the regulations do fall short of defining common standards for the API, which means that certified EHRs could choose to expose these data with proprietary formats and incompatible interfaces. Through a National Institutes of Health–funded and Office of the National Coordinator for Health Information Technology–supported effort called Sync for Science, Dr. Mandel, Mr. Kreda, and colleagues are undertaking a pilot project with six commercial EHR vendors to establish and test a common, nonproprietary interface for sharing data with research. Building on open standards established through SMART on FHIR and the Argonaut Project,38 Sync for Science has defined a focused set of APIs for EHR vendors to implement. These APIs are published alongside new functionality in each vendor’s patient portal, giving patients the means to approve sharing their data with apps. Under this model, a research study can create an app that asks participants for access to their EHR data. The Sync for Science technology delivers the patient’s approval in the form of an access token following the OAuth 2.0 specification61 (Fig. 3), allowing a research app access to a participant’s EHR data for a designated period of time (typically 1 year). To support vendor implementation of the Sync for Science APIs, developer documentation (http://syncfor.science/apicalls) and a test suite that connects to each vendor’s portal and provides a compatibility report have been developed. The test suite verifies the availability of sample data, validating that payloads conform to the FHIR specification and checking coded terms against a set of expected vocabularies. Although errors and warnings are produced when data fail to match expectations, the tests are permissive, allowing

HOW INFORMATICS CAN HELP YOUR PRACTICE

FIGURE 3. Workflow for a Patient Sharing Data With a Research Study, Using an App Built Using Sync for Science Technology

researchers to obtain a richer set of real-world (if occasionally messy) data rather than a smaller set of cleaner data. Over the course of 2017, EHR vendors are working to deploy this technology at approximately 15 pilot sites around the United States. Although these APIs are designed to support any patient-selected application, the pilot deployment focuses on the Precision Medicine Initiative’s All of Us research program as an initial testbed. During the pilot phase, a known set of provider sites has been engaged to enable access to a single, well-respected research app, which will provide crucial experience with API-driven data sharing. That said, three impediments in scaling this technology to support a wider ecosystem of research studies are anticipated: (1) building a robust sharemy-data feature requires a high-quality national provider directory that includes API endpoints for each provider, (2) connecting an app to a provider system still requires registration, a step for which not all vendors have provided an automated approach, and (3) despite regulations that empower patients to access and share their data as they

choose, many health care providers are not yet comfortable enabling connections to unknown apps.

CONCLUSION

Many data systems have evolved to support improved quality of care with greater efficiencies. Despite the richness of available data and the life-threatening nature of cancer, their use throughout oncology practice remains more limited today than in other chronic diseases. Translating innovations developed in the informatics research space into clinical practice is every bit as important as traditional bench-to-bedside translational science. To facilitate such knowledge transfer, ASCO recently launched two journals to explicitly link the cancer and informatics and bioinformatics communities together: JCO Precision Oncology and JCO Clinical Cancer Informatics. Sharing information through these and similar venues, as well as through presentation at conferences, will remain paramount in helping us all benefit from this innovation faster and ultimately allowing us to deliver more medicine with fewer clicks.

References 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2017. CA Cancer J Clin. 2017;67:7-30.

5. Murthy VH, Krumholz HM, Gross CP. Participation in cancer clinical trials: race-, sex-, and age-based disparities. JAMA. 2004;291:2720-2726.

2. Friedberg MW, Chen PG, Van Busum KR, et al. Factors Affecting Physician Professional Satisfaction and Their Implications for Patient Care, Health Systems, and Health Policy. Santa Monica, CA: Rand Corporation; 2013.

6. Tejeda HA, Green SB, Trimble EL, et al. Representation of AfricanAmericans, Hispanics, and whites in National Cancer Institute cancer treatment trials. J Natl Cancer Inst. 1996;88:812-816.

3. Lyman GH, Moses HL. Biomarker tests for molecularly targeted therapies: laying the foundation and fulfilling the dream. J Clin Oncol. 2016;34:2061-2066.

7. Sherman RE, Anderson SA, Dal Pan GJ, et al. Real-world evidence— what is it and what can it tell us? N Engl J Med. 2016;375: 2293-2297.

4. National Cancer Institute, Division of Extramural Activities. President’s Cancer Panel—Reports. https://deainfo.nci.nih.gov/advisory/pcp/ annualReports/. Accessed January 3, 2017.

8. Center for Devices and Radiological Health. Use of Real-World Evidence to Support Regulatory Decision-Making for Medical Devices: Draft Guidance for Industry and Food and Drug Administration Staff. http://www.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 457

PATT ET AL

fda.gov/downloads/medicaldevices/deviceregulationandguidance/ guidancedocuments/ucm513027.pdf. Accessed January 29, 2017. 9. Sledge GW Jr, Miller RS, Hauser R. CancerLinQ and the future of cancer care. Am Soc Clin Oncol Educ Book. 2013;33:430-434. 10. ASCO forges ahead with CancerLinQ. Cancer Discov. 2014;4:OF4. 11. Patt D. Better screening using big data. J Oncol Pract. 2016;12:699700. 12. Amarasingham R, Patel PC, Toto K, et al. Allocating scarce resources in real-time to reduce heart failure readmissions: a prospective, controlled study. BMJ Qual Saf. 2013;22:998-1005. 13. Warner JL, Zhang P, Liu J, et al. Classification of hospital acquired complications using temporal clinical information from a large electronic health record. J Biomed Inform. 2016;59:209-217. 14. Antonuzzo A, Vasile E, Sbrana A, et al. Impact of a supportive care service for cancer outpatients: management and reduction of hospitalizations. Preliminary results of an integrated model of care. Support Care Cancer. 2017;25:209-212. 15. Chirgwin JH, Giobbie-Hurder A, Coates AS, et al. Treatment adherence and its impact on disease-free survival in the Breast International Group 1-98 trial of tamoxifen and letrozole, alone and in sequence. J Clin Oncol. 2016;34:2452-2459. 16. Jain P, Keating M, Wierda W, et al. Outcomes of patients with chronic lymphocytic leukemia after discontinuing ibrutinib. Blood. 2015;125:2062-2067. 17. Jain P, Thompson PA, Keating M, et al. Causes of discontinuation and long-term outcomes of patients with cll after discontinuing ibrutinib. https://ash.confex.com/ash/2016/webprogram/Paper95542.html. Accessed January 19, 2017. 18. Dorsey ER, Topol EJ. State of telehealth. N Engl J Med. 2016;375:154161. 19. Congalton AT, Oakley AM, Rademaker M, et al. Successful melanoma triage by a virtual lesion clinic (teledermatoscopy). J Eur Acad Dermatol Venereol. 2015;29:2423-2428. 20. Sharma M, Marple E, Reichenberg J, et al. Design and characterization of a novel multimodal fiber-optic probe and spectroscopy system for skin cancer applications. Rev Sci Instrum. 2014;85:083101. 21. Marshall CL, Petersen NJ, Naik AD, et al. Implementation of a regional virtual tumor board: a prospective study evaluating feasibility and provider acceptance. Telemed J E Health. 2014;20:705-711.

28. Warner JL, Cowan AJ, Hall AC, et al. HemOnc.org: a collaborative online knowledge platform for oncology professionals. J Oncol Pract. 2015;11:e336-e350. 29. Zon RT, Frame JN, Neuss MN, et al. American Society of Clinical Oncology policy statement on clinical pathways in oncology. J Oncol Pract. 2016;12:261-266. 30. Balmaña J, Digiovanni L, Gaddam P, et al Conflicting interpretation of genetic variants and cancer risk by commercial laboratories as assessed by the Prospective Registry of Multiplex Testing. J Clin Oncol. 2016;34:4071-4078. 31. Ritter DI, Roychowdhury S, Roy A, et al; ClinGen Somatic Cancer Working Group. Somatic cancer variant curation and harmonization through consensus minimum variant level data. Genome Med. 2016;8:117. 32. Swanton C. My Cancer Genome: a unified genomics and clinical trial portal. Lancet Oncol. 2012;13:668-669. 33. Rehm HL, Berg JS, Brooks LD, et al; ClinGen. ClinGen—the Clinical Genome Resource. N Engl J Med. 2015;372:2235-2242. 34. U.S. Food and Drug Administration. Precision Medicine Initiative. http:// www.fda.gov/ScienceResearch/SpecialTopics/PrecisionMedicine/ default.htm. Accessed January 29, 2017. 35. Li MM, Datto M, Duncavage EJ, et al. Standards and guidelines for the interpretation and reporting of sequence variants in cancer: a joint consensus recommendation of the Association for Molecular Pathology, American Society of Clinical Oncology, and College of American Pathologists. J Mol Diagn. 2017;19:4-23. 36. Warner JL, Jain SK, Levy MA. Integrating cancer genomic data into electronic health records. Genome Med. 2016;8:113. 37. Warner JL, Rioth MJ, Mandl KD, et al. SMART precision cancer medicine: a FHIR-based app to provide genomic information at the point of care. J Am Med Inform Assoc. 2016;23:701-710. 38. Mandel JC, Kreda DA, Mandl KD, Kohane IS, Ramoni RB. SMART on FHIR: a standards-based, interoperable apps platform for electronic health records. J Am Med Inform Assoc. 2016;23:899-908. 39. Bates DW, Kuperman GJ, Wang S, et al. Ten commandments for effective clinical decision support: making the practice of evidencebased medicine a reality. J Am Med Inform Assoc. 2003;10:523-530. 40. Slight SP, Seger DL, Nanji KC, et al. Are we heeding the warning signs? Examining providers’ overrides of computerized drug-drug interaction alerts in primary care. PLoS One. 2013;8:e85071.

22. Collins FS, Varmus H. A new initiative on precision medicine. N Engl J Med. 2015;372:793-795.

41. Yeh M-L, Chang Y-J, Wang P-Y, et al. Physicians’ responses to computerized drug-drug interaction alerts for outpatients. Comput Methods Programs Biomed. 2013;111:17-25.

23. Durack DT. The weight of medical knowledge. N Engl J Med. 1978;298:773-775.

42. Mandl KD, Mandel JC, Kohane IS. Driving innovation in health systems through an apps-based information economy. Cell Syst. 2015;1:8-13.

24. Warner JL. Grappling with the data explosion in oncology. Oncol Hematol Rev US. 2015;11:102-103.

43. Tversky A, Kahneman D. Judgment under uncertainty: heuristics and biases. Science. 1974;185:1124-1131.

25. Blumenthal D, Tavenner M. The “meaningful use” regulation for electronic health records. N Engl J Med. 2010;363:501-504.

44. Clauson KA, Polen HH, Boulos MN, et al. Scope, completeness, and accuracy of drug information in Wikipedia. Ann Pharmacother. 2008;42:1814-1821.

26. U.S. Food and Drug Administration. Mobile medical applications. http:// www.fda.gov/MedicalDevices/DigitalHealth/MobileMedicalApplications/ ucm255978.htm. Accessed January 29, 2017. 27. Shortliffe EH, Scott AC, Bischoff MB, et al. An expert system for oncology protocol management. In Buchanan BG, Shortliffe EH (eds). Rule-Based Expert Systems: The MYCIN Experiments of the Stanford Heuristic Programming Project. Reading, MA: Addison-Wesley; 1984: 653-665.

458 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

45. Kräenbring J, Monzon Penza T, Gutmann J, et al. Accuracy and completeness of drug information in Wikipedia: a comparison with standard textbooks of pharmacology. PLoS One. 2014;9:e106930. 46. Maskalyk J. Modern medicine comes online: how putting Wikipedia articles through a medical journal’s traditional process can put free, reliable information into as many hands as possible. Open Med. 2014;8:e116-e119.

HOW INFORMATICS CAN HELP YOUR PRACTICE

47. Reilly T, Jackson W, Berger V, et al. Accuracy and completeness of drug information in Wikipedia medication monographs. J Am Pharm Assoc. Epub 2016 Nov 17.

54. Harpaz R, Vilar S, Dumouchel W, et al. Combing signals from spontaneous reports and electronic health records for detection of adverse drug reactions. J Am Med Inform Assoc. 2013;20:413-419.

48. Rosenbloom ST, Denny JC, Xu H, et al. Data from clinical notes: a perspective on the tension between structure and flexible documentation. J Am Med Inform Assoc. 2011;18:181-186.

55. Bot BM, Suver C, Neto EC, et al. The mPower study, Parkinson disease mobile data collected using ResearchKit. Sci Data. 2016;3:160011.

49. Yim W-W, Yetisgen M, Harris WP, et al. Natural language processing in oncology: a review. JAMA Oncol. 2016;2:797-804. 50. Middleton B, Bloomrosen M, Dente MA, et al; American Medical Informatics Association. Enhancing patient safety and quality of care by improving the usability of electronic health record systems: recommendations from AMIA. J Am Med Inform Assoc. 2013;20: e2-e8. 51. Office of the National Coordinator for Health Information Technology. Office-based physician electronic health record adoption. https:// dashboard.healthit.gov/quickstats/pages/physician-ehr-adoptiontrends.php. Accessed January 29, 2017. 52. Office of the National Coordinator for Health Information Technology. Non-federal acute care hospital electronic health record adoption. https://dashboard.healthit.gov/quickstats/pages/FIG-Hospital-EHRAdoption.php. Accessed January 29, 2017. 53. Jensen PB, Jensen LJ, Brunak S. Mining electronic health records: towards better research applications and clinical care. Nat Rev Genet. 2012;13:395-405.

56. Patient-Centered Outcomes Research Institute. Clinical data and patient-powered research networks—awarded projects. http://www. pcori.org/research-results/pcornet-national-patient-centered-clinicalresearch-network/clinical-data-and-0. Accessed January 29, 2017. 57. Multiple Myeloma Research Foundation. The MMRF CoMMunity Gateway. https://www.themmrf.org/living-with-multiple-myeloma/ community-gateway/. Accessed January 29, 2017. 58. National Institutes of Health. All of Us Research Program. https:// www.nih.gov/research-training/allofus-research-program. Accessed January 29, 2017. 59. U.S. Department of Health and Human Services. Individuals’ right under HIPAA to access their health information 45 CFR § 164.524. https://www.hhs.gov/hipaa/for-professionals/privacy/guidance/ access/index.html. Accessed January 29, 2017. 60. Office of the National Coordinator of Health Information Technology. 2015 edition final rule. https://www.healthit.gov/policy-researchersimplementers/2015-edition-final-rule. Accessed January 29, 2017. 61. Hardt D. The OAuth 2.0 authorization framework. https://tools.ietf. org/html/rfc6749. Accessed January 29, 2017.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 459

KLINE ET AL

The Oncology Care Model: Perspectives From the Centers for Medicare & Medicaid Services and Participating Oncology Practices in Academia and the Community Ron Kline, MD, Kerin Adelson, MD, Jeffrey J. Kirshner, MD, Larissa M. Strawbridge, MPH, Marsha Devita, RN, MS, Naralys Sinanis, MPH, Patrick H. Conway, MD, MSc, and Ethan Basch, MD OVERVIEW Cancer care delivery in the United States is often fragmented and inefficient, imposing substantial burdens on patients. Costs of cancer care are rising more rapidly than other specialties, with substantial regional differences in quality and cost. The Centers for Medicare & Medicaid Services (CMS) Innovation Center (CMMIS) recently launched the Oncology Care Model (OCM), which uses payment incentives and practice redesign requirements toward the goal of improving quality while controlling costs. As of March 2017, 190 practices were participating, with approximately 3,200 oncologists providing care for approximately 150,000 unique beneficiaries per year (approximately 20% of the Medicare Fee-for-Service population receiving chemotherapy for cancer). This article provides an overview of the program from the CMS perspective, as well as perspectives from two practices implementing OCM: an academic health system (Yale Cancer Center) and a community practice (Hematology Oncology Associates of Central New York). Requirements of OCM, as well as implementation successes, challenges, financial implications, impact on quality, and future visions, are provided from each perspective.

ncology is a complex and expensive medical specialty with costs rising faster than other medical specialties. The care is often fragmented and inefficient, imposing substantial burdens upon patients. Importantly, data show major differences in the cost of care in different regions of the United States without appreciable differences in outcome,1 thus identifying opportunities for improvement. For these reasons, the CMMI recognized oncology as an important specialty for a patient-focused model emphasizing care coordination and enhanced services and worked to create the OCM.2 As of March 2017, 190 practices are participating in OCM, with approximately 3,200 oncologists included in the model, providing care for an estimated 150,000 unique beneficiaries (and 190,000 episodes) per year, or approximately 20% of the Medicare Fee-for-Service (FFS) population receiving chemotherapy for the treatment of cancer. The goal of OCM is to use payment incentives and required practice redesign activities to transform oncology care in the United States so that it becomes universally high quality, high value, and patient focused. In addition to usual fee-for-service payments, OCM provides a $160 per beneficiary per month (Monthly Enhanced Oncology Services [MEOS]) payment to practices to support enhanced services for Medicare bene-

ficiaries receiving chemotherapy. A retrospective analysis is done on each 6-month episode of care to generate a performance-based payment (PBP) for practices that successfully reduce expenditures while providing high-quality care. In addition to the payment methodology that incentivizes high-value care, there are six required practice redesign activities intended to move practices toward coordinated, patient-focused care: (1) access to a provider on a 24/7 basis with access to the patient’s clinical record, (2) use of data for clinical quality improvement, (3) use of certified electronic health record (EHR) technology, (4) treatment of patients according to national guidelines, (5) provision of care navigation services, and (6) documentation of a care plan incorporating the 13 elements of the Institute of Medicine (IOM) care plan cited in the 2013 consensus report on cancer care.3

PERSPECTIVE FROM CMS

CMS appreciates the difficult work that practices throughout the country are undertaking to transform cancer care. Although even early objective analysis of the program’s impact to date is still several months away, we are gratified by the anecdotal reports of improvements in patient-centered care. These include improved care attributed to the man-

From the Centers for Medicare & Medicaid Services, Washington, DC; Smilow Cancer Hospital at Yale-New Haven, Yale Cancer Center, New Haven, CT; Hematology Oncology Associates of Central New York, East Syracuse, NY; Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Ethan Basch, MD, University of North Carolina, 170 Manning Dr., CB #7305, Chapel Hill, NC 27516; email: [emailprotected]. © 2017 American Society of Clinical Oncology

460 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PERSPECTIVES ON THE ONCOLOGY CARE MODEL

dated use of the IOM care plan and the creation of interdisciplinary teams formed to coordinate patient care. Our communications with participating practices often focus on the changes these practices have made to their care processes to improve quality and patient focus. In the design of OCM, such as the payment incentives and through the inclusion of other payers , our goal has always been whole-practice transformation, so we have been pleased to hear many practices report that their enhanced and newly coordinated services are offered to all of their patients, not just Medicare FFS beneficiaries. OCM is a model test intended to transform and improve the way oncology care is delivered in the United States. The model must work across diverse geographic regions, business models, and practice types. It must function within the existing frameworks of CMS claims and ICD-10 while providing complex care to patients with a diverse array of diseases and comorbidities. Given these challenges, the model is not static; it has already adapted to address early lessons learned, and it will evolve over time as problems are identified and solutions developed.

Early Lessons

Tracking OCM beneficiaries. To be eligible for MEOS payments for a 6-month episode of care, OCM beneficiaries must have a qualifying cancer diagnosis and a qualifying chemotherapy trigger. These beneficiaries must receive the enhanced services described above, including the initial completion of the IOM care plan, with an update to the care plan during subsequent episodes if applicable. These payments and care requirements direct practices to track beneficiaries with specific diagnoses receiving specific therapies, including the dates those therapies were received. This has

KEY POINTS • The OCM was recently launched by the CMS Innovation Center. • OCM uses payment incentives and practice redesign requirements toward the goal of improving quality while controlling costs. • As of March 2017, 190 practices are participating, with approximately 3,200 oncologists providing care for about 150,000 unique beneficiaries per year (approximately 20% of the Medicaid Fee-for-Service population receiving chemotherapy for cancer). • Key requirements for practices in OCM are to: (1) provide patients with 24/7 access to a clinician with real-time access to health records; (2) use of electronic health records certified by the Office of the National Coordinator for Health Information Technology; (3) use data for continuous quality improvement; (4) provide core functions of patient navigation; (5) document a care plan that contains the 13 components in the Institute of Medicine Care Management Plan; and (6) treat patients with therapies consistent with nationally recognized clinical practice guidelines.

required practices to put in place processes that track these data to identify when claims for MEOS payments should be filed, as well as to ensure that enhanced services have been provided to OCM beneficiaries (and that these activities have been documented). Particular attention has focused on tracking episodes for OCM beneficiaries (with Part D coverage) receiving only oral chemotherapy. The episode commences on the fill date of the chemotherapy (in association with a Part B cancer service in the previous 2 months). CMS cannot provide real-time Part D data to practices, though these data are available to practices on a several-month time lag as part of their quarterly feedback reports. To date, some practices with patients who do not fill their prescriptions in house have contacted pharmacies to obtain the fill dates of oral chemotherapy drugs for their OCM patients, though this is a manual process. We continue to work on identifying best practices and possible solutions to this challenge. IOM care plan. One of the practice redesign activities requires practices to document a care plan that includes the 13 elements recommended by the IOM consensus committee. These elements were identified as the foundations necessary to provide comprehensive, high-quality care to the oncology patient and promote shared decision making. There has been much discussion about one of the elements—patients’ out-of-pocket costs for cancer treatment—specifically how to estimate these costs. Although not traditionally an aspect of health care, increasing concerns about financial toxicity, especially in oncology, have made this an important issue. Practices are working diligently to understand not only the costs they generate specific to chemotherapy, but also costs generated from other aspects of oncology care such as radiation therapy, imaging, and laboratory diagnostics. Adoption of EHR standards. OCM requires the entry of anatomic staging and other clinically relevant data into its data registry (e.g., molecular mutations that enable the use of targeted therapies). These data will inform the creation of subsequent payment bundles that are narrower and more clinically focused. Collection of quality measurement data is necessary for the calculation of PBPs and for practice quality improvement. The ultimate goal for reporting data to OCM is that required data elements will be seamlessly exported from practice EHRs to the OCM data registry with minimal provider burden. CMS surveyed the EHR landscape and identified heterogeneity in capabilities, data capture fields, and electronic export standards. Several EHR vendors stated they were waiting for OCM to release such standards before building their EHRs to those specifications. CMS therefore identified the Health Level Seven (HL-7) standard for export, referred to as “Reporting to Public Health Cancer Registries from Ambulatory Healthcare Providers,” to support submission of staging and clinical data. Additionally, we aligned our quality measures with nationally validated measures and existing registry reporting programs wherever possible. Given asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 461

KLINE ET AL

the spectrum of both practice and EHR capabilities, and the variety of existing registries, there have been some early difficulties. Accordingly, CMS decided to reduce practice reporting requirements in the first year of the model to allow time for continued practice process improvement and for EHR capabilities to further align with OCM requirements. Work continues with the data registry contractor and EHR vendors to make the data registry more user friendly and to improve the automated data export process. Bladder and prostate cancer care. Target prices for broad cancer bundles inherently include low-cost patients for whom the cost of treatment is lower than the target price and high-cost patients for whom the cost of treatment is higher than the target price. In OCM, the target price is based on the average costs of all patients in each bundle in the historical baseline period adjusted by each practice’s baseline experience. In a practice that treats a random distribution of all cancer stages and molecular subtypes, this methodology is appropriate. When separate practices consistently treats patients of different stages, then this methodology may not be appropriate. CMS noted that, in general, urologists cared for a greater proportion of patients with low-risk bladder and prostate cancer, whereas medical oncologists cared for a greater proportion of high-risk patients. To ensure equity in the model, CMS created separate target prices for high- and low-risk bladder and prostate cancer for episodes beginning after July 1, 2017. CMS identified drugs typically used in the treatment of these different stages of cancer to generate separate target prices.

Future Directions

In the first year of OCM, participating practices have invested considerable energy and resources implementing the model, and CMS has made adaptations where necessary to respond to identified problems. We view the model test as an opportunity to learn about how care and health outcomes can be improved for Medicare beneficiaries with cancer who receive chemotherapy in diverse practice environments. As noted above, one of the limitations of OCM, as currently designed, are its broad clinical bundles, because anatomic staging and relevant molecular markers are not a part of existing Medicare FFS claims data. By collecting detailed staging and molecular information in the data registry, we plan to link these data with claims to design more clinically refined payment bundles for different stages and molecular subtypes of cancer where meaningful cost and outcome variations exist. Part of this process will involve remaining current in clinical oncology so that molecular mutations with new targeted therapies are incorporated into the data registry as quickly as possible. OCM also has a robust learning and diffusion component incorporated into the model. Among other activities, such as OCM’s online collaboration platform, our webinars will highlight practices that develop successful approaches to practice transformation so that others may benefit from this 462 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

innovative work. In addition, CMS has launched a palliative care affinity group, and future affinity groups will focus on topics such as using data for quality improvement to allow practices with specific interests or needs to learn from one another. CMS looks forward to engaging with OCM practices and other stakeholders during the remaining 4 years of OCM to ensure that this is a successful model test that will improve the quality of cancer care in the United States.

ACADEMIC HEALTH SYSTEM OCM PARTICIPANT PERSPECTIVE (YALE CANCER CENTER)

Smilow Cancer Hospital at Yale New Haven Hospital is the clinical facility of the Yale Cancer Center. Today, we deliver care to one in four patients with cancer in the state of Connecticut. We have 10 community medical oncology and hematology practices and an academic main campus where care is delivered in multispecialty disease teams. We serve as the largest academic referral center in the state and care for the largest proportion of uninsured and underinsured patients.

Rationale for Joining OCM

The transition toward value-based care presents different challenges for a large health system compared with ambulatory oncology practices. Although reducing hospitalizations and emergency department visits represent an opportunity for savings for payers and society at large, for a health system, this savings represents a loss of revenue. For the Smilow Cancer Hospital, OCM served as a catalyst to move toward value-based payment. OCM’s MEOS payments would fund clinical infrastructure that would improve oncology care, and the potential for PBPs would offset potential losses in revenue.4 In a best-case scenario, OCM would allow us to transform how we care for patients through implementation of new programs in care management, oncology urgent care, implementation of clinical pathways, and expansion of palliative care into the ambulatory setting, while allowing us to earn PBPs for reduced cost and higherquality care. In a worst-case scenario, OCM would allow us to build this essential clinical infrastructure and gain experience with value-based payment, even if we did not achieve savings or PBPs. With either scenario, Smilow Cancer Hospital leadership believed OCM would enhance the quality of care while providing early experience with an alternative payment model. Finally, as a National Cancer Institute–designated comprehensive cancer center, our mission is to improve outcomes for patients with cancer; to that end, we must participate in, learn from, and help inform new, valueoriented models for cancer care delivery. Academic centers must have a voice in national conversation that will ultimately redefine quality cancer care and inform the restructuring of our national payment system. OCM gave us this opportunity.

PERSPECTIVES ON THE ONCOLOGY CARE MODEL

Steps to Prepare for and Implement OCM

Our clinical transformation and cost-saving strategy is focused on keeping patients out of the hospital by providing care management while patients are home, expanding access to urgent visits and symptom management services, and integrating palliative care earlier in the disease process. In many ways, this has meant creating the clinical infrastructure to function as an oncology medical home. Achieving transformation in care delivery requires uniting multiple stakeholders, including the clinical arm of the school of medicine (Yale Medicine), which employs the physicians and is responsible for MEOS billing, and the hospital (Yale New Haven Health), which is funding most of the infrastructure. We created an OCM executive committee to serve as the decision-making and funding body of the program. We then organized our work into six thematic projects: 1. Patient identification and MEOS billing: We built a team that included an Epic report writer, lead pharmacist, lead physician, program manager, and billing representative to translate the detailed patient eligibility criteria into ongoing patient eligibility reports. After multiple iterations, this final patient list was translated into EHR flags and then into work queues for care management, financial counseling, and billing. 2. IOM care plan: We worked with our Epic team to centralize the 13 IOM care plan elements into one document. We made a deliberate decision not to burden our physicians and advanced practice providers with additional documentation demands. Instead, we required providers to enter patients’ stage and treatment goals when ordering chemotherapy (curative vs. noncurative intent). With this documentation in place, our nurse care managers could fill out the care plan. 3. Open an oncology extended care clinic: We developed a business plan to build and staff a new extended care clinic that would be open 16 hours a day, 7 days a week. This center should open in Spring 2017. 4. Launch a care management program: The goal of this program is to improve contact with patients when they are home and identify and stabilize early symptom exacerbations before they lead to hospitalizations. We have hired four out of a total of eight OCM care managers. 5. Integrate clinical pathways into practice: We have committed to use Via Oncology clinical pathways with the goal of reducing unnecessary variation and reducing the use of high-cost drugs in situations where they do not improve efficacy. 6. Quality and registry reporting: We partnered with our tumor registrars and data analysts to define registry requirements. We created hard stops in our EHR to ensure that required documentation would be accessible in structured fields. Our tumor registry has begun abstracting in real-time, a radical change to their workflow.

Successes and Challenges

Participation in OCM requires time-intensive resources from across the organization. There is a constant tension between working to meet the reporting requirements and meaningfully transforming care. Although checklists and EHR tools may help in an audit or improve chance of PBPs, they are unlikely to change patterns of care or reduce cost. Although we are behind on completing each component of the IOM care plan for our more than 3,000 eligible patients, we have made real strides in building the infrastructure we believe will ultimately achieve clinical transformation. Timeline challenges. Due to the complexity of eligibility requirements described below, it took more than 4 months to finalize our initial patient list; initiation of downstream services (financial counseling, IOM care plan completion, care management) and MEOS billing was delayed until this process was complete. Barriers with patient identification and MEOS. The patient identification process was rigorous and time intensive and required an iterative report build. Because patients taking oral drugs often received multiple refills when first prescribed, we could not rely on a new prescription to trigger enrollment and instead created a candidate list of patients who received oral prescriptions in the last year. Our pharmacists manually checked disparate data systems for Medicare Part D status and prescription fill verification for thousands of patients. This resource and time-consuming process continues today. There is a critical need for CMS to make this information easily accessible to OCM sites. We had challenges in the MEOS billing and payment process that have delayed revenue earmarked to support new clinical infrastructure. Because our hospital committed to OCM participation, we have moved forward with clinical program building despite the delayed revenue. Achieving resolution of the billing issues has been slow and labor and time intensive. Going forward, it would be helpful if there were real-time problem resolution at the OCM and CMS support lines. IOM care plan challenges. Epic did not provide us with an out-of-the-box solution for IOM care plan. Our internal discussions have revolved around whether we should meet the program requirements by creating check boxes—such as, “I closed the referral loop,” or “Treatment benefits and harms discussed with the patient”—or whether we should focus on the spirit of the program and use the care plan to facilitate meaningful discussions with patients. We have chosen the latter and believe that, in the long run, this will facilitate better prognostic understanding and influence downstream health care utilization. In the meantime, we are challenged with completing these care plans and sharing them with more than 3,000 patients. Reporting. Reporting processes are proving to be more time intensive and manual than we had hoped. For example, classifying patients as having “very high–risk” or “highrisk” prostate cancer is challenging because the data does not exist in structured format in either our EHR or in the tumor registry. Initially, CMS reporting timeframes required asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 463

KLINE ET AL

that our tumor registrar begin concurrent abstraction, a dramatic change in their workflow. However, in response to concerns from participating sites, CMS has revised reporting requirements, which has been appreciated by our tumor registrars.

Future Impact on Practice

Like other academic health centers with a strong research mission, we are challenged with balancing our role as a destination hospital for patients seeking the latest treatment options while ensuring that we elicit their true preferences, provide realistic expectations for treatment benefit, and support their quality of life.1 We believe that OCM will serve as the catalyst to shift care from an inpatient to an outpatient setting.

Financial Feasibility

Under OCM, health systems face revenue loss from reduced inpatient services; our finance team studied the impact that success in the program would have on revenue from Medicare and private payers who would also benefit from the clinical infrastructure we sought to build. Although achieving the 4% reduction in costs required to achieve PBPs would impact the contribution margin, we found the effects would be tolerable over time.4 We have not formally modeled how we will fair with PBPs, which depend first on achieving a greater than 4% savings and then on performing well compared with the national average on multiple quality metrics. However, our financial justification for participation in OCM relied entirely on MEOS revenue and modeling of the impact that care transformation would have on our contribution margin. Thus, even without guarantee of PBPs, we felt the program was sustainable.

Impact on Quality

We believe that participation in OCM should improve clinical quality through better care coordination, access to urgent care services, reduction in variability of chemotherapy choice, and earlier integration of palliative care. Furthermore, OCM will ensure ongoing access to total cost of care claims data, which will allow us to provide physicians, disease teams, and community practices with detailed feedback on patterns of care, including hospital admission rates, emergency department utilization, intensive care unit use, chemotherapy near the end of life, and timeliness of hospice. Participating in OCM has made investments in clinical infrastructure possible that were not feasible before. For 3 years, we attempted to create a workable business model for an oncology extended care clinic. Each time, the model incorporated loss of inpatient revenue and could not be financially justified. Similarly, we wanted to implement clinical pathways to diminish variation in care but had no riskbased contracts, and thus there was no financial incentive to warrant it. In the context of OCM, MEOS revenue could offset these infrastructure costs, and the potential to earn back savings as PBPs could offset some revenue losses. 464 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

COMMUNITY PRACTICE OCM PARTICIPANT PERSPECTIVE (HEMATOLOGY ONCOLOGY ASSOCIATES OF CENTRAL NEW YORK)

Hematology Oncology Associates of Central New York is a hematology/oncology practice comprised of 14 medical oncologists, three radiation oncologists, 20 midlevel providers (17 nurse practitioners and three physician assistants), and a total of 280 employees. Our main office is in East Syracuse, New York, with satellite offices in Onondaga Hill-Syracuse and Auburn. The catchment area is approximately one million. There is an infusion center at all three locations, and two sites have radiation oncology. The great majority of chemotherapy is administered in our offices, although we do have admitting privileges at three local hospitals. We have an outpatient pharmacy at our main office to provide and monitor oral oncolytic agents. We actively participate in clinical research and are a main member of the Alliance for Clinical Trials in Oncology, one of the major National Cancer Institute–sponsored cooperative groups.

Rationale for Joining OCM

Ensuring high quality of care for our patients has always been a high priority for our practice. We are Quality Oncology Practice Initiative–certified through ASCO and are one of nine Oncology Medical Homes certified by the American College of Surgeons Commission on Cancer. We have successfully participated in the CMS Meaningful Use program and continue to report quality data through the Physicians Quality Reporting System. We decided to apply for participation in OCM for many reasons. The bottom line is that we believe that participation will help us provide better care to our patients. Our practice always strives to be progressive and up to date. We truly believe OCM is a better payment model because quality is incorporated rather than just fee-for-service. We also see participation as a way to prepare for the future.

Steps to Prepare for and Implement OCM

We began preparing for OCM in early 2015 when we hired a quality coordinator to help with the Physicians Quality Reporting System; a year later, we hired an incentive coordinator. Our EHR is regularly updated to meet quality reporting. Our chief clinical officer oversees the entire program and reports to our chief executive officer and board of directors, which is comprised of our physician partners. We also have created a quality care committee with representation from multiple departments. We were fortunate to be accepted in the OCM program initiated in July 2016. As detailed in our application, we have completed the practice transformation plan as required by CMS: 1. Provide and attest to 24/7 patient access to an appropriate clinician who has real-time access to the practice’s medical records. 2. Attest to the use of ONC-certified EHRs. 3. Use data for continuous quality improvement. 4. Provide core functions of patient navigation.

PERSPECTIVES ON THE ONCOLOGY CARE MODEL

5. Document a care plan that contains the 13 components in the IOM Care Management Plan. 6. Treat patients with therapies consistent with nationally recognized clinical guidelines. Prior to the start date of the OCM program, we had extensive training for our entire staff, including the physicians. Our EHR was updated to include OCM reporting requirements, and the health care providers had to become proficient in incorporating these changes. For example, chemotherapy could no longer be ordered without answering four questions on a dropdown bar that popped up on the screen: prognosis, goals, expected response, and advanced care plan. Pain had to be graded on a scale of 1 to 10, with a treatment plan entered. There is a tab for referral to our survivorship program. Eligible OCM patients are identified in a number of ways, including review of health records and pharmacy ordering of chemotherapy. There have been initial challenges in identifying patients who were already receiving treatment. Patients receiving oral agents are more difficult to identify, but the EHR is monitored regularly by a dedicated information technology individual. Once eligible patients are identified and entered into the OCM program, billing and financial services are promptly notified. A dedicated financial services advocate contacts the patient on the telephone and/or in person to explain the program and distribute the required notification from CMS.

Successes and Challenges

Internal monitoring of individual provider performance is performed regularly, and, for the most part, each individual has exceeded 90% compliance. Patients report over 90% satisfaction in monthly surveys. Our first report to CMS was in February 2017. Five measures were reported, and these are our results from July 1 to December 31, 2016: 1. Prostate cancer (adjuvant hormonal therapy for highrisk or very high–risk disease): None of these patients were seen during the reporting period. 2. Adjuvant chemotherapy is recommended or admin istered within 4 months of diagnosis to patients under age 80 with stage III colon cancer: 100% compliance (35 patients). 3. Combination chemotherapy is recommended or administered within 4 months of diagnosis for women under age 70 with stage IB-III hormone receptor–negative breast cancer: 100% compliance (27 patients). 4. Trastuzumab administered to patients with stage I (T1c)-III and HER2-positive breast cancer who receive adjuvant chemotherapy: 100% compliance (34 patients). 5. Hormonal therapy for stage I-III estrogen receptor/ progesterone receptor–positive breast cancer: 100% compliance (more than 800 patients).

Financial Feasibility

Enrollment in the OCM program has ranged from 311 to 755 patients. Net revenue for the first 6 months of the program

was $459,958. Expenses are estimated by multiplying the salaries and benefits of the dedicated employees by the percentage of time devoted to OCM activities. Annualized expenses amount to $616,317. There are many other expenses that are more difficult to quantitate, including the many additional hours of work provided by our clinical staff and those who work in the financial services and information technology departments.

Future Impact on Practice

We will be changing to a new EHR, OncoEMR, and are working with the engineers to insure incorporation of OCM parameters and requirements into the EHR. We feel that the transition from Mosaiq should be fairly straightforward and seamless. This new EHR will be more user friendly and easier to stage new patients and document care plans, meeting the documentation requirements of CMS. We will be adding two new medical oncologists who will be trained in OCM and EHR use. Our quality committee will continue to monitor our programs. We will update and add in-house clinical pathways consistent with national guidelines.

Impact on Quality

To date, we have been very pleased with our participation in OCM. The amount of work to implement the program has been substantial, but doable, largely as a result of assembling a dedicated and competent team of individuals. They are well prepared and have been learning on the job. We believe that our first year has been a success in terms of maintaining and improving quality. The program appears to be financially viable, and although it is difficult to quantitate, we may realize a profit. In terms of the ultimate goals of CMS, we have indeed demonstrated an improvement in record keeping and compliance with the stated requirements of OCM, which should lead to better quality of care for our patients and overall less expenditure of health care dollars, as hospital admissions and emergency department visits will decrease. Alternative payment models appear to be here to stay, and we plan on continuing our participation in OCM and future programs as they become available.

CONCLUSION

OCM provides a path to improving care quality and controlling costs of care in the United States through a partnership between CMS and practices built on the backbone of the current system for reimbursem*nt and care delivery. OCM has prompted practices to enact patient-centered delivery approaches focused on quality that are intended to improve the patient experience with care as well as measurable outcomes. This program is in its initial phase of implementation, and the ongoing experience of CMS and participating practices will provide further insights about feasibility of various aspects of the model, financial feasibility, impact on outcomes, and sustainability. CMS will be monitoring progress of OCM and a host of metrics at participating and comparator sites. Despite initial challenges at sites to implement various aspects of the model, the asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 465

KLINE ET AL

aspiration of this program is to provide insights toward future approaches that optimize resources, quality, and patient centeredness in cancer care delivery.

ACKNOWLEDGMENT

The views expressed in the manuscript represent the authors and not necessarily the views or policies of CMS.

References 1. Brooks GA, Li L, Sharma DB, et al. Regional variation in spending and survival for older adults with advanced cancer. J Natl Cancer Inst. 2013;105:634-642.

3. IOM (Institute of Medicine). Delivering high-quality cancer care: Charting a new course for a system in crisis. Washington, DC: The National Academies Press; 2013.

2. Kline RM, Bazell C, Smith E, et al. CMS—using an episode-based payment model to improve oncology care. J Oncol Pract. 2015;11:114116.

4. Adelson KB, Velji S, Patel K, et al. Preparing for value-based payment: a stepwise approach for cancer centers. J Oncol Pract. 2016;12:e924-e932.

466 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

HEMATOLOGIC MALIGNANCIES—LEUKEMIA, MYELODYSPLASTIC SYNDROMES, AND ALLOTRANSPLANT

KHOURY ET AL

Chronic Myeloid Leukemia: What Every Practitioner Needs to Know in 2017 Hanna Jean Khoury, MD, FACP, Loretta A. Williams, PhD, APRN, AOCN, Ehab Atallah, MD, and Rüdiger Hehlmann, MD, PhD OVERVIEW The prognosis of chronic phase chronic myeloid leukemia (CML) has improved so that life expectancy for patients responding to tyrosine kinase inhibitors (TKIs) is now equivalent to age-matched controls. Attention should be paid to comorbidities that impact survival. The success of TKI therapy can be easily and reliably assessed at well-accepted time points using quantitative polymerase chain reaction (PCR) standardized to the international scale. Patient-reported outcome (PRO) tools are readily available for use in the clinic and provide complementary information on the tolerance of TKIs. Effectively managing adverse events of TKIs can improve compliance and quality of life. Discontinuation of TKIs is the next frontier in CML. In select patients with sustained deep molecular remission, a discontinuation of TKI is associated with a durable treatment-free remission in approximately 50%. Patient engagement in their discontinuation can be achieved through a provider multi-team coaching, is complementary to the available guidelines, and may provide an additional safety net so that these discontinuations remain safe when applied in general practices.

nderstanding the molecular pathophysiology of BCRABL undeniably led to a scientific breakthrough that completely transformed the landscape of CML. Indeed, in 2017, we are witnessing the dramatic effects of these discoveries with five very effective commercially available TKIs, a life expectancy that is comparable to age-matched population, and prolonged treatment-free remission in some patients, so one may even consider using the word “cure.”1 These impressive successes are broadly achievable when management is based on a set of commonly accepted treatment and monitoring guidelines—European LeukemiaNet (ELN)/National Comprehensive Center Network (NCCN). This article reviews key variables that impact outcomes of CML, provides patients’ perspective on TKI tolerance, and discusses THE potential role of TKI discontinuation in daily practice.

is sophisticated but also robust enough to be standardized and used by every hematologist, provided the necessary infrastructure is available.3 A standardized and well-equipped laboratory with a turnaround time of no more than 14 days is necessary, in addition to a care facility that can reliably and regularly follow patients.4

MANAGEMENT OF CML IN PRACTICE: VARIABLES TO CONSIDER

Table 2 summarizes long-term survival of patients with chronic phase CML treated with imatinib on clinical trials.6 The 10-year survival data are now available for imatinib, and 5-year data for the second-generation TKIs (2G-TKIs) dasatinib and nilotinib are also available.7,8 Excellent 10- year molecular response (MR) rates are achieved with first-line imatinib: 92% for MR2 (corresponding to complete cytogenetic remission), 89% for MR3 (or major MR [MMR]), 81% for MR4 (reduction of residual BCR-ABL transcripts by ≥ 4 logs), 72% for MR4.5 (≥ 4.5 log reduction), and 59% for MR5

Current Management Needs

Even with such an excellent prognosis that cure appears to be a realistic perspective,1 more patients are currently dying of comorbidities than of their CML.2 Thus, the skills of a general internist in addition to the expertise of the CML hematologist are needed now more than ever to effectively treat these patients. Careful monitoring of CML response is key to the success of therapy. The monitoring technology

Initial Testing and Monitoring

Current requirements for initial testing and follow-up are listed in Table 1. Figure 1A shows the correlation between cytogenetic and molecular findings with leukemic cell mass and response milestones as defined by ELN5 and NCCN. The international scale was designed to make results comparable between different laboratories as illustrated in Fig. 1B.

Outcomes of Chronic Phase CML Following Treatment With Imatinib

From the Winship Cancer Institute, Emory University, Atlanta, GA; The University of Texas MD Anderson Cancer Center, Houston, TX; Medizinische Fakultät Mannheim der Universität Heidelberg, Mannheim, Germany. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Hanna Jean Khoury, MD, FACP, Winship Cancer Institute, Emory University, 1365 Clifton Rd., Atlanta, GA 30322; email: [emailprotected]. © 2017 American Society of Clinical Oncology

468 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

WHAT EVERY PRACTITIONER NEEDS TO KNOW ABOUT CML

TABLE 1. Evaluation of Chronic Myeloid Leukemia at Diagnosis and During Follow-up Evaluations at Diagnosis

Follow-up

Record spleen size

Spleen size

Complete blood count (basophils, eosinophils) and basic metabolic profile

Complete blood count and basic metabolic profile

Risk score (Sokal, Euro, EUTOS, or ELTS) Bone marrow aspirate (blasts, karyotype)

Marrow karyotype (12 months)

Baseline molecular genetics (quantitative PCR for BCR-ABL) with transcript type

Molecular monitoring by quantitative PCR (every 3 months until MMR, then every 6 months)

Abbreviations: EUTOS, European Treatment and Outcome Study; ELTS, EUTOS Long-Term Survival score; PCR, polymerase chain reaction.

(≥ 5 log reduction).9 Progression to blast crisis is 5% to 7% by 8 to 10 years in the randomized trial comparing imatinib with interferon/cytarabine (IRIS trial) and in the CML-study IV,10-12 and relative survival compared with the matched general population was 92% in the CML-study IV, 96% across clinical trials in Europe (Fig. 2),12 and around 90% in Dutch and Swedish population-based registries.13,14

First-Line Treatment Options

First-line treatment options include imatinib at 400 mg or higher dosages (600–800 mg), dasatinib at 100 mg, and nilotinib at 300 mg twice a day.5 Although the secondgeneration TKIs dasatinib and nilotinib achieve response milestones earlier and faster than imatinib, no survival advantage has yet been reported for any TKI. In contrast to imatinib, 2G-TKIs have been associated with serious and potentially fatal adverse events. In choosing the most suitable front-line treatment in chronic phase CML, the following variables have to be considered: 1. CML risk score: Patients with high-risk disease must be monitored more closely. 2. Cytogenetics: Additional chromosomal aberrations (ACA), in particular so-called major-route ACA (+8, +Ph, i(17)(q10), +19), indicate accelerated phase and are associated with a poorer prognosis.22,23 3. Comorbidities: Patients with pre-existing vascular disease should not receive nilotinib, and patients with

lung disease or a history of pleural effusion should not receive dasatinib. Comorbidities have been identified as the major cause of death for patients with CML in the TKI era.2,12 They have no or little impact on progression of CML but determine survival particularly in patients with good-risk disease on stable first-line therapy. 4. Costs: Generic imatinib is now generally available at reasonable costs. In the face of the favorable efficacy and safety profile of imatinib and the increasing

FIGURE 1. Correlation of Cytogenetic and Molecular Data With Leukemia Cell Mass and Response Milestones (A) and Comparability of Molecular Data Between Laboratories by Calculation of Conversion Factors (B)

KEY POINTS • Current treatment of CML—if done right—results in normal, or near normal, life expectancy. • Treatment choice should consider patients’ comorbidities, adverse events profile of drugs, and patients’ preference. • PRO tools are available for use in clinical practice, can easily be completed at the time of a clinic visit, can alert clinicians to specific areas of concern, and can help clinicians follow symptom trends over time. • PROs can help clinicians identify and better manage side effects of TKIs, which may lead to better adherence to therapy and improved clinical outcomes. • TKI discontinuation, if done according to guidelines and in select patients, is safe and associated with a treatment-free remission of 40% to 50%.

Abbreviations: ELN, European LeukemiaNet; CCR, complete cytogenetic response; MMR, major molecular response.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 469

KHOURY ET AL

FIGURE 2. Relative and Overall Survival in Clinical Trials

Abbreviations: CML, chronic myeloid leukemia; IFN, interferon; SCT, stem cell transplantation; HU, hydroxyurea. Courtesy of Dr. Pfirrmann.12

Therefore, experts in the field consider it appropriate to start with imatinib and to switch to 2G-TKIs only in the case of intolerance or if response milestones are not reached.

prevalence of CML as a consequence of successful therapy (Fig. 3), generic imatinib represents a rational first-line treatment option for most patients with CML.24

TABLE 2. Long-term Survival of Patients With Chronic Phase Chronic Myeloid Leukemia Following Treatment With Imatinib on Clinical Trials Study

IM-Dose mg

No. of Patients

Age at Diagnosis, Median, Years

5-Year Survival %

10-Year Survival %

Median Observation Time, Years

CML-IV15

IM 400–800

1,536

9.5

IRIS10

IM 400

553

85 (8 years)

GIMEMA

IM 400–800

559

Hammersmith17

IM 400

204

46.3

3.2

PETHEMA

IM 400

210

97.5

4.2

IM 400

157

94 (4 years)

93.4 (4 years)

TOPS19 MDACC20 ILTE21 (CCR only) ENESTnd8

Dasision7

IM 800

319

IM 400

M 800

201

IM NR

832

51a

98 (6 years)

IM 400

283

Nilo 600

282

8.3

Nilo 800

281

IM 400

260

Dasa 100

259

Median (estimate) Abbreviations: IM, imatinib; NA, not applicable; NR, not reported; yr, year; min, minimum; CCR, complete cytogenetic response.

470 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

3.5 3.5

9.9

(min 8)

95 (8 years)

5.8

WHAT EVERY PRACTITIONER NEEDS TO KNOW ABOUT CML

Second-Line Treatment

Prior to any TKI switch for secondary treatment decisions, the following variables must be considered: 1. Adherence to the prescribed drug: Poor adherence has been determined to be the most frequent cause of suboptimal response. In a review by Noens and colleagues,25 varying degrees of nonadherence were reported in clinical trials, and nonadherence was associated with poorer event-free survival. A coordinated team approach might help to overcome problems with adherence.26 2. Resistance mutations dictating TKI choice: It is helpful to obtain an ABL kinase domain mutation at the time of resistance to a TKI and prior to switching to have a rational basis for choosing the appropriate TKI. 3. Clonal evolution: ACA newly arising under therapy have been defined as a sign of resistance indicating a need for a change of therapy.5 As a rule, stem cell transplantation is considered the treatment of choice in this situation if a donor is available and the patient is suitable for transplantation.27 4. Intolerance: Adverse events and comorbidities predisposing adverse events must be considered when choosing the new TKI (e.g., avoid dasatinib in patients with pulmonary conditions, avoid nilotinib in patients with a history of vascular events, caution with nilotinib in the presence of liver disease or diabetes mellitus).

Safety of TKIs

Adverse drug reactions to imatinib are frequent but generally mild.9 Most adverse drug reactions appear early and are reversible. No serious late toxicities have surfaced up to now with imatinib. Sequential analyses of glomerular filtration rates in patients receiving treatment with imatinib

have indicated a reduction of filtration rates in 6% to 8% of patients, particularly in patients with preexisting renal failure.28 These patients can be candidates for switching to nilotinib, as nilotinib has been reported beneficial for renal function.29 Published evidence shows that approximately 2.5% of patients treated with nilotinib experience serious vascular events, which become more frequent over time.8 Nilotinib should be avoided or used with caution in patients with vascular risk factors. Dasatinib is frequently associated with mostly benign pleural effusion and may rarely be associated with potentially fatal pulmonary arterial hypertension. Bosutinib is associated with transient diarrhea that usually resolves in the first week.30 Lastly, ponatinib is associated with cardiovascular side effects in approximately 25% of patients.31 An overview of adverse TKI reactions is provided in Table 3.

Impact of Health Care Setting

Health care infrastructure may affect quality of care and survival. Survival of patients managed at academic centers has been reported superior to office-based management or management at municipal hospitals.36 A study on frequency of molecular monitoring in Europe and the United States has reported serious deficits, pointing to the need for standardized laboratories and better education of doctors and their patients with CML.37

HOW WELL ARE TKIS TOLERATED? THE PATIENT PERSPECTIVE

Patient-Reported Outcomes

Capturing patient perspectives of disease and treatment in a valid, reliable, and reproducible way is the goal of PRO measures.38,39 PROs most often are assessed as questionnaires that measure concepts such as health-related quality of life (HRQoL), health status, symptoms, functional abilities,

FIGURE 3. Increase in Prevalence of CML in the TKI Era as Determined for Three Incidences (1, 1.5, or 2 per 100,000 per Year)

Abbreviations: CML, chronic myeloid leukemia; TKI, tyrosine kinase inhibitor.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 471

KHOURY ET AL

TABLE 3. Overview of Adverse Drug Reactions Associated With Tyrosine Kinase Inhibitors Approved for Chronic Myeloid Leukemia Imatinib9

Nilotinib32

Dasatinib33

Bosutinib34

Ponatinib35

Myelosuppression

+++

Fluid Retention

—

+++

—

Rash

—

Diarrhea

+++

Increased Glucose/Cholesterol

—

Vascular Occlusion

—

+++

Renal Insufficiency

—

(+)

treatment adherence, and satisfaction with treatment.38 Table 4 summarizes commonly used PRO measures. Routine assessment with PROs can provide valuable information about patients’ experiences of their condition to clinicians and researchers in much the same way that laboratory tests, scans, and physical findings do.40 For clinicians, PRO assessments can be useful in improving communication with patients, making diagnoses, deciding on treatments, assessing treatment response, managing treatment toxicities to improve tolerability and compliance, and identifying targets to improve patient HRQoL.40,41 PROs can also be useful in clinical research, in assessing quality of care, and in monitoring safety and treatment outcomes in clinical effectiveness and postmarketing registry studies.42-44 The purpose of a PRO is to capture patients’ evaluation of their experience, so by definition, PROs are subjective reports. However, the science of instrument development and psychometrics has progressed to the point that well-developed PROs can be trusted as valid and reliable measures when used for their intended purpose in their intended populations.45 Studies in recent years have shown that PROs may be more accurate and complete measures of patient experience than clinician report and can provide complementary information to clinician assessments.46

Patient and Health Care Provider Perceptions

Differences between patient and health care provider perceptions of symptoms and HRQoL issues have been found

in CML.47,48 As shown in Fig. 4, patients rated the relevance of symptom issues higher than clinicians, whereas clinicians rated psychosocial issues as having more relevance than did patients.47 Similar results were reported in another study that enrolled 422 pairs of patients with CML receiving imatinib and their treating physicians who assessed the severity of nine imatinib-specific symptoms. Patients rated individual symptoms as more severe more frequently than did their physicians. Agreement on symptom severity ranged from 34% for muscular cramps to 66% for nausea. Fifty-one percent of patients rated fatigue higher than their physicians, whereas 10% of physicians rated fatigue as more severe than their patients. Physicians rated patient health status better than the patients 67% of the time and the same as the patients 26% of the time.48

Clinical Use of PROs

There are easy and reliable ways to capture and use PROs in the clinic. Selecting the correct PRO for clinical use is important. The health care provider should consider what information will be the most useful at a clinical visit, how easy the PRO will be for patients to understand and complete in a reasonable amount of time, and how easy the PRO will be for the clinician to review and interpret quickly. PROs that assess symptoms and functional abilities are often the most useful in clinical settings as these are simple concepts most directly related to a patient’s physical condition (Fig. 5).49

TABLE 4. Commonly Used Patient-Reported Outcomes Measures Type of PRO

Common PRO

Health-related quality of life

EORTC Quality of Life Questionnaire (QLQ)-C30 Functional Assessment of Cancer Therapy (FACT)

Health status

Euro Quality of Life (QoL)-5D (EQ-5D) Medical Outcomes Study Short Form-12 or 36 (MOS SF-12 or 36)

Symptoms

MD Anderson Symptom Inventory (MDASI) Memorial Symptom Assessment Scale (MSAS)

Functional abilities

Work Productivity and Activity Impairment Questionnaire (WPAI)

Treatment adherence

Morisky Adherence Scale (MAS)

Satisfaction with treatment

Cancer Therapy Satisfaction Questionnaire (CTSQ)

Abbreviations: PRO, patient-reported outcomes; EORTC, European Organisation for Research and Treatment of Cancer.

472 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

WHAT EVERY PRACTITIONER NEEDS TO KNOW ABOUT CML

FIGURE 4. Differences in Patients With CML and Health Care Provider Valuations of Aspects of HRQoL47

Abbreviations: CML, chronic myeloid leukemia; HRQoL, health-related quality of life; HCP, health care provider.

More complex concepts such as HRQoL require longer questionnaires, often have complicated scoring algorithms, and are better suited for research than for clinical use. However, to give clinicians an overall impression of patients’ HRQoL, a single item can be valid and reliable.50 The traditional method of PRO administration is paper and pencil. Although this is still an easy, effective, and cost-efficient method, newer technologies allow PROs to be completed electronically.51 Computer kiosks or electronic tablets allow PRO completion in waiting area. Other methods, such as automated phone systems, mobile apps/e-diaries, and electronic medical record patient portals, allow patients to complete PROs at home prior to or in between clinic visits. Electronic applications can remind patients an assessment is due, allow direct transfer of the PRO data to the electronic medical record, provide automated scoring and reporting of results in clinician-friendly displays, and send the clinician alerts when results require follow-up.51

PROs for CML

Currently there are two PROs developed specifically for use by patients with CML: the MD Anderson Symptom Inventory for CML (MDASI-CML),52 which measures the symptom burden of CML, and the European Organization for Research and Treatment of Cancer (EORTC) QLQ-CML24, which measures the HRQoL of patients with CML.53 More general PROs can also be used in patients with CML, but they may be less sensitive to patient concerns and changes in patients’ conditions. Although the use of PROs in hematologic malignancy

FIGURE 5. Model of HRQoL

research has lagged solid tumor research,54 PROs have been incorporated into clinical trials and have provided insight into the tolerability of treatments for CML, including TKIs.55 Studies have shown that most patients tolerate TKI therapy with few symptomatic effects or negative impact on HRQoL,52,56 and some may have an improvement in some aspects of HRQoL.57,58 However, a minority of patients will experience symptoms, such as fatigue and muscle soreness or cramping, at moderate to severe levels that impact HRQoL, and these symptoms may persist indefinitely. In one study, 25% to 30% of patients treated with imatinib reported severe fatigue, edema, musculoskeletal pain, and muscle cramps, with women being more affected than men.56 Younger patients reported substantial role impairments because of physical and emotional problems, compared with matched normal controls.56 In a year-long longitudinal study that assessed the symptom burden of CML using the MDASI-CML,52 the highest severity symptoms were fatigue, muscle soreness and cramping, drowsiness, disturbed sleep, and trouble remembering things. Although these symptoms were mild (< 4 on a 0–10 scale),52 trajectory analysis of these symptoms identified a high symptom group of 32% of patients with moderately severe symptoms (mean, 4.21, standard deviation, 1.58) that persist over time (Fig. 6).52 Given that most patients are recommended to remain on indefinite treatment with a TKI,5,59 tolerability of long-term therapy—especially symptoms, functional impairments, and decreased HRQoL—can be an important issue for patients. Even mild deficits that are not considered important for patients on short-term therapies can become major deterrents to treatment adherence in patients on continuous therapy60,61 and can lead to poorer treatment outcomes.62,63 For example, imatinib-induced nausea and muscle cramps were associated with intentional nonadherence and worse HRQoL,64 symptoms that may not befully appreciated by treating physicians.47,48

TKI DISCONTINUATION IN CML: IS IT READY FOR PRIME TIME?

Although the common thinking that prevailed in the CML scientific community was life-long TKIs, a team of French investigators who previously pioneered discontinuation of interferon in CML65 launched a daring trial of TKI discontinuation in patients with sustained molecular remission and undetectable molecular residual disease.66 Surprisingly, loss of undetectable molecular residual disease was only observed in 40% to 50% and occurred early after the TKI was stopped (6 months), whereas the other 50% to 60% enjoyed a sustained treatment-free remission. These data have since been replicated by several groups, as summarized in Table 5, with striking similarities in outcomes, despite different criteria for enrollment and/or for restarting TKI. This remarkable consistency in outcomes across all publications worldwide likely reflects the underlying biology but also the rigor and consistency with which these patients were monitored following TKI discontinuation. Why discontinue the TKI given the excellent outcomes of patients with CML on chronic oral TKI therapy? As discussed asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 473

KHOURY ET AL

TABLE 5. Summary of Tyrosine Kinase Discontinuation Trials Study

TKI

Min. TKI Treatment Duration (Years) No. of Patients

Depth of MR

Min. Duration of MR (Years)

RFS With at Least MMR

Euro-SKI67

750

MR4

52% at 2 years

STIM66

100

MR5

38% at 7 years

TWISTER

MR4.5

45% at 42 mos.

A-STIM69

UMRD

64% at 23 mos.

KIDS

MR4.5

58% at 2 years

HOVON71

20 mos.

MR4.5

33% at 3 years

STIM2

200

MR4.5

46% at 2 years

108

UMRD

1.5

52% at 22 mos.

ISAV73 STOP 2G-TKI

Dasa/nilo

MR4.5

ca. 55% at 4 years

DADI75

Dasa second line

MR4

49% at 6 mos.

NILST

Nilo

MR4.5

59% at 1 year

IM/dasa

MR4.5

58% at 6 mos.

Dasfree

Dasa

130

MR4.5

63% at 1 year

ENESTop79

IM/nilo

126

MR4.5

58% at 4 years

STAT2

TRAD77 78

IM/nilo

MR4.5

68% at 1 year

ENEST freedom81

Nilo

190

MR4.5

52% at 4 years

D-STOP

IM/dasa

(Second stop)

35 mos.

RE-STIM83

Total: 18

MR4

63% at 1 year

Mostly UMRD

31 mos

44% at 22 mos.

2,334

33%–68% after 0.5–7 years

*Duration of dasatinib therapy Abbreviations: TKI, tyrosine kinase inhibitors; MR, molecular remission; RFS, relapse-free survival; MMR, major molecular remission; IM, imatinib; UMRD, undetectable molecular residual disease; nilo, nilotinib; dasa, dasatinib.

above, TKIs are overall well tolerated, but PROs have shown that 30% to 40% of patients suffer low-grade but substantial adverse drug reactions that affect quality of life and compliance52—not to mention the financial burden that affects both patient (i.e., copays) and the society (i.e., costs of TKI).24,84

FIGURE 6. Trajectory of the Five Most Severe Symptoms in the High and Low Symptom Groups Over 1 Year

What Would It Take for TKI Discontinuation to Become Standard?

Criteria for TKI discontinuation are now part of the NCCN guidelines (Fig. 7),59 as well as in position papers by CML experts.4,85 In addition to the technical prerequisites that must be in place prior to considering TKI discontinuation (e.g., the availability of a molecular laboratory that quantitatively reports BCR-ABL1 levels with a reasonable turnaround time [within 2 weeks]), a key component is the CML knowledge within the practice that helps to interpret the molecular monitoring results without creating unnecessary anxiety or, on the other extreme, complaisance. An important complementary operational aspect is a multi-team approach where nonoverlapping function exists between team members who provide care, communication, and patient education—all key factors for patients’ engagement in their care.26 Indeed, such a team approach ensures patients come for their scheduled molecular monitoring and provides timely results with an interpretation on how the patient is faring with regard to the restart the TKI or not. A coordinated workforce and patients’ engagement are the top two Institute of Medicine recommendations for high-quality cancer care.86

What Type of Patients Can Be Offered TKI Discontinuation? This research was originally published in Williams et al.52 © American Society of Hematology.

474 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Patients with chronic phase CML with no prior resistance to TKI, treated with TKI for at least 5 years with a sustained

WHAT EVERY PRACTITIONER NEEDS TO KNOW ABOUT CML

FIGURE 7. TKI Discontinuation Guidelines

Abbreviations: CML, chronic myeloid leukemia; TKI, tyrosine kinase inhibitor; ALL, acute lymphoblastic leukemia; QPCR, qualitative polymerase chain reaction; IS, international scale; MMR, major molecular response; NCCN, National Comprehensive Cancer Network.

undetectable BCR-ABL1 or MR4.5 for at least 2 years appear to be the best candidates. Based on these criteria, approximately 30% to 40% of chronic phase CML treated with imatinib, nilotinib, or dasatinib achieve these landmarks.7,8,87 So far, no reliable marker is available that helps identify those patients who will not lose MMR following TKI discontinuation. Recent studies have suggested that a higher NK-cell count at the time of TKI discontinuation88 and a lower expression of the T-cell inhibitory receptor-ligand CD86 on plasmacytoid dendritic cells are associated with better probability of treatment-free remission.89 The frequency with which patients must be monitored has not been established. The original STIM monitoring schedule is quite intense (monthly PCRs for 6 months, then every other month) and has been the most commonly used regimen.66 A recent report suggests that a less intense monitoring frequency following TKI discontinuation yields comparable outcomes to current standards.90

Are There Risks Associated With TKI Discontinuation?

So far with current criteria and careful monitoring, TKI discontinuation appears safe with no reported blast transformations occurring off TKI. A rheumatologic syndrome commonly called “TKI-withdrawal syndrome” presenting with joint and muscle pain occurs in approximately 30% of patients in the first 4 weeks after stopping TKI.91 These symp-

toms usually last 4 to 6 weeks, are managed with nonsteroidal or short-courses of steroids, and resolve in the majority of cases, but in rare cases, they require reinstitution of TKI treatment. Better identification of patients who can discontinue successfully and remain in treatment-free remission are needed and will reduce the number of patients who go through this process unsuccessfully. Additionally, if any predictor of successful treatment-free remission is identified and able to be influenced, prediscontinuation interventions may increase the pool of patients who can successfully go through a first, and perhaps a second, TKI discontinuation trial.

Is TKI Discontinuation Ready for Prime Time?

Practically speaking, this verdict will be reached by practicing physicians and patients/patient advocacy groups. Indeed, a solid enough degree of confidence from both the provider and the patient sides is needed to proceed with this intervention. To ensure that this process remains as safe as it has been so far, we need CML expert guidelines, complemented by patients and their caretakers’ engagement in their discontinuation through education, communication, and technology.

CONCLUSION

TKI therapy for patients with CML has transformed cancer care and exemplifies the paradigm of precision medicine. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 475

KHOURY ET AL

Previously requiring a stem cell transplantation for any hope of cure, now patients taking TKIs can expect survival similar to the general population. However, TKI therapy is not without bothersome side effects for some patients that can impair quality of life and decrease adherence to prescribed treatment. It is important for clinicians to regularly assess the patient’s perspective of therapy, ideally with a routine patient-reported outcome measure, and to involve the pa-

tient in treatment decision making. The next threshold in patient care will be to define who can safely stop therapy and be cured. Evidence to date still suggests that the majority of patients will require lifelong therapy. Perhaps it is time now for CML to lead the transformation of cancer care again—setting our sights on ways to eradicate this disease and curing all patients with CML (cure defined as off therapy with no evidence of disease).

References 1. Hehlmann R. Innovation in hematology. Perspectives: CML 2016. Haematologica. 2016;101:657-659. 2. Saussele S, Krauss MP, Hehlmann R, et al; Schweizerische Arbeitsgemeinschaft für Klinische Krebsforschung and the German CML Study Group. Impact of comorbidities on overall survival in patients with chronic myeloid leukemia: results of the randomized CML study IV. Blood. 2015;126:42-49. 3. Cross NCP, White HE, Müller MC, et al. Standardized definitions of molecular response in chronic myeloid leukemia. Leukemia. 2012;26:2172-2175. 4. Hughes TP, Ross DM. Moving treatment-free remission into mainstream clinical practice in CML. Blood. 2016;128:17-23. 5. Baccarani M, Deininger MW, Rosti G, et al. European LeukemiaNet recommendations for the management of chronic myeloid leukemia: 2013. Blood. 2013;122:872-884. 6. Hehlmann R. CML--Where do we stand in 2015? Ann Hematol. 2015;94(Suppl 2):S103-S105. 7. Cortes JE, Saglio G, Kantarjian HM, et al. Final 5-year study results of DASISION: the dasatinib versus imatinib study in treatment-naïve chronic myeloid leukemia patients trial. J Clin Oncol. 2016;34:23332340. 8. Hochhaus A, Saglio G, Hughes TP, et al. Long-term benefits and risks of frontline nilotinib vs imatinib for chronic myeloid leukemia in chronic phase: 5-year update of the randomized ENESTnd trial. Leukemia. 2016;30:1044-1054. 9. Kalmanti L, Saussele S, Lauseker M, et al. Safety and efficacy of imatinib in CML over a period of 10 years: data from the randomized CML-study IV. Leukemia. 2015;29:1123-1132. 10. Deininger M, O’Brien SG, Guilhot F, et al. International randomized study of interferon Vs STI571 (IRIS) 8-year follow up: sustained survival and low risk for progression or events in patients with newly diagnosed chronic myeloid leukemia in chronic phase (CML-CP) treated with imatinib. Blood. 2009;114:1126. 11. Hehlmann R. How I treat CML blast crisis. Blood. 2012;120:737-747. 12. Pfirrmann M, Baccarani M, Saussele S, et al. Prognosis of long-term survival considering disease-specific death in patients with chronic myeloid leukemia. Leukemia. 2016;30:48-56.

15. Hehlmann R, Müller MC, Lauseker M, et al. Deep molecular response is reached by the majority of patients treated with imatinib, predicts survival, and is achieved more quickly by optimized high-dose imatinib: results from the randomized CML-study IV. J Clin Oncol. 2014;32:415423. 16. Baccarani M, Rosti G, Castagnetti F, et al. Sokal score and response to imatinib in early chronic phase CML: The GIMEMA CML Working Party experience on 559 patients. Haematologica. 2009;94:254 (abstr. 0626). 17. de Lavallade H, Apperley JF, Khorashad JS, et al. Imatinib for newly diagnosed patients with chronic myeloid leukemia: incidence of sustained responses in an intention-to-treat analysis. J Clin Oncol. 2008;26:3358-3363. 18. Cervantes F, López-Garrido P, Montero MI, et al. Early intervention during imatinib therapy in patients with newly diagnosed chronicphase chronic myeloid leukemia: a study of the Spanish PETHEMA group. Haematologica. 2010;95:1317-1324. 19. Baccarani M, Druker BJ, Branford S, et al. Long-term response to imatinib is not affected by the initial dose in patients with Philadelphia chromosome-positive chronic myeloid leukemia in chronic phase: final update from the Tyrosine Kinase Inhibitor Optimization and Selectivity (TOPS) study. Int J Hematol. 2014;99:616-624. 20. Sasaki K, Kantarjian HM, Jain P, et al. Ten-year follow up of patients with newly diagnosed chronic myeloid leukemia in chronic phase treated with 400 mg or 800 mg of imatinib daily. J Clin Oncol. 2014;32:5s (suppl; abstr 7024). 21. Gambacorti-Passerini C, Antolini L, Mahon FX, et al. Multicenter independent assessment of outcomes in chronic myeloid leukemia patients treated with imatinib. J Natl Cancer Inst. 2011;103:553-561. 22. Fabarius A, Leitner A, Hochhaus A, et al; Schweizerische Arbeitsgemeinschaft für Klinische Krebsforschung (SAKK) and the German CML Study Group. Impact of additional cytogenetic aberrations at diagnosis on prognosis of CML: long-term observation of 1151 patients from the randomized CML Study IV. Blood. 2011;118:6760-6768.

13. Höglund M, Sandin F, Simonsson B. Epidemiology of chronic myeloid leukaemia: an update. Ann Hematol. 2015;94(Suppl 2):S241-S247.

23. Fabarius A, Kalmanti L, Dietz CT, et al; SAKK and the German CML Study Group. Impact of unbalanced minor route versus major route karyotypes at diagnosis on prognosis of CML. Ann Hematol. 2015;94:2015-2024.

14. Thielen N, Visser O, Ossenkoppele G, et al. Chronic myeloid leukemia in the Netherlands: a population-based study on incidence, treatment, and survival in 3585 patients from 1989 to 2012. Eur J Haematol. 2016;97:145-154.

24. Padula WV, Larson RA, Dusetzina SB, et al. Cost-effectiveness of tyrosine kinase inhibitor treatment strategies for chronic myeloid leukemia in chronic phase after generic entry of imatinib in the United States. J Natl Cancer Inst. 2016;108:djw003.

476 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

WHAT EVERY PRACTITIONER NEEDS TO KNOW ABOUT CML

25. Noens L, Hensen M, Kucmin-Bemelmans I, et al. Measurement of adherence to BCR-ABL inhibitor therapy in chronic myeloid leukemia: current situation and future challenges. Haematologica. 2014;99:437447.

Measures: Use in Medical Product Development to Support Labeling Claims. http://www.fda.gov/downloads/Drugs/GuidanceComplia nceRegulatoryInformation/Guidances/UCM193282.pdf. Accessed January 18, 2017.

26. Holloway S, Lord K, Bethelmie-Bryan B, et al. Managing chronic myeloid leukemia: a coordinated team care perspective. Clin Lymphoma Myeloma Leuk. 2012;12:88-93.

40. Gilbert A, Sebag-Montefiore D, Davidson S, et al. Use of patientreported outcomes to measure symptoms and health related quality of life in the clinic. Gynecol Oncol. 2015;136:429-439.

27. Gratwohl A, Pfirrmann M, Zander A, et al; SAKK; German CML Study Group. Long-term outcome of patients with newly diagnosed chronic myeloid leukemia: a randomized comparison of stem cell transplantation with drug treatment. Leukemia. 2016;30:562-569.

41. Basch E. The rationale for collecting patient-reported symptoms during routine chemotherapy. Am Soc Clin Oncol Educ Book. 2014;161165:161-165.

28. Marcolino MS, Boersma E, Clementino NC, et al. Imatinib treatment duration is related to decreased estimated glomerular filtration rate in chronic myeloid leukemia patients. Ann Oncol. 2011;22:20732079. 29. Steegmann JL, Baccarani M, Breccia M, et al. European LeukemiaNet recommendations for the management and avoidance of adverse events of treatment in chronic myeloid leukaemia. Leukemia. 2016;30:1648-1671. 30. Kantarjian HM, Cortes JE, Kim DW, et al. Bosutinib safety and management of toxicity in leukemia patients with resistance or intolerance to imatinib and other tyrosine kinase inhibitors. Blood. 2014;123:1309-1318. 31. Cortes JE, Kim D-W, Pinilla-Ibarz J, et al; PACE Investigators. A phase 2 trial of ponatinib in Philadelphia chromosome-positive leukemias. N Engl J Med. 2013;369:1783-1796. 32. Nicolini FE, Alimena G, Al-Ali HK, et al. Final safety analysis of 1793 CML patients from ENACT (Expanding Nilotinib Access in Clinical Trials) study in adult patients with imatinib-resistant or –intolerant chronic myeloid leukemia. Haematologica. 2009;94 (abstr 0630). 33. Shah NP, Kim D-W, Kantarjian HM, et al. Dasatinib dose-optimization in chronic phase chronic myeloid leukemia (CML-CP): 2-year data from CA180-034 show equivalent long-term efficacy and improved safety with 100 mg once daily dose. Blood. 2008;112:3225. 34. Cortes JE, Kantarjian H, Brümmendorf T, et al. Safety and efficacy of bosutinib (SKI-606) in patients (pts) with chronic phase (CP) chronic myeloid leukemia (CML) following resistance or intolerance to imatinib (IM). J Clin Oncol. 2015;28:15s (suppl; abstr 6502).

42. Ellis LM, Bernstein DS, Voest EE, et al. American Society of Clinical Oncology perspective: Raising the bar for clinical trials by defining clinically meaningful outcomes. J Clin Oncol. 2014;32:1277-1280. 43. Hassett MJ, McNiff KK, Dicker AP, et al. High-priority topics for cancer quality measure development: results of the 2012 American Society of Clinical Oncology Collaborative Cancer Measure Summit. J Oncol Pract. 2014;10:e160-e166. 44. Basch E. New frontiers in patient-reported outcomes: adverse event reporting, comparative effectiveness, and quality assessment. Annu Rev Med. 2014;65:307-317. 45. Basch E, Wu A, Moinpour C, et al. Steps for assuring rigor and adequate patient representation when using patient-reported outcome performance measures. https://www.qualitymeasures.ahrq.gov/ expert/expert-commentary/47059. Accessed January 18, 2017. 46. Basch E, Jia X, Heller G, et al. Adverse symptom event reporting by patients vs clinicians: relationships with clinical outcomes. J Natl Cancer Inst. 2009;101:1624-1632. 47. Efficace F, Breccia M, Saussele S, et al. Which health-related quality of life aspects are important to patients with chronic myeloid leukemia receiving targeted therapies and to health care professionals? GIMEMA and EORTC Quality of Life Group. Ann Hematol. 2012;91:1371-1381. 48. Efficace F, Rosti G, Aaronson N, et al. Patient- versus physicianreporting of symptoms and health status in chronic myeloid leukemia. Haematologica. 2014;99:788-793. 49. Wilson IB, Cleary PD. Linking clinical variables with health-related quality of life. A conceptual model of patient outcomes. JAMA. 1995;273:59-65.

35. Lipton JH, Shah D, Tongbram V, et al. Comparative efficacy among 3rd line post-imatinib chronic phase-chronic myeloid leukemia (CPCML) patients after failure of dasatinib or nilotinib tyrosine kinase inhibitors. Blood. 2014;124:4551.

50. Sloan JA, Aaronson N, Cappelleri JC, et al; Clinical Significance Consensus Meeting Group. Assessing the clinical significance of single items relative to summated scores. Mayo Clin Proc. 2002;77:479-487.

36. Lauseker M, Hasford J, Pfirrmann M, et al; German CML Study Group. The impact of health care settings on survival time of patients with chronic myeloid leukemia. Blood. 2014;123:2494-2496.

51. Jensen RE, Snyder CF, Abernethy AP, et al. Review of electronic patientreported outcomes systems used in cancer clinical care. J Oncol Pract. 2014;10:e215-e222.

37. Goldberg SL, Cortes J, Gambacorti-Passerini C, et al. Predictors of performing response monitoring in patients with chronic-phase chronic myeloid leukemia (CP-CML) in a prospective observational study (SIMPLICITY). J Clin Oncol. 2014;32 (suppl 30; abstr 116).

52. Williams LA, Garcia Gonzalez AG, Ault P, et al. Measuring the symptom burden associated with the treatment of chronic myeloid leukemia. Blood. 2013;122:641-647.

38. European Medicines Agency. Reflection Paper on the Use of Patient Reported Outcome (PRO) Measures in Oncology Studies. http:// www.ema.europa.eu/docs/en_GB/document_library/Scientific_ guideline/2014/06/WC500168852.pdf. Accessed January 18, 2017. 39. U. S. Department of Health and Human Services Food and Drug Administration. Guidance for Industry Patient-Reported Outcome

53. Efficace F, Baccarani M, Breccia M, et al. International development of an EORTC questionnaire for assessing health-related quality of life in chronic myeloid leukemia patients: the EORTC QLQ-CML24. Qual Life Res. 2014;23:825-836. 54. Williams LA, Yucel E, Cortes JE, et al. Measuring symptoms as a critical component of drug development and evaluation in hematological diseases. Clin Investig (Lond). 2013;3:1127-1138.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 477

KHOURY ET AL

55. Hahn EA, Glendenning GA, Sorensen MV, et al; IRIS Investigators. Quality of life in patients with newly diagnosed chronic phase chronic myeloid leukemia on imatinib versus interferon alfa plus low-dose cytarabine: results from the IRIS Study. J Clin Oncol. 2003;21:21382146. 56. Efficace F, Baccarani M, Breccia M, et al; GIMEMA. Health-related quality of life in chronic myeloid leukemia patients receiving long-term therapy with imatinib compared with the general population. Blood. 2011;118:4554-4560. 57. Trask PC, Cella D, Besson N, et al. Health-related quality of life of bosutinib (SKI-606) in imatinib-resistant or imatinib-intolerant chronic phase chronic myeloid leukemia. Leuk Res. 2012;36:438-442. 58. Aziz Z, Iqbal J, Aaqib M, et al. Assessment of quality of life with imatinib mesylate as first-line treatment in chronic phase-chronic myeloid leukemia. Leuk Lymphoma. 2011;52:1017-1023. 59. National Comprehensive Cancer Network. Chronic Myelogenous Leukemia. NCCN Clinical Practice Guidelines in Oncology. https:// www.nccn.org/professionals/physician_gls/pdf/cml.pdf. Accessed January 24, 2017. 60. Pinilla-Ibarz J, Cortes J, Mauro MJ. Intolerance to tyrosine kinase inhibitors in chronic myeloid leukemia: Definitions and clinical implications. Cancer. 2011;117:688-697. 61. De Marchi F, Medeot M, Fanin R, et al. How could patient reported outcomes improve patient management in chronic myeloid leukemia? Expert Rev Hematol. 2017;10:9-14. 62. Marin D, Bazeos A, Mahon FX, et al. Adherence is the critical factor for achieving molecular responses in patients with chronic myeloid leukemia who achieve complete cytogenetic responses on imatinib. J Clin Oncol. 2010;28:2381-2388. 63. Ibrahim AR, Eliasson L, Apperley JF, et al. Poor adherence is the main reason for loss of CCyR and imatinib failure for chronic myeloid leukemia patients on long-term therapy. Blood. 2011;117:37333736. 64. Efficace F, Rosti G, Cottone F, et al. Profiling chronic myeloid leukemia patients reporting intentional and unintentional non-adherence to lifelong therapy with tyrosine kinase inhibitors. Leuk Res. 2014;38:294298. 65. Mahon FX, Delbrel X, Cony-Makhoul P, et al. Follow-up of complete cytogenetic remission in patients with chronic myeloid leukemia after cessation of interferon alfa. J Clin Oncol. 2002;20:214-220. 66. Mahon FX, Réa D, Guilhot J, et al; Intergroupe Français des Leucémies Myéloïdes Chroniques. Discontinuation of imatinib in patients with chronic myeloid leukaemia who have maintained complete molecular remission for at least 2 years: the prospective, multicentre Stop Imatinib (STIM) trial. Lancet Oncol. 2010;11:1029-1035. 67. Mahon FX, Richter J, Guilhot J, et al. Cessation of tyrosine kinase inhibitors treatment in chronic myeloid leukemia patients with deep molecular response: results of the Euro-Ski Trial. Blood. 2016;128:787. 68. Ross DM, Branford S, Seymour JF, et al. Safety and efficacy of imatinib cessation for CML patients with stable undetectable minimal residual disease: results from the TWISTER study. Blood. 2013;122:515-522. 69. Rousselot P, Charbonnier A, Cony-Makhoul P, et al. Loss of major molecular response as a trigger for restarting tyrosine kinase inhibitor therapy in patients with chronic-phase chronic myelogenous leukemia

478 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

who have stopped imatinib after durable undetectable disease. J Clin Oncol. 2014;32:424-430. 70. Lee SE, Choi SY, Bang JH, et al. Predictive factors for successful imatinib cessation in chronic myeloid leukemia patients treated with imatinib. Am J Hematol. 2013;88:449-454. 71. Thielen N, van der Holt B, Cornelissen JJ, et al. Imatinib discontinuation in chronic phase myeloid leukaemia patients in sustained complete molecular response: a randomised trial of the Dutch-Belgian Cooperative Trial for Haemato-Oncology (HOVON). Eur J Cancer. 2013;49:3242-3246. 72. Nicolini FE, Nicolini FE, Noël M-P, et al. Preliminary report of the STIM2 study: a multicenter stop imatinib trial for chronic phase chronic myeloid leukemia de novo patients on imatinib. Blood. 2013;122:654. 73. Mori S, Vagge E, le Coutre P, et al. Age and dPCR can predict relapse in CML patients who discontinued imatinib: the ISAV study. Am J Hematol. 2015;90:910-914. 74. Rea D, Nicolini FE, Tulliez M, et al; France Intergroupe des Leucémies Myéloïdes Chroniques. Discontinuation of dasatinib or nilotinib in chronic myeloid leukemia: interim analysis of the STOP 2G-TKI study. Blood. 2017;129:846-854. 75. Imagawa J, Tanaka H, Okada M, et al; DADI Trial Group. Discontinuation of dasatinib in patients with chronic myeloid leukaemia who have maintained deep molecular response for longer than 1 year (DADI trial): a multicentre phase 2 trial. Lancet Haematol. 2015;2:e528-e535. 76. Kadowaki N, Kawaguchi T, Kuroda J, et al. Discontinuation of nilotinib in patients with chronic myeloid leukemia who have maintained deep molecular responses for at least 2 years: a multicenter phase 2 stop nilotinib (Nilst) trial. Blood. 2016;128:790. 77. Dong D, Bence-Bruckler I, Forrest DL, et al. Treatment-free remission accomplished by dasatinib (TRAD): preliminary results of the PanCanadian Tyrosine Kinase Inhibitor Discontinuation trial. Blood. 2016;128:1922. 78. Shah NP, Paquette R, Müller MC, et al. Treatment-free remission (TFR) in patients with chronic phase chronic myeloid leukemia (CML-CP) and in stable deep molecular response (DMR) to dasatinib – the Dasfree Study. Blood. 2016;128:1895. 79. Hughes TP, Boquimpani CM, Takahashi N, et al. Treatment-free remission in patients with chronic myeloid leukemia in chronic phase according to reasons for switching from imatinib to nilotinib: subgroup analysis from ENESTop. Blood. 2016;128:792. 80. Takahashi N, Nakaseko C, Nishiwaki K, et al. Two-year consolidation by nilotinib is associated with successful treatment free remission in chronic myeloid leukemia with MR4.5: subgroup analysis from STAT2 trial in Japan. Blood. 2016;128:1889. 81. Hochhaus A, Casares MT, Stentoft J, et al. Patient-reported quality of life before and after stopping treatment in the ENESTfreedom trial of treatment-free remission for patients with chronic myeloid leukemia in chronic phase. Blood. 2016;128:3066. 82. Kumagai T, Nakaseko C, Nishiwaki K, et al. Discontinuation of dasatinib after deep molecular response for over 2 years in patients with chronic myelogenous leukemia and the unique profiles of lymphocyte subsets for successful discontinuation: a prospective, multicenter Japanese trial (D-STOP Trial). Blood. 2016;128:791.

WHAT EVERY PRACTITIONER NEEDS TO KNOW ABOUT CML

83. Pagliardini T, Nicolini FE, Giraudier S, et al. Second TKI discontinuation in CML patients that failed first discontinuation and subsequently regained deep molecular response after TKI re-challenge. Blood. 2016;128:788. 84. Experts in Chronic Myeloid Leukemia. The price of drugs for chronic myeloid leukemia (CML) is a reflection of the unsustainable prices of cancer drugs: from the perspective of a large group of CML experts. Blood. 2013;121:4439-4442. 85. Saußele S, Richter J, Hochhaus A, et al. The concept of treatment-free remission in chronic myeloid leukemia. Leukemia. 2016;30:16381647. 86. Institute of Medicine. Delivering High-Quality Cancer Care: Charting a New Course for a System in Crisis. Washington, DC: National Academies Press; 2013. 87. Branford S, Seymour JF, Grigg A, et al. BCR-ABL messenger RNA levels continue to decline in patients with chronic phase chronic myeloid leukemia treated with imatinib for more than 5 years and approximately

half of all first-line treated patients have stable undetectable BCR-ABL using strict sensitivity criteria. Clin Cancer Res. 2007;13:7080-7085. 88. Ilander M, Olsson-Strömberg U, Schlums H, et al. Increased proportion of mature NK cells is associated with successful imatinib discontinuation in chronic myeloid leukemia. Leukemia. Epub 2016 Dec 16. 89. Schütz C, Inselmann S, Sausslele S, et al. Expression of the CTLA4 ligand CD86 on plasmacytoid dendritic cells (pDC) predicts risk of disease recurrence after treatment discontinuation in CML. Leukemia. Epub 2017 Jan 27. 90. Kong JH, Winton EF, Heffner LT, et al. Does the frequency of molecular monitoring following tyrosine kinase inhibitor discontinuation affect outcomes of chronic myeloid leukemia? Cancer. Epub 2017 Feb 27. 91. Richter J, Söderlund S, Lübking A, et al. Musculoskeletal pain in patients with chronic myeloid leukemia after discontinuation of imatinib: a tyrosine kinase inhibitor withdrawal syndrome? J Clin Oncol. 2014;32:2821-2823.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 479

HAIDER ET AL

New Insight Into the Biology, Risk Stratification, and Targeted Treatment of Myelodysplastic Syndromes Mintallah Haider, MD, Eric J. Duncavage, MD, Khalid F. Afaneh, MD, Rafael Bejar, MD, PhD, and Alan F. List, MD OVERVIEW In myelodysplastic syndromes (MDS), somatic mutations occur in five major categories: RNA splicing, DNA methylation, activated cell signaling, myeloid transcription factors, and chromatin modifiers. Although many MDS cases harbor more than one somatic mutation, in general, there is mutual exclusivity of mutated genes within a class. In addition to the prognostic significance of individual somatic mutations, more somatic mutations in MDS have been associated with poor prognosis. Prognostic assessment remains a critical component of the personalization of care for patients with MDS because treatment is highly risk adapted. Multiple methods for risk stratification are available with the revised International Prognostic Scoring System (IPSS-R), currently considered the gold standard. Increasing access to myeloid gene panels and greater evidence for the diagnostic and predictive value of somatic mutations will soon make sequencing part of the standard evaluation of patients with MDS. In the absence of formal guidelines for their prognostic use, well-validated mutations can still refine estimates of risk made with the IPSS-R. Not only are somatic gene mutations advantageous in understanding the biology of MDS and prognosis, they also offer potential as biomarkers and targets for the treatment of patients with MDS. Examples include deletion 5q, spliceosome complex gene mutations, and TP53 mutations.

DS are bone marrow stem cell malignancies characterized by inefficient hematopoiesis, abnormal myeloid morphology, and cytopenias with risk of progression to secondary acute myeloid leukemia (AML). MDS is the most common hematopoietic myeloid cancer in adults with an average annual incidence of up to 75 per 100,000 persons 65 years or older.1,2 The diagnosis of MDS requires persistent cytopenias in the presence of dysplasia in one or more cell lineages and/or increased myeloblasts or clonal cytogenetic abnormalities and is classified according to the World Health Organization (WHO) criteria (Table 1).3 Over the last several years, many advances have been made in understanding the biology of MDS, most notably through the use of newer high-throughput DNA sequencing methods.

THE BIOLOGY OF MDS

Cytogenetic Findings in MDS

The earliest known molecular alterations in MDS were cytogenetic abnormalities detected by metaphase cytogenetics.5 Approximately 45% of patients with MDS harbor a recurrent cytogenetic abnormality (Table 2).6,7 In contrast to AML, copy number alterations including chromosomal deletions and amplifications are more common than translocations.

Certain cytogenetic findings in MDS are associated with changes in prognosis and are incorporated into the IPSS-R.8 Changes including complex karyotype (more than three cytogenetic abnormalities) and monosomal karyotype (one autosomal monosomy in the presence of a structural abnormality) have been associated with a poor prognosis.9,10 In addition to metaphase cytogenetics, fluorescence in situ hybridization may be used to detect recurrent cytogenetic alterations, providing increased sensitivity over conventional cytogenetics and permitting more accurate monitoring of disease burden in patients with MDS undergoing treatment.

Somatic Gene Mutations in MDS

The advent of massively paralleled digital sequencing methods (often colloquially grouped as next-generation sequencing) has provided rapid growth in our understanding of the molecular biology of myeloid neoplasms.11-13 These methods allow for the sequencing of small gene panels, the exome (the coding portion of the genome), or the entire genome with high sensitivity and at minimal cost.14,15 Over the last 8 years, numerous studies have demonstrated the following: (1) approximately 90% of patients with MDS will harbor at least one mutation from a set of approximately

From the Department of Hematology and Medical Oncology, Moffitt Cancer Center and the University of South Florida, Tampa, FL; Department of Pathology and Immunology, Washington University School of Medicine, St. Louis, MO; Moores Cancer Center, Division of Hematology and Oncology, University of California, San Diego, CA; Department of Malignant Hematology, Moffitt Cancer Center, Tampa, FL. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Mintallah Haider, MD, Department of Hematology and Medical Oncology, Moffitt Cancer Center/University of South Florida, GME Office, 12902 USF Magnolia Dr., Tampa, FL 33612; email: [emailprotected]. © 2017 American Society of Clinical Oncology

480 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

INsIGHTS INTO THE BIOLOGY, RISK STRATIFICATION, AND TREATMENT OF MDS

TABLE 1. 2016 WHO Classification of MDS

Classification

Dysplastic Lineages

Cytopenias*

RS as % of Marrow Erythroid Elements

MDS with SLD

1 or 2

MDS with MLD

2 or 3

MDS-RS with SLD

BM and PB Blasts

Cytogenetics by Conventional Karyotype Analysis

< 15%/< 5%**

BM < 5%, PB < 1%, no Auer rods

Any, unless fulfills all criteria for MDS with isolated del(5q)

1–3

< 15%/< 5%**

BM < 5%, PB < 1%, no Auer rods

Any, unless fulfills all criteria for MDS with isolated del(5q)

1 or 2

≥ 15%/≥ 5%**

BM < 5%, PB < 1%, no Auer rods

Any, unless fulfills all criteria for MDS with isolated del(5q)

MDS-RS with MLD

2 or 3

1–3

≥ 15%/≥ 5%**

BM < 5%, PB < 1%, no Auer rods

Any, unless fulfills all criteria for MDS with isolated del(5q)

MDS with isolated del(5q)

1–3

1–2

None or any

BM < 5%, PB < 1%, no Auer rods

del(5q) alone or with one additional abnormality, except −7 or del (7q)

MDS-EB-1

0–3

1–3

None or any

BM 5%–9% or PB 2%–4%, no Auer rods

Any

MDS-EB-2

0–3

1–3

None or any

BM 10%–19% or PB 5%–19%, or Auer rods

Any

With 1% Wood blasts

1–3

None or any

BM < 5%, PB = < 1%†, no Auer rods

Any

With SLD and pancytopenia

None or any

BM < 5%, PB < 1%, no Auer rods

Any

Based on defining cytogenetic abnormality

1–3

< 15%‡

BM < 5%, PB < 1%, no Auer rods

MDS-defining abnormality

Refractory cytopenia of childhood

1–3

None

BM < 5%, PM < 2%

Any

MDS With RS

MDS-EB

MDS-U

*Cytopenias defined as: hemoglobin (Hb) less than 10 g/dL, platelets less than 100,000/μL, and absolute neutrophils count less than 1,800/μL. PB monocytes must be less than 1,000/μL. **If SF3B1 mutation is present. †1% PB blasts must be recorded on at least two separate occasions. ‡Cases with at least 15% RS by definition have significant erythroid dysplasia and are classified as MDS-RS-SLD. Abbreviations: BM, bone marrow; EB, excess blasts; MDS, myelodysplastic syndromes; MLD, multilineage dysplasia; PB, peripheral blood; RS, ring sideroblasts; SLD, single-lineage dysplasia; U, unclassifiable. Adopted from Arber et al.4

KEY POINTS • Over the past decade, high-throughput DNA sequencing methods have advanced the understanding of MDS biology. • Somatic MDS mutations occur in five major categories: RNA splicing, DNA methylation, activated cell signaling, myeloid transcription factors, and chromatin modfication. • Increasing access to myeloid gene panels and greater evidence for the diagnostic and predictive value of somatic mutations will soon make sequencing part of the standard evaluation of patients with MDS. • In the absence of formal guidelines for their prognostic use, well-validated mutations can still refine estimates of risk made with the IPSS-R. • Not only are somatic gene mutations advantageous in understanding the biology of MDS and prognosis, they also offer potential as biomarkers and targets for the treatment of patients with MDS.

40 recurrently mutated MDS genes16-18 (Table 3; Fig. 1); (2) certain somatic mutations are associated with changes in prognosis19; (3) somatic mutations can be used to decipher the clonal architecture of MDS; and (4) a similar spectrum of somatic mutations can be seen in older patients without dysplasia (discussed in a later section), precluding the use of sequencing-based studies to replace morphologic evaluation.20,21 Sequencing-based “MDS gene panels” have now become commonplace in the clinical setting and can be used to better stratify patient risk and monitor clonal populations.22 Somatic MDS mutations occur in five major categories, including RNA splicing, DNA methylation, activated cell signaling, myeloid transcription factors, and chromatin modifiers. Although many MDS cases harbor more than one somatic mutation, in general, there is mutual exclusivity of mutated genes within a class. In addition to the prognostic significance of individual somatic mutations, more somatic mutations in MDS have been associated with poor prognosis. Splicing mutations. Mutations involving RNA splicing are present in up to 45% of MDS; however, they appear to be rare in de novo AML.13,24 These mutations affect 3′ splice asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 481

HAIDER ET AL

TABLE 2. Prognostically Significant Recurrent Cytogenetic Findings in Patients With MDS

Cytogenetic Abnormalities

Median Survival (Years)

Median AML Evolution, 25% (Years)

HRs OS/AML

−Y, del(11q)

5.4

0.7/0.4

Good (66%–72%)

Normal, del(5q), del(12p), del(20q), double including del(5q)

4.8

9.4

1/1

Intermediate (13%– 19%)

del(7q), +8, +19, i(17q), any other single or double independent clones

2.7

2.5

1.5/1.8

Poor (4%–5%)

−7, inv(3)/t(3q)/del(3q), double including –7/ del(7q), complex: three abnormalities

1.5

1.7

2.3/2.3

Very poor (7%)

Complex: more than three abnormalities

0.7

3.8/3.6

Prognostic Subgroups, % of Patients Very good (3%–4%)

Abbreviations: AML, acute myeloid leukemia; HR, hazard ratio; MDS, myelodysplastic syndromes; NR, not reached; OS, overall survival. Hazard ratios are reported as overall survival/risk of AML transformation. Adopted from Greenberg et al.8

recognition sites, although the exact mechanisms by which the mutations cause dysplasia and the RNA targets of aberrant gene splicing are unknown. Mutations in the splicing factor 3b, subunit 1 (SF3B1) gene are present in approximately 20% of MDS and are associated with ring sideroblast morphology, lower-grade disease, and better prognosis.25 Interestingly, mutations in SF3B1 have also been found in patients with chronic lymphocytic leukemia (CLL) and breast cancer.26-28 Patients with MDS with ring sideroblasts without mutated SF3B1 (approximately 20% of patients) are thought to have an inferior prognosis compared with patients with mutated SF3B1.29 Mouse models have demonstrated that mutated SF3B1 causes an MDS phenotype and that cells carrying the

mutation are sensitive to spliceosome modulator drugs.30 SF3B1 mutations have also been associated with chronic lymphocytic leukemia.30 Mutations in the U2 small nuclear RNA auxiliary factor 1 gene (U2AF1 or U2AF35) are present in approximately 8% to 12% of MDS and are associated with a poor prognosis. U2AF1 mutations have been shown to alter splice recognition sites and specificity of precursor messenger RNA binding, eliciting changes in thousands of RNA transcripts; however, the exact mechanism by which U2AF1 mutations give rise to dysplasia have yet to be determined.31 Interestingly, in vitro and animal studies have demonstrated increased sensitivity to precursor messenger RNA splicing modulator drugs.32 Mutations in the Serine/arginine-rich

TABLE 3. Recurrently Mutated MDS Genes Gene

Function

Chromosome

Incidence (%)

Clinical Significance

NRAS

Activated signaling

1p13.2

5–10

Associated with poor prognosis

CBL

Activated signaling

11q23.3

1,000 radiation oncologists. Am J Clin Oncol. Epub 2015 Feb 2. 92. Weickhardt AJ, Scheier B, Burke JM, et al. Local ablative therapy of oligoprogressive disease prolongs disease control by tyrosine kinase inhibitors in oncogene-addicted non–small cell lung cancer. J Thorac Oncol. 2012;7:1807-1814. 93. Gan GN, Weickhardt AJ, Scheier B, et al. Stereotactic radiation therapy can safely and durably control sites of extra-central nervous system oligoprogressive disease in anaplastic lymphoma kinase–positive

618 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

lung cancer patients receiving crizotinib. Int J Radiat Oncol Biol Phys. 2014;88:892-898. 94. Yu HA, Sima CS, Huang J, et al. Local therapy with continued EGFR tyrosine kinase inhibitor therapy as a treatment strategy in EGFRmutant advanced lung cancers that have developed acquired resistance to EGFR tyrosine kinase inhibitors. J Thorac Oncol. 2013;8:346-351. 95. Iyengar P, Kavanagh BD, Wardak Z, et al. Phase II trial of stereotactic body radiation therapy combined with erlotinib for patients with limited but progressive metastatic non–small cell lung cancer. J Clin Oncol. 2014;32:3824-3830.

HISTOLOGIC CHARACTERIZATION AND TISSUE/PLASMA GENOTYPING IN NSCLC

Pathology Issues in Thoracic Oncology: Histologic Characterization and Tissue/Plasma Genotyping May Resolve Diagnostic Dilemmas Ibiayi Dagogo-Jack, MD, Andreas Saltos, MD, Alice T. Shaw, MD, PhD, and Jhanelle E. Gray, MD OVERVIEW Lung cancer is a heterogeneous diagnosis that encompasses a spectrum of histologic and molecular subgroups. A paradigm shift favoring selection of treatment based on histologic and molecular makeup has positively affected prognosis for patients with metastatic lung cancer, with select patients experiencing durable responses to treatment. However, prognosis remains poor for the majority of patients. Furthermore, oncologists are increasingly faced with challenging dilemmas related to histopathologic and molecular characterization of tumors, both at diagnosis and during treatment. In this review, we focus on three particular challenges: (1) management of mixed histology tumors, a particularly aggressive group of lung cancers, (2) distinguishing multiple primary lung tumors from intrapulmonary metastases, and (3) incorporation of liquid biopsies into the diagnostic algorithm and subsequent follow-up of patients with advanced lung cancer. This review will summarize the existing literature and highlight the potential for molecular genotyping to help refine approaches to each of these challenges.

ung cancer, the leading cause of cancer-related mortality worldwide, accounts for one-quarter of deaths from cancer.1 In 2017, it is estimated that 222,500 people will be diagnosed with and 155,870 people will die of lung cancer in the United States.1 Despite significantly improved prognosis for all-comers with cancer in the past 2 decades, the 5-year survival rate for lung cancer has been stagnant at 18%.1,2 Over the years, adoption of newer chemotherapy combinations and optimization of supportive care has modestly improved outlook for patients diagnosed with metastatic lung cancer.3 However, the greatest impact on outcome is likely to come with the recent shift from uniform prescriptions of chemotherapy to use of molecular and pathologic features to tailor treatments in the metastatic setting. The diagnosis of lung cancer encompasses a spectrum of histologic subtypes, most notably non–small cell lung cancer (NSCLC) and small cell lung cancer (SCLC).4 Although the disparate biologic behavior of these two major subtypes of lung cancer has long been recognized, the demonstration that NSCLC histologic subtype (i.e., adenocarcinoma, squamous, and large cell carcinoma) dictates outcomes to treatment is a more recent finding.5 Indeed, the realization that histology is an independent predictor of response provided strong rationale for using pathology to guide treatment and reset the standard of care for NSCLC.5,6 However, in a subset

of patients, presence of more than one histologic subtype poses a challenge for histology-guided treatment strategies, as the optimal management of these mixed-histology tumors remains to be established. In addition, another challenge often encountered in clinical practice is the presence of multiple lung nodules. Although the histopathology of these nodules can sometimes distinguish metastatic lung cancer from multiple synchronous primaries, other testing modalities may be required, including molecular profiling of separate lesions and additional imaging studies. Characterization of recurring molecular alterations has been instrumental in cementing the understanding of NSCLC as a heterogeneous disease comprised of distinct molecular subgroups with unique clinical features and prognostic outcomes.7-12 Remarkably, patients belonging to select molecular subgroups are now expected to survive 3 to 4 years after diagnosis.12-14 As such, it is now recommended that diagnostic biopsy specimens from all newly diagnosed patients with metastatic nonsquamous NSCLC undergo molecular testing for EGFR mutations and ALK and ROS1 rearrangements, preferentially within the context of broad molecular panels.15 Although the superior efficacy of treatment strategies that rely on defining biologic features (e.g., high PD-L1 expression or presence of molecular drivers) has shifted some focus away from histology, it should be noted that histology

From the Massachusetts General Hospital Cancer Center, Harvard Medical School, Boston, MA; Hematology/Oncology Fellowship Program, Moffitt Cancer Center, Tampa, FL; University of South Florida, Tampa, FL; Department of Thoracic Oncology, Moffitt Cancer Center, Tampa, FL. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Alice T. Shaw, MD, PhD, Massachusetts General Hospital Cancer Center, 32 Fruit St., Boston, MA, 02114; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 619

DAGOGO-JACK ET AL

is still highly relevant because roughly one-half of patients will not have tumors bearing these markers.9,12,16 It is now widely accepted that histologic and molecular characterization offers greater insight into the biologic behavior of metastatic NSCLC than imaging features. Despite this consensus, molecular profiling is most often an isolated event that occurs at diagnosis with subsequent clinical decisions predicated on serial imaging. Imaging, however, can be an imprecise biomarker. Given the limitations of imaging as a barometer of underlying biology, several studies have advocated for repeat biopsy of progressive or suspicious sites and demonstrated its utility in refining treatment and improving understanding of the biology of resistance.17-20 Based on these studies, repeat biopsy is now becoming a fundamental part of routine clinical practice for oncogene-driven NSCLC (e.g., EGFR-mutant and ALK-positive NSCLC). Serial tissue sampling, however, is not always feasible or informative.21 Thus, there is growing interest in developing noninvasive technologies, particularly plasma genotyping assays, that establish molecular genotype at diagnosis and facilitate serial and comprehensive molecular profiling. Although prioritizing pathology and molecular findings has rapidly transformed clinical practice and allowed for standardization of treatment approaches by molecular subtype, treatment of NSCLC is far from algorithmic. Incorporation of plasma genotyping into management strategies is likely to add another layer of complexity. In this review, we will summarize the existing literature and describe our approach to several diagnostic and therapeutic conundrums, namely optimal treatment of mixed histology tumors, differentiating between intrathoracic metastasis and second primary tumors, and interpreting plasma genotyping results within the context of current clinical practice.

KEY POINTS • Lung cancer is a heterogeneous disease comprised of distinct histologies and unique molecular subgroups, each with characteristic clinical features and prognostic outcomes. • Mixed histology lung cancers, accounting for 5% of lung cancers, are associated with aggressive biology and poor outcomes. • Comprehensive assessment that integrates radiographic, immunohistochemical, morphologic, and molecular findings may improve accuracy when attempting to distinguish multiple primary lung cancers from intrathoracic metastasis. • Liquid biopsies are a promising method for detecting clinically relevant and novel molecular alterations, improving understanding of molecular response dynamics, and elucidating genetic determinants of resistance. • The sensitivity of current liquid biopsy platforms is not sufficient to discount a molecular alteration if it is not present in plasma, but high specificity supports initiating treatment based on plasma-detected alterations. 620 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

APPROACH TO MIXED-HISTOLOGY TUMORS

Although the vast majority of lung cancers can be classified into four major histologies (adenocarcinoma, squamous, large cell, or small cell carcinoma), there are a wide variety of histologic subtypes and considerable heterogeneity among less common histologies.4 Mixed-histology tumors comprise approximately 5% of lung cancers.22-25 Due to the rareness of these tumors and exclusion of patients with mixed-histology tumors from many of the clinical trials that have shaped the treatment landscape, these cases present unique diagnostic and therapeutic challenges.

Combined Small Cell Lung Cancer

Combined SCLC (c-SCLC) is defined by World Health Organization classification as a subtype of SCLC characterized by an admixture of elements of SCLC with NSCLC.4 These combined tumors are believed to arise from a pluripotent stem cell capable of differentiating into either SCLC or NSCLC. This theory is supported by the identification of SCLC transformation as a mechanism of resistance to EGFR inhibition in tumors that were initially characterized as EGFR-mutant NSCLC.18 Estimates of the incidence of c-SCLC are based on a limited number of studies and range from under 5% to as many as 30% of cases of SCLC.26-28 Tumors that contain any component of SCLC should be classified as c-SCLC. Clinical characteristics and presentation do not differ significantly among patients with c-SCLC and those with pure SCLC.29 Given the aggressive biology of SCLC, c-SCLCs are typically treated with SCLC regimens. Standard-of-care treatment involves doublet chemotherapy with platinum plus etoposide for patients with extensive-stage cancer and this doublet combined with radiotherapy in limited-stage disease. As in SCLC, surgery is reserved for select patients with very earlystage c-NSCLC who do not have nodal involvement. The prognosis for c-SCLC is generally comparable to pure SCLC, although there have been conflicting reports.29-31

Other Mixed Neuroendocrine Tumors

The spectrum of neuroendocrine differentiation includes carcinoid tumors, large cell carcinoma, and small cell carcinoma. Mixed neuroendocrine tumors with components of large cell carcinoma or carcinoid are rare and, consequently, not well described. In one large retrospective study featuring 2,501 patients, mixed neuroendocrine tumors were identified in less than 1% of cases.32 Data are limited regarding optimal management and prognosis for patients with these mixed neuroendocrine tumors. However, tumors with carcinoid components are expected to have a more indolent course than those harboring higher-grade small cell or large cell components.33 For localized disease, particularly tumors with a low-grade neuroendocrine component, surgery is the mainstay of treatment and may be helpful for definitive grading.32,34

Adenosquamous Carcinoma

Adenosquamous carcinomas are defined as NSCLCs that have at least 10% histologic components of both adenocarcinoma and squamous carcinoma.4 Although relatively uncommon,

HISTOLOGIC CHARACTERIZATION AND TISSUE/PLASMA GENOTYPING IN NSCLC

they represent 2% to 4% of lung cancers and often present at a more advanced stage. A study of stage I and II NSCLCs found that adenosquamous lung cancer has a worse prognosis (5-year overall survival 59.4%) than that of pure adenocarcinoma (5-year overall survival 71.7%) or pure squamous carcinoma (5-year overall survival 66.1%),35 an association that has been observed in other published series.25,36 Although these observations might suggest that adjuvant therapies should be used for treatment of this subtype of lung cancer, in the absence of confirmation of this hypothesis, adenosquamous carcinomas are typically treated using standard NSCLC treatment regimens. Recently, molecular profiling of these tumors has demonstrated a high incidence of driver mutations, including EGFR mutations and activating mutations involving the phosphoinositide 3-kinase signaling pathway.37 Interestingly, these mutations are often present in both the adenocarcinoma and squamous components of the tumor, suggesting that these tumors may respond to treatment with molecularly targeted agents.38 As identification of actionable alterations may allow patients to access additional therapies, we recommend molecular profiling for all metastatic adenosquamous carcinomas.

Biphasic Tumors

By definition, biphasic pulmonary tumors contain mixed epithelial and mesenchymal components. The most common examples include carcinosarcoma (0.2%–0.3% of all lung cancers) and pulmonary blastoma (0.25%–0.5% of all lung cancers), both of which are subtypes of pulmonary sarcomatoid carcinoma.4,39,40 Carcinosarcoma is defined as a combination of a typical lung carcinoma with a sarcomatous element and is most frequently observed in middle-aged patients with extensive tobacco exposure. Pulmonary blastomas have an adenocarcinoma component with fetal features as well as a primitive mesenchymal stroma. The average age at diagnosis of pulmonary blastoma is 40 years, and occurrence is more frequent in males. Both carcinosarcomas and pulmonary blastomas are aggressive tumors that frequently exhibit rapid growth and locoregional and distant spread. Not surprisingly given these biologic characteristics, prognosis of these tumors is poorer than that of more common NSCLC histologies.39,40 Treatment of these tumors is a challenge, as systemic therapies do not produce robust responses. Given the poor efficacy of systemic treatments, surgical resection is recommended when feasible. Adjuvant radiation therapy has been shown to reduce the rate of local recurrence by 15% in some cases.39 As these tumors are characterized by a poor response to chemotherapy, there is interest in exploring targeted therapy and immunotherapy for these patients. Interestingly, mutations leading to MET exon 14 skipping, an alteration that confers sensitivity to MET inhibitors, have been detected in 8% to 31% of sarcomatoid carcinomas in three published series.41-43 In addition, high-level expression of PD-L1 was identified by quantitative immunofluorescence in 9 of 13 (69.2%) sarcomatoid carcinomas in a small study.44 There is limited experience using checkpoint inhibitors in this patient population, and, as such, efficacy is unknown.

As carcinosarcomas and blastomas have been excluded from genotyping studies, little is known about the molecular makeup of these tumors. As these tumors are also poorly differentiated, it is possible that they might share molecular features with sarcomatoid carcinomas.

DIFFERENTIATING METASTASIS IN THE CHEST FROM SECOND PRIMARIES

Identification of multiple malignant lesions within the lungs presents an increasingly common clinical challenge. Published series have estimated an incidence of multiple synchronous tumors of up to 15% among cases of NSCLC.45,46 Several studies analyzing these cases have concluded that, whereas a majority of tumors represent metastatic lesions, approximately one-third may represent distinct primary tumors.46,47 The high incidence of second primary lung cancers may be explained by a field effect in smokers.48 Despite the therapeutic and prognostic ramifications, reaching a definitive conclusion regarding the origins of the lung lesions (i.e., metastatic disease or multiple primary cancers) is often difficult for pathologists and clinicians. There are several published criteria to aid in distinguishing intrathoracic metastasis from second primary tumors. The Martini-Melamed criteria (Sidebar 1), which use anatomic and histologic features to classify tumors, have traditionally been used in these cases.49 However, these criteria were developed prior to the advent of modern imaging and molecular diagnostic technologies. In addition, these criteria may not be sufficient in situations in which the cancers are histologically indistinguishable. Indeed, several recent studies have shown that, even among histologically similar NSCLCs, approximately one-third of cases exhibit distinct molecular and/or genetic profiles.45,50,51

Histologic and Molecular Assessment

As reliably differentiating between these two entities requires adequate tissue for downstream analyses, biopsy of multiple suspicious lesions is essential. Notably, updated criteria have been published that incorporate detailed histologic subtyping and molecular and genetic signatures to improve accuracy when assessing clonality of separate tumors (Sidebar 1).44,52 Comprehensive histologic assessment involves meticulous characterization of specimens, including identification and quantification of histologic subtypes and analysis of cytologic features such as grade, necrosis, and stromal appearance. Although time-intensive, the depth of analysis allows for more accurate classification of multiple tumors than the MartiniMelamed criteria.52 Current guidelines also champion a multipronged approach to resolving this clinical conundrum. For example, the 7th edition of the AJCC Cancer Staging Manual recommends classifying tumors as synchronous primaries “based on features such as differences in morphology, immunohistochemistry and/or molecular studies” and absence of “evidence of mediastinal nodal metastases or of nodal metastases within a common nodal drainage.”53 Multidimensional assessment that integrates imaging, histologic, and molecular features not only increases accuracy but also decreases the potential that interobserver variation may bias conclusions. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 621

DAGOGO-JACK ET AL

SIDEBAR 1. Summary of Published Criteria for Definitions of Distinct Primary Lung Cancer

Martini and Melamed49

Synchronous Tumors: A. Tumors physically distinct and separate B. Histologic type 1. Different 2. Same, but in different segment, lobe, or lung if: a. Origin from carcinoma in situ b. No carcinoma in common lymphatics c. No extrapulmonary metastases Metachronous Tumors: A. Histologic type different, or B. Histologic type same if: 1. Interval between cancers ≥ 2 years, or 2. Origin from carcinoma in situ, or 3. In different lobe or lung, so long as: a. No carcinoma in common lymphatics b. No extrapulmonary metastases

Girard et al52

A. Histologic type different, or B. For squamous carcinomas: 1. Cytologic/stromal features different C. For adenocarcinomas: 1. Major histologic subtype different, or 2. Major histologic subtype same, but: a. Other histologic subtype percentages different, and b. Cytologic/stromal features different

Detterbeck et al45

A. Histologic type different unless: 1. Clearly different by comprehensive histologic assessment, or 2. Both squamous carcinoma arising from carcinoma in situ B. Comparative genomic hybridization, if performed, should not identify matching breakpoints C. Relative arguments favoring separate tumors: 1. Different radiographic appearance or metabolic uptake 2. Different pattern of biomarkers (e.g., driver mutations) 3. Different rates of growth (if prior imaging available) 4. Absence of nodal or systemic metastases D. Relative arguments favoring a single tumor source: 1. Same radiographic appearance 2. Similar growth patterns 3. Significant nodal or systemic metastases 4. Same biomarker pattern 5. Matching appearance on comprehensive histologic assessment As molecular profiling is now a fundamental component of the diagnostic paradigm for NSCLC, routine molecular testing results alone may adequately discriminate between metastatic tumors and second primaries. Indeed, several studies have reported discordant driver mutation status in separate lesions.46,54 As it is possible that tumors may not contain well-characterized lung cancer alterations or that synchronous tumors of distinct origin may still share a 622 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

common driver mutation, next-generation sequencing (NGS) panels that enable simultaneous interrogation of multiple genes may be the most ideal platform for establishing the molecular footprint of spatially separated lesions. In addition to these standard molecular profiling strategies, several studies have used alternative approaches, specifically microsatellite analysis and assessment of loss of heterozygosity, to distinguish metastatic foci from second primary cancers.46,47,51

HISTOLOGIC CHARACTERIZATION AND TISSUE/PLASMA GENOTYPING IN NSCLC

Treatment Approach and Prognosis

When synchronous primary tumors are confirmed, the individual tumors should be staged independently and treated as distinct cancers. If the distinction between metastasis and second primary is unclear, surgical resection of both tumors is encouraged whenever appropriate, based on tumor staging and the patient’s cardiopulmonary reserve. Surgical resection offers the best survival outcome and ensures sufficient tissue for comprehensive histologic assessment and molecular profiling, which might not be possible with a limited biopsy specimen.55 For metachronous lung tumors, a biopsy or resection is recommended. Similar methods to those described above can be used to distinguish local recurrence from a second primary.55 As synchronous and metachronous lesions with discordant EGFR mutation status have been observed, molecular characterization of multiple lesions is encouraged prior to pursuing investigational targeted approaches.54 Survival estimates for multiple primary lung cancer are considerably better than that of intrapulmonary metastasis. For example, a recent meta-analysis of stage I and II tumors reported 5-year overall survival rates for multiple primary lung cancer ranging from 15% to 81%.56 Although survival rates are comparable between synchronous and metachronous lesions, survival after a diagnosis of multiple primary lung cancer is significantly longer than that observed for patients with intrapulmonary metastasis (hazard ratio 2.66).

LIQUID BIOPSIES: A COMPLEMENTARY APPROACH TO TISSUE BIOPSIES FOR GENOTYPING NSCLC

Rationale for Liquid Biopsies

Biopsy of a suspected involved site, the longstanding gold standard for confirming a lung cancer diagnosis, allows for histologic and molecular characterization of tumors and provides invaluable guidance for designing rational therapeutic approaches.12 However, up to one in four patients will not have sufficient tissue for molecular testing and, as such, cannot access potentially effective therapies.57 Moreover, increasing appreciation of the evolution of lung cancers under therapeutic selective pressure suggests that serial molecular profiling may be more advantageous than a single diagnostic biopsy.17-20 Repeat tissue sampling, however, has considerable limitations, including patient-specific factors (e.g., risk and discomfort) and lesion-specific factors (e.g., inaccessible sites and intrapatient tumor heterogeneity).21,58,59 Recognition of the inadequacy of an isolated diagnostic biopsy as a blueprint for clinical decision making has generated broad interest in developing reliable, noninvasive methods for molecular surveillance. Several studies have shown that profiling tumor-derived free-floating nucleic acids in plasma or molecular material contained in circulating tumor cells or exosomes can provide valuable insight into a cancer’s dynamic molecular trajectory.60-66 Regardless of the analytic platform or source of genetic material, molecular profiling of plasma contents is collectively referred to as a liquid biopsy.

Of the liquid biopsy options that have been explored to date, analysis of plasma cell-free tumor-derived deoxyribonucleic acid (ctDNA) has had the most clinical impact. Notably, the initial description of circulating cell-free DNA in plasma and the realization that cell-free DNA was more abundant in patients with malignancies occurred more than half a century before the approval of the first liquid biopsy by the U.S. Food and Drug Administration.67,68 The predominance of non–tumor-derived DNA in plasma relative to tumor-derived DNA posed the greatest hindrance to tapping into the potential of plasma profiling and likely accounted for much of the delay in translating these early observations into a robust, clinically relevant technology.69 Groundbreaking molecular diagnostic and analytical advances in the past decade, however, have facilitated the development and incorporation of liquid biopsies into clinical practice.

Performance Characteristics of Available Liquid Biopsy Technologies

With optimization of diagnostics to suppress input from the wild-type allele and benign cells, the sensitivity of current platforms approaches 80%. Sensitivity is highest for ubiquitously present molecular alterations, particularly base substitutions or short insertions/deletions (e.g., EGFR exon 19 deletion and EGFR L858R).63,64,70-78 For example, one study observed that plasma genotyping using allele-specific polymerase chain reaction (PCR) technology (Table 1) had a sensitivity approaching 90% for detecting sensitizing EGFR mutations compared with sensitivity of approximately 40% for the EGFR T790M resistance mutation (Table 2).70 The performance of this Cobas allele-specific plasma genotyping assay in patients with sensitizing EGFR mutations ultimately led to approval of the first liquid biopsy by the U.S. Food and Drug Administration in 2016.79 Subsequent studies have demonstrated that assays that use emulsion-based digital PCR techniques (Table 1) that enrich for tumor DNA relative to background DNA produce superior results when compared with allele-specific PCR.63,70,74 Although the performance characteristics of digital PCRbased liquid biopsies have been encouraging, these assays are still prone to false-negative results. In addition, these assays are not optimally equipped to detect complex alterations (e.g., fusions) and can only interrogate hot spots in a handful of genes at a given time. These shortcomings are particularly relevant, as identification of certain gene fusions (e.g., ALK or ROS1 rearrangements) is of critical prognostic and therapeutic consequence.80,81 The migration toward comprehensive molecular profiling of diagnostic biopsy specimens using multiplex or NGS panels and away from testing for individual alterations suggests that NGSbased liquid biopsy platforms may be the most optimal surrogate for tissue genotyping. Indeed, the handful of studies published to date using NGS-based plasma assays demonstrates that this method can identify a broad range of clinically relevant alterations in multiple genes with comparable sensitivity to digital PCR in patients with NSCLC.62,64 Although only a small number of patients with fusion-driven NSCLC asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 623

DAGOGO-JACK ET AL

TABLE 1. Techniques for Genotyping Plasma DNA Assay Technique

Method of Detecting Mutant Variants

Detected Alterations

Allele-specific PCR (Cobas EGFR assays)

Uses PCR primers that selectively target and amplify the mutant variant

Insertions/deletions and base substitutions

Digital/emulsion PCR (digital droplet PCR, BEAMing)

Employs surfactant technology to partition DNA into thousands of individual PCR reactions contained in droplets; this technique allows for quantification of the number of mutant and wild-type variants and permits detection of variants present at very low levels.

Insertions/deletions and base substitutions

Amplicon-based NGS

Multiplexed detection of a broad range of alterations involving multiple genes using sequence-specific PCR primers designed to selectively target and simultaneously amplify genomic regions of interest

Insertions/deletions, base substitutions, copy number gains

Capture-based NGS

Simultaneous detection of a broad range of alterations involving multiple genes using DNA oligonucleotides complementary to sequences of interests in exons and introns

Insertions/deletions, base substitutions, copy number gains, gene fusions

Abbreviation: PCR, polymerase chain reaction; BEAM, beads, emulsification, amplification, and magnetics; NGS, next-generation sequencing.

were included in these studies, one study demonstrated high sensitivity for detecting ALK fusions using a hybrid capture–based NGS assay (Tables 1 and 2).77

INCORPORATING LIQUID BIOPSIES INTO CLINICAL PRACTICE

Utility of Liquid Biopsy for Identifying Clinically Relevant Targets at Diagnosis or Relapse

Multiple studies have established high concordance between tissue and plasma genotyping using various techniques (Table 2). Although the PCR-based Cobas assays are the only U.S. Food and Drug Administration–approved liquid biopsies, NGS-based genotyping reliably identifies established drivers of resistance in plasma, including ALK resistance mutations in crizotinib-resistant plasma specimens and EGFR T790M and MET amplification in plasma samples from patients with lung cancers that are resistant to

first- and second-generation EGFR tyrosine kinase inhibitors (TKIs).64 In addition, NGS-based plasma profiling successfully captures relevant molecular alterations in cases with inadequate diagnostic biopsies, suggesting that it may be a powerful tool for guiding initial management.62 For example, in a single-center study, plasma genotyping by NGS was successful for 50 patients with an insufficient tissue sample or an inaccessible site for biopsy, including eight cases with EGFR T790M.62 Plasma genotyping with digital PCR yields quicker results relative to tissue genotyping using comprehensive panels.63 Use of NGS-based assays for plasma genotyping may eliminate this advantage. However, the reporting delays that arise from the increased complexity of input material may be offset by the potential to simultaneously query multiple actionable genes and types of alterations using NGS. Moreover, with the exception of the U.S. Food and

TABLE 2. Concordance Between Tissue and Plasma NSCLC Genotyping in PublishedStudies Alteration

Technique

Sensitivity (%)

Specificity (%)

References

EGFR T790M

Allele-specific PCR

100

EGFR L858R or exon 19 deletion

Allele-specific PCR

46–90

97–100

70-73

EGFR T790M

BEAMing

70 to 71

67–69

70,74

EGFR L858R or exon 19 deletion

BEAMing

82.3–100

96.5–100

70,74

EGFR T790M

Digital droplet PCR

EGFR L858R or exon 19 deletion

Digital droplet PCR

74–82

100

KRAS G12X

Digital droplet PCR

100

EGFR mutations

Amplicon-based

87–100

94–100

Multiple (no fusions)

Amplicon-based

ALK rearrangement

Capture-based

79.2

100

Multiple

Capture-based

72–85

96–100

64,78

Allele-Specific PCR

Digital PCR

NGS

Abbreviation: NSCLC, non–small cell lung cancer; PCR, ploymerase chain reaction; BEAM, beads, emulsification, amplification, and magnetics; NGS, next-generation sequencing.

624 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

HISTOLOGIC CHARACTERIZATION AND TISSUE/PLASMA GENOTYPING IN NSCLC

Drug Administration–approved plasma EGFR assays, using ctDNA as a sole determinant of tumor genotype without attempting tissue confirmation is not supported by current guidelines and should ideally only be done within the context of a clinical trial. As targeted therapies can result in rapid clinical improvement, the authors of this review, however, acknowledge that in urgent situations, it may be in a patient’s best interest to initiate treatment based on plasma results alone. If such measures are undertaken, we recommend close surveillance and expedited response assessment if tissue results are not aligned with plasma results. The current performance of ctDNA platforms is not sufficient to discount an alteration if it is not detected in plasma. Rather, if suspicion remains high for a potentially actionable alteration (e.g., molecular testing in a never-smoker) after an insufficient tissue biopsy and negative plasma results, repeat tissue biopsy should be attempted if feasible (Fig. 1). The high specificity of available plasma assays, however, supports selection of treatment based on alterations identified in plasma. Indeed, two recent studies observed comparable outcomes to treatment with third-generation EGFR TKIs when patients with plasma-detected EGFR T790M mutation were compared with those with tissue-detected T790M.74,82 These studies highlight the clinical utility of plasma biomarkers and provide the rationale for studies exploring efficacy of targeted agents in plasma biomarker-selected patients. Encouragingly, in a recent study in which plasma was exclusively used to direct patients who did not undergo repeat biopsies toward treatment with osimertinib, the reported response rate for plasma-positive patients was consistent with the studies in which dual genotyping was performed.83 Notably, a phase II clinical trial using digital droplet PCR to identify EGFR mutations and enable initiation of treatment with erlotinib prior to confirmation of tissue genotype opened to enrollment in early 2017 (NCT02770014).

Serial Plasma Sampling as a Tool for Monitoring Response and Resistance

As a noninvasive, low-risk technology, liquid biopsy is ideally suited for longitudinal analysis and detection of genetic mechanisms of resistance. Indeed, using plasma assays to improve understanding of response and resistance to molecularly targeted therapies is an active area of research. As an example of the potential for liquid biopsies to uncover resistance mechanisms, serial analysis of ctDNA from patients with EGFR-mutant NSCLC treated with third-generation EGFR T790M-specific TKIs was instrumental in identifying novel resistance mechanisms that were subsequently confirmed in tissue specimens. These mechanisms include the tertiary EGFR C797S mutation as well as “loss” of T790M.19,66 Additionally, liquid biopsies identified MET amplification, an alteration previously implicated in resistance to firstand second-generation EGFR TKIs, as a prevalent acquired event among patients who develop resistance to thirdgeneration TKIs.65 Similarly, several studies have confirmed the ability of plasma genotyping to identify acquisition of wellcharacterized alterations conferring resistance to targeted

agents, including secondary ALK resistance mutations in crizotinib-resistant plasma specimens and the EGFR T790M resistance mutation, amplification of MET or HER2, and PIK3CA mutations in plasma samples collected at resistance to first- and second-generation EGFR TKIs.63,65,77 Several studies have demonstrated that plasma indicators of progression may predate radiographic or clinical progression by weeks to months.84,85 However, it is unclear whether molecular relapse, defined as appearance of resistance alterations in plasma or increase in the plasma allelic frequency of pretreatment alterations, should trigger therapeutic decisions. As isolated plasma progression may become a common occurrence with widespread use of commercially available liquid biopsies, it is essential to interpret findings within the context of the current treatment landscape. Specifically, the precedent for continuing treatment beyond radiographic progression and availability of effective agents for treating relapsed or resistant NSCLC should be considered. Although the prognostic implications of resistance mutations that emerge in plasma or increase in plasma mutant allele frequency have not been systematically explored, small studies have shown that clearance of oncogenic alterations from plasma may correlate with depth of response to treatment.63,85 Due to the lack of evidence to support discontinuing therapy for plasma progression, we advise against using plasma dynamics as a primary determinant of progression. Instead, we recommend that the decision to terminate a therapy be based on supporting radiographic and/or clinical signs.

Management Strategies for Cases With TissuePlasma Discordance

As liquid biopsies become integrated into clinical practice, it is conceivable that tissue-plasma discordance may be encountered in cases in which paired genotyping is used (Fig. 1). Because tissue is the accepted gold standard for genotyping and liquid biopsies are an investigational technology in most situations, we anticipate that tissue-only alterations (i.e., failure to detect a known tissue alteration in plasma) will not create diagnostic or therapeutic uncertainty. Although tissue-only alterations may limit the role of longitudinal plasma profiling in molecular surveillance, several factors affect the yield of ctDNA, including metastatic burden and location of metastases.63,74,82 The opposite scenario (i.e., plasma-only detection), however, can be more perplexing, as findings may potentially represent false positives, and indiscriminate use of targeted agents can be detrimental in patients who do not actually harbor the target.86 Reassuringly, in a recent study, confirmatory plasma testing with an alternative assay was able to validate plasma findings for most patients with isolated plasma alterations.74 Durable responses to osimertinib in some patients positive for EGFR T790M by plasma alone support the notion that these alterations are not false positives but, rather, true mutations present at a subclonal fraction in lesions or areas not sampled.74 As molecularly targeted approaches are most successful when directed against ubiquitous alterations that confer asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 625

DAGOGO-JACK ET AL

global susceptibility, it is plausible that responses may be less pronounced for patients with discordant positive plasma. Indeed, studies suggest that some patients with plasma-only alterations will not respond to treatment with molecularly targeted therapy.74 For example, although small numbers limit definitive conclusions, the response rate to osimertinib was considerably lower for patients positive for T790M by plasma alone compared with those positive by tissue and plasma in one study.74 Similarly, several studies have demonstrated that the ratio of a mutant allele to the wild-type allele or other ubiquitous driver events in plasma correlates with response to treatment.65,74,85 Despite these intriguing findings, these studies have been limited in size and have not established a consistent threshold for predicting response to treatment. As such, additional studies are necessary to determine which patients within this heterogeneous group benefit most from treatment with targeted agents. Although this hypothesis is best explored using a large, prospective study to assess the impact of plasma allelic fractions on treatment outcomes, a study that selectively or exclusively enrolls patients that are tissue-negative and plasma-positive for target alterations may be difficult to execute. Given the compelling evidence suggesting that plasma-only alterations are viable targets, it is our practice to offer approved therapies to patients with plasma-only alterations. In contrast, if plasma detects alterations without approved therapies, we strongly favor repeating a tissue biopsy, if feasible (Fig. 1). If repeat tissue sampling is

not feasible, we recommend pursuing standard chemotherapy or immunotherapy prior to exploring investigational agents.

CONCLUSION

Lung cancer is a heterogeneous disease with a variety of clinical and radiographic presentations and diverse underlying histologic and molecular features. Approaches to treatment of lung cancer are similarly versatile, with agent selection heavily influenced by pathologic evaluation. Comprehensive histologic and molecular assessment is a necessary component of successful treatment strategies. Although insight into underlying biology has improved prognosis for subgroups of patients with NSCLC, viable targets remain elusive for many patients. The strides made in treating common histology tumors should inspire efforts to characterize rarer and mixed histology tumors, entities for which generic therapies have been minimally effective. Even among cancers with typical features and defining molecular drivers, the inherently complex and dynamic nature of lung cancer limits the chances of cure. To ensure ongoing success, diagnostic approaches must be as flexible as therapeutic maneuvers. Indeed, therapeutic victory is reliant on an accurate understanding of a cancer’s vulnerability. Given the limitations of serial sampling and the growing body of evidence establishing the reliability of liquid biopsies, this technology will likely be integrated into clinical practice, particularly for those patients with oncogene-addicted lung cancers. Despite its

FIGURE 1. Proposed Framework for Incorporating Plasma Genotyping Results Into Clinical Practice

626 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

HISTOLOGIC CHARACTERIZATION AND TISSUE/PLASMA GENOTYPING IN NSCLC

current strengths, additional modifications are necessary to improve sensitivity of liquid biopsy to level the playing field between tissue and plasma genotyping. Nonetheless, the potential for liquid biopsy to capture relevant alterations outside of the scope of single-site sampling and identify

actionable alterations in patients with inadequate or inaccessible tissue specimens suggests that this technology will have a durable presence in clinical practice as an adjunct for the treatment of patients with advanced NSCLC at diagnosis and during the course of treatment.

References 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2017. CA Cancer J Clin. 2017;67:7-30. 2. Groome PA, Bolejack V, Crowley JJ, et al; IASLC International Staging Committee; Cancer Research and Biostatistics; Observers to the Committee; Participating Institutions. The IASLC Lung Cancer Staging Project: validation of the proposals for revision of the T, N, and M descriptors and consequent stage groupings in the forthcoming (seventh) edition of the TNM classification of malignant tumours. J Thorac Oncol. 2007;2:694-705. 3. Schiller JH, Harrington D, Belani CP, et al; Eastern Cooperative Oncology Group. Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. N Engl J Med. 2002;346: 92-98. 4. Travis WD, Brambilla E, Burke A, et al (eds). WHO Classification of Tumours of the Lung, Pleura, Thymus and Heart. Geneva, Switzerland: WHO Press; 2015. 5. Scagliotti GV, Parikh P, von Pawel J, et al. Phase III study comparing cisplatin plus gemcitabine with cisplatin plus pemetrexed in chemotherapy-naive patients with advanced-stage non-small-cell lung cancer. J Clin Oncol. 2008;26:3543-3551. 6. Scagliotti G, Brodowicz T, Shepherd FA, et al. Treatment-by-histology interaction analyses in three phase III trials show superiority of pemetrexed in nonsquamous non-small cell lung cancer. J Thorac Oncol. 2011;6:64-70. 7. Lynch TJ, Bell DW, Sordella R, et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of nonsmall-cell lung cancer to gefitinib. N Engl J Med. 2004;350:2129-2139. 8. Paez JG, Jänne PA, Lee JC, et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science. 2004;304:1497-1500.

receptor and anaplastic lymphoma kinase tyrosine kinase inhibitors: American Society of Clinical Oncology endorsem*nt of the College of American Pathologists/International Association for the study of lung cancer/association for molecular pathology guideline. J Clin Oncol. 2014;32:3673-3679. 16. Reck M, Rodríguez-Abreu D, Robinson AG, et al; KEYNOTE-024 Investigators. Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung cancer. N Engl J Med. 2016;375:1823-1833. 17. Gainor JF, Dardaei L, Yoda S, et al. Molecular mechanisms of resistance to first- and second-generation ALK inhibitors in ALK-rearranged lung cancer. Cancer Discov. 2016;6:1118-1133. 18. Yu HA, Arcila ME, Rekhtman N, et al. Analysis of tumor specimens at the time of acquired resistance to EGFR-TKI therapy in 155 patients with EGFR-mutant lung cancers. Clin Cancer Res. 2013;19:22402247. 19. Piotrowska Z, Niederst MJ, Karlovich CA, et al. Heterogeneity underlies the emergence of EGFR T790 wild-type clones following treatment of T790M-positive cancers with a third-generation EGFR inhibitor. Cancer Discov. 2015;5:713-722. 20. Ortiz-Cuaran S, Scheffler M, Plenker D, et al. Heterogeneous mechanisms of primary and acquired resistance to third-generation EGFR inhibitors. Clin Cancer Res. 2016;22:4837-4847. 21. Redig AJ, Costa DB, Taibi M, et al. Prospective study of repeated biopsy feasibility and acquired resistance at disease progression in patients with advanced EGFR mutant lung cancer treated with erlotinb in a phase 2 trial. JAMA Oncol. 2016;2:1240-1242. 22. Cakir E, Demirag E, Aydin M, et al. Clinicopathologic features and prognostic significance of lung tumours with mixed histologic patterns. Acta Chir Belg. 2009;109:489-493.

9. Cancer Genome Atlas Research Network. Comprehensive molecular profiling of lung adenocarcinoma. Nature. 2014;511:543-550.

23. Chilosi M, Murer B. Mixed adenocarcinomas of the lung: place in new proposals in classification, mandatory for target therapy. Arch Pathol Lab Med. 2010;134:55-65.

10. Soda M, Choi YL, Enomoto M, et al. Identification of the transforming EML4-ALK fusion gene in non-small-cell lung cancer. Nature. 2007;448:561-566.

24. Deng P, Hu C, Zhou L, et al. Clinical characteristics and prognostic significance of 92 cases of patients with primary mixed-histology lung cancer. Mol Clin Oncol. 2013;1:863-868.

11. Takeuchi K, Soda M, Togashi Y, et al. RET, ROS1 and ALK fusions in lung cancer. Nat Med. 2012;18:378-381.

25. Ruffini E, Rena O, Oliaro A, et al. Lung tumors with mixed histologic pattern. Clinico-pathologic characteristics and prognostic significance. Eur J Cardiothorac Surg. 2002;22:701-707.

12. Kris MG, Johnson BE, Berry LD, et al. Using multiplexed assays of oncogenic drivers in lung cancers to select targeted drugs. JAMA. 2014;311:1998-2006. 13. Lin JJ, Cardarella S, Lydon CA, et al. Five-year survival in EGFR-mutant lung adenocarcinoma treated with EGFR-TKIs. J Thorac Oncol. 2016;11:556-565. 14. Gainor JF, Tan DS, De Pas T, et al. Progression-free and overall survival in ALK-positive NSCLC patients treated with sequential crizotinib and ceritinib. Clin Cancer Res. 2015;21:2745-2752. 15. Leighl NB, Rekhtman N, Biermann WA, et al. Molecular testing for selection of patients with lung cancer for epidermal growth factor

26. Nicholson SA, Beasley MB, Brambilla E, et al. Small cell lung carcinoma (SCLC): a clinicopathologic study of 100 cases with surgical specimens. Am J Surg Pathol. 2002;26:1184-1197. 27. Wagner PL, Kitabayashi N, Chen YT, et al. Combined small cell lung carcinomas: genotypic and immunophenotypic analysis of the separate morphologic components. Am J Clin Pathol. 2009;131:376382. 28. Men Y, Hui Z, Liang J, et al. Further understanding of an uncommon disease of combined small cell lung cancer: clinical features and prognostic factors of 114 cases. Chin J Cancer Res. 2016;28:486-494.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 627

DAGOGO-JACK ET AL

29. Babakoohi S, Fu P, Yang M, et al. Combined SCLC clinical and pathologic characteristics. Clin Lung Cancer. 2013;14:113-119. 30. Adelstein DJ, Tomashefski JF Jr, Snow NJ, et al. Mixed small cell and non-small cell lung cancer. Chest. 1986;89:699-704. 31. Mangum MD, Greco FA, Hainsworth JD, et al. Combined small-cell and non-small-cell lung cancer. J Clin Oncol. 1989;7:607-612. 32. Li DH, Wang C, Chen HJ, et al. Clinical characteristics of the mixed form of neuroendocrine tumor in the lung: a retrospective study in 2501 lung cancer cases. Thorac Cancer. 2015;6:25-30. 33. Travis WD, Rush W, Flieder DB, et al. Survival analysis of 200 pulmonary neuroendocrine tumors with clarification of criteria for atypical carcinoid and its separation from typical carcinoid. Am J Surg Pathol. 1998;22:934-944. 34. Gridelli C, Rossi A, Airoma G, et al. Treatment of pulmonary neuroendocrine tumours: state of the art and future developments. Cancer Treat Rev. 2013;39:466-472.

48. Slaughter DP, Southwick HW, Smejkal W. Field cancerization in oral stratified squamous epithelium; clinical implications of multicentric origin. Cancer. 1953;6:963-968. 49. Martini N, Melamed MR. Multiple primary lung cancers. J Thorac Cardiovasc Surg. 1975;70:606-612. 50. van Rens MT, Eijken EJ, Elbers JR, et al. p53 mutation analysis for definite diagnosis of multiple primary lung carcinoma. Cancer. 2002;94:188-196. 51. Dacic S, Ionescu DN, Finkelstein S, et al. Patterns of allelic loss of synchronous adenocarcinomas of the lung. Am J Surg Pathol. 2005;29:897-902. 52. Girard N, Deshpande C, Lau C, et al. Comprehensive histologic assessment helps to differentiate multiple lung primary nonsmall cell carcinomas from metastases. Am J Surg Pathol. 2009;33:1752-1764. 53. Edge S, Byrd D, Compton C, et al. AJCC Cancer Staging Manual, 7th Ed. New York: Springer; 2010.

35. Cooke DT, Nguyen DV, Yang Y, et al. Survival comparison of adenosquamous, squamous cell, and adenocarcinoma of the lung after lobectomy. Ann Thorac Surg. 2010;90:943-948.

54. Chuang JC, Shrager JB, Wakelee HA, et al. Concordant and discordant EGFR mutations in patients with multifocal adenocarcinomas: implications for EGFR-targeted therapy. Clin Ther. 2016;38:1567-1576.

36. Takamori S, Noguchi M, Morinaga S, et al. Clinicopathologic characteristics of adenosquamous carcinoma of the lung. Cancer. 1991;67:649-654.

55. Shen KR, Meyers BF, Larner JM, et al. Special treatment issues in lung cancer: ACCP evidence-based clinical practice guidelines (2nd edition). Chest. 2007;132:290s-305s.

37. Vassella E, Langsch S, Dettmer MS, et al. Molecular profiling of lung adenosquamous carcinoma: hybrid or genuine type? Oncotarget. 2015;6:23905-23916.

56. Jiang L, He J, Shi X, et al. Prognosis of synchronous and metachronous multiple primary lung cancers: systematic review and meta-analysis. Lung Cancer. 2015;87:303-310.

38. Kurishima K, Ohara G, Kagohashi K, et al. Adenosquamous cell lung cancer successfully treated with gefitinib: A case report. Mol Clin Oncol. 2014;2:282-284.

57. Sholl LM, Aisner DL, Varella-Garcia M, et al; LCMC Investigators. Multi-institutional oncogenic driver mutation analysis in lung adenocarcinoma: the lung cancer mutation consortium experienced. J Thorac Oncol. 2015;10:768-777.

39. Braham E, Ben Rejeb H, Aouadi S, et al. Pulmonary carcinosarcoma with heterologous component: report of two cases with literature review. Ann Transl Med. 2014;2:41. 40. Smyth RJ, Fabre A, Dodd JD, et al. Pulmonary blastoma: a case report and review of the literature. BMC Res Notes. 2014;7:294. 41. Liu X, Jia Y, Stoopler MB, et al. Next-generation sequencing of pulmonary sarcomatoid carcinoma reveals high frequency of actionable MET gene mutations. J Clin Oncol. 2016;34:794-802. 42. Tong JH, Yeung SF, Chan AW, et al. MET amplification and exon 14 splice site mutation define unique molecular subgroups of non-small cell lung carcinoma with poor prognosis. Clin Cancer Res. 2016;22:3048-3056. 43. Schrock AB, Frampton GM, Suh J, et al. Characterization of 298 patients with lung cancer harboring MET exon 14 skipping alterations. J Thorac Oncol. 2016;11:1493-1502. 44. Velcheti V, Rimm DL, Schalper KA. Sarcomatoid lung carcinomas show high levels of programmed death ligand-1 (PD-L1). J Thorac Oncol. 2013;8:803-805. 45. Detterbeck FC, Franklin WA, Nicholson AG, et al. The IASLC Lung Cancer Staging Project: background data and proposed criteria to distinguish separate primary lung cancers from metastatic foci in patients with two lung tumors in the forthcoming eighth edition of the TNM Classification for Lung Cancer. J Thorac Oncol. 2016;11:651-665. 46. Warth A, Macher-Goeppinger S, Muley T, et al. Clonality of multifocal nonsmall cell lung cancer: implications for staging and therapy. Eur Respir J. 2012;39:1437-1442. 47. Wang X, Wang M, MacLennan GT, et al. Evidence for common clonal origin of multifocal lung cancers. J Natl Cancer Inst. 2009;101:560-570.

628 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

58. Sundaresan TK, Sequist LV, Heymach JV, et al. Detection of T790M, the acquired resistance EGFR mutation, by tumor biopsy versus noninvasive blood-based analyses. Clin Cancer Res. 2016;22:11031110. 59. Niederst MJ, Sequist LV, Poirier JT, et al. RB loss in resistant EGFR mutant lung adenocarcinomas that transform to small-cell lung cancer. Nat Commun. 2015;6:6377. 60. Maheswaran S, Sequist LV, Nagrath S, et al. Detection of mutations in EGFR in circulating lung-cancer cells. N Engl J Med. 2008;359: 366-377. 61. Krug AK, Karlovich C, Koestler T, et al. Plasma EGFR mutation detection using a combined exosomal RNA and circulating tumor DNA approach in patients with acquired resistance to first-generation EGFR-TKIs. Presented at: 26th AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics. Boston, MA; November 2015. Abstract B136. 62. Thompson JC, Yee SS, Troxel AB, et al. Detection of therapeutically targetable driver and resistance mutations in lung cancer patients by next-generation sequencing of cell-free circulating tumor DNA. Clin Cancer Res. 2016;22:5772-5782. 63. Sacher AG, Paweletz C, Dahlberg SE, et al. Prospective validation of rapid plasma genotyping for detection of EGFR and KRAS mutations in advanced lung cancer. JAMA Oncol. 2016;2:1014-1022. 64. Paweletz CP, Sacher AG, Raymond CK, et al. Bias-corrected targeted next-generation sequencing for rapid, multiplexed detection of actionable alterations in cell-free DNA from advanced lung cancer patients. Clin Cancer Res. 2016;22:915-922.

HISTOLOGIC CHARACTERIZATION AND TISSUE/PLASMA GENOTYPING IN NSCLC

65. Chabon JJ, Simmons AD, Lovejoy AF, et al. Circulating tumour DNA profiling reveals heterogeneity of EGFR inhibitor resistance mechanisms in lung cancer patients. Nat Commun. 2016;7:11815.

sequencing of circulating free DNA in lung cancer from never-smokers: a proof-of-concept study from BioCAST/IFCT-1002. Clin Cancer Res. 2014;20:4613-4624.

66. Thress KS, Paweletz CP, Felip E, et al. Acquired EGFR C797S mutation mediates resistance to AZD9291 in non-small cell lung cancer harboring EGFR T790M. Nat Med. 2015;21:560-562.

77. Wang Y, Tian PW, Wang WY, et al. Noninvasive genotyping and monitoring of anaplastic lymphoma kinase (ALK) rearranged nonsmall cell lung cancer by capture-based next-generation sequencing. Oncotarget. 2016;7:65208-65217.

67. Mandel P, Metais P. Les acides nucléiques du plasma sanguin chez l’homme. C R Seances Soc Biol Fil. 1948;142:241-243. 68. Koffler D, Agnello V, Winchester R, et al. The occurrence of singlestranded DNA in the serum of patients with systemic lupus erythematosus and other diseases. J Clin Invest. 1973;52:198-204. 69. Jahr S, Hentze H, Englisch S, et al. DNA fragments in the blood plasma of cancer patients: quantitations and evidence for their origin from apoptotic and necrotic cells. Cancer Res. 2001;61:1659-1665. 70. Thress KS, Brant R, Carr TH, et al. EGFR mutation detection in ctDNA from NSCLC patient plasma: a cross-platform comparison of leading technologies to support the clinical development of AZD9291. Lung Cancer. 2015;90:509-515. 71. Marchetti A, Palma JF, Felicioni L, et al. Early prediction of response to tyrosine kinase inhibitors by quantification of EGFR mutations in plasma of NSCLC patients. J Thorac Oncol. 2015;10:1437-1443. 72. Reck M, Hagiwara K, Han B, et al. ctDNA determination of EGFR mutation status in European and Japanese patients with advanced NSCLC: the ASSESS study. J Thorac Oncol. 2016;11:1682-1689. 73. Karachaliou N, Mayo-de las Casas C, Queralt C, et al; Spanish Lung Cancer Group. Association of EGFR L858R mutation in circulating free DNA with survival in the EURTAC trial. JAMA Oncol. 2015;1:149-157. 74. Oxnard GR, Thress KS, Alden RS, et al. Association between plasma genotyping and outcomes of treatment with osimertinib (AZD9291) in advanced non-small-cell lung cancer. J Clin Oncol. 2016;34:3375-3382.

78. Newman AM, Bratman SV, To J, et al. An ultrasensitive method for quantitating circulating tumor DNA with broad patient coverage. Nat Med. 2014;20:548-554. 79. U.S. Food and Drug Administration. FDA approves first blood test to detect gene mutation associated with non-small cell lung cancer. http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ ucm504488.htm. Accessed March 14, 2017. 80. Solomon BJ, Mok T, Kim DW, et al; PROFILE 1014 Investigators. Firstline crizotinib versus chemotherapy in ALK-positive lung cancer. N Engl J Med. 2014;371:2167-2177. 81. Shaw AT, Ou SH, Bang YJ, et al. Crizotinib in ROS1-rearranged nonsmall-cell lung cancer. N Engl J Med. 2014;371:1963-1971. 82. Wakelee HA, Gadgeel SM, Goldman JW, et al. Epidermal growth factor receptor (EGFR) genotyping of matched urine, plasma and tumor tissue from non-small cell lung cancer (NSCLC) patients (pts) treated with rociletinib. J Clin Oncol. 2016;34 (suppl; abstr 9001). 83. Remon J, Caramella C, Jovelet C, et al. Osimertinib benefit in EGFRmutant NSCLC patients with T790M-mutation detected by circulating tumour DNA. Ann Oncol. Epub 2017 Jan 18. 84. Zheng D, Ye X, Zhang MZ, et al. Plasma EGFR T790M ctDNA status is associated with clinical outcome in advanced NSCLC patients with acquired EGFR-TKI resistance. Sci Rep. 2016;6:20913.

75. Reckamp KL, Melnikova VO, Karlovich C, et al. A highly sensitive and quantitative test platform for detection of NSCLC EGFR mutations in urine and plasma. J Thorac Oncol. 2016;11:1690-1700.

85. Karlovich C, Goldman JW, Sun JM, et al. Assessment of EGFR mutation status in matched plasma and tumor tissue of NSCLC patients from a phase I study of rociletinib (CO-1686). Clin Cancer Res. 2016;22:23862395.

76. Couraud S, Vaca-Paniagua F, Villar S, et al; BioCAST/IFCT-1002 investigators. Noninvasive diagnosis of actionable mutations by deep

86. Mok TS, Wu YL, Thongprasert S, et al. Gefitinib or carboplatin-pacl*taxel in pulmonary adenocarcinoma. N Engl J Med. 2009;361:947-957.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 629

SHIRISH M. GADGEEL

Role of Chemotherapy and Targeted Therapy in Early-Stage Non–Small Cell Lung Cancer Shirish M. Gadgeel, MD OVERVIEW On the basis of several randomized trials and meta-analyses, adjuvant chemotherapy is the accepted standard of care for certain patients with early-stage non–small cell lung cancer (NSCLC). Patients with stage II, IIIA, or large (≥ 4 cm) IB tumors are candidates for adjuvant chemotherapy. The survival improvement with adjuvant chemotherapy is approximately 5% at 5 years, though certain trials have suggested that it can be 8% to 10%. Neoadjuvant chemotherapy also has shown a survival advantage, though the volume of data with this approach is far less than that of adjuvant chemotherapy. The combination of cisplatin and vinorelbine is the most well-studied regimen, but current consensus is to use four cycles of any of the platinum-based chemotherapy regimens commonly used as front-line therapy for patients with advanced-stage NSCLC. Trials to define biomarkers that can predict benefit from adjuvant chemotherapy have not been successful, but results of other such trials are still awaited. On the basis of the benefit observed with targeted agents in patients with advanced-stage disease and driver genetic alterations in their tumors, ongoing trials are evaluating the utility of these targeted agents as adjuvant therapy. Similarly, clinical benefit observed with checkpoint inhibitors has prompted assessment of these drugs in patients with early-stage NSCLC. It is very likely, in the future, that factors other than the anatomy of the tumor will be used to select patients with early-stage NSCLC for systemic therapy and that the choice of systemic therapy will extend beyond platinum-based chemotherapy.

he 5-year survival in patients with resected NSCLC ranges from 25% to 75%.1 The primary reason for death in these patients is recurrence of the cancer, which suggests that a proportion of patients with early-stage NSCLC have micrometastatic disease that remains untreated with surgery alone. One of the major advances in the management of NSCLC during the last 15 years has been that adjuvant chemotherapy has become the standard of care on the basis of clinical trials data that showed survival improvement with the use of adjuvant chemotherapy. This review discusses adjuvant chemotherapy, the accepted criteria for the use of adjuvant chemotherapy, the available data about the use of neoadjuvant chemotherapy and adjuvant targeted therapy, and ongoing clinical trials.

ADJUVANT CHEMOTHERAPY

Clinical trials to assess the benefits of adjuvant chemotherapy in patients with resected lung cancer have been conducted since the late 1960s. None of these trials demonstrated a survival advantage. However, a metaanalysis published in 1995 suggested a benefit with use of platinum-based chemotherapy, with a 13% reduction in the risk of death that did not reach statistical significance.2 This

meta-analysis spurred an interest in conducting more trials to assess adjuvant chemotherapy, particular with newer agents approved for use in the late 1990s. A possible reason that adjuvant trials of the past did not demonstrate a survival advantage was the lack of chemotherapy drugs with sufficient efficacy. A Southwest Oncology Group (SWOG) study that evaluated the addition of vinorelbine to cisplatin was among the first studies to demonstrate that combination chemotherapy provided superior survival compared with single-agent cisplatin in patients with advanced NSCLC.3 Subsequently, other chemotherapy agents introduced in the late 1990s, such as pacl*taxel, gemcitabine, and docetaxel, also improved outcomes when combined with a platinum agent compared with single-agent cisplatin in patients with metastatic NSCLC.4-6 These results prompted the evaluation of these newer combinations, such as cisplatin/vinorelbine, as adjuvant therapy. Two trials, one conducted in North America and the other in Europe, assessed the chemotherapy regimen of cisplatin and vinorelbine as adjuvant therapy (Table 1). The North American trial, JBR-10, conducted by National Cancer Institute of Canada, enrolled 482 patients with completely resected stage IB and II NSCLC, and these patients were

From the Department of Oncology, Karmanos Cancer Institute, Wayne State University, Detroit, MI. Disclosures of potential conflicts of interest provided by the author are available with the online article at asco.org/edbook. Corresponding author: Shirish M. Gadgeel, MD, Department of Oncology, Karmanos Cancer Institute, Wayne State University, 4100 John Rd, 4HWCRC, Detroit, MI 48201; email: [emailprotected]. © 2017 American Society of Clinical Oncology

630 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

SYSTEMIC THERAPY IN PATIENTS WITH EARLY-STAGE NSCLC

TABLE 1. Adjuvant Trials Using Newer Platinum-Based Combinations

Chemotherapy

Median Follow-up (Years)

5-Year Survival Benefit (%)

Hazard Ratio, p

Cisplatin/vinorelbine

9.3

0.78, .04

I to IIIA

Cisplatin/vinorelbine

6.3

8.6

0.80, .017

Carboplatin/pacl*taxel

6.1

0.83, .12

Several

5.2

5.4

0.89, .005

Trial

No. of Patients

Stage

JBR-107

482

IB to II

ANITA8

840

CALGB 96339

344

LACE

4,584

I to IIIA

10*

*LACE was a meta-analysis that included trials to evaluate cisplatin-based adjuvant chemotherapy that were conducted after 1995 and had enrolled at least 300 patients.

randomly assigned either to four cycles of cisplatin and vinorelbine chemotherapy or to observation.7 At a median follow-up time of 9.3 years, adjuvant chemotherapy resulted in a significant improvement in survival (adjusted hazard ratio [HR], 0.79; 95% CI, 0.62–1.00; p = .05) and a 5-year survival improvement of 11% (67% with chemotherapy vs. 56% with observation). Though the study design planned for four cycles of chemotherapy, the median number of cycles delivered was three. The major adverse events were fatigue, anorexia, nausea, and febrile neutropenia. Two patients died as a result of chemotherapy-related toxicity. In the European ANITA trial, 840 patients with resected stage IB/IIIA disease were randomly assigned to observation or cisplatin/vinorelbine chemotherapy.8 After a median follow-up time of 76 months, the median survival was 65.7 months in the patients who received chemotherapy and was 43.7 months in the patients who did not (adjusted HR, 0.80; 95% CI, 0.66–0.96; p = .017). Survival at 5 years was improved by 8.6%. Toxicities observed in this trial were similar to those in the JBR-10 trial, and seven patients died as a result of chemotherapy-related toxicities. To gain a better perspective, several meta-analyses have been conducted. The LACE (Lung Adjuvant Cisplatin Evaluation) meta-analysis included all adjuvant trials that evaluated cisplatin-based chemotherapy, were conducted after 1995, and enrolled more than 300 patients (Table 1).10 This analysis showed that the HR for death was 0.89 (95% CI,

KEY POINTS • Adjuvant chemotherapy improves survival by at least 5% at 5 years in patients with early-stage NSCLC. • The stage of the cancer determines if a patient is a candidate for adjuvant chemotherapy. Patients with stage II or IIIA disease are candidates for adjuvant chemotherapy. Certain patients with stage IB NSCLC (tumors ≥ 4 cm) may also be considered for adjuvant therapy. • Four cycles of a platinum-based chemotherapy regimen is the accepted standard. Cisplatin can be substituted with carboplatin in appropriate patients. • Neoadjuvant chemotherapy can result in survival improvement similar to adjuvant chemotherapy. • Ongoing trials will determine the role of targeted therapy and immune therapy in patients with early-stage NSCLC.

0.82–0.96; p = .005) and that the absolute benefit in survival at 5 years with adjuvant chemotherapy was 5.4%. There was a trend toward greater benefit with cisplatin/vinorelbine chemotherapy compared with older regimens used in these trials. A cumulative dose of greater than 300 mg/m2 of cisplatin also improved survival compared with lower doses. Fifty-nine percent of the patients received at least 240 mg/m2 of cisplatin. Among the patients who received chemotherapy, the rate of grade 3 or 4 toxicities was 66%, and the rate of toxicity related death was 0.9%. Similar results were observed in the other meta-analyses.11,12 One trial has evaluated a carboplatin-based treatment regimen as adjuvant therapy. In CALGB 9633, patients with resected stage IB NSCLC were randomly assigned to carboplatin and pacl*taxel or to observation.9 At a median follow-up time of 74 months, there was a trend for survival advantage, but it was not statistically significant (HR, 0.83; 95% CI, 0.64–1.08; p = .12). Because most adjuvant trials were conducted with cisplatin-based regimens, there is a preference to use cisplatin for adjuvant therapy. However, in patients with NSCLC who cannot receive cisplatin because of comorbid illnesses or intolerance of cisplatin, carboplatin is an accepted substitute. Long-term follow-up of the IALT (International Adjuvant Lung Trial), one of the first trials to demonstrate a survival advantage with adjuvant therapy, showed that the benefit was not sustained.13 The HR for survival advantage at 5 years was 0.86, and it declined to an HR of 0.91 at a median follow-up time of 7.5 years. This decline in benefit during longer follow-up was caused by an increase in non–lung cancer-related deaths, which raised the concern of late toxicities from chemotherapy used in this trial. Patients in this study were treated with chemotherapy drugs that are not commonly used. A similar decline in efficacy was not observed with long-term follow-up of either the ANITA or the JBR-10 study, which suggests that choice of chemotherapy agents maybe relevant in terms of long-term toxicities.

VARIABLES INFLUENCING THE USE OF CHEMOTHERAPY

The data from adjuvant trials show that the 5-year survival in patients with resected NSCLC without adjuvant chemotherapy is 25% to 75% and that this rate is improved by 5% to 10% with adjuvant therapy. These data suggest that not everybody needs chemotherapy and that the benefits of adjuvant chemotherapy are modest. Therefore, it is important asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 631

SHIRISH M. GADGEEL

FIGURE 1. Treatment Schema for Adjuvant Therapy

Certain patients with stage IB (particularly patients with tumors ≥ 4 cm) can be considered for adjuvant therapy. Four cycles of adjuvant chemotherapy are usually administered. Chemotherapy is usually started 6 to 12 weeks after surgery. Delayed chemotherapy also may provide survival advantage. Neoadjuvant chemotherapy has shown benefit and can be considered if thought to be in the best interest of the patient. Carboplatin-based therapy can be used if there is a concern about cisplatin use.

to be familiar with factors that predict a greater potential for benefit from adjuvant therapy.

Clinical Variables Influencing Benefit From Adjuvant Therapy

Stage. The LACE meta-analysis showed that the benefit with adjuvant therapy varied according to stage. In patients with stage IA disease (104 patients), the HR was 1.40 (95% CI, 0.95–2.06), which suggests a detriment with adjuvant chemotherapy. These data must be viewed with a level of caution, because the numbers are relatively small. Nonetheless, adjuvant chemotherapy should not be considered in patients with stage IA disease. There was a trend for benefit in patients with stage IB disease (HR, 0.93; 95% CI, 0.78–1.10). The benefit was notable in patients with stage II and stage III disease (HR, 0.83; 95% CI, 0.72–0.95). The test for interaction between survival benefit from adjuvant chemotherapy and stage of NSCLC was significant (test for trend p = .04). Thus, the benefit from adjuvant chemotherapy may be proportionally greater in patients with higher-stage disease, particularly in patients with nodal metastases. The benefits of adjuvant therapy in patients with stage IB disease were assessed in retrospective analyses of two trials. In CALB9633, patients with stage IB disease were randomly assigned to adjuvant carboplatin and pacl*taxel. The overall results did not show a survival benefit in these patients. However, a post hoc analysis demonstrated that adjuvant chemotherapy improved survival in patients with tumors of 4 cm greater from a median of 77 months to a median of 99 months (HR, 0.69; 90% CI, 0.48–0.99; p = .043).9 In patients with tumors less than 4 cm, there was a trend toward inferior survival among patients who received chemotherapy (HR, 1.01; 90% CI, 0.69–1.48; p = .49). A similar retrospective analysis was conducted by the investigators of JBR-10. This study included patients with both stage IB and stage II disease. Although the overall results showed improved survival with adjuvant therapy, the benefit was restricted to patients with stage II disease (HR, 0.68; 95% CI, 0.55–0.97; p = .01), and there was no benefit in patients with stage IB disease (HR, 1.03; 95% CI, 0.7– 1.52; p = .87). Furthermore, analysis restricted to patients with stage IB disease showed that patients with tumors of 4 cm or greater had improved survival with chemotherapy 632 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

(HR, 0.66; 95% CI, 0.39–1.14; p = .13), whereas patients with smaller tumors did not (HR, 1.73; 95% CI, 0.98–3.04; p = .06). It is important to remember that these data from CALGB 9633 and JBR-10 are based on retrospective analyses. However, on the basis of these data and the LACE meta-analysis, the consensus that has emerged is to use adjuvant chemotherapy in patients with tumors that have metastasized to regional lymph nodes (hilar or mediastinal) and in patients with large (≥ 4 cm) tumors (Fig. 1). Age. The median age of U.S. patients with lung cancer is 71, and the median age of patients enrolled in adjuvant trials is approximately 61. Thus, the applicability of the data from these trials to the population of older adults may be limited. Population-based studies have shown that the use of adjuvant chemotherapy is lower in older adults.14,15 In an analysis of the National Cancer Database the odds ratio of receipt of adjuvant chemotherapy in patients older than age 75 was only 0.11 and, among patients age 66 to 75, was 0.38 compared with patients age 55 or younger. Similar results were reported in a population-based study from Ontario, Canada. The JBR-10 investigators evaluated the benefits of adjuvant therapy in patients age 65 or older and found that these patients did derive survival benefit with adjuvant therapy (HR, 0.61; 95% CI, 0.38–0.98; p = .04).7 The LACE meta-analysis investigators also evaluated the benefits of adjuvant chemotherapy in patients age 70 or older and found a trend toward survival advantage in these patients (HR, 0.90; 95% CI, 0.70–1.16; p = .29).10 On the basis of these retrospective analyses, adjuvant chemotherapy should be considered in older adults. However, a level of caution should be exercised in patients age 75 or older. Only 23 patients in this age group were enrolled in the JBR10 trial, and this group experienced a trend toward inferior outcomes with adjuvant chemotherapy (HR, 2.35; 95% CI, 0.84–6.58; p = .09). A proper assessment of the risk-benefit ratio is important before adjuvant chemotherapy is initiated in very old adults. Cisplatin may have a greater propensity to cause toxicities in these patients, also; therefore, carboplatin should be considered in older patients. Performance status. Stage and performance status are the most important prognostic factors in patients with NSCLC. All of the adjuvant trials excluded patients who had poor

SYSTEMIC THERAPY IN PATIENTS WITH EARLY-STAGE NSCLC

performance statuses. Some patients do experience a decline in performance status after thoracic surgery. Utility of adjuvant therapy in these patients is unclear. However, a recent analysis of data in the National Cancer Database showed that delaying chemotherapy from the accepted standard of 6 to 9 weeks after surgery does not reduce the survival benefit from adjuvant therapy.16 Similar data were also reported by Booth et al.17 In their retrospective analysis of the Ontario Cancer Registry, the median time to start of adjuvant chemotherapy was 8 weeks. However, a third of the patients started adjuvant chemotherapy after 10 weeks, and their survival was not inferior to patients who received chemotherapy within 10 weeks of surgery. Thus, even if patients require a longer time to recover from lung cancer surgery, adjuvant therapy could be considered after the patient has recovered.

Pathologic Variables

Histology. Histology has emerged as an important determinant of choice of chemotherapy in patients with NSCLC, particularly with the use of pemetrexed. The combination of cisplatin and pemetrexed was evaluated in a randomized phase II trial; its primary objective was to assess the clinical feasibility of delivering this combination as adjuvant therapy compared with cisplatin/vinorelbine.18 Clinical feasibility was defined as no grade 4 neutropenia/thrombocytopenia; grade 3 or 4 febrile neutropenia or thrombocytopenia with bleeding; and no grade 3 or 4 nonhematologic toxicity. The study showed that feasibility rates were 95.5% with the pemetrexed combination versus 75.4% with the vinorelbine combination. Dose delivery was 90% of planned dose with pemetrexed combination and was 66% with the vinorelbine combination. Overall toxicity also was less with the cisplatin/ pemetrexed combination. ECOG1505 evaluated the benefits of added bevacizumab to a variety of cisplatin-based chemotherapy regimens.19 The study failed to show any benefit with the addition of bevacizumab. In a retrospective analysis of the trial, no difference in outcomes was observed with any of the chemotherapy regimens used. However, among patients with nonsquamous disease (patients whose tumors had any non–small cell histology other than squamous), the cisplatin/pemetrexed combination had significantly less grades 3 to 5 toxicities compared with other regimens (p < .001). No notable differences in toxicities were observed in regimens used in patients with squamous cell disease. Thus, results of two separate studies, TREAT and E1505, suggest that the combination of cisplatin/pemetrexed is better tolerated as adjuvant therapy. The data from E1505 also suggest that platinum-based chemotherapy combinations used in patients with advanced-stage disease can be considered in the adjuvant setting. Histology also may have relevance for the choice of the platinum analog. CISCA (cisplatin vs. carboplatin) was a meta-analysis of randomized trials to compare cisplatin- with carboplatin-based treatments in patients with advanced NSCLC.20 This meta-analysis showed that patients who received carboplatin-based therapy had a nonsignificant

increase in the hazard of mortality compared with patients who received cisplatin-based chemotherapy (HR, 1.07; 95% CI, 0.99–1.15; p = .100). However, in patients with nonsquamous disease, carboplatin-based regimens were associated with statistically significant inferior survival (HR, 1.12; 95% CI, 1.01–1.23; p = .098); there was no difference in outcome among patients with squamous histology. Though these data are in patients with advanced cancer, the CISCA metaanalysis suggests that cisplatin should be the preferred platinum for adjuvant therapy, particularly in patients with nonsquamous histology. It is important to note that this meta-analysis included trials conducted before the availability of pemetrexed, an effective chemotherapeutic agent in patients with nonsquamous NSCLC. Other pathologic features. Certain tumor pathologic features are associated with worse prognosis. These include visceral pleural invasion and angiolymphatic invasion.21-24 Though it is accepted that presence of these features is associated with worse outcomes, it is not known whether adjuvant therapy in patients with tumors that have these features, but do not meet any other criteria for adjuvant therapy, provides any survival benefit. If these features are present, the prognostic implications of these features with the patient should be discussed; adjuvant therapy occasionally may be considered in patients with tumors of 4 cm or smaller without lymph node metastases that have these pathologic features. The National Comprehensive Cancer Network guidelines state that pathologic features of poorly differentiated tumor, vascular invasion, visceral pleural invasion, and lung neuroendocrine tumors should be considered high-risk features and that the presence of these high-risk features should be included in the decision-making process about adjuvant therapy.25

NEOADJUVANT CHEMOTHERAPY

Systemic chemotherapy in patients with NSCLC has the capacity to treat not only the micrometastatic disease but also clinically evident cancer in the chest. Thus, neoadjuvant chemotherapy has the potential to address all sites of disease simultaneously. In addition, preoperative chemotherapy may be better tolerated than postoperative chemotherapy. For these reasons, neoadjuvant chemotherapy in patients with early-stage NSCLC has generated interest. In the early 1990s, two small randomized trials demonstrated that neoadjuvant chemotherapy can improve survival. Rosell et al26 evaluated neoadjuvant cisplatin, mitomycin C, and ifosfamide in patients with stage IIIA disease. Similarly, Roth et al27 evaluated neoadjuvant cisplatin, etoposide, and cyclophosphamide in the same group of patients. Both trials showed a survival advantage and sparked additional assessment in the neoadjuvant approach. One of the largest studies conducted in the United States to evaluate neoadjuvant chemotherapy was S9900.28 Patients with stage IB/IIIA disease were randomly assigned to surgery alone or to three cycles of carboplatin/pacl*taxel followed by surgery. The plan was to enroll 300 patients per asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 633

SHIRISH M. GADGEEL

arm, but the study closed after 354 patients were enrolled in the study because data about benefits of adjuvant therapy became available during the conduct of the study; it became challenging to continue to randomly assigned patients to surgery alone. Among the patients enrolled, there was an improvement in survival (median, 62 vs. 41 months; HR, 0.79; p = .11) and progression-free survival (median, 33 vs. 20 months; HR, 0.80; p = .10) with neoadjuvant chemotherapy, but this improvement was not statistically significant, possibly because of lower-than-planned enrollment. The surgical resection rate among randomly assigned patients was similar at 87% among patients who had surgery alone and 84% among patients who had neoadjuvant chemotherapy. Postoperative adverse events were similar among patients who did and did not undergo neoadjuvant chemotherapy. However, the postoperative mortality was higher among patients in the neoadjuvant chemotherapy arm who underwent pneumonectomy (4 of 24 vs. 0 of 26 in the surgery-alone arm). Another trial, CHEST (chemotherapy in early-stage NSCLC trial) evaluated neoadjuvant cisplatin and gemcitabine followed by surgery versus surgery alone.29 This study enrolled only 270 of the planned 700 patients. Both progression-free survival (HR, 0.70; p = .003) and overall survival (HR, 0.63; p = .02) were significantly improved with neoadjuvant chemotherapy. The study found that the survival benefit with neoadjuvant chemotherapy was restricted to patients with stage IIB/IIIA disease (HR, 0.42; p < .001). In patients with stage IB/IIA disease, the majority of whom were stage IB, there was no improvement in survival with neoadjuvant chemotherapy (HR, 1.02; p = .94). Several meta-analyses of studies to evaluate neoadjuvant chemotherapy have been performed.30-32 These meta-analyses suggest that neoadjuvant chemotherapy does improve survival, with an absolute benefit of 5% to 6% at 5 years; this rate is similar to the benefit observed with adjuvant chemotherapy. Few trials have compared neoadjuvant and adjuvant chemotherapy. A prominent study was the NATCH trial that randomly assigned patients with stage I or stage II, T3N1 NSCLC to surgery alone, to neoadjuvant chemotherapy with carboplatin and pacl*taxel followed by surgery, or to surgery followed by adjuvant chemotherapy with the same regimen.33 The study failed to demonstrate improvement in survival either with neoadjuvant or adjuvant chemotherapy. It is possible that the reason for lack of survival advantage with chemotherapy was that greater than 70% of the patients in each arm had stage I disease—a group that the LACE meta-analysis showed may not derive benefit from chemotherapy. However, outcomes with neoadjuvant chemotherapy were similar to adjuvant chemotherapy, which suggested that there is neither an advantage nor a disadvantage with neoadjuvant chemotherapy compared with adjuvant therapy. However, 97% of the patients in the neoadjuvant group started chemotherapy, compared with 66% in the adjuvant group (p < .0001), which suggests that neoadjuvant chemotherapy may be better tolerated than adjuvant chemotherapy. 634 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

An indirect meta-analysis to compare preoperative and postoperative chemotherapy was conducted by Lim et al.34 It is important to note that the majority of the trials included in this analysis did not directly compare the two approaches. More than 10,000 patients were included in this analysis, though the number of adjuvant trials far exceeded trials that assessed neoadjuvant chemotherapy. This analysis showed that both overall survival and disease-free survival were similar with adjuvant or neoadjuvant chemotherapy.34 In summary, chemotherapy improves survival in patients with stage IB/IIIA NSCLC who have undergone surgical resection, irrespective of administration before or after surgery. The data for adjuvant chemotherapy are far more robust than the data for neoadjuvant chemotherapy; therefore, adjuvant chemotherapy should be the preferred approach. However, if there are concerns about surgical resection, then in the author's opinion neoadjuvant chemotherapy could be considered and the feasibility of surgical resection could be re-evaluated after chemotherapy.

PREDICTORS OF CHEMOTHERAPY SENSITIVITY

Only a proportion of patients with advanced-stage and earlystage NSCLC benefit from currently available chemotherapy drugs. Therefore, there has been an interest in identifying markers that can predict for benefit of, or lack thereof from, chemotherapy drugs. Several molecular markers, such as ERCC1, RRM1, BRCA1, and thymidylate synthase (TS), have been assessed independently and collectively to identify patients most likely to benefit from specific chemotherapy drugs. The expectation was that assessment of such molecules may spare patients who are unlikely to derive benefit from toxicities of chemotherapy drugs. However, despite promising results in pilot studies, randomized studies have failed to demonstrate the predictive utility of these markers.35-37 The ITACA (International Tailored Chemotherapy Adjuvant Therapy) phase III trial is evaluating the predictive utility of the mRNA expression levels of molecular markers ERCC1 and TS.38 Patients will undergo assessment of both markers by quantitative reverse transcriptase polymerase chain reaction. Patients were randomly assigned to investigator’s choice of platinum-based chemotherapy or chemotherapy defined by the molecular markers. Patients with tumors that had high ERCC1 and high TS received single-agent docetaxel; patients with high ERCC1 and low TS received single-agent pemetrexed; patients with low ERCC1 and high TS received cisplatin and gemcitabine; finally, patients with low ERCC1 and low TS received cisplatin and pemetrexed. The study has completed enrollment, and the results are awaited. Currently, no factor other than histology is predictive of benefit of, or lack thereof from, specific chemotherapy in patients with NSCLC. Despite this, several laboratories continue to perform assessments of these markers in tumors of patients with NSCLC. For now, therapy should not be based on the results of these markers.

SYSTEMIC THERAPY IN PATIENTS WITH EARLY-STAGE NSCLC

Genomic Markers

Lung cancer, like most cancers, is a result of genetic alterations that initiate an oncogenic phenotype in the affected tissue. There is a notable interest in identification of specific gene signatures that can provide prognostic and predictive guidance. Genomic assessment is conducted routinely to determine adjuvant therapy for breast cancer.39 Various investigators have proposed different genomic signatures as prognostic markers and/or predictive markers in resected NSCLC by using different testing platforms. Chen et al40 analyzed 125 patients with resected stage I to III NSCLC with different histologic subtypes by microarray gene expression analysis and identified a prognostic score that was based on the expression of five genes. All of the tissues analyzed were fresh-frozen. The patients with high gene scores had a lower median survival than patients with low gene scores (20 vs. 40 months; p < .001). In a multivariable analysis, the five-gene score was significantly (p = .03) associated with death, as were patient age and tumor stage. Kratz et al41 defined a 14-gene signature on paraffin-embedded tumor specimens of patients with nonsquamous NSCLC. The investigators were able to categorize patients into three distinct prognostic categories. The 5-year survival rates in the three different categories were 74% in low-risk group, 58% in the intermediate-risk group, and 45% in the high-risk group. Among 330 total patients with stage I NSCLC, the median survival times were 113 months in the 78 low-risk patients, 88 months in the 104 intermediate-risk patients, and 70 months in the 151 high-risk patients. The investigators confirmed the results in an independent cohort of resected NSCLC obtained from a database in China. The ability to use formalin-fixed and paraffin-embedded tissue to generate such a gene expression–based score has greater clinical utility than signatures derived from studies that used fresh-frozen tissues. Apart from assessment of genetic signatures, investigators also have assessed prognosis on the basis of levels of certain microRNAs (miRNA), which are small noncoding RNAs that function in regulation of gene expression by targeting either the 3-prime or 5-prime region of specific mRNAs. Expression levels of various miRNAs are altered in cancers, including lung cancer.42 Several investigators have identified that expression of certain miRNAs may have prognostic relevance. This is a new and exciting area of research and could complement the prognostic utility of gene expression signatures. Although assessment of gene expression to predict prognosis and benefit from adjuvant chemotherapy is a rational approach, none of the studies to date have proven the clinical value of gene signatures in prospective trials. In addition, the technologies, the tissue sources, and gene sets have varied among reports, and this variance severely limits the clinical applicability of gene signatures in current practice.

ADJUVANT TARGETED THERAPY

EGFR Inhibitors as Adjuvant Therapy

One of the major advances in the management of NSCLC is the identification of driver genetic alterations that can

be targeted for therapeutic benefit. The first driver genetic alteration that was successfully targeted for therapeutic benefit was EGFR mutation. Approximately 10% to 15% of patients with NSCLC, particularly those with adenocarcinomas of the lung, have mutations in the tyrosine kinase domain of the EGFR gene. Exon 19 deletions and the point mutation L858R in exon 21 constitute 90% of all EGFR mutations identified in NSCLC. Randomized trials have shown that patients with advanced NSCLC who are positive for these EGFR mutations derive greater clinical benefit from EGFR tyrosine kinase inhibitors (TKIs) than standard front-line chemotherapy.43 On the basis of these data, there is a clear interest in evaluation of these drugs as adjuvant therapy in patients with EGFR mutation–positive NSCLC. Retrospective analysis has suggested that adjuvant EGFR TKIs can provide clinical benefit in patients with EGFR mutation–positive NSCLC.44 However, as yet, there are no conclusive prospective data to support the use of these drugs as adjuvant therapy. The largest trial to evaluate EGFR TKIs as adjuvant therapy was the RADIANT trial.45 In this trial, patients were eligible if their tumors were positive for EGFR expression as assessed by immunohistochemistry or fluorescent in situ hybridization. Patients were randomly assigned in a 2:1 fashion to receive 150 mg daily of erlotinib for 2 years or to placebo. Patients who were candidates for adjuvant chemotherapy received this treatment before start of the study therapy. The primary endpoint of the study was improvement in disease-free survival. The study failed to demonstrate a disease-free survival or survival advantage with the use of adjuvant erlotinib. Of the 973 patients enrolled in the study, 161 patients had exon 19 deletion or L858R EGFR mutations. In this population of patients, the disease-free survival was superior with erlotinib (HR, 0.61; 95% CI, 0.384–0.981; p = .0391). However, the difference could not be considered statistically significant because of hierarchical testing that allowed assessment of statistical significance of secondary endpoints only if the primary endpoint was statistically significant. At the time of data analysis, the follow-up was too limited to assess survival differences. In another randomized trial, BR19, gefitinib was evaluated as adjuvant therapy in all patients and was not restricted to EGFR mutation–positive NSCLC. The study closed early after gefitinib failed to show survival advantage in patients with advanced NSCLC in the ISEL trial. The study enrolled 503 patients with resected stage IB/IIIA NSCLC, and patients were randomly assigned in a 1:1 manner to gefitinib or placebo. Patients received gefitinib for a median of 4.8 months. Gefitinib did not improve survival in these patients. Only 15 patients had EGFR mutations—seven of whom received gefitinib, and eight of whom received placebo. There was a suggestion of worse survival among the seven patients with EGFR mutation–positive tumors who received gefitinib (HR, 3.16; 95% CI, 0.61–16.45; p = .15). These results have to be viewed with some level of caution because of the small number of patients with EGFR mutations, early closure of the study, and the short duration of gefitinib administered to these patients. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 635

SHIRISH M. GADGEEL

FIGURE 2. ALCHEMIST (NCT02194738) Trial Design

The SELECT trial was a single-arm, multicenter, phase II study to assess erlotinib as adjuvant therapy in patients with stage IA to IIIA EGFR mutation–positive NSCLC.46 Patients received erlotinib for 2 years. The study was designed to assess the ability of adjuvant erlotinib to improve 2-year disease-free survival from 76% (on the basis of historical data) to 86%. The 2-year disease-free survival in the 100 patients enrolled in the study was 89%. Of the 29 patients who experienced recurrence, 25 developed disease recurrence after erlotinib treatment was stopped. The median time to recurrence after treatment stopped was 8.5 months. In addition, the duration of treatment was significantly shorter in patients who had recurrence compared with those who were recurrence free at the time of data analysis (p = .027). These data suggest that duration of therapy with an EGFR TKI may have relevance in the adjuvant setting. Data from studies in gastrointestinal stromal tumors (GIST), which also have driver genetic alteration, have shown that 3 years of adjuvant therapy with imatinib provides greater benefit than 1 year of therapy.47 In that trial, only 69% of the patients completed at least 22 months of therapy, whereas only half of the patients in the RADIANT trial completed all 24 months of therapy. Thus, it is possible that prolonged delivery of adjuvant EGFR TKIs is essential for improved outcomes in EGFR mutation–positive NSCLC. Several ongoing or recently completed trials are addressing the use of adjuvant EGFR TKIs. The largest such effort is the ALCHEMIST trial (NCT02194738), conducted by all of the cooperative oncology groups in the United States under the leadership of the National Cancer Institute (Fig. 2). Patients with stage IB to IIIA disease will undergo molecular analysis. If their tumor has an EGFR mutation, then they could 636 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

enter the EGFR mutation substudy that plans to enroll 410 patients and will randomly assign patients to erlotinib for 2 years or to placebo. The primary endpoint of the study is overall survival. The ALCHEMIST trial also has a substudy to evaluated the role of adjuvant crizotinib in patients with resected ALK-positive NSCLC. Other trials (C-TONG 1104, NCT01405079; WJOG6401L) are comparing adjuvant EGFR TKIs to adjuvant platinum-based chemotherapy by using disease-free survival as the primary endpoint. Finally, a trial in the United States (NCT01746251) is randomly assigning patients to receive adjuvant afatinib for either 3 months or 2 years to assess the relevance of duration of therapy. The available data do not conclusively demonstrate that adjuvant EGFR TKIs provide a survival advantage in patients with early-stage lung cancer. Therefore, adjuvant EGFR TKIs currently are not considered the standard of care.

Bevacizumab

Bevacizumab is a monoclonal antibody that targets VEGF. Angiogenesis is an important component of carcinogenic phenotype; therefore, targeting oncogenic angiogenesis for therapeutic benefit has been of interest for a long time. E4599 demonstrated that the addition of bevacizumab to the chemotherapy of carboplatin and pacl*taxel improved survival in patients with advanced nonsquamous NSCLC.48 Other studies conducted did not document an overall survival benefit in patients who received bevacizumab with chemotherapy. The survival advantage observed with the addition of bevacizumab to chemotherapy in E4599 was the basis of E1505, a study that evaluated the efficacy of bevacizumab added to adjuvant chemotherapy.19 The study randomly

SYSTEMIC THERAPY IN PATIENTS WITH EARLY-STAGE NSCLC

assigned 1,501 patients with stage IB to IIIA disease who had undergone surgical resection with appropriate lymph node sampling to receive four cycles of platinum-based adjuvant chemotherapy with or without 1 year of bevacizumab. Among the patients randomly assigned to receive bevacizumab, only 37% of the patients completed an entire year of treatment. Neutropenia, hypertension, and overall grade 3 to 5 toxicities were more frequent in patients who received bevacizumab. There was no difference in survival among patients who were randomly assigned to receive bevacizumab and patients who were not (HR, 0.99).

IMMUNOTHERAPY

The ability to evade immune surveillance is an important aspect of the oncogenic phenotype. Thus, activation or restoration of immune surveillance could treat and eradicate cancers.

Vaccines

Therapeutic vaccines have been evaluated as adjuvant therapy in patients with resected lung cancer. Both tumor-based vaccines and peptide-based vaccines have been evaluated in lung cancer. The largest study to test this strategy in the adjuvant setting was the MAGRIT trial, a study that evaluated the melanoma-associated antigen (MAGE)–A3 vaccine.49 MAGE-A3 is expressed on the surface of several cancers, including NSCLC, and is not expressed on normal tissues other than the placenta and testis. A randomized phase II study suggested that an adjuvant MAGE-A3 vaccine could enhance both disease-free survival and overall survival.50 MAGRIT was a randomized phase III study that evaluated more than 13,000 patients; 2,312 of these patients were randomly assigned in a 2:1 manner to receive the vaccine or placebo during 27 months. Of the patients randomly assigned, approximately 50% had received adjuvant chemotherapy. The schedule of the vaccine and placebo was similar to the phase II trial of 13 doses administered over 27 months. The study failed to show an improvement in disease-free survival in the overall population (HR, 1.02; p = .74) or in patients who had received adjuvant chemotherapy (HR, 1.10; p = .36).

Other vaccines have been evaluated for the management of patients with NSCLC, both in early and advanced stages.51-54 Though some of the trials have shown very promising results, none have conclusively demonstrated survival improvement in a randomized phase III study. Whether this approach will prove to provide meaningful benefit as adjuvant therapy remains to be seen.

Checkpoint Inhibitors

Tumors can evade immune surveillance by activating inhibitory checkpoints on T cells. Two of the most well-studied checkpoints are PD-1 and its ligand, PD-L1. Inhibition of the PD-1 signaling pathway by antibodies directed against either PD-1 or PD-L1 has led to dramatic clinical benefits in a minority of patients with advanced NSCLC. Three drugs that target the PD-1 signaling pathway have been approved for the treatment of advanced NSCLC. Pembrolizumab is a fully humanized IgG4 antibody against PD-1. On the basis of randomized phase III studies, the drug has been approved for front-line therapy in patients with advanced NSCLC whose tumors have high PD-L1 expression.55 It is also approved for patients with NSCLC who have PD-L1–positive cancers and who were previously treated with a platinum-based chemotherapy.56 Nivolumab is also a fully humanized IgG4 antibody against PD-1. Currently, it is approved for patients with advanced NSCLC who were previously treated with platinum-based chemotherapy irrespective of the tumor PD-L1 expression.57,58 Recently, atezolizumab, a fully humanized antibody that targets PD-L1, was approved for use in patients with advanced NSCLC who were previously treated with platinum-based chemotherapy.59 There is a great deal of interest in the evaluation of these agents in early-stage NSCLC. Several randomized trials are evaluating these agents in the adjuvant setting. The ALCHEMIST trial was modified recently to randomly assign patients whose tumors are not EGFR mutation–positive or ALK mutation–positive to nivolumab or placebo. Results of these trials are eagerly awaited. Whether the benefit will be restricted to patients with tumors that express a high level of PD-L1 or those that have other biomarkers remains to be seen.

References 1. Goldstraw P, Crowley K, Chansky K, et al. The IASLC lung cancer staging project: proposals for the revision of the TNM stage groupings in the forthcoming (seventh) edition of the TNM classification of malignant tumors. J Thorac Oncol. 2007;2:706-714. 2. Non–Small Cell Lung Cancer Collaborative Group. Chemotherapy in non– small cell lung cancer: a meta-analysis using updated data on individual patients from 52 randomised clinical trials. BMJ. 1995;311:899-909. 3. Wozniak AJ, Crowley JJ, Balcerzak SP, et al. Randomized trial comparing cisplatin with cisplatin plus vinorelbine in the treatment of advanced non–small cell lung cancer: a Southwest Oncology Group study. J Clin Oncol. 1998;16:2459-2465.

4. Bonomi P, Kim K, Fairclough D, et al. Comparison of survival and quality of life in advanced non-small-cell lung cancer patients treated with two dose levels of pacl*taxel combined with cisplatin versus etoposide with cisplatin: results of an Eastern Cooperative Oncology Group trial. J Clin Oncol. 2000;18:623-631. 5. Le Chevalier T, Scagliotti G, Natale R, et al. Efficacy of gemcitabine plus platinum chemotherapy compared with other platinum containing regimens in advanced non-small-cell lung cancer: a meta-analysis of survival outcomes. Lung Cancer. 2005;47:69-80. 6. Fossella F, Pereira JR, von Pawel J, et al. Randomized, multinational, phase III study of docetaxel plus platinum combinations versus

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 637

SHIRISH M. GADGEEL

vinorelbine plus cisplatin for advanced non-small-cell lung cancer: the TAX 326 study group. J Clin Oncol. 2003;21:3016-3024. 7. Butts CA, Ding K, Seymour L, et al. Randomized phase III trial of vinorelbine plus cisplatin compared with observation in completely resected stage IB and II non–small cell lung cancer: updated survival analysis of JBR-10. J Clin Oncol. 2010;28:29-34. 8. Douillard JY, Rosell R, De Lena M, et al. Adjuvant vinorelbine plus cisplatin versus observation in patients with completely resected stage IB-IIIA non–small-cell lung cancer (Adjuvant Navelbine International Trialist Association [ANITA]): a randomised controlled trial. Lancet Oncol. 2006;7:719-727. 9. Strauss GM, Herndon JE, Maddaus MA, et al. Adjuvant pacl*taxel plus carboplatin compared with observation in stage IB non–small cell lung cancer: CALGB 9633 with the Cancer and Leukemia Group B, Radiation Therapy Oncology Group, and North Central Cancer Treatment Group study groups. J Clin Oncol. 2008;26:5043-5051. 10. Pignon JP, Tribodet H, Scagliotti GV, et al. Lung adjuvant cisplatin evaluation: a polled analysis by the LACE collaborative group. J Clin Oncol. 2008;26:3552-3559. 11. NSCLC Meta-Analyses Collaborative Group. Adjuvant chemotherapy with or without postoperative radiotherapy in operable non–small cell lung cancer: two meta-analyses of individual patient data. Lancet. 2010;375:1267-1277. 12. Zhong C, Liu H, Jiang L, et al. Chemotherapy plus best supportive care versus best supportive care in patients with non–small cell lung cancer: a meta analysis of randomized controlled trials. PLoS One. 2013;8:e58466. 13. Arriagada R, Dunant A, Pignon JP. Long-term results of the international adjuvant trial evaluating adjuvant cisplatin-based adjuvant chemotherapy in resected lung cancer. J Clin Oncol. 2010;28:35-42. 14. Rajaram R, Paruch JL, Mohanty S, et al. Patterns and predictors of chemotherapy use for resected non–small cell lung cancer. Ann Thorac Surg. 2016;101:533-540. 15. Booth CM, Shepherd FA, Peng Y, et al. Adjuvant chemotherapy for non–small cell lung cancer: practice patterns and outcomes in general population of Ontario, Canada. J Thorac Oncol. 2012;7:559-566. 16. Salazar MC, Rosen JE, Wang Z, et al. Association of delayed adjuvant chemotherapy with survival after lung cancer surgery. JAMA Oncol. Epub 2017 Jan 5.

22. Fibla JJ, Cassivi SD, Brunelli A, et al. Re-evaluation of the prognostic value of visceral pleura invasion in Stage IB non–small cell lung cancer using the prospective multicenter ACOSOG Z0030 trial data set. Lung Cancer. 2012;78:259-262. 23. Schuchert MJ, Schumacher L, Kilic A, et al. Impact of angiolymphatic and pleural invasion on surgical outcomes for stage I non–small cell lung cancer. Ann Thorac Surg. 2011;91:1059-1065. 24. Kato T, Ishikawa K, Aragaki M, et al. Angiolymphatic invasion exerts a strong impact on surgical outcomes for stage I lung adenocarcinoma, but not non-adenocarcinoma. Lung Cancer. 2012;77:394-400. 25. National Comprehensive Cancer Network. Non–small Cell Lung Cancer, version 4.2017. https://www.nccn.org/professionals/physician_gls/ pdf/nscl.pdf. Accessed February 8, 2017. 26. Rosell R, Gómez-Codina J, Camps C, et al. A randomized trial comparing preoperative chemotherapy plus surgery with surgery alone in patients with non–small cell lung cancer. N Engl J Med. 1994;330:153-158. 27. Roth JA, Fossella F, Komaki R, et al. A randomized trial comparing perioperative chemotherapy and surgery with surgery alone in resectable stage IIIA non–small cell lung cancer. J Natl Cancer Inst. 1994;86:673-680. 28. Pisters KM, Vallieres E, Crowley JJ, et al. Surgery with or without preoperative pacl*taxel and carboplatin in early stage non–small cell lung cancer: Southwest Oncology Group Trial S9900, an intergroup randomized phase III trial. J Clin Oncol. 2010;28:1843-1849. 29. Scagliotti GV, Pastorino U, Vansteenkiste JF, et al. Randomized phase III study of surgery alone or surgery plus preoperative cisplatin and gemcitabine in stages IB to IIIA non-small-cell lung cancer. J Clin Oncol. 2012;30:172-178. 30. Berghmans T, Paesmans M, Meert AP, et al. Survival improvement in resectable non–small cell lung cancer with (neo)adjuvant chemotherapy: results of a meta-analysis of the literature. Lung Cancer. 2005;49:13-23. 31. Burdett SS, Stewart LA, Rydzewska L. Chemotherapy and surgery versus surgery alone in non–small cell lung cancer. Cochrane Database Syst Rev. 2007;(3):CD006157. 32. NSCLC Meta-analysis Collaborative Group. Preoperative chemotherapy for non–small cell lung cancer: a systematic review and meta-analysis of individual patient data. Lancet. 2014;383:1561-1571.

17. Booth CM, Shepherd FA, Peng Y, et al. Time to adjuvant chemotherapy and survival in non–small cell lung cancer. Cancer. 2013;119:1243-1250.

33. Felip E, Rosell R, Maestre JA, et al. Preoperative chemotherapy plus surgery versus surgery plus adjuvant chemotherapy versus surgery alone in early-stage non–small cell lung cancer. J Clin Oncol. 2010;28:3138-3145.

18. Kreuter M, Vansteenkiste J, Fischer JR, et al Randomised phase 2 trial on refinement of early-stage NSCLC adjuvant chemotherapy with cisplatin and pemetrexed versus cisplatin and vinorelbine: the TREAT trial. Ann Oncol. 2013;24:986-992.

34. Lim E, Harris G, Patel A, et al. Preoperative versus postoperative chemotherapy in patients with resectable non–small cell lung cancer: systematic review and indirect comparison meta-analysis of randomized trials. J Thorac Oncol. 2009;4:1380-1388.

19. Wakelee HA, Dahlberg SE, Keller SM, et al. E1505: adjuvant chemotherapy +/− bevacizumab for early stage NSCLC—outcomes based on chemotherapy subsets. J Clin Oncol. 2016;34 (abstract 8507).

35. Massuti B, Rodriguez-Paniagua JM, Cobo Dols M, et al. Results phase III trial customized adjuvant CT after resection of NSCLC with lymph node metastases SCAT: a Spanish lung cancer group trial. J Thorac Oncol. 2015;10:s180.

20. Ardizzoni A, Boni L, Tiseo M, et al. Cisplatin- versus carboplatin-based chemotherapy in first-line treatment of advanced non–small cell lung cancer: an individual patient data meta-analysis. J Natl Cancer Inst. 2007;99:847-857.

36. Wislez M, Barlesi F, Besse B, et al. Customized adjuvant phase II trial in patients with non–small cell lung cancer: IFCT-0801 TASTE. J Clin Oncol. 2014;32:1256-1261.

21. Kudo Y, Saji H, Shimada Y, et al. Impact of visceral pleural invasion on the survival of patients with non–small cell lung cancer. Lung Cancer. 2012;78:153-160.

638 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

37. Bepler G, Williams C, Schell MJ, et al. Randomized international phase III trial of ERCC1 and RRM1 expression-based chemotherapy versus gemcitabine/carboplatin in advanced non–small cell lung cancer. J Clin Oncol. 2013;31:2404-2412.

SYSTEMIC THERAPY IN PATIENTS WITH EARLY-STAGE NSCLC

38. Novello S, Grohe C, Geissler M, et al. Preliminary results of the international tailored chemotherapy adjuvant trial: the ITACA trial. J Thorac Oncol. 2015;10:s179.

resected MAGE-A3-positive non–small cell lung cancer (MAGRIT): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol. 2016;17:822-835.

39. Harris LN, Ismaila N, McShane LM, et al. Use of biomarkers to guide decisions on adjuvant systemic therapy for women with early-stage invasive breast cancer: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol. 2016;34:1134-1150.

50. Vansteenkiste J, Zielinski M, Linder A, et al. Adjuvant MAGE-A3 immunotherapy in resected non–small cell lung cancer: phase II randomized study results. J Clin Oncol. 2013;31:2396-2403.

40. Chen HY, Yu SL, Chen CH, et al. A five-gene signature and clinical outcome in non–small cell lung cancer. N Engl J Med. 2007;356:11-20. 41. Kratz JR, He J, Van Den Eeden SK, et al. A practical molecular assay to predict survival in resected nonsquamous, non–small cell lung cancer: development and international validation studies. Lancet. 2012;379:823-832. 42. Boeri M, Pastorino U, Sozzi G. Role of microRNAs in lung cancer: microRNA signatures in cancer prognosis. Cancer J. 2012;18:268-274. 43. Haspinger ER, Agustoni F, Torri V, et al. Is there evidence for different effects among EGFR-TKIs? Systematic review and meta-analysis of EGFR tyrosine kinase inhibitors (TKIs) versus chemotherapy as firstline treatment for patients harboring EGFR mutations. Crit Rev Oncol Hematol. 2015;94:213-227. 44. D’Angelo SP, Janjigian YY, Ahye N, et al. Distinct clinical course of EGFR mutant resected lung cancers: results of testing of 1118 surgical specimens and effects of adjuvant gefitinib and erlotinib. J Thorac Oncol. 2012;7:1815-1822. 45. Kelly K, Altorki NK, Eberhardt WEE, et al. Adjuvant erlotinib versus placebo in patients with stage IB-IIIA non–small cell lung cancer (RADIANT): a randomized, double-blind, phase III trial. J Clin Oncol. 2015;33:4007-4014. 46. Pennell NA, Neal JW, Chaft JE, et al. SELECT: a multicenter phase II trial of adjuvant erlotinib in resected early-stage EGFR mutation–positive NSCLC. J Clin Oncol. 2014; 32 (suppl; abstr 7514). 47. Joensuu H, Eriksson M, Sundby K, et al. One vs three years of adjuvant imatinib for operable gastrointestinal stromal tumor. JAMA. 2012;307:1265-1272. 48. Sandler A, Gray R, Perry MC, et al. Pacl*taxel-carboplatin alone or with bevacizumab for non–small cell lung cancer. N Engl J Med. 2006;355:2542-2550. 49. Vansteenkiste JF, Cho BC, Vanakesa T, et al. Efficacy of the MAGE-A3 cancer immunotherapeutic as adjuvant therapy in patients with

51. Butts C, Socinski MA, Mitchell PL, et al. Tecemotide (L-BLP25) versus placebo after chemoradiotherapy for stage III non–small cell lung cancer (START): a randomised, double-blind, phase 3 trial. Lancet Oncol. 2014;15:59-68. 52. Giaccone G, Bazhenova LA, Nemunaitis J, et al. A phase III study of belagenpumatucel-L, an allogeneic tumour cell vaccine, as maintenance therapy for non–small cell lung cancer. Eur J Cancer. 2015;51:2321-2329. 53. Quoix E, Lena H, Losonczy G, et al. TG4010 immunotherapy and first-line chemotherapy for advanced non–small-cell lung cancer (TIME): results from the phase 2b part of a randomised, double-blind, placebo-controlled, phase 2b/3 trial. Lancet Oncol. 2016;17:212-223. 54. Kotsakis A, Papadimitraki E, Vetsika EK, et al. A phase II trial evaluating the clinical and immunologic response of HLA-A2(1) non–small cell lung cancer patients vaccinated with an hTERT cryptic peptide. Lung Cancer. 2014;86:59-66. 55. Reck M, Rodríguez-Abreu D, Robinson AG, et al; KEYNOTE-024 Investigators. Pembrolizumab versus chemotherapy for PD-L1-positive non–small cell lung cancer. N Engl J Med. 2016;375:1823-1833. 56. Herbst R, Baas P, Kim DW, et al. Pembrolizumab versus docetaxel for previously treated, PD-L1 positive, advanced non–small cell lung cancer (Keynote-010): a randomised controlled trial. Lancet. 2016;387:1540-1550. 57. Borghaei H, Paz-Ares L, Horn L, et al. Nivolumab versus docetaxel in advanced nonsquamous non–small cell lung cancer. N Engl J Med. 2015;373:1627-1639. 58. Brahmer J, Reckamp KL, Baas P, et al. Nivolumab versus docetaxel in advanced squamous-cell non–small cell lung cancer. N Engl J Med. 2015;373:123-135. 59. Rittmeyer A, Barlesi F, Waterkamp D, et al; OAK Study Group. Atezolizumab versus docetaxel in patients with previously treated non-small-cell lung cancer (OAK): a phase 3, open-label, multicentre randomised controlled trial. Lancet. 2017;389:255-265.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 639

MELANOMA/SKIN CANCERS

MANAGEMENT OF ADVANCED EXTRACUTANEOUS MELANOMAS AND NONMELANOMA SKIN CANCERS

Advances in the Treatment of Advanced Extracutaneous Melanomas and Nonmelanoma Skin Cancers Kimberly M. Komatsubara, MD, Joanne Jeter, MD, Richard D. Carvajal, MD, Kim Margolin, MD, Dirk Schadendorf, MD, and Axel Hauschild, MD OVERVIEW Cutaneous malignancies make up the greatest proportion of all human cancers and include melanomas as well as nonmelanoma skin cancers (NMSCs) such as basal cell carcinoma (BCC) and cutaneous squamous cell carcinoma (cSCC), as well as less common Merkel cell carcinoma (MCC), cutaneous lymphomas, cutaneous adnexal tumors, Kaposi sarcomas, and other sarcomas. Each of these NMSCs differ significantly in biology, clinical behavior, and optimal treatment recommendations from each other and from cutaneous melanoma. Similarly, less common extracutaneous melanomas, such as mucosal (MMs) and uveal (UMs), are unique biologic and clinical entities that require distinct diagnostic and management considerations. In this review, we summarize recent advances in biology and treatment of extracutaneous melanomas and NMSCs, including MMs, UMs, cSCC, BCC, and MCC.

utaneous malignancies make up the greatest proportion of all human cancers and include melanomas as well as NMSCs such as BCC and cSCC, as well as less common malignancies such as MCC, cutaneous lymphomas, cutaneous adnexal tumors, Kaposi sarcomas, and other sarcomas. Each of these NMSCs differ significantly in biology, clinical behavior, and optimal treatment recommendations from each other and from cutaneous melanoma. Similarly, less common extracutaneous melanomas such as MM and UM are unique biological and clinical entities from cutaneous melanoma and require distinct management considerations. In this review, we summarize recent advances in our understanding and management of a subset of advanced extracutaneous melanomas and NMSCs, including MMs, UMs, cSCC, BCC, and MCC.

EXTRACUTANEOUS MELANOMAS

Melanoma is a heterogeneous collection of diseases arising from melanocytes within the skin, uveal tract of the eye, and mucosal surfaces of the body. Although the majority of cases arise from cutaneous surfaces, approximately 3%–5% of cases arise from the uveal tract of the eye, and approximately 2% arise from the mucosal surfaces of the body.1,2 UM arises from melanocytes anywhere along the uveal tract, with the majority of cases arising from the choroid (approximately 85% of cases) and the remaining cases from the iris or ciliary body.3 Despite definitive primary therapy

with enucleation or radiotherapy, nearly 50% of patients will develop metastatic disease.4 Survival from the time of diagnosis of metastatic disease is poor, with overall survival (OS) ranging from 6 to 13 months.5,6 The biology of UM is distinct from that of cutaneous melanoma; thus, treatments that have been effective for advanced cutaneous melanoma, including chemotherapy, molecularly targeted therapies, and immunotherapies, have been less effective in UMs and have not impacted outcomes in this rare disease. MM arises in any mucosal epithelium containing melanocytes, such as that of the respiratory, gastrointestinal, and urogenital tracts. The most commonly affected sites include the head and neck, anorectal region, and vulvovagin*l region. In general, management of localized MM consists of wide local excision if negative margins can be achieved with or without adjuvant radiotherapy; however, because of anatomic constraints, this approach is not always possible. Sentinel lymph node biopsy is unproven in these cancers, although elective regional lymph node dissection may be considered in some subtypes. Data are lacking on the benefit of systemic adjuvant therapy with interferon or ipilimumab, the currently approved adjuvant treatment options available for cutaneous melanoma at high risk of recurrence. Adjuvant therapy with the combination of temozolomide and cisplatin as well as single-agent high-dose interferon showed survival benefit when compared with observation in a single-institution study that has not been validated in subsequent larger trials.7 Overall, most patients

From the Columbia University Medical Center, New York, NY; Ohio State University Medical Center, Columbus, OH; City of Hope, Duarte, CA; Department of Dermatology, University Hospital Essen, Essen, Germany; Department of Dermatology, University Hospital Schleswig-Holstein, Kiel, Germany. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Axel Hauschild, MD, University Hospital Schleswig-Holstein, Schittenhelmstrasse 7, Kiel 24105, Germany; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 641

KOMATSUBARA ET AL

develop and die of disseminated disease. Disease-specific survival for MM at 5 years is markedly decreased compared with that of cutaneous melanoma (25% vs. 80%).2

Uveal Melanoma

In contrast to cutaneous melanoma, which is subdivided into those that harbor activating mutations in BRAF, RAS, or loss of function of NF1, UM is characterized by mutations in the G-α-protein subunits GNAQ or GNA11 in approximately 80%–95% of cases.8-11 Mutations in GNAQ or GNA11 result in disabling of their intrinsic GTPase activity and constitutive activation of downstream pathways, including the mitogen-activated protein kinase (MAPK) and phosphoinositide 3-kinase (PI3K)/Akt/mTOR pathways. More recently, mutations in PLCB4, a downstream effector of GNAQ and GNA11,12 and recurrent activating mutations in the G-protein–coupled receptor CYSLTR2, have been identified as oncogenic drivers that contribute in a mutually exclusive manner from GNAQ and GNA11 mutations to UM.13 BRCA1-associated protein 1 (BAP1) is a tumor suppressor that is mutated in approximately 47% of primary UMs and associated with metastasis and poor prognosis.14 SF3B1 mutations have been identified in 18% of primary UMs and EIF1AX mutations in 48% of primary UMs and are associated with an overall more favorable prognosis, although SF3B1 mutant UM is a subset characterized by atypical presentations and late occurrences of distant disease.15,16

KEY POINTS • Extracutaneous melanomas and NMSCs represent a biologically and clinically heterogeneous group of diseases, each of which requires unique management considerations. • Although single-agent immunologic checkpoint blockades have limited efficacy in advanced uveal melanoma, other immune-based treatment strategies including the use of novel T-cell redirection as well as adoptive T-cell transfer therapies are being pursued based upon promising preliminary data. Preclinical data demonstrate the efficacy of epigenetic targeting of uveal melanoma via histone deacetylase inhibition as well as bromodomain targeting, with clinical trials testing this strategy ongoing. • Both targeted therapy and immunologic checkpoint blockade have activity in mucosal melanoma but with less favorable outcomes than for cutaneous disease, and continued investigation of novel therapies is needed. • Immune checkpoint blockade with agents targeting the PD-1/PD-L1 pathway has shown antitumor activity in two clinical trials for advanced Merkel cell carcinoma. There is also promise for immune checkpoint blockade in combination with radiotherapy and other treatments for advanced Merkel cell carcinoma. In addition, testing in the neoadjuvant and/or adjuvant setting is also critical and should be prioritized. • Further study of immunotherapy with PD-1 and PD-L1 inhibitors is ongoing for advanced SCC and BCC. 642 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Emerging therapies for primary uveal melanoma. Treatment of primary UM can be subdivided into either globe-preserving therapy or enucleation. In the United States, the majority of primary UMs are treated with plaque brachytherapy based on results of the Collaborative Ocular Melanoma Study that evaluated plaque brachytherapy versus enucleation for medium-sized choroidal melanomas.17 Although current treatment modalities for primary UM achieve excellent local control, complications leading to vision loss in the affected eye are common, and new approaches for local therapy are still needed. ICON-1 is one novel therapy for primary UM currently in development. This human immunoconjugate protein targets a modified version of human factor VII, the ligand of tissue factor, which is commonly overexpressed in primary UM.18 ICON-1 binds to UM cells overexpressing tissue factor and signals an immune response to eliminate pathologic tissue while leaving normal tissue intact. A phase I clinical study of ICON-1 in patients with UM who are planned to undergo enucleation or brachytherapy is currently enrolling (NCT02771340). AU-011, another investigational therapy for primary UM, is a viruslike particle that binds to cancer cells and is conjugated to infrared molecules that can be activated by an ophthalmic laser, allowing for selective targeting of UM cells. This therapy is based on preclinical data in ovarian and lung cancer models demonstrating that human papillomavirus-like particles selectively bind to heparan sulfate proteoglycans on the disrupted epithelium of cancer cells while leaving the surrounding healthy tissue unharmed.19 AU-011 has been granted an orphan drug designation by the U.S. Food and Drug Administration, and a clinical study of AU-011 in UM is currently in development. Emerging therapies for metastatic uveal melanoma. Treatment strategies adapted from cutaneous melanoma have generally been ineffective in UM, and thus, no standard of care therapy exists for advanced UM. Systemic chemotherapy with dacarbazine- or cisplatin-based regimens demonstrated response rates of 0%–10% in metastatic UM.20-23 More recently, molecularly targeted therapies for the MAPK and/or the PI3K/Akt pathways have been conducted in metastatic UM. A randomized phase II study of the MAPK kinase inhibitor, selumetinib, versus chemotherapy in advanced UM demonstrated a modest improvement in progression-free survival (PFS) with selumetinib treatment, but no OS benefit. A subsequent phase III study of selumetinib and dacarbazine versus dacarbazine alone showed no improvement in either PFS or OS.24-26 Liver-directed therapies such as embolization or hepatic arterial chemotherapy infusion have controlled UM in selected patient populations but without OS benefit.27-29 Treatment approaches currently being investigated for advanced disease include combination targeted therapies, immune-based strategies, and epigenetic agents. Given the poor outcomes in this disease and current lack of effective therapies, rationally designed clinical trials investigating novel therapeutic approaches for advanced UM are urgently needed and should be prioritized in this disease.

MANAGEMENT OF ADVANCED EXTRACUTANEOUS MELANOMAS AND NONMELANOMA SKIN CANCERS

Immune-based therapies in uveal melanoma. Immune checkpoint inhibitors have demonstrated limited efficacy in metastatic UM, with durable responses in less than 5% of patients. Two prospective clinical trials of CTLA-4 blockade with ipilimumab, as well as a prospective study of tremelimumab, demonstrated very low response rates and short PFS of less than 3 months: the phase II GEM-1 trial evaluated ipilimumab, at a dose of 10 mg/kg, in treatment-naive patients with metastatic UM and reported only one response in 13 evaluable patients.30 A subsequent multicenter phase II trial by the Dermatologic Cooperative Oncology Group (DeCOG) evaluated ipilimumab at a dose of 3 mg/kg in 45 pretreated and eight treatment-naive patients with metastatic UM and reported no responses. The median PFS was 2.8 months, and median OS was 6.8 months.31 A prospective multicenter phase II study of tremelimumab in advanced UM was stopped early for futility with no responses in the first 11 patients and a median PFS of only 2.9 months and median OS 12.8 months, likely reflecting the natural history of advanced UM.32 Similarly, results of PD-1/PD-L1 checkpoint blockade in metastatic UM have been disappointing. Studies evaluating PD-1/PD-L1 inhibition in UM thus far have been limited to small retrospective case series. A retrospective analysis of 25 patients with metastatic UM treated with pembrolizumab through an expanded access program reported two partial responses and stable disease in six patients.33 The largest case series to date analyzed 58 patients with metastatic UM treated with anti–PD-1 or anti–PD-L1 therapy across nine different academic centers.23 The response rate was 3.6% (two partial responses), and stable disease for 6 months or longer was observed in 8.9% (five of 56 evaluable patients). Median PFS and OS were 2.8 (95% CI, 2.4–2.8 months) and 7.6 months (95% CI, 0.7–14.6 months), respectively.23 PD-L1 status was not reported in this retrospective analysis. Overall, clinical benefit with immune checkpoint inhibition in metastatic UM is rare, and these therapies should not be standard in metastatic UM. Differences in tumor mutational landscape, with UM being characterized by fewer genetic mutations, may explain in part the inferior outcomes of immune checkpoint inhibition in UM versus cutaneous melanoma.34,35 Further investigation of the biology of this distinct melanoma variant and its profound resistance to single-agent immunotherapy, possibly because of an unfavorable microenvironment in its commonest metastatic site, the liver, require further exploration. There is an ongoing study of pembrolizumab in metastatic UM (NCT02359851) and two ongoing studies of combined CTLA-4 and PD-1 blockade with ipilimumab and nivolumab in metastatic UM (NCT01585194 and NCT02626962). Investigation of other immune-based therapeutic strategies are ongoing with promising early results. Glycoprotein 100 (gp100) is a tumor-associated antigen that is strongly expressed in both cutaneous and UM.36 IMCgp100 is a recombinant T-cell receptor currently in development that recognizes the gp100 antigen presented by HLA-A2 on its targeting end and binds and activates CD3+ T lymphocytes

on its effector end, thus allowing cytotoxic T cells to be redirected toward the gp100-expressing UM cells. The first-inhuman phase I study of IMCgp100 enrolled 84 HLA-A*02– positive patients with advanced melanoma, including 16 patients with advanced UM.37 Subjects were treated with intravenous IMCgp100 at two separate dosing regimens: weekly or daily for 4 days, every 3 weeks. The most common adverse events included rash (100%), pruritus (64%), pyrexia (50%), and periorbital edema (46%). Grade 3 or 4 adverse events were observed predominantly in the first 3 weeks of study treatment and included rash (23%), lymphopenia (8%), and hypotension (6%), which was associated with trafficking of CD3+ T cells to the tumor environment and chemokine release. An intrapatient dose-escalation design was subsequently implemented to mitigate the hypotension observed during the first few weeks of therapy.37 Results of the 15 evaluable patients with UM enrolled in this first-in-human study were recently presented at the 2016 Society for Melanoma Research Congress. The majority of patients with UM enrolled in this study had liver metastases and elevated lactate dehydrogenase. A partial response was achieved in 20% (three patients) and stable disease in 47% (seven patients) of patients with UM at 8 weeks. The disease control rate was 53% at 16 weeks and 40% at 24 weeks.38 Based on these promising results, there is an ongoing phase I/II study of IMCgp100 in patients with advanced UM using an intrapatient dose-escalation regimen (NCT02570308), with a pivotal randomized clinical trial in development. Investigation of other immune-based therapies in UM is ongoing. A study of dendritic cell vaccination in 14 patients with metastatic UM demonstrated four responses and a median OS of 19.2 months.39 A phase II study of autologous tumor-infiltrating lymphocytes in metastatic UM is currently enrolling (NCT01814046). Epigenetic therapies in uveal melanoma. Given that UM is a genetically simple disease characterized by few somatic insults compared with cutaneous melanoma,34,35 other factors such as epigenetic alterations may be important in the pathogenesis of UM. This is supported by evidence that genes associated with a high-risk, class 2 phenotype, such as PHLDA1, seem to be regulated through epigenetic modifications.40 Additionally, preclinical data in UM cell lines support the role of histone deacetylase inhibitors (HDACi) in reversing the phenotypic and biochemical cell changes associated with BAP1 loss and metastatic potential in UM cells.41,42 In UM cell lines, epigenetic modification with four different HDACi (including valproic acid, trichostatin A, LBH-589 or panobinostat, and vorinostat), induced G1 cell-cycle arrest, melanocytic differentiation, and gene expression changes consistent with reversion to a class I phenotype.42 Additionally, valproic acid was capable of inhibiting growth of UM tumors in vivo. There is an ongoing clinical study of vorinostat in metastatic UM (NCT01587352). Additionally, HDACi is also being explored as a strategy in the adjuvant setting in an ongoing clinical study of adjuvant sunitinib or valproic acid in high-risk patients with UM (NCT02068586). asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 643

KOMATSUBARA ET AL

Recent preclinical data suggest that epigenetic therapies targeting the bromodomain and extra-terminal domain (BET) family of proteins may be a promising new strategy in UM. The BET family of proteins, including BRD2, BRD3, BRD4, and BRDT, are epigenetic regulators that bind to acetylated lysine residues on histones and direct the assembly of nuclear complexes that regulate DNA replication, chromatin remodeling, and transcription.43,44 BRD4 is a key regulator of transcriptional elongation by recruiting positive transcriptional elongation factor complex to chromatin and activating RNA polymerase II–dependent transcription. It is considered a nononcogenic regulator of cancer growth, in part through activation of the Myc transcriptome.45 BET inhibition has shown antitumor activity in preclinical studies in hematologic malignancies45,46 and selected solid tumors.47-50 It is hypothesized that UM may be particularly susceptible to BET targeting given the relatively high incidence of Myc amplifications observed and preclinical data supporting epigenetic targeting in UM.51 In preclinical studies, JQ1, a first-generation BET inhibitor that competitively displaces BRD4 from acetylated histones, demonstrated potent cytotoxic activity in GNAQ and GNA11 mutant cell lines, but not wild-type cells.52 Microarray analysis of cell lines treated with JQ1 revealed changes in expression in genes involved in cell cycle regulation, apoptosis, and the DNA damage response. Interestingly, concomitant silencing of Bcl-xL and Rad51, regulators of apoptosis and the DNA damage response, respectively, was sufficient to induce apoptosis independent of Myc expression.52 Small-molecule inhibitors of BET proteins are currently in clinical development. Based on intriguing preclinical data, clinical investigation of BET protein inhibition in UM may be warranted.

Mucosal Melanoma

The biology and clinical outcomes of MM differs significantly from that of cutaneous melanoma. A recent analysis of updated survival data for melanoma subtypes was based on information from 3,454 patients diagnosed with metastatic melanoma from 2000 to 2013, including 237 patients with MM.6 The median OS for patients with advanced MM was 9.1 months (95% CI, 7.6–9.8), which was significantly shorter than that for patients with other subtypes of melanoma, including UM. This trend appears to have continued even in more recent years, as it persisted even in the cohort of patients diagnosed in 2011 to 2013. No notable differences in survival were found for the subsets of MM defined by the primary site (anorectal, head and neck, vulvovagin*l, or other). BRAF-activating mutations occur far less frequently in mucosal than in cutaneous melanoma, whereas KIT mutations are found more often in MM. Genomic sequencing of 10 MMs demonstrated that somatic mutation rates were significantly lower than those found in sun-exposed cutaneous melanoma and that more copy number and structural variations were present in the mucosal tumors.53 Although data with direct comparisons are lacking, one summary review suggests that KIT mutations are observed more frequently 644 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

in vulvovagin*l and anal melanoma than in sinonasal melanoma.54 A study of 467 anorectal melanomas identified driver mutations in 95% of tumors.55 Affected genes included KIT, NF1, other elements of the MAPK pathway, and SF3B1. Conversely, a small study of melanomas from the female genital tract showed a low mutational burden for genes in the MAPK/ERK, PI3K/AKT, and GNAQ/11 pathways.56 Another study of melanoma from the female urogenital tract showed that NRAS mutations were more prevalent than KIT mutations in this tumor type (21% vs. 4%).57 Of the mutations identified in this study, three of the NRAS mutations were in exon 3 (codon 61), one was in exon 2 (codon 12), and the KIT mutation was in exon 17 (codon 820). Molecularly targeted therapy directed at KIT as the oncogenic driver. Because of the frequency of KIT mutations and other genetic aberrations such as gene amplification in MMs, imatinib has been investigated as a potential therapeutic agent. In a multicenter, single-arm phase II trial, 28 patients with advanced or unresectable melanoma with KIT mutation or amplification were treated with imatinib 400 mg twice daily.58 Among the 13 patients with MM in this study, four had stable disease (reported at 10–20 weeks), one had a transient partial response, one had a partial response lasting 53 weeks, and one had a complete response lasting 95 weeks. The relative number of responses was similar to those found in the patients in the study with KIT alterations in acral melanomas (one out of 10 with complete response and two out of 10 with partial response). In a subsequent phase II trial, 43 patients with KIT mutation or amplification, including 11 with MM, were treated with imatinib 400 mg daily.59 Four patients with MM had imatinib escalated to 800 mg daily at the time of disease progression on the lower dose but did not respond to the higher dose. Toxicity of the regimen was dose-limiting grade 3 to 4 edema, nausea, fatigue, and anorexia. Subgroup analysis for the mucosal group was not reported, but for the overall cohort, 10 patients achieved partial response (23%), and 13 had stable disease at 8 weeks or later. The median PFS was 3.5 months. In a later phase II trial of imatinib 400 mg daily in 24 patients with KIT mutated or amplified melanoma, seven patients had partial responses, and five had stable disease ranging in duration from approximately 3 to 11 months.60 The only objective responses were reported among the 17 patients with MM. Of note, all responders had KIT mutations, and none had amplification of KIT. The response rates to imatinib have prompted investigation of the effects of other tyrosine kinase inhibitors in the MM population. An early report of the use of dasatinib in two patients with L576P KIT-mutated MMs showed a rapid radiologic response in both, as well as improvement in symptomatic control in one of the patients, but the responses were short-lived (3 to 4 months).61 More recently, data from the second stage of the U.S. cooperative group study E2607 were presented at the 2016 ASCO Annual Meeting.62 In this study, subjects received 70 mg of dasatinib orally twice a day. The first stage included 51 subjects with acral melanoma, MM, or melanoma of chronically sun-damaged

MANAGEMENT OF ADVANCED EXTRACUTANEOUS MELANOMAS AND NONMELANOMA SKIN CANCERS

skin regardless of mutation status, of whom three turned out to have a KIT mutation, and the second phase had 22 patients with KIT-mutated melanoma. Of note, accrual to this study suffered due part to the rapid adoption of drugs such as imatinib and dasatinib for patients with advanced MM and KIT mutations based on the prior experience and general enthusiasm about molecularly targeted therapy, particularly for a disease that had historically been so difficult to treat. Twenty-nine of the subjects across the two stages had MM, and three had partial responses, but responses occurred both in individuals with KIT-mutated and KIT wildtype tumors, presumably because dasatinib is a “dirty” kinase inhibitor without specificity for KIT. Nilotinib was also tested in a small study of KIT-mutated advanced melanoma refractory to at least one prior KIT inhibitor or with brain metastases. Among 19 patients with MM in this study, only two partial responses and a high rate of dose-limiting toxicity were reported.63 The overall status of KIT-targeted therapies in MM remains a subject of investigation, including efforts to better understand the dependence of cells carrying this mutation on KIT and other pathways that may be targetable with small molecules or other classes of agent. So far, however, the responses to these agents are not durable, as they are in single pathway-driven malignancies like chronic myeloid leukemia and gastrointestinal stromal tumors; therefore, it is critical to study mechanisms of both intrinsic and acquired resistance to KIT inhibition. Melanoma cells with acquired resistance to KIT inhibition have been found to have activation of MAPK and PI3K signaling and remain sensitive to concurrent inhibition of these pathways.64 Opportunities may be identified for vertical or horizontal inhibition of more than one pathway with molecularly targeted therapy. Although the tolerance of these single-agent kinase inhibitors at full doses may be limited, many toxicities are nonoverlapping and may permit investigation of full or near-full doses of each class of agent. Immune checkpoint inhibitors in mucosal melanoma. Ipilimumab use in the population with MM has been described in several studies, all of which were only for patients previously treated with systemic therapy, generally consisting of single-agent or combination chemotherapy, which provides very low (less than 10%) objective response rates in this subset of melanoma. In one small report with six evaluable patients with MM, one subject had a partial response, which is the number expected from the data in cutaneous melanoma.31,65 Disease control rate, defined as stable disease lasting at least 12 weeks plus all objective responses, was 50%; however, no patients survived at the 2-year endpoint. Another retrospective case series of 33 patients with metastatic MM treated with ipilimumab after failure of cytotoxic chemotherapy showed an overall response rate of 6.7%, with one complete responder, one partial responder, and six with stable disease. The disease control rate of 24% was very similar to that which has been reported in much larger series for unselected melanoma.66 The largest study reported to date consisted of 71 patients with pretreated

metastatic MM in the Italian ipilimumab expanded access program.67 In this group, the disease control rate was 36% and the overall response rate 12%. The rate of immune-related toxicity was similar for MM to that reported for unselected patients with melanoma. A small case series detailed the use of combination ipilimumab and external beam radiation therapy in four women with locally recurrent MMs of the vagin* or cervix.68 Patients received up to four doses of ipilimumab concurrently with radiation to 3,000 to 6,020 cGy. Two subjects had grade 3 adverse effects (colitis and dermatitis, both of which could have been radiotherapy-related, but the contribution of the checkpoint-blocking antibody could not be determined). One subject had a complete response to the combined-modality therapy and did not proceed to surgery. Three subjects underwent resection after combination therapy, and two were found to have residual melanoma at that time. After resection of viable tumor, two patients remained disease-free at 20 and 38 months. Recent data have shown that PD-1 inhibitors have efficacy in patients with MM, although the rates of response may be somewhat lower than in cutaneous melanoma. Shoushtari et al69 reported an objective response rate of 23% (95% CI, 10%–40%) to first- or subsequent-line use of nivolumab or pembrolizumab in 35 patients with MM. The majority of these patients had M1c disease and brain metastases; most were wild-type for BRAF, NRAS, or KIT mutations. Over 75% had prior therapy, most often with ipilimumab, and had progressed without an initial response.69 Of the 24 patients with MM and no benefit from prior ipilimumab, five patients had an objective response to a PD-1 inhibitor. Evaluation of 84 patients with MM treated with pembrolizumab in the KEYNOTE-001, 002, and 006 studies has been presented in abstract form. This study showed an overall response rate of 19% (95% CI, 12%–29%) with durable responses of up to 27 months. Of note, activity was seen both in patients with prior treatment with ipilimumab and in those who were treatment-naive.70 A recent pooled analysis of 86 patients with MM who received nivolumab, either alone or in combination with ipilimumab, reported a similar response rate to single-agent nivolumab (23%) but a better response to the combination of nivolumab and ipilimumab (37%).71 Toxicity in the patients with MM was similar to that seen in patients with cutaneous melanoma on these regimens. Based on these data, combination therapy with nivolumab and ipilimumab may be the preferred regimen in patients with MM who can tolerate it. The role of tumor biomarkers such as immunohistochemical expression of PD-L1 to predict benefit remains no more clear in MM than it is currently for cutaneous melanoma and a variety of other malignancies.

Research Directions

Current research directions include investigation of novel tyrosine kinases as well as the combination of targeted therapies with immunotherapies in this population. As with targeted therapies in other oncologic settings, responses to tyrosine kinase inhibitors may be of limited duration asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 645

KOMATSUBARA ET AL

because of primary or acquired resistance to these agents. Use of a multikinase inhibitor may bypass or prevent the development of resistance to targeted agents. In contrast, the natural history of rapid progression of mucosal disease may not allow the time required for an initial response to immunotherapy as well. Therefore, a combination of agents from these categories may provide a synergistic effect to overcome these obstacles to disease response.

NONMELANOMA SKIN CANCERS

NMSC is the most common cancer in humans and the most frequently observed malignancy in whites. Approximately 57%–80% of NMSCs are BCC, and 20%–25% are cSCC. MCC is a rather rare, but increasingly observed type of NMSC even more associated with immunosuppression, older age, and ultraviolet (UV) damage than other NMSCs. NMSC represents a major global economic and health burden. More than 2.1 million individuals in the United States are diagnosed with NMSC each year, with the vast majority (80%–90%) localized in the sun-exposed areas of the head and neck.72 The overall mortality rates for NMSCs are low in general, although MCC is a highly aggressive malignancy with a disease-specific mortality rate in a range of 25%–50%, reflecting a high rate of dissemination at the time of diagnosis, especially for large primaries, which are often mistaken for benign lesions or BCCs. All three types of NMSC discussed in this study are characterized by the risk of local recurrence, which, in some cases, can be predicted by specific clinical and pathologic features such as the size and location of the primary tumor.73 The risk of aggressive cSCCs and MCCs is extremely high in immunosuppressed patients, such as those with solid-organ transplants, particularly if these patients already have a history of sun-damaged skin.74 Based on current guidelines, the primary treatment of NMSC is surgical. The complete excision, ideally with Mohs surgery or micrographic surgery of the primary tumors, is mandatory to prevent local relapses. The safety margins for cSCC and BCC are typically in a range of 0.5 to 1 cm, whereas in MCC, a 1- to 2-cm minimum excision margin is recommended. Mohs and micrographic surgery allow an evaluation of the completeness of tumor resections on cryostat-fixed or paraffin-embedded tumors, respectively. Particularly in NMSC of the face and in relapsing tumors, this technique is the standard of care. Selective lymph node dissection (SLND) is not recommended for the surgical management of cSCC or BCC. In contrast, an SLND has been established as a routine in patients with MCC. The SLND offers the potential to assess regional nodes for occult disease and an appropriate selection for further treatment. Typically, patients with a positive SLND are referred to a complete lymphadenectomy. Adjuvant radiotherapy is typically recommended for patients with a high-risk MCC, resulting in a better local control with a significantly decreased number of local relapses by the use of adjuvant radiotherapy with a typical dose of 50–60 Gy in conventionally fractionated 2-Gy doses. There 646 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

is no proven effective adjuvant treatment of cSCC and BCC. Therefore, neither adjuvant irradiation nor systemic treatment is offered in most centers to those patients, even if they are estimated to have a high risk of relapse.

Cutaneous Squamous Cell Carcinoma

Despite the use of routine surgery and radiotherapy for advanced cSCC, not all patients are cured. In a recent paper,75 clinical features such as tumor thickness of more than 6 mm, localization on or near the ear, and immunosuppression have been associated with increased risk of local relapse and death. A satisfactory systemic treatment of advanced SCC has not been established so far, because of the poor therapeutic index of cytotoxic agents and insufficient information about immunotherapy. In patients with medical contraindications to surgery or radiotherapy, platinum-based regimens, either as single agents or in combinations, have been typically used in the past. However, despite a high response rate of up to 80%, the median response duration are only 4 to 6 months. There is no evidence that chemotherapies have an impact on OS for NMSC, and prospective randomized clinical trials are lacking. Because cSCCs frequently demonstrate EGFR overexpression, antibodies to EGFR have been studied for the treatment of advanced disease. A 36-patient phase II study of cetuximab showed a response rate of 45%, but with a relative short response duration of 4 months.76 In some cases, cetuximab has been used in combination with radiotherapy as in head and neck cancer, with some evidence of disease control but an unknown contribution to OS. Lapatinib showed activity in a neoadjuvant trial (two of eight patients had disease regression) and may warrant further evaluation in this setting as well as in immunocompromised patients with more advanced cSCC.77 Because most cSCCs carry a high mutational burden,78 a specific genetic UV signature, and overexpression of PD-L1 in keratinocytes, blockade of the PD-1/PD-L1 pathway is a promising new treatment approach. Very recently, some case reports suggest the usefulness of PD-1 antibodies in patients with advanced or metastatic cSCC, with some demonstrating dramatic and complete responses even in heavily pretreated patients.79-81 In 2016, a phase II trial on a new PD-1 antibody, REGN 2810, in patients with locally advanced and metastatic cSCCs who are not candidates for surgery or radiotherapy has been initiated (NCT02760498).

Basal Cell Carcinoma

The vast majority of BCCs are easy to treat by conventional surgery but very difficult to treat when unresectable or, rarely, metastatic. The “sonic hedgehog” signal transduction pathway was identified as crucial for the progression of BCCs and is commonly associated with mutations in the tumor-suppressor gene PATCHED and the tumor oncogene SMOOTHENED. Competitive inhibitors of SMOOTHENED, namely vismodegib and sonidegib, have been developed with some success, and are now approved based on responses in unresectable and/or metastatic BCC.

MANAGEMENT OF ADVANCED EXTRACUTANEOUS MELANOMAS AND NONMELANOMA SKIN CANCERS

The multicenter, international, nonrandomized “ERIVANCE” study enrolled 104 patients with locally advanced and metastatic BCC and treated them with vismodegib at a flat dose of 150 mg daily.82 In patients with metastatic BCC, the response rate was 30%, whereas in the patients with locally advanced BCC, the response rate was 43%, including a complete response rate of 21%. Typical adverse events included muscle spasms, alopecia, dysgeusia, weight loss, and fatigue. The results of this study led to the approval of vismodegib for advanced BCCs by the U.S. Food and Drug Administration (FDA) and European Medicines Agency (EMA). A subsequent large safety study of vismodegib enrolled 1,227 patients with locally advanced or metastatic BCC (STEVIE).83 The interim analysis confirmed the previous results with an objective response seen in 302 of 453 (66.7%) patients with locally advanced disease, of which half were complete responses. Of the 29 patients with metastatic BCC the response rate was 37.9% (two complete and nine partial responses). No previously unreported treatment-related adverse events have been observed in this safety study. More recently, another hedgehog inhibitor, sonidegib, was evaluated in a multicenter, randomized, double-blind phase II trial (BOLT).84 This study compared two different dosages of sonidegib (200 mg or 800 mg orally daily) using a primary endpoint of objective response. Twenty of 55 patients (36%) receiving sonidegib at a dose of 200 mg and 39 of 116 patients (34%) receiving sonidegib at a dose of 800 mg achieved an objective response. The 200mg dose was better tolerated compared to the 800-mg dose, with treatment discontinuations in 22% and 36% of the treated patients in each dosing group, respectively. Only typical adverse events specific for sonic-hedgehog inhibitors have been observed in this trial. These results led to the FDA and EMA approval of sonidegib at a dose of 200 mg daily for patiens with locally advanced and metastatic BCC. There are no clear differences between vismodegib and sonidegib in terms of response rates and tolerability. Although hedgehog inhibitors provide responses in roughly 60% of patients with unresectable BCCs and multiple BCCs from the BCC syndrome, including approximately 30% with complete responses, many patients discontinue treatment despite ongoing response because of toxicities, mainly fatigue, dysgeusia, and muscle cramps. Because platinum-based chemotherapies and EGFR inhibitors are only occasionally used in BCC and do not achieve long-term benefit, there remains a critical need for better advanced BCC treatment. The rationale to use PD-1 antibodies for BCCs is the known high mutational burden and a clear genetic UV signature.85 Case reports on successful treatment with PD-1 antibodies in advanced BCCs have been published recently.79,81 An expansion arm for patients with BCC with progression or intolerance of hedgehog inhibitor therapy has been recently initiated on the ongoing phase I study of REGN2810 in patients with advanced malignancies (NCT02383212).

Merkel Cell Carcinoma

Local, regional, and distant metastases are frequently observed in patients with MCC. For patients who are not good candidates for surgery or radiotherapy, systemic chemotherapy is typically administered based on the biologic similarity of MCC to small cell lung cancer in its aggressiveness and high (over 50% response rate) but short-term (average less than 6 months) responsiveness to platinum-based chemotherapies with or without radiotherapy. The advanced age and comorbidities of many patients with advanced MCC limit their tolerance of chemotherapy. Two studies published in 2016 showed the effectiveness of immune checkpoint inhibitors for patients with advanced and metastatic MCC. Avelumab, an anti–PD-L1 monoclonal antibody, was investigated in a phase II trial for stage IV chemotherapy-refractory MCC and showed a response rate of 31.8% in 88 patients, with eight complete and 20 partial responses. Of interest, responses were ongoing in 82% of the patients at the time of analysis (median follow-up of 10.4 months).86 Another phase II trial studied pembrolizumab, an anti–PD-1 antibody, in treatment-naive patients with MCC. Among 25 evaluable patients, 56% responded, including four complete responses. At a median follow-up of 33 weeks, only two of the 14 responders relapsed. Of interest, the responses were independent of the presence of the Merkel cell polyomavirus in tumor.87 These encouraging data led to a recent uptake of the available PD-1 antibodies pembrolizumab and nivolumab (not yet studied in MCC) as the new standard for advanced MCC.88,89 It is very likely that these PD-1 (pembrolizumab)/ PD-L1 (avelumab) antibodies will become the new backbone for development of even more powerful immunotherapy regimens, which may include radiotherapy and may be applied in the adjuvant setting, for this aggressive but highly immunogenic cancer.

CONCLUSION

The biologic and clinical heterogeneity of cutaneous malignancies and noncutaneous melanomas provide a number of unique opportunities and challenges for preclinical and clinical investigators. The differential response to molecularly targeted and immunomodulatory therapies provides the opportunity to assess variable biomarkers of sensitivity and mechanisms of primary and secondary resistance. The rarity of some of these malignancies, particularly in the advanced disease setting, is a challenge that investigators can overcome with growing awareness of these diseases and increasing collaboration between investigators. As described above, there are important advances being made in our understanding of the biology and treatment of patients with advanced extracutaneous melanomas and nonmelanoma skin cancers that will lead to improved outcomes for these patients.

ACKNOWLEDGMENT

K. M. Komatsubara and J. Jeter contributed equally to the development of this manuscript. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 647

KOMATSUBARA ET AL

References 1. Singh AD, Turell ME, Topham AK. Uveal melanoma: trends in incidence, treatment, and survival. Ophthalmology. 2011;118:1881-1885. 2. Chang AE, Karnell LH, Menck HR; The American College of Surgeons Commission on Cancer and the American Cancer Society. The National Cancer Data Base report on cutaneous and noncutaneous melanoma: a summary of 84,836 cases from the past decade. Cancer. 1998;83:1664-1678. 3. McLaughlin CC, Wu XC, Jemal A, et al. Incidence of noncutaneous melanomas in the U.S. Cancer. 2005;103:1000-1007. 4. Kujala E, Mäkitie T, Kivelä T. Very long-term prognosis of patients with malignant uveal melanoma. Invest Ophthalmol Vis Sci. 2003;44:46514659. 5. Diener-West M, Reynolds SM, Agugliaro DJ, et al; Collaborative Ocular Melanoma Study Group. Development of metastatic disease after enrollment in the COMS trials for treatment of choroidal melanoma: Collaborative Ocular Melanoma Study Group Report No. 26. Arch Ophthalmol. 2005;123:1639-1643.

18. Walker TM, Van Ginkel PR, Gee RL, et al. Expression of angiogenic factors Cyr61 and tissue factor in uveal melanoma. Arch Ophthalmol. 2002;120:1719-1725. 19. Kines RC, Cerio RJ, Roberts JN, et al. Human papillomavirus capsids preferentially bind and infect tumor cells. Int J Cancer. 2016;138:901911. 20. Bedikian AY, Papadopoulos N, Plager C, et al. Phase II evaluation of temozolomide in metastatic choroidal melanoma. Melanoma Res. 2003;13:303-306. 21. Flaherty LE, Unger JM, Liu PY, et al. Metastatic melanoma from intraocular primary tumors: the Southwest Oncology Group experience in phase II advanced melanoma clinical trials. Am J Clin Oncol. 1998;21:568-572. 22. Kivelä T, Suciu S, Hansson J, et al. Bleomycin, vincristine, lomustine and dacarbazine (BOLD) in combination with recombinant interferon alpha-2b for metastatic uveal melanoma. Eur J Cancer. 2003;39:11151120.

Kuk D, Shoushtari AN, Barker CA, et al. Prognosis of mucosal, uveal, 6. acral, nonacral cutaneous, and unknown primary melanoma from the time of first metastasis. Oncologist. 2016;21:848-854.

23. Algazi AP, Tsai KK, Shoushtari AN, et al. Clinical outcomes in metastatic uveal melanoma treated with PD-1 and PD-L1 antibodies. Cancer. 2016;122:3344-3353.

7. Lian B, Si L, Cui C, et al. Phase II randomized trial comparing high-dose IFN-α2b with temozolomide plus cisplatin as systemic adjuvant therapy for resected mucosal melanoma. Clin Cancer Res. 2013;19:4488-4498.

24. Carvajal RD, Sosman JA, Quevedo JF, et al. Effect of selumetinib vs chemotherapy on progression-free survival in uveal melanoma: a randomized clinical trial. JAMA. 2014;311:2397-2405.

8. Van Raamsdonk CD, Griewank KG, Crosby MB, et al. Mutations in GNA11 in uveal melanoma. N Engl J Med. 2010;363:2191-2199. Griewank KG, van de Nes J, Schilling B, et al. Genetic and clinico9. pathologic analysis of metastatic uveal melanoma. Mod Pathol. 2014;27:175-183. 10. Van Raamsdonk CD, Bezrookove V, Green G, et al. Frequent somatic mutations of GNAQ in uveal melanoma and blue naevi. Nature. 2009;457:599-602. 11. Piperno-Neumann S, Kapiteijn E, Larkin J, et al. Landscape of genetic alterations in patients with metastatic uveal melanoma. J Clin Oncol. 2014;32:5s (suppl; abstr 9043). 12. Johansson P, Aoude LG, Wadt K, et al. Deep sequencing of uveal melanoma identifies a recurrent mutation in PLCB4. Oncotarget. 2016;7:4624-4631. 13. Moore AR, Ceraudo E, Sher JJ, et al. Recurrent activating mutations of G-protein-coupled receptor CYSLTR2 in uveal melanoma. Nat Genet. 2016;48:675-680. Harbour JW, Onken MD, Roberson ED, et al. Frequent mutation of 14. BAP1 in metastasizing uveal melanomas. Science. 2010;330:14101413. 15. Harbour JW, Roberson ED, Anbunathan H, et al. Recurrent mutations at codon 625 of the splicing factor SF3B1 in uveal melanoma. Nat Genet. 2013;45:133-135. 16. Martin M, Maßhöfer L, Temming P, et al. Exome sequencing identifies recurrent somatic mutations in EIF1AX and SF3B1 in uveal melanoma with disomy 3. Nat Genet. 2013;45:933-936. 17. Collaborative Ocular Melanoma Study Group. The COMS randomized trial of iodine 125 brachytherapy for choroidal melanoma: V. Twelveyear mortality rates and prognostic factors: COMS report No. 28. Arch Ophthalmol. 2006;124:1684-1693.

648 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

25. Carvajal RD, Piperno-Neumann S, Kapiteijn E, et al. SUMIT: Phase III, randomized, placebo-controlled, double-blind trial of selumetnib in combination with dacarbazine in patients with metastatic uveal melanoma. Presented at: Society for Melanoma Research Congress. San Francisco, CA; November 18–21, 2015. 26. Shoushtari AN, Kudchadkar R, Panageas KS, et al. A randomized phase 2 study of trametinib with or without GSK2141795 in patients with advanced uveal melanoma. J Clin Oncol. 2016;34 (suppl; abstr 9511). 27. Carvajal RD, Schwartz GK, Tezel T, et al. Metastatic disease from uveal melanoma: treatment options and future prospects. Br J Ophthalmol. 2017;101:38-44. Leyvraz S, Piperno-Neumann S, Suciu S, et al. Hepatic intra-arterial 28. versus intravenous fotemustine in patients with liver metastases from uveal melanoma (EORTC 18021): a multicentric randomized trial. Ann Oncol. 2014;25:742-746. 29. Valsecchi ME, Terai M, Eschelman DJ, et al. Double-blinded, randomized phase II study using embolization with or without granulocytemacrophage colony-stimulating factor in uveal melanoma with hepatic metastases. J Vasc Interv Radiol. 2015;26:523-532.e522. 30. Piulats Rodriguez JM, Ochoa de Olza M, Codes M, et al. Phase II study evaluating ipilimumab as a single agent in the first-line treatment of adult patients (Pts) with metastatic uveal melanoma (MUM): The GEM-1 trial. J Clin Oncol. 2014;32:5s (suppl; abstr 9033). Zimmer L, Vaubel J, Mohr P, et al. Phase II DeCOG-study of ipilimumab 31. in pretreated and treatment-naïve patients with metastatic uveal melanoma. PLoS One. 2015;10:e0118564. 32. Joshua AM, Monzon JG, Mihalcioiu C, et al. A phase 2 study of tremelimumab in patients with advanced uveal melanoma. Melanoma Res. 2015;25:342-347. 33. Karydis I, Chan PY, Wheater M, et al. Clinical activity and safety of Pembrolizumab in Ipilimumab pre-treated patients with uveal melanoma. OncoImmunology. 2016;5:e1143997.

MANAGEMENT OF ADVANCED EXTRACUTANEOUS MELANOMAS AND NONMELANOMA SKIN CANCERS

34. Furney SJ, Pedersen M, Gentien D, et al. SF3B1 mutations are associated with alternative splicing in uveal melanoma. Cancer Discov. 2013;3:1122-1129.

52. Ambrosini G, Sawle AD, Musi E, et al. BRD4-targeted therapy induces Myc-independent cytotoxicity in Gnaq/11-mutatant uveal melanoma cells. Oncotarget. 2015;6:33397-33409.

35. Royer-Bertrand B, Torsello M, Rimoldi D, et al. Comprehensive genetic landscape of uveal melanoma by whole-genome sequencing. Am J Hum Genet. 2016;99:1190-1198.

53. Furney SJ, Turajlic S, Stamp G, et al. Genome sequencing of mucosal melanomas reveals that they are driven by distinct mechanisms from cutaneous melanoma. J Pathol. 2013;230:261-269.

36. van Dinten LC, Pul N, van Nieuwpoort AF, et al. Uveal and cutaneous melanoma: shared expression characteristics of melanoma-associated antigens. Invest Ophthalmol Vis Sci. 2005;46:24-30.

54. Schoenewolf NL, Bull C, Belloni B, et al. Sinonasal, genital and acrolentiginous melanomas show distinct characteristics of KIT expression and mutations. Eur J Cancer. 2012;48:1842-1852.

37. Middleton MR, Steven NM, Evans TJ, et al. Safety, pharmaco*kinetics and efficacy of IMCgp100, a first-in-class soluble TCR-antiCD3 bispecific t cell redirector with solid tumour activity: Results from the FIH study in melanoma. J Clin Oncol. 2016;34:15s (suppl; abstr 3016).

55. Yang HM, Hsiao SJ, Schaeffer DF, et al. Identification of recurrent mutational events in anorectal melanoma. Mod Pathol. 2017;30:286296.

38. Shoushtari AN, Evans J, Corrie P, et al. A phase I study of IMCgp100, a soluble HLA-A2 restricted gp100-specific T cell receptor-CD3 therapeutic with solid tumor activity in patients with advanced uveal melanoma. Presented at: Late-breaking Abstract and Oral Presentation at theSociety for Melanoma Research Congress. Boston, MA; November 6–9, 2016. Bol KF, Mensink HW, Aarntzen EH, et al. Long overall survival after 39. dendritic cell vaccination in metastatic uveal melanoma patients. Am J Ophthalmol. 2014;158:939-947. Sisley K, Rennie IG, Parsons MA, et al. Abnormalities of chromosomes 40. 3 and 8 in posterior uveal melanoma correlate with prognosis. Genes Chromosomes Cancer. 1997;19:22-28. Matatall KA, Agapova OA, Onken MD, et al. BAP1 deficiency causes 41. loss of melanocytic cell identity in uveal melanoma. BMC Cancer. 2013;13:371. Landreville S, Agapova OA, Matatall KA, et al. Histone deacetylase 42. inhibitors induce growth arrest and differentiation in uveal melanoma. Clin Cancer Res. 2012;18:408-416. Fu LL, Tian M, Li X, et al. Inhibition of BET bromodomains as a therapeutic 43. strategy for cancer drug discovery. Oncotarget. 2015;6:5501-5516.

56. Pappa KI, Vlachos GD, Roubelakis M, et al. Low mutational burden of eight genes involved in the MAPK/ERK, PI3K/AKT, and GNAQ/11 pathways in female genital tract primary melanomas. Biomed Res Int. 2015;2015:303791. 57. van Engen-van Grunsven AC, Küsters-Vandevelde HV, De Hullu J, et al. NRAS mutations are more prevalent than KIT mutations in melanoma of the female urogenital tract--a study of 24 cases from the Netherlands. Gynecol Oncol. 2014;134:10-14. Carvajal RD, Antonescu CR, Wolchok JD, et al. KIT as a therapeutic 58. target in metastatic melanoma. JAMA. 2011;305:2327-2334. 59. Guo J, Si L, Kong Y, et al. Phase II, open-label, single-arm trial of imatinib mesylate in patients with metastatic melanoma harboring c-Kit mutation or amplification. J Clin Oncol. 2011;29:2904-2909. 60. Hodi FS, Corless CL, Giobbie-Hurder A, et al. Imatinib for melanomas harboring mutationally activated or amplified KIT arising on mucosal, acral, and chronically sun-damaged skin. J Clin Oncol. 2013;31:31823190. 61. Woodman SE, Trent JC, Stemke-Hale K, et al. Activity of dasatinib against L576P KIT mutant melanoma: molecular, cellular, and clinical correlates. Mol Cancer Ther. 2009;8:2079-2085.

Delmore JE, Issa GC, Lemieux ME, et al. BET bromodomain inhibition 45. as a therapeutic strategy to target c-Myc. Cell. 2011;146:904-917.

62. Kalinsky K, Lee SJ, Rubin KM, et al. A phase II trial of dasatinib in patients with unresectable locally advanced or stage IV mucosal, acral, and vulvovagin*l melanomas: A trial of the ECOG-ACRIN Cancer Research Group (E2607). J Clin Oncol. 2016;34 (suppl; abstr 9501).

46. Dawson MA, Prinjha RK, Dittmann A, et al. Inhibition of BET recruitment to chromatin as an effective treatment for MLL-fusion leukaemia. Nature. 2011;478:529-533.

63. Carvajal RD, Lawrence DP, Weber JS, et al. Phase II study of nilotinib in melanoma harboring KIT alterations following progression to prior KIT inhibition. Clin Cancer Res. 2015;21:2289-2296.

47. Lenhart R, Kirov S, Desilva H, et al. Sensitivity of small cell lung cancer to BET inhibition is mediated by regulation of ASCL1 gene expression. Mol Cancer Ther. 2015;14:2167-2174.

64. Carlino MS, Todd JR, Rizos H. Resistance to c-Kit inhibitors in melanoma: insights for future therapies. Oncoscience. 2014;1:423426.

48. Shimamura T, Chen Z, Soucheray M, et al. Efficacy of BET bromodomain inhibition in Kras-mutant non-small cell lung cancer. Clin Cancer Res. 2013;19:6183-6192.

Zimmer L, Eigentler TK, Kiecker F, et al. Open-label, multicenter, 65. single-arm phase II DeCOG-study of ipilimumab in pretreated patients with different subtypes of metastatic melanoma. J Transl Med. 2015;13:351.

Filippakopoulos P, Qi J, Picaud S, et al. Selective inhibition of BET 44. bromodomains. Nature. 2010;468:1067-1073.

49. Henssen A, Althoff K, Odersky A, et al. Targeting MYCN-driven transcription by BET-bromodomain inhibition. Clin Cancer Res. 2016;22:2470-2481.

66. Postow MA, Luke JJ, Bluth MJ, et al. Ipilimumab for patients with advanced mucosal melanoma. Oncologist. 2013;18:726-732.

50. Asangani IA, Dommeti VL, Wang X, et al. Therapeutic targeting of BET bromodomain proteins in castration-resistant prostate cancer. Nature. 2014;510:278-282.

67. Del Vecchio M, Di Guardo L, Ascierto PA, et al. Efficacy and safety of ipilimumab 3mg/kg in patients with pretreated, metastatic, mucosal melanoma. Eur J Cancer. 2014;50:121-127.

51. Parrella P, Caballero OL, Sidransky D, et al. Detection of c-myc amplification in uveal melanoma by fluorescent in situ hybridization. Invest Ophthalmol Vis Sci. 2001;42:1679-1684.

68. Schiavone MB, Broach V, Shoushtari AN, et al. Combined immunotherapy and radiation for treatment of mucosal melanomas of the lower genital tract. Gynecol Oncol Rep. 2016;16:42-46.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 649

KOMATSUBARA ET AL

69. Shoushtari AN, Munhoz RR, Kuk D, et al. The efficacy of anti-PD-1 agents in acral and mucosal melanoma. Cancer. 2016;122:3354-3362. 70. Butler M, Hamid O, Ribas A, et al. Efficacy of pembrolizumab in patients with advanced mucosal melanoma enrolled in the KEYNOTE-001, 002, and 006 studies. Presented at: European Cancer Congress; Amsterdam, the Netherlands. January 27-30, 2017. 71. D’Angelo SP, Larkin J, Sosman JA, et al. Efficacy and safety of nivolumab alone or in combination with ipilimumab in patients with mucosal melanoma: a pooled analysis. J Clin Oncol. 2017;35:226-235. 72. Porceddu SV, Veness MJ, Guminski A. Nonmelanoma cutaneous head and neck cancer and Merkel cell carcinoma: current concepts, advances, and controversies. J Clin Oncol. 2015;33:3338-3345. 73. Burton KA, Ashack KA, Khachemoune A. Cutaneous squamous cell carcinoma: a review of high-risk and metastatic disease. Am J Clin Dermatol. 2016;17:491-508. 74. Traywick C, O’Reilly FM. Management of skin cancer in solid organ transplant recipients. Dermatol Ther (Heidelb). 2005;18:12-18. 75. Brantsch KD, Meisner C, Schönfisch B, et al. Analysis of risk factors determining prognosis of cutaneous squamous-cell carcinoma: a prospective study. Lancet Oncol. 2008;9:713-720. Maubec E, Petrow P, Scheer-Senyarich I, et al. Phase II study 76. of cetuximab as first-line single-drug therapy in patients with unresectable squamous cell carcinoma of the skin. J Clin Oncol. 2011;29:3419-3426. Jenni D, Karpova MB, Mühleisen B, et al. A prospective clinical trial to 77. assess lapatinib effects on cutaneous squamous cell carcinoma and actinic keratosis. ESMO Open. 2016;1:e000003. Pickering CR, Zhou JH, Lee JJ, et al. Mutational landscape of aggressive 78. cutaneous squamous cell carcinoma. Clin Cancer Res. 2014;20:65826592. Falchook GS, Leidner R, Stankevich E, et al. Responses of metastatic 79. basal cell and cutaneous squamous cell carcinomas to anti-PD1 monoclonal antibody REGN2810. J Immunother Cancer. 2016;4:70.

650 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

80. Chang AL, Kim J, Luciano R, et al. A case report of unresectable cutaneous squamous cell carcinoma responsive to pembrolizumab, a programmed cell death protein 1 inhibitor. JAMA Dermatol. 2016;152:106-108. 81. Borradori L, Sutton B, Shayesteh P, et al. Rescue therapy with antiprogrammed cell death protein 1 inhibitors of advanced cutaneous squamous cell carcinoma and basosquamous carcinoma: preliminary experience in five cases. Br J Dermatol. 2016;175:1382-1386. 82. Sekulic A, Migden MR, Oro AE, et al. Efficacy and safety of vismodegib in advanced basal-cell carcinoma. N Engl J Med. 2012;366:21712179. Basset-Seguin N, Hauschild A, Grob JJ, et al. Vismodegib in patients 83. with advanced basal cell carcinoma (STEVIE): a pre-planned interim analysis of an international, open-label trial. Lancet Oncol. 2015;16:729-736. 84. Migden MR, Guminski A, Gutzmer R, et al. Treatment with two different doses of sonidegib in patients with locally advanced or metastatic basal cell carcinoma (BOLT): a multicentre, randomised, double-blind phase 2 trial. Lancet Oncol. 2015;16:716-728. Jayaraman SS, Rayhan DJ, Hazany S, et al. Mutational landscape of 85. basal cell carcinomas by whole-exome sequencing. J Invest Dermatol. 2014;134:213-220. Kaufman HL, Russell J, Hamid O, et al. Avelumab in patients with 86. chemotherapy-refractory metastatic Merkel cell carcinoma: a multicentre, single-group, open-label, phase 2 trial. Lancet Oncol. 2016;17:1374-1385. Nghiem PT, Bhatia S, Lipson EJ, et al. PD-1 blockade with pembroli 87. zumab in advanced Merkel-cell carcinoma. N Engl J Med. 2016;374: 2542-2552. Killock D. Immunotherapy: overcoming checkpoints in Merkel-cell 88. carcinoma. Nat Rev Clin Oncol. 2016;13:328-329. 89. Hauschild A, Schadendorf D. Checkpoint inhibitors: a new standard of care for advanced Merkel cell carcinoma? Lancet Oncol. 2016;17:13371339.

OPERABLE MELANOMA MANAGEMENT

Operable Melanoma: Screening, Prognostication, and Adjuvant and Neoadjuvant Therapy Ahmad A. Tarhini, MD, PhD, Paul Lorigan, MD, and Sancy Leachman, MD, PhD OVERVIEW The importance of reducing the numbers of patients with late-stage melanoma, identifying which patients are most likely to progress, and treating these patients at the earliest possible stage cannot be overemphasized. Improved screening of patients prior to diagnosis has the advantage of identifying early-stage disease that is for the most part treatable by surgical methods. The process of melanoma screening is rapidly evolving through population-based programs, mobile health technologies, and advanced imaging tools. For patients with newly diagnosed melanoma, accurately estimating disease prognosis has important implications for management and follow-up. Prognostic factors are individual host- or tumor-related factors or molecules that correlate with genetic predisposition and clinical course. These include clinical covariates and host and tumor proteomic/genomic markers that allow the prognostic subclassification of patients. Adjuvant therapy for high-risk surgically resected melanoma targets residual micrometastatic disease with the goal of reducing the risk of relapse and mortality. In the United States, three regimens have achieved regulatory approval for adjuvant therapy, including high-dose interferon alpha, pegylated interferon alpha, and ipilimumab at 10 mg/kg. Phase III trials have reported benefits in relapse-free survival (all regimens) and overall survival (high-dose interferon alpha and ipilimumab). The management of locally/regionally advanced melanoma may benefit from neoadjuvant therapy, which is the subject of several ongoing studies. Recent studies have shown promising clinical activity and yielded important biomarker findings and mechanistic insights.

nvestigators in the field of melanoma research have recently developed effective targeted and immune therapies for advanced disease, creating an increased awareness for the role that melanoma can play as a model for the management of other cancers. As a model for cancer care, melanoma has several advantages. For the most part, the primaries are visible to the naked eye and can be easily removed, studied, followed for response to therapy, and used as an in vivo source of immune stimulation. Genetic and clinical risk factors for melanoma are well established, a clear causal environmental factor (ultraviolet radiation) is known for melanoma, and melanoma has a high mutational load. These characteristics allow melanoma to serve as a model cancer for the development and optimization of processes, technologies, and therapeutic regimens that will enhance screening, prognostication, and adjuvant/neoadjuvant therapy for melanoma. It is highly likely that these methods will prove useful for many other, less accessible tumors as well. Screening is important because early detection of melanoma, like many cancers, is associated with substantial improvements in survival.1,2 Because of the visibility of most melanomas on the skin, approximately 70% are detected prior to metastatic spread to lymph nodes or distant sites.2

Screening programs have the potential to improve early detection and reduce mortality from melanoma. Successful screening requires transdisciplinary teams and approaches, including (1) population and public health teams to identify and reach the at-risk population; (2) mobile health technology teams to develop, test, and implement phonebased screening tools; and (3) imaging, computer vision, and photonics teams to enhance the sensitivity and specificity of melanoma identification. Through the use of these screening programs and methodologies, it may be possible to substantially reduce the number of individuals requiring further treatment, resulting in reduced morbidity and mortality and improved quality of life for the individual, and reduced costs to society as well. Surgery remains the mainstay of curative treatment for patients with operable melanoma. Thereafter, patients are treated in a risk-adjusted way, largely based on the American Joint Committee on Cancer (AJCC) staging system and outcome of a sentinel lymph node biopsy. These practices have allowed the identification of different risk groups and have informed decisions on intensity of follow-up, adjuvant therapy, and involvement in clinical trials. Prognosis-driven clinical care has the major advantage of being robust and

From the University of Pittsburgh, Pittsburgh, PA; University of Manchester, Manchester, United Kingdom; Oregon Health & Science University, Portland, OR. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Ahmad A. Tarhini, MD, PhD, University of Pittsburgh, UPMC Cancer Pavilion, 5150 Centre Ave. (555), Pittsburgh, PA 15232; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 651

TARHINI, LORIGAN, AND LEACHMAN

thoroughly validated, and these factors are generally able to be assessed at the time of diagnosis. Current prognostic discriminators tend to divide patients in broad groups and do not provide enough information to accurately predict the likely outcome on an individual patient basis. As an example, it is clear that sentinel node positivity confers a higher risk of recurrence than a negative sentinel node biopsy, yet the majority of patients who develop metastatic disease have had a negative sentinel node biopsy.3 Melanoma is a clear example of how understanding the molecular basis of the tumor and the interaction with the host (i.e., immune system and tumor microenvironment) has led to unprecedented advances in the treatment of advanced disease and the potential for similar benefits in the adjuvant setting. Large clinically annotated tumor banks of primary tumors are being interrogated to better understand the biology of melanoma, predict its behavior, and develop new preventative and adjuvant strategies.4 Systemic adjuvant therapy may benefit patients with resected melanoma who carry a high postoperative risk of relapse and death. Patients with melanoma AJCC stages IIB– IIC, III, or IV whose risk of mortality exceeds 35%–40% at 5 years have historically been categorized as high risk.1 For these patients, residual micrometastatic disease believed to be the source of future disease recurrence may be eliminated with adjuvant therapy. Interferon alpha (IFN-α) has been extensively tested in randomized controlled trials (RCTs) of adjuvant evaluating multiple regimens varying by the formulation of interferon (IFN), dose level (high, intermediate, or low), treatment duration, and route of IFN administration. IFN was shown to have a consistent effect in reducing the risk of relapse across most adjuvant RCTs as well as in four major meta-analyses of IFN adjuvant trials. A reduction in mortality risk was only significantly shown in two of three Eastern Cooperative Oncology Group and U.S. Intergroup RCTs that investigated the 1-year high-dose interferon (HDI) regimen versus observation (E1684 trial) and the ganglio-

KEY POINTS • Various organizations generally agree that individuals who have an increased risk for melanoma should be screened regularly by a provider. • There are a number of important clarifications of definitions and changes in some of the classifications in the forthcoming version 8 of the American Joint Committee on Cancer staging system of melanoma. • Tumor gene expression profiling is becoming an important prognostic tool with promising emerging data but is not yet a standard of care. • In the United States, three regimens have achieved regulatory approval for adjuvant therapy, including highdose interferon alpha, pegylated interferon alpha, and ipilimumab at 10 mg/kg. • The management of locally/regionally advanced melanoma may benefit from neoadjuvant therapy, which is the subject of several ongoing studies. 652 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

side GMK vaccine (E1694 trial). This overall survival (OS) advantage, although small, was also observed in the three largest meta-analyses of adjuvant IFN RCTs. Ipilimumab at a dose of 10 mg/kg as adjuvant therapy has been shown to significantly reduce the risk of relapse and the risk of death in stage III melanoma, but with a relatively high risk of serious toxicities experienced at this dose. This is notable because the dose of ipilimumab approved for the treatment of inoperable metastatic melanoma is 3 mg/kg, which is almost half as toxic as the 10-mg/kg dose. The ongoing U.S. Intergroup Trial E1609 is currently testing ipilimumab at 3 mg/kg or 10 mg/kg compared with HDI among patients with stage III and IV melanoma and is expected to guide the field on the relative safety and efficacy of ipilimumab at 3 and 10 mg/kg versus HDI in the adjuvant setting. Adjuvant studies investigating PD-1 blockade or molecularly targeted therapy are either ongoing or have completed accrual. Ongoing neoadjuvant studies are also testing novel immunotherapeutic and molecularly targeted agents and combinations. Here, we review the latest updates in melanoma screening, prognostication, and adjuvant and neoadjuvant therapy. We also review the current evidence and identify what we feel to be the likely clinical developments in the next few years.

SCREENING FOR PRIMARY MELANOMA

Screening Guidelines

The American Academy of Dermatology5 (AAD) currently recommends that every individual perform a regular skin self-examination and that anyone who notices “any unusual spots on their skin, including those that are changing, itching, or bleeding, should make an appointment with a board-certified dermatologist” and that “…people with an increased risk of melanoma or a history of skin cancer should talk to a dermatologist to determine how often they should receive a skin exam from a doctor.” The U.S. Preventive Services Task Force6 concluded that “the current evidence is insufficient to assess the balance of benefits and harms of visual skin examination by a clinician to screen for skin cancer in adults,” but it also states that “This recommendation applies to asymptomatic adults who do not have a history of premalignant or malignant skin lesions. Patients who present with a suspicious skin lesion or who are already under surveillance because of a high risk of skin cancer, such as those with a familial syndrome (e.g., familial atypical mole and melanoma syndrome), are outside the scope of this recommendation statement.” Furthermore, “For people aged 20 or older who get periodic health examinations, a cancer-related check-up should include health counseling and, depending on a person’s age and gender, examinations for cancers of the thyroid, oral cavity, skin, lymph nodes, testes, and ovaries…,” according to an American Cancer Society7 statement on screening that includes melanoma. International recommendations for melanoma screening exist for several countries, including the United Kingdom, Germany, the Netherlands, Australia, and New Zealand.8-14 In general, these guidelines recommend a complete skin examination

OPERABLE MELANOMA MANAGEMENT

by a medical provider every 3–12 months for individuals with an increased risk of skin cancer, including those with features such as a vulnerable phenotype (fair skin, freckles, red or blonde hair, numerous and/or atypical melanocytic nevi, etc.), a personal history of melanoma, a family history of melanoma or nonmelanoma skin cancer, actinic damage, or immunosuppression. The Royal Australian College of General Practitioners15 suggests that regular screening be provided to individuals who have a sixfold or higher risk of melanoma and that those with average or moderate risk (two- to fivefold elevation) should be screened “opportunistically” when they are being seen for some other purpose by a medical provider. Taken together, these various organizations generally agree that individuals who have an increased risk for melanoma should be screened regularly by a provider.

Screening Methods

Despite general agreement that screening of high-risk patients is important, methods recommended for screening are not as uniform. The AAD16 suggests screening individuals during a visit with a provider and also encourages its membership to participate in community-based free skin cancer screenings. Screenings may also be performed by primary care or nondermatologist specialty providers, but the rate of skin cancer screenings that occur in a nondermatologist clinic-based setting is low, at approximately 10%–15%.17,18 As part of a basic dermatology curriculum, the AAD has an online learning module on performance of a total body skin examination, which includes examination of the entire skin surface (including the scalp, hair, nails, and mucous membranes of the eyes, mouth, anus, and genitalia; www. aad.org/education/basic-derm-curriculum/suggestedorder-of-modules/the-skin-exam). Although this module is designed for medical providers, the AAD also has online resources and brochures to teach lay people how to perform a skin self-examination, which includes the same principles (www.aad.org/public/spot-skin-cancer/learn-about-skincancer/detect). Skin self-screening is important because approximately 53% of melanomas are detected by the patient. Interestingly, dermatologists detect melanoma at an earlier stage, but 80% of melanomas that are detected by dermatologists are seen incidentally rather than during a screening examination.19 Overall, methods of comprehensive skin examination are accessible and can be readily applied to melanoma screening.

Detecting Suspicious Lesions

With respect to the identification of melanoma during a skin examination, the AAD has long promoted the ABCDs of melanoma, a mnemonic for visual signs that suggest a pigmented lesion is at risk for being melanoma, including asymmetry, border irregularity, color variation, and diameter greater than 6 mm.20 The AAD has extended the ABCDs to include an E for evolution, to capture the fact that a changing lesion can be suspicious.21 The ugly duckling sign has also gained acceptance as a clue to malignancy, capturing the concept

that a pigmented lesion with a different clinical appearance relative to others on the same individual may be a more sensitive marker of melanoma than the ABCDEs.22 In addition to these clinical tools for detecting suspicious lesions, dermoscopic evaluation, using polarized light and magnification, has been refined and has become an essential component of a good skin screening examination.23 Multispectral and hyperspectral dermoscopy as well as in vivo confocal microscopy and optical coherence tomography are in development and may become next-generation gold standards for melanoma screening.24,25 The final frontier may well be the application of machine learning and artificial intelligence to the problem, creating a more objective and reliable interpretation of digital images of nevi and melanoma that surpass that of the human eye.26

PROGNOSTIC FACTORS FOR PATIENTS WITH OPERABLE MELANOMA: WHAT IS NEW?

AJCC Staging

The AJCC staging for melanoma has been revised (cancerstaging.org/references-tools/deskreferences/pages/default.aspx) and version 8 will come into use on January 1, 2018. At the time of this writing, the survival outcomes data on which this revision is based have not yet been published. Notable changes for primary and locoregional disease include the following: • All principal T-category tumor thickness ranges are maintained, but T1 is now subcategorized by tumor thickness strata at 0.8-mm thickness. • The tumor mitotic rate was removed from staging criteria for T1 tumors. • Sentinel lymph node tumor burden is not used to determine N-category group. • Nodal disease is classified as clinically occult or clinically detected. • Microsatellites, satellites, and in-transit metastases are clarified. • The contribution of lactate dehydrogenase in designating the M subcategories is revised. Clinical implications and likely developments. These revisions clarify several areas of uncertainty and make more accurate staging easier. The development of staging algorithms linked to outcomes data, readily accessible in the form of an app, will be a key resource for clinicians and patients. Further, investing in the discovery and validation of prognostic biomarkers will allow us to more accurately stage our patients and several exciting developments are emerging in this area as we summarize next.

Circulating Tumor Cells, Cell-Free DNA, and Circulating Tumor DNA

Techniques used to detect circulating tumor cells (CTCs) in the blood of patients with melanoma are based on the expression of melanocyte-specific markers, distinctive physical properties, or melanocyte-specific nucleic sequences.27 Marker-based technologies have been shown to be prognostic in metastatic melanoma but are limited by the fact asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 653

TARHINI, LORIGAN, AND LEACHMAN

that there is no single marker uniformly expressed in melanoma and that the number of CTCs is lower than in other tumors.28 The isolation by size of epithelial tumor cells is a direct method for CTC identification and has been validated for melanoma but is labor intensive and operator dependent.29 A major focus has been on the detection of CTCs by the analysis of mRNA melanocyte-specific transcripts; tyrosinase, Melan-A/Mart-1, gp-100, and the MAGE proteins are commonly used, with reverse transcription polymerase chain reaction. A number of studies examined this strategy in stage I–III melanoma.30 In the Sunbelt trial, reverse transcription polymerase chain reaction was performed on peripheral blood mononuclear cells at baseline for 207 patients with stage III disease using four markers: tyrosinase, Melan-A/Mart-1, MAGE3, and gp-100. Baseline reverse transcription polymerase chain reaction status was not associated with substantial differences in outcomes.31 Hoshimoto et al32 evaluated patients with stage III disease entering a randomized adjuvant melanoma vaccine program. Samples were drawn only once after the radical surgery and CTCs were detected using a multimarker reverse transcription polymerase chain reaction. The presence of two or more positive biomarkers was significantly associated with shorter distant metastasis–free survival (hazard ratio [HR], 2.13; 95% CI, 1.20–3.76; p = . 009). The analysis of cell-free DNA in plasma in the metastatic setting has been shown to be predictive of survival and can be used to monitor response to treatment and identify the emergence and mechanisms of resistance.33 These techniques are being applied to high-risk patients with resected disease in the context of randomized adjuvant trials. Clinical implications. The results of cell-free DNA analysis from large adjuvant trials are awaited.

Gene Expression Profiling

Gene expression profiling (GEP) to determine prognostic and predictive factors has enormous potential in both the metastatic and adjuvant settings. Several profiles have been identified but, apart from incorporation of the 15 gene–based DecisionDX-UM assay into the staging of uveal melanoma, none have yet received overwhelming support. DecisionDx-Melanoma is a GEP based on 28 genes and is available commercially. This assay was initially developed on a discovery panel of 107 cases with stage I–IV cutaneous melanoma and was expanded to a training set of 268 cases, including the discovery set.34 A further 104 independent cases were then studied as validation. Patients were classified in a binary way as having either low risk (class 1) or high risk (class 2). GEP and AJCC stage were independent predictors of metastatic risk (HR, 9.55 vs. 5.40, respectively). A second study of the subgroup of 217 patients with a sentinel node (SLN) biopsy (58 SLN+ and 159 SLN−) showed that both SLN+ and GEP class 2 were important predictors of disease-free survival, distant metastasis-free survival, and OS.35 In multivariate analysis, for each point, GEP had a higher HR and only GEP was notable for OS. The greatest discrimination was seen for the 42% of patients who had an 654 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

SLN− biopsy but were identified as class 2 and had a 5-year OS of 55%. This was very similar to the 57% 5-year OS for the patients with a SLN+ biopsy who were identified as class 2. More recently, a study on a further 334 primary tumors in which 29% of patients went on to develop distant metastases reported that the sensitivity, specificity, and positive and negative predictive power of both SLN and GEP were very similar.36 Of note, 13 of 83 patients with an SLN− biopsy (16%) went on to have a distant metastatic event, and 10 of these patients (77%) were class 2. The authors concluded that DecisionDX-Melanoma deserves prospective evaluation. This is a fair assessment. The use of samples retrospectively identified to develop a molecular predictor is entirely justifiable, but there are many examples in which early results are not borne out in randomized trials. The recent failure of the MAGE3 ASCI vaccine for patients with stage III disease, and the failure of the molecular predictor to identify those likely to benefit, despite a very strong signal in a phase II study, is a salutatory lesson.37 Furthermore, we must consider the consequences of adopting this technology. A validation study of the clinical impact on 156 patients reported that changes in management (e.g., changes in the use of imaging as follow-up, and changes in the frequency of clinical review) were observed for 82 patients (53%), with the majority of class 2 patients (77%) undergoing management changes compared with 37% of class 1 patients (p < .0001). The role of imaging in the follow-up of patients with melanoma remains unclear, and the variation in clinical guidelines reflects this. European Society for Medical Oncology guidelines indicate that there is currently no consensus on the frequency of follow-up examinations and the use of imaging techniques. The National Comprehensive Cancer Network guidelines advise considering imaging patients with stage IIB–IV NED. More important will be whether decisions on adjuvant therapy can be made on the basis of GEP, given that this has not been included as a stratification factor in the pivotal studies. Given the multitude of randomized controlled adjuvant trials that have been carried out in melanoma, there is an opportunity for independent validation of this technology in a randomized sample of patients. Clinical implications. GEP is becoming an important tool for clinical decision making but is not yet a standard of care.

Vitamin D

Most human diets provide little vitamin D, so humans rely on synthesis from sunlight. Over the last 8 years, the role of vitamin D in melanoma has been increasingly recognized. Results from three large cohort studies indicate that vitamin D has an effect on both melanoma susceptibility and outcome.38 The initial report from the Leeds group described a prospective cohort study of 872 patients with a Breslow thickness greater than 0.75 mm.39 Higher vitamin D levels at diagnosis were associated with a lower Breslow thickness (p = .002), were protective of risk of relapse (HR, 0.79; 95% CI, 0.6–0.96; p = .01) for a 20-nmol/L increase in serum level,

OPERABLE MELANOMA MANAGEMENT

and were associated, to a lesser extent, with a reduced risk of death (HR, 0.83; 95% CI, 0.68–1.02). In multivariable analyses (adjusted for tumor thickness), death from melanoma was associated with a low vitamin D level at recruitment (< 20 nmol/L vs. 20–60 nmol/L; HR, 1.52; 95% CI, 0.97–2.40; p = .07) and smoking duration at diagnosis (HR, 1.11; 95% CI, 1.03–1.20; p = .009). The Melan Cohort Study measured vitamin D levels for 1,171 patients.40 The investigators found that vitamin D levels were inversely related to AJCC stage (p, .001), Breslow thickness (p < .001), and ulceration (p, .001). However, they found no association with risk of relapse. In a large study from MD Anderson Center, investigators prospectively collected samples from 1,042 patients with melanoma to assess C reactive protein (CRP) and vitamin D levels. After adjustment for age, sex, disease stage, blood-draw season, and log-transformed CRP, vitamin D levels remained significantly associated with outcome measures for OS (HR, 1.02; 95% CI, 1.01–1.04; p = .0051), melanoma-specific survival (HR, 1.02; 95% CI, 1.00–1.04: p = .048), and disease-free survival (HR, 1.02; 95% CI, 1.00–1.04; p = .0427). Similar results have been reported for smaller series of patients.41,42 Clinical implications. There is currently insufficient evidence to establish a cause-and-effect relationship between vitamin D and melanoma recurrence and death, or to define the potential mechanisms for such an effect. Nevertheless, recommendations to measure vitamin D levels at baseline and advise supplementation if they are low have been added to some national guidelines and do not seem unreasonable.

Other Prognostic Factors

Many other potential prognostic factors have been studied and some have made their way into clinical guidelines and clinical practice, despite a lack of randomized data. These include S-100β, melanoma inhibitory activity, CRP, lactate dehydrogenase, and serum cytokines.43 In a retrospective study, investigators examined 127 patients with primary melanoma who were followed up with regular imaging and assessment of S-100β levels and subsequently had a recurrence of the disease.44 Beyeler et al44 reported that 37% of patients with recurrent disease had elevated S-100 at the time of recurrence, and it was the first indicator in 5.5%. Increased S-100β was more common for patients with regional nodal disease or metastatic disease than local recurrence or in-transit metastases. A further study of 296 patients with stage II or III disease in which metastasis occurred for 41% reported a sensitivity of detection of relapse of 29% for S-100β, 22% for melanoma inhibitory activity, and 2% for lactate dehydrogenase.45 Serum S-100β protein was found to be an important prognostic marker among high-risk patients with melanoma participating in the E1694 adjuvant trial and may improve patient selection for adjuvant therapy.43 Multiplex analysis of serum cytokines in high-risk patients treated with IFNα in the E1694 adjuvant trial showed that baseline proinflammatory cytokine levels may predict 5-year relapse-free survival (RFS) with IFNα.46 A follow-up study reported a four-marker signature consisting of baseline

serum tumor necrosis factor-RII, transforming growth factor-alpha, TIMP-1, and CRP that is prognostic of worse survival in high-risk surgically resected melanoma in the E1694 trial.47 In addition, in the E1697 adjuvant trial, an early on-treatment (after 1 month of adjuvant IFNα) proinflammatory cytokine signature consisting of interleukin 2 receptor alpha, interleukin 12p40, and IFN predicted 1-year RFS with IFNα but not observation in intermediate-risk surgically resected melanoma.48 Overall, some melanoma follow-up guidelines (e.g., National Comprehensive Cancer Network and National Institute for Health and Care Excellence) state that blood tests are not indicated in follow-up, whereas others (e.g., European Society for Medical Oncology) present the data but do not make a recommendation. Clinical implications. Although lactate dehydrogenase has a defined role in the AJCC staging of melanoma, the roles of S-100β, MIA, CRP, and cytokine profiles remain unclear and unproven and efforts to validate them in the context of randomized studies (e.g., E1609 trial) are ongoing.

ADJUVANT AND NEOADJUVANT THERAPY OF MELANOMA

Randomized Clinical Trials of Adjuvant Therapy That Led to Regulatory Approval

The immunomodulatory effects of IFNα have been widely studied and include antiangiogenic activity, differentiation-inducing and proinflammatory effects, as well as direct antitumor proapoptotic and antiproliferative activity.49 Studies testing IFNα at a high dose (> 10 MU/dose) as adjuvant therapy first included the North Central Cancer Treatment Group NCCTG 83-7052 study50 and the Eastern Cooperative Oncology Group ECOG E1684 trial.51 In the E1684 trial, the HDI regimen consisted of an intravenous induction phase administered at 20 MU/m2 for 5 consecutive days a week for 4 weeks, followed by a subcutaneous maintenance phase at 10 MU/m2 three times a week for 48 weeks. Eligibility criteria required regional elective lymph node dissection for patients without clinical evidence of nodal involvement. This study enrolled 287 patients and the majority had clinically detectable nodal disease or recurrent melanoma after prior surgery. The study was reported after a median follow-up of 6.9 years, showing that HDI improved both RFS and OS compared with observation. Five-year RFS was 37% (95% CI, 30%–46%) versus 26% (95% CI, 19%–34%) and 5-year OS was 46% (95% CI, 39%–55%) versus 37% (95% CI, 30%–46%) in favor of HDI.52 The HRs and key findings are summarized in Table 1. In terms of toxicity, the incidence of grade 3 and 4 adverse events was 67% with HDI and two early hepatotoxicity grade 5 events were reported. The U.S. Food and Drug Administration approved HDI as an adjuvant therapy in 1995.8 The E1690 trial investigated HDI and low-dose IFN for 2 years compared with postoperative observation was first reported after a median follow-up of 4.3 years. Five-year RFS was 44% with HDI, 40% with low-dose IFN, and 35% with observation.53 Compared with observation, HDI significantly improved RFS (p = .03). On the other hand, no OS benefit was seen, with 5-year OS rates of 52%, 53%, and 55% with asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 655

TARHINI, LORIGAN, AND LEACHMAN

TABLE 1. High-Risk Melanoma Adjuvant Trials Leading to Regulatory Approval in the United States

Study

Stage at Study Entry Size

E1684

T4, N+

287

Hazard Ratio

Adjuvant Regimen vs. Control

Median Followup at Reporting (Years) RFS

p Value

Key Findings

HDI vs. Observation

6.9

.001

0.67

.01

Majority bulky nodal or recurrent disease

0.61

Greatest benefit in high tumor burden (N1)

E1690

E1694

EORTC 18991

EORTC 18071

T4, N+

642

880

N1 1,256 (occult), N2 (bulky)

N1,2,3 (except in transit)

951

HDI or LDI vs. observation

12.6

0.72

.02

0.82

4.3

0.78

.05

1.0

.18

At 12.6 years, competing causes of death may have affected OS analysis Consistent RFS benefit Unlike E1684, E1690 did not require ELND

6.6

0.81

.09

1.0

HDI vs. GMK 1.3 vaccine for 96 weeks

0.67

.0004

0.72

2.1

0.75

.006

0.76

3.8

0.82

.011

0.98

RFS benefit seen in N1 (occult) but not in N2 (bulky) disease

7.6

0.87

.055

0.96

Greatest benefit seen in N1 with ulcerated primary

5.3

0.76

.0008

0.72

Pegylated IFN-α vs. observation

Ipilimumab 10 mg/kg vs. placebo

Cross-over of 38 Obs patients to HDI at nodal relapse .023

Minority bulky nodal disease Greatest benefit in lower tumor burden (T4, N−)

.04

.001

RFS correlates with OS (as in E1684)

Consistent RFS and OS benefits Substantial toxicity at this dose requiring close follow-up and expertise in irAE management

Abbreviations: ELND, elective lymph node dissection; EORTC, European Organisation for Research and Treatment of Cancer; HDI, high-dose interferon; IFN-α, interferon alpha; irAE, immune-related adverse event; LDI, low-dose interferon; Obs, observation; OS, overall survival; RFS, relapse-free survival.

HDI, low-dose IFN, and observation, respectively (Table 1). It is noteworthy that unlike the E1684 trial, the E1690 trial did not require elective lymph node dissection, and a retrospective analysis showed cross-over of 38 patients at nodal recurrence from the observation arm to standard-of-care HDI adjuvant therapy. This cross-over may have affected the OS analysis, which is supported by the observation that OS in the observation arm of the E1690 trial was superior to that in the E1684 trial (median 6 vs. 2.8 years). U.S. Intergroup Trial E1694 followed and tested patients who received HDI versus the ganglioside vaccine GMK, which consisted of ganglioside GM2 coupled to keyhole limpet hemocyanin and was combined with the adjuvant QS-21. HDI was significantly better than GMK in regard to RFS (HR, 0.67) and OS (HR, 0.72) in the intent-to-treat analysis.52,54 A pooled analysis of HDI trials with follow-up through April 2001 was later conducted and included the two observation controlled trials (E1684 and E1690). At a median follow-up of 12.6 years and 6.6 years for the E1684 and E1690 trials, respectively, HDI maintained substantial RFS benefits.55 However, no substantial improvement in OS was seen. In the E1684 trial with the 656 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

relatively very long follow-up in the intent-to-treat analysis of OS, competing causes of death were not accounted for and may have confounded OS data.51,53-55 In the European Organisation for Research and Treatment of Cancer EORTC 18991 trial, pegylated interferon alfa-2b was compared with observation as adjuvant therapy for patients with AJCC stage III melanoma.22 The adjuvant 5-year subcutaneous regimen consisted of an induction phase (6 μg/kg a week for 8 weeks) followed by a maintenance phase (3 μg/kg a week). The study showed significant improvement in the primary endpoint of RFS (HR, 0.87; 95% CI, 0.76–1.00; p = .05; median follow-up 7.6 years) in favor of pegylated interferon alfa-2b, whereas no substantial differences in distant metastasis–free survival or OS were seen. In the subgroup analysis, patients with ulcerated primary melanoma and microscopic nodal metastasis appeared to derive the greatest RFS and OS benefits. The median duration of treatment was 14 months and the toxicity attrition rate was 37%. Four major meta-analyses of adjuvant IFN RCTs have been reported since 2002, and all have concluded that adjuvant IFN therapy has substantial RFS benefits.56-58 The

OPERABLE MELANOMA MANAGEMENT

largest and most recent was the Cochrane Analysis of Adjuvant Melanoma Trials, which included 17 RCTs and 10,499 participants.58 This meta-analysis estimated HRs of 0.83 (95% CI, 0.78–0.87) and 0.91 (95% CI, 0.85–0.97) in terms of RFS and OS, respectively. Ipilimumab is an anti-CTLA4 fully humanized immunoglobulin G1 kappa monoclonal antibody. In the treatment of patients with inoperable sage III/IV melanoma, ipilimumab significantly improved OS when tested at 3 mg/kg versus the Gp100 peptide vaccine,59 and at 10 mg/kg combined with dacarbazine versus dacarbazine alone.60 EORTC 18071 was an adjuvant trial of the 10-mg/kg dose of ipilimumab versus placebo for patients with resected stage III melanoma.61 This trial demonstrated significant benefits in RFS and OS with adjuvant ipilimumab. At a median follow-up of 5.3 years, median RFS was 27.6 months (95% CI, 19.3, 37.2) versus 17.1 (95% CI, 13.6, 21.6), with an HR of 0.76 (95% CI, 0.64, 0.89; p = .0008). For OS, the HR was 0.72 (95% CI, 0.58, 0.88; p = .001). Five-year survival rates were 65% versus 54% for OS and 41% versus 30% for RFS. Safety results demonstrated a high rate of immune-related adverse events, including a 41% rate of grade 3/4 immune-related adverse events and five grade 5 events secondary to immune-related adverse events after treatment with ipilimumab.

The Leading Ongoing Adjuvant Studies in High-Risk Resected Melanoma

U.S. Intergroup Trial E1609, lead by the Eastern Cooperative Oncology Group/American College of Radiology Imaging Network, is testing ipilimumab at 3 mg/kg or 10 mg/kg as adjuvant therapy in high-risk resected stage III (IIIB, IIIC) and IV (M1a, M1b) melanoma versus HDI (NCT01274338). This study has two coprimary endpoints (RFS and OS) and is trying to answer important questions related to the relative safety of ipilimumab at 3 and 10 mg/kg as well as their efficacy relative to HDI. Early results from this trial are expected to be presented during the 2017 ASCO Annual Meeting. CheckMate 238 is an adjuvant trial testing PD-1 blockade with nivolumab versus 10 mg/kg of ipilimumab and accrual was completed in 2015 (NCT02388906). KEYNOTE-054 is testing adjuvant therapy with pembrolizumab compared with placebo (NCT02362594). U.S. Intergroup Trial S1404 is comparing adjuvant therapy with pembrolizumab to the control arm of standard adjuvant therapy with HDI or ipilimumab at the dose of 10 m/k. Adjuvant trials of molecularly targeted therapy in BRAF-mutant high-risk resected melanoma are also ongoing and have completed accrual. These include COMBI-AD, which is comparing the combination of dabrafenib and trametinib to placebo (NCT01682083), and BRIM8, which is comparing vemurafenib to placebo (NCT01667419).

Neoadjuvant Therapy of Locally and Regionally Advanced Melanoma

Neoadjuvant therapy has the potential to significantly improve the clinical outcome of patients with locally/regionally advanced melanoma, particularly in this era of newer and

effective targeted and immunotherapeutic agents. Such studies also provide access to biospecimens before and during therapy, allowing for the conduct of biomarker and mechanistic studies that may have an important impact in drug development. On the other hand, neoadjuvant therapy carries the risk of toxicity from systemic therapy and the risk of delaying the indicated surgical procedure, although the chances of cure with surgery alone are low in patients who may be eligible for neoadjuvant therapy. Previous neoadjuvant studies tested chemotherapy with temozolomide in a phase II study, in which oral temozolomide was given at 75 mg/m2 per day for 6 weeks of an 8-week cycle with two cycles administered preoperatively. The clinical activity was limited and similar to the response rates in metastatic disease.62 Biochemotherapy was tested in two studies in the neoadjuvant setting. The biochemotherapy regimen consisted of cisplatin, vinblastine, dacarbazine, interleukin-2, and IFNα.63,64 The response rates approached 40%–50%, including pathologic complete remission of 6%–11%. However, biochemotherapy therapy was subsequently abandoned after the results of RCTs of metastatic disease showed no survival advantage over chemotherapy.65 Neoadjuvant immunotherapy studies in melanoma reported to date included HDI, 10 mg/kg of ipilimumab, and the combinations of HDI with ipilimumab (3 or 10 mg/kg), nivolumab with ipilimumab, and dabrafenib with trametinib.66-68 In the neoadjuvant HDI study, among 20 patients, 3 had pathologic complete remission and the overall clinical response rate was 55%. Substantial nodal infiltration by CD3+/CD11+ monocyte-derived dendritic cells was found among responders.69 In the neoadjuvant ipilimumab study, no pathologic complete remission was seen and the clinical response rate approached 10%. Mechanistically, substantial findings were reported, including the impact of ipilimumab on downregulating myeloid-derived suppressor cells and inducing tumor-specific T-cell responses and T-cell memory, found to be associated with clinical benefit.70 Notably, the combination of HDI and ipilimumab yielded a 39% rate of pathologic complete remission or microscopic residual disease after 6 weeks of neoadjuvant therapy. Although 10 mg/kg of ipilimumab was associated with increased toxicity compared with 3 mg/kg, the clinical activity was similar.66 These studies yielded evidence of promising clinical activity and important biomarker and biologic findings that further illuminate the underlying mechanisms of action.70,71 These findings support later combination studies of IFNα and pembrolizumab and a modified regimen of ipilimumab and nivolumab (both ongoing). Studies of other molecularly targeted and immunotherapeutic agents and combinations are ongoing in the neoadjuvant setting, and updates from ongoing studies are expected to be presented during the annual meeting.

CONCLUSION

Screening for melanoma includes performing a clinical screening examination among appropriate high-risk individuals, applying a rigorous method of examination, and asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 657

TARHINI, LORIGAN, AND LEACHMAN

using strategies and technologies to enhance detection of suspicious melanocytic lesions before metastasis. The field of biomarkers of prognosis is progressing at a rapid pace, with several important leads reported in recent years. Large validation studies are needed, given the important implications that such biomarkers may have in the care of patients with melanoma. As adjuvant therapy for high-risk resected melanoma, substantial benefits were shown with IFNα (HDI, pegylated IFN) and ipilimumab in RCTs. Phase III trials have reported benefits in RFS (all regimens) and OS (HDI and 10 mg/kg of ipilimumab). The toxicity of ipilimumab

is dose dependent; after the recent regulatory approval of adjuvant ipilimumab at 10 mg/kg, it has become urgent to evaluate the relative safety and efficacy of ipilimumab at the two dose levels that were tested in the E1609 trial. Other ongoing adjuvant trials are testing BRAF/MEK inhibitors for patients with BRAF mutant melanoma and anti–PD-1 antibodies, with early results expected in the coming 2–3 years. Neoadjuvant therapy of locally/regionally advanced melanoma offers the potential to significantly improve the clinical outcome of these patients and several studies are ongoing in this area.

References 1. Balch CM, Gershenwald JE, Soong SJ, et al. Final version of 2009 AJCC melanoma staging and classification. J Clin Oncol. 2009;27:6199-6206.

melanoma without distant metastases (summary report)]. Bull Cancer. 2006;93:371-384.

2. National Cancer Institute Surveillance, Epidemiology, and End Results Program. Cancer Stat Facts: melanoma of the skin, 2016. http://seer. cancer.gov/statfacts/html/melan.html. Accessed January 31, 2017.

12. German Guideline Program in Oncology. Evidence-based Guideline on Prevention of Skin Cancer. http://leitlinienprogramm-onkologie.de/ uploads/media/Short_version_-_Guideline_on_prevention_of_skin_ cancer.pdf. Accessed January 30, 2017.

3. Morton DL, Thompson JF, Cochran AJ, et al; MSLT Group. Sentinelnode biopsy or nodal observation in melanoma. N Engl J Med. 2006;355:1307-1317. 4. Kruit WH, Suciu S, Dreno B, et al. Selection of immunostimulant AS15 for active immunization with MAGE-A3 protein: results of a randomized phase II study of the European Organisation for Research and Treatment of Cancer Melanoma Group in Metastatic Melanoma. J Clin Oncol. 2013;31:2413-2420. 5. Lebwohl MG. AAD statement on skin cancer screening. https:// www.aad.org/media/news-releases/aad-statement-on-skin-cancerscreening. Accessed January 30, 2017. 6. Bibbins-Domingo K, Grossman DC, Curry SJ, et al. Screening for skin cancer: U.S. Preventive Services Task Force recommendation statement. JAMA. 2016;316:429-435. 7. American Cancer Society. Cancer Facts & Figures, 2015. http://www. cancer.org/acs/groups/content/@editorial/documents/document/ acspc-044552.pdf. Accessed January 30, 2017. 8. Marsden JR, Newton-Bishop JA, Burrows L, et al; British Association of Dermatologists (BAD) Clinical Standards Unit. Revised UK guidelines for the management of cutaneous melanoma 2010. J Plast Reconstr Aesthet Surg. 2010;63:1401-1419. 9. Cancer Council Australia; Australian Cancer Network. Clinical Practice Guidelines for the Management of Melanoma in Australia and New Zealand. http://www.cancer.org.au/content/pdf/Health Professionals/ClinicalGuidelines/ClinicalPracticeGuidelinesManagementofMelanoma.pdf. Accessed January 30, 2017. 10. Cancer Council Australia. Position statement - Screening and early detection of skin cancer. http://wiki.cancer.org.au/policy/Position_ statement_-_Screening_and_early_detection_of_skin_cancer. Accessed January 30, 2017. 11. Négrier S, Saiag P, Guillot B, et al; Standards, Options and Recommendations; Société Française de Dermatologie; Fédération Nationale des Centres de Lutte contre le Cancer; Institut National du Cancer; Ligue Nationale contre le Cancer; Fédération Hospitalière de France; Fédération Nationale de Cancerologie des CHRU; Fédération Francaise de Cancérologie. [Clinical practice guideline: 2005 update of recommendations for the management of patients with cutaneous

658 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

13. Johnson M. Skin cancer screening: recommendations for datadriven screening guidelines and a review of the USPSTF controversy. Melanoma Manag. In press. 14. Dutch Working Group on Melanoma. Melanoma Guideline 2012 (2013). http://www.oncoline.nl/uploaded/docs/melanoom/201208_ vertalingRichtlijnmelanoomdef.pdf. Accessed January 30, 2017. 15. Royal Australian College of General Practitioners. Guideline 9.1:Skin cancer. http://www.racgp.org.au/your-practice/guidelines/redbook/ early-detection-of-cancers/skin-cancer. Accessed January 30, 2017. 16. American Academy of Dermatology. Detect skin cancer (2015). https://www.aad.org/public/spot-skin-cancer/learn-about-skincancer/detect. Accessed February 27, 2017. 17. LeBlanc WG, Vidal L, Kirsner RS, et al. Reported skin cancer screening of US adult workers. J Am Acad Dermatol. 2008;59:55-63. 18. Coups EJ, Geller AC, Weinstock MA, et al. Prevalence and correlates of skin cancer screening among middle-aged and older white adults in the United States. Am J Med. 2010;123:439-445. 19. Avilés-Izquierdo JA, Molina-López I, Rodríguez-Lomba E, et al. Who detects melanoma? Impact of detection patterns on characteristics and prognosis of patients with melanoma. J Am Acad Dermatol. 2016;75:967-974. 20. Friedman RJ, Rigel DS. The clinical features of malignant melanoma. Dermatol Clin. 1985;3:271-283. 21. Tsao H, Olazagasti JM, Cordoro KM, et al; American Academy of Dermatology Ad Hoc Task Force for the ABCDEs of Melanoma. Early detection of melanoma: reviewing the ABCDEs. J Am Acad Dermatol. 2015;72:717-723. 22. Grob JJ, Bonerandi JJ. The ‘ugly duckling’ sign: identification of the common characteristics of nevi in an individual as a basis for melanoma screening. Arch Dermatol. 1998;134:103-104. 23. Forsea AM, Tschandl P, Del Marmol V, et al; Eurodermoscopy Working Group. Factors driving the use of dermoscopy in Europe: a panEuropean survey. Br J Dermatol. 2016;175:1329-1337. 24. Song E, Grant-Kels JM, Swede H, et al. Paired comparison of the sensitivity and specificity of multispectral digital skin lesion analysis

OPERABLE MELANOMA MANAGEMENT

and reflectance confocal microscopy in the detection of melanoma in vivo: a cross-sectional study. J Am Acad Dermatol. 2016;75:11871192.

42. Bade B, Zdebik A, Wagenpfeil S, et al. Low serum 25-hydroxyvitamin d concentrations are associated with increased risk for melanoma and unfavourable prognosis. PLoS One. 2014;9:e112863.

25. Leachman SA, Cassidy PB, Chen SC, et al. Methods of melanoma detection. Cancer Treat Res. 2016;167:51-105. 26. Esteva A, Kuprel B, Novoa RA, et al. Dermatologist-level classification of skin cancer with deep neural networks. Nature. 2017;167:51-105.

43. Tarhini AA, Stuckert J, Lee S, et al. Prognostic significance of serum S100B protein in high-risk surgically resected melanoma patients participating in Intergroup Trial ECOG 1694. J Clin Oncol. 2009;27:3844.

27. Krebs MG, Sloane R, Priest L, et al. Evaluation and prognostic significance of circulating tumor cells in patients with non-small-cell lung cancer. J Clin Oncol. 2011;29:1556-1563.

44. Beyeler M, Waldispuhl S, Strobel K, et al. Detection of melanoma relapse: first comparative analysis on imaging techniques versus S100 protein. Dermatology. 2006;213:187-191.

28. Khoja L, Lorigan P, Zhou C, et al. Biomarker utility of circulating tumor cells in metastatic cutaneous melanoma. J Invest Dermatol. 2013;133:1582-1590.

45. Garbe C, Leiter U, Ellwanger U, et al. Diagnostic value and prognostic significance of protein S-100beta, melanoma-inhibitory activity, and tyrosinase/MART-1 reverse transcription-polymerase chain reaction in the follow-up of high-risk melanoma patients. Cancer. 2003;97:17371745.

29. Khoja L, Shenjere P, Hodgson C, et al. Prevalence and heterogeneity of circulating tumour cells in metastatic cutaneous melanoma. Melanoma Res. 2014;24:40-46. 30. Khoja L, Lorigan P, Dive C, et al. Circulating tumour cells as tumour biomarkers in melanoma: detection methods and clinical relevance. Ann Oncol. 2015;26:33-39. 31. Scoggins CR, Ross MI, Reintgen DS, et al. Prospective multiinstitutional study of reverse transcriptase polymerase chain reaction for molecular staging of melanoma. J Clin Oncol. 2006;24:2849-2857. 32. Hoshimoto S, Shingai T, Morton DL, et al. Association between circulating tumor cells and prognosis in patients with stage III melanoma with sentinel lymph node metastasis in a phase III international multicenter trial. J Clin Oncol. 2012;30:3819-3826. 33. Girotti MR, Gremel G, Lee R, et al. Application of sequencing, liquid biopsies, and patient-derived xenografts for personalized medicine in melanoma. Cancer Discov. 2016;6:286-299. 34. Gerami P, Cook RW, Wilkinson J, et al. Development of a prognostic genetic signature to predict the metastatic risk associated with cutaneous melanoma. Clin Cancer Res. 2015;21:175-183.

46. Yurkovetsky ZR, Kirkwood JM, Edington HD, et al. Multiplex analysis of serum cytokines in melanoma patients treated with interferonalpha2b. Clin Cancer Res. 2007;13:2422-2428. 47. Tarhini AA, Lin Y, Yeku O, et al. A four-marker signature of TNF-RII, TGF-α, TIMP-1 and CRP is prognostic of worse survival in high-risk surgically resected melanoma. J Transl Med. 2014;12:19. 48. Kunitomo K, Irie RF, Kern DH. Modulation of ganglioside expression in human melanoma cell lines; increased resistance to chemo- and radiation treatment. Tokushima J Exp Med. 1992;39:55-62. 49. Kirkwood JM, Richards T, Zarour HM, et al. Immunomodulatory effects of high-dose and low-dose interferon alpha2b in patients with highrisk resected melanoma: the E2690 laboratory corollary of intergroup adjuvant trial E1690. Cancer. 2002;95:1101-1112. 50. Creagan ET, Dalton RJ, Ahmann DL, et al. Randomized, surgical adjuvant clinical trial of recombinant interferon alfa-2a in selected patients with malignant melanoma. J Clin Oncol. 1995;13:2776-2783.

35. Gerami P, Cook RW, Russell MC, et al. Gene expression profiling for molecular staging of cutaneous melanoma in patients undergoing sentinel lymph node biopsy. J Am Acad Dermatol. 2015;72:780-5.e3.

51. Kirkwood JM, Strawderman MH, Ernstoff MS, et al. Interferon alfa2b adjuvant therapy of high-risk resected cutaneous melanoma: the Eastern Cooperative Oncology Group Trial EST 1684. J Clin Oncol. 1996;14:7-17.

36. Zager JS, Messina J, Sondak VK, et al. Performance of a 31-gene expression profile in a previously unreported cohort of 334 cutaneous melanoma patients. J Clin Oncol. 2016;34 (suppl; abstr 9581).

52. Tarhini AA, Kirkwood JM. How much of a good thing? What duration for interferon alfa-2b adjuvant therapy? J Clin Oncol. 2012;30:37733776.

37. Dreno B, Thompson JF, Smithers B. DERMA, a double-blind, randomized, placebo-controlled phase III study to assess the efficacy of MAGE-A3 cancer immunotherapeutic as adjuvant therapy in patients with resected MAGE-A3-positive stage III melanoma. Paper presented at: Society for Melanoma 2015 Research Congress; November 2015; San Francisco, CA.

53. Kirkwood JM, Ibrahim JG, Sondak VK, et al. High- and low-dose interferon alfa-2b in high-risk melanoma: first analysis of intergroup trial E1690/S9111/C9190. J Clin Oncol. 2000;18:2444-2458.

38. Reichrath J, Rech M, Moeini M, et al. In vitro comparison of the vitamin D endocrine system in 1,25(OH)2D3-responsive and -resistant melanoma cells. Cancer Biol Ther. 2007;6:48-55.

54. Kirkwood JM, Ibrahim JG, Sosman JA, et al. High-dose interferon alfa2b significantly prolongs relapse-free and overall survival compared with the GM2-KLH/QS-21 vaccine in patients with resected stage IIBIII melanoma: results of intergroup trial E1694/S9512/C509801. J Clin Oncol. 2001;19:2370-2380.

39. O'Shea SJ, Davies JR, Newton-Bishop JA. Vitamin D, vitamin A, the primary melanoma transcriptome and survival. Br J Dermatol. 2016;Suppl 2:30-34.

55. Kirkwood JM, Manola J, Ibrahim J, et al; Eastern Cooperative Oncology Group. A pooled analysis of eastern cooperative oncology group and intergroup trials of adjuvant high-dose interferon for melanoma. Clin Cancer Res. 2004;10:1670-1677.

40. Saiag P, Aegerter P, Vitoux D, et al. Prognostic value of 25-hydroxyvitamin D3 levels at diagnosis and during follow-up in melanoma patients. J Natl Cancer Inst. 2015;107:djv264.

56. Wheatley K, Ives N, Hanco*ck B, et al. Does adjuvant interferon-alpha for high-risk melanoma provide a worthwhile benefit? A meta-analysis of the randomised trials. Cancer Treat Rev. 2003;29:241-252.

41. Wyatt C, Lucas RM, Hurst C, et al. Vitamin D deficiency at melanoma diagnosis is associated with higher Breslow thickness. PLoS One. 2015;10:e0126394.

57. Mocellin S, Pasquali S, Rossi CR, et al. Interferon alpha adjuvant therapy in patients with high-risk melanoma: a systematic review and meta-analysis. J Natl Cancer Inst. 2010;102:493-501.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 659

TARHINI, LORIGAN, AND LEACHMAN

58. Mocellin S, Lens MB, Pasquali S, et al. Interferon alpha for the adjuvant treatment of cutaneous melanoma. Cochrane Database Syst Rev. 2013;6:CD008955. 59. Hodi FS, O’Day SJ, McDermott DF, et al. Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med. 2010;363:711-723. 60. Robert C, Thomas L, Bondarenko I, et al. Ipilimumab plus dacarbazine for previously untreated metastatic melanoma. N Engl J Med. 2011;364:2517-2526. 61. Eggermont AM, Chiarion-Sileni V, Grob JJ, et al. Adjuvant ipilimumab versus placebo after complete resection of high-risk stage III melanoma (EORTC 18071): a randomised, double-blind, phase 3 trial. Lancet Oncol. 2015;16:522-530. 62. Shah GD, Socci ND, Gold JS, et al. Phase II trial of neoadjuvant temozolomide in resectable melanoma patients. Ann Oncol. 2010;21:1718-1722. 63. Buzaid AC, Colome M, Bedikian A, et al. Phase II study of neoadjuvant concurrent biochemotherapy in melanoma patients with localregional metastases. Melanoma Res. 1998;8:549-556. 64. Gibbs P, Anderson C, Pearlman N, et al. A phase II study of neoadjuvant biochemotherapy for stage III melanoma. Cancer. 2002;94:470-476. 65. Tarhini AA, Edington H, Butterfield LH, et al. Immune monitoring of the circulation and the tumor microenvironment in patients with

660 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

regionally advanced melanoma receiving neoadjuvant ipilimumab. PLoS One. 2014;9:e87705. 66. Kanazawa J, Ohta S, sh*tara K, et al. Therapeutic potential of chimeric anti-(ganglioside GD3) antibody KM871: antitumor activity in xenograft model of melanoma and effector function analysis. Cancer Immunol Immunother. 2000;49:253-258. 67. Moschos SJ, Edington HD, Land SR, et al. Neoadjuvant treatment of regional stage IIIB melanoma with high-dose interferon alfa-2b induces objective tumor regression in association with modulation of tumor infiltrating host cellular immune responses. J Clin Oncol. 2006;24:3164-3171. 68. Tarhini AA. Neoadjuvant therapy for melanoma: a promising therapeutic approach and an ideal platform in drug development. Am Soc Clin Oncol Educ Book. 2015;35:e535-e542. 69. Tarhini AA, Pahuja S, Kirkwood JM. Neoadjuvant therapy for high-risk bulky regional melanoma. J Surg Oncol. 2011;104:386-390. 70. Tarhini AA, Zahoor H, Lin Y, et al. Baseline circulating IL-17 predicts toxicity while TGF-β1 and IL-10 are prognostic of relapse in ipilimumab neoadjuvant therapy of melanoma. J Immunother Cancer. 2015;3:39. 71. Scott AM, Liu Z, Murone C, et al. Immunological effects of chimeric anti-GD3 monoclonal antibody KM871 in patients with metastatic melanoma. Cancer Immun. 2005;5:3.

TREATMENT OF INOPERABLE MELANOMA

Systemic Therapy Options for Patients With Unresectable Melanoma Melinda Yushak, MD, MPH, Paul Chapman, MD, Caroline Robert, MD, PhD, and Ragini Kudchadkar, MD OVERVIEW There has been a therapeutic revolution in the treatment of metastatic melanoma over the past decade. Patients presenting with inoperable disease have several therapeutic options, which can include both targeted and immune therapy. Immune checkpoint inhibitors have demonstrated an improvement in overall survival and led to some durable responses. However, toxicity, especially in combination regimens, can be severe. Adverse events should be anticipated, diagnosed as early as possible, monitored, and managed. Combination BRAF and MEK inhibition has also been shown to improve overall survival in patients with V600E-mutated melanoma. Responses to therapy are often rapid, and treatment is not associated with immune-related adverse events. Current trials are under way to determine which option is optimal as frontline therapy for patients with V600E melanoma. In patients with progressive disease despite standard therapies, clinical trials are recommended. There are several promising agents in development.

he MAPK or ERK pathway has been well described, and a detailed discussion of the pathway is beyond the scope of this review. However, as an overview, Fig. 1A outlines the ERK pathway in a normal cell. A receptor tyrosine kinase is activated by binding of its ligand, which ultimately leads to RAS activation and the generation of RAS-GTP. This promotes RAF hetero- and hom*odimerization among the three isoforms of RAF: A-RAF, B-RAF, and C-RAF. In Fig. 1A, a B-RAF/C-RAF heterodimer is shown, along with RAF monomers. The RAF dimer can activate MEK, which activates ERK, leading to cellular proliferation through several pathways. ERK activation also leads to negative feedback mediated through DUSP6, which suppresses RAS-GTP formation and serves to modulate the activity of the pathway under normal circ*mstances.

TARGETED THERAPY

General Pathway Overview

In a melanoma cell harboring a BRAF V600E mutation, the pathway is quite different. The V600E mutant BRAF is sufficient to hyperactive MEK even as a monomer. This leads to hyperactivated ERK and cellular proliferation. It also leads to strong negative feedback on to RAS, which helps prevent dimerization of RAF molecules. Thus, in BRAF V600-mutated melanoma cells, ERK activation is driven by BRAF V600 monomers, whereas in the normal cell, ERK activation takes place through RAF dimers. This distinction is critical and explains the unique therapeutic index of RAF inhibitors. In a BRAF V600E–mutated

melanoma cell, the RAF inhibitor binds to the monomers and inhibits their function. This suppresses MEK and ERK and leads to cell death. However, in a normal cell, when a RAF inhibitor binds to one of the RAF molecules of a dimer, it causes activation of the other RAF molecule, which leads to some increase in ERK activation. This paradoxical activation is thought to play a role in the cutaneous toxicities caused by RAF inhibitors. RAF inhibitors have also been found to increase proliferation of malignancies driven by RAS mutations,1,2 underscoring the importance of verifying that the melanoma harbors a BRAF V600 mutation before treating with a RAF inhibitor.

Medical Efficacy of RAF Inhibitors and RAF/MEK Combinations

Two RAF inhibitors are currently approved by the U.S. Food and Drug Administration (FDA) for the treatment of BRAF V600-mutant melanoma: vemurafenib and dabrafenib. Both drugs have been shown in randomized trials to have superior progression-free survival (PFS) compared with dacarbazine. The improvements in median PFS were 5.5 months3 and 5.1 months,4 respectively. However, the estimated PFS at 1 year was less than 30% for vemurafenib (not reported for dabrafenib). This highlights the observation that BRAF V600E–mutated melanomas quickly develop resistance to single-agent RAF inhibitors. We now know that common mechanisms of resistance are amplification and/or overexpression of the mutated BRAF allele, a splicing variant of mutated BRAF that permits dimerization with

From the Winship Cancer Institute of Emory University, Atlanta, GA; Memorial Sloan Kettering Cancer Center, New York, NY; Institute Gustave Roussy, Villejuif, France. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Ragini Kudchadkar, MD, 1365C Clifton Rd. NE, Atlanta, GA 30322; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 661

YUSHAK ET AL

FIGURE 1. MAPK Kinase Pathway

In a BRAF wild-type cell (A), activation of a receptor tyrosine kinase at the cell surface results in NRAS activation. This promotes RAF activation by hom*o- and heterodimerization, leading to MEK and ERK activation. Negative feedback mechanisms serve to modulate NRAS activation and thus output through the pathway. In a BRAF V600E–mutated melanoma cell (B), MEK is activated directly by BRAF V600E monomers. This leads to hyperactivation of ERK and strong negative feedback, which inhibits RAF dimerization.

wild-type RAF protein even in the absence of activated RAS, and the appearance of an upstream activating NRAS mutation.5,6 In an attempt to foil the melanoma cell’s ability to develop resistance, RAF inhibitors have been combined with MEK inhibitors. Two different RAF inhibitor/MEK inhibitor combinations are currently FDA approved for use in patients with melanoma harboring a BRAF V600 mutation: dabrafenib/ trametinib and vemurafenib/cobimetinib. In randomized phase III trials comparing the combination with the RAF inhibitor alone, these combinations have shown median improvements in PFS of 1.2 months7 and 5.1 months,8 respectively (Table 1). Although these differences were statistically significant, the magnitude of benefit was perhaps disappointing. Despite this, both combinations were associated with improved overall survival (OS) compared with RAF inhibitor alone. The improvement in OS despite a relatively minor improvement in PFS, especially for the dabrafenib/trametinib combination, suggests the possibility

KEY POINTS • Immune checkpoint inhibitors have demonstrated an improvement in overall survival and led to some durable responses. • Combination BRAF/MEK inhibition leads to rapid responses and has shown an improvement in overall survival. • Both combination BRAF/MEK inhibition and immunotherapy are first-line options for patients with BRAF-mutated V600E melanoma. • Adverse events should be anticipated, diagnosed as early as possible, monitored, and managed. • There are several promising agents in development targeting both BRAF resistance mechanisms and immune checkpoint agonists and antagonists. 662 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

that a minority of patients experience prolonged PFS. This is supported by the PFS curves from the dabrafenib/trametinib trial, which do indeed suggest a plateau at about the 18-month time point.7 A plateau in the PFS curve from the vemurafenib/cobimetinib trial was less evident, but there was minimal follow up beyond 18 months. Therefore, the data are consistent with the notion that the combination of a RAF and MEK inhibitor can lead to prolonged PFS in a minority of patients harboring a BRAF V600E mutation and that this can improve OS. As result, standard of care is currently a RAF/MEK inhibitor combination rather than a single-agent RAF inhibitor. A randomized trial testing a third RAF/MEK inhibitor combination, encorafenib/binimetinib, has been completed. Array Biopharma announced in a press release in September 2016 that combination encorafenib plus binimetinib had a PFS of 14.9 months as compared to 7.3 months for vemurafenib.

Toxicity of RAF/MEK Inhibitor Combinations and Management Guidelines

It is not surprising that, in general, RAF inhibitor plus MEK inhibitor combinations are associated with more toxicity than single-agent RAF inhibitors. Compared with dabrafenib alone, the dabrafenib/trametinib combination was associated

TABLE 1. Improvement in PFS and OS With RAF/MEK Inhibition Compared to RAF Inhibition Alone PFS (Months)

OS (Months)

Reference

Dabrafenib/ Trametinib

1.2

6.4

Long et al7

Vemurafenib/ Cobimetinib

5.1

4.9

Ascierto et al8

Abbreviations: OS, overall survival; PFS, progression-free survival.

TREATMENT OF INOPERABLE MELANOMA

with a higher incidence of fever, chills, and diarrhea.7 However, there were some adverse events (AEs) that were less frequent with the combination, such as arthralgia, handfoot syndrome, alopecia, keratoacanthoma, and cutaneous squamous cell carcinoma, which are thought to be mediated by paradoxical activation by RAF inhibitors and thus blocked by MEK inhibition. Overall, 7% of patients in the dabrafenib group had to discontinue therapy because of AEs, compared with 11% in the combination group. Fever and chills were the most common reasons for discontinuing the dabrafenib/ trametinib combination. Experience has shown that this toxicity needs to be managed actively by discontinuing treatment until the fever has resolved. Usually treatment can be restarted, but over the first 1 to 2 months, several drug holidays may be needed. Also, nonsteroidal antiinflammatory drugs or low-dose prednisone (5–10 mg/day) can be helpful. Similarly, the combination of cobimetinib/vemurafenib was associated with more toxicity than vemurafenib alone. Patients receiving the combination were more likely to experience nausea, vomiting, photosensitivity, and elevations in transaminases. Patients receiving cobimetinib/trametinib also experienced more creatine phosphokinase elevations with decreased ejection fractions and serous retinopathy, both of which are considered MEK inhibitor–associated AEs. Distinct from dabrafenib/trametinib, the cobimetinib/ vemurafenib combination is not associated with pyrexia. However, similar to the dabrafenib/trametinib experience, AEs led 14% of patients to discontinue combination therapy, compared with only 7% of patients receiving vemurafenib alone. Also, similar to dabrafenib/trametinib, the combination of cobimetinib/vemurafenib was associated with a lower incidence of arthralgia, alopecia, keratoacanthoma, and cutaneous squamous cell carcinoma.

Treatment With BRAF/MEK Inhibitors

First-line therapy for patients with BRAF V600E–mutated melanoma. Patients with melanoma who harbor a BRAF V600E mutation have two first-line treatment options: RAF/ MEK inhibition therapy and checkpoint inhibition. Both are known to have high response rates and to be associated with improved OS. One of the remaining questions in the field is which treatment should be recommended first? Clinical trials combining checkpoint inhibitors with MAPK pathway inhibitors are under way, but until these trials provide data for guidance, first-line therapy for most patients in this situation will be either RAF/MEK or checkpoint inhibition. Checkpoint inhibitors appear to lead to a higher percentage of durable responses, and for this reason, checkpoint inhibitor therapy is often selected as the first-line therapy, even if the melanoma is known to harbor a BRAF V600E mutation. On the other hand, there are clinical situations in which a RAF inhibitor may be preferred as first-line therapy, such as a patient with rapidly progressive disease or an active autoimmune disease or a patient who requires immunosuppressive therapy for some other reason. At present,

patients with brain or bone metastases are often offered RAF inhibitors as first-line therapy because RAF inhibitors are known to have activity in these disease sites, while the response rate to checkpoint inhibitors at these sites is not well characterized. However, this could change as additional studies are performed. Ipilimumab was shown to have a 22% response rate in patients with brain metastases not on steroids,9 and a recent small trial with pembrolizumab in patients with melanoma with brain metastases reported responses in four of 16 patients.10 The response rate of brain metastases to the nivolumab/ipilimumab combination has not yet been reported, but an ongoing study is evaluating this concept (CheckMate 204). Treatment of melanoma with non-V600 BRAF mutations. Clinical research in melanoma has naturally focused largely on the V600E and K mutations, as these are by far the most common BRAF mutations. In addition, these mutations result in the highest ERK activation and, through feedback mechanisms, presumably are the most suppressive of RASGTP. This ensures that RAF remains monomeric and, as a result, highly susceptible to inhibition by vemurafenib or dabrafenib. However, as more melanoma tumors are sequenced by next-generation sequencing techniques, we are learning that many melanomas harbor BRAF mutations at codons other than 600. Figure 2 shows data from 167 patients with melanoma in the MSKCC cBioPortal database11,12 in whom a BRAF mutation was identified. In 33% of cases, the mutation was not at codon 600. Many of these mutations were within the kinase domain, and some are known to be oncogenic, although the level of ERK activation is thought to be lower than for a V600E mutation.13 Most of the other BRAF mutations are of unknown oncogenic potential, and many may be simply passenger mutations. It is common for these less activating non-V600 BRAF mutations to be found with concomitant RAS mutations or loss of NF1. This is consistent with in vitro data showing that cells harboring these mutations are not sensitive to RAF inhibitors.13 Treatment of these tumors with pathway inhibitors remains a subject for clinical trials. These cells may be sensitive to MEK inhibitors but will probably require inhibition at more than one level of the pathway.

IMMUNOTHERAPY WITH CHECKPOINT INHIBITORS

General Pathway PD-1/CTLA-4 Overview

CTLA-4 and PD-1 (CD279) belong to the so-called immune checkpoint molecules, meaning that they physiologically participate in the negative control of the immune response. These molecules exert their action at different levels. CTLA4 is expressed on T cells and regulates the early stages of T-cell activation in the lymphoid organs. Approximately 22 days after naive T cells have been activated by the binding of their antigen, expression of CTLA-4 counteracts the costimulation interaction replacing CD28 and binding to both CD80 and CD86 with a much higher affinity than that of CD28, thus ending the process of T-cell activation.14,15 TREG cells also express CTLA-4 and are susceptible to antibody-dependent asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 663

YUSHAK ET AL

cell-mediated cytotoxicity after the binding of anti-CTLA-4 antibody to Fcγ receptors expressed on tumor-associated monocytes or macrophages. Thus, the anticancer effect of the blockade of CTLA-4 results from the enhancement of T-cell activation in lymphoid organs, as well as from TREG depletion in the tumor microenvironment. As a result of this nonspecific activation of the immune response, by broad effector T-cell activation and TREG depletion, especially given that TREG cells have a critical role in self-tolerance, a broad spectrum of immune-related AEs was expected. Indeed, in murine models, knocking out CTLA-4 induced lymphoproliferation and serious autoimmune diseases. Certain polymorphisms in the human CTLA-4 gene are associated with an increased risk of autoimmune diseases,16,17 including rheumatoid arthritis, Addison disease, celiac disease, Crohn disease, type I diabetes, and thyroid disorders. PD-1 is also negative regulator of T-cell activity but acts mainly within the peripheral tissues. In the context of cancer, PD-1 is expressed on activated tumor-infiltrating lymphocytes (mainly CD4+ T cells) as well as on B cells, natural killer cells, monocytes, and dendritic cells.18 When engaged with one of its two ligands, PD-1 phosphorylation activates intracellular phosphatases that lead to a dramatic downregulation of antigen receptor signaling with decreased proliferation as well as decreased cytokine production. The two PD-1 ligands are PD-L1 (B7-1H and CD274) and PD-L2 (B7-DC and CD273).19 PD-L1 can be expressed on tumor cells and is induced by interferon gamma but can also be expressed within the tumor microenvironment by immune-infiltrating cells, including infiltrating macrophages. PD-L2 is expressed by antigen-presenting cells. In preclinical models, inhibition of the interaction between PD-1 and PD-L1 generates antitumor activity and enhances autoimmunity with an autoimmune phenotype distinct from that observed in mice that are deficient in CTLA-4.20

Review of Clinical Efficacy Data

Single-agent CTLA-4 inhibition: ipilimumab. CTLA-4 was the first immune checkpoint receptor to be clinically targeted. Initially two fully humanized monoclonal antibodies directed

against CTLA-4 were evaluated in patients with metastatic melanoma, an IgG1, ipilimumab, and an IgG2, tremelimumab. Only ipilimumab development was successful in this population of patients, with the demonstration of a significant survival benefit compared with a peptide-base vaccine in pretreated patients.21,22 Although the response rate was low, about 10%, and median PFS was only 2 to 3 months, the reduction of the risk for death was significant. A second phase III trial, comparing ipilimumab in combination with dacarbazine versus dacarbazine monotherapy, demonstrated a similar survival benefit, with a significant advantage over chemotherapy.23 However, the combination of dacarbazine did not seem to add any benefit, and the hepatotoxicity was higher, so this combination was not pursued. Ipilimumab is given in four infusions at 3-week intervals, and the two major phase III trials used two different doses of ipilimumab, 3 and 10 mg/kg. The drug was approved in 2011 at the lower dose of 3 mg/kg, but a recent trial demonstrated an OS benefit of the dose of 10 mg/kg compared with 3 mg/kg).24 However, the toxicity is also higher with a higher dose. With the hindsight we have today, we know that the survival curve of ipilimumab reaches a plateau after 3 years, with a stable survival rate of 23% to 25%, and that patients who respond to ipilimumab and are alive at 3 years have a high probability of remaining alive. Thus we can reasonably hope that some patients who were treated several years ago are definitely cured. Both ipilimumab and tremelimumab are currently being actively investigated for the treatment of other cancer types. Single-agent anti–PD-1 inhibition: nivolumab and pembrolizumab. The human and humanized PD-1–blocking monoclonal IgG4 antibodies nivolumab and pembrolizumab, respectively, were rapidly and successively developed for patients with metastatic melanoma in recent years. Pembrolizumab was evaluated in KEYNOTE-001, a multicohort phase I study that enrolled 655 ipilimumab-naive or ipilimumab-pretreated patients, and was approved by the FDA on the basis of the results of this study in September 2014.25,26 KEYNOTE-002, a phase II trial in patients pretreated with ipilimumab, and KEYNOTE-006, a large phase III trial testing two doses of pembrolizumab compared with

FIGURE 2. BRAF Mutations Among 167 Patients With Melanoma With BRAF Mutations on the MSKCC cBioPortal11,12

Mutations at codon 600 accounted for 67% of BRAF mutations, but gene alterations were seen throughout the gene, as depicted by the green circles. Red arrows indicate alterations that are known to be oncogenic. All are within the kinase domain. Data are lacking regarding most of the other mutations shown.

664 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

TREATMENT OF INOPERABLE MELANOMA

ipilimumab, confirmed the spectacular results of the phase I study.27,28 When analyzing these studies together, we observed an objective response rate of about 35% to 40%, median PFS of about 6 months, and median OS of about 2 years with pembrolizumab. Nivolumab followed a parallel and similarly successful development in patients with metastatic melanoma, with a large phase I study enrolling 117 patients,29,30 a phase II study, and the phase III Checkmate 037 trial comparing nivolumab with chemotherapy in patients pretreated with ipilimumab.31 All these trials were largely positive and confirmed the benefit of nivolumab, with response rates, median PFS, and OS similar to those seen in pembrolizumabtreated patients. Nivolumab obtained FDA and European Medicines Agency approval in 2014 and 2015, respectively. Combination PD-1/CTLA-4 inhibition: ipilimumab and nivolumab. Because CTLA-4 and PD-1 have distinct mechanisms of action, combined blockade of these two receptors was evaluated in patients with metastatic melanoma. In a phase I trial combining ipilimumab and nivolumab, promising results were observed in a cohort of 53 patients, with response rates of about 50% and dramatic and rapid tumor regression in most of the responders.32 In a double-blind, randomized phase II study, Checkmate 069, evaluating ipilimumab and nivolumab compared with ipilimumab and in a phase III trial evaluating the same combination compared with each of the agents separately, a similarly high response rate and prolonged PFS of close to 1 year were demonstrated.33,34 Results on OS will soon be available.

Toxicity of Immunotherapy and Management Guidelines

Checkpoint inhibitors represent a revolution in the treatment of metastatic melanoma and more generally in the field of cancer, with several FDA and European Medicines Agency approvals obtained and many others pending in various cancer types. However, their use is not devoid of AEs, and as expected from their mechanisms of action, most of these AEs result from an exacerbated immune activation, some of them mirroring genuine autoimmune diseases. The term immune-related AEs (irAEs) is now commonly used to describe these AEs. Grade 3 to 5 AEs are more frequent with ipilimumab than with anti–PD-1 monotherapies, 23% to 25% versus 13% to 15%, respectively. When used in combination, toxicities are much more frequent and severe, with AEs of any grade observed in almost every patient and grade 3 to 5 AEs in 55% to 60%, and 40% of patients interrupting treatment of toxicity.35 Because some of these irAEs can be severe and even fatal, their early diagnosis and management is of paramount importance. Ipilimumab. The most frequent AEs associated with ipilimumab are pruritus (in 25%–35% of patients), diarrhea (in 23%–33% of patients), rash (in 15%–33% of patients), and fatigue (in 15%–28% of patients). Grade 3 and 4 AEs are reported in 20% to 27% of patients, the most frequent being diarrhea (in 3%–6% of patients).35 At the dose of 10 mg/kg,

AEs are more frequent, as recently published in a European Organisation for Research and Treatment of Cancer trial evaluating ipilimumab 10 mg/kg in the adjuvant setting in a population of patients at high risk for relapse, in which five patients died in relation to the treatment.36 Enterocolitis and/or diarrhea. Colitis is reported in 8% to 22% of patients treated with ipilimumab, but the diagnosis of colitis is not standardized and does not always rely on a colonoscopic examination. The incidence of diarrhea and colitis increases with the dose of ipilimumab.35 Grade 3 diarrhea is the most frequent AE leading to discontinuation of treatment by patients receiving ipilimumab. Arthralgia is observed in up to one-fourth of patients presenting with ipilimumab-induced enterocolitis. Endoscopic investigations show erythema, mucosal friability, or ulceration, predominantly in the distal colon. Histologic features of ipilimumab-induced colitis include neutrophilic inflammation, lymphocytic infiltration, or both.37 Inflammation of the oral mucosa, esophagus, stomach, duodenum, and ileum might also occur. Several lines of evidence suggest that ipilimumabinduced enterocolitis is a peculiar form of inflammatory bowel disease with features of ulcerative colitis (inflammation predominating in the colon) and Crohn disease (reflecting possible involvement of the distal ileum and granuloma).38,39 Fatal bowel perforation can rarely occur, especially if the colitis is not readily recognized and treated. Skin-related events. Skin-related irAEs occur in 43% to 45% of patients receiving ipilimumab, with nonspecific maculopapular rash, pruritus, and vitiligo being the most commonly observed skin AEs. They are usually of low grade and do not impair treatment continuation, although rare cases of patients presenting with life-threatening StevensJohnson syndrome and toxic epidermal necrolysis have been reported.35 Endocrine-related events. Ipilimumab can affect the endocrine system, especially the pituitary gland, and 6% to 8% of patients present with panhypopituitarism or isolated anterior pituitary hormone deficiency. Symptoms can be quite pleomorphic and include fatigue, headache, vertigo, memory difficulties, and visual disturbances that can be confounded with brain metastases. When an endocrinopathy is suspected, a complete endocrine work-up is necessary to determine pituitary, thyroid, adrenal, and gonadal functions, and imaging of the brain and pituitary gland can also be performed to look for an enlarged pituitary gland and potential brain metastases.35,40 More rarely, irAEs such as pancreatitis, hepatitis, neurologic toxicities including Guillain-Barré syndrome, meningoradiculoneuritis, granulomatous inflammation of the central nervous system, and aseptic meningitis; ocular toxicities including episcleritis, uveitis, and autoimmune polymyositis; and immune cytopenia have been reported.35,41 Anti–PD-1 antibodies. Anti–PD-1 monotherapies with pembrolizumab or nivolumab are associated with fewer irAEs than ipilimumab. Drug-related AEs of any grade reportedly occurred in 70% to 80% of patients, but only 13% to 15% present with grade 3 or worse AEs.35,41 asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 665

YUSHAK ET AL

The most common AEs of any grade are fatigue (in 24% to 35% of patients), pruritus (in 10% to 23%), rash (in 12% to 21%), and diarrhea (in 11% to 20%) and are of low grade in the vast majority of patients (≥ 95%). Unlike with ipilimumab, anti–PD-1-induced diarrhea and colitis are rare, and fewer than 2% of patients present with grade 3 or 4 diarrhea or colitis. Overall, pembrolizumab-related grade 3 and 4 AEs occurred in 12% to 14% of patients. The spectrum of autoimmune AEs reported with anti–PD-1 monoclonal antibodies is also distinct from that observed with ipilimumab. Thyroid dysfunction is more frequent with anti–PD-1 therapy; about 10% of patients have definitive hypothyroidism requiring hormone replacement therapy, often preceded by transient hyperthyroidism. In contrast, colitis and hypophysitis are more frequent with ipilimumab. Pneumonitis is also more frequent with anti–PD-1 treatment than with ipilimumab (2%–4% vs. 1%) but is rarely severe. Treatment-related AEs rarely lead to discontinuation of anti–PD-1 treatment, in fewer than 10% of patients. Toxicity of ipilimumab/nivolumab combined therapy. The combination of ipilimumab and nivolumab results in more frequent, rapid, and severe AEs. Any grade treatment-related AEs are observed in more than 90% of patients and reach grades 3 to 5 in 53% of patients. The most common AEs are rash (55%), pruritus (47%), fatigue (38%), diarrhea (34%), nausea (21%), and pyrexia (21%). Elevated levels of lipase (13%), aspartate aminotransferase (13%), and alanine aminotransferase (11%) are the most common grade 3 and 4 AEs. The onset of the majority of the AEs occurs during the 12-week induction phase, when the two drugs are administered concomitantly.32,35,42,43 Treatment interruption for toxicity is required in 40% of patients, which does not seem to affect the benefit of treatment of those patients who are responding. Management of AEs. The successful management of toxicities requires that AEs be anticipated, diagnosed as early as possible, monitored, and treated appropriately.35,40,44 Obtaining clear information from patients and the physicians involved in their management is critical. Close collaboration with expert specialists, such as gastroenterologists, hepatologists, endocrinologists, neurologists, and dermatologists, can be useful. In general, grade 1 and 2 AEs are managed symptomatically and do not require treatment discontinuation. For persistent grade 2 AEs, dose skipping and symptomatic treatments are prescribed. Treatment discontinuation is recommended when grade 2 AEs persist despite the symptomatic measures after 1 to 2 weeks and for patients with grade 3 or 4 AEs. In the latter cases, referral to an organ specialist can be considered. If AEs are immune mediated, and after ruling out an active infection, corticosteroids are indicated in patients with persistent grade 2 or grade 3 or 4 irAEs. Typically 0.5 to 1 mg/kg of prednisone is prescribed and should be continued until symptoms resolve, then very progressively tapered over at least 4 weeks. Rash and pruritus are usually mild to moderate and treated symptomatically with emollient antihistamines and/or topical steroids or systemic steroids when refractory or severe. 666 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

For grade 1 and 2 diarrhea, antidiarrheal agents (loperamide) and oral hydration are prescribed. In case of persistent grade 2 or grade 3 or 4 diarrhea, sigmoidoscopy or a colonoscopy is recommended. Other causes of enterocolitis, such as ischemia and/or infection, should be excluded by performing a stool test for viral (cytomegalovirus), bacterial pathogens, and Clostridium difficile toxins, and checkpoint inhibitor treatment should be discontinued and rehydration and systemic steroids initiated. Steroid treatment can be orally administered in grade 2 disease but may require intravenous administration of high doses of steroids (1–2 mg/kg) for more severe cases, also with intravenous rehydration. In severe cases, treatment with checkpoint inhibitors should be permanently discontinued. If symptoms do not improve significantly after 5 days of intravenous corticosteroids, treatment with 5 mg/kg doses of the anti-TNF antibody infliximab should be initiated, and patients should be closely monitored because of the risk of bowel perforation. Endocrine disorders. Plasma levels of cortisol, adrenocorticotropic hormone, and thyroid hormones should be regularly monitored and checked immediately in the case of symptoms or biology suggestive of endocrinopathy. Hormone replacement can be a therapeutic emergency in cases of hypophysitis and should sometimes be initiated without waiting for a confirmed diagnosis, with immediate hospitalization and intravenous administration of corticosteroids with mineralocorticoid activity. Endocrinopathies are usually irreversible and require lifelong hormonal replacement. Symptoms suggestive of rare AEs such as pneumonitis, uveitis, and/or neuropathies should be looked for by monitoring patients for any sign of dyspnea, eye pain or blurred vision, or neurologic abnormality. In patients with preexisting autoimmune disorders, treatment should be discussed on a case-by-case basis, and a potential flare of the autoimmune disease should be put in balance with the potential benefit from the treatment in the context of a metastatic and fatal disease. The potential relationship between irAEs and clinical response to checkpoint inhibitors is not completely elucidated. A positive correlation between vitiligo and patients’ objective response rates was found in a prospective study of 67 patients treated with pembrolizumab.45

In Which Patients Is Immunotherapy Appropriate?

Many new questions are arising with this new treatment paradigm. The identification of biomarkers would allow us to address the critical questions of which patients to treat with immunotherapy and, more specifically, with monotherapy as opposed to a combination of immunotherapies or, in the case of BRAF mutation, the optimal way to sequence or combine targeted anti-BRAF agents with immunotherapy. Role of PD-L1 testing in selecting patients for single-agent PD-1 versus combination. PD-L1 expression on tumor cells and/or immune cells has been well studied, and in almost every study, high expression of PD-L1 is associated with a better clinical outcome during treatment with anti–PD-1, nivolumab, or pembrolizumab.46,47 This association is weaker

TREATMENT OF INOPERABLE MELANOMA

in the context of the ipilimumab/nivolumab combination, although the response rates are higher in the subgroup of patients with high PD-L1 expression.48 However, the positive or predictive value of PD-L1 expression is not sufficient as a single marker to guide treatment. Many reasons can be put forward to explain this. First, there are technical issues that are not standardized: several antibodies are used by different companies, these distinct antibodies do not recognize the same portion of the molecule, and their respective performance in recognizing PD-L1 expressed on tumor or immune cells is variable.46 The threshold of positivity is different from one antibody to another, and standardization of techniques has not been performed. In addition, we know that the single expression of PD-L1 can result from an intrinsic intracellular molecular pathway and does not always result from the surrounding presence of activated T cells secreting interferon gamma. Indeed, the presence of T cells and, more precisely of CD8+ T cells, in the active margin of the tumor seems a critical parameter for response to PD-1 immunotherapy.49 Furthermore, expression of PD-L1 is variable in distinct metastases, and even expresssion in various areas of the same metastasis render different results. Thus, selection based on PD-L1 expression is very difficult. Although we have data suggesting that the expression of PD-L1 has a less important impact on response to the combination of ipilimumab and nivolumab than on the benefit of anti–PD-1 monotherapy, PD-L1 immunostaining alone currently does not appear to be a strong and reliable enough marker to guide treatment decisions, and more predictable biomarkers or marker combinations are actively being investigated. Genetic instability and neoantigens. Mutations generated in the cancer cell genome that favor the genetic instability of cancer cells can give rise to neoantigens. In accordance with murine experimental data that demonstrated that these neoantigens could be involved in cancer immunosurveillance,50,51 recent data in patients have shown that the mutational landscape influences response to immunotherapy with checkpoint inhibitors and that tumors with the highest rates of mutation and generating numerous neoantigens were more sensitive to immunotherapy with CTLA-4 and PD-1 inhibitors.52,53 However, not all neoantigens can generate T-cell clones able to eliminate the tumor cells, and a process of tumor clone immunoediting induced by specific T cells results in complex mutational landscape dynamics.50 Treatment strategies in patients with BRAF-mutant melanoma. In patients with BRAF-mutant melanoma, whether to initiate treatment with a combination of targeted antiBRAF+ and anti-MEK agents compared with immunotherapy is a frequent question. The combination of these two strategies is promising on the basis of several preliminary studies recently presented at international meetings. Several randomized phase II and phase III trials are presently exploring these questions. Today, in countries where both targeted agents and checkpoint inhibitors are available, until controlled data from randomized trials are available, most

physicians prescribe targeted agents for aggressive and rapidly progressive metastatic disease, when a fast response is critical to obtain, and favor immunotherapy for slowly progressive disease. In the intermediate situation, physician choices vary. One element that would favor immunotherapy over targeted agents is that we now have some evidence that patients in complete response following immunotherapy with checkpoint inhibitors can stop treatment, and, with close to 3 years of follow-up, many patients do not relapse, whereas most patients who stop targeted therapy after responding to the treatment seem to relapse.54 Patients with high lactate dehydrogenase remain a high medical need. Finally, although we have dramatically improved the prognosis of patients with metastatic melanoma, some populations of patients present a particularly challenging situation and have very low benefit from both targeted agents and immunotherapy. This is the case for patients with high lactate dehydrogenase, which is usually, but not always, associated with a high tumor load. For these patients, new treatment strategies are urgently needed.

OPTIONS WHEN APPROVED TARGETED AND IMMUNOTHERAPY AGENTS FAIL

Role of Alternative Approved Agents

Although immunotherapy and targeted therapy are the most efficacious therapies for patients with metastatic melanoma, many patients will develop disease progression while on these therapies. Additional therapeutic options are often needed. Given the many promising agents in clinical trial development, the ideal scenario is that patients progressing on first- and second-line therapies would be enrolled in a clinical trial. However, this is not always feasible for a variety of reasons, including access to clinical trials, patient eligibility, and other reasons. In this scenario, there are approved agents for the treatment of metastatic melanoma in certain clinical settings. These options include interleukin-2 (IL-2), talimogene laherparepvec (T-VEC), chemotherapy, and radiation. High-dose IL-2 was one of the major therapies before the approval of targeted therapy and immune checkpoint blockade. However, this therapy was available only to patients with excellent performance status, good organ function, and access to specialized inpatient units administering this therapy. It was approved on the basis of durable remission in 1% to 5% of the treated population. Considerable toxicities include capillary leak syndrome, fever, infection, and cardiopulmonary and renal failure.55 The role of IL-2 in the era of newer therapies is generally reserved for patients who have progressed on other lines of therapy and are not eligible for clinical trials. Current clinical trials are exploring combination therapies with IL-2 and other agents, but its future role in treatment combinations is still unclear. T-VEC is an attenuated oncolytic herpes simplex virus containing a granulocyte macrophage colony-stimulating factor (GM-CSF) gene. Antitumor benefit occurs through the production of GM-CSF within the tumor, which enhances cellular immunity along with a direct effect from viral infection and asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 667

YUSHAK ET AL

lytic replication. T-VEC was approved by the FDA in 2015 for the treatment of patients with unresectable and injectable cutaneous, subcutaneous, or nodal disease. It is only for patients with limited or no visceral disease and patients must have an easily accessible (nodal or skin) metastases. In a phase III trial of T-VEC versus GM-CSF, the overall response rate was 26.4% versus 5.7%, respectively. The durable response rate was higher in patients treated with T-VEC compared with GM-CSF (16.3% vs. 2.1%). Subset analysis demonstrated an improvement in OS among stage III and IVM1a patients. However, OS among all subsets was not significantly improved. Responses have been seen in both injected and noninjected lesions. Common toxicities included fatigue, chills, pyrexia, and injection-site pain.56 Before the approval of immune checkpoint inhibitors and targeted therapy, cytotoxic chemotherapy was the mainstay of therapy in patients with metastatic disease who were not candidates for or had progressed on IL-2. Despite the common use, no clinical trial has shown a survival benefit for cytotoxic therapy. Commonly used regimens include dacarbazine, temozolomide, and carboplatin and pacl*taxel. Dacarbazine has a response rate of 8% to 20%. Toxicities include nausea and emesis and bone marrow suppression.57,58 Temozolomide is an oral analog of dacarbazine and can cross the blood-brain barrier. Clinical trials comparing it with dacarbazine have not shown any statistically significant improvement in survival or progression.59,60 Carboplatin and pacl*taxel are both widely used agents but have never been compared with dacarbazine or temozolomide in a randomized trial. Response rates in a phase III trial comparing carboplatin and pacl*taxel to carboplatin and pacl*taxel with sorafenib demonstrated no benefit to adding sorafenib. However, the response rates in both arms were 18% to 20%, demonstrating the clinical activity of the combination regimen.61 Radiation therapy (RT) has long been used as a palliative measure for pain control and management of brain metastases. Various fractionation schemes have shown effectiveness in treating painful sites of extracranial metastatic disease.62 The brain is a frequent site of melanoma metastases. Autopsy studies have demonstrated up to 75% of patients will develop evidence of brain metastases during their course of illness.63 Whole-brain RT used to be the standard treatment of these patients. However, now that systemic therapies have shown dramatic improvements in OS, whole-brain RT should not be considered as standard of care for most patients given the concern about long-term toxicities.64 Alternatives to whole-brain RT include stereotactic RT. Additionally, both immune therapy and targeted therapy have shown response in the brain.65,66 The optimal management of patients with melanoma brain metastases is yet to be established. Several ongoing clinical trials are exploring this area.

Alternative Genomic Targets

As discussed above, the treatment of patients with metastatic melanoma who harbor a BRAF mutation with BRAF 668 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

and MEK inhibitors has resulted in improvements in both PFS and OS. Unfortunately, most patients will develop resistance to targeted therapy and develop disease progression. Several mechanisms have been proposed. Three main areas of resistance have been targeted for potential therapeutic intervention to help overcome resistance. These include alteration of the immune system by targeted therapy, activation of the PI3K-mTOR pathway, and reactivation of the MAPK pathway leading to continued ERK activation. MAPK reactivation has been demonstrated in 79% of tumors developing resistance67 through a variety of mechanisms, including BRAF splice variant, BRAF amplification, secondary activating mutations of NRAS or MEK1 and MEK2, and overexpression of RAF1 and MAP3K8-COT kinases.6,68-71 ERK inhibitors are currently in clinical trials.72 In addition, various compounds are in trials in combination with BRAF inhibitors targeting parallel signaling pathways, reviewed by Welsh et al.72 Increasingly, BRAF and MEK inhibitors are being recognized for their role in the tumor microenvironment. Treated patients have shown increases in CD4+, CD8+, and PD-1+ lymphocytes.73 Because of these changes, several ongoing trials are combining targeted therapy with immune checkpoint inhibitors. NRAS mutations occur in approximately 20% of patients with melanoma. With this mutation, activation of the MAPK pathway occurs upstream of BRAF. Although to date, directly inhibiting mutated NRAS has not been possible, an alternative strategy is to block the pathway downstream at the level of MEK. The MEK inhibitor binimetinib has shown an objective response rate of 20% and median PFS of 3.7 months in a phase I/II trial.74 The NEMO trial randomized 402 patients with melanoma harboring an NRAS mutation in a 2:1 ratio to binimetinib, a MEK inhibitor, or dacarbazine. The primary endpoint was PFS by independent review. This trial was presented at the 2016 ASCO Annual Meeting and demonstrated an improvement in PFS of 2.8 versus 1.5 months (p < 0.001). However, OS was not significantly improved.75 These data are consistent with the understanding that NRAS-mutated melanomas are MEK dependent, but there appears to be little long-term clinical benefit with MEK inhibition so far. We await longer follow-up on this trial and the formal publication. Additional observations in the NRAS cohort of patients have shown dysregulation in the CDK4/6-RRB1 pathway. A phase I/II trial with ribociclib (a CDK4/6 inhibitor) and binimetinib demonstrated PFS of 6.7 months and an overall response rate of 41%.76,77 When presenting the data at the 2014 ASCO Annual Meeting, Sosman et al76 reported that among 22 patients treated with binimetinib and LEE011, there were seven partial responders but no complete responders. Toxicity was tolerable and consisted mostly of cutaneous toxicity and creatine phosphokinase elevations. Further investigation of this combination is ongoing. Fewer than 10% of newly diagnosed melanomas are either acral or mucosal. These patients generally have a poorer prognosis compared with those with cutaneous melanomas. Approximately 20% of these patients will harbor

TREATMENT OF INOPERABLE MELANOMA

a mutation in the growth factor receptor c-KIT. A phase II trial of imatinib in patients with metastatic mucosal, acral, or chronically sun-damaged skin with mutation or amplification of KIT demonstrated a best overall response rate of 29%. The overall disease control rate was 50% but varied on the basis of KIT status (77% mutated vs. 18% amplified). Median OS was 12.5 months with median time to progression of 3.7 months.78,79 A phase II study examined nilotinib in patients with a KIT mutation or activation who were either intolerant or refractory to a prior KIT inhibitor (cohort A) or with brain metastases (cohort B). Three of 11 patients in cohort A reached the primary endpoint of 4-month disease control. Median OS was 14.2 months in cohort A. Cohort B demonstrated limited efficacy in KIT-mutated patients with brain metastases.79 The National Cancer Institute’s MATCH (Molecular Analysis for Therapy Choice) trial seeks to pair patients with molecular abnormalities in their tumor tissue with agents targeting their mutation. The trial is not unique to melanoma but includes mutations that can be present in melanoma. Patients enrolling in the match trial agree to undergo a biopsy for DNA sequencing of tumor tissue. If a molecular abnormality is found in the tumor tissue that is targeted by one of the drugs in the trial, patients are further screened for enrollment. If they meet eligibility requirements, they begin therapy on the specific arm targeting the mutation (NCT02465060).

Alternative Immune Targets

Although the anti–PD-1 drugs nivolumab and pembrolizu mab are the most commonly used medications targeting the PD-1/PD-L1 interaction, several anti–PD-L1 drugs are being studied in combination and as single agents. A phase I study of atezolizumab showed an overall response rate of 26%.80 Most current immunotherapy options in melanoma attempt to expand or induce tumor antigen–specific immune responses in vivo. Adoptive cell therapy isolates tumor antigen–specific T cells from patients, either from peripheral blood or resected tumor, expands the cells, then reinfuses them to the patients. Initial studies of adoptive cell therapy demonstrated limited clinical benefit. However, when lymphoablation occurred before transfer of tumor-infiltrating lymphocyte and was followed by IL-2, response rates up to 50% were demonstrated. Of these responses, up to 20% were durable.81-86 Similar to IL-2, adoptive cell therapy is available only to a limited number of patients with excellent performance status, ability to travel to a specialized center, and the ability to wait without treatment of several weeks while the cells expand. As technology improves for this technique and the expansion of cells is available at multiple centers, the use of this therapy is expected to increase. The term abscopal effect is used to describe the impact of RT on areas outside of the radiation treatment field. The decrease in the sizes of tumors outside the radiation field was hypothesized to be related to a systemic inflammatory

or immune response from the radiation. Early in the use of immunotherapy in melanoma, several case reports showed a response to immunotherapy following the addition of RT after the patient had progressed on immunotherapy alone or an increase in T-cell activation.46,87 This has led to speculation that RT may have a role in patients who have failed to respond to immunotherapy or developed progression after an initial response. However, this has not yet been proved in a controlled clinical trial. Additionally, ongoing trials are administering RT in combination with immunotherapy to see if the initial response rates are higher. There are several targets in current development both preclinically and in early-phase trials. This includes both immune checkpoint agonists and antagonists. Immunostimulatory targets include 4-1BB, CD27, and OX40. Of these, 4-1BB potentiates effector responses in lymphocytes necessary for tumor immunity. However, initial development was slowed because of the incidence of hepatitis and cytopenia. Additional clinical trials are now investigating this agent.88 Immunosuppressive targets include IDO, LAG-3, and TIM3. LAG-3 is an immune checkpoint inhibitor that maintains immune tolerance. It binds to effector T cells and acts as a ligand for MHC class II proteins. Drugs focusing on these targets are currently being studied as a single agents and in combination with anti–PD-1 therapy.

CONCLUSION

The treatment of advanced melanoma has drastically changed over the past decade. The previously standard therapies, dacarbazine and high-dose IL-2, are now reserved only for situations in which all immunotherapy and targeted agents have failed and no clinical trials are available. Targeted therapy for patients whose tumors harbor the BRAF mutation achieves high response rates and OS benefit with combination BRAF/MEK inhibition. Single-agent BRAF inhibitors should not be used, as randomized trials have established increased clinical benefit (OS and PFS). Treatments containing PD-1 antibodies are clearly superior to CTLA-4 antibodies in the frontline setting; however, the question of upfront BRAF/MEK inhibition versus immunotherapy has yet to be answered. Combination immunotherapy with ipilimumab/nivolumab shows higher response rates and higher toxicity rates compared with single-agent immunotherapy, and survival data will be available in the near future. Biomarkers such as PD-L1 status are still controversial and cannot yet be used for routine clinical decision making. Novel targets in both immunotherapy and targeted therapy are currently being explored in a variety of clinical trials. Studies with new combinations such as immunotherapy with intralesional or targeted therapy are also being evaluated. The future of melanoma management is rapidly changing, although the backbone of treatment at this time is PD-1 antibodies and BRAF inhibitors (for tumors with BRAF mutations).

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 669

YUSHAK ET AL

References 1. Callahan MK, Rampal R, Harding JJ, et al. Progression of RASmutant leukemia during RAF inhibitor treatment. N Engl J Med. 2012;367:2316-2321. 2. Andrews MC, Behren A, Chionh F, et al. BRAF inhibitor-driven tumor proliferation in a KRAS-mutated colon carcinoma is not overcome by MEK1/2 inhibition. J Clin Oncol. 2013;31:e448-e451. 3. McArthur GA, Chapman PB, Robert C, et al. Safety and efficacy of vemurafenib in BRAF(V600E) and BRAF(V600K) mutation-positive melanoma (BRIM-3): extended follow-up of a phase 3, randomised, open-label study. Lancet Oncol. 2014;15:323-332. 4. Hauschild A, Grob J-J, Demidov LV, et al. Dabrafenib in BRAF-mutated metastatic melanoma: a multicentre, open-label, phase 3 randomised controlled trial. Lancet. 2012;380:358-365. 5. Lito P, Rosen N, Solit DB. Tumor adaptation and resistance to RAF inhibitors. Nat Med. 2013;19:1401-1409.

18. Keir ME, Butte MJ, Freeman GJ, et al. PD-1 and its ligands in tolerance and immunity. Annu Rev Immunol. 2008;26:677-704. 19. Latchman Y, Wood CR, Chernova T, et al. PD-L2 is a second ligand for PD-1 and inhibits T cell activation. Nat Immunol. 2001;2:261-268. 20. Nishimura H, Nose M, Hiai H, et al. Development of lupus-like autoimmune diseases by disruption of the PD-1 gene encoding an ITIM motif-carrying immunoreceptor. Immunity. 1999;11:141-151. 21. Ribas A. Clinical development of the anti-CTLA-4 antibody tremelimumab. Semin Oncol. 2010;37:450-454. 22. Hodi FS, O’Day SJ, McDermott DF, et al. Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med. 2010;363:711-723. 23. Robert C, Thomas L, Bondarenko I, et al. Ipilimumab plus dacarbazine for previously untreated metastatic melanoma. N Engl J Med. 2011;364:2517-2526.

6. Poulikakos PI, Persaud Y, Janakiraman M, et al. RAF inhibitor resistance is mediated by dimerization of aberrantly spliced BRAF(V600E). Nature. 2011;480:387-390.

24. Ascierto PA, Del Vecchio M, Robert C, et al. Overall Survival (OS) and safety results from phase 3 trial of ipilimumab (IPI) at 3 mg/kg vs 10 mg/kg in patients with metastatic melanoma (MEL). Ann Oncol. 2016;27(suppl_6):379-400.

7. Long GV, Stroyakovskiy D, Gogas H, et al. Dabrafenib and trametinib versus dabrafenib and placebo for Val600 BRAF-mutant melanoma: a multicentre, double-blind, phase 3 randomised controlled trial. Lancet. 2015;386:444-451.

25. Robert C, Ribas A, Wolchok JD, et al. Anti-programmed-deathreceptor-1 treatment with pembrolizumab in ipilimumab-refractory advanced melanoma: a randomised dose-comparison cohort of a phase 1 trial. Lancet. 2014;384:1109-1117.

8. Ascierto PA, McArthur GA, Dréno B, et al. Cobimetinib combined with vemurafenib in advanced BRAF(V600)-mutant melanoma (coBRIM): updated efficacy results from a randomised, double-blind, phase 3 trial. Lancet Oncol. 2016;17:1248-1260.

26. Ribas A, Hamid O, Daud A, et al. Association of pembrolizumab with tumor response and survival among patients with advanced melanoma. JAMA. 2016;315:1600-1609.

9. Margolin K, Ernstoff MS, Hamid O, et al. Ipilimumab in patients with melanoma and brain metastases: an open-label, phase 2 trial. Lancet Oncol. 2012;13:459-465.

27. Ribas A, Puzanov I, Dummer R, et al. Pembrolizumab versus investigator-choice chemotherapy for ipilimumab-refractory melanoma (KEYNOTE-002): a randomised, controlled, phase 2 trial. Lancet Oncol. 2015;16:908-918.

10. Goldberg SB, Gettinger SN, Mahajan A, et al. Pembrolizumab for patients with melanoma or non-small-cell lung cancer and untreated brain metastases: early analysis of a non-randomised, open-label, phase 2 trial. Lancet Oncol. 2016;17:976-983.

28. Robert C, Schachter J, Long GV, et al; KEYNOTE-006 Investigators. Pembrolizumab versus ipilimumab in advanced melanoma. N Engl J Med. 2015;372:2521-2532.

11. Gao J, Aksoy BA, Dogrusoz U, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal. 2013;6:pl1. 12. Cerami E, Gao J, Dogrusoz U, et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov. 2012;2:401-404. 13. Yao Z, Torres NM, Tao A, et al. BRAF mutants evade ERK-dependent feedback by different mechanisms that determine their sensitivity to pharmacologic inhibition. Cancer Cell. 2015;28:370-383. 14. Freeman GJ, Gribben JG, Boussiotis VA, et al. Cloning of B7-2: a CTLA4 counter-receptor that costimulates human T cell proliferation. Science. 1993;262:909-911.

29. Weber JS, Kähler KC, Hauschild A. Management of immune-related adverse events and kinetics of response with ipilimumab. J Clin Oncol. 2012;30:2691-2697. 30. Topalian SL, Hodi FS, Brahmer JR, et al. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N Engl J Med. 2012;366:2443-2454. 31. Weber JS, D’Angelo SP, Minor D, et al. Nivolumab versus chemotherapy in patients with advanced melanoma who progressed after antiCTLA-4 treatment (CheckMate 037): a randomised, controlled, openlabel, phase 3 trial. Lancet Oncol. 2015;16:375-384. 32. Wolchok JD, Kluger H, Callahan MK, et al. Nivolumab plus ipilimumab in advanced melanoma. N Engl J Med. 2013;369:122-133.

15. Azuma M, Ito D, Yagita H, et al. B70 antigen is a second ligand for CTLA4 and CD28. Nature. 1993;366:76-79.

33. Larkin J, Chiarion-Sileni V, Gonzalez R, et al. Combined nivolumab and ipilimumab or monotherapy in untreated melanoma. N Engl J Med. 2015;373:23-34.

16. Vijayakrishnan L, Slavik JM, Illés Z, et al. An autoimmune diseaseassociated CTLA-4 splice variant lacking the B7 binding domain signals negatively in T cells. Immunity. 2004;20:563-575.

34. Postow MA, Chesney J, Pavlick AC, et al. Nivolumab and ipilimumab versus ipilimumab in untreated melanoma. N Engl J Med. 2015;372:2006-2017.

17. Ueda H, Howson JMM, Esposito L, et al. Association of the T-cell regulatory gene CTLA4 with susceptibility to autoimmune disease. Nature. 2003;423:506-511.

35. Boutros C, Tarhini A, Routier E, et al. Safety profiles of anti-CTLA-4 and anti-PD-1 antibodies alone and in combination. Nat Rev Clin Oncol. 2016;13:473-486.

670 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

TREATMENT OF INOPERABLE MELANOMA

36. Eggermont AMM, Chiarion-Sileni V, Grob J-J, et al. Prolonged survival in stage III melanoma with ipilimumab adjuvant therapy. N Engl J Med. 2016;375:1845-1855.

54. Robert C, Ribas A, Hamid O, et al. Three-year overall survival for patients with advanced melanoma treatment with pembrolizumab in KEYNOTE-001. J Clin Oncol. 2016;34 (suppl; abstr 9503).

37. Beck KE, Blansfield JA, Tran KQ, et al. Enterocolitis in patients with cancer after antibody blockade of cytotoxic T-lymphocyte-associated antigen 4. J Clin Oncol. 2006;24:2283-2289.

55. Atkins MB, Lotze MT, Dutcher JP, et al. High-dose recombinant interleukin 2 therapy for patients with metastatic melanoma: analysis of 270 patients treated between 1985 and 1993. J Clin Oncol. 1999;17:2105-2116.

38. Carbonnel F, Soularue E, Coutzac C, et al. Inflammatory bowel disease and cancer response due to anti-CTLA-4: is it in the flora? Semin Immunopathol. Epub 2017 Jan 16. 39. Marthey L, Mateus C, Mussini C, et al. Cancer immunotherapy with anti-CTLA-4 monoclonal antibodies induces an inflammatory bowel disease. J Crohn’s Colitis. 2016;10:395-401. 40. Champiat S, Lambotte O, Barreau E, et al. Management of immune checkpoint blockade dysimmune toxicities: a collaborative position paper. Ann Oncol. 2016;27:559-574. 41. Michot JM, Bigenwald C, Champiat S, et al. Immune-related adverse events with immune checkpoint blockade: a comprehensive review. Eur J Cancer Oxf Engl. 2016;54:139-148. 42. Hodi FS, Chesney J, Pavlick AC, et al. Combined nivolumab and ipilimumab versus ipilimumab alone in patients with advanced melanoma: 2-year overall survival outcomes in a multicentre, randomised, controlled, phase 2 trial. Lancet Oncol. 2016;17:15581568.

56. Andtbacka RH, Kaufman HL, Collichio F, et al. Talimogene laherparepvec improves durable response rate in patients with advanced melanoma. J Clin Oncol. 2015;33:2780-2788. 57. Atkins MB. The role of cytotoxic chemotherapeutic agents either alone or in combination with biological response modifiers. In Kirkwood JK (ed). Molecular Diagnosis, Prevention & Therapy of Melanoma. New York, NY: Marcel Dekker; 1997;219. 58. Houghton AN, Legha S, Bajorin DF. Chemotherapy for metastatic melanoma. In Balch CM, Houghton M, Milton GW, Sober AJ, Soong S-J (eds). Cutaneous Melanoma, 2nd ed. Philadelphia, PA: J.B. Lippincott; 1992;498. 59. Middleton MR, Grob JJ, Aaronson N, et al. Randomized phase III study of temozolomide versus dacarbazine in the treatment of patients with advanced metastatic malignant melanoma. J Clin Oncol. 2000;18:158166.

43. Postow MA, Callahan MK, Barker CA, et al. Immunologic correlates of the abscopal effect in a patient with melanoma. N Engl J Med. 2012;366:925-931.

60. Patel PM, Suciu S, Mortier L, et al; EORTC Melanoma Group. Extended schedule, escalated dose temozolomide versus dacarbazine in stage IV melanoma: final results of a randomised phase III study (EORTC 18032). Eur J Cancer. 2011;47:1476-1483.

44. Weber JS, Hodi FS, Wolchok JD, et al. Safety profile of nivolumab monotherapy: a pooled analysis of patients with advanced melanoma. J Clin Oncol. Epub 2016 Nov 14.

61. Flaherty KT, Lee SJ, Zhao F, et al. Phase III trial of carboplatin and pacl*taxel with or without sorafenib in metastatic melanoma. J Clin Oncol. 2013;31:373-379.

45. Hua C, Boussemart L, Mateus C, et al. Association of vitiligo with tumor response in patients with metastatic melanoma treated with pembrolizumab. JAMA Dermatol. 2016;152:45-51.

62. Konefal JB, Emami B, Pilepich MV. Analysis of dose fractionation in the palliation of metastases from malignant melanoma. Cancer. 1988;61:243-246.

46. Festino L, Botti G, Lorigan P, et al. Cancer treatment with anti-PD-1/ PD-L1 agents: is PD-L1 expression a biomarker for patient selection? Drugs. 2016;76:925-945.

63. Flanigan JC, Jilaveanu LB, Faries M, et al. Melanoma brain metastases: is it time to reassess the bias? Curr Probl Cancer. 2011;35:200-210.

47. Daud AI, Wolchok JD, Robert C, et al. Programmed death-ligand 1 expression and response to the anti-programmed death 1 antibody pembrolizumab in melanoma. J Clin Oncol. 2016;34:4102-4109. 48. Long GV, Lark J, Ascierto PA. PD-L1 expression as a biomarker for nivolumab (NIVO) plus ipilimumab (IPI) and NIVO alone in advanced melanoma (MEL): A pooled analysis. Ann Oncol. 2016; 27 (suppl_6):1112. 49. Tumeh PC, Harview CL, Yearley JH, et al. PD-1 blockade induces responses by inhibiting adaptive immune resistance. Nature. 2014;515:568-571. 50. Schumacher TN, Schreiber RD. Neoantigens in cancer immunotherapy. Science. 2015;348:69-74. 51. Ward JP, Gubin MM, Schreiber RD. The role of neoantigens in naturally occurring and therapeutically induced immune responses to cancer. Adv Immunol. 2016;130:25-74. 52. Rizvi NA, Hellmann MD, Snyder A, et al. Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in nonsmall cell lung cancer. Science. 2015;348:124-128. 53. McGranahan N, Furness AJS, Rosenthal R, et al. Clonal neoantigens elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade. Science. 2016;351:1463-1469.

64. Flanigan JC, Jilaveanu LB, Chiang VL, et al. Advances in therapy for melanoma brain metastases. Clin Dermatol. 2013;31:264-281. 65. Gibney GT, Gauthier G, Ayas C, et al. Treatment patterns and outcomes in BRAF V600E-mutant melanoma patients with brain metastases receiving vemurafenib in the real-world setting. Cancer Med. 2015;4:1205-1213. 66. Knisely JP, Yu JB, Flanigan J, et al. Radiosurgery for melanoma brain metastases in the ipilimumab era and the possibility of longer survival. J Neurosurg. 2012;117:227-233. 67. Rizos H, Menzies AM, Pupo GM, et al. BRAF inhibitor resistance mechanisms in metastatic melanoma: spectrum and clinical impact. Clin Cancer Res. 2014;20:1965-1977. 68. Shi H, Moriceau G, Kong X, et al. Melanoma whole-exome sequencing identifies (V600E)B-RAF amplification-mediated acquired B-RAF inhibitor resistance. Nat Commun. 2012;3:724. 69. Carlino MS, Fung C, Shahheydari H, et al. Preexisting MEK1P124 mutations diminish response to BRAF inhibitors in metastatic melanoma patients. Clin Cancer Res. 2015;21:98-105. 70. Wagle N, Van Allen EM, Treacy DJ, et al. MAP kinase pathway alterations in BRAF-mutant melanoma patients with acquired resistance to combined RAF/MEK inhibition. Cancer Discov. 2014;4: 61-68.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 671

YUSHAK ET AL

71. Johannessen CM, Boehm JS, Kim SY, et al. COT drives resistance to RAF inhibition through MAP kinase pathway reactivation. Nature. 2010;468:968-972. 72. Welsh SJ, Rizos H, Scolyer RA, et al. Resistance to combination BRAF and MEK inhibition in metastatic melanoma: where to next? Eur J Cancer. 2016;62:76-85. 73. Kakavand H, Wilmott JS, Menzies AM, et al. PD-L1 expression and tumor-infiltrating lymphocytes define different subsets of MAPK inhibitor-treated melanoma patients. Clin Cancer Res. 2015;21:31403148. 74. Ascierto PA, Schadendorf D, Berking C, et al. MEK162 for patients with advanced melanoma harbouring NRAS or Val600 BRAF mutations: a non-randomised, open-label phase 2 study. Lancet Oncol. 2013;14:249-256. 75. Dummer RSD, Ascierto PA, et al. Results of NEMO: a phase III trial of binimetinib (BINI) vs dacarbazine (DTIC) in NRAS-mutant cutaneous melanoma. J Clin Oncol. 2016;34 (suppl; abstr 9500). 76. Sosman J, Kittaneh M, Lolkema MPJK, et al. A phase 1b/2 study of LEE011 in combination with binimetinib (MEK162) in patients with NRAS-mutant melanoma: early encouraging clinical activity. J Clin Oncol. 2014;32:9009-9009. 77. Hodis E, Watson IR, Kryukov GV, et al. A landscape of driver mutations in melanoma. Cell. 2012;150:251-263. 78. Hodi FS, Corless CL, Giobbie-Hurder A, et al. Imatinib for melanomas harboring mutationally activated or amplified KIT arising on mucosal, acral, and chronically sun-damaged skin. J Clin Oncol. 2013;31:31823190. 79. Carvajal RD, Lawrence DP, Weber JS, et al. Phase II study of nilotinib in melanoma harboring KIT alterations following progression to prior KIT inhibition. Clin Cancer Res. 2015;21:2289-2296.

672 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

80. Hamid O, Sosman JA, Lawrence DP, et al. Clinical activity, safety, and biomarkers of MPDL3280A, an engineered PD-L1 antibody in patients with locally advanced or metastatic melanoma (mM). J Clin Oncol. 2013;31 (suppl; abstr 9010). 81. Rosenberg SA, Packard BS, Aebersold PM, et al. Use of tumorinfiltrating lymphocytes and interleukin-2 in the immunotherapy of patients with metastatic melanoma. A preliminary report. N Engl J Med. 1988;319:1676-1680. 82. Rosenberg SA, Yannelli JR, Yang JC, et al. Treatment of patients with metastatic melanoma with autologous tumor-infiltrating lymphocytes and interleukin 2. J Natl Cancer Inst. 1994;86:1159-1166. 83. Klebanoff CA, Gattinoni L, Palmer DC, et al. Determinants of successful CD8+ T-cell adoptive immunotherapy for large established tumors in mice. Clin Cancer Res. 2011;17:5343-5352. 84. Dudley ME, Yang JC, Sherry R, et al. Adoptive cell therapy for patients with metastatic melanoma: evaluation of intensive myeloablative chemoradiation preparative regimens. J Clin Oncol. 2008;26:52335239. 85. Besser MJ, Shapira-Frommer R, Treves AJ, et al. Clinical responses in a phase II study using adoptive transfer of short-term cultured tumor infiltration lymphocytes in metastatic melanoma patients. Clin Cancer Res. 2010;16:2646-2655. 86. Rosenberg SA, Yang JC, Sherry RM, et al. Durable complete responses in heavily pretreated patients with metastatic melanoma using T-cell transfer immunotherapy. Clin Cancer Res. 2011;17:4550-4557. 87. Stamell EF, Wolchok JD, Gnjatic S, et al. The abscopal effect associated with a systemic anti-melanoma immune response. Int J Radiat Oncol Biol Phys. 2013;85:293-295. 88. Bartkowiak T, Curran MA. 4-1BB agonists: multi-potent potentiators of tumor immunity. Front Oncol. 2015;5:117.

PATIENT AND SURVIVOR CARE

JACOBSEN, NIPP, AND GANZ

Addressing the Survivorship Care Needs of Patients Receiving Extended Cancer Treatment Paul B. Jacobsen, PhD, Ryan D. Nipp, MD, and Patricia A. Ganz, MD OVERVIEW Cancer survivorship care and research has typically focused on the health care needs of people with cancer following the acute phase of treatment. Work in this area, however, has faced challenges in identifying when treatment is complete for many forms of cancer. Acknowledging this challenge, the scope of survivorship research is often expanded to include patients also receiving maintenance or prophylactic therapy. Inherent in this expanded definition is the recognition that for many individuals, cancer is a chronic disease requiring extended treatment over many years. Three distinct patient populations can be identified for which extended treatment poses important survivorship care needs that, to date, have not been adequately addressed. The first group includes patients receiving extended endocrine therapy, such as women with breast cancer receiving tamoxifen and/or aromatase inhibitors as well as men with prostate cancer receiving androgen deprivation therapy. The second group includes patients receiving extended targeted therapy to control disease, as exemplified by patients with chronic myelogenous leukemia receiving treatment with tyrosine kinase inhibitors. A key issue in both of these patient groups is the need to identify and address factors that contribute to difficulties in maintaining high levels of adherence to the prescribed therapy over extended periods of time. The third group includes patients receiving novel therapies for advanced or metastatic cancer that can extend life for prolonged periods. A key issue for this group is the need to understand and address their unique supportive care needs.

he number of cancer survivors (i.e., people living with diagnoses of cancer) continues to grow and is expected to reach 18 million individuals in the United States by 2022.1 The growth in the number of survivors has, in part, stimulated the development of a field known as cancer survivorship. Although definitions vary, cancer survivorship care and research is widely viewed as focusing on the health and life of a person with cancer beyond the acute diagnosis and treatment phase. According to the National Cancer Institute’s Office of Cancer Survivorship, research in this area seeks to “prevent and control adverse cancer diagnosis and treatment-related outcomes…, to provide a knowledge base regarding optimal follow-up care and surveillance of cancers, and to optimize health after cancer treatment.”2 Although the focus of cancer survivorship has been on the period following acute diagnosis and treatment, work in this area has acknowledged the challenges inherent in identifying the end of the acute or primary phase for many forms of cancer treatment.3 Indeed, attempts to summarize cancer survivorship research often refer to studies of individuals who have completed curative treatment or have transitioned to maintenance or prophylactic therapy.4

Inherent in this expanded definition is recognition that for many individuals, cancer will be a chronic disease requiring extended treatment over many years.5 With growing acceptance of the need to expand the scope of cancer survivorship care and research to include patients on extended treatment, we offer an in-depth examination of three distinct patient populations. In each instance, we identify major survivorship issues related to receiving extended treatment and the current status of efforts to understand and address these issues. First, we discuss the patient population receiving extended treatment to prevent disease recurrence or progression. Issues encountered by early-stage patients receiving hormonal therapy for breast and prostate cancer are used to illustrate the survivorship care needs of this patient population. Second, we focus on the population receiving extended treatment to control disease. Issues encountered by patients receiving targeted therapy for chronic myelogenous leukemia (CML) are used to illustrate the survivorship care needs of this patient population. Third, we discuss the population with advanced or metastatic disease receiving extended treatment to slow disease

From the Healthcare Delivery Research Program, Division of Cancer Control and Population Sciences, National Cancer Institute, Bethesda, MD; Department of Medicine, Division of Hematology and Oncology, Massachusetts General Hospital Cancer Center and Harvard Medical School, Boston, MA; Jonsson Comprehensive Cancer Center, Fielding School of Public Health and the David Geffen School of Medicine, University of California, Los Angeles, Los Angeles, CA. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Paul B. Jacobsen, PhD, Division of Cancer Control and Population Sciences, National Cancer Institute, 9609 Medical Center Dr., Bethesda, MD 20892; email: [emailprotected]. © 2017 American Society of Clinical Oncology

674 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

SURVIVORSHIP CARE NEEDS

progression, control symptoms, and maintain quality of life (QOL). Issues encountered by patients receiving novel therapies for advanced or metastatic cancers are used to illustrate the survivorship care needs of this patient population. A concluding section identifies cross-cutting themes and directions for future research.

EXTENDED ADJUVANT ENDOCRINE THERAPY

Adjuvant therapy for cancer is prescribed for patients in whom all known disease has been treated with either primary surgery or radiation therapy. With control of the local disease completed, the focus of adjuvant therapy is to eliminate occult metastatic cancer that may have disseminated in the months and years before the primary tumor was discovered. In this setting, the treatments are directed against preventing a recurrence (local or distant) of disease. Cytotoxic chemotherapy or radiation therapy are usually time limited in duration (6–12 months); however, endocrine therapies can extend over many years. Indications for the use of adjuvant therapy are related to the risk for recurrence, and this is usually dictated by the size of the tumor, its local extent, and other histologic and biologic features (e.g., grade, hormone receptors, gene expression profile). Breast and prostate cancers represent two very common adult cancers in which extended hormonal (endocrine) therapies are applied, and for which excellent data from large randomized trials support the value of targeting the hormonal milieu of the patient’s body as a means of preventing recurrent disease from manifesting itself. Endocrine therapies may be given as the sole adjuvant therapy in both breast and prostate cancer or may follow chemotherapy and radiation. For both patients with breast cancer and those with prostate cancer, there may be untoward side effects from antagonizing the normal hormonal environment, and these may limit the ability to maintain adherence to these treatments over a long period of time.

Breast Cancer Adjuvant Endocrine Therapy

All women with nonmetastatic primary breast cancer, whose tumor expresses estrogen and/or progesterone receptor, are considered appropriate candidates for adjuvant endocrine therapy. This also includes women with stage 0, noninvasive, intraductal cancers for whom the primary indication is breast cancer prevention in the ipsilateral or contralateral breast.6 This approach to management evolved

KEY POINTS • There are challenges to identifying when treatment is complete for many forms of cancer. • Cancer is a chronic disease requiring extended treatment over many years. • Extended treatment poses important survivorship care needs for patients receiving extended endocrine therapy, extended targeted therapy to control disease, and novel therapies for advanced or metastatic cancer.

over many years of randomized trials, in which initially only women with relatively high absolute risks for recurrence were exposed to adjuvant endocrine therapy. However, over the past 30 years, and as supported in a National Institutes of Health consensus conference in 2000, all women with hormone receptor positive tumors larger than 1 cm were deemed to benefit from 5 years of adjuvant tamoxifen.7 Subsequently, additional studies supported the use of aromatase inhibitors as an alternative for 5 years in postmenopausal women, and further studies identified settings in which adjuvant endocrine therapy should be continued for up to 10 years.8 Although the specific choice of endocrine therapy may differ according to the woman’s menopausal status, all of the current approaches to endocrine therapy require daily oral therapy for a minimum of 5 years. From the earliest clinical use of adjuvant tamoxifen, adherence to ongoing therapy was recognized as an issue.9 In this report from Partridge et al,9 up to 50% of women were nonadherent to tamoxifen therapy by 4 years, and this was most common among younger, older, and nonwhite women. Even in clinical trials of adjuvant tamoxifen therapy, diminished adherence rates at 5 years were noted in highly motivated patients; for both tamoxifen and placebo, only 23% of women were taking their study medications at 5 years in the NSABP B-14 trial.10 In the recently reported NSABP B-35 trial comparing anastrozole with tamoxifen in patients with ductal carcinoma in situ,6 only 64% of participants completed the study medications at 5 years, with no difference between the two endocrine therapies. Finally, in a large observational study within an integrated health system, adherence to clinically prescribed endocrine therapy11 was reduced, with the finding of early discontinuation (in the first year of therapy) among younger and older women and poor persistence into the fifth year of treatment. An additional report from this same cohort showed increased mortality among women who were nonadherent.12 What are the barriers to initiation and adherence among women who are prescribed adjuvant endocrine therapy? In Sidebar 1, we list the most common issues identified in the literature13-17 and in clinical experience. Interventions to enhance adherence should focus on these common issues, with an important focus on effective communication about the clinical value and magnitude of treatment benefit, with the assurance that potential side effects will be addressed. These are issues that should be addressed as part of initial treatment discussions and decisions. Furthermore, ongoing follow-up should be conducted by a member of the clinical team to assess any challenges to continued adherence, especially addressing ongoing complaints and any financial concerns that may make continued persistence with treatment an issue.

Prostate Cancer Endocrine Therapy

Unlike the setting of breast cancer, endocrine therapy of prostate cancer is often limited to neoadjuvant use before radiation therapy or short-term adjuvant therapy after radiation. This is usually provided only to patients with high-risk asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 675

JACOBSEN, NIPP, AND GANZ

SIDEBAR 1. Common Barriers to Adherence to Endocrine Therapy in Patients With Breast Cancer • Lack of knowledge about the role and benefits of endocrine therapy • Uncontrolled treatment-associated symptoms (vasomotor symptoms, arthralgia, vagin*l symptoms) • Concerns about rare but serious toxicities (e.g., blood clots, stroke, endometrial cancer, fracture) • Cost of the medications • Distrust of health system and poor communication with medical staff • Lack of perceived risk for recurrence local disease. However, there is another larger group of patients, who have undergone prostatectomy or radiation therapy for local disease, in whom endocrine therapy is initiated for a rising prostate-specific antigen level without evidence of definitive metastatic disease. These patients are likely to be on long-term endocrine therapy, without evidence of clinically symptomatic disease, for which longterm adherence is an issue. Of course, men with metastatic prostate cancer are also on long-term endocrine therapy, and this situation is more comparable to patients with CML described in the next section. Similar to breast cancer, endocrine therapy for prostate cancer manipulates a man’s hormonal environment and focuses on androgen deprivation as a first maneuver. This can be accomplished either with orchiectomy or regular injection of a gonadotropin-releasing hormone analog. Sometimes this is combined with an oral anti-androgen agent (e.g., bicalutamide), but more often antiandrogen therapies are added at the time of prostate-specific antigen level progression or for locally recurrent disease. The patient is said to have “castrate-resistant prostate cancer,” and a variety of other androgen-targeted oral therapies are added. All of these therapies are associated with a variety of symptoms associated with low testosterone, including hot flashes, breast enlargement or tenderness, weight gain, decreased libido, and impotence. In addition, there may be body image and mood changes. Although the symptoms of androgen therapies have been described in clinical trials and observational studies, relatively little is known about how these symptoms affect adherence to endocrine therapy. In a recent study, Jung et al18 found that many men did not have adequate information about their treatments and that their reports of symptoms to their physicians were not addressed. Many of the findings in this survey study were similar to what has been reported about women with breast cancer. However, we could find no other reports in the literature on this topic, and thus this is an important gap. In addition, the newer antiandrogen therapies are very expensive, with monthly costs for enzalutamide and abiraterone estimated to be in excess of $8,000.19 This is certainly another major barrier to long-term adherence. 676 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

EXTENDED TARGETED TREATMENT TO CONTROL DISEASE

Targeted cancer therapies represent a new generation of drugs designed to treat cancer by interfering with molecular targets that play a critical role in growth, progression, and spread of the disease. One of the first and most successful examples of how targeted therapies can improve outcomes occurred in the treatment of CML. With approximately 8,000 new diagnoses annually in the United States,20 CML accounts for 20% of new adult leukemias. On the basis of research showing the cause to be formation of the BCR-ABL oncogene that produces a constitutively active tyrosine kinase,21 the oral medication imatinib was evaluated because it is a potent inhibitor of this enzyme.22 Clinical trials confirmed its efficacy23,24 and demonstrated clinically important differences in QOL favoring imatinib25 over the existing regimen of interferon and cytarabine. Eight-year survival rates for patients with CML have since improved from less than 20% historically to 87% in the imatinib era.26 The success of imatinib is widely considered a model for the development of other targeted cancer therapies.27 Several second-generation tyrosine kinase inhibitors (TKIs) have since been approved for use against CML,27 and other oral medications targeting tyrosine kinase pathways have been or are being developed for many other forms of cancer.28 Treatment of CML typically requires daily oral administration of a TKI over an extended period of time. Discontinuation of medication is generally not recommended unless patients achieve a deep molecular response.29 Among those patients who do achieve a deep molecular response, studies suggest that only 40% will stay in remission after stopping first-line treatment.29

Adherence Issues

The necessity of taking an oral medication for an extended period, combined with the potential for missed doses that can result in impaired cytogenetic and molecular responses,30,31 points to the importance of understanding and promoting medication adherence in patients with CML prescribed TKIs. A recent meta-analysis of 40 studies concluded that, depending on the assessment method, 25% to 33% of patients with CML are not adhering to their prescribed regimens.32 Research on predictors of adherence in this patient population is more limited. A systematic review of this literature identified drug-related adverse events and forgetfulness as common reasons for intentional and unintentional nonadherence, respectively.33 These findings suggest that efforts to maintain adherence should include reminders to take medication as well as effective management of medication side effects.

Side Effects and Symptoms

Although imatinib and similar TKIs are better tolerated than many of the regimens they replaced,25 evidence suggests they are not without side effects. Common side effects of imatinib, dasatinib, and nilotinib observed in clinical trials include pain, diarrhea, nausea, and fatigue.34 On the basis of the National Cancer Institute’s Common Toxicity Criteria,

SURVIVORSHIP CARE NEEDS

adverse event rates for any grade for these symptoms among patients taking imatinib were found to be 43.7% (nausea), 36.5% (pain), 34.5% (fatigue), and 32.8% (diarrhea).24 Similar adverse rates for these symptoms have been observed for nilotinib and dasatinib.27 These toxicities were generally low grade, but even low-grade toxicities that persist for months or years in patients on chronic therapy have the potential to greatly impair function and overall QOL and contribute to nonadherence. It should be noted that Common Toxicity Criteria adverse event reports are based on clinician ratings and may represent underestimates of symptoms for which assessment through patient self-report represents the “gold standard” (e.g., fatigue).35 More recent studies using patient self-report measures suggest that fatigue is among the most common and problematic symptoms experienced by patients with CML.36-40 One study in particular speaks to the clinical importance of fatigue in patients with CML. Efficace et al37 evaluated the relationship of symptoms, clinical features, and demographics to QOL. Fatigue (as measured by the Functional Assessment of Chronic Illness Therapy Fatigue Scale41) was independently associated with worse QOL on all scales of the SF-3642 and had the highest inclusion frequency of all variables examined. The presence of fatigue was also associated with greater symptom burden, a factor related to poorer TKI adherence in patients with CML.31 These findings led the investigators to conclude that fatigue is the main factor limiting QOL in patients with CML who receive long-term TKI therapy.37

Interventions to Promote Adherence and Manage Symptoms in Patients With CML Prescribed TKIs

Despite its importance, we are aware of only one published randomized controlled trial evaluating an intervention designed to promote oral medication adherence in patients with CML.43 In this study, 86 patients with CML who had been on TKIs for at least 6 months were randomized to an intervention group or a usual-care control group. The intervention combined a nurse-conducted medication counseling session with supporting educational materials and access to daily text message reminders to take medication. At 9-month follow-up, self-reported medication adherence had increased significantly more often in the intervention group than in the control group (60% vs. 33%, respectively). Seventy percent or more of patients rated the counseling and educational materials as useful. In contrast, only onethird of patients chose to receive text message reminders, and only 27% of them perceived them as useful. Despite the high prevalence of treatment-related symptoms, we could identify no published randomized controlled trials evaluating symptom management interventions for patients with CML on TKI therapy. Given research suggesting its clinical importance, the development of interventions to address fatigue should be viewed as a high priority. In the absence of an understanding of the pathophysiology of TKI-related fatigue,37 clinical practice guidelines for addressing fatigue in cancer survivors44 may suggest promising strategies.

Although evidence was viewed as insufficient to recommend pharmacologic therapies, evidence was sufficient to recommend physical activity interventions and cognitive behavioral therapy.44 An example of the latter is an intervention demonstrated to be effective against severe fatigue in disease-free survivors following completion of cancer treatment.45 This intervention addresses six possible contributory factors (insufficient coping with cancer, fear of disease recurrence, dysfunctional fatigue-related cognitions, sleep dysregulation, activity dysregulation, and low social support or negative social interactions) and is delivered in a series of face-to-face sessions by a trained therapist. A recent publication describes the successful adaptation of this intervention for use in patients with TKI-related fatigue and for internet delivery to improve patient access.46 A small-scale randomized controlled trial of this adapted intervention is currently under way.47

Summary of Survivorship Care Needs of Patients With CML on Extended Therapy

Patients with CML are in the vanguard in that they are one of the first cancer populations for whom disease control is typically achieved exclusively with the use a targeted therapy agent prescribed over an extended period of time. Although much less toxic than earlier regimens, TKIs for CML have been found to commonly produce side effects that adversely affect QOL and contribute to intentional nonadherence to daily oral dosing. Accordingly, survivorship care needs of this patient population include effective management of common treatment-related symptoms (e.g., fatigue) and assistance in maintaining high levels of medication adherence. Development of interventions to effectively meet the needs of this population is still at a very early stage.

EXTENDED TREATMENT OF PATIENTS WITH ADVANCED OR METASTATIC DISEASE

Patients with advanced or metastatic cancer often receive treatments with the goal of slowing disease progression, controlling symptoms, and maintaining QOL. Historically, treatment of many patients with advanced cancers rarely extended life beyond 1 year.48-54 However, the expected survival of numerous patients with advanced cancer has improved significantly in recent years, following the advent of novel genotype-directed therapies55-60 and immune checkpoint–targeting agents.61-66 Thus, as the paradigm has shifted toward more effective treatment of patients with advanced cancer, so too have the supportive care needs of this unique group of cancer survivors.

Supportive Care Needs of Survivors With Advanced Cancer

Patients with advanced cancer often experience multiple symptoms, both physical and psychological, as well as issues related to prognostic uncertainty, financial distress, and the need for caregiving support from family and friends. Notably, patients with advanced cancer who receive extended treatments and survive for prolonged periods likely experience asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 677

JACOBSEN, NIPP, AND GANZ

SIDEBAR 2. Unique Supportive Care Needs of Survivors With Advanced Cancer • Physical and psychological symptoms • Maintaining quality of life • Prognostic uncertainty • Making informed treatment decisions • Financial burden • Family caregiving demands issues similar to all patients with advanced cancer, but limited research has focused on the unique survivorship needs of this population (Sidebar 2).

Symptoms Experienced by Patients With Advanced Cancer

Patients with advanced cancer frequently experience numerous physical and psychological symptoms that are often under-recognized by their clinicians.67-70 Symptoms such as pain, dyspnea, fatigue, nausea, and lack of appetite can lead to poor QOL and psychological distress for patients and their family members.71-73 However, research demonstrates that clinicians often fail to reliably detect their patients’ symptoms and frequently underestimate their severity.74-77 In addition, studies suggest that patients may underreport their symptoms to their clinicians, often resulting in worse symptom management.70,78-80 Thus, there is a critical need to recognize and address symptoms in patients with advanced cancer. Moreover, research is needed to better understand the symptom support needs of patients with advanced cancer who receive extended treatments and survive for prolonged periods. Patients’ symptoms represent a modifiable target for interventions aimed at improving patient outcomes.81-83 A randomized trial of an intervention in which patients in the outpatient setting completed electronic self-reports of their symptoms and had their symptom reports delivered to their clinicians demonstrated better symptom control for those receiving the intervention compared with usual care.82 More recently, a web-based patient-reported symptom monitoring intervention with automated reporting to clinicians for severe or worsening symptoms was compared with usual care in 766 patients with cancer in the outpatient setting.81 Patients who received the intervention reported better QOL and experienced fewer hospitalizations. These studies highlight the importance of symptom monitoring interventions, and future work should further investigate the efficacy of these interventions among long-term survivors living with advanced cancer.

Prognostic Uncertainty

Patients with advanced cancer often misunderstand their prognosis and the goals of their treatment.84,85 Recent advancements in cancer therapies have further complicated oncologists’ ability to effectively communicate an accurate assessment of their patient’s prognosis.86 However, little research exists regarding the increasing challenge of how 678 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

clinicians should communicate with patients about their prognosis in the modern era of targeted anticancer therapies. This is important, because patients with accurate perceptions of their prognosis are more empowered to make informed treatment decisions and plan for their future.87-90 Moreover, research has shown that patients with advanced cancer prefer their oncologists to provide honest and accurate prognostic disclosure early in the disease course.91 Consistent with these preferences, expert groups have recommended that clinicians initiate communication about prognosis at the time of diagnosis and that these discussions continue longitudinally, throughout the cancer trajectory.92,93 By initiating these discussions early, and incorporating new information as it becomes available, clinicians can help patients better understand their prognosis and make more effective decisions about their care.

The Financial Burden Experienced by Cancer Survivors

Increasingly, studies have shown that patients with cancer experience substantial financial burden related to the disease and its treatment, yet financial burden among cancer survivors remains understudied.94-98 Prior work demonstrates that patients with histories of cancer experience financial issues such as job loss, missed work, and trouble obtaining affordable health insurance.99-102 Notably, cancer survivors often need long-term health care for years after their initial diagnosis, and the high out-of-pocket medical costs coupled with the loss of income can further compound their economic hardship.103-105 This financial burden can negatively affect their health outcomes, including poorer QOL, increased symptom burden, and potentially higher mortality.98,106,107 Importantly, in the modern era of cancer therapeutics, with patients living longer and drug prices increasing exponentially increasing, patients with advanced cancers are particularly vulnerable to the adverse financial consequences of their cancer.103,108 Survivors’ financial burden may influence their decision to forgo needed care or to not properly adhere to prescribed therapies in an effort to defray costs109-111 and thereby jeopardize their health.12,112,113 Thus, the financial burden experienced by cancer survivors is an important issue with the potential to impact the quality of their survivorship care.

Family Caregivers of Patients With Cancer

Patients with advanced cancer often require assistance from friends and family as they navigate their cancer course.114,115 Unfortunately, family caregivers are often neglected when considering the unique supportive care needs of patients with advanced cancer. Family caregivers often experience a substantial symptom burden, including fatigue, sleep disturbance, depression, and anxiety.116-118 Caregiving demands can negatively affect family caregivers’ QOL and affect their ability to effectively care for their loved ones.119,120 As novel cancer therapies continue to change the survival trajectory for patients with advanced cancer, efforts are needed to understand how best to support family caregivers throughout the patient’s course.

SURVIVORSHIP CARE NEEDS

Summary of Survivorship Care Needs of Patients With Advanced Cancer Receiving Extended Therapies

Patients with advanced cancer often experience issues related to symptoms, prognostic uncertainty, financial distress, and caregiving. Importantly, numerous studies have demonstrated the efficacy of palliative care interventions to address the unique supportive care needs of these patients.121-126 On the basis of ample evidence, ASCO guidelines recommend dedicated palliative care services for patients with advanced cancer early in the disease course, concurrent with active treatment.127 However, minimal data exist to determine the role of palliative care interventions for patients with advanced cancer who receive extended treatments and survive for prolonged periods. Future studies are needed that focus on the unique survivorship needs of this population to develop effective ways to support these patients throughout their cancer trajectory.

CONCLUSION AND FUTURE DIRECTIONS

For many people with cancer, their care will involve extended treatment over considerable periods of time. This reality challenges the paradigm that has defined survivorship care and research as focusing on the period after patients complete a relatively brief period of active treatment. Organizations such as ASCO have already acknowledged that survivorship care and research should also include patients receiving maintenance or prophylactic therapy for cancer. As described above, patient populations receiving extended treatment include individuals receiving hormonal therapy for breast and prostate cancer, as well as individuals

receiving targeted therapy such as TKIs for CML. A major survivorship issue for these patients is the need to identify and address factors that contribute to difficulties in maintaining high levels of adherence to prescribed therapies over extended periods of time. This situation is especially challenging, because patients control when they take their medicine. A key driver of nonadherence with these agents is adverse side effects that impair QOL (e.g., arthralgia and fatigue). This situation underscores the importance of achieving adequate symptom control among patients receiving extended treatment. Unfortunately, there has been relatively little research addressing issues of adherence and symptom management with oral anticancer agents. Recognizing this gap, the National Cancer Institute recently released a funding opportunity announcement designed to encourage research on oral anticancer medication utilization, delivery, and adherence.128 The other population described above includes patients with advanced or metastatic cancer who are receiving novel therapies that can extend life for prolonged periods of time. This population has a number of survivorship needs, including effective symptom management, help in dealing with prognostic uncertainty and financial distress, and family caregiver support. None of these issues has been systematically studied, despite the growing numbers of patients with advanced or metastatic disease experiencing longer survival with novel therapies. An immediate and important research goal is to evaluate the potential benefits of palliative care services for these individuals, given ample evidence regarding the efficacy of these services for patients who are newly diagnosed with advanced disease.

References 1. de Moor JS, Mariotto AB, Parry C, et al. Cancer survivors in the United States: prevalence across the survivorship trajectory and implications for care. Cancer Epidemiol Biomarkers Prev. 2013;22:561-570. 2. Office of Cancer Survivorship. Definitions. https://cancercontrol. cancer.gov/ocs/statistics/definitions.html. Accessed December 21, 2016. 3. Feuerstein M. Defining cancer survivorship. J Cancer Surviv. 2007;1: 5-7. 4. Jacobsen PB, Rowland JH, Paskett ED, et al. Identification of key gaps in cancer survivorship research: findings from the American Society of Clinical Oncology survey. J Oncol Pract. 2016;12:190-193. 5. Surbone A, Tralongo P. Categorization of cancer survivors: why we need it. J Clin Oncol. 2016;34:3372-3374. 6. Margolese RG, Cecchini RS, Julian TB, et al. Anastrozole versus tamoxifen in postmenopausal women with ductal carcinoma in situ undergoing lumpectomy plus radiotherapy (NSABP B-35): a randomised, double-blind, phase 3 clinical trial. Lancet. 2016;387:849856. 7. Eifel P, Axelson JA, Costa J, et al. National Institutes of Health Consensus Development Conference statement: adjuvant therapy for breast cancer, November 1-3, 2000. J Natl Cancer Inst. 2001;93:979989.

8. Burstein HJ, Temin S, Anderson H, et al. Adjuvant endocrine therapy for women with hormone receptor-positive breast cancer: American Society of Clinical Oncology clinical practice guideline focused update. J Clin Oncol. 2014;32:2255-2269. 9. Partridge AH, Wang PS, Winer EP, et al. Nonadherence to adjuvant tamoxifen therapy in women with primary breast cancer. J Clin Oncol. 2003;21:602-606. 10. Fisher B, Dignam J, Bryant J, et al. Five versus more than five years of tamoxifen therapy for breast cancer patients with negative lymph nodes and estrogen receptor-positive tumors. J Natl Cancer Inst. 1996;88:1529-1542. 11. Hershman DL, Kushi LH, Shao T, et al. Early discontinuation and nonadherence to adjuvant hormonal therapy in a cohort of 8,769 early-stage breast cancer patients. J Clin Oncol. 2010;28:41204128. 12. Hershman DL, Shao T, Kushi LH, et al. Early discontinuation and non-adherence to adjuvant hormonal therapy are associated with increased mortality in women with breast cancer. Breast Cancer Res Treat. 2011;126:529-537. 13. Bright EE, Petrie KJ, Partridge AH, et al. Barriers to and facilitative processes of endocrine therapy adherence among women with breast cancer. Breast Cancer Res Treat. 2016;158:243-251.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 679

JACOBSEN, NIPP, AND GANZ

14. Sedjo RL, Devine S. Predictors of non-adherence to aromatase inhibitors among commercially insured women with breast cancer. Breast Cancer Res Treat. 2011;125:191-200. 15. Demissie S, Silliman RA, Lash TL. Adjuvant tamoxifen: predictors of use, side effects, and discontinuation in older women. J Clin Oncol. 2001;19:322-328. 16. Owusu C, Buist DSM, Field TS, et al. Predictors of tamoxifen discontinuation among older women with estrogen receptor-positive breast cancer. J Clin Oncol. 2008;26:549-555. 17. Grunmark B, Garmo H, Zethelius B, et al. Anti-androgen prescribing patterns, patient treatment adherence and influencing factors: results from the nationwide PCBaSe Sweden. Eur J Clin Pharmacol. 2012;68:1619-1630. 18. Jung B, Stoll C, Feick G, et al. Prostate cancer patients’ report on communication about endocrine therapy and its association with adherence. J Cancer Res Clin Oncol. 2016;142:465-470. 19. Pilon D, Queener M, Lefebvre P, et al. Cost per median overall survival month associated with abiraterone acetate and enzalutamide for treatment of patients with metastatic castration-resistant prostate cancer. J Med Econ. 2016;19:777-784. American Cancer Society. Cancer Facts & Figures 2016. Atlanta, GA: 20. American Cancer Society; 2015. 21. Ren R. Mechanisms of BCR-ABL in the pathogenesis of chronic myelogenous leukaemia. Nat Rev Cancer. 2005;5:172-183. 22. Druker BJ, Tamura S, Buchdunger E, et al. Effects of a selective inhibitor of the Abl tyrosine kinase on the growth of Bcr-Abl positive cells. Nat Med. 1996;2:561-566. 23. Kantarjian H, Sawyers C, Hochhaus A, et al; International STI571 CML Study Group. Hematologic and cytogenetic responses to imatinib mesylate in chronic myelogenous leukemia. N Engl J Med. 2002;346:645-652. 24. O’Brien SG, Guilhot F, Larson RA, et al; IRIS Investigators. Imatinib compared with interferon and low-dose cytarabine for newly diagnosed chronic-phase chronic myeloid leukemia. N Engl J Med. 2003;348:994-1004. Hahn EA, Glendenning GA, Sorensen MV, et al; IRIS Investigators. 25. Quality of life in patients with newly diagnosed chronic phase chronic myeloid leukemia on imatinib versus interferon alfa plus low-dose cytarabine: results from the IRIS Study. J Clin Oncol. 2003;21:21382146. 26. Kantarjian H, O’Brien S, Jabbour E, et al. Improved survival in chronic myeloid leukemia since the introduction of imatinib therapy: a singleinstitution historical experience. Blood. 2012;119:1981-1987. 27. Giles FJ, O’Dwyer M, Swords R. Class effects of tyrosine kinase inhibitors in the treatment of chronic myeloid leukemia. Leukemia. 2009;23:1698-1707. Zhang J, Yang PL, Gray NS. Targeting cancer with small molecule kinase 28. inhibitors. Nat Rev Cancer. 2009;9:28-39. 29. Saußele S, Richter J, Hochhaus A, et al. The concept of treatment-free remission in chronic myeloid leukemia. Leukemia. 2016;30:16381647. 30. Marin D, Bazeos A, Mahon FX, et al. Adherence is the critical factor for achieving molecular responses in patients with chronic myeloid leukemia who achieve complete cytogenetic responses on imatinib. J Clin Oncol. 2010;28:2381-2388.

680 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Noens L, van Lierde MA, De Bock R, et al. Prevalence, determinants, 31. and outcomes of nonadherence to imatinib therapy in patients with chronic myeloid leukemia: the ADAGIO study. Blood. 2009;113:54015411. 32. Alrabiah Z, Alhossan A, Yun S, et al. Adherence to tyrosine kinase inhibitor therapy in patients with chronic myeloid leukemia: metaanalyses of prevalence rates by measurement method. Blood. 2016;128:3610. 33. Noens L, Hensen M, Kucmin-Bemelmans I, et al. Measurement of adherence to BCR-ABL inhibitor therapy in chronic myeloid leukemia: current situation and future challenges. Haematologica. 2014;99:437447. Mauro MJ. Tailoring tyrosine kinase inhibitor therapy in chronic 34. myeloid leukemia. Cancer Contr. 2009;16:108-121. 35. Basch E, Jia X, Heller G, et al. Adverse symptom event reporting by patients vs clinicians: relationships with clinical outcomes. J Natl Cancer Inst. 2009;101:1624-1632. 36. Efficace F, Baccarani M, Breccia M, et al; GIMEMA. Health-related quality of life in chronic myeloid leukemia patients receiving long-term therapy with imatinib compared with the general population. Blood. 2011;118:4554-4560. 37. Efficace F, Baccarani M, Breccia M, et al. Chronic fatigue is the most important factor limiting health-related quality of life of chronic myeloid leukemia patients treated with imatinib. Leukemia. 2013;27:1511-1519. 38. Efficace F, Breccia M, Saussele S, et al. Which health-related quality of life aspects are important to patients with chronic myeloid leukemia receiving targeted therapies and to health care professionals? GIMEMA and EORTC Quality of Life Group. Ann Hematol. 2012;91:1371-1381. 39. Phillips KM, Pinilla-Ibarz J, Sotomayor E, et al. Quality of life outcomes in patients with chronic myeloid leukemia treated with tyrosine kinase inhibitors: a controlled comparison. Support Care Cancer. 2013;21:1097-1103. 40. Williams LA, Garcia Gonzalez AG, Ault P, et al. Measuring the symptom burden associated with the treatment of chronic myeloid leukemia. Blood. 2013;122:641-647. 41. Yellen SB, Cella DF, Webster K, et al. Measuring fatigue and other anemia-related symptoms with the Functional Assessment of Cancer Therapy (FACT) measurement system. J Pain Symptom Manage. 1997;13:63-74. 42. Ware JE. SF-36 Health Survey. Boston, MA: The Health Institute; 1993. 43. Kekäle M, Söderlund T, Koskenvesa P, et al. Impact of tailored patient education on adherence of patients with chronic myeloid leukaemia to tyrosine kinase inhibitors: a randomized multicentre intervention study. J Adv Nurs. 2016;72:2196-2206. 44. Bower JE, Bak K, Berger A, et al; American Society of Clinical Oncology. Screening, assessment, and management of fatigue in adult survivors of cancer: an American Society of Clinical Oncology clinical practice guideline adaptation. J Clin Oncol. 2014;32:1840-1850. 45. Gielissen MFM, Verhagen S, Witjes F, et al. Effects of cognitive behavior therapy in severely fatigued disease-free cancer patients compared with patients waiting for cognitive behavior therapy: a randomized controlled trial. J Clin Oncol. 2006;24:4882-4887. 46. Poort H, Meade CD, Knoop H, et al. Adapting an evidence-based intervention to address targeted therapy-related fatigue in chronic myeloid leukemia patients. Cancer Nurs. Epub 2016 Nov 8.

SURVIVORSHIP CARE NEEDS

47. ClinicalTrials.gov. Cognitive Behavioral Intervention for Targeted Therapy Fatigue (CBT-TTF) Intervention. https://clinicaltrials.gov/ct2/ show/NCT02592447. Accessed February 16, 2017.

65. Larkin J, Chiarion-Sileni V, Gonzalez R, et al. Combined nivolumab and ipilimumab or monotherapy in untreated melanoma. N Engl J Med. 2015;373:23-34.

48. Conroy T, Desseigne F, Ychou M, et al; Groupe Tumeurs Digestives of Unicancer; PRODIGE Intergroup. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med. 2011;364:1817-1825.

66. Postow MA, Chesney J, Pavlick AC, et al. Nivolumab and ipilimumab versus ipilimumab in untreated melanoma. N Engl J Med. 2015;372:2006-2017.

49. Von Hoff DD, Ervin T, Arena FP, et al. Increased survival in pancreatic cancer with nab-pacl*taxel plus gemcitabine. N Engl J Med. 2013;369:1691-1703.

67. Davis MP, Dreicer R, Walsh D, et al. Appetite and cancer-associated anorexia: a review. J Clin Oncol. 2004;22:1510-1517.

50. Llovet JM, Ricci S, Mazzaferro V, et al; SHARP Investigators Study Group. Sorafenib in advanced hepatocellular carcinoma. N Engl J Med. 2008;359:378-390. 51. Enzinger PC, Burtness BA, Niedzwiecki D, et al. CALGB 80403 (Alliance)/ E1206: a randomized phase II study of three chemotherapy regimens plus cetuximab in metastatic esophageal and gastroesophageal junction cancers. J Clin Oncol. 2016;34:2736-2742. 52. Valle J, Wasan H, Palmer DH, et al; ABC-02 Trial Investigators. Cisplatin plus gemcitabine versus gemcitabine for biliary tract cancer. N Engl J Med. 2010;362:1273-1281. Schiller JH, Harrington D, Belani CP, et al; Eastern Cooperative 53. Oncology Group. Comparison of four chemotherapy regimens for advanced non-small-cell lung cancer. N Engl J Med. 2002;346: 92-98. Atkins MB, Lotze MT, Dutcher JP, et al. High-dose recombinant 54. interleukin 2 therapy for patients with metastatic melanoma: analysis of 270 patients treated between 1985 and 1993. J Clin Oncol. 1999;17:2105-2116. Lynch TJ, Bell DW, Sordella R, et al. Activating mutations in the 55. epidermal growth factor receptor underlying responsiveness of nonsmall-cell lung cancer to gefitinib. N Engl J Med. 2004;350:2129-2139. Shaw AT, Kim DW, Mehra R, et al. Ceritinib in ALK-rearranged non56. small-cell lung cancer. N Engl J Med. 2014;370:1189-1197. Shaw AT, Kim DW, Nakagawa K, et al. Crizotinib versus chemotherapy 57. in advanced ALK-positive lung cancer. N Engl J Med. 2013;368:23852394. Robert C, Karaszewska B, Schachter J, et al. Improved overall survival 58. in melanoma with combined dabrafenib and trametinib. N Engl J Med. 2015;372:30-39. 59. Long GV, Stroyakovskiy D, Gogas H, et al. Combined BRAF and MEK inhibition versus BRAF inhibition alone in melanoma. N Engl J Med. 2014;371:1877-1888. 60. Chapman PB, Hauschild A, Robert C, et al; BRIM-3 Study Group. Improved survival with vemurafenib in melanoma with BRAF V600E mutation. N Engl J Med. 2011;364:2507-2516. 61. Garon EB, Rizvi NA, Hui R, et al; KEYNOTE-001 Investigators. Pembrolizumab for the treatment of non-small-cell lung cancer. N Engl J Med. 2015;372:2018-2028. Borghaei H, Paz-Ares L, Horn L, et al. Nivolumab versus docetaxel in 62. advanced nonsquamous non-small-cell lung cancer. N Engl J Med. 2015;373:1627-1639. 63. Brahmer J, Reckamp KL, Baas P, et al. Nivolumab versus docetaxel in advanced squamous-cell non-small-cell lung cancer. N Engl J Med. 2015;373:123-135. 64. Robert C, Long GV, Brady B, et al. Nivolumab in previously untreated melanoma without BRAF mutation. N Engl J Med. 2015;372:320330.

68. Miovic M, Block S. Psychiatric disorders in advanced cancer. Cancer. 2007;110:1665-1676. 69. Teunissen SC, Wesker W, Kruitwagen C, et al. Symptom prevalence in patients with incurable cancer: a systematic review. J Pain Symptom Manage. 2007;34:94-104. Nekolaichuk CL, Maguire TO, Suarez-Almazor M, et al. Assessing the 70. reliability of patient, nurse, and family caregiver symptom ratings in hospitalized advanced cancer patients. J Clin Oncol. 1999;17:36213630. 71. Cooley ME, Short TH, Moriarty HJ. Symptom prevalence, distress, and change over time in adults receiving treatment for lung cancer. Psychooncology. 2003;12:694-708. 72. LeBlanc TW, Nipp RD, Rushing CN, et al. Correlation between the international consensus definition of the cancer anorexia-cachexia syndrome (CACS) and patient-centered outcomes in advanced nonsmall cell lung cancer. J Pain Symptom Manage. 2015;49:680-689. 73. Paice JA, Portenoy R, Lacchetti C, et al. Management of chronic pain in survivors of adult cancers: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol. 2016;34:3325-3345. 74. Fromme EK, Eilers KM, Mori M, et al. How accurate is clinician reporting of chemotherapy adverse effects? A comparison with patient-reported symptoms from the Quality-of-Life Questionnaire C30. J Clin Oncol. 2004;22:3485-3490. 75. Laugsand EA, Sprangers MA, Bjordal K, et al. Health care providers underestimate symptom intensities of cancer patients: a multicenter European study. Health Qual Life Outcomes. 2010;8:104. 76. Atkinson TM, Li Y, Coffey CW, et al. Reliability of adverse symptom event reporting by clinicians. Qual Life Res. 2012;21:1159-1164. 77. Di Maio M, Gallo C, Leighl NB, et al. Symptomatic toxicities experienced during anticancer treatment: agreement between patient and physician reporting in three randomized trials. J Clin Oncol. 2015;33:910-915. 78. Bernabei R, Gambassi G, Lapane K, et al. Management of pain in elderly patients with cancer. SAGE Study Group. Systematic assessment of geriatric drug use via epidemiology. JAMA. 1998;279:1877-1882. Breetvelt IS, Van Dam FS. Underreporting by cancer patients: the case 79. of response-shift. Soc Sci Med. 1991;32:981-987. 80. Ward SE, Goldberg N, Miller-McCauley V, et al. Patient-related barriers to management of cancer pain. Pain. 1993;52:319-324. Basch E, Deal AM, Kris MG, et al. Symptom monitoring with patient81. reported outcomes during routine cancer treatment: a randomized controlled trial. J Clin Oncol. 2016;34:557-565. 82. Berry DL, Hong F, Halpenny B, et al. Electronic self-report assessment for cancer and self-care support: results of a multicenter randomized trial. J Clin Oncol. 2014;32:199-205. Strasser F, Blum D, von Moos R, et al; Swiss Group for Clinical Cancer 83. Research (SAKK). The effect of real-time electronic monitoring of

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 681

JACOBSEN, NIPP, AND GANZ

patient-reported symptoms and clinical syndromes in outpatient workflow of medical oncologists: E-MOSAIC, a multicenter clusterrandomized phase III study (SAKK 95/06). Ann Oncol. 2016;27:324332. 84. Weeks JC, Cook EF, O’Day SJ, et al. Relationship between cancer patients’ predictions of prognosis and their treatment preferences. JAMA. 1998;279:1709-1714. 85. Weeks JC, Catalano PJ, Cronin A, et al. Patients’ expectations about effects of chemotherapy for advanced cancer. N Engl J Med. 2012;367:1616-1625. 86. Temel JS, Shaw AT, Greer JA. Challenge of prognostic uncertainty in the modern era of cancer therapeutics. J Clin Oncol. 2016;34:36053608. 87. Steinhauser KE, Christakis NA, Clipp EC, et al. Preparing for the end of life: preferences of patients, families, physicians, and other care providers. J Pain Symptom Manage. 2001;22:727-737. 88. Steinhauser KE, Christakis NA, Clipp EC, et al. Factors considered important at the end of life by patients, family, physicians, and other care providers. JAMA. 2000;284:2476-2482. 89. Steinhauser KE, Clipp EC, McNeilly M, et al. In search of a good death: observations of patients, families, and providers. Ann Intern Med. 2000;132:825-832. 90. Epstein AS, Prigerson HG, O’Reilly EM, et al. Discussions of life expectancy and changes in illness understanding in patients with advanced cancer. J Clin Oncol. 2016;34:2398-2403. 91. Hagerty RG, Butow PN, Ellis PA, et al. Cancer patient preferences for communication of prognosis in the metastatic setting. J Clin Oncol. 2004;22:1721-1730. Jackson VA, Jacobsen J, Greer JA, et al. The cultivation of prognostic 92. awareness through the provision of early palliative care in the ambulatory setting: a communication guide. J Palliat Med. 2013;16:894-900. 93. Dying in America: improving quality and honoring individual preferences near the end of life. Mil Med. 2015;180:365-367. 94. Ubel PA, Abernethy AP, Zafar SY. Full disclosure—out-of-pocket costs as side effects. N Engl J Med. 2013;369:1484-1486. Zafar SY, Peppercorn JM, Schrag D, et al. The financial toxicity of cancer 95. treatment: a pilot study assessing out-of-pocket expenses and the insured cancer patient’s experience. Oncologist. 2013;18:381-390. 96. Lathan CS, Cronin A, Tucker-Seeley R, et al. Association of financial strain with symptom burden and quality of life for patients with lung or colorectal cancer. J Clin Oncol. 2016;34:1732-1740. 97. Ramsey S, Blough D, Kirchhoff A, et al. Washington State cancer patients found to be at greater risk for bankruptcy than people without a cancer diagnosis. Health Aff (Millwood). 2013;32:1143-1152. 98. Ramsey SD, Bansal A, Fedorenko CR, et al. Financial insolvency as a risk factor for early mortality among patients with cancer. J Clin Oncol. 2016;34:980-986.

102. de Boer AG, Taskila T, Ojajärvi A, et al. Cancer survivors and unemployment: a meta-analysis and meta-regression. JAMA. 2009;301:753-762. 103. Narang AK, Nicholas LH. Out-of-pocket spending and financial burden among Medicare beneficiaries with cancer. JAMA Oncol. Epub 2016 Nov 3. 104. Guy GP Jr, Ekwueme DU, Yabroff KR, et al. Economic burden of cancer survivorship among adults in the United States. J Clin Oncol. 2013;31:3749-3757. 105. Zajacova A, Dowd JB, Schoeni RF, et al. Employment and income losses among cancer survivors: Estimates from a national longitudinal survey of American families. Cancer. 2015;121:4425-4432. 106. Kale HP, Carroll NV. Self-reported financial burden of cancer care and its effect on physical and mental health-related quality of life among US cancer survivors. Cancer. 2016;122:283-289. 107. Steel JL, Geller DA, Kim KH, et al. Web-based collaborative care intervention to manage cancer-related symptoms in the palliative care setting. Cancer. 2016;122:1270-1282. 108. Meropol NJ, Schulman KA. Cost of cancer care: issues and implications. J Clin Oncol. 2007;25:180-186. 109. Neugut AI, Subar M, Wilde ET, et al. Association between prescription co-payment amount and compliance with adjuvant hormonal therapy in women with early-stage breast cancer. J Clin Oncol. 2011;29:25342542. 110. Streeter SB, Schwartzberg L, Husain N, et al. Patient and plan characteristics affecting abandonment of oral oncolytic prescriptions. J Oncol Pract. 2011; 7 (3, Suppl) 46s-51s. 111. Zullig LL, Peppercorn JM, Schrag D, et al. Financial distress, use of costcoping strategies, and adherence to prescription medication among patients with cancer. J Oncol Pract. 2013;9:60s-63s. 112. Nipp R, Zullig L, Peppercorn J, et al. Coping with cancer treatmentrelated financial burden. J Clin Oncol. 2014,32 (suppl 31; abstr 161). 113. Kent EE, Forsythe LP, Yabroff KR, et al. Are survivors who report cancerrelated financial problems more likely to forgo or delay medical care? Cancer. 2013;119:3710-3717. 114. Siegel K, Raveis VH, Houts P, et al. Caregiver burden and unmet patient needs. Cancer. 1991;68:1131-1140. 115. Given BA, Given CW, Kozachik S. Family support in advanced cancer. CA Cancer J Clin. 2001;51:213-231. 116. Palos GR, Mendoza TR, Liao KP, et al. Caregiver symptom burden: the risk of caring for an underserved patient with advanced cancer. Cancer. 2011;117:1070-1079. 117. Nipp RD, El-Jawahri A, Fishbein JN, et al. Factors associated with depression and anxiety symptoms in family caregivers of patients with incurable cancer. Ann Oncol. 2016;27:1607-1612.

Hewitt M, Rowland JH, Yancik R. Cancer survivors in the United States: 99. age, health, and disability. J Gerontol A Biol Sci Med Sci. 2003;58:82-91.

118. Grov EK, Dahl AA, Moum T, et al. Anxiety, depression, and quality of life in caregivers of patients with cancer in late palliative phase. Ann Oncol. 2005;16:1185-1191.

100. Yabroff KR, Lawrence WF, Clauser S, et al. Burden of illness in cancer survivors: findings from a population-based national sample. J Natl Cancer Inst. 2004;96:1322-1330.

119. Wadhwa D, Burman D, Swami N, et al. Quality of life and mental health in caregivers of outpatients with advanced cancer. Psychooncology. 2013;22:403-410.

101. Kirchhoff AC, Kuhlthau K, Pajolek H, et al. Employer-sponsored health insurance coverage limitations: results from the Childhood Cancer Survivor Study. Support Care Cancer. 2013;21:377-383.

120. Northouse LL, Mood D, Kershaw T, et al. Quality of life of women with recurrent breast cancer and their family members. J Clin Oncol. 2002;20:4050-4064.

682 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

SURVIVORSHIP CARE NEEDS

121. Temel JS, Greer JA, Muzikansky A, et al. Early palliative care for patients with metastatic non-small-cell lung cancer. N Engl J Med. 2010;363:733-742.

125. Bakitas MA, Tosteson TD, Li Z, et al. Early versus delayed initiation of concurrent palliative oncology care: patient outcomes in the ENABLE III randomized controlled trial. J Clin Oncol. 2015;33:1438-1445.

122. Zimmermann C, Swami N, Krzyzanowska M, et al. Early palliative care for patients with advanced cancer: a cluster-randomised controlled trial. Lancet. 2014;383:1721-1730.

126. Temel JS, Greer JA, El-Jawahri A, et al. Effects of early integrated palliative care in patients with lung and GI cancer: a randomized clinical trial. J Clin Oncol. Epub 2016 Dec 28.

123. Temel JS, Jackson VA, Billings JA, et al. Phase II study: integrated palliative care in newly diagnosed advanced non-small-cell lung cancer patients. J Clin Oncol. 2007;25:2377-2382.

127. Ferrell BR, Temel JS, Temin S, et al. Integration of palliative care into standard oncology care: American Society of Clinical Oncology clinical practice guideline update. J Clin Oncol. 2017;35:96-112.

124. Bakitas M, Lyons KD, Hegel MT, et al. Effects of a palliative care intervention on clinical outcomes in patients with advanced cancer: the Project ENABLE II randomized controlled trial. JAMA. 2009;302:741-749.

128. Department of Health and Human Services. Oral Anticancer Agents: Utilization, Adherence, and Health Care Delivery (R01). https://grants.nih. gov/grants/guide/pa-files/PA-17- 060.html. Accessed January 19, 2017.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 683

JONES, EVES, AND SCOTT

Bench-to-Bedside Approaches for Personalized Exercise Therapy in Cancer Lee W. Jones, PhD, Neil D. Eves, PhD, and Jessica M. Scott, PhD OVERVIEW The past 2 decades have witnessed a growing body of work investigating the feasibility and efficacy of exercise therapy on a broad array of outcomes in many different oncology scenarios. Despite this heterogeneity, the exercise therapy prescription approach and the dose tested has been largely similar. Thus, current exercise therapy prescriptions in the oncology setting adopt a one-size-fits-all approach. In this article, we provide an overview of personalization of exercise therapy in cancer using the principles of training as an overarching framework. Specifically, we first review the fundamentals of exercise prescription in chronic disease before focusing attention on application of these principles to optimize the safety and efficacy of exercise therapy on (1) cancer treatment–induced cardiovascular toxicity and (2) tumor progression and metastasis.

he field of exercise oncology is now recognized as an emergent subdiscipline of oncology research and practice.1-5 Evidence supporting the efficacy of structured exercise therapy in the other major chronic diseases (e.g., coronary artery disease, heart failure) became standard of care in the 1960s and 1970s.6-8 In sharp contrast, the first investigation of exercise therapy for patients with cancer did not occur until the late 1980s, although consistent publications in this field did not occur until the late 1990s.1 Since these first investigations, the past 20 years have witnessed a relative explosion in the area of exercise oncology, with well over 200 studies examining the role of exercise (or physical activity) for patients with solid and hematologic malignancies. Of importance, despite considerable heterogeneity in study endpoints and study populations/settings, the exercise therapy prescriptions tested across these studies are relatively similar.1 Specifically, almost all studies closely adhered to the national exercise guidelines for patients with cancer (guidelines that are identical to exercise recommendations for all adults): endurance (aerobic) exercise either alone or in combination with resistance training, prescribed at a moderate intensity (50%–75% of a predetermined physiologic parameter, typically age-predicted heart rate maximum or reserve), achieved in two to five sessions per week for 10–60 minutes per session, with the ultimate objective of achieving at least 150 minutes of moderate-intensity or at least 75 minutes of vigorous-intensity exercise per week.3,9 Thus, current exercise therapy prescriptions in the

oncology setting adopt a one-size-fits-all approach, which is essentially analogous to all patients with cancer, regardless of age, histology, or oncogenic somatic genotype, receiving a similar type, dose, and schedule of anticancer therapy.10 Systematic reviews and meta-analyses do, however, indicate that exercise therapy following a generic prescription is safe, tolerable, and efficacious (at improving symptom control outcomes) for patients with cancer both during and following primary therapy.3,11 Nevertheless, several important caveats must be considered when interpreting these data: (1) the effects of exercise therapy are compared against a nonintervention (usual care) control group and the marked deleterious consequences of a sedentary lifestyle (i.e., deconditioning) are well established; (2) meta-analyses and systematic reviews do not include more recent data, including data from three large randomized controlled trials, which contrast current conclusions; and, most importantly, (3) there are insufficient data to comprehensively evaluate whether the safety or efficacy of exercise therapy differs as a function of cancer-related medical (e.g., type, treatment, stage) or even exercise prescription characteristics (e.g., dose, schedule). Thus, it is not known whether alternative dosing and scheduling prescriptions that adopt a more personalized approach confer superior efficacy. A strong rationale to test alternative approaches in cancer is provided by more than 50 years of research and practice in the athletic arena.10 Exercise scientists have continued to elucidate the determinants of human performance to continually refine and personalize exercise training dosing and scheduling to

From the Memorial Sloan Kettering Cancer Center, New York, NY; Centre for Heart, Lung, and Vascular Health, School of Health and Exercise Sciences, University of British Columbia, Kelowna, British Columbia, Canada. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Lee W. Jones, PhD, Department of Medicine, Memorial Sloan Kettering Cancer Center, 1275 York Ave., New York, NY 10065; email: [emailprotected]. © 2017 American Society of Clinical Oncology

684 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PERSONALIZED EXERCISE IN CANCER

minimize injury and maximize performance. The fundamental basis of all athletic training prescriptions adheres to the tenets of human exercise physiology known as the principles of training.10 This article aims to provide an overview of the evidence supporting personalization of exercise therapy in cancer using the principles of training as an overarching framework. Specifically, we first review the fundamentals of exercise prescription in chronic disease before focusing attention on the application of these principles to optimize the safety and efficacy of exercise therapy on (1) cancer treatmentinduced cardiovascular toxicity and (2) tumor progression and metastasis.

FUNDAMENTALS OF EFFECTIVE EXERCISE PRESCRIPTION IN CHRONIC DISEASE

Adoption of a personalized approach to exercise therapy prescription in clinical populations requires fundamental understanding of the underlying disease pathophysiology in question, providing insight into the unique limitations to exercise.12,13 Such insights can be obtained from objective evaluation of a patient’s baseline physiologic status to identify appropriate exercise intensities to personalize the prescription.10,14,15 For aerobic training prescriptions, an incremental cardiopulmonary exercise test (CPET) is the assessment of choice for the accurate quantification of cardiorespiratory fitness and integrative evaluation of cardiovascular, respiratory, skeletal muscle, and neuromuscular responses to exercise.16-18 Furthermore, exercise contraindications, adverse cardiovascular responses (e.g., hypertension, ischemia), and exertional symptoms can also be evaluated; such parameters are critical to prescribe safe as well as effica-

KEY POINTS • There is a growing body of work investigating the feasibility and efficacy of exercise therapy in cancer. • Exercise therapy prescriptions can be personalized to the physiologic status of each individual using the principles of training as a framework. • A nonlinearized approach that utilizes undulating training stress (achieved through varying the intensity and duration of training load) is safe and efficacious for improving cardiorespiratory fitness as well as other important clinical outcomes for several different clinical populations, including patients with cancer. • The importance of performing the appropriate amount of exercise at intensities that permit recovery and adaptation is underscored by clinical and preclinical studies indicating that systems already under stress may have increased susceptibility to exercise-induced pathology. • The CHALLENGE and INTERVAL trials will provide important and novel insights addressing the fundamental question of whether increasing exercise exposure following a cancer diagnosis alters disease outcomes.

cious prescriptions.16,19 Assessments of muscular strength, muscular endurance, and balance can also offer important insight regarding a variety of additional functional limitations20-22 and how the patient is adapting to the exercise prescription.23 Thus, a single or small battery of tests performed at baseline during short-duration training programs (4 to 12 weeks) or at regularly chosen intervals throughout longer interventions can help optimize the training prescription. The utilization of specific metabolic or ventilatory responses to CPET is shown to be superior to more generic prescriptions, such as those that use a percentage of maximal heart rate or maximal oxygen consumption (%VO2max),24-26 because considerable individual variability exists for the metabolic stress imposed by any given exercise load (i.e., percent of VO2max).27 More specifically, the utilization of blood lactate or ventilatory responses to incremental exercise can identify unique training zones that anchor different intensities of training to the two lactate or ventilatory thresholds (LTs or VTs, respectively) and VO2max for each individual independent of disease severity or baseline fitness.28 This approach allows for the generation of three (Fig. 1A) or sometimes five (Fig. 1B) unique training zones related to specific metabolic events that are associated with different types of endurance performance.28,29 Following objective evaluation of a patient’s baseline physiologic status, the next step for effective exercise prescription is to systematically design the training regimen using specific principles of training. The most important aspects of these principles have been described previously10 and will be briefly discussed here.

Individualization

All exercise training programs should be tailored to the physiologic status of each individual.10,26 Even among international elite-caliber athletes, considerable differences exist in the initial fitness level (including VO2max), the adaptive response to training, nutritional/sleep status, and thus the exercise stimulus dose required to achieve the desired physiologic adaptation.30,31 In chronic disease states, the level of heterogeneity is substantially amplified with comorbidities, chronic low-grade inflammation, altered sympathovagal balance, and pharmaceutical interactions that collectively alter training dose tolerability and adaptive response.

Specificity

Different types, intensities, durations, and frequencies of exercise therapy can confer markedly distinct physiologic adaptations. However, few studies specifically tailor the training prescription to optimally target the desired physiologic or performance outcome. Unlike molecularly targeted therapies, exercise therapy can cause markedly different adaptations depending on the exercise load and/or stimulus.32 For example, high-intensity, short-duration exercise sessions (i.e., Zone 3; Fig. 1A) above LT2/VT2 interval exercise for approximately 30 minutes activate mitochondrial biogenesis and capillarization within skeletal muscle, which do not commonly occur in response to low-to-moderate asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 685

JONES, EVES, AND SCOTT

intensity, longer-duration training (i.e., Zone 1; Fig. 1A) at LT1/VT1 and below for approximately 60 minutes or longer.33-35 Although the specific prescription for every desired physiologic adaptation is not known, a considerable body of research has identified the necessary dose, intensity, and/or duration of exercise to achieve a large number of physiologic adaptations to guide this fundamental approach to exercise therapy dosing.32

Progressive Overload

Physiologic adaptation requires an exercise load to challenge system homeostasis beyond that habitually performed.10,36 Repetitive exposure to a training load above habitual levels promotes adaptation (allostasis),37,38 with subsequent increases in training load required for continued adaptation. It is important to stress that progressive overload only confers physiologic adaptation with adequate rest and recovery (another key principle of training) to maximize the adaptive response.10,36,39 Rest and recovery are essential elements of any exercise therapy prescription (similar to drug-free breaks) because biologic resynthesis only occurs during rest, allowing the affected system(s) to adapt (supercompensation).39 Chronically overloading a system without adequate rest and recovery can lead to fatigue, maladaptation, and illness (allostatic overload or overtraining).37,40 This could stimulate worsening symptoms or inferior clinical outcomes in certain clinical populations. Several other principles, including the variety of the training type and stimulus, regularity, reversibility of the training adaptation, and maintenance and accommodation of exercise load, are also important considerations but are beyond the scope of this review.

Exercise Sequencing/Scheduling

A key component of exercise training in line with the principles of training is the systematic sequencing or scheduling of training to optimize physiologic adaptation and

enhance performance (known as periodization).41 In athletes, training stress is structured in a cyclic pattern with planned changes in training volume and intensity,42,43 with the goal of optimizing performance for a specific competition. Clearly, in most circ*mstances, clinical patients are not preparing for a specific competition, although analogous events in oncology could be preparing for surgery or cytotoxic therapy. Irrespective of the scenario, periodization is a key aspect for appropriately managing training stress to optimize treatment while helping to avoid excessive fatigue44 even with similar training volumes or loads being performed.45-50 Few studies have used periodization in the clinical setting, and there are many periodization models that could be used.41,46,51,52 However, in the limited available data, a nonlinearized approach that utilizes undulating training stress (achieved through varying the intensity and duration of training load) is safe and efficacious for improving cardiorespiratory fitness as well as other important clinical outcomes for several different clinical populations, including patients with cancer.53-56 A recent randomized controlled trial in chronic obstructive pulmonary disease (110 patients) demonstrated that patients treated with a nonlinear periodized prescription led to superior improvements in exercise tolerance (measured as constant load exercise time) and disease-specific, health-related quality of life compared with a generic exercise prescription.56 Further work testing and adopting different periodized approaches to optimize exercise training appears warranted.

Training-Intensity Distribution

Although it is somewhat counterintuitive, superior performance in endurance events has consistently been associated with higher volumes of lower-intensity training (i.e., exercise in Zone 1; below LT1/VT1 in Fig. 1A).28,57,58 Furthermore, performance of the majority of training (approximately 75%) in Zone 1 (Fig. 1A) offset with 15%–20% of training load at high intensities (i.e., Zone 3; Fig. 1A) confers superior

FIGURE 1. Exercise Intensity Zones

(A) Three-zone model: Zone 1 below ventilatory (VT)/lactate (LT) threshold 1; Zone 2 above VT1/LT1 and below VT2/LT2; Zone 3 above VT2 and below peak oxygen consumption. (B) Five-zone model: Zone 1 below VT1/LT1; Zone 2 equal to VT1/LT1; Zone 3 below VT2/LT2; Zone 4 equal to VT2/LT2; Zone 5 above VT2 and below peak oxygen consumption. Abbreviations: LT, lactate threshold; VT, ventilatory threshold.

686 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PERSONALIZED EXERCISE IN CANCER

improvements in endurance performance outcomes (including VO2max) compared with utilizing one intensity of training, such as high-intensity interval training, training near the LT/VT, or high-volume training at low intensities.29 This polarized approach to training28,29 demonstrates the importance of utilizing individualized exercise intensities with higher relative training loads to improve cardiorespiratory fitness, but it also highlights the importance of performing an appropriate amount of exercise at intensities that permit recovery and adaptation. In summary, personalized training based on a framework of key principles of training that use evidence-based approaches is postulated to result in superior physiologic adaptations leading to improved clinical outcomes. The remainder of this review will provide an overview of the application of these principles to effective exercise prescription and scheduling in the oncology setting.

PERSONALIZATION OF EXERCISE THERAPY TO MITIGATE TREATMENT-INDUCED CARDIOVASCULAR TOXICITY

Five-year relative cancer-specific survival rates for the 10 most common malignancies improved from 58% in 1977 to 73% in 2012.59 As a result, patients diagnosed with these malignancies now have sufficient longevity (with the exception of lung cancer) to be at risk for normal agerelated pathologies, predominantly cardiovascular diseases (CVDs) including heart failure, coronary artery disease, and stroke.60,61 Moreover, CVD is more common (twofold to fourfold higher risk) and occurs at an earlier age than that observed in the general population.62,63 The underlying pathogenesis of this heightened and accelerated CVD-risk phenotype relates to the direct (e.g., cytotoxic/ radiation-induced injury) and indirect (e.g., effects secondary to therapy, such as deconditioning) effects of adjuvant therapy.60 Importantly, the risk of cardiovascular-related toxicity is likely to increase with continual improvements in cancer-specific mortality,59 trends toward extended adjuvant therapy in which patients are exposed to drug therapy for longer periods of time,64 and testing and approval of novel targeted therapies with unique cardiovascular safety profiles.65 Despite the importance and changing landscape of the cardiovascular safety profile in cancer, an established standard-of-care approach to either prevent and/or treat such disorders does not yet exist.66 Exercise therapy is an established cornerstone of CVD prevention and treatment in multiple nononcologic settings.67 Exercise therapy improves a patient’s CVD risk profile via favorable alterations in insulin sensitivity, lipid profile, and blood pressure with concomitant improvements in the reserve capacity of the skeletal muscle/vasculature/cardiovascular axis.68-72 In contrast, the efficacy of exercise therapy to prevent/offset CVD in the oncology setting has received minimal attention. Recent epidemiologic data from our group indicate that postdiagnosis exercise exposure is independently associated with graded reductions in CVD events for patients with primary

breast cancer73 and adult survivors of childhood Hodgkin lymphoma.74 These data provide the initial rationale to develop a program of work to comprehensively elucidate the efficacy and mechanisms of exercise therapy on cardiovascular toxicity for patients with cancer. Building on the concepts introduced in the first section, here we draw on work from cardiovascular medicine as well as exercise oncology to provide an overview of novel approaches to the optimization of exercise therapy on cardiovascular toxicity via adoption of the principles of training.

Individualization

Use of generic prescriptions that fail to consider the baseline physiologic status of any individual increases the propensity for underdosing and/or overdosing of exercise therapy; such considerations may be particularly important in cancer. For example, use of an estimated or age-predicted maximum heart rate may provoke overtraining among patients with primary breast cancer receiving or previously treated with polychemotherapy. Such agents increase the risk of autonomic dysfunction,23,24 wherein an elevated resting heart rate decreases heart rate reserve.75,76 To circumvent use of heart rates to individualize training for patients with cancer, researchers have a number of alternative options, such as workloads (e.g., treadmill speed and power output) corresponding to a specific percentage of VO2peak (e.g., 55%, 65%) elicited during CPET.77,78 Moreover, utilization of the aforementioned blood lactate or ventilatory responses to incremental exercise to identify unique training zones has been used to further refine individualized exercise prescriptions.54,79 This approach abrogates therapy-induced declines, or improvements in VO2peak.54,79 Nevertheless, even with the adoption of this approach, notable heterogeneity in VO2peak is evident. For example, in a study of 50 men with localized prostate cancer who were randomly assigned to 24 weeks of aerobic training (5 days per week; 30–60 minutes at 55%–100% VO2peak) or to the usual-care control following radical prostatectomy,80 we found that the mean change in VO2peak in the exercise group was +2.4 mL/kg per minute, but the individual change in VO2peak ranged from approximately −18% to +32%. Similarly, Leon et al81 reported a significant group mean increase in high-density lipoprotein cholesterol (p < .001) and a decrease in triglycerides (p < .01) among 675 sedentary individuals who completed 20 weeks of aerobic training (3 days per week; 30–50 minutes at 55%–75% VO2peak). However, marked individual variability in high-density lipoprotein cholesterol (−24% to +66%) and triglyceride (−3% to +8%) delta change was observed. Thus, the heterogeneity in response supports, by definition, the adoption of personalized medicine approaches in exercise oncology. Similar to oncology-targeted agents, exercise prescriptions must be specific to the pathophysiology of the primary subclinical (e.g., exercise intolerance, left ventricular dysfunction, hypertension) or clinical (e.g., heart failure, coronary artery disease) CVD phenotype to improve patient morbidity and mortality. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 687

JONES, EVES, AND SCOTT

SIDEBAR 1. Clinical Vignettes Patient 1 is a 65-year-old woman with pre-existing hypertension with stage II, node-positive primary breast cancer undergoing anthracycline and pacl*taxel chemotherapy. On a CPET with echocardiography, her VO2peak was 15.2 mL/kg per minute (16% below sex- and age-matched sedentary values), and she had a peak heart rate of 150 beats per minute, a peak CO of 9 L/min, and a calculated peak cardiac output and arterial venous difference of 11 mL/dL. It was determined that impaired VO2peak was reduced primarily because of inotropic incompetence resulting in reduced peak stroke volume. Accordingly, the exercise therapy prescription is designed with a focus on high-intensity aerobic training (as part of a larger prescription that also includes adequate rest and recovery as well as lower-intensity training sessions), given the evidence that high-intensity exercise has been shown to be superior to low-intensity exercise at improving left ventricular contraction78 and in lowering blood pressure.79 Patient 2 is a 55-year-old man with hyperlipidemia undergoing androgen-deprivation therapy for prostate cancer. On a CPET with echocardiography, his VO2peak was 20.8 ml/kg per minute (35% below age-matched sedentary values), and he had a peak heart rate of 162 beats per minute, a peak CO of 15 L/min, and a calculated peak cardiac output and arterial venous difference of 11 mL/dL. It was determined that VO2peak was reduced primarily because of impaired peak cardiac output and arterial venous difference. As a result, the targeted exercise prescription is designed to focus on combined moderate-intensity aerobic and resistance training, based on the findings that moderate-intensity exercise is superior to vigorous exercise in lowering triglycerides80 and, in the setting of reduced peripheral oxygen uptake, targeted muscle training is critical to improving VO2peak.81

Specificity

Clearly, application of generic exercise prescriptions to offset different CVD-cancer phenotypes will be associated with heterogeneous responses. To facilitate practical understanding of precision exercise therapy for patients with an exercise-intolerance phenotype, Fig. 2 presents a three-step framework. The first step is to perform a CPET with gas exchange to determine VO2peak and to obtain an echocardiogram to determine peak cardiac output. The second step is to determine the underlying limitation to VO2peak. In accordance with the Fick equation,16 where VO2 is equal to the product of cardiac output and arterial venous difference, the primary limitation can be identified as a central limitation (e.g., decreased cardiac output owing to lusitropic, inotropic, or chronotropic incompetence)

or as a peripheral limitation (e.g., decreased cardiac output and arterial venous difference owing to decreased capillary density or impaired oxygen utilization by the exercising skeletal muscles). The third step is to use exercise prescriptions targeted to the primary limitation to unveil the full therapeutic potential of exercise therapy to augment VO2peak (Fig. 2). To illustrate the potential applicability of personalized exercise therapy, data from two representative patients are presented (Sidebar 1).

Progressive Overload

Nonlinear prescriptions vary between low, moderate, and high intensity to target various physiologic systems involved in the cardiovascular response to exercise therapy.10 To date, approximately six trials, in various oncology settings, have

FIGURE 2. Three-Step Framework to Facilitate Practical Application of a Personalized Approach for Patients With Cancer With Exercise Intolerance

Abbreviations: A-VO2 Diff, cardiac output and arterial venous difference; CO, cardiac output; CPET, cardiopulmonary exercise test; O2, oxygen.

688 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PERSONALIZED EXERCISE IN CANCER

explored the safety, tolerability, and initial efficacy of nonlinear prescriptions.10 For example, our group conducted a phase II randomized trial to determine the efficacy of 12 weeks of nonlinear aerobic training (3 days per week, 15–45 minutes per session ranging in intensity from 60%–100% peak workload) in attenuating anthracycline-induced changes in VO2peak.79 Intention-to-treat analysis indicated that VO2peak decreased by 1.5 ± 2.2 mL/kg per minute (−9%) in the usual-care group and increased by 2.6 ± 3.5 mL/kg per minute (+13%) in the exercise group (between-group difference, p = .001). Importantly, this was the first study to show significant improvements in VO2peak among patients with breast cancer receiving adjuvant chemotherapy; other studies using linear prescriptions reported nonsignificant improvements in VO2peak in exercise groups.77,82 The importance of performing the appropriate amount of exercise at intensities that permit recovery and adaptation is underscored by clinical and preclinical studies indicating that systems already under stress may have increased susceptibility to exercise-induced pathology. For example, an ancillary analysis of 90 patients with cancer from the Heart Failure: A Controlled Trial Investigating Outcomes of Exercise Training (HF-ACTION) trial who were randomly assigned to aerobic training (three times per week, 20–45 minutes at 60%–70% heart rate reserve) or the usual-care control revealed that the incidence of cardiovascular mortality or cardiovascular hospitalization was significantly higher in the exercise group versus the usual-care group (41% vs. 67%; adjusted hazard ratio, 1.94; 95% CI, 1.12–3.16; p = .017).83 In preclinical work, Huang et al84 demonstrated that strenuous exercise (90 minutes twice a day for 14 days) resulted in left ventricular hypertrophy, myofibril disarray, and increased fibrosis in juvenile mice exposed to anthracycline-containing therapy. It is important to note that neither study adhered to a nonlinear or progressive overload approach. Whether a nonlinear approach is superior to a linear exercise prescription in improving CVD phenotypes and conferring reductions in clinical events in cancer remains unknown. As an initial step, our group is comparing the effects of nonlinear versus linear aerobic training on changes in VO2peak among 174 patients with primary breast cancer with exercise intolerance.85

PERSONALIZATION OF EXERCISE THERAPY TO MODULATE TUMOR PROGRESSION AND METASTASIS

The majority of research efforts in exercise oncology to date have, and continue to, focus on the efficacy of exercise therapy to mitigate acute and chronic patient-reported and/or physiologic toxicities associated with cytotoxic therapy.1 However, in recent years, a parallel line of investigation is focusing on the antitumor effects of exercise among patients with or who at risk for cancer.4 Studies in this arena were launched by work showing for the first time that (selfreported) exercise was inversely associated with the risk of recurrence and cancer-specific mortality in primary breast cancer. Specifically, Holmes et al86 found that compared with

less than 3 MET-hrs·wk−1 of physical activity (i.e., all types of physical activity including structured exercise were evaluated), the adjusted relative risk of death from breast cancer was 0.50 (95% CI, 0.31–0.82) for 9–14.9 MET-hrs·wk−1 (i.e., equivalent to approximately 150–250 minutes of moderate-intensity exercise per week) among 2,987 patients with primary breast cancer. Since the publication of this seminal work, approximately 26 studies have evaluated whether exposure to exercise and general physical activity following a cancer diagnosis alters disease pathogenesis. In a recent systematic review, postdiagnosis exercise was associated with, on average, a 37% reduction (95% CI, 0.54–0.73) in the risk of cancer-specific mortality, comparing the most- versus the least-active patients.87 The available observational data, together with ancillary data from small randomized trials showing that exercise therapy alters circulating levels of various factors postulated to underpin the exercise/cancer progression relationship (e.g., sex and metabolic-steroid hormones and growth factors, immune/inflammation axis effectors),5 have led to the popular assertion that sufficient data exists to initiate largescale, phase III trials to definitively test the efficacy of exercise on disease outcomes among patients with cancer.87-90 Indeed, two such trials are currently underway: the Colon Health and Life-Long Exercise Change (CHALLENGE) trial includes 962 patients with resected stage III colorectal cancer,91 and the Intense Exercise for Survival Among Men With Metastatic Castrate-Resistant Prostate Cancer (INTERVAL) trial includes 866 patients with metastatic prostate cancer.92 There is little argument that adequately powered randomized controlled trials remain the gold standard and provide the best evidence of causality. In this respect, the CHALLENGE and INTERVAL trials will provide important and novel insights addressing the fundamental question of whether increasing exercise exposure following a cancer diagnosis alters disease outcomes. Nevertheless, several important knowledge gaps exist that preclude the optimal design of definitive exercise trials. These include, but are not limited to, the following: (1) identification of the optimal treatment dose and schedule, (2) mechanistic understanding of how exercise might delay and/or prevent cancer progression, and (3) predictors of response to guide patient selection. Parallel translational studies that address these research gaps will permit continual refinement of exercise dosing to optimize the safety and efficacy of exercise therapy as a candidate antitumor strategy.4 To facilitate such efforts, we provide an overview of these knowledge gaps in the following section, using the concepts of individualization, specificity, and progressive overload from the principles of training.

Individualization

Classically, individualization of exercise is considered only from the perspective of improvements in exercise performance (using parameters such as CPET, age, and concomitant comorbidities). Clearly, consideration of these factors is essential when designing any exercise prescription irrespective of the therapeutic target; however, we postulate that asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 689

JONES, EVES, AND SCOTT

additional factors must be considered when designing exercise trials with therapeutic intent (i.e., the primary endpoint of interest is tumor specific). These factors can be broadly categorized into tumor-related and host-related factors. Tumor-related factors. Human cancer is a biologically heterogeneous disease with well-defined clinical and molecular subgroups. Stratification by clinical and molecular subgroups in all solid tumors reveals stark differences in prognosis as well as response to conventional and novel anticancer treatments. Given this, the notion that tumor response to exercise may also differ by such characteristics appears plausible. To this end, investigators have started to explore whether the relationship differs as a function of tumor features. In terms of clinicopathologic features (e.g., tumor stage, estrogen receptor status), Holmes et al86 found that exercise exposure (≥ 9 metabolic equivalents of task hours of physical activity per week) was associated with a substantial 50% reduction in breast cancer death in estrogen receptor (ER)–positive tumors compared with a nonsignificant 9% reduction in ER-negative tumors. At least three other independent observational studies also found that ER-positive tumors were more responsive to exercise.93-95 In corroboration and extension of this work, we found that increasing exercise exposure was not associated with a reduction in the risk of recurrence or breast cancer death for 6,211 patients (i.e., unselected cohort).94 However, stratification by clinical subtype indicated that the hormone receptor–positive, HER2-negative clinical subtype was preferentially responsive to exercise; exercise did not reduce the risk of breast cancer death in the other two clinical subtypes (i.e., HER2-positive or triple-negative clinical). In terms of molecular features, a series of studies from the same group reported that tumor PTGS2 positivity, CTNNB1 negativity, and expression of CDKN1B (p27) predict sensitivity to exercise in colorectal cancer.96-98 Clearly, such findings are hypothesis generating, requiring validation in independent cohorts and biologic confirmation in preclinical studies. Nevertheless, as in oncology drug trials, tumor-related factors likely contribute to exercise efficacy and therefore may, in turn, inform patient selection (into exercise trials) as well as provide further information on how to individualize the exercise therapy prescription. Host-related factors. To date, personalized oncology medicine has focused predominantly on elucidation of tumorcentric factors to predict therapeutic response. More recent work, however, has highlighted the importance of how host-related factors (e.g., genetic predisposition, circulating concentrations and function of immune surveillance phenotypes, inflammatory or metabolic effectors, gut microbiota) contribute to and/or modify the antitumor activity of conventional and novel agents. To our knowledge, whether host-related factors modify or predict tumor response to exercise therapy has not been investigated, although initial insights can be gleaned from related work. For instance, Bonanni et al99 investigated the effects of 4 weeks of metformin compared with placebo on markers of tumor proliferation (Ki-67) for 200 patients with primary breast 690 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

cancer prior to surgical resection. Intention-to-treat analyses indicated no differences in Ki-67 between study arms; however, a differential effect of metformin was observed in secondary, unplanned analyses as a function of baseline insulin resistance. Specifically, pretreatment insulin resistance (i.e., homeostasis model assessment index > 2.8, fasting glucose [mmol/L] × insulin [mU/L]/22.5) was associated with a nonsignificant mean proportional decrease in Ki-67 of 10.5%, whereas noninsulin resistance was associated with a nonsignificant increase of 11.1% in Ki-67. These data were corroborated in a subsequent trial, indicating that the efficacy of metformin differed as a function of homeostasis model assessment index as well as other circulating metabolic factors and tumor clinical subtype.100 Together, these data suggest that the efficacy of metformin differs as a function of host and tumor-related characteristics in primary breast cancer. There is a long history of work in general exercise physiology focused on exercise genomics, which is essentially, the application of genome-wide association studies as well as targeted approaches to further understanding of the interaction between genetic predisposition and related pathways to predict human response to physical activity and exercise training interventions. Not surprisingly, this work has focused on genetic predictors of exercise or sports performance, with a paucity of work examining whether the germline DNA profile predicts the primary incidence of cancer or outcomes after a cancer diagnosis. Indeed, to our knowledge, only two studies to date have investigated this question. Nkondjock et al101 found no association between physical activity levels and risk of breast cancer among women with a BRCA mutation, whereas King et al102 reported that exercise delayed/reduced the lifetime risks of ovarian cancer by 54% in BRCA1 mutation carriers and 23% for BRCA2 mutation carriers. To our knowledge, whether the association between exercise and disease outcomes for patients with cancer differs on the basis of genetic predisposition (sequencing of germline DNA) has not been evaluated.

Specificity

Adoption of a generic prescription for oncology therapeuticintent studies is problematic because it assumes that the exercise load required to optimally modulate the cardiovascular system and tumor outcomes is the same, or that exerciseinduced modulation of tumor outcomes is dependent on improvements in cardiorespiratory fitness. This assumption has not been directly tested in the oncology setting and has surprisingly received little attention in other clinical settings; nevertheless, at least one trial provides some initial insight. Kraus et al103 investigated the effects of three aerobic training prescriptions that differed in total weekly duration and intensity on cardiorespiratory fitness and lipoproteins for 84 overweight men and women with mild-to-moderate dyslipidemia for 6–8 months. Of interest, improvements in fitness were similar in the high-intensity prescriptions (i.e., high-duration/high-intensity [65%–80% of VO2peak]; lowduration/high-intensity), yet improvements in lipoprotein

PERSONALIZED EXERCISE IN CANCER

profiles were consistently superior with high-duration/ high-intensity training, indicating that the effects of exercise on different end points differ as a function of the exercise stimulus. Thus, the exercise prescription must be specific and targeted to the primary endpoint or system(s) or pathway(s) known or postulated to underpin the effects of exercise on the primary therapeutic target. For example, in the context of breast cancer, current observational data suggest that ER-positive tumors are particularly responsive to exercise,94 creating the rational hypothesis that exercise prescriptions should be designed to optimally inhibit ER activity, its ligands, or coactivating pathways. Similarly, in tumors with PIK3CA mutations, prescriptions could be designed to optimally modulate circulating metabolic growth factor ligands that regulate the phosphoinositide 3-kinase/ AKT/mTOR signaling cascade.104 Rational dosing loads and schedules could be conceived, tested, and validated in clinically relevant animal models prior to testing and validation in human investigations. Such translational studies will be an exciting area of future research in exercise oncology.

Progressive Overload

There is minimal understanding of the complex interplay between the underlying concept of progressive overload and alterations in whole-organism, tissue, and cellular biology to regulate disease phenotypes,105 including cancer pathogenesis. Observational studies indicate an inverse linear dose-response relationship between exercise and cardiovascular events, supporting the hypothesis that higher exercise doses cause linear improvements in cardiovascular risk profiles.73,74 However, parallel observational data with cancer outcomes as the endpoints of interest indicate, in general, a nonlinear relationship.87 In other words, increasing exercise is associated with linear reductions in the risk of recurrence and cancer mortality but only up to a specific threshold; exercise exposure beyond this threshold is associated with an attenuated effect on cancer outcomes, suggesting that

an upper threshold or optimal dose of exercise exists to impact cancer outcomes. The importance of elucidating the dose-response effect in exercise and cancer outcomes is underscored by preclinical data studies showing that moderate-intensity exercise (forced swimming, 8 minutes per day, 9 weeks) reduced metastatic burden in the lung, whereas strenuous exercise (forced swimming, 16 or 32 minutes per day, 9 weeks) accelerated metastasis.106 Similarly, the conventional treatment approach is delivery of the maximum tolerated dose, because the maximum tolerated dose is considered to provide optimal cancer cell killing. However, elegant preclinical work demonstrated that the conventional maximum tolerated dose accelerates the emergence of resistant clones, whereas evolution-based approaches (e.g., initial tumor control with intense doses then maintenance with smaller, variable doses) may prolong progressionfree survival.107 Such a dosing schedule is somewhat analogous to a nonlinear exercise prescription that adheres to the concept of progressive overload; no study to date has compared the antitumor activity of standard versus nonlinear exercise dosing on tumor outcomes.

CONCLUSION

On the basis of progress over the past decade, it is anticipated that exercise therapy will become an increasingly important strategy in cancer prevention and control efforts over the next 2 decades. Continued progress in this arena will require close attention to the adoption of the concepts presented here to optimize the safety and efficacy of ex ercise in cancer. We focused attention on cardiovascular toxicity and oncology endpoints in cancer, but the concepts and approach described can be applied to essentially any endpoint, including patient-reported outcomes. It is hoped that attention to these issues will provide the platform for constructive dialogue with the view toward the development of standardized exercise guidelines for patients with cancer.

References 1. Jones LW, Alfano CM. Exercise-oncology research: past, present, and future. Acta Oncol. 2013;52:195-215.

7. Bethell HJ. Cardiac rehabilitation: from Hellerstein to the millennium. Int J Clin Pract. 2000;54:92-97.

2. Courneya KS, Friedenreich CM. Framework PEACE: an organizational model for examining physical exercise across the cancer experience. Ann Behav Med. 2001;23:263-272.

8. Mampuya WM. Cardiac rehabilitation past, present and future: an overview. Cardiovasc Diagn Ther. 2012;2:38-49.

3. Rock CL, Doyle C, Demark-Wahnefried W, et al. Nutrition and physical activity guidelines for cancer survivors. CA Cancer J Clin. 2012;62:243274. 4. Jones LW. Precision oncology framework for investigation of exercise as treatment for cancer. J Clin Oncol. 2015;33:4134-4137. 5. Betof AS, Dewhirst MW, Jones LW. Effects and potential mechanisms of exercise training on cancer progression: a translational perspective. Brain Behav Immun. 2013;30(Suppl):S75-S87. 6. Naughton J, Lategola MT, Shanbour K. A physical rehabilitation program for cardiac patients: a progress report. Am J Med Sci. 1966;252:545-553.

9. Schmitz KH, Courneya KS, Matthews C, et al; American College of Sports Medicine. American College of Sports Medicine roundtable on exercise guidelines for cancer survivors. Med Sci Sports Exerc. 2010;42:1409-1426. 10. Sasso JP, Eves ND, Christensen JF, et al. A framework for prescription in exercise-oncology research. J Cachexia Sarcopenia Muscle. 2015;6:115-124. 11. Speck RM, Courneya KS, Mâsse LC, et al. An update of controlled physical activity trials in cancer survivors: a systematic review and meta-analysis. J Cancer Surviv. 2010;4:87-100. 12. Bourjeily G, Rochester CL. Exercise training in chronic obstructive pulmonary disease. Clin Chest Med. 2000;21:763-781.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 691

JONES, EVES, AND SCOTT

13. Cooper CB. Exercise in chronic pulmonary disease: limitations and rehabilitation. Med Sci Sports Exerc. 2001;33 (Suppl):S643-S646.

32. Wasfy MM, Baggish AL. Exercise dose in clinical practice. Circulation. 2016;133:2297-2313.

14. Cooper CB. Exercise in chronic pulmonary disease: aerobic exercise prescription. Med Sci Sports Exerc. 2001;33 (Suppl):S671-S679.

33. Egan B, Carson BP, Garcia-Roves PM, et al. Exercise intensitydependent regulation of peroxisome proliferator-activated receptor coactivator-1 mRNA abundance is associated with differential activation of upstream signalling kinases in human skeletal muscle. J Physiol. 2010;588:1779-1790.

15. Ong KC, Benedicto JP, Chan AH, et al. Cardiopulmonary exercise testing in heart transplant candidates. Ann Acad Med Singapore. 2000;29:442-446. 16. American Thoracic Society; American College of Chest Physicians. ATS/ ACCP statement on cardiopulmonary exercise testing. Am J Respir Crit Care Med. 2003;167:211-277.

34. MacInnis MJ, Gibala MJ. Physiological adaptations to interval training and the role of exercise intensity. J Physiol. Epub 2016 Oct 17.

17. Jones LW, Eves ND, Haykowsky M, et al. Cardiorespiratory exercise testing in clinical oncology research: systematic review and practice recommendations. Lancet Oncol. 2008;9:757-765.

35. MacInnis MJ, Haikalis ME, Martin BJ, et al. Mitochondrial adaptation to training: superior effect of interval versus continuous exercise when work is matched. Med Sci Sports Exerc. 2016;48:747 (suppl; abstr 2673).

18. Myers J, Arena R, Cahalin LP, et al. Cardiopulmonary exercise testing in heart failure. Curr Probl Cardiol. 2015;40:322-372. 19. Fletcher GF, Ades PA, Kligfield P, et al; American Heart Association Exercise, Cardiac Rehabilitation, and Prevention Committee of the Council on Clinical Cardiology, Council on Nutrition, Physical Activity and Metabolism, Council on Cardiovascular and Stroke Nursing, and Council on Epidemiology and Prevention. Exercise standards for testing and training: a scientific statement from the American Heart Association. Circulation. 2013;128:873-934.

36. Kraemer WJ, Adams K, Cafarelli E, et al; American College of Sports Medicine. American College of Sports Medicine position stand. Progression models in resistance training for healthy adults. Med Sci Sports Exerc. 2002;34:364-380. 37. McEwen BS. Stressed or stressed out: what is the difference? J Psychiatry Neurosci. 2005;30:315-318. 38. McEwen BS, Wingfield JC. The concept of allostasis in biology and biomedicine. Horm Behav. 2003;43:2-15.

20. Beebe JA, Lang CE. Relationships and responsiveness of six upper extremity function tests during the first six months of recovery after stroke. J Neurol Phys Ther. 2009;33:96-103.

39. Hickson RC, Hagberg JM, Ehsani AA, et al. Time course of the adaptive responses of aerobic power and heart rate to training. Med Sci Sports Exerc. 1981;13:17-20.

21. Nyberg A, Saey D, Maltais F. Why and how limb muscle mass and function should be measured in patients with chronic obstructive pulmonary disease. Ann Am Thorac Soc. 2015;12:1269-1277.

40. Kreher JB, Schwartz JB. Overtraining syndrome: a practical guide. Sports Health. 2012;4:128-138.

22. Yelnik A, Bonan I. Clinical tools for assessing balance disorders. Neurophysiol Clin. 2008;38:439-445. 23. Larose J, Sigal RJ, Khandwala F, et al; Diabetes Aerobic and Resistance Exercise (DARE) trial investigators. Associations between physical fitness and HbA1(c) in type 2 diabetes mellitus. Diabetologia. 2011;54:93-102. 24. Mann T, Lamberts RP, Lambert MI. Methods of prescribing relative exercise intensity: physiological and practical considerations. Sports Med. 2013;43:613-625. 25. Serres I, Varray A, Vallet G, et al. Improved skeletal muscle performance after individualized exercise training in patients with chronic obstructive pulmonary disease. J Cardiopulm Rehabil. 1997;17:232-238. 26. Vallet G, Varray A, Fontaine JL, et al. Value of individualized rehabilitation at the ventilatory threshold level in moderately severe chronic obstructive pulmonary disease. Rev Mal Respir. 1994;11:493-501. 27. Scharhag-Rosenberger F, Meyer T, Gässler N, et al. Exercise at given percentages of VO2max: heterogeneous metabolic responses between individuals. J Sci Med Sport. 2010;13:74-79. 28. Seiler KS, Kjerland GO. Quantifying training intensity distribution in elite endurance athletes: is there evidence for an “optimal” distribution? Scand J Med Sci Sports. 2006;16:49-56. 29. Stöggl T, Sperlich B. Polarized training has greater impact on key endurance variables than threshold, high intensity, or high volume training. Front Physiol. 2014;5:33. 30. Bacon AP, Carter RE, Ogle EA, et al. VO2max trainability and high intensity interval training in humans: a meta-analysis. PLoS One. 2013;8:e73182. 31. Bonafiglia JT, Rotundo MP, Whittall JP, et al. Inter-individual variability in the adaptive responses to endurance and sprint interval training: a randomized crossover study. PLoS One. 2016;11:e0167790.

692 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

41. Lorenz DS, Reiman MP, Walker JC. Periodization: current review and suggested implementation for athletic rehabilitation. Sports Health. 2010;2:509-518. 42. Smith DJ. A framework for understanding the training process leading to elite performance. Sports Med. 2003;33:1103-1126. 43. Fleck SJ. Non-linear periodization for general fitness & athletes. J Hum Kinet. 2011;29A:41-45. 44. Buford TW, Rossi SJ, Smith DB, et al. A comparison of periodization models during nine weeks with equated volume and intensity for strength. J Strength Cond Res. 2007;21:1245-1250. 45. Hartmann H, Wirth K, Keiner M, et al. Short-term periodization models: effects on strength and speed-strength performance. Sports Med. 2015;45:1373-1386. 46. Rønnestad BR, Ellefsen S, Nygaard H, et al. Effects of 12 weeks of block periodization on performance and performance indices in welltrained cyclists. Scand J Med Sci Sports. 2014;24:327-335. 47. Rønnestad BR, Hansen J, Ellefsen S. Block periodization of highintensity aerobic intervals provides superior training effects in trained cyclists. Scand J Med Sci Sports. 2014;24:34-42. 48. Rønnestad BR, Hansen J, Thyli V, et al. 5-week block periodization increases aerobic power in elite cross-country skiers. Scand J Med Sci Sports. 2016;26:140-146. 49. Rhea MR, Ball SD, Phillips WT, et al. A comparison of linear and daily undulating periodized programs with equated volume and intensity for strength. J Strength Cond Res. 2002;16:250-255. 50. Rhea MR, Phillips WT, Burkett LN, et al. A comparison of linear and daily undulating periodized programs with equated volume and intensity for local muscular endurance. J Strength Cond Res. 2003;17:82-87.

PERSONALIZED EXERCISE IN CANCER

51. Lorenz D, Morrison S. Current concepts in periodization of strength and conditioning for the sports physical therapist. Int J Sports Phys Ther. 2015;10:734-747.

69. Flynn KE, Piña IL, Whellan DJ, et al; HF-ACTION Investigators. Effects of exercise training on health status in patients with chronic heart failure: HF-ACTION randomized controlled trial. JAMA. 2009;301:1451-1459.

52. Prestes J, da Cunha Nascimento D, Tibana RA, et al. Understanding the individual responsiveness to resistance training periodization. Age (Dordr). 2015;37:55.

70. Erbs S, Höllriegel R, Linke A, et al. Exercise training in patients with advanced chronic heart failure (NYHA IIIb) promotes restoration of peripheral vasomotor function, induction of endogenous regeneration, and improvement of left ventricular function. Circ Heart Fail. 2010;3:486-494.

53. Jones LW, Eves ND, Peterson BL, et al. Safety and feasibility of aerobic training on cardiopulmonary function and quality of life in postsurgical nonsmall cell lung cancer patients: a pilot study. Cancer. 2008;113:3430-3439. 54. Jones LW, Hornsby WE, Freedland SJ, et al. Effects of nonlinear aerobic training on erectile dysfunction and cardiovascular function following radical prostatectomy for clinically localized prostate cancer. Eur Urol. 2014;65:852-855. 55. Jones LW, Peddle CJ, Eves ND, et al. Effects of presurgical exercise training on cardiorespiratory fitness among patients undergoing thoracic surgery for malignant lung lesions. Cancer. 2007;110:590-598. 56. Klijn P, van Keimpema A, Legemaat M, et al. Nonlinear exercise training in advanced chronic obstructive pulmonary disease is superior to traditional exercise training. A randomized trial. Am J Respir Crit Care Med. 2013;188:193-200. 57. Muñoz I, Cejuela R, Seiler S, et al. Training-intensity distribution during an ironman season: relationship with competition performance. Int J Sports Physiol Perform. 2014;9:332-339. 58. Esteve-Lanao J, Foster C, Seiler S, et al. Impact of training intensity distribution on performance in endurance athletes. J Strength Cond Res. 2007;21:943-949. 59. Miller KD, Siegel RL, Lin CC, et al. Cancer treatment and survivorship statistics, 2016. CA Cancer J Clin. 2016;66:271-289. 60. Jones LW, Haykowsky MJ, Swartz JJ, et al. Early breast cancer therapy and cardiovascular injury. J Am Coll Cardiol. 2007;50:1435-1441. 61. Mozaffarian D, Benjamin EJ, Go AS, et al; Writing Group Members; American Heart Association Statistics Committee; Stroke Statistics Subcommittee. Heart disease and stroke statistics-2016 update: a report from the American Heart Association. Circulation. 2016;133:e38-e360. 62. Scott JM, Adams SC, Koelwyn GJ, et al. Cardiovascular late effects and exercise treatment in breast cancer: current evidence and future directions. Can J Cardiol. 2016;32:881-890. 63. Scott JM, Armenian S, Giralt S, et al. Cardiovascular disease following hematopoietic stem cell transplantation: pathogenesis, detection, and the cardioprotective role of aerobic training. Crit Rev Oncol Hematol. 2016;98:222-234. 64. Koelwyn GJ, Khouri M, Mackey JR, et al. Running on empty: cardiovascular reserve capacity and late effects of therapy in cancer survivorship. J Clin Oncol. 2012;30:4458-4461. 65. Moslehi JJ. Cardiovascular toxic effects of targeted cancer therapies. N Engl J Med. 2016;375:1457-1467. 66. Khouri MG, Douglas PS, Mackey JR, et al. Cancer therapy-induced cardiac toxicity in early breast cancer: addressing the unresolved issues. Circulation. 2012;126:2749-2763. 67. Gielen S, Schuler G, Adams V. Cardiovascular effects of exercise training: molecular mechanisms. Circulation. 2010;122:1221-1238. 68. Jones LW, Eves ND, Haykowsky M, et al. Exercise intolerance in cancer and the role of exercise therapy to reverse dysfunction. Lancet Oncol. 2009;10:598-605.

71. Eisele JC, Schaefer IM, Randel Nyengaard J, et al. Effect of voluntary exercise on number and volume of cardiomyocytes and their mitochondria in the mouse left ventricle. Basic Res Cardiol. 2008;103:12-21. 72. Kavazis AN, McClung JM, Hood DA, et al. Exercise induces a cardiac mitochondrial phenotype that resists apoptotic stimuli. Am J Physiol Heart Circ Physiol. 2008;294:H928-H935. 73. Jones LW, Habel LA, Weltzien E, et al. Exercise and risk of cardiovascular events in women with nonmetastatic breast cancer. J Clin Oncol. 2016;34:2743-2749. 74. Jones LW, Liu Q, Armstrong GT, et al. Exercise and risk of major cardiovascular events in adult survivors of childhood Hodgkin lymphoma: a report from the childhood cancer survivor study. J Clin Oncol. 2014;32:3643-3650. 75. Lakoski SG, Jones LW, Krone RJ, et al. Autonomic dysfunction in early breast cancer: Incidence, clinical importance, and underlying mechanisms. Am Heart J. 2015;170:231-241. 76. Scott JM, Jones LW, Hornsby WE, et al. Cancer therapy-induced autonomic dysfunction in early breast cancer: implications for aerobic exercise training. Int J Cardiol. 2014;171:e50-e51. 77. Courneya KS, Segal RJ, Mackey JR, et al. Effects of aerobic and resistance exercise in breast cancer patients receiving adjuvant chemotherapy: a multicenter randomized controlled trial. J Clin Oncol. 2007;25:4396-4404. 78. Dolan LB, Campbell K, Gelmon K, et al. Interval versus continuous aerobic exercise training in breast cancer survivors--a pilot RCT. Support Care Cancer. 2016;24:119-127. 79. Hornsby WE, Douglas PS, West MJ, et al. Safety and efficacy of aerobic training in operable breast cancer patients receiving neoadjuvant chemotherapy: a phase II randomized trial. Acta Oncol. 2014;53:6574. 80. Jones LW, Hornsby WE, Freedland SJ, et al. Effects of non-linear aerobic training on erectile function and cardiovascular function in men following prostatectomy for clinically-localized prostate cancer. Eur Urol. 2014;65:852-855. 81. Leon AS, Gaskill SE, Rice T, et al. Variability in the response of HDL cholesterol to exercise training in the HERITAGE Family Study. Int J Sports Med. 2002;23:1-9. 82. Courneya KS, McKenzie DC, Mackey JR, et al. Effects of exercise dose and type during breast cancer chemotherapy: multicenter randomized trial. J Natl Cancer Inst. 2013;105:1821-1832. 83. Jones LW, Douglas PS, Khouri MG, et al. Safety and efficacy of aerobic training in patients with cancer who have heart failure: an analysis of the HF-ACTION randomized trial. J Clin Oncol. 2014;32:2496-2502. 84. Huang C, Zhang X, Ramil JM, et al. Juvenile exposure to anthracyclines impairs cardiac progenitor cell function and vascularization resulting in greater susceptibility to stress-induced myocardial injury in adult mice. Circulation. 2010;121:675-683.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 693

JONES, EVES, AND SCOTT

85. Jones LW, Douglas PS, Eves ND, et al. Rationale and design of the Exercise Intensity Trial (EXCITE): a randomized trial comparing the effects of moderate versus moderate to high-intensity aerobic training in women with operable breast cancer. BMC Cancer. 2010;10:531.

96. Yamauchi M, Lochhead P, Imamura Y, et al. Physical activity, tumor PTGS2 expression, and survival in patients with colorectal cancer. Cancer Epidemiol Biomarkers Prev. 2013;22:1142-1152.

86. Holmes MD, Chen WY, Feskanich D, et al. Physical activity and survival after breast cancer diagnosis. JAMA. 2005;293:2479-2486.

97. Morikawa T, Kuchiba A, Yamauchi M, et al. Association of CTNNB1 (betacatenin) alterations, body mass index, and physical activity with survival in patients with colorectal cancer. JAMA. 2011;305:1685-1694.

87. Friedenreich CM, Neilson HK, Farris MS, et al. Physical activity and cancer outcomes: a precision medicine approach. Clin Cancer Res. 2016;22:4766-4775.

98. Meyerhardt JA, Ogino S, Kirkner GJ, et al. Interaction of molecular markers and physical activity on mortality in patients with colon cancer. Clin Cancer Res. 2009;15:5931-5936.

88. Ballard-Barbash R, Hunsberger S, Alciati MH, et al. Physical activity, weight control, and breast cancer risk and survival: clinical trial rationale and design considerations. J Natl Cancer Inst. 2009;101:630643.

99. Bonanni B, Puntoni M, Cazzaniga M, et al. Dual effect of metformin on breast cancer proliferation in a randomized presurgical trial. J Clin Oncol. 2012;30:2593-2600.

89. Chlebowski RT, Reeves MM. Weight loss randomized intervention trials in female cancer survivors. J Clin Oncol. 2016;34:4238-4248. 90. Goodwin PJ, Ambrosone CB, Hong CC. Modifiable lifestyle factors and breast cancer outcomes: current controversies and research recommendations. Adv Exp Med Biol. 2015;862:177-192. 91. Courneya KS, Booth CM, Gill S, et al. The Colon Health and Life-Long Exercise Change trial: a randomized trial of the National Cancer Institute of Canada Clinical Trials Group. Curr Oncol. 2008;15:279-285. 92. Movember Foundation. Intense Exercise for Survival Among Men With Metastatic Castrate-Resistant Prostate Cancer (INTERVAL), ClinicalTrials.gov identifier NCT02730338. https://clinicaltrials.gov/ ct2/show/NCT02730338. Accessed February 13, 2017. 93. Irwin ML, Smith AW, McTiernan A, et al. Influence of pre- and postdiagnosis physical activity on mortality in breast cancer survivors: the health, eating, activity, and lifestyle study. J Clin Oncol. 2008;26:3958-3964. 94. Jones LW, Kwan ML, Weltzien E, et al. Exercise and prognosis on the basis of clinicopathologic and molecular features in early-stage breast cancer: the LACE and Pathways studies. Cancer Res. 2016;76:54155422. 95. Sternfeld B, Weltzien E, Quesenberry CP Jr, et al. Physical activity and risk of recurrence and mortality in breast cancer survivors: findings from the LACE study. Cancer Epidemiol Biomarkers Prev. 2009;18:8795.

694 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

100. DeCensi A, Puntoni M, Gandini S, et al. Differential effects of metformin on breast cancer proliferation according to markers of insulin resistance and tumor subtype in a randomized presurgical trial. Breast Cancer Res Treat. 2014;148:81-90. 101. Nkondjock A, Robidoux A, Paredes Y, et al. Diet, lifestyle and BRCArelated breast cancer risk among French-Canadians. Breast Cancer Res Treat. 2006;98:285-294. 102. King MC, Marks JH, Mandell JB; New York Breast Cancer Study Group. Breast and ovarian cancer risks due to inherited mutations in BRCA1 and BRCA2. Science. 2003;302:643-646. 103. Kraus WE, Houmard JA, Duscha BD, et al. Effects of the amount and intensity of exercise on plasma lipoproteins. N Engl J Med. 2002;347:1483-1492. 104. Guerrero-Zotano A, Mayer IA, Arteaga CL. PI3K/AKT/mTOR: role in breast cancer progression, drug resistance, and treatment. Cancer Metastasis Rev. 2016;35:515-524. 105. Neufer PD, Bamman MM, Muoio DM, et al. Understanding the cellular and molecular mechanisms of physical activity-induced health benefits. Cell Metab. 2015;22:4-11. 106. Zhang QB, Zhang BH, Zhang KZ, et al. Moderate swimming suppressed the growth and metastasis of the transplanted liver cancer in mice model: with reference to nervous system. Oncogene. 2016;35:4122-4131. 107. Enriquez-Navas PM, Kam Y, Das T, et al. Exploiting evolutionary principles to prolong tumor control in preclinical models of breast cancer. Sci Transl Med. 2016;8:327ra24.

IMPROVING CANCER CARE THROUGH THE PATIENT EXPERIENCE

Improving Cancer Care Through the Patient Experience: How to Use Patient-Reported Outcomes in Clinical Practice Kathi Mooney, RN, PhD, FAAN, Donna L. Berry, RN, PhD, FAAN, Meagan Whisenant, RN, PhD, and Daniel Sjoberg, PhD OVERVIEW Poorly controlled symptoms are common and debilitating during cancer treatment and can affect functional status and quality of life, health care resource utilization, treatment adherence, and cancer survivorship. Historically, the patient experience, including symptoms during treatment, has not been tracked or documented in the patient health record. Measurement of patient-reported outcomes (PROs), including symptoms, is an essential component to cancer care focused on the illness impact to the patient and family. PROs can be useful at the individual level for monitoring and promoting symptom care both in the clinic and remotely and at the population level for aggregating population data for use in research and quality improvement initiatives. Implementation of PROs in cancer clinical care requires a carefully thought out process to overcome challenges related to integrating PROs into existing electronic health records and clinical work flow. Issues with implementing PRO collection may include making decisions about measurement tools, modes of delivery, frequency of measurement, and interpretation that are guided by a clarification of the purpose for collecting PROs. We focus on three aspects of PRO use: (1) improving care for individual patients, (2) analyzing aggregated data to improve care and outcomes overall, and (3) considerations in implementing PRO collection.

ontinuing advances in cancer treatment and targeted therapies have improved cancer survival and outcomes. However, cancer and its treatment are accompanied by distressing symptoms and serious toxicities that affect functioning and quality of life.1 Patients arrive in the therapeutic setting with varying levels of symptoms. Once cancer treatment begins, another profile of symptoms commences as toxicities and treatment-related complications develop. Symptom burden is negatively correlated with a patient’s quality of life, and distressing symptoms can persist long after treatment.2,3 Dealing with the demands of treatment and the accompanying symptoms, toxicities, and worries dominates the patient and family experience. Standard symptom care includes providing patients with a variety of prescriptions for symptom treatment and written educational materials on symptom management at the beginning of treatment and instructions to call the oncology clinic if symptoms are not well controlled. Despite this, there is evidence that patients rarely call and that symptom burden remains considerable.4 When symptoms are poorly controlled, they can result in emergency department visits, unplanned hospitalizations, delays in treatment, and lack of adherence and persistence with an effective treatment course.5-8 Symptoms commonly linger after initial treatment

is complete. Survivors requiring prolonged maintenance therapy after initial treatment, such as hormonal therapy, often discontinue their medication due to symptoms, even though it has clearly been shown to prolong disease free survival.9,10 Improving cancer outcomes requires a focus not only on the tumor but also the illness experience and its impact on patients and their families. With an increasing emphasis on value-based care rather than fee-for-service, the patient’s perspective on what brings value is central to improving outcomes.11 Measurement of outcomes, including the patient experience, is also an essential component to systematically monitor and improve care. Historically the patient experience, including symptom presence and severity, has not been systematically tracked or consistently documented in the electronic health record (EHR) in contrast to other data elements, such as laboratory values or tumor markers. Adding patient-reported outcomes (PROs) and data to routine clinical care requires substantial planning, logistics, and adjustment in care delivery practices. Technology now permits electronic capture of patient-reported symptoms, functioning, and quality of life, but adoption into routine care is slow. In a recent perspective, Basch12 identified three challenges limiting adoption: lack of integration of PRO data

From the University of Utah, Salt Lake City, UT; Hunstman Cancer Institute, Salt Lake City, UT; Harvard Medical School, Boston, MA; Dana-Farber Cancer Institute, Boston, MA; Memorial Sloan Kettering Cancer Center, New York, NY. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Kathi Mooney, RN, PhD, FAAN, 10 S. 2000 E, University of Utah, Salt Lake City, UT 84112; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 695

MOONEY ET AL

into EHR systems, lack of reimbursem*nt for implementation and monitoring PROs, and lack of effective processes for integration into the clinical care work flow. Despite current barriers that dampen adoption, a growing number of cancer care organizations are implementing electronic PRO measurement, sharing their experiences, and improving care and outcomes on the basis of the data.12 As the benefits of systematic PRO collection and integration in clinical care become more widely known, the tipping point for adoption will rapidly occur. PROs have several important uses. They can be used during the clinical encounter to intensify symptom care and improve quality of life, and they can be used to remotely monitor patients and intervene in between clinic visits. In addition, they can be aggregated into population-level data and used to guide quality improvement initiatives. Through analysis of large PRO data sets, they can also be used to provide patients with information and decision aids in choosing among treatment options or understanding the likely patient experience and recovery course of a particular treatment approach. In this article we summarize our session at the 2017 ASCO Annual Meeting on the use of PROs in clinical practice. We focus on three aspects of PRO use: (1) improving care for individual patients, (2) analyzing aggregated data to improve care and outcomes overall, and (3) considerations in implementing PRO collection.

USING PROs AT THE POINT OF CARE FOR INDIVIDUAL PATIENTS

Many cancer clinicians and researchers are aware of the importance of measuring both the tumor response as well as the individual’s experienced response. Analytic reports have emphasized the relationships between quality of life and survival outcomes. Today’s rapid expansion of genomic profiling adds another dimension to what has been termed personalized or precision medicine. Cancer care that attends to genetic risk, tumor profiles, and biologic responses, yet omits systematic assessment and treatment of the patient’s personal experience, is incomplete. Too often, the

KEY POINTS • Measurement of PROs is an essential component of cancer care. • PROs are useful at the individual level for monitoring and promoting symptom care both in the clinic and remotely. • PROs are useful at the population level for aggregating population data for use in research and quality improvement initiatives. • Issues with implementing PRO collection may include making decisions about measurement tools, modes of delivery, frequency of measurement, and interpretation that are guided by a clarification of the purpose for collecting PROs. • Clinician champions are essential to accelerate the adoption of PROs in clinical practice. 696 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

care system priorities of logistics and cost take precedence, and patient-centered care remains a frequently espoused ideal without meaningful implementation and evaluation. The first step in addressing the priority concerns of a patient treated for cancer is to assess those priorities. Since the 1960s, cancer clinicians and researchers have used various approaches to patient-reported information and data as the subjective component of a comprehensive assessment. Although there is a universal understanding that the patient’s self-appraisal does not always match the clinician’s appraisal, we still grapple with how to reconcile differing perspectives. The path to a reliable and valid patient-reported symptom or quality-of-life instrument is neither simple nor rapid. Contemporary understanding of usability, literacy, and cultural sensitivity issues demands instrument and program testing in diverse settings and populations. Patient-provider communication is required for adequate symptom management. Clinicians obtain information about patients using several methods, including physical examination, imaging, clinical chemistry, and direct questioning of the patient to obtain history and symptoms. There is considerable evidence that patients and physicians do not communicate well with respect to this last category of patient-reported data. In a study from the Memorial Sloan Kettering Cancer Center, 467 patients with breast, lung, genitourinary, or gynecologic cancer completed symptom questionnaires at a total of 4,034 clinic visits. Their reports were compared with those recorded by doctors and nurses treating those patients at the same visits as a part of standard institutional documentation. Clinicians dramatically underestimated symptom incidence. For instance, at 1 year, appetite loss was reported by about a third of patients but was documented in the case notes of fewer than one in 20.13 This study is complemented by an extensive literature. Xiao et al14 conducted a systematic review that included no fewer than 36 papers comparing physician- and patientreported symptoms in cancer and documented consistent evidence that clinicians “underestimate the incidence, severity, or distress of symptoms experienced by cancer patients.” Thus to be accurate, the patient experience, including symptoms, needs to be reported by the patient and clearly documented and tracked in the patient’s health record. Thoroughly discussing symptoms and quality-of-life issues in the face-to-face clinic visit can promote partnership between clinicians and patients,15 validate the patient’s experiences, enhance communication and satisfaction16 and reduce symptom distress.17 However, our current health system is characterized by limited face-to-face patientclinician contacts. Time constraints within the context of an exam visit and patients’ hesitancy to verbally report certain symptoms18 can result in missed or undercommunicated symptoms and quality-of-life issues of important clinical significance.19 PRO assessment prior to the actual face-to-face encounter and summarized data and graphs displaying trends over time greatly improve the likelihood that symptoms can be addressed efficiently during the visit.

IMPROVING CANCER CARE THROUGH THE PATIENT EXPERIENCE

Various strategies to enhance patient-clinician communication have been studied. Trials in the United States, Canada, Australia, and northern Europe have shown symptom and quality-of-life clinical screening with or without supportive intervention to be feasible and clinically beneficial with regard to communication and, most important, patient outcomes. Table 1 provides a summary of some of these trials. The methods of delivery vary widely, and only a minority conducted usability and feasibility testing. Several large trial20-23 interventions included a substantial component of personal contact by study nurses or coordinators, minimizing the practicality of such interventions outside of the research setting. Yet we see evidence that clinicians can readily use PRO summaries in practice,24,25 and such use results in significantly enhanced communication and improved patient experience.23,26 Questions remain, however, on the cost-effectiveness of programs in which patients monitor symptoms and quality of life, and feedback is given to clinicians who may or may not intervene appropriately.27 There is copious evidence from high-quality randomized trials that beyond being valid and feasible, integrating electronic patient-reported data in clinical care improves both care processes and care outcomes. Berry et al17,26 conducted two randomized trials in a total of 1,512 ambulatory patients starting active cancer therapy at two comprehensive cancer centers in Seattle and Boston. All patients completed online symptom questionnaires but, in the first trial, were randomly assigned to have a graphical summary of symptoms and quality-of-life concerns reported or not reported to the clinical team. The probability that a symptom was discussed during a consultation differed between groups only if the patient reported the symptom as problematic on the electronic questionnaire (p = .03), providing clear evidence that reporting of electronically gathered patient-reported data to doctors did influence the subsequent consultation. Of particular interest, there was no difference in consultation time. In other words, patient-reported data appear to improve the quality of the consultation without increasing the duration of the consultation.26 The intervention in the second randomized trial added a patient-facing intervention to the graphical clinician summary: self-monitoring between visits and communication coaching and self-management instructions tailored to each problematic symptom. Again, the intervention enhanced patient provider communication,34 and the intervention patients reported significantly less symptom distress and depression than the control group patients.17 Cleeland et al38 examined the effects of electronic patient-reported outcomes on postoperative outcomes. One hundred patients undergoing thoracotomy for lung cancer or lung metastasis, 60% of whom were aged over 60, received automatic telephone calls and completed interactive voice response system symptom reports twice weekly for 4 weeks. Using a similar approach to that of Berry et al,26 patients were then randomized to have their reports forwarded to clinical staff members or not. Patients in the experimental arm had a far greater reduction in severe symptoms

over time than controls. This was particularly apparent for the pain endpoint, with 60 severe pain events in controls compared with only 20 in patients on the intervention arm. There were also statistically significant differences between groups in symptom interference and patient satisfaction.38 The largest trial, which was recently reported by Basch et al,23 involved 766 participants undergoing chemotherapy for advanced solid tumors. Participants were randomized to electronic symptom reporting on tablets in clinic and via email from home or to routine symptom monitoring from clinicians. Health-related quality of life was measured for all patients at 6 months. In the usual care control group, 53% of patients experienced worsening of quality of life during the trial, 18% improved, and 29% were unchanged. In contrast, only 38% of patients in the intervention group had poorer quality of life at 6 months, with 34% improving and 28% unchanged (p < .001 for difference between groups). Given the high morbidity in this population, these quality-of-life differences translated to a statistically significant difference is emergency department visits. Survival was also higher in the electronic symptom reporting group, with 75.1% 1-year survival compared with 68.6% in controls. If a drug were found that could reduce mortality while improving quality of life and decreasing urgent care visits, we would consider such a drug to be standard of care. Patient-reported data can identify patients at risk for missed chemotherapy, adverse events, and even shortened survival. For example, nonadherence to oral chemotherapy or hormonal agents has been related to severity of cancer symptoms and side effects36,39 as well as demographic variables such as gender,34,40 marital status,36,40,41 and working status.31 PRO tracking systems can monitor side effects and facilitate adherence and resolution of unpleasant side effects. PRO use can also decrease inappropriate health care utilization. Symptom and quality-of-life monitoring and interventions have been shown to reduce unplanned hospital admissions and emergency department visits.23,42,43 Finally, PRO monitoring that results in improved symptom and quality-of-life outcomes may contribute to extending survival, as there is emerging evidence that depression and anxiety44,45 and fatigue46 are significant independent predictors of survival.

USING PROs FOR CLINICAL DECISION MAKING THROUGH USE OF PRO DATABASES

The adoption of systematic PRO use in clinical practice allows PRO data to be aggregated and used for clinical decision support and quality improvement initiatives, or what has been termed data integration. The theory behind data integration is that if a patient is asked a question once electronically, the response can be reused for multiple different purposes: clinical care, research, and quality assurance. The Division of Urology at the Memorial Sloan Kettering Cancer Center has played a leading role in piloting the concept of data integration. Urology patients complete electronic questionnaires about recovery of erectile and urinary function after radical prostatectomy at home via an email link asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 697

N = 286; mixed dx, 73% women, 83% metastatic in cancer specialty clinic

N = 80; mixed dx, 60% women, stage NR, in cancer specialty clinic

N = 213; breast, colorectal, lung; 67% women; 56% stage IV

N = 1,134; lung and breast cancer, 73% women; 19% stage IV in cancer specialty clinic

N = 219; mixed dx, 70% women, 32% receiving palliative therapy in community hospital clinic

N = 145; lymphoma and leukemia, 38% women in cancer specialty units and clinics

N = 200; breast cancer; 100% women; mixed stages in community hospitals

Boyes, 200630; Australia

Rosenbloom, 200721; U.S.

Carlson, 201031; Canada

Hilarius, 200832; Netherlands

Ruland, 201033; Norway

Klinkhammer-Schalke, 201220; Germany

Sample

Velikova, 2004,28 201029; U.K.

First Author, Year; Country

698 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

RCT; PR in hospital just before discharge; referral to experts for issues related to physical, psychosocial, pain and nutrition/fitness issues

RCT; PR on tablet PC in clinic or hospital unit with printed summary to MDs and RNs

Pre-post; PR in clinic on touchscreen PC with graphed results to RN

RCT; in clinic PR using touchscreen PCs, screening with triage; graphed results to MD

RCT; in clinic PR on paper, research RN interview plus info passed to clinic RN

RCT; in clinic PR using touchscreen PCs with graphed results to MD

Design; Intervention

EORTC QlQ C-30 & Breast-25

19 cancer symptom categories

PRO survey regarding whether particular issue discussed; SF-36; FACT-BCS; FACT-C; FACT-L

DT; PSSCAN

FACT-G; FLIC; brief POMS-17;MOSPSQ III

Physical symptoms; HADS; SCNS

EORTC QLQ-C30); HADS; FACT-G; MCQ

PROs

Yes

Usability Testing

Yes

Feasibility testing

Yes

Satisfaction or Acceptability Measure

Not tested

More topics discussed at fourth clinic/study visit

Not tested

Communication subscale of MCQ at 3 months

Significantly Enhanced Communication Outcomes

Continued

Better Global QOL and emotional function at 6 months; physical and emotional function at 9 months; arm symptoms at 12 months

Less discomfort, better eating/ drinking, sleep/rest and sexuality up to 1 year

None

In breast cancer, lower DT scores at 3 months

None

Better physical symptoms at second clinic/study visit

General, emotional, physical and functional well-being of FACT-G; over time 3+ months

Significantly Improved Outcomes on Specific PROs

TABLE 1. Selected Cancer Symptom and Quality-of-Life Assessment and Intervention Studies With Multisymptom PRO Outcome Evaluation in Patients With Cancer at the Point of Service During Active Curative or Palliative Treatment, 2004 to 2016

MOONEY ET AL

RCT; in clinic PR using home access or touchscreen PCs with graphed results to clinicians, plus electronic communication coaching and tailored self-care instruction

N = 752; mixed dx, 48% women, 28% stage IV in cancer specialty clinic

N = 766; mixed dx, 58% women, advanced solid tumors in cancer specialty clinic

N = 261; liver, gallbladder cancers, primary and metastatic; 27% women, 100% advanced in cancer specialty clinic

N = 1,274; mixed dx; gender % NR; stage NR; population based, provincial specialty clinics

Berry, 2014,17,34 2015,35 201536; U.S.

Basch, 201623; U.S.

Steel, 201622; U.S.

Watson, 201637; Canada

ESAS; CPC; FACT-G

CES-D; BPI; FACT-G; FACT-Fatigue; FACT-Anemia; FACT-Hep

EuroQol EQ-5D

SDS; EORTC QLQ C-30; EORTC-CIPN-20; PHQ-9; PINS; PROMIS pain interfere; skin changes

PROs

Yes

Usability Testing

Yes

Feasibility testing

Yes

Satisfaction or Acceptability Measure

Abbreviations: PRO, patient-reported outcomes; dx, diagnosis; RCT, randomized controlled trial; PR, patient report; PC, personal computer; MD, medical doctor; NR, not reported; RN, nurse.

Pre-post; paper PRO results given to RN or radiation therapist for discussion with patient; referrals made as needed

RCT; in clinic PR via interview + home access or with face—face care coordination in clinic + telephone support

RCT; in clinic PR touchscreen PCs + home access with graphed results to RN+MD and auto-alerts to nurses who provided telephone support

Design; Intervention

Sample

First Author, Year; Country

Not tested

Number of patient to provider statements for problematic symptoms and QOL

Significantly Enhanced Communication Outcomes

Fewer severe symptoms on ESAS and fewer problems on CPC reported in 10 month post period

Depression on CES-D and QOL on FACT-G at 6 months

Better QOL on EuroQol EQ-5D

Lower symptom distress on SDS and lower depression on PHQ9 at end of treatment

Significantly Improved Outcomes on Specific PROs

IMPROVING CANCER CARE THROUGH THE PATIENT EXPERIENCE

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 699

MOONEY ET AL

or in the clinic on tablets. The data are presented to surgeons at follow-up visits in the form of a report. This allows surgeons to focus the consultation on relevant aspects of patients’ recovery. Take, for instance, a patient who has recovered urinary but not erectile function. Instead of starting the consultation with a list of general questions (e.g., “Are you using pads?”, “Do you have to rush to the bathroom?”, “Are you able to get an erection?”), the surgeon is able to say, “Your urinary function seems reasonable but you seem to be having erectile dysfunction. Do you want to talk about that?” The use of electronically reported patient data in prediction modeling aids in clinical decision making. In a report provided to urologists at Memorial Sloan Kettering Cancer Center following patients after radical prostatectomy, there are several prediction models that inform clinical decision making. First, in Fig. 1A, actual patient recovery is plotted against expected recovery for the individual patient. For instance, the patient is an older man with only moderate erectile function at baseline. The graph shows that his expected erectile function was estimated using linear regression, predicting postoperative function using patients' age and erectile function before surgery. Second, we can make predictions about future progress based on a patient's progress to date. The patient, for instance could be told even at 6 months that he was unlikely to recover erectile function and that a referral to sexual medicine might be appropriate. As another example, Fig. 1B shows the life expectancy calculation based on data electronically reported by a patient with prostate cancer about his comorbidity and general health status. Patients and providers are given the probability that a man will die of other causes within 10 and 15 years and then the probability that he will die of prostate cancer, taking into account the risk for other-cause mortality.47 This life expectancy information aids in deciding whether active treatment of the patient is warranted or whether the patient’s disease is better followed through an active surveillance program. Additionally, there is increasing use of PROs for developing quality improvement initiatives focused on clinical care of symptoms and improvement of patient quality of life. PROs provide unique information about the patient’s perspective on what brings value.11 Measurement of outcomes, from the perspective of the patient, is an essential component to systematically monitoring the care provided in any institution or care setting. PROs may be useful for studying patients’ experiences with care, for assessing hospital care quality, and for developing standing methods for monitoring symptomatic adverse effects to medications.48-50 Troeschel et al51 described the use of PROs to develop symptom management quality improvement reports, demonstrating feasibility and acceptability. At a clinical care level, PROs provide valuable information about clinician symptom management effectiveness and value from the patient perspective.51,52 Importantly, PRO databases provide an efficient method for collecting and tracking patient-based data related to care effectiveness, care outcomes, and care satisfaction. 700 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

CONSIDERATIONS IN IMPLEMENTING SYSTEMATIC PRO MEASUREMENT

Successful implementation of PRO measurement and routine use in clinical practices or to track outcomes requires a number of choices and a carefully thought through process. The International Society for Quality of Life Research offers a very helpful guide for planning PRO implementation.53 To begin the process, practices should clarify the purpose for collecting PROs. For example, is it to improve individual patient care or to track outcomes using pooled data for quality improvement? Design decisions will vary on the basis of the primary purpose. Subsequent choices, such as questionnaire selection, frequency of delivery, and immediacy of scoring, depend on a clear understanding of purposes. If both patient-level and population-level data are desired, some compromise in selection will be needed. Assessment of resources that will be needed for PRO implementation and availability of technology and technical support for patient and for staff and clinician users is also an important early consideration.54 Choosing the questionnaire(s) to use should be based on the domains to be measured, which may include symptoms, functional performance, and/or quality of life. Patient burden must be balanced with the completeness of measurement. Another consideration is whether to use generic measures that can be compared with population norms and used across cancers throughout the continuum of care. Generic measures are often used if the primary purpose is to track outcomes. If, however, the primary purpose is to improve care at the individual patient level, using disease- or treatment-specific measures is recommended because they better capture the pertinent symptoms the patient is experiencing. Knowing the expected symptoms in a particular patient population and treatment scenario also influences questionnaire choice. Patients become annoyed when they complete lengthy symptom questionnaires and yet key symptoms they find bothersome are not assessed. Symptom assessment questionnaires can be a series of single-item measures of individual symptoms or multiple items for each symptom. Single items are concise, address patient burden, and allow the greatest number of symptoms to be addressed, whereas multiple items for each symptom may be more precise and valid but reduces the number of symptoms that can be assessed because of burden. Some multi-item scales, such as the PROMIS measures, come in computerized adaptive testing formats that allow rapid assessment with fewer questions per assessment than the static version. Choosing particular questionnaires should match the focus of the desired assessment such as symptom presence, severity, frequency, or burden. Although studies have found that patient acceptability of PRO measurement is generally high, especially if they see the data being used to address their needs, questionnaire selection needs to address health literacy, readability, availability in various languages, and the quality of visual display or format.55 Patients need an explanation of the purpose of collecting PROs, especially if frequent assessments are

IMPROVING CANCER CARE THROUGH THE PATIENT EXPERIENCE

FIGURE 1. Electronic Patient-Reported Data Can Be Used Immediately by Clinicians to Aid in Medical Decision Making

(A) Expected versus observed erectile and urinary function after radical prostatectomy for an individual patient. Expected recovery is estimated using data collected from other patients with similar characteristics (e.g., similar age and baseline function). (B) Probability of death of prostate cancer and other causes using data reported by the patient regarding general health status.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 701

MOONEY ET AL

planned. PRO measurement can increase patients’ satisfaction and engagement in their care if the patients understand the purpose.56 An important consideration in implementation design is the mode of delivery. Technology advances make electronic delivery feasible, but there is cost associated with deploying tablets or installing computer kiosks or other devices to collect PROs in clinic settings. To be useful at the point of care, PRO data should be immediately available and integrated into the EHR, because it is the single source of all other patient data. EHR vendors have been slow to facilitate PRO integration, although patient portals have evolved to include some PRO questionnaires that can be pushed to patients for completion prior to clinic visits.53 Frequency of measurement is another decision point. When tracking outcomes and change over time is the primary purpose, periodic measurement is appropriate. Consideration should be given whether assessments should be based on calendar dates, such as quarterly, or timed to phases in care transition or some combination. However, when the primary purpose is improving the individual patient experience, more frequent assessment is needed. Commonly, measurement is paired with a clinic visit and can be completed at home prior to the appointment or at checkin for the visit. Collection in the clinic involves consideration of work flow and participation of front-end staff members in facilitating collection. Adoption of PRO measurement can be hampered by clinicians’ concern that it will be disruptive to work flow. Attention to integration is critical. If improving individual patient care is the objective, monitoring between clinic visits may also be considered. This overcomes several key issues in symptom management, namely, that patients may not initiate calls to clinicians about poorly controlled symptoms, and symptoms normally peak at various times during the interim period between visits and are therefore missed if measurement is only timed with a scheduled visit. Although feasible, remote monitoring does add burden to clinicians in monitoring and responding to intensify care. Severity thresholds can be set to automatically alert clinicians of poorly controlled symptoms, thereby decreasing the burden of reviewing all PRO data reported. This, combined with automated self-care coaching based on the symptom severity reported, can significantly improve symptom outcomes.4 Mooney et al4 recently reported on a clinical trial of automated home monitoring of PROs in which patients receiving chemotherapy reported daily symptom presence and severity for 11 symptoms. The intervention group immediately received automated self-care coaching based on the specific symptom pattern reported, and automated alerts were sent to a nurse practitioner who used a guideline-based decision support system to call patients and adjust care for poorly controlled symptoms. The intervention group had significantly less symptom severity across all symptoms (p < .001), with a symptom reduction burden of nearly 43% compared with usual care. Examining days when participants reported one or more severe or one or more moderate symptoms, intervention participants had 702 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

67% fewer severe days and 39% fewer moderate days compared with the usual care group (p < .001 for both). As telehealth approaches become more widespread, remote PRO monitoring may extend care beyond the walls of the health systems to patients and families at home. Scoring and interpretation of PROs also requires substantial planning when used for individual patient care.57 Immediacy of the data and scoring is imperative. Integration in the EHR is ideal. Protocols must be designed to clarify who will receive the reports, what are clinically actionable thresholds, who will be responsible for follow-up, and whether any automatic referrals are generated. The design of reports is also important. Use will improve if interpretation is easy and concisely addresses and displays the data. For example, will only numeric scores be presented, or can they be accompanied by simple visual graphs to clearly spot out-of-norm values and trends over time? Other considerations include whether a copy of the data will be provided to the patient as a part of the care planning and to engage the patient in self-care. A final consideration is whether guideline-based decision support recommendations should be provided to clinicians so that they can efficiently take the next steps to improve care for poorly controlled symptoms and quality-of-life concerns.58 Measurement of PROs alone with not improve patient outcomes unless clinicians act on the data.59 Clinical champions are essential to create enthusiasm and accelerate the adoption process. Involvement of clinicians and staff members in thinking through the many decisions and designing and adapting processes to fit with work flow and clinic characteristics is exceedingly important. Greater value will be gained beyond the individual patient level, by involving clinicians in examining aggregated data and generating quality improvement initiatives as needed. Ongoing clinical analysis of the real-world patient experience of cancer and its treatment through PROs is an important component for a rapid learning system to improve cancer care.60

CONCLUSION

There is considerable evidence that patient-reported data are poorly documented by clinicians. Collection of patient-reported data using electronic tools has been shown to be accurate and feasible in both the clinical and research settings and has been demonstrated, in randomized trials, to improve both quality of life and mortality endpoints. An added benefit of collecting patient-reported data is the documentation of the patient perspective on care endpoints, which then can be used to direct quality improvement initiatives. Collection of PROs is now feasible and generally well accepted by patients. It is consistent with a patient-centered philosophy and a value-based care framework. Systematic PRO collection, integration in the EHR, and use of the data to improve care now provide both a broader and richer approach to evaluating cancer outcomes. Implementation requires the commitment of resources, thoughtful planning and monitoring, and clinical champions who see the value and are willing to work through the process of adoption.

IMPROVING CANCER CARE THROUGH THE PATIENT EXPERIENCE

References 1. Esther Kim JE, Dodd MJ, Aouizerat BE, et al. A review of the prevalence and impact of multiple symptoms in oncology patients. J Pain Symptom Manage. 2009;37:715-736.

19. Kai J, Beavan J, Faull C. Challenges of mediated communication, disclosure and patient autonomy in cross-cultural cancer care. Br J Cancer. 2011;105:918-924.

2. Deshields TL, Potter P, Olsen S, et al. The persistence of symptom burden: symptom experience and quality of life of cancer patients across one year. Support Care Cancer. 2014;22:1089-1096.

20. Klinkhammer-Schalke M, Koller M, Steinger B, et al; Regensburg QoL Study Group. Direct improvement of quality of life using a tailored quality of life diagnosis and therapy pathway: randomised trial in 200 women with breast cancer. Br J Cancer. 2012;106: 826-838.

3. Deshields TL, Potter P, Olsen S, et al. Documenting the symptom experience of cancer patients. J Support Oncol. 2011;9:216-223. 4. Mooney KH, Beck SL, Wong B, et al. Automated home monitoring and management of patient-reported symptoms during chemotherapy: results of the symptom care at home RCT. Cancer Med. 2017;6:537546. 5. Barbera L, Atzema C, Sutradhar R, et al. Do patient-reported symptoms predict emergency department visits in cancer patients? A populationbased analysis. Ann Emerg Med. 2013;61:427-437.e5. 6. Eliasson L, Clifford S, Barber N, et al. Exploring chronic myeloid leukemia patients’ reasons for not adhering to the oral anticancer drug imatinib as prescribed. Leuk Res. 2011;35:626-630. 7. Land SR, Walcott FL, Liu Q, et al. Symptoms and QOL as predictors of chemoprevention adherence in NRG Oncology/NSABP Trial P-1. J Natl Cancer Inst. 2016;108:djv365. 8. Spoelstra SL, Given CW, Sikorskii A, et al. Treatment with oral anticancer agents: symptom severity and attribution, and interference with comorbidity management. Oncol Nurs Forum. 2015;42:80-88. 9. van Herk-Sukel MP, van de Poll-Franse LV, Voogd AC, et al. Half of breast cancer patients discontinue tamoxifen and any endocrine treatment before the end of the recommended treatment period of 5 years: a population-based analysis. Breast Cancer Res Treat. 2010;122:843851. 10. Simon R, Latreille J, Matte C, et al. Adherence to adjuvant endocrine therapy in estrogen receptor-positive breast cancer patients with regular follow-up. Can J Surg. 2014;57:26-32. 11. Porter ME. What is value in health care? N Engl J Med. 2010;363:24772481. 12. Basch E. Patient-reported outcomes—harnessing patients’ voices to improve clinical care. N Engl J Med. 2017;376:105-108. 13. Basch E, Jia X, Heller G, et al. Adverse symptom event reporting by patients vs clinicians: relationships with clinical outcomes. J Natl Cancer Inst. 2009;101:1624-1632. 14. Xiao C, Polomano R, Bruner DW. Comparison between patientreported and clinician-observed symptoms in oncology. Cancer Nurs. 2013;36:E1-E16. 15. Underhill ML, Sheldon LK, Halpenny B, et al. Communication about symptoms and quality of life issues in patients with cancer: provider perceptions. J Cancer Educ. 2014;29:753-761. 16. Takeuchi EE, Keding A, Awad N, et al. Impact of patient-reported outcomes in oncology: a longitudinal analysis of patient-physician communication. J Clin Oncol. 2011;29:2910-2917. 17. Berry DL, Hong F, Halpenny B, et al. Electronic self-report assessment for cancer and self-care support: results of a multicenter randomized trial. J Clin Oncol. 2014;32:199-205. 18. Jenssen BP, Mitra N, Shah A, et al. Using digital technology to engage and communicate with patients: a survey of patient attitudes. J Gen Intern Med. 2016;31:85-92.

21. Rosenbloom SK, Victorson DE, Hahn EA, et al. Assessment is not enough: a randomized controlled trial of the effects of HRQL assessment on quality of life and satisfaction in oncology clinical practice. Psychooncology. 2007;16:1069-1079. 22. Steel JL, Geller DA, Kim KH, et al. Web-based collaborative care intervention to manage cancer-related symptoms in the palliative care setting. Cancer. 2016;122:1270-1282. 23. Basch E, Deal AM, Kris MG, et al. Symptom monitoring with patientreported outcomes during routine cancer treatment: a randomized controlled trial. J Clin Oncol. 2016;34:557-565. 24. Mullen KH, Berry DL, Zierler BK. Computerized symptom and qualityof-life assessment for patients with cancer part II: acceptability and usability. Oncol Nurs Forum. 2004;31:E84-E89. 25. Basch E, Wood WA, Schrag D, et al. Feasibility and clinical impact of sharing patient-reported symptom toxicities and performance status with clinical investigators during a phase 2 cancer treatment trial. Clin Trials. 2016;13:331-337. 26. Berry DL, Blumenstein BA, Halpenny B, et al. Enhancing patientprovider communication with the electronic self-report assessment for cancer: a randomized trial. J Clin Oncol. 2011;29:1029-1035. 27. Kroenke K, Cheville AL. Symptom improvement requires more than screening and feedback. J Clin Oncol. 2016;34:3351-3352. 28. Velikova G, Booth L, Smith AB, et al. Measuring quality of life in routine oncology practice improves communication and patient well-being: a randomized controlled trial. J Clin Oncol. 2004;22:714-724. 29. Velikova G, Keding A, Harley C, et al. Patients report improvements in continuity of care when quality of life assessments are used routinely in oncology practice: secondary outcomes of a randomised controlled trial. Eur J Cancer. 2010;46:2381-2388. 30. Boyes A, Newell S, Girgis A, et al. Does routine assessment and realtime feedback improve cancer patients’ psychosocial well-being? Eur J Cancer Care (Engl). 2006;15:163-171. 31. Carlson LE, Groff SL, Maciejewski O, et al. Screening for distress in lung and breast cancer outpatients: a randomized controlled trial. J Clin Oncol. 2010;28:4884-4891. 32. Hilarius DL, Kloeg PH, Gundy CM, et al. Use of health-related qualityof-life assessments in daily clinical oncology nursing practice: a community hospital-based intervention study. Cancer. 2008;113:628637. 33. Ruland CM, Holte HH, Røislien J, et al. Effects of a computer-supported interactive tailored patient assessment tool on patient care, symptom distress, and patients’ need for symptom management support: a randomized clinical trial. J Am Med Inform Assoc. 2010;17:403-410. 34. Berry DL, Hong F, Halpenny B, et al. The electronic self report assessment and intervention for cancer: promoting patient verbal reporting of symptom and quality of life issues in a randomized controlled trial. BMC Cancer. 2014;14:513.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 703

MOONEY ET AL

35. Berry DL, Blonquist TM, Patel RA, et al. Exposure to a patient-centered, Web-based intervention for managing cancer symptom and quality of life issues: impact on symptom distress. J Med Internet Res. 2015;17:e136.

49. O’Malley AJ, Zaslavsky AM, Elliott MN, et al. Case-mix adjustment of the CAHPS Hospital Survey. Health Serv Res. 2005;40(6p2): 2162-2181.

36. Berry DL, Blonquist TM, Hong F, et al. Self-reported adherence to oral cancer therapy: relationships with symptom distress, depression, and personal characteristics. Patient Prefer Adherence. 2015;9:1587-1592.

50. Kluetz PG, Chingos DT, Basch EM, et al. Patient-reported outcomes in cancer clinical trials: measuring symptomatic adverse events with the National Cancer Institute’s Patient-Reported Outcomes Version of the Common Terminology Criteria for Adverse Events (PRO-CTCAE). Am Soc Clin Oncol Educ Book. 2016;35:67-73.

37. Watson L, Groff S, Tamagawa R, et al. Evaluating the impact of provincial implementation of screening for distress on quality of life, symptom reports, and psychosocial well-being in patients with cancer. J Natl Compr Canc Netw. 2016;14:164-172. 38. Cleeland CS, Wang XS, Shi Q, et al. Automated symptom alerts reduce postoperative symptom severity after cancer surgery: a randomized controlled clinical trial. J Clin Oncol. 2011;29:994-1000. 39. Lebovits AH, Strain JJ, Schleifer SJ, et al. Patient noncompliance with self-administered chemotherapy. Cancer. 1990;65:17-22. 40. Noens L, van Lierde MA, De Bock R, et al. Prevalence, determinants, and outcomes of nonadherence to imatinib therapy in patients with chronic myeloid leukemia: the ADAGIO study. Blood. 2009;113:5401-5411. 41. Hershman DL, Kushi LH, Shao T, et al. Early discontinuation and nonadherence to adjuvant hormonal therapy in a cohort of 8,769 early-stage breast cancer patients. J Clin Oncol. 2010;28:4120-4128. 42. Berry DL, Hong F, Blonquist T, et al. Self report assessment and support for cancer symptoms: Impact on hospital admissions and emergency department visits. J Clin Oncol. 2013;31 (suppl, abstr e20552). 43. Barbera L, Sutradhar R, Howell D, et al. Does routine symptom screening with ESAS decrease ED visits in breast cancer patients undergoing adjuvant chemotherapy? Support Care Cancer. 2015;23:3025-3032. 44. Vodermaier A, Lucas S, Linden W, et al. Anxiety after diagnosis predicts lung-cancer specific and overall survival in patients with stage III non-small cell lung cancer. A population-based cohort study. J Pain Symptom Manage. Epub 2017 Jan 4. 45. Antoni MH, Jacobs JM, Bouchard LC, et al Post-surgical depressive symptoms and long-term survival in non-metastatic breast cancer patients at 11-year follow-up. Gen Hosp Psychiatry. 2017;44:16-21. 46. Hsu T, Speers CH, Kennecke HF, et al. The utility of abbreviated patientreported outcomes for predicting survival in early stage colorectal cancer. Cancer. Epub 2017 Jan 12. 47. Kent M, Vickers AJ. A systematic literature review of life expectancy prediction tools for patients with localized prostate cancer. J Urol. 2015;193:1938-1942. 48. Johnson ML, Rodriguez HP, Solorio MR. Case-mix adjustment and the comparison of community health center performance on patient experience measures. Health Serv Res. 2010;45:670-690.

704 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

51. Troeschel A, Smith T, Castro K, et al. The development and acceptability of symptom management quality improvement reports based on patient-reported data: an overview of methods used in PROSSES. Qual Life Res. 2016;25:2833-2843. 52. Tribett EL, Tun S, Winget M, et al. From PRO screening to improved wellness: A nurse-led intervention. J Clin Oncol. 2015;33 (suppl, abstr 72). 53. Aaronson N, Elliott T, Greenhalgh J, et al (eds). User’s Guide to Implementing Patient-Reported Outcomes Assessment in Clinical Practice Version 2. Milwaukee: International Society for Quality of Life Research; 2015. 54. Rose M, Bezjak A. Logistics of collecting patient-reported outcomes (PROs) in clinical practice: an overview and practical examples. Qual Life Res. 2009;18:125-136. 55. Howell D, Molloy S, Wilkinson K, et al. Patient-reported outcomes in routine cancer clinical practice: a scoping review of use, impact on health outcomes, and implementation factors. Ann Oncol. 2015;26:1846-1858. 56. Kotronoulas G, Kearney N, Maguire R, et al. What is the value of the routine use of patient-reported outcome measures toward improvement of patient outcomes, processes of care, and health service outcomes in cancer care? A systematic review of controlled trials. J Clin Oncol. 2014;32:1480-1501. 57. Jensen RE, Rothrock NE, DeWitt EM, et al. The role of technical advances in the adoption and integration of patient-reported outcomes in clinical care. Med Care. 2015;53:153-159. 58. Hughes EF, Wu AW, Carducci MA, et al. What can I do? Recommendations for responding to issues identified by patientreported outcomes assessments used in clinical practice. J Support Oncol. 2012;10:143-148. 59. Mooney KH, Beck SL, Friedman RH, et al. Automated monitoring of symptoms during ambulatory chemotherapy and oncology providers’ use of the information: a randomized controlled clinical trial. Support Care Cancer. 2014;22:2343-2350. 60. Abernethy AP, Etheredge LM, Ganz PA, et al Rapid-learning system for cancer care. J Clin Oncol. 2010;28:4268-4274.

PAIN AND OPIOIDS IN CANCER CARE

Pain and Opioids in Cancer Care: Benefits, Risks, and Alternatives Mike Bennett, MD, FRCP, FFPMRCA, Judith A. Paice, PhD, RN, and Mark Wallace, MD OVERVIEW Pain remains common in the setting of malignancy, occurring as a consequence of cancer and its treatment. Several high-quality studies confirm that more than 50% of all patients with cancer experience moderate to severe pain. The prevalence of pain in cancer survivors is estimated to be 40%, while close to two-thirds of those with advanced disease live with pain. Progress has occurred in the management of cancer pain, yet undertreatment persists. Additionally, new challenges are threatening these advances. These challenges are numerous and include educational deficits, time restraints, and limited access to all types of care. New challenges to access are occurring as a result of interventions designed to combat the prescription drug abuse epidemic, with fewer clinicians willing to prescribe opioids, pharmacies reluctant to stock the medications, and payers placing strict limits on reimbursem*nt. A related challenge is our evolving understanding of the risks of long-term adverse effects associated with opioids. And reflective of the opioid abuse epidemic affecting the general population, the potential for misuse or abuse exists in those with cancer. Guidelines have been developed to support oncologists when prescribing the long-term use of opioids for cancer survivors. The challenges surrounding the use of opioids, and the need for safe and effective alternative analgesics, are leading to intense interest in the potential benefits of cannabis for cancer-related pain. Oncologists are faced with questions regarding the types of cannabis available, differences between routes of administration, data concerning safety and efficacy, and legal and regulatory dynamics.

ain remains a disturbingly common consequence of cancer and its treatment. In a large study of more than 5,000 adults with cancer, 56% suffered moderate to severe pain at least monthly.1 A large systematic review of 52 studies confirmed this high prevalence, with 53% of people at all stages of cancer experiencing pain.2 Although there is evidence that the management of cancer pain has improved, undertreatment remains common and new challenges are threatening the fragile progress that has previously been made.3 These challenges are numerous and include educational deficits, time restraints, and limited access to all types of care.4 Comprehensive cancer pain management includes a thorough assessment along with the use of pharmacologic, nonpharmacologic, integrative, and interventional therapies.5 Reimbursem*nt for many of these therapies is limited, particularly for nonpharmacologic techniques such as mental health counseling, physical or occupational therapy, massage, and integrative medicine. As a result, access to cancer pain management is often restricted to pharmacologic therapies. Opioids are the mainstay of this pharmacologic management and are essential for those with pain from advanced disease. However, our evolving understanding of the risks of

long-term adverse effects, including the potential for misuse or abuse, raises concerns about the long-term use of opioids for cancer survivors.6 These challenges surrounding the use of opioids, and the need for safe and effective alternative analgesics, are leading to intense interest in the potential benefits of cannabis for cancer-related pain.

OPIOIDS IN ADVANCED CANCER: ACCESS, EFFICACY, AND OUTCOMES

Access

Cancer that is locally progressive or has metastasized is frequently painful. A systematic review by van den Beuken-van Everdingen showed that pain prevalence rises with disease progression and affects about 64% of patients with advanced cancer.2 About 45% of all patients with advanced cancer experience pain of moderate to severe intensity (at least 5 on a 0–10 pain rating scale).1,2 Morphine and other strong opioids are key to managing pain in advanced cancer. Since 1986, the focus of cancer pain treatment has been the use of strong opioids based on the World Health Organization’s (WHO’s) “analgesic ladder.”7 Globally, 8.2 million people die of advanced cancer each year, and the WHO estimates that around 6 million of

From the Institute of Health Sciences, School of Medicine, University of Leeds, Leeds, United Kingdom; Division of Hematology-Oncology, Feinberg School of Medicine, Northwestern University; Chicago, IL; Department of Anesthesiology, University of California, San Diego, San Diego, CA. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Judith A. Paice, PhD, RN, Feinberg School of Medicine, Northwestern University, 676 North St. Clair St., Suite 850, Chicago, IL 60611; email: j-paice@ northwestern.edu. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 705

BENNETT, PAICE, AND WALLACE

these patients have inadequate or no access to strong opioids largely because there has been no increase in availability of opioids for decades in the world’s poorest but most populated countries.8,9 This is largely because of government regulations that restrict supply and access.10 Even in developed countries in which there is good access to opioids, at least 32% of patients with cancer are undertreated for their pain.3 There is understandable concern about abuse of prescription opioids in contexts other than advanced disease, and there is increased tightening of prescribing regulations for opioids in the United States in particular.11 This restrictive attitude toward opioids should not be allowed to exacerbate the existing undertreatment of pain in advanced cancer.12 Retrospective cohort analyses estimate that only 43%– 48% of U.K. patients with cancer receive a strong opioid before their death,13,14 though this might be as high as 60% in Norway.15 Ziegler et al14 demonstrated that median time between initiation of strong opioids and death for 6,080 patients was 9 weeks, with increasing age associated with significantly later initiation of treatment, consistent with other studies.13,15 Patients who died in the hospital were less likely to be prescribed a strong opioid while at home, compared with those who died in hospice, and were more likely to commence strong opioids late. These variations were not explained by cancer type, duration of disease, or socioeconomic deprivation. This suggests that poor pain control at home may result in admission to and subsequent death in the hospital. Therefore, earlier pain assessment might lead to improved access to opioids and improved outcomes for patients. Even in developed countries, patients with cancer appear to access strong opioids relatively late in their disease. One methodological issue with these epidemiologic data are that they cannot be matched with individual patient-reported pain data. This means that it remains uncertain whether

KEY POINTS • Cancer pain is prevalent throughout the disease trajectory, yet undertreatment continues to be a significant problem. • Clinical experience, research, and systematic reviews all demonstrate the efficacy of opioids in relieving cancer pain, particularly in the setting of advanced disease. • Despite efficacy in relieving cancer pain, the long-term use of opioids is associated with previously unrecognized adverse effects, including endocrinopathy, neurotoxicity, sleep-disordered breathing, and, in some circ*mstances, misuse or abuse. • In an aim to provide comfort, improve function, and limit adverse effects, multimodal interventions that include pharmacologic and nonpharmacologic therapies are needed to treat cancer pain. • Although not recommended as first-line therapy, cannabis may be considered as an adjuvant analgesic in the management of refractory cancer pain. 706 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

this pattern of opioid access closely matches the patients’ onset of pain before death or highlights undertreatment. The latter seems more likely based on known epidemiology of duration of cancer pain in large cohorts.1

Efficacy

How effective are strong opioids for patients with cancer pain? Initial observational studies that evaluated the WHO ladder suggested that this approach could control pain in around 73% of patients with cancer.16,17 In absolute terms, one randomized trial that compared morphine with oxycodone in patients with cancer pain showed that both strong opioids provided good pain control in 75% of patients.18 Both strong opioids produced approximately a 3-point mean reduction on a 0–10 pain rating scale at a group level, although these data were not compared with response to placebo. There were no differences in adverse effects. A meta-analysis of clinical trials of strong opioids has provided more detailed and comparative data. The National Institute for Health and Clinical Excellence (NICE) in the United Kingdom published a detailed meta-analysis confirming that there were no significant differences in efficacy between morphine, oxycodone, transdermal fentanyl, and transdermal buprenorphine.19 Specifically, there were no differences in efficacy or adverse effects between morphine and oxycodone—probably the most frequently prescribed strong opioids for cancer pain globally.20 Overall, there were no differences in burden of adverse effects across all strong opioids, though transdermal opioids were significantly less likely to cause constipation than oral opioids (odds ratio 0.43; p < .002). Another recently published randomized trial directly compared these same four strong opioids (morphine, oxycodone, transdermal fentanyl, and transdermal buprenorphine). Corli et al21 showed no differences in efficacy between all opioids (all of which produced approximately a 3-point mean reduction on a 0-10 pain rating scale at a group level) and, interestingly, showed no differences in prevalence of constipation. The only significant differences occurred between morphine and fentanyl in the incidence of hallucinations (13.2% vs. 2.4%; p = .001) and severe confusion (15.5% vs. 6.3%; p = .018) that favored transdermal fentanyl. In summary, strong opioids are very effective interventions for cancer pain resulting in a 75% response rate and reducing average pain intensity from 6 to 3 on a 0–10 pain scale.20,21 When compared with the early evaluations of the WHO analgesic ladder, these more recent data imply that the effectiveness of the WHO approach is based entirely on strong opioids, with no substantial contribution from other approaches. These studies underpin international guidance on strong opioids for cancer pain that advises first-line treatment with either morphine, oxycodone, transdermal fentanyl, or transdermal buprenorphine based on efficacy.5,22 In the United Kingdom, NICE guidance recommends morphine as a first-line opioid because of its substantially cheaper cost.19

PAIN AND OPIOIDS IN CANCER CARE

Outcomes

What is a meaningful outcome for a patient with cancer pain? Understandably, patients express that they want to be pain free, although, in general, they do not actually expect their pain to be relieved completely.23 Bender et al24 identified that patients are keen to understand the cause of their cancer pain, what to expect, options for pain control (including addressing concerns about strong opioids), and how to cope with cancer pain including talking with others and finding help. A number of qualitative studies have revealed that patients seem to determine whether their pain is controlled by whether they can perform activities or tasks and maintain relationships with family or friends.24,25 To perform these activities, patients frequently try to reduce interference from both pain and the cognitive effects of analgesia to maintain as much function as possible.26 This commonly leads to trade-offs between pain and analgesia, impacting medication adherence. The concept of trading off has not been well described in medical literature, but it is clearly seen as important by patients in reaching outcomes that are meaningful for them. For this reason, clinicians should seek to understand patient preferences for cancer pain management when initiating and managing strong opioids. Key priorities for clinicians regarding pain management strategies for patients with advanced cancer should be to help them achieve a balance between pain and adverse effects of analgesia to optimize physical function and to provide support for self-management.27,28 Overly simplifying these important outcomes to a numerical rating of pain intensity is likely to be poorly sensitive (patients may be content with the balance of their pain management yet report higher pain scores) or poorly specific (patients may judge their pain control unsatisfactory despite lower pain scores because they struggle with opioid adverse effects, which severely limits their function). A focus on determining interference from pain or analgesia in daily activities and understanding the degree of self-efficacy (ability to cope) are more important measures of successful pain management.

OPIOIDS IN CANCER SURVIVORS: BENEFITS, RISKS, AND CHALLENGES

Currently, there are approximately 15.5 million cancer survivors in the United States, and this number is expected to grow to 26.1 million by 2040.29 Of these individuals, more than two-thirds have lived 5 years or more after diagnosis, and 44% have survived 10 years or more.29 Much of these impressive figures in survivorship are because of extraordinary advances in the development of more effective cancer therapies. Unfortunately, many of these highly effective treatments also lead to persistent pain syndromes. As a result, studies suggest the prevalence of pain in cancer survivors may be 40% or higher.2,30,31

Benefits and Risks

Although opioids have a clear and primary role in the care of pain associated with advanced disease, their role in relieving

pain in cancer survivors is less apparent. The recent ASCO Clinical Practice Guideline Management of Chronic Pain in Survivors of Adult Cancer outlines the results of a systematic review of studies investigating chronic pain management in cancer survivors.6 One systematic review of randomized controlled trials of opioids for relief of cancer pain found few high-quality, long-term trials.32 In a large study of more than 500 subjects randomly assigned to receive one of four opioids for 28 days, the worst and average pain intensity decreased over 4 weeks with no significant differences between drugs.21 Changes in therapy, including dose escalation, switches to other opioids, and the addition of adjuvant analgesics, were common, and close to 15% of patients were nonresponders.21 Although clinical experience suggests that select patients may obtain safe and effective pain control with opioids, there are no studies that guide clinicians as they consider a trial of opioids in cancer survivors. Aligned with unclear long-term efficacy is an increasing awareness of the adverse effects associated with prolonged use of opioids (Sidebar). Mental clouding, effects on libido and fertility, hyperalgesia, and sleep disorders can all affect employment, relationships, and overall quality of life.33-35 Provocative and troubling early data from laboratory models suggest that opioids may affect immune function36,37 and tumor progression,38,39 yet it is too early to determine if these findings are clinically meaningful. Of particular concern in the face of the current opioid abuse epidemic is that cancer survivors treated with opioids may also develop opioid or other substance abuse disorders as has been documented in people with chronic noncancer pain.11,40,41

Risk Mitigation

Methods to mitigate the risk of harm include careful assessment and awareness of adherence monitoring. Screening tools are available to determine risk of misuse, although none have been validated in an oncology population to date. Key factors that have been found to be associated with opioid misuse/abuse in those with a noncancer diagnosis include male sex, age younger than 65, opioid misuse history, depression, family history of substance use disorders, current smoking, past or current incarceration, and posttraumatic stress disorder.42-45 Adherence monitoring may include use of a controlled substance agreement, review of data from prescription drug monitoring programs, periodic drug testing, pill counts, and education.46-51 After review of assessment data and information obtained from urine drug screening and a prescription drug monitoring program, a decision is made to prescribe based on risk stratification. If risk is low and the pain warrants use of an opioid, the oncologist may decide to prescribe. If the risk of abuse is moderate or high, the oncologist must decide if the severity of the pain is seriously affecting the patient’s physical or mental well-being and if there are reasonable alternatives. If the effect of pain is severe and there are no other reasonable alternatives, and the risk of abuse and/or diversion is manageable, a trial of opioids may be considered. Regardless of level of asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 707

BENNETT, PAICE, AND WALLACE

SIDEBAR. Adverse Effects Associated With Long-Term Opioid Use6,33-35 • Constipation • Mental clouding • Upper gastrointestinal symptoms (pyrosis, nausea, bloating) • Endocrinopathy (hypogonadism/hyperprolactinemia) • Fatigue • Infertility • Osteoporosis/osteopenia • Reduced libido • Reduced frequency/duration or absence of menses • Neurotoxicity • Myoclonus • Other changes in mental status (including mood effects, memory problems, increased risk of falls in elderly patients) • Risk of opioid-induced hyperalgesia (incidence and phenomenology uncertain, but escalating pain in tandem with dose-escalation raises concern) • Sleep-disordered breathing • Increased risk of concurrent benzodiazepine in patients predisposed to sleep apnea • New-onset sleep apnea • Worsening of sleep apnea syndromes risk, nonopioid and nonpharmacologic therapies should always be optimized. If opioids are prescribed, adherence monitoring should continue, with the frequency dictated by the level of risk (Table 1).6 If opioids are ineffective or serious adverse effects occur, careful reconsideration of therapy must occur. Given the severity of the opioid misuse/abuse epidemic, oncology clinicians must be attentive to the potential for diversion, and those with cancer may be targeted as potential sources of prescription drugs. Education about safe storage and disposal has been shown to increase awareness and improve safe practice by patients.52

CANNABINOIDS IN CANCER PAIN MANAGEMENT

Over the past 20 years, the public and medical community’s attitude toward cannabinoids has been shifting. Although it remains illegal federally, over half of the U.S. states have legalized the medicinal use, with a handful legalizing recreational use. With this changing landscape comes many challenging questions from oncology providers and patients: • What role do cannabinoids play in alleviating pain? • Should physicians recommend cannabinoids for the treatment of pain, particularly pain related to cancer? • What are the risks and side effects? • How do patients obtain and use a cannabinoid drug? • How should physicians who choose to recommend 708 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

cannabinoids select appropriate patients and monitor them? • What are the legal and regulatory issues that providers and researchers face in dealing with cannabinoids?

Cannabis Content

The cannabis plant contains over 400 chemical compounds, of which at least 80 are cannabinoids.53 Delta-9tetrahydrocannabinol (Δ9-THC) is the most well-known and the primary psychoactive compound in cannabis.54-56 It mimics the action of anandamide, an endogenous cannabinoid in humans, having approximately equal affinity for the CB1 and CB2 receptors. Cannabidiol (CBD) is the second most abundant compound in cannabis after THC.55 It is thought to have wider medical applications than THC, which has fueled the demand for medical marijuana and pharmaceuticals with higher CBD concentrations compared with recreational marijuana.57 CBD is generally considered to have no psychoactive effects, but clinically it has been reported to reduce seizures, improve muscle spasm, and reduce inflammation.54,56 It has a very low affinity for the CB1 and CB2 receptors and may act as an inverse agonist/antagonist.54,55 These interactions with the CB receptors may attenuate some of the psychotropic effects of THC.54 In addition to whole-leaf cannabis plants, there is another class of phytocannabinoids collectively referred to as cannabis-based medicine extracts. These are derived by extracting compounds directly from cannabis plants. There are two subtypes of cannabis-based medicine extracts: (1) those produced by pharmaceutical companies under well-regulated, controlled conditions and undergoing rigorous clinical trials and (2) those produced and sold in medical marijuana dispensaries without any regulatory oversight or clinical trials.

Pharmacology of Cannabinoids

The precise pharmacology of most cannabinoids remains unknown. However, researchers have elicited major mechanisms underlying the active compounds in cannabis including THC, CBD, and cannabinol. Complicating this is the highly variable absorption because of the plethora of delivery forms and routes. These include inhalation and ingestion as well as absorption via oral, sublingual, topical, or rectal application.58 For centuries, the primary means of delivering cannabinoids has been via the inhaled smoke of cannabis or hashish. The variable concentration of THC and other cannabinoids in cannabis, lack of controlled production and testing in most medical marijuana products, and the diversity of delivery routes makes prediction of pharmacologic effects difficult.56 In general, the inhalation of cannabis results in a fast predictable plasma concentration of cannabinoids that is short lived allowing for fine titration to affect. Ingestion results in a delayed, variable peak plasma concentration that is more prolonged. Transmucosal delivery results in peak plasma levels similar to ingestion but more rapid and of shorter duration. These different delivery methods can be clinically

PAIN AND OPIOIDS IN CANCER CARE

used depending on the situation. For example, most patients prefer ingestion at night for the prolonged effect, while inhalation is the preferred method during the daytime (Wallace M, personal experience).

Cannabinoid/Opioid System Interactions

With the prescription opioid overdose crisis in the United States, there is concern over the increasing use of medicinal cannabis and its effects on this crisis. Studies have demonstrated that states with medicinal cannabis legalization have actually seen a reduction in opioid analgesic overdose.59 A

recent retrospective cross-sectional survey of patients with chronic pain using medicinal cannabis showed a 64% decreased opioid use, decreased side effects of medications, and an improved quality of life.60 In another study of cannabinoid-opioid interaction, 21 subjects with chronic pain taking twice-daily sustained-release morphine or oxycodone inhaled vaporized cannabis three times daily for 5 days. This resulted in a 27% reduction in pain with no altered plasma opioid levels. Pulse oximetry did not show any lowered oxygen saturation suggesting that cannabinoids do not worsen opioid-induced respiratory depression.61

TABLE 1. Risk Stratification and Adherence Monitoring6 Action

Low Risk

Moderate Risk

High Risk

Risk stratification*

No history of alcohol abuse or drug abuse, no family history of alcohol or drug abuse

Remote history of alcohol or drug abuse

Recent history, or multiple episodes, of alcohol or drug abuse

No history of a major psychiatric disorder

History of addiction with a sustained period of recovery and a strong system to help sustain recovery

History of addiction with limited or no system to sustain recovery

Older age

Questionable family history of alcohol or drug abuse

Strong family history of alcohol or drug abuse

No smoking

History of major psychiatric disorder that has been managed effectively

History of major psychiatric disorder

Stable social support

Younger age Smoking History of physical or sexual abuse Lack of social support Involvement with others engaging in drug abuse

Adherence monitoring and mitigation

At least annual adherence monitoring

At least semiannual adherence monitoring (more frequent at higher levels of assessed risk)

Adherence monitoring at least every 2–3 months and more frequent visits

Monitoring should usually include:

•D etailed interviewing about drugrelated behavior

• Detailed interviewing about drugrelated behavior

• Detailed interviewing about drug-related behavior

•Q uestioning of family members and record review from other treating physicians

• Questioning of family members and record review from other treating physicians

•C heck of prescription monitoring program

• Check of prescription monitoring program

• Urine drug screen

• Pill counts

Reconsideration of treatment to determine whether nonopioid therapies can be better used

Reconsideration of treatment to determine whether nonopioid therapies can be better

Reconsideration of treatment to determine whether nonopioid therapies can be better used

• Urine drug screen Respond to aberrant behaviors

Refills limited or not permitted Small frequent prescriptions No concurrent use of more than one opioid (e.g., no prescription of a second short-acting drug for breakthrough pain in those prescribed a long-acting drug for daily use) Mandated consultation with addiction specialists/psychiatrist *The level of risk conferred is indicated by the presence of one or more factors itemized in the corresponding risk categories.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 709

BENNETT, PAICE, AND WALLACE

Cannabinoids as Analgesics

There are a limited number of randomized controlled trials involving cannabinoids for the treatment of pain. Stimulated by the burden of chronic pain globally and the need to find safer, nonopioid therapeutic targets, the number of studies has been rising. Complicating this area of research, however, are complex federal regulatory issues because of the Schedule I status of cannabinoids and the lack of standards for cannabinoid form and administration across various studies. The studies differ in the type of cannabinoid (i.e., plant, extract, synthetic), route of administration (i.e., inhalation, ingestion, mucosal absorption), and dosing that create unique challenges in interpretation. All of the current studies have focused on THC. There are no studies focusing on CBD, although there is increasing interest given its lack of psychoactivity. To date, all of the cannabis supplied by the National Institutes of Health that has been used in current studies had CBD levels of less than 1%. A summary of select randomized controlled trials across several pain conditions is highlighted in Table 2.61-75

Risks and Side Effects of Cannabinoids

As with any potential therapy, cannabinoids carry risks and adverse effects. The most common cannabinoid side effects include sedation, dizziness, dry mouth, and dysphoria. Other significant side effects include cognitive impairment, anxiety, and psychosis. It is important to note that most of the published side effects of cannabis and cannabinoids come from the study of their recreational use. A recent study of cannabis for the treatment of chronic pain had no more adverse effects than matched controls.77

The abuse potential of cannabis is controversial. Although cannabis abuse is prevalent, animal studies show that cannabinoids do not seem to be as robust as other agents (e.g., heroin, cocaine, nicotine).78 There appears to be opposing effects of high- and low-dose THC, with high-dose producing aversion and low-dose producing pleasure.79 This therapeutic window has been demonstrated in human studies.80 Plasma levels of THC between 5 and 15 ng/mL appear to be therapeutic for pain relief; however, this relief is lost at levels above 15 ng/mL (Wallace M et al, unpublished data). With chronic cannabis use, tolerance develops to the physiologic (i.e., cardiovascular) and subjective (i.e., high) effects, and abrupt termination in habitual users will result in withdrawal symptoms similar to opioids. However, withdrawal is less likely to occur or is associated with fewer symptoms when the dose of THC consumed is low.81,82

Regulatory, Professional, and Legal Considerations

Possession and use of cannabis remains illegal under U.S. federal law. Since 1970, cannabis has been listed by the U.S. Drug Enforcement Administration as a Schedule I drug with “high potential for abuse,” “no currently accepted medical use,” and “lack of acceptable safety for use under medical supervision."83 This is in direct contrast to its legal status within many U.S. states for medicinal and recreational use. This has created confusion for many providers and patients. Neither the U.S. Food and Drug Administration nor any other federal regulatory agency currently oversees or regulates the production and distribution of cannabis or the myriad cannabis-based products sold in medical marijuana dispensaries. Moreover, there is no national oversight, and

TABLE 2. Summary of Published Cannabis-Based Studies on Pain61-76 Pain Type

Cannabinoid Tested

Outcome

Adverse Effects

Reference

Chronic pain

THC/Cannabidiol (SL spray)

Decreased pain

Mild

Blake, 2005

Decreased pain

Mild

Notcutt, 2004

THC (SL spray)

Decreased pain

Mild

Notcutt, 2004

Cannabidiol (SL spray)

No effect

Mild

Notcutt, 2004

Cannabis (smoked)

Decreased pain (dose dependent)

Mild

Wallace, 2015

Neuropathic pain

Decreased pain (high dose)

Mild

Ware, 2010

Decreased pain (high dose)

2 cases of toxic psychosis

Ellis, 2008

Decreased pain (high dose)

Mild

Wilsey, 2008

Decreased pain

Mild

Abrams, 2007

Decreased pain

Mild

Serpell, 2014

Decreased pain

Mild

Nurmikko, 2007

Decreased pain

Mild

Rog, 2005

Decreased pain

Mild

Berman, 2004

Cannador (oral)

Decreased pain

Mild

Zajicek, 2003, 2005

Nabiximols (SL spray)

Decreased pain with low and middle dose; no effect with high dose

Mild

Portenoy, 2012

Decreased pain, nabiximols; increased pain, THC

Mild

Johnson, 2010

Nabiximols (SL spray)

Cancer pain

710 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PAIN AND OPIOIDS IN CANCER CARE

limited state regulation of the labeling, concentration, dosing, or purity of cannabis and cannabis-based products. Thus, it is often left up to the growers, processors, and distributors of medical cannabis in states where it has been legalized to self-regulate. Lastly, neither cannabis nor any of these products have undergone the large-scale clinical trials necessary for showing clear efficacy for a particular indication. Efforts are emerging to provide better oversight of herbal marijuana processing and distribution. Marijuana laws vary widely among those states that have passed some form of legalization, and each clinician must be familiar with the state in which they practice. Because marijuana is not approved by the U.S. Food and Drug Administration, no state requires a physician to write a prescription. There has been a push by the American Medication Association to enact federal legislation protecting physicians who prescribe cannabis. Some states provide guidelines for recommending medicinal cannabis. In the absence of guidelines, clinicians

who choose to recommend cannabis should manage their patients in accordance with good medical practice. This involves becoming familiar with the safety and efficacy of medical cannabis and counseling patients on their responsibilities and on the side effects. Patients should then be monitored to assess the clinical effects, adverse effects, and impact on function and quality of life. Appropriate documentation in the patient’s medical record should be made.

CONCLUSION

Cancer pain remains prevalent, yet undertreatment continues, in part due to concerns regarding the use of opioids. The efficacy of opioids in advanced disease has been clearly established, however, questions remain about the safety and effectiveness of opioids in long-term survivors of cancer. As a result of challenges surrounding opioids, alternative analgesics, including cannabis, are being studied. Risks and benefits, as well as regulatory and legal issues, must be carefully considered when recommending these treatment options.

References 1. Breivik H, Cherny N, Collett B, et al. Cancer-related pain: a panEuropean survey of prevalence, treatment, and patient attitudes. Ann Oncol. 2009;20:1420-1433. 2. van den Beuken-van Everdingen MH, de Rijke JM, Kessels AG, et al. Prevalence of pain in patients with cancer: a systematic review of the past 40 years. Ann Oncol. 2007;18:1437-1449. 3. Greco MT, Roberto A, Corli O, et al. Quality of cancer pain management: an update of a systematic review of undertreatment of patients with cancer. J Clin Oncol. 2014;32:4149-4154.

13. Higginson IJ, Gao W. Opioid prescribing for cancer pain during the last 3 months of life: associated factors and 9-year trends in a nationwide United Kingdom cohort study. J Clin Oncol. 2012;30: 4373-4379. 14. Ziegler L, Mulvey M, Blenkinsopp A, et al. Opioid prescribing for patients with cancer in the last year of life: a longitudinal population cohort study. Pain. 2016;157:2445-2451.

4. Kwon JH. Overcoming barriers in cancer pain management. J Clin Oncol. 2014;32:1727-1733.

15. Fredheim OM, Brelin S, Hjermstad MJ, et al. Prescriptions of analgesics during complete disease trajectories in patients who are diagnosed with and die from cancer within the five-year period 2005-2009. Eur J Pain. 2017;21:530-540.

5. National Comprehensive Cancer Network. Adult Cancer Pain: NCCN Clinical Practice Guidelines in Oncology, Version 2.2016. https://www. nccn.org/professionals/physician_gls/pdf/pain.pdf. Accessed March 9, 2017.

16. Bennett M. What evidence do we have that the WHO analgesic ladder is effective in cancer pain? In McQuay HJ and Kelso E (eds). Systematic Reviews in Pain Research: Methodology Refined. Seattle: IASP Press; 2008.

6. Paice JA, Portenoy R, Lacchetti C, et al. Management of chronic pain in survivors of adult cancers: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol. 2016;34:3325-3345.

17. Zech DF, Grond S, Lynch J, et al. Validation of World Health Organization Guidelines for cancer pain relief: a 10-year prospective study. Pain. 1995;63:65-76.

7. World Health Organization. Cancer Pain Relief—With a Guide to Opioid Availability, 2nd ed. Geneva: World Health Organization; 1996.

18. Riley J, Branford R, Droney J, et al. Morphine or oxycodone for cancerrelated pain? A randomized, open-label, controlled trial. J Pain Symptom Manage. 2015;49:161-172.

8. World Health Organization. Impact of Impaired Access to Controlled Medications. http://www.who.int/medicines/areas/quality_safety/ Impaired_Access/en/. Accessed January 30, 2017. 9. Berterame S, Erthal J, Thomas J, et al. Use of and barriers to access to opioid analgesics: a worldwide, regional, and national study. Lancet. 2016;387:1644-1656. 10. Maurer MA, Gilson AM, Husain SA, et al. Examining influences on the availability of and access to opioids for pain management and palliative care. J Pain Palliat Care Pharmacother. 2013;27:255-260. 11. Ballantyne JC, Kolodny A. Preventing prescription opioid abuse. JAMA. 2015;313:1059. 12. Tanco K, Bruera SE, Bruera E. Insurance company denial of payment and enforced changes in the type and dose of opioid analgesics for patients with cancer pain. Palliat Support Care. 2014;12:515-518.

19. Bennett MI, Graham J, Schmidt-Hansen M, et al; Guideline Development Group. Prescribing strong opioids for pain in adult palliative care: summary of NICE guidance. BMJ. 2012;344:e2806. 20. Schmidt-Hansen M, Bennett MI, Hilgart J. Oxycodone for cancer pain in adult patients. JAMA. 2015;314:1282-1283. 21. Corli O, Floriani I, Roberto A, et al. Are strong opioids equally effective and safe in the treatment of chronic cancer pain? A multicenter randomized phase IV ‘real life’ trial on the variability of response to opioids. Ann Oncol. 2016;27:1107-1115. 22. Caraceni A, Hanks G, Kaasa S, et al; European Palliative Care Research Collaborative (EPCRC); European Association for Palliative Care (EAPC). Use of opioid analgesics in the treatment of cancer pain: evidence-based recommendations from the EAPC. Lancet Oncol. 2012;13:e58-e68.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 711

BENNETT, PAICE, AND WALLACE

23. Gibbins J, Bhatia R, Forbes K, et al. What do patients with advanced incurable cancer want from the management of their pain? A qualitative study. Palliat Med. 2014;28:71-78.

42. Boscarino JA, Rukstalis M, Hoffman SN, et al. Risk factors for drug dependence among out-patients on opioid therapy in a large US health-care system. Addiction. 2010;105:1776-1782.

24. Bender JL, Hohenadel J, Wong J, et al. What patients with cancer want to know about pain: a qualitative study. J Pain Symptom Manage. 2008;35:177-187.

43. Liebschutz JM, Saitz R, Weiss RD, et al. Clinical factors associated with prescription drug use disorder in urban primary care patients with chronic pain. J Pain. 2010;11:1047-1055.

25. Hackett J, Godfrey M, Bennett MI. Patient and caregiver perspectives on managing pain in advanced cancer: A qualitative longitudinal study. Palliat Med. 2016;30:711-719.

44. Rice JB, White AG, Birnbaum HG, et al. A model to identify patients at risk for prescription opioid abuse, dependence, and misuse. Pain Med. 2012;13:1162-1173.

26. Manzano A, Ziegler L, Bennett M. Exploring interference from analgesia in patients with cancer pain: a longitudinal qualitative study. J Clin Nurs. 2014;23:1877-1888.

45. White AG, Birnbaum HG, Schiller M, et al. Analytic models to identify patients at risk for prescription opioid abuse. Am J Manag Care. 2009;15:897-906.

27. Bennett MI, Bagnall AM, José Closs S. How effective are patientbased educational interventions in the management of cancer pain? Systematic review and meta-analysis. Pain. 2009;143:192-199.

46. Christo PJ, Manchikanti L, Ruan X, et al. Urine drug testing in chronic pain. Pain Physician. 2011;14:123-143.

28. Sheinfeld Gorin S, Krebs P, Badr H, et al. Meta-analysis of psychosocial interventions to reduce pain in patients with cancer. J Clin Oncol. 2012;30:539-547. 29. Bluethmann SM, Mariotto AB, Rowland JH. Anticipating the “silver tsunami”: prevalence trajectories and comorbidity burden among older cancer survivors in the United States. Cancer Epidemiol Biomarkers Prev. 2016;25:1029-1036. 30. Davidsen M, Kjøller M, Helweg-Larsen K. The Danish National Cohort Study (DANCOS). Scand J Public Health. 2011;39:131-135. 31. Jensen MP, Chang HY, Lai YH, et al. Pain in long-term breast cancer survivors: frequency, severity, and impact. Pain Med. 2010;11: 1099-1106.

47. Gugelmann HM, Perrone J. Can prescription drug monitoring programs help limit opioid abuse? JAMA. 2011;306:2258-2259. 48. Manchikanti L, Atluri S, Trescot AM, et al. Monitoring opioid adherence in chronic pain patients: tools, techniques, and utility. Pain Physician. 2008;11:S155-S180. 49. Manchikanti L, Manchukonda R, Damron KS, et al. Does adherence monitoring reduce controlled substance abuse in chronic pain patients? Pain Physician. 2006;9:57-60. 50. Reifler LM, Droz D, Bailey JE, et al. Do prescription monitoring programs impact state trends in opioid abuse/misuse? Pain Med. 2012;13:434-442. 51. Wang J, Christo PJ. The influence of prescription monitoring programs on chronic pain management. Pain Physician. 2009;12:507-515.

32. Koyyalagunta D, Bruera E, Solanki DR, et al. A systematic review of randomized trials on the effectiveness of opioids for cancer pain. Pain Physician. 2012;15:ES39-ES58.

52. de la Cruz M, Reddy A, Balankari V, et al. The impact of an educational program on patient practices for safe use, storage, and disposal of opioids at a comprehensive cancer center. Oncologist. 2017;22:115-121.

33. Correa D, Farney RJ, Chung F, et al. Chronic opioid use and central sleep apnea: a review of the prevalence, mechanisms, and perioperative considerations. Anesth Analg. 2015;120:1273-1285.

53. Mechoulam R, Devane WA, Breuer A, et al. A random walk through a cannabis field. Pharmacol Biochem Behav. 1991;40:461-464.

34. Dev R, Hui D, Del Fabbro E, et al. Association between hypogonadism, symptom burden, and survival in male patients with advanced cancer. Cancer. 2014;120:1586-1593. 35. Lee M, Silverman SM, Hansen H, et al. A comprehensive review of opioid-induced hyperalgesia. Pain Physician. 2011;14:145-161. 36. Ninković J, Roy S. Role of the mu-opioid receptor in opioid modulation of immune function. Amino Acids. 2013;45:9-24. 37. Sacerdote P, Franchi S, Panerai AE. Non-analgesic effects of opioids: mechanisms and potential clinical relevance of opioid-induced immunodepression. Curr Pharm Des. 2012;18:6034-6042.

54. Pertwee RG. Ligands that target cannabinoid receptors in the brain: from THC to anandamide and beyond. Addict Biol. 2008;13:147-159. 55. Appendino G, Chianese G, Taglialatela-Scafati O. Cannabinoids: occurrence and medicinal chemistry. Curr Med Chem. 2011;18: 1085-1099. 56. Borgelt LM, Franson KL, Nussbaum AM, et al. The pharmacologic and clinical effects of medical cannabis. Pharmacotherapy. 2013;33:195-209. 57. Maa E, Figi P. The case for medical marijuana in epilepsy. Epilepsia. 2014;55:783-786. 58. Baggio S, Deline S, Studer J, et al. Routes of administration of cannabis used for nonmedical purposes and associations with patterns of drug use. J Adolesc Health. 2014;54:235-240.

38. Lennon FE, Mirzapoiazova T, Mambetsariev B, et al. The Mu opioid receptor promotes opioid and growth factor-induced proliferation, migration and epithelial mesenchymal transition (EMT) in human lung cancer. PLoS One. 2014;9:e91577.

59. Bachhuber MA, Saloner B, Cunningham CO, et al. Medical cannabis laws and opioid analgesic overdose mortality in the United States, 1999-2010. JAMA Intern Med. 2014;174:1668-1673.

39. Nguyen J, Luk K, Vang D, et al. Morphine stimulates cancer progression and mast cell activation and impairs survival in transgenic mice with breast cancer. Br J Anaesth. 2014;113:i4-i13.

60. Boehnke KF, Litinas E, Clauw DJ. Medical cannabis use is associated with decreased opiate medication use in a retrospective cross-sectional survey of patients with chronic pain. J Pain. 2016;17:739-744.

40. Koyyalagunta D, Bruera E, Aigner C, et al. Risk stratification of opioid misuse among patients with cancer pain using the SOAPP-SF. Pain Med. 2013;14:667-675.

61. Abrams DI, Jay CA, Shade SB, et al. Cannabis in painful HIV-associated sensory neuropathy: a randomized placebo-controlled trial. Neurology. 2007;68:515-521.

41. Kwon JH, Tanco K, Hui D, et al. Chemical coping versus pseudoaddiction in patients with cancer pain. Palliat Support Care. 2014;12:413-417.

62. Berman JS, Symonds C, Birch R. Efficacy of two cannabis based medicinal extracts for relief of central neuropathic pain from brachial

712 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PAIN AND OPIOIDS IN CANCER CARE

plexus avulsion: results of a randomised controlled trial. Pain. 2004;112:299-306. 63. Blake DR, Robson P, Ho M, et al. Preliminary assessment of the efficacy, tolerability and safety of a cannabis-based medicine (Sativex) in the treatment of pain caused by rheumatoid arthritis. Rheumatology (Oxford). 2006;45:50-52. 64. Ellis RJ, Toperoff W, Vaida F, et al. Smoked medicinal cannabis for neuropathic pain in HIV: a randomized, crossover clinical trial. Neuropsychopharmacology. 2009;34:672-680. 65. Johnson JR, Burnell-Nugent M, Lossignol D, et al. Multicenter, doubleblind, randomized, placebo-controlled, parallel-group study of the efficacy, safety, and tolerability of THC:CBD extract and THC extract in patients with intractable cancer-related pain. J Pain Symptom Manage. 2010;39:167-179. 66. Notcutt W, Price M, Miller R, et al. Initial experiences with medicinal extracts of cannabis for chronic pain: results from 34 ‘N of 1’ studies. Anaesthesia. 2004;59:440-452. 67. Nurmikko TJ, Serpell MG, Hoggart B, et al. Sativex successfully treats neuropathic pain characterised by allodynia: a randomised, doubleblind, placebo-controlled clinical trial. Pain. 2007;133:210-220. 68. Portenoy RK, Ganae-Motan ED, Allende S, et al. Nabiximols for opioid-treated cancer patients with poorly-controlled chronic pain: a randomized, placebo-controlled, graded-dose trial. J Pain. 2012;13: 438-449. 69. Rog DJ, Nurmikko TJ, Friede T, et al. Randomized, controlled trial of cannabis-based medicine in central pain in multiple sclerosis. Neurology. 2005;65:812-819. 70. Serpell M, Ratcliffe S, Hovorka J, et al. A double-blind, randomized, placebo-controlled, parallel group study of THC/CBD spray in peripheral neuropathic pain treatment. Eur J Pain. 2014;18:999-1012.

73. Wilsey B, Marcotte T, Tsodikov A, et al. A randomized, placebocontrolled, crossover trial of cannabis cigarettes in neuropathic pain. J Pain. 2008;9:506-521. 74. Zajicek J, Fox P, Sanders H, et al; UK MS Research Group. Cannabinoids for treatment of spasticity and other symptoms related to multiple sclerosis (CAMS study): multicentre randomised placebo-controlled trial. Lancet. 2003;362:1517-1526. 75. Zajicek JP, Sanders HP, Wright DE, et al. Cannabinoids in multiple sclerosis (CAMS) study: safety and efficacy data for 12 months follow up. J Neurol Neurosurg Psychiatry. 2005;76:1664-1669. 76. GW Pharmaceuticals. GW Pharmaceuticals and Otsuka Announce Results From Two Remaining Sativex(R) Phase 3 Cancer Pain Trials. http://ir.gwpharm.com/releasedetail.cfm?releaseid=938554. Accessed January 31, 2017. 77. Lewis MA, Zhao F, Jones D, et al. Neuropathic symptoms and their risk factors in medical oncology outpatients with colorectal vs. breast, lung, or prostate cancer: results from a prospective multicenter study. J Pain Symptom Manage. 2015;49:1016-1024. 78. Cooper ZD, Haney M. Actions of delta-9-tetrahydrocannabinol in cannabis: relation to use, abuse, dependence. Int Rev Psychiatry. 2009; 21:104-112. 79. Braida D, Pozzi M, Cavallini R, et al. Conditioned place preference induced by the cannabinoid agonist CP 55,940: interaction with the opioid system. Neuroscience. 2001;104:923-926. 80. Wallace M, Schulteis G, Atkinson JH, et al. Dose-dependent effects of smoked cannabis on capsaicin-induced pain and hyperalgesia in healthy volunteers. Anesthesiology. 2007;107:785-796. 81. Hart CL, Haney M, Ward AS, et al. Effects of oral THC maintenance on smoked marijuana self-administration. Drug Alcohol Depend. 2002;67:301-309.

71. Wallace MS, Marcotte TD, Umlauf A, et al. Efficacy of inhaled cannabis on painful diabetic neuropathy. J Pain. 2015;16:616-627.

82. Haney M, Hart CL, Vosburg SK, et al. Marijuana withdrawal in humans: effects of oral THC or divalproex. Neuropsychopharmacology. 2004;29: 158-170.

72. Ware MA, Wang T, Shapiro S, et al. Smoked cannabis for chronic neuropathic pain: a randomized controlled trial. CMAJ. 2010;182: E694-E701.

83. U.S. Department of Justice Drug Enforcement Agency. Controlled Substances Schedules. www.deadiversion.usdoj.gov/schedules. Accessed March 28, 2017.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 713

SMITH, PHILLIPS, AND SMITH

Using the New ASCO Clinical Practice Guideline for Palliative Care Concurrent With Oncology Care Using the TEAM Approach Cardinale B. Smith, MD, PhD, Tanyanika Phillips, MD, MPH, and Thomas J. Smith, MD, FACP, FASCO, FAAHPM OVERVIEW Palliative care alongside usual oncology care is now recommended by ASCO as the standard of care for any patient with advanced cancer on the basis of multiple randomized trials that show better results with concurrent care than with usual oncology care. Some benefits include better quality of life, better symptom management, reduced anxiety and depression, less caregiver distress, more accordance of care with the wishes of the patient, and less aggressive end-of-life care. Several studies show a survival advantage of several months, and many show considerable cost savings: better care at an affordable cost. However, there are not enough palliative care specialists available, so oncologists must practice exemplary primary palliative care. Protocols used in the clinical trials, similar to those designed for new chemotherapy agents, help oncologists use the TEAM approach of extra time, typically an hour a month spent with the palliative care team; education, especially about prognostic awareness and realistic options, which include formal setting of goals of care and discussion of advance directives; formal assessments for symptoms and for spiritual and psychosocial health; and management by an interdisciplinary team. These are all potentially accomplished by an oncology practice to replicate the services provided by concurrent palliative care.

very patient with advanced cancer should be treated by a multidisciplinary palliative care team—in addition to her or his oncologist—within 8 weeks of diagnosis.1 The guidelines are summarized in Sidebar 1 at the end of this document. The main conclusions that led to the ASCO Guideline are straightforward: 1. Do not wait to refer all patients with advanced cancer to an interdisciplinary palliative care team until the end of life. Waiting is still the norm for most oncologists, who refer either not at all (68% in one recent series) or late in the last month of life (32%). Patients referred earlier received better care and saved the health system $6,687 per person.2 2. Caregivers may be referred, too. 3. An interdisciplinary team is best. However, many oncology practices do not have available hospice and palliative medicine–trained specialists and teams available, given a shortage of 6,000 to 10,000 palliative care practitioners. Others only have hospice

available, and these programs may not participate in the Medicare Choices Program that allows concurrent care.3

ON WHAT EVIDENCE ARE THE GUIDELINES BASED? HOW COMPELLING IS THE EVIDENCE?

To oncologists, this guidelines update may be similar to the incorporation of trastuzumab as treatment in the adjuvant breast cancer setting—a major advance that was based on several landmark trials, but without clarity about how to use the treatment or which regimen was best (a full year, as used in the CALGB-NCCTG study, or just 12 weeks, as used in the Finn-HER study, or even 2 years?). Regardless, it was an advance that should be incorporated into practice. The evidence is striking: multiple randomized trials in patients with advanced pancreatic, lung, and gastrointestinal cancers, and in even patients who underwent hematopoietic stem cell transplantations, show the benefits of concurrent care. Not a single trial shows harm (Table 1).4-19

From the Tisch Cancer Institute, Brookdale Department of Geriatrics and Palliative Medicine, Icahn School of Medicine at Mount Sinai, New York, NY; CHRISTUS St. Frances Cabrini Hospital, Alexandria, LA; Harry J. Duffey Family Patient and Family Services Program, The Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins Medical Institutions, Baltimore, MD. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Thomas J. Smith, MD, FACP, FASCO, FAAHPM, The Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins, 600 N. Wolfe St., Blalock 369, Baltimore, MD 21287; email: [emailprotected]. © 2017 American Society of Clinical Oncology

714 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

ONCOLOGY PALLIATIVE CARE

ACADEMIC MODELS To improve the quality cancer care, standardized criteria, or triggers for palliative care consultation, were developed on an inpatient solid tumor service. Professionals on the service performed a 3-month pilot intervention in which patients who met eligibility criteria automatically received a palliative care consultation. For 6 weeks before the intervention, the inpatient solid tumor census was examined daily to identify patients who met eligibility criteria. This was the usual-care group, so palliative care consultations were not mandated but could be obtained if requested by the primary team. Patients were eligible if they met one or more of the following criteria (Fig. 1): (1) advanced cancer (stage IV solid tumor or stage III lung or pancreatic cancer); (2) prior hospitalization within 30 days; hospitalization lasting longer than 7 days; and (3) any active symptom, including pain, nausea/ vomiting, dyspnea, delirium, and psychological distress. Patients who were admitted for routine chemotherapy or a planned procedure, or who were unable to speak English, were excluded. The palliative care consultation followed a standardized approach that used the National Comprehensive Cancer Network guidelines and core elements of palliative care as detailed in the National Quality Forum. Specifically, this approach included the following: (1) symptom assessment and treatment using the Edmonton Symptom Assessment Scale; (2) establishment of goals of care and advance care plans using standardized communication protocols; and (3) transition planning. The palliative care team was composed of (at a minimum) one board-certified palliative care physician, one nurse practitioner, one social worker, a chaplain, and one or two trainees. Recommendations to consulting physicians were made using standardized palliative care team chart notes and in person or by telephone. Patients were seen daily to monitor implementation and results of treatment recommendations and to assess new and

KEY POINTS • Concurrent palliative care alongside usual oncology care is now recommended by ASCO for all patients with advanced cancer. • Concurrent specialty care should start within 8 weeks of diagnosis and be delivered by an interdisciplinary team. • There are not enough palliative care specialists, therefore oncologists can adapt the methods used by the palliative care teams in the randomized trials. • Dedicate an extra hour a month for this. Start with symptom, psychosocial, and spiritual assessments. Inquire about their understanding of their situation. Then bring up goals of care and create advance directives. Whenever the prognosis changes or when reviewing scan results, ask the patient, "Would you like to talk about what this means?" • Set up a hospice information visit when it is possible the patient could die within the next 6 months to ensure that the transition is planned and smooth.

ongoing symptoms. The palliative care teams conducted or assisted with discussions about new or changing goals of care, communicated bad news, and conducted or assisted with associated treatment adjustments. The teams also worked with the social workers and families of the patients to facilitate transition management consistent with goals of care according to available resources. Overall, 39% of patients in the pre-intervention group and 80% in the intervention group (p < .0001) received a palliative care consultation (Table 2). Univariable analysis to compare the pre-intervention group with the intervention group demonstrated a decrease in 30-day readmission rates decreased from 35% to 18%, respectively (p = .04). Hospice referral rates increased from 14% to 26% in the preintervention and intervention groups, respectively (p = .03), and receipt of chemotherapy after discharge decreased from 44% to 18%, respectively (p = .03). In addition, there was no significant change in length of stay (p = .15) or use of the intensive care unit (p = .11) between the two groups. Overall, place of discharge was different between the groups (p = .004). Patients in the intervention group were more likely to be discharged to home with home-based services (32% vs. 19%), home hospice (15% vs. 8%), or inpatient hospice (11% vs. 6%) and were less likely to be discharged to subacute rehabilitation facilities (3% vs. 13%).21

COMMUNITY PRACTICE MODELS

Where you live when you are diagnosed with cancer matters.20 In 2013, the Centers for Disease Control and Prevention reported that the top five states with the highest cancer mortality rates were geographically outlined in southern United States, and Louisiana was among them. Patients with cancer in Louisiana, unfortunately, are diagnosed with high rates of advanced cancer, have high symptom burdens, and have high palliative care needs. Central Louisiana services a large rural area, and small community hospitals have limited resources to provide formal palliative care programs. Limited resources include time to deliver palliative care services in busy ambulatory oncology practices and staffing with expertise in palliative care to deliver inpatient and outpatient palliative care services. CHRISTUS St. Frances Cabrini Hospital in Alexandria, Louisiana, is a 240-bed facility with a comprehensive cancer center to deliver cancer services in one location. The oncology outpatient team includes four medical oncologists, a radiation oncologist, a nurse practitioner, a nurse navigator, and a social worker. Within the cancer center, palliative care tools used by the social worker include the Distress Thermometer and the Patient Health Questionnaireafter patient referral at a pivotal moment during diagnosis. Patients are referred at the time of diagnosis and/or at the initiation of cancer-directed therapy, which often is within 8 weeks of diagnosis, but patients may be referred at any point along the trajectory of care. Unlike academic programs, small community hospitals often have limited resources to provide formal palliative care programs. Limited resources here also include time and staffing. CHRISTUS St. Frances Cabrini Hospital in Alexandria asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 715

SMITH, PHILLIPS, AND SMITH

TABLE 1. Summary of Recent Studies Comparing Usual Care to Usual Care With Palliative Care Patient Experience Study First Author and Year (Population)

QOL

Symptoms

Anxiety Depression

Caregiver Distress

Survival

Cost

Brumley et al, 2007 (one of three cancers)4

NM; satisfaction increased

Equal

−$7,550 per person (p = .03); more likely to die at home, less likely to visit ED, admit to hospice

Gade et al, 2008 (one of three cancers)5

Increased (p = .04)

Equal

−$4,885 per person (p = .001); fewer ICU admissions (p = .04), longer hospice stays (p = .04)

Bakitas et al, 2009 (cancer)6

Increased (p = .02)

p = .06

Less depressed mood (p = .02)

Longer by 5.5 months (p = .14 [NS])

Equal

Temel et al, 2010 (lung cancer)7

Increased (p = .03)

Less depression (p = .01)

Longer by 2.7 months (p = .02)

No change in costs despite the longer survival; cost per day was $117 lower.8

Farquhar et al (cancer as cause of breathlessness)9

Increased

Improved: reduced patient distress from breathlessness (p = .049)

Equal

Total costs, £354 ($444) less; better QOL; dominates cost-effectiveness

Zimmermann, 2014 (cancer)10

Increased (p = .05)

Equal (3 months; p = .33) improved (4 months; p = .05)

Equal

Improved (p = .003)

Equal

Higginson et al, 2014 (dyspnea, most cancer)11

Equal

Improved: mastery of breathlessness (p = .048); equal dyspnea

Equal

Bakitas et al, 2015 (cancer)12,13

Equal (p = .30)

Equal (p = .09)

Equal mood

Lower depression and stress, (p = .02 and .01, respectively) but not better QOL

Longer by 6.5 months; 1-year OS, 63% vs. 48% (p = .038)

NR; equal resource use

Ferrell et al, 2015 (lung cancer)14,15

Increased (p < .001)

Improved (p < .001)

Improved: better well-being and less distress (p = .001); less burden (p = .008)

Longer by 6 months (NS)

NR; more advance directives: 44% vs. 9% (p < .001)

Grudzen et al, 2016 (patients with cancer in emergency departments)16

Increased (p = .03)

Equal

Longer by 5.2 months (p = .20 [NS])

Equal: only 25%–28% use of hospice in both groups

Temel et al, 2016 (lung or gastrointestinal cancer)17

Increased at week 12 (p = .34) and at week 24 (p = .01)

Improved (p = .048)

Too early to tell

NR; more likely to discuss end of life wishes: 30% vs. 14.5% (p = .004)

El-Jawahri et al, 2016 (bone marrow transplantation)18

Smaller decrease (p = .045)

Less increase (p = .03 at 2 weeks); equal at 3 months

Improved depression and anxiety (p < .001)

No change in QOL or anxiety; less increase in depression (p = .03)

Too early to tell

Continued

716 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

ONCOLOGY PALLIATIVE CARE

TABLE 1. Summary of Recent Studies Comparing Usual Care to Usual Care With Palliative Care (Cont'd) Patient Experience Study First Author and Year (Population)

QOL

Symptoms

Maltoni et al, 2016 (pancreatic cancer)19,20

Increased (p = .04)

NR; FACT-Hep, HCS, and TOI all better with palliative care

Anxiety Depression

Caregiver Distress

Survival

Cost

Equal

Equal: OS 32%–37% at 1 year

NR; NS improvements in chemotherapy use in the last 30 days, hospice LOS, place of death

Abbreviations: FACT-Hep, Functional Assessment of Cancer Therapy—Hepatobiliary; HCS, Hepatobiliary Cancer Subscale; ICU, intensive care unit; LOS, length of stay; ND, not determined; NM, not measured; NR, not reported; NS, not significant; OS, overall survival; QOL quality of life; TOI, trial outcome index.

FIGURE 1. Eligibility Criteria for Automatic Palliative Care Consultation Among Hospitalized Patients With Solid Tumors

cancer-intensive pharmacotherapy such as intravenous chemotherapy or immunotherapy to symptom burden–designed therapeutic plans or hospice), this has not yet been implemented. At this time, palliative care specialists provide urgent ambulatory services during high disease–burden symptom crises, existential crises, or family and caregiver management crises. Access to experts in palliative care provides tremendous stewardship for the delivery of complex

TABLE 2. Outcomes of Patients: Comparison of Pre-Intervention Control to Intervention Group No. (%) of Patients Characteristic

Pre-Intervention (n = 48)

Intervention (n = 65)

p Value

Palliative care consultation

19 (39)

52 (80)

< .0001

30-day readmission

17 (35)

13 (18)

.04

Hospice referral

7 (14)

17 (26)

.03

Mean (SD) length of stay, days

11 ± 12

14 ± 14

.15

ICU use

5 (10)

2 (3)

.11

Median (IQR) number of days in the ICU

1 (3)

0.3 (2)

.08

Disposition

.004

Home without services

25 (52)

16 (25)

Home with services

9 (19)

21 (32)

Home with hospice

4 (8)

10 (15)

Subacute rehabilitation

6 (13)

2 (3)

Inpatient hospice

3 (6)

7 (11)

Chemotherapy after discharge

.03

Yes

21 (44)

12 (18)

27 (56)

53 (82)

Abbreviations: IQR, interquartile range; SD, standard deviation.

Louisiana offers a formal palliative care inpatient consultation service and an inpatient hospice unit run by a board-certified palliative care physician and nurse practitioner. Although a formal combined palliative oncology ambulatory program has been proposed, in which the patient would visit with both an oncology provider and a palliative care specialist for symptom control visits or transitional management visits (e.g., when a patient is transitioning from aggressive

care among patients, especially those with comorbidities, such as cardiovascular disease, diabetes, kidney disease, and liver disease. The greatest challenge for guideline implementation in a small community hospital is time (scheduling a combined or same-day visit or a different-day separate palliative care visit). The current inpatient palliative care program and hospice unit remain busy, which thus inhibits burgeoning asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 717

SMITH, PHILLIPS, AND SMITH

TABLE 3. Components of TEAM-Based Palliative Care Component

Use in Trials

What Is Known

Time

A structured palliative care visit at least an extra hour a month, every month; not just once.

Provider does not have to be the doctor. Advanced practice nurses may be better for this.

Education

Structured and recurrent discussions about prognosis, symptom management, and communication with the health care team.

Usual topics include, on a recurring basis, the following: (1) Medically appropriate options for treatment. This is best done by listing the treatments used and their outcomes, especially with recurrent disease. If appropriate, reinforce patient and family work as advocates. (2) After a scan that shows progressive disease is the perfect time to revisit prognosis and advance care planning. “Would you like to talk about what this means?” (3) Advance care planning. Recent data suggest DPMA does not make any difference, there has to be a living will or advance directive.24 (4) Use of hospice for best possible care, and arranging a hospice information visit when the disease is predictably going to take the person’s life within 3–6 months, or even longer.

Assessment

Formal assessments for symptoms (ESAS, MSAS-C, CAPC rounding tool), spirituality (FICA, or “Are you a religious or spiritual person?”), and psychosocial status (Distress Thermometer, others)

After these formal symptom assessments, move onto goal setting. Use questions such as: (1) How do you like to get medical information? (2) What is your understanding of your disease? (3) What is important to you? (4) What are you hoping for? (5) Have you thought about a time when you might become sicker, such that you would need and advance directive or living will? (Some motivational interviewing)

It is not sufficient to just ask “How are you?” Patients and families are sometimes reluctant to share their problems for fear that nothing can be done, the oncologist will stop treating their cancer, or they will be labeled as a complainer.

The questioning was incorporated into a temporary tattoo that gives oncologists a script to start the most difficult discussions (Fig. 1).

Use set protocols and an interdisciplinary team (advance practice nurses, social work, chaplain, doctors).

Giving people knowledge of their realistic options and a plan of action was shared across all the studies.

Management

An oncology office that does not have established social work or chaplain ties should develop them, much as ties are developed with surgeons or radiation oncologists. Abbreviations: APN, advance practice nurse; CAPC, Center to Advance Palliative Care; DPMA, durable power of medical attorney; ESAS, Edmonton Symptom Assessment Scale; FICA, faith, importance, community, actions; MSAS-C, Memorial Symptom Assessment Scale-Condensed; TEAM, time, education, assessment, and management.

growth and development of combination programs for these resources in an ambulatory clinic. Oncology provider visits often occur at 15- to 20-minute intervals. The volume of patients that must move through the practice and treatment areas of intravenous infusion and radiation requires management of space and time efficiently. A 1-hour palliative care visit is a challenging component to include within the existing oncology operational model for a same-day visit. Conversely, more than one third of patients at CHRISTUS St. Frances Cabrini Hospital travel 30 minutes or more to the facility; some travel 1 hour and 30 minutes for care, and this makes separate-day palliative care visits less desirable and increases the risk for noncompliance. There has been a surge in the number of agencies delivering hospice care in Louisiana; more than 200 agencies statewide offer the opportunity for concurrent care. Although this offers choice for providers and patient/families, it also offers competing interests for smaller communities/hospitals 718 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

that seek to provide a systematic approach to palliative care with valid and reliable tools for assessment and delivery of care.

WHERE CAN I GET THE TOOLS AND TECHNIQUES FOR OUR PRACTICE TO OFFER PALLIATIVE CARE METHODS USED IN RANDOMIZED CLINICAL TRIALS?

The CHRISTUS St. Frances Cabrini Hospital service has tried to take the salient points from all the recent trials and incorporate them into the TEAM (time, education, assessments, and management) approach. A recent meta-analysis about palliative care included 43 randomized controlled trials with data from 12,731 patients and found improvements in patient quality of life and symptom burden, but the range and intensity of the interventions studied varied so much that this study is not comparable to this series of more cancerfocused team interventions.22 The methods of a multi- or

ONCOLOGY PALLIATIVE CARE

FIGURE 2. A Communication Tattoo Used at Johns Hopkins

Provided by Rebecca Kirch, JD. Tattoos available from Dr. Smith for $0.50 each.

interdisciplinary team approach to concurrent palliative care used in these trials are in Table 3.23 Time is one factor to the success of palliative care; an extra hour a month, after initial consultation, was recorded in all of the studies. Oncologists cannot complete a palliative care visit within a 20-minute visit that concentrates on response to chemotherapy. These palliative care visits can be in person or by phone/telemedicine. Underlying principles, though, are that the visit must be structured and that it should last at least an hour a month, regardless of which practitioner is involved. Most trials have included palliative care advance practice nurses and doctors on the team, as we do in practice. Education was a component of all of the clinical trials. In the monthly visits with the palliative care team, the patient and family can explore realistic options. Prognostic awareness (or the ability to admit a potential life-ending illness) appears to be key and requires coaching as well as direct communication by the health care provider. More than twothirds of patients with stage IV incurable lung and colorectal cancer thought their palliative chemotherapy,24 radiation,25 and/or surgery26 could cure them. An excellent communication guide is available.27 Knowledge works: patients with prognostic awareness, especially those who completed advanced medical directives more than 30 days before death, die less often in the hospital (19% vs. 50% in Australia28) and use hospice care more and for longer durations.29 In the non–small cell lung cancer trial by Temel et al,7 those in the palliative care group who had prognostic awareness received (ineffective fourthor fifth-line) intravenous chemotherapy near the end of life 9% of the time versus 50% in the usual care group. Those who have end-of-life discussions (about goals of care, understanding of illness) are more likely to be satisfied, die at the place of their choosing, and have less distressed relatives.30 However, physicians must start the conversations. Those patients who had prognostic discussions with

their physicians revised their self-reported estimates by a 17.2-month decrease, which more accurately reflected reality (months not years). These patients expressed no more depression, sadness, or anxiety; completed advance directives more often; and received better end-of-life care.31 The palliative care team must work with the oncologist, especially about prognosis, although the two teams may have starkly different views on prognosis and medically appropriate treatment. Most palliative care practitioners use a script to approach what the patient and family knows, and wants to know, before they talk about prognosis. Our service found that a temporary tattoo on the inner forearm, visible to the oncologist or advance practice nurse, was helpful to remember how to start difficult conversations32 (Fig. 1). After physicians review questions and understand the comprehension and goals of the patient and family, motivational interviewing is easier (Fig. 2). An example of motivational interviewing is, “You are doing okay now, but have you thought about a time in the future when you might be sicker and need and advance directive or living will?” It is simple for oncologists to address understanding and prognosis with a patient after any scan that shows progressive disease. A study found that only four of 64 oncologist discussions about scan results had frank prognosis discussions; the authors suggested addition of the question “Would you like to talk about what this means?” to allow the patient some control about providing permission to disclose crucial information.33 Formal assessment tools, as used in nearly all of the trials, also are key to the process of identifying physical symptoms, psychosocial distress, and spiritual distress and are key to identifying how the patient and family are coping. Although oncologists may believe that spiritual assessment is not part of their job description, 87% of patients with cancer want physicians to know their spiritual needs; yet, only 6% were ever asked. Receiving spiritual care from the medical team was associated with a doubling of the use of hospice, and asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 719

SMITH, PHILLIPS, AND SMITH

TABLE 4. Goals of Care Discussion Template for Epic, Cerner, or Other Electronic Medical Records Care Discussion Point

Template Question

General questions

How do you like to get medical information? Full and completely honest, or something else? How about prognosis? What is your understanding of your situation? What is important to you? What are you hoping for? Are you getting the best care possible? We don’t want to leave medical stones unturned. Recognize that not all things have a medical fix.

Questions studied in clinical trials

Do you have a will? Do you have a living will or advanced directive? What does it say about CPR? (For patients imminently dying in the hospital as a result of their cancer, the success rate of CPR is zero.) Who do you want to make medical decisions, if you can’t? Have you discussed this with her/him? Are there spiritual issues to be settled? Are there family issues to be settled? Are there financial issues to be settled? Have you met with hospice yet? (Plan for at least 3–6 months before death, which, for most diseases, is predictable. This really helps the transition if and when hospice is needed, and it helps people with congestive heart failure who use hospice live longer.) Have you thought about where you would like to be for your death, if and when? Legacy work: (1) Let’s start doing a life review: what do you want people to remember about you? (2) What's important to you? (3) What are you hoping for? (4) What do you want to accomplish in the time you have?

Living day to day Other instructions

Exercise Diet How to call or reach me: Office Days Nights Cell Email

the number of patients who died in the intensive care unit (a marker of poor quality of care) decreased from 22% to 0%.34 Yet, as oncologists, we fail at this task: in an audit of care given to patients with glioblastoma on our service, no patient had a formal outpatient symptom, spiritual, or coping assessment or a formal statement of prognosis. Perhaps as a consequence, 37% were hospitalized in the last month of life for an average of 9 days, only 17% had any advance directives in the charts, and nearly 40% received chemotherapy in the last month of life.35 We hope to do better by using the formal tools used in the randomized trials. Management by a consultant interdisciplinary team also was a key component of the randomized trials. In one 720 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Australian trial for patients who received palliative care, a structured regular meeting of the interdisciplinary team with recommendations to the primary care physician, was the only way that care actually improved.36 Most practices will have some components of the interdisciplinary team in place: social workers, chaplains, advance practice and oncology nurses, and physicians. A key step is to identify patients at risk for complications and discuss their care at a weekly interdisciplinary team meeting to troubleshoot. Such forward thinking of anticipatory care, like calling patients the day after a new chemotherapy regimen, has been a successful technique used by oncology medical home models that led to reduced hospitalizations and lower costs.

ONCOLOGY PALLIATIVE CARE

SIDEBAR 1. The Integration of Palliative Care Into Standard Oncology Care: ASCO Clinical Practice Guideline

Guideline Question

Should palliative care concurrent with oncology care be standard practice? Answer: Yes, unequivocally. And EARLY, within 8 weeks, not at the end of life.

Key Recommendation

• Patients with advanced cancer, inpatient and outpatient, should receive dedicated palliative care services early in the disease course, concurrent with active treatment. Referring patients to interdisciplinary palliative care teams is optimal, and services may complement existing programs. Providers may refer caregivers of patients with early or advanced cancer to palliative care services.

Specific Recommendations

• Patients with advanced cancer should be referred to interdisciplinary palliative care teams (consultation) that provide inpatient and outpatient care early in the course of disease, alongside active treatment of their cancer. (Type: evidence based, benefit outweighs harms; Evidence quality: intermediate; Strength of recommendation: strong). • Palliative care for patients with advanced cancer should be delivered through interdisciplinary palliative care teams with consultation available in both outpatient and inpatient settings (Type: evidence based, benefits outweigh harms; Evidence quality: intermediate; Strength of recommendation: moderate). • Patients with advanced cancer should receive palliative care services, which may include a referral to a palliative care provider. Essential components of palliative care include may include: ◦◦ rapport and relationship building with patient and family caregiver(s) ◦◦ symptom, distress, and functional status management (i.e., pain, dyspnea, fatigue, sleep disturbance, mood, nausea, or constipation) ◦◦ exploration of understanding and education about illness and prognosis ◦◦ clarification of treatment goals ◦◦ assessment and support of coping needs (i.e., provision of dignity therapy) ◦◦ assistance with medical decision making ◦◦ coordination with other care providers ◦◦ provision of referrals to other care providers as indicated. • For newly diagnosed patients early palliative care involvement, within 8 weeks of diagnosis, is suggested. (Type: informal consensus, Evidence quality: intermediate; Strength of recommendation: moderate). • Among patients with cancer with high symptom burden, high expectant needs, or great anticipation of experiencing overlapping phases of care, (diagnosis, staging, treatment, and end of life), outpatient programs of cancer care should provide and use dedicated resources (palliative care clinicians) to deliver palliative care services to complement existing program tools (Type: evidence-based, benefits outweigh harms; Evidence quality: intermediate; Strength of recommendation: moderate). • For patients with early or advanced cancer for whom family caregivers will provide care in the outpatient setting, nurses, social workers, or other providers may initiate caregiver-tailored palliative care support, which could include telephone coaching, education, referrals, and face-to-face meetings. For family caregivers who may live in rural areas or be unable to travel to clinic, offering telephone support over face-to-face support may be offered (Type: evidence-based; Evidence quality: low; Strength of recommendation: Weak).

Additional Resources

More information, including a Data Supplement with additional evidence tables, a Methodology Supplement with information about evidence quality and strength of recommendations, slide sets, and clinical tools and resources, is available at www.asco.org/palliative-care-guideline and www.asco.org/guidelineswiki. Patient information is available at www. cancer.net. The essential components of the multidisciplinary team are still unknown. When Muir et al37 offered palliative care in oncology offices with just a physician and advance practice nurse, patients experienced better symptom management—a 21% decrease in symptom burden, an increase in oncologist satisfaction (necessary for the palliative care team to continue

to work with oncologists), and an 87% increase of consultations in 2 years. Their efforts saved each oncologist more than 4 weeks of time so that the practice could offer more regular oncology services, or so that oncologists could take some time off to avoid burnout.35 We believe that these services are essential and that providers, patients, and families asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 721

SMITH, PHILLIPS, AND SMITH

greatly appreciate them, so these services are incorporated within the budget. Most offices in the Louisiana service had substantial financial counseling available but did not have social workers or chaplains, so some of the nonreimbursable services, such as chaplaincy or social worker, may be omitted; however, the omission would need formal testing before it could be endorsed. US Oncology has adopted the best practice model of appointing someone in the office, usually a social worker or nurse, to review advance care planning within the first visits of diagnosis of a life-ending illness, and the result was advance care planning increases to 80%. The Johns Hopkins service adopted a formal goals of care discussion format in EPIC to capture some of the practical parts of these difficult conversations. Just as laboratory values or radiographs appear on the screen, so does this format, as an EPIC SmartPhrase, appear on the screen; practitioners type in the answers and print the patient information, which contributes to meaningful use and is easy to send to referring health care practitioners so that all can be on the same page (Table 4). Although this

tool was not used in the randomized controlled trials, it is being used in them now for patients who receive concurrent care in phase I trials, and it appears to be a useful work-simplifying tool.

CONCLUSION

The TEAM approach works in practice like it did in the clinical trials, if the protocol is followed. Appoint someone, if not the oncologist, to perform the assessments. Use every worse scan or change in Eastern Cooperative Oncology Group performance status to ask “Would you like to talk about what this means?” and give real numbers about prognosis and options. Make a point of having a goal of care discussion at several points as the prognosis changes, using the template in Table 4. Ensure that most of your patients have advance medical directives completed months before they die. Last, if you involve specialists in palliative care, as suggested by the ASCO guideline for every patient with advanced cancer, ensure that it happens within 8 weeks of diagnosis.

References 1. Ferrell BR, Temel JS, Temin S, et al. Integration of palliative care into standard oncology care: American Society of Clinical Oncology clinical practice guideline update. J Clin Oncol. 2017;35:96-112. 2. Scibetta C, Kerr K, Mcguire J, et al. The costs of waiting: implications of the timing of palliative care consultation among a cohort of decedents at a comprehensive cancer center. J Palliat Med. 2016;19:69-75. 3. Centers for Medicare & Medicaid Services. Medicare Care Choices Model. https://innovation.cms.gov/initiatives/Medicare-Care-Choices/. Accessed January 21, 2017. 4. Brumley R, Enguidanos S, Jamison P, et al. Increased satisfaction with care and lower costs: results of a randomized trial of in-home palliative care. J Am Geriatr Soc. 2007;55:993-1000. 5. Gade G, Venohr I, Conner D, et al. Impact of an inpatient palliative care team: a randomized control trial. J Palliat Med. 2008;11:180-190. 6. Bakitas M, Lyons KD, Hegel MT, et al. Effects of a palliative care intervention on clinical outcomes in patients with advanced cancer: the project ENABLE II randomized controlled trial. JAMA. 2009;302:741-749. 7. Temel JS, Greer JA, Muzikansky A, et al. Early palliative care for patients with metastatic non–small cell lung cancer. N Engl J Med. 2010;363:733-742. 8. Greer JA, Tramontano AC, McMahon PM, et al. Cost analysis of a randomized trial of early palliative care in patients with metastatic non–small cell lung cancer. J Palliat Med. 2016;19:842-848. 9. Farquhar MC, Prevost AT, McCrone P, et al. Is a specialist breathlessness service more effective and cost-effective for patients with advanced cancer and their careers than standard care? Findings of a mixedmethod randomised controlled trial. BMC Med. 2014;12:194. 10. Zimmermann C, Swami N, Krzyzanowska M, et al. Early palliative care for patients with advanced cancer: a cluster-randomised controlled trial. Lancet. 2014;383:1721-1730. 11. Higginson IJ, Bausewein C, Reilly CC, et al. An integrated palliative and respiratory care service for patients with advanced disease and

722 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

refractory breathlessness: a randomised controlled trial. Lancet Respir Med. 2014;2:979-987. 12. Bakitas MA, Tosteson TD, Li Z, et al. Early versus delayed initiation of concurrent palliative oncology care: patient outcomes in the ENABLE III randomized controlled trial. J Clin Oncol. 2015;33:1438-1445. 13. Dionne-Odom JN, Azuero A, Lyons KD, et al. Benefits of early versus delayed palliative care to informal family caregivers of patients with advanced cancer: outcomes from the ENABLE III randomized controlled trial. J Clin Oncol. 2015;33:1446-1452. 14. Ferrell B, Sun V, Hurria A, et al. Interdisciplinary palliative care for patients with lung cancer. J Pain Symptom Manage. 2015;50:758-767. 15. Sun V, Grant M, Koczywas M, et al. Effectiveness of an interdisciplinary palliative care intervention for family caregivers in lung cancer. Cancer. 2015;121:3737-3745. 16. Grudzen CR, Richardson LD, Johnson PN, et al. Emergency departmentinitiated palliative care in advanced cancer: a randomized clinical trial. JAMA Oncol. Epub 2016 Jan 14. 17. Temel JS, Greer JA, El-Jawahri A, et al. Effects of early integrated palliative care in patients with lung and gi cancer: a randomized clinical trial. J Clin Oncol. 2017;35:834-841. 18. El-Jawahri A, LeBlanc T, VanDusen H, et al. Effect of inpatient palliative care on quality of life 2 weeks after hematopoietic stem cell transplantation: a randomized clinical trial. JAMA. 2016;316:2094-2103. 19. Maltoni M, Scarpi E, Dall’Agata M, et al; Early Palliative Care Italian Study Group (EPCISG). Systematic versus on-demand early palliative care: results from a multicentre, randomised clinical trial. Eur J Cancer. 2016;65:61-68. 20. Mokdad AH, Dwyer-Lindgren L, Fitzmaurice C, et al. Trends and patterns of disparities in cancer mortality among US counties, 19802014. JAMA. 2017;317:388-406. 21. Adelson K, Paris J, Horton JR, et al. Standardized criteria for palliative care consultation on a solid tumor oncology service reduces downstream health care use. J Oncol Pract. Epub 2017 Mar 17.

ONCOLOGY PALLIATIVE CARE

22. Kavalieratos D, Corbelli J, Zhang D, et al. Association between palliative care and patient and caregiver outcomes: a systematic review and meta-analysis. JAMA. 2016;316:2104-2114. 23. Narang AK, Wright AA, Nicholas LH. Trends in advance care planning in patients with cancer: results from a national longitudinal survey. JAMA Oncol. 2015;1:601-608. 24. Weeks JC, Catalano PJ, Cronin A, et al. Patients’ expectations about effects of chemotherapy for advanced cancer. N Engl J Med. 2012;367:1616-1625. 25. Chen AB, Cronin A, Weeks JC, et al. Expectations about the effectiveness of radiation therapy among patients with incurable lung cancer. J Clin Oncol. 2013;31:2730-2735. 26. Kim Y, Winner M, Page A, et al. Patient perceptions regarding the likelihood of cure after surgical resection of lung and colorectal cancer. Cancer. 2015;121:3564-3573. 27. Jackson VA, Jacobsen J, Greer JA, et al. The cultivation of prognostic awareness through the provision of early palliative care in the ambulatory setting: a communication guide. J Palliat Med. 2013;16:894-900.

30. Kumar P, Temel JS. End-of-life care discussions in patients with advanced cancer. J Clin Oncol. 2013;31:3315-3319. 31. Enzinger AC, Zhang B, Schrag D, et al. Outcomes of prognostic disclosure: associations with prognostic understanding, distress, and relationship with physician among patients with advanced cancer. J Clin Oncol. 2015;33:3809-3816. 32. Leong M, Shah M, Smith TJ. How to avoid late chemotherapy. J Oncol Pract. 2016;12:1208-1210. 33. Singh S, Cortez D, Maynard D, et al. Characterizing the nature of scan results discussions: insights into why patients misunderstand their prognosis. J Oncol Pract. 2017;13:e231-e239. 34. Balboni TA, Balboni M, Enzinger AC, et al. Provision of spiritual support to patients with advanced cancer by religious communities and associations with medical care at the end of life. JAMA Intern Med. 2013;173:1109-1117. 35. Kuchinad K, Strowd R, Evans A. End of life care for glioblastoma patients at a large academic center. J Neurooncol. 2017. In press.

28. Stein RA, Sharpe L, Bell ML, et al. Randomized controlled trial of a structured intervention to facilitate end-of-life decision making in patients with advanced cancer. J Clin Oncol. 2013;31:3403-3410.

36. Abernethy A, Currow DC, Shelby-James T, et al. Delivery strategies to optimize resource utilization and performance status for patients with advanced life-limiting illness: results from the "palliative care trial" [ISRCTN 81117481]. J Pain Symptom Manage. 2013;45:488-505.

29. Mack JW, Walling A, Dy S, et al. Patient beliefs that chemotherapy may be curative and care received at the end of life among patients with metastatic lung and colorectal cancer. Cancer. 2015;121:1891-1897.

37. Muir JC, Daly F, Davis MS, et al. Integrating palliative care into the outpatient, private practice oncology setting. J Pain Symptom Manage. 2010;40:126-135.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 723

PEDIATRIC ONCOLOGY

ADVANCES IN THE TREATMENT OF PEDIATRIC BONE SARCOMAS

Advances in the Treatment of Pediatric Bone Sarcomas Patrick J. Grohar, MD, PhD, Katherine A. Janeway, MD, MMSc, Luke D. Mase, DO, and Joshua D. Schiffman, MD OVERVIEW Bone tumors make up a significant portion of noncentral nervous system solid tumor diagnoses in pediatric oncology patients. Ewing sarcoma and osteosarcoma, both with distinct clinical and pathologic features, are the two most commonly encountered bone cancers in pediatrics. Although mutations in the germline have classically been more associated with osteosarcoma, there is recent evidence germline alterations in patients with Ewing sarcoma also play a significant role in pathogenesis. Treatment advances in this patient population have lagged behind that of other pediatric malignancies, particularly targeted interventions directed at the biologic underpinnings of disease. Recent advances in biologic and genomic understanding of these two cancers has expanded the potential for therapeutic advancement and prevention. In Ewing sarcoma, directed focus on inhibition of EWSR1-FLI1 and its effectors has produced promising results. In osteosarcoma, instead of a concentrated focus on one particular change, largely due to tumor heterogeneity, a more diversified approach has been adopted including investigations of growth factors inhibitors, signaling pathway inhibitors, and immune modulation. Continuing recently made treatment advances relies on clinical trial design and enrollment. Clinical trials should include incorporation of biological findings; specifically, for Ewing sarcoma, assessment of alternative fusions and, for osteosarcoma, stratification utilizing biomarkers. Expanded cancer genomics knowledge, particularly with solid tumors, as it relates to heritability and incorporation of family history has led to early identification of patients with cancer predisposition. In these patients through application of cost-effective evidence-based screening techniques the ultimate goal of cancer prevention is becoming a realization.

wing sarcoma (ES) is a small, round blue cell tumor characterized by oncogenic fusions between EWSR1 or, less often, FUS and genes of the ETS family (FLI1 being the most common; Table 1).1,2 In pediatric patients, ES arises in bone in 80% of patients with occurrence in axial bones slightly more common than occurrence in appendicular bones; conversely, in adults as many as 75% of primary ES arise in soft tissue. The remaining cases of ES arise in soft tissue locations. ES occurs in patients age 0 to 50 with the median age somewhere between age 13 and 17. Poor prognostic factors include presence of metastatic disease at diagnosis, age 18 or older at diagnosis, primary site in the pelvis, large tumor, and poor histologic necrosis after induction chemotherapy.3 Diagnosis of ES is usually straightforward when biopsy of a typical-appearing mass in a patient of the appropriate age demonstrates a small, round blue cell tumor with intense membranous CD99 staining, and cytogenetics, and fluorescent in situ hybridization, or reverse-transcription polymerase chain reaction demonstrate an associated fusion. It is important to note that fusions involving EWSR1 and FUS are seen in a variety of other sarcomas, as well (Table 1). Thus, a fluorescent in situ hybridization result indicating a

fusion involving EWSR1 is not pathognomonic for ES. In addition, there is increasing recognition of the so-called Ewinglike sarcomas. This ill-defined group of malignancies is characterized by the presence of alternative fusions such as CIC-DUX4 and CCNB3-BCOR and histopathology not entirely classic for ES, including less uniform CD99 immunohistochemistry. The Ewing-like sarcomas appear to represent as many as 5% of the Ewing family of sarcomas, and are thought to occur more often in soft tissue locations and in older patients, and they may have a worse outcome.2,4 Successive trials of chemotherapy intensification in ES have resulted in improved outcomes with 5-year overall survival in 1975 to 1977 versus 2002 to 2008 increasing from 58% to 83%. Chemotherapy treatment of ES includes vincristine, doxorubicin, etoposide, and ifosfamide and/or cyclophosphamide. In the United States, all patients receive intensively timed (cycles of every 2 weeks) vincristine, doxorubicin, and cyclophosphamide alternating with ifosfamide and etoposide with growth factor support. In much of Europe, patients receive induction with vincristine, ifosfamide, doxorubicin, and etoposide with consolidation therapy depending on risk factors. Patients with localized disease and

From the Van Andel Research Institute/Helen DeVos Children’s Hospital, Grand Rapids, MI; Harvard Medical School, Boston, MA; Dana-Farber/Boston Children’s Cancer and Blood Disorders Center, Boston, MA; Department of Pediatrics and Department of Oncological Sciences, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Katherine A. Janeway, MD, MMSc, Pediatric Oncology, Dana-Farber Cancer Institute, 450 Brookline Ave., Dana 3-130, Boston, MA 02215; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 725

GROHAR ET AL

TABLE 1. Translocations in Ewing and Ewing-like Sarcomas and EWSR1- and FUS-Containing Translocations in Other Sarcomas Diagnosis

Translocation

Frequency of Ewing (%)

Ewing sarcoma

EWSR1-FLI1

EWSR1-ERG

EWSR1-FEV

H3.1) and nearly half of DMGs arising in the thalamus and spinal cord.3 Most K27M-mutant DMGs are hypointense on T1 and hyperintense on T2/fluid-attenuated inversion recovery MRI studies. Intratumoral contrast enhancement, necrosis, or hemorrhage may be present. DIPG presents as an expansile lesion centered within the pons (Fig. 1A) and often shows

From the Cincinnati Children's Hospital Medical Center, Cincinnati, OH; German Cancer Research Center, Heidelberg, Germany; Dana-Farber Cancer Institute, Boston, MA. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Christine E. Fuller, MD, Division of Pathology and Laboratory Medicine, Cincinnati Children’s Hospital Medical Center, 3333 Burnet Ave., MLC 1035, Cincinnati, OH 45229-3026; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 753

FULLER, JONES, AND KIERAN

FIGURE 1. Example of a DMG

H3 K27M-mutant presenting as a diffuse intrinsic pontine glioma, hyperintense on this T2-weighted axial MRI (A). Histology was that of a high-grade astrocytoma with moderate pleomorphism (B). Mutantspecific immunohistochemistry for H3 K27M showed strong staining of the tumor cell nuclei (C).

KEY POINTS • The WHO 2016 represents a major restructuring of brain tumor classification, incorporating both microscopic and molecular parameters. • H3 K27M diffuse midline glioma, RELA fusion ependymoma, and molecular-based medulloblastoma subgrouping are new WHO 2016 additions important in pediatric neuro-oncology. • The WHO 2016 also introduces a molecular definition for atypical teratoid/rhabdoid tumors, and the new entity embryonal tumor with multilayered rosettes, C19MCaltered. • Gene expression and DNA methylation profiling have been and continue to be powerful tools in identifying molecular subgroups and therapeutic targets in pediatric brain tumors. • Current and upcoming clinical trials are incorporating a variety of genetic determinants (mutation status and molecular-based tumor subgrouping) into treatment stratification schema. 754 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

locoregional infiltration. A gliomatosis-cerebri–type pattern of diffuse parenchymal involvement, leptomeningeal, and/ or intraventricular spread may also be present. K27M-mutant DMGs present histologically as infiltrative gliomas, typically with an astrocytic cytomorphology. Though some contain deceptively bland tumor cells, others harbor cells with striking pleomorphism. The WHO 2016 specifies that “mitotic activity is present in most cases, but is not necessary for the diagnosis; microvascular proliferation and necrosis may be seen.”3 That said, DMG, H3 K27M-mutant may histologically resemble diffuse glioma ranging from low-grade diffuse astrocytoma to glioblastoma (Fig. 1B). For DIPGs, there is evidence to support that these variable histologic appearances are not predictive of clinical outcome independent of H3 K27M-mutant status, although this same association has not been established for DMGs arising elsewhere.7,8 Detectable H3 K27M mutation in a pediatric DMG correlates with a much worse prognosis in comparison with DMGs lacking this signature.7,9 In DIPGs, some groups have identified distinct molecular subgroups, and there is evidence to suggest that the type of H3 K27M mutation (H3.3 vs. H3.1) may convey variable tumor phenotype and prognosis.4,10

NEW CLASSIFICATION FOR CNS TUMORS: IMPLICATIONS FOR DIAGNOSIS AND THERAPY

A variety of additional alterations of genes involved in chromatin and transcription regulation (ATRX, BCOR, and MYC) and the RAS–phosphoinositide 3-kinase, Rb, and TP53 pathways has been demonstrated in pediatric DMG, either in addition to or independent of H3 K27M mutation.11,12 This emphasizes the fact that not all pediatric DMGs qualify for the pathologic diagnosis of DMG, H3 K27M-mutant. To be designated as such requires demonstration of H3 K27M mutation. This may be accomplished by direct sequencing, but it is also readily demonstrated via immunohistochemistry; mutant-specific antibody targeting H3 K27M (which detects both H3.3 and H3.1 K27M mutation) will show diffuse nuclear positivity (Fig. 1C), whereas H3 K27me3 antibody will show loss of nuclear staining.13,14 Pediatric DMGs lacking the signature H3 K27M mutation should be histomorphologically classified (astrocytoma vs. oligodendroglioma) and graded (WHO grade II to IV), followed by assessment of 1p/19q codeletion and/or isocitrate dehydrogenase status as appropriate based on histologic findings. This is therefore similar to the histologic-molecular workup of adult diffuse gliomas in accordance with the WHO 2016 classification schema.3

Pediatric Ependymomas

Ependymomas represent the third most common pediatric brain tumor, accounting for 30% of intracranial tumors in children younger than 3 years. The focus to refine the accurate classification of ependymomas has shifted from one centered primarily upon developing objective histologic grading criteria to one that takes advantage of genetic and/ or epigenetic classifiers. Ependymomas do not share a unifying molecular signature. On the contrary, ependymomas have been shown to represent multiple genetically distinct subsets, relative to age of occurrence, location, and biologic potential.15-17 To that end, the WHO 2016 introduced a new molecularly defined entity: ependymoma, RELA fusion-positive.3 This genetically defined ependymoma accounts for 70%

of pediatric supratentorial ependymomas, although it has occasionally been encountered in adults.18,19 The oncogenic fusion C11orf95-RELA fusion is the most commonly demonstrated alteration, resulting in aberrant activation of the nuclear factor-κB signaling pathway.18-20 These fusions often arise through chromothripsis, and occasionally C11orf95 or RELA may fuse with other partners.19 RELA fusion-positive ependymomas may histologically resemble other conventional ependymomas, though clear cell morphology and branching capillaries are often present20 (Fig. 2A). L1CAM expression detectable by immunohistochemistry (Fig. 2B) correlates well with the presence of RELA fusion19; however, definitive diagnosis requires demonstration of signature fusions. Given the varied array of breakpoints and rare alternate fusion partners, fluorescence in situ hybridization (FISH) represents the current method of choice for assessment. Typically, break-apart FISH assays using locus-specific probe pairs targeting opposing ends of the RELA and C11orf95 regions are used, with fusions demonstrated as split signals as seen in Fig. 2C. The importance in identification of RELA fusion-positive ependymomas stems from a large multi-institutional study in which this group of genetically defined supratentorial ependymomas represented the most biologically aggressive of the supratentorial ependymal tumors. In that study, supratentorial subependymomas and ependymomas with Yes-associated protein 1 fusions both had comparably better prognoses.15 The WHO 2016 does not prescribe specific molecular testing to either diagnose or otherwise subgroup other ependymal tumors; however, it emphasizes that multiple groups have independently identified biologically and molecularly distinct subgroups of pediatric infratentorial ependymomas; posterior fossa (PF)–group A tumors arise in infants/young children and are biologically aggressive akin to RELA fusion-positive ependymomas, whereas PF-group B tumors arise in older children and have a better prognosis.21-23 Recent studies indicate that PF-group A ependymomas exhibit a distinct epigenetic phenotype, and

FIGURE 2. Example of a Pediatric Supratentorial Ependymoma, RELA-Fusion Positive

On histologic examination (A), these lesions often exhibit delicate branched capillaries and clear cell appearance. (B) Dual-color FISH using probes flanking the RELA locus show single red and green signals (break-apart), confirming the presence of RELA fusion.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 755

FULLER, JONES, AND KIERAN

assessment of H3 K27me3 status by IHC shows promise in differentiating between group A and B tumors.24,25 It is likely that subsequent WHO updates will incorporate additional ependymoma genetic subgrouping in line with the aforementioned findings.

Pediatric Embryonal Tumors

Medulloblastoma. Medulloblastoma (MB) is a primitive CNS embryonal tumor arising in the cerebellum. Historically, the management of patients with MB has centered upon traditional prognostic factors for patient risk stratification: age, extent of resection, presence of metastatic disease, and histology.26,27 On the basis of these parameters, current multimodality treatments afford standard-risk patients with MB 5-year event-free survival rates of 79% to 85%, whereas high-risk patients have a 5-year event-free survival rate of 55% to 70%.28-31 A limitation of traditional MB risk stratification is that it did not account for the molecular heterogeneity of this disease. Multiple groups (Pomeroy et al,32 Cho et al,33 Kool et al,34 and Northcott et al35) have elucidated the MB (epi)genomic landscape. This led to a consensus recognition of four genetically defined subgroups, each with a distinct gene expression, mutation/copy-number alteration, and methylome profile, as well as typical patient demographic, histologic, and prognostic features.36 In response, the WHO 2016 included two distinct diagnostic classifications for MBs: one genetically defined and the other histologically defined.3 The WHO 2016 histologically defined MB classification is essentially a recapitulation of well-defined microscopic definitions present in prior classifications2; these include classic, desmoplastic/nodular (DN), large-cell/anaplastic (LCA) MB, and MB with extensive nodularity.3 Histologic classification provides relevant clinical information in cases in which molecular classification cannot be performed. For instance, extensively nodular and DN MBs correlate exclusively with a sonic hedgehog (SHH)–activated signature, whereas the majority of LCA MB are either group 3 or SHH activated.37 LCA MB has repeatedly been shown to correspond to aggressive biologic behavior.38,39 The four WHO 2016 genetically defined MB subgroups are as follows: Medulloblastoma, WNT-activated Accounting for 10% of MB, WNT-activated tumors arise in older children and adults.40,41 Most are classic histology; survival rates range from 90% to 95% with standard therapy, with rare LCA WNT-activated MB retaining a good prognosis.37,42,43 IHC demonstration of nuclear β-catenin in tumor cells is a useful method to identify WNT-activated MB in the clinical laboratory, although the WHO 2016 states optimal evaluation combines IHC β-catenin analysis with detection of monosomy 6 or CTNNB1 mutation.34,39 Other recurrent alterations include APC, DDX3X, SMARCA4, and TP53 mutations.44 TP53 mutations do not appear to confer a worse prognosis in the WNT-activated group.45 756 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Medulloblastoma, SHH-activated SHH-activated tumors account for 30% of MBs and exhibit a heterogeneous biologic potential based on several additional factors, especially TP53 mutation status.36,45 Although gene expression or methylation profiling remain the gold standard for MB subgroup identification, GAB1 and Yes-associated protein 1 detection by IHC provides a surrogate assay for identification of SHH-activated status for purposes of pathologic workup37 (Fig. 3). Medulloblastoma, SHH-activated and TP53 wild-type SHH-activated, TP53 wild-type MBs are typified by mutations involving PTCH1, SMO, or SUFU; frequent copy-number alterations include PTCH1 or chromosome 10q loss.46 These MBs exhibit a bimodal age distribution, targeting infants and adults.44 Histologically, extensively nodular and DN MBs predominate and are associated with a good prognosis45,47; the biologic potential of those with LCA or classic histomorphology is less defined. Medulloblastoma, SHH-activated and TP53-mutant In addition to a defining somatic or germline mutation of TP53, SHH-activated, TP53-mutant MBs often harbor amplifications of GLI2, MYCN, or SHH.46 LCA morphology, 17p loss, and metastatic disease at presentation are more common in this subgroup, most tumors arising in children 4 to 17 years.45,46 Diffuse nuclear p53 accumulation by IHC correlates well with the presence of TP53 mutation.48 These tumors are associated with poor clinical outcomes and are typically quite refractory to conventional and even SMO-targeted therapies.45,46 Medulloblastoma, non-WNT/non-SHH The WHO 2016 includes MB group 3 and group 4 as provisional variants in the non-WNT/non-SHH subgroup, as these two groups are not as genetically well separated as WNT- and SHH-activated subgroups.33,49 Although they have been shown to cluster into two groups based upon their gene expression or methylomic signatures, neither group 3 nor group 4 MB have specific driving signaling pathways. IHC assessment of GAB1, Yes-associated protein 1, and β-catenin are useful in excluding these MBs from SHH and WNT subgroups; however, gene expression or methylation profiling is required for definitive subgroup identification. Group 3 Group 3 tumors account for approximately 20% of MBs, arising nearly exclusively in children, particularly infants.34 This MB group carries the worst prognosis, with almost 50% of patients presenting with metastatic disease.34,36 Most are classic or LCA histology. MYC amplification and isochromosome 17q are frequent cytogenetic findings.34 Group 4 Group 4 accounts for the majority of MB cases (approximately 40%), occurs more commonly in children at a

NEW CLASSIFICATION FOR CNS TUMORS: IMPLICATIONS FOR DIAGNOSIS AND THERAPY

FIGURE 3. Example of an MB, SHH-Activated, and TP53-Mutant

This MB exhibited large-cell/anaplastic histology (A). Immunohistochemistry for β-catenin (B) showed only cytoplasmic staining; however, the lesion was diffusely positive for GAB1 (C). Pathologic diffuse nuclear staining for p53 was also seen by IHC (D), and TP53 mutation was subsequently demonstrated by targeted sequencing.

peak age of 10s, and affects males three times as often as females.36 Most tumors are classic histology. Up to 80% harbor copy-number alterations on chromosome 17 including 17p deletion, 17q gain, or isochromosome 17q.36 MYCN amplifications may also be encountered. Prognosis is variable and overall intermediate; almost one-third of patients present with metastatic disease at diagnosis.34,36 Other embryonal tumors. Embryonal tumor with multilayered rosettes, C19MC-altered is another new entity introduced in WHO 2016. These aggressive embryonal tu-

mors arise in infants throughout the CNS, sharing a genetic signature unique to this group, namely, amplification involving a cluster of microRNAs termed C19MC.50,51 High-resolution molecular techniques established that most tumors previously classified as ependymoblastoma, medulloepithelioma, and embryonal tumor with abundant neuropil and true rosettes, exhibit these C19MC alterations and comprise a single clinicopathologic entity.52,53 In fact, any CNS embryonal tumor with demonstrable C19MC alteration qualifies for this diagnosis under the WHO 2016.53 C19MC amplification is readily identified by FISH analysis and tends to correlate with increased expression of LIN28A by IHC; only asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 757

FULLER, JONES, AND KIERAN

demonstration of the former is diagnostic, however, as LIN28A expression may rarely be encountered with alternate pathologies.53,54 The diagnostic categories of Embryonal Tumor with Multilayered Rosettes, Not Otherwise Specified (NOS), and Medulloepithelioma are reserved for those tumors lacking C19MC abnormalities but with pathology otherwise typical of these diagnoses. The WHO 2016 made additional updates relative to CNS embryonal tumors. The diagnosis of atypical teratoid/rhabdoid tumor (AT/RT) now requires demonstration of inactivation of either SMARCB1 (INI1) or SMARCA4 (BRG1). This is readily accomplished through IHC assessment of the respective nuclear proteins. The phrase “CNS embryonal tumor with rhabdoid features” is applied for pathologically similar tumors when INI1 and BRG1 are found intact. Lastly, embryonal tumors that do not qualify for any of the aforementioned diagnoses are classified as CNS embryonal tumor, NOS; the phrase “primitive neuroectodermal tumor” has been completely abandoned.

As noted above, the WHO 2016 took steps forward in acknowledging the importance of molecular subgrouping in selected entities, with several classes now linked with defined alterations. Although this is an important first step, the relationship between genetic alterations and defined molecular classes is not always a clear-cut one-to-one match. For example, BRAF V600E mutation occurs in a variety of distinct histologic and molecular groups. Thus, more comprehensive methods of subgrouping, looking at (epi) genome- or transcriptome-wide profiles, are becoming an increasingly important part of the diagnostic toolbox. In this review, we provide a short overview of some of the molecular classification approaches that have been or are currently being applied to pediatric brain tumors and an outlook as to where these methods may lead us in the future.

core groups. An attempt to make this expression-based subgrouping more applicable to routine practice involved the development of a NanoString-based, targeted expression assay, which demonstrated robust performance on suboptimal samples.57 This has been somewhat superseded by other methodologies, however, and is not currently widely used. Global transcriptome analysis has identified molecular subgroups in a variety of other pediatric brain tumors. In glioblastoma, gene expression analysis revealed prognostically distinct subgroups and also demonstrated the dramatic differences compared with adult counterparts that would later be corroborated by other profiling techniques.58,59 In ependymoma, a global overview of tumors across various anatomic sites indicated distinct molecular subgroups linked with defined cellular compartments and patterns of oncogenic drivers.60 A large study of PF ependymoma further defined two principle subsets of ependymoma occurring in this location, termed PF-A and PF-B, differing in terms of copy-number alterations, age distribution, and prognosis.21 Overall, gene expression profiling benefited from an earlymover advantage in terms of the maturity of the technology becoming the first method of choice for molecular subgrouping; it remains a powerful tool for exploring tumor heterogeneity. Several groups originally identified using this method have stood the test of time and remain the backbone of current schema. Limitations, however, include (1) a need for good-quality RNA to conduct the profiling (restricting the use of archival samples), and (2) the risk that extremes of expression in stromal or inflammatory cells can dilute some of the signal in samples with lower tumor cell purity. Thus, gene expression analysis has not truly entered routine diagnostic use. It remains to be seen whether newer technologies such as RNA sequencing as a method of transcriptome-based classification (independent of its clear usefulness in terms of, for example, fusion gene identification) may enter the future diagnostic arena.

Gene Expression Profiling

DNA Methylation Analysis

APPROACHES TO MOLECULAR AND GENOMIC CLASSIFICATION IN PEDIATRIC BRAIN TUMORS

One of the earliest technologies to bear fruit in terms of molecular subgrouping of pediatric brain tumors was the gene expression microarray, allowing for a global profiling of the bulk tumor transcriptome. A seminal study by Pomeroy et al32 demonstrated the power of such profiling to distinguish clear biologic subtypes of embryonal tumors (MB, AT/RT, and primitive neuroectodermal tumors [PNETs]) that also had strong prognostic power. This study was the first to define a uniform SHH-activated MB subgroup that would later become one of the core subgroups of this disease. Subsequent transcriptome-profiling studies defined slightly different numbers (between four and six) of MB molecular groups, with WNT- and SHH-activated groups proving more consistent than non-WNT/non-SHH tumors.49,55,56 The current international consensus defines four groups, with generically named group 3 and group 4 as well as WNTand SHH-driven tumors,36 although new studies on larger cohorts are starting to identify substructure within these 758 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

The technology that has now somewhat overtaken expression profiling in the molecular subgrouping arena is DNA methylation analysis. The range of pediatric brain tumors that have been subjected to such profiling is steadily increasing, with most of the main entities covered to a greater or lesser extent. In MB, it was shown that DNA methylation analysis can recapitulate the expression-based subgroups in a highly robust manner.61,62 Subsequent studies in other tumors either confirmed this close association with gene expression or revealed further structure that had not previously been identified through transcriptomic analyses. For example, location-specific differences in the genomic profiles of pilocytic astrocytoma, as previously alluded to through gene expression,63,64 were confirmed at an epigenetic level.65 DNA methylation groups of pediatric glioblastoma were found to precisely match to the distinct histone 3 mutation types seen in this disease.66 Furthermore, a subset of histone wild-type glioblastomas were found to show

NEW CLASSIFICATION FOR CNS TUMORS: IMPLICATIONS FOR DIAGNOSIS AND THERAPY

methylation patterns more closely resembling pleomorphic xanthoastrocytoma, with BRAF V600E mutations, 9p21 deletions, and a better outcome.9 In ependymoma, a large study looking at tumors across ages and locations led to an integrated scheme of nine molecular subgroups: three each in the spinal, PF, and supratentorial compartments.15 Many of the subgroups revealed in these different tumor types have been shown to be linked with defined genetic alterations that may represent drug targets, convey prognostic significance, or both. The power of methylation-based analysis to identify novel tumor entities in an unbiased way was demonstrated on a recent study of samples histologically diagnosed as PNETs.67 As well as highlighting a high degree of misdiagnoses of other known entities, four novel groups were defined that each had a striking association with a clear pathogenetic mechanism, thereby expanding the catalog of CNS tumor types. This was part of the reason why the term PNET has now been abandoned in WHO 2016. Based on these positive examples, work is ongoing to produce a broadly applicable, pan-brain cancer classification scheme based on DNA methylation array data. It is thought that the principle of epigenetic subgrouping works so well because the chromatin structure and DNA marks retain a fingerprint or memory of the developmental decisions made in the life history of the tumor cell of origin, making it principally suitable across all tumor types. To improve access to such a system for any center generating such data, the Heidelberg groups have made this tool freely available online in the form of a web-based analysis and reporting system (www.molecularneuropathology.org). The site, and the algorithm behind it, will be continuously updated as new insights into novel subgroups are obtained. The benefits of array-based DNA methylation analysis include minimal tissue requirements and robust performance from archival, paraffin-embedded material. Although it suffers from the same problems of normal cell contamination as gene expression profiling, DNA methylation data are partly buffered from this effect by the largely binary nature of most CpG sites in the genome, being either fully methylated or completely unmethylated. Although there may be exceptions in which somatic alterations impose an extreme shift in DNA methylation, such as in the CpG island methylator phenotype seen in the presence of isocitrate dehydrogenase mutations, the fingerprint of cellular origins appears to be clear enough in most cases that it remains constant both spatially throughout a tumor tissue and temporally from primary to relapse samples. It is currently therefore the method of choice for many molecular stratification applications.

distinct DNA copy-number alterations on prognosis within the consensus MB subgroups and identified marker combinations that may provide an added level of predictive power.68 Such alterations have the added benefit of being detectable by FISH, a more widely available assay technique. Though some changes are almost pathognomonic, such as isodicentric 17q in MB or 7q34 duplication in pilocytic astrocytoma, other alterations are found across a spectrum of entities (e.g., gain of 1q in multiple histologies). Thus, it is likely that copy-number changes will predominantly be of use only in concert with other subgrouping approaches. Next-generation sequencing–based approaches also show tremendous promise in terms of both classification and the ability to detect potential therapeutic targets. Although deep whole-genome sequencing is currently restricted in its clinical applications for both technical and cost reasons, sequencing targeted panels of cancer-specific genes are becoming widely available (for example, Kline et al,69 Sahm et al,70 and Ramkissoon et al71). Global mutational profiling can also provide information about signatures of mutagenic processes acting in a given cell,72 which may prove to be of diagnostic or therapeutic use in the near future. For example, signatures of BRCAness have been suggested as predictive markers for poly-ADP ribose polymerase inhibitors (reviewed in Lord and Ashworth73), whereas hypermutated tumors may respond better to immune checkpoint inhibitors.74 Sequencing of histone modifications in their chromatin context via chromatin immunoprecipitation sequencing has recently displayed its use in classifying tumors based on an epigenetic fingerprint of cellular wiring. Mapping of enhancer elements using the H3K27Ac mark, for example, confirmed the previously described structure of MB subgroups75 and delineated among three different subsets of AT/RT.76 Looking in closer detail at so-called superenhancer elements,77 with substantial accumulation of K27Ac in a defined region, can also provide extremely valuable information as to the core regulatory transcription factor networks active in a cell and give additional hints as to precise cells of origin for the different molecular subgroups.75 Finally, the application of single-cell sequencing techniques, such as those described by Patel et al78 and elsewhere, has the potential to inform on intertumoral as well as intratumoral heterogeneity at an exquisitely detailed resolution. This step-change in the power to interrogate functionally distinct cellular subclones and aspects such as tumor-microenvironmental interplay will likely further modify the landscape of molecular classification moving forward.

Further Classification Approaches

IMPACT OF WHO 2016 ON CLINICAL TRIALS IN PEDIATRIC NEURO-ONCOLOGY

Although gene expression and DNA methylation have been the most widely adopted characterization tools because of their relative ease of use, low cost, and scalability, multiple alternative tools are being used that promise to give an ever more detailed picture of molecular heterogeneity. For example, one recent study investigated the impact of

The ability to translate the notable advancements in the classification of pediatric CNS tumors into improved patient treatment, defined by improved survival and/or decreased morbidity, unfortunately continues to lag significantly behind. The next decade will likely see a major change in asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 759

FULLER, JONES, AND KIERAN

how clinical trials are performed as we recognize the importance of specific pathways in tumorigenesis and the increasing availability of therapies that target those pathways. We are moving away from the nonspecific and toxic approaches of radiation and chemotherapy and toward precision medicine; only a few examples of this are currently available. Many of the drugs beginning to impact clinical trials are pathway inhibitors, not true targeted drugs. Perhaps the best example of a true targeted inhibitor exemplifying the goals of precision medicine is the BRAF V600E inhibitors vemurafenib or dabrafenib. These agents recognize the BRAF V600E point mutation with exquisite specificity, resulting in inhibition of its signaling. By contrast, many downstream mitogen-activated protein kinase kinase or nonspecific epigenetic agents such as histone deacetylase inhibitors affect broad pathways but are not specific for the mutations they are targeting (i.e., BRAF-KIAA1549, H3K27M, or INI1).

High-Grade Gliomas

The improved understanding of the mutational profiles of pediatric high-grade gliomas (HGGs) and their distinction from adult HGGs has become an important guidepost in the development of pediatric clinical trials. The differences between the tumors in these two age groups highlight the need for separate trials in many circ*mstances, especially in relation to isocitrate dehydrogenase 1 and epidermal growth factor receptor VIII mutations that are common in adult tumors but rare in their pediatric counterparts.79 Similarly, although the role of methylguanine-DNA methyltransferase expression is an important prognostic and treatment component of adult HGG,80 it has not been as clearly impactful in pediatric patients, perhaps secondary to the lower incidence of methylguanine-DNA methyltransferase promoter methylation that supports a more resistant phenotype to agents such as temozolomide. Other than BRAF V600E mutation in a small percentage of pediatric HGGs, targetable mutations in these tumors are limited and thus, so are meaningful clinical trials. The WHO 2016 recognition of H3K27M-mutated midline glioma adds a molecular component to the pathologic classification of these tumors,3 but not a clear treatment option. Pathway targets such as histone deacetylase inhibitors are being tested in several upfront and relapsed clinical trials, but to date have not changed the outcome of these aggressive pediatric gliomas. The remainder of the molecular analyses routinely being performed (such as p53, ATRX, platelet-derived growth factor receptor, etc.) does not have currently available effective inhibitors that penetrate the CNS. One important change in the WHO 20163 is the recognition of anaplastic pleomorphic xanthoastrocytoma, a WHO grade III tumor that can now be treated as an HGG. Previously, pleomorphic xanthoastrocytoma, regardless of anaplasia, was considered grade II and thus not eligible for HGG trials. Many of these patients also possess the BRAF V600E mutation,81 most diverted from typical HGG trials to those focused on targeting this specific mutation. 760 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Low-Grade Gliomas

The WHO 2016 made very few changes in the low-grade glioma (LGG) classification schema. Pilomyxoid astrocytoma, previously grade II, was reassigned as a variant of grade I pilocytic astrocytoma3 but this change is unlikely to affect most clinical trials because grade I and II LGGs are treated similarly. The major advances in the molecular classification of pediatric LGGs (presence of the BRAF V600E point mutation in 10% and the BRAF-KIAA1549 truncated fusion duplication in 75%) as well as rarer mutations are mentioned in the WHO 2016, though were not included as entity or subgroup-defining classifiers.82 Nonetheless, many pediatric LGG clinical trials are now incorporating mutation status into treatment stratification. For example, BRAF V600E-mutated LGGs are going on to targeted trials through several different consortia, predominantly at the time of relapse. For the truncated fusion BRAF-KIAA1549 forms, which are paradoxically stimulated by BRAF V600E drugs and thus contraindicated for this group of patients, downstream inhibitors of this pathway are undergoing clinical testing. This includes mitogen-activated protein kinase kinase and mTOR inhibitors, both of which are being used in recurrent/ progressive disease. The one exception is the use of mTOR inhibitors in tuberous sclerosis-associated subependymal giant cell astrocytomas. A smaller percentage of pediatric LGGs have non-BRAF–mutated targets such as FGFR1, NTRK, or others. Though not diagnostic of specific tumors, status for these mutations will likely be included in basket trials of multiple agents.

Ependymoma

The WHO 2016 recognized the new molecularly defined RELA fusion-positive ependymoma. Although this molecular subgroup has some prognostic implication, there are no specific treatment options that currently affect clinical trials. Other recently recognized molecular subtypes of ependymoma,83 including Yes-associated protein 1 supratentorial and group A and B infratentorial tumors, were mentioned in the WHO 2016, though were not codified as diagnostic entities/subgroups. Ependymomas are therefore still staged and treated on standard existing criteria (degree of resection, location, and presence of anaplasia).

Medulloblastoma

The WHO 2016 introduced considerable changes to MB classification, some of which may affect clinical trials. Four new MB molecular subgroups were introduced as noted above; this WHO 2016 schema differs somewhat from the four genomic consensus categories widely used around the world.36 MB WNT-activated, the best prognostic group, is now being evaluated in a series of radiation, chemotherapy, or combined radiation/chemotherapy reduction strategies. The treatment of these patients requires not only demonstration of nuclear β-catenin staining by IHC, but also sequencing confirmation of CTNNB1 mutation along with monosomy 6. Similarly, although SHH tumors can be classified by a number of validated approaches,84 not all tumors will be

NEW CLASSIFICATION FOR CNS TUMORS: IMPLICATIONS FOR DIAGNOSIS AND THERAPY

appropriate for SHH-targeted therapy, as the current array of approved drugs targeting this pathway does not inhibit signaling for three (SuFu, Gli, and Mycn) of the five (Ptch and Smo) common genes involved.46 The WHO 2016 identifies a new category of MB SHH-activated, p53-mutated based on a number of papers suggesting a very poor outcome in this group,45 and an international amendment of the consensus grouping will likely soon follow. The final WHO 2016 category includes all non-WNT, non-SHH MBs (combined group 3 and 4) in one class. Targeted approaches for groups 3 and 4 have not yet been part of major clinical trials. The recognized poor prognosis of MYC-amplified group 3 MBs68 is not included in the WHO 2016; however, as new therapeutic interventions targeting MYC become available, differentiating this subgroup from other non-MYC–amplified patients in groups 3 and 4, in whom the prognosis is intermediate, will become important. A major change in the WHO 2016 is the reclassification of what were previously referred to as CNS PNETs and are now grouped under multiple headings. One recognized subgroup of these, those with the C19MC amplification, has become a new entity. For the other tumors of this class, the general category of embryonal tumor NOS must suffice for clinical trial entry. Historically, clinical trials have treated these tumors (under the heading of PNET) like high-risk MB and pineoblastoma. What will continue to complicate the clinical trial development of this class of heterogeneous tumors is the growing recognition that many of them are probably

not true ETs, but rather undifferentiated glioblastoma multiforme, ependymomas, and others.67

CONCLUSION

The many examples in the literature, and touched upon above, highlight the power of molecular subgrouping as an additional tool in the diagnostician’s armamentarium. In most cases, this additional biologic information clearly enhances patient stratification in terms of outcome prediction and identification of therapeutic vulnerabilities. It is almost inevitable that more complex genome-wide profiling of methylomes/transcriptomes and molecular alterations will increasingly enter the diagnostic routine and likewise upcoming versions of the WHO classification. This naturally raises issues of access to necessary technology and expertise and the broad applicability of classification schema in different socioeconomic regions. These issues will hopefully be resolved, as costs come down and more laboratories offer such testing. In the meantime, the added value of molecular stratification means that it is incumbent on us to ensure that diagnostic standards follow and adapt to the latest research findings as much as is practicable, without resorting to a lower common denominator. Although this may make it challenging to compare with historical epidemiologic data and previous clinical trial outcomes, these hurdles are outweighed by the long-term benefit to patients that can be expected from precision diagnostics and better matching to targeted therapies.

References 1. Louis DN, Perry A, Burger P, et al; International Society Of Neuropathology--Haarlem. International Society Of Neuropathology-Haarlem consensus guidelines for nervous system tumor classification and grading. Brain Pathol. 2014;24:429-435.

9. Korshunov A, Ryzhova M, Hovestadt V, et al. Integrated analysis of pediatric glioblastoma reveals a subset of biologically favorable tumors with associated molecular prognostic markers. Acta Neuropathol. 2015;129:669-678.

2. Louis DN, Ohgaki H, Wiestler OD, et al. WHO Classification of Tumours of the Central Nervous System. Lyon: IARC Press; 2007.

10. Castel D, Philippe C, Calmon R, et al. Histone H3F3A and HIST1H3B K27M mutations define two subgroups of diffuse intrinsic pontine gliomas with different prognosis and phenotypes. Acta Neuropathol. 2015;130:815-827.

3. Louis DN, Ohgaki H, Wiestler OD, et al. WHO Classification of Tumours of the Central Nervous System. Lyon: IARC Press; 2016. 4. Buczkowicz P, Hoeman C, Rakopoulos P, et al. Genomic analysis of diffuse intrinsic pontine gliomas identifies three molecular subgroups and recurrent activating ACVR1 mutations. Nat Genet. 2014;46:451456. 5. Fontebasso AM, Papillon-Cavanagh S, Schwartzentruber J, et al. Recurrent somatic mutations in ACVR1 in pediatric midline high-grade astrocytoma. Nat Genet. 2014;46:462-466. 6. Bender S, Tang Y, Lindroth AM, et al. Reduced H3K27me3 and DNA hypomethylation are major drivers of gene expression in K27M mutant pediatric high-grade gliomas. Cancer Cell. 2013;24:660-672.

11. Wu G, Broniscer A, McEachron TA, et al; St. Jude Children’s Research Hospital–Washington University Pediatric Cancer Genome Project. Somatic histone H3 alterations in pediatric diffuse intrinsic pontine gliomas and non-brainstem glioblastomas. Nat Genet. 2012;44:251253. 12. Schwartzentruber J, Korshunov A, Liu XY, et al. Driver mutations in histone H3.3 and chromatin remodelling genes in paediatric glioblastoma. Nature. 2012;482:226-231. 13. Bechet D, Gielen GG, Korshunov A, et al. Specific detection of methionine 27 mutation in histone 3 variants (H3K27M) in fixed tissue from high-grade astrocytomas. Acta Neuropathol. 2014;128:733-741.

7. Khuong-Quang DA, Buczkowicz P, Rakopoulos P, et al. K27M mutation in histone H3.3 defines clinically and biologically distinct subgroups of pediatric diffuse intrinsic pontine gliomas. Acta Neuropathol. 2012;124:439-447.

14. Venneti S, Santi M, Felicella MM, et al. A sensitive and specific histopathologic prognostic marker for H3F3A K27M mutant pediatric glioblastomas. Acta Neuropathol. 2014;128:743-753.

8. Buczkowicz P, Bartels U, Bouffet E, et al. Histopathological spectrum of paediatric diffuse intrinsic pontine glioma: diagnostic and therapeutic implications. Acta Neuropathol. 2014;128:573-581.

15. Pajtler KW, Witt H, Sill M, et al. Molecular classification of ependymal tumors across all CNS compartments, histopathological grades, and age groups. Cancer Cell. 2015;27:728-743.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 761

FULLER, JONES, AND KIERAN

16. Palm T, Figarella-Branger D, Chapon F, et al. Expression profiling of ependymomas unravels localization and tumor grade-specific tumorigenesis. Cancer. 2009;115:3955-3968. 17. Taylor MD, Poppleton H, Fuller C, et al. Radial glia cells are candidate stem cells of ependymoma. Cancer Cell. 2005;8:323-335.

34. Kool M, Korshunov A, Remke M, et al. Molecular subgroups of medulloblastoma: an international meta-analysis of transcriptome, genetic aberrations, and clinical data of WNT, SHH, Group 3, and Group 4 medulloblastomas. Acta Neuropathol. 2012;123:473-484. 35. Northcott PA, Korshunov A, Witt H, et al. Medulloblastoma comprises four distinct molecular variants. J Clin Oncol. 2011;29:1408-1414.

18. Pietsch T, Wohlers I, Goschzik T, et al. Supratentorial ependymomas of childhood carry C11orf95-RELA fusions leading to pathological activation of the NF-κB signaling pathway. Acta Neuropathol. 2014;127:609-611.

36. Taylor MD, Northcott PA, Korshunov A, et al. Molecular subgroups of medulloblastoma: the current consensus. Acta Neuropathol. 2012;123:465-472.

19. Parker M, Mohankumar KM, Punchihewa C, et al. C11orf95-RELA fusions drive oncogenic NF-κB signalling in ependymoma. Nature. 2014;506:451-455.

37. Ellison DW, Dalton J, Kocak M, et al. Medulloblastoma: clinicopathological correlates of SHH, WNT, and non-SHH/WNT molecular subgroups. Acta Neuropathol. 2011;121:381-396.

20. Figarella-Branger D, Lechapt-Zalcman E, Tabouret E, et al. Supratentorial clear cell ependymomas with branching capillaries demonstrate characteristic clinicopathological features and pathological activation of nuclear factor-kappaB signaling. Neuro-oncol. 2016;18:919-927.

38. Brown HG, Kepner JL, Perlman EJ, et al. “Large cell/anaplastic” medulloblastomas: a Pediatric Oncology Group Study. J Neuropathol Exp Neurol. 2000;59:857-865.

21. Witt H, Mack SC, Ryzhova M, et al. Delineation of two clinically and molecularly distinct subgroups of posterior fossa ependymoma. Cancer Cell. 2011;20:143-157. 22. Wani K, Armstrong TS, Vera-Bolanos E, et al; Collaborative Ependy moma Research Network. A prognostic gene expression signature in infratentorial ependymoma. Acta Neuropathol. 2012;123:727-738. 23. Hoffman LM, Donson AM, Nakachi I, et al. Molecular sub-groupspecific immunophenotypic changes are associated with outcome in recurrent posterior fossa ependymoma. Acta Neuropathol. 2014;127:731-745. 24. Mack SC, Witt H, Piro RM, et al. Epigenomic alterations define lethal CIMP-positive ependymomas of infancy. Nature. 2014;506:445-450. 25. Bayliss J, Mukherjee P, Lu C, et al. Lowered H3K27me3 and DNA hypomethylation define poorly prognostic pediatric posterior fossa ependymomas. Sci Transl Med. 2016;8:366ra161. 26. Zeltzer PM, Boyett JM, Finlay JL, et al. Metastasis stage, adjuvant treatment, and residual tumor are prognostic factors for medullo blastoma in children: conclusions from the Children’s Cancer Group 921 randomized phase III study. J Clin Oncol. 1999;17:832-845. 27. Eberhart CG, Kepner JL, Goldthwaite PT, et al. Histopathologic grading of medulloblastomas: a Pediatric Oncology Group study. Cancer. 2002;94:552-560. 28. Packer RJ, Goldwein J, Nicholson HS, et al. Treatment of children with medulloblastomas with reduced-dose craniospinal radiation therapy and adjuvant chemotherapy: a Children’s Cancer Group Study. J Clin Oncol. 1999;17:2127-2136. 29. Packer RJ, Gajjar A, Vezina G, et al. Phase III study of craniospinal radia tion therapy followed by adjuvant chemotherapy for newly diagnosed average-risk medulloblastoma. J Clin Oncol. 2006;24:4202-4208. 30. Packer RJ, Sutton LN, Elterman R, et al. Outcome for children with medulloblastoma treated with radiation and cisplatin, CCNU, and vincristine chemotherapy. J Neurosurg. 1994;81:690-698. 31. Gajjar AJ, Robinson GW. Medulloblastoma-translating discoveries from the bench to the bedside. Nat Rev Clin Oncol. 2014;11:714-722. 32. Pomeroy SL, Tamayo P, Gaasenbeek M, et al. Prediction of central nervous system embryonal tumour outcome based on gene expression. Nature. 2002;415:436-442. 33. Cho YJ, Tsherniak A, Tamayo P, et al. Integrative genomic analysis of medulloblastoma identifies a molecular subgroup that drives poor clinical outcome. J Clin Oncol. 2011;29:1424-1430.

762 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

39. Ellison DW, Kocak M, Dalton J, et al. Definition of diseaserisk stratification groups in childhood medulloblastoma using combined clinical, pathologic, and molecular variables. J Clin Oncol. 2011;29:1400-1407. 40. Pietsch T, Schmidt R, Remke M, et al. Prognostic significance of clinical, histopathological, and molecular characteristics of medulloblastomas in the prospective HIT2000 multicenter clinical trial cohort. Acta Neuropathol. 2014;128:137-149. 41. Gajjar A, Pfister SM, Taylor MD, et al. Molecular insights into pediatric brain tumors have the potential to transform therapy. Clin Cancer Res. 2014;20:5630-5640. 42. Fattet S, Haberler C, Legoix P, et al. Beta-catenin status in paediatric medulloblastomas: correlation of immunohistochemical expression with mutational status, genetic profiles, and clinical characteristics. J Pathol. 2009;218:86-94. 43. Ellison DW, Onilude OE, Lindsey JC, et al; United Kingdom Children’s Cancer Study Group Brain Tumour Committee. beta-Catenin status predicts a favorable outcome in childhood medulloblastoma: the United Kingdom Children’s Cancer Study Group Brain Tumour Committee. J Clin Oncol. 2005;23:7951-7957. 44. Northcott PA, Jones DT, Kool M, et al. Medulloblastomics: the end of the beginning. Nat Rev Cancer. 2012;12:818-834. 45. Zhukova N, Ramaswamy V, Remke M, et al. Subgroup-specific prognostic implications of TP53 mutation in medulloblastoma. J Clin Oncol. 2013;31:2927-2935. 46. Kool M, Jones DT, Jäger N, et al; ICGC PedBrain Tumor Project. Genome sequencing of SHH medulloblastoma predicts genotype-related response to smoothened inhibition. Cancer Cell. 2014;25:393-405. 47. Rutkowski S, von Hoff K, Emser A, et al. Survival and prognostic factors of early childhood medulloblastoma: an international meta-analysis. J Clin Oncol. 2010;28:4961-4968. 48. Tabori U, Baskin B, Shago M, et al. Universal poor survival in children with medulloblastoma harboring somatic TP53 mutations. J Clin Oncol. 2010;28:1345-1350. 49. Kool M, Koster J, Bunt J, et al. Integrated genomics identifies five medulloblastoma subtypes with distinct genetic profiles, pathway signatures and clinicopathological features. PLoS One. 2008;3:e3088. 50. Korshunov A, Remke M, Gessi M, et al. Focal genomic amplification at 19q13.42 comprises a powerful diagnostic marker for embryonal tumors with ependymoblastic rosettes. Acta Neuropathol. 2010;120:253-260.

NEW CLASSIFICATION FOR CNS TUMORS: IMPLICATIONS FOR DIAGNOSIS AND THERAPY

51. Kleinman CL, Gerges N, Papillon-Cavanagh S, et al. Fusion of TTYH1 with the C19MC microRNA cluster drives expression of a brain-specific DNMT3B isoform in the embryonal brain tumor ETMR. Nat Genet. 2014;46:39-44. 52. Korshunov A, Sturm D, Ryzhova M, et al. Embryonal tumor with abundant neuropil and true rosettes (ETANTR), ependymoblastoma, and medulloepithelioma share molecular similarity and comprise a single clinicopathological entity. Acta Neuropathol. 2014;128:279289. 53. Spence T, Sin-Chan P, Picard D, et al. CNS-PNETs with C19MC amplification and/or LIN28 expression comprise a distinct histogenetic diagnostic and therapeutic entity. Acta Neuropathol. 2014;128:291303. 54. Weingart MF, Roth JJ, Hutt-Cabezas M, et al. Disrupting LIN28 in atypical teratoid rhabdoid tumors reveals the importance of the mitogen activated protein kinase pathway as a therapeutic target. Oncotarget. 2015;6:3165-3177. 55. Northcott PA, Korshunov A, Witt H, et al. Medulloblastoma comprises four distinct molecular variants. J Clin Oncol. 2011;29:1408-1414. 56. Thompson MC, Fuller C, Hogg TL, et al. Genomics identifies medulloblastoma subgroups that are enriched for specific genetic alterations. J Clin Oncol. 2006;24:1924-1931. 57. Northcott PA, Shih DJH, Remke M, et al. Rapid, reliable, and reproducible molecular sub-grouping of clinical medulloblastoma samples. Acta Neuropathol. 2012;123:615-626. 58. Faury D, Nantel A, Dunn SE, et al. Molecular profiling identifies prognostic subgroups of pediatric glioblastoma and shows increased YB-1 expression in tumors. J Clin Oncol. 2007;25:1196-1208. 59. Paugh BS, Qu C, Jones C, et al. Integrated molecular genetic profiling of pediatric high-grade gliomas reveals key differences with the adult disease. J Clin Oncol. 2010;28:3061-3068. 60. Johnson RA, Wright KD, Poppleton H, et al. Cross-species genomics matches driver mutations and cell compartments to model ependy moma. Nature. 2010;466:632-636. 61. Hovestadt V, Remke M, Kool M, et al. Robust molecular subgrouping and copy-number profiling of medulloblastoma from small amounts of archival tumour material using high-density DNA methylation arrays. Acta Neuropathol. 2013;125:913-916. 62. Schwalbe EC, Williamson D, Lindsey JC, et al. DNA methylation profiling of medulloblastoma allows robust subclassification and improved outcome prediction using formalin-fixed biopsies. Acta Neuropathol. 2013;125:359-371. 63. Sharma MK, Mansur DB, Reifenberger G, et al. Distinct genetic signatures among pilocytic astrocytomas relate to their brain region origin. Cancer Res. 2007;67:890-900. 64. Tchoghandjian A, Fernandez C, Colin C, et al. Pilocytic astrocytoma of the optic pathway: a tumour deriving from radial glia cells with a specific gene signature. Brain. 2009;132:1523-1535. 65. Lambert SR, Witt H, Hovestadt V, et al. Differential expression and methylation of brain developmental genes define location-specific subsets of pilocytic astrocytoma. Acta Neuropathol. 2013;126:291301. 66. Sturm D, Witt H, Hovestadt V, et al. Hotspot mutations in H3F3A and IDH1 define distinct epigenetic and biological subgroups of glioblastoma. Cancer Cell. 2012;22:425-437.

67. Sturm D, Orr BA, Toprak UH, et al. New brain tumor entities emerge from molecular classification of CNS-PNETs. Cell. 2016;164:10601072. 68. Shih DJH, Northcott PA, Remke M, et al. Cytogenetic prognostication within medulloblastoma subgroups. J Clin Oncol. 2014;32:886-896. 69. Kline CN, Joseph NM, Grenert JP, et al. Targeted next-generation sequencing of pediatric neuro-oncology patients improves diagnosis, identifies pathogenic germline mutations, and directs targeted therapy. Neuro-oncol. Epub 2016 Nov 14. 70. Sahm F, Schrimpf D, Jones DT, et al. Next-generation sequencing in routine brain tumor diagnostics enables an integrated diagnosis and identifies actionable targets. Acta Neuropathol. 2016;131:903-910. 71. Ramkissoon SH, Bandopadhayay P, Hwang J, et al. Clinical targeted exome-based sequencing in combination with genome-wide copy number profiling: precision medicine analysis of 203 pediatric brain tumors. Neuro-oncol. Epub 2017 Jan 19. 72. Alexandrov LB, Nik-Zainal S, Wedge DC, et al; Australian Pancreatic Cancer Genome Initiative; ICGC Breast Cancer Consortium; ICGC MMML-Seq Consortium; ICGC PedBrain. Signatures of mutational processes in human cancer. Nature. 2013;500:415-421. 73. Lord CJ, Ashworth A. BRCAness revisited. Nat Rev Cancer. 2016;16:110120. 74. Bouffet E, Larouche V, Campbell BB, et al. Immune checkpoint inhibition for hypermutant glioblastoma multiforme resulting from germline biallelic mismatch repair deficiency. J Clin Oncol. 2016;34:2206-2211. 75. Lin CY, Erkek S, Tong Y, et al. Active medulloblastoma enhancers reveal subgroup-specific cellular origins. Nature. 2016;530:57-62. 76. Johann PD, Erkek S, Zapatka M, et al. Atypical teratoid/rhabdoid tumors are comprised of three epigenetic subgroups with distinct enhancer landscapes. Cancer Cell. 2016;29:379-393. 77. Whyte WA, Orlando DA, Hnisz D, et al. Master transcription factors and mediator establish super-enhancers at key cell identity genes. Cell. 2013;153:307-319. 78. Patel AP, Tirosh I, Trombetta JJ, et al. Single-cell RNA-seq highlights intratumoral heterogeneity in primary glioblastoma. Science. 2014;344:1396-1401. 79. Sturm D, Bender S, Jones DT, et al. Paediatric and adult glioblastoma: multiform (epi)genomic culprits emerge. Nat Rev Cancer. 2014;14:92107. 80. Hegi ME, Diserens AC, Gorlia T, et al. MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med. 2005;352:9971003. 81. Dias-Santagata D, Lam Q, Vernovsky K, et al. BRAF V600E mutations are common in pleomorphic xanthoastrocytoma: diagnostic and therapeutic implications. PLoS One. 2011;6:e17948. 82. Packer RJ, Pfister S, Bouffet E, et al. Pediatric low-grade gliomas: implications of the biologic era. Neuro-oncol. Epub 2016 Sept 28. 83. Pajtler KW, Mack SC, Ramaswamy V, et al. The current consensus on the clinical management of intracranial ependymoma and its distinct molecular variants. Acta Neuropathol. 2017;133:5-12. 84. Shou Y, Robinson DM, Amakye DD, et al. A five-gene hedgehog signature developed as a patient preselection tool for hedgehog inhibitor therapy in medulloblastoma. Clin Cancer Res. 2015;21:585593.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 763

PROFESSIONAL DEVELOPMENT

COLLABORATING WITH ADVANCED PRACTICE PROVIDERS

Collaborating With Advanced Practice Providers: Impact and Opportunity Heather M. Hylton, MS, PA-C, DFAAPA, and G. Lita Smith, DNP, RN, ACNP-C OVERVIEW Although significant progress has been made in cancer care, access to coordinated, high-quality care across the cancer care continuum remains a challenge for many patients. With significant workforce shortages in oncology anticipated, physician assistants (PAs) and nurse practitioners (NPs)—known collectively as advanced practice providers (APPs)—are considered to be a part of the solution to bridging the gap between the supply of and demand for oncology services. APPs are integral to the provision of team-based care in oncology, and optimizing the roles of all members of the patient’s care team is vital to ensuring the teams are cost-effective and that each team member is performing at the functional level intended. Studies have shown significant patient, physician, and APP satisfaction with collaborative care models, and APPs are well positioned to enhance value for patients in the oncology setting. Understanding the full scope of APP impact can be challenging as it extends well beyond direct patient care. As rapid progress in cancer care continues, innovative approaches to care delivery will be necessary to ensure patients’ access. Effective oncologist–APP partnerships will be key to providing optimal, value-centered care to patients.

n 2017, an estimated 1,688,780 new cases of cancer will be diagnosed in the United States.1 As of January 2016, there were more than 15 million cancer survivors in the United States, and this number is expected to exceed 20 million by 2026.2 Although significant progress has been made in cancer care, leading to marked improvement in survival, access to coordinated, high-quality care across the cancer care continuum remains a challenge for many patients. Additional challenges to the system, as eloquently described in the Institute of Medicine’s 2013 report, Delivering High-Quality Cancer Care Charting a New Course for a System in Crisis, include3: • An aging population, • The escalating cost of cancer care, • Complexity related to the marked advances in the understanding of cancer biology, • Limitations on tools to improve quality of care, and • Workforce shortages. These challenges are interwoven with the uncertainty of the current health care (and health care reform) landscape, changing reimbursem*nt models, productivity and access pressures, and provider burnout risk and have the potential to stress a health care system that is already overwhelmed. This, in turn, may lead to additional barriers to patients’ access to care. Significant workforce shortages are anticipated in the near future with the Association of American Medical Colleges projecting a shortage of 61,700 to 94,700 physicians

by 2025.4 In 2007, Erikson et al projected a shortfall of approximately 2,550 to 4,080 oncologists by 2020 based on their analysis of the supply of and demand for oncology services.5 A follow-up study, conducted by Yang and colleagues, projected the supply of and demand for oncology services extending out to 2025. This study was largely confirmatory of the work by Erikson and colleagues, with the anticipated workforce shortages being somewhat delayed than initially anticipated.6 PAs and NPs, known collectively as APPs, are highly trained and skilled health care providers and, as such, are an integral part of the health care team. APPs are able to provide a broad range of services in the oncology space and have been consistently identified as a part of the solution for bridging the anticipated gap between the supply of and demand for oncology services.5-7 Services APPs provide include, but are not limited to, those related to prevention, screening, surveillance, and diagnosis; supportive care during the course of active treatment; long-term follow-up and survivorship care, and counseling on disease specifics, treatment, and prevention. APPs are instrumental in conducting goals of care and prognosis discussions and providing care at the end of life. APPs also often participate in clinical research activities. APP educational preparation is of a more generalist nature (i.e., PAs) or population-focused nature (i.e., NPs). As such, specialty knowledge for APPs is largely acquired through a blend of self-directed learning, practice-based training, and mentoring. Effective and

From Memorial Sloan Kettering Cancer Center, New York, NY; University of Michigan, Ann Arbor, MI. Corresponding author: Heather M. Hylton, MS, PA-C, DFAAPA, Memorial Sloan Kettering Cancer Center, 1275 York Ave., Box 124, New York, NY 10065; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK e1

HYLTON AND SMITH

appropriate training of APPs in the oncology practice setting is key to APPs developing the necessary competencies for taking on expanded roles in the practice, such as ordering chemotherapy and performing a range of procedures.8 Successful integration of APPs has occurred within community and academic oncology settings, and patient satisfaction with a collaborative care model has been noted to be high.7,9 A recent retrospective study looking at adherence to ASCO’s Quality Oncology Practice Initiative (QOPI) measures at a single center showed oncology attending physicians, fellows, and APPs performed similarly across the quality measures tracked.10 Although this was a small single institution study, it suggests consistency of performance across oncology providers with opportunity to both reinforce positive practice behaviors and improve as a team together. Additionally, APPs have been shown to increase oncologists’ productivity.5,7,11 Interestingly, 73.1% of American Society of Clinical Oncology census practices reported employing APPs in 2015, a marked increase from 52% reported by the 2014 census practices.12 Although an increase in the utilization of APPs in oncology practices to help expand access to care is encouraging, the total oncology APP workforce capacity is unknown. The American Academy of Physician Assistants indicated that as of December 2016, there are more than 115,000 certified PAs in clinical practice, and the American Association of Nurse Practitioners indicated that as of October 2016, at least 220,000 NPs are licensed in the United States.13,14 Although an exact head count of APPs practicing in oncology is unclear, previous analysis suggests it is less than 5% of the total number of APPs in clinical practice.15 Research is currently underway to mitigate the gap in knowledge about the oncology APP workforce capacity and to better understand this workforce.

TEAM-BASED CARE IN ONCOLOGY

Health care delivery has historically occurred in more of a siloed fashion whereby each person does his or her part for the care of the patient but does not work interdependently with others in the process of providing that care. What distinguishes a team of providers from a group of providers caring for a patient is that within a true team construct,

KEY POINTS • APPs are highly trained and skilled health care providers who are integral to providing team-based care in oncology. • Cancer care is complex, which underscores the need for highly effective teams to deliver this care. • Role optimization for all members of the team is important to ensure each team member is performing at the functional level intended. • APP contributions to practices and patient care extend well beyond direct patient care activity. • APPs are well positioned to enhance value for patients in the oncology setting by positively impacting outcomes that matter to patients. e2 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

all providers work interdependently to achieve a common goal.16 In a group of providers caring for a patient, each person contributes independently to reach a common goal or product.16 When groups of providers care for a patient, the risk of that care being fragmented is high, which can lead to suboptimal outcomes for patients. Cancer care is becoming increasingly complex for a multitude of reasons, underscoring the need for highly effective teams to deliver this care. Pillars of team-based health care, as described in the Institute of Medicine’s discussion paper Core Principles & Values of Effective Team-Based Health Care, include17: • Shared and valued goals, which are clearly understood by all team members, • Clear roles, • Mutual trust, • Effective communication, and • Measurable processes and outcomes. Team-based care must be patient-centric and coordinated, with all the tasks needing team completion to be well defined. Roles and responsibilities of each team member, including patients and their caregivers, should be clear. Cultivating support among all members of the team is a key part of team formation and ongoing development. The patient’s needs should drive the composition his or her care team, and defining the tasks that must be completed for the patient ensures completeness of care. Once these tasks are defined, identifying the member or members of the team who have the appropriate training and experience to carry out each function is key. As part of the groundwork for constructing the team, developing a responsibility assignment matrix may be helpful in delineating roles and responsibilities of team members, and appropriate workflows can then be established accordingly. For optimal team functioning, each member of the team performs those duties consistent with the fullest extent of his or her license (as applicable), education, training, experience, and competency. This leads to the formation of teams that are cost-effective, provides assurance that the patient’s and caregiver’s needs are being met by the most appropriate members of the team, establishes accountability, eliminates duplicative work effort, and ensures each member of the team is performing at the functional level intended. A thoughtfully planned introduction of the team to the patient and caregiver should occur at the start of engagement to appropriately frame expectations. This allows the patient and caregiver, at the initiation of care, to develop a clear understanding of each team member’s role and to recognize how their needs will be met. Furthermore, this step helps foster the development of trust between the patient, caregiver, and care team, which enhances the effectiveness of the therapeutic relationship. When determining roles and responsibilities for APPs, it is important to think of this in the context of whatever duties the APP is taking care of or services the APP is providing would otherwise be done or provided by the physician in the absence of the APP. To better understand APP time and effort allocation, Moote and colleagues performed a self-reported time study of 2 weeks’ duration in an academic

COLLABORATING WITH ADVANCED PRACTICE PROVIDERS

medical center. This showed APPs were spending only approximately 36% of their time in direct patient care, defined as billable and bundled services.18 Care facilitation services, defined as services that would otherwise require a physician to perform, made up about 49% of APP time and effort and included such activities as assisting with rounds, providing patient education, writing progress notes and other patient documentation, and preparing discharge summaries.18 Although the amount of APP time and effort devoted to providing direct billable services was limited in this group, this study illuminates the extended amount of time spent by APPs in activities that do not necessarily generate work relative value units (wRVUs) but that are needed as part of patient care and which would otherwise be done by the physician if the APP were not carrying out these functions. As such, the authors felt there was room to further optimize APP practice in this setting through closer alignment of the roles and responsibilities of the APPs with that of physicians.18 Optimizing roles may not necessarily require adding additional resources. Elnahal and colleagues recently developed a strategy to improve workflows in a multidisciplinary clinic without increasing their human resources.19 This was accomplished by incorporating the military acuity model, ultimately realigning workload to be commensurate with the skill set and competency of each team member.19 This strategy enabled them to increase their clinic volume by more than 30%, decrease the number of postclinic emergency department visits from 9.9% to 7.9%, and decrease the number of postclinic patient phone calls with unresolved issues from 34% to 22%.19 Although some teams are relatively high functioning almost organically, investing time and effort in team training has important implications as it has been demonstrated to improve patient safety.20 In an outpatient oncology setting, Bunnell and colleagues instituted team training with their breast cancer staff including physicians, APPs, nurses, pharmacists, and support staff.21 The team training intervention was notable for improving communication, perceptions of improved efficiency, quality, and patient care safety; relationships/interactions among team members; and patients’ perception of how well their care was coordinated.21 Deliberately working as a team takes time, effort, and patience. The team must go through natural developmental stages that include forming, “storming,” “norming,” and performing as psychologist Bruce Tuckman has described.22 Establishing and maintaining a high level of performance as a team requires ongoing development and nurturing of the team to sustain both the team itself and the individuals that it comprises.

CARE DELIVERY MODELS

APPs practice in two categories of clinical practice models: comanagement and autonomous. In the comanagement model, physicians and APPs are jointly involved in each patient encounter, providing direct patient care. In the autonomous model, APPs provide medical services to

patients without the physical presence of the physician (in accordance with state laws and regulations and facility or practice policy). In a single institutional analysis, Buswell and colleagues evaluated collaborative practice models deployed in their center.9 Three general models were identified and described as the independent visit model (IVM), the shared visit model (SVM), and the mixed visit model (MVM).9 In the IVM, as defined by Buswell et al, APPs and physicians each saw their patients as independent visits for at least two-thirds of the time.9 In the SVM, as defined by Buswell et al, at least two-thirds of patient visits were seen as shared visits between the physicians and APPs.9 The MVM thus represented a blend of visits seen as either independent visits or shared visits, with neither type of visit predominant (Fig. 1).9 In this analysis, physicians were very satisfied with both the IVM and SVM; APPs were very satisfied with the IVM and moderately satisfied with the SVM. Patient satisfaction scores, although variable across models, were generally high for both the IVM and SVM.9 Productivity across all three models was similar. Many factors can influence the care delivery model(s) a practice utilizes, including the patient population itself, facility or practice policy, overarching goals of integrating an APP into the practice, and the nature of the physician–APP partnership. Patients’ access to oncology services can be problematic as the demand for oncology services increases. By jointly identifying appropriate cohorts of patients for the APP to see as independent visits and creating same-day access for patients who must be seen urgently but who do not need to be seen in an emergent care setting, physicians and APPs can together expand access to oncology services. Symptom management is often a cornerstone of APP practice in oncology, and having APPs focus on this aspect of patient care can help make patients more comfortable and improve quality of life for patients and their caregivers.23 Within the respective patient population, it is important to balance the patient panels for both the physician and the APP as a means of preserving the workforce. Specifically, although shifting a significant proportion of patient follow-up visits to the APP may improve access to the oncologist for new patient visits or more complicated patient follow-up visits, it is easy to imagine a diminished level of professional satisfaction for an oncologist whose panel is exclusively the highest acuity patients. Anecdotally, oncologists have described the importance of a balanced template and appreciate the opportunity to see patients who are doing well or who have less complicated courses in addition to seeing more complex patient visits. As a context for this concern, it is essential to note Shanafelt and colleagues’ observation that although oncologists largely indicated satisfaction with their choice of both career and specialty, those who spent the greatest amount of time in direct patient care were identified as being at the highest risk for professional burnout.24 Professional burnout is an entity that bears close surveillance and, even more importantly, prevention, in the face of impending workforce shortages. Recent studies have shown the incidence of burnout in oncologists and oncology PAs to asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK e3

HYLTON AND SMITH

FIGURE 1. Models of Collaborative Practice Between Physicians and NPs and PAs

be 45% and 35%, respectively, and prevention strategies will be vital for ensuring patients’ access to oncology care.24,25 It is also crucial that care teams and care delivery models nimbly adjust to the shifting nature of how and where oncology care is provided. With more care that was traditionally delivered in the inpatient setting now being conducted in the outpatient setting and more oral therapies now available for patients, novel means of delivering care and monitoring patients is key and can include such entities as telemedicine and home visits, as appropriate. APPs, as collaborating partners with oncologists, can leverage their knowledge, skills, and expertise to lead innovative care delivery efforts to provide high-quality, patient-centric care.

VALUE AS THE PLATFORM FOR CARE DELIVERY

Although operational effectiveness remains paramount to optimizing the delivery of health care services, quality—not quantity—of the services provided and the value of those services to patients are rightly the focus. Any strategic plan deployed in the current era must, at its very foundation, be centered on value. Value in health care is defined as the outcomes achieved as proportionate to the dollars spent to achieve those outcomes.26,27 Care for a medical condition such as cancer typically involves a number of multidisciplinary providers rendering a variety of services. Value, in this context, is derived from the combined efforts of the providers across the full cycle of patient care, and all providers are accountable for that value.27 With this in mind, never has team-based care in oncology been as important as it is now. As we shift from a fragmented system focused on all we can deliver to patients to a patient-centric system with care teams constructed based on patient and caregiver needs, we have the opportunity to optimize the value of the care we provide to our patients.28 APPs are well positioned to enhance value for patients by improving outcomes while keeping costs relatively flat, decreasing costs without diminishing outcomes, or both. In a e4 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

retrospective study looking at patients with acute myelogenous leukemia admitted for reinduction chemotherapy, outcomes of a PA/attending physician team were compared with a house staff/attending physician service. Although intensive care unit transfers and in-hospital mortality were similar across both groups, length of stay, readmission rates, and consulting service utilization were decreased in the PA/ attending physician team compared with the house staff/ attending physician team.29 This study suggests the PA/attending service model could deliver care with increased operational efficiency while decreasing health service use but without negatively impacting outcomes.29 In another retrospective study, outcomes for patients with locally advanced oropharyngeal cancer receiving 7 weeks of concurrent chemotherapy and radiation therapy were evaluated prior to and following the initiation of an NP-led clinic. With the implementation of the NP-led clinic, patients were seen weekly to evaluate the profound toxicities they experienced in association with combination therapy. Previously, patients were seen at the beginning of treatment, in the middle of therapy, and at the end of treatment. The data demonstrated that with the implementation of the NP-led clinic, rates of hospitalization for toxicities lowered (12% vs. 28%), chemotherapy dose reductions decreased (6% vs. 48%), and 90% of patients seen in the weekly clinic went on to complete all seven cycles of chemoradiation, compared with 46% prior to the implementation of this intervention.30 Although longterm outcomes could not be determined by a retrospective review, one could hypothesize that maintaining dose density and intensity would translate into improved long-term outcomes. A key component of improving value is measuring outcomes and, more specifically, outcomes that matter to patients. Porter describes an outcomes measure hierarchy composed of three tiers.27 The first tier refers to the patient’s health status attained or retained; the dimensions encompassed in this tier include survival and the degree of recovery or health attained or retained by the patient.27

COLLABORATING WITH ADVANCED PRACTICE PROVIDERS

The second tier, which focuses on the recovery process, encompasses the dimensions of the amount of time it takes to recover and for the patient to return to his or her normal level of functioning or the best level of functioning that can be achieved.27 Tier two also includes a focus on mitigation of issues the patient may experience from treatment or care, ranging from adverse effects and discomfort from treatment to errors in diagnosis and other complications.27 In the third tier, the sustainability of health is the focal point with the dimensions of this including recurrence of disease and/or long-term complications of the original disease and/ or treatment of the disease.27 Although APPs may have a limited ability to influence survival, they can have a significant impact on other outcomes important to patients, including functional status, return to usual activities, symptom control and reduction of suffering, management of late effects, and minimizing wait times.

UNDERSTANDING APP IMPACT

Understanding APP impact in its entirety remains challenging as APP clinical activity can be hidden in shared visits, and APPs are often partaking in activities that do not generate billing activity or wRVUs but that would otherwise be done by the physician in the absence of the APP, thus reducing the physician’s availability to see and evaluate patients. When evaluating the impact of APPs, it is helpful to think beyond the productivity aspect to fully understand the value proposition of APPs. The concept of productivity is relatively concrete and relates to the volume and work intensity of services provided.31 It can be assessed using such standardized measurements such as wRVUs, volume of patients seen, and billing and collections data. Although this can be helpful in understanding productivity, it has its limitations as severity of illness and patient acuity are not factored into the measurements.31 Establishing and tracking direct billable services provided by an APP is one important metric to consider but can be challenging given the variety of collaborative practice models used and varying APP roles. In particular, within the SVM, services are billed and revenue is generated under the physician’s name and national provider index (NPI) number, and thus the services provided by the APP may be less transparent and not easily captured or tracked. This likewise makes looking exclusively at wRVUs of APPs a problematic approach for understanding the scope of their contributions. This can, however, be a very useful metric if an APP’s practice is primarily composed of independent visits and/or procedures. Additionally, services rendered during the global period, such as postoperative visits, do not generate incremental revenue or wRVUs. These visits, however, would need to be conducted by the physician if the APP was not seeing the visits, thus preventing the physician from partaking in other revenue/wRVU-generating clinical activity such as surgery. Tracking productivity in an inpatient setting is equally complex as care and costs may be entangled among many different health system providers and departments. Although national productivity benchmarking metrics from

organizations such as the American Medical Group Association and the Medical Group Management Association provide robust productivity metrics associated with APPs working in the primary care setting, there is a relative lack of national benchmarking productivity metrics available for specialty care such as oncology. Capturing the number of visits or encounters by the APP may be a way of increasing understanding of the APP’s workload, providing there is attribution for shared visits. Although billing and collections data may have some relevance in understanding productivity, it is essential to recognize that practices set charges and payers set fee schedules. This can also become an issue in the SVM with the relative invisibility of the APP in this construct. As another challenge, not all payers enroll APPs. This does not necessarily mean, however, that the payer will not reimburse for services provided by APPs. The payer contracts may stipulate the services provided by the APP be billed under the physician’s name, even when provided as independent visits. It is imperative to review payer contracts to ensure compliance and optimization of reimbursem*nt. An understanding of the APP’s distribution of time and effort, and hence the “big picture,” is essential to comprehending the spectrum of ways in which the APP is contributing to the practice. Care facilitation, teaching and training students and new employees, administrative responsibilities, and research are all activities that bring value to the practice but may not result in billing activity or wRVUs. In response to this issue, Gilbert and Sherry convened a group of APPs within a single comprehensive cancer center to create and pilot metrics related to APP practice and performance.32 The overarching goal of this initiative was to demonstrate APP contributions through a standardized framework that addressed the salient areas of quality of care and productivity. One of the challenges in creating this metrics card was to ensure the metrics appropriately encompassed the many facets of the APPs’ role across multiple oncology subspecialties.32 Likewise, data pertaining to the specific metrics identified needed to be easily trackable, which, in any setting, is often a difficult endeavor. The metrics were organized into four performance categories: financial impact, professional development, patient satisfaction, and quality indicators (Fig. 2).32 Future work in this area beyond the pilot entails establishing a benchmark for each metric with an expectation that all APPs achieve 80% or higher on each metric.32 Establishing effective tracking of all APP clinical activity and sharing this data with the APP is critical to ensure accuracy, bearing in mind the data are only as reliable as the systems set up to collect and analyze the data. It is also important to consider that the APP will likely need some time to get up and running in the clinical practice once integrated, and productivity measurement should be interpreted accordingly. As part of the onboarding process, APPs would benefit from training on billing practices and requisite documentation to ensure both compliance with payer policies and the appropriate level of billing. In some cases, it may be asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK e5

HYLTON AND SMITH

FIGURE 2. Metrics Categories, Definitions, and Measurement Devices

more helpful to evaluate the productivity of the APP–physician unit or the practice as a whole and compare this to the corresponding productivity data prior to the APP starting in the practice, looking for an incremental increase. If team productivity falls short of established goals, it is necessary to engage in thoughtful exploration of this to establish the root cause(s) so that fitting interventions can be designed accordingly.

CONCLUSION

With the growing complexity of cancer care and impending workforce shortages, team-based care in oncology provides the opportunity to deliver coordinated, high-quality, high-value, patient-centered oncology care. Thoughtfully constructed teams of individuals who work interdependently to accomplish shared and valued goals can positively impact outcomes that matter to patients. As an integral part of the patient’s care team, APPs contribute to practices and patient care in many ways. Although their contributions may not always be easily measured, they are central to the care of the patient. Studies have demonstrated significant patient, physician, and APP satisfaction with collaborative care models, and collaboration with APPs has been shown to increase oncologists’ productivity. As rapid progress in cancer care continues, innovative approaches to care delivery will be necessary to ensure patients’ access. Effective oncologist–APP partnerships will be key to providing optimal, high-value care to patients.

Reprinted with permission from the Journal of the Advanced Practitioner in Oncology (JADPRO).

References 1. American Cancer Society. Cancer Facts & Figures 2017. https://www. cancer.org/content/dam/cancer-org/research/cancer-facts-andstatistics/annual-cancer-facts-and-figures/2017/cancer-facts-andfigures-2017.pdf. Accessed February 22, 2017. 2. American Cancer Society. Cancer Treatment & Survivorship Facts & Figures 2016-2017. https://www.cancer.org/content/dam/cancerorg/research/cancer-facts-and-statistics/cancer-treatment-andsurvivorship-facts-and-figures/cancer-treatment-and-survivorshipfacts-and-figures-2016-2017.pdf. Accessed February 22, 2017.

5. Erikson C, Salsberg E, Forte G, et al. Future supply and demand for oncologists: challenges to assuring access to oncology services. J Oncol Pract. 2007;3:79-86. 6. Yang W, Williams JH, Hogan PF, et al. Projected supply of and demand for oncologists and radiation oncologists through 2025: an aging, betterinsured population will result in shortage. J Oncol Pract. 2014;10:39-45. 7. Towle EL, Barr TR, Hanley A, et al. Results of the ASCO study of collaborative practice arrangements. J Oncol Pract. 2011;7:278-282.

3. Institute of Medicine. Delivering High-Quality Cancer Care: Charting a New Course for a System in Crisis. Washington, DC: National Academies Press; 2013.

8. Hinkel JM, Vandergrift JL, Perkel SJ, et al. Practice and productivity of physician assistants and nurse practitioners in outpatient oncology clinics at national comprehensive cancer network institutions. J Oncol Pract. 2010;6:182-187.

4. IHS, Inc. The Complexities of Physician Supply and Demand 2016 Update: Projections From 2014 to 2025. Washington, DC: Association of American Medical Colleges; 2016.

9. Buswell LA, Ponte PR, Shulman LN. Provider practice models in ambulatory oncology practice: analysis of productivity, revenue, and provider and patient satisfaction. J Oncol Pract. 2009;5:188-192.

e6 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

COLLABORATING WITH ADVANCED PRACTICE PROVIDERS

10. Zhu J, Zhang T, Shah R, et al. Comparison of quality oncology practice initiative (QOPI) measure adherence between oncology fellows, advanced practice providers, and attending physicians. J Cancer Educ. 2015;30:774-778. 11. Akscin J, Barr TB, Towle EL. Benchmarking practice operations: results from a survey of office-based oncology practices. J Oncol Pract. 2007;3:9-12. 12. American Society of Clinical Oncology. The state of cancer care in America, 2016: a report by the American Society of Clinical Oncology. J Oncol Pract. 2016;12:339-383. 13. American Academy of PAs. What Is a PA? https://www.aapa.org/ what-is-a-pa/. Accessed March 21, 2017. 14. American Association of Nurse Practitioners. NP fact sheet. https:// www.aanp.org/all-about-nps/np-fact-sheet. Accessed March 21, 2017. 15. Polansky M, Ross AC, Coniglio D. Physician assistant perspective on the ASCO workforce study regarding the use of physician assistants and nurse practitioners. J Oncol Pract. 2010;6:31-33. 16. Taplin SH, Weaver S, Chollette V, et al. Teams and teamwork during a cancer diagnosis: interdependency within and between teams. J Oncol Pract. 2015;11:231-238. 17. Mitchell P, Wynia M, Golden R, et al. Core principles & values of effective team-based health care. Institute of Medicine: Washington, DC, 2012. 18. Moote M, Nelson R, Veltkamp R, et al. Productivity assessment of physician assistants and nurse practitioners in oncology in an academic medical center. J Oncol Pract. 2012;8:167-172.

21. Bunnell CA, Gross AH, Weingart SN, et al. High performance teamwork training and systems redesign in outpatient oncology. BMJ Qual Saf. 2013;22:405-413. 22. Tuckman BW. Developmental sequence in small groups. Psychol Bull. 1965;63:384-399. 23. Shulman LN. Efficient and effective models for integrating advanced practice professionals into oncology practice. Am Soc Clin Oncol Educ Book. 2013;33:e377-e379. 24. Shanafelt TD, Gradishar WJ, Kosty M, et al. Burnout and career satisfaction among US oncologists. J Clin Oncol. 2014;32:678-686. 25. Tetzlaff ED, Hylton HM, Ruth K, et al Provider characteristics and their association with burnout and career satisfaction among physician assistants in oncology. J Clin Oncol. 2016;34 (suppl; abstr 6521). 26. Porter ME, Lee TH. Why strategy matters now. N Engl J Med. 2015;372:1681-1684. 27. Porter ME. What is value in health care? N Engl J Med. 2010;363:24772481. 28. Porter ME, Lee TH. The Strategy That Will Fix Health Care. Harvard Business Review. https://hbr.org/2013/10/the-strategy-that-will-fixhealth-care. Accessed March 30, 2017. 29. Glotzbecker BE, Yolin-Raley DS, DeAngelo DJ, et al. Impact of physician assistants on the outcomes of patients with acute myelogenous leukemia receiving chemotherapy in an academic medical center. J Oncol Pract. 2013;9:e228-e233. 30. Mason H, DeRubeis MB, Foster JC, et al. Outcomes evaluation of a weekly nurse practitioner-managed symptom management clinic for patients with head and neck cancer treated with chemoradiotherapy. Oncol Nurs Forum. 2013;40:581-586.

19. Elnahal SM, Moningi S, Wild AT, et al. Improving safe patient throughput in a multidisciplinary oncology clinic. Physician Leadersh J. 2015;2:56-60, 62, 64-65.

31. Pickard T. Calculating your worth: understanding productivity and value. J Adv Pract Oncol. 2014;5:128-133.

20. Salas E, Rosen MA. Building high reliability teams: progress and some reflections on teamwork training. BMJ Qual Saf. 2013;22:369-373.

32. Gilbert E, Sherry V. Applying metrics to outpatient oncology advanced practice providers. J Adv Pract Oncol. 2016;7:192-202.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK e7

THE VALUE OF PERSONAL REFLECTION IN ONCOLOGY

For Our Patients, for Ourselves: The Value of Personal Reflection in Oncology Lidia Schapira, MD, Jane Lowe Meisel, MD, and Ranjana Srivastava, FRACP, OAM OVERVIEW Caring for patients with cancer is a great privilege as well as an emotionally and intellectually challenging task. Stress and burnout are prevalent among oncology clinicians, with serious repercussions for the care of patients. Professional societies must provide guidance for trainees and practicing physicians to mitigate the negative consequences of stress on their personal lives and medical practice. Reflection, reading, and writing about personal experiences provide outlets for fortifying personal reserves and promoting resilience to allow us to recognize the joy and meaning of our work and to forge connections with our peers. Herein, we present some of our own reflections on how and why one might take time to write, and about the power of the written word in oncology and medicine.

n the field of oncology, we are all connected by our desire to improve the lives of patients afflicted with cancer. However, whether one is primarily clinical or spends most of his or her time conducting research, the work can be inspiring and humbling, while proving tremendously difficult. In a professional landscape that values clinical and research productivity by numbers of patients seen and manuscripts published, reflection can serve to provide perspective and maintain some balance in our personal and professional lives. I attended a progressive junior high school, governed by the philosophy that 12- and 13-year-olds are often so focused on their emotional lives that this reality needed to be incorporated into the curriculum rather than ignored. Therefore, in class, we were assigned to write about our personal experiences. Delving into our emotions in our writing and then sharing our pieces with the class allowed us not only to understand the power of the written word and to aspire to harness it but also to connect with one another at a time that can be inherently insecure and lonely. Oncologists are very different from seventh graders, but the process of putting our experiences into words and sharing them with others is, in many ways, even more important. This process can help us make sense of difficult conversations with patients, the impact of our work on our home lives, or challenges encountered in the laboratory. In a recent survey conducted by ASCO evaluating burnout and career satisfaction, almost 45% of the nearly 1,500 oncologists surveyed were burned out on the emotional

exhaustion and/or depersonalization domain of the Maslach Burnout Inventory.1 More hours spent on patient care were positively correlated with the risk for burnout, a concerning finding given the projected shortage of oncologists over the coming years and the need for many of us to start seeing higher volumes of patients. Burnout happens not just because we are too busy but also because many of us do not have a way to process, either by ourselves or with our colleagues, the gravity of our experiences and what they mean to us. Consider this: many of us, as our stacks of medical journals arrive in the mail, turn first to the New England Journal of Medicine’s "Perspective" section, JAMA’s "A Piece of My Mind," or the Journal of Clinical Oncology’s "Art of Oncology" (AOO). We devour these pieces, hungry for stories and looking for connections. It was through this lens that our session on the use of narrative in oncology was conceptualized. We believe that reading and writing about our experiences may allow us to achieve greater self-awareness and more of a sense of community among colleagues and, through this, allow us to be better at what we do and to derive greater enjoyment from it. What follows here is a personal account of the journey of one inspiring and prolific oncologist-writer, Dr. Ranjana Srivastava (Part I), and a piece on the power of stories from Dr. Lidia Schapira, the current editor of AOO (Part II). Our hope is that these reflections will inspire our readers to think more deeply about the power of narrative in our field and the different ways they might use it to further their own personal goals.

From the Stanford Cancer Institute, Stanford University School of Medicine, Stanford, CA; Winship Cancer Institute, Emory University School of Medicine, Atlanta, GA; Department of Oncology, Monash Health, Victoria, Australia. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Ranjana Srivastava, FRACP, OAM, Chemotherapy Day Unit, Dandenong Hospital, David St., Victoria 3175, Australia; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 765

SCHAPIRA, MEISEL, AND SRIVASTAVA

PART I: THE WRITER’S JOURNEY It was nearing the 10th anniversary of the loss of my twin pregnancy when I felt an urge to write about it.2 I can’t say that I had been dwelling substantially on the loss or that the 10th year felt any more significant than, say, the 1-year or the 5-year mark. If truth be told, life had been very good to me after that tragedy, with the healthy arrival of three children, a fulfilling profession, and much more. The valuable perspective gained from a career as an oncologist meant that my grief wasn’t as paralyzing as I had feared. But clearly, as the anniversary approached, the event must have been somewhere in my subconscious because I felt the need to expunge it. That column ended up becoming one of the world’s most widely read and shared columns in the Guardian that year. What touched me most was the tenderness and humanity of exchanges the column elicited in what truly felt like a global village. Complete strangers sent me their wishes and forwarded the essay to others going through the same experience. Voltaire was right: writing is the painting of the voice. I am a medical oncologist and writer. I have written books and essays, and for the past few years, I have been a regular columnist on medicine and society for the Guardian, which was founded in 1821 as the Manchester Guardian and now has a global reach. I am also an essayist for the New England Journal of Medicine. In this personal reflection, I will track my own journey while answering some of the commonly asked questions of why, what, and when to write.

Why Write?

This is the easiest one. As oncologists, we are witness to life’s deepest and most intimate moments. These moments move, inspire, frighten, teach, and challenge us. Who do we tell about the pregnant mother with advanced breast cancer or the successful businessman with metastatic melanoma who goes from diagnosis to death in 4 weeks? Who will share our heartache at looking after a grandfather whose greatest lament is not that he is dying but that his children can’t find the time to visit? Who will admire with the same intensity the patient whose face glows with dignity and courage even as cancer invades her skin? Our patients stir

KEY POINTS • Writing about our experiences as oncologists can help us understand them with greater clarity. • Learning to set aside small blocks of time for writing, reading, and reflection is a useful strategy. • Reading about the experiences of others in our field can help us feel more connected to colleagues, providing a form of social support. • Reflecting upon one’s reactions to the written word facilitates self-awareness. • Reading and writing narratives in oncology may strengthen us emotionally, allowing us to be more fully present for our patients and our loved ones. 766 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

a range of emotions with us, not all of which we necessarily feel like speaking aloud. We fear that our family and friends may not understand us or that they may find our stories gloomy or upsetting. But we know that acknowledging our raw emotions, our learnings and feelings, is critical if we are to be better doctors. Human beings find meaning through stories, we connect through stories—and our stories demand to be written, though not everything we write needs to be published. I started writing when I was 11, but it is only in the past decade that I have started publishing widely.3 Most of what I write is for private consumption, catharsis, and making sense of the world. My writing centers me; knowing this means that if the market for my writing were to fall away, I’d still gain personal satisfaction from the habit.

What to Write

The history of medicine is replete with fine writers, and it’s really no wonder when you think of the fertile grounds for writing that being a doctor provides. We just have to turn up to work to stumble upon stories. The lives of our patients and our own lives intertwine to provide us with rich experiences and powerful learnings, and as long as we tune into human stories, there will never be a shortage of ideas. However, one thing that concerns doctor-writers is the matter of consent. Is it ethical to write about our patients, who trust us with their secrets? Should one always seek consent when writing? What happens if we unintentionally end up offending a patient, or for that matter a colleague, through our writing? The impetus, and the temptation, to be published can exert such a pull that it’s easy to cross the line between telling a story and breaching patient confidentiality. Something every modern writer must be aware of nowadays is that writing has an unprecedented digital footprint. Once you hit send, you can’t control the ways your work is read, interpreted, and used. It is also always and readily available, even if you’d like it to go away. This is something I have become increasingly aware of in writing for high-profile platforms such as the Guardian and the New England Journal of Medicine. Editorial assistance is important, but it’s just that, assistance; as the author, one must own and defend one’s writing. It is impractical and unnecessary to always get consent to write. Furthermore, I think that the very act of seeking consent changes the nature of writing—it’s difficult to render a totally honest interpretation of an event and write without fear or favor. At the same time, no doctor wants to hurt a patient or jeopardize a valuable and therapeutic relationship. Because I write almost exclusively about patients, here are some rules of thumb I follow. I ask myself why it’s important to write about what happened. Is there a meaningful and universal message to share? Could what I write inform, educate, or empower someone? Or is it because I am annoyed and need to vent? I work mostly in a highly socioeconomically disadvantaged community with a high proportion of non-English-speaking

THE VALUE OF PERSONAL REFLECTION IN ONCOLOGY

refugees and asylum seekers. It’s safe to say that the vast majority of my patients would never come across my writing, but I always ask myself how they would feel if they were to stumble upon it. Would they be hurt, or would they feel heard? Would they feel exploited or understood? Would they say I had misrepresented them, or would they consider me their advocate? There are things I never write about without prior consent. These have included attending the funeral of a patient I was fond of, acknowledging a gift from a dying patient, reporting an intimate but unique consultation, and encounters for which it would be immediately obvious to a reader that the story was about him or her or a loved one. No one has ever withheld consent when I have explained the reason for my writing; patients and their relatives are very generous and thoughtful in offering their experiences as teaching moments. Across many years of writing, I have attracted the ire of only one patient, who believed that I had been loose with the facts of her case. She chastised me for abusing my position and refused to accept my apology. In fact, her story was an all too common one, but in telling it, I had obviously skirted unacceptably close to her personal experience. This was one of the lowest points of my writing career, as I felt guilty about causing a dying patient distress and sad that I had not had an opportunity to make amends. But her rebuke has stayed with me and made me more cautious and more considerate. Ultimately, writing about medicine relies on personal integrity and having a moral compass that detects right from wrong before an editor or one’s audience has the opportunity to do it. It means thinking deeply about one’s intent, endeavoring to set aside personal bias, and then having the courage of one’s conviction. Finally, this is a one-line mental checklist I tick each time I write: “Will I be able to hold my head high in clinic tomorrow if I publish this?”

When to Write

“How do you find the time?” I once asked a famous writer. “And what do you do about writer’s block?” “Nonsense,” she said briskly. “When you show up to work, do you suffer from oncologist’s block? Writing is a job. It takes commitment.” Several of my colleagues lament that they used to write well until careers in medicine put waste to their dreams of becoming authors. Now, between juggling patients, configuring career progression, and raising their families, they just don’t have the time to write. A barrier I identified early on in my writing career is that the idea of having unlimited time, no distractions, a spotless desk, a cabin in the woods, or a house overlooking the ocean was never going to be my reality! With a busy clinical load and young children, there was never a good time to write. I spent the day doing my regular job, and by nighttime, I was too exhausted to write. But I never gave up writing a journal, filling it with mostly mundane observations and reflections, not realizing that

the mere habit of writing was important. I stuck to nice pens and sought out beautiful leather-bound journals to enhance the meditative quality of longhand writing. But it didn’t feel like enough. Finally, the urge to write more and communicate with an audience became so great that I had to confront the reality: I could either write amid the chaos of work and home or not write at all. So, slowly, I trained myself to write among the chatter of children, keeping an eye on the trampoline and another on the screen. I became adept at stealing moments to write: between school pickups and sports drop-offs, while waiting in the car for swim school to finish, or perched on the edge of a bath. I learned to write a few lines if a patient unexpectedly canceled or if a meeting was delayed. I also learned to write in my head when I went jogging. Going for a run in the early morning before the hustle and bustle of the day begins is a fine way of sifting through my thoughts. Now, with a deadline every fortnight, I must and can write almost anywhere. I have no set time to write, but I do know that when an encounter lingers in my mind, it’s a signal to write. I turn the encounter in my mind, let myself feel uncomfortable or challenged or gratified, until gradually the essence of the experience becomes clearer and I am ready to write. Then, the words seem to tumble out. The hardest part of writing is getting started. Now, I worry less about perfection and more about getting the words down on paper. It’s much easier to edit than get started. I have had to make some compromises. I love the slowness of writing by hand, which allows you to turn your thoughts in your mind, but I can write like this only in my journal now. The rest of my work is done on a laptop, but because I don’t like carrying it everywhere, I save my work in the cloud so I can access it from anywhere in the world. In the same way as many people work on talks and presentations in the airport lounge or on a flight, I write wherever I can. But perhaps the most deliberate, and the hardest, decision I have had to make is to not undertake full-time clinical work to make some room for writing and its necessary companion, reflection. This has inevitably meant somewhat restricted career opportunities, with academic and financial ramifications, but for me it seems like a fair price to pay for the tremendous job satisfaction of being a doctor and a writer, able to serve not only my patients but a world of people. To have a few hours in the week to read widely, experiment with different forms of writing, and reflect upon the meaning of being a doctor seems like a luxury that many of our time-starved, emotionally fatigued colleagues are eager to embrace. They need to know that if good medicine is about advocacy, we can serve society through various means. Research and clinical work are two time-honored ways, but writing and public speaking are legitimate means of democratizing medicine. The celebrated physician and writer Anton Chekov observed, “Medicine is my lawfully wedded wife and writing my mistress. When I tire of one, I spend the night with the other. Though it’s disorderly, it’s not so dull, and besides neither of them loses anything from my infidelity.”4 asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 767

SCHAPIRA, MEISEL, AND SRIVASTAVA

Nurturing the art of medicine through reflection and writing is important. It allows the development of a therapeutic, creative, and educational outlet. We must not consider it an unaffordable luxury but an essential tool for improving our own lives and those of our patients.

PART II: THE POWER OF STORIES

A Case for Reading

Reading essays published in medical journals gives us the opportunity to reflect, alone and collectively, on important aspects of practice that are essential but often overlooked. Ethical dilemmas come into sharper focus, and the emotional toll of practice is assuaged by feeling connected to peers or to the writer. Even if we read when we are alone, the act of reading establishes a virtual connection to the writer, editor, and fellow readers. It provides a form of social support that is so often lacking in the workplace and offers validation of the experiential and intellectual aspects of our complex professional lives. Doctors have traditionally kept their worries to themselves and paid a price for their stoicism and emotional isolation. By stimulating reflection and conversation, reading can foster self-awareness and self-expression. Furthermore, reflecting on one’s reaction to text nurtures our sense of purpose and vocation, helping us maintain perspective and balance in our lives. Reading a story or personal reflection forces us to slow down and inspires us to daydream. In those moments when time seems suspended, we allow our minds to roam, occasionally stumbling or straying, always searching for what is normally tucked behind conscious thoughts and hardly ever allowed to surface. We connect with our sense of vocation, with ideas we once cherished and then discarded or forgot, and with desires we may not have known existed. In the act of reading, we are lifted and transported by the creativity of our peers, whose artistry gives us new insights into old problems and language to describe what seemed beyond the reach of words. Stories, poems, photo essays, and commentaries provide a platform for reflection that allow us to explore other perspectives and other ways of being in the world. In other words, reading stimulates our empathic abilities, bringing ideas and dreams into sharper focus. Stories and opinion pieces serve another useful professional function, in that they shape our professional discourse.5 I can easily quote essays that shaped my views on important topics; their messages remain fresh and powerful years after publication. We also learn, from our colleagues’ experience, how to frame and discuss challenging topics so that our communication is clear and supportive. Stories and reflections expand our vocabulary and our mind-set, at times providing guidance and focus that can improve our clinical performance. Reading can, by all of these mechanisms, help reduce perceived levels of work-related stress and even contribute to our well-being. Perhaps reading even helps reduce the risk of professional burnout, although this is impossible to prove. What is clear, however, is that the ability to remain curious and to imagine something that does not yet exist is indispensable for success in research and innovation. 768 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

Composing a persuasive and coherent narrative is also essential for achieving one’s goals on personal and professional fronts and may contribute to professional satisfaction. Successful grant writers persuade readers to invest in their dreams. Trusted mentors help junior colleagues bring their ambitions and projects into focus. Clinicians function as coeditors for their patients’ narratives through attentive listening and deliberate communication. Thus reading serves to prepare us for the work of empathic listening. Reading also brings us into contact with talented storytellers. Listening to stories provides another powerful venue for enjoyment and has become easily accessible in the era of podcasts and audio books. Stories can keep us company on our commute to work or while sitting in the car waiting for a child to finish practice or during workouts. We can read privately, quietly or out loud, and reread at our own pace. Stories are extraordinary tools for teaching ethics and interpersonal and relational skills. They give us access to complex emotions and deepen our appreciation for another’s suffering or heroism. Stories surprise and entertain us, expand our intellectual reach, and challenge our creativity. Persuasive commentaries can influence opinion and have a transformative effect on education and practice. Because we inevitably spend so much of our time reading medical notes, scientific papers, and manuscripts, reading stories provides a welcome escape into a world of colorful characters, poignant storylines, and insightful messages. Essays published in medical journals are often personal narratives written by one individual, although increasingly these narratives have multiple authors, suggesting a collaboration and a team effort. They are selected for publication on the basis of their messages, originality, and artistry, as well as their perceived relevance for the community of readers, and this varies depending on the orientation of the journal. Editors envision that readers turn to text as a springboard for reflection and that they appreciate writes’ willingness to share personal doubts, to expose their vulnerabilities, and to let us peek at their inner landscapes.

"Art of Oncology" in the Journal of Clinical Oncology: The Stories We Tell

The Journal of Clinical Oncology has invited the submission of personal essays, poems, and art forms for publication within the "Art of Oncology" section since 2000. Since its inception, AOO has published essays on a broad range of topics that represent the human side of cancer from the perspectives of patients, advocates, and clinicians. Under the skillful leadership of Charles Loprinzi and then David Steensma, AOO struck a chord with the global readership of the journal. I had the privilege of succeeding Dr. Steensma as consultant editor at the end of 2014, and I enjoy working with a brilliant and wise editorial board to select submissions we feel contribute to shaping our professional culture. We look for essays that have timely and relevant messages, written with humility and candor.

THE VALUE OF PERSONAL REFLECTION IN ONCOLOGY

Great essays capture our attention from the start. Some are funny or whimsical, others sorrowful or nostalgic. Through an assortment of storylines and scenarios, we travel imaginary roads and grapple with common dilemmas. Essays help us witness others’ suffering and celebrate their heroism and provide a safe release for the emotional toll of working in oncology. Despite enormous scientific progress, those of us involved in the care of patients know the grief and sorrow that accompany a career in oncology. Reading can help us get through a tough day. Writers write about what they know. Doctors and nurses spend a lot of time listening to stories and are familiar with plot, protagonist, setting, dialogue, and theme. Oncologists struggle to find meaning in tragedy and humor in daily minutiae and to maintain a healthy balance between their work and home lives. Several essays addressed these issues in the past year. In “What Mommy Does,” Melissa Mark6 describes her struggle to shield her young daughter from learning that her mother’s work involves the care of children who are dying and how this changed after the child overheard a telephone conversation with a hospice nurse while taking her evening bath at home. Megan Caram7 coins a new term, “oncologist’s guilt,” to describe the conflicting emotions she experienced during her 3-month maternity leave. She describes feeling as if she were “abandoning” patients and contrasts the healthy period of attachment between a newborn and his mother with the feelings of dependence that are inherent to close therapeutic relationships. William Meyer8 shares his heartbreak over the death of his own grandchild from cancer. His inner pain is almost palpable as he writes about feeling a sense of “abject failure to help the ones most dear [to you] despite years of training and supposed ‘expertise.’” He concludes the essay on an unsettled note: “these are not easy feelings to come to grips with, and perhaps the sharing of further insight on these experiences will require the passage of time.” Indeed, with time we can find meaning and integrate painful experiences into the larger tapestry of our lives, as told by Jonathan Finlay9 in “A Ruby Anniversary.” On the 40th anniversary of his last “encounter with seminoma,” he embraces his fortune and believes he is a better physician because of his own experience as a patient with cancer. Coming to terms with grief and loss is a recurrent theme for AOO. In her remarkable essay “Pieces of Grief,” Erica Kaye describes the visceral reaction she experienced after the death of a patient in the intensive care unit, a death that forced her to face her emotional exhaustion.10 Katherine Reeder-Hayes11 describes being overcome by emotion and crying, as she is standing alone in her new, empty home at midnight, listening to bluegrass playing on the radio, a paintbrush in her hand. Reeder-Hayes writes about beloved patients, whose passing affects us very deeply. Daniel Rayson12 explores both sides of the clinical relationship in his wonderful essay “White Knuckling.” He delves into the lived experience of a young, dedicated oncology nurse who is experiencing symptoms of burnout and trying her best to encourage and comfort her patient, a tough, retired neo-

natal intensive care unit nurse who voices her ambivalence about continuing to fight her metastatic cancer, fully aware that she will die of this disease. Rayson captures the imaginary dialogue that occurs in the infusion unit, while the patient receives her infusion of bisphosphonate, giving voice to the trauma experienced by oncology nurses who are on the front lines of cancer teams, delivering solace and cheer together with powerful anticancer therapies. Authors write to share their stories and to give advice. Laura Melton13 draws a parallel between a patient who successfully compartmentalized his illness until the very end and clinicians who cope with loss by compartmentalizing their feelings. She acknowledges that this emotional distancing provides a buffer that allows “us to be fully present without feeling overwhelmed” and also warns us that artificial boundaries are porous and may crumble during transitions between work and home life. David Korones14 writes about caring for an adolescent with a pontine glioma who insisted that she did not want to know her prognosis, describing the tension he experienced in trying to reconcile his patient’s request not to know with the evidence supporting full disclosure of prognostic information. Reading through these essays, we find common ground with colleagues we have never met. Essays also serve to express regret and remorse, as in Nikhil Barot’s15 tale of a patient who died of complications of an unwanted diagnostic bronchoscopy. The author wishes he had listened more carefully when she refused the procedure and asked to be allowed to go home to die, ending his story with a very simple and effective “and you sit and think and think.” Reena George16 expresses remorse at having judged and dismissed the unreasonable requests from the daughter of a patient with advanced cancer, until she understood that they stemmed from a desperate and loving desire to help her dying mother. These sincere reflections can be therapeutic for both writer and reader.

CONCLUSION

Cancer clinicians need stories to recalibrate their emotional lives, to make sense of their experiences, and to learn from one another. Writing can serve as an outlet for self-expression or a mechanism for making sense of complex experiences. Writers write for fun, for therapy, to share stories and opinions, to honor a patient or colleague, for atonement, and sometimes for the glory of being published. Essays bring joy and insight to readers, allowing them to slow down and reflect and to refuel their emotional reservoirs. It is our hope that reading also stimulates dialogue and helps foster a culture of collegiality among oncologists. Talking about our reactions to the written word can help us get to know one another and contribute to the professional development of junior colleagues and trainees. Reading, reflecting, and sharing stories serve an important role in the professional development of oncologists. Stories can guide us to find our own sources of inspiration and support and strengthen our therapeutic skills. In turn, this will affect the lives of patients and family caregivers struggling to cope with the unwanted burden of illness. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 769

SCHAPIRA, MEISEL, AND SRIVASTAVA

References 1. Shanafelt TD, Gradishar WJ, Kosty M, et al. Burnout and career satisfaction among US oncologists. J Clin Oncol. 2014;32:678-686. 2. Srivastava R. Losing my twin baby boys for ever changed the way I treat my patients. The Guardian. https://www.theguardian.com/ commentisfree/2015/jun/15/losing-my-twin-baby-boys-foreverchanged-the-way-i-treat-my-patients. Accessed January 30, 2017.

7. Caram ME. Oncologist’s guilt. J Clin Oncol. 2016;34:3221-3222. 8. Meyer WH. Regarding Beau. J Clin Oncol. 2016;34:2669-2670. 9. Finlay JL. A ruby anniversary. J Clin Oncol. 2016;34:2312-2313. 10. Kaye EC. Pieces of grief. J Clin Oncol. 2015;33:2923-2924. 11. Reeder-Hayes K. Haunted. J Clin Oncol. 2017;35:113-114. 12. Rayson D. White knuckling. J Clin Oncol. 2016;34:1419-1420.

3. Rajana Srivastava. The Guardian. https://www.theguardian.com/ profile/ranjana-srivastava. Accessed February 1, 2017.

13. Melton L. Compartmentalizing cancer. J Clin Oncol. 2016;34:15581559.

4. Anton Pavlovich Chekhov (1860-1904). http://www.eldritchpress. org/ac/chekhov.html. Accessed January 30, 2017.

14. Korones DN. Talking to children with cancer: sometimes less is more. J Clin Oncol. 2016;34:3477-3479.

5. Steensma DP. Stories we tell one another: narrative reflection and the art of oncology. Am Soc Clin Oncol Educ Book. 2013;33:e331-e335.

15. Barot N. You have seen her before. J Clin Oncol. 2016;34:1012.

6. Mark M. What mommy does. J Clin Oncol. Epub 2016 Dec 5.

770 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

16. George R, Kandasamy R. A space to heal. J Clin Oncol. 2016;34:33493350.

RESILIENCE: APPROACH TO PREVENT AND COMBAT BURNOUT IN ONCOLOGY

Mastering Resilience in Oncology: Learn to Thrive in the Face of Burnout Fay J. Hlubocky, PhD, MA, Miko Rose, MD, and Ronald M. Epstein, MD OVERVIEW Oncology clinician burnout has become a noteworthy issue in medical oncology directly affecting the quality of patient care, patient satisfaction, and overall organizational success. Due to the increasing demands on clinical time, productivity, and the evolving medical landscape, the oncology clinician is at significant risk for burnout. Long hours in direct care with seriously ill patients/families, limited control over daily responsibilities, and endless electronic documentation, place considerable professional and personal demands on the oncologist. As a result, the oncology clinician's wellness is adversely impacted. Physical/emotional exhaustion, cynicism, and feelings of ineffectiveness evolve as core signs of burnout. Unaddressed burnout may affect cancer clinician relationships with their patients, the quality of care delivered, and the overall physical and emotional health of the clinician. Oncology clinicians should be encouraged to build upon their strengths, thrive in the face of adversity and stress, and learn to positively adapt to the changing cancer care system. Fostering individual resilience is a key protective factor against the development of and managing burnout. Empowering clinicians at both the individual and organizational level with tailored resilience strategies is crucial to ensuring clinician wellness. Resilience interventions may include: burnout education, work-life balance, adjustment of one’s relationship to work, mindful practice, and acceptance of the clinical work environment. Health care organizations must act to provide institutional solutions through the implementation of: team-based oncology care, communication skills training, and effective resiliency training programs in order to mitigate the effects of stress and prevent burnout in oncology.

r. A is 11 years past his medical oncology fellowship training and remains motivated to provide the optimal oncologic care for every patient and family member he sees. He works in a vast urban health care system with a patient panel of 110 to 120 patients per week. Dr. A is affable, has a hardy personality, and is admired by patients, nurses, staff, and his partners. Recently, Dr. A became partner, working long hours to achieve this lifelong dream. However, Dr. A is feeling physically exhausted of late, irritable, sad, and ineffective, as it seems as though his clinical duties never cease. At home, he calls his patients and spends most evenings in front of a computer completing patient notes or orders. Dr. A is unable to sleep most nights and spends little time engaging in leisure activities, such as running or attending his son’s piano recitals. Currently, Dr. A is on in-patient service and gives weekly hour-long lectures to oncology fellow trainees at an affiliated academic hospital. He reports feeling cynical regarding the future to his colleague Dr. Z and questions, “Is any of this worth it?” Although the oncology clinician, like Dr. A, is adequately equipped and expert at providing benevolent care to patients with cancer and their families, sadly, the greater majority of

clinicians like Dr. A fail to provide self-compassion and care when it is most needed as symptoms associated with burnout arise. Dedicated empathic clinicians like Dr. A respond with self-blame when he is unable to perform at optimal levels. Little if any sympathy has been given to the physician especially the oncologist, who, despite best efforts at “toughing it out,” fails to meet all work duties, with his role as physician directly conflicting with his role as parent. As a result, Dr. A feels physically and emotionally depleted, cynical, and ineffective. However, Dr. A may readily face these challenges and address burnout by developing and mastering resilience skills.

A BRIEF OVERVIEW OF BURNOUT IN ONCOLOGY: FOCUS ON RESILIENCE

A comprehensive review and analysis of burnout, including prevalence, symptoms, risk factors, related concepts, as well as individual and organizational interventions for consideration for both the practicing oncology clinician and healthcare institution was presented at the ASCO Annual Meeting in 2016 and documented.1 A brief succinct overview of the seminal concepts and issues associated with burnout will be presented in this review with a focus on resilience.

From the Department of Medicine, Section of Hematology/Oncology, The University of Chicago, Chicago, IL; Department of Adult Psychiatry, Michigan State University, East Lansing, MI; University of Rochester School of Medicine and Dentistry, Rochester, NY. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Fay J. Hlubocky, PhD, MA, Department of Medicine, Section of Hematology/Oncology, The University of Chicago, 5841 S. Maryland Ave., MC2115, Chicago, IL 60637; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 771

HLUBOCKY, ROSE, AND EPSTEIN

Burnout: What It Is and Why It Matters For over a decade now, it is estimated that approximately 45% to 80% of practicing oncologists worldwide in countries such as the United States, Europe, and Australia experience symptoms associated with burnout.2-4 Specifically, burnout has been identified as a work-related syndrome that manifests as a result of the interaction between the oncology clinician and the organization.5-11 Burnout is characterized by three core symptom domains: physical and emotional exhaustion, cynicism and depersonalization (sense of disengagement), and low sense of professional accomplishment (ineffectiveness; Sidebar 1).5-14 These three-dimensional signs of burnout exist along a continuum characterized by distinctly unique symptoms and an overlap of symptoms.5-14 For example, cynicism and depersonalization is traditionally characterized by pessimism or depression (which are also key symptoms of emotional exhaustion), isolation, detachment, and demoralization. Burnout is not a disease. Burnout is a stable, chronic,and insidious process with the initial core physical exhaustion and negative emotional symptoms slowly developing over the course of 1 year as interpersonal and occupational stressors arise and persist.5-14

Risk Factors

Multiple individual and organizational factors have been identified as contributing factors responsible for clinician burnout in health care.6,9-27 Individual contributors are internal dispositional risk factors consisting of sociodemographic (e.g., younger age; female gender presents with emotional exhaustion, whereas male physicians present with cynicism, single/unmarried marital status, and medical trainee status) as well as personality (e.g., extraversion and conscientiousness) characteristics. Recent evidence revealed that physicians who experience unaddressed burnout are less likely to identify with medicine as a calling, a duty to serve the greater good, adversely affecting both the clinician and patient.27 However, given the changing landscape of the present-day health care system, recent research equally centers on specific external, occupational, and organizational risk

KEY POINTS • Burnout has three domains: physical and emotional exhaustion, cynicism and depersonalization, and feelings of ineffectiveness. • Resilience has three components: strength of the individual, rise above adversity, and positive adaptation. • Resilience is the key protective factor against burnout, as it shapes the individual’s efficacy, engagement, and personal accomplishment. • Resilience interventions include: education, integrate work/personal life, adjust relationship to work, mindfulness training, and work environment. • Organizations must address oncology clinician burnout through the direct implementation of successful, feasible, effective resilience model interventions. 772 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

SIDEBAR 1. Three Domains of Burnout 1. Physical and emotional exhaustion 2. Cynicism and depersonalization 3. Ineffectiveness factors that are important contributors driving oncologist burnout.6,9-27 These stressors are work-related factors that do not meet the clinician’s interpretation of the job or work expectations. For example, today, the oncology clinician is exposed to extended work hours, increased time in direct patient care, high occupational demands, lack of control over daily tasks, increased administrative responsibilities, increased time and use of electronic health record systems, limited decision making regarding patient care services, unclear job expectations, lack of social support, educational debt, and the evolving medical landscape.19-21 The identification of these internal and external factors is of extreme importance to help promote and tailor individual, and primarily organizational, interventions designed to prevent and target unaddressed burnout and build resilience.

When Does It Start?

Although less understood, it is entirely possible that the risk for burnout for some oncology clinicians begins early in their career during medical training.12-27 Several studies demonstrate that residents and medical students have high rates of burnout and disproportionate rates of depression and suicide. Every year, the United States loses approximately 400 physicians to suicide, the equivalent of at least one entire medical school.28 Though all medical schools provide a course in psychiatry to provide student insight into behavioral issues as related to patients, traditional curricula ignore these issues as related to medical students own development. Residency curricula are even less attuned to these issues. In fact, medical students are at higher risk for some psychiatric disorders than the general population, and suicidal ideation among them is estimated to be a very high 11.2% to 20%,29,30 with higher rates among African-American respondents.31 As a result of stigma, self-reported data likely underestimate these numbers. The prevalence and severity of depressive symptoms increases throughout school and rates of depression are higher in females than their male counterparts.30,32 Additional risk factors may include that 31% of medical students have a low sense of personal accomplishment, and 22% demonstrate at risk behaviors for alcohol use.32 Regarding mental health disorders, an estimated 12% of all medical trainees had probable major depression and 9.2% had probable mild to moderate depression with higher rates among medical students (versus residents) and women.31 Although medical students demonstrate higher physical quality of life scores than the general population, they also report overall lower mental quality of life scores.5 Even after completion of formal medical training, physicians continue to have elevated rates of psychiatric

RESILIENCE: APPROACH TO PREVENT AND COMBAT BURNOUT IN ONCOLOGY

disorders in comparison with the general population. Of note, male physicians complete suicide at a rate 70% higher than the population at large and female physicians at a startlingly high rate of 400%. To date, suicide is the only cause of death with risks greater for physicians than the general population.15,25,26,29 Burnout itself is not formally diagnosed as a disorder, given it is primarily recognized as an occupational-related condition; however, it shares similar symptomology with psychiatric disorders such as depression and post-traumatic stress disorders that are identified as precursors to the development of burnout development as well as consequences of burnout.6,10,11,19-21,33-38 The long-term personal and professional consequences associated with unaddressed burnout are of primary concern. Long-term unaddressed burnout may lead to personal consequences such as chronic health conditions (heart disease and obesity) or mental health conditions (depression, anxiety, substance use, and suicide).6,10,11,19-21,33-38 Professionally, long-term burnout leads to diminished quality care, reduced professional satisfaction, and overall accomplishment.19,20 For Dr. A, symptoms of physical exhaustion and negative emotions arise coupled with cynicism as work responsibilities increase and quality family time decreases. His feelings of ineffectiveness in the role of an oncologist adversely affect and directly conflict with his role as father and husband. Such symptoms indicate that Dr. A is in need of developing resilience skills to enhance his quality of life as well as optimize professional satisfaction. This evidence reveals the complex yet salient aspects and issues associated with burnout warranting intervention.

WHAT IS RESILIENCE: THEORY AND SCIENCE

Resilience, specifically psychological resilience, is a multifaceted theory that places emphasis on the human capacity to cope with, overcome, and become strengthened by adversity.39-52 Current clinical and research efforts center on the strengths of the individual, rather than the individual’s vulnerability, as a means of empowerment to rise above adversity, and persevere, resulting in positive adaptation (Sidebar 2). To date, the theory and study of resilience has shifted from a focus on the long-term adverse consequences of trauma to a focus on strength, triumph, and competence to build interventions tailored to foster resilience.42,47,49-51 The concept of resilience grew from within the developmental psychology by the study of children who were able to thrive, survive, and overcome negative abusive childhood environments with poor parenting.43,45 Resilience has also been applied to survivors’ populations of war, trauma, and the military.42,45,48,49 SIDEBAR 2. Three Components of Resilience 1. Strength of the individual 2. Rise above adversity 3. Positive adaptation

Resilience: Supports Health and Enhances Coping Through a Psychobiologic Mechanism

Current research approaches enhance our understanding of the concept of resilience by placing an emphasis on specific factors that support human health and enhance coping rather than highlighting stress-related factors associated with disease.39,41,42,47,49 Although evidence indicates environmental, neurologic, social and cultural factors are associated with the development of resilience, from a psychobiological perspective, resilience is believed to be a physiological positive adaptation to stress as it is associated with maintenance of the following: somatic, autonomic (sympathetic and parasympathetic), and central nervous systems.39,53-55 The specific brain regions associated with resilience involve the prefrontal cortical region and amygdala. Additionally, decreases in the stress hormone cortisol, neuropeptide Y (an anxiety neurotransmitter), and 5-dehydroepiandrosterone prevent initiation of the stress response by decreasing sympathetic nervous system activation.39,53-55 Also, elevated levels of the neurotransmitters serotonin and dopamine (“the reward center”) and neuropeptide oxytocin have also been linked to resilience.53-55 Positive emotions (e.g., happiness; optimism) play a crucial role in the development of resilience. Although it may appear that certain individuals are genetically predisposed to effectively cope with stressful situations, resilience is not necessarily an inherited trait, but rather a skill that can be learned and mastered. Yet, despite this strong scientific evidence, questions surround how to adequately describe and define resilience due to its complexity.

How Is Resilience Defined?

No universally accepted definition of resilience exists given its complex nature encompassing social, psychological, biologic, and cultural factors that act together to determine how the individual responds to stress.6,39,42,50,51 The definitions of resilience continue to advance and grow. However, most definitions and researchers agree that for resilience to be demonstrated, both adversity and positive adaptation must be present.42,44-52 Resilience is a positive response to adversities in the form of everyday minor stressors to key life-altering events. Resilience has been described as both a trait and a process, either present or absent, inherited or learned; however, according to Southwick, a well-known resilience expert, and colleagues,42,47,49 it likely exists on a continuum ever present to differing dimensions across several life domains influenced by psychological characteristics within the stress process. Ideally, resilient individuals persevere in the face of adversity and life stress leading to transformative positive growth, acceptance, and a sense of greater meaning in life. For example, a clinician who is unable to positively adapt to work stress may successfully adapt to his personal life, or theoncology clinician may be more resilient during the late phase of career, yet not another phase such as in early residency.42,47,49 As a result of interaction with the environment, resilience may change depending on the individual’s response to stress and interactions with others in the environment.42,47,49 asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 773

HLUBOCKY, ROSE, AND EPSTEIN

The Interplay Between Burnout and Resilience

Although several protective factors shield the individual against the development of burnout, such as peer support, communication skills training, and self-care, resilience is the key protective factor against burnout, as it shapes and enhances the individual’s efficacy, engagement, and personal accomplishment.6-8 Christine Maslach, a psychologist who has studied burnout extensively believes that burnout involves not simply the interaction between the individual and organization, but also the individual’s attitudes, selfappraisal, and appraisal of others.6-8 As such, burnout can be viewed as a barometer that measures a potentially toxic environment which did not support the clinician to manage his needs and emotions.6-8 Moreover, Maslach and colleagues found that consideration of the individual’s emotions promotes the individual’s sense of control, commitment, and self-efficacy that further protects the individual from burnout.6-8 In addition, several key emotional personality variables associated with resilience significantly minimizes the potential vulnerability to developing burnout, including a sense of coherence, thriving, hardiness (commitment, control), optimism, emotional competence, learned resourcefulness, self-efficacy, locus of control, potency, stamina, and personal causation.6-8,45 The individual’s ability to sustain and activate these resources in response to stress leads to a transformative active coping style required to directly address stressors and adversity.6-8 Research on physician resilience supports Maslach’s hypothesis. Zwack and Schweitzer56 conducted an interview study of 200 physicians in Germany to identify health-promotion strategies used by senior physicians to maintain resilience. Three core domains were identified to illustrate strategies and attitudes used to activate resources that lead to active coping and the promotion of resilience, including: job-related fulfillment; behavioral practice (e.g., leisure activities, limit work hours, and professional development activities); and shift in attitudes (e.g., acceptance and attention to positive work endeavors).56 In summary, despite stressful work conditions, physicians were able to activate resources to engage in positive coping strategies needed to foster resilience. As the cancer clinician like Dr. A learns to gain self-awareness and self-regulation of his emotions, which include thoughts and feelings, this enables him to build resources to find solutions to the issues at hand in a complex, ever-changing medical environment. Resilience, in the face of adversity, enables the cancer clinician to be armed with a broad spectrum of skills to develop more solutions to problems and positively adapt to the situation.

Resilience Interventions

Evidence-based, resilience-focused approaches have been promoted as burnout-prevention programs for clinicians tailored to enhance clinicians’ individual skills building and workplace engagement factors.57-59 These approaches, as well as mindfulness-based stress reduction programs, are believed to help foster clinician wellness by preventing and targeting unaddressed burnout directly. Therefore, from an 774 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

institutional perspective, it is in the best interests of any health care organization to implement and support oncology clinician wellness efforts aimed at promoting clinician resilience as a means to maximize value and improve overall quality of care.

HEALTH PROFESSIONAL RESILIENCE: TACKLING BURNOUT IN ONCOLOGY

Dr. A had promised his son he would attend tonight’s piano recital; however, one of his patients coded in clinic. The use of mindfulness training to build resilience skills would be of benefit for Dr. A. Rather than becoming angry and engaging in self-criticism with statements such as, “How does this always happen to me? I’m the worst father,” reframing critical thoughts and providing self-compassion with gratitude and acceptance would be beneficial to Dr. A in this situation. In response, phrases such as “I will try to end clinic earlier on recital days. I’ll ask Z if he will cover for me. I am a good father and love my son. I’m glad I was here for my patient” are reflective of resilience training. Although the toll of burnout has been clearly described, it is not as clear what to do to help clinicians become more resilient, engaged with work, and truly thriving in their professional roles. Resilience is not merely restoration to a prior (balanced) state of being. Resilient organisms not only bounce back, they also grow in ways to prevent future trauma and promote growth.59 Resilience does not necessarily lead to greater engagement; it is possible to be both resilient and burned out, surviving but not thriving—the walking wounded. Interventions should not merely try to prevent and mitigate burnout, they should also promote positive mood, physical and psychological health, joy, and flourishing in their clinical roles. Recently, West and colleagues57 reviewed 15 randomized trials and 37 cohort studies to address burnout. On average, interventions reduced overall burnout from 54% to 44%, as measured on the Maslach Burnout Inventory. Although individual (e.g., mindfulness, discussion, and stress management) and organizational (e.g., work environment changes and reduction in work hours) interventions produced similar improvements in burnout in the review by West et al,57 Panagioti et al60 suggested that institutional interventions might be more effective. Both expressed a need for testing of a wider range (and combinations) of interventions with larger sample sizes. Studies of resilience in the general population have mostly focused on people who experienced extreme trauma that had a beginning, middle, and end, unlike the ongoing vicarious trauma experienced by oncologists and other clinicians dealing with serious illness and death. Yet, there are lessons to be learned. Psychiatrists Southwick and Charney49 interviewed former prisoners of war, Special Forces instructors, and civilians who had experienced severe psychological traumas such as rape, sexual abuse, the loss of a limb, or cancer. They found that in spite of these extreme events, remarkably only a small percentage developed depression or post-traumatic stress disorder. They identified 10 resilience

RESILIENCE: APPROACH TO PREVENT AND COMBAT BURNOUT IN ONCOLOGY

factors: realistic optimism, facing fear, moral compass, religion and spirituality, social support, role models, physical fitness, brain fitness, cognitive and emotional flexibility, and a sense of meaning and purpose. Personality was found to be important also. The ability to form warm and caring relationships with others, so-called secure attachment, is associated with greater resilience, as is a perception of personal autonomy and perceiving oneself as competent.61 Conversely, cognitive rigidity, excessive need for certainty, and low emotional intelligence are associated with lower resilience. These traits, to some extent, are determined by early life experience and genetics, but are mutable.

Education and Training Are Important

For example, stress inoculation is a principle of applying graded and increasing levels of stress during training to ensure that the individual progressively adapts to stressors. Clearly, stress inoculation is not the modus operandi in clinical training in which the introduction to human suffering is more intense and uncontrolled. Better integration of work life and personal life confers greater resilience through helping individuals use the strengths developed in one domain to inform the other. This integration is not merely a balance between work (presumed to be aversive and stressful) and life, that which happens only when outside of work. Integration refers to finding meaning in work, setting appropriate but not rigid boundaries, and finding ways to engage more fully with work when the going gets rough. Adjusting one’s relationship to work is key. Many wellness programs emphasize healthy activities outside of work— time with family, vacations, exercise, yoga, etc. However, these approaches outside of work may have limited effect on resilience at work unless they are accompanied by a fundamental change in the workplace or one’s relationship to it, especially one’s attitudes and orientation toward the challenges in the workplace. Mindfulness training is one of the most widely studied approaches. Mindfulness refers to intentional awareness of one’s own thoughts and feelings, nonjudgmentally, with the goal of promoting clarity and compassion (Sidebar 3). By focusing on awareness and not relaxation, mindfulness training can help individuals be more aware of burnout in its early phases—noting changes in the body (e.g., headaches and muscle tension), emotions (irritability and sarcasm), or thoughts (blaming self or others)—before it becomes unmanageable, name it, and accept that it is present.62 Being more mindful of one’s own inner experience can build

SIDEBAR 3. Three Components of Mindfulness 1. Intention: intentional awareness of thoughts 2. Attention: ability to pay attention in the present, nonjudgmentally 3. Attitude: goal to promote acceptance and selfcompassion skills to mitigate burnout and enhance resilience, such as perspective-taking and cognitive reappraisal.49,59 Mindfulness also addresses some of the biologic underpinnings of resilience. For example, mindfulness programs for military recruits promoted self-awareness and enhanced “healthy” gene expression, providing one plausible pathway toward enhanced resilience.63 Mindfulness approaches emphasize “turning toward” difficult and potentially aversive challenges, identifying the earliest signs of stress, adopting an attitude of curiosity and beginner’s mind, the capacity to see a familiar situation with new eyes. Turning toward distressing situations is possible only if one can lower one’s level of reactivity and wait momentarily before reacting, mitigating stress before it becomes overwhelming. Various contemplative practices, including formal meditation and “mindful moments” during the workday, can help individuals recognize stressors more readily, respond to them sooner, and develop positive attitudes rather than fearful avoidance (Sidebar 4). Our study of 70 primary care physicians included mindfulness meditation, structured narrative exercises and appreciative inquiry (a strength-based interview approach), and discussion of key topics such as errors, grief and loss, meaningful moments, self-care, witnessing suffering, and communication with patients. After the program, physicians were not only less burned out and experienced less psychological distress, but they also reported greater empathy and better relationships with their patients.62,64 Their personalities changed to be more attentive and resilient, and the effects endured. Key elements of the program, according to participants, were a greater sense of community, having acquired self-awareness skills, and giving themselves permission to care for themselves in the interest of being more available to their patients. Subsequent studies suggest that patient ratings of their physicians also improved.65

The Work Environment

Healthy clinical teams promote resilience; supportive social environments lead to greater resilience. A supportive

SIDEBAR 4. Take a Mindful Moment During Your Workday Oncology clinicians routinely wash their hands between patients multiple times a day. Now is the time for a mindful moment: simply focus, pay attention to the sound of the water: its temperature, weight, and the way it feels on your hands. Look at the water, how it falls. Your thoughts may wander—do not worry, acknowledge them, and return your attention back to the water. Notice the smell of the hand soap, its texture, and weight on your skin. Your thoughts may wander—do not worry, return your focus to the water. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 775

HLUBOCKY, ROSE, AND EPSTEIN

SIDEBAR 5. Principles of Well-Functioning Primary Care Practices That Might Be Adopted in Oncology (adapted from Sinsky et al68) • Proactive planned care, with previsit planning and previsit laboratory tests • Sharing clinical care among a team, with expanded rooming protocols, standing orders, and panel management • Sharing clerical tasks with collaborative documentation (scribing), nonphysician order entry, and streamlined prescription management • Improving communication by verbal messaging and in-box management • Improving team functioning through colocation, team meetings, and work flow mapping social environment is associated with increases in neurotransmitters and hormones associated with well-being (and their corresponding receptors), presumably because of social epigenetic processes. For example, supportive environments lead to increased production of dopaminergic receptors in key areas of the brain, receptors that are involved in the brain’s reward circuits. Conversely, the toxic combination of high responsibility, low sense of control, and isolation sets the stage for a sense of exhaustion, powerlessness, and helplessness.66 Putting clinicians in morally compromising situations, excessive cognitive load because of interruptions and dysfunctional electronic health record systems, the increase in meaningless documentation and regulatory requirements, and placing increasing pressure on clinicians to see more patients without regard to quality are environmental influences that must be addressed by health care teams and health care institutions.67,68 Merely reducing work hours will likely not be effective in promoting resilience without enhancing the work environment. Sinsky et al68 suggest a set of changes to enhance the work environment that may hold promise in reducing burnout and enhancing clinician resilience and well-being. Their suggestions revolve around shared care and teamwork and are based on observations of primary care physicians who report greater joy in practice. However, many of these changes could be adopted in oncology outpatient settings. These are listed in Sidebar 5. Although not directed at resilience per se, these enhance the quality of clinicians’ workday and merit further investigation.

Just as individuals can be mindful of their level of burnout and well-being, health care organizations can monitor these as quality indicators and disseminate findings to raise collective awareness and resolve.69 In this case, leadership is key; individual practitioners are more likely to thrive in those organizations in which the leadership has a demonstrated commitment to clinician well-being. Case reports of health care organizations that have implemented organizational approaches to clinician resilience emphasize principles that are summarized in Sidebar 6.70 Although there are few controlled trials and institutions tend to report their own positive outcomes (improvements in burnout, distress, and the clinical environment), these suggestions are sensible and pragmatic; we cannot afford to delay until results of larger randomized trials are available.

THE JOY INITIATIVE: A STUDY OF POSITIVE PSYCHIATRY AND MINDFULNESS TRAINING ON LEVELS OF LIFE SATISFACTION AND WELLNESS

Among medical students, mindfulness meditation has been demonstrated to decrease symptoms of anxiety.71 Mindfulness-based stress reduction interventions also decrease tension/anxiety, depression, severity of stress, and mood disturbance scores on the Profile of Mood States and confusion/bewilderment in this population. Similarly, these studies have also revealed increases in vigor/activity, trainees feeling more effective in managing stressful situations, and increased empathy.72,73 Cognitive behavioral therapy and positive psychology exercises have also proven effective

SIDEBAR 6. Nine Principles of Organizational Leadership That Can Promote Clinician Resilience and Well-Being (adapted from Shanafelt and Noseworthy70) 1. Acknowledging and assessing the problem 2. Recognizing the behaviors of leaders that can increase or decrease burnout 3. Using a systems approach to develop targeted interventions to improve efficiency and reduce clerical work 4. Cultivating community at work 5. Using rewards and incentives strategically 6. Assessing whether the organizations actions are aligned with the stated values and mission 7. Implementing organizational practices and policies that promote flexibility and work-life balance 8. Providing resources to help individuals promote self-care 9. Supporting organizational science (study the factors in your own institution that contribute to the problem, and invest in solutions) 776 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

RESILIENCE: APPROACH TO PREVENT AND COMBAT BURNOUT IN ONCOLOGY

in decreasing depressive symptoms and improving positive attitude and happiness/outlook on life for clinically ill individuals and the general population,71-75 but there have been no studies to date demonstrating efficacy in the medical student population. A systematic review of stress management programs designed for medical students did identify three mindfulness-based interventions for medical trainees that demonstrated positive outcomes; however, to date, no interventions focusing on emotional resilience training skills using cognitive behavioral therapy combined with mindfulness training have been identified.76 The investigators at Michigan State University developed an easily pilot-deployable programmatic intervention to help students and residents discuss and address their own burnout issues to enhance trainee well-being and emotional health with a focus on developing strengths to face the emotional challenges of medical training.

Programmatic Intervention Design

Michigan State University Department of Psychiatry resident physicians created and taught 60-minute weekly classes for 10 weeks for students at the Michigan State University College of Osteopathic Medicine. Half of each session was devoted to mindfulness therapy and the other half to cognitive behavioral therapy exercises. The cognitive behavioral therapy exercises were created and written by our lead resident physician (M. Rose), primarily based on the philosophy and writings of Victor Frankl. Individual weekly topics included selecting and practicing joyful activities, identifying one’s core strengths and virtues, creating a vision, naming goals, daily compassion, and practice of gratitude. Each session had weekly mindfulness exercises and homework. The mindfulness topics included breathing, body awareness, eating, walking, and sound. Each class session concluded with brief homework assignments that reinforced the week’s theme. Seven female and seven male students elected to participate in the intervention. A control group was approved near the end of the intervention consisting of 79 medical students who did not participate in the intervention. The Beck Anxiety Inventory, Fordyce Happiness Scale, and the Authentic Happiness Inventory were administered to both groups to assess the impact of the intervention. The Beck Anxiety Inventory is a 21-item self-report measure of anxiety. Higher scores reveal greater levels of anxiety. The Fordyce scale is a self-report happiness scale consisting of two parts. Section one measures the overall perception of mood (rating of 0–10), and the second part measures the percentage of time a subject estimates feeling happy, unhappy, or neutral. The Authentic Happiness Inventory is a 25-question survey assessing aspects of well-being including self-esteem, life purpose, and emotional supports. For the intervention group, these surveys were administered at the outset, midpoint, and termination of the 10-week intervention. For the control group, the surveys were administered at the termination (10-week intervention point) of the study. All data were analyzed using SPSS (IBM) and MYSTAT (SYSTAT; San Jose, CA).

RESULTS

Fourteen students participated in the Joy Initiative project (Fig. 1). None of the intervention participants withdrew from this study. Figure 1 depicts study participants results. The analyses revealed the mean Beck Anxiety Inventory scores of participating (intervention) students declined from 13.8 at the first session (standard deviation [SD] 8.1) to 6.8 (SD 6.8) after the last session. This decrease in scores was statistically significant (p = .007; 95% CI, −8.089, −1.711; degrees of freedom [df] 9; SD 4.4). There were no statistically significant differences between the means of male or female (intervention group) participants. The mean Beck Anxiety Inventory score for the 79 students in the control group was higher of 9.6 (SD 7.5), compared with the intervention group mean of 6.8 (SD 6.8). However, this difference between groups was not statistically significant (p = .326; 95% CI, −4.423, 11.756; df 8; SD 10.5). The mean Authentic Happiness Inventory Scores of participating (intervention) students improved, increasing from 79.2 (SD 9.6) to 87.3 (SD 13.9). The increase in Authentic Happiness Inventory scores between the beginning and endpoint of this intervention was a statistically significant change (p = .046; 95% CI, 0.186, 16.214; df 9; SD 11.2). There was a statistically significant difference between the female and male changes in scores, with female mean score difference 13.3 points higher than that of male mean scores (p = .007; 95% CI, 5.061, 21.605; df 7; SD [female] 4.676; SD [male] 5.586). At the conclusion of the 10-week intervention, the mean Authentic Happiness Scale Score of the Intervention Group was 87 (SD 13.9) compared with the Control Group 75 (SD 12.3; analysis of variance 8.8; df 1; p = .004). For the Fordyce Happiness Scale, Part One: for the intervention participants, the mean Fordyce Part One scores increased from 7.6 (SD 1.0) to 7.8 (SD 0.4). This difference was not statistically significant (p = .182; 95% CI, −0.419, 1.419; df 3; SD 0.6). There were no statistically significant differences between the means of male or female (intervention group) participants. The control group had a lower Fordyce Part One happiness score of 6.6 (SD 2.0) compared with that of the intervention group of 7.8 (SD 0.4). However, this difference between groups was not statistically significant (p = .178; 95% CI, −0.280, 1.080; df 4; SD 0.5). For Fordyce Part Two, the mean Fordyce Part Two scores improved for the intervention participants, increasing from 56.2 (SD 18.0) to 69.8 (SD 18.7). The difference between the two data sets was not statistically significant (p = .090; 95% CI, −2.058, 23.658; df 9; SD 18.0). There were no statistically significant differences between the means of male or female (intervention group) participants. The intervention group had a higher Fordyce Part Two happiness score of 69.8 (SD 18.7) compared with the control group of 54.5 (SD 23.9). However, this difference between groups was not statistically significant (p = .102; 95% CI, −37.699, 4.099; df 9; SD 29.2). The availability of classroom space limited our intervention timing with respect to the academic schedule. The first measures were taken from the intervention group as soon as students returned from a 2-week winter break, at a time they might be expected to naturally feel most relaxed and asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 777

HLUBOCKY, ROSE, AND EPSTEIN

happy. In addition, the students provided reports via qualitative feedback that they very much enjoyed the class and looked forward to each week’s session. None of the students provided negative feedback regarding the intervention. Students reported improved life satisfaction and increased ability to cope with stressors. One student recommended additional videos and interactive sessions, but reported overall satisfaction from participation. None of the participants withdrew from the study. The homework and techniques provided to students were intentionally brief, high-yield exercises, allowing them to practice these techniques while going to class and studying. Indeed, students also reported using the techniques demonstrated each week.

Institutional Response With a Long-term Impact

After the Joy Initiative pilot intervention study, the medical school administration provided support and funding for continuation of the project. Since these monthly “Joy Initiative Focus Group” meetings began, changes have been made on an administrative level. As a direct result of com-

munication during these meetings, a new medical college staff position was created, a Program Officer for Outreach and Inclusion, with duties including coordination and provision of administrative support to continue the Joy Initiative monthly meetings. Student representatives from the medical school diversity committee spearheaded a lead role in the organization of the Joy Initiative events, and a minority student event related to the Joy Initiative was incorporated into student orientation activities for incoming medical students. The Joy Initiative Focus Group meetings continue on a monthly basis, with average attendance ranging from 50 to 70 students across 3 campus sites. In addition, the interventions used in this pilot study are now incorporated into formal elective classes offered at both the osteopathic and allopathic medical schools at Michigan State University, (“Happiness and Emotional Resilience Training for Health Care Providers Elective,” Course PSC 591 301, Michigan State University College of Osteopathic Medicine; and “Resilience and Happiness Promotion for Health Care Providers,” HM 590 Section 304, Michigan State University

FIGURE 1. Joy Initiative Participant Outcomes for Anxiety and Happiness

Outset, midpoint, and termination of the 10-week intervention compared with control group.

778 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

RESILIENCE: APPROACH TO PREVENT AND COMBAT BURNOUT IN ONCOLOGY

College of Human Medicine). Future programs should include a larger sample size with long-term follow-up to investigate whether students continue to use these tools years after the intervention to determine if they maintain high levels of satisfaction and low levels of depression, stress, and burnout. In addition, core components from the intervention could be used not only for all forms of trainees, but also for other health care professionals, including cancer clinicians, hematologist/oncologists, radiation oncologists, surgical oncologists, oncology nurses, and physician assistants.

CONCLUSION

Oncology clinicians are at increased risk for burnout; however, building resilience in the face of adversity to positively adapt to the changing health care system is key. Although the optimal program to address burnout requires additional research, organizations must not delay to act. Organizations must set a precedent and address oncology clinician burnout through the direct implementation of successful, feasible, effective resilience interventions such as the Joy Initiative.

References 1. Hlubocky FJ, Back AL, Shanafelt TD. Addressing burnout in oncology: Why cancer care clinicians are at risk, what individuals can do, and how organizations can respond. Am Soc Clin Oncol Educ Book. 2016;35:271-279. 2. Allegra CJ, Hall R, Yothers G. Prevalence of burnout in the U.S. oncology community: results of a 2003 survey. J Oncol Pract. 2005;1:140-147. 3. Shanafelt TD, Gradishar WJ, Kosty M, et al. Burnout and career satisfaction among US oncologists. J Clin Oncol. 2014;32:678-686. 4. Blanchard P, Truchot D, Albiges-Sauvin L, et al. Prevalence and causes of burnout amongst oncology residents: a comprehensive nationwide cross-sectional study. Eur J Cancer. 2010;46:2708-2715. 5. Freudenberger HJ. Staff burn-out. J Soc Issues. 1974;30:159-165. 6. Bahrer-Kohler S (ed). Burnout for Experts: Prevention in the Context of Lving and Working. New York: Springer; 2014. 7. Schaufeli WB, Leiter MP, Maslach C. Burnout: 35 years of research and practice. Career Dev Int. 2009;14:204-220. 8. Maslach C, Schaufeli WB, Leiter MP. Job burnout. Annu Rev Psychol. 2001;52:397-422. 9. Cohn KH, Panasuk DB, Holland JC. Workplace burnout. In Cohn KH (ed). Better Communication for Better Care: Mastering PhysicianAdministration Collaboration. Chicago, IL: Health Administration Press; 2005;56-62. 10. Kash KM, Holland JC, Breitbart W, et al. Stress and burnout in oncology. Oncology (Williston Park). 2000;14:1621-1633, discussion 1633-1634, 1636-1637. 11. Trufelli DC, Bensi CG, Garcia JB, et al. Burnout in cancer professionals: a systematic review and meta-analysis. Eur J Cancer Care (Engl). 2008;17:524-531.

18. Figley CR. Compassion Fatigue: Coping With Secondary Traumatic Stress Disorder in Those Who Treat the Traumatized. New York: Brunner/Maze; 2005. 19. Shanafelt T, Dyrbye L. Oncologist burnout: causes, consequences, and responses. J Clin Oncol. 2012;30:1235-1241. 20. Shanafelt TD, Raymond M, Kosty M, et al. Satisfaction with work-life balance and the career and retirement plans of US oncologists. J Clin Oncol. 2014;32:1127-1135. 21. McManus IC, Keeling A, Paice E. Stress, burnout and doctors’ attitudes to work are determined by personality and learning style: a twelve year longitudinal study of UK medical graduates. BMC Med. 2004;2:29. 22. Alarcon G, Escheleman KJ, Bowling NA. Relationships between personality variables and burnout: a meta-analysis. Work Stress. 2009;23:244-263. 23. Shanafelt TD, Bradley KA, Wipf JE, et al. Burnout and self-reported patient care in an internal medicine residency program. Ann Intern Med. 2002;136:358-367. 24. Oreskovich MR, Shanafelt T, Dyrbye LN, et al. The prevalence of substance use disorders in American physicians. Am J Addict. 2015;24:30-38. 25. Tyssen R, Hem E, Vaglum P, et al. The process of suicidal planning among medical doctors: predictors in a longitudinal Norwegian sample. J Affect Disord. 2004;80:191-198. 26. Shanafelt TD, Balch CM, Dyrbye L, et al. Special report: suicidal ideation among American surgeons. Arch Surg. 2011;146: 54-62.

12. Shanafelt TD, Sloan JA, Habermann TM. The well-being of physicians. Am J Med. 2003;114:513-519.

27. Jager AJ, Tutty MA, Kao AC. Association between physician burnout and identification with medicine as a calling. Mayo Clin Proc. 2017;92:415-422.

13. Kakiashvili T, Leszek J, Rutkowski K. The medical perspective on burnout. Int J Occup Med Environ Health. 2013;26:401-412.

28. Andrew L, Brenner B. Physician suicide. http://emedicine.medscape. com/article/806779-overview. Accessed July 19, 2015.

14. Bourg Carter S. High-Octane Women: How Superachievers Can Avoid Burnout. Amherst, NY: Prometheus; 2011.

29. Dyrbye LN, Thomas MR, Massie FS, et al. Burnout and suicidal ideation among U.S. medical students. Ann Intern Med. 2008;149: 334-341.

15. Center C, Davis M, Detre T, et al. Confronting depression and suicide in physicians: a consensus statement. JAMA. 2003;289:3161-3166. 16. Ramirez AJ, Graham J, Richards MA, et al. Mental health of hospital consultants: the effects of stress and satisfaction at work. Lancet. 1996;347:724-728. 17. Ramirez AJ, Graham J, Richards MA, et al. Burnout and psychiatric disorder among cancer clinicians. Br J Cancer. 1995;71:1263-1269.

30. Dyrbye L, Thomas M, Shanafelt T. Systematic review of depression, anxiety, and other indicators of psychological distress among U.S. and Canadian medial students. Acad Med. 2006;81:354-373. 31. Goebert D, Thompson D, Takesh*ta J, et al. Depressive symptoms in medical students and residents: a multischool study. Acad Med. 2009;84:236-241.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 779

HLUBOCKY, ROSE, AND EPSTEIN

32. Dyrbye LN, Thomas MR, Huntington JL, et al. Personal life events and medical student burnout: a multicenter study. Acad Med. 2006;81:374-384.

51. Fletcher D, Sarkar M. Psychological resilience: a review and critique of definitions, concepts, and theory. Eur Psychol. 2013;18: 12-23.

33. Bianchi R, Schonfeld IS, Laurent E. Is burnout a depressive disorder? A reexamination with special focus on atypical depression. Int J Stress Manag. 2014;21:307-324.

52. Johnson J, Panagioti M, Bass J, et al. Resilience to emotional distress in response to failure, error or mistakes: A systematic review. Clin Psychol Rev. 2017;52:19-42.

34. Bianchi R, Boffy C, Hingray C, et al. Comparative symptomatology of burnout and depression. J Health Psychol. 2013;18:782-787.

53. Charney DS. Psychobiological mechanisms of resilience and vulnerability: implications for successful adaptation to extreme stress. Am J Psychiatry. 2004;161:195-216.

35. Kraft U. Burned out: your job is extremely fulfilling. It is also extremely demanding--and you feel overwhelmed. You are not alone. Sci Am Mind. 2006:29-33. 36. Toker S, Melamed S, Berliner S, et al. Burnout and risk of coronary heart disease: a prospective study of 8838 employees. Psychosom Med. 2012;74:840-847. 37. Honkonen T, Ahola K, Pertovaara M, et al. The association between burnout and physical illness in the general population--results from the Finnish Health 2000 Study. J Psychosom Res. 2006;61:5966. 38. Shirom A, Melamed S. Does burnout affect physical health? A review of the evidence. In Alexander-Stamatios AG, Cooper CL (eds). Research Companion to Organizational Health Psychology. Cheltenham: Edward Elgar Publishing; 2005;599-622. 39. Epstein RM. Attending: Medicine, Mindfulness, and Humanity. Delran, NJ: Simon & Schuster; 2016. 40. Nedrow A, Steckler NA, Hardman J. Physician resilience and burnout: can you make the switch? Fam Pract Manag. 2013;20:25-30. 41. Antonovsky A. The sense of coherence: An historical and future perspective. In McCubbin HI, Thompson EA, Thompson AI, et al (eds). Stress, coping, and health in families: Sense of Coherence and resiliency. Thousand Oaks, CA: Sage; 1998a;3-20. 42. Southwick SM, Pietrzak RH, Tsai J, et al. Resilience: an update. PTSD Research Quarterly. 2015;25:1-10. 43. Garmezy N. Resiliency and vulnerability to adverse developmental outcomes associated with poverty. Am Behav Scientist. 1991;34:416430. 44. Luthar SS, Cicchetti D, Becker B. The construct of resilience: a critical evaluation and guidelines for future work. Child Dev. 2000;71:543562. 45. VanBreda AD. Resilience Theory: A Literature Review. Pretoria, South Africa: South African Military Health Service; 2001. 46. Bonanno GA. Loss, trauma, and human resilience: have we underestimated the human capacity to thrive after extremely aversive events? Am Psychol. 2004;59:20-28. 47. Southwick SM, Bonanno GA, Masten AS, et al. Resilience definitions, theory, and challenges: interdisciplinary perspectives. Eur J Psychotraumatol. 2014;5:1-14. 48. Pietrzak RH, Southwick SM. Psychological resilience in OEFOIF Veterans: application of a novel classification approach and examination of demographic and psychological correlates. J Affect Disord. 2011;133:560-568. 49. Southwick SM, Charney DS. Resilience: The Science of Mastering Life’s Greatest Challenges. Cambridge, U.K.: Cambridge University Press; 2012. 50. Aburn G, Gott M, Hoare K. What is resilience? An integrative review of the empirical literature. J Adv Nurs. 2016;72:980-1000.

780 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

54. Ozbay F, Fitterling H, Charney D, et al. Social support and resilience to stress across the life span: a neurobiologic framework. Curr Psychiatry Rep. 2008;10:304-310. 55. Puglisi-Allegra S, Andolina D. Serotonin and stress coping. Behav Brain Res. 2015;277:58-67. 56. Zwack J, Schweitzer J. If every fifth physician is affected by burnout, what about the other four? Resilience strategies of experienced physicians. Acad Med. 2013;88:382-389. 57. West CP, Dyrbye LN, Erwin PJ, et al. Interventions to prevent and reduce physician burnout: a systematic review and meta-analysis. Lancet. 2016;388:2272-2281. 58. Back AL, Steinhauser KE, Kamal AH, et al. Building resilience for palliative care clinicians: an approach to burnout prevention based on individual skills and workplace factors. J Pain Symptom Manage. 2016;52:284-291. 59. Epstein RM, Krasner MS. Physician resilience: what it means, why it matters, and how to promote it. Acad Med. 2013;88: 301-303. 60. Panagioti M, Panagopoulou E, Bower P, et al. Controlled interventions to reduce burnout in physicians: a systematic review and metaanalysis. JAMA Intern Med. 2017;177:195-205. 61. Ng JYY, Ntoumanis N, Thøgersen-Ntoumani C, et al. Self-determination theory applied to health contexts: a meta-analysis. Perspect Psychol Sci. 2012;7:325-340. 62. Krasner MS, Epstein RM, Beckman H, et al. Association of an educational program in mindful communication with burnout, empathy, and attitudes among primary care physicians. JAMA. 2009;302:1284-1293. 63. Johnson DC, Thom NJ, Stanley EA, et al. Modifying resilience mechanisms in at-risk individuals: a controlled study of mindfulness training in Marines preparing for deployment. Am J Psychiatry. 2014;171:844-853. 64. Beckman HB, Wendland M, Mooney C, et al. The impact of a program in mindful communication on primary care physicians. Acad Med. 2012;87:815-819. 65. Schroeder DA, Stephens E, Colgan D, et al. A brief mindfulness-based intervention for primary care physicians: a pilot randomized controlled trial. Am J Lifestyle Med. 2016;1-9. 66. Abramson LY, Seligman ME, Teasdale JD. Learned helplessness in humans: critique and reformulation. J Abnorm Psychol. 1978;87:4974. 67. Epstein RM, Privitera MR. Doing something about physician burnout. Lancet. 2016;388:2216-2217. 68. Sinsky CA, Willard-Grace R, Schutzbank AM, et al. In search of joy in practice: a report of 23 high-functioning primary care practices. Ann Fam Med. 2013;11:272-278.

RESILIENCE: APPROACH TO PREVENT AND COMBAT BURNOUT IN ONCOLOGY

69. Wallace JE, Lemaire JB, Ghali WA. Physician wellness: a missing quality indicator. Lancet. 2009;374:1714-1721.

73. Shapiro SL, Schwartz GE, Bonner G. Effects of mindfulness-based stress reduction on medical and premedical students. J Behav Med. 1998;21:581-599.

70. Shanafelt TD, Noseworthy JH. Executive leadership and physician wellbeing: nine organizational strategies to promote engagement and reduce burnout. Mayo Clin Proc. 2017;92:129-146.

74. Seligman MEP, Rashid T, Parks AC. Positive psychotherapy. Am Psychol. 2006;61:774-788.

71. Britton WB, Shahar B, Szepsenwol O, et al. Mindfulness-based cognitive therapy improves emotional reactivity to social stress: results from a randomized controlled trial. Behav Ther. 2012;43:365-380.

75. Pace TW, Negi LT, Adame DD, et al. Effect of compassion meditation on neuroendocrine, innate immune and behavioral responses to psychosocial stress. Psychoneuroendocrinology. 2009;34:87-98.

72. Rosenzweig S, Reibel DK, Greeson JM, et al. Mindfulness-based stress reduction lowers psychological distress in medical students. Teach Learn Med. 2003;15:88-92.

76. Shiralkar MT, Harris TB, Eddins-Folensbee FF, et al. A systematic review of stress-management programs for medical students. Acad Psychiatry. 2013;37:158-164.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 781

MARKHAM, GENTILE, AND GRAHAM

Social Media for Networking, Professional Development, and Patient Engagement Merry Jennifer Markham, MD, Danielle Gentile, PhD, and David L. Graham, MD OVERVIEW Social media has become an established method of communication, and many physicians are finding these interactive tools and platforms to be useful for both personal and professional use. Risks of social media, or barriers to its use, include perceived lack of time, privacy concerns, and the risk of damage to one’s reputation by unprofessional behavior. Of the social media platforms, Twitter has become favored by physicians and other health care professionals. Although one of the most obvious uses of social media is for rapid dissemination and receipt of information, oncologists are finding that social media is important for networking through blogs, Facebook, and Twitter. These platforms also have potential for providing opportunities for professional development, such as finding collaborators through networking, participation in Twitter journal clubs, and participating in online case-based tumor boards. Social media can also be used for patient engagement, such as through participation in tweet chats. There is emerging data that patient engagement through these platforms may lead to improvement in some health-related outcomes; however, data are sparse for oncology-specific outcomes. Efforts are underway to determine how to assess how social media engagement impacts health outcomes in oncology patients.

ocial media has evolved over the years to become an established method of communication in our current society. Eighty-six percent of Americans are internet users, and of those, almost 80% use Facebook, 32% use Instagram, and 24% use Twitter.1 Sixty-two percent of Americans get their news on social media, primarily through Facebook, but increasingly through Twitter.2 The awareness of social media and its relevance in society continue to grow, certainly bolstered by the fact that the newest U.S. president communicates regularly with the public through Twitter. Social media refers to tools or platforms for the interactive, or social, sharing of user-generated content. Social media includes a wide variety of platforms for the sharing of words (e.g., Twitter, blogs), images (e.g., Pinterest, Instagram), and video (e.g., YouTube, Snapchat, Periscope). Sites such as LinkedIn tend to be used more for professional networking, and others, such as Doximity, are geared specifically toward social networking between physicians and other health care professionals. As social media technology evolves, so does the potential for personal and professional use of these platforms. Twitter has become a favored forum for health care communication for physicians, patient advocates, and health care organizations. Through Twitter, a user can post messages (“tweets”) of up to 140 characters, and assuming the user’s account is public rather than private, these messages

can be shared (“retweeted”) by other Twitter users to their followers. The use of a hashtag (a word or phrase preceded by the # sign, such as #breastcancer or #myeloma) in a tweet serves to link the message to a conversation or a virtual community. Hashtags also are useful for searching for information about a topic on Twitter. Thompson et al describe in more detail the anatomy of a tweet and some basics of using Twitter, including valuable resources for those physicians just getting started, or wanting to get started, using the platform.3 It is difficult to estimate how many oncologists are active users of social media. A survey conducted of Canadian oncology physicians and oncology trainees found that 72% of respondents used social media.4 The authors found that social media use varied by age, a typical finding in social media use surveys, such that 93% of oncology fellows and 72% of early-career oncologists reported social media use, compared with 39% of midcareer oncologists. When these oncologists and oncology trainees used social media for professional development, they reported that their goals were for networking (55% of respondents), sharing research (17%), and leadership development (13%). One of the most obvious uses of social media is for rapid dissemination and receipt of information. Breaking news commonly appears on Twitter prior to appearing in newspapers or television news broadcasts. Medical research shared

From the Division of Hematology and Oncology, Department of Medicine, University of Florida College of Medicine, Gainesville, FL; Levine Cancer Institute, Charlotte, NC; Western Region, Levine Cancer Institute, Charlotte, NC. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Merry Jennifer Markham, MD, Division of Hematology and Oncology, Department of Medicine, University of Florida College of Medicine, P.O. Box 100278, 1600 SW Archer Rd., Gainesville, FL 32610; email: [emailprotected]. © 2017 American Society of Clinical Oncology

782 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

SOCIAL MEDIA FOR NETWORKING, PROFESSIONAL DEVELOPMENT, AND PATIENT ENGAGEMENT

through social media also has potential to reach much broader audiences in a more rapid, real-time fashion. Because of this potential, many journals now have a presence on Twitter. For example, the journals of the American Society of Clinical Oncology (ASCO)—the Journal of Clinical Oncology (@JCO_ASCO), Journal of Oncology Practice (@JOP_ ASCO), and Journal of Global Oncology (@JGO_ASCO)—are all represented on Twitter. There are numerous blog posts and articles written by physicians, including oncologists, that describe why the use of social media has value for both personal and professional uses.5-9 Here, we will discuss some of those potential uses for social media, including using social media for professional development, networking, and patient engagement. First, however, it is important to address the potential risks of social media use.

BARRIERS AND RISKS OF SOCIAL MEDIA USE

Adilman et al found that the most frequently cited barrier to using social media was not having enough time, as reported by 59% of participants.4 Campbell and colleagues also identified a lack of time as a potential barrier; however, their research indicated that this is an area of disparate views among the physicians they studied. Although some physicians felt the time needed to use social media was an impediment to patient care, others felt time was not problematic.10 Other potential barriers along this theme that are anecdotally cited by physicians include decreased productivity that may result from time spent on social media, lack of time to learn how to use social media effectively, and being overwhelmed by social media and technology overload. Privacy concerns are frequently reported as barriers to social media use.4,10,11 Although most physicians who use social media, and especially Twitter, enjoy the engagement with the general public, patients, and patient advocates, some

KEY POINTS • Social media participation allows for rapid and realtime information sharing and receiving, and physicians are finding platforms such as Twitter to be valuable for health care communication. • Networking through social media allows physicians to make connections with others with similar interests, foster collaboration, and gain support for personal and professional growth. • Professional development opportunities exist through social media, such as through networking (including through Twitter at medical meetings) and participation in Twitter journal clubs and online case-based discussions and tumor boards. • Patients and patient advocates are engaging with each other, physicians, and health care organizations through social media using platforms such as Facebook and Twitter, including participation in tweet chats. • Patient engagement in social media may lead to improvement in some health-related outcomes.

physicians are concerned about engaging with patients on social media and avoid social media use for this reason. Another concern is the permanence of anything shared on social media. For example, a deleted tweet on Twitter is not truly gone. A common expression is that posts shared on social media are written in pen, not pencil. Health care organizations who employ or work with physicians are concerned about the potential harms from unprofessional or unethical behavior on social media. Physicians themselves are worried about inadvertently sharing misinformation or sharing something unprofessional.10 Unethical or unprofessional information shared on social media could pose a risk to a physician’s or a health care organization’s reputation. An early study by Chretien and colleagues examined the tweets of self-identified physicians on Twitter to determine whether physicians were behaving unprofessionally.12 Of the 260 users they collected data on, a total of 5,156 tweets were analyzed. One hundred forty tweets (3% of total tweets) were categorized as unprofessional. Thirty-eight of the tweets (0.7%) contained potential patient privacy violations, 33 (0.6%) contained profanity, 14 (0.3%) contained sexually explicit material, and four (0.1%) included discriminatory statements. Twelve tweets contained possible conflicts of interest, such as promoting health products sold on their website, and 10 tweets were statements about medical therapies that were counter to existing medical knowledge or guidelines. Many health care organizations have established policies for social media use to be proactive in establishing rules and guidelines for online professionalism. Dizon and colleagues catalog some of the common concepts in social media policies and give practical guidance for using social media in the oncology practice.13 ASCO published online its “Ten Tips for Use of Social Media” that serves as a quick resource for responsible physician use of Twitter and other social media platforms (www.asco.org/sites/www.asco.org/files/asco_ socialmedia_card.pdf). Additional ASCO-related resources for social media use can be found in Table 1.

SOCIAL MEDIA FOR NETWORKING AND PROFESSIONAL DEVELOPMENT

Professional development and networking go hand in hand on social media. Networking and connecting with other physicians on social media is one of the main benefits of participation. Traditionally, physicians have interacted and engaged with physicians in their own communities and medical centers, with networking limited by physical location. With Twitter, however, a physician can meet and interact with physicians around the world who may have similar professional or research interests, thus creating opportunities for the sharing of ideas, collaboration, and connection. By using cancer-specific hashtags on Twitter, oncologists can participate in discussions and network with colleagues in these virtual communities centered around common cancer interests.14 Some examples of common cancer-specific hashtags are listed in Table 2. Interacting through blogs or online forums provides additional opportunities for networking with colleagues. ASCO asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 783

MARKHAM, GENTILE, AND GRAHAM

TABLE 1. ASCO-Related Resources for Social Media Resource

Twitter Handle or Website

ASCO

@ASCO

ASCO publications Journal of Clinical Oncology

@JCO_ASCO

Journal of Oncology Practice

@JOP_ASCO

Journal of Global Oncology

@JGO_ASCO

The ASCO Post

@ASCOPost

Cancer.Net

@CancerDotNet

Conquer Cancer Foundation

@ConquerCancerFd

ASCO University Course: Use of Social Media

https://goo.gl/cYqH6J

Ten Tips for Use of Social Media

https://goo.gl/m11SDL

Social Media for Cancer Care Providers 101

https://goo.gl/JsE8C6

Practical Guidance: The Use of Social Media in Oncology Practice13

https://goo.gl/sKG2KG

Roundtable: The Use of Social Media in Oncology Practice (Podcast)

Connection (http://connection.asco.org) is a relevant hub of information and networking opportunities for ASCO members, from commenting on ASCO Connection blogs (thus engaging with the authors and other commenters) to participating in the ASCO Connection Discussion forums. Doximity and LinkedIn are other sites often used for professional networking. Twitter has become the next frontier for the traditional journal club, moving the discussion about a journal article out of the classroom and into the public, international space.15 Thangasamy and colleagues describe their experience with the international urology journal club (#urojc) on Twitter.16 Each month, the moderators of #urojc host a

TABLE 2. Examples of Common Cancer-Specific Hashtags Hashtag

Topic

#AYACSM

Adolescent and young adult cancer

#BCSM

Breast cancer

#CRCSM

Colorectal cancer

#GynCSM

Gynecologic cancer

#HNCSM

Head and neck cancer

#KCSM

Kidney cancer

#LCSM

Lung cancer

#LeuSM

Leukemia

#LymSM

Lymphoma

#MMSM

Multiple myeloma

#PallOnc

Palliative oncology

#PancSM

Pancreatic cancer

#PCSM

Prostate cancer

#PedCSM

Pediatric cancer

#SCSM

Sarcoma

#SurvOnc

Cancer survivorship

784 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

48-hour discussion focusing on recently published journal articles. The extended time allows for Twitter users in different time zones to participate. Over a 12-month period, 189 unique users representing 19 different countries participated in their monthly #urojc Twitter discussion. Two oncologyspecific Twitter journal clubs include the radiation oncology (#radonc) journal club moderated by @Rad_Nation and the bone marrow transplant journal club (#bmtojc) hosted by the American Society for Blood and Marrow Transplantation (@ASBMT). Tweet chats are regularly held on Twitter by several groups, always organized around a hashtag and led by a moderator or several moderators. For example, the Breast Cancer Social Media (#BCSM) chat began in 2011 as a conversation on Twitter and has grown into a robust virtual community (@BCSMChat; http://bcsm.org/). The #GYNCSM monthly tweet chat (@gyncsm; http://gyncsm.blogspot.com/) centers around discussions about gynecologic cancers and was established in 2013. Participants in these chats, and in other cancer-related tweet chats, include medical oncologists, surgeons, radiation oncologists, nononcology physicians, nurses, psychologists and other health care professionals, patients, patient advocates, and health care organizations. Physician participation in tweet chats provides the opportunity to network with colleagues doing similar work, meet potential research collaborators, advocate for patients, and engage with patients and patient advocates. Participation in online case-based discussion is another opportunity for professional development. Located on the ASCO Connection Discussion (https://connection.asco. org/discussion), the Molecular Oncology Tumor Board has been an active online tumor board since January 2015. These educational modules are presented in blog post form and consist of a case presentation followed by several discussion questions, with the discussion moderated by specialist physicians. Users of the ASCO Connection can provide answers and further discussion in the comments

SOCIAL MEDIA FOR NETWORKING, PROFESSIONAL DEVELOPMENT, AND PATIENT ENGAGEMENT

of the blog post. A similar case-based educational opportunity is provided by the moderators of the TeamHaem blog (https://teamhaem.com/), with a focus on hematology cases. The moderators (@TeamHaem) present a case on their blog and then request discussion on Twitter using the hashtag #TeamHaem to organize the discussion. Follow-up blog posts include updates about the case based on discussions held on Twitter. Networking with colleagues and other health professionals at meetings through Twitter is now mainstream. Over the last several years, Twitter use at the annual meeting of ASCO has grown significantly.17,18 Attendees of the meetings routinely share information about abstracts being presented, scientific breakthroughs, or observations about the meeting. This allows for a much broader audience for the scientific research being shared, extending the reach of the information presented. Those who are not in attendance— or even those attendees who are attending different sessions in different rooms—can stay up to date on the news coming out of the meeting halls. Those attendees who are sharing tweets may find that composing tweets during a meeting—which requires editing the content to be shared to a maximum of 140 characters—can allow for reflection and better understanding of the information.19 Connecting virtually with colleagues at the meeting has the potential to foster broader discourse on research studies, provide opportunities to collaborate, and create new friendships. And, importantly, opportunities exist through “tweet-ups” to meet Twitter friends in person at social gatherings geared specifically for that purpose. The opportunity for support and online mentorship is not to be overlooked. Reaching out on Twitter, for example, with a simple message of frustration or joy can generate responses that provide support and acknowledgment that we are not alone. For example, when one of our authors (Markham) sent a tweet after an emotionally challenging week caring for patients with cancer, it was met with supportive responses and, ultimately, a collaboration and friendship.20 In addition to connecting through Twitter, there is potential for networking in private or closed Facebook groups. The European Society for Medical Oncology (ESMO) sponsors a closed Facebook group, the ESMO Young Oncologists group, for early-career oncologists. Examples of two robust, interactive groups that exist for women physicians are the Physicians Mom Group (PMG) and the Hematology and Oncology Women Physician Group. As of February 2017, the PMG Facebook group had over 66,000 members, and the Hematology and Oncology Women Physician Group had 646 members. Because they are out of the public eye, the conversation among physicians and oncologists in these groups can be more in depth and personal, and these groups have become a place for support, both personal and professional, and friendship. Radiation oncologist Miriam Knoll (@ MKnoll_MD) described her experience with the PMG Facebook group as follows: “As physicians and individuals, we need to give and accept support from our fellow colleagues.

Think about it: Where else could a physician share a moment of frustration or achievement with 50,000 colleagues and receive immediate feedback including 2,000 likes and hundreds of supportive comments? This is the unique platform of PMG.”21

SOCIAL MEDIA FOR PATIENT ENGAGEMENT

“Patient engagement” is a term that is gaining great use in health care discussions. Reading the term on the surface, it is hard to argue against encouraging patients to take a more active role in their care. A difficulty in the discussions is that there is not a clearly accepted definition of patient engagement. This is further complicated by the near-synonymous use of the phrases “patient activation” along with “patientand family-centered care.” An often-used definition of patient engagement has been advanced by Angela Coulter and focuses on the activities by patients and health care professionals to “promote and support active patient and public involvement in health and health care and to strengthen their influence on health care decisions.”22 The expanded use of social media platforms by various health care institutions has raised the question of their impact on patient engagement and whether that impact could be shown to translate to improved outcomes. There are certainly some social media platforms that may lend themselves to improving patient engagement more than others. The simple presence of a website is likely not adequate. A study of patient’s perception of specialty society websites in Australia, Europe, and the United States gave an average rating of 3.2 out of 10, with the majority of patients rating the websites as failing to meet an “adequate” standard of information delivery.23 More health care institutions are moving toward mechanisms with a greater level of interactivity. Facebook, having a monthly active user base of more than 1.5 billion and a high level of potential interactivity, is an attractive platform. Twitter, with more than 300 million accounts, has the capacity for interaction with the use of retweets, but only 2% of original tweets are retweeted. Pinterest, in contrast, may have fewer users, with less than 70 million visits per day, but the rate of “re-pins” can be as high as 80%, suggesting a greater level of engagement.24 YouTube, with more than a billion active users but less interactivity, has a potential for education and information dissemination. As an example, one review of YouTube videos in 2014 found more than 280 videos on preparation prior to colonoscopy, each with more than 5,000 views.25 Blogs and webcasts/podcasts provide the lowest level of interactivity but can still be useful as educational platforms. The practice of patients using the internet to find health information has been long recognized. Thackery et al in 2013 reported that nearly 75% of patients will begin looking for health care information via a search engine, but nearly 33% will ultimately use social media sites as well by the time their search is completed.26 Although patients turn to social media for gathering health information, they may be less likely to actively interact with other social media users asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 785

MARKHAM, GENTILE, AND GRAHAM

to share information. A survey of more than 3,300 patients found that less than 4% were willing to communicate with their physician regarding health goals or test results via social media. Only 11.7% were willing to engage in peer coaching with other patients through Facebook.27 These results have led to the recognition that patients who are “information altruists” are required for these communities to truly succeed. These patients must be willing to engage with other patients, caregivers, researchers, and other stakeholders within social media platforms.28 Objective data of the impact of social media in improving patient engagement and outcome results specifically in the area of oncology are lacking, so a more generalized review is needed. One measure of patient engagement is the “Patient Activation Measure,” a tool that uses responses to 13 statements to assess a patient’s level of engagement. Patients defined as “less activated” by this tool are more likely to have unmet medical needs and delay medical care.29 Chronically ill patients who are “more activated” are more likely to adhere to treatment and obtain regular chronic care.30 What was not assessed by these studies, however, was the impact of different interventions on the activation scores. Grosberg et al has reported a positive impact of social media use and patient activation.31 Camoni is a Hebrew-language social media site established in 2008. Participants in the four largest communities on the site, namely diabetes, pain, depression, and hypertension, were surveyed from 2012 through 2013. Their survey found that increased frequency and duration of visits to the site were associated with increased Patient Activation Measure scores. Interestingly, no difference was seen between active participants and “lurkers” (i.e., those who visited the site but did not interact with other participants). Research regarding the impact of social media interventions on nononcology health outcomes, however, are more abundant. Lelutiu-Weinberger et al reported in 2015 the results of an online intervention program to reduce HIV risk in young men who have sex with men.32 Their program modified an in-office program recognized as effective in reducing risky behaviors such as failure to use condoms. Although the studied population of 41 men was small, significant reductions in risk behaviors were seen between baseline and follow-up. Improvements were also observed in knowledge of the connection between substance use and sexual risk. Saberi and Johnson reported a correlation between internet use for health care engagement purposes and improved clinical outcomes in HIV-positive individuals.33 They recruited nearly 1,500 respondents via a multitude of social media platforms. Use of the internet for health care engagement was associated with a significantly higher chance of antiretroviral therapy adherence and chance of an undetectable HIV viral load. The significance impact was confirmed on multivariate analysis. Attai et al reported outcomes related to participation in “tweet chats” for patients with breast cancer in 2015.34 The Breast Cancer Social Media tweet chat was established in 2011. The number of Twitter users using the #BCSM hashtag 786 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

increased to more than 14,000 in 2014. A survey of those users obtained 206 responses. Participation was associated with a significantly lower rate of extreme/high anxiety levels. An interesting impact was that 28.4% of participants reported subsequent volunteer efforts, representing a surrogate marker for increased engagement. The potential use of social media to assist with clinical trial improvement is also of interest. Descriptions of mechanisms in place to assist with accrual through social media exist, but little research exists regarding their effectiveness. Khatri et al reported the impact of free social media efforts on a U.K. trial determining the impact of nonsteroidal antiinflammatory drugs following gastrointestinal surgery.35 They reported 18.2% of the needed accrual occurring in a short time period. Their click-through rate was 33.7% compared with previously reported rates of 3.2% for paid Facebook advertising of a breast cancer trial.36 Online communities have also been associated with cost improvements. A U.K. mental health community, the Big White Wall, was established in 2007. This community allows patients to perform self-assessments, join guided support programs, and even receive live therapy via Skype. An economic evaluation of the program reported in 2011 showed that members of the community had, on average, one less visit to a general care practitioner, hospital, or emergency department. Factoring in the per patient cost of the program, this led to a net savings of $615 per patient per year.37 Sawesi et al reported a systematic review of the literature regarding the impact of health information technology on patient engagement and health behavior change.38 In their summary of 160 papers, 82.9% of papers reported improvement in patient engagement after using IT platforms. The only statistically significant impact, however, was seen in those platforms that were internet based. Seventy-five percent of internet-based IT interventions defined as “usable” showed positive health outcomes. Eleven percent of studies showed no impact on health behavior. Undesirable effects, including increased anxiety, were noted in 18%. The data that exist are intriguing and suggest a substantial impact on patient engagement and subsequent improvement in health care outcomes. More research is needed, however, to define the impact of social media interventions in the oncology population. We are at a nascent enough point that the questions to be addressed and the mechanism to address them are as of yet undefined, and what is needed is a mechanism to identify the most appropriate mechanism to study the issue. To that end, a group of oncology health professionals interested in social media for improving cancer care has been established to start to explore these questions.39 COSMO, the Collaboration for Outcomes of Social Media on Oncology, aims to define a mechanism whereby we can better assess the ongoing impact of efforts involving social media to benefit oncology patients. Through these mechanisms and subsequent trials, we can aim to better define whether these are worthwhile efforts in which to continue.

SOCIAL MEDIA FOR NETWORKING, PROFESSIONAL DEVELOPMENT, AND PATIENT ENGAGEMENT

References 1. Greenwood S, Perrin A, Duggan M. “Social Media Update 2016.” Pew Research Center, November 11, 2016. www.pewinternet. org/2016/11/11/social-media-update-2016/.

22. Carman KL, Dardess P, Maurer M, et al. Patient and family engagement: a framework for understanding the elements and developing interventions and policies. Health Aff (Millwood). 2013;32:223-231.

2. Gottfried A, Shearer E. “News Use Across Social Media Platforms 2016.” Pew Research Center, May 26, 2016. www.journalism. org/2016/05/26/news-use-across-social-media-platforms-2016/.

23. Ow D, Wetherell D, Papa N, et al. Patients’ perspectives of accessibility and digital delivery of factual content provided by official medical and surgical specialty society websites: a qualitative assessment. Interact J Med Res. 2015;4:e7.

3. Thompson MA, Majhail NS, Wood WA, et al. Social media and the practicing hematologist: Twitter 101 for the busy healthcare provider. Curr Hematol Malig Rep. 2015;10:405-412.

24. Timimi FK. The shape of digital engagement: health care and social media. J Ambul Care Manage. 2013;36:187-192.

4. Adilman R, Rajmohan Y, Brooks E, et al. Social media use among physicians and trainees: results of a national medical oncology physician survey. J Oncol Practice. 2016;12:79-80, e52-60.

25. Basch CH, Hillyer GC, Reeves R, et al. Analysis of YouTube videos related to bowel preparation for colonoscopy. World J Gastrointest Endosc. 2014;6:432-435.

5. Pennell NA. “The draw of social media for oncology professionals.” ASCO Connection, April 19, 2016. https://connection.asco.org/blogs/ draw-social-media-oncology-professionals. 6. Snipelisky D. Social media in medicine: a podium without boundaries. J Am Coll Cardiol. 2015;65:2459-2461. 7. Choo EK, Ranney ML, Chan TM, et al. Twitter as a tool for communication and knowledge exchange in academic medicine: a guide for skeptics and novices. Med Teach. 2014;37:411-416. 8. Thompson MA. Using social media to learn and communicate: it is not about the tweet. Am Soc Clin Oncol Educ Book. 2015;35:206-211. 9. Fisch MJ, Chung AE, Accordino MK. Using technology to improve cancer care: social media, wearables, and electronic health records. Am Soc Clin Oncol Educ Book. 2016;36:200-208.

26. Thackeray R, Crookston BT, West JH. Correlates of health-related social media use among adults. J Med Internet Res. 2013;15:e21. 27. Jenssen BP, Mitra N, Shah A, et al. Using digital technology to engage and communicate with patients: a survey of patient attitudes. J Gen Intern Med. 2015;31:85-92. 28. Kohane IS, Altman RB. Health-information altruists—a potentially critical resource. N Engl J Med. 2005;353:2074-2077. 29. Hibbard JH, Cunningham PJ. How engaged are consumers in their health and health care, and why does it matter? Res Brief. 2008;8:1-9. 30. Hibbard JH, Greene J. What the evidence shows about patient activation: better health outcomes and care experiences; fewer data on costs. Health Aff (Millwood). 2013;32:207-214.

10. Campbell L, Evans Y, Pumper M, et al. Social media use by physicians: a qualitative study of the new frontier of medicine. BMC Med Inform Decis Mak. 2016;16:91.

31. Grosberg D, Grinvald H, Reuveni H, et al. Frequent surfing on social health networks is associated with increased knowledge and patient health activation. J Med Internet Res. 2016;18:e212.

11. Alpert JM, Womble FE. Just what the doctor tweeted: physicians’ challenges and rewards of using Twitter. Health Commun. 2015;31:824-832.

32. Lelutiu-Weinberger C, Pachankis JE, Gamarel KE, et al. Feasibility, acceptability, and preliminary efficacy of a live-chat social media intervention to reduce HIV risk among young men who have sex with men. AIDS Behav. 2014;19:1214-1227.

12. Chretien KC, Azar J, Kind T. Physicians on Twitter. JAMA. 2011;305:566568. 13. Dizon DS, Graham D, Thompson MA, et al. Practical guidance: the use of social media in oncology practice. J Oncol Pract. 2012;8:e114-e124. 14. Katz MS. “Hashtags in Cancer Care: Embedding Meaning in Digital Health.” Cancer Tag Ontology, November 4, 2013. www.symplur.com/ blog/hashtags-cancer-care-embedding-meaning-digital-health. 15. Roberts MJ, Perera M, Lawrentschuk N, et al. Globalization of continuing professional development by journal clubs via microblogging: a systematic review. J Med Internet Res. 2015;17:e103. 16. Thangasamy IA, Leveridge M, Davies BJ, et al. International Urology Journal Club via Twitter: 12-month experience. Eur Urol. 2014;66:112-117. 17. Chaudhry A, Glodé LM, Gillman M, et al. Trends in twitter use by physicians at the American Society of Clinical Oncology annual meeting, 2010 and 2011. J Oncol Pract. 2012;8:173-178. 18. Katz M. “Twitter Use at Three Annual Professional Meetings (20122014).” November 24, 2014. https://socialmedia.mayoclinic. org/discussion/twitter-use-at-three-annual-professionalmeetings-2012-2014. 19. McGuckin DG. Live tweeting: a tool for learning and reflection. BMJ. 2016;354:i3975. 20. Dizon DS. “Help from an Online Community.” ASCO Connection, September 13, 2012. https://connection.asco.org/blogs/help-onlinecommunity. 21. Knoll M. “The Case for Connectivity.” ASCO Connection, February 29, 2016. http://connection.asco.org/blogs/case-connectivity.

33. Saberi P, Johnson MO. Correlation of internet use for health care engagement purposes and HIV clinical outcomes among HIV-positive individuals using online social media. J Health Commun. 2015;20:10261032. 34. Attai DJ, Cowher MS, Al-Hamadani M, et al. Twitter social media is an effective tool for breast cancer patient education and support: patientreported outcomes by survey. J Med Internet Res. 2015;17:e188. 35. Khatri C, Chapman SJ, Glasbey J, et al; STARSurg Committee. Social media and internet driven study recruitment: evaluating a new model for promoting collaborator engagement and participation. PLoS One. 2015;10:e0118899. 36. Fenner Y, Garland SM, Moore EE, et al. Web-based recruiting for health research using a social networking site: an exploratory study. J Med Internet Res. 2012;14:e20. 37. Laurance J, Henderson S, Howitt PJ, et al. Patient engagement: four case studies that highlight the potential for improved health outcomes and reduced costs. Health Aff (Millwood). 2014;33:1627-1634. 38. Sawesi S, Rashrash M, Phalakornkule K, et al. The impact of information technology on patient engagement and health behavior change: a systematic review of the literature. JMIR Med Inform. 2016;4:e1. 39. Attai DJ, Sedrak MS, Katz MS, et al; Collaboration for Outcomes on Social Media in Oncology (COSMO). Social media in cancer care: highlights, challenges & opportunities. Future Oncol. 2016;12:15491552.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 787

THE ROAD OF MENTORSHIP

The Road of Mentorship Kelly J. Cooke, DO, Debra A. Patt, MD, MPH, MBA, and Roshan S. Prabhu, MD, MS OVERVIEW Mentorship can be the cornerstone of professional development and career satisfaction. There is literature to support that mentorship not only improves job satisfaction, but also improves productivity, facilitates personal growth, and can rekindle our passion while lessening the risk of compassion fatigue. Mentorship is a developmental relationship that changes as the relationship evolves. There are two broad categories of mentorship: traditional and transformational. There are four subtypes within each of those areas: formal, informal, spot, or peer. Mentorship is critical to the professional development of junior colleagues. Good mentorship is guiding and steering younger partners and other colleagues toward paths of success. As a mentor, one should be looking for opportunities for formal professional development and engagement of mentees. Self-motivation is the hallmark of the successful mentee. The mentee should be able to set his or her own goals, strive to actively seek feedback, ask questions, and keep an accurate record of progress. Although the onus is on the mentee to reach out, mentorship has bidirectional value directly related to the efforts of both parties. There are many benefits to mentorship, such as the promotion of learning, personal development, improved job satisfaction, and improved job performance. Barriers exist, including the rapidly changing landscape of oncology, time constraints, lack of self-awareness, and generational differences. Through a career, mentoring needs will change, as will mentors.

entorship can be the cornerstone of professional development and career satisfaction. Although there is not a lot of mentorship research specific to oncology, there is literature to support that mentorship not only improves job satisfaction, but also improves productivity, facilitates personal growth, and can rekindle our passion while lessening the risk of compassion fatigue.1 Mentorship is a developmental relationship that changes as the relationship evolves. Like any relationship, there is risk of dysfunction. Mentorship is intended to be a learning relationship to guide individuals in their own way to sort through their career challenges whether directly related to oncology practice or psychosocial functions as an oncologist. Both mentees and mentors must have self-awareness. There needs to be a balance of support and challenge. A mentorship contract can be used to clarify expectations, set boundaries, and define objectives.2 It is important to understand what mentorship is not. Mentorship should not be confused with preceptorship. A preceptor is a teacher as in the fellowship model of training. Nor is a mentor a faculty advisor. Mentorship is different from sponsorship. A sponsor is more of a coach or advocate in the work place who has some leadership power who can lean in with you, whereas a mentor listens and guides while providing practical insight and constructive criticism. Ideally, a good mentor helps the mentee achieve his or her full

potential. Successful mentorship is a two-way street that requires clear expectations on both the mentee and mentor’s parts, open communication, dedication, and feedback along the journey. There are two broad categories of mentorship: traditional and transformational. Traditional mentorship is the model of the older and wiser physician sharing knowledge and guiding the young and inexperienced physician, as we more often see in academics and research. Conversely, transformational mentorship lacks the hierarchy. The mentor and mentee are considered equals and learn from one another as we often see in the community setting.1 Within each of those areas, one may engage in formal, informal, spot, or peer mentorship opportunities. Given different needs, most people will be exposed to all four subtypes of mentorship. Formal mentorship is more structured and may be initiated through a professional organization or institution. For those in research, there is often a formal mentoring relationship in which the mentor is appointed. Informal mentorship often occurs on an ad hoc basis and may exist over a long period. Informal mentoring may be done by colleagues, individuals more senior to you, or even those outside your department or institution. Spot mentoring is typically a single conversation with someone with whom you seek expert advice. For example, you have a complicated patient with a rare malignancy, and you seek

From the UW Cancer Center at ProHealth Care, Waukesha, WI; The US Oncology Network/McKesson Specialty Health, Austin, TX; Southeast Radiation Oncology Group/Levine Cancer Institute, Carolinas Healthcare System, Charlotte, NC. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Kelly J. Cooke, DO, UW Cancer Center at ProHealth Care, N16 W24131 Riverwood Dr., Waukesha, WI 53188; email: [emailprotected]. © 2017 American Society of Clinical Oncology

788 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

COOKE, PATT, AND PRABHU

out your department chair for advice. Lastly is peer mentorship. Peer mentoring is typically a small group of individuals at a similar career stage who meet regularly to support one another. Although this article focuses on the roles of the mentor and mentee, it is important to recognize that the institution in which they practice also plays a key role. Today, oncologists practice in a variety of settings that impacts options and support for mentorship. Ideally, the institution or practice helps to foster mentorship opportunities. In some settings that means dedicated time or funding, and in other settings, it is formally connecting mentees and mentors. Regardless of the size of the practice or type of setting (academic, government, community, etc.), it is about creating a culture the promotes mentorship and champions the recognition of the mentorship process and the value to the mentee, mentor, and institution. Fostering a culture for mentorship can start in the trenches.

GETTING STARTED

During one’s career, mentorship needs change. It is important to assess your needs. For example, during early career, the focus often is in the transition from trainee to attending, technical skills needed for your institution, having difficult conversations, and work-life balance. During midcareer, needs may include professional development, leadership skills, keeping up with the literature, and volunteering (such as on ASCO committees). During late career, needs may focus on becoming a mentor, leadership in the community, and transitioning to retirement. Once you understand your needs, you seek mentors. It is not about one person meeting all needs. Often you ask mentors for specific areas based on their expertise or your view of them as a role model for that need. For example, as an early career oncologist with less technical skill in end-of-life communication, you may seek guidance from an oncologist who the nurses view as good at those discussions or even the local palliative care provider. It can provide an opportunity for you to exchange expertise. You teach the palliative care provider something about prognostication from an oncology viewpoint, and the palliative care provider helps you

KEY POINTS • Mentorship is a developmental relationship that changes as the relationship evolves and can serve as the cornerstone of professional development and career satisfaction. • There are two broad categories of mentorship: traditional and transformational. Within each of those areas, there are four subtypes of mentorship: formal, informal, spot, or peer. • Mentorship is critical to the professional development of our young colleagues. • Good mentorship is guiding and steering junior partners and other colleagues toward paths of success. • Self-motivation is the hallmark of the successful mentee.

improve your skill with having difficult conversations. That type of mentorship could be formal, in which you ask the provider to enter a partnership with that expectation, or the experience could be informal, in which you ask to observe during a family meeting. As you begin on the journey of mentorship, it is important to be open to formal, informal, spot, and peer mentoring opportunities. Additionally, not all mentors will be oncologists. For example, as a midcareer oncologist wanting to be more active at your local hospital, you may find that your hospital’s chief of staff could be a good mentor for leadership skills. Likewise, you may find peers from other institutions at a similar stage in their career that provide an opportunity for peer mentorship in which you learn from one another, sharing knowledge as you grow together. Working as an ASCO volunteer creates many opportunities to network and find mentors or mentees. Attending the ASCO annual meeting provides the chance for spot mentoring as well. Finally, do not give up. Mentorship to an extent is about chemistry and trust. Often you may find a mentorship opportunity once a friendship has developed. Although similar personalities may create an opportunity to build rapport and foster comfort for open communication, the down side is that it is easier to stay within your comfort zone. Often you may learn more from someone who looks at things from a different perspective.2

TENETS OF A GOOD MENTOR

Mentorship is critical to the professional development of our young colleagues. Mentorship relationships may differ from highly structured with very specific goals, assignments, and timelines to less clearly articulated relationships with variable meeting schedules and less deliverables. It must be a relationship that is based on mutual trust and value. It is important to be aligned regarding the goals of the relationship, as the goals in professional development can be variable. Professional success for our colleagues may be based on satisfying certain criteria for advancement—key positions, publications, or managing collaboration. In private practice, success is initially measured by your ability to build a practice, be a good partner, and contribute to your community and later your ability to lead. Mentorship in an academic setting may be a more formalized relationship, whereas mentorship in private practice is often an informal process benchmarked by communication, education, dissemination of social capital, support, and presentation of opportunities for professional growth and development. Leaders in oncology have new skills to learn in managing the organization’s business and development needs, leadership, and strategy and managing change. Good mentorship is guiding and steering junior partners and other colleagues toward paths of success. This may mean introducing them to critical relationships within the institution or community. It also means introducing them to referring physicians, endorsing their addition to the practice, and giving them opportunities to contribute to community or organizational efforts and lead. Sometimes it also means asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 789

THE ROAD OF MENTORSHIP

helping them navigate obstacles and derailing behaviors that could affect professional success. Not all mentorship relationships are formalized. They do not have a particular cadence or duration, though sometimes in more structured relationships, they will. Although formal mentorship relationships may have structured communication timelines, informal mentorship relationships may have varied communication, sometimes communicating several times a week, a few times a month, or only a few months out of the year. Most mentorship relationships span over many years, even decades. Most mentorship relationships have value to both mentor and mentee. In community practice, these relationships are more collaborative, as there is an egalitarian nature to the organizational structure, and in academic settings, these relationships are more hierarchical. Some of the best advice a mentor can give to a mentee in early practice is the importance of “the four A’s” of practice success: ability, availability, affability, and alacrity.3 There is tremendous value in being ready and happily willing to give good counsel to your referring providers. When you easily help people solve problems, you become their partner in problem solving. With changes in oncology, an individual’s success is less about the individual and more about the teams they lead. Expertise in leading teams of clinical and research collaborators and optimizing communication within the team is critical to professional development. With the advent of the oncology care model and other alternative payment models in our practice, we are dependent on well-integrated team-based care. As a practitioner, we are more dependent than ever on the many hands that help the patients we serve.

MEETING NEW LEADERSHIP CHALLENGES: WHAT GOT THEM HERE WILL NOT GET THEM THERE

Mentors help identify sources of professional growth. As a new clinician enters practice, there are a new set of challenges, new competencies that medical training does not prepare you well to navigate, and new obstacles that even mastery of “the four A’s” and great team dynamics will not adequately prepare you to tackle. In clinical practice, you have to work with many collaborators: hospital systems, community support organizations, and referring providers. There are also new challenges in understanding the business of medical practice. For most young doctors, the business challenges are new and require some supplemental knowledge. Monthly review of financial statements and understanding the structures of collaboration with these partnering organizations frequently requires additional knowledge of finance, operations, strategic planning, and negotiation. Certainly, leading a practice and managing conflict and challenges internal and external to your practice requires new skill sets. Some doctors pick these new competencies up very naturally, but most of us require additional training. Becoming competent in these areas for a physician leader can help dramatically with leadership success. Advising junior oncologists to consider supplemental education 790 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

SIDEBAR. Vital Aspects for Successful Mentorship Partnerships per Allen and Poteet9 • Establish an open communication system with reciprocal feedback • Set standards, goals, and expectations • Establish trust • Care for and enjoy each other • Allow mistakes • Participate willingly • Demonstrate flexibility • Consider constraints to mentoring • Learn from others • Work on common tasks • Be open and comfortable in finance and operations via remote courses (such as www. coursera.org/) and to read books on leadership, conflict resolution, managing change, and influence (such as the Crucial Conversations series by Patterson and colleagues). Some physicians may elect to pursue additional degrees, such as a master’s degree in business administration or health care administration. In the academic world, the challenges in leadership are also new and require new skill sets. They may include more business knowledge, but certainly require knowledge of leading teams, strategy, and organizational development. As a mentor, one should be looking for opportunities for formal professional development and engagement of mentees.

FORMAL PROFESSIONAL DEVELOPMENT

Identify and recommend formal leadership development for mentees. Many large academic institutions, hospital systems, and large practices have formal leadership development programs. Sometimes these programs can be accessed through professional organizations, like ASCO’s leadership development program (www.asco.org/training-education/ professional-development/leadership-development-program) or possibly internally within your own group or health system. Some community oncology practices offer formal professional development. For example, Texas Oncology has developed a formal leadership development course that is statewide and resembles a mini-MBA that is managed in collaboration with a local business school. In the US Oncology Network, there are tier I, II, and III leadership-development courses designed for incremental leadership-development training for incrementally invested physicians. Participation in these programs is costly to the mentee in time and money, but formal leadership development is an investment in the future. Physicians are not the only ones who can benefit from this kind of leadership development. When thinking about mentees, we should include advanced practice practitioners, nurse practitioners, and physician assistants. In addition to leadership-development courses, mentees with high development potential may benefit from working with an executive coach.

COOKE, PATT, AND PRABHU

OPPORTUNITIES FOR ENGAGEMENT

There is not one clear path for engagement, but meeting with your mentee and knowing their professional goals will help you identify opportunities for them to engage and lead. This requires a mentor to engage with their mentees about their goals of professional development, look at the landscape ahead, and know what opportunities they cannot see. Frequently, a mentor will have to leverage his or her social capital on behalf of the mentee to offer opportunities to lead in new arenas. This may come in the form of your recommendation to work with a local hospital or philanthropic group; it may come in the form of fostering engagement with professional organizations like ASCO, American Society for Radiation Oncology, Community Oncology Alliance, and American Association for Cancer Research, making the connections they need so they can lead research within a collaborative group and giving talks at national meetings to make a name for themselves in cancer care.

TENETS OF BEING A GOOD MENTEE

The traditional concept of mentorship in the medical field is primarily derived from academic practice. Academic medicine has long-standing formalized career paths (i.e., clinical track, medical education track, tenure/research track, etc.) with specific timelines, expectations, checklists, and requirements for career advancement that are generally consistent between institutions. These career-advancement guidelines can be readily found through the Office of Faculty Affairs or its equivalent at academic medical schools (e.g., Emory University: http://med.emory.edu/administration/ faculty_affairs_dev/promotions.html). The requirements for promotion are some mix of scholarship, teaching, and service based on your specific track. There are even readily available guidelines and recommendations for mentor/ mentee conversations according to timeline and academic track (e.g., University of Pennsylvania: www.med.upenn. edu/mentee/documents/mentor_guide.pdf). However, this type of formalism does not exist in the community medical setting, and there is significantly more variability in the concept of what career development means and the role of the mentee and mentor in nonacademic practice. In community oncology practice, career development has no standard and can represent a variety of different scenarios depending on many factors including: practice type (hospital employed vs. physician owned), practice size, location (urban vs. not), role of research in the growing trend of hybrid-type community-academic practices, practice structure (existence of a cancer center or formal cancer program with physician administration), and involvement with accreditation organizations such as ASCO Quality Oncology Practice Initiative, American Society for Radiation Oncology, and Commission on Cancer, to name a few. An important initial step for the mentee in community practice is to know your specific interests and strengths and have a vision for what you would consider to be a successful career while considering the needs of your practice, group, or organization. In most community practices, the expectation is for a busy

clinical practice with additional responsibilities performed either with relatively small amounts of protected time or “on your own time.” Because of this structure, it is important not to overextend yourself by making too many commitments or getting involved in too many endeavors that will lead to failure, physician burnout, or both. It is imperative for the mentee to set personal goals and have an idea of individual strengths and what “you can bring to the table” to begin identifying a career path and, consequently, who the ideal mentor would be. It is typical to have multiple mentors because of the various aspects of community oncology practice. It is common to have a different mentor for patient care/referral relationships, for hospital leadership/committee access and networking, and for research efforts. Another key distinction between academic and community mentorship is motivation. Academic mentors, especially midcareer faculty, are incentivized to provide quality mentorship as part of their advancement criteria, and their track record of successful mentorship is a metric that is scrutinized during the promotion process. That is generally not the case in the community setting, where these formalized systems are not in place. This underscores that the mentee should be aware that their mentors are providing their time, energy, and expertise for little to no external benefit, and as such, the mentee should be self-motivated, take initiative, and have an active role in the relationship, recognize and acknowledge the time and effort their mentor is providing, be flexible and understanding of the mentor’s schedule, and be prompt for all interactions. There is published literature on the characteristics of successful or failed mentoring relationships. A recent study surveyed medical mentors and mentees and found that successful mentoring relationships were characterized by reciprocity, mutual respect, clear expectations, personal connection, and shared values. Failed mentoring relationships were characterized by poor communication, lack of commitment, personality differences, perceived (or real) competition, conflicts of interest, and the mentor’s lack of experience.4 There are many areas of community oncology practice in which strong mentorship can be beneficial. A recent study published survey results of physicians in a community-based mentoring program and demonstrated that participants reported a variety of benefits, including setting goals (62%), planning next steps in their career (60%), gaining new insights (52%), completing a long-deferred goal (30%), reducing stress (19%), and improving self-confidence (19%).5 Self-motivation is the hallmark of the successful community-based mentee. The mentee should be able to set his or her own goals, strive to actively seek feedback, ask questions, and keep an accurate record of progress. The mentee cannot expect or rely on the mentor to do the heavy lifting. One of the most important aspects of the mentor is to be a guide on the road of mentorship, directing the mentee toward opportunities, but not doing the work for the mentee. A downside of nonacademic practice is the relatively reduced access to networks that form the governing bodies of asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 791

THE ROAD OF MENTORSHIP

national organizations and journal editorial boards. A mentor can facilitate crossing the initial barrier to entering these organizations, which is usually the most difficult obstacle for successful engagement. There are pitfalls to avoid as a mentee. Many pitfalls stem from being conflict adverse or lacking confidence. For example, a mentee who eludes conflict may over commit and agree to tasks that are irrelevant to his or her career or even allows himself or herself to be walked over. Someone who lacks confidence may not be comfortable asking for help or questioning the mentor. Vaughn and colleagues describe several mentee missteps to avoid.6,7 Clear communication is key. Avoid assumptions. Instead, ask for clarification when needed. Feedback and constructive criticism are invaluable. Although the mentee should actively seek and be open to feedback, receiving constructive feedback can be a learned skill to help avoid being defensive or sensitive to criticism. In the community setting, there is generally less hierarchy between the mentor and mentee. It is helpful to know something about your mentor’s life outside of work to develop a relationship and improve communication. However, do not try to force a friendship or become artificially close, as that can potentially detract from the intended tone of mutual respect.

The mentor-mentee relationship is meant to be mutually beneficial and has been shown to help with work-life balance and reduce rates of physician stress and burnout.8 A successful relationship requires investment of time and effort from both the mentee and mentor, and emphasizing certain positive characteristics and avoiding known pitfalls can help maximize the success of both parties. Allen and Poteet9 assembled details about vital aspects for successful mentorship relationships, which are outlined in the Sidebar.

CONCLUSION

Although the onus is on the mentee to reach out, mentorship has bidirectional value directly related to the efforts of both parties. There are many benefits to mentorship, such as the promotion of learning, personal development, improved job satisfaction, and improved job performance. Barriers exist, including the rapidly changing landscape of oncology, time constraints, lack of self-awareness, and generational differences. Through a career, mentoring needs will change, as will mentors. Nonetheless, mentorship over the long haul will likely result in your transition from mentee to mentor and hopefully maintain your passion for oncology while inspiring other young physicians.

References 1. Thomas-Maclean R, Hamoline R, Quinlan E, et al. Discussing mentorship: an ongoing study for the development of a mentorship program in Saskatchewan. Can Fam Physician. 2010;56:e263-e272. 2. MacLeod S. The challenge of providing mentorship in primary care. Postgrad Med J. 2007;83:317-319. 3. Patterson K, Grenny J, Maxfield D, et al. Change Anything: The New Science of Professional Success. New York: Grand Central Publishing; 2012. 4. Straus SE, Johnson MO, Marquez C, et al. Characteristics of successful and failed mentoring relationships: a qualitative study across two academic health centers. Acad Med. 2013;88:82-89.

792 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

5. Tietjen P, Griner PF. Mentoring of physicians at a community-based health system: preliminary findings. J Hosp Med. 2013;8:642-643. 6. Chopra V, Edelson DP, Saint S. A piece of my mind. Mentorship malpractice. JAMA. 2016;315:1453-1454. 7. Vaughn V, Saint S, Chopra V. Mentee missteps: tales from the academic trenches. JAMA. 2017;317:475-476. 8. Griner PF. Burnout in health care providers. Integr Med. 2013;12:22-24. 9. Allen TD, Poteet ML. Developing effective mentoring relationships: strategies from the mentor’s viewpoint. Career Dev Q. 1999;48:59-73.

SARCOMA

ANDREW E. ROSENBERG

Bone Sarcoma Pathology: Diagnostic Approach for Optimal Therapy Andrew E. Rosenberg, MD OVERVIEW The pathologic interpretation of malignant bone tumors is one of the more challenging areas in surgical pathology. This is based on the reality that primary bone sarcomas are uncommon, demonstrate significant morphologic heterogeneity, and have a broad spectrum of biology. Accordingly, it is difficult for pathologists to acquire the necessary experience to confidently and accurately diagnose bone sarcomas. The task is further complicated by the fact that it requires the integration of clinical and radiologic information into the diagnostic process. Lastly, molecular aberrations in sarcomas are being newly discovered and their identification is often critical to make specific diagnoses. The pathologist’s role in guiding optimal treatment in biopsy specimens is to make an accurate diagnosis and provide the grade and molecular aberrations when appropriate. The pathology report of resected tumors must confirm this information and assess the surgical resection margins and the percentage of necrosis if the sarcoma has been treated with neoadjuvant systemic therapy.

arcomas of the bone often pose diagnostic challenges to pathologists and clinicians. This reality is not surprising, as the tumors are uncommon and morphologically heterogeneous, possess a broad spectrum of biologic behavior, and require specific and complex therapeutic strategies to effect cure. Accurate diagnosis requires an integrated approach that assesses and correlates the clinical, radiologic, histologic, molecular, and prognostic characteristics of the malignancy. In most instances, this is best accomplished when members of a sarcoma multidisciplinary team collaborate to diagnose and stage the tumor and design and implement an optimal treatment plan.1 Also important in guiding effective treatment is the assessment of tumor necrosis in neoplasms treated with neoadjuvant chemotherapy. This discussion provides fundamental knowledge about bone sarcomas and information that should be included in a pathology report that forms the foundation of patient management.

EPIDEMIOLOGY

The overall frequency of bone tumors is unknown, as most benign tumors are asymptomatic and are only detected as incidental findings. Some benign tumors are quite common, for example, fibrous cortical defects develop in 50% of boys and 20% of girls older than age 2, and hemangiomas of the spine can be identified in at least 10% of the population, indicating that benign tumors of the bone affect many millions of individuals.2 On the basis of this information, it is estimated that benign bone tumors outnumber their

malignant counterparts by at least 10,000 to 1. Accordingly, bone sarcomas are uncommon: they account for 0.2% of all malignancies with approximately 3,020 bone sarcomas newly diagnosed annually in the United States, and they are aggressive, resulting in 1,460 deaths each year.3 The adjusted incidence rate for all bone and joint malignancies is 0.9 per 100,000 persons per year. Sarcomas of the bone develop in all age groups. In many instances, however, there is a relationship between the patient’s age and the specific location and type of tumor. As a group, bone sarcomas have a bimodal age distribution; the first peak occurs in patients age 10 to 20, and the second develops during the seventh decade of life. The risk of developing a bone sarcoma is equal in both of these age groups, but in absolute numbers, more bone sarcomas are diagnosed during the second decade of life. Statistically, the younger the patient, the more likely a bone tumor is to be benign, because benign tumors outnumber sarcomas, commonly occur in childhood, and diminish in frequency with age. Bone sarcomas affect males and females at a ratio of 1:0.7, and they develop in all parts of the skeleton. Most demonstrate a predilection for the pelvis, axial skeleton, and proximal long bones and rarely affect the small bones of the hands and feet.4

CLINICAL PRESENTATION

The clinical presentation of malignant bone tumors is highly variable and generally nonspecific. Symptoms are usually

From the Miller School of Medicine, University of Miami, Miami, FL. Disclosures of potential conflicts of interest provided by the author are available with the online article at asco.org/edbook. Corresponding author: Andrew E. Rosenberg, MD, Department of Pathology, University of Miami Hospital, 1400 NW 12th Ave., Suite 4061, Miami, FL 33136; email: arosenberg@ miami.edu. © 2017 American Society of Clinical Oncology

794 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

BONE SARCOMA PATHOLOGY

localized to the affected site and include pain, swelling, and mechanical disorders. The pain may be intermittent, constant, progressive, and radiating. Long-duration swelling is usually associated with benign lesions, whereas rapid swelling in conjunction with skin changes, such as red violaceous discoloration and the development of prominent blood vessels, is commonly a manifestation of malignancies. Mechanical dysfunction is usually in the form of restricted movement and may result from tumor bulk or synovitis caused by a periarticular mass. Systemic symptoms of fever, fatigue, and weight loss are usually associated with malignant bone neoplasms and are frequently indicative of advanced disease. A minority (approximately 10%) of malignant primary bone tumors are complicated by a pathologic fracture. The fracture may be the heralding event, and it results from an enlarging tumor that destroys the underlying bone. Minimal trauma eventually causes the bone to fail and break, producing sudden excruciating pain, swelling, and hemorrhage.

CLASSIFICATION

The classification of bone sarcomas is based on the normal cell or tissue type that they recapitulate (Sidebar). The vast majority differentiates along the cell lines or tissue types that compose the skeletal system; only a small number have consistent and distinctive clinicopathologic features but lack a normal tissue counterpart. Further subclassification of bone sarcomas is based on their specific histologic characteristics, their relationship to the underlying bone, the presence of pre-existing conditions, and their biologic potential (i.e., grade). The classification system most commonly used is that presented in the World Health Organization’s Classification of Tumours of Soft Tissue and Bone.5

GRADING AND PATHOLOGIC STAGING BONE SARCOMAS

The pathologist’s attempt to predict the biologic behavior of bone sarcomas is reflected in the histologic grade. Grading systems similar to the National Cancer Institute and Frente Nacional de Combate ao Câncer schemes devised for soft

KEY POINTS • Bone sarcomas are uncommon forms of neoplasms. • An experienced musculoskeletal sarcoma multidisciplinary team should perform the diagnosis and treatment of bone sarcomas. • Bone sarcomas are classified according to their normal tissue counterpart. • Diagnosing bone sarcomas includes the integration of clinical, radiologic, and pathologic information. • Pathology report should include the name and grade of the sarcoma, and for resected tumors, the margin status and assessment of neoadjuvant treatment effect, if used, must be identified.

SIDEBAR. Classification of Primary Bone Sarcomas

Chondrosarcoma

• Conventional • Dedifferentiated • Clear cell • Mesenchymal • Secondary

Osteosarcoma

• Conventional • Chondroblastic, fibroblastic, osteoblastic, mixed • Parosteal • Periosteal • Intramedullary well differentiated • Surface high grade • Small cell • Telangiectatic • Secondary

Fibrosarcoma Ewing sarcoma and other round cell sarcomas Chordoma Malignant vascular tumors

• Epithelioid hemangioendothelioma • Pseudomyogenic hemangioendothelioma • Angiosarcoma • Kaposi sarcoma

Other uncommon entities

• Adamantinoma • Liposarcoma • Leiomyosarcoma • Malignant peripheral nerve sheath tumor • Primary non-Hodgkin lymphoma • Synovial sarcoma

Undifferentiated high-grade pleomorphic sarcoma tissue sarcomas have not been developed and universally applied to bone sarcomas. There are, however, grading schemes that some investigators have proposed for specific types of sarcomas, especially chondrosarcoma.6 Regardless, all bone sarcomas—exclusive of Ewing sarcoma and other poorly differentiated round cell/spindle cell sarcomas, adamantinoma, and chordoma—are typically graded. The three-tiered grading system currently used is based on the assessment of standard morphologic criteria, including the degree of differentiation, cytologic atypia, mitotic activity, and necrosis. The goal of grading sarcoma is to distinguish sarcomas with a low probability of dissemination asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 795

ANDREW E. ROSENBERG

(low grade; < 10% chance of metastasis) from those that are aggressive and have a significant risk of systemic spread (high grade; > 10% chance of metastasis). Accordingly, in the three-tiered system, grade 1 sarcomas are low grade and are usually hypocellular to moderately cellular. The tumor cells demonstrate mild cytologic atypia, closely resemble their normal tissue counterparts, and have few if any mitoses and minimal necrosis. For treatment purposes, grade 2 and 3 sarcomas are considered high grade and are moderately to densely cellular. The cells are moderately to severely pleomorphic and hyperchromatic and mitotically active with atypical forms, and a grade 2 or 3 tumor contains areas of necrosis. Generally, the focus of treatment of low-grade sarcomas is local control, whereas systemic therapy combined with local control is used to attempt to cure patients with high-grade sarcomas. Staging bone sarcomas provides important prognostic information and offers guidelines for effective treatment. The two major staging systems used are those endorsed by the American Joint Commission on Cancer and the Musculoskeletal Tumor Society. The American Joint Commission on Cancer system incorporates tumor grade, size, location in the body, and status and location of metastases.7 In contrast, the Musculoskeletal Tumor Society staging scheme is more focused on surgical staging and integrates tumor grade, anatomic extent, and presence of metastases.

BONE TUMOR SPECIMENS

The ability to make a primary diagnosis, document recurrence or metastasis, and assess treatment effect is based on the assessment of bone tumor specimens. The different types of tissue specimens include fine-needle aspiration cytology, needle core biopsy, open curettage (which yields multiple, irregular fragments of tissue), and en bloc resection. In specific instances, frozen section analysis can be performed to facilitate diagnosis and assess margin status.

Fine-Needle Aspiration

Cytologic evaluation has been reliably and successfully used for many years in the investigation and diagnosis of metastases to the skeleton. Fine-needle aspiration diagnosis of primary bone tumors is challenging because of the morphologic heterogeneity of the tumors and their relative rarity. Studies have shown that the fine-needle aspiration diagnosis of primary bone tumors has an accuracy rate of 70% to 90% when the goal is distinguishing benign from malignant lesions. Accordingly, it is not a technique that is typically used to render the primary diagnosis and grade the tumor, except in the hands of the most experienced cytologists. Knowledge of the cytologic appearance of primary bone tumors is important, however, because they may be inadvertently aspirated during the work-up of suspected metastatic disease, from which they must be distinguished.

Needle Core Biopsy

Needle core biopsy is often performed with CT or ultrasound guidance; we recommend that a minimum of three 796 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

cores of tumor-bearing tissue be obtained for diagnosis. A frozen section can be performed on one core to confirm that diagnostic tissue is present, provide a provisional diagnosis, and facilitate triage of the remaining tissue, including generating touch preparation slides for fluorescent in situ hybridization analysis, if appropriate. A portion of the second core can be submitted for cytogenetic karyotype analysis when needed, and the remaining cores can be fixed in formalin and processed routinely for standard hematoxylin and eosin–stained slides. If the tissue requires decalcification, it should be done with solutions that preserve RNA and DNA such as EDTA.

Open Biopsy

Open biopsy specimens often provide abundant tissue for analysis. These specimens may undergo frozen section analysis (see below) to help provide the surgeon with information that guides therapy at the time of surgery. Definitive curettage specimens should be fixed, decalcified, and thoroughly sampled (minimum of 10 cassettes if enough tissue is present).

Resections

Most malignancies are widely resected en bloc with a rim of normal tissue. The specimen should be oriented, and the soft tissue and bone margins should be carefully assessed grossly. The margins should be inked and the specimen transected with a bone saw along the plane of the greatest dimension of the tumor and its relationship to the closest soft tissue and bone margins. If needed, fresh tumor can be frozen for both diagnostic purposes and tissue triage. Subsequently, in most instances, a longitudinal slab 0.5- to 1-cm thick can be cut from the center of the specimen through the greatest dimension of the tumor. The remaining two hemispheres of tissue can then be “breadloafed” at 0.5- to 1-cm intervals in the plane perpendicular to the cut surface of the slab. Sections demonstrating the proximity of the tumor to the closest soft tissue and bone margins should be submitted, and the tumor should be carefully dissected and sampled. This usually requires processing a minimum of one cassette per centimeter of tumor. The relationship between the tumor and the surrounding cancellous bone, cortex, articular surfaces, and neighboring soft tissues should be illustrated in some of these sections. Resected tumors that have been treated with preoperative chemotherapy (osteosarcoma, fibrosarcoma, Ewing sarcoma, and other poorly differentiated round cell/spindle cell sarcomas, dedifferentiated chondrosarcoma, and mesenchymal chondrosarcoma) require determination of the percentage of tumor necrosis. To accomplish this, the central slab of tissue can be imaged and the tumor mapped and blocked out in its entirety (Fig. 1). A section of tumor per centimeter (as determined by its greatest dimension) should be processed from each of the remaining two hemispheres of the specimen. During histologic review, the amount of tumor necrosis on each slide can

BONE SARCOMA PATHOLOGY

FIGURE 1. Diagram for Mapping the Slab for Pathologic Assessment of Tumor Necrosis

Histologic Distinction of Benign and Malignant Tumors

be estimated, and these scores can then be averaged to calculate the overall percentage of tumor necrosis (Fig. 2). The location of the areas of viable and necrotic tumor can then be located on the map of the slab section, if necessary.

FIGURE 2. Osteosarcoma Treated With Preoperative Chemotherapy Show Areas of Necrotic and Viable Tumor

Distinguishing benign from malignant bone tumors is not always easily accomplished through the assessment of conventional histologic features such as the degree of cellularity, mitotic activity, and necrosis. This is because benign tumors such as chondroblastoma, osteoblastoma, giant cell tumor, and solid aneurysmal bone cyst can be densely cellular, demonstrate many mitoses, and have large areas of necrosis, whereas variants of osteosarcoma, chondrosarcoma, and fibrosarcoma may be relatively hypocellular and have few mitoses and little or no necrosis. Significant pleomorphism and atypia is a telltale sign of malignancy when accompanied by concurrent cellularity and mitotic activity. The absence of these features, however, should be interpreted with caution, as degenerative nuclear atypia, similar to that seen in ancient schwannoma, may be present infrequently in a variety of benign neoplasms. A very important morphologic feature indicative of malignancy is an infiltrative growth pattern in which the tumor replaces the marrow elements, encases pre-existing bony trabeculae and percolates within haversian systems. This finding is strongly suspicious of malignancy, especially for bone- and cartilage-forming neoplasms. Hemangioma is the only benign tumor that routinely infiltrates the marrow cavity, although infiltration may also be present infrequently in desmoplastic fibroma. Other processes that can cause confusion with infiltration are fracture callus and a tangential plane of section through a well-delineated, but undulating, interface between tumor and surrounding bone. The converse is also important, in that most benign tumors have well-circ*mscribed margins, and it is uncommon for bone sarcomas to be well delineated along their entire margin. The pathologic interpretation of bone tumor specimens is best accomplished by surgical pathologists experienced in this subspecialty. Even in the hands of experienced muscu-

Frozen Section

Bone tumor specimens often undergo frozen section analysis. The tissue, including bone (except for pieces of cortex), can be frozen to construct a working diagnosis and allow for the appropriate triage of tissue. If the surgeon is going to perform a definitive procedure based on the frozen section diagnosis during the same operation, then all the tissue submitted for initial diagnosis should undergo frozen section analysis to avoid errors based on sampling. The pathologist should also understand the algorithm used by the surgeon to prevent patient mismanagement, and if there is uncertainty with the diagnosis, then it should be deferred until formalin-fixed tissue is available for histologic interpretation. asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 797

ANDREW E. ROSENBERG

loskeletal multidisciplinary teams, the accuracy of needle biopsy in diagnosing has been reported to be 80.8%, with a diagnostic error rate of 7.1% and nondiagnostic rates of 12.1%.8

CLINICAL REPORTING OF BONE TUMOR SPECIMENS

The surgical pathology report of biopsied and resected sarcomas should follow the guidelines proposed by the College of American Pathologists and include tumor type and grade, and if pretreated with cytotoxic therapy, the percentage of tumor necrosis should be indicated. The presence of precursor lesions or other conditions should be identified. The relationship of the tumor to important anatomic structures, such as large neurovascular bundles, articular surfaces, synovium, cruciate ligaments, etc., should be commented on.

The status of the closest soft tissue and bone margins and the distances of the tumor to these surfaces must be carefully identified, measured, assessed, and recorded. If special histochemical stains, immunohistochemistry, electron microscopy, karyotype, or molecular analyses have been performed, then the results should be integrated into the report.

CONCLUSION

The accurate diagnosis of bone sarcoma is critical to optimal patient care. This process requires the integration of clinical, radiologic, pathologic, and molecular information so that the clinical treatment team is using optimal therapy based on the precise diagnosis, grade, percentage of necrosis, and surgical margin status of the sarcoma.

References 1. Weber K, Damron TA, Frassica FJ, et al. Malignant bone tumors. Instr Course Lect. 2008;57:673-688. 2. Unni KK, Inwards CY. Dahlin’s Bone Tumors: General Aspects and Data on 10,165 Cases, 6th Ed. Philadelpha: Lippincott Williams & Wilkins Publishers; 2010. 3. Siegel R, Ma J, Zou Z, et al. Cancer statistics, 2014. CA Cancer J Clin. 2014;64:9-29. 4. Dorfman HD, Czerniak B. Bone cancers. Cancer. 1995;75: 203-210.

798 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

5. Fletcher CDM, Bridge JA, Hogendoorn CW, et al. WHO Classification of Tumours of Soft Tissue and Bone, 4th Ed., Vol. 5. Lyon: International Agency for Research on Cancer; 2013. 6. Inwards CY, Unni KK. Classification and grading of bone sarcomas. Hematol Oncol Clin North Am. 1995;9:545-569. 7. Amin MB, Edge S, Greene F, et al (eds.) AJCC Cancer Staging Manual, 8th Ed. Chicago: Springer; 2017. 8. Trieu J, Schlicht SM, Choong PF. Diagnosing musculoskeletal tumours: How accurate is CT-guided core needle biopsy? Eur J Surg Oncol. 2016;42:1049-1056.

FERTILITY, CARDIAC, AND ORTHOPEDIC CHALLENGES IN SARCOMA SURVIVORS

Fertility, Cardiac, and Orthopedic Challenges in Survivors of Adult and Childhood Sarcoma Emma R. Lipshultz, Ginger E. Holt, MD, Ranjith Ramasamy, MD, Raphael Yechieli, MD, and Steven E. Lipshultz, MD OVERVIEW The combination of cisplatin, doxorubicin, and methotrexate was established as the standard backbone of contemporary osteosarcoma therapy in 1986. Since then, however, further improving the survival of patients with osteosarcoma has been challenging—30% to 40% of patients with osteosarcoma still die of this disease. In addition, these patients often experience loss of fertility at a young age, short- and long-term treatment-related cardiotoxicity, and adverse orthopedic effects from surgical resection of the tumor or endoprosthetic reconstructions. Cancer treatment often markedly increases the risk of infertility later in life, causing many patients substantial distress and regret. Sperm banking and oocyte cryopreservation are standard of care and should be available to all at-risk patients. Newer techniques, such as autologous gonadal tissue transplant for prepubertal children, are being developed, and newer systemic agents have infertility risk profiles that remain undefined and warrant further study. Cost and access remain barriers to these options. The late effects of anthracycline-induced cardiotoxicity are also increasingly a problem for these patients. These effects are often progressive and can be disabling. Adding dexrazoxane to doxorubicin therapy significantly reduces the risk for most adverse cardiac outcomes without compromising the efficacy of induction chemotherapy. Limb salvage surgery remains the standard of care for treatment in the majority of patients with extremity sarcomas. Modular metal prostheses and allograft reconstructions comprised the majority of surgical procedures for limb salvage surgery. The most common mechanism of failure of these implants is infection and mechanical failure of the implant.

urrently, more than 640,000 adolescent and young adult (age 15 to 39) cancer survivors live in the United States. This number is expected to rise sharply during the next decade.1 Survival depends greatly on the type and stage of cancer, but with a combination of local and systemic treatment, cure rates range from 85% among patients with stage I disease to 10% to 20% for patients with stage IV disease.2 However, the cure has a cost: the same life-saving surgery, radiation, and anthracycline chemotherapy used to treat cancer often comes with the loss of fertility, early and late cardiotoxicity, and orthopedic problems. These outcomes include loss of spermatogenesis and premature ovarian failure; acute cardiomyopathy during chemotherapy and late cardiomyopathy in subsequent decades; infections and complications of limb-salvage surgery; and death.3,4

PRESERVING FERTILITY IN PATIENTS WITH SARCOMA

Oncofertility is a term coined in 2006 to describe the specific care patients with cancer require to preserve their present or future reproductive capacity.5 The field is at the

intersection of reproductive specialists and oncologists, and is designed to bring more and better reproductive options to cancer survivors. Infertility can be distressing for adolescents and young adults, particularly those who have not started their own families. The effects of cancer-related infertility are longstanding, with increased grief and decreased quality of life reported even 10 years after diagnosis.6,7 Up to 75% of nulliparous patients report wanting to have children.8 Soft tissue sarcoma comprises 7% of the total number of cancer cases, and disproportionately affects children: it is the third most common childhood cancer,9 accounting for 20% of cancers in children and 10% in adolescents and young adults.10 It can present in a wide variety of locations and histologic types, with rhabdomyosarcoma accounting for almost one-half of cases.9 As a result of the wide prevalence of soft tissue sarcomas among adolescents, and because treatment can typically result in infertility, efforts to preserve fertility before therapy can be beneficial in young patients with sarcoma. As tumors of muscle and bone, sarcomas can arise anywhere in the body, especially in the thigh, pelvis, and

From the Dana-Farber Cancer Institute, Boston, MA; Vanderbilt-Ingram Cancer Center, Nashville, TN; University of Miami Miller School of Medicine, Miami, FL; Wayne State University, Children’s Hospital of Michigan, Karmanos Cancer Institute, Detroit, MI. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Steven E. Lipshultz, MD, Department of Pediatrics, Wayne State University School of Medicine, Children’s Hospital of Michigan, 3901 Beaubien Blvd., Suite 1K40, Detroit, MI 48201; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 799

LIPSHULTZ ET AL

retroperitoneum. Therefore, reproductive organs can potentially be in the radiation field or within the scatter region for radiation therapy. Increasingly complex treatment techniques, such as intensity-modulated radiation therapy, may improve clinical outcomes, although a larger area can be exposed to radiation, albeit at lower doses.11 For males, even cumulative doses as low as 2 Gy to the gonads can affect fertility. For females, the damaging dose is age dependent; lower radiation doses can affect fertility more as age increases,12 probably because of the natural decrease in the number of follicles with aging. Thus, the radiation dose most likely to cause ovarian failure decreases from 15 Gy in girls to 6 Gy in women.13,14 Systemic multiagent therapy, which includes alkylating agents, for patients with high-grade sarcoma is individualized, yet a concern for later fertility issues. Systemic therapy is considered a principle backbone of therapy for certain sarcomas, including Ewing sarcoma and rhabdomyosarcoma. It is well known that alkylating agents as part of systematic therapy can decrease fertility in both male and female cancer survivors.15 In males, it inhibits spermatogenesis and can cause prolonged azoospermia. Although this effect is dose-dependent,16,17 and though efforts are always made to avoid a toxic dose, the effects in combination regimens are additive and not fully understood.18 Dacarbazine and taxanes usually cause only temporary sterility, but they can also have an additive effect when combined with alkylating agents.19 In females with sarcoma, abdominal or pelvic irradiation decreases ovarian reserves. Any radiation exposure to the uterus or ovaries increases the risk of infertility, and higher doses of radiation further increase this risk.15 Additionally, alkylating agents commonly used for sarcoma, such as ifosfamide, carry a high risk of amenorrhea and subsequent infertility.20 All patients with sarcoma should be informed of the increased risk of infertility from treatment. If they are interested, or even undecided, they should be offered a fertility consult and a referral to a specialist as soon as possible, ideally before treatment is initiated.21 Options for preserving fertility continue to evolve. For males, sperm banking before treatment is the best way to

Key Points • Preserving fertility, treating cardiotoxicity, and minimizing orthopedic interventions and complications are integral aspects of caring for patients with sarcoma. • Sperm banking and oocyte cryopreservation should be considered for all patients at risk of losing fertility. • Adding the cardioprotectant dexrazoxane to cancer treatment can reduce anthracycline-related cardiotoxicity without compromising efficacy. • Tumor resection and failed limb-preserving surgeries can result in serious infections and complications. • The increasing number of survivors of sarcoma makes studying the late effects of treatment more important. 800 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

preserve fertility. Before chemotherapy, viable sperm may be collected by masturbation, penile vibrostimulation, electroejacul*tion, or, in rare cases, testicular sperm extraction. sem*n samples must be analyzed to ensure the presence of sperm. If sperm are absent or in sexually immature patients, sperm can be extracted by testicular biopsy.22 Both testicular sperm and cryopreserved ejacul*ted sperm require assisted reproduction with intracytoplasmic sperm injection for conception. After chemotherapy, nonobstructive azoospermia can also be treated with testicular sperm extraction, but success rates are limited, reaching 20% after exposure to alkylating agents.23 Unfortunately, spermatogenesis is not always recovered, and a pregnancy can only be achieved through a sperm donor. In females, age is critically important because follicular reserve decreases with time and decreases further with cancer treatment.24 Given the potential impact of cancer treatment on female fertility, the risk of infertility and fertility preservation options should be discussed before cancer therapy is initiated. After cancer therapy is complete, fertility treatments may be less successful and many patients will often require donor eggs or surrogates.21 Oncofertility options for women most commonly include embryo and oocyte cryopreservation. Despite a high success rate and being a validated method for preserving fertility, embryo cryopreservation presents a unique problem: it requires sperm from the patient’s partner or a donor, which is an unrealistic option for minors.21 Oocyte cryopreservation is a practical alternative and should be recommended to all female patients at risk for infertility, with appropriate counseling. The process requires injecting follicle-stimulating hormone for egg retrieval, and it is only possible in post-pubertal females.25 Delay in starting cancer therapy is a strong concern because a complete ovulation-and-egg retrieval cycle can take up to 4 weeks. However, with options such as natural cycle stimulation, egg retrieval can be done in less than 2 weeks.26 Other options for preserving fertility in prepubertal patients remain experimental. These options include in vitro maturation of immature eggs, autologous transplantation of cryopreserved ovarian tissue, and cryopreservation of testicular tissue.21 Oncofertility is a developing field for which the future still holds several challenges, from educating providers to determining the effects of new therapeutic agents on fertility. Newer agents that have improved survival in patients with sarcoma include trabectedin, a new DNA-binding molecule,27 and several molecular-targeted agents including pazopanib and sunitinib, which are multikinase angiogenesis inhibitors,28,29 and crizotinib and imatinib, which inhibit tyrosine kinases.30,31 Animal studies show that targeted molecules are generally safer for fertility than conventional chemotherapy,32 but long-term studies in humans are still required. The biggest challenge of oncofertility is providing access to care to all patients. Costs can be prohibitive and are currently not covered by most insurance companies.33

FERTILITY, CARDIAC, AND ORTHOPEDIC CHALLENGES IN SARCOMA SURVIVORS

TABLE 1. Cardiotoxic Effects of Selected Cytotoxic Agents Treatment

Cardiotoxic Effect

Anthracyclines Daunorubicin, doxorubicin, epirubicin, idarubicin, mitoxantrone

Arrhythmias, pericarditis, myocarditis, HF, LV dysfunction

Liposomal anthracyclines Pegylated liposomal doxorubicin hydrochloride (DOXIL, CAELYX)

HF, LV dysfunction, arrhythmias

Antimetabolites Capecitabine, carmustine, clofarabine, cytarabine, 5-fluorouracil, methotrexate

Ischemia, chest pain, MI, HF, arrhythmias, pericardial effusions, pericarditis, hemodynamic abnormalities

Antimicrotubule agents Pacl*taxel, vinca alkaloids

Hypotension or hypertension, ischemia, angina, MI, bradycardia, arrhythmias, conduction abnormalities, HF

Alkylating agents Busulfan, chlormethine, cisplatin, cyclophosphamide, ifosfamide, mitomycin

Endomyocardial fibrosis, pericarditis, tamponade, ischemia, MI, hypertension, myocarditis, HF, arrhythmias

Small-molecule tyrosine kinase inhibitors Dasatinib, gefitinib, imatinib mesylate, lapatinib, erlotinib, sorafenib, sunitinib

HF, edema, pericardial effusion, pericarditis, hypertension, arrhythmias, prolonged QT interval, ischemia, chest pain

Monoclonal antibodies Alemtuzumab, bevacizumab, cetuximab, rituximab, trastuzumab

Hemodynamic abnormalities, LV dysfunction, HF, thromboembolism, angioedema, arrhythmias

Interleukins Denileukin, IL-2

Hypotension, capillary leak syndrome, arrhythmias, coronary artery thrombosis, ischemia, LV dysfunction

Miscellaneous agents All-retinoic acid, arsenic trioxide, asparaginase, etoposide, IFN-α, lenalidomide, 6-mercaptopurine, pentostatin, teniposide, thalidomide

Electrocardiographic changes, QT prolongation, torsades de pointes, other arrhythmias, ischemia, angina, MI, HF, edema, hypotension, bradycardia, thromboembolism, and retinoid acid syndrome that includes fever, hypotension, respiratory distress, weight gain, peripheral edema, pleural-pericardial effusions

Abbreviations: HF, heart failure; LV, left ventricular; MI, myocardial infarction. Reproduced from Amdani et al42 with permission from Elsevier.

CARDIO-ONCOLOGY IN SARCOMA SURVIVORSHIP

Table 1 summarizes cardiotoxic effects of select cytotoxic therapies. Doxorubicin is a major agent for treating osteosarcoma. Event-free survival is lower in regimens with lower cumulative doses or dose-intensity, but higher doses increase the risk of cardiotoxicity.34,35 The cumulative doxorubicin dose (up to 450 mg/m2) currently used in the United States to treat osteosarcoma is associated with acute cardiomyopathy during chemotherapy, late cardiomyopathy in subsequent decades, and death (Table 2). The hazard ratio of adverse cardiac outcomes in survivors who receive more than 250 mg/m2 of anthracycline is two to five times as high as it is in those receiving doses less than 250 mg/m2.36 After 300 to 450 mg/m2 of doxorubicin, the incidence of cardiomyopathy is readily apparent, with more than 25% of patients experiencing left ventricular systolic dysfunction beyond 15 years of follow-up.37 Many long-term survivors are now between age 40 and 50, and they remain at risk for cardiac deterioration for the rest of their lives. Trials using doxorubicin for osteosarcoma have reported a substantial incidence of acute cardiotoxicity. In one trial of 31 children and adults, cardiotoxicity required stopping doxorubicin administration in four patients.39 In another, six of 164 patients experienced severe cardiotoxicity; five patients experienced events within 12 weeks of completing therapy.40

In 120 children and adults treated with bolus doses of doxorubicin in childhood (87 for acute lymphoblastic leukemia and 33 for nonmetastatic osteogenic sarcoma), 12 had transient early heart failure during or within 1 year after completing doxorubicin treatment.41 Heart failure occurred later in 12 patients, seven of whom had also had early heart failure 3 to 16 years before. In three of the 12 patients with late heart failure, medical treatment failed; one underwent heart transplantation, one underwent heart-lung transplantation, and one died of documented ventricular fibrillation. Another five had initial episodes of heart failure at a mean of 10 years after completing doxorubicin treatment, including two women during the peripartum period and one during nonanthracycline chemotherapy for a relapse of cancer. When patients with clinical evidence of cardiotoxicity were excluded, the results were similar.41 Risk factors for cardiotoxicity with anthracycline therapy are described in Table 3. Multivariate analysis in this trial revealed that female sex and higher cumulative doxorubicin doses were associated with depressed left ventricular contractility (p < .001) and that these two variables interacted.41 Independent and significant associations were found between a higher rate of administration of doxorubicin and increased left ventricular afterload (p < .001), left ventricular dilation, and depressed left ventricular function; between a higher cumulative doxorubicin dose and depressed left ventricular function (p < .001); between younger age at diagnosis and reduced left ventricular wall asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 801

LIPSHULTZ ET AL

TABLE 2. Characteristics of Different Types of Anthracycline Cardiotoxicity Characteristic

Acute Cardiotoxicity

Early-Onset Progressive Cardiotoxicity

Late-Onset Progressive Cardiotoxicity

Onset

Within the first week of anthracycline treatment

< 1 year after completion of anthracycline treatment

≥ 1 year after completion of anthracycline treatment

Risk factor dependence

Unknown

Yesa

Clinical features in adults

Transient depression of myocardial contractility

Dilated cardiomyopathy

Clinical features in children

Transient depression of myocardial contractility

Restrictive cardiomyopathy and/or dilated cardiomyopathy

Course

Usually reversible after discontinuation of anthracycline

Can be progressive

See Table 3 for risk factors. Reproduced from Amdani et al42 with permission from Elsevier.

thickness and mass and increased afterload; and between a longer time since completing doxorubicin therapy and reduced left ventricular wall thickness and increased afterload (p < .001).41 Late cardiotoxic effects of doxorubicin are increasingly a problem for survivors of childhood cancer. This cardiotoxicity is often progressive and can be disabling. However, given the efficacy of doxorubicin in treating childhood cancers, including osteosarcoma, many treatment initiatives have focused on preventing doxorubicin-related cardiotoxicity. Dexrazoxane is a topoisomerase II inhibitor that protects against anthracycline-related cardiotoxicity, probably by scavenging free radicals and chelating heavy metals or by preventing the topoisomerase IIB–mediated DNA and mitochondrial damage induced by doxorubicin.43,44 Used initially for cardioprotection in clinical trials of women with breast cancer receiving doxorubicin, dexrazoxane decreased the expected cardiotoxicity.43 A recent meta-analysis of dexrazoxane use in children found that it substantially reduced the risk for most adverse cardiac outcomes.45,46 In a study of 101 children with newly diagnosed metastatic osteosarcoma treated with trastuzumab, a humanized monoclonal antibody targeting HER2, in combination with cytotoxic chemotherapy and dexrazoxane, no patient developed clinical evidence of congestive heart failure after an average of 41.6 months of follow-up time.47 Dexrazoxane protects against cardiotoxicity without adverse outcomes in a wide range of cancers.48 Its use has been endorsed by the American Heart Association and the American Academy of Pediatrics as a cardioprotectant in children and adolescents undergoing anthracyclinecontaining treatment protocols.48 Doxorubicin has been used as the standard of good clinical care for all DanaFarber Cancer Institute high-risk childhood acute lymphoblastic leukemia protocols involving anthracycline therapy since 2000 and on all Children’s Oncology Group protocols involving treatment with at least 150 mg/m2 doxorubicin or anthracycline administration at any dose with planned radiation treatment portals that may impact the heart since 2015.49 802 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

In a trial of children with osteosarcoma randomly assigned to receive doxorubicin with or without dexrazoxane, the dexrazoxane-treated children maintained higher mean left ventricular fractional shortening and were able to receive more doxorubicin.50 In another trial, dexrazoxane reduced acute cardiotoxicity in young patients with sarcoma, but sample size limited the assessment of oncologic efficacy.51 In a preliminary analysis of Children’s Oncology Group protocols with random dexrazoxane assignments, long-term survivors of childhood cancer treated with doxorubicin and dexrazoxane appeared to have more preserved systolic function and reduced myocardial wall stress compared with

TABLE 3. Risk Factors for Anthracycline-Induced Cardiotoxicity Risk Factors

Features

Total cumulative dose

Most important predictor of abnormal cardiac function

Age

For similar cumulative doses, younger age predisposes to greater cardiotoxicity (especially < 5 years)

Length of follow-up

Longer follow-up reveals higher prevalence of myocardial impairment

Sex

Females more vulnerable than males for similar doses

Concomitant mantle irradiation

Evidence of enhanced cardiotoxicity; not clear whether additive or synergistic

Others

Concomitant exposure to cyclophosphamide, bleomycin, vincristine, amsacrine, or mitoxantrone may predispose to cardiotoxicity; trisomy 21 and black race have been associated with a higher risk of early clinical cardiotoxicity

Rate of anthracycline administration

Higher rate was thought to predispose to greater toxicity, but current trials in children do not support this finding

Reproduced from Amdani et al42 with permission from Elsevier.

FERTILITY, CARDIAC, AND ORTHOPEDIC CHALLENGES IN SARCOMA SURVIVORS

survivors treated with doxorubicin alone.52 Schwartz et al showed that dexrazoxane did not interfere with the tumor cytotoxicity of preoperative induction chemotherapy in 242 children with leukemia enrolled in Children’s Oncology Group protocol P9754.53 Dexrazoxane was also not associated with acute cardiotoxicity in patients receiving either standard (450 mg/m2) or intensified (600 mg/m2) doses of doxorubicin.53 Thus, dexrazoxane does not compromise response to induction chemotherapy. In the study by Schwartz et al,53 dexrazoxane was safely administered. It did not impair tumor response or interfere with cancer treatment efficacy. It also did not significantly increase the risk of secondary malignancy, and it allowed the cumulative doxorubicin dose to be increased in standard responders to induction chemotherapy. As well, in a randomized study of dexrazoxane administration with more than 12 years of follow-up, overall mortality did not differ by dexrazoxane status in three childhood cancer trials (1,008 patients).54 These findings support the use of dexrazoxane in children and adolescents with osteosarcoma as it permits anthracycline dose-density increases without compromising overall long-term survival. Cardiotoxicity secondary to anthracycline chemotherapy can be a devastating late effect of osteosarcoma treatment. Not only may it cause death and increase health care costs, but cardiac death was the second most common cause of late mortality in childhood cancer survivors reported by the Children’s Cancer Survivor Study.55 Heart failure, myocardial infarction, pericardial disease, and valvar abnormalities were substantially more prevalent in these patients than they were in siblings of cancer survivors.36,56

LONG-TERM OUTCOMES OF PATIENTS WITH SARCOMA AFTER ORTHOPEDIC SURGERY

Limb-salvage surgery remains the standard of care for treating patients with sarcomas of the long bones and is successful in about 90% of cases.57 Innovations in implant design have increased the longevity of modular metal prostheses. Progress in allograft donation and processing has increased availability and survivability of allograft reconstructions. Despite advances in infection management and antibiotic development, the most common mechanism of failure in orthopedic interventions after treatment of sarcoma is infection. The second most common mechanism is failure of the construct from mechanical failure of the implant, whether from loosening of the implant away from the host or the fracture of the implant itself.57-63 Amputation remains an option for these patients, and advances in prosthetic limbs allow a more active lifestyle, making this option more acceptable to patients.64

Bone Sarcomas

In bone sarcomas, after the primary lesion is removed, bones can be treated with what are termed the Five As: allograft, arthrodesis, arthroplasty (implanting metal modular oncologic endoprostheses), autograft, and amputation. The most common methods are arthroplasty and allograft. The

method of reconstruction depends on the type of surgical resection (intercalary or intra-articular), the degree of residual bone loss, and the age of the patient. Adults more commonly receive modular endoprostheses, which allow immediate stability, immediate weight bearing, and do not rely on osseous integration as heavily as do allografts. The small bones and joints of younger and skeletally immature patients, as well as their potential growth, pose additional challenges.59,61 Allografts often allow unique surgical approaches that can spare growth plates and thus are more commonly used in children. Overall implant survival at 15 years for all endoprosthetic reconstructions is 80%.57-63 Success is generally better in the upper arm than in the lower arm or the leg. Most limb-salvage procedures involve the proximal femur, the knee, and the distal femur or proximal tibia. Survival for proximal femur replacement is 93% at 5 years and 85% at 15 years.62 Modular oncologic prostheses about the knee have about a 22% failure rate at 10 years. Infection is the most common mechanism of failure and has been up to 10% in large series.62 The mechanism is aseptic loosening in about 5% of cases and implant fracture or failure in about 2%.62 Functional outcome scores average 91%. Causes of prosthetic failure include soft-tissue failure, aseptic loosening, structural failure of the implant, infection, and local tumor recurrence.57-63 Overall allograft survival at 15 years is 70%.65 As with modular prostheses, most reconstructions occur about the knee. Allograft reconstructions about the knee have about a 32% risk of failure at 10 years.58,65 The primary mechanism of failure is infection. The 10-year risk of amputation is 11%. Functional outcome scores average 88%.58 The most common mechanisms of failure are infection and failure of the reconstruction secondary to septic loosening and failure of the implant from infection, wear, or mechanical stresses at the host-bone interface that precludes long-term healing and osseous integration. Infection is the most common method of failure for any long-bone reconstruction after sarcoma surgery.57-65 Patients with a bone sarcoma typically undergo resection and reconstruction in combination after chemotherapy and are accordingly at risk for infection while immunosuppressed. Large surgical wounds are at particular risk for wound-healing complications. Breaching the skin can increase the risk of wound breakdown and infection, both of which endanger the reconstruction. The large bone defect left after resecting implants or allografts often precludes limb-salvage surgery and thus results in amputation.58,59,64 All methods of reconstruction are subject to failure-ofconstruct. In the setting of allograft, autograft and arthrodesis, failure of the allograft to heal to the host bone will allow the hardware to fail over time. The resulting hardware failure, pain, and subsequent revision surgeries may lead to amputation. Arthroplasty, especially to place a modular oncologic prosthesis or mega prosthesis, is uniquely susceptible to aseptic loosening. The large volume of bone loss places substantial stress on the bond between the implant asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 803

LIPSHULTZ ET AL

and the host site. This bond is often fixed with cement, and rotation forces incurred at this site are particularly likely to loosen it. Implant fixation techniques that foster osseous integration may improve long-term outcomes, but the weight bearing surfaces of these implants are still subject to wear at the bearing surfaces, especially in the young population who have more cycles on an implant. Unique surgical procedures for children include vascularized bone grafting, rotationplasty, and growing prostheses.59,61 The unique nature and small number of these procedures limits the amount of data about them. However, outcomes, such as retaining the limb with the use of an allograft, arthroplasty, or vascularized autograft, are better for both physical functioning and emotional acceptance than they are for amputation, which includes ablative surgery and rotationplasty.59 Pelvic reconstruction after sarcoma surgery also has unique circ*mstances and complications.66 These reconstructions can include allografts, metal prostheses, or, alternatively, no reconstruction at all. Patient satisfaction varies, depending on volume of the pelvis removed and the age of patient at the time of resection. Patients who do not undergo reconstruction after pelvic resection (flail limb) have fewer complications and higher satisfaction scores.66

Soft Tissue Sarcomas

Most soft tissue sarcomas occur in adults, although the complications are quite similar in some soft tissue sarcomas in children, except for those related to growth. Soft tissue sarcomas are commonly treated with surgical resection and radiation therapy, which are responsible for complications. Radiation therapy involving the growth plate of a bone may halt growth in that bone. In addition, short- and long-term complications from treating soft tissue sarcomas can be related to surgery when combined with radiation therapy. Wound healing complications and fibrosis are the most common complications. Preoperative radiation therapy is delivered in a lower dose to a smaller field, but the short-term insult to the

skin interferes with early wound healing, which can lead to further stiffness, given the need for surgical debridement and wound care. Stiffness reduces the range of motion, limiting mobility and increasing pain. Radiation therapy after surgery provides a larger field and a higher dose, which can contribute to larger areas of stiffness and lymphedema. Radiation to large, deep, high-grade sarcomas can also contribute to radiation necrosis, with short-or longterm effects, including bone fractures requiring intramedullary fixation and the need to remove necrotic bone.64 Once the bone has become necrotic, attempts to support bone healing without resection are fraught with complications, including multiple surgical procedures, pain, and unsatisfactory results.67

CONCLUSION

Worldwide, more than 28 million people live with cancer. This number could triple by 2030. With the increasing number of patients and improvements in cancer management that continue to reduce cancer death rates, the number of survivors is projected to increase rapidly, especially among those afflicted during childhood. In children and adolescents, the survival rate has jumped from fewer than 50% in the mid-1970s to 80% today. The growing population of childhood survivors is notable for its vulnerability to adverse health outcomes, many of which may not become clinically apparent until years after therapy has been completed.38 Loss of fertility, cardiotoxicity, and orthopedic complications are three such adverse outcomes. For prepubertal patients, preserving and perhaps transplanting testicular and ovarian immature tissue should be discussed as experimental options. The data support the use of the cardioprotectant dexrazoxane for all children who require anthracycline therapy for treatment of osteosarcoma to mitigate or prevent the development of cardiotoxicity, and developments in limb-salvage surgery should improve the orthopedic outcomes in these patients.

References 1. Jeys LM, Kulkarni A, Grimer RJ, et al. Endoprosthetic reconstruction for the treatment of musculoskeletal tumors of the appendicular skeleton and pelvis. J Bone Joint Surg Am. 2008;90:1265-1271. 2. Albergo JI, Gaston CL, Aponte-Tinao LA, et al. Proximal tibia reconstruction after bone tumor resection: are survivorship and outcomes of endoprosthetic replacement and osteoarticular allograft similar? Clin Orthop Relat Res. 2017;475:676-682. 3. Benedetti MG, Okita Y, Recubini E, et al. How much clinical and functional impairment do children treated with knee rotationplasty experience in adulthood? Clin Orthop Relat Res. 2016;474: 995-1004. 4. Bernthal NM, Greenberg M, Heberer K, et al. What are the functional outcomes of endoprosthestic reconstructions after tumor resection? Clin Orthop Relat Res. 2015;473:812-819.

804 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

5. Woodruff TK. The emergence of a new interdiscipline: oncofertility. Cancer Treat Res. 2007;138:3-11. 6. Loscalzo MJ, Clark KL. The psychosocial context of cancer-related infertility. Cancer Treat Res. 2007;138:180-190. 7. Duffy CM, Allen SM, Clark MA. Discussions regarding reproductive health for young women with breast cancer undergoing chemotherapy. J Clin Oncol. 2005;23:766-773. 8. Schover LR, Rybicki LA, Martin BA, et al. Having children after cancer. A pilot survey of survivors’ attitudes and experiences. Cancer. 1999;86:697-709. 9. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J Clin. 2016;66:7-30.

FERTILITY, CARDIAC, AND ORTHOPEDIC CHALLENGES IN SARCOMA SURVIVORS

10. Ballinger ML, Goode DL, Ray-Coquard I, et al; International Sarcoma Kindred Study. Monogenic and polygenic determinants of sarcoma risk: an international genetic study. Lancet Oncol. 2016;17:1261-1271. 11. Folkert MR, Singer S, Brennan MF, et al. Comparison of local recurrence with conventional and intensity-modulated radiation therapy for primary soft-tissue sarcomas of the extremity. J Clin Oncol. 2014;32:3236-3241. 12. Wallace WH, Thomson AB, Saran F, et al. Predicting age of ovarian failure after radiation to a field that includes the ovaries. Int J Radiat Oncol Biol Phys. 2005;62:738-744. 13. Levine JM, Kelvin JF, Quinn GP, et al. Infertility in reproductive-age female cancer survivors. Cancer. 2015;121:1532-1539. 14. Green DM, Sklar CA, Boice JD Jr, et al. Ovarian failure and reproductive outcomes after childhood cancer treatment: results from the Childhood Cancer Survivor Study. J Clin Oncol. 2009;27:2374-2381. 15. Antal Z, Sklar CA. Gonadal function and fertility among survivors of childhood cancer. Endocrinol Metab Clin North Am. 2015;44:739-749. 16. Servitzoglou M, De Vathaire F, Oberlin O, et al. Dose-effect relationship of alkylating agents on testicular function in male survivors of childhood lymphoma. Pediatr Hematol Oncol. 2015;32:613-623. 17. Green DM, Liu W, Kutteh WH, et al. Cumulative alkylating agent exposure and sem*n parameters in adult survivors of childhood cancer: a report from the St Jude Lifetime Cohort Study. Lancet Oncol. 2014;15:1215-1223. 18. Meistrich ML. Effects of chemotherapy and radiotherapy on spermatogenesis in humans. Fertil Steril. 2013;100:1180-1186.

of a phase III randomized multicenter clinical trial. J Clin Oncol. 2016;34:786-793. 28. Kasper B, Sleijfer S, Litière S, et al. Long-term responders and survivors on pazopanib for advanced soft tissue sarcomas: subanalysis of two European Organisation for Research and Treatment of Cancer (EORTC) clinical trials 62043 and 62072. Ann Oncol. 2014;25:719-724. 29. Stacchiotti S, Negri T, Zaffaroni N, et al. Sunitinib in advanced alveolar soft part sarcoma: evidence of a direct antitumor effect. Ann Oncol. 2011;22:1682-1690. 30. Butrynski JE, D’Adamo DR, Hornick JL, et al. Crizotinib in ALK-rearranged inflammatory myofibroblastic tumor. N Engl J Med. 2010;363:17271733. 31. Cassier PA, Gelderblom H, Stacchiotti S, et al. Efficacy of imatinib mesylate for the treatment of locally advanced and/or metastatic tenosynovial giant cell tumor/pigmented villonodular synovitis. Cancer. 2012;118:1649-1655. 32. Yildiz C, Kacan T, Akkar OB, et al. Effects of pazopanib, sunitinib, and sorafenib, anti-VEGF agents, on the growth of experimental endometriosis in rats. Reprod Sci. 2015;22:1445-1451. 33. Campbell JE, Assanasen C, Robinson RD, et al. Fertility preservation counseling for pediatric and adolescent cancer patients. J Adolesc Young Adult Oncol. 2016;5:58-63. 34. Smith MA, Ungerleider RS, Horowitz ME, et al. Influence of doxorubicin dose intensity on response and outcome for patients with osteogenic sarcoma and Ewing’s sarcoma. J Natl Cancer Inst. 1991;83:1460-1470.

19. Howell SJ, Shalet SM. Spermatogenesis after cancer treatment: damage and recovery. J Natl Cancer Inst Monogr. 2005;(34):12-17.

35. Bacci G, Picci P, Ferrari S, et al. Influence of adriamycin dose in the outcome of patients with osteosarcoma treated with multidrug neoadjuvant chemotherapy: results of two sequential studies. J Chemother. 1993;5:237-246.

20. Lee SJ, Schover LR, Partridge AH, et al; American Society of Clinical Oncology. American Society of Clinical Oncology recommendations on fertility preservation in cancer patients. J Clin Oncol. 2006;24:29172931.

36. Mulrooney DA, Yeazel MW, Kawashima T, et al. Cardiac outcomes in a cohort of adult survivors of childhood and adolescent cancer: retrospective analysis of the Childhood Cancer Survivor Study cohort. BMJ. 2009;339:b4606.

21. Loren AW, Mangu PB, Beck LN, et al; American Society of Clinical Oncology. Fertility preservation for patients with cancer: American Society of Clinical Oncology clinical practice guideline update. J Clin Oncol. 2013;31:2500-2510.

37. Brown TR, Vijarnsorn C, Potts J, et al. Anthracycline induced cardiac toxicity in pediatric Ewing sarcoma: a longitudinal study. Pediatr Blood Cancer. 2013;60:842-848.

22. Tournaye H, Dohle GR, Barratt CL. Fertility preservation in men with cancer. Lancet. 2014;384:1295-1301. 23. Hsiao W, Stahl PJ, Osterberg EC, et al. Successful treatment of postchemotherapy azoospermia with microsurgical testicular sperm extraction: the Weill Cornell experience. J Clin Oncol. 2011;29:16071611.

38. Lipshultz SE, Adams MJ. Cardiotoxicity after childhood cancer: beginning with the end in mind. J Clin Oncol. 2010;28:1276-1281. 39. Postma A, Bink-Boelkens MTE, Beaufort-Krol GCM, et al. Late cardiotoxicity after treatment for a malignant bone tumor. Med Pediatr Oncol. 1996;26:230-237.

24. Petrek JA, Naughton MJ, Case LD, et al. Incidence, time course, and determinants of menstrual bleeding after breast cancer treatment: a prospective study. J Clin Oncol. 2006;24:1045-1051.

40. Bacci G, Picci P, Ferrari S, et al. Primary chemotherapy and delayed surgery for nonmetastatic osteosarcoma of the extremities. Results in 164 patients preoperatively treated with high doses of methotrexate followed by cisplatin and doxorubicin. Cancer. 1993;72: 3227-3238.

25. Practice Committees of American Society for Reproductive Medicine; Society for Assisted Reproductive Technology. Mature oocyte cryopreservation: a guideline. Fertil Steril. 2013;99:37-43.

41. Lipshultz SE, Lipsitz SR, Mone SM, et al. Female sex and higher drug dose as risk factors for late cardiotoxic effects of doxorubicin therapy for childhood cancer. N Engl J Med. 1995;332:1738-1743.

26. Goodman LR, Balthazar U, Kim J, et al. Trends of socioeconomic disparities in referral patterns for fertility preservation consultation. Hum Reprod. 2012;27:2076-2081.

42. Amdani SM, Bansal N, Franco VI, et al. Clinical Cardio-Oncology: Cardiovascular effects of anthracycline chemotherapy and radiation therapy in children with cancer. Clinical Cardio-oncology; 2016.

27. Demetri GD, von Mehren M, Jones RL, et al. Efficacy and safety of trabectedin or dacarbazine for metastatic liposarcoma or leiomyosarcoma after failure of conventional chemotherapy: results

43. Green MD. Rationale and strategy for prevention of anthracycline cardiotoxicity with the bisdioxopiperazine, ICRF-187. Pathol Biol (Paris). 1987;35:49-53.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 805

LIPSHULTZ ET AL

44. Vavrova A, Jansova H, Mackova E, et al. Catalytic inhibitors of topoisomerase II differently modulate the toxicity of anthracyclines in cardiac and cancer cells. PLoS One. 2013;8:e76676.

54. Chow EJ, Asselin BL, Schwartz CL, et al. Late mortality after dexrazoxane treatment: a report from the Children’s Oncology Group. J Clin Oncol. 2015;33:2639-2645.

45. Shaikh F, Dupuis LL, Alexander S, et al. Cardioprotection and second malignant neoplasms associated with dexrazoxane in children receiving anthracycline chemotherapy: a systematic review and metaanalysis. J Natl Cancer Inst. 2015;108:pii:djv357.

55. Armstrong GT, Liu Q, Yasui Y, et al. Late mortality among 5-year survivors of childhood cancer: a summary from the Childhood Cancer Survivor Study. J Clin Oncol. 2009;27:2328-2338.

46. Kopp LM, Bernstein M, Schwartz CL, et al. Dexrazoxane in treating pediatric osteosarcoma. J Clin Oncol. 2012;30 (suppl; abstr 9503).

56. Nagarajan R, Kamruzzaman A, Ness KK, et al. Twenty years of follow-up of survivors of childhood osteosarcoma: a report from the Childhood Cancer Survivor Study. Cancer. 2011;117:625-634.

47. Ebb D, Meyers P, Grier H, et al. Phase II trial of trastuzumab in combination with cytotoxic chemotherapy for treatment of metastatic osteosarcoma with human epidermal growth factor receptor 2 overexpression: a report from the children’s oncology group. J Clin Oncol. 2012;30:2545-2551.

57. Gilg MM, Gaston CL, Parry MC, et al. What is the morbidity of a noninvasive growing prosthesis? Bone Joint J. 2016;98-B:1697-1703.

48. Lipshultz SE, Adams MJ, Colan SD, et al; American Heart Association Congenital Heart Defects Committee of the Council on Cardiovascular Disease in the Young, Council on Basic Cardiovascular Sciences, Council on Cardiovascular and Stroke Nursing, Council on Cardio vascular Radiology. Long-term cardiovascular toxicity in children, adolescents, and young adults who receive cancer therapy: patho physiology, course, monitoring, management, prevention, and research directions: a scientific statement from the American Heart Association. Circulation. 2013c;128:1927-1995.

59. Pala E, Trovarelli G, Calabrò T, et al. Survival of modern knee tumor megaprostheses: failures, functional results, and a comparative statistical analysis. Clin Orthop Relat Res. 2015;473:891-899.

49. Lipshultz SE, Franco VI, Sallan SE, et al. Dexrazoxane for reducing anthracycline-related cardiotoxicity in children with cancer: An update of the evidence. Prog Pediatr Cardiol. 2014;36:39-49. 50. de Matos Neto RP, Petrilli AS, Silva CM, et al. Left ventricular systolic function assessed by echocardiography in children and adolescents with osteosarcoma treated with doxorubicin alone or in combination with dexrazoxane. Arq Bras Cardiol. 2006;87:763-771. 51. Wexler LH, Andrich MP, Venzon D, et al. Randomized trial of the cardioprotective agent ICRF-187 in pediatric sarcoma patients treated with doxorubicin. J Clin Oncol. 1996;14:362-372. 52. Chow EJ, Doody DR, Armenian SH, et al. Effect of dexrazoxane on heart function among long-term survivors of childhood leukemia and lymphoma: a report from the Children’s Oncology Group (COG). Paper presented at: 58th American Society of Hematology Annual Meeting and Exposition; December 2016; San Diego, CA. 53. Schwartz CL, Wexler LH, Krailo MD, et al. Intensified chemotherapy with dexrazoxane cardioprotection in newly diagnosed non-metastatic osteosarcoma: a report from the Children’s Oncology Group. Pediatr Blood Cancer. 2016;63:54-61.

806 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

58. Henderson ER, Groundland JS, Pala E, et al. Failure mode classification for tumor endoprostheses: retrospective review of five institutions and a literature review. J Bone Joint Surg Am. 2011;93:418-429.

60. Rougraff BT, Simon MA, Kneisl JS, et al. Limb salvage compared with amputation for osteosarcoma of the distal end of the femur. A longterm oncological, functional, and quality-of-life study. J Bone Joint Surg Am. 1994;76:649-656. 61. Aponte-Tinao LA, Ayerza MA, Muscolo DL, et al. What are the risk factors and management options for infection after reconstruction with massive bone allografts? Clin Orthop Relat Res. 2016;474:669673. 62. Dramis A, Grimer RJ, Malizos K, et al. Non-metastatic pelvic ewing’s sarcoma: oncologic outcomes and evaluation of prognostic factors. Acta Orthop Belg. 2016;82:216-221. 63. Riad S, Biau D, Holt GE, et al. The clinical and functional outcome for patients with radiation-induced soft tissue sarcoma. Cancer. 2012;118:2682-2692. 64. Sternheim A, Saidi K, Lochab J, et al. Internal fixation of radiationinduced pathological fractures of the femur has a high rate of failure. Bone Joint J. 2013;95-B:1144-1148. 65. Miller KD, Siegel RL, Lin CC, et al. Cancer treatment and survivorship statistics, 2016. CA Cancer J Clin. 2016;66:271-289. 66. Skubitz KM, D’Adamo DR. Sarcoma. Mayo Clin Proc. 2007;82:14091432. 67. Wasilewski-Masker K, Seidel KD, Leisenring W, et al. Male infertility in long-term survivors of pediatric cancer: a report from the childhood cancer survivor study. J Cancer Surviv. 2014;8:437-447.

EARLY DRUG DEVELOPMENT FOR PATIENTS WITH SARCOMA

The Current Landscape of Early Drug Development for Patients With Sarcoma Breelyn A. Wilky, MD, Robin L. Jones, BSc, MB, MRCP, MD, and Vicki L. Keedy, MD, MSCI OVERVIEW Until recently, advancements in the treatment of patients with adult soft tissue sarcomas have been relatively slow. This is, in part, due to their heterogeneity and rarity. A better understanding of the biology and differences among the various histologies has led to substantial growth in novel strategies. In addition to novel cytotoxic chemotherapies, agents targeting platelet-derived growth factor receptor-α (PDGFRα), mTOR, and angiogenesis are areas of active investigation. Additionally, with the success of checkpoint inhibitors in other malignancies and early encouraging results of checkpoint inhibitors in some sarcoma subtypes, this approach is being widely investigated in various sarcomas. As we increasingly recognize and treat each sarcoma histology as a separate disease, it is important to spread awareness of the exciting clinical trials available to our patients with these rare malignancies.

dult sarcomas are a complex and heterogeneous group of neoplasms. This complexity and their rarity hinder drug development, making advancements for patients with sarcoma frustratingly slow. Not only do these malignancies arise from distinct mesenchymal tissues such as adipocytic and smooth muscle, but within each subset exist histologies that behave very differently. Historically, sarcoma clinical trials have included all adult soft histologies, making it nearly impossible to see the efficacy of a particular treatment by diluting any potential effect. Fortunately, there has been major progress in the last few years. By recognizing sarcomas as the separate entities that they are and narrowing the types of sarcomas studied, two drugs have been approved by the U.S. Food and Drug Administration in the last year. Despite the fact that sarcomas account for about 1% of all adult cancers, patients with sarcomas have accounted for a higher proportion of patients entered into phase I clinical trials. Classically, phase I trials have been dose-finding and toxicity-defining studies open to patients with all cancer types. However, recently, there has been greater emphasis on the initiation of phase I trials with an underlying biologic rationale that are limited to specific tumor types. The sarcoma community has made major advancements in the understanding of the biology and drivers of several sarcomas. Notable examples are the approval of the tyrosine kinase inhibitor, imatinib, in gastrointestinal stromal tumors (GISTs) and dermatofibrosarcoma protuberans, and denosumab in giant-cell tumor of bone. These success stories have led to an increased interest in drug development

in individual sarcomas, with many trials now enrolling specific sarcoma subtypes (Table 1).

EARLY-PHASE CHEMOTHERAPY AND TARGETED THERAPY TRIALS FOR PATIENTS WITH ADVANCED SARCOMA

Gastrointestinal Stromal Tumors

There are now three approved agents for patients with metastatic GISTs: imatinib, sunitinib, and regorafenib. However, there remains an unmet medical need for patients with GISTs whose disease is resistant to these three drugs and for patients with tumors harboring mutations resistant to the approved drugs. One of the mutations known to be resistant is the platelet-derived growth factor receptor-α (PDGFRα) D842V mutation. Although this is a rare molecular subtype of a rare disease, the treatment of patients is challenging. There are currently two drugs in development specifically for patients with tumors harboring this mutation. Crenolanib is a type I, small-molecule inhibitor of FLT3 and PDGFGRα (including the D842V mutation). In a phase I/II trial, crenolanib demonstrated activity in PDGFRα D842V mutant GISTs, with three out of 16 partial responses and three out of 16 patients achieving stable disease. Furthermore, seven patients continued receiving crenolanib for over 6 months, and one patient each for 1 and 2 years, respectively. In addition, this agent was well tolerated. Consequently, a randomized, placebo-controlled trial of crenolanib in patients with GIST with tumors harboring the D842V mutation is in development (NCT02847429).

From the Sylvester Comprehensive Cancer Center, Miami, FL; Royal Marsden Hospital, The Institute of Cancer Research, London, United Kingdom; Vanderbilt University Medical Center, Nashville, TN. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Vicki L. Keedy, MD, MSCI, Vanderbilt University Medical Center, 2220 Pierce Ave., 777 Preston Research Building, Nashville, TN 37232; email: vicki.keedy@ vanderbilt.edu. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 807

WILKY, JONES, AND KEEDY

TABLE 1. Select Targeted and Immunotherapy Trials in Soft Tissue Sarcomas Trial Number

Histology

Drug

Target

NCT00942877

ASPS

Cediranib

VEGFR1-3

NCT01755195

STS

Cabozantinib

VEGFR2/MET

NCT01879085

STS

Vorinostat/gemcitabine/doectaxel

HDAC/cyctoxic

NCT02584647

MPSNT/STS

Pexidartinib/sirolimus

CSF-1R and KIT and Flt3/mTOR

NCT02846987

DD-LPS

Abemaciclib

CDK4/6

NCT03009201

STS

Ribociclib/doxorubicin

CDK4/6 and cytotoxic

NCT02601950

Synovial/INI1− tumors

Tazemetostat

EZH2

NCT02048371

LPS/bone

Regorafenib

VEGFR/PDGFR

NCT01391962

ASPS

Sunitinib vs. cediranib

VEGFR1-3

NCT02609984

Synovial/RC-LPS

CMB305/atezolizumab

NY-ESO-1 and PD-L1

NCT02979899

Angiosarcoma

TRC105/pazopanib

Endoglin/TRC105

NCT00902044

HER2+ sarcoma

HER2 CAR T cells

HER2

NCT02636725

ASPS/STS

Axitinib/pembrolizumab

VEGFR1-3 and PD-1

NCT01803152

STS

Dendritic cell vaccine with or without gemcitabine

T-cell proliferation

NCT02180698

STS

GLA-SE/radiation

TLR4

NCT02888665

STS

Doxorubicin/pembrolizumab

Cytotoxic and PD-1

NCT02815995

Multiple cohorts

Durvalumab/tremelimumab

PD-L1 and CTLA-4

Abbreviations: ASPS, alveolar soft part sarcoma; CAR, chimeric antigen receptor; DD-LPS, dedifferentiated liposarcoma; EZH2, enhancer of zeste hom*olog 2; GLA-SE, glucopyranosyl lipid adjuvant-stable emulsion; HDAC, histone deacetylases; LPS, liposarcoma; MPNST; malignant peripheral nerve sheath tumor; PDGFR, platelet-derived growth factor receptor; RC-LPS, round cell liposarcoma; STS, soft tissue sarcoma; TLR4, toll-like receptor 4.

Blu-285 is an oral mutation-specific inhibitor of PDGFRα D842V and KIT D816V and is currently in phase I development (NCT02508532). Although still in dose escalation, this drug has shown antitumor activity, with reduction in tumor size in 11 out of 12 patients with D842V tumors. The drug is well tolerated. Furthermore, updated results will be presented at the 2017 ASCO Annual Meeting. DCC-2618 is a pan-KIT and PDFRα inhibitor in phase I development that has also reported encouraging clinical

KEY POINTS • Histology-based preclinical and clinical research has led to recent advancements in the treatment of patients with sarcoma. • Early-phase clinical trials provide important treatment options for patients with certain sarcoma subtypes. • Important molecularly defined subsets of gastrointestinal stromal tumors (GISTs) are primarily resistant or develop rapid secondary resistance to imatinib, which has led to several novel tyrosine kinase inhibitor trials for patients with GISTs. • Novel approaches of targeting the immune system in sarcomas include targeting NY-ESO–expressing tumors and adoptive chimeric antigen receptor T-cell strategies. • Single-agent checkpoint inhibitors have shown modest efficacy in only select sarcoma subtypes; thus, much of the focus is now on combined strategies such as radiation or tyrosine kinase inhibitors in combination with immunotherapies. 808 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

efficacy in pretreated metastatic GISTs (NCT02571036). An initial presentation of 24 enrolled patients reported seven metabolic responses by PET in seven KIT mutant GISTs. The most common treatment-emergent adverse events have included fatigue, lipase elevation, dyspnea, anemia, and decreased appetite.

Angiosarcoma

Angiosarcomas are aggressive tumors of endothelial origin associated with a very poor outcome. Endoglin is a protein that is overexpressed on endothelial cells and is essential for angiogenesis. Endoglin is upregulated, following VEGF inhibition, and enables continued angiogenesis despite this inhibition. Therefore, by targeting endoglin (which is upregulated following VEGF inhibition) and because endoglin is highly expressed in angiosarcoma, there is a clear rationale for combing anti-endoglin therapy with pazopanib. TRC105 is an anti-endoglin antibody that has shown promise in a phase IB/II trial. Of five originally enrolled patients with angiosarcoma, the progression-free survival was equal to or greater than 16.6 months.1 An additional nine patients with angiosarcoma were treated with a combination of tRC105 and pazopanib with a median progression-free survival of 5.59 months, and notably, three of these patients had progressed on prior pazopanib. Based on these data, a randomized phase III trial of pazopanib with or without TRC105 will be performed in patients with metastatic angiosarcoma. The Italian Sarcoma Group is conducting a trial of trabectedin in combination with the PARP-inhibitor olaparib (NCT02398058).

EARLY DRUG DEVELOPMENT FOR PATIENTS WITH SARCOMA

Anthracyclines

GPX-150 is a doxorubicin analog that has shown promise in early-stage clinical trials in patients with metastatic sarcoma.2 This compound has been modified in two locations, with the aim of reducing the cardiotoxicity of doxorubicin. Subsequently, an open-label phase II trial has been opened specifically for patients with metastatic soft tissue sarcoma (STS). The maximum tolerated dose in the phase I trial was 265 mg/m2; in the phase II trial, patients with sarcoma were treated at this dose every 21 days to a maximum of 16 cycles.

PDGFRα

There has been substantial evidence of the role of PDGFRα in several sarcomas. Recently, the PDGFRα antibody olaratumab showed significant activity when given in combination with doxorubicin in a randomized phase II trial, leading to U.S. Food and Drug Administration approval and ongoing interest in this compound.3 There are a number of ongoing early-phase trials of olaratumab, including a phase I/II trial of gemcitabine and docetaxel with or without olaratumab (NCT02659020), with overall survival as the primary endpoint of the phase II component, and a phase I trial in combination with a PD-1 inhibitor.

mTOR

mTOR inhibitors have been studied in several sarcoma studies with relatively limited success, with the exception of perivascular epithelioid cell tumors, which are rare malignancies characterized by activation of the mTOR pathway. A number of previous studies have reported the activity of mTOR inhibitors in this disease. ABI-009 is a nanoparticle, albumin-bound version of the mTOR inhibitor rapamycin. A phase II registration trial of ABI-009 has been commenced in patients with advanced perivascular epithelioid cell tumors (NCT02494570). TAK-228 is a TORC1/2 inhibitor. There is an ongoing phase II trial of this agent in patients with complex genomic sarcomas exhibiting PI3 kinase pathway dysregulation (NCT02987959). This agent is administered orally at a dose of 3 mg.

EARLY-PHASE IMMUNOTHERAPY TRIALS FOR PATIENTS WITH ADVANCED SARCOMA

With promising results of immunotherapy in other cancer types, there are a number of ongoing trials in sarcomas investigating immunotherapy, including checkpoint inhibitors as well as adoptive T-cell therapy. With modest results reported for single-agent pembrolizumab in selective bone and STS subtypes in the phase II SARC028 study, novel trials are focusing more on combination approaches to target potential resistance mechanisms to checkpoint blockade.4

Combination Therapies

Evidence in other tumors has supported that combining checkpoint inhibitors with chemotherapy and radiation may improve responses which is believed to be related to the

increased release of tumor neoantigens after necrosis from chemotherapy or radiation. Combination studies with chemotherapy include a phase II study of doxorubicin plus pembrolizumab for advanced STS (NCT0288665) and a phase II study of gemcitabine-based regimens or pegylated liposomal doxorubicin combined with pembrolizumab for solid tumors (NCT02331251). An additional study combining pembrolizumab with radiation for upfront treatment of patients with extremity sarcomas is planned through the Sarcoma Alliance for Research through Collaboration consortium. Additionally, dual-checkpoint inhibition with drugs targeting both the PD-1/PD-L1 axis as well as CTLA-4 have shown superior activity in melanoma and non–small cell lung cancer compared with monotherapy. A phase II clinical trial of 80 patients with metastatic bone and soft tissue sarcomas randomly assigned to nivolumab versus nivolumab plus ipilimumab has completed accrual, with preliminary results expected at the 2017 ASCO Annual Meeting. This combination is also being investigated for pediatric solid tumors, including sarcoma, through the Children’s Oncology Group (NCT02304458). A large, multiarm phase II study combining the CTLA-4 inhibitor tremelimumab plus the PD-L1 inhibitor durvalumab is also ongoing for patients with bone and soft tissue sarcomas (NCT02815995). Tyrosine kinase inhibitors, including imatinib, pexidartinib, pazopanib, and axitinib, not only serve to disrupt cellular pathways critical for sarcomas, but have also been shown to impact the immune microenvironment within tumors. Preclinical studies of imatinib in GIST mouse models demonstrated an increased ratio of CD8+ cytotoxic T cells to suppressive T-regulatory cells through inhibition of IDO1, as well as decreased PD-L1 expression on GIST tumor cells.5,6 Pexidartinib blocks CSF1 signaling, leading to depletion of immunosuppressive macrophages and suppression of tumor growth in malignant peripheral nerve sheath tumor mouse models.7 Pazopanib and axitinib suppress VEGF signaling, which is well known to promote accumulation of suppressive myeloid subtypes as well as inhibit T-cell migration and activation within tumors.8 In light of these findings, several other clinical trials are evaluating tyrosine kinase inhibitors in combination with checkpoint inhibitors in sarcomas. A phase I/II clinical trial is now ongoing for patients with malignant peripheral nerve sheath tumor and advanced STS with pexidartinib and sirolimus (NCT02584647), as well as a study of pexidartinib with pembrolizumab in solid tumors, including sarcoma and GISTs (NCT02452424). A phase I study of imatinib plus ipilimumab with an expansion cohort for patients with GIST is accruing (NCT01738139). Finally, a phase II study of axitinib plus pembrolizumab is ongoing for patients with advanced bone and soft tissue sarcomas, with a focus on patients with alveolar soft part sarcoma (NCT02636725).

Immunotherapy for NY-ESO-1–Positive Sarcomas

NY-ESO-1 is one of the best-characterized and most immunogenic cancer testis antigens. It is well documented that the asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 809

WILKY, JONES, AND KEEDY

majority of synovial sarcomas express NY-ESO-1, and this opens up the potential for targeted immunotherapy for patients with this particular subtype. Although this disease is relatively sensitive to chemotherapy, particularly ifosfamide, the outcome for patients with pretreated metastatic disease is poor. A number of studies have reported the feasibility of targeting NY-ESO-1, with promising results. In addition, there are a number of ongoing trials. These include a pilot trial of genetically engineered NY-ESO-1–specific (c259) T cells in patients with HLA-A2–positive synovial sarcoma. A separate study is also ongoing for NY-ESO-1–positive myxoid/round cell liposarcomas, which also have a high rate of NY-ESO-1 expression. Another approach is to extract native NY-ESO-1–specific T cells using tetramer-based cell sorting after peptide-pulsed, dendritic cell–based stimulation.9 Cells are then expanded in the presence of interleukin-21 and returned to the patients for adoptive therapy. This protocol is ongoing in a phase I trial for patients with NY-ESO-1–expressing synovial sarcomas and myxoid/ round cell liposarcomas (NCT01477021), as well as in combination with radiation therapy (NCT02319824). Finally, a randomized, open-label phase II trial of CMB305 (sequentially administered LV305 and G305) that targets NY-ESO-1 with the PD-L1 inhibitor atezolizumab is accruing patients with locally advanced, relapsed, or metastatic synovial or myxoid liposarcomas expressing NY-ESO-1.

Adoptive Chimeric Antigen Receptor T-Cell Therapy for HER2-Positive Sarcomas

Subsets of sarcomas overexpress HER2, which has been targeted using a chimeric antigen receptor (CAR) T cell. Transduction of a self-activating CAR into the patient’s harvested T cells avoids the requirement for HLA matching needed for engineered T-cell receptor approaches like the NY-ESO-1 strategy mentioned earlier. This approach demonstrated safety and tolerability with modest clinical responses in a phase I trial for 19 patients with osteosarcoma, Ewing sarcoma, and desmoplastic small round blue cell tumors.10 Expansion with lymphodepleting chemotherapy is ongoing (NCT00902044).

CONCLUSION

In summary, recent years have seen remarkable growth in novel treatment strategies for sarcomas, with an increased emphasis on understanding genetic and molecular biology of various sarcoma subtypes to guide design of clinical trials and optimal patient enrollment for new targeted therapies. Although biomarkers of response to immunotherapy are just beginning to be explored, the observations of remarkable benefit in some patients provide hope for immunotherapy as a future established treatment paradigm. It is critical that patients and providers are aware of the rich opportunities for clinical trials for patients with sarcoma refractory to standard therapies.

References 1. Attia S, Sankhala KK, Riedel RF, et al. A phase 1B/phase 2A study of TRC105 (endoglin antibody) in combination with pazopanib (P) in patients (pts) with advanced soft tissue sarcoma (STS). J Clin Oncol. 2016;34 (suppl; abstr 11016).

6. Seifert AM, Zeng S, Zhang JQ, et al. PD-1/PD-L1 blockade enhances T cell activity and antitumor efficacy of imatinib in gastrointestinal stromal tumors. Clin Cancer Res. 2017;23:454-465.

2. Van Tine B, Agulnik M, Olson RD, et al. A phase 2 trial of 5-imino-12deoxydoxorubicin (GPX-150) in metastatic and non-resectable soft tissue sarcomas. J Clin Oncol. 2016;34 (suppl; abstr 11019).

7. Patwardhan PP, Surriga O, Beckman MJ, et al. Sustained inhibition of receptor tyrosine kinases and macrophage depletion by PLX3397 and rapamycin as a potential new approach for the treatment of MPNSTs. Clin Cancer Res. 2014;20:3146-3158.

3. Tap WD, Jones RL, Van Tine BA, et al. Olaratumab and doxorubicin versus doxorubicin alone for treatment of soft-tissue sarcoma: an open-label phase 1b and randomised phase 2 trial. Lancet. 2016;388:488-497.

8. Kumar V, Gabrilovich DI. Hypoxia-inducible factors in regulation of immune responses in tumour microenvironment. Immunology. 2014;143:512-519.

4. Tawbi HA, Burgess MA, Crowley J, et al. Safety and efficacy of PD-1 blockage using pembrolizumab in patients with advanced soft tissue (STS) and bone sarcomas (BS): results of SARC028—a multicenter study. J Clin Oncol. 2016;34 (suppl; abstr 11006).

9. Pollack SM, Jones RL, Farrar EA, et al. Tetramer guided, cell sorter assisted production of clinical grade autologous NY-ESO-1 specific CD8(+) T cells. J Immunother Cancer. 2014;2:36.

5. Balachandran VP, Cavnar MJ, Zeng S, et al. Imatinib potentiates antitumor T cell responses in gastrointestinal stromal tumor through the inhibition of Ido. Nat Med. 2011;17:1094-1100.

810 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

10. Ahmed N, Brawley VS, Hegde M, et al. Human epidermal growth factor receptor 2 (HER2)–specific chimeric antigen receptor–modified T cells for the immunotherapy of HER2-positive sarcoma. J Clin Oncol. 2015;33:1688-1696.

TUMOR BIOLOGY

VELCHETI, RADIVOYEVITCH, AND SAUNTHARARAJAH

Higher-Level Pathway Objectives of Epigenetic Therapy: A Solution to the p53 Problem in Cancer Vamsidhar Velcheti, MD, Tomas Radivoyevitch, PhD, and Yogen Saunthararajah, MD OVERVIEW Searches for effective yet nontoxic oncotherapies are searches for exploitable differences between cancer and normal cells. In its core of cell division, cancer resembles normal life, coordinated by the master transcription factor MYC. Outside of this core, apoptosis and differentiation programs, which dominantly antagonize MYC to terminate cell division, necessarily differ between cancer and normal cells, as apoptosis is suppressed by biallelic inactivation of the master regulator of apoptosis, p53, or its cofactor p16/CDKN2A in approximately 80% of cancers. These genetic alterations impact therapy: conventional oncotherapy applies stress upstream of p53 to upregulate it and causes apoptosis (cytotoxicity)—a toxic, futile intent when it is absent or nonfunctional. Differentiation, on the other hand, cannot be completely suppressed because it is a continuum along which all cells exist. Neoplastic evolution stalls advances along this continuum at its most proliferative points—in lineage-committed progenitors that have division times measured in hours compared with weeks for tissue stem cells. This differentiation arrest is by mutations/deletions in differentiation-driving transcription factors or their coactivators that shift balances of gene-regulating protein complexes toward corepressors that repress instead of activate hundreds of terminal differentiation genes. That is, malignant proliferation without differentiation, also referred to as cancer “stem” cell self-renewal, hinges on druggable corepressors. Inhibiting these corepressors (e.g., DNMT1) releases p53-independent terminal differentiation in cancer stem cells but preserves self-renewal of normal stem cells that express stem cell transcription factors. Thus, epigenetic-differentiation therapies exploit a fundamental distinction between cancer and normal stem cell self-renewal and have a pathway of action downstream of genetic defects in cancer, affording favorable therapeutic indices needed for clinical progress.

he search for solutions to the fundamental problems of toxicity and resistance in oncotherapy reduces to a search for druggable differences between cancer and normal self-replication. Self-replication is the engine that drives all biologic evolution, including neoplastic evolution. Huge public and private efforts have focused on investigations of the mechanisms of cancer self-renewal and the development of candidate drugs that target this as the heart of the malignancy.1 Fundamental differences between malignant and normal self-renewal have been identified. These distinctions have opened the door to novel treatments that target one, but not the other, and that are rational in the overall genetic and epigenetic context of cancer, including near universal p53-system inactivation.

LOSS OF DIFFERENTIATION AND CANCER

“Anaplasia” (loss of differentiation) and “dedifferentiation” were coined in 1890 during the earliest histologic examinations of cancer by Hansemann.2 Today, we routinely use differentiation failure to distinguish malignant from

benign tumors (e.g., adenocarcinoma from adenoma), while the degree of differentiation failure identifies more from less aggressive transformation (e.g., Richter syndrome from chronic lymphocytic leukemia, and acute myeloid leukemia [AML] from myelodysplastic syndromes [MDS]). Even when loss of differentiation is not readily apparent by light microscopy, it is evident by gene expression analyses. For example, grade 1 hepatocellular carcinomas, although “well-differentiated” by light microscopy, demonstrate suppression of hundreds of hepatocyte specialization genes relative to normal liver cells (Fig. 1).

Why?

Multicellularity, defined by cell specialization/differentiation, arose approximately 600 million years ago, after about 3 billion years of unicellular cell growth and division. More ancient cell growth and division, coordinated by MYC or its paralogs, thus had to be conquered by differentiation genes for ordered multicellularity to succeed.5,6 The nature of this dominant regulation varies with differentiation stage.

From the Department of Hematology & Medical Oncology, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH; Department of Quantitative Health Sciences, Cleveland Clinic, Cleveland, OH; Department of Translational Hematology & Oncology Research, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Yogen Saunthararajah, MD, Taussig Cancer Institute Cleveland Clinic, Case Comprehensive Cancer Center, 9500 Euclid Ave., R40, Cleveland, OH 44195; email: [emailprotected]. © 2017 American Society of Clinical Oncology

812 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PATHWAY OBJECTIVES OF EPIGENETIC THERAPY

In developed tissues, cells are organized into functionally distinct differentiation stages (Fig. 2): 1. Tissue stem cells can self-replicate, but slowly. MYC activation is not vigorous; intervals between cell divisions extend to weeks or months, and proliferation kinetics are quiescent or linear9-11 (reviewed in Li12). Stem cells also produce daughter cells committed to various lineages (multipotency). 2. Lineage-committed progenitors are lineage-committed daughter cells of stem cells that activate and stabilize MYC to high levels that result in intervals between cell divisions measured in hours9-11 (reviewed in Li12) and exponential growth kinetics.13-17 These cell divisions are coupled to advances along differentiation axes that activate, as governors of growth, terminaldifferentiation programs that antagonize MYC and force cell cycle exits.18-25 The coupling that exists between master transcription factor drivers of differentiation and those of cell growth and division can be observed biochemically. 3. Terminally differentiated cells do not actively divide. They focus instead on the execution of specialized functions to serve the interests and needs of the overall multicellular organism. Thus, cancers suppress differentiation because progressive differentiation dominantly antagonizes MYC and terminates replication. The cause-effect relationship is, however, an area of scientific debate. One view is that increases in proliferation (e.g., by stabilization/amplification of MYC by RAS mutations, MYC copy number gains) cause decreases in differentiation. This mechanism is expected to occur in stem cells that can proliferate without differentiating (i.e., selfrenew).26,27 Another possibility is that loss of differentiation in lineage-committed progenitors (e.g., by disruption of transcription factor circuits that activate terminaldifferentiation programs) converts exponential proliferation limited by terminal differentiation into exponential proliferation without differentiation (i.e., self-replication).28 These divergent views should be reconciled by phenotypes of self-replicating, accumulating cancer cells, being either more stem cell–like or lineage-committed progenitor–like.

KEY POINTS • Neoplastic evolution stalls advances along differentiation continuums at its most proliferative points—in lineagecommitted progenitors. • This converts exponential proliferation usually coupled with differentiation into exponential self-renewal (proliferation without differentiation). • The arrest hinges on corepressors, which are needed to epigenetically repress hundreds of terminaldifferentiation genes. • Corepressor inhibition releases p53-independent terminal-differentiation in cancer stem cells poised for these fates, but preserves self-renewal of normal stem cells.

WHERE IN THE DIFFERENTIATION CONTINUUM ARE DIFFERENTIATION ADVANCES STALLED IN CANCER? Because self-renewal is an inherent property of tissue stem cells, an intuitive expectation was that differentiation arrest is at the level of self-replicating tissue stem cells. The earliest investigation into this found that leukemia cells that initiated leukemia in immune-compromised mice were rare, with surface phenotype features resembling hematopoietic stem cells (HSCs; CD34+CD38–) of the normal hematopoietic hierarchy.26 Leukemia thus seemed to recapitulate the hierarchical structure of normal hematopoiesis from which it is derived. The rare self-replicating stem cell–like cells were coined “leukemia stem cells” or LSCs (also called “leukemia-initiating cells”). A gain-of-function hit in this compartment presumably caused a decrease in differentiation.26,27 Contradicting this initial report, however, several groups found that LSCs were much more common, having surface-phenotype features of lineage-committed progenitors (e.g., CD34+38+, CLL-1+, CD71+, CD90–, c-Kit–).29-40 In fact, with the incorporation of additional parameters into the sorting strategy (e.g., CD90) to better discriminate HSCs from downstream committed progenitors, leukemiainitiating capacity was found to be absent from HSC-like cells but present in committed progenitors.40-42 That is, LSCs were abundant and phenocopied lineage-committed progenitors, not rare HSC-like cells.40-42 Reinforcing this conclusion, highly recurrent transforming genetic alterations exclusively linked with AML and not normal hematopoiesis (e.g., NPM1, FLT3, and RAS mutations) were detected only in cells with committed progenitor phenotype and not in HSCs.41-44 Moreover, 85% to 97% of bone marrow cells in patients with de novo AML have granulocyte-monocyte progenitor surface phenotypes, not HSC phenotypes, accumulated at the expense of downstream mature cells.40,45 Thus, self-replicating, accumulating, leukemia-initiating AML cells are stalled at a lineage-committed, intrinsically proliferative level of the hematopoietic hierarchy, whether at diagnosis or at relapse.40-43 To mitigate controversies arising from reliance on surface phenotypes, it is useful to examine functionally deterministic biology. Although there are hundreds of transcription factors in a cell, only a handful are masters that command cell fates. This was demonstrated by studies of murine knock-outs, enhancers, and, most strikingly, lineage-conversion, in which introduction of a few transcription factors converts any cell into a stem cell or a completely different lineage.46-50 The master transcription factors that govern hematopoiesis are well documented. This facilitates analyses and interpretations for AML.47 Counter to intuitive expectations, LSCs express miniscule levels of stem cell master transcription factors (e.g., HLF).28,51,52 Instead, LSCs and AML cells from across the genetic spectrum of disease express very high, supra-normal levels of lineage differentiation–driving master transcription factors (e.g., CEBPA, PU.1) that usually drive granulomonocytic fate,51,52 with lineage destinations, nonetheless, not being achieved (Fig. 1). asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 813

VELCHETI, RADIVOYEVITCH, AND SAUNTHARARAJAH

FIGURE 1. Differentiation Failure Features in Cancer, Whether It Is Obvious by Histologic Examination (AML) or Not (HCC)

(A) Hundreds of specialized hepatocyte differentiation genes with functions in lipid metabolism, coagulation factor synthesis, etc., are suppressed in HCC compared with normal liver (rows: 353 genes, columns: samples; TCGA RNA-Seq). (B) Different HCC-initiating insults produce similar relative preservation of master progenitor TF, but multifold repression of key terminal differentiation TF. (C) Leukemia “stem” cells express low levels of master stem cell TF, but supra-normal levels of master lineage differentiation-driving TF. Even so, key terminal differentiation genes (e.g., CEBPE) are repressed. The normal hematopoietic hierarchy expresses expected levels of these factors. Gene expression by microarray GSE24006.3 Error bars = median ± range. (D) Different AML genetics but similar differentiation arrest of committed progenitors. AML morphologic subtypes (M0-M7) correspond to relative amounts of CEBPA, PU.1, and GATA1 that drive granulocytic, monocytic, and erythroid lineage-fates, respectively. Master stem cell TF: HLF, PRDM5, ZFP37; master commitment/early-differentiation TF: CEBPA, PU.1, GATA1; master late/terminal differentiation TF: CEBPE (rows: genes, columns: samples; TCGA RNA-Seq, n = 174).4 Abbreviations: AML, acute myeloid leukemia; HCC, hepatocellular carcinoma; TCGA, The Cancer Genome Atlas; TF, transcription factors; RNA-Seq, RNA sequencing; NHSC, normal hematopoietic stem cells; NMPP, normal multipotent progenitors; NCMP, normal common myeloid progenitor; NRAPos, normal mature myeloid cells; NMEP, normal megakaryocyte erythroid progenitor; LSC, leukemia stem cells; LPC, leukemia progenitor cells.

Master transcription factors that produce stem cells compared with lineage-committed progenitors in solid tissues are not as comprehensively characterized as for hematopoiesis. Nonetheless, where the identity of these transcription factors is known, the cancers that arise from these tissues 814 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

also express very high levels of master lineage-differentiation drivers. For example, malignant melanoma cells express high levels of the melanocyte differentiation drivers MITF and SOX10,53,54 rhabdomyosarcomas express high levels of the muscle-specifying transcription factor MYOD,55 clear

PATHWAY OBJECTIVES OF EPIGENETIC THERAPY

FIGURE 2. Terminal Differentiation Is the Apex Control on Proliferation

Loss of terminal differentiation causes malignant self-renewal, even as other genetic alterations suppress apoptosis and promote MYC protein levels. (A) Exponential proliferation in committed progenitors is self-limited by coupling to progressive maturation, culminating in activation of terminal differentiation programs. (B) Malignant self-renewal (red-dotted oval) is caused by loss of terminal differentiation, converting proliferation with differentiation into proliferation without differentiation (self-renewal). This contrasts fundamentally with normal tissue homeostasis, in which self-renewal is restricted to mostly quiescent stem cells (pink-dotted oval). (C) Apoptosis and proliferation have the same master transcription factors across histologies and species (p53 and MYC, respectively); differentiation has various master transcription factors/preferred coactivators, depending on lineage and maturation stage. Thus, genetic alterations that repress terminal differentiation are varied, but with a common theme of inactivating master transcription factors and/or their coactivators (TCGA pan-cancer, data from Xena browser). (D) Proliferation genes have chromatin that is poised for gene activation (low CpG methylation) even in the earliest tissue precursors (ESC), while renal and hepatocyte epithelial differentiation genes have repressed chromatin (high CpG methylation). This epigenetic gradient to activation of terminal differentiation likely interacts with coactivator loss for selective repression of terminal differentiation programs. MYC target genes = 5,716 CpG linked with 356 genes.7 Hepatocyte genes = 4,729 CpG linked with 353 genes suppressed in HCC compared with normal liver. Renal genes = 9,496 CpG linked with 394 genes suppressed in RCC compared with normal liver. Plotted are medians of methylation values (β-values) by Illumina 450k CpG array for the three categories of CpG. β-values in ESC from GSE31848 (n = 19).8 Abbreviations: TCGA, The Cancer Genome Atlas; ESC, embryonic stem cells.

cell renal cell cancers (RCC) express very high levels of the renal epithelial-fate driving transcription factors PAX2 and PAX8, and hepatocellular carcinomas express very high levels of hepatocyte fate transcription factors FOXA1, FOXA3, and, to some extent, GATA4. Yet, morphologically and/or by gene expression analyses, hundreds of target differentiation genes are suppressed, not activated. Again, cancer cells are at intermediate, intrinsically proliferative points in differentiation continuums. For example, the medulloblastoma gene expression profile corresponds to a normal maturation stage that has a high proliferation rate and low levels of late cerebellar differentiation genes,56 and squamous lung carcinoma cells also have gene expression profiles of intermediate stages of normal lung development and differentiation.56,57

Differentiation suspension is thus not at the level of stem cells but at the most proliferative point in the differentiation continuum, in committed lineage progenitors. Beneath superficial differences in histology and genetics, this is a common core to cancers.

Why?

Replication drives evolution and repopulation. Thus the logarithmically higher replication rate of lineage-committed progenitors compared with stem cells, together with less rigorous policing of the genome during this replication,58 likely determine the committed-progenitor context of transformation and cellular accumulation. With intrinsic proliferation rates so skewed, even the slightest advantage is asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 815

VELCHETI, RADIVOYEVITCH, AND SAUNTHARARAJAH

amplified tremondously. Moreover, differentiation programs limiting exponential replication are evolutionarily recent overlays on substantially more ancient cell growth and division programs and, as described below, have many ways to fail.5,6

HOW IS DIFFERENTIATION STALLED DESPITE LINEAGE COMMITMENT?

A fundamental, observable property of cancers, common across histologies and genetics, is high expression of master transcription factor drivers of lineage differentiation, yet anomalously, suppression of terminal differentiation genes usually invariably induced by these commanders. The detailed molecular mechanisms underlying this incongruity have been characterized in several instances. There is a

shared motif. Transcription factors integrate gene-regulating signaling inputs via dynamic interchange of opposing coactivators and corepressors59-61: coactivators create the chromatin modifications that facilitate gene activation, while corepressors execute the opposite function. The shared motif in cancer is that master transcription factor hub stoichiometry shifts toward corepressors and away from coactivators (Fig. 3) via genetic alterations described below.

Genetic Inactivation of Coactivators

A major revelation of the genomics revolution is the high rate at which various SWI/SNF family coactivators are inactivated in cancer genomes (Fig. 2). SWI/SNF are ATP-dependent chromatin remodelers that execute the most energetically expensive work in epigenetics, that of repositioning

FIGURE 3. Therapeutic Implications of Frequent Inactivating Mutations of p53 (TP53) and p16 (CDKN2A) and of Master Differentiation-Driving TF and Their Coactivators

(A) Most current treatments apply stress to cells upstream of p53/p16 circuitry with the goal of increasing p53 protein. This is a problem when p53 or p16 are missing or nonfunctional. (B) Overall survival TCGA Pan-Cancer (n = 5,364) stratified by TP53 mutation or bi-allelic deletion, CDKN2A biallelic deletion, or both alterations. (C) TP53 mutation rates in curable compared with incurable disseminated malignancies. (D) Terminal differentiation suppression is by altered master differentiation-driving transcription factor hub composition, which favors CoR over CoA (E). Inhibiting druggable CoR (e.g., DNMT1) can rebalance toward CoA function and release terminal-differentiation programs, because cancer cells (including cancer “stem” cells) are poised for these fates with high master differentiation-driving transcription factor expression. These cell cycle exits do not require missing p53/p16. Abbreviations: TF, transcription factors; CoR, corepressors; CoA, coactivator.

816 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PATHWAY OBJECTIVES OF EPIGENETIC THERAPY

nucleosomes. Different SWI/SNF coactivators are inactivated in different cancers,59-61 a phenomenom likely explained by preferences of lineage-driving transcription factors for specific coactivator subunits.60,62 One example is the SWI/SNF coactivator PBRM1, which is universally inactivated in one or both alleles in clear cell RCC. PBRM1 coactivates for the highly expressed PAX2/PAX8 master transcription factor circuit that drives renal epithelial differentiation . Another example is ARID1A, which is frequently biallelically inactivated in hepatocellular carcinoma and a coactivator for GATA4/FOXA1 circuitry that drives hepatocyte epithelial differentiation. Master transcription factors also recruit and use splicing factors, cohesins and other coactivators to activate genes, and recurrent inactivating mutations/ translocations in these and related factors (e.g., TET2) likely also contribute to the incongruity in cancer of high master differentiation-driving transcription factor expression yet also the repression of hundreds of differentiation target genes.62

haploinsufficiency in hepatocellular carcinoma, as with RUNX1 haploinsufficiency in AML, shifts coregulator stoichiometry of the lineage differentiation–driving circuitry toward corepressors to repress instead of activate hundreds of terminal-differentiation genes. Because GATA4 is a master transcription factor driver of differentiation in all three germ layers, highly recurrent GATA4 haploinsufficiency via frequent chromosome 8p deletions in several solid tumor malignancies could similarly underlie differentiation arrest. The gene most recurrently altered in de novo AMLs is NPM1. Mutant NPM1 dislocates the master transcription factor PU.1 but not CEBPA from the nucleus to the cytoplasm, again disrupting differentiation-driving transcription factor hub balances toward corepressors.74 The second most altered gene in de novo AMLs is FLT3; FLT3-activating mutations compromise transactivations by the differentiation driver CEBPA.75

Metabolic Inactivation of Coactivators

Proliferation genes (MYC target genes) are already poised or on, that is, DNA CpG hypomethylated, in the earliest tissue precursors, embryonic stem cells (Fig. 2D). By contrast, terminal-differentiation genes require substantial chromatin remodeling from hypermethylated (off) to hypomethylated (on) states during tissue ontogeny. This epigenetic gradient to activation of terminal-differentiation genes interacts with corepressor/coactivator balances in differentiation-driving transcription factor hubs to selectively suppress terminal-differentiation programs characteristic of cancers.

Abnormal gain-of-function mutations in isocitrate dehydrogenases (IDHs) that are highly recurrent in glioma and AML produce an oncometabolite—the R-enantiomer of 2-hydroxyglutarate—that inhibits alpha-ketoglutarate– dependent enzymes such as the TET family of DNA demethylation enzymes and the chromatin-remodeling lysine demethylases KDM4A and KDM4C, mutations of which have been linked to differentiation arrest.63

Translocations, Mutations, and Dislocations of Transcription Factors That Disrupt Corepressor/ Coactivator Exchange

Usually, the retinoic acid receptor (RARA) exchanges corepressors for coactivators upon binding of its ligand retinoic acid to activate terminal-differentiation programs including the granulocyte differentiation program. For the leukemia fusion protein PML-RARA, corepressor/coactivator exchange is no longer achieved by physiologic concentrations of retinoic acid, and hundreds of granulocyte differentiation genes are repressed instead of activated.64-66 RUNX1 is a key hematopoietic transcription factor that cooperates with master myeloid lineage differentiation–driving transcription factors such as PU.1 to exchange corepressors for coactivators.62,67 In the leukemia fusion protein RUNX1-ETO, the domain of RUNX1 that effects this cooperation is replaced with corepressor-recruiting domains of the ETO protein.68-70 Similarly, the EWS-FLI1 fusion protein, found in more than 85% of Ewing sarcomas, recruits corepressor complexes to arrest osteogenic differentiation.71,72 MLL (KMT2A) translocations in leukemia and cancer invariably remove the histone methyltransferase domain of MLL that usually creates the epigenetic activation mark H3K4me2 or me3. Inactivating mutations of RUNX1 that are highly recurrent in myeloid malignancies also disrupt usual cooperation with PU.1 to exchange corepressors for coactivators.62,67,68,73 GATA4 cooperates with FOXA1 in a similar way. Thus, GATA4

Epigenetic Gradient to Activation of the TerminalDifferentiation Program

Founder Mutations (“First-Hits”) Can Originate Many Commitment Decisions Antecedent to the Progenitor Cell of Transformation

Familial AML pedigrees demonstrate how founder mutations can originate in one compartment yet manifest their transforming properties far downstream, in daughter cells many commitment decisions removed from the cell of origin. For example, the most common cause of familial AML is loss-of-function mutations in RUNX1. RUNX1-deficient HSCs lineage-commit normally, but RUNX1 deficiency then compromises PU.1/CEBPA master transcription factor function, thus retarding maturation after commitment.51,73,76 Similarly, point mutations in CEBPA that cause familial AML permit commitment but impede subsequent maturation.30 In short, mutations may originate in germline or tissue stem cells but produce ectopic cell divisions/expansion in already highly proliferative lineage-committed daughter cells, where even a small advantage might be amplified tremondously, facilitating further evolution and transformation.

HOW THE APOPTOSIS PROGRAM IS SUPPRESSED IN CANCER

Apoptosis coemerged with differentiation/multicellularity to mediate orderly cell cycle arrests and suicide of stressed or damaged cells, in the interests of the larger cellular aggregate.5,6 Thus, in common with differentiation, apoptosis asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 817

VELCHETI, RADIVOYEVITCH, AND SAUNTHARARAJAH

dominantly regulates MYC-coordinated cell growth and division. The master transcription factor p53 (TP53) and its key cofactor p16 (CDKN2A) are pan-histology, pan-species, master regulators of apoptosis. Dominance over MYC explains why TP53 and/or CDKN2A are biallelically inactivated by mutation and deletion in approximately 80% of cancers, even as cancers simultaneously stabilize or amplify MYC by RAS mutations, PI3K/AKT pathway alterations, etc. p53/p16 inactivation in cancer has major treatment implications, discussed below.

THERAPEUTIC IMPLICATIONS

The way in which the three major metazoan programs of proliferation, differentiation, and apoptosis disconnect in cancer compared with normal cells is fundamental to treatment and its outcomes.

Treatment Implications of p53/p16 Deletion/ Inactivation

Conventional medical and radiation therapy intends to use apoptosis (cytotoxicity), via stress applied upstream of p53, to antagonize MYC and terminate malignant replication (Fig. 3). The drugs may have different proximal molecular actions (e.g., topoisomerase inhibition [daunorubicin] or termination of DNA chain synthesis [cytarabine]), but the downstream objective is shared: to upregulate p53/p16 (reviewed in Kinzler et al77). This is a futile intent when p53 and/or p16 are absent/nonfunctional, but normal cells with intact p53/p16 are meanwhile destroyed.78-83 Questions of drug sensitivity/resistance are usually investigated by looking for differences between sensitive and resistant cancer cells. Thus, the fact that more than 80% of cancer cell lines are p53- and/or p16-deficient to begin with can cause in vitro studies to underappreciate the role of p53/p16 in clinical resistance. Similarly, inactivation of p53/p16 is so commonplace in some cancer histologies that its impact on outcomes is hard to discern by studies of the individual cancer histology in isolation (reviewed in Kinzler et al77). It is highly illustrative, however, to note that the p53/p16 inactivation rate is close to zero in those few disseminated cancer types that are routinely cured by cytotoxic therapy (testicular cancer, pediatric acute lymphoblastic leukemia [ALL]), whereas rates approximate 80% in notoriously incurable disseminated cancers (e.g., pancreatic), and also to note relatively high rates ofp53/p16 inactivation in the few testicular cancer or pediatric ALL cases that are relapsed/refractory (e.g., > 90% in incurable low-hypodiploid ALL; Fig. 3C).84-87 In sum, p53/p16 inactivation is a difference between cancer and normal cells that hurts rather than helps when treatment goals are to activate apoptosis. Alternative pathways for antagonizing MYC and inducing irreversible cell cycle exits are thus needed for the majority of cancers, which lack functional p53 and/or p16.

Treatment Implications of Corepressor/Coactivator Imbalances in Master Transcription Factor Hubs

Malignant self-replication, also referred to as LSC/CSC self-renewal, needs corepressors to decouple differentiation 818 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

from proliferation. Thus, pharmacologic reduction of corepressor activity compensates for genetic reduction of coactivators, releases terminal-differentiation fates intended by the master transcription factors resident in LSCs/CSCs, and thus terminates malignant cell self-renewal.19,51,52,64,81,88,89 The same treatments increase self-renewal of normal tissue stem cells because these express high levels of master stem cell transcription factors, not differentiation drivers.51,88,90-98 Normal committed progenitors, such as CSCs and LSCs, also differentiate.51,88,90 Because MYC is subservient to terminal differentiation, replication is terminated even if MYC is stabilized and/or amplified by other genetic alterations typical of cancer.19,51,52,81,88,89 Also, this pathway of terminating malignant self-replication does not require p53/p16, which is often missing, but is not needed for differentiation19,51,52,81,89 (p53-null and p16/Cdkn2a-null mice develop/differentiate normally99,100). Finally, corepressor reduction operates downstream of the genetic defects in cancer cells that stalled differentiation in the first place. Cancers shrink and resolve because expansion derives from relentless self-replication that exceeds a high spontaneous death rate. The best illustration of these properties is, of course, clinical results. First example of clinical differentiation-restoring therapy. Acute promyelocytic leukemia (APL) was converted from the AML subtype with the worst overall survival to the best by using retinoic acid and arsenic to reverse differentiation failure mediated by corepressors recruited by the leukemia fusion protein PML-RARA.64-66,101 Overall survival of APL treated with differentiation therapy is better than for any other disseminated malignancy, including pediatric ALL, using only two drugs compared with the five or more used to treat ALL.64,101 Differentiation restoration using different agents could yield similar benefits against other cancers. DNMT1-targeting therapy. The corepressor DNMT1 is aberrantly enriched in master transcription factor hubs of multiple cancer types and has been scientifically validated as a pan-cancer target for differentiation-restoring therapy (reviewed in Saunthararajah et al28; Fig. 3).19,51,52,81,89,102-126 DNMT1 can be depleted by decitabine, a deoxycytidine analog with a modified base that binds to and depletes DNMT1 after incorporation into DNA, without terminating DNA chain elongation and, thus, at low useful doses, without cytotoxicity. A favorable therapeutic index that spares normal stem cells is especially critical when treating MDS/ AML in the elderly, as these cells are needed to reverse low blood counts that cause morbidity and death. Decitabine regimens approved to treat MDS, however, administer high doses that cause off-target cytotoxicity that requires pulse-cycled administration for recoveries. To generate clinical proof of concept of p53/p16-independent, noncytotoxic epigenetic-differentiation therapy, we treated patients with MDS with reduced decitabine doses (0.1–0.2 mg/kg/day compared with the U.S. Food and Drug Administration– approved 20–45 mg/m2/day—a 75%–90% reduction) that avoid cytotoxicity. These well-tolerated doses were administered 1 to 3 days/week nonstop, instead of pulse-cycled 3 to 5 days straight once every 4 to 6 weeks. This increases

PATHWAY OBJECTIVES OF EPIGENETIC THERAPY

probabilities that cancer S-phase entries coincide with drug presence in cells, which is required because DNMT1 depletion by decitabine is S-phase dependent. The patients were elderly with a median age of 73. Many had disease that was relapsed/refractory to standard first-line treatments. Adverse events were related to neutropenia present at baseline, and antiemetics were not needed. Responses meeting MDS working group criteria occurred in 44% of subjects and were highly durable, with treatment-induced freedom from transfusion lasting a median of 1,025 days with several still ongoing at the time of the data analysis; 20% of the subjects were treated for more than 3 years, including several patients older than age 80.52 Complete cytogenetic remissions were produced even in cases with biallelic inactivation of TP53 and complex chromosome abnormalities. Noncytotoxic DNMT1 depletion was confirmed by serial bone marrow γ-H2AX and DNMT1 analyses. MYC master oncoprotein levels were markedly decreased by treatment.52 In a subsequent report, a 100% response rate was observed in 21 patients with TP53-mutated/deleted MDS and AML.127 Interestingly, p53 loss biases pyrimidine metabolism toward decitabine uptake, further facilitating the use of this agent to treat p53-null malignancies.128 Selective effects on LSC compared with HSC self-replication and the p53-independent mechanism of action of DNMT1depleting therapy explains why 5-azacytidine and decitabine are the only two drugs approved for treatment of all MDS and are also routinely used to treat AMLs.52,81,129,130 Unfortunately, these observations in myeloid malignancies are not readily extended to p53/p16-null solid tumor malignancies, not because of diminished validity of DNMT1 as a therapeutic target, but because decitabine and 5azacytidine are inactivated within minutes by the pyrimidine metabolism enzyme cytidine deaminase (CDA) that is highly expressed in solid tissues.102 CDA upregulation within malignant cells is also a mechanism of resistance in myeloid malignancies.131,132 We thus combine decitabine with a CDA-inhibitor (tetrahydrouridine) for orally administered, noncytotoxic DNMT1-depleting treatment of TP53-mutated solid and liquid cancers (NCT02664181, NCT02847000, and NCT02846935). Other examples of clinical epigenetic-differentiation therapy. IDH2 inhibitors in clinical trials are able to salvage chemorefractory (apoptosis-resistant) AML by terminal differentiation.133 Inhibitors of FLT3 and of mutant-NPM1 nuclear export also restore terminal-differentiation observations in active clinical translation for chemorefractory AML.74 KDM1A (LSD1) inhibitors are in clinical trials to treat refractory/relapsed myeloid malignancies and small cell lung cancer.134-137 Liquid or solid tumor malignancies? Although clinical differentiation-restoring treatments are most advanced in myeloid malignancies, decades worth of preclinical research has documented terminal differentiation of solid cancer cells in response to corepressor inhibition. Examples include aggressive, differentiation-impaired melanoma and breast cancer cells that resumed differentiation completely through cell

cycle exits when exposed to an embryonic cell microenvironment that opens chromatin138,139; oocyte extracts, another microenvironment that induces DNA hypomethylation and removes repressive histone marks, terminating breast cancer cell tumorigenicity140; histone deacetylase inhibitors (HDACi) inducing terminal differentiation in a spectrum of solid cancer primary cells and cell lines, as well as leukemia cells68,141-151; genetic or pharmacologic suppression of KDM1A (LSD1), a component of the NURD corepressor complex, inducing terminal maturation in several solid cancer as well as leukemia models72,134-137; and DNMT1, validated by several groups as a molecular target for differentiation restoration of solid and liquid malignancies (reviewed in Saunthararajah et al28).51,52,64,81,88,103-126 In short, new drugs and pharmacologies are needed for clinical translation, but molecular targets for normal stem cell sparing, p53-independent, epigenetic-differentiation treatment of solid malignancies have been identified and validated.

RESISTANCE

Corepressor inhibition exploits a distinction between malignant and normal stem cell self-renewal and hence offers a solution for toxicity. Nonetheless, all drugs are metabolized, must distribute into target cells, and have to successfully engage their molecular targets, providing multiple opportunities for escape from treatment effects. That is, treatment resistance still must be addressed. For example, decitabine and 5-azacytidine must traverse pyrimidine metabolism pathways to reach their target. These pathways have regulators in place that are designed to minimize nucleotide imbalances. Thus, the nucleotide load of administered decitabine or 5-azacytidine is countered by reflexive metabolic shifts that decrease drug uptake (reviewed in Saunthararajah et al152).132 Nontoxic treatments targeting malignant selfreplication can, however, be rationally combined. This point is illustrated by the more than 95% cure rate of APL when treated with only two such drugs, retinoic acid and arsenic. Other than in APL, there have been no clinical trials of combination therapy for explicit noncytotoxic epigeneticdifferentiation goals, an omission that will be corrected moving forward (HDACi combined with 5-azacytidine or decitabine in clinical trials have been cytotoxic/cytostatic, antagonizing S-phase–dependent DNMT1-depletion by 5-azacytidine or decitabine [reviewed in Saunthararajah et al152]).153

CONCLUSION

Hippocrates said, “Natural forces within us are the true healers of disease.” LSCs/CSCs contain very high levels of master transcription factor drivers of lineage fate and are poised for terminal differentiation. Corepressors are the druggable barriers that suspend execution of these naturally intended fates. Although differentiation therapy is a decades-old idea,154-156 that corepressor/coactivator imbalance causes differentiation failure and thus malignant selfreplication is a relatively recent concept. Drug development and clinical applications, which have dwelled on self-renewal driving failed differentiation, have thus been lagging. Even so, asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 819

VELCHETI, RADIVOYEVITCH, AND SAUNTHARARAJAH

clinical proof of principle that epigenetic, differentiationrestoring treatment can be a broad solution to toxicity, and to resistance from p53/p16-inactivation, already exists.

Rational combinations of such treatments can solve other resistance problems to keep patients well and alive for even longer.

References 1. Kaiser J. The cancer stem cell gamble. Science. 2015;347:226-229. 2. Bignold LP, Coghlan B, Jersmann H. David Paul Hansemann: chromosomes and the origin of the cancerous features of tumor cells. Cell Oncol. 2009;31:61. 3. Gentles AJ, Plevritis SK, Majeti R, et al. Association of a leukemic stem cell gene expression signature with clinical outcomes in acute myeloid leukemia. JAMA. 2010;304:2706-2715. 4. Cancer Genome Atlas Research Network. Genomic and epigenomic landscapes of adult de novo acute myeloid leukemia. N Engl J Med. 2013;368:2059-2074. 5. Domazet-Loso T, Tautz D. Phylostratigraphic tracking of cancer genes suggests a link to the emergence of multicellularity in metazoa. BMC Biol. 2010;8:66. 6. Srivastava M, Simakov O, Chapman J, et al. The Amphimedon queenslandica genome and the evolution of animal complexity. Nature. 2010;466:720-726. 7. Kim J, Woo AJ, Chu J, et al. A Myc network accounts for similarities between embryonic stem and cancer cell transcription programs. Cell. 2010;143:313-324. 8. Nazor KL, Altun G, Lynch C, et al. Recurrent variations in DNA methylation in human pluripotent stem cells and their differentiated derivatives. Cell Stem Cell. 2012;10:620-634. 9. Nygren JM, Bryder D, Jacobsen SE. Prolonged cell cycle transit is a defining and developmentally conserved hemopoietic stem cell property. J Immunol. 2006;177:201-208. 10. Schwartz GN, Vance BA, Levine BM, et al. Proliferation kinetics of subpopulations of human marrow cells determined by quantifying in vivo incorporation of [2H2]-glucose into DNA of S-phase cells. Blood. 2003;102:2068-2073. 11. van der Wath RC, Wilson A, Laurenti E, et al. Estimating dormant and active hematopoietic stem cell kinetics through extensive modeling of bromodeoxyuridine label-retaining cell dynamics. PLoS One. 2009;4:e6972. 12. Li J. Quiescence regulators for hematopoietic stem cell. Exp Hematol. 2011;39:511-520. 13. Wilson A, Murphy MJ, Oskarsson T, et al. c-Myc controls the balance between hematopoietic stem cell self-renewal and differentiation. Genes Dev. 2004;18:2747-2763.

17. Arnold I, Watt FM. c-Myc activation in transgenic mouse epidermis results in mobilization of stem cells and differentiation of their progeny. Curr Biol. 2001;11:558-568. 18. Acosta JC, Ferrándiz N, Bretones G, et al. Myc inhibits p27-induced erythroid differentiation of leukemia cells by repressing erythroid master genes without reversing p27-mediated cell cycle arrest. Mol Cell Biol. 2008;28:7286-7295. 19. Negrotto S, Hu Z, Alcazar O, et al. Noncytotoxic differentiation treatment of renal cell cancer. Cancer Res. 2011;71:1431-1441. 20. Grote D, Souabni A, Busslinger M, et al. Pax 2/8-regulated Gata 3 expression is necessary for morphogenesis and guidance of the nephric duct in the developing kidney. Development. 2006;133:53-61. 21. Green LM, Wagner KJ, Campbell HA, et al. Dynamic interaction between WT1 and BASP1 in transcriptional regulation during differentiation. Nucleic Acids Res. 2009;37:431-440. 22. Lucas B, Grigo K, Erdmann S, et al. HNF4alpha reduces proliferation of kidney cells and affects genes deregulated in renal cell carcinoma. Oncogene. 2005;24:6418-6431. 23. Ramaswamy S, Nakamura N, Sansal I, et al. A novel mechanism of gene regulation and tumor suppression by the transcription factor FKHR. Cancer Cell. 2002;2:81-91. 24. Aschauer L, Gruber LN, Pfaller W, et al. Delineation of the key aspects in the regulation of epithelial monolayer formation. Mol Cell Biol. 2013;33:2535-2550. 25. Kojima T, Shimazui T, Horie R, et al. FOXO1 and TCF7L2 genes involved in metastasis and poor prognosis in clear cell renal cell carcinoma. Genes Chromosomes Cancer. 2010;49:379-389. 26. Lapidot T, Sirard C, Vormoor J, et al. A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature. 1994;367:645-648. 27. Dick JE. Stem cell concepts renew cancer research. Blood. 2008;112:4793-4807. 28. Saunthararajah Y, Triozzi P, Rini B, et al. p53-Independent, normal stem cell sparing epigenetic differentiation therapy for myeloid and other malignancies. Semin Oncol. 2012;39:97-108. 29. Taussig DC, Miraki-Moud F, Anjos-Afonso F, et al. Anti-CD38 antibody-mediated clearance of human repopulating cells masks the heterogeneity of leukemia-initiating cells. Blood. 2008;112:568-575.

14. Reavie L, Della Gatta G, Crusio K, et al. Regulation of hematopoietic stem cell differentiation by a single ubiquitin ligase-substrate complex. Nat Immunol. 2010;11:207-215.

30. Kirstetter P, Schuster MB, Bereshchenko O, et al. Modeling of C/ EBPalpha mutant acute myeloid leukemia reveals a common expression signature of committed myeloid leukemia-initiating cells. Cancer Cell. 2008;13:299-310.

15. Laurenti E, Varnum-Finney B, Wilson A, et al. Hematopoietic stem cell function and survival depend on c-Myc and N-Myc activity. Cell Stem Cell. 2008;3:611-624.

31. Huntly BJ, Shigematsu H, Deguchi K, et al. MOZ-TIF2, but not BCRABL, confers properties of leukemic stem cells to committed murine hematopoietic progenitors. Cancer Cell. 2004;6:587-596.

16. Zhang J, Xiao Y, Guo Y, et al. Differential requirements for c-Myc in chronic hematopoietic hyperplasia and acute hematopoietic malignancies in Pten-null mice. Leukemia. 2011;25:1857-1868.

32. Somervaille TC, Cleary ML. Identification and characterization of leukemia stem cells in murine MLL-AF9 acute myeloid leukemia. Cancer Cell. 2006;10:257-268.

820 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

PATHWAY OBJECTIVES OF EPIGENETIC THERAPY

33. van Rhenen A, Moshaver B, Kelder A, et al. Aberrant marker expression patterns on the CD34+CD38- stem cell compartment in acute myeloid leukemia allows to distinguish the malignant from the normal stem cell compartment both at diagnosis and in remission. Leukemia. 2007;21:1700-1707. 34. Blair A, Hogge DE, Ailles LE, et al. Lack of expression of Thy-1 (CD90) on acute myeloid leukemia cells with long-term proliferative ability in vitro and in vivo. Blood. 1997;89:3104-3112. 35. Blair A, Sutherland HJ. Primitive acute myeloid leukemia cells with long-term proliferative ability in vitro and in vivo lack surface expression of c-kit (CD117). Exp Hematol. 2000;28:660-671. 36. Wunderlich M, Chou FS, Link KA, et al. AML xenograft efficiency is significantly improved in NOD/SCID-IL2RG mice constitutively expressing human SCF, GM-CSF and IL-3. Leukemia. 2010;24:17851788. 37. Feuring-Buske M, Gerhard B, Cashman J, et al. Improved engraftment of human acute myeloid leukemia progenitor cells in beta 2-microglobulin-deficient NOD/SCID mice and in NOD/SCID mice transgenic for human growth factors. Leukemia. 2003;17:760-763. 38. Agliano A, Martin-Padura I, Mancuso P, et al. Human acute leukemia cells injected in NOD/LtSz-scid/IL-2Rgamma null mice generate a faster and more efficient disease compared to other NOD/scid-related strains. Int J Cancer. 2008;123:2222-2227. 39. Sarry JE, Murphy K, Perry R, et al. Human acute myelogenous leukemia stem cells are rare and heterogeneous when assayed in NOD/SCID/ IL2Rγc-deficient mice. J Clin Invest. 2011;121:384-395. 40. Goardon N, Marchi E, Atzberger A, et al. Coexistence of LMPP-like and GMP-like leukemia stem cells in acute myeloid leukemia. Cancer Cell. 2011;19:138-152. 41. Corces-Zimmerman MR, Hong WJ, Weissman IL, et al. Preleukemic mutations in human acute myeloid leukemia affect epigenetic regulators and persist in remission. Proc Natl Acad Sci USA. 2014;111:2548-2553.

50. Forrest AR, Kawaji H, Rehli M, et al; FANTOM Consortium and the RIKEN PMI and CLST (DGT). A promoter-level mammalian expression atlas. Nature. 2014;507:462-470. 51. Negrotto S, Ng KP, Jankowska AM, et al. CpG methylation patterns and decitabine treatment response in acute myeloid leukemia cells and normal hematopoietic precursors. Leukemia. 2012;26:244-254. 52. Saunthararajah Y, Sekeres M, Advani A, et al. Evaluation of noncytotoxic DNMT1-depleting therapy in patients with myelodysplastic syndromes. J Clin Invest. 2015;125:1043-1055. 53. McGill GG, Horstmann M, Widlund HR, et al. Bcl2 regulation by the melanocyte master regulator Mitf modulates lineage survival and melanoma cell viability. Cell. 2002;109:707-718. 54. Cronin JC, Wunderlich J, Loftus SK, et al. Frequent mutations in the MITF pathway in melanoma. Pigment Cell Melanoma Res. 2009;22:435-444. 55. Yang Z, MacQuarrie KL, Analau E, et al. MyoD and E-protein heterodimers switch rhabdomyosarcoma cells from an arrested myoblast phase to a differentiated state. Genes Dev. 2009;23:694-707. 56. Kho AT, Zhao Q, Cai Z, et al. Conserved mechanisms across development and tumorigenesis revealed by a mouse development perspective of human cancers. Genes Dev. 2004;18:629-640. 57. Warburton D, Schwarz M, Tefft D, et al. The molecular basis of lung morphogenesis. Mech Dev. 2000;92:55-81. 58. Walter D, Lier A, Geiselhart A, et al. Exit from dormancy provokes DNA-damage-induced attrition in haematopoietic stem cells. Nature. 2015;520:549-552. 59. Shain AH, Pollack JR. The spectrum of SWI/SNF mutations, ubiquitous in human cancers. PLoS One. 2013;8:e55119. 60. Lemon B, Inouye C, King DS, et al. Selectivity of chromatin-remodelling cofactors for ligand-activated transcription. Nature. 2001;414:924928.

42. Shlush LI, Zandi S, Mitchell A, et al; HALT Pan-Leukemia Gene Panel Consortium. Identification of pre-leukaemic haematopoietic stem cells in acute leukaemia. Nature. 2014;506:328-333.

61. Ho L, Jothi R, Ronan JL, et al. An embryonic stem cell chromatin remodeling complex, esBAF, is an essential component of the core pluripotency transcriptional network. Proc Natl Acad Sci USA. 2009;106:5187-5191.

43. Jan M, Snyder TM, Corces-Zimmerman MR, et al. Clonal evolution of preleukemic hematopoietic stem cells precedes human acute myeloid leukemia. Sci Transl Med. 2012;4:149ra118.

62. Gu X, Hu Z, Ebrahem Q, et al. Runx1 regulation of Pu.1 corepressor/ coactivator exchange identifies specific molecular targets for leukemia differentiation therapy. J Biol Chem. 2014;289:14881-14895.

44. Jaiswal S, Ebert BL. MDS is a stem cell disorder after all. Cancer Cell. 2014;25:713-714.

63. Chowdhury R, Yeoh KK, Tian YM, et al. The oncometabolite 2-hydroxyglutarate inhibits histone lysine demethylases. EMBO Rep. 2011;12:463-469.

45. Quek L, Otto GW, Garnett C, et al. Genetically distinct leukemic stem cells in human CD34- acute myeloid leukemia are arrested at a hemopoietic precursor-like stage. J Exp Med. 2016;213: 1513-1535. 46. Vierbuchen T, Wernig M. Molecular roadblocks for cellular reprogramming. Mol Cell. 2012;47:827-838. 47. Riddell J, Gazit R, Garrison BS, et al. Reprogramming committed murine blood cells to induced hematopoietic stem cells with defined factors. Cell. 2014;157:549-564. 48. Iwasaki H, Akashi K. Myeloid lineage commitment from the hematopoietic stem cell. Immunity. 2007;26:726-740. 49. Yamanaka R, Barlow C, Lekstrom-Himes J, et al. Impaired granulopoiesis, myelodysplasia, and early lethality in CCAAT/enhancer binding protein epsilon-deficient mice. Proc Natl Acad Sci USA. 1997;94:13187-13192.

64. Lo-Coco F, Avvisati G, Vignetti M, et al; Gruppo Italiano Malattie Ematologiche dell’Adulto; German-Austrian Acute Myeloid Leukemia Study Group; Study Alliance Leukemia. Retinoic acid and arsenic trioxide for acute promyelocytic leukemia. N Engl J Med. 2013;369:111-121. 65. Hu Z, Saunthararajah Y. CEBPE activation in PML-RARA cells by arsenic. Blood. 2012;119:2177-2179. 66. Huang ME, Ye YC, Chen SR, et al. All-trans retinoic acid with or without low dose cytosine arabinoside in acute promyelocytic leukemia. Report of 6 cases. Chin Med J (Engl). 1987;100:949-953. 67. Hu Z, Gu X, Baraoidan K, et al. RUNX1 regulates corepressor interactions of PU.1. Blood. 2011;117:6498-6508. 68. Wang J, Saunthararajah Y, Redner RL, et al. Inhibitors of histone deacetylase relieve ETO-mediated repression and induce differentiation of AML1-ETO leukemia cells. Cancer Res. 1999;59:2766-2769.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 821

VELCHETI, RADIVOYEVITCH, AND SAUNTHARARAJAH

69. Amann JM, Nip J, Strom DK, et al. ETO, a target of t(8;21) in acute leukemia, makes distinct contacts with multiple histone deacetylases and binds mSin3A through its oligomerization domain. Mol Cell Biol. 2001;21:6470-6483. 70. Wang J, Hoshino T, Redner RL, et al. ETO, fusion partner in t(8;21) acute myeloid leukemia, represses transcription by interaction with the human N-CoR/mSin3/HDAC1 complex. Proc Natl Acad Sci USA. 1998;95:10860-10865. 71. Sankar S, Bell R, Stephens B, et al. Mechanism and relevance of EWS/ FLI-mediated transcriptional repression in Ewing sarcoma. Oncogene. 2013;32:5089-5100. 72. Komura S, Semi K, Itakura F, et al. An EWS-FLI1-induced osteosarcoma model unveiled a crucial role of impaired osteogenic differentiation on osteosarcoma development. Stem Cell Rep. 2016;6:592-606. 73. Ng KP, Hu Z, Ebrahem Q, et al. Runx1 deficiency permits granulocyte lineage commitment but impairs subsequent maturation. Oncogenesis. 2013;2:e78. 74. Gu X, Mahfouz RZ, Enane F, et al. A specific mechanism by which NPM1 mutations impede myeloid differentiation also explains the link with DNMT3A mutation. Blood. 2013;122:1254. 75. Radomska HS, Bassères DS, Zheng R, et al. Block of C/EBP alpha function by phosphorylation in acute myeloid leukemia with FLT3 activating mutations. J Exp Med. 2006;203:371-381. 76. Sun W, Downing JR. Haploinsufficiency of AML1 results in a decrease in the number of LTR-HSCs while simultaneously inducing an increase in more mature progenitors. Blood. 2004;104:3565-3572. 77. Kinzler KW, Vogelstein B. Cancer therapy meets p53. N Engl J Med. 1994;331:49-50. 78. Roberts DA, Wadleigh M, McDonnell AM, et al. Low efficacy and high mortality associated with clofarabine treatment of relapsed/ refractory acute myeloid leukemia and myelodysplastic syndromes. Leuk Res. 2015;39:204-210. 79. Brennig S, Rattmann I, Lachmann N, et al. In vivo enrichment of cytidine deaminase gene-modified hematopoietic cells by prolonged cytosine-arabinoside application. Cytotherapy. 2012;14:451-460. 80. Ben-Ishay Z, Barak V. Bone marrow stromal dysfunction in mice administered cytosine arabinoside. Eur J Haematol. 2001;66:230-237. 81. Ng KP, Ebrahem Q, Negrotto S, et al. p53 independent epigeneticdifferentiation treatment in xenotransplant models of acute myeloid leukemia. Leukemia. 2011;25:1739-1750. 82. Mandelli F, Petti MC, Ardia A, et al. A randomised clinical trial comparing idarubicin and cytarabine to daunorubicin and cytarabine in the treatment of acute non-lymphoid leukaemia. A multicentric study from the Italian Co-operative Group GIMEMA. Eur J Cancer. 1991;27:750-755. 83. Schoch C, Kern W, Schnittger S, et al. The influence of age on prognosis of de novo acute myeloid leukemia differs according to cytogenetic subgroups. Haematologica. 2004;89:1082-1090. 84. Houldsworth J, Xiao H, Murty VV, et al. Human male germ cell tumor resistance to cisplatin is linked to TP53 gene mutation. Oncogene. 1998;16:2345-2349. 85. Ding LW, Sun QY, Tan KT, et al. Mutational landscape of pediatric acute lymphoblastic leukemia. Cancer Res. 2017;77:390-400. 86. Hof J, Krentz S, van Schewick C, et al. Mutations and deletions of the TP53 gene predict nonresponse to treatment and poor outcome in

822 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

first relapse of childhood acute lymphoblastic leukemia. J Clin Oncol. 2011;29:3185-3193. 87. Holmfeldt L, Wei L, Diaz-Flores E, et al. The genomic landscape of hypodiploid acute lymphoblastic leukemia. Nat Genet. 2013;45:242252. 88. Hu Z, Negrotto S, Gu X, et al. Decitabine maintains hematopoietic precursor self-renewal by preventing repression of stem cell genes by a differentiation-inducing stimulus. Mol Cancer Ther. 2010;9:15361543. 89. Alcazar O, Achberger S, Aldrich W, et al. Epigenetic regulation by decitabine of melanoma differentiation in vitro and in vivo. Int J Cancer. 2012;131:18-29. 90. Milhem M, Mahmud N, Lavelle D, et al. Modification of hematopoietic stem cell fate by 5aza 2'deoxycytidine and trichostatin A. Blood. 2004;103:4102-4110. 91. De Felice L, Tatarelli C, Mascolo MG, et al. Histone deacetylase inhibitor valproic acid enhances the cytokine-induced expansion of human hematopoietic stem cells. Cancer Res. 2005;65:1505-1513. 92. Bug G, Gül H, Schwarz K, et al. Valproic acid stimulates proliferation and self-renewal of hematopoietic stem cells. Cancer Res. 2005;65:25372541. 93. Young JC, Wu S, Hansteen G, et al. Inhibitors of histone deacetylases promote hematopoietic stem cell self-renewal. Cytotherapy. 2004;6:328-336. 94. Lee JH, Hart SR, Skalnik DG. Histone deacetylase activity is required for embryonic stem cell differentiation. Genesis. 2004;38:32-38. 95. Araki H, Mahmud N, Milhem M, et al. Expansion of human umbilical cord blood SCID-repopulating cells using chromatin-modifying agents. Exp Hematol. 2006;34:140-149. 96. Suzuki M, Harashima A, Okochi A, et al. 5-Azacytidine supports the long-term repopulating activity of cord blood CD34(+) cells. Am J Hematol. 2004;77:313-315. 97. Chung YS, Kim HJ, Kim TM, et al. Undifferentiated hematopoietic cells are characterized by a genome-wide undermethylation dip around the transcription start site and a hierarchical epigenetic plasticity. Blood. 2009;114:4968-4978. 98. Chaurasia P, Gajzer DC, Schaniel C, et al. Epigenetic reprogramming induces the expansion of cord blood stem cells. J Clin Invest. 2014;124:2378-2395. 99. Attardi LD, Donehower LA. Probing p53 biological functions through the use of genetically engineered mouse models. Mutat Res. 2005;576:4-21. 100. Serrano M, Lee H, Chin L, et al. Role of the INK4a locus in tumor suppression and cell mortality. Cell. 1996;85:27-37. 101. Smith ML, Hills RK, Grimwade D. Independent prognostic variables in acute myeloid leukaemia. Blood Rev. 2011;25:39-51. 102. Ebrahem Q, Mahfouz RZ, Ng KP, et al. High cytidine deaminase expression in the liver provides sanctuary for cancer cells from decitabine treatment effects. Oncotarget. 2012;3:1137-1145. 103. Shakya R, Gonda T, Quante M, et al. Hypomethylating therapy in an aggressive stroma-rich model of pancreatic carcinoma. Cancer Res. 2013;73:885-896. 104. Cecconi D, Astner H, Donadelli M, et al. Proteomic analysis of pancreatic ductal carcinoma cells treated with 5-aza-2′-deoxycytidine. Electrophoresis. 2003;24:4291-4303.

PATHWAY OBJECTIVES OF EPIGENETIC THERAPY

105. Yamada T, Ohwada S, Saitoh F, et al. Induction of Ley antigen by 5-aza-2′-deoxycytidine in association with differentiation and apoptosis in human pancreatic cancer cells. Anticancer Res. 1996;16: 735-740. 106. Belinsky SA, Klinge DM, Stidley CA, et al. Inhibition of DNA methylation and histone deacetylation prevents murine lung cancer. Cancer Res. 2003;63:7089-7093. 107. Belinsky SA, Grimes MJ, Picchi MA, et al. Combination therapy with vidaza and entinostat suppresses tumor growth and reprograms the epigenome in an orthotopic lung cancer model. Cancer Res. 2011;71:454-462. 108. Zöchbauer-Müller S, Minna JD, Gazdar AF. Aberrant DNA methylation in lung cancer: biological and clinical implications. Oncologist. 2002;7:451-457. 109. Liu CC, Lin JH, Hsu TW, et al. IL-6 enriched lung cancer stem-like cell population by inhibition of cell cycle regulators via DNMT1 upregulation. Int J Cancer. 2015;136:547-559. 110. Kim HJ, Kim JH, Chie EK, et al. DNMT (DNA methyltransferase) inhibitors radiosensitize human cancer cells by suppressing DNA repair activity. Radiat Oncol. 2012;7:39. 111. Rauch T, Wang Z, Zhang X, et al. Homeobox gene methylation in lung cancer studied by genome-wide analysis with a microarraybased methylated CpG island recovery assay. Proc Natl Acad Sci USA. 2007;104:5527-5532.

122. Ushmorov A, Leithäuser F, Sakk O, et al. Epigenetic processes play a major role in B-cell-specific gene silencing in classical Hodgkin lymphoma. Blood. 2006;107:2493-2500. 123. Tsai HC, Li H, Van Neste L, et al. Transient low doses of DNAdemethylating agents exert durable antitumor effects on hematological and epithelial tumor cells. Cancer Cell. 2012;21:430-446. 124. Momparler RL, Côté S, Momparler LF. Epigenetic action of decitabine (5-aza-2′-deoxycytidine) is more effective against acute myeloid leukemia than cytotoxic action of cytarabine (ARA-C). Leuk Res. 2013;37:980-984. 125. Liu Y, Tabarroki A, Billings S, et al. Successful use of very low dose subcutaneous decitabine to treat high-risk myelofibrosis with Sweet syndrome that was refractory to 5-azacitidine. Leuk Lymphoma. 2014;55:447-449. 126. Tabarroki A, Saunthararajah Y, Visconte V, et al. Ruxolitinib in combination with DNA methyltransferase inhibitors; clinical responses in symptomatic myelofibrosis patients with cytopenias and elevated blasts counts. Leuk Lymphoma. 2015;56:497-499. 127. Welch JS, Petti AA, Miller CA, et al. TP53 and decitabine in acute myeloid leukemia and myelodysplastic syndromes. N Engl J Med. 2016;375:2023-2036. 128. Radivoyevitch T, Saunthararajah Y, Pink J, et al. dNTP supply gene expression patterns after P53 loss. Cancers (Basel). 2012;4:1212-1224.

112. Peters SL, Hlady RA, Opavska J, et al. Essential role for Dnmt1 in the prevention and maintenance of MYC-induced T-cell lymphomas. Mol Cell Biol. 2013;33:4321-4333.

129. Fenaux P, Mufti GJ, Hellström-Lindberg E, et al. Azacitidine prolongs overall survival compared with conventional care regimens in elderly patients with low bone marrow blast count acute myeloid leukemia. J Clin Oncol. 2010;28:562-569.

113. Höglund A, Nilsson LM, Forshell LP, et al. Myc sensitizes p53-deficient cancer cells to the DNA-damaging effects of the DNA methyltransferase inhibitor decitabine. Blood. 2009;113:4281-4288.

130. Dombret H, Seymour JF, Butrym A, et al. International phase 3 study of azacitidine vs conventional care regimens in older patients with newly diagnosed AML with >30% blasts. Blood. 2015;126:291-299.

114. Guan H, Xie L, Klapproth K, et al. Decitabine represses translocated MYC oncogene in Burkitt lymphoma. J Pathol. 2013;229:775-783.

131. Mahfouz RZ, Jankowska A, Ebrahem Q, et al. Increased CDA expression/activity in males contributes to decreased cytidine analog half-life and likely contributes to worse outcomes with 5-azacytidine or decitabine therapy. Clin Cancer Res. 2013;19:938-948.

115. Hassler MR, Klisaroska A, Kollmann K, et al. Antineoplastic activity of the DNA methyltransferase inhibitor 5-aza-2′-deoxycytidine in anaplastic large cell lymphoma. Biochimie. 2012;94:2297-2307. 116. Kalac M, Scotto L, Marchi E, et al. HDAC inhibitors and decitabine are highly synergistic and associated with unique gene-expression and epigenetic profiles in models of DLBCL. Blood. 2011;118: 5506-5516. 117. Leshchenko VV, Kuo PY, Jiang Z, et al. Integrative genomic analysis of temozolomide resistance in diffuse large B-cell lymphoma. Clin Cancer Res. 2014;20:382-392. 118. Iqbal J, Kucuk C, Deleeuw RJ, et al. Genomic analyses reveal global functional alterations that promote tumor growth and novel tumor suppressor genes in natural killer-cell malignancies. Leukemia. 2009;23:1139-1151. 119. Kozłowska A, Jagodziński PP. Inhibition of DNA methyltransferase activity upregulates Fyn tyrosine kinase expression in Hut-78 T-lymphoma cells. Biomed Pharmacother. 2008;62:672-676. 120. Ripperger T, von Neuhoff N, Kamphues K, et al. Promoter methylation of PARG1, a novel candidate tumor suppressor gene in mantle-cell lymphomas. Haematologica. 2007;92:460-468. 121. Han Y, Amin HM, Frantz C, et al. Restoration of shp1 expression by 5-AZA-2′-deoxycytidine is associated with downregulation of JAK3/ STAT3 signaling in ALK-positive anaplastic large cell lymphoma. Leukemia. 2006;20:1602-1609.

132. Ebrahem Q, Mahfouz RZ, Durkin L, et al. Mechanisms of resistance to 5-azacytidine/decitabine in MDS-AML and pre-clinical in vivo proof of principle of rational solutions to extend response. Blood. 2015;126:678. 133. Wang F, Travins J, DeLaBarre B, et al. Targeted inhibition of mutant IDH2 in leukemia cells induces cellular differentiation. Science. 2013;340:622-626. 134. Mohammad HP, Smitheman KN, Kamat CD, et al. A dna hypomethylation signature predicts antitumor activity of lsd1 inhibitors in SCLC. Cancer Cell. 2015;28:57-69. 135. Mould DP, McGonagle AE, Wiseman DH, et al. Reversible inhibitors of LSD1 as therapeutic agents in acute myeloid leukemia: clinical significance and progress to date. Med Res Rev. 2015;35:586-618. 136. Schenk T, Chen WC, Göllner S, et al. Inhibition of the LSD1 (KDM1A) demethylase reactivates the all-trans-retinoic acid differentiation pathway in acute myeloid leukemia. Nat Med. 2012;18:605-611. 137. Harris WJ, Huang X, Lynch JT, et al. The histone demethylase KDM1A sustains the oncogenic potential of MLL-AF9 leukemia stem cells. Cancer Cell. 2012;21:473-487. 138. Postovit LM, Margaryan NV, Seftor EA, et al. Human embryonic stem cell microenvironment suppresses the tumorigenic phenotype of aggressive cancer cells. Proc Natl Acad Sci USA. 2008;105:4329-4334.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 823

VELCHETI, RADIVOYEVITCH, AND SAUNTHARARAJAH

139. Cowan CA, Atienza J, Melton DA, et al. Nuclear reprogramming of somatic cells after fusion with human embryonic stem cells. Science. 2005;309:1369-1373. 140. Allegrucci C, Rushton MD, Dixon JE, et al. Epigenetic reprogramming of breast cancer cells with oocyte extracts. Mol Cancer. 2011;10:7. 141. Kumagai T, Wakimoto N, Yin D, et al. Histone deacetylase inhibitor, suberoylanilide hydroxamic acid (Vorinostat, SAHA) profoundly inhibits the growth of human pancreatic cancer cells. Int J Cancer. 2007;121:656-665. 142. Gozzini A, Rovida E, Dello Sbarba P, et al. Butyrates, as a single drug, induce histone acetylation and granulocytic maturation: possible selectivity on core binding factor-acute myeloid leukemia blasts. Cancer Res. 2003;63:8955-8961. 143. Kosugi H, Towatari M, Hatano S, et al. Histone deacetylase inhibitors are the potent inducer/enhancer of differentiation in acute myeloid leukemia: a new approach to anti-leukemia therapy. Leukemia. 1999;13:1316-1324. 144. Nowak D, Stewart D, Koeffler HP. Differentiation therapy of leukemia: 3 decades of development. Blood. 2009;113:3655-3665. 145. Spira AI, Carducci MA. Differentiation therapy. Curr Opin Pharmacol. 2003;3:338-343. 146. Gore SD, Samid D, Weng LJ. Impact of the putative differentiating agents sodium phenylbutyrate and sodium phenylacetate on proliferation, differentiation, and apoptosis of primary neoplastic myeloid cells. Clin Cancer Res. 1997;3:1755-1762. 147. Moldenhauer A, Frank RC, Pinilla-Ibarz J, et al. Histone deacetylase inhibition improves dendritic cell differentiation of leukemic blasts with AML1-containing fusion proteins. J Leukoc Biol. 2004;76: 623-633.

824 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

148. Jones PA, Taylor SM. Cellular differentiation, cytidine analogs and DNA methylation. Cell. 1980;20:85-93. 149. Pinto A, Attadia V, Fusco A, et al. 5-Aza-2′-deoxycytidine induces terminal differentiation of leukemic blasts from patients with acute myeloid leukemias. Blood. 1984;64:922-929. 150. Creusot F, Acs G, Christman JK. Inhibition of DNA methyltransferase and induction of Friend erythroleukemia cell differentiation by 5-azacytidine and 5-aza-2′-deoxycytidine. J Biol Chem. 1982;257:2041-2048. 151. Niitsu N, Hayashi Y, Sugita K, et al. Sensitization by 5-aza-2′deoxycytidine of leukaemia cells with MLL abnormalities to induction of differentiation by all-trans retinoic acid and 1alpha,25dihydroxyvitamin D3. Br J Haematol. 2001;112:315-326. 152. Saunthararajah Y. Key clinical observations after 5-azacytidine and decitabine treatment of myelodysplastic syndromes suggest practical solutions for better outcomes. Hematology (Am Soc Hematol Educ Program). 2013;2013:511-521. 153. Prebet T, Sun Z, Figueroa ME, et al. Prolonged administration of azacitidine with or without entinostat for myelodysplastic syndrome and acute myeloid leukemia with myelodysplasia-related changes: results of the US Leukemia Intergroup trial E1905. J Clin Oncol. 2014;32:1242-1248. 154. Pierce GB Jr, Verney EL. An in vitro and in vivo study of differentiation in teratocarcinomas. Cancer. 1961;14:1017-1029. 155. Michalewicz R, Lotem J, Sachs L. Cell differentiation and therapeutic effect of low doses of cytosine arabinoside in human myeloid leukemia. Leuk Res. 1984;8:783-790. 156. Seilern-Aspang F, Kratochwil K. Induction and differentiation of an epithelial tumour in the newt (Triturus cristatus). J Embryol Exp Morphol. 1962;10:337-356.

TARGETING CANCER METABOLISM

Metabolic Alterations in Cancer and Their Potential as Therapeutic Targets Jamie D. Weyandt, PhD, Craig B. Thompson, MD, Amato J. Giaccia, PhD, and W. Kimryn Rathmell, MD, PhD OVERVIEW Otto Warburg’s discovery in the 1920s that tumor cells took up more glucose and produced more lactate than normal cells provided the first clues that cancer cells reprogrammed their metabolism. For many years, however, it was unclear as to whether these metabolic alterations were a consequence of tumor growth or an adaptation that provided a survival advantage to these cells. In more recent years, interest in the metabolic differences in cancer cells has surged, as tumor proliferation and survival have been shown to be dependent upon these metabolic changes. In this educational review, we discuss some of the mechanisms that tumor cells use for reprogramming their metabolism to provide the energy and nutrients that they need for quick or sustained proliferation and discuss the potential for therapeutic targeting of these pathways to improve patient outcomes.

etabolic pathways are the means by which cells break down nutrients to acquire the energy and building blocks that they need for growth, proliferation, and the maintenance of critical cellular processes. Energy within cells is stored by adenosine triphosphate (ATP) molecules, which are both required and produced by metabolic pathways and thus are referred to as cellular energy currency. Cells generate ATP through respiration, of which there are two distinct mechanisms: aerobic and anaerobic. Both of these pathways require the initial uptake of glucose, which is converted through a series of steps known as glycolysis to pyruvate. However, at this point, what happens to pyruvate is typically dependent upon the environmental conditions surrounding the cell. Aerobic respiration, most often used by normal cells under normal, nonproliferating conditions, requires oxygen and results in the conversion of the glycolytic product pyruvate to acetyl coenzyme A (acetylCoA). The primary function of acetyl-CoA is to donate an acetyl group to the citric acid (also known as tricarboxylic acid, TCA, or Krebs) cycle. By continuing through the TCA cycle and electron transport chain (ETC) reactions, all of which take place in the mitochondria, the downstream metabolism of a single molecule of glucose by aerobic respiration yields a net gain of about 36 molecules of ATP and releases carbon dioxide as a byproduct. This process is often collectively referred to as oxidative phosphorylation (Fig. 1A). Anaerobic respiration, on the other hand, is far less efficient, producing a net gain of only two molecules of

ATP per molecule of glucose metabolized and thus is typically only used during hypoxic or stressful conditions, as it does not require the presence of oxygen. During anaerobic respiration—also referred to as fermentation—pyruvate is converted to lactate and ethyl alcohol entirely within the cytosol. Although inefficient, this pathway can keep the cell alive during stressful conditions in which the supply of oxygen is low by generating enough ATP to continue sustained cycling through glycolysis (Fig. 1A). Although normal or quiescent cells rely primarily on aerobic respiration/oxidative phosphorylation to meet their energy requirements, cancer cells appear to meet their increased demands for energy quite differently. Because tumor cells grow rapidly, they must increase the import of nutrients from their environment in an effort to maintain the pools of ATP and, even more importantly, carbon intermediates that serve as building blocks for the assembly of DNA, proteins, and lipids needed during cell growth and division. In the 1920s, Otto Warburg first made the observation that tumors took up markedly higher levels of glucose in comparison with normal tissues.1 Furthermore, Warburg showed that even in the presence of ample oxygen, cancer cells produced much more lactate than normal tissues, suggesting that these cells were shuttling glucose through the glycolytic fermentation pathway.2 The sustained use of this pathway to meet energy requirements under normoxic conditions is now termed "aerobic glycolysis," and the increased dependence on this pathway by cancer cells has

From the Department of Medicine, Division of Hematology and Oncology, Vanderbilt-Ingram Cancer Center, Vanderbilt University Medical Center, Nashville, TN; Cancer Biology and Genetics Program, Memorial Sloan Kettering Cancer Center, New York, NY; Department of Radiation Oncology, Stanford University School of Medicine, Stanford, CA. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: W. Kimryn Rathmell, MD, PhD, Department of Medicine, Division of Hematology and Oncology, Vanderbilt-Ingram Cancer Center, 2220 Pierce Avenue, Preston Research Building, Suite 777, Vanderbilt University Medical Center, Nashville, TN 37232; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 825

WEYANDT ET AL

FIGURE 1. Metabolic Reprogramming in Tumor Cells: The Warburg Effect

(A) Under normoxic conditions, normal tissues convert the glycolytic product of pyruvate to cetyl coenzyme A, which is used in the mitochondria for the tricarboxylic acid cycle to begin the process of oxidative phosphorylation. In the absence of oxygen, pyruvate is converted to lactate, and sustained anaerobic glycolysis is used to meet requirements for energy and nutrients. (B) Tumor cells convert the majority of the glycolytic product pyruvate to lactate and replenish their nutrients and energy through sustained aerobic glycolysis, but maintain mitochondrial function and some oxidative respiration. Abbreviation: ATP, adenosine triphosphate.

come to be known as the Warburg effect (Fig. 1A). In more recent years, this phenomenon has been confirmed in a number of different tumor types in different tissues and has proven useful for diagnostic imaging using 18F-deoxyglucose positron emission tomography to detect the higher levels of glucose uptake observed in tumors in comparison with surrounding normal stroma.3 It seems counterintuitive that rapidly dividing cancer cells would prefer the less efficient glycolytic pathway for meeting their energy demands. For this reason, Warburg originally hypothesized that the increased rates of aerobic glycolysis in cancer cells were attributable to impaired function of the mitochondria in these cells, requiring them to rely solely upon glycolysis to make ATP needed for survival.2,4 This theory has been disproven in more recent years, however, as the majority of cancer cells

KEY POINTS • Warburg metabolism is a common feature of tumor biology, but it only represents one way that tumors adapt metabolic processes to survival advantage. • Mutations and modifications of Krebs cycle and electron transport function also underlie tumor cell physiology. • Biometabolites find functional use in energy generation and as essential components of epigenetic features. • Numerous directions are being investigated to harness energetic processes as therapeutic strategies for cancer. 826 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

have been found to maintain functioning mitochondria.5 It has also become increasingly clear that tumor cells continue to carry out oxidative respiration in addition to sustained aerobic glycolysis (Fig. 1B) and that a likely advantage of this altered metabolic profile is the sustained production of glycolytic carbon intermediates required for the production of macromolecules needed by the rapidly dividing cells.6 Although the Warburg effect is perhaps the most recognized metabolic characteristic of many cancer cells, a broad range of metabolic alterations has been observed in tumors. In addition to increased glucose uptake, tumor cells have also been commonly shown to have higher levels of dependence on glutamine, which is a source of nitrogen for the synthesis of nucleotides and amino acids.7 Interactions involving various intermediates of glycolysis, the TCA cycle, the ETC, and the pentose phosphate pathway, as well as lipid metabolism pathways, have all been shown to be altered in tumor cells and to play a role in tumorigenesis.8 These metabolic changes can result from genetic aberrations in metabolic enzymes themselves, but can also be a downstream consequence of activating mutations in numerous growth factors and oncogenes, loss of tumor suppressor signaling, or epigenetic alterations,7 all of which we will discuss in more detail in later sections of this review. Recent findings demonstrating the influence of metabolic pathways on tumor cell proliferation, growth, and differentiation have renewed interest in identifying susceptibilities of these pathways to therapeutic intervention, and thus the investigation of metabolic

TARGETING CANCER METABOLISM

reprogramming as a hallmark of cancer has become an extremely active area of research in the last decade.3

METABOLIC ALTERATIONS IN RENAL CELL CARCINOMAS

The renal cell carcinomas (RCCs) are prime examples of tumor types that are highly linked to alterations in metabolic pathways. There are three main subtypes of RCC: clear cell (ccRCC), papillary (pRCC), and chromophone (chRCC), each distinguished by unique histology and driver mutations.9 Interestingly, although the overall mutational burden is relatively low in RCC in comparison with many other tumor types,9 the vast majority of mutations identified in these tumors are in some way involved in the cell’s ability to sense or respond to nutrients, oxygen, iron, or energy, suggesting that metabolic pathway alterations are key drivers of proliferation in all subsets of RCC.10 Mutations resulting in dysregulation of specific steps of glycolysis, the TCA cycle, and the ETC pathways have all been found in subtypes of RCC, illustrating the diversity of metabolic alterations that may contribute to tumorigenesis (Fig. 2). Here, we discuss three examples of mutations that alter different metabolic pathways in RCC.

VHL Mutations in Clear Cell Renal Cell Carcinmoa

Mutations in the von Hippel-Lindau gene (VHL) are associated with a hereditary form of RCC found in patients with

germline VHL disease but are also observed in nearly 90% of patients with sporadic clear cell kidney cancer (ccRCC).11 The VHL protein is considered a tumor suppressor, and under normal circ*mstances, when there is enough oxygen and iron in the cell, it is part of a complex that binds to the hypoxia-inducible factors (HIFs) and targets them for degradation by ubiquitination.12 In the majority of cases of ccRCC, inactivating mutations in VHL inhibit its ability to interact with the HIF proteins, and consequently the HIF proteins are stabilized, even during normoxic conditions. The HIF proteins are transcription factors that regulate the activity of a number of downstream genes, including glucose transporters GLUT1 and GLUT3, endothelial growth factor, vascular endothelial growth factor (or VEGF), and platelet-derived growth factor.10 The aberrant activation of these proteins and growth factors is believed to contribute to tumor growth and proliferation downstream of inactivating mutations in VHL. The up-regulation of GLUT1 and GLUT3 likely contributes to the faster rates and increased levels of glycolysis in these tumors. A number of glycolytic enzymes are also transcriptionally regulated by HIFs, including HK1, HK2, GPI, PFKL, ALDA, ALDC, TPI, GAPDH, PGK, ENO1, and PKM. Increased expression of these enzymes may also contribute to the increased glycolytic activity in the VHL mutant ccRCC tumors.13 Finally, lactate dehydrogenase A (LDHA) expression is also transcriptionally regulated by HIFs. This enzyme converts the glycolytic product pyruvate to lactate,

FIGURE 2. Metabolic Alterations in RCC Subsets

Clear cell renal cell carcinoma (left panel) frequently exhibits mutations in VHL, resulting in stabilization of HIFs and their transcriptional targets, including VEGF and GLUT1, and thus is characterized by increased angiogenesis and up-regulated glycolysis. Mutations in FH and SDH in papillary renal cell carcinoma (middle panel) inhibit completion of the TCA cycle and result in accumulation of fumarate and/ or succinate. Chromophobe renal cell carcinoma (right panel) is rare but associated with mutations in mitochondrial complex I enzymes, such as MT-ND5, leading to an inhibition of electron transport chain reactions and an accumulation of defective mitochondria. Abbreviations: VHL, von Hippel-Lindau, HIF, hypoxia-inducible factor; TCA, tricarboxylic acid; ETC, electron transport chain.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 827

WEYANDT ET AL

and thus up-regulation of this enzyme contributes to the increased levels of lactic acid observed in tumors, a result of sustained aerobic glycolysis at the expense of the conversion of pyruvate to acetyl-CoA for use in mitochondrial oxidative phosphorylation.13 Thus, VHL mutant tumors exhibit classic features of pseudo-hypoxic Warburg metabolism: up-regulated glycolysis, high levels of lactate production, and lower levels of oxidative phosphorylation. An understanding of how the VHL and HIF pathways contribute to ccRCC tumorigenesis has provided the basis for most current treatments of patients with advanced ccRCC. Most current therapies target the VEGF signaling pathway, inhibiting angiogenesis.10,14 Increased knowledge of the metabolic dependencies of RCC cells has also led to increased interest in targeting the HIF pathways and their metabolism-regulating targets. Recently, a HIF-2 agonist showed promise in reducing growth in a subset of cell lines in patients with ccRCC.15 Agonists of GLUT1 and glycolytic pathway enzymes have also been investigated as potential therapeutic inhibitors of glycolysis in RCC.16,17 Further characterization of the metabolic reprogramming that occurs in ccRCC has the potential to identify additional vulnerabilities of therapeutic value.

SDH and FH Mutations in Papillary Renal Cell Carcinoma

Mutations in several TCA cycle enzymes have been observed in pRCC. Succinate dehydrogenase (SDH) catalyzes the oxidation of succinate to fumarate in the TCA cycle. Germline mutations in the succinate dehydrogenase family subunits SDHB, SDHC, and SDHD have been identified in patients with familial paraganglioma/pheochromocytoma who are predisposed to developing pRCC tumors and in other patients with a family history of pRCC.18 Likewise, germline mutations in fumarate hydratase (FH), the enzyme that catalyzes the conversion of fumarate to malate in the TCA cycle, have been found in patients with hereditary leiomyomatosis RCC and, very rarely, in sporadic cases of pRCC.19 Because both SDH and FH mutations block normal TCA cycle and ETC activity, cells from these tumors take up almost no oxygen and rely primarily on glycolysis to supply energy and macromolecules needed for replication and growth. These tumors thus also exhibit Warburg metabolism and produce high levels of lactate.10 In the case of SDH-deficient tumors, succinate accumulates in the mitochondrial matrix as a result of loss of SDH function. Succinate, however, can also leak out into the cytosol, where it can inhibit the prolyl hydroxylation of HIF complexes, preventing them from being targeted for proteasomal degradation. In this way, succinate accumulation stabilizes the HIF transcription factors, thus promoting the activation of their downstream targets, creating a pseudohypoxic expression signature.20 Therefore, in addition to defective mitochondrial respiration, SDH mutant cells also have increased expression of GLUT1, VEGF, and other growth factors and glycolytic enzymes, promoting cell growth and proliferation, angiogenesis, and means for up-regulating 828 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

glycolysis. These tumors have also been shown to have increased vasculature.18 Although it has not yet been well investigated because of the rarity of this tumor type, targeting the HIFs or the glycolytic pathway in these cells may have potential therapeutic value in these tumors.10 FH mutations similarly result in the accumulation of both succinate and fumarate as a result of the malfunction of the FH enzyme in the TCA cycle. Like succinate, fumarate can also move from the mitochondria into the cytoplasm, where it can interact with prolyl hydroxylases and prevent the degradation of HIF proteins.21 Similarly to SDH mutant tumors, pRCC tumors with FH mutations have up-regulated expression of the HIF target genes involved in proliferation, glycolysis, and angiogenesis. Highly vascularized, these tumors grow very aggressively and have a pseudo-hypoxic gene expression profile.22 Patients with these tumors typically have a poor prognosis, and more research is needed to identify improved therapies. The malfunctions of mitochondrial respiration and up-regulation of glycolysis in these cells appear to be key factors in their proliferation, and thus investigation of these pathways may be important for improving outcome for patients with FH mutations.

Electron Transport Chain-Complex I Mutations in Chromophobe Renal Cell Carcinoma

A third subset of RCC, known as chromophobe RCC (or chRCC), is the least common type of RCC. Like many of the RCCs, this type of tumor is associated with a hereditary disorder, Birt-Hogg-Dubé syndrome. Until more recent years, however, it was not known what genetic alterations contributed to sporadic cases of chRCC. Interestingly, PET/CT scans have demonstrated that, in contrast to other types of RCC, chRCC tumors are nonglycolytic, taking up very limited amounts of glucose.23 In addition, gene expression profiling of these tumors indicated that genes involved in the TCA cycle and ETC pathways were up-regulated in these tumors.24 Mitochondrial DNA sequencing has revealed that many chRCC tumors have mutations in genes involved in the ETC complex I, particularly in MT-ND5, and that these mitochondrial gene mutations also correlate with samples exhibiting an eosinophilic histologic phenotype.24 This phenotype also correlates with an increase in mitochondrial mass resulting from an accumulation of mitochondria, possibly in compensation for hindered mitochondrial functioning.9 Thus, the metabolic profile of chRCC appears to be very different from that of other types of kidney cancer. Although the mechanisms behind the accumulation of mitochondria in this tumor type remain to be investigated, it is clear that metabolic alterations may play an important role in growth of this rare tumor type, and hence further study of these pathways for potential use as biomarkers and therapeutic targeting is warranted. In summary, the RCCs provide an illustration of the varied strategies used by cancer cells to augment growth through manipulations of their metabolic activities. These activities reveal possible critical dependencies, which, as has been referenced above, have been examined in terms of using

TARGETING CANCER METABOLISM

altered glycolysis for diagnostic as well as potentially therapeutic intervention. Below, we will highlight the emerging strategies to intervene in cellular metabolism for therapeutic benefit, including strategies currently approved in renal cancers and other malignancies, and new concepts that may apply in the future alone, or as adjuncts to treatments.

OTHER METABOLIC ALTERATIONS IN CANCER: CURRENT THERAPEUTIC TARGETS

A number of different kinds of genetic mutations have been associated with the dysregulation of metabolic pathways in various tumor types. The reprogramming of metabolism and alterations in metabolic flux of tumor cells compared with normal cells confers unique properties to these cells, which may prove to be useful for therapeutic targeting in patients wtih cancer. Here, we describe some of the pathways currently identified as regulators of metabolism in tumors and the current therapies targeting these alterations.

mTOR Inhibition

The mechanistic (previously mammalian) target of rapamycin (mTOR) is a serine-threonine protein kinase that forms complexes with other proteins and is involved in a number of cellular processes related to growth, proliferation, survival, motility, and protein translation.25 mTOR signaling is commonly dysregulated in cancer through several different mechanisms. Although mutations in the MTOR gene itself can occur, it is more commonly activated downstream of gain-of-function mutations in the PI3K-AKT pathway or growth factors, or through inactivation of tumor suppressors such as PTEN. mTOR is also activated downstream of activation of 5′-adenosine monophosphate–activated protein kinase (AMPK), a protein that serves as an intracellular sensor of nutrients.26 mTOR activation plays a key role in controlling intracellular metabolism through its involvement in protein translation and autophagy. The mTOR pathway has been shown to stimulate glutaminolysis by up-regulating the expression of MYC, which in turn up-regulates glutaminase, which converts glutamine to glutamate that can be used to make alpha-ketoglutarate for use in the TCA cycle.27 mTOR activation is also known to play a role in the stabilization of HIF proteins, resulting in increased activation of their transcriptional targets, including GLUT1, VEGF, and other glycolysis enzymes.28 Thus, activation of the mTOR pathway plays a role in the up-regulation of glycolysis, glutamine uptake, and angiogenesis in cancer cells. Two mTOR inhibitors, temsirolimus and everolimus, have been approved in the United States and Europe for the treatment of solid tumors. These drugs bind to the mTOR complex 1 (mTORC1) by associating with FK506-binding protein12 (FKBP12), blocking the correct alignment of substrates to the catalytic cleft of this complex.29 These drugs have shown benefits in delaying the progression and extending survival in advanced RCC, breast, and pancreatic cancers. However, resistance to these inhibitors appears to develop over time, possibly as a result of the accumulation of additional mutations in the mTOR pathway30 or through

negative feedback of the pathway itself, as inhibiting mTOR signaling can also up-regulate AKT signaling through insulinlike growth factor receptor 1 (IGF-1R).31 Thus, continued research to find less resistant mechanisms for inhibiting mTOR is needed.

Metformin/Phenformin Inhibition of Oxidative Phosphorylation

As cancer cells are frequently known to be metabolically active and have very high levels of glucose uptake, it has been postulated that hypoglycemic drugs that have been used for treating diabetes could help to restore normal metabolism in these cells and prevent tumor growth. Two such drugs, metformin and phenformin, have shown some promise in targeting cancer cell metabolism. These organic compounds are known as biguanides, and it has been shown that diabetic patients taking them have a reduced risk of developing cancer.32,33 The exact mechanisms by which biguanides regulate cellular metabolism is not yet well understood, but they are believed to interfere with mitochondrial complex I, inhibiting oxidative phosphorylation, while activating the AMPK signaling pathway.34 Metformin and phenformin have been shown to delay progression of tumor cell growth in breast cancer35 and melanoma32 and have also exhibited antiangiogenic properties.35 One disadvantage of treatment with these compounds is that they can induce severe acidosis in patients. More research is needed to determine the most effective dosage levels and which tumor types may be most susceptible to biguanide treatment.

Glutaminase Inhibition

Although cancer cells have been shown to have highly up-regulated glycolysis, demonstrated by the Warburg effect, they also maintain oxidative phosphorylation. In addition to glycolysis, many cancer cells appear to be dependent upon glutamine metabolism to supply the nutrients and biosynthetic precursors that they need for macromolecule synthesis.36 The glutaminase enzyme converts glutamine to glutamate, which can be used to make alpha-ketoglutarate (α-KG), important in the TCA cycle. The TCA cycle intermediates are used in the synthesis of nucleic and fatty acids, and thus interfering with glutamine metabolism can have a profoundly detrimental effect on replicating cells. Several mechanisms have been proposed for inhibiting glutaminase, including targeting ASCT2, the transporter that mediates glutamine uptake into cells, and the use of glutamine mimetics to competitively inhibit glutamine uptake and activity. Unfortunately, early clinical trials testing glutamine mimetics resulted in high levels of toxicity in patients.37 More recently, however, several small molecule allosteric inhibitors of glutaminase activity have been identified, including CB-839, currently in clinical trials, and bis-2-(5phenylacetamido-1,3,4-thiadiazol-2-yl) ethyl sulfide (BPTES). Targeting glutaminase activity has been shown to reduce oncogenic transformation in cancer cells,38 and allosteric inhibitors of glutaminase have been used in combination with other chemotherapeutics to reduce tumor cell growth asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 829

WEYANDT ET AL

in lymphoma,39 lung,40 and breast41 cancer cell lines. Further research will likely focus on determining which tumors are most glutamine dependent and thus most susceptible to glutaminase inhibition. It will also be important to investigate the most efficient methods for targeting glutaminase activity in cancer cells while minimizing toxicity to others.

Inhibition of Isocitrate Dehydorgenase Enzymes 1 and 2

The isocitrate dehydrogenase enzymes (IDH1 and IDH2) are important metabolic enzymes that convert isocitrate to alpha-ketoglutarate by oxidative decarboxylation. α-KG is a key player in the TCA cycle, and thus these enzymes play an important role in oxidative phosphorylation. IDH1 and IDH2 also play a role in the generation of NADPH, a reducing factor that helps to protect the cell against oxidative damage.42 Therefore, mutations in IDH1 and IDH2 are believed to both alter cellular metabolism and potentially increase rates of DNA damage attributable to altered NADPH protection. Mutations in IDH1 and IDH2 have been observed in several types of tumors, including leukemias, lymphomas, and gliomas. The mutations identified in IDH1 and IDH2 in cancers appear to be gain-of-function point mutations that occur at specific arginine residues that presumably alter the structure of these proteins. These mutations lead to increased conversion of α-KG to D-2-hydroxyglutarate.43 High levels of D-2-hydroxyglutarate have been associated with increases in histone and DNA methylation, contributing to tumor progression.42 It has also been shown that mutant IDH1 heterodimerizes with wild-type IDH1, inhibiting the activity of the wild-type enzyme and reducing levels of α-KG, which may play a role in the degradation of HIF proteins. Thus, mutant IDH1 may also play a role in the stabilization of HIFs and increased activation of their transcription factors involved in tumorigenesis and angiogenesis.43 Several targeted chemical inhibitors of the activity of specific IDH1 and IDH2 point mutants have been designed and have been shown to reduce D-2-hydroxyglutarate and growth in cells and mouse models.43 Clinical trials using these inhibitors are ongoing in early stages. Another possible mechanism for inhibiting IDH1/IDH2 signaling is to deprive them of α-KG using glutaminase inhibitors as described above.44 The study of IDH inhibition is ongoing in an effort to identify patients that may benefit from these therapies and which compounds and dosages are most effective.

Targeting Lipid Metabolism

Although altered glucose and glutamine metabolism have been the primary focus of work studying the changes in metabolism of cancer cells, another aspect of cellular metabolism that is unique in proliferating and cancer cells is the oxidation and synthesis of lipids. Lipids such as fatty acids serve as an additional energy source for cells and are required for membrane synthesis during cellular growth and division. Lipids can also play roles in cellular signaling by functioning as second messengers and as hormones.45 Fatty acids, the primary building block of cellular membranes, 830 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

can be obtained from environmental sources or the cells can synthesize these molecules de novo. Most normal adult cells prefer to get fatty acids from exogenous sources, but observations in cancer cells indicate that the de novo synthesis of fatty acids is highly up-regulated in many types of tumors.46 The shift in fatty acid synthesis in tumors has been suggested as a potential target to limit cancer cell growth. One potential target for limiting lipid synthesis in tumor cells is the fatty acid synthase enzyme complex, FASN, which has also been found to be up-regulated in some breast tumors.47 Currently, several chemical inhibitors, as well as genetic ablation of FASN by RNA interference, are being tested for effectiveness in reducing tumor cell growth and proliferation.47 These studies have been extended to other enzymes involved in fatty acid synthesis as well. Other potential targets for inhibiting fatty acid synthesis are the sterol regulatory element-binding proteins (SREBPs), which are upstream regulators of lipid synthesis.47 So far, inhibitors of these molecules are in preclinical trial stages, as the investigation of potential side effects is necessary before they are given to patients.

INTERACTIONS BETWEEN METABOLISM AND EPIGENETICS

The metabolic reprogramming that occurs in cancer has far-reaching effects. In addition to altering metabolic pathways in response to nutrient uptake, metabolic changes also influence the epigenetic regulation of gene expression. Epigenetics are heritable changes in DNA that are not the result of an alteration in sequence and include histone modifications such as methylation, acetylation, and phosphorylation. These epigenetic changes can influence gene expression by enhancing or repressing the transcription of genes. Thus, epigenetic changes downstream of metabolic alterations can influence the expression levels of many genes in cancer cells, possibly giving them a survival and growth advantage. In addition, the reverse could also be true: Epigenetic alterations can influence cellular metabolism by altering the transcription of genes involved in metabolic pathways.48 These processes are tightly linked, and we will discuss several possible mechanisms for these interactions here. The methylation of DNA at CpG sites in promoters is a mechanism by which epigenetic modifications repress the expression of genes. In cancer, DNA methylation is often observed in the promoter sites of tumor suppressor genes. DNA methylation is mediated by DNA methyltransferases, and histone methylation is mediated by histone methyltransferases, both of which use an activated methyl donor from S-adenosylmethionine (SAM), a product of one-carbon metabolism. Dysregulation of carbon metabolism pathways in cancer can alter the levels of SAM and methyl donors available, thus influencing the epigenetic modifications and expression of genes in these cells.49 Another mechanism by which metabolic pathways can effect epigenetics is through the TCA cycle metabolites. Several histone demethylases require the TCA cycle protein α-KG as a cofactor for activation, and thus the levels of TCA cycle

TARGETING CANCER METABOLISM

intermediates may influence demethylase protein activity by competitive inhibition. Likewise, D-2-hydroxyglutarate, the protein made from α-KG by cancer cells with IDH1/IDH2 mutations (discussed above), inhibits the activity of α-KG– dependent demethylases. SDH and FH mutations that result in accumulation of succinate and fumarate in cancer cells can also act as competitive antagonists for inhibiting these α-KG–dependent demethylases. Thus, inhibition of demethylases in cancer cells by TCA cycle intermediates can result in hypermethylation of a variety of genes,49 contributing to the repression of tumor suppressor genes and others. Another metabolic molecule that contributes to epigenetic programming is acetyl-CoA. Acetyl-CoA fuels the TCA cycle and is involved in nearly all aspects of cellular metabolism, but is also used as a cofactor by enzymes that transfer acetyl groups, including histone acetyltransferases, which catalyze the addition of an acetyl group to histones. Histone acetylation is associated with transcriptional activation of genes. Thus, the availability of acetyl-CoA, highly influenced by cellular metabolism pathways, also plays a role in the epigenetic regulation of gene expression.48,49 These are just a few examples of ways in which metabolic alterations can influence the epigenetic regulation of gene expression. Thus, it must be considered that targeting metabolic pathways can also alter the epigenetic control of gene expression. Likewise, targeting epigenetic modification pathways also holds potential to alter gene

expression, including that of metabolism pathway enzymes. Future research investigating the links between epigenetics and metabolism will hopefully provide greater understanding of the complexity of the interactions between metabolism and chromatin dynamics in both normal and cancer cells.

CONCLUSION

Proliferating cancer cells must maintain both sufficient energy and pools of metabolic intermediates for building macromolecules needed for proliferation, including DNA, proteins, and lipids. These tasks are accomplished in most cancer cells by adapting their metabolism to be more dependent upon aerobic glycolysis and glutaminolysis. The mechanisms behind the metabolic reprogramming that takes place in most tumor cells are diverse and include oncogenic activation, the repression of tumor suppressor signaling, epigenetic modifications, and mutations in metabolic enzymes themselves. As the metabolic profiles of tumor cells distinguish them from normal cells and are critical for their growth and survival, the metabolic signaling pathways have become desirable targets for therapeutic intervention in patients with cancer. Recent work has focused on identifying inhibitors of critical metabolism pathways and shows promise in targeting metabolism to improve patient outcomes, either alone or in combination with other targeted therapies.

References 1. Warburg OP, Negelein E. Uber den stoffwechsel der carcinomzelle. Biochem Zeitschr. 1927;152:309-344.

12. Gossage L, Eisen T, Maher ER. VHL, the story of a tumour suppressor gene. Nat Rev Cancer. 2014;15:55-64.

2. Warburg O. On the origin of cancer cells. Science. 1956;123:309-314.

13. sem*nza, GL. HIF-1 mediates the Warburg effect in clear cell renal carcinoma. J Bioenerg Biomembr. 2007;39:231-234.

3. Ward PS, Thompson CB. Metabolic reprogramming: a cancer hallmark even Warburg did not anticipate. Cancer Cell. 2012;21:297-308. 4. Warburg O. On respiratory impairment in cancer cells. Science. 1956;124:269-270. 5. Martínez-Reyes I, Diebold LP, Kong H, et al. TCA cycle and mitochondrial membrane potential are necessary for diverse biological functions. Mol Cell. 2016;61:199-209. 6. Dang CV. Links between metabolism and cancer. Genes Dev. 2012;26:877-890.

14. Jonasch E, Futreal PA, Davis IJ, et al. State of the science: an update on renal cell carcinoma. Mol Cancer Res. 2012;10:859-880. 15. Chen W, Hill H, Christie A, et al. Targeting renal cell carcinoma with a HIF-2 antagonist. Nature. 2016;539:112-117. 16. Chan DA, Sutphin PD, Nguyen P, et al. Targeting GLUT1 and the Warburg effect in renal cell carcinoma by chemical synthetic lethality. Sci Transl Med. 2011;3:94ra70.

7. Pavlova, NN, Thompson, CB. The emerging hallmarks of cancer metabolism. Cell Metab. 2016;23:27-47.

17. van Der Mijn, JC, Panka, DJ, Geissler, AK, et al. Novel drugs that target the metabolic reprogramming in renal cell cancer. Cancer Metab. 2016;4:14.

8. Cantor JR, Sabatini DM, Liations A. Cancer cell metabolism: one hallmark, many faces. Cancer Discov. 2012;2:881-898.

18. Bardella, C, Pollard, PJ, Tomlinson, I. SDH mutations in cancer. Biochim Biophys Acta. 2011;1807:1432-1443.

9. Haake SM, Weyandt JD, Rathmell WK. Insights into the genetic basis of the renal cell carcinomas from The Cancer Genome Atlas. Mol Cancer Res. 2016;14:589-598.

19. Trpkov K, Hes O, Agaimy A, et al. Fumarate hydratase-deficient renal cell carcinoma is strongly correlated with fumarate hydratase mutation and hereditary leiomyomatosis and renal cell carcinoma syndrome. Am J Surg Pathol. 2016;40:865-875.

10. Linehan WM, Ricketts CJ. The metabolic basis of kidney cancer. Semin Cancer Biol. 2013;23:46-55. 11. Moore LE, Nickerson ML, Brennan P, et al. Von Hippel-Lindau (VHL) inactivation in sporadic clear cell renal cancer: associations with germline VHL polymorphisms and etiologic risk factors. PLoS Genet. 2011;7:e1002312.

20. Selak MA, Armour SM, MacKenzie ED, et al. Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-alpha prolyl hydroxylase. Cancer Cell. 2005;7:77-85. 21. Sudarshan S, Sourbier C, Kong HS, et al. Fumarate hydratase deficiency in renal cancer induces glycolytic addiction and hypoxia-inducible

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 831

WEYANDT ET AL

transcription factor 1alpha stabilization by glucose-dependent generation of reactive oxygen species. Mol Cell Biol. 2009;29:40804090. 22. King A, Selak MA, Gottlieb E. Succinate dehydrogenase and fumarate hydratase: linking mitochondrial dysfunction and cancer. Oncogene. 2006;25:4675-4682.

35. Orecchioni S, Reggiani F, Talarico G, et al. The biguanides metformin and phenformin inhibit angiogenesis, local and metastatic growth of breast cancer by targeting both neoplastic and microenvironment cells. Int J Cancer. 2015;136:E534-E544. 36. Deberardinis RJ, Chandel NS. Fundamentals of cancer metabolism. Sci Adv. 2016;2:e1600200.

23. Rathmell KW, Chen F, Creighton CJ. Genomics of chromophobe renal cell carcinoma: implications from a rare tumor for pan-cancer studies. Oncoscience. 2015;2:81-90.

37. Lukey MJ, Wilson KF, Cerione RA. Therapeutic strategies impacting cancer cell glutamine metabolism. Future Med Chem. 2013;5:16851700.

24. Davis CF, Ricketts CJ, Wang M, et al; Cancer Genome Atlas Research Network. The somatic genomic landscape of chromophobe renal cell carcinoma. Cancer Cell. 2014;26:319-330.

38. Wang JB, Erickson JW, Fuji R, et al. Targeting mitochondrial glutaminase activity inhibits oncogenic transformation. Cancer Cell. 2010;18:207219.

25. Laplante M, Sabatini DM. mTOR signaling at a glance. J Cell Sci. 2009;122:3589-3594.

39. Xiang Y, Stine ZE, Xia J, et al. Targeted inhibition of tumor-specific glutaminase diminishes cell-autonomous tumorigenesis. J Clin Invest. 2015;125:2293-2306.

26. Kennedy BK, Lamming DW. The mechanistic target of rapamycin: the grand conducTOR of metabolism and aging. Cell Metab. 2016;23:9901003. 27. Jin L, Alesi GN, Kang S. Glutaminolysis as a target for cancer therapy. Oncogene. 2015;35:3619-3625. 28. Land SC, Tee AR. Hypoxia-inducible factor 1alpha is regulated by the mammalian target of rapamycin (mTOR) via an mTOR signaling motif. J Biol Chem. 2007;282:20534-20543. 29. Weichhart T, Hengstschläger M, Linke M. Regulation of innate immune cell function by mTOR. Nat Rev Immunol. 2015;15:599-614. 30. Grabiner BC, Nardi V, Birsoy K, et al. A diverse array of cancerassociated MTOR mutations are hyperactivating and can predict rapamycin sensitivity. Cancer Discov. 2014;4:554-563. 31. Wan X, Harkavy B, Shen N, et al. Rapamycin induces feedback activation of Akt signaling through an IGF-1R-dependent mechanism. Oncogene. 2006;26:1932-1940. 32. Petrachi T, Romagnani A, Albini A, et al. Therapeutic potential of the metabolic modulator phenformin in targeting the stem cell compartment in melanoma. Oncotarget. 2016;8:6914-6928. 33. Liu Z, Ren L, Liu C, et al. Phenformin induces cell cycle change, apoptosis, and mesenchymal-epithelial transition and regulates the AMPK/mTOR/p70s6k and MAPK/ERK pathways in breast cancer cells. PLos One. 2015;10:e0131207. 34. Hur KY, Lee MS. New mechanisms of metformin action: focusing on mitochondria and the gut. J Diabetes Investig. 2015;6:600-609.

832 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

40. Momcilovic M, Bailey ST, Lee JT, et al. Targeted inhibition of EGFR and glutaminase induces metabolic crisis in EGFR mutant lung cancer. Cell Reports. 2017;18:601-610. 41. Gross MI, Demo SD, Dennison JB, et al. Antitumor activity of the glutaminase inhibitor CB-839 in triple-negative breast cancer. Mol Cancer Ther. 2014;13:890-901. 42. Dang L, Yen K, Attar EC. IDH mutations in cancer and progress toward development of targeted therapeutics. Ann Oncol. 2016;27:599-608. 43. Fujii T, Khawaja MR, DiNardo CD, et al. Targeting isocitrate dehydrogenase (IDH) in cancer. Discov Med. 2016;21:373-380. 44. Laurenti G, Tennant, DA. Isocitrate dehydrogenase (IDH), succinate dehydrogenase (SDH), fumarate hydratase (FH): three players for one phenotype in cancer? Biochem Soc Trans. 2016;44:1111-1116. 45. Santos CR, Schulze A. Lipid metabolism in cancer. FEBS J. 2012;279:2610-2623. 46. Currie E, Schulze A, Zechner R, et al. Cellular fatty acid metabolism and cancer. Cell Metab. 2013;18:153-161. 47. Röhrig F, Schulze A. The multifaceted roles of fatty acid synthesis in cancer. Nat Rev Cancer. 2016;16:732-749. 48. Lu C, Thompson CB. Metabolic regulation of epigenetics. Cell Metab. 2012;16:9-17. 49. Wong C, Qian Y, Yu J. Interplay between epigenetics and metabolism in oncogenesis: mechanisms and therapeutic approaches. Oncogene. Epub 2017 Jan 16.

VALUE-BASED MEDICINE AND INTEGRATION OF TUMOR BIOLOGY

Value-Based Medicine and Integration of Tumor Biology Gabriel A. Brooks, MD, MPH, Linda D. Bosserman, MD, Isa Mambetsariev, and Ravi Salgia, MD, PhD OVERVIEW Clinical oncology is in the midst of a genomic revolution, as molecular insights redefine our understanding of cancer biology. Greater awareness of the distinct aberrations that drive carcinogenesis is also contributing to a growing armamentarium of genomically targeted therapies. Although much work remains to better understand how to combine and sequence these therapies, improved outcomes for patients are becoming manifest. As we welcome this genomic revolution in cancer care, oncologists also must grapple with a number of practical problems. Costs of cancer care continue to grow, with targeted therapies responsible for an increasing proportion of spending. Rising costs are bringing the concept of value into sharper focus and challenging the oncology community with implementation of value-based cancer care. This article explores the ways that the genomic revolution is transforming cancer care, describes various frameworks for considering the value of genomically targeted therapies, and outlines key challenges for delivering on the promise of personalized cancer care. It highlights practical solutions for the implementation of value-based care, including investment in biomarker development and clinical trials to improve the efficacy of targeted therapy, the use of evidence-based clinical pathways, team-based care, computerized clinical decision support, and value-based payment approaches.

ancer is a heterogeneous disease with variations that extend molecularly, clinically, and therapeutically, and the complex diagram of cancer treatment is evolving exponentially as more and more treatments are developed. This presents a challenge for physicians as they attempt to decipher the proper treatment profile and timeline for each patient with the lowest burden on the patient. Value-based care is a critical component of cancer treatment and should include an emphasis on quality of care as well as patient experience. This requires coordination and communication among all physicians and facilities involved in a patient’s care to ensure that the patient is fully informed and engaged in the treatment approach across the trajectory of care. Ultimately, value-based cancer care requires the complicated tasks of balancing standardization and individualization of care, with transparency about the expected clinical and financial implications of care.

TUMOR BIOLOGY, CLINICAL IMPLICATIONS, AND REASONS FOR VALUE-BASED MEDICINE

The costs of cancer care continue to increase,1 but, on average, the costs of cancer drugs amount to only 5% to 20% of the total costs of cancer care.2 However, the average cost of some of the newer treatment options, such as combinations of checkpoint inhibitors, can cost as much as $100,000 per month.2 This cost led the Institute of Medicine to define six elements of value in cancer care: safety, effectiveness, patient-centeredness, timeliness, efficiency, and equity. ASCO developed a value framework that focuses on three

elements that are easily measured and frequently reported in clinical trials: clinical benefit (effectiveness), toxicity (safety), and cost (efficiency).3,4 These factors are vital to incorporate into care processes as oncology faces a growing and aging cancer population and increasing costs of oncology drugs—up 30% over 4 years.5 This creates an opportunity to leverage the value of genomics to arrive at personalized medicine and precision medicine that aims to cure cancers and improve quality of life without creating a financial burden for patients. The key for the success of precision medicine will be to balance the current system of organ-focused cancer classification and therapy with the new transforming model in which cancers are defined by their genetic makeup. As an example, lung cancer is not only histologically split between small cell and non–small cell lung cancer (15% and 85%, respectively) but also additionally delineated within non–small cell lung cancer into adenocarcinoma, squamous cell carcinoma, and large cell carcinoma (40%, 25% to 30%, and 5% to 10% of all lung cancer incidences, respectively).6 This heterogeneity extends further with the introduction of complex molecular profiling within each subtype of disease. For example, in adenocarcinoma, the oncogene makeup consists of KRAS (32.2%), EGFR (11.3%), BRAF (7.0%), NF1 (8.3%), MET ex14 (4.3%), ALK fusion (1.3%), ROS1 fusion (1.7%), and numerous other oncogene mutations.7 Meanwhile, in squamous cell lung carcinoma, the key candidate genes are FGFR1 (20%), SOX (20%), PIK3CA (20%), MDM2 (10%), PDGFRA (8% to 10%), MET (6%), and several other mutations.8

From the Norris Cotton Cancer Center, Dartmouth-Hitchco*ck Medical Center, Lebanon, NH; City of Hope Comprehensive Cancer Center, Duarte, CA. Disclosures of potential conflicts of interest provided by the authors are available with the online article at asco.org/edbook. Corresponding author: Ravi Salgia, MD, PhD, City of Hope Comprehensive Cancer Center, 1500 E. Duarte Rd., Duarte, CA 91010; email: [emailprotected]. © 2017 American Society of Clinical Oncology

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 833

BROOKS ET AL

This complex array in the frequency and variety of mutations within cancer subtypes is the driver behind the new era of targeted therapies that originally began with EGFR tyrosine kinase inhibitors, such as erlotinib and gefitinib. As an example of clinical benefit and efficiency, erlotinib has been shown to have a progression-free survival benefit in patients with advanced EGFR mutation–positive non–small cell lung cancer and was associated with more tolerability than standard chemotherapy for first-line treatment.9 The identification of EGFR and other driver mutations within lung cancer and other cancer types has revolutionized cancer treatment to true personalized medicine in which the genetic makeup of a tumor is analyzed by next-generation sequencing or liquid biopsy to truly individualize cancer treatment.10 In our clinics, we consistently face the need for genomics and other omics analysis as well as their pairing with clinically effective and cost-effective therapies. As an example, a 67-year-old man with a history of localized squamous cell esophageal cancer was first diagnosed and treated with chemoradiation. Six months later, during his surveillance workup, an isolated left lower lung nodule was noted on imaging. The biopsy reportedly showed squamous cell carcinoma, and he underwent a left lower lobectomy with lymph node dissection in 2003. The pathology confirmed T1N0Mx well-differentiated squamous cell carcinoma, and he was treated with adjuvant fluorouracil and carboplatin followed by radiotherapy. He did well for 7 years after treatment and had no evidence of disease on serial imaging and endoscopies until a follow-up scan showed a new left upper lung nodule suspicious for primary adenocarcinoma along with multiple, nonspecific micronodules. Surveillance continued for 3 months, when a repeat chest CT showed steady progression of size and density of the left upper lobe nodule associated with increased uptake on PET imaging. Pathology showed squamous cell carcinoma in situ in the right lower lung, but the left lung nodule displayed well-differentiated adenocarcinoma (pT2aNX). Molecular marker testing on this tissue showed EGFR and KRAS wild-type genes, but there was not enough tissue for EML4-ALK testing. The patient began treatment with five cycles of carboplatin and gemcitabine, which he tolerated well, and his dis-

KEY POINTS • Genomic, and other omic platforms, are currently utilized in oncology for certain diseases. • The value of the omic platforms is not consistent throughout practices. • As we think about value in oncology, pathways should also be considered. • The financial implications of the various platforms and decision analysis must also be taken into account. This has to reflect patient benefit. • As we move into value based medicine, we have to consider the heterogeneity of cancer and ultimately optimize therapy based on as much information available. 834 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

ease was clinically stable. Twelve months later, however, a CT scan found an increase in the size of pulmonary nodules in the right lung as well as new nodules at the left base that confirmed progression of disease. A bronchoscopy was performed and showed squamous cell carcinoma. Treatment was switched to systemic chemotherapy with carboplatin and docetaxel, but the patient developed neutropenic fever that required hospitalization after cycle 1. This prompted a 20% dose reduction for cycle 2. Tissue from the bronchoscopy was sent for molecular testing, which identified seven genomic alterations and 17 variants of unknown significance. The patient subsequently stopped chemotherapy and was monitored with follow-up CT scans for 12 months. At the next disease progression, a VeriStrat test was done; the result was VeriStrat Good, which indicated a potential benefit from EGFR inhibitor therapy. Thus, the patient received 150 mg of erlotinib daily. During erlotinib treatment, the patient developed an acneiform rash and diarrhea, which were managed supportively. A follow-up CT scan showed disease progression with increased lymphadenopathy as well as new lung and liver lesions. Erlotinib treatment was replaced with nab-pacl*taxel for two cycles, which had to be stopped for recurrent bacterial infections and disease progression. Then, photodynamic therapy was started; although the patient responded well, with better breathing and fewer symptoms, there was little disease response. The patient subsequently died within 4 months. Figure 1 summarizes the oncologic history of the patient as a timeline. This patient case is not only a good example of the clinical benefit of targeted therapies but also allows us to understand the challenges of their use without reliable biomarker testing. Although it was determined by proteomic analysis that our patient would be a reasonable candidate for EGFR inhibitor therapy, the clinical results of erlotinib treatment showed no evidence of response. This extends to other patient cases, in which the presence of an EGFR mutation and treatment with an EGFR tyrosine kinase inhibitor or other targeted therapy does not guarantee response or absence of toxicity. However, it also highlights the value in understanding the entire omic structure of lung cancer, for which it is not only the tissue molecular testing by next-generation sequencing that plays a vital role but also the liquid biopsy, which may offer options for patients with advanced non– small cell lung cancer to detect mechanisms of acquired resistance, such as T790M.11,12 As shown in this case, the next-generation sequencing was performed—with a test that costs anywhere between $5,500 and $5,800—but did not offer the patient any clinical options in terms of targeted therapy.13 However, this approach to clinically understand the efficacy and the benefit of these omic tests highlights the importance of verifying and validating all omics testing in the new paradigm of precision medicine so that testing is affordable, in the best interest of the patient, appropriate, and also widely understood and accepted across national institutions (Fig. 2). As oncology trends grow and evolve, it will be essential to reconsider the traditional clinical pathways to incorporate

VALUE-BASED MEDICINE AND INTEGRATION OF TUMOR BIOLOGY

FIGURE 1. Timeline of Oncologic History of a Patient

Abbreviations: 5-FU, 5-fluorouracil; LUL, left upper lobe; NGS, next-generation sequencing; RLL, right lower lobe; RUL, right upper lobe.

the transformation of cancer from an overarching disease into a larger number of molecularly defined diseases that have individualized therapeutic options.5 Ongoing national clinical trials, including TAPUR, Basket, QUILT, and MATCH, are excellent examples of pooling national patient access to genomically determined targeted agents and capturing response and toxicity data to understand clinical benefit. It is hoped that, by including patients with many different pathologic cancers into targeted therapy trials that are based on similar genomic mutations, we can better understand efficacy and the tumor types that may or may not respond similarly on the basis of similar driver mutations.

ACTUALIZING INDIVIDUALIZED THERAPIES WITH PATHWAYS, TOOLS, AND PROCESSES TO ACHIEVE VALUE-BASED CARE

The foundations of any treatment plan are personal health information, performance status, diagnosis, and staging

information, which are then paired with molecular data. Mutation status may include one or more targetable mutations, depending on the primary cancer site, and can be heterogeneous at diagnosis (across primary and metastatic sites). In addition, both mutation status and extent of disease can change with disease progression. The comprehensive treatment plan also is based on the setting (e.g., prevention, neoadjuvant, adjuvant, metastatic, induction, consolidation, and maintenance phases) as well as the line of each therapy, type and time of any past response, and sites of metastases. In addition, treatment plans may be modified for specific populations, such as adolescent/young adults, geriatric populations, and those with particular comorbidities or inherited gene mutations. If oncologists expect to fully integrate these numerous data points to help guide the best care for each patient, clinical systems are needed to prompt for order and collection of discrete data to offer real-time decision support and

FIGURE 2. Omics Architecture Detailing Available Methods for Patient Oncogenic Testing

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 835

BROOKS ET AL

extractable data for outcome reporting. These systems will need rapid updatability, given the frequency of new discoveries that have clinical relevance for patients. Value-based care must be delivered by innovative clinical teams that work collaboratively with payer systems to avoid the growing regulatory burden that is increasing the administrative oncology work load14 and straining a challenged oncology work force that faces increasing burnout.15 Given the explosion of molecular data to guide optimal treatment options, given the approval of some therapies only in a set sequence, and given more and more targetable mutations identified and linked to effective treatments for common and rare cancers, even doctors who specialize in one type of cancer can benefit from the collective wisdom of national experts at the point of care. Most oncologists, in fact, see patients whose diseases span the entire range of cancer diagnoses, stages, and molecular features. Having high-quality, expert decision support at the point of care can ensure appropriate molecular and other testing is done at the best time to optimize the chance of giving a patient the right therapy, including targeted therapies with supportive and palliative care, to achieve the best and most cost-effective health outcome.16,17 Pathways, thus, have become the tools to empower evidence-based cancer care plans. In addition, they have been shown to lower costs and still ensure the delivery of evidence-based care. Work published by the US Oncology group showed equivalent outcomes and significantly lower costs with their pathway program in both metastatic colon and lung cancers.18,19 A pilot study of the UnitedHealthcare episode-based payment model, with practice-chosen pathways for breast, colon, and lung cancers, showed a 34% reduction in costs compared with their fee-for-service database.20 The 3-year results (2009–2012) from an expanded pilot study in five practices reported the same 34% overall reduction in medical costs from before and after the pilot study, even with higher chemotherapy costs, because hospitalizations were markedly reduced.21 Adopting practice or group practice pathways, addressing and standardizing care processes, and using team-based care have been the pillars of groups who work with payers to

achieve value-based care, which is generally accepted as a measure of outcomes achieved per monetary expenditure. Several groups have reported various aspects of improved outcomes with these approaches.22-25 Team-based care, led by physicians, is another process to ensure caregiver teams work to the top of their license so that data are collected to empower and improve care while clinicians are allowed time to compassionately care for vulnerable patients.26,27 As pathways have proliferated, however, payers, patients, and clinicians have struggled to manage practical implementation into daily clinical practice, especially when different payers require different pathways to authorize payments. As pathways have helped payers understand complexities of care and guide their authorization processes, different payers have adopted different pathway systems and rules for coverage. A majority of cancer practices have or are adopting pathway programs, but they may have to use several in one practice to get authorizations and coverage for different patients who have the same disease types. This proliferation of payer pathways and the increase in the administrative burden led the ASCO board to empanel a Pathway Task Force in 2014. With extensive stakeholder input, the task force developed a policy statement of clinical pathways in oncology,14 followed more recently by criteria for high-quality clinical pathways in oncology, depicted in Fig. 3. The goal is for all stakeholders to have criteria to ensure the pathway program a practice chooses is developed transparently and implemented efficiently, with analytic capabilities to evaluate short- and long-term impacts. New clinical processes also are critical to ensure appropriate tumor testing at diagnosis and relapse to customize the treatment plan for a patient. Oncologists are networking with primary care physicians, interventional radiologists, and pathologists to ensure that tumors are tested accurately. Although technology has helped lower the cost of complex molecular and genetic assessments over time, batch testing, for now, is not standard, and high-cost molecular tests are often required in sequence, which requires tracking by busy clinicians. As costs continue to decline, larger molecular and genetic panels may be done at the time of either a liquid biopsy or tumor biopsy to ensure identification of any effective targetable mutations.

FIGURE 3. ASCO Criteria for High-Quality Oncology Pathway Programs

Reproduced with permission.28

836 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

VALUE-BASED MEDICINE AND INTEGRATION OF TUMOR BIOLOGY

With the major impact of improved cancer outcomes for patients from targeted therapy comes a rapid change phase in oncology to develop, study, and deploy new care processes, tools, and teams to incorporate best-practice options so that patients have access to timely and effective care. We can look forward to electronic medical record enhancements that facilitate collaboration and collection of discrete data about patients, diagnoses, and molecular mutations, as well as tracking of tests and therapies and their changes over time. Real-time decision support could provide pathway prompting for the best therapies and sequences of targeted and other treatments that consider specific patient populations, patient preferences, comorbidities, toxicities, and costs. Groups that use similar electronic medical record systems and larger big data collaborations with analytic systems, like ASCO CancerLinQ, Flatiron Health, and NantHealth, are all working to collect validated data to analyze clinical and financial outcomes partnered with patient and clinician satisfaction. These systems will guide oncologists as we continue to discover new targetable pathways, diagnostics, and therapies to achieve the triple aim outlined by the Institute of Medicine and the fourth aim, now recognized as essential for success: better health, better patient experience, lower costs, and improved clinician satisfaction.29

CANCER CARE DELIVERY AND COSTEFFECTIVENESS ANALYSIS IN THE ERA OF OMICS

One of the hoped-for benefits of precision medicine is that genomically targeted therapies will improve the value of cancer care. Cost-effectiveness analysis is an economic approach to assess the value of medical therapies, and it generally takes a societal perspective. The societal perspective is particularly relevant for policy makers, because the costs of medical care are diffused across society (e.g., in the form of higher taxes or more expensive health insurance premiums). Implicitly, opportunity costs also are experienced at the societal level: the decision to adopt higher-cost cancer therapies should translate to improved health outcomes for patients with cancer, but the opportunity cost of this decision may preclude government spending for other societal programs. The value formula used in cost-effectiveness analysis is intuitively simple; costs associated with a new therapy or technology are placed in the numerator of the value equation, and an outcome measure of effectiveness (usually quality-adjusted life-years [QALYs]) is placed in the denominator. In most cases, the key assessment is an incremental analysis to compare a new therapy against a standard-of-care comparator. The quotient of incremental cost versus incremental benefit is then expressed as the incremental costeffectiveness ratio (e.g., cost per QALY). Cost-effectiveness is only one of multiple approaches to define value; however, the simple principles of cost-effectiveness analysis provide an important starting point for any discussion of value. At least three properties that are theoretically shared by targeted therapies should nominally enhance the

cost-effectiveness of targeted agents relative to untargeted therapies. First, only a subset of the population will receive a targeted agent, (e.g., only patients with breast cancer that overexpresses HER2 should receive trastuzumab).30 In this way, patients who are unlikely to benefit from a targeted therapy are spared treatment and any accompanying toxicities. A second property of targeted agents is that they are designer drugs, which have a hypothesis-based mechanism of action to target an important oncologic process. Last, and related to the second property, targeted agents should have fewer off-target toxicities, which make them theoretically easier to tolerate. Although the promise of enhanced efficacy from targeted therapies is increasingly being realized for many cancer types, improvements in efficacy only address the numerator of the value equation. The denominator of this equation, cost, usually is not included in journal articles that report drug efficacy. Nevertheless, cost is a critical real-world determinant of access to drug therapy. In some cases, access is rationed at the systemic or societal level, as in the United Kingdom, Canada, and many other nations that directly incorporate cost-effectiveness into drug coverage decisions3,31,32; in other cases, access is rationed at the individual level, as in the United States, where high medication copays are associated with nonadherence to life-sustaining therapy.33 The case of pertuzumab in metastatic breast cancer provides context for the relative contributions of both efficacy and cost toward the overall cost-effectiveness of therapy. The CLEOPATRA study demonstrated a 15.7-month improvement in overall survival for patients with HER2-positive, metastatic breast cancer who received pertuzumab as part of first-line chemotherapy.34 This impressive efficacy result underscores the power of molecularly targeted therapy to improve patient outcomes, and very few oncologists or patients would discount the benefit of this survival improvement. However, drug cost is a key driver of cost-effectiveness, and the average sales price of pertuzumab in the third quarter of 2016 was $4,000 to $5,000 per 3-week treatment cycle.35 A peer-reviewed cost-effectiveness analysis by Durkee and colleagues36 estimated incremental costs associated with first-line pertuzumab use of $294,747 per patient and an incremental cost-effectiveness ratio of $472,668 per QALY ($206,335 per life-year).36 Although no absolute thresholds exist in U.S. health care policy, these estimates generally are accepted as poor cost-effectiveness—even in the context of highly impressive treatment efficacy. Pertuzumab is hardly an outlier in terms of pricing for targeted therapies. Another instructive case is that of necitumumab, a monoclonal antibody inhibitor of EGFR that was approved by the U.S. Food and Drug Administration (FDA) for the treatment of squamous cell lung cancer in November 2015. Because necitumumab is approved only for advanced squamous cell lung cancer, it stands to reason that a value-based price for necitumumab is best defined in that clinical setting. Unlike the case of pertuzumab, the survival benefit reported for necitumumab in advanced squamous cell lung cancer was small (overall survival, 11.5 months asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 837

BROOKS ET AL

in the necitumumab group, which was 1.6 months longer than in the control group).37 Grade 3 or greater toxicities were more common with necitumumab than with control (72% vs. 62% of patients). After the study results were reported, but before necitumumab was approved by the FDA, Goldstein and colleagues38 sought to define a value-based price for necitumumab. They calculated that necitumumab would be cost-effective at a willingness-to-pay threshold of $100,000 per QALY if priced at $563 per 3-week cycle, and that it would be cost-effective at a higher threshold of $200,000 per QALY if priced at $1,309 per cycle. Now that necitumumab has been approved and marketed, the actual cost per cycle is approximately $4,200.35 As a result of this glaring desynchrony between value and price, necitumumab was recently removed from the National Comprehensive Cancer Network list of guideline-recommended therapies for metastatic squamous cell lung cancer.39 These examples demonstrate that targeted therapies are not any more inherently cost-effective than traditional cytotoxic therapies. Cost is an essential component of the value calculation under any framework. Even highly effective, minimally toxic therapies are of low value to patients and to society at large if exorbitant costs limit or prevent access. How, then, can the value of targeted therapies be improved? We submit that strategies to improve the value of targeted therapies should be directed at both the numerator and the denominator of the value equation. To focus on the denominator of the value equation, strategies are needed to enhance efficacy and reduce toxicity of targeted therapies. Fortunately, this is an area in which cancer researchers are moving ahead full steam. A targeted therapy is only as good as the biomarker target, and strengthening the link between biomarker and efficacy is one example of a strategy to improve the value of targeted therapies. Cetuximab and panitumumab were initially approved in 2004 and 2006, respectively, for the treatment of chemotherapy-refractory colorectal cancer. After a growing body of data demonstrated that EGFR inhibitors were ineffective in roughly 50% of patients with colorectal cancer who had KRAS mutations,40,41 the FDA restricted its approval of these agents in 2009 to patients who did not have KRAS mutations. Further research has continued to narrow the population of patients with colorectal cancer who are candidates for EGFR inhibitor therapy, after the finding that NRAS mutations42 and right-sided tumors43,44 also are predictors of poor efficacy from EGFR inhibitor therapy. With every step to narrow the population of patients with colorectal cancer who are eligible to receive EGFR inhibitor therapy, the average efficacy among treated patients should improve alongside the value of the drug. Much more work is needed to improve biomarkers for other molecularly targeted therapies; however, we are optimistic that the medical and scientific communities are up to this task. Changing the numerator of the value equation for targeted therapies is a more complicated task. There is increasing anecdotal evidence that patients,45 physicians,46 and society at large will not tolerate a continuation of current trends in 838 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

drug pricing, particularly in the United States, where drug costs are highest. Whether these sentiments will translate to legislative action to regulate drug prices is uncertain, although the current U.S. president and prominent members of Congress recently have discussed or proposed legislation to regulate drug prices. Alternative approaches for bringing drug prices into alignment with societal resources include value-based payment schemes, such as indication-specific pricing or performance-based pricing. Indication-specific pricing is an approach that is particularly relevant for targeted therapies, which often hold many distinct indications, both approved and off label.47 For example, trastuzumab has accepted indications that include the treatment of early-stage breast cancer, metastatic breast cancer, salivary duct cancer, gastroesophageal cancer, and lung cancer. The value of trastuzumab therapy, as measured in a cost-effectiveness framework, varies greatly across these indications. However, drug payments for trastuzumab are the same regardless of indication. In indication-specific pricing, the payment for a specified drug is allowed to vary by indication. Pharmaceutical companies could garner higher payments in settings in which the drug has a high proven efficacy, such as adjuvant trastuzumab therapy for HER2positive breast cancer (34% reduction in the risk of death at 2 years after completion of therapy).48 In situations in which trastuzumab has a lower efficacy, such as metastatic esophagogastric cancer (2.7-month improvement in median overall survival),49 a lower price would be required to maintain cost-effectiveness. The incentive structure of this system would encourage pharmaceutical companies to develop drugs in a way that maximizes efficacy for distinct targeted populations, rather than a way that seeks the lowest efficacy threshold achievable in the largest possible patient population. Indication-specific pricing and other forms of value-based payment would require specific reimbursem*nt system changes to permit implementation47 and likely would require a governmental mandate. Legislative price controls are considered politically challenging in the United States, although implementation of controls in developed countries around the world attests to their widespread acceptability. Nevertheless, finding ways to restrain uncontrolled growth in the prices of cancer therapies is imperative at the societal level to maintain access to treatment for patients. The efforts of ASCO and other societies to call attention to the cost and value of cancer therapies are an important step toward moderating drug costs and delivering on the promise of high-value, highly targeted cancer therapies.2-4,50

CONCLUSION

It is exhilarating to practice oncology with the current and rapidly expanding ability to deploy effective and welltolerated molecularly targeted therapies to prevent, control, and sometimes cure malignant diseases. Challenges remain, however, in applying principles of value-based care in this rapidly evolving landscape. These challenges are compounded by the fragmentation of classic diagnostic categories into a

VALUE-BASED MEDICINE AND INTEGRATION OF TUMOR BIOLOGY

greater number of molecularly defined disease entities. In clinical research, key questions include which and how many molecular targets are needed, and whether they are needed alone or in combination. In clinical practice, best practices are emerging for the development and deployment of the integrated tools, teams, and processes to provide value-based care. At the societal level, increasing costs associated with cancer treatment threaten access to therapies, and strategies are needed to ensure that costs are commensurate with benefits at the individual and societal levels. To meet these challenges and achieve true value-based care will require the support of government, industry, and

health systems for continued research and development of effective therapies. We need investments in real-time decision support, expanded access to clinical trials, and new payment and team-based care models. The scientific advancements that have enabled the genomic revolution are truly remarkable; however, the promise of this new era cannot be fully realized for our patients until the implementation of value-based care is accomplished.

ACKNOWLEDGMENT

G. A. Brooks and L. D. Bosserman contributed equally to this article.

References 1. Mariotto AB, Yabroff KR, Shao Y, et al. Projections of the cost of cancer care in the United States: 2010-2020. J Natl Cancer Inst. 2011;103:117128.

14. Zon RT, Frame JN, Neuss MN, et al. American Society of Clinical Oncology policy statement on clinical pathways in oncology. J Oncol Pract. 2016;12:261-266.

2. Young RC. Value-based cancer care. N Engl J Med. 2015;373:25932595.

15. American Society of Clinical Oncology. The state of cancer care in America, 2015: a report by the American Society of Clinical Oncology. J Oncol Pract. 2015;11:79-113.

3. Schnipper LE, Davidson NE, Wollins DS, et al; American Society of Clinical Oncology. American Society of Clinical Oncology statement: a conceptual framework to assess the value of cancer treatment options. J Clin Oncol. 2015;33:2563-2577.

16. Ellis PG, Brufsky AM, Beriwal S, et al. Pathways clinical decision support for appropriate use of key biomarkers. J Oncol Pract. 2016;12:e681-e687.

4. Schnipper LE, Davidson NE, Wollins DS, et al. Updating the American Society of Clinical Oncology value framework: revisions and reflections in response to comments received. J Clin Oncol. 2016;34:2925-2934.

17. Bach PB, Saltz LB, Wittes RE. In cancer care, cost matters. The New York Times. October 14, 2012. http://www.nytimes.com/2012/10/15/ opinion/a-hospital-says-no-to-an-11000-a-month-cancer-drug.html. Accessed February 7, 2017.

5. Aitken M BL, Mawrie R. Development in Cancer Treatments, Market Dynamics, Patient Access and Value. Global Oncology Trend Report 2015. Parsippany, NJ: IMS Institute For Healthcare Informatics; 2015.

18. Neubauer MA, Hoverman JR, Kolodziej M, et al. Cost effectiveness of evidence-based treatment guidelines for the treatment of non–small cell lung cancer in the community setting. J Oncol Pract. 2010;6:12-18.

6. Zappa C, Mousa SA. Non–small cell lung cancer: current treatment and future advances. Transl Lung Cancer Res. 2016;5:288-300.

19. Hoverman JR, Cartwright TH, Patt DA, et al. Pathways, outcomes, and costs in colon cancer: retrospective evaluations in two distinct databases. J Oncol Pract. 2011; 7:52s-59s.

7. Cancer Genome Atlas Research Network. Comprehensive molecular profiling of lung adenocarcinoma. Nature. 2014;511:543-550. 8. Heist RS, Sequist LV, Engelman JA. Genetic changes in squamous cell lung cancer: a review. J Thorac Oncol. 2012;7:924-933. 9. Zhou C, Wu YL, Chen G, et al. Erlotinib versus chemotherapy as firstline treatment for patients with advanced EGFR mutation-positive non–small cell lung cancer (OPTIMAL, CTONG-0802): a multicentre, open-label, randomised, phase 3 study. Lancet Oncol. 2011;12:735742. 10. Khoo C, Rogers TM, Fellowes A, et al. Molecular methods for somatic mutation testing in lung adenocarcinoma: EGFR and beyond. Transl Lung Cancer Res. 2015;4:126-141. 11. Yu HA, Riely GJ. Second-generation epidermal growth factor receptor tyrosine kinase inhibitors in lung cancers. J Natl Compr Canc Netw. 2013;11:161-169.

20. Newcomer LN. UnitedHealthcare’s episode-based payment model program cuts cost. Am Health Drug Benefits. 2015;8:11-12. 21. Newcomer LN, Gould B, Page RD, et al. Changing physician incentives for affordable, quality cancer care: results of an episode payment model. J Oncol Pract. 2014;10:322-326. 22. Hoverman JR, Klein I, Harrison DW, et al. Opening the black box: the impact of an oncology management program consisting of level I pathways and an outbound nurse call system. J Oncol Pract. 2014;10:63-67. 23. Bosserman LD, Verrilli D, McNatt W. Partnering with a payer to develop a value-based medical home pilot: a West Coast practice’s experience. J Oncol Pract. 2012; 8:38s-40s. 24. JD S. Oncology patient-centered medical home and accountable cancer care. Community Oncol. 2010;7:565-572.

12. Stasi I, Cappuzzo F. Second generation tyrosine kinase inhibitors for the treatment of metastatic non–small cell lung cancer. Transl Respir Med. 2014;2:2.

25. Balogh EP, Bach PB, Eisenberg PD, et al. Practice-changing strategies to deliver affordable, high-quality cancer care: summary of an Institute of Medicine workshop. J Oncol Pract. 2013;9:54s-59s.

13. Goldstein DA, Shaib WL, Flowers CR. Costs and effectiveness of genomic testing in the management of colorectal cancer. Oncology (Williston Park). 2015;29:175-183.

26. Krzyzanowska MK, Blayney DW, Bosserman LD, et al. Models that work: incorporating quality principles in different clinical settings. J Oncol Pract. 2013;9:135-137.

asco.org/edbook | 2017 ASCO EDUCATIONAL BOOK 839

BROOKS ET AL

27. Zhang B, Wright AA, Huskamp HA, et al. Health care costs in the last week of life: associations with end-of-life conversations. Arch Intern Med. 2009;169:480-488. 28. Zon RT, Edge SB, Page RD, et al. American Society of Clinical Oncology Criteria for High-Quality Clinical Pathways in Oncology. J Oncol Pract. 2017;13:207-210. 29. Bodenheimer T, Sinsky C. From triple to quadruple aim: care of the patient requires care of the provider. Ann Fam Med. 2014;12:573-576. 30. Schnipper LE, Lyman GH, Blayney DW, et al. American Society of Clinical Oncology 2013 top five list in oncology. J Clin Oncol. 2013;31: 4362-4370. 31. Raftery J.NICE and the challenge of cancer drugs. Br Med J. 2009;338:b67. 32. Clement FM, Harris A, Li JJ, et al. Using effectiveness and costeffectiveness to make drug coverage decisions: a comparison of Britain, Australia, and Canada. JAMA. 2009;302:1437-1443.

https://www.nccn.org/professionals/physician_gls/pdf/nscl.pdf. Accessed February 7, 2017. 40. Lièvre A, Bachet J-B, Boige V, et al. KRAS mutations as an independent prognostic factor in patients with advanced colorectal cancer treated with cetuximab. J Clin Oncol. 2008;26:374-379. 41. Bokemeyer C, Bondarenko I, Hartmann JT, et al. KRAS status and efficacy of first-line treatment of patients with metastatic colorectal cancer (mCRC) with FOLFOX with or without cetuximab: the OPUS experience. J Clin Oncol. 2008;26 (suppl; abstr 4000) 42. Van Cutsem E, Lenz H-J, Köhne C-H, et al. Fluorouracil, leucovorin, and irinotecan plus cetuximab treatment and RAS mutations in colorectal cancer. J Clin Oncol. 2015;33:692-700. 43. Venook A, Niedzwiecki D, Innocenti F, et al. Impact of primary tumor location on overall survival (OS) and progression-free survival (PFS) in patients (pts) with metastatic colorectal cancer (mCRC): Analysis of CALGB/SWOG 80405 (Alliance). J Clin Oncol. 2016;34 (suppl; abstr 3504).

33. Dusetzina SB, Winn AN, Abel GA, et al. Cost sharing and adherence to tyrosine kinase inhibitors for patients with chronic myeloid leukemia. J Clin Oncol. 2014;32:306-311.

44. Tejpar S, Stintzing S, Ciardiello F, et al. Prognostic and predictive relevance of primary tumor location in patients with ras wild-type metastatic colorectal cancer: retrospective analyses of the CRYSTAL and FIRE-3 trials. JAMA Oncol. Epub 2008 Oct 10.

34. Swain SM, Baselga J, Kim S-B, et al; CLEOPATRA Study Group. Pertuzumab, trastuzumab, and docetaxel in HER2-positive metastatic breast cancer. N Engl J Med. 2015;372:724-734.

45. Tefferi A, Kantarjian H, Rajkumar SV, et al. In support of a patientdriven initiative and petition to lower the high price of cancer drugs. Mayo Clin Proc. 2015;90:996-1000.

35. Center for Medicare & Medicaid Services. Medicare Part B drug average sales price (ASP): 2017 ASP drug pricing files. https://www. cms.gov/Medicare/Medicare-Fee-for-Service-Part-B-Drugs/McrPartB DrugAvgSalesPrice/2017ASPFiles.html. Accessed February 7, 2017.

46. Experts in Chronic Myeloid Leukemia. The price of drugs for chronic myeloid leukemia (CML) is a reflection of the unsustainable prices of cancer drugs: from the perspective of a large group of CML experts. Blood. 2013;121:4439-4442.

36. Durkee BY, Qian Y, Pollom EL, et al. Cost-effectiveness of pertuzumab in human epidermal growth factor receptor 2-positive metastatic breast cancer. J Clin Oncol. 2016;34:902-909.

47. Bach PB. Indication-specific pricing for cancer drugs. JAMA. 2014;312:1629-1630.

37. Thatcher N, Hirsch FR, Luft AV, et al; SQUIRE Investigators. Necitumumab plus gemcitabine and cisplatin versus gemcitabine and cisplatin alone as first-line therapy in patients with stage IV squamous non–small cell lung cancer (SQUIRE): an open-label, randomised, controlled phase 3 trial. Lancet Oncol. 2015;16:763-774. 38. Goldstein DA, Chen Q, Ayer T, et al. Necitumumab in metastatic squamous cell lung cancer: establishing a value-based cost. JAMA Oncol. 2015;1:1293-1300. 39. National Comprehensive Cancer Network. NCCN clinical practice guidelines in oncology: non–small cell lung cancer (version 4.2017).

840 2017 ASCO EDUCATIONAL BOOK | asco.org/edbook

48. Smith I, Procter M, Gelber RD, et al; HERA study team. Two-year followup of trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer: a randomised controlled trial. Lancet. 2007;369:29-36. 49. Bang YJ, Van Cutsem E, Feyereislova A, et al; Toga Trial Investigators. Trastuzumab in combination with chemotherapy versus chemotherapy alone for treatment of HER2-positive advanced gastric or gastrooesophageal junction cancer (ToGA): a phase 3, open-label, randomised controlled trial. Lancet. 2010;376:687-697. 50. Meropol NJ, Schrag D, Smith TJ, et al; American Society of Clinical Oncology. American Society of Clinical Oncology guidance statement: the cost of cancer care. J Clin Oncol. 2009;27:3868-3874.

Author Index Adelson, Kerin Afaneh, Khalid F. Agarwal, Neeraj Al-Sukhun, Sana Aokage, Keiju Aragon-Ching, Jeanny B. Arreola-Ornelas, Hector Asamura, Hisao Atallah, Ehab Atreya, Chloe E. Barr, Paul M. Basch, Ethan Beatty, Gregory L. Beaver, Julia A. Bedard, Philippe L. Bejar, Rafael Beltran, Himisha Bennett, Mike Benson, Al B. III Bernstam, Elmer V. Berry, Donna L. Bhadelia, Afsan Blank, Stephanie V. Borrello, Ivan Bosserman, Linda D. Brant, Jeannine M. Brentjens, Renier J. Brewer, Molly A. Briganti, Alberto Brooks, Gabriel A. Bustoros, Mark Cairo, Jamie Canin, Beverly Carvajal, Richard D. Catoe, Heath Chao, Samuel Chapman, Andrew E. Chapman, Paul Chaudhary, Rekha Chiang, Anne C. Chu, Edward Chuk, Meredith K. Cohen, Adam D. Cohen, Jonathon B. Cohn, Susan L. Conway, Patrick H. Cooke, Kelly J. Cordeiro, Peter G. Coughlin, Steven S. Cox, Suzanne M. Crippa, Stefano Csik, Valerie Curtin, John P. Dagogo-Jack, Ibiayi Dale, William Dalurzo, Mercedes Liliana de Bock, G. H. De Mello, Ramon Andrade DeMichele, Angela Denduluri, Neelima

.....................................................................................................................................................................

............................................................................................................................................................

................................................................................................................................................................

...............................................................................................................................................................

........................................................................................................................................................................

..................................................................................................................................

....................................................................................................................................

.................................................................................................................................................................

..........................................................................................................................................................................

...............................................................................................................................................................

.........................................................................................................................................................................

...........................................................................................................................................................................

........................................................................................................................................................

..................................................................................................................................................................

.....................................................................................................................................................

............................................................................................................................................................................

..............................................................................................................................................................

......................................................................................................................................................................

................................................................................................................................................................

....................................................................................................................................................

.................................................................................................................................................................

..................................................................................................................................................................

460 480 319 395 426 330 416 426 468 246 535 460 267 216 106 480 358 705 232 450 695 416 23 561 833 416 193 435 370 833 548 40 383 641 29 171 383 661 175 155 246 139 561 512 746 460 788 93 128 746 301 383 23 619 370 403 124 261 106 57

.......................................................................................................................................................

.........................................................................................................................................................................

................................................................................................................................................

.....................................................................................................................................................

...................................................................................................................................................

............................................................................................................................................................

..............................................................................................................................................................

........................................................................................................................................................

..................................................................................................................................................................

.................................................................................................................................................................................

......................................................................................................................................................................

...................................................................................................................................................

..............................................................................................................................................................................

........................................................................................................................................................................

..............................................................................................................................................

...................................................................................................................................................................

........................................................................................................................................................

................................................................................................................................................................

...........................................................................................................................................................................

........................................................................................................................................................

..............................................................................................................................................................

...................................................................................................................................................

....................................................................................................................................................................

.....................................................................................................................................................

.....................................................................................................................................................................

............................................................................................................................................................

....................................................................................................................................................

.............................................................................................................................................................

..................................................................................................................................................................

.............................................................................................................................................................................

........................................................................................................................................................................

.....................................................................................................................................................

.........................................................................................................................................................................

.............................................................................................................................

.....................................................................................................................................................................

.........................................................................................................................

......................................................................................................................................................

........................................................................................................................................................

Dent, Rebecca Detappe, Alexandre Devita, Marsha Dhodapkar, Madhav V. Dicker, Adam P. Dienstmann, Rodrigo DiNardo, Courtney D. Dispenzieri, Angela Duncavage, Eric J. Eghbali, Shabnam Eiber, Matthias El-Khoueiry, Anthony Ellis, Peter Eniu, Alexandru E. Epstein, Ronald M. Ersek, Jennifer L. Eves, Neil D. Falconi, Massimo Ficke, Deanna Flaherty, Keith T. Flowers, Christopher R. Ford-Pierce, Shaunta Freeman-Daily, Janet Frey, Melissa K. Friedlander, Terence W. Fuld Nasso, Shelley Fuller, Christine E. Gadgeel, Shirish M. Gainor, Justin F. Ganz, Patricia A. Garralda, Elena Gaspar, Laurie E. Gelmon, Karen A. Gentile, Danielle Ghobrial, Irene M. Giaccia, Amato J. Giles, Francis Gill, David M. Ginsburg, Ophira Giralt, Sergio Gnant, Michael Goetzke, Katrina Goldberg, Kirsten B. Gong, Jun Gospodarowicz, Mary Graham, David L. Graham, John Gray, Jhanelle E. Grohar, Patrick J. Grossman, Robert Gschwend, J¨urgen E. Gupta, Sudeep Haider, Mintallah Hauschild, Axel Haykowsky, Mark J. Heath, Allison Hehlmann, R¨udiger Henderson, Tara O. Henry, N. Lynn Hirsch, Fred R.

65 548 460 561 144 210, 232 495 575 480 267 344 311 155 409 771 597 684 301 40 222 18 40 597 23 358 35 753 630 607 674 210 45 76 782 548 825 139 319 29, 395 575 116 40 216 337 395 782 344 619 725 746 344 29 480 641 57 746 468 736 106 403

.........................................................................................................................................................................

..................................................................................................................................................

...................................................................................................................................................................

......................................................................................................................................

...............................................................................................................................................................

............................................................................................................................

............................................................................................................................................

.....................................................................................................................................................

.........................................................................................................................................................

..................................................................................................................................................................

..............................................................................................................................................

...................................................................................................................................................................................

.......................................................................................................................................................

.....................................................................................................................................................

............................................................................................................................................................

............................................................................................................................................................................

...........................................................................................................................................................

...........................................................................................................................................................................

............................................................................................................................................................

.........................................................................................................................................