Vol. 21, No. 3, July 2014 - Moffitt Cancer Center

rg hip /c s cj u rv ey

O YO C PI U s o OU NI R at ur O M re NT N of ad S! fit e t.o rs es

Ac c

MOFFITT.org/ccj

Vol. 21, No. 3, July 2014

H. LEE MOFFITT CANCER CENTER & RESEARCH INSTITUTE, AN NCI COMPREHENSIVE CANCER CENTER

Risks and Benefits of Phase 1 Clinical Trial Participation Amit Mahipal, MD, and Danny Nguyen, MD

Phase 1 Trial Design: Is 3 + 3 the Best? Aaron R. Hansen, MBBS, Donna M. Graham, MBBCh, Gregory R. Pond, PhD, and Lillian L. Siu, MD

Participation of the Elderly Population in Clinical Trials: Barriers and Solutions Aaron C. Denson, MD, and Amit Mahipal, MD

Studying Cancer Treatment in the Elderly Patient Population Lodovico Balducci, MD

BRAF Mutations: Signaling, Epidemiology, and Clinical Experience in Multiple Malignancies Richard D. Hall, MD, and Ragini R. Kudchadkar, MD

PD-1 Pathway Inhibitors: Changing the Landscape of Cancer Immunotherapy Dawn E. Dolan, PharmD, and Shilpa Gupta, MD

Cancer Control is included in Index Medicus/MEDLINE

H. Lee Moffitt Cancer Center and Research Institute

H LEE MOFFITT CANCER CENTER & RESEARCH INSTITUTE INC 12902 Magnolia Drive Tampa FL 33612-9416

PAID NON-PROFIT ORG U.S. POSTAGE

3 RD A N N U A L S TAT E - O F -T H E - A R T

Neuro-Oncology CONFERENCE

SAVE THE DATE

March 19-20,2015 Sheraton Sand Key Clearwater Beach, Florida

CONFERENCE HIGHLIGHTS • Renowned faculty representing the country’s leading institutions • Updates on brain and spine tumors • Preconference dinner presentation on March 19 • Interactive case presentations with discussion • Nurse track session • Call for abstracts

COURSE DIRECTORS Peter Forsyth, MD • Frank Vrionis, MD, MPH, PhD CONFERENCE CONTACT [email protected] • MOFFITT.org/NeuroOncology2015

SPONSORED BY

Editorial Board Members Editor: Lodovico Balducci, MD Senior Member Program Leader, Senior Adult Oncology Program Moffitt Cancer Center

Deputy Editor: Julio M. Pow-Sang, MD Senior Member Chair, Department of Genitourinary Oncology Director of Moffitt Robotics Program Moffitt Cancer Center

Editor Emeritus: John Horton, MB, ChB Professor Emeritus of Medicine & Oncology

Moffitt Cancer Center Journal Advisory Committee: Aliyah Baluch, MD Assistant Member Infectious Diseases Matthew C. Biagioli, MD Assistant Member Radiation Oncology Dung-Tsa Chen, PhD Associate Member Biostatistics

Bela Kis, MD, PhD Assistant Member Diagnostic Radiology Rami Komrokji, MD Associate Member Malignant Hematology Conor C. Lynch, PhD Assistant Member Tumor Biology Amit Mahipal, MD, MPH Assistant Member Clinical Research Unit Gastrointestinal Oncology Kristen J. Otto, MD Assistant Member Head & Neck Oncology Michael A. Poch, MD Assistant Member Genitourinary Oncology Jeffery S. Russell, MD, PhD Assistant Member Endocrine Tumor Oncology Elizabeth M. Sagatys, MD Assistant Member Pathology - Clinical Jose E. Sarria, MD Assistant Member Anesthesiology Saïd M. Sebti, PhD Senior Member Drug Discovery

Hey Sook Chon, MD Assistant Member Gynecological Oncology

Bijal D. Shah, MD Assistant Member Malignant Hematology

Jasreman Dhillon, MD Assistant Member Pathology - Anatomic

Lubomir Sokol, MD, PhD Associate Member Hematology/Oncology

Jennifer S. Drukteinis, MD Associate Member Diagnostic Radiology

Hatem H. Soliman, MD Assistant Member Breast Oncology

Timothy J. George, PharmD Pharmacy Residency Director Clinical Pharmacist - Malignant Hematology

Jonathan R. Strosberg, MD Assistant Member Gastrointestinal Oncology

Clement K. Gwede, PhD Associate Member Health Outcomes & Behavior

Sarah W. Thirlwell, RN Nurse Director Supportive Care Medicine Program

Sarah E. Hoffe, MD Associate Member Radiation Oncology

Eric M. Toloza, MD, PhD Assistant Member Thoracic Oncology

John V. Kiluk, MD Associate Member Breast Oncology

Nam D. Tran, MD Assistant Member Neuro-Oncology

Richard D. Kim, MD Assistant Member Gastrointestinal Oncology

Jonathan S. Zager, MD Associate Member Sarcoma and Cutaneous Oncology

Production Staff: Veronica Nemeth Editorial Coordinator Don Buchanan Graphic Designer

Consultants: Sherri Damlo Medical Copy Editor Michele R. Sassi Art Coordinator

Associate Editor of Educational Projects John M. York, PharmD Akita Biomedical Consulting 1111 Bailey Drive Paso Robles, CA 93446 Phone: 805-238-2485 Fax: 949-203-6115 E-mail: [email protected]

For Consumer and General Advertising Information: Veronica Nemeth Editorial Coordinator Cancer Control: Journal of the Moffitt Cancer Center 12902 Magnolia Drive – MBC-JRNL Tampa, FL 33612 Phone: 813-745-1348 Fax: 813-449-8680 E-mail: [email protected]

Cancer Control is a member of the Medscape Publishers’ Circle®, an alliance of leading medical publishers whose content is featured on Medscape (www.medscape.com). Most issues and supplements of Cancer Control are available at MOFFITT.org/ccj

CANCER CONTROL: JOURNAL OF THE MOFFITT CANCER CENTER (ISSN 1073-2748) is published by H. Lee Moffitt Cancer Center & Research Institute, 12902 Magnolia Drive, Tampa, FL 33612. Telephone: 813-745-1348. Fax: 813-449-8680. E-mail: [email protected]. Internet address: MOFFITT.org/ccj. Cancer Control is included in Index Medicus ®/MEDLINE ® and EMBASE ®/ Excerpta Medica, Thomson Reuters Science Citation Index Expanded (SciSearch®) and Journal Citation Reports/Science Edition. Send subscription requests to the publisher. Single copy: $10 US, $15 Canada and foreign. Subscription rates: nonqualified individuals $60 per year US, $75 per year outside US, institutions/libraries $120 per year US ($135 per year foreign). Send change of address to the publisher with old address label and new address. Publisher is not responsible for undelivered copies. Copyright 2014 by H. Lee Moffitt Cancer Center & Research Institute. All rights reserved. Printed on acid-free paper. Cancer Control: Journal of the Moffitt Cancer Center is a peer-reviewed journal that is published to enhance the knowledge needed by professionals in oncology to help them minimize the impact of human malignancy. Each issue emphasizes a specific theme relating to the detection or management of cancer. The objectives of Cancer Control are to define the current state of cancer care, to integrate recently generated information with historical practice patterns, and to enlighten readers through critical reviews, commentaries, and analyses of recent research studies. DISCLAIMER: All articles published in this journal, including editorials and letters, represent the opinions of the author(s) and do not necessarily reflect the opinions of the editorial board, the H. Lee Moffitt Cancer Center & Research Institute, Inc, or the institutions with which the authors are affiliated unless clearly specified. The reader is advised to independently verify the effectiveness of all methods of treatment and the accuracy of all drug names, dosages, and schedules. Dosages and methods of administration of pharmaceutical products may not be those listed in the package insert and solely reflect the experience of the author(s) and/or clinical investigator(s).

July 2014, Vol. 21, No. 3

Cancer Control 185

Table of Contents Editorial Clinical Trials and Drug Development

188

Amit Mahipal, MD

Articles Risks and Benefits of Phase 1 Clinical Trial Participation

193

Amit Mahipal, MD, and Danny Nguyen, MD

Phase 1 Trial Design: Is 3 + 3 the Best?

200

Aaron R. Hansen, MBBS, Donna M. Graham, MBBCh, Gregory R. Pond, PhD, and Lillian L. Siu, MD

Participation of the Elderly Population in Clinical Trials: Barriers and Solutions

209

Aaron C. Denson, MD, and Amit Mahipal, MD

Studying Cancer Treatment in the Elderly Patient Population

215

Lodovico Balducci, MD

BRAF Mutations: Signaling, Epidemiology, and Clinical Experience in Multiple Malignancies

221

Richard D. Hall, MD, and Ragini R. Kudchadkar, MD

PD-1 Pathway Inhibitors: Changing the Landscape of Cancer Immunotherapy

231

Dawn E. Dolan, PharmD, and Shilpa Gupta, MD

186 Cancer Control

July 2014, Vol. 21, No. 3

Table of Contents Departments Ten Best Readings Relating to Clinical and Translational Research

238

Translational Medicine Simplified: AKT Goes Cycling

239

Kiran N. Mahajan, PhD, and Nupam P. Mahajan, PhD

Special Report: Novel Pancreatic Cancer Vaccines Could Unleash the Army Within

242

Gregory M. Springett, MD, PhD

Pathology Report: Histopathological and Immunophenotypical Features of Intestinal-Type Adenocarcinoma of the Gallbladder and its Precursors

247

Yan You, MD, Katherine Bui, Marilyn M. Bui, MD, PhD, Mokenge Malafa, MD, and Domenico Coppola, MD

Case Report: Follicular Lymphoma With Progression to Diffuse Large B-Cell Lymphoma and Concurrent CD5-Negative Mantle Cell Lymphoma-3 Entities in a Lymph Node

251

Janese A. Trimaldi, MD, Jeremy W. Bowers, MD, Celeste Bello, MD, and Elizabeth M. Sagatys, MD

About the art in this issue: Henry Domke, MD, is an artist who creates nature art for hospitals. He took a nontraditional path to get where he is today. While still in high school, Domke enrolled in art school but decided to pursue medicine instead, and he worked as a family physician in Jefferson City, Missouri, for 25 years. By the mid-1990s, his enduring passion for art led him to formalize his art training. He studied independently with Annette Weintraub, an art professor at City College in New York, and William Hawk, a painting professor at the University of Missouri-Columbia. Hawk convinced him to enter the university’s graduate arts program, where he studied from 1998 to 2001 while he temporarily reduced his hours at his medical practice. Another defining factor in Domke’s life is the Prairie Garden Trust, a 600-acre property that he has lived on since 1981. It is a nature restoration project begun by his parents in central Missouri, and much of his art is based on images taken there. Recently he has been expanding beyond his backyard to find inspiration and has photographs from coast to coast. In addition to local wildflowers and wildlife, his collection includes coastal landscapes, mountainous terrains, desert scenes, and sea shells. His life had come full circle and he said goodbye to medicine in 2007 and pursued art full time. Today he tries to find beauty in the ordinary and create images that inspire and heal. For further information or to contact the artist, visit www.henrydomke.com or www.healthcarefineart.com. Cover: Cherokee View_0376K. Page186: Copper Iris_0328. Nautilus Spiral_4592. Cattails_0269. Daffodil_9767. Cut Nautilus Shell_4499. Toohey Lake Panorama_0104K (detail). July 2014, Vol. 21, No. 3

Cancer Control 187

Editorial

Clinical Trials and Drug Development Clinical trials are the principal means through which novel therapeutic agents are approved in the field of oncology. Clinical research on human volunteers is allowed for the benefit of society; at the same time, we must safeguard individual rights and interests. The goals of clinical trials do not always align with the primary reasons of volunteer participation in clinical trials. This is especially problematic in early-phase clinical trials in which the primary aim generally is to find the dose acceptable for future clinical trials and evaluate toxicities. A participant may enter such a clinical trial in the hopes that treatment will provide a meaningful clinical benefit. Despite an increased awareness, fewer than 1% of patients with cancer participate in clinical trials within the United States. Low accrual to clinical trials hampers the development of novel drugs targeted at potentially lethal malignancies, thus questioning the generalizability of such trial results. Much uncertainty exists among health care professionals with regard to referring patients for a phase 1 clinical trial. Frequently, by the time patients are referred to a phase 1 study, their performance status has declined so much so that it precludes them from enrolling in a clinical trial. In this issue of Cancer Control, Drs Mahipal and Nguyen discuss the harms and benefits of participating in a phase 1 clinical trial. The response rates for patients participating in such clinical trials are approximately 10%, and one-third of the patients achieve stable disease. The clinical benefit is similar to later lines of therapies for many malignancies. Increased risk of death (< 1%) and increased adverse events are seen in early-phase clinical trials compared with phase 3 trials. Phase 1 clinical trials should be considered as a therapeutic option earlier in the treatment plan rather than as a last resort when the options are hospice care or clinical trial participation. By the nature of the design of clinical trials, most of the patients participating in a phase 1 clinical trial will receive a suboptimal dose of a medication, being either “too high” or “too low.” Such a dose results in either subtherapeutic dosing in which patients may not derive benefit (even if the drug is effective treatment for their disease type) or they are unable to tolerate the drug secondary to adverse events. Traditional 3 + 3 design is the most common dose-escalation scheme employed and has been utilized in more than 95% of phase 1 clinical trials. Dr Hansen and colleagues review the literature on the drawbacks and advantages of traditional and novel, pharmaco188 Cancer Control

kinetically guided, and model-based dose-escalation designs. A paucity of patient-based data demonstrates whether one dose escalation scheme is superior to another; however, computer simulations suggest that the 3 + 3 design fails to correctly estimate the maximum tolerated dose in the majority of trials. The primary concern is that the 3 + 3 design is conservative; therefore, many patients are likely treated at subtherapeutic doses. Novel designs may help reach the maximum tolerated dose with fewer patients, but such designs are more complicated and require extensive statistical input that may not be available at every institute. Traditionally, conservative designs may be best suited for clinical trials with drugs that have a narrow therapeutic index. Drs Denson and Mahipal address the issue of clinical trial participation among patients who are elderly. Although many clinical trials do not exclude patients based on age, elderly patients are frequently not considered for enrollment in a clinical trial. Valid concern exists for some patients secondary to comorbidities and performance status. However, much misconception exists among oncologists, patients and their family members, and researchers who design clinical trials. Many of the patients with cancer are elderly; therefore, limiting their participation in clinical trials prevents us from making valid assumptions about the applicability of study results to the general population. Some of the solutions proposed to increase trial enrollment among the elderly include educating patients and physicians, designing trials specific for this population, and providing logistic support and effective communication. As a follow-up to clinical trial participation among the elderly, Dr Balducci discusses the challenges of cancer care in this population. The age cut-off for elderly patients is unclear, with various studies using a threshold somewhere between 60 and 75 years of age. Physiological and chronological age may differ; therefore, although an assessment of physiological age is complicated, doing so may help guide the most appropriate treatment plan for a patient. Validated tools, such as the Comprehensive Geriatric Assessment, have been developed but are frequently underutilized. Future clinical trials may incorporate instruments such as ePrognosis (University of California, San Francisco, CA) to estimate life expectancy and the Chemotherapy Risk Assessment Scale for High-Age Patients to estimate the risk of high-grade adverse events. With the help of these tools, better risk stratification of elderly July 2014, Vol. 21, No. 3

patients can be performed in order to choose the most appropriate treatment for each patient. This issue also includes 2 articles discussing the cancer pathways that are changing the landscape of oncology treatment, ie, the v-raf murine sarcoma viral oncogene homolog B (BRAF) signaling and programmed death 1 (PD-1) pathways. Discovery of BRAF mutations and the subsequent development of BRAF and MEK inhibitors have revolutionized the treatment of metastatic melanoma. Drs Hall and Kudchadkar review the importance of BRAF mutation in melanoma and other malignancies, including colorectal, thyroid, and lung cancers, as well as hairy cell leukemia. With the advent of BRAF inhibitors, for the first time in the history of melanoma treatment, response rates were reaching 50%. However, similar results have not been replicated in BRAF-mutated colorectal cancer. It has been previously debated whether tumor histology will lose relevance to the molecular characterization of the cancer. The differential benefit seen in patients with various tumor types and BRAF mutation suggests that histology matters — at least for now. It has also been suggested that in colorectal cancer, other signaling pathways, including phosphatidylinositol 3 kinase and epidermal growth factor receptor, are activated, thus leading to BRAF inhibitor resistance. Clinical trials with dual inhibitors are underway. After decades of few successes with immunotherapy, blocking the PD-1 immune checkpoint pathway has resulted in impressive results in melanoma, lung cancer, and renal cell carcinoma. Drs Dolan and Gupta review the PD-1 and programmed cell death ligand 1 (PDL-1) inhibitors currently in development. At the

July 2014, Vol. 21, No. 3

time of publication, at least 7 drugs are being evaluated in various clinical trials that target this pathway. None of these agents have been approved by the US Food and Drug Administration, but some of these are expected to receive such approval once the data are mature. In general, these agents are well tolerated, but significant and potentially life-threatening, immune-related adverse events (rash, colitis, ophthalmitis, hepatitis, pneumonitis, hypophysitis) can occur in a small proportion of patients. An urgent need exists to increase the number of patients with cancer enrolled into clinical trials. This is especially true for elderly patients because, although they are frequently under-represented in clinical trials, they have the highest incidence of cancer compared with younger patient populations; therefore, elderly patients would have the largest impact on trial results. With the advent of molecular target–driven therapies, it is of the utmost importance that patients with specific molecular alteration are able to receive these therapeutic agents in a clinical trial setting. Hopefully, trials that are smarter and faster will be developed that will then provide huge benefit to a select group of patients, rather than a minor benefit to a general population. By replicating the success of BRAF and PD-1/PDL-1 inhibitors in various malignancies using different targets, we may be able to achieve the holy grail of cancer treatment: cure. Amit Mahipal, MD Assistant Member Medical Director, Clinical Research Unit H. Lee Moffitt Cancer Center & Research Institute Tampa, Florida [email protected]

Cancer Control 189

Clinical Research Unit H. Lee Moffitt Cancer Center & Research Institute Moffitt Cancer Center is the only NCI-designated comprehensive cancer center in Florida. Early therapeutic trials are of particular interest at Moffitt and more than 350 patients are enrolled in phase 1/2 trials every year. The cancer mission of the Clinical Research Unit at Moffitt Cancer Center is to support clinical investigational research by monitoring patients and administering investigational agents. The Clinical Research Unit/Phase 1 program consists of a team of physicians from various oncology disciplines and other health care professionals, including nurse practitioners, clinical trial coordinators, pharmacists, research nurses, data managers, and regulatory specialists. There are more than 30 active phase 1 clinical trials in solid tumors and hematological malignancies at Moffitt to help patients with cancer receive novel therapies. Many of these trials are biomarker driven that preselect patients with a higher likelihood of response to therapy.

Associate Center Director, Clinical Investigations Dan M. Sullivan, MD

Medical Director Amit Mahipal, MD

NURSE PRACTITIONERS Barbara Bertels, MSN, ARNP-C Georgine Blum Wapinsky, MSN, ARNP-C

CLINICAL RESEARCH COORDINATORS Nancy L. Burke Jennifer L. Cooksey Barbara Padilla-Oliver Irene J. Williams Elson

DATA MANAGERS Lawrence P. McKinney Laura A. Narney

Selected Phase 1 Clinical Trial Faculty Director, Protocol Review and Regulatory Affairs Richard M. Lush, III, PhD BREAST ONCOLOGY Hyo S. Han, MD Susan E. Minton, DO Hatem Soliman, MD

HEAD AND NECK AND ENDOCRINE ONCOLOGY Jeffery Russell, MD, PhD

GASTROINTESTINAL ONCOLOGY Khaldoun Almhanna, MD Richard Kim, MD Amit Mahipal, MD Jonathan R. Strosberg, MD

GENITOURINARY ONCOLOGY Shilpa Gupta, MD Mayer N. Fishman, MD, PhD

MALIGNANT HEMATOLOGY Dan M. Sullivan, MD Rachid Baz, MD Jeffrey E. Lancet, MD Kenneth H. Shain, MD, PhD Eric Padron, MD

NEURO-ONCOLOGY Solmaz Sahebjam, MD

THORACIC ONCOLOGY Scott J. Antonia, MD, PhD Jhanelle Gray, MD

190 Cancer Control

July 2014, Vol. 21, No. 3

Selected Active Phase 1 Clinical Trials at Moffitt Cancer Center MCC 16002

Open Label, Phase 1/2 Study of MEDI-551, a Humanized Monoclonal Antibody Directed Against CD19, in Adult Subjects With Relapsed or Refractory Advanced B-Cell Malignancies

MCC 16254

Phase I Study of the HDAC Inhibitor Vorinostat With Chemotherapy and Radiation Therapy for the Treatment of Locally Advanced Non-Small Cell Lung Cancer (NSCLC)

MCC 16434

Study of the Anti-EphA3 Monoclonal Antibody KB004 in Subjects with EphA3-Expressing Hematologic Malignancies

MCC 16523

A Sequential Two-Stage Dose Escalation Study to Evaluate the Safety and Efficacy of Eltrombopag in Myelodysplastic Syndrome (MDS) Patients With Thrombocytopenia Who Progressed or Are Resistant to Hypomethylating Agents

MCC 16658

Phase I Multicenter Open Label Dose Escalation Study of Pasireotide (SOM230) LAR in Patients With Advanced Neuroendocrine Tumors (NET)

MCC 16683

Phase I Pharmacokinetic Study of Belinostat for Solid Tumors and Lymphomas in Patients With Varying Degrees of Hepatic Dysfunction

MCC 16699

Phase IB/II Trial of ALT-801 in Combination With Cisplatin and Gemcitabine in Muscle Invasive or Metastatic Urothelial Cancer

MCC 16738

Phase I Study to Evaluate the Safety and Tolerability and Pharmacokinetic/Pharmacodynamics of MK-8242 in Patients With Advanced Solid Tumors

MCC 17017

Phase 1A/1B, Multicenter, Open Label, Dose-Finding Study to Assess the Safety, Tolerability, Pharmacokinetics and Preliminary Efficacy of the Dual DNA-PK and Tor Kinase Inhibitor, CC-115, Administered Orally to Subjects With Advanced Solid Tumors and Hematologic Malignancies

MCC 17061

Phase 1B/2 Study to Evaluate the Safety and Efficacy of PF-04449913, an Oral Hedgehog Inhibitor, in Combination With Intensive Chemotherapy, Low Dose Ara-C or Decitabine in Patients With Acute Myeloid Leukemia or High-Risk Myelodysplastic Syndrome

MCC 17088

Phase I Study of the Safety, Pharmacokinetics and Pharmacodynamics of Escalating Doses of the Selective Inhibitor of Nuclear Export (SINE) KPT-330 in Patients With Advanced Hematological Malignancies

MCC 17096

Phase I Study of the Safety, Pharmacokinetics and Pharmacodynamics of Escalating Doses of the Selective Inhibitor of Nuclear Export (SINE) KPT-330 in Patients With Advanced or Metastatic Solid Tumor Malignancies

MCC 17114

Phase I/II Study of Ruxolitinib in Combination With Nilotinib in CML Patients With Evidence of Molecular Disease

MCC 17136

Phase I/II Study of Oral Bicarbonate as Adjuvant for Pain Reduction in Patients With Tumor Related Pain

MCC 17148

Phase 1 Study to Evaluate the Safety, Tolerability, and Pharmacokinetics of MEDI4736 in Subjects With Advanced Solid Tumors

MCC 17155

Phase I/II Trial of Combination Plerixafor (AMD3100), Bortezomib and Dexamethasone in Relapsed/Refractory Multiple Myeloma

MCC 17164

Phase IB, Multi-Center, Open Label, Dose Escalation Study of Oral LDE225 in Combination With BKM120 in Patients With Advanced Solid Tumors

MCC 17176

Phase I Trial Evaluating Safety and Tolerability of the Irreversible Epidermal Growth Factor Receptor Inhibitor Afatinib (BIBW 2992) in Combination With the SRC Kinase Inhibitor Dasatinib for Patients With Non–Small-Cell Lung Cancer (NSCLC)

MCC 17208

Phase I, Open-Label, Dose-Escalation Study of SGN-CD19A in Patients With B-Lineage Acute Lymphoblastic Leukemia and Highly Aggressive Lymphomas

MCC 17223

Phase 1/2 Open-Label Study to Assess the Safety, Tolerability and Preliminary Efficacy of TH-302, a Hypoxia-Activated Prodrug, and Dexamethasone With or Without Bortezomib in Subjects With Relapsed/Refractory Multiple Myeloma continued on page 192

July 2014, Vol. 21, No. 3

Cancer Control 191

Selected Active Phase 1 Clinical Trials at Moffitt Cancer Center (continued) MCC 17250

Phase IB Study of SAR650984 (Anti-CD38 mAb) in Combination With Lenalidomide and Dexamethasone for the Treatment of Relapsed or Refractory Multiple Myeloma

MCC 17259

Sequential Two-Stage Dose Escalation Study to Evaluate the Safety and Efficacy of Ruxolitinib for the Treatment of Chronic Myelomonocytic Leukemia (CMML)

MCC 17303

Phase 1, Open-Label, Dose-Escalation Study of SGN-CD19A in Patients With Relapsed or Refractory B-Lineage Non-Hodgkin Lymphoma

MCC 17340

Study of HSP90 Inhibitor AT13387 Alone and in Combination With Crizotinib in the Treatment of Non–Small-Cell Lung Cancer (NSCLC)

MCC 17378

Phase IB Multi-Cohort Study of MK-3475 in Subjects With Advanced Solid Tumors

MCC 17396

Single-Arm, Multicenter, Nilotinib Treatment-Free Remission Study in Patients With BCR-ABL1 Positive Chronic Myelogenous Leukemia in Chronic Phase Who Have Achieved Durable Minimal Residual Disease (MRD) Status on First Line Nilotinib Treatment

MCC 17419

Phase IB/2, Multicenter, Open-Label Study of Oprozomib and Dexamethasone in Patients With Relapsed and/or Refractory Multiple Myeloma

MCC 17453

Early Phase I Study of ABT-888 in Combination With Carboplatin and Paclitaxel in Patients With Hepatic or Renal Dysfunction and Solid Tumors

MCC 17461

Phase IB, Open-Label Study of Oral BGJ398 in Combination With Oral BYL719 in Adult Patients With Select Advanced Solid Tumors

MCC 17576

Phase I Trial of SGN-CD33A in Patients With CD33-Positive Acute Myeloid Leukemia

MCC 17608

Phase 1B Open-Label Study to Evaluate the Safety and Tolerability of MEDI4736 in Combination With Tremelimumab in Subjects With Advanced Non–Small-Cell Lung Cancer

MCC 17622

Phase I Open Label, Multicenter, Multiple Ascending Dose Trial Evaluating the Safety, Tolerability and Immunogenicity of Intramuscular Recombinant NY-ESO-1 Protein With GLA-SE Adjuvant in Patients With Unresectable or Metastatic Cancer Expressing NY-ESO-1 Antigen

MCC 17630

Phase 1/2 Clinical Trial of NPI-0052 in Patients With Relapsed or Relapsed/Refractory Multiple Myeloma

MCC 17682

Phase 1B, Multi-Center, Open-Label Study of Novel Combinations of CC-122, CC-223, CC-292, and Rituximab in Diffuse Large B-Cell Lymphoma

MCC 17687

Phase 1/2, Open-Label Study of Nivolumab Monotherapy or Nivolumab Combined With Ipilimumab in Subjects With Advanced or Metastatic Solid Tumors

MCC 17690

Phase I Trial of Single Agent Trametinib (GSK1120212) in Advanced Cancer Patients With Hepatic Dysfunction

MCC 17736

Phase I, Open-Label, Multicentre Study to Assess The Safety, Tolerability, Pharmacokinetics and Preliminary Anti-Tumor Activity of Gefitinib in Combination With Medi4736 (Anti Pd-L1) in Subjects With Non–Small-Cell Lung Cancer (NSCLC)

To schedule a patient appointment with a physician at Moffitt Cancer Center, call the New Patient Appointment Center at 813-745-3980 or 1-888-860-2778 (during normal business hours). For information about clinical trials, call Cheryl Maker in the Clinical Research Unit at 813-745-4106 or e-mail [email protected]. www.MOFFITT.org 192 Cancer Control

July 2014, Vol. 21, No. 3

In the era of molecular-targeted therapy, “effective” dose is sometimes measured through the inhibition of the intended target, which can prove to be problematic.

Copper Iris_0328. Photograph courtesy of Henry Domke, MD. www.henrydomke.com

Risks and Benefits of Phase 1 Clinical Trial Participation Amit Mahipal, MD, and Danny Nguyen, MD Background: The results from phase 1 clinical trials can allow new treatments to progress further in drug development or halt that process altogether. At the forefront of phase 1 clinical trials is the safety of every patient participant, which is particularly true when testing new oncologic treatments in which patients may risk potentially toxic treatments in the hope of slowing the progression of or even curing their disease. Methods: We explore the benefits and risks that patients experience when participating in phase 1 clinical trials. Results: Rules and regulations have been put into place to protect the safety and interests of patients while undergoing clinical trials. Selecting patients with cancer who will survive long enough to accrue data for these trials continues to be challenging. New prognostic models have been validated to help health care professionals select those patients who will likely benefit from participation in phase 1 trials. There also are long-lasting positive and negative impacts on those patients who choose to participate in phase 1 clinical trials. Conclusions: Modern phase 1 clinical trials represent a therapeutic option for many patients who progress through frontline therapy for their malignancies. Recent phase 1 clinical trials testing targeted therapies have increased responses in many diseases in which other lines of therapy have failed. Patients still face many risks and benefits while enrolled in a phase 1 trial, but the likelihood of treatment response in the era of rational, targeted therapy is increased when compared with the era of cytotoxic therapy.

Introduction Results from clinical trials help to answer questions and provide guidance for practicing health care professionals. The regimented clinical trial design was not standardized until the twentieth century1; however, physicians have been employing concepts of modern clinical trials for centuries. An ancient medical text, From the Clinical Research Unit at the H. Lee Moffitt Cancer Center & Research Institute, Tampa, Florida. Submitted February 12, 2014; accepted March 6, 2014. Address correspondence to Amit Mahipal, MD, Clinical Research Unit, H. Lee Moffitt Cancer Center, 12902 Magnolia Drive, Tampa, FL 33647. E-mail: [email protected] No significant relationship exists between the authors and the companies/organizations whose products or services may be referenced in this article. July 2014, Vol. 21, No. 3

The Canon of Medicine, established guidelines for the proper conduct of medical experimentation.2 In this text, the principles for testing the efficacy of a new medication were laid out, including that the drug must be free from any extraneous accidental quality and that the experimentation must be performed with the human body.2 The essence of these guidelines became the scientific method for testing of medications, and, for the most part, the medical field regulated itself when it came to new medications, elixirs, “cure-alls,” panaceas, and the like. The turning point in medication development that resulted in the rigorous, regimented development of clinical trials in the United States occurred in 1937 when pharmaceutical manufacturer S.E. Massengill Company (Bristol, Tennessee) released the first elixir Cancer Control 193

formulation of sulfanilamide, an antibiotic that, at the time, had been shown to have activity against streptococcal throat infections.3 The elixir was available to consumers without undergoing animal or human testing of any kind prior to its release. However, the antibiotic was suspended in diethylene glycol, also known colloquially as antifreeze. The product was so extensively disseminated into US stores that the US Food and Drug Administration (FDA) and S.E. Massengill could not fully recall the product, which had caused the deaths of at least 100 people.1 Even then, the FDA was empowered to recall the drug only because the label was misleading (ie, it was labeled as an “elixir” and, therefore, had to contain alcohol, but this “elixir” did not have any). Due in part to this series of deaths, the FDA was granted new powers in 1938 under the Federal Food, Drug, and Cosmetic Act, which required drug sponsors to submit safety data to the FDA for it to evaluate prior to marketing of the drug, thus planting the seed for the modern clinical trial structure4; this was later modernized by Hill in 1948.1 Hill, who was a British statistician, performed one of the first randomized controlled studies that showed that streptomycin could cure tuberculosis.5 However, in 1962, thalidomide, a drug popular as a hypnotic in Europe and suspected to cause birth defects, was supplied to US physicians who subsequently gave the drug to expectant mothers as a remedy for morning sickness.6 This act resulted in nearly a dozen infants being born with birth defects, far less than the approximately 10,000 infants worldwide born with thalidomide-related defects. The smaller impact of thalidomide in the United States was due in part to the efforts of the FDA, which denied the thalidomide application on grounds that more evidence of

safety was required.1 The amendments in 1962 that followed on the heels of the thalidomide incident further strengthened the control of the FDA over new investigational drugs, thus requiring pharmaceutical companies to demonstrate that their investigational drug could be safely given to patients in the preclinical setting, thereby setting the stage for the formation of phase 1 clinical trials (Table 1).1,7

Purpose of Phase 1 Trials Historically, the focus of phase 1 clinical trials has been to demonstrate that a new drug can be safely given to humans at the maximum tolerated dose (MTD),8 which is associated with dose-limiting toxicities (DLTs). The MTD, which could be a therapeutic dose or the maximum dose that can safely be administered, is then carried on to further phases of clinical trials. In the era of targeted agents, the biologically effective dose is now frequently used rather than the MTD. Because the primary purpose is not efficacy, maintaining patient population homogeneity and obtaining measurable tumor response is not required; however, many investigators include these factors in their protocols.9 Understanding the emphasis on safety in phase 1 studies requires an understanding of the history of drug development in the United States and why the FDA is concerned with establishing safety followed by efficacy. The field of oncology has matured during the last 20 years due in part to the understanding of the various molecular pathways involved in tumorigenesis. Because of the advent of molecularly targeted therapies due to this evolution, the standard dosing regimen, which consists of “cycles” of chemotherapy at the MTD, may need to be reconsidered.10 In fact, se-

Table 1. — Phases of Clinical Trials Phase

Primary Goal

Primary Researcher

Subject Type

Comment

Preclinical

Nonhuman efficacy Toxicity PK

PhD, MD, PharmD, or any researcher

Cell lines (animal)

0

Determining PK and PD

Clinical researcher

Human

Focuses on determining oral bioavailability and half-life Often combined with phase 1

1

Evaluation of safety and adverse events

Clinical researcher

Human

May be expanded or combined with phase 2

2

Examine efficacy and dose range

Clinical researcher

Human

May help in optimizing dose, schedule, and select disease types

3

Expanded study to substantiate efficacy and safety

Clinical researcher

Human (N = large range)

Generally includes multiple sites and investigators

4

Postmarketing surveillance

Primary physician

Human (N = all patients taking the drug)

Determines long-term effects

PD = pharmacodynamic, PK = pharmacokinetic.

194 Cancer Control

July 2014, Vol. 21, No. 3

lected molecularly targeted therapies such as tyrosine kinase inhibitors (eg, imatinib, ibrutinib, sorafenib) are not given in cycles but instead are given orally every day.10 The goal in such cases may not be tumor regression but rather tumor control. As such, dosing at the MTD may not be the dose associated with the most effectiveness. As in the case of ibrutinib, the MTD was never reached because the drug was well-tolerated and the dose selected for further clinical trials was based on the dose that caused near complete occupation of all Bruton tyrosine kinase receptors.11 This calls into question whether toxicity can continue to be the primary goal for phase 1 trial design.10 For a particular agent, its effects on its purported molecular target may serve as another marker for efficacy. Logistically, this may become a complicated matter, such as repeatedly obtaining tissue or routine blood work. For the patient, this may result in more invasive procedures, which carry their own inherent risks, or more frequent blood work, which

one may expect to negatively impact patient enrollment. However, study results indicate that patients are willing to undergo multiple biopsies if needed.12

Study Design The difficulty in designing a phase 1 clinical trial is the decision of whether to escalate the dose of the study drug quickly (such that patients develop toxicities sooner) or whether to escalate the dose slowly (such that patients are treated at subtherapeutic doses for longer).13 However, study design protocols that attempt to answer this question are out of the scope of this review article, but they may be of interest because investigators must consider the impact of the study design on patient safety. For instance, one study examining phase 1 patients enrolled between 2002 and 2004 demonstrated that aggressive dose-escalation schemes did not have a response advantage for cytotoxic agents but were associated with more toxicity when compared with conservative dose-escalation schemas.14 In

Table 2. — Selected Dose Escalation Designs Dose-Escalation Method

Description

Advantages

Disadvantages

Rule-Based Designs 3 + 3 (including 2 + 4, 3 + 3 + 3, and 3 + 1 + 1)

Dose escalation follows a modified Fibonacci sequence (dose escalation sequence 100%  67%  50%  40%, and so on) If 1 patient has a DLT, 3 more patients are added (+ 3) Escalation continues until 2 patients among the same cohort experience a DLT

Simple Safe Adding 3 more patients per dose level supplies more PK data

Excessive number of escalation steps means more patients potentially treated at subtherapeutic doses

Accelerated titration

Assignment of patients to dose levels follows specific rules according to observed toxicities at each dose level Allows intrapatient dose escalation

Reduces the amount of patients treated at subtherapeutic doses Eventual phase 2 dose can be interpreted from data from all patients

May mask cumulative toxic effects of treatment if model does not fit data

Pharmacologic-guided dose escalation

Assumes that DLT is predicted by plasma drug concentrations and an animal model Area under the curve predicted from preclinical data

Reduces the amount of patients treated at subtherapeutic doses (100% dose increment escalation) Provides PK data

Logistics behind obtaining real-time PK data Interpatient variability in drug metabolism may affect results

Based on the Bayesian model Initial dose based on preclinical data All patients treated at predicted maximum tolerated dose Probability of reaching DLT updated for every patient who enters the study at every dose level Stopping rules vary (eg, when 6 patients are assigned to the same dose level)

Reduces amount of patients treated at subtherapeutic doses Uses all data gathered from all patients Phase 2 dose estimated with a confidence interval Late toxicities are accounted for

Logistics and manpower behind calculations for every patient for every cohort Requires strong support from a statistician for dose escalation

Model-Based Designs Continual reassessment

DLT = dose-limiting toxicity, PK = pharmacokinetic. Adapted from Le Tourneau C, Lee JJ, Siu LL. Dose escalation methods in phase I cancer clinical trials. J Natl Cancer Inst. 2009;101(10):708-720. Published in its adapted form by permission of Oxford University Press.

July 2014, Vol. 21, No. 3

Cancer Control 195

this study, investigators reported a death rate of 1.1%,14 which, in general, is more than double the typically accepted risk of death for phase 1 studies.15 Innovative, more efficient, and safer designs are being developed compared with the traditional 3 + 3 dose-escalation design,16 which was designed in the era of cytotoxic therapy. During this time, higher doses were assumed to result in higher efficacy rates, but these doses also resulted in higher toxicity rates. Another main drawback of the traditional 3 + 3 design is that each escalation step may represent a group of patients treated with subtherapeutic levels of a particular medication. An analysis of 21 trials of cancer therapies using the 3 + 3 design between 1992 and 2008 (therapies eventually approved by the FDA) revealed that more than one-half of these designs had at least 6 dose-escalation levels.17 Many different dose-escalation schemes exist, although the predominant scheme used is the 3 + 3 design. Table 2 lists the advantages and disadvantages of selected dose-escalation designs.17 Ultimately, the primary goal of newer dose-escalation schemes is to maximize the number of patients receiving the most efficacious dose. In the era of molecular-targeted therapies, new questions arise as to what constitutes an “effective” dose. Oftentimes, this concept is measured through the inhibition of the intended target, which can pose several obstacles, such as access and assessment of tissue (eg, tumor, peripheral blood) and the determination of the level of inhibition required to obtain a clinical response.17 In these situations, dose-escalation designs may not be as relevant as during the era of cytotoxic therapy. However, generally speaking, toxicity is still used as an end point for molecular-targeted therapies. In addition, emphasis is placed on the preclinical setting and the so-called phase 0 trial in which the demonstration of a targeted effect is the primary goal. Pharmacokinetic and pharmacodynamic data are also obtained during phase 0 trials. The advantage of phase 0 trials is that having data upfront helps expedite new drugs through other phases of clinical testing.7

Patient Selection From our experience, the largest risk to patients who participate in phase 1 trials is death; secondary risks include adverse events associated with the study drug that may or may not be reversible. Our experience also suggests that oncologists generally offer patients with progressive, refractory malignancies the opportunity to participate in phase 1 studies as a “last ditch effort.” Consequently, many patients may be frail and will have experienced end-organ dysfunction and have short life expectancies. Early reports suggested that approximately 20% of patients passed away during the first 90 days of entry into a phase 1 196 Cancer Control

trial.18 Because of this, modern phase 1 studies use arguably biased stringent inclusion criteria, which exclude approximately 33% of participants screened for entry.19 Moreover, criteria are so stringent that a study published by Seidenfeld et al20 concluded that 93% of participants of phase 1 trials nearly matched the performance status (PS) of the general population. Other inclusion criteria, along with Eastern Cooperative Oncology Group (ECOG) PS, Karnofsky PS, or both, generally look at organ function (eg, creatinine, liver enzymes), age, lactate dehydrogenase (LDH), and other comorbidities.21 In an effort to select which patients might reasonably survive long enough to accrue safety data for phase 1 studies, many scoring systems have been formulated to help select patients with the lowest risk of mortality.22,23 For instance, Wheler et al24 retrospectively determined that a history of thromboembolism, the presence of liver metastasis, and thrombocytosis predicted a shorter survival rate in patients enrolled in phase 1 clinical trials, with each parameter bearing comparable risk of death and weighed equally. From these data, they developed a risk score with corresponding risk groups and 6- and 12-month survival rates (low risk = 73%, 51%; intermediate risk = 65%, 34%; high risk = 35%, 6%, respectively).24 This study was the first to report the survival rate of phase 1 participants in the era of biologically and molecularly targeted therapy. A median overall survival (OS) rate of 9 months was reported in this study,24 which is in contrast to the median OS rate of 5 months in the era of cytotoxic therapy and ECOG PS and LDH levels.21 Arkenau et al22 from the Royal Marsden Hospital (RMH) developed a prognostic score using retrospective data of 212 patients enrolled in their phase 1 program (Table 3). In this study, 3 variables associated with poor outcomes were isolated, including an elevated level of LDH (> upper limit of normal), low Table 3. — Royal Marsden Hospital Prognostic Score Variable

Score

Lactate Dehydrogenase < ULN > ULN

0 1

Albumin (g/dL) > 3.5 < 3.5

0 1

Sites of Metastases 0–2 >2

0 1

Hazard Ratio 1.85

1.83

1.54

Scores 0–1 = good prognosis, 2–3 = poor prognosis. ULN = upper limit of normal. Data from reference 22.

July 2014, Vol. 21, No. 3

level of albumin (< 3.5 g/dL), and more than 2 sites of metastasis. Patients with a score of 0 to 1 had a median OS rate of 74.1 weeks, whereas patients with a score of 2 to 3 had a median OS rate of 24.9 weeks across all tumor types.22 These data were prospectively studied at the same institution and validated in a follow-up study.25 Using the RMH score, Arkenau et al25 demonstrated that nearly 90% of patients who died within the first 90 days of entry into a phase 1 trial had a prognostic score of 2 to 3. At the time of the study, those with a score of 0, 1, 2, or 3 had a median OS rate that was not reached: 25.7 weeks, 15.7 weeks, and 14.1 weeks, respectively. This scoring system was further modified and validated at the phase 1 clinic at the University of Texas MD Anderson Cancer Center in Houston.23 Wheler et al23 added gastrointestinal tumor type and ECOG PS (≥ 1) to the RMH score as factors associated with a poor prognosis (Table 4). Using their prognostic score, they found that median survival rates for the low-risk (0), low-intermediate (1), intermediaterisk (2), high-intermediate risk (3), and high-risk (4–5) groups were 24.0 months, 15.2 months, 8.4 months, 6.2 months, and 4.1 months, respectively.23 The relative risk of having more than 2 sites of metastasis and ECOG PS of at least 1 was lower than the other variables, a finding likely due to stringent inclusion criteria and clinical judgment. Also of note is the median survival rate of 10 months, with 86% patients having received a targeted therapy/biological agent and 32% having received a cytotoxic agent. These results Table 4. — MD Anderson Clinical Center Prognostic Score Variable

Score

Lactate Dehydrogenase < ULN > ULN

0 1

Albumin (g/dL) > 3.5 < 3.5

0 1

Sites of Metastases 0–2 >2

0 1

ECOG PS 0 ≥1 Tumor Type Non-GI GI

Relative Risk for Death 1.74

1.58

1.26

0 1

1.32

1.42 0 1

Scores: 0 = low risk, 1 = low-intermediate risk, 2 = intermediate risk, 3 = high-intermediate risk, 4–5 = high risk. ECOG PS = Eastern Cooperative Oncology Group performance status, GI = gastrointestinal, ULN = upper limit of normal. Data from reference 23.

July 2014, Vol. 21, No. 3

further demonstrate the increased clinical benefit of phase 1 clinical trials in the era of targeted therapies.

Phase 1 Trial Participation as a Therapeutic Option Although the goal of phase 1 studies has primarily focused on safety profiles, most patients with cancer participate in these trials with the hope of deriving clinical benefit, and health care professionals are beginning to integrate participation in a phase 1 study as part of a patient’s plan of care.26 Historically, health care professionals expected that phase 1 studies would yield a response rate of approximately 6% and a death rate due to the study drug of approximately 0.5%.27 With the advent of molecular targets and immunotherapy, this expectation of efficacy has changed. Horstmann et al15 updated these findings using data from the Cancer Therapy Evaluation Program, which consisted of data from 10,402 participants of phase 1 trials that took place between 1991 and 2002. They found an overall response rate of 10.6% and partial response and complete response rates of 7.5% and 3.1%, respectively. They reported that 0.49% of patients died while participating in a trial (0.21% of patient deaths were attributed to the study drug). Italiano et al26 reviewed the efficacy of phase 1 trials from their own institution between the years 2003 and 2006. The researchers found an objective response rate of 7.2%, a rate of stable disease of 41%, a progression-free survival rate of 2.3 months, and a median OS rate of 8.7 months.26 In addition, 56.6% of participants went on to pursue different treatment options after exiting the phase 1 study, demonstrating that clinicians at that institution were incorporating participation in a phase 1 study as part of treatment pathways, particularly for malignancies without a clear, preferred treatment option with good effectiveness.26 Moreover, in some malignancies (eg, progressive head and neck cancers), participating in a phase 1 clinical trial could potentially mean that patients would have progression-free survival rates similar to those seen in third-line therapies already approved by the FDA.28 Considering the evidence of efficacy behind selected approvals by the FDA,29,30 these results are significant. For instance, the addition of cetuximab to leucovorin/fluorouracil/irinotecan compared with leucovorin/fluorouracil/irinotecan alone in KRAS wild-type patients increased the progression-free survival rate from 8.7 months to 9.9 months29 and the addition of nab-paclitaxel to gemcitabine increased the progression-free survival rate from 3.7 months to 5.5 months.30 Further expanding on the benefit of targeted therapy, one study found that the risk of death during a phase 1 trial testing a cytotoxic agent was nearly quadruple that of a trial testing a targeted agent.31 Cancer Control 197

Phase 1 studies that include cytotoxic agents that have received approval from the FDA also tend to have lower risks of death and toxicity than novel cytotoxic agents.14

Patient Benefits In addition to the possibility of controlling or reducing disease burden when other lines of therapy have failed, patients derive other benefits from participating in phase 1 trials. Per our experience, some researchers view early access to a potentially helpful drug as a benefit; however, by the very nature of phase 1 clinical trials, this early access may also prove to be a risk. Dealing with a refractory, recurrent, nonoperable, and/ or metastatic malignancy can be taxing on patients and family members. Most patients are overall satisfied with their experiences in phase 1 trials.32,33 The regimented routine that patients undergo as part of a phase 1 trial, which may include routine physical examinations, laboratory draws, biopsies, and radiological examinations, help alleviate some fear about the gravity of their disease, and patients were generally satisfied to receive more information about their disease as well as to supply information themselves.33 Given the very nature of toxicity reporting in a phase 1 trial — with every symptom scrutinized — patients unexpectedly viewed this as being positive. Many of these patients developed trust in the trial physician and were content knowing that their participation might contribute to future patients’ health.11,32 Many also felt empowered by attempting to control their disease.13

Patient Risks Participation in a phase 1 clinical trial has known and unknown risks. The primary risk is death from the investigational agent or death from malignancy progression or malignancy-related complications. In addition to this risk of death, other risks that patients may experience include acute toxicities (eg, nausea, fatigue, diarrhea) or delayed toxicities, which may not be detected until the completion of the phase 1 trial and further studies progress. For instance, ponatinib was approved by the FDA after study results indicated great response rates.34 However, the correlation between the risk of thrombotic events was made only after ponatinib had been approved by the FDA for the treatment of chronic myelogenous leukemia blast crisis or those with a T315I mutation.35 With regard to patient satisfaction, many patients do not feel better about their disease once their trial participation is completed.32 With the advent of biomarker-driven trials that require the testing of tumor tissue, the unavailability of tissue for future trials can become an issue. Moreover, many early-phase trials require fresh biopsies that subject individual patients to risks without any direct benefit to them. 198 Cancer Control

Conclusions Study design was originally focused on dosing and safety; however, the design of trials is becoming increasingly sophisticated and includes ways in which to maximize clinical benefit through dosing schemes, incorporate randomized trial designs36 (including elements of phase 2 trial designs), and to use more targeted and biological therapies. During the last 20 years, the advent of targeted therapies in phase 1 trials has improved clinical benefit in terms of overall survival rates and toxicity profiles when compared with the era of cytotoxic agents. It is because of the progression and refinement of our knowledge of cancer that such therapies are available for testing, and it is because of better therapeutic targets for drugs that clinical trials can often be considered treatment options for patients with recurring, relapsed, or refractory malignancies. Phase 1 clinical trials should not be viewed as a last resort for patients who have failed current therapy. Rather, enrollment in clinical trials should be viewed as another therapeutic option. The field of oncology will continue to accumulate more knowledge and be able to rationally target molecular pathways as they become elucidated. Indeed, it is an exciting time to be in the field of oncology, and phase 1 clinical trials are one of the ways that both physicians and patients can help reinforce its foundation. References 1. Junod SW. FDA and clinical drug trials: a short history. http://www. fda.gov/AboutFDA/WhatWeDo/History/Overviews/ucm304485.htm. Accessed March 26, 2014. 2. Gruner C, transl. A Treatise on the Canon of Medicine of Avicenna. New York: AMS Press; 1973. 3. Wax PM. Elixirs, diluents, and the passage of the 1938 Federal Food, Drug and Cosmetic Act. Ann Intern Med. 1995;122(6):456-461. 4. US Food and Drug Administration. About FDA: FDA history part II. The 1938 Food, Drug, and Cosmetic Act. http://www.fda.gov/ aboutFDA/WhatWeDo/Histor y/origin/ucm054826.htm. Accessed April 8, 2014. 5. Yoshioka A. Use of randomisation in the Medical Research Council’s clinical trial of streptomycin in pulmonary tuberculosis in the 1940s. BMJ. 1998;317(7167):1220-1223. 6. Avorn J. Learning about the safety of drugs--a half-century of evolution. N Engl J Med. 2011;365(23):2151-2153. 7. Kummar S, Rubinstein L, Kinders R, et al. Phase 0 clinical trials: conceptions and misconceptions. Cancer J. 2008;14(3):133-137. 8. Gad SC. Clinical Trials Handbook. Hoboken, NJ: John Wiley & Sons; 2009. 9. Rubinstein LV, Simon RM. Phase I Clinical Trial Design. Bethesda, MD: National Cancer Institute; 2003. http://linus.nci.nih.gov/techreport/phaseIctd. pdf. Accessed March 26, 2014. 10. Eisenhauer EA. Phase I and II trials of novel anti-cancer agents: endpoints, efficacy and existentialism. The Michel Clavel Lecture, held at the 10th NCI-EORTC Conference on New Drugs in Cancer Therapy, Amsterdam, 16-19 June 1998. Ann Oncol. 1998;9(10):1047-1052. 11. Advani RH, Buggy JJ, Sharman JP, et al. Bruton tyrosine kinase inhibitor ibrutinib (PCI-32765) has significant activity in patients with relapsed/refractory B-cell malignancies. J Clin Oncol. 2013;31(1):88-94. 12. Abadie R. The Professional Guinea Pig: Big Pharma and the Risky World of Human Subjects. Durham, NC: Duke University Press; 2010. 13. Collins JM, Zaharko DS, Dedrick RL, et al. Potential roles for preclinical pharmacology in phase I clinical trials. Cancer Treat Rep. 1986;70(1):73-80. 14. Koyfman SA, Agrawal M, Garrett-Mayer E, et al. Risks and benefits associated with novel phase 1 oncology trial designs. Cancer. 2007;110(5): 1115-1124. 15. Horstmann E, McCabe MS, Grochow L, et al. Risks and benefits of phase 1 oncology trials, 1991 through 2002. N Engl J Med. 2005;352(9): 895-904. July 2014, Vol. 21, No. 3

16. Ivy SP, Siu LL, Garrett-Mayer E, Rubinstein L. Approaches to phase 1 clinical trial design focused on safety, efficiency, and selected patient populations: a report from the Clinical Trial Design Task Force of the National Cancer Institute Investigational Drug Steering Committee. Clin Cancer Res. 2010;16(6):1726-1736. 17. Le Tourneau C, Lee JJ, Siu LL. Dose escalation methods in phase I cancer clinical trials. J Natl Cancer Inst. 2009;101(10):708-720. 18. Arkenau HT, Olmos D, Ang JE, et al. 90-days mortality rate in patients treated within the context of a phase-I trial: how should we identify patients who should not go on trial? Eur J Cancer. 2008;44(11):1536-1540. 19. Karavasilis V, Digue L, Arkenau T, et al. Identification of factors limiting patient recruitment into phase I trials: a study from the Royal Marsden Hospital. Eur J Cancer. 2008;44(7):978-982. 20. Seidenfeld J, Horstmann E, Emanuel EJ, Grady C. Participants in phase 1 oncology research trials: are they vulnerable? Arch Intern Med. 2008;168(1):16-20. 21. Bachelot T, Ray-Coquard I, Catimel G, et al. Multivariable analysis of prognostic factors for toxicity and survival for patients enrolled in phase I clinical trials. Ann Oncol. 2000;11(2):151-156. 22. Arkenau HT, Olmos D, Ang JE, et al. Clinical outcome and prognostic factors for patients treated within the context of a phase I study: the Royal Marsden Hospital experience. Br J Cancer. 2008;98(6):1029-1033. 23. Wheler J, Tsimberidou AM, Hong D, et al. Survival of 1,181 patients in a phase I clinic: the MD Anderson Clinical Center for targeted therapy experience. Clin Cancer Res. 2012;18(10):2922-2929. 24. Wheler J, Tsimberidou AM, Hong D, et al. Survival of patients in a phase 1 clinic: the M. D. Anderson Cancer Center experience [published correction appears in Cancer. 2009;115(7):1588]. Cancer. 2009;115(5):1091-1099. 25. Arkenau HT, Barriuso J, Olmos D, et al. Prospective validation of a prognostic score to improve patient selection for oncology phase I trials. J Clin Oncol. 2009;27(16):2692-2696. 26. Italiano A, Massard C, Bahleda R, et al. Treatment outcome and survival in participants of phase I oncology trials carried out from 2003 to 2006 at Institut Gustave Roussy. Ann Oncol. 2008;19(4):787-792. 27. Smith TL LJ, Kantarjian HM, Legha SS, et al. Design and results of phase I cancer clinical trials: three-year experience at M.D. Anderson Cancer Center. J Clin Oncol. 1996;14(1):287-295. 28. Garrido-Laguna I, Janku F, Falchook GS, et al. Patients with advanced head and neck cancers have similar progression-free survival on phase I trials and their last Food and Drug Administration-approved treatment. Clin Cancer Res. 2010;16(15):4031-4037. 29. 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(14):1408-1417. 30. Von Hoff DD, Ervin T, Arena FP, et al. Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine. N Engl J Med. 2013;369(18):16911703. 31. Roberts TG Jr, Goulart BH, Squitieri L, et al. Trends in the risks and benefits to patients with cancer participating in phase 1 clinical trials. JAMA. 2004;292(17):2130-2140. 32. Verheggen FW, Nieman FH, Reerink E, et al. Patient satisfaction with clinical trial participation. Int J Qual Health Care. 1998;10(4):319-330. 33. Zaric B, Perin B, Ilic A, et al. Clinical trials in advanced stage lung cancer: a survey of patients’ opinion about their treatment. Multidiscip Respir Med. 2011;6(1):20-27. 34. Cortes JE, Kim DW, Pinilla-Ibarz J, et al. A phase 2 trial of ponatinib in Philadelphia chromosome-positive leukemias. N Engl J Med. 2013;369(19):1783-1796. 35. Prasad V, Mailankody S. The accelerated approval of oncologic drugs: lessons from ponatinib. JAMA. 2014;311(4):353-354. 36. Djulbegovic B, Hozo I, Ioannidis JP. Improving the drug development process: more not less randomized trials. JAMA. 2014;311(4):355-356.

July 2014, Vol. 21, No. 3

Cancer Control 199

Innovative trial designs addressing the limitations of traditional dose-escalation methods have yet to establish their clinical superiority in the phase 1 trial setting. Nautilus Spiral_4592. Photograph courtesy of Henry Domke, MD. www.henrydomke.com

Phase 1 Trial Design: Is 3 + 3 the Best? Aaron R. Hansen, MBBS, Donna M. Graham, MBBCh, Gregory R. Pond, PhD, and Lillian L. Siu, MD Background: Concerns have been recognized about the operating characteristics of the standard 3 + 3 dose-escalation design. Various innovative phase 1 trial designs have been proposed to address the issues and new challenges posed by molecularly targeted agents. However, in spite of these proposals, the conventional design is still the most widely utilized. Methods: A review of the literature of phase 1 trials and relevant statistical studies was performed. Results: Beyond statistical simulations, sparse clinical data exist to support or refute many of the shortcomings ascribed to the 3 + 3 rule method. Data from phase 1 trials demonstrate that traditional designs identified the correct dose and relevant toxicities with an acceptable level of precision in some instances; however, no single escalation method was proven superior in all circumstances. Conclusions: Design selection should be guided by the principle of slow escalation in the face of toxicity and rapid dose increases in the setting of minimal or no adverse events. When the toxicity of a drug is uncertain or a narrow therapeutic window is suggested from preclinical testing, then a conservative 3 + 3 method is generally appropriate. However, if the therapeutic window is wide and the expected toxicity is low, then rapid escalation with a novel rule- or model-based design should be employed.

Introduction The primary objective of a phase 1 oncology trial is to define the recommended phase 2 dose (RP2D) of a new drug or multiagent combination in the schedFrom the Princess Margaret Cancer Centre, University Health Network, Division of Medical Oncology and Hematology (ARH, DMG, LLS) and the Department of Medicine, University of Toronto, Ontario, Canada (ARH, DMG, LLS), the Department of Oncology, McMaster University (GRP), and the Ontario Clinical Oncology Group, Hamilton, Ontario, Canada (GRP). Submitted December 18, 2013; accepted January 24, 2014. Address correspondence to Lillian L. Siu, MD, Drug Development Program, Princess Margaret Cancer Centre, 610 University Avenue, Suite 5-718, Toronto, Ontario, Canada M5G 2M9. E-mail: [email protected] No significant relationships exist between the authors and the companies/organizations whose products or services may be referenced in this article. 200 Cancer Control

ule tested. Although many crucial components make up a phase 1 study, this article will focus on various dose-escalation methods that can be incorporated into such trials. The conventional 3 + 3 design, as described by Storer1 in 1989, was originally introduced in the 1940s2 and was among the earliest dose-escalation and de-escalation schemes utilized. However, several concerns have been raised about the quality of the operating characteristics of the 3 + 3 design. Statistical simulations have demonstrated that a trial using the 3 + 3 design identifies the maximum tolerated dose (MTD) in as few as 30% of trials.3 Furthermore, some argue that this method of dose escalation may result in a high proportion of patients being treated at subtherapeutic doses.4 Innovative trial designs that offer potentially more superior operating characteristics have been July 2014, Vol. 21, No. 3

proposed. However, despite these advances, the 3 + 3 dose-escalation method remains the most popular method employed by researchers of phase 1 trials. In a review of more than 1,200 phase 1 studies from 1991 to 2006, more than 98% of trials utilized the 3 + 3 dose escalation scheme.5 This figure was confirmed in a review of 181 phase 1 trials from January 2007 to December 2008, with more than 96% of trials using the traditional 3 + 3 design or a variation.6 Given the time frame of these reviews, most of the agents included in these 2 reports were cytotoxic chemotherapeutic agents,5,6 with molecularly targeted agents (MTAs) comprising only 18% in the latter report.6 MTAs are defined as anticancer agents that selectively target molecular pathways, as opposed to DNA, tubulin, or cell division machinery.7 The recent surge in the development of MTAs has challenged early-phase trial design. Because MTAs were purported to have a more specific therapeutic index on tumor tissue while sparing normal tissue, it was believed that studies of MTAs would have resulted in a shift toward newer methods with less conservative dose escalations. However, in a review of 155 MTA phase 1 trials published between 2000 and 2010, more than 60% of them incorporated the conventional 3 + 3 dose escalation.8 Logistical simplicity of the 3 + 3 design and clinician familiarity with the escalation rules may explain this observation and preclude the transition to novel designs on a larger scale. Moreover, it is not clear whether the advantages offered by newer methodologies are negated by their complexity and difficulty to implement. This article will review the current phase 1 trial design landscape and evaluate which dose-escalation methods are optimal for determining dose and safety in an efficient manner, in addition to addressing several challenges faced by modern phase 1 trials.

Dose-Escalation Designs Phase 1 trials must prioritize safety while attempting to maintain efficiency. A typical dose-escalation phase 1 study selects a safe starting dose based on preclinical data from in vitro and in vivo testing of the drug. Incremental dose increases for assigned patient cohorts occur until a prespecified end point is reached, which, in general, is the incidence of dose-limiting toxicities (DLTs). DLTs are severe but ideally reversible adverse events that occur within a protocol-defined period, usually the first cycle. Geographical variations in nomenclature for the final dose levels in phase 1 trials can be confusing. In Europe, the MTD is defined as the dose level in which an allowable frequency of DLTs has been exceeded, and the dose level immediately below is usually expanded to confirm that its incidence of DLTs is within an acceptable threshold.9 If such is the case, then that penultimate dose level is considered July 2014, Vol. 21, No. 3

the RP2D. In North America, the highest dose level reached, in most instances due to an unacceptable incidence of DLTs, is referred to as the maximum administered dose (MAD). The MTD is typically defined as the dose level immediately below the MAD and corresponds with the RP2D.9 In the event that the MTD is not reached, other non–toxicity-based end points can be considered in order to recommend a dose, such as a pharmacodynamic (PD) marker of sufficient pathway inhibition of a putative molecular target. Phase 1 trial designs are broadly divided into rule- or model-based methods. Rule-based methods utilize prespecified rules based on actual observations of target events (eg, DLT) from the clinical data to assign patients to dose levels and determine the MTD or RP2D. Model-based designs use a statistical estimation of the target toxicity level by depicting the dose-toxicity relationship. The model is used to assign patients to dose levels and define the MTD or RP2D. Safeguards are established in most model-based designs to limit escalation above the MTD and patient exposure to excessively toxic treatment doses.4 The Table2,4,10-18 outlines selected examples of these phase 1 trial designs. The most commonly used rule-based design is the traditional 3 + 3 design, which guides “up-and-down” decisions, using the modified Fibonacci mathematical series to determine the amount of dose increase for cohorts of sequentially enrolled patients. The Fibonacci series ensures that dose increases are initially large but increments are smaller at higher dose levels (Fig 1A).2 Newer trial designs have explored modifications of the process of dose escalation using statistical and empirical methods.7 In an effort to expedite the dose-escalation process to more rapidly and efficiently determine the RP2D, accelerated titration designs have been proposed (Fig 1B).4 Although they are not commonly used in clinical practice, these designs use single-patient cohorts for initial dose escalation until DLT or 2 moderate toxicities occur during the first treatment cycle. Following the initial acceleration phase, the traditional 3 + 3 design is applied, which theoretically reduces the number of patients treated at subtherapeutic doses. In this design, intrapatient dose escalation is permitted if no or minimal toxicities are observed.4 Accelerated titration designs are still rule based, but they can incorporate model-based designs following the initial single-patient cohorts to create hybrid methods. Interests in exploring nontoxicity end points have been fuelled by the observation that MTAs do not have the classical monotonic dose-toxicity curve as seen with cytotoxic chemotherapy. Therefore, escalating doses solely based on toxicity may not be the most appropriate strategy, and using alternative end points such as pharmacokinetic (PK) or PD data may be more appropriate for determining the optimal dose of these Cancer Control 201

agents. Robust preclinical data to support the use of these end points are imperative if such methodology is to be used. Pharmacokinetically guided dose escalation (PGDE) involves regular PK assessments to determine subsequent dose modifications (Fig 1C).10 The main practical limiting factors of PGDE include interpatient variability and the need for timely PK analysis.10 Where a PD end point of target inhibition is used, there must be a predefined target, available tumor tissue for testing, and a validated assay to determine and quantitate the extent of target inhibition. These are difficult criteria to meet, and the ability to reliably correlate PD readouts with clinical outcomes based on limited data further challenges the use of this strategy.

Model-based designs were developed to improve precision in estimating the RP2D as well as efficiency during dose escalation. Model-based designs establish a dose-toxicity curve prior to patient enrollment and then use toxicity data from enrolled patients and Bayesian statistical methods to modify and update this curve as the study proceeds. The first model-based method was the continual reassessment method (CRM).2 Modifications to this design have been developed with the aim of improving safety and avoiding an overestimation of the MTD. These include modified CRM,11,12 escalation with overdose control (Fig 1D),13 and the time-to-event continual reassessment method (TITE-CRM).14 It has been suggested that model-

Table. — Selected Examples of Trial Escalation Designs Trial Design

Method

Comments

Rule-Based Designs Key Points No prior assumptions about dose-toxicity curve Decision to escalate based on toxicity results from first course administration of current level 3 + 32 (Fig 1A)

3 patients treated per dose level If no DLT, dose is escalated for the next cohort of 3 patients If 1 DLT, 3 additional patients are treated at this level with dose escalation only if no additional DLTs If ≥ 2 DLTs, prior dose level is defined as MTD MTD decided when 6 patients are treated at a dose level with < 2 DLTs

Potentially more patients are treated at subtherapeutic doses Statistical simulations suggest RP2D often at lower doses than other designs

Accelerated titration4 (Fig 1B)

A series of designs have been proposed. All have fixed increments for dose escalation: Design 1 is as for 3 + 3 design but with 40% dose increments Design 2 has single patient cohorts during accelerated phase. When a first-course DLT or second first-course intermediate toxicity occurs, cohort expands and reverts to design 1 Design 3 has single patient cohorts with double-dose escalation steps (80% dose increments). Revert to design 1 with same trigger as design 2 Design 4 is per design 3 but trigger to revert to design 1 is any course DLT or second instance of any course intermediate toxicity

Acceleration and escalation phase in one design Intrapatient dose escalation, where permitted, may mask delayed or cumulative toxicity

PGDE Key Points Requires real-time PK measurement and assessment for dose modification Assumes DLT can be predicted by plasma drug concentration PGDE10 (Fig 1C)

Starting dose is determined by animal data as standard First cohort treated as standard with average AUC measured for this cohort Dose escalation occurs according to distance to target AUC either: Initially by a factor equal to the square root of the ratio of the target AUC to the AUC associated with the initial dose and subsequently follow a Fibonacci scheme; OR Initially by a factor of 2 until AUC is 40% of target AUC and subsequently follows Fibonacci scheme

Interpatient variability may limit dose escalation Differences between species used for dose estimation and humans may affect utility of method

continued on page 203 202 Cancer Control

July 2014, Vol. 21, No. 3

Table. — Selected Examples of Trial Escalation Designs (continued) Trial Design

Method

Comments

Model-Based Designs Key Points Establishes dose-toxicity curve prior to patient enrollment Uses toxicity data from enrolled patients to modify curve as study proceeds Requires good biostatistical support for constructing and updating dose-toxicity estimates CRM17

A target level of toxicity defined at baseline. Increased dose levels defined with initial expectations of the probability of DLT at these doses to construct a statistical dose-toxicity model Single patient cohorts; fixed sample size With treatment of successive patients the statistical model is recalculated using Bayesian principles to update estimated probability of a DLT and increase certainty associated with dose-toxicity relationship Dose associated with the target DLT rate according to the final dose-toxicity model at trial completion is defined as the MTD If a patient experiences no toxicity, dose may be escalated to the next dose level for subsequent cycles

May overestimate dose for MTD Uncertainty about toxicity of investigational agent may be reflected in initial dose-toxicity model

Modified CRM11,12

Conservative starting dose, as with 3 + 3 design Dose escalation may only occur by a single dose level per patient cohort Following DLT, the dose for the next patient may not be escalated Cohorts may be larger than single patient Stopping rule defined rather than fixed sample size

Safety improved compared with CRM

EWOC13 (Fig 1D)

Dose-toxicity curve modeled to minimize the probability a patient will be treated at an unacceptably high dose, ie, a dose where the probability of a DLT is greater than some value

Dose-toxicity curve constantly remodeled requiring additional statistical support

TITE-CRM14

Data from all treated patients, including partial data, incorporated into dose-toxicity curve and subsequent dose calculations Patients experiencing DLT are fully weighted Patients not experiencing toxicity are weighted by the proportion of time observed on study

Allows toxicity information of patients to contribute to dose recommendation before all patients are fully followed

mTPI18

Bayesian framework used to calculate posterior probabilities of intervals, reflecting relative distance between toxicity rate of each dose level to the target probability with a fixed sample size Dose-escalation decisions depend on category of toxicity rate: categorization as underdosing results in dose escalation and overdosing results in de-escalation When toxicity rate is categorized as proper dose, dose is not modified

Software provided online Fewer patients treated at doses above MTD

Mixed-effect POM15

Repeated measurements of graded toxicities are used to generate per-cycle toxicity estimates Modified definition of RP2D: dose associated with a predefined probability of severe toxicity per cycle

May be more useful for agents with chronic toxicities (eg, MTAs)

Fractional dose-finding methods16

May be applied to 3 + 3 design or CRM Time to toxicity modeled using Kaplan-Meier estimator. Mass of each censored observation redistributed to the right Censored observation is taken if patient has not experienced a DLT at the time of observation. Fractional contribution of each patient used in dose-escalation decisions

May shorten duration of trial while maintaining accurate determination of RP2D

AUC = area under the curve, CRM = continuous reassessment method, DLT = dose-limiting toxicity, EWOC = escalation with overdose control, MTA = molecularly targeted agent, MTD = maximum tolerated dose, mTPI = modified toxicity probability interval, PGDE = pharmacologically guided dose escalation, PK = pharmacokinetic, POM = proportional odds model, RP2D = recommended phase 2 dose, TITE = time to event.

July 2014, Vol. 21, No. 3

Cancer Control 203

Phase 1 Design Elements

based designs achieve better estimations of the target probability of a DLT at the RP2D while minimizing suboptimal dosing.19 These designs require biostatistics expertise and adequate software for modeling. By contrast to cytotoxic chemotherapy, which is generally delivered intermittently and for a predefined duration, MTAs are often continuously dosed until the development of resistance and, thus, chronic or late toxicity can emerge. Recently, model-based methods have been employed to incorporate these factors into dose-escalation decisions. The modified toxicity probability interval,20 proportional odds model (POM),21 mixed-effect POM,15 and fractional dose-finding methods16 utilize toxicity information from treated patients, including those who have not experienced a DLT at the time of observation, to accurately reflect the ongoing effects of the agents under investigation.

Critical aspects of a phase 1 study include the accurate determination of RP2D, a comprehensive assessment of toxicity, and efficiency in time to study completion, patient accrual, and logistical requirements. Study designs that can precisely determine RP2D and drug toxicity profile, shorten trial duration, subject fewer patients to subtherapeutic or overly toxic doses, and place fewer demands on clinical and trial resources are preferred. Determining the Recommended Dose Beyond statistical simulations, sparse clinical data exist that compare the accuracy of different dose-escalation schemes in predicting the correct RP2D. In a review from 1990 to 2012 of registration trials of cancer drugs approved by the US Food and

A

Standard 3 + 3 Design

B

C

Pharmacokinetically Guided Dose-Escalation Design

D

Accelerated Titration Design

Adaptive Model-Based Design

Fig 1. — (A) Schematic of the standard 3 + 3 design. (B) Schematic of the rule-based accelerated titration design. (C) Schematic of the pharmacokinetically guided dose-escalation design. (D) Schematic of an adapative model-based design (eg, escalation with overdose control). DLT = dose-limiting toxicity, EU = European Union, MAD = maximum administered dose, MTD = maximum tolerated dose, PK = pharmacokinetics, RP2D = recommended phase 2 dose, US = United States. 204 Cancer Control

July 2014, Vol. 21, No. 3

Drug Administration and their corresponding phase 1 studies, the registered doses of the drugs were within 20% of the RP2D in 73% (62 of 85) of the matched phase 1 trials.22 An important caveat reported in this study was that phase 1 trials of MTAs performed poorly in predicting the registered dose to within 20% of the RP2D when compared with phase 1 trials of cytotoxic treatments (odds ratio = 0.2 [0.03–0.8], P = .025).22 The type of dose escalation employed in these phase 1 studies was not reported, and uncertainty remains as to whether the registered dose is the optimal dose of drug. Furthermore, drugs that failed to achieve registration were not included; thus, their impact on the results is unknown. The broad implication of this study is that, overall, phase 1 trials are reasonably accurate in predicting the correct RP2D in most — but not all — cases. Given that most phase 1 trials utilize the 3 + 3 design, it is plausible to extrapolate that these figures reflect the accuracy of this type of methodology in most instances. However, the lack of trial design details from this review renders it difficult to draw definitive conclusions. These empirical data contrast with the many statistical simulations that report the inaccuracy of the 3 + 3 method compared with novel rule- or model-based strategies.11,17,23-26 In a recent comparative simulation study of the modified toxicity probability interval and the 3 + 3 designs using matched sample sizes, the former was found to reliably select the true MTD in 32 out of 42 scenarios.18 The reason for the discrepancy between empirical and statistical simulation data is unclear, but many factors can influence the RP2D that may not be incorporated into the simulation; in addition, computational modeling has its own set of limitations. Rather than ascribing a particular escalation scheme as the most appropriate for all situations, it is more informative for clinical trialists to determine the contexts in which specific methods perform better or are more fit-for-purpose. A key principle of any phase 1 study is to escalate slowly in the face of toxicity and rapidly in the setting of minimal or no adverse events. In situations in which the toxicity of a drug is uncertain or preclinical testing suggests that the therapeutic window is narrow, then a conservative 3 + 3 escalation may be reasonable. By contrast, rapid escalation designs using either novel rule- or model-based schemes would be appropriate if the expected toxicity was low or high-quality preclinical data indicated a wide therapeutic index. Failing to accurately define the true RP2D of a drug may compromise further drug development by testing a subtherapeutic dose and schedule in phase 2 or 3 trials. Agents not successfully registered may be either intrinsically ineffective or not evaluated at the correct RP2D. July 2014, Vol. 21, No. 3

Determining the Relevant Toxicities A critical assumption with the early-phase drug development of cytotoxic agents is that toxicity and efficacy monotonically increase with dose. In the era of MTAs, this assumption is challenged by the premise of differential target saturation of the drug in tumor versus normal tissue, whereby target blockade can become saturated in tumor, thus leading to a plateau of antitumor effect. However, toxicity may continue to rise with an increasing drug dose due to unsaturated targets in normal tissue. In addition, off-target effects may occur such that dosing to toxicity for MTAs may not be appropriate. Despite this, determining the toxicity level remains a critical goal in all phase 1 trials, regardless of the dose-escalation method. It has been reported that around 70% of all clinically significant adverse events are identified in the phase 1 study.22 The probability of observing an adverse event related to a study drug is increased by using a larger sample size. A trial has a 90% chance of detecting an adverse event occurring in 5% of patients with a sample size of 57 patients, a rate that increases to 99% with 82 patients.27 Increasing the sample size of a trial can be achieved by using prespecified expansion cohorts after the dose-escalation component of the trial is completed. Often referred to as a phase 1b or dose-expansion phase study, this part of the trial typically tests a particular dose and schedule, which may be restricted to a population of patients with specific tumor types, molecular aberrations, or some other criteria.28 Study Size, Timing, and Other Logistics Several factors can affect the efficiency of a phase 1 trial (eg, number of patients enrolled and the fraction of those treated at subtherapeutic doses, trial duration, number of dose levels tested). Statistical simulations have attempted to assess the efficiency of various dose-escalation designs.28,29 In a comparative simulation of 3 + 3 and CRMs, the latter treated fewer patients at subtherapeutic doses and, depending on the true MTD, required a smaller sample size than the 3 + 3 design.31 However, real-world experience does not appear to demonstrate substantial differences in efficiency across dose-escalation methods, possibly because of neutralization by other factors. In a retrospective review of 81 trials evaluating 60 MTAs, the ratio of MTD or MAD to starting dose was calculated for each trial and compared with the number of dose levels tested.32 Trials with a high MTD or MAD to starting dose ratio (presumably representing trials of agents with wider therapeutic indices) versus those with a low ratio (presumably representing trials of agents with narrower therapeutic indices) had similar numbers of tested dose levels. This phenomenon was not explained by the use of different dose-escalation Cancer Control 205

methods. Instead, it appears that higher increments between dose levels were selected for agents deemed to have wide therapeutic indices based on preclinical data, regardless of the dose-escalation method used.32 Furthermore, in a review of 84 phase 1 trials of MTAs from 2000 to 2010, trial size and the number of patients treated at doses below MTD were similar across 3 + 3, accelerated titration, and CRM designs.33 One criticism of the conventional 3 + 3 design is that the escalation is unnecessarily slow, thus prolonging the duration of the trial.4 Similar complaints have been leveled at the CRM, because results from the last patient are required to dose subsequent patients.11 Modifications to CRM and accelerated titration designs have sought to address these concerns, and simulations have demonstrated that they speed up the completion of a phase 1 study but also provide more information.4 However, clinical data to substantiate these simulations are sparse. It is important to recognize that the 3 + 3 method does not require statistical software support or modeling, and it is widely acknowledged as being uncomplicated and safe to implement.6 Processes underlying model-based designs are often perceived by clinicians as being nontransparent.34 Furthermore, the software that these systems utilize is often considered difficult for those who are not technically inclined.34 Although adaptations to these models have been suggested, such logistical barriers may limit their widespread implementation in phase 1 studies.

Challenges to Phase 1 Designs

Delayed or Chronic Toxicities Many MTAs or biological agents are chronically administered as either continuous oral or frequent intravenous doses, thus predisposing patients to delayed or chronic toxicities. These cumulative adverse events can affect the tolerability of the drug and should be considered when recommending a phase 2 dose. Assessing for DLTs in the initial cycle or first few weeks of treatment will not account for late or cumulative adverse events. In a retrospective review of 36 phase 1 trials of MTAs, approximately one-half of severe toxicities occur after cycle 1.35 In an initiative to redefine criteria for determining the DLT, adverse events from more than 2,000 patients, representing 54 monotherapy MTA studies, concluded that 49.5% of patients experienced their first grade 3 or higher toxicity after cycle 1, and a significant proportion of patients required a dose reduction for selected grade 2 or lower events as early as cycle 1.36 Several designs have been proposed to account for chronic adverse events, including TITE-CRM, mixed-effect POM, and fractional dose-finding methods.15,16 These latter designs have been tested in statistical simulations but have not been widely applied in 206 Cancer Control

clinical trials. In studies of drugs with delayed toxicities, the TITE-CRM and 3 + 3 design were compared utilizing the Monte Carlo simulation.37 The former was reported to result in trials of shorter length and a higher number of patients treated at or around the MTD. When the TITE-CRM was incorporated into 2 phase 1 studies of concurrent chemoradiotherapy for pancreatic cancer, these trials failed to reproduce the predicted methodological benefits, supporting the notion that all designs must be implemented into actual trials before their alleged advantages can be accepted.38,39 Nontoxicity End Points and Optimal Biological Dosing Some MTAs have different dose–effect relationships than traditional cytotoxics; therefore, dosing these agents to toxicity may not be appropriate. In a review of 24 phase 1 trials of MTAs, 683 patients were retrospectively assigned to a cohort based on a comparative dose to the MTD (low [≤ 25% MTD], medium [25%–75% MTD], or high [≥ 75% MTD]), and their outcomes were compared.40 No significant difference in response rate or survival was seen across the cohorts, implying that these agents have a nonmonotonic dose-efficacy curve. Under this assumption, escalating doses until predefined PK or PD parameters demonstrate when a target is saturated, or when a biological pathway is maximally altered, would be an alternative to the traditional toxicity end point. This type of strategy has been termed optimal biological dosing (OBD) and is defined as the dose that produces the most favorable prespecified effect on a biomarker. However, toxicity remains the predominant end point as evidenced by several reviews of phase 1 trials, which have reported that the proportion of studies using nontoxicity end points ranges from 24% to 48%.6,7,22 Because the 3 + 3 design typically uses toxicity to guide dose escalation, it is unlikely that such a design would be optimized to ascertain the OBD. The use of real-time PK monitoring in the PGDE method makes it an appealing strategy to determine the OBD. Several adaptive dose-finding designs have been proposed to identify the OBD. The nonparametric and semi-parametric methods have been proven to have the best operating characteristics via computational testing and are recommended for use.41 However, to date, no trial reporting an OBD has utilized either adaptive methodology. Given the enthusiasm surrounding OBD, it is anticipated that more trials utilizing novel trial designs will emerge, providing a deeper understanding of which escalation methods are the most appropriate.42 Combination Therapy The development of treatment resistance has long been recognized with cytotoxic therapies. To circumvent this phenomenon, combination chemotherapy regimens July 2014, Vol. 21, No. 3

with non–cross-reactive compounds, administered either concurrently or sequentially, were tested.43 This rationale also forms the basis for testing MTA combinations. Several unbiased genome sequencing studies have shown that many tumors have a complex molecular profile of low frequency mutations.44 Single-agent MTAs have a limited benefit in patients with advanced cancer, with progressive disease eventually developing in all patients, even among those with significant or prolonged responses.45 Complex tumor biology implicates the involvement of multiple interconnected signalling networks that drive cancer initiation and metastasis.46,47 Thus, rationally designed combination MTA therapy is an attractive strategy to prevent — or, at the very least, delay — the emergence of treatment resistance by perturbing multiple cellular networks, potentially thwarting the development of molecular escape pathways.45 Major concerns in administering these combination regimens include the cumulative impact of overlapping toxicities, the potentiation of adverse events due to drug–drug interactions, and the occurrence of unexpected adverse events.48,49 As with delayed toxicities and OBD, no single dose-escalation method has proven to be superior for evaluating combination regimens in clinical testing. Combination studies are more complex because a number of schedules may be recommended, and different schedules can produce the same level of toxicity. These types of trials often have RP2D and toxicity information from the monotherapy phase 1 trials of each agent that can be utilized by either rule- or model-based designs, possibly improving the efficiency of the trial. However, preclinical data describing PK or PD interactions or potential synergy with regard to antitumor activity or toxicity may be lacking. Given that the MTD of each drug in the combination is known, it is unlikely that either rule- or model-based designs will escalate patients to overly toxic dose levels. The challenge facing both types of designs is the optimal method by which to escalate each drug. Careful selection of starting doses and subsequent dose levels is crucial, regardless of the escalation method. In rule-based methods, the agents can be escalated to prespecified doses either sequentially, in parallel, or with one drug fixed at either a high or low dose while the other is increased toward the RP2D. The zone method illustrates a rule-based escalation method in which consecutive dosing zones are comprised of sequentially increasing dose combinations and patients are randomized to these preset doses within an individual zone and escalation decisions are made using a modified 3 + 3 method.50 These dose levels are then compared and the combinations that are either too toxic or unlikely to be efficacious are eliminated. Statistical simulations have reported that this design requires a smaller sample size, has better July 2014, Vol. 21, No. 3

power, and patients are more likely to be treated at efficacious doses compared with conventional escalation methods.50 Various Bayesian parametric models have been designed and they purport to recommend a more accurate dose and schedule for combination regimens.51-53 As with monotherapy trials, the escalation method selected for these types of studies should be individualized and guided by the anticipated level of toxicity from the combination.

Conclusion This article is neither a defense nor homage to the standard 3 + 3 design, but rather it is a review of the clinical evidence. The conventional escalation method is slow, inaccurate in recommending doses, and enrolls a significant proportion of patients at subtherapeutic doses. However, sparse empirical data exist to support or refute the assertion that any one trial design is superior to another. The evidence from this review suggests that the 3 + 3 design identifies the recommended phase 2 dose and toxicities with an acceptable level of precision in some circumstances, while being straightforward to operate with few logistical demands. Newer escalation methods have had limited incorporation into the clinical trial setting and, consequently, have had little opportunity to demonstrate favorable clinical application beyond statistical simulations. Although such designs may have clinical utility, simulations performed under ideal circumstances may not reflect the true clinical reality in which, for example, protocol deviations for dose escalations are permitted due to specific patient-, institutional-, or treatment-related factors. Thus, novel trial designs demonstrating superiority over the 3 + 3 method in statistical simulations without corroborating clinical evidence are of theoretical value alone. Following the explosion of molecularly targeted agents under investigation, delayed toxicities, optimal dosing, and combination treatments have presented challenges to the design of phase 1 trials. Time-toevent designs may be the most suitable for testing agents with delayed or chronic toxicities; however, recently proposed designs, such as the mixed-effect proportional odds model, may prove appropriate with clinical experience. The pharmacokinetically guided dose-escalation scheme is a strategy that may account for optimal biological dosing, although the ability to confidently define optimal biological activity in the phase 1 setting remains elusive. A study of an investigational drug or combination with an uncertain or high probability of producing serious adverse events should adopt a conservative escalation scheme such as the 3 + 3 design. By contrast, studies of agents predicted to have less toxicity can be quickly escalated with either an accelerated titration- or model-based design until an acceptable toxicity range is reached. Regardless of Cancer Control 207

the type of agent or setting, the guiding principles of safe starting dose selection, minimizing the number of patients treated at subtherapeutic doses, rapid escalation in the absence of toxicity, and slow escalation in the presence of adverse events will ensure the sound design of any phase 1 trial. Therefore, although the 3 + 3 design may not be the best in all circumstances, in this context it should not be abandoned and may still have a place in the design of phase 1 studies. References 1. Storer BE. Design and analysis of phase I clinical trials. Biometrics. 1989:45(3):925-937. 2. Dixon WJ, Mood AM. A method for obtaining and analyzing sensitivity data. J Am Stat Assoc. 1948;43(241):109-126. 3. Reiner E, Paoletti X, O’Quigley J. Operating characteristics of the standard phase I clinical trial design. Comput Stat Data Anal. 1999;30(3):303-315. 4. Simon R, Rubinstein L, Arbuck SG, et al. Accelerated titration designs for phase I clinical trials in oncology. J Natl Cancer Inst. 1997;89(15):1138-1147. 5. Rogatko A, Schoeneck D, Jonas W, et al. Translation of innovative designs into phase I trials. J Clin Oncol. 2007 2007;25(31):4982-4986. 6. Le Tourneau C, Lee JJ, Siu LL. Dose escalation methods in phase I cancer clinical trials. J Natl Cancer Inst. 2009;101(10):708-720. 7. Parulekar WR, Eisenhauer EA. Phase I trial design for solid tumor studies of targeted, non-cytotoxic agents: theory and practice. J Natl Cancer Inst. 2004;96(13):990-997. 8. Le Tourneau C, Razak AR, Gan HK, et al. Heterogeneity in the definition of dose-limiting toxicity in phase I cancer clinical trials of molecularly targeted agents: a review of the literature. Eur J Cancer. 2011;47(10):1468-1475. 9. Eisenhauer EA, Twelves C, Buyse ME. Phase 1 Cancer Clinical Trials: A Practical Guide. New York: Oxford University Press; 2006. 10. Collins JM, Zaharko DS, Dedrick RL, et al. Potential roles for preclinical pharmacology in phase I clinical trials. Cancer Treat Rep. 1986;70(1):73-80. 11. Goodman SN, Zahurak ML, Piantadosi S. Some practical improvements in the continual reassessment method for phase I studies. Stat Med. 1995;14(11):1149-1161. 12. Faries D. Practical modifications of the continual reassessment method for phase I cancer clinical trials. J Biopharm Stat. 1994;4(2):147-164. 13. Babb J, Rogatko A, Zacks S. Cancer phase I clinical trials: efficient dose escalation with overdose control. Stat Med. 1998;17(10):1103-1120. 14. Cheung YK, Chappell R. Sequential designs for phase I clinical trials with late-onset toxicities. Biometrics. 2000;56(4):1177-1182. 15. Doussau A, Thiébaut R, Paoletti X. Dose-finding design using mixedeffect proportional odds model for longitudinal graded toxicity data in phase I oncology clinical trials. Stat Med. 2013;32(30):5430-5447. 16. Yin G, Zheng S, Xu J. Fractional dose-finding methods with late-onset toxicity in phase I clinical trials. J Biopharm Stat. 2013;23(4):856-870. 17. O’Quigley J, Pepe M, Fisher L. Continual reassessment method: a practical design for phase 1 clinical trials in cancer. Biometrics. 1990;46(1):33-48. 18. Ji Y, Wang S-J. Modified toxicity probability interval design: a safer and more reliable method than the 3 + 3 design for practical phase I trials. J Clin Oncol. 2013;31(14):1785-1791. 19. Mick R, Ratain MJ. Model-guided determination of maximum tolerated dose in phase I clinical trials: evidence for increased precision. J Natl Cancer Inst. 1993;85(3):217-223. 20. Westblade LF, Jennemann R, Branda JA, et al. Multicenter study evaluating the Vitek MS system for identification of medically important yeasts. J Clin Microbiol. 2013;51(7):2267-2272. 21. Van Meter EM, Garrett-Mayer E, Bandyopadhyay D. Proportional odds model for dose-finding clinical trial designs with ordinal toxicity grading. Stat Med. 2011;30(17):2070-2080. 22. Jardim DL, Hess KR, Lorusso PM, et al. Predictive value of phase I trials for safety in later trials and final approved dose: analysis of 61 approved cancer drugs. Clin Cancer Res. 2014;20(2):281-288. 23. O’Quigley J. Another look at two phase I clinical trial designs. Stat Med. 1999;18(20):2683-2692. 24. Korn EL, Midthune D, Chen TT, et al. A comparison of two phase I trial designs. Stat Med. 1994;13(18):1799-1806. 25. Neuenschwander B, Branson M, Gsponer T. Critical aspects of the Bayesian approach to phase I cancer trials. Stat Med. 2008;27(13):2420-2439. 26. Berry SM, Carlin BP, Lee JJ, et al. Bayesian Adaptive Methods for Clinical Trials. Boca Raton, FL: CRC Press; 2010. 27. DeMichele A, Berry DA, Zujewski J, et al. Developing safety criteria for introducing new agents into neoadjuvant trials. Clin Cancer Res. 2013;19(11):2817-2823. 28. Manji A, Brana I, Amir E, et al. Evolution of clinical trial design in early drug development: systematic review of expansion cohort use in single-agent phase I cancer trials. J Clin Oncol. 2013;31(33):4260-4267. 208 Cancer Control

29. Eckhardt S, Siu L, Clark G, et al. The continual reassessment method (CRM) for dose escalation in phase I trials in San Antonio does not result in more rapid study completion. Paper presented at the Annual Meeting of the American Society for Clinincal Oncology, Chicago, Illinois, 1999. Abstract 627. 30. Walling J, Zervos P, McCarthy S, et al. Dose escalation methodology in phase I clinical trials: a comparison of the modified continual reassessment method (MCRM) and a traditional method. Experience with the multitargeted antifolate (MTA). Paper presented at the Annual Meeting of the American Society for Clinincal Oncology, Chicago, Illinois, 1997. 31. Iasonos A, Wilton AS, Riedel ER, et al. A comprehensive comparison of the continual reassessment method to the standard 3 + 3 dose escalation scheme in phase I dose-finding studies. Clin Trials. 2008;5(5):465-477. 32. Le Tourneau C, Stathis A, Vidal L, et al. Choice of starting dose for molecularly targeted agents evaluated in first-in-human phase I cancer clinical trials. J Clin Oncol. 2010;28(8):1401-1407. 33. Le Tourneau C, Gan HK, Razak AR, et al. Efficiency of new dose escalation designs in dose-finding phase I trials of molecularly targeted agents. PloS One. 2012;7(12):e51039. 34. Sweeting M, Mander A, Sabin T. Bcrm: Bayesian continual reassessment method designs for phase I dose-finding trials. J Stat Software. 2013;54(13). http://www.jstatsoft.org/v54/i13/paper. Accessed April 11, 2014. 35. Postel-Vinay S, Gomez-Roca C, Molife LR, et al. Phase I trials of molecularly targeted agents: should we pay more attention to late toxicities? J Clin Oncol. 2011;29(13):1728-1735. 36. Postel-Vinay S, Le Tourneau C, Olmos D, et al. Towards new methods for the determination of dose limiting toxicities and recommended dose of molecularly targeted agents. Presented at the European Cancer Congress (ECCO-ESMO-ESTRO), Amsterdam, The Netherlands, 2013. 37. Normolle D, Lawrence T. Designing dose-escalation trials with lateonset toxicities using the time-to-event continual reassessment method. J Clin Oncol. 2006;24(27):4426-4433. 38. Muler JH, McGinn CJ, Normolle D, et al. Phase I trial using a time-toevent continual reassessment strategy for dose escalation of cisplatin combined with gemcitabine and radiation therapy in pancreatic cancer. J Clin Oncol. 2004;22(2):238-243. 39. Desai SP, Ben-Josef E, Normolle DP, et al. Phase I study of oxaliplatin, full-dose gemcitabine, and concurrent radiation therapy in pancreatic cancer. J Clin Oncol. 2007;25(29):4587-4592. 40. Jain RK, Lee JJ, Hong D, et al. Phase I oncology studies: evidence that in the era of targeted therapies patients on lower doses do not fare worse. Clin Cancer Res. 2010;16(4):1289-1297. 41. Zang Y, Lee JJ, Yuan Y. Adaptive designs for identifying optimal biological dose for molecularly targeted agents. http://odin.mdacc.tmc.edu/~yyuan/ Software_release/TargetAgent/targetAgentDF.pdf. Accessed April 16, 2014. 42. Kummar S, Gutierrez M, Doroshow JH, Murgo AJ. Drug development in oncology: classical cytotoxics and molecularly targeted agents. Br J Clin Pharmacol. 2006;62(1):15-26. 43. Goldie JH, Coldman AJ. A mathematic model for relating the drug sensitivity of tumors to their spontaneous mutation rate. Cancer Treat Rep. 1979;63(11-12):1727-1733. 44. Vogelstein B, Papadopoulos N, Velculescu VE, et al. Cancer genome landscapes. Science. 2013;339(6127):1546-1558. 45. Dancey JE, Chen HX. Strategies for optimizing combinations of molecularly targeted anticancer agents. Nat Rev Drug Discov. 2006;5(8):649-659. 46. Fidler IJ, Gersten DM, Hart IR. The biology of cancer invasion and metastasis. Adv Cancer Res. 1978;28:149-250. 47. Hanahan D1, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646-674. 48. Dy GK, Adjei AA. Understanding, recognizing, and managing toxicities of targeted anticancer therapies. CA Cancer J Clin. 2013;63(4):249-279. 49. Park SR, Davis M, Doroshow JH, et al. Safety and feasibility of targeted agent combinations in solid tumours. Nat Rev Clin Oncol. 2013;10(3):154-168. 50. Huang X, Biswas S, Oki Y, et al. A parallel phase I/II clinical trial design for combination therapies. Biometrics. 2007;63(2):429-436. 51. Yin G, Li Y, Ji Y. Bayesian dose-finding in phase I/II clinical trials using toxicity and efficacy odds ratios. Biometrics. 2006;62(3):777-787. 52. Thall PF, Millikan RE, Mueller P, et al. Dose-finding with two agents in phase I oncology trials. Biometrics. 2003;59(3):487-496. 53. Yuan Y, Yin G. Sequential continual reassessment method for twodimensional dose finding. Stat Med. 2008;27(27):5664-5678.

July 2014, Vol. 21, No. 3

Health care professionals and researchers must work together to recognize and overcome barriers to clinical trial enrollment among elderly patients.

Cattails_0269. Photograph courtesy of Henry Domke, MD. www.henrydomke.com

Participation of the Elderly Population in Clinical Trials: Barriers and Solutions Aaron C. Denson, MD, and Amit Mahipal, MD Background: Despite the fact that cancer disproportionately affects the elderly, most participants of clinical cancer trials are relatively young. This misrepresentation greatly affects the oncology treatment of the elderly population (> 65 years of age). Few studies have attempted to identify the problems related to discrepancy based on age for clinical trial participation. Methods: A literature review was performed to identify barriers and solutions to enrollment of elderly persons for clinical cancer trials. Results: Physician-related barriers include perception about treatment tolerance, drug metabolism, a lack of evidence for efficacy, and age bias. Lack of autonomy, concerns about quality of life and toxicities, accessibility to clinical trials, and logistical and financial difficulties are common patient-related barriers. Trial-related barriers include eligibility criteria based on performance status, organ dysfunction, and the presence of comorbidities. Solutions, such as improved communication, and coordinating logistical challenges may help overcome some of these challenges. Studies designed for the geriatric population could modify the perception and behavior of health care professionals and improve patient participation in clinical trials. Conclusions: Implementing some of these solutions and increased research may help overcome shortfalls in elderly enrollment, thus allowing for more effective treatment of older patients.

Introduction

From the Department of Internal Medicine at the University of South Florida Morsani College of Medicine, Tampa, Florida (ACD), and the Department of Gastrointestinal Oncology and the Clinical Research Unit at the H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida (AM). Submitted December 19, 2013; accepted January 10, 2014. Address correspondence to Amit Mahipal, MD, Clinical Research Unit, H. Lee Moffitt Cancer Center, 12902 Magnolia Drive, MCC-3177, Tampa, FL 33647. E-mail: [email protected] No significant relationship exists between the authors and the companies/organizations whose products or services may be referenced in this article. July 2014, Vol. 21, No. 3

Cancer is the second most common cause of death in the United States, exceeded only by heart disease, and accounts for nearly 1 out of 4 deaths.1 The total cost of cancer treatment by the year 2020 is projected to be as high as $173 billion, which represents a 39% increase from 2010.2 Finding innovative and cost-effective ways to treat patients with cancer is more relevant than ever due to the increasing age of the American population.3 It is estimated that the elderly population (defined as age ≥ 65 years) in the United States will reach 70 million by 2030.3 As the “baby boomer” generation gets older, the US health care system will be stressed by the increasing age of the population.4 Cancer Control 209

Table. — Selected Barriers and Solutions Patient-Related

Physician-Related

Trial-Related

Barriers

Logistics Finances Lack of understanding of benefits Autonomy

Perceptions Culture Complex pharmacokinetics/pharmacodynamics Lack of evidence

Strict inclusion criteria Poor methods for evaluating functional status Lack of funding dedicated to elderly population

Solutions

Provide transportation Provide lodging Research nurses, trial coordinators Improved communication

Elder-focused studies Improved communication Increase physician training in geriatrics specialty

Create geriatric-focused trials Increase/fund studies of elderly population

Breakthroughs in cancer treatment require many steps, including large randomized clinical trials to evaluate dosing, safety, and effectiveness. A total of 2% to 4% of patients with cancer are enrolled in a clinical trial, most of whom are younger than 65 years of age.5 The overall proportion of older patients (> 65 years of age) in trials conducted by the Southwest Oncology Group was 25%.6 This same age group accounts for 63% of new cases of cancer in the United States, while comprising 14% of the entire US population.6 Cancer is a disease that predominantly affects the elderly, yet patients older than 65 years are not well represented in clinical trials.6,7 This misrepresentation in clinical trials impacts the care of elderly patients.3,8,9 Multiple reasons exist for this discrepancy (Table). Compared with their younger counterparts, elderly people have increased rates of comorbidities and complications.10 Elderly patients also undergo complicated physiological and pathological changes and have higher rates of drug interactions, which may complicate treatment.11 Furthermore, the perceptions of health care professionals and family members can make it difficult for elderly patients to participate in clinical trials.11 Other obstacles to enrollment include trial eligibility criteria, financial, and logistic issues. Such enrollment barriers for older patients have gained interest in the last 20 years.8,12-14 Although numerous solutions have been proposed, those that have been implemented have had limited success. It has been suggested that a larger, more succinct effort is necessary from the medical community.8,9,13,14

Methods The goals of this review are to clarify possible barriers and solutions to participating in clinical trials among the elderly. To do this, we reviewed the medical literature from the last 10 years through PubMed, using the keywords “cancer,” “clinical trials,” and “elderly.” Additional published studies were included based on manual searches as well as references from reviews or original articles. Articles included were original research primarily focused on the barriers or solu210 Cancer Control

tions faced by elderly patients with cancer with regard to clinical trial enrollment. It is worth noting that only publications in English were reviewed and may not be representative of situations seen outside of English-speaking countries.

Barriers

Physician-Related The understanding of physician-related barriers to elderly enrollment is evolving and becoming increasingly complex. Some studies have implicated physician fears of possible toxicity and comorbidity interactions in the elderly as the main barriers to enrollment, suggesting that response to treatment has been poorly understood due to low numbers of elderly patients in clinical trials.14,15 In fact, data on toxicity are limited for patients older than 75 years, and a paucity of data exists for patients older than 80 years of age.16-18 However, several studies demonstrate that treatment tolerance in clinical trials is similar across different age groups,19,20 while others suggest that age bias among physicians is an independent barrier to enrollment among the elderly patient population.21,22 One study reported that 11% of physicians explicitly stated age as a reason for not enrolling a patient in a clinical trial.21 Perceptions among physicians about age and clinical trial participation are complicated.23 Physicians acknowledge that they recommend less aggressive treatment to elderly patients, yet their reasonings are multifactorial and not well understood.23 The biological changes of aging, in addition to the complex pharmacokinetics involved, require more extensive care when administering systemic therapy, and such a perception about treatment tolerance may decrease elderly patient enrollment in clinical trials.17 Findings from 1 study demonstrated that, among patients eligible for the study, clinical trial participation was discussed with 76% of those younger than 65 years of age and 58% of patients older than 65 years,21 and advanced age may deter oncologists from choosing intensive cancer therapy, even if patients are highly functional and lack comorbidities.21,23 Although this July 2014, Vol. 21, No. 3

disparity has been documented, the reasons for this gap have not been well elucidated, but possible barriers to enrollment may be physician fear of a lack of survival benefit among older patients.21 However, a paucity of evidence supports the belief that elderly people do not derive benefit from participating in a clinical trial. It remains a matter of controversy that physician bias could be based on toxicity difference or age alone, but both of these factors are likely to play a role as a barrier to clinical trial participation among the elderly.21 Patient-Related When patients eligible for a clinical trial consider whether to participate, many cite a lack of autonomy over treatment choice as being a reason for foregoing participation.24 Many patients also want to choose their own treatment and fear that their participation in a clinical trial will mean they will lose that decision-making capacity. Increased survival was less important than improved quality of life for older patients when declining trial enrollment.24 These preferences illustrate a cultural or philosophical difference that may exist between age groups, which may be a barrier, yet it is difficult to describe.24 Elderly patients also cited other reasons for declining to enroll in a clinical trial, including such concerns as adverse events, friends who oppose participation, or a belief that participation in a clinical trial would provide no value to future generations.21 Family opposition to enrollment is a more important issue for older patients compared with their younger counterparts. The role of altruism in trial participation has also been explored, but conclusions about altruism are difficult to make.25 Moreover, compared with their younger counterparts, elderly patients are less likely to actively seek participation and less informed of the availability of clinical trials, which may be related to the differences seen in literacy rates among the age groups.24 Other cultural implications from both patients and physicians may also be present and are difficult to assess but do lead to decreased trial enrollment.24 When patients were surveyed, the reason given for enrolling in a trial was “I trusted the doctor treating me,” although it was also given as a reason for those who chose not to enroll in a trial.25 Patient perceptions of trial efficacy appear to play a role in patient enrollment. One study found that 44% of patients who declined enrollment did so when offered participation in a trial comparing standard treatment with a novel agent.25 Patients surveyed reported a higher likelihood to enroll in trials that included standard treatment with or without the addition of a novel drug. As many as 20% of surveyed patients reported that they chose to enroll in a trial because they felt that the trial provided the best treatment options.25 July 2014, Vol. 21, No. 3

In addition, some patients felt that they must receive at a minimum standard treatment and said they could not afford to be assigned to a placebo group. Logistical barriers present another obstacle to enrollment in clinical trials among the elderly, making a difficult-to-recruit population even smaller.26 Participation in a clinical trial may require patients to travel to cancer or academic centers, and elderly persons may have a smaller support network than their younger counterparts; however, in 1 study, this factor was not significantly different between older and younger patients.21 Financial issues may also play a role in elderly patients choosing not to enroll in clinical trials, although no difference was seen among age groups in this study.21 Trial-Related Generally, clinical trials do not limit eligibility based on age alone, but other criteria, including performance status (PS), organ dysfunction, and disease status, may preclude older patients from participating in a study. Previous studies have demonstrated that survival correlates well with PS and comorbidities, and study exclusion criteria often are based on these data.27,28 Although these are logical exclusion criteria, they limit elderly enrollment in clinical trials because older patients generally have a higher number of comorbidities than their younger counterparts. In multiple studies, trial ineligibility was the greatest barrier to clinical trial enrollment among older persons, with both patients and physicians perceiving this barrier as a major obstacle, and up to 60% of elderly patients who did not enroll in a clinical trial stated they failed to do so because of trial unavailability or ineligibility.21,29 One study reported that 65% of patients 65 years of age or older were eligible to participate in the trial compared with 78% of younger patients.21 However, if patients were eligible, trial participation rates did not significantly differ by age (34% for age ≥ 65 years vs 40% for age < 65 years). After considering other factors, overall survival and toxicity rates were similar among the younger and older patients. PS was the most significant determinant of overall 30-day (PS 0–1, 97.5%; PS 2–3, 79%) and 1-year (PS 0–1, 21%; PS 2–3, 9.5%) survival rates (P = .029). Similarly, PS was the most important factor for the development of serious toxicities (P = .034). One study found that the most common reasons for nonenrollment due to ineligibility were poor PS (13.7% [32 patients]), the need for emergent radiotherapy (8.6% [20 patients]), patient refusal (6.0% [14 patients]), geographical issues (4.3% [10 patients]), and insurance issues (4.3% [10 patients]).29 These results support the claim that poor PS is the most common reason for nonenrollment into a clinical trial, while also emphasizing the complex nature of trial Cancer Control 211

enrollment.29 Therefore, criteria should be thoughtfully revised to include these large numbers of patients being excluded because of poor PS and trials should be aimed at elderly patients to increase enrollment among this difficult-to-recruit population.

Solutions

Physician-Related Enrollment in clinical trials relies heavily on physicians, creating many barriers for the elderly.24 Both physician bias and perception have been shown to be impediments to enrollment of older persons in clinical trials.14,26 Therefore, it is imperative to create a cultural shift among oncologists to boost trial enrollment of older patients.26 The factors influencing physician culture can be complex and difficult to manipulate. One commonly cited issue is the lack of data on toxicity and survival rates among elderly patients, a challenge that may enable bias.21 Physicians report that how elderly patients are likely to tolerate a specific treatment has not been well elucidated, a barrier that creates unknown variables favoring conservative treatment for this patient population.21 Because of the low numbers of participation among elderly patients in clinical trials, the problem of lacking data on treatment tolerance is further compounded. Although some studies have demonstrated that age itself does not change tolerance to treatment, additional studies are needed to further clarify this issue.4,27 As more specific data become available, physician attitudes toward trial participation are likely to change. It is important to increase trials specifically targeting elderly patients,14 because older patients with cancer may require more thorough care when instituting systemic therapy compared with younger patients with cancer.21 This is due to the biological changes of aging and uncertainties of the pharmacokinetic profiles of some medications, including chemotherapy, which is a concern common among oncologists that may hinder patient enrollment.21 Studies of pharmacodynamics and pharmacokinetics directed at elderly populations will be important for solving these challenges. Late-stage clinical trials can also stratify patients based on age, and increasing data on the elderly patient population may improve treatment and decrease physician-related barriers. The prognoses of elderly patients referred to a phase 1 study are comparable with the rest of the study population.17 In fact, elderly patients enrolled in phase 1 trials had improved survival rates when compared with elderly patients who did not receive treatment during a phase 1 trial.17 However, some physicians do not perceive clinical trials as being beneficial for their patients.24 Therefore, increasing data on the older population, as well as changing physician perceptions, will be important in 212 Cancer Control

increasing the numbers of trials specifically targeting the elderly, possibly acting as the key to shifting physician attitudes away from age bias. Patient-Related The most common patient-related challenges relate to understanding the benefits of clinical trials and the logistics of clinical trial enrollment.17,30 Solutions to these problems are complex and can be approached in different manners.26 For example, controversy exists as to whether increased patient information will increase levels of enrollment in clinical trials among the elderly.17,26 By contrast, logistical issues have been an easier challenge to address.30 Transportation is difficult among older patients who, compared with younger patients, more frequently require help from a family member or friend for travel. In addition, older patients have increased time requirements for transportation. Housing is also more complicated among elderly patients. Among older persons, a small inconvenience can become a major issue, such as having access to an elevator.30,31 Communication is a key factor in facilitating clinical trial participation, with research indicating that more time is often required for nurses to effectively communicate with the geriatric population, in particular elderly patients who are frail.30 Ensuring that trials are accessible is important with any study population, but this is especially true among older patients. Accessibility can be achieved by providing funding for transportation, housing, and coordination, provided that no ethical dilemmas are presented. Although it is unethical to provide direct monetary incentives for trial enrollment, financial support to offset logistical barriers is considered appropriate.31 Solutions such as home visits and flexible scheduling have also been proposed.30 Additional research staffing may be needed to account for the extra time and resources required for enrolling older patients into clinical trials. Increasing the number of research personnel was rated by oncologists as the most important method to increase trial accrual among the elderly.32 A team approach involving family members, physicians, support staff, and others provides the most effective method to overcoming logistical barriers to patient enrollment.30 An increase in logistical support will be a key feature in attracting more elderly patients to clinical trials, and, although the data on patient-related solutions are sparse, improving logistical support, follow-up methods, and patient education are likely to increase enrollment among this patient population.17 Trial-Related Issues of eligibility and availability to clinical trials continue to be the most obvious trial-related barriers July 2014, Vol. 21, No. 3

to enrollment among elderly patients.30 Therefore, increasing the number of trials aimed at this target population, with protocols specifically written to include elderly patients, will help address these challenges.30 Trial design must adapt to fit the needs of this unique population. For example, assessing patient status through the use of a comprehensive geriatric assessment rather than through traditional methods might improve cancer treatment in the geriatric population.23 Researchers should also aim to create study criteria that allow the inclusion of additional elderly participants without interfering with survival statistics.14,17 Several methods have been identified for evaluating life expectancy rates and functional status among patients with cancer,16,33,34 and, although these improve the ability of researchers to evaluate patient eligibility, these methods must be further studied and refined. There is a trend in the right direction, but more is needed to address the problem.25,29,30 Designers of clinical trials must also anticipate the increased costs and time associated with treating an elderly population.30 Providing extra funding for trials aimed at older populations has the potential to offset these limitations, thus improving data, which could then be used in clinical practice. Members of the team should also have an affinity with older patients and be cognizant that extra time and financial resources will be required when conducting research on frail patients.30 Resource barriers are a key target when considering clinical trials in an elderly population.26 Data are lacking on practical and specific solutions to trial-related barriers, which is indicative of the overall issues limiting enrollment in this population. Increased funding for these studies as well as involving the elderly population in breaching these barriers will be crucial when moving forward.

Discussion What began in 1993 with the National Institutes of Health Revitalization Act was a movement to develop evidence about participation barriers in the elderly population.35 The scientific community has made some efforts to define these barriers, yet solutions continue to be poorly defined.5,8,30 It is obvious that more information is needed to further refine barriers to identify practical and effective solutions to them. The largest physician barrier identified is the culture of the medical community, which is a broad and complex area of study that the health care industry may find difficult to change. Currently, the most promising solution is increasing data on the outcomes and tolerances of various therapeutic regimens among the elderly, including increasing clinical trials or providing alternative methods. Currently, models expanded from the analyses of prospectively obtained information are the most effective strategies for gathering information July 2014, Vol. 21, No. 3

about chemotherapy tolerance in the elderly population.12,19,36 Understanding tumor biology and treatment tolerance could also provide concrete evidence for physicians to base enrollment decisions on, thus decreasing the impact of age bias. Patient barriers continue to be poorly defined and difficult to assess.14,16,24 This is in part due to the cultural and psychological differences among the elderly population; therefore, such issues are more difficult to generalize and may be best managed on an individual level. Increased communication with older patients is key to increasing clinical trial enrollment among the elderly as well as exploring patient-centered barriers to enrollment. Barriers related to trial design are the most studied and well understood of the barriers to patient enrollment. Historically, trials excluded patients on the basis of age alone; however, age is no longer an exclusion criterion for most trials. Trial eligibility is now mostly based on PS, comorbidities, and organ function.5,17,22 However, clinical trials do not distinguish methods for evaluating elderly or younger patient functional status. More research is needed to effectively evaluate elderly patients and target appropriate populations. Trials specifically designed for the elderly have increased in recent years,30,37 yet more data are needed to understand how to treat this patient population. As the number of research teams comfortable with elderly patients increases, the solutions to these problems are likely to become obvious. For example, some researchers have discovered they do not fully understand the implications of working with an elderly population, and specifically designed trials for an older patient population may be more complex than previously thought.26 It will also be important to address logistical barriers to trial participation, which are some of the most obvious and well-known barriers, yet often they become the most difficult to overcome.19,30 Appropriate trial management and allocation of funds are imperative in order to avoid logistical barriers to enrollment.

Conclusions The issues facing our health are constantly changing and increasingly complex. Enrolling diverse patient populations in clinical trials is just one example of the many challenges we face as health care professionals. Focusing clinical trials on the elderly population is of increasing importance, particularly with the aging population of the United States. To address the shortage of older patient enrollment, physicians and patients must work together to recognize and overcome barriers. Physicians should be encouraged to reflect on their own practice and consider changes to help alleviate the shortage of elCancer Control 213

derly patients in clinical trials. Although more research is needed to elucidate these barriers and solutions to enrollment issues among the older patient population, obvious causes should be addressed. Increasing eligibility of elderly patients, decreasing logistical barriers to participation, and instituting a cultural shift are all major improvements that can be immediately enacted. It is imperative that data collected from clinical trials are applicable to the patient population to be treated. Otherwise, health care professionals will continue to treat the majority of patients with cancer based on perceptions and best clinical judgment rather than on conclusive data. References 1. American Cancer Society. Cancer Facts & Figures 2014. Atlanta, GA: American Cancer Society; 2014. 2. 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(2):117-128. 3. Knickman JR, Snell EK. The 2030 problem: caring for aging baby boomers. Health Serv Res. 2002;37(4):849-884. 4. Janssen-Heijnen ML, Houterman S, Lemmens VE, et al. Prognostic impact of increasing age and co-morbidity in cancer patients: a populationbased approach. Crit Rev Oncol Hematol. 2005;55(3):231-240. 5. 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(6):1728-1733. 6. Unger JM, Coltman CA Jr, Crowley JJ, et al. Impact of the year 2000 Medicare policy change on older patient enrollment to cancer clinical trials. J Clin Oncol. 2006;24(1):141-144. 7. Kimmick GG, Peterson BL, Kornblith AB, et al. Improving accrual of older persons to cancer treatment trials: a randomized trial comparing an educational intervention with standard information: CALGB 360001. J Clin Oncol. 2005;23(10):2201-2207. 8. Hutchins LF, Unger JM, Crowley JJ, et al. Underrepresentation of patients 65 years of age or older in cancer-treatment trials. N Engl J Med. 1999;341(27):2061-2067. 9. Aapro MS, Köhne CH, Cohen HJ, et al. Never too old? Age should not be a barrier to enrollment in cancer clinical trials. Oncologist. 2005;10(3):198-204. 10. Rao A, Cohen HJ. Symptom management in the elderly cancer patient: fatigue, pain, and depression. J Natl Cancer Inst Monogr. 2004;2004(32): 150-157. 11. Bernabei R, Gambassi G, Lapane K, et al; SAGE Study Group. Management of pain in elderly patients with cancer. Systematic assessment of geriatric drug use via epidemiology [published correction appears in JAMA. 1999;281(2):136]. JAMA. 1998;279(23):1877-1882. 12. Lewis JH, Kilgore ML, Goldman DP, et al. Participation of patients 65 years of age or older in cancer clinical trials. J Clin Oncol. 2003;21(7):1383-1389. 13. Trimble EL, Carter CL, Cain D, et al. Representation of older patients in cancer treatment trials. Cancer. 1994;74(7 suppl):2208-2214. 14. Kemeny MM, Peterson BL, Kornblith AB, et al. Barriers to clinical trial participation by older women with breast cancer [published correction appears in J Clin Oncol. 2004;22(23):4811]. J Clin Oncol. 2003;21(12):2268-2275. 15. Yee KW, Pater JL, Pho L, et al. Enrollment of older patients in cancer treatment trials in Canada: why is age a barrier? J Clin Oncol. 2003;21(8): 1618-1623. 16. 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(25):3457-3465. 17. Zafar SF, Heilbrun LK, Vishnu P, et al. Participation and survival of geriatric patients in phase I clinical trials: the Karmanos Cancer Institute (KCI) experience. J Geriatr Oncol. 2011;2(1):18-24. 18. Palaia I, Loprete E, Musella A, et al. Chemotherapy in elderly patients with gynecological cancer. Oncology. 2013;85(3):168-172. 19. Giovanazzi-Bannon S, Rademaker A, Lai G, et al. Treatment tolerance of elderly cancer patients entered onto phase II clinical trials: an Illinois Cancer Center study. J Clin Oncol. 1994;12(11):2447-2452. 20. LoConte NK, Smith M, Alberti D, et al. Amongst eligible patients, age and comorbidity do not predict for dose-limiting toxicity from phase I chemotherapy. Cancer Chemother Pharmacol. 2010;65(4):775-780. 21. Javid SH, Unger JM, Gralow JR, et al. A prospective analysis of the influence of older age on physician and patient decision-making when considering enrollment in breast cancer clinical trials (SWOG S0316). Oncologist. 2012;17(9):1180-1190. 22. Townsley CA, Selby R, Siu LL. Systematic review of barriers to the 214 Cancer Control

recruitment of older patients with cancer onto clinical trials. J Clin Oncol. 2005;23(13):3112-3124. 23. Foster JA, Salinas GD, Mansell D, et al. How does older age influence oncologists’ cancer management? Oncologist. 2010;15(6):584-592. 24. Townsley CA, Chan KK, Pond GR, et al. Understanding the attitudes of the elderly towards enrolment into cancer clinical trials. BMC Cancer. 2006;6:34. 25. Jenkins V, Farewell V, Farewell D, et al; TTT Steering Committee. Drivers and barriers to patient participation in RCTs. Br J Cancer. 2013;108(7): 1402-1407. 26. Umutyan A, Chiechi C, Beckett LA, et al. Overcoming barriers to cancer clinical trial accrual: impact of a mass media campaign. Cancer. 2008;112(1):212-219. 27. Garrido-Laguna I1, Janku F, Vaklavas C, et al. Validation of the Royal Marsden Hospital prognostic score in patients treated in the Phase I Clinical Trials Program at the MD Anderson Cancer Center. Cancer. 2012;118(5): 1422-1428. 28. van Heeckeren WJ, Fu P, Barr PM, et al. Safety and tolerability of phase I/II clinical trials among older and younger patients with acute myelogenous leukemia. J Geriatr Oncol. 2011;2(3):215-221. 29. Baggstrom MQ, Velcheti V, Malik M, et al. Barriers for accrual to clinical trials in adult patients (pts) with non-small cell lung cancer: P1-250. J Thorac Oncol. 2007;2(8):S838. 30. Hempenius L, Slaets JP, Boelens MA, et al. Inclusion of frail elderly patients in clinical trials: solutions to the problems. J Geriatr Oncol. 2013;4(1):26-31. 31. Herrera AP, Snipes SA, King DW, et al. Disparate inclusion of older adults in clinical trials: priorities and opportunities for policy and practice change. Am J Public Health. 2010;100(suppl 1):S105-S112. 32. Kornblith AB, Kemeny M, Peterson BL, et al. Survey of oncologists’ perceptions of barriers to accrual of older patients with breast carcinoma to clinical trials. Cancer. 2002;95(5):989-996. 33. Yourman LC, Lee SJ, Schonberg MA, et al. Prognostic indices for older adults: a systematic review. JAMA. 2012;307(2):182-192. 34. Extermann M, Boler I, Reich RR, et al. Predicting the risk of chemotherapy toxicity in older patients: the Chemotherapy Risk Assessment Scale for High-Age Patients (CRASH) score. Cancer. 2012;118(13):3377-3386. 35. National Institues of Health (NIH) Revitalization Act. Subtitle B: §131133; 1993. 36. Extermann M, Chen H, Cantor AB, et al. Predictors of tolerance to chemotherapy in older cancer patients: a prospective pilot study. Eur J Cancer. 2002;38(11):1466-1473. 37. Hamaker ME, Stauder R, van Munster BC. On-going clinical trials for elderly patients with a hematological malignancy: are we addressing the right end points? Ann Oncol. 2014;25(3):675-681.

July 2014, Vol. 21, No. 3

A multipronged approach is needed when caring for elderly patients with cancer.

Daffodil_9767. Photograph courtesy of Henry Domke, MD. www.henrydomke.com

Studying Cancer Treatment in the Elderly Patient Population Lodovico Balducci, MD Background: Data relating to cancer treatment in the older patient population are limited because older individuals have been under-represented in clinical trials. The goal of this review was to establish which factors hinder the participation of older individuals to clinical trials and to examine possible solutions. Methods: The literature relating to cancer treatment in the older patient population was reviewed. Results: The benefit of systemic cancer treatment may decrease with age, and risks may be increased due to reduced life expectancy and reduced tolerance of stress in the older population. Therefore, a multipronged approach is recommended for clinical studies in these patients, including phase 2 studies limited to persons 70 years of age and older, stratification by life expectancy and predicted treatment tolerance in phase 3 studies, and registration studies to establish predictive variables for treatment-related toxicity in older individuals. Conclusions: A combination of prospective and registration studies may supply adequate information to study cancer treatments in the older patient population.

Introduction The study of cancer care in the older population is a complex task. The word “complex” derives from the Latin cum plexere, meaning to weave together. In an older person, many interwoven conditions may conspire to reduce life expectancy, the tolerance of stress, and the ability to live independently.1 Anemia in an older person is an example of such a complexity, because it may be affected by multiple causes, including hemopoietic insufficiency, chronic renal dysfuncFrom the Senior Adult Oncology Program at the H. Lee Moffitt Cancer Center & Research Institute, Tampa, Florida, and the Department of Oncologic Sciences at the University of South Florida Morsani College of Medicine, Tampa, Florida. Submitted January 26, 2014; accepted March 13, 2014. Address correspondence to Lodovico Balducci, MD, 12902 Magnolia Drive, Tampa, FL 33612. E-mail: [email protected] No significant relationship exists between the author and the companies/organizations whose products or services may be referenced in this article. July 2014, Vol. 21, No. 3

tion, chronic inflammation, and iron deficiency from chronic bleeding and iron malabsorption.2 The study of cancer care in the older population must take into account the complexity of aging; therefore, as described in this article, a multipronged approach is needed for this patient population.

Definition of Age Aging is associated with common trends that include a decreased functional reserve of multiple organ systems and an increased susceptibility to diseases and injuries.1 These changes occur at different rates in different individuals and are poorly reflected in chronological age. The assessment of physiological, rather than chronological, age is paramount to the enrollment of older individuals in clinical trials of cancer treatment. Chronological age may be used as a landmark to establish when the assessment of physiological age becomes necessary, and this landmark is commonly established to be 70 years of age3; howCancer Control 215

ever, this statement does not imply that all individuals 70 years of age and older are elderly.

dependent due to a treatable neoplastic condition, such as lymphoma, then treatment is indicated. In these situations, treatment may reverse ADL depenAssessment of Physiological Age dence. Instrumental activities of daily living (IADLs) Age is associated with decreased life expectancy and are activities necessary to live independently and intolerance of stress. Thus, the determination of physiclude the use of transportation, ability to take medicaological age may be based on the assessment of mortions, to use the telephone, to manage one’s finances, tality risk and stress-related complications. For this and to provide to one’s meals. A person dependent purpose, the best validated instrument is a Comprein one or more IADLs will require assistance. The hensive Geriatric Assessment (CGA; Table 1).3-5 determination of IADLs is relevant to this review, beActivities of daily living (ADLs) are activities cause it implies an increased incidence of therapeutic necessary to basic survival and include transferring, complications in addition to an increased mortality eating, grooming, dressing, going to the bathroom risk.6,7 Presently, polymorbidity is a more popular term alone, and continence. A person dependent in at least than comorbidity, implying that different diseases 1 ADL requires a full-time caregiver or admission to may influence both the treatment and the behavior an assisted-living facility. In general, patients depenof other diseases.8 Polymorbidity is associated with dent in at least 1 ADL have a limited life expectancy, a decreased life expectancy, decreased tolerance to a limited tolerance for stress, and are candidates for antineoplastic treatment and, in general, a poor canpalliative care; however, exceptions do exist. If a pacer prognosis.3 Geriatric syndromes include common tient who was previously independent is now ADL conditions, although not all are unique to aging.9 Patients with cancer, ADL dependence, and at least 1 geriatric syndrome are candidates Table 1. — Examples of Domains of the Comprehensive for symptom control, unless the geriatric Geriatric Assessment and Potential Clinical Applications syndrome is reversible. Malnutrition is a Domain Clinical Application common complication of both cancer and aging and is associated with a decreased Functional Status tolerance of chemotherapy and decreased Activities of daily living Relation to life expectancy immune function.10 Instrumental activities of daily living Functional dependence In addition to the CGA, other forms Tolerance of stress of assessing physiological age are noteComorbidity worthy. The frailty index is calculated by Number of comorbid conditions Relation to life expectancy summing the functional deficits in an agand comorbidity indices Tolerance of stress ing person.11 Physiological age is assessed based on the average number of deficits Mental Status accumulated by a person of that chronoMini-Mental State Examination Relation to life expectancy logical age. For example, if a 75-year-old (Folstein test) and dependence woman has the number of deficits that Emotional Conditions correspond to an average 61-year-old, the Geriatric Depression Scale Relation to survival woman’s physiological age is assessed as May indicate motivation to receive treatment 61 years. However, determining the frailty index is too laborious for clinical applicaNutritional Status tions because it requires the clinician to Mini Nutritional Assessment Reversible condition evaluate 70 conditions11; in addition, it is Possible relationship to survival not clear whether the index predicts morPolypharmacy tality risks and tolerance of stress. Risk of drug interactions Numerous laboratory tests have been proposed for the assessment of physioGeriatric Syndromes logical aging. Of these, the assessment of Delirium Relationship to survival and stress tolerance inflammatory markers in the circulation12 Dementia Functional dependence and the length of leukocyte telomeres13 Depression May be reversible to some extent bear a relation to mortality risk. Telomere Falls length also predicts the risk of adverse Incontinence events from cytotoxic chemotherapy.14 Spontaneous bone fractures Although these tests are of interest, they Neglect and abuse have not been validated. The determinaFailure to thrive tion of inflammatory markers lacks ad216 Cancer Control

July 2014, Vol. 21, No. 3

equate sensitivity; moreover, the telomere length varies among different persons, thus making the comparison of physiological age based on telomere length problematic.

Influence of Aging on Cancer Treatment Factors related to aging that may influence cancer treatment include cancer biology, which may be different among younger and older patients; decreased life expectancy of the older person, which may reduce the benefits of cancer treatment; and increased vulnerability to complications due to cancer therapies. Biology of Cancer and Aging Cancer growth and aggressiveness are influenced by 2 factors, namely, the tumor cell and the tumor host. For example, acute myeloid leukemia is less susceptible to treatment in the elderly patient population, and this is due — at least in part — to the higher prevalence of unfavorable prognostic factors, including complex cytogenetic changes, MDR-1 gene expression, and the involvement of the early multipotential hemopoietic progenitors.15 By contrast, in the setting of breast cancer, the prevalence of favorable prognostic factors, such as hormone receptor concentration and good cellular differentiation, increases with age.16 Genomic and proteomic analysis may help account for these factors in clinical trials. Assessment of patient-related factors is difficult. Such factors may include immunosenescence, endocrine senescence, proliferative senescence, and chronic inflammation.17 Animal data suggest that immunosenescence may have different effects on the growth of various neoplasms; for example, immunosenescence may enhance the growth of highly immunogenic neoplasms, while disfavoring the growth of poorly immunogenic neoplasms. Decreased production of sexual hormones may inhibit hormone-dependent cancers, such as breast and prostate cancers. Age is associated with increased insulin resistance, which results in an increased concentration of insulin, a powerful growth factor for several tumors, in the circulation.18 The aging of stromal tissues involves the proliferative senescence of fibroblasts, facilitating neoplastic growth with the production of tumor growth factors and enzymes that dissolve basal membranes.19 Age is also associated with progressive and chronic inflammation, which may contribute to immunosenescence and tumor growth.17 Polypharmacy is another patient-related factor among the older population because the number of medications used and the prevalence of polypharmacy increase with age.8 For example, the use of metformin, a drug that decreases insulin resistance and, consequently, circulating levels of insulin, is associated with prolonged survival in patients with prostate or July 2014, Vol. 21, No. 3

breast cancer.20,21 As mentioned previously, polymorbidity may also affect cancer growth. For example, the prognosis of breast, prostate, or large bowel cancer is worse in individuals with diabetes than in those without diabetes.8 At present, these factors cannot be accounted for in randomized clinical trials. Treatment Goals The risk–benefit ratio of antineoplastic treatment may be reduced in the majority of older individuals. The expected benefits are lower in this population due to a progressive decline in life expectancy. Even in the most fit of older persons, individual age is a risk factor for some complications of chemotherapy, including myelosuppression, mucositis, cardiomyopathy, and peripheral neuropathy.22 The risk of such complications increases in individuals with compromised function and multiple morbidities. It is reasonable to aim for a cure when facing a rapidly lethal but curable disease, such as large B-cell lymphoma or acute leukemia, despite the high risk of serious complications. Currently, it is reasonable not to submit older individuals with limited life expectancies and a chronic, but not life-threatening disease, such as chronic lymphocytic leukemia, to the toxicity of fludarabine, cyclophosphamide, and rituximab,23 which may add a few months of survival at a time when most patients might have died of a disease other than cancer. Cure, prolongation of survival, and symptom management are the main goals of treatment; however, the preservation of function and active life expectancy should also be goals for older patients.24 Active life expectancy is a period of time during which a person remains functionally independent. Loss of functional independency is a significant threat to the quality of life of older individuals.24 Barriers to Treatment Numerous social factors may preclude cancer treatment in older patients, including accessibility (eg, many older individuals may not be able to negotiate their way alone to a treatment center), difficulty with finances (recipients of Medicare may have to pay unaffordable co-payments for cancer treatment), and inadequate home support. However, it is important to remember that ageism can be a hindrance to the reception of adequate cancer treatment.25 A study conducted by the Cancer and Leukemia Group B (CALGB) demonstrated that the main obstacle to clinical trial participation of older patients with breast cancer was the reluctance of physicians to offer experimental treatment to older individuals.25 Clinical studies of cancer treatment among older patients must account for the factors outlined in this brief review, including a poor understanding of the Cancer Control 217

interaction between tumor and patient, a reduced risk–benefit ratio, the increased risk of treatment complications, the inclusion of active life expectancy among the treatment goals, and the socioeconomic barriers to treatment.

Clinical Trials in Older Patients With Cancer Aging may be associated with a number of pharmacological changes that render the study of new drugs in the older population necessary (Table 2).22 Data on drug absorption are wanted, but the bioavailability of oral drugs is expected to decrease with age. Decreased total body water content is associated with a decreased volume of distribution and an increased level of water-soluble drugs in the circulation, which may purportedly increase toxicity. Renal excretion and hepatic metabolism of drugs are universally decreased with age. Although the decline in the glomerular filtration rate may be accounted for by calculating the level of creatinine clearance, a clinical test of hepatic metabolism is still needed. As already mentioned, numerous age-related changes in target organs may be associated with increased hemopoietic, mucosal, cardiac, and neurological toxicities. The question of

whether a new agent is effective and safe in both the younger and older patient populations must be addressed in appropriate clinical trials. Advanced age should never be a criterion to exclude older individuals from participating in clinical trials designed for adults. Rather, as clinicians, we must ask whether certain clinical trials should be exclusively dedicated to older individuals. Phase 1 and 2 Trials Older individuals should have access to phase 1 trials, but reserving clinical trials for older individuals alone is not a productive strategy. Due to the increased risk of adverse effects in the older population, phase 1 trials dedicated to older individuals may unnecessarily delay the development and approval of life-saving drugs. Instead, a representation of individuals 70 years or older in phase 2 trials should be adequate to establish the activity and the safety of a new drug among this older population. In my opinion, phase 2 trials represent a convenient way to study the pharmacology of new agents among the elderly without delaying drug development.

Table 2. — Pharmacological Changes of Aging Type of Change

Comments

Pharmacokinetics Absorption

Effects of aging on absorption are unknown Reasonable to assume a progressive decrease in absorption due to atrophic gastritis, decreased gastric motility, and decreased splanchnic circulation

Volume of distribution

Changes in body composition; increased fat and decreased water content

Metabolism

Hepatic metabolism reduced from progressive loss of liver mass and decreased splanchnic circulation

Renal excretion

Glomerular filtration rate declines with age in nearly all individuals

Hepatic excretion

Biliary excretion appears to remain intact

Pharmacodynamics Hematopoietic system

Decreased concentration of early hematopoietic progenitors Decreased lymphocytic production Homing abnormality may reduce concentration of early progenitors in bone marrow

Mucosa epithelium

Decreased epithelial stem cells Increased proliferation of differentiated cells

Heart

Reduction in myocardial sarcomeres Increased fibrosis and degenerative processes (amyloid)

Peripheral nervous system

Increased degenerative processes

Central nervous system

Atrophy Increase in degenerative processes with decreased circulation

218 Cancer Control

Phase 3 Trials Phase 3 trials dedicated to the elderly may be conducted for 2 reasons: (1) to determine whether the benefits of a treatment strategy decline with age, and (2) whether the use of new drugs may improve the treatment of elderly patients who are not candidates for more aggressive treatment. In a seminal study of the CALGB, women 65 years of age or older with early-stage breast cancer were randomized to receive combination chemotherapy or single-agent capecitabine.26 The study showed that combination chemotherapy in the adjuvant setting reduced recurrence rates and improved survival among these patients. This study resolved a decade-long controversy of whether adjuvant chemotherapy was beneficial to older women. Other important studies of this type demonstrated that doublet chemotherapy was superior to single-agent chemotherapy in older patients with metastatic non–small-cell lung cancer27 and that full-dose chemotherapy with cyclophosphamide, doxorubicin, vincristine, and prednisone was superior to chemotherapy in reduced doses among elderly patients with large cell lymphoma.28 These triJuly 2014, Vol. 21, No. 3

als determined that age alone is not a contraindication to active cancer treatment; however, these trials included healthy elderly patients, and, thus, shed no light on the best treatment options for elderly persons affected by disability and multiple morbidities. A 2014 study demonstrated that elderly patients with diseases and disability can be studied in randomized controlled trials.29 Individuals with chronic lymphocytic leukemia who were 70 years of age or older and not eligible for more aggressive treatment (eg, combination fludarabine, cyclophosphamide, and rituximab) were randomized to receive chlorambucil or chlorambucil plus rituximab or obinutuzumab. Nearly 1,500 patients were stratified during randomization according to comorbidity severity. Disease-free and overall survival rates were improved in patients treated with obinutuzumab. The results from this recent study represent a model for future randomized controlled studies among older patients with an estimated life expectancy of several years.

person’s risk of grade 3/4 toxicity, I recommend the Chemotherapy Risk Assessment Scale for High-Age Patients and the Cancer and Aging Research Group instruments, both of which have been validated in older individuals.6,7 Of course, the risk of toxicity that disqualifies an individual from a trial may vary from disease to disease. For example, the threshold may be higher in cases in which the disease is rapidly lethal and the treatment option offers an opportunity for cure or prolonged survival.

Other Studies in Older Patients With Cancer Clinical trials exclusively dedicated to older patients will still be unable to embrace the diversity of this patient population. The questions that persist include: • Is the treatment beneficial to the majority of older individuals?

• Which individual factors determine the benefit and risk of treatment in the older population?

Studies that match Surveillance Epidemiology and End Results data with Medicare records may provide a partial answer to the first question above. Because of these data, it was possible to discern that individuals 75 years and older benefit from the adjuvant treatment of colon and rectal cancers30 and that adjuvant chemotherapy is leukemogenic in older women.31 However, it is impossible to establish from these results exactly which patients may benefit and which ones may be harmed by Life expectancy cytotoxic chemotherapy, as it is not possible to analyze certain prognostic factors from these data. A prospective evaluation of these paLonger than life Shorter than life tients alone is the way to build prognostic expectancy with cancer expectancy with cancer models that encompass the variabilities among the older patient population, including all of the factors with the potential to interact with treatment options. Through Study regimen this process, models were derived to estimate the risk of chemotherapy complications in older individuals.6,7 The advent of the electronic medical Estimated risk Estimated risk record offers a unique opportunity to pergrade 3/4 grade 3/4 form studies in such a way that the majortoxicity ≤ 50% toxicity > 50% ity of older patients with cancer can be included in research trials. As long as all relevant information related to function, comorbidity, emotional status, and social Palliative care support is included within the medical reRandomize cord, it may become possible to create preStratify according to: cise prognostic models that help deliver • Life expectancy personalized care to older patients with • Toxicity risk cancer. Such is the aim of the project CancerLinQ, which is a system designed by the Fig. — Suggested future stratification of older patients with cancer into clinical trials. American Society of Clinical Oncology.32 Future Directions The Figure illustrates how future studies might be conducted. To estimate a patient’s life expectancy, the use of ePrognosis (http://eprognosis.ucsf.edu; University of California, San Francisco) is recommended, which is available free of charge.4 To estimate an older

July 2014, Vol. 21, No. 3

Cancer Control 219

Conclusions Age should not be considered a criterion for disqualifying a person with cancer from a clinical trial. Phase 2 trials dedicated to older individuals are necessary to establish the pharmacology of novel agents in the older patient population. Phase 3 trials that study the older patient population should stratify patients according to life expectancy and treatment risks. Registry studies with prospective data collection are necessary to encompass the diversity of all older individuals. References 1. Balducci L, Aapro M. Complicated and complex: helping the older cancer patients to exit the labyrinth. J Ger Oncol. 2014;5(1):116-118. 2. Tettamanti M, Lucca U, Gandini F, et al. Prevalence, incidence, and types of mild anemia in the elderly: the “Health and Anemia” population-based study. Hematologica. 2010;95(11):1849-1856. 3. Hurria A, Wildes T, Blair ST, et al. Senior adult oncology, version 2.2014: clinical practice guidelines in oncology. J Natl Compr Cancer Netw. 2014;12(1):82-126. 4. Yourman LG, Lee SJ, Schonberg MA, et al. Prognostic indices for older adults: a systematic review. JAMA. 2012;307(2):182-192. 5. Lee SJ, Leipzig RM, Walter LC. Incorporating lag time to benefits into prevention decisions for older adults. JAMA. 2013;310(24):2609-2610. 6. Extermann M, Boler I, Reich RR, et al. Predicting the risk of chemotherapy toxicity in older patients: the Chemotherapy Risk Assessment Scale for High-Age Patients (CRASH) score. Cancer. 2012;118(13):3377-3386. 7. 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(25):3457-3465. 8. Balducci L, Goetz-Parten D, Steinman MA. Polypharmacy and the management of the older cancer patient. Ann Oncol. 2013;(24 suppl 7):vii36-40. 9. Mohile SG, Fan L, Reeve E, et al. Association of cancer and geriatric syndromes in older Medicare beneficiaries. J Clin Oncol. 2011;29(11): 1458-1464. 10. Van Cutsem E, Arends J. The causes and consequences of cancerassociated malnutrition. Eur J Oncol Nurs. 2005;(9 suppl 2):S51-S63. 11. Mitnitski A, Song X, Rockwood K. Assessing biological aging: the origin of deficit accumulation. Biogerontology. 2013;14(6):709-717. 12. Huffman KM, Pieper CF, Kraus VB, et al. Relations of a marker of endothelial activation (s-VCAM) to function and mortality in community-dwelling older adults. J Gerontol A Biol Sci Med Sci. 2011;66(12):1369-1375. 13. Sanders JL, Newman AB. Telomere length in epidemiology: a biomarker of aging, age-related disease, both, or neither? Epidemiol Rev. 2013;35(1): 112-131. 14. Falandry C, Gilson E, Rudolph KL. Are aging biomarkers clinically relevant in oncogeriatrics? Crit Rev Oncol Hematol. 2013;85(3):257-265. 15. Juliusson G, Antunovic P, Derolf A, et al. Age and acute myeloid leukemia: real world data on decision to treat and outcomes from the Swedish Acute Leukemia Registry. Blood. 2009;113(18):4179-4187. 16. Turner N, Zafarana E, Becheri D, et al. Breast cancer in the elderly: which lessons have we learned? Future Oncol. 2013;9(12):1871-1881. 17. Balducci L, Ershler WB. Cancer and aging: a nexus at several levels. Nat Rev Cancer. 2005;5(8):655-662. 18. Evans JL, Goldfine ID. Aging and insulin resistance: just say iNOS. Diabetes. 2013;62(2):346-348. 19. Velarde MC, Demaria M, Campisi J. Senescent cells and their secretory phenotype as targets for cancer therapy. Interdiscip Top Gerontol. 2013; 38:17-27. 20. Margel D, Urbach DR, Lipscombe LI, et al. Metformin use and all-cause and prostate cancer-specific mortality among men with diabetes. J Clin Oncol. 2013;31(25):3069-3075. 21. Chlebowski RT, McTiernan A, Wactawski-Wende J, et al. Diabetes, metformin, and breast cancer in postmenopausal women. J Clin Oncol. 2012;30(23):2844-2852. 22. Hoffe S, Balducci L. Cancer and age: general considerations. Clin Geriatr Med. 2012;28(1):1-18. 23. Tadmor T, McLaughlin P, Polliack A. Chemoimmunotherapy with fludarabine, cytoxan and rituximab regimen: to use, not to use, or give it as “FCRLITE”? Leuk Lymphoma. 2014;55(4):733-734. 24. Freedman VA. Research gaps in the demography of aging with disability. Disabil Health J. 2014;7(1 suppl):S60-S63. 25. Kemeny MM, Peterson BL, Komblith AB, et al. Barriers to clinical trial participation by older women with breast cancer. J Clin Oncol. 2003;21(12):2268-2275. 26. Muss HB, Berry DA, Cirrincione CT, et al. Adjuvant chemotherapy in 220 Cancer Control

older women with early-stage breast cancer. N Engl J Med. 2009;360(20): 2055-2065. 27. Meoni G, Cecere FL, Lucherini E, et al. Medical treatment of advanced non-small cell lung cancer in elderly patients: a review of the role of chemotherapy and targeted agents. J Geriatr Oncol. 2013;4(3):282-290. 28. Bastion Y, Blay JY, Divine M, et al. Elderly patients with aggressive non-Hodgkin’s lymphoma: disease presentation, response to treatment, and survival--a Groupe d’Etude des Lymphomes de l’Adulte study on 453 patients older than 69 years. J Clin Oncol. 1997;15(8):2945-2953. 29. Goede V, Fischer K, Busch R, et al. Obinutuzumab plus chlorambucil in patients with CLL and coexisting conditions. N Engl J Med. 2014;370(12): 1101-1110. 30. 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(21):2624-2634. 31. Lyman GH, Dale DC, Wolff DA, et al. Acute myeloid leukemia or myelodysplastic syndrome in randomized controlled trials of cancer chemotherapy with granulocyte colony-stimulating factor: a systematic review. J Clin Oncol. 2010;28(17):2914-2924. 32. American Society of Clinical Oncology. CancerLinQ. http://www.asco. org/quality-guidelines/cancerlinq. Accessed April 22, 2014.

July 2014, Vol. 21, No. 3

Detecting BRAF mutations in a wide array of cancers represents an advance in delivering molecularly targeted therapies to patients with cancer.

Cut Nautilus Shell_4499. Photograph courtesy of Henry Domke, MD. www.henrydomke.com

BRAF Mutations: Signaling, Epidemiology, and Clinical Experience in Multiple Malignancies Richard D. Hall, MD, and Ragini R. Kudchadkar, MD Background: Mutations in BRAF were first reported in 2002. Since that time, the molecular basis for oncogenic signaling has been elucidated in multiple malignancies. The development of v-raf murine sarcoma viral oncogene homolog B (BRAF) inhibitors has helped improve clinical outcomes in malignant melanoma and is suggested by case reports in other malignancies. Methods: A review of pertinent articles examining the mechanisms of BRAF signaling in various cancer types and an update on clinical trials of BRAF inhibitions are presented. Results: Clinical response to BRAF inhibition varies by malignancy. In melanoma, single-agent vemurafenib or dabrafenib prolongs overall survival compared with chemotherapy, but both are limited by the development of acquired resistance in many patients. Results of early-phase clinical trials and case reports demonstrate responses in V600E-mutant non–small-cell lung cancer, thyroid cancer, and hairy cell leukemia. However, no significant difference in progression-free survival was seen in colorectal cancer with single-agent vemurafenib. Overcoming resistance to BRAF inhibition with combination therapy is an active area of research. Conclusions: The detection of BRAF mutations represents an advance in delivering molecularly targeted therapies to patients with a variety of cancers. Acquired resistance limits the ability of BRAF inhibitors to produce long-term remissions; however, combining BRAF inhibitors with the mitogen-activated protein kinase pathway and/or other pathway inhibitors represents a promising method to improve long-term outcomes.

From the Hematology Oncology Fellowship Program (RDH) and the Department of Cutaneous Oncology (RRK) at the H. Lee Moffitt Cancer Center & Research Institute, Tampa, Florida. Submitted January 13, 2014; accepted March 15, 2014. Address correspondence to Richard D. Hall, MD, Department of Internal Medicine, Division of Medical Oncology, University of Virginia Health System, Hematology/Oncology, PO Box 800716, Charlottesville, VA 22908. E-mail: [email protected] Dr Kudchadkar has received honoraria and consultant fees from Bristol-Myers Squibb and Genentech. Dr Hall reports no significant relationship with the companies/organizations whose products or services may be referenced in this article. Dr Kudchadkar is now affiliated with the Winship Cancer Institute of Emory University, Atlanta, Georgia. Dr Hall is now affiliated with the University of Virginia Health System, Charlottesville, Virginia. July 2014, Vol. 21, No. 3

Introduction The discovery of mutations in BRAF, part of the mitogen-activated protein kinase (MAPK) signaling pathway, heralded a new era of therapeutic options for patients with malignant melanoma, colorectal cancer (CRC), and non–small-cell lung cancer (NSCLC).1 Additional mutations in BRAF have been described in other malignancies as well, including thyroid cancer, hairy cell leukemia (HCL), and multiple myeloma (where they were initially thought to be absent).2-6 Significant variation exists in the incidence and epidemiology of BRAF mutations across cancers. Mutations in this gene have been found to be universal in Cancer Control 221

HCL, in about one-half of patients with melanoma and thyroid cancer, and in about 10% or less in CRC and NSCLC.1,5,6 Although using small molecule inhibitors of v-raf murine sarcoma viral oncogene homolog B (BRAF) in melanoma has produced improved clinical outcomes, their use in CRC has not produced clinical benefit.7-11 The V600E mutation results in an amino acid substitution from valine (V) to glutamic acid (E), and it is the most common BRAF mutation detected in human cancer; however, among tumors known to harbor BRAF mutations, lung cancer is notable for a high fraction of non-V600E mutations.1,12,13 In the 11 years since mutations in BRAF were first reported, vemurafenib, dabrafenib, and trametinib have received approval from the US Food and Drug Administration for the treatment of V600-mutated melanoma.14,15 This review will examine the current understanding of BRAF cell signaling and will highlight disease-specific epidemiology and clinical experience using BRAF inhibitors across a disparate group of human cancers.

BRAF Signaling Constitutive activation of the MAPK pathway is a common event in many cancers that leads to sustained proliferative signaling.16 The MAPK pathway is best defined as the group of kinases comprised of the rapidly accelerated fibroblast (RAF) family of serine/ threonine kinases, the MAPK/extracellular-signalregulated kinase MEK1/2, and terminating with the extracellular signal-regulated kinase (ERK).17 Binding of ERK to nuclear protein transcription factors, including the E26 transformation specific (ETS) family, leads to gene expression that promotes cell growth and survival.18 In normal conditions, upstream activation of the MAPK pathway occurs most often through ligand binding to receptor tyrosine kinases. For example, binding of the epidermal growth factor family of ligands to the epidermal growth factor receptors (EGFRs) leads to receptor dimerization followed by autophosphorylation and subsequent downstream signaling through both the MAPK pathway and the phosphatidylinositol 3 kinase/protein kinase B/mammalian target of rapamycin (PI3K/Akt/mTOR) pathway.19,20 Following receptor dimerization, adaptor proteins undergo phosphorylation that ultimately leads to the activation of the rat sarcoma (RAS) family of GTPases.21 Binding of RAS to one of the RAF proteins leads to subsequent downstream MAPK signaling. The 3 RAS isoforms, HRAS, KRAS, and NRAS, comprise a group of highly conserved GTPases and are the most frequently mutated oncogenes in human cancers.22 KRAS mutations are detected in large percentages of CRC, NSCLC, and pancreatic adenocarcinoma, and NRAS is the second most commonly mutated gene in melanoma, occurring in approximately 40% of cases 222 Cancer Control

of BRAF wild-type melanoma.22,23 Similar to KRAS, there are 3 RAF isoforms that are serine/threonine kinases, which lead to MEK and ERK phosphorylation when activated via RAS. Under normal conditions, RAS proteins bind to cytosolic RAF dimers, upon which they undergo phosphorylation.24 Activated RAF then recruits MEK, ERK, and scaffolding proteins to the cell membrane, thus leading to the phosphorylation of MEK and ERK.25,26 RAF mutations represent another opportunity for malignant cells to sustain MAPK signaling. Mutations in BRAF, first described in 2002,1 occur most often at nucleotide 1796, leading to a valine to glutamic acid change at codon 599 (V599E; subsequently renamed to V600E due to a nomenclature change). The V600E mutation leads to a conformational change in the G-loop activation segment of BRAF, rendering it constitutively active and able to bind MEK and ERK as a monomer.25 Mutated BRAF results in persistently elevated ERK phosphorylation and target gene transcription. In addition, it is resistant to negative feedback signals that attempt to counterbalance the ERK activation.27 The multiple tyrosine kinase signaling pathways within a cell are interconnected and do not exist in isolation. It has been noted that V600E-mutant BRAF activates the mTOR pathway.28 The first report on BRAF mutations described them as being detected in 59% of melanomas, 18% of CRCs, 11% of gliomas, and 4% of lung adenocarcinomas and ovarian carcinomas.1 All mutations occurred in either exon 11 or 15 across all malignancies, and the V600E mutation was the most commonly detected mutation.1 In addition to melanoma, CRC, and NSCLC, BRAF mutations have been detected in thyroid cancers, HCL, and multiple myeloma.2,5,6 The wide spectrum of cancers in which BRAF mutations are detected highlights the prominent role MAPK signaling plays in promoting oncogenesis. The following sections will highlight the epidemiology and scientific understanding of BRAF mutations in 5 malignancies: malignant melanoma, CRC, NSCLC, thyroid cancer, and HCL. Clinical experience using small molecule inhibitors to inhibit BRAF signaling are reviewed and summarized in the Table.

Melanoma Malignant melanoma ranks as the fifth and seventh most commonly diagnosed malignancy in men and women, respectively, with an estimated incidence of more than 76,000 persons in the United States in 2013.29 Traditionally, metastatic melanoma has carried a dismal prognosis, with 10-year survival rates of less than 10% as recently as 2009, with a historic 1-year survival rate of 25%.30 However, the discovery of BRAF mutations in melanoma changed the pathologic understanding of the disease, leading to new July 2014, Vol. 21, No. 3

treatment options for patients with metastatic disease. Following the discovery of BRAF mutations in a large fraction of primary cutaneous melanoma cases, BRAF mutations were identified in a similarly high percentage of dysplastic nevi, implicating the BRAF mutation as a necessary but insufficient oncogenic driver in early melanoma.31 In the metastatic setting, BRAF mutations are found in 46% to 48% of metastatic biopsy specimens, with V600E as the most common mutation (73%–91%) followed by V600K (7%–20%) and, less commonly, V600D and L597R mutations.32-34 There appear to be differences in mutation type (V600E vs non-V600E) according to patient age, primary disease site, type of melanoma, and response to BRAF inhibition. In a cohort of 302 patients who had melanoma with activating BRAF mutations, Bucheit et al35 reported a V600K mutation rate of 24%, with statistically significant differences in median age (60.0 vs 44.7 years), male sex, and truncal location compared with patients with the V600E mutation. Menzies et al34 detected a similar trend between V600E and non-V600E mutations, with non-V600E mutations found in fewer than 20% of patients younger than 50 years and more than 40% in patients 70 years of age or older. They also reported a decreasing incidence of BRAF mutations by decade of life. A total of 25% of tumors from patients 70 years of age or older had a mutation, while the tumors of patients younger than 30 years of age almost universally possessed the BRAF mutation. V600E mutations occur more common-

ly on intermittently sun-exposed skin and are found most often in superficial spreading melanoma, while non-V600E mutations occur more frequently on chronically sun-exposed areas such as the head and neck.34,36,37 Non-V600E mutations respond to BRAF inhibition in clinical trials; however, retrospective evidence suggests that patients with these mutations have a shorter disease-free interval (defined as the duration of time from primary site diagnosis to metastatic disease) and a trend toward inferior survival rates compared with those who have V600E-mutant melanoma.7,10,34,35 Efforts to inhibit mutant BRAF in melanoma using small molecule inhibitors began following the characterization of BRAF mutations. Based on preclinical data that showed the inhibition of growth among melanoma tumor xenografts, sorafenib entered clinical trials in BRAF-mutated melanoma.38 Overall, the clinical trial results of sorafenib were disappointing, although initial studies showed some promise in combination with chemotherapy. In a phase 1 trial of sorafenib combined with carboplatin and paclitaxel chemotherapy, patients with melanoma had a longer median progression-free survival (PFS) rate compared with patients who had other tumor types (307 vs 104 days, respectively).39 However, a phase 2 trial of sorafenib monotherapy given to 34 patients with stage 4 melanoma revealed a low response rate (2.8%) and no difference in response rate between BRAF mutant and wild type, suggesting that, as single-agent therapy, sorafenib had minimal activity against BRAF-mutant melanoma.40 A multina-

Table. — BRAF Mutation Prevalence, Clinical Characteristics, and Selected Active Clinical Trials by Cancer Type Cancer Type

Mutation Frequency and Type

Clinical Characteristics

Selected Active Clinical Trials

Melanoma

46%–48%, V600E more common than V600K; other rare exon 15 mutations reported

BRAF V600E mutations more common in younger persons and in tumors arising from intermittently sun-exposed skin Mutually exclusive with NRAS

NCT01726738 NCT01841463 NCT01826448 NCT01616199 NCT01754376 NCT01682083

Colorectal

7.9%–15.2%; predominantly V600E

Associated with inferior outcomes compared with BRAF wild-type Mutually exclusive with KRAS and PIK3CA

NCT01791309 NCT01750918 NCT01719380 NCT01902173

Thyroid

44% (papillary) and 24% (anaplastic); predominantly V600E

In papillary, associated with increased risk of lymph node invasion and metastasis

NCT01723202 NCT01709292 NCT01534897

Non–Small-Cell Lung

1.6%–4.9% (adenocarcinoma), 4% (squamous). Approximately equal numbers with V600E and non-V600E in adenocarcinoma

No difference in clinical outcomes between V600E and non-V600E Non-V600E does not respond to BRAF inhibitors

NCT01336634 NCT01514864

Hairy Cell Leukemia

Approximately 100%; all V600E

Durable complete response reported in 2 patients after short treatment with BRAF inhibitors

NCT01711632

BRAF = v-raf murine sarcoma viral oncogene homolog B.

July 2014, Vol. 21, No. 3

Cancer Control 223

tional, randomized, double-blind, placebo-controlled phase 3 trial of sorafenib or placebo with carboplatin and paclitaxel failed to demonstrate a difference in PFS rate (17.9 weeks with placebo, 17.4 weeks with sorafenib; hazard ratio [HR], 0.91; 99% confidence interval [CI], 0.63–1.31; 2-sided log rank test P = .49), though BRAF status was not reported in the baseline characteristics of trial participants.41 Although additional studies explored combination strategies with sorafenib and either temozolomide or temsirolimus, the development of selective BRAF inhibitors largely supplanted the work to use sorafenib as a therapeutic agent in BRAF-mutated melanoma.42-44 Eight years after the discovery and characterization of BRAF mutations in melanoma, a dose-escalation, extension phase 1 trial of vemurafenib in 81 patients with melanoma was undertaken by Flaherty et al.9 Of the 32 patients with melanoma who were treated with the recommended phase 2 dose, 24 patients experienced a partial response (PR) and 2 patients had a complete response (CR).9 In 2011, Chapman et al7 reported the results of an international, randomized, open-label phase 3 trial comparing vemurafenib with dacarbazine in patients with the V600E mutation. The trial met the prespecified early stopping rule at the time of interim analysis in December 2010, with both overall survival (OS) and PFS rates favoring vemurafenib. Response rates (PR or CR) approached 48% in the vemurafenibtreated cohort compared with 5% in the dacarbazine cohort.7 Following publication, the US Food and Drug Administration (FDA) approved vemurafenib for the treatment of BRAF-mutated melanoma, along with a companion diagnostic test, in August 2011.15 Updated OS results from the Chapman et al7 study were presented in 2012, and showed a continued significant difference in OS rates (13.2 months [95% CI, 12.0–15.0] for vemurafenib and 9.6 months [95% CI, 7.9–11.8] for dacarbazine) and 12-month OS rates of 55% for vemurafenib and 43% for dacarbazine.45 The BRAF inhibitor dabrafenib has also been studied as a single agent as well as in combination with trametinib, a MEK1/2 inhibitor.8,10,46 In a phase 3 openlabeled, randomized controlled trial comparing dabrafenib with dacarbazine, dabrafenib had significantly improved PFS (5.1 and 2.7 months for dacarbazine; HR, 0.30; 95% CI, 0.18–0.51; P < .0001).10 In addition to the direct inhibition of BRAF, evidence exists to support MEK inhibition in BRAF-mutant melanoma.47-49 Several clinical trials have examined MEK inhibitors as single agents in BRAF-mutant melanoma. In a phase 1 study by Falchook et al,50 treatment with trametinib resulted in a 33% overall response rate in patients with melanoma who were treatment naive compared with a 10% response rate in patients with BRAF wild-type melanoma. A phase 2 trial subsequently compared overall response 224 Cancer Control

(CR or PR) in 2 cohorts of patients, ie, those naive to BRAF inhibition (previously treated with chemotherapy or immunotherapy) or patients previously treated with a BRAF inhibitor.51 Interestingly, overall responses were only seen in patients naive to BRAF inhibitors (25% vs 0%), and stable disease was higher in the group naive to BRAF treatment (51% vs. 28%). The results of this trial suggested that the development of acquired resistance following BRAF inhibition also affected response to single-agent MEK inhibition. Therefore, single-agent MEK inhibition following treatment failure by either single-agent BRAF inhibitor is not recommended. Although the clinical experience involving BRAF and MEK inhibitors as single agents represents a significant advance in the treatment of metastatic melanoma, secondary or acquired resistance to single-agent therapy appears to be universal. Acquired resistance to BRAF inhibitors was predicted by preclinical studies that suggested combination therapies would be needed to treat melanoma.52,53 Unlike the experience derived from the use of small molecule inhibitors in chronic myeloid leukemia, in which progressive disease most often results from kinase domain gatekeeper mutations, Nazarian et al54 reported the absence of acquired mutations in BRAF among resistant cell lines. Resistance developed through the reactivation of MAPK signaling via upregulated platelet-derived growth factor receptor beta (PDGFR β) and NRAS mutations in 5 out of 12 patients (mutations in NRAS were mutually exclusive with increased protein expression of PDGFR β).54 Johannessen et al55 identified a third mechanism by which melanoma expressed MAP3K8 (the gene-encoding cancer Osaka thyroid kinase [COT]/Tpl2), leading to MEK and ERK signaling independent of RAF. In cell-line models, combined RAF and MEK inhibition led to decreased levels of phospho-ERK (p-ERK) and reduced cell growth, suggesting that combined RAF and MEK inhibition may allow cells to circumvent COT-mediated resistance.55 Using serial biopsies obtained as part of a phase 2 clinical study, Trunzer et al56 used immunohistochemistry for p-ERK to demonstrate the reactivation of MAPK signaling at the time of progression. They identified NRAS mutations in 3 of 13 tumors at progression and MEK1 mutations in 4 of 20 tumors at progression. In all tumors with NRAS or MEK1 mutations, BRAF V600E mutation persisted.56 Combining BRAF inhibitors with MEK inhibitors was a logical next step in the effort to prevent or delay the development of acquired resistance. Work by Paraiso et al57 demonstrated that melanoma cell lines exhibited increased p-ERK signaling prior to the development of BRAF inhibitor resistance, but they also noted that combined treatment with BRAF and MEK inhibitors enhanced apoptosis and prevented the develJuly 2014, Vol. 21, No. 3

opment of resistance in cell lines. In a phase 1/2 study of dabrafenib and trametinib in patients with metastatic melanoma who were naive to BRAF inhibitors, combined treatment with both agents at full doses (dabrafenib 150 mg twice daily and trametinib 2 mg once daily) resulted in improved PFS rates compared with dabrafenib monotherapy (9.4 vs 5.8 months; HR, 0.39; 95% CI, 0.25–0.62; P < .001) and a reduced incidence of cutaneous squamous cell carcinoma and rash in the combination therapy cohort.44 OS data for a phase 2 study comparing dabrafenib alone and dabrafenib plus trametinib were recently reported and showed the OS rate approaching 2 years in the group treated with dabrafenib 150 mg twice daily and trametinib 2 mg daily.58 Two phase 3 trials studying the combination of BRAF and MEK inhibitors have completed accrual, and the final results are anxiously anticipated (NCT01584648, NCT01597908). Contingent upon the successful trial completion and phase 3 study results, in January 2014, the FDA approved combination dabrafenib/trametinib for the treatment of metastatic BRAF-mutant melanoma.14 Based on these results, it is anticipated that combination therapy will become the new standard of care over single-agent BRAF-inhibitor therapy for BRAF-mutant metastatic melanoma. Wagle et al59 recently reported results from whole exome sequencing and whole transcriptome sequencing of a case series of tumor tissue obtained from 5 patients prior to combined BRAF/MEK treatment and after progression. They detected a MEK2 mutation (MEK2Q60P), a novel BRAF splice variant, and BRAF amplification in tumor tissue at progression in 3 patients but were unable to find a resistance mechanism in 2 patients. MAPK signaling reactivation appears to be a primary driver of clinical resistance to both singleagent therapy and combined BRAF and MEK inhibitor in BRAF-mutated melanoma.54-56,59 The study by Wagle et al59 highlights the importance of obtaining paired biopsy samples from patients enrolled in clinical trials prior to treatment and at the time of disease progression to elucidate mechanisms of acquired resistance and develop more effective therapeutic strategies. Current clinical trials in BRAF-mutant melanoma study BRAF inhibitors in combination with novel agents and in the adjuvant setting. LCCC 1128 is an open-label phase 2 study of dabrafenib and trametinib that will evaluate tumor tissue of patients with stage 3 or 4 BRAF-mutant melanoma at study entry and at the time of progression to study mechanisms of acquired resistance (NCT01726738). Vemurafenib is the subject of combination trials with P1446A-05, an oral cyclindependent kinase inhibitor (NCT01841463), PLX3397, and oral multikinase inhibitor (NCT01826448), PX866, an irreversible PI3K inhibitor (NCT01616199), XL888, a HSP90 inhibitor (NCT01657591) and interleukin-2 in patients with metastatic disease who are July 2014, Vol. 21, No. 3

naive to BRAF-targeted therapy (NCT01754376). A placebo-controlled, randomized, double-blind study comparing dabrafenib and trametinib versus 2 placebos in patients with surgically resected, high-risk, BRAF V600 mutation-positive melanoma is ongoing to study the role of dual BRAF/MEK inhibition in the adjuvant setting (NCT01682083).

Colorectal Adenocarcinoma Colorectal adenocarcinoma is the third most commonly diagnosed malignancy in both men and women and led to approximately 50,830 deaths in the United States in 2013.29 Mutations in KRAS were discovered in 1983 and are used as a biomarker to predict response to anti-EGFR monoclonal antibodies in patients with metastatic CRC.60,61 Although KRAS mutations are commonly observed in CRC with a frequency approaching 40%, additional mutations in BRAF, PIK3CA, and PTEN have been described.62 Knowledge of the mutational background of CRC has generated significant interest in developing combinatorial therapeutics to target the MAPK pathway. Notable differences exist in the clinical behavior and pathogenesis of BRAF-mutant CRC when compared with melanoma and other cancers that harbor BRAF mutations. Most cases of colon cancers arise from chromosomal instability and aneuploidy; however, 15% of CRCs develop in the setting of microsatellite instability (MSI), leading to the accumulation of base pair substitutions, frameshift mutations, and small deletions.63 Hereditary nonpolyposis colorectal cancer (HNPCC) is a predisposition syndrome responsible for 3% of CRC and is associated with defective mismatch repair (MMR) machinery leading to MSI.64,65 However, most tumors with MSI arise sporadically and are not associated with HNPCC.64 Such tumors frequently arise from hypermethylation of CpG-rich regions within the promoter region of the MLH1 gene, leading to the CpG island methylator phenotype (CIMP).66 Rajagopalan et al67 were the first to link BRAF status and defective MMR in CRC based on their analysis of 330 cases of CRC. Two years later, Kambara et al68 reported a significant association between BRAF mutations in CIMP-high (20 of 26; 77%), CIMP-low (8 of 44; 18%), and CIMP-negative (0 of 34; 0%) CRCs (P < .0001), as well as between BRAF status and sporadic MSI-high cancers (16 of 17 CIMP-high harbored the V600E mutation compared with 5 of 9 CIMP-low and 0 of 2 CIMP negative; P = .004). Using a novel technique to detect CIMP-positive tumors, Weisenberger et al69 described a highly significant association between CIMP positivity and BRAF mutation and MLH1 methylation. Thus, tumors arising from MSI without genetic predisposition were shown to be highly associated with CIMP and BRAF mutation, defining a unique subset of CRC. Cancer Control 225

Mutations in BRAF were first reported in CRC pression and produced greater growth inhibition in tumors by Yuen et al70 with an observed incidence of all CRC cell lines when combined with the BRAF 5.1% in adenocarcinoma tissue. A population-based inhibitor vemurafenib than with the BRAF inhibitor study detected BRAF mutations in 78 of 513 CRC alone.81 Prahallad et al82 recently demonstrated that tumors (15.2%), and studies of BRAF mutations in BRAF-mutant CRC cell lines treated with vemurafenib patients with metastatic disease report an incidence experienced EGFR feedback activation, and a comof 7.9% to 8.7%.71-73 Similar to thyroid cancer and bined inhibition of EGFR using either cetuximab or melanoma, yet unlike lung adenocarcinoma, BRAF gefitinib (a small molecule EGFR inhibitor) with BRAF mutations almost exclusively affect codon 600.70-72 In inhibition produced synergistic growth inhibition in multiple studies BRAF mutations are mutually excluboth cell lines and mouse xenograft models. sive to KRAS or PIK3CA mutations.72,74,75 A signifiBased on clinical experience using single-agent cant body of literature supports the observation that BRAF inhibitors in CRC and recent preclinical data, BRAF-mutated CRC has a more aggressive clinical combinatorial therapeutic strategies will be necessary behavior than BRAF wild-type CRC. Early evidence to improve outcomes in patients with BRAF-mutant by Ogino et al76 reported that mutations in this gene CRC. The Figure summarizes a selection of ongoing were associated with an increased cancer-specific efforts to combine pathway inhibitors in CRC as well mortality rate (multivariate HR = 1.97; 95% CI, 1.13– as melanoma, NSCLC, and thyroid cancer. In CRC, 3.42) among patients with stages 1 to 4 CRC. BRAFclinical trials targeting both BRAF and EGFR are acmutant CRC was associated with inferior OS rates tively recruiting patients, including a pilot study of among patients with stage 3 resected CRC treated vemurafenib and panitumumab (NCT01791309) and in a large, prospective adjuvant chemotherapy trial, an open label, 3-part, phase 1/2 study combining daband a meta-analysis of 26 CRC studies revealed a sigrafenib with or without trametinib with panitumumab nificantly increased risk of overall mortality among (NCT01750918). Two combination trials studying dual patients with a BRAF mutation (HR = 2.25; 95% CI, PI3K pathway and BRAF inhibition are ongoing, in1.37–2.12).77,78 Studies have also documented inferior cluding a phase 1/2 study of the BRAF inhibitor enPFS rates in BRAF-mutant CRC treated with anticorafenib with or without the PI3K inhibitor BYL719 EGFR antibodies.79,80 plus panitumumab (NCT01719380) and a phase 1/2 Clinical experience with BRAF inhibition in CRC suggests significant differences in response compared with melanoma. Kopetz et al11 reported results of vemurafenib in a phase 1 study of 21 patients with metastatic CRC, with 1 confirmed PR out of 19 evaluable patients. Several studies have subsequently examined the etiology of intrinsic (primary) resistance to BRAF inhibition in patients with CRC. By comparing BRAF-mutant CRC and melanoma cell lines, Mao et al81 reported increased levels of PI3K/Akt activation and lower levels of MEK pathway activation. Inhibition of both the BRAF and PI3K pathways resulted in synergistic growth inhibiFigure. — Approved and investigational agents for the treatment of BRAF-mutated malignancies. Dabrafenib, tion in CRC cell lines. The trametinib, and vemurafenib have been approved by the US Food and Drug Administration for the treatment of authors also reported that metastatic melanoma, either alone or in combination. In the setting of colorectal cancer, current trials are studying the use of 5-azacytidine (a the effects of anti–epidermal growth factor receptor therapy in combination with panitumumab and dabrafenib with or without trametinib and panitumumab and encorafenib with or without BYL719. Dabrafenib and trametinib are hypomethylating agent) under investigation in both thyroid and non–small-cell lung cancers harboring the V600 mutation. Case reports reduced phospho-Akt ex- suggest activity exists with either single-agent dabrafenib or vemurafenib in hairy cell leukemia. 226 Cancer Control

July 2014, Vol. 21, No. 3

study combining dabrafenib with the Akt inhibitor GSK214795 (NCT01902173).

Differentiated Thyroid Cancer More than 60,000 new cases of thyroid malignancies were diagnosed in 2013, many of which were differentiated thyroid tumors in women.29 Thyroid malignancies are commonly categorized according to their aggressiveness. Well-differentiated tumors (papillary and follicular) comprise the majority of new cases each year and have the least aggressive clinical behavior. Intermediate tumors (medullary thyroid carcinoma, Hürthle cell, and poorly differentiated) and undifferentiated tumors (anaplastic thyroid carcinoma [ATC]) make up 10% of newly diagnosed cases each year but have more aggressive clinical behavior.83 Evidence for BRAF mutations in differentiated thyroid cancers was first reported by Kimura et al5 who noted that 28 of their 78 studied patients (35.8%) with papillary thyroid cancer (PTC) possessed the BRAF V600E mutation independent of mutations in RET and RAS. Notably, no BRAF mutations were detected in small cohorts of follicular or Hürthle cell carcinomas.5 Nikifirova et al84 were the first to report that BRAF mutations were restricted to patients with PTC or with poorly differentiated or undifferentiated thyroid cancers arising from previous PTC. A recent compilation of data from 29 studies reported BRAF V600E mutations in 44% of patients with PTC and 24% with ATC.85 Similar to melanoma, the majority of BRAF mutations in PTC or ATC occurs at codon 600, although rare mutations in codons near codon 600 have been described.86 In addition to characterizing BRAF mutations, additional work has demonstrated that genetic alterations in the PI3K/Akt pathway are common in thyroid malignancies, occur with increasing frequency in more aggressive tumors, and are mutually exclusive.87,88 BRAF V600E mutation predicts for a more aggressive clinical course in patients with PTC.89-91 Using multivariate analysis, BRAF mutations in PTC are associated with an increased risk of lymph node invasion and metastasis as well as a more advanced stage of disease at initial surgery.85 In addition, BRAF mutations render tumors less responsive to repeat radioactive iodine treatment in the event of recurrent disease.85,92,93 A large retrospective study consisting of 1,849 patients with PTC found an increased overall mortality rate among patients with the V600E mutation (12.87 deaths per 1,000 person-years; 95% CI, 9.61–17.24) compared with patients not carrying the mutation (2.52 deaths per 1,000 person-years; 95% CI, 1.40–4.55) with an HR of 2.66 (95% CI, 1.30–5.43) after adjusting for age at diagnosis, sex, and medical center.94 However, in the same study, when additional factors associated with worse prognosis, including lymph node metastasis, July 2014, Vol. 21, No. 3

extrathyroid invasion, and distant metastasis, were included in the model, BRAF mutations were no longer associated with increased mortality. Preclinical work in thyroid cancer cell lines using MEK1/2 and BRAF inhibitors, along with clinical experience using these drugs in melanoma, has led to clinical trials in patients with thyroid cancer targeting the MAPK and PI3K/Akt pathways.95-97 In a phase 2 study of sorafenib for metastatic thyroid cancer, 6 out of 41 patients with PTC achieved a PR with a median duration of 7.5 months, while none of the 17 patients with a different type of thyroid cancer achieved a PR.98 Among patients with PTC and V600E mutation, 3 out of 9 evaluable patients participating in a phase 1, dose-escalation trial of dabrafenib experienced a PR.8 One case reported highlighted a rapid clinical response to treatment with vemurafenib in a patient with ATC found to have V600E mutation.99 Active clinical trials in thyroid cancers include dabrafenib with or without the MEK inhibitor trametinib (NCT01723202), neoadjuvant vemurafenib in patients with locally advanced thyroid cancer (NCT01709292), and the use of the BRAF inhibitor dabrafenib to resensitize patients with BRAF-mutated thyroid cancer to radioactive iodine (NCT01534897).

Non–Small-Cell Lung Cancer Lung cancer remains the leading cause of annual cancer-related mortality in the United States.29 Among patients with lung adenocarcinoma, driver mutations have been identified in 62% of patients undergoing testing for at least 1 genomic alteration.100 Mutations in BRAF arise in 1.6% to 4.9% of adenocarcinomas.1,12,13,101,102 Unlike melanoma, PTC, and HCL in which most BRAF mutations are V600E, mutations in lung adenocarcinomas can be separated into either V600E (50%–56.8%) and non-V600E (43.2%–50%).12,13 Although V600E mutations directly phosphorylate MEK, mutations in exon 11 (non-V600E) possess impaired kinase activity and are dependent on CRAF to mediate downstream signaling through MEK and ERK.103 BRAF mutations are considered to be mutually exclusive with common driver mutations in NSCLC such as EGFR, KRAS, and the EML4-ALK rearrangement.104 However, one recent analysis detected 1 out of 18 patients with a coexisting V600E mutation and PIK3CA E545K mutation and 2 out of 18 patients with non-V600E mutation and coexisting KRAS mutation.12 In addition to lung adenocarcinoma, recent data from the Cancer Genome Atlas Research Network105 revealed a 4% mutation rate in squamous cell carcinoma of the lung. Clinical predictors of BRAF mutations in lung adenocarcinomas are controversial. Paik et al104 reported a statistically significant association between the presence of BRAF mutations and current or previous smoking status, while a larger study by Cardarella et al12 Cancer Control 227

reported no significant association between smoking status and BRAF mutations. No difference in PFS was seen after first-line platinum doublet chemotherapy between patients with V600E mutation compared with patients without a driver mutation (EGFR, KRAS, and BRAF wild-type and EML4-ALK nontranslocated). In addition, no difference in PFS was detected between patients with V600E and non-V600E mutations.12 Nonetheless, differences in response to BRAF inhibition with vemurafenib have been reported based on V600E mutation status (objective response in 1 patient with V600E mutation and primary progressive disease in a patient with non-V600E mutation).106,107 Two patients with NSCLC and V600E mutation have also experienced treatment response to dabrafenib.8,108 Yet another patient with Y472C BRAF mutation experienced a prolonged remission after treatment with dasatinib.109 Another mechanism to inhibit downstream activation of ERK in patients with both V600E and non-V600E mutations is through MEK inhibition, which was the subject of a completed phase 1 study using selumetinib, a MEK1/2 inhibitor, in nonmelanoma, BRAF-mutated solid tumors (NCT00888134).110 Active clinical trials in patients with NSCLC and both V600E and non-V600E mutant BRAF are ongoing. Dabrafenib is the subject of an ongoing international phase 2 trial in V600E-mutant NSCLC (NCT01336634). Interim results were recently presented and revealed an overall response rate of 54% (7 PRs of 13 patients evaluable for response).111 Dasatinib is currently the subject of an ongoing phase 2 trial in patients with non-V600E inactivating BRAF mutations (NCT01514864).

Hairy Cell Leukemia HCL is a rare clonal disorder of B cells characterized by progressive splenomegaly, pancytopenia, and the absence of peripheral lymphadenopathy.112 Its annual incidence is estimated to be 3.3 persons per 1 million person-years in the United States.113 The underlying genomic etiology of HCL remained elusive until a landmark paper by Tiacci et al.6 Using massively paralleled whole exome sequencing to compare leukemia cells and nonleukemia cells in a single patient with HCL, the researchers identified 5 unique nonsynonymous mutations, one of which included the BRAF V600E mutation.6 Forty-six patients with HCL underwent polymerase chain reaction and Sanger sequencing for the V600E mutation, all of whom were found to harbor the V600E mutation.6 In a second cohort of 62 patients with HCL and 178 patients with splenic marginal zone lymphoma, Waldenström macroglobulinemia, or chronic B-cell lymphoproliferative disorders, V600E was detected in all 62 patients with HCL and only 2 of the remaining 178 patients.114 When patients require treatment for HCL, a single course of continuous infusion cladribine over 7 days 228 Cancer Control

induces CRs in more than 90% of patients with an OS rate of 96% at 48 months.115 After 7 years of follow-up in the same cohort of patients, the median duration of first response was 98 months, with 37% of patients experiencing relapsed disease.116 In patients with relapsed disease, re-treatment with cladribine induced complete remission in the majority of patients.115-117 Thus, while cladribine is a highly effective treatment for HCL, eventual relapse is common. After the discovery of the V600E mutation, Dietrich et al118 offered off-label vemurafenib to a patient with refractory HCL, massive splenomegaly, and cytopenias. The patient was treated for 56 days after dose escalation to a maximum of 1440 mg/day and experienced significant reduction in spleen size (24.8 × 8.3 cm pretreatment to 14 × 5 cm at day 16) and hematological CR at 43 days.118,119 Six months after completing vemurafenib treatment, the patient remained in complete remission.119 Another patient with refractory HCL was treated with vemurafenib 960 mg orally twice daily for 3 weeks and was recently reported to have a normalization of platelets, white blood cells, and neutrophils at 5 weeks; this normalization persisted for 4 months after treatment discontinuation.120 The optimal dosing and duration of treatment using BRAF inhibitors is unknown for patients with treatment refractory HCL119,120; however, a multicenter, phase 2 study of vemurafenib in treatment refractory HCL is ongoing (NCT01711632).

Conclusion The characterization and discovery of BRAF mutations in both epithelial and hematological malignancies illustrates the promise of personalized medicine in oncology. The detection of BRAF mutations in a variety of malignancies is still ongoing, and it is leading to rapid drug development across a wide range of cancers. Although we have seen significant improvements in patient outcomes within the last 4 years, particularly in melanoma, more research is needed to understand the mechanisms of intrinsic and acquired resistance in BRAF inhibitors. Applying whole exome sequencing and improving the capabilities of bioinformatics will contribute to continued gains in the development of combinatorial therapeutic strategies against mutated BRAF and the effects of its downstream signaling. References 1. Davies H, Bignell GR, Cox C, et al. Mutations of the BRAF gene in human cancer. Nature. 2002;417(6892):949-954. 2. Andrulis M, Lehners N, Capper D, et al. Targeting the BRAF V600E mutation in multiple myeloma. Cancer Discov. 2013;3(8):862-869. 3. Bonello L, Voena C, Ladetto M, et al. BRAF gene is not mutated in plasma cell leukemia and multiple myeloma. Leukemia. 2003;17(11): 2238-2240. 4. Chapman MA, Lawrence MS, Keats JJ, et al. Initial genome sequencing and analysis of multiple myeloma. Nature. 2011;471(7339):467-472. 5. Kimura ET, Nikiforova MN, Zhu Z, et al. High prevalence of BRAF mutations in thyroid cancer: genetic evidence for constitutive activation of the RET/PTC-RAS-BRAF signaling pathway in papillary thyroid carcinoma. Cancer Res. 2003;63(7):1454-1457. July 2014, Vol. 21, No. 3

6. Tiacci E, Trifonov V, Schiavoni G, et al. BRAF mutations in hairy-cell leukemia. N Engl J Med. 2011;364(24):2305-2315. 7. 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(26):2507-2516. 8. Falchook GS, Long GV, Kurzrock R, et al. Dabrafenib in patients with melanoma, untreated brain metastases, and other solid tumours: a phase 1 dose-escalation trial. Lancet. 2012;379(9829):1893-1901. 9. Flaherty KT, Puzanov I, Kim KB, et al. Inhibition of mutated, activated BRAF in metastatic melanoma. N Engl J Med. 2010;363(9):809-819. 10. Hauschild A, Grob JJ, Demidov LV, et al. Dabrafenib in BRAF-mutated metastatic melanoma: a multicentre, open-label, phase 3 randomised controlled trial. Lancet. 2012;380(9839):358-365. 11. Kopetz S, Desai J, Chan E, et al. PLX4032 in metastatic colorectal cancer patients with mutant BRAF tumors. Paper presented at: American Society of Clinical Oncology Annual Meeting, June 4–8, 2010, Chicago, IL. 12. Cardarella S, Ogino A, Nishino M, et al. Clinical, pathologic, and biologic features associated with BRAF mutations in non-small cell lung cancer. Clin Cancer Res. 2013;19(16):4532-4540. 13. Marchetti A, Felicioni L, Malatesta S, et al. Clinical features and outcome of patients with non-small-cell lung cancer harboring BRAF mutations. J Clin Oncol. 2011;29(26):3574-3579. 14. US Food and Drug Administration. FDA approves Mekinist in combination with Tafinlar for advanced melanoma [press release]. Published January 10, 2014. http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ ucm381159.htm. Accessed April 10, 2014. 15. US Food and Drug Administration. Vemurafenib. http://www.fda.gov/ aboutfda/centersoffices/officeofmedicalproductsandtobacco/cder/ucm268301. htm. Accessed April 10, 2014. 16. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646-674. 17. Chang L, Karin M. Mammalian MAP kinase signalling cascades. Nature. 2001;410(6824):37-40. 18. Sharrocks AD. The ETS-domain transcription factor family. Nat Rev Mol Cell Biol. 2001;2(11):827-837. 19. Salomon DS, Brandt R, Ciardiello F, et al. Epidermal growth factorrelated peptides and their receptors in human malignancies. Crit Rev Oncol Hematol. 1995;19(3):183-232. 20. Schlessinger J. Cell signaling by receptor tyrosine kinases. Cell. 2000;103(2):211-225. 21. Downward J. Targeting RAS signalling pathways in cancer therapy. Nat Rev Cancer. 2003;3(1):11-22. 22. Bos JL. ras oncogenes in human cancer: a review [published correction appears in Cancer Res. 1990;50(4):1352]. Cancer Res. 1989;49(17):4682-4689. 23. Omholt K, Platz A, Kanter L, et al. NRAS and BRAF mutations arise early during melanoma pathogenesis and are preserved throughout tumor progression. Clin Cancer Res. 2003;9(17):6483-6488. 24. Chong H, Vikis HG, Guan KL. Mechanisms of regulating the Raf kinase family. Cell Signal. 2003;15(5):463-469. 25. Lito P, Rosen N, Solit DB. Tumor adaptation and resistance to RAF inhibitors. Nat Med. 2013;19(11):1401-1409. 26. Tuveson DA, Weber BL, Herlyn M. BRAF as a potential therapeutic target in melanoma and other malignancies. Cancer Cell. 2003;4(2):95-98. 27. Pratilas CA, Taylor BS, Ye Q, et al. (V600E)BRAF is associated with disabled feedback inhibition of RAF-MEK signaling and elevated transcriptional output of the pathway. Proc Natl Acad Sci U S A. 2009;106(11):4519-4524. 28. Liu J, Suresh Kumar KG, Yu D, et al. Oncogenic BRAF regulates betaTrcp expression and NF-kappaB activity in human melanoma cells. Oncogene. 2007;26(13):1954-1958. 29. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA Cancer J Clin. 2013;63(1):11-30. 30. Balch CM, Gershenwald JE, Soong SJ, et al. Final version of 2009 AJCC melanoma staging and classification. J Clin Oncol. 2009;27(36):6199-6206. 31. Pollock PM, Harper UL, Hansen KS, et al. High frequency of BRAF mutations in nevi. Nat Genet. 2003;33(1):19-20. 32. Colombino M, Capone M, Lissia A, et al. BRAF/NRAS mutation frequencies among primary tumors and metastases in patients with melanoma. J Clin Oncol. 2012;30(20):2522-2529. 33. Long GV, Menzies AM, Nagrial AM, et al. Prognostic and clinicopathologic associations of oncogenic BRAF in metastatic melanoma. J Clin Oncol. 2011;29(10):1239-1246. 34. Menzies AM, Haydu LE, Visintin L, et al. Distinguishing clinicopathologic features of patients with V600E and V600K BRAF-mutant metastatic melanoma. Clin Cancer Res. 2012;18(12):3242-3249. 35. Bucheit AD, Syklawer E, Jakob JA, et al. Clinical characteristics and outcomes with specific BRAF and NRAS mutations in patients with metastatic melanoma. Cancer. 2013;119(21):3821-3829. 36. Lee JH, Choi JW, Kim YS. Frequencies of BRAF and NRAS mutations are different in histological types and sites of origin of cutaneous melanoma: a meta-analysis. Br J Dermatol. 2011;164(4):776-784. 37. Sasaki Y, Niu C, Makino R, et al. BRAF point mutations in primary melanoma show different prevalences by subtype. J Invest Dermatol. 2004;123(1):177-183. July 2014, Vol. 21, No. 3

38. Karasarides M, Chiloeches A, Hayward R, et al. B-RAF is a therapeutic target in melanoma. Oncogene. 2004;23(37):6292-6298. 39. Flaherty KT, Schiller J, Schuchter LM, et al. A phase I trial of the oral, multikinase inhibitor sorafenib in combination with carboplatin and paclitaxel. Clin Cancer Res. 2008;14(15):4836-4842. 40. Ott PA, Hamilton A, Min C, et al. A phase II trial of sorafenib in metastatic melanoma with tissue correlates. PLoS One. 2010;5(12):e15588. 41. Hauschild A, Agarwala SS, Trefzer U, et al. Results of a phase III, randomized, placebo-controlled study of sorafenib in combination with carboplatin and paclitaxel as second-line treatment in patients with unresectable stage III or stage IV melanoma. J Clin Oncol. 2009;27(17):2823-2830. 42. Amaravadi RK, Schuchter LM, McDermott DF, et al. Phase II trial of temozolomide and sorafenib in advanced melanoma patients with or without brain metastases. Clin Cancer Res. 2009;15(24):7711-7718. 43. Davies MA, Fox PS, Papadopoulos NE, et al. Phase I study of the combination of sorafenib and temsirolimus in patients with metastatic melanoma. Clin Cancer Res. 2012;18(4):1120-1128. 44. Margolin KA, Moon J, Flaherty LE, et al. Randomized phase II trial of sorafenib with temsirolimus or tipifarnib in untreated metastatic melanoma (S0438). Clin Cancer Res. 2012;18(4):1129-1137. 45. Chapman PB, hauschild A, robert C, et al. Updated overall survival (OS) results for BRIM-3, a phase III randomized, open-label, multicenter trial comparing BRAF inhibitor vemurafenib (vem) with dacarbazine (DTIC) in previously untreated patients with BRAFV600E-mutated melanoma. Paper presented at: American Society of Clinical Oncology Annual Meeting, June 1–5, 2012, Chicago, IL. 46. Flaherty KT, Infante JR, Daud A, et al. Combined BRAF and MEK inhibition in melanoma with BRAF V600 mutations. N Engl J Med. 2012; 367(18):1694-1703. 47. Haass NK, Sproesser K, Nguyen TK, et al. The mitogen-activated protein/extracellular signal-regulated kinase kinase inhibitor AZD6244 (ARRY142886) induces growth arrest in melanoma cells and tumor regression when combined with docetaxel. Clin Cancer Res. 2008;14(1):230-239. 48. Smalley KS, Contractor R, Haass NK, et al. Ki67 expression levels are a better marker of reduced melanoma growth following MEK inhibitor treatment than phospho-ERK levels. Br J Cancer. 2007;96(3):445-449. 49. Solit DB, Garraway LA, Pratilas CA, et al. BRAF mutation predicts sensitivity to MEK inhibition. Nature. 2006;439(7074):358-362. 50. Falchook GS, Lewis KD, Infante JR, et al. Activity of the oral MEK inhibitor trametinib in patients with advanced melanoma: a phase 1 dose-escalation trial. Lancet Oncol. 2012;13(8):782-789. 51. Kim KB, Kefford R, Pavlick AC, et al. Phase II study of the MEK1/ MEK2 inhibitor Trametinib in patients with metastatic BRAF-mutant cutaneous melanoma previously treated with or without a BRAF inhibitor. J Clin Oncol. 2013;31(4):482-489. 52. Montagut C, Sharma SV, Shioda T, et al. Elevated CRAF as a potential mechanism of acquired resistance to BRAF inhibition in melanoma. Cancer Res. 2008;68(12):4853-4861. 53. Smalley KS, Haass NK, Brafford PA, et al. Multiple signaling pathways must be targeted to overcome drug resistance in cell lines derived from melanoma metastases. Mol Cancer Ther. 2006;5(5):1136-1144. 54. Nazarian R, Shi H, Wang Q, et al. Melanomas acquire resistance to B-RAF(V600E) inhibition by RTK or N-RAS upregulation. Nature. 2010;468(7326):973-977. 55. Johannessen CM, Boehm JS, Kim SY, et al. COT drives resistance to RAF inhibition through MAP kinase pathway reactivation. Nature. 2010;468(7326):968-972. 56. Trunzer K, Pavlick AC, Schuchter L, et al. Pharmacodynamic effects and mechanisms of resistance to vemurafenib in patients with metastatic melanoma. J Clin Oncol. 2013;31(14):1767-1774. 57. Paraiso KH, Fedorenko IV, Cantini LP, et al. Recovery of phospho-ERK activity allows melanoma cells to escape from BRAF inhibitor therapy. Br J Cancer. 2010;102(12):1724-1730. 58. Daud A, Weber JS, Sosman JA, et al. Overall Survival (OS) Update for BRF113220 Part C, a Phase II three-arm randomized study of dabrafenib (D) alone vs. combination of dabrafenib and trametinib (D+T) in pts with BRAF V600 mutation-positive (+) metastatic melanoma (MM). Paper presented at: Society of Melanoma Research, November 17–20, 2013, Philadelphia, PA. 59. 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(1):61-68. 60. Feinberg AP, Vogelstein B, Droller MJ, et al. Mutation affecting the 12th amino acid of the c-Ha-ras oncogene product occurs infrequently in human cancer. Science. 1983;220(4602):1175-1177. 61. Karapetis CS, Khambata-Ford S, Jonker DJ, et al. K-ras mutations and benefit from cetuximab in advanced colorectal cancer. N Engl J Med. 2008;359(17):1757-1765. 62. De Roock W, De Vriendt V, Normanno N, et al. KRAS, BRAF, PIK3CA, and PTEN mutations: implications for targeted therapies in metastatic colorectal cancer. Lancet Oncol. 2011;12(6):594-603. 63. Kinzler KW, Vogelstein B. Lessons from hereditary colorectal cancer. Cell. 1996;87(2):159-170. 64. Cunningham D, Atkin W, Lenz HJ, et al. Colorectal cancer. Lancet. Cancer Control 229

2010;375(9719):1030-1047. 65. Hampel H, Frankel WL, Martin E, et al. Feasibility of screening for Lynch syndrome among patients with colorectal cancer. J Clin Oncol. 2008;26(35):5783-5788. 66. Toyota M, Ahuja N, Ohe-Toyota M, et al. CpG island methylator phenotype in colorectal cancer. Proc Natl Acad Sci U S A. 1999;96(15):8681-8686. 67. Rajagopalan H, Bardelli A, Lengauer C, et al. Tumorigenesis: RAF/ RAS oncogenes and mismatch-repair status. Nature. 2002;418(6901):934. 68. Kambara T, Simms LA, Whitehall VL, et al. BRAF mutation is associated with DNA methylation in serrated polyps and cancers of the colorectum. Gut. 2004;53(8):1137-1144. 69. Weisenberger DJ, Siegmund KD, Campan M, et al. CpG island methylator phenotype underlies sporadic microsatellite instability and is tightly associated with BRAF mutation in colorectal cancer. Nat Genet. 2006;38(7):787-793. 70. Yuen ST, Davies H, Chan TL, et al. Similarity of the phenotypic patterns associated with BRAF and KRAS mutations in colorectal neoplasia. Cancer Res. 2002;62(22):6451-6455. 71. Richman SD, Seymour MT, Chambers P, et al. KRAS and BRAF mutations in advanced colorectal cancer are associated with poor prognosis but do not preclude benefit from oxaliplatin or irinotecan: results from the MRC FOCUS trial. J Clin Oncol. 2009;27(35):5931-5937. 72. Tol J, Nagtegaal ID, Punt CJ. BRAF mutation in metastatic colorectal cancer. N Engl J Med. 2009;361(1):98-99. 73. Vaughn CP, Zobell SD, Furtado LV, et al. Frequency of KRAS, BRAF, and NRAS mutations in colorectal cancer. Genes Chromosomes Cancer. 2011;50(5):307-312. 74. Simi L, Pratesi N, Vignoli M, et al. High-resolution melting analysis for rapid detection of KRAS, BRAF, and PIK3CA gene mutations in colorectal cancer. Am J Clin Pathol. 2008;130(2):247-253. 75. Tian S, Simon I, Moreno V, et al. A combined oncogenic pathway signature of BRAF, KRAS and PI3KCA mutation improves colorectal cancer classification and cetuximab treatment prediction. Gut. 2013;62(4):540-549. 76. Ogino S, Nosho K, Kirkner GJ, et al. CpG island methylator phenotype, microsatellite instability, BRAF mutation and clinical outcome in colon cancer. Gut. 2009;58(1):90-96. 77. Ogino S, Shima K, Meyerhardt JA, et al. Predictive and prognostic roles of BRAF mutation in stage III colon cancer: results from intergroup trial CALGB 89803. Clin Cancer Res. 2012;18(3):890-900. 78. Safaee Ardekani G, Jafarnejad SM, Tan L, et al. The prognostic value of BRAF mutation in colorectal cancer and melanoma: a systematic review and meta-analysis. PLoS One. 2012;7(10):e47054. 79. Douillard JY, Oliner KS, Siena S, et al. Panitumumab-FOLFOX4 treatment and RAS mutations in colorectal cancer. N Engl J Med. 2013;369(11): 1023-1034. 80. Van Cutsem E, Kohne CH, Lang I, et al. Cetuximab plus irinotecan, fluorouracil, and leucovorin as first-line treatment for metastatic colorectal cancer: updated analysis of overall survival according to tumor KRAS and BRAF mutation status. J Clin Oncol. 2011;29(15):2011-2019. 81. Mao M, Tian F, Mariadason JM, et al. Resistance to BRAF inhibition in BRAF-mutant colon cancer can be overcome with PI3K inhibition or demethylating agents. Clin Cancer Res. 2013;19(3):657-667. 82. Prahallad A, Sun C, Huang S, et al. Unresponsiveness of colon cancer to BRAF(V600E) inhibition through feedback activation of EGFR. Nature. 2012;483(7387):100-103. 83. Carling TU, Robert. Thyroid Tumors. In: Jr. JWP, ed. Devita, Hellman, and Rosenberg’s Cancer: Principles & Practice of Oncology. 9th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2011: 1457-1472. 84. Nikiforova MN, Kimura ET, Gandhi M, et al. BRAF mutations in thyroid tumors are restricted to papillary carcinomas and anaplastic or poorly differentiated carcinomas arising from papillary carcinomas. J Clin Endocrinol Metab. 2003;88(11):5399-5404. 85. Xing M. BRAF mutation in thyroid cancer. Endocr Relat Cancer. 2005;12(2):245-262. 86. Hou P, Liu D, Xing M. Functional characterization of the T1799-1801del and A1799-1816ins BRAF mutations in papillary thyroid cancer. Cell Cycle. 2007;6(3):377-379. 87. Hou P, Liu D, Shan Y, et al. Genetic alterations and their relationship in the phosphatidylinositol 3-kinase/Akt pathway in thyroid cancer. Clin Cancer Res. 2007;13(4):1161-1170. 88. Wang Y, Hou P, Yu H, et al. High prevalence and mutual exclusivity of genetic alterations in the phosphatidylinositol-3-kinase/akt pathway in thyroid tumors. J Clin Endocrinol Metab. 2007;92(6):2387-2390. 89. Kim TH, Park YJ, Lim JA, et al. The association of the BRAF(V600E) mutation with prognostic factors and poor clinical outcome in papillary thyroid cancer: a meta-analysis. Cancer. 2012;118(7):1764-1773. 90. Xing M. BRAF mutation in papillary thyroid cancer: pathogenic role, molecular bases, and clinical implications. Endocr Rev. 2007;28(7):742-762. 91. Xing M, Westra WH, Tufano RP, et al. BRAF mutation predicts a poorer clinical prognosis for papillary thyroid cancer. J Clin Endocrinol Metab. 2005;90(12):6373-6379. 92. Barollo S, Pennelli G, Vianello F, et al. BRAF in primary and recurrent papillary thyroid cancers: the relationship with (131)I and 2-[(18)F]fluoro2-deoxy-D-glucose uptake ability. Eur J Endocrinol. 2010;163(4):659-663. 230 Cancer Control

93. Riesco-Eizaguirre G, Gutierrez-Martinez P, Garcia-Cabezas MA, et al. The oncogene BRAF V600E is associated with a high risk of recurrence and less differentiated papillary thyroid carcinoma due to the impairment of Na+/I- targeting to the membrane. Endocr Relat Cancer. 2006;13(1):257-269. 94. Xing M, Alzahrani AS, Carson KA, et al. Association between BRAF V600E mutation and mortality in patients with papillary thyroid cancer. JAMA. 2013;309(14):1493-1501. 95. Ball DW, Jin N, Rosen DM, et al. Selective growth inhibition in BRAF mutant thyroid cancer by the mitogen-activated protein kinase kinase 1/2 inhibitor AZD6244. J Clin Endocrinol Metab. 2007;92(12):4712-4718. 96. Liu D, Xing M. Potent inhibition of thyroid cancer cells by the MEK inhibitor PD0325901 and its potentiation by suppression of the PI3K and NFkappaB pathways. Thyroid. 2008;18(8):853-864. 97. Xing J, Liu R, Xing M, et al. The BRAFT1799A mutation confers sensitivity of thyroid cancer cells to the BRAFV600E inhibitor PLX4032 (RG7204). Biochem Biophys Res Commun. 2011;404(4):958-962. 98. Kloos RT, Ringel MD, Knopp MV, et al. Phase II trial of sorafenib in metastatic thyroid cancer. J Clin Oncol. 2009;27(10):1675-1684. 99. Rosove MH, Peddi PF, Glaspy JA. BRAF V600E inhibition in anaplastic thyroid cancer. N Engl J Med. 2013;368(7):684-685. 100. Johnson B, Kris M, Berry L, et al. A multicenter effort to identify driver mutations and employ targeted therapy in patients with lung adenocarcinomas: the Lung Cancer Mutation Consortium (LCMC). Paper presented at: American Society of Clinical Oncology Annual Meeting, May 31–June 4, 2013, Chicago, IL. 101. Naoki K, Chen TH, Richards WG, et al. Missense mutations of the BRAF gene in human lung adenocarcinoma. Cancer Res. 2002;62(23):7001-7003. 102. Sasaki H, Kawano O, Endo K, et al. Uncommon V599E BRAF mutations in Japanese patients with lung cancer. J Surg Res. 2006;133(2):203-206. 103. Wan PT, Garnett MJ, Roe SM, et al. Mechanism of activation of the RAF-ERK signaling pathway by oncogenic mutations of B-RAF. Cell. 2004;116(6):855-867. 104. Paik PK, Arcila ME, Fara M, et al. Clinical characteristics of patients with lung adenocarcinomas harboring BRAF mutations. J Clin Oncol. 2011;29(15):2046-2051. 105. Cancer Genome Atlas Research Network. Comprehensive genomic characterization of squamous cell lung cancers. Nature. 2012;489(7417): 519-525. 106. Gautschi O, Pauli C, Strobel K, et al. A patient with BRAF V600E lung adenocarcinoma responding to vemurafenib. J Thorac Oncol. 2012;7(10):e23-24. 107. Gautschi O, Peters S, Zoete V, et al. Lung adenocarcinoma with BRAF G469L mutation refractory to vemurafenib. Lung Cancer. 2013;82(2):365-367. 108. Rudin CM, Hong K, Streit M. Molecular characterization of acquired resistance to the BRAF inhibitor dabrafenib in a patient with BRAF-mutant non-small-cell lung cancer. J Thorac Oncol. 2013;8(5):e41-42. 109. Sen B, Peng S, Tang X, et al. Kinase-impaired BRAF mutations in lung cancer confer sensitivity to dasatinib. Sci Transl Med. 2012;4(136):136ra170. 110. Pratilas CA, Hanrahan AJ, Halilovic E, et al. Genetic predictors of MEK dependence in non-small cell lung cancer. Cancer Res. 2008;68(22): 9375-9383. 111. Planchard D, Mazieres J, Riely G, et al. Interim Results of phase II study BRF113928 of dabrafenib in BRAF V600E mutation-positive non-small cell lung cancer. Paper presented at: American Society of Clinical Oncology Annual Meeting, May 31–June 4, 2013, Chicago, IL. 112. Swerdlow S, Campo E, NL H, eds. WHO Classification of Tumours of Haematopoietic and Lymphoid Tissues. Lyon, France: International Agency for Research on Cancer; 2008:188-190. 113. Morton LM, Wang SS, Devesa SS, et al. Lymphoma incidence patterns by WHO subtype in the United States, 1992-2001. Blood. 2006;107(1):265-276. 114. Arcaini L, Zibellini S, Boveri E, et al. The BRAF V600E mutation in hairy cell leukemia and other mature B-cell neoplasms. Blood. 2012;119(1):188-191. 115. Saven A, Burian C, Koziol JA, et al. Long-term follow-up of patients with hairy cell leukemia after cladribine treatment. Blood. 1998;92(6):1918-1926. 116. Goodman GR, Burian C, Koziol JA, et al. Extended follow-up of patients with hairy cell leukemia after treatment with cladribine. J Clin Oncol. 2003;21(5):891-896. 117. Jehn U, Bartl R, Dietzfelbinger H, et al. An update: 12-year follow-up of patients with hairy cell leukemia following treatment with 2-chlorodeoxyadenosine. Leukemia. 2004;18(9):1476-1481. 118. Dietrich S, Glimm H, Andrulis M, et al. BRAF inhibition in refractory hairy-cell leukemia. N Engl J Med. 2012;366(21):2038-2040. 119. Dietrich S, Hullein J, Hundemer M, et al. Continued response off treatment after BRAF inhibition in refractory hairy cell leukemia. J Clin Oncol. 2013;31(19):e300-303. 120. Munoz J, Schlette E, Kurzrock R. Rapid response to vemurafenib in a heavily pretreated patient with hairy cell leukemia and a BRAF mutation. J Clin Oncol. 203;31(20):e351-352.

July 2014, Vol. 21, No. 3

Anti–PD-1/PD-L1 agents have a good safety profile and have resulted in durable responses in a variety of cancers.

Toohey Lake Panorama_0104K (detail). Photograph courtesy of Henry Domke, MD. www.henrydomke.com

PD-1 Pathway Inhibitors: Changing the Landscape of Cancer Immunotherapy Dawn E. Dolan, PharmD, and Shilpa Gupta, MD Background: Immunotherapeutic approaches to treating cancer have been evaluated during the last few decades with limited success. An understanding of the checkpoint signaling pathway involving the programmed death 1 (PD-1) receptor and its ligands (PD-L1/2) has clarified the role of these approaches in tumor-induced immune suppression and has been a critical advancement in immunotherapeutic drug development. Methods: A comprehensive literature review was performed to identify the available data on checkpoint inhibitors, with a focus on anti–PD-1 and anti–PD-L1 agents being tested in oncology. The search included Medline, PubMed, the ClinicalTrials.gov registry, and abstracts from the American Society of Clinical Oncology meetings through April 2014. The effectiveness and safety of the available anti–PD-1 and anti–PD-L1 drugs are reviewed. Results: Tumors that express PD-L1 can often be aggressive and carry a poor prognosis. The anti–PD-1 and anti–PD-L1 agents have a good safety profile and have resulted in durable responses in a variety of cancers, including melanoma, kidney cancer, and lung cancer, even after stopping treatment. The scope of these agents is being evaluated in various other solid tumors and hematological malignancies, alone or in combination with other therapies, including other checkpoint inhibitors and targeted therapies, as well as cytotoxic chemotherapy. Conclusions: The PD-1/PD-L1 pathway in cancer is implicated in tumors escaping immune destruction and is a promising therapeutic target. The development of anti–PD-1 and anti–PD-L1 agents marks a new era in the treatment of cancer with immunotherapies. Early clinical experience has shown encouraging activity of these agents in a variety of tumors, and further results are eagerly awaited from completed and ongoing studies.

Introduction From the Pharmacy Service (DED) and the Genitourinary Oncology Program (SG) at the H. Lee Moffitt Cancer Center & Research Institute, Tampa, Florida. Submitted February 26, 2014; accepted April 29, 2014. Address correspondence to Shilpa Gupta, MD, Genitourinary Oncology Program, 12902 Magnolia Drive, Tampa, FL 33612. E-mail: [email protected] No significant relationships exist between the authors and the companies/organizations whose products or services may be referenced in this article. The authors have disclosed that this article discusses unlabeled/ unapproved uses of anti–PD-1/PD-L1 drugs. July 2014, Vol. 21, No. 3

An intact immune system is capable of recognizing and eliminating tumor cells through immune checkpoints; however, tumors can adapt and circumvent these natural defense mechanisms.1-3 Over the last several decades, significant efforts have targeted and activated the immune system to treat cancers; presently, increasing evidence exists that tumors can evade adaptive immunity and disrupt T-cell checkpoint pathways. The interaction between the programmed death 1 (PD-1) receptor and its ligand 1 and 2 (PD-L1/2) is a key pathway hijacked by tumors to Cancer Control 231

suppress immune control.2,4-7 Reversing the inhibition of adaptive immunity can lead to active stimulation of a patient’s immune systems; one such approach utilizes antagonistic antibodies to block checkpoint pathways, thus releasing tumor inhibition. These antagonistic antibodies target cytotoxic T-lymphocyte antigen 4 (CTLA-4), the PD-1 receptor and PD-L1, block immune checkpoints, and facilitate antitumor activity. These agents are unique among antagonistic antibodies because they target lymphocyte receptors or their ligands.8,9 In this review, we discuss the role of the PD-1/ PD-L1 pathway and the drug development efforts to block this pathway in cancer, focusing on the currently available data from completed and ongoing clinical trials. The clinical development of several anti–PD-1 and anti–PDL-1 agents, their efficacy, toxicity, and scope in these cancers as single agents, or in combination with other therapies, will also be discussed.

Role of PD-1/PD-L1 Pathway PD-1 is an immunoinhibitory receptor that belongs to the CD28 family and is expressed on T cells, B cells, monocytes, natural killer cells, and many tumor-infiltrating lymphocytes (TILs)10; it has 2 ligands that have been described (PD-L1 [B7H1] and PD-L2 [B7-DC]).11 Although PD-L1 is expressed on resting T cells, B cells, dendritic cells, macrophages, vascular endothelial cells, and pancreatic islet cells, PD-L2 expression is seen on macrophages and dendritic cells alone.10 Certain tumors have a higher expression of PD-L1.12 PD-L1 and L2 inhibit T-cell proliferation, cytokine production, and cell adhesion.13 PD-L2 controls immune T-cell activation in lymphoid organs, whereas PD-L1 appears to dampen T-cell function in peripheral tissues.14 PD-1 induction on activated T cells occurs in response to PD-L1 or L2 engagement and limits effector T-cell activity in peripheral organs and tissues during inflammation, thus preventing autoimmunity. This is a crucial step to protect against tissue damage when the immune system is activated in response to infection.15-17 Blocking this pathway in cancer can augment the antitumor immune response.18 Like the CTLA-4, the PD-1 pathway down-modulates Tcell responses by regulating overlapping signaling proteins that are part of the immune checkpoint pathway; however, they function slightly differently.14,16 Although the CTLA-4 focuses on regulating the activation of T cells, PD-1 regulates effector T-cell activity in peripheral tissues in response to infection or tumor progression.16 High levels of CTLA-4 and PD-1 are expressed on regulatory T cells and these regulatory T cells and have been shown to have immune inhibitory activity; thus, they are important for maintaining self-tolerance.16 The role of the PD-1 pathway in the interaction of tumor cells with the host immune response and the 232 Cancer Control

PD-L1 tumor cell expression may provide the basis for enhancing immune response through a blockade of this pathway.16 Drugs targeting the PD-1 pathway may provide antitumor immunity, especially in PD-L1 positive tumors. Various cancers, such as melanoma, hepatocellular carcinoma, glioblastoma, lung, kidney, breast, ovarian, pancreatic, and esophageal cancers, as well as hematological malignancies, have positive PD-L1 expression, and this expression has been correlated with poor prognosis.8,19 Melanoma and kidney cancer are prototypes of immunogenic tumors that have historically been known to respond to immunotherapeutic approaches with interferon alfa and interleukin 2. The CTLA-4 antibody ipilimumab is approved by the US Food and Drug Administration for use in melanoma. Clinical activity of drugs blocking the PD-1/PD-L1 pathway has been demonstrated in melanoma and kidney cancer.20-24 In patients with kidney cancer, tumor, TIL-associated PD-L1 expression, or both were associated with a 4.5-fold increased risk of mortality and lower cancer-specific survival rate, even after adjusting for stage, grade, and performance status.18,19,25,26 A correlation between PD-L1 expression and tumor growth has been described in patients with melanoma, providing the rationale for using drugs that block the PD-1/PD-L1 pathway.19,27 Historically, immunotherapy has been ineffective in cases of non–small-cell lung cancer (NSCLC), which has been thought to be a type of nonimmunogenic cancer; nevertheless, lung cancer can evade the immune system through various complex mechanisms.28 In patients with advanced lung cancer, the peripheral and tumor lymphocyte counts are decreased, while levels of regulatory T cells (CD4+), which help suppress tumor immune surveillance, have been found at higher levels.29-32 Immune checkpoint pathways involving the CTLA-4 or the PD-1/PD-L1 are involved in regulating T-cell responses, providing the rationale for blocking this pathway in NSCLC with antibodies against CTLA-4 and the PD-1/PD-L1 pathway.32 Triple negative breast cancer (TNBC) is an aggressive subset of breast cancer with limited treatment options. PD-L1 expression has been reported in patients with TNBC. When PD-L1 expression was evaluated in TILs, it correlated with higher grade and larger-sized tumors.33 Tumor PD-L1 expression also correlates with the infiltration of T-regulatory cells in TNBC, findings that suggest the role of PD-L1–expressing tumors and the PD-1/PD-L1–expressing TILs in regulating immune response in TNBC.34 The PD-1/PD-L1 interaction may create an initial site for viral infection followed by an adaptive immune resistance, and PD-1 levels may positively correlate with a favorable outcome.35,36 It is hypothesized that human papilloma virus (HPV)–associated oropharynJuly 2014, Vol. 21, No. 3

geal cancers express PD-L1 as an immune evasion mode and PD-L1–expressing tumors were more likely to be HPV positive, thus pointing to the potential role of this pathway as a therapeutic target in HPV-associated head and neck cancer. No correlation existed between PD-L1 expression and disease recurrence, but a correlation was seen between PD-L1 expression and the development of distant metastases.37

Drugs Targeting the PD-1 vs PD-L1 Pathway The anti–PD-1 antibody blocks interactions between PD-1 and its ligands, PD-L1 and PD-L2, while the anti–PD-L1 antibody blocks interactions between PD-L1 and both PD-1 and B7-1 (CD80), which is implicated in the down-modulation of T-cell responses. Several PD-1 and PD-L1 inhibitors are in clinical development in early- and late-stage clinical trials across a wide variety of cancers (Tables 1 and 2). Patterns and Evaluation of Response A finding related to response to the anti–PD-1/PD-L1 drugs is that a flare response can be seen, with transient worsening of disease or its progression before stabilization or tumor regression occurs. Patients may exhibit durable responses, and, after discontinuing therapy, they may respond to re-treatment with these therapies in cases of progression.23 From early clinical experience, both the anti–PD-1 and the anti– PD-L1 drugs appear to have activity in various cancers, but no definitive conclusions can be drawn regarding

the differences in their effectiveness.20-24,38-41 However, looking at available results from several studies, it appears that objective responses for anti–PD-L1 antibodies may be somewhat lower than those with anti–PD-1 antibodies, because the latter blocks signaling via both the PD-L1 and PD-L2.20-24,38-41 Safety The anti–PD-1/PD-L1 agents are relatively well tolerated. However, drug-related adverse events with potential immune-related causes, such as pneumonitis, vitiligo, colitis, hepatitis, hypophysitis, and thyroiditis, can occur. The incidence of immune-related adverse events with anti–PD-1/PD-L1 agents is similar to that seen with ipilimumab but is less severe.20-24,38-42 A comparison of immune-related adverse events with anti–PD-1/PD-L1 drugs, including ipilimumab, is shown in Table 3.20,21,23,39,42 An often severe adverse event that has emerged with these agents is pneumonitis; high levels of PD-L1–expressing antigen-presenting cells seen in the lung may give relevance not only to the toxicity across cancers but also the observed responses in NSCLC.43 Pneumonitis may be associated with anti–PD-1 drugs, not with anti–PD-L1 drugs, making the latter potentially safer.20-24,38-42 PD-L1 Inhibitors BMS-936559/MDX-1105 is a fully human, high affinity, immunoglobulin (Ig) G4 monoclonal antibody to PD-L1. Initial results from a phase 1 trial of 207 patients

Table 1. — Selected Ongoing Clinical Trials of Anti–PD-L1 Drugs Indication

Compound

Clinical Trials No.

Phase

Advanced solid tumors

BMS-936559 MEDI4736

NCT00729664 NCT01693562

1 1

Melanoma

MPDL3280A + vemurafenib MEDI4736 + dabrafenib + trametinib or trametinib alone

NCT01656642 NCT02027961

1b 1/2

NSCLC

MPDL3280A + erlotinib MPDL3280A MPDL3280A MPDL3280A vs docetaxel MPDL3280A vs docetaxel MEDI4736 + tremelimumab

NCT02013219 NCT01846416 NCT02031458 NCT01903993 NCT02008227 NCT02000947

1b 2 2 2 3 1b

RCC

MPDL3280A ± bevacizumab vs sunitinib

NCT01984242

2

Solid or hematological malignancies

MPDL3280A

NCT01375842

1

Solid tumors

MPDL3280A + bevacizumab and/or chemotherapy MPDL3280A + cobimetinib MEDI4736 MEDI4736 + tremelimumab MSB0010718C MSB0010718C

NCT01633970 NCT01988896 NCT01938612 NCT01975831 NCT01943461 NCT01772004

1 1 1 1 1 1

PD-L1 = programmed death ligand 1, NSCLC = non–small-cell lung cancer, RCC = renal cell carcinoma.

July 2014, Vol. 21, No. 3

Cancer Control 233

Table 2. — Ongoing Clinical Trials of Anti–PD-1 Drugs for Solid Tumors Indication

Compound

Clinical Trials No.

Phase

Advanced cancer

AMP-224

NCT01352884

1

Advanced solid tumors

Nivolumab + iliolumbar (anti-KIR)

NCT01714739

1

Castration-resistant prostate cancer, melanoma, NSCLC, RCC

Nivolumab

NCT00730639

1b

Colon

Pembrolizumab

NCT01876511

2

Gastric, head and neck, TNBC, urothelial

Pembrolizumab

NCT01848834

1

Gastric, pancreatic, small-cell lung cancer, TNBC

Nivolumab ± ipilimumab

NCT01928394

1/2

Glioblastoma

Nivolumab ± ipilimumab vs bevacizumab

NCT02017717

2

Hepatocellular

Nivolumab

NCT01658878

1

Hodgkin lymphoma, myeloma, myelodysplastic syndrome, non-Hodgkin lymphoma

Pembrolizumab

NCT01953692

1

Malignant gliomas

Pidilizumab

NCT01952769

1/2

Melanoma

Nivolumab ± ipilimumab vs ipilimumab Nivolumab + ipilimumab vs ipilimumab Nivolumab + ipilimumab Nivolumab sequentially with ipilimumab Nivolumab vs DTIC or carboplatin/paclitaxel after ipilumumab Nivolumab vs DTIC Nivolumab + multiple class 1 peptides and montanide ISA 51 VG Nivolumab + multiple class 1 peptides and montanide ISA 51 VG Nivolumab Pembrolizumab vs chemotherapy Pembrolizumab vs ipilimumab

NCT01844505 NCT01927419 NCT01024231 NCT01783938 NCT01721746 NCT01721772 NCT01176461 NCT01176474 NCT01621490 NCT01704287 NCT01866319

3 2 1 2 3 3 1 1 1 2 3

Melanoma, NSCLC

Pembrolizumab

NCT01295827

1

NSCLC

Nivolumab ± gemcitabine/cisplatin, pemetrexed/cisplatin, carboplatin/paclitaxel, bevacizumab, erlotinib, ipilimumab Nivolumab vs docetaxel Nivolumab vs docetaxel Nivolumab Nivolumab Pembrolizumab vs docetaxel Pembrolizumab

NCT01454102

1

NCT01673867 NCT01642004 NCT01721759 NCT01928576 NCT01905657 NCT02007070

3 3 3 2 2/3 1

Pancreatic

Pidilizumab + gemcitabine

NCT01313416

2

Prostate

Pidilizumab + sipuleucel-T + cyclophosphamide

NCT01420965

2

RCC

Nivolumab + sunitinib, pazopanib, or ipilimumab Nivolumab Nivolumab vs everolimus Nivolumab Pembrolizumab + pazopanib Pidilizumab ± dendritic cell/RCC fusion cell vaccine

NCT01472081 NCT01354431 NCT01668784 NCT01358721 NCT02014636 NCT01441765

1 2 2 1 1 2

Solid tumors

Anti-LAG3 (BMS-986016) ± nivolumab Nivolumab Nivolumab + interleukin-21 AMP-554

NCT01968109 NCT00836888 NCT01629758 NCT02013804

1 1 1 1

Solid tumors, NSCLC

Pembrolizumab

NCT01840579

1

PD-1 = programmed death 1, NSCLC = non–small-cell lung cancer, RCC = renal cell carcinoma, TNBC = triple negative breast cancer.

234 Cancer Control

July 2014, Vol. 21, No. 3

showed durable tumor regression (objective response rate of 6%–17%) and prolonged stabilization of disease (12%–41% at 24 weeks) in patients with advanced cancers, including NSCLC, melanoma, and kidney cancer.20 MPDL3280A is an engineered human monoclonal antibody targeting PD-L1. In a phase 1 study of 171 patients with advanced solid tumors, an overall response rate of 21% was observed in nonselected solid tumors among several patients exhibiting delayed responses following initial radiological progression.39 The 24-week progression free survival rate was 44%. Patients with PD-L1 expressing tumors had an overall response rate of 39% and 12% had progressive disease. Those with PD-L1 tumors had an overall response rate of 13% and 59% had progressive disease.39 Additional anti–PD-L1 agents, including MSB0010718C and MEDI473, are being tested in early-phase trials (see Table 1). PD-1 Inhibitors CT-011/pidilizumab is a humanized IgG1 monoclonal antibody that binds to PD-1. A phase 1 study in 17 patients with advanced stage hematologic malignancies (acute myeloid leukemia, chronic lymphocytic leukemia, Hodgkin lymphoma, multiple myeloma, non-Hodgkin lymphoma) showed a clinical benefit in 33% patients and a prolonged complete response of longer than 68 weeks in 1 patient.38 Several phase 1 and 2 trials are ongoing to study the use of this agent in various solid tumors, including prostate and renal cell cancers (see Table 2).

BMS-936558/MDX-1106/nivolumab is a fully human IgG4 monoclonal antibody against PD-1. The first human study evaluated its safety and tolerability in 39 patients with advanced refractory solid tumors.22 Results of a larger phase 1 study in 296 patients have also been reported.23,24,40 Objective responses were seen in 31% of patients with melanoma, 17% in patients with NSCLC, and 29% in patients with RCC.40 A total of 65% of responders had durable responses lasting for more than 1 year. Stable disease lasting 24 weeks was seen in patients with melanoma (7%), NSCLC (10%), and RCC (27%). The median overall survival rate for patients with melanoma was 16.8 months. PD-L1 expression was tested in 42 patients; 9 out of 25 (36%) patients had PD-L1–expressing tumors and experienced an objective response to PD-1 blockade, while the remaining 17 patients had PD-L1–negative tumors that were nonresponsive.23 Pembrolizumab is a highly selective, humanized IgG4-kappa monoclonal antibody with activity against PD-1. Its safety and efficacy were evaluated in a phase 1 trial in solid tumors.41 The rate of median progression-free survival was more than 7 months; however, the median overall survival rate was not been reached.

Rationale for Combination Therapies Thus far, anti–PD1 and anti–PD-L1 antibodies have yielded promising results with durable responses in several tumors and a reasonable safety profile. Given that these agents produce durable responses despite treatment discontinuation, it is thought that the

Table 3. — Comparison of Immune-Related Adverse Events Between Anti–PD-1/PD-L1 Drugs and Ipilimumab Ipilimumab (%)42

Nivolumab/ BMS-936558 (%)23

Pembrolizumab/ MK-3475 (%)21

Pidilizumab/ CT-011 (%)35

BMS-936559 (%)20

MPDL3280A (%)39

Colitis

7.6/5.3

14

13

0

9

39

Dermatological

43/1.5

23

21/2

Diarrhea

33/5

18

20/1

Fatigue

42/7

32

30/1

Hepatic

13/1

Hypothyroid

8/1

Hypophysitis

1.5/1.5

Infusion reactions

10

Pneumonitis

/1

Pruritus

4/0

0/0

21

Total grade 3/4

45.8

Total immune-related

96.9

0

79

61

39

0

PD = programmed death 1, PD-L1 = programmed death ligand 1, NSCLC = non–small-cell lung cancer, RCC = renal cell carcinoma.

July 2014, Vol. 21, No. 3

Cancer Control 235

re-education of the immune system helps it adapt to tumor manipulation to develop resistance.16 Preclinical evidence exists for the complementary roles of CTLA-4 and PD-1 in regulating adaptive immunity, and this provides rationale for combining drugs targeting these pathways.44-46 Paradoxically and originally believed to be immunosuppressive, new data allow us to recognize that cytotoxic agents can antagonize immunosuppression in the tumor microenvironment, thus promoting immunity based on the concept that tumor cells die in multiple ways and that some forms of apoptosis may lead to an enhanced immune response.8,15 For example, nivolumab was combined with ipilimumab in a phase 1 trial of patients with advanced melanoma.46 The combination had a manageable safety profile and produced clinical activity in the majority of patients, with rapid and deep tumor regression seen in a large proportion of patients. Based on the results of this study, a phase 3 study is being undertaken to evaluate whether this combination is better than nivolumab alone in melanoma (NCT01844505). Several other early-phase studies are underway to explore combinations of various anti–PD-1/PD-L1 drugs with other therapies across a variety of tumor types (see Tables 1 and 2), possibly paving the way for future combination studies.

PD-L1 as a Predictive Biomarker Tumor PD-L1 expression has been shown to correlate with poor prognosis in many cancers.47 Available early data allude to PD-L1 expression in tumors as a possible predictive biomarker of response to anti–PD-1/ PD-L1 drugs; however, these data must be confirmed, and the role of tumor expression of PD-L1 must be further elucidated.

Conclusions The discovery of agents targeted at the anti–programmed death 1 and anti–programmed death ligand 1 pathway, as well as their remarkable activity in several cancers, has launched an era of effective immunotherapeutic drugs that will change the landscape of cancer treatment. These agents also produce responses in nonimmunogenic cancers such as non–small-cell lung and colon cancers, broadening their scope beyond classic immunogenic tumors like melanoma and renal cell cancer.20 The activity of these agents has been suggested in early-phase studies of melanoma, renal cell, and non–small-cell lung cancers, and the results from completed and ongoing phase 3 studies are eagerly awaited. In addition, these agents are being explored alone or in combination across other difficult-to-treat tumor types. In summary, the programmed death 1/programmed death ligand 1 pathway inhibitors have 236 Cancer Control

made an addition to the armamentarium of currently available immunotherapeutic drugs and carry great potential for treating immunogenic as well as nonimmunogenic cancers. References 1. Disis ML. Immune regulation of cancer. J Clin Oncol. 2010;28(29): 4531-4538. 2. Vesely MD, Kershaw MH, Schreiber RD, et al. Natural innate and adaptive immunity to cancer. Annu Rev Immunol. 2011;29:235-271. 3. Dunn GP, Bruce AT, Ikeda H, et al. Cancer immunoediting: from immunosurveillance to tumor escape. Nat Immunol. 2002;3(11):991-998. 4. Drake CG, Jaffee E, Pardoll DM. Mechanisms of immune evasion by tumors. Adv Immunol. 2006;90:51-81. 5. Dong H, Strome SE, Salomao DR, et al. Tumor-associated B7-H1 promotes T-cell apoptosis: a potential mechanism of immune evasion. Nat Med. 2002;8(8):793-800. 6. Francisco LM, Sage PT, Sharpe AH. The PD-1 pathway in tolerance and autoimmunity. Immunol Rev. 2010;236:219-242. 7. Thompson RH, Dong H, Lohse CM, et al. PD-1 is expressed by tumorinfiltrating immune cells and is associated with poor outcome for patients with renal cell carcinoma. Clin Cancer Res. 2007;13(6):1757-1761. 8. Mellman I, Coukos G, Dranoff G. Cancer immunotherapy comes of age. Nature. 2011:480(7378);480-489. 9. Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer. 2012;12(4):252-264. 10. Keir ME, Butte MJ, Freeman GJ, et al. PD-1 and its ligands in tolerance and immunity. Annu Rev Immunol. 2008;26:677-704. 11. 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(3):261-268. 12. Gajewski TF, Louahed J, Brichard VG, et al. Gene signature in melanoma associated with clinical activity: a potential clue to unlock cancer immunotherapy. Cancer J. 2010;16(4):399-403. 13. Ghiotto M, Gauthier L, Serriari N, et al. PD-L1 and PD-L2 differ in their molecular mechanisms of interaction with PD-1. Int Immunol. 2010;22(8): 651-660. 14. Hiraoka N. Tumor-infiltrating lymphocytes and hepatocellular carcinoma: molecular biology. Int J Clin Oncol. 2010;15(6):544-551. 15. Ramsay AG. Immune checkpoint blockade immunotherapy to activate anti-tumour T-cell immunity. Br J Haematol. 2013;162(3): 313-325. 16. Pardoll DM. The blockade of immune checkpoints in cancer immunotherapy. Nat Rev Cancer. 2012;12(4): 252-264. 17. Barber DL, Wherry EJ, Masopust D, et al. Restoring function in exhausted CD8 T cells during chronic viral infection. Nature. 2006;439(7077):682-687. 18. Tang PA, Heng DY. Programmed death 1 pathway inhibition in metastatic renal cell cancer and prostate cancer. Curr Oncol Rep. 2013;15(2): 98-104. 19. Zitvogel L, Kroemer G. Targeting PD-1/PD-L1 interactions for cancer immunotherapy. Oncoimmunology. 2012;1(8):1223-1225. 20. Brahmer JR, Tykodi SS, Chow LQ, et al. Safety and activity of anti-PD-L1 antibody in patients with advanced cancer. N Engl J Med. 2012;366(26): 2455-2465. 21. Hamid O, Robert C, Daud A, et al. Safety and tumor responses with lambrolizumab (anti-PD-1) in melanoma. N Engl J Med. 2013;369(2):134-144. 22. Brahmer JR, Drake CG, Wollner I, et al. Phase I study of single-agent anti-programmed death-1 (MDX-1106) in refractory solid tumors: safety, clinical activity, pharmacodynamics, and immunologic correlates. J Clin Oncol. 2010;28(19):3167-3175. 23. 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(26): 2443-2454. 24. Topalian SL, Sznol M, Brahmer JR, et al. Nivolumab (anti-PD-1; BMS936558; ONO-4538) in patients with advanced solid tumors: survival and long-term safety in a phase I trial. J Clin Oncol. 2013;31(suppl):3002. 25. Thompson RH, Gillett MD, Cheville JC, et al. Costimulatory B7-H1 in renal cell carcinoma patients: Indicator of tumor aggressiveness and potential therapeutic target. Proc Natl Acad Sci U S A. 2004;101(49):17174-17179. 26. Thompson RH, Kuntz SM, Leibovich BC, et al. Tumor B7-H1 is associated with poor prognosis in renal cell carcinoma patients with long-term follow-up. Cancer Res. 2006;66(7):3381-3385. 27. Hino R, Kabashima K, Kato Y, et al. Tumor cell expression of programmed cell death-1 ligand 1 is a prognostic factor for malignant melanoma. Cancer. 2010;116(7):1757-1766. 28. Dasanu CA, Sethi N, Ahmed N. Immune alterations and emerging immunotherapeutic approaches in lung cancer. Expert Opin Biol Ther. 2012;12(7):923-937. 29. Sato Y, Mukai K, Watanabe S, et al. Lymphocyte subsets in pulmonary venous and arterial blood of lung cancer patients. Jpn J Clin Oncol. 1989;19(3):229-236. 30. Wesselius LJ, Wheaton DL, Manahan-Wahl LJ, et al. Lymphocyte subJuly 2014, Vol. 21, No. 3

sets in lung cancer. Chest. 1987;91(5):725-729. 31. Woo EY, Yeh H, Chu CS, et al. Cutting edge: Regulatory T cells from lung cancer patients directly inhibit autologous T cell proliferation. J Immunol. 2002;168(9):4272-4276. 32. Brahmer JR. Harnessing the immune system for the treatment of nonsmall-cell lung cancer. J Clin Oncol. 2013;31(8):1021-1028. 33. Ghebeh H, Barhoush E, Tulbah A, et al. FOXP3+ Tregs and B7-H1+/ PD-1+ T lymphocytes co-infiltrate the tumor tissues of high-risk breast cancer patients: implication for immunotherapy. BMC Cancer. 2008;8:57. 34. Ghebeh, Barhoush E, Tulbah A, et al. FOXP3+ Tregs and B7-H1+/ PD-1+ T lymphocytes co-infiltrate the tumor tissues of high-risk breast cancer patients: Implication for immunotherapy. BMC Cancer. 2008;8:57. 35. Lyford-Pike S, Peng S, Young GD, et al. Evidence for a role of the PD1:PD-L1 pathway in immune resistance of HPV-associated head and neck squamous cell carcinoma. Cancer Res. 2013;73(6):1733-1741. 36. Badoual C, Hans S, Merillon N, et al. PD-1-expressing tumor-infiltrating T cells are a favorable prognostic biomarker in HPV-associated head and neck cancer. Cancer Res. 2013;73(1):128-138. 37. Ukpo OC, Thorstad WL, Lewis JS Jr. B7-H1 expression model for immune evasion in human papillomavirus-related oropharyngeal squamous cell carcinoma. Head Neck Pathol. 2013;7(2):113-121. 38. Berger R, Rotem-Yehudar R, Slama G, et al. Phase I safety and pharmacokinetic study of CT-011, a humanized antibody interacting with PD1, in patients with advanced hematologic malignancies. Clin Cancer Res. 2008;14(10):3044-3051. 39. Herbst RS, Gordon MS, Fine GF, et al. A study of MPDL3280A, an engineered PD-L1 antibody in patients with locally advanced or metastatic tumors. J Clin Oncol. 2013;31(suppl):3000. 40. Drake CG, McDermott DF, Sznol M, et al. Survival, safety, and response duration results of nivolumab (anti-PD-1; BMS-936558; ONO-4538) in a phase I trial in patients with previously treated metastatic renal cell carcinoma (mRCC): long-term patient follow-up. J Clin Oncol. 2013;31(suppl):4514. 41. Patnaik A, Kang SP, Tolcher AW, et al. Phase I study of MK-3475 (anti-PD-1 monoclonal antibody) in patients with advanced solid tumors. J Clin Oncol. 2012;30(suppl):2512. 42. Hodi FS, O’Day SJ, McDermott DF, et al. Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med. 2010;363(8): 711-723. 43. Quezada SA, Peggs KS. Exploiting CTLA-4, PD-1 and PD-L1 to reactivate the host immune response against cancer. Br J Cancer. 2013;108(8): 1560-1565. 44. Perez-Gracia JL, Berraondo P, Martinez-Forero I, et al. Clinical development of combination strategies in immunotherapy: are we ready for more than one investigational product in an early clinical trial? Immunotherapy. 2009;1(5):845-853. 45. Curran MA, Montalvo W, Yagita H, et al. PD-1 and CTLA-4 combination blockade expands infiltrating T cells and reduces regulatory T and myeloid cells within B16 melanoma tumors. Proc Natl Acad Sci U S A. 2010;107(9): 4275-4280. 46. Wolchok JD, Kluger H, Callahan MK, et al. Nivolumab plus ipilimumab in advanced melanoma. N Engl J Med. 2013;369(2):122-133. 47. Thompson RH, Dong H, Lohse CM, et al. PD-1 is expressed by tumorinfiltrating immune cells and is associated with poor outcome for patients with renal cell carcinoma. Clin Cancer Res. 2007;13(6):1757-1761.

July 2014, Vol. 21, No. 3

Cancer Control 237

Ten Best Readings Relating to Clinical and Translational Research Sleijfer S, Bogaerts J, Siu LL. Designing transformative clinical trials in the cancer genome era. J Clin Oncol. 2013;31(15):1834-1841. The use of molecular profiling in clinical practice is increasing. Using genomic sequencing, cancers can be classified into molecular subsets. However, designing clinical trials for these molecular subsets is challenging secondary to the few numbers of available patients. Newer clinical trial designs are needed to identify novel anticancer therapy based on molecularly defined subsets. Ji Y, Wang SJ. Modified toxicity probability interval design: a safer and more reliable method than the 3 + 3 design for practical phase I trials. J Clin Oncol. 2013;31(14):1785-1791. The 3 + 3 design is the most commonly used dose escalation scheme in phase 1 clinical trials. In simulation studies, adaptive designs, including the modified toxicity probability interval (mTPI), have been shown to be more efficient in identifying the maximum tolerated dose. Software is provided free of cost for utilizing the mTPI design. Jardim DL, Hess KR, Lorusso P, et al. Predictive value of phase I trials for safety in later trials and final approved dose: analysis of 61 approved cancer drugs. Clin Cancer Res. 2014;20(2):281-288. Dosing based on phase 1 trials was associated with a low toxicity-related death rate in later trials. The ability to predict relevant toxicities correlates with the number of patients in the initial phase 1 trial. The final dose approved was within 20% of the recommended phase 2 dose in 73% of assessed trials. Manji A, Brana I, Amir E, et al. Evolution of clinical trial design in early drug development: systematic review of expansion cohort use in single-agent phase I cancer trials. J Clin Oncol. 2013;31(33):4260-4267. The use of expansion cohorts has increased with time. Safety and efficacy are common objectives, but 26% of these cohorts fail to report explicit aims. Expansion cohorts may provide useful supplementary data for phase 1 trials, particularly with regard to toxicity and defining the recommended dose for phase 2 studies. Townsley CA, Selby R, Siu LL. Systematic review of barriers to the recruitment of older patients with cancer onto clinical trials. J Clin Oncol. 2005;23(13): 3112-3124. Physician education to dispel unfounded perceptions, improved access to available clinical trials, and the provision of personnel and resources to accommodate the unique requirements of an older population are possible solutions to remove the barriers of ageism. 238 Cancer Control

Goede V, Fischer K, Busch R, et al. Obinutuzumab plus chlorambucil in patients with CLL and coexisting conditions. N Engl J Med. 2014;370(12):1101-1110. Combining an anti-CD20 antibody with chemotherapy improved outcomes in patients with chronic lymphocytic leukemia and coexisting conditions. In this patient population, obinutuzumab was superior to rituximab when each drug was combined with chlorambucil. Flaherty KT, Infante JR, Daud A, et al. Combined BRAF and MEK inhibition in melanoma with BRAF V600 mutations. N Engl J Med. 2012;367(18): 1694-1703. Dabrafenib and trametinib were safely combined at full monotherapy doses. The rate of pyrexia was increased with combination therapy, whereas the rate of proliferative skin lesions was not significantly reduced. Progression-free survival rates were significantly improved. 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(1):61-68. The continued MAPK signaling–based resistance identified in patients with BRAF-mutant melanoma suggests that an alternative dosing of current agents, more potent RAF/MEK inhibitors, and/or inhibition of the downstream kinase ERK may be needed for durable control of BRAF-mutant melanoma. Sosman JA, Kim KB, Schuchter L, et al. Survival in BRAF V600-mutant advanced melanoma treated with vemurafenib. N Engl J Med. 2012;366(8):707-714. Vemurafenib induces clinical responses in more than one-half of patients with previously treated BRAF V600-mutant metastatic melanoma. In this study with lengthy follow-up, the median overall survival rate was approximately 16 months. Okazaki T, Honjo T. The PD-1-PD-L pathway in immunological tolerance. Trends Immunol. 2006;27 (4):195-201. The authors review recent studies on the role of programmed cell death 1 (PD-1) in immunological tolerance and discuss the possible clinical applications of manipulating PD-1.

July 2014, Vol. 21, No. 3

Translational Medicine Simplified

AKT Goes Cycling Kiran N. Mahajan, PhD, and Nupam P. Mahajan, PhD

Protein kinase B (AKT) is an Table. — AKT Phosphorylations and Corresponding Kinases important signaling molecule Study Kinase Site of Phosphorylation Domain in multiple cell types, relaying Guo13 IKBKE Serine 137 Intradomain region between growth, proliferation, and surpleckstrin-homology and kinase 1-3 14 Xie vival cues. AKT interacts with regulatory networks to feed exMahajan4,15 TNK2 (ACK1) Tyrosine 176 Kinase tracellular signals into transcrip11 Joung TANK-binding kinase 1 Threonine 195 Kinase tional programs within cells that eventually dictate the cell state. Stephens8 PDK1 Threonine 308 Kinase Its activation impacts a number Chen16 PTK6, SRC Tyrosine 315 Kinase of physiological processes such 17 Zheng as glucose metabolism, protein synthesis, and regulated apopChen16 PTK6, SRC Tyrosine 326 Kinase tosis. In normal cells, the actiZheng17 vation of AKT is thought to be Sarbassov10 mTORC2 Serine 473 C-terminal region critically dependent on phosphorylations, which are tightly Liu12 Cdk2/cyclin A Serine 477 C-terminal region regulated by disparate upstream Threonine 479 kinases, responding to different stimuli.4,5 Two of the best studied AKT phosphorylation events occur at threonine Such efforts have revealed that AKT hyperactivation (Thr) 308 and serine (Ser) 473 residues, mediated by in cancer cells is not mediated by the RTK/PI3K/PTEN the receptor tyrosine kinase/phosphatidylinositol 3-kisignaling alone, but rather a diverse group of kinases nase/phosphatase and tensin homolog (RTK/PI3K/ might target and activate AKT to promote uncontrolled PTEN) signaling nexus.6-10 The PI3K/PTEN pathway proliferation and resistance to chemotherapeutic is one of the most deregulated pathways in human agents (Table).4,5,8,10-17 For example, some oncogenic cancers. However, cancer cells often develop resiskinases, such as ACK1 (also known as TNK2) and tance to PI3K inhibitors or do not utilize the PI3K/ TANK-binding kinase 1 bypass the PI3K dependence PTEN pathway for AKT activation. Therefore, numerto activate AKT and promote tumor growth and resisous laboratories have invested considerable efforts into tance to PI3K inhibitors.4,11 Therefore, keeping in mind understanding the mechanisms of AKT activation and the multiple regulatory networks that feed into AKT its pathological role in driving human malignancies. signaling and the complexity of signaling, pinpointing the mechanisms by which AKT signaling is activated is crucial to specifically target this pathway to achieve From the Drug Discovery Program at the H. Lee Moffitt Cancer maximum clinical benefit. Center & Research Institute, Tampa, Florida, and the Department A report by Liu et al12 is another addition in the of Oncology at the University of South Florida, Tampa, Florida. already long list of AKT phosphorylations, highlightSubmitted April 29, 2014; accepted May 5, 2014. ing a previously unknown mode of AKT activation. Address correspondence to Kiran N. Mahajan, PhD, or Nupam P. Mahajan, PhD, Drug Discovery Program, Moffitt Cancer Center, The researchers identified a novel phosphorylation 12902 Magnolia Drive, SRB-2, Tampa, FL 33612. E-mail: event at the AKT carboxy terminus tail residues, [email protected] or [email protected] Ser-477/Thr-479, which occurs in a cell cycle–depenNo significant relationships exist between the authors and the comdent manner. The key to this study is the generation panies/organizations whose products or services may be referenced in this article. of high-specificity antibodies that cross-react with the Dr K.N. Mahajan is a recipient of Grant #W81XWH-12-1-0248 novel phosphorylated-AKT Ser-477/Thr-479 residues from the US Department of Defense. Dr N.P. Mahajan is supported but not with phosphorylation-AKT Ser 473, which is a by Grant #1R01CA135328 from the National Institutes of Health/ fairly robust phosphorylation.12 Using a synchronized National Cancer Institute and Grants #W81XWH-14-1-0002 and #W81XWH-14-1-0003 from the Department of Defense. population of cycling cells, the researchers uncovered July 2014, Vol. 21, No. 3

Cancer Control 239

that phosphorylated-AKT Ser-477/Thr-479 oscillates during the cell cycle. It mirrors the periodic cyclin A2 expression and is catalyzed by the Cdk2/cyclin A2 complex, whose activity is regulated during the cell cycle. Liu et al12 also identified 4 evolutionarily conserved RXL cyclin A-binding motifs in all of the 3 human AKT isoforms as well as in mouse and rat AKT. Mechanistically, the Cdk2/cyclin A2–mediated AKT phosphorylation at Ser-477/Thr-479 enhanced AKT activity by promoting the activating Ser-473 phosphorylation. Accordingly, a phosphomimetic AKT1-DE mutant (AKT1-Ser-477D/Thr-479E), displayed increased Ser-473 phosphorylation and had an enhanced ability to promote tumorigenesis in a mouse xenograft tumor model compared with the wild-type enzyme, while a double-alanine mutant (AKT1-Ser-477AD/Thr-479A) showed loss of Ser-473 phosphorylation, loss of substrate phosphorylation, and decreased tumor development. Further, the researchers provided evidence of Cdk2/cyclin A2 complex as being the prime regulator of phosphorylation-AKT Ser-477/Thr-479 by ectopically expressing AKT1-DE in mouse embryo fibroblasts derived from quadruple knockout mice (cyclin E1–/–/cyclin E2–/–/ cyclin A1–/–/cyclin A2f/f ) after transfection with Cre. AKT1-DE partly rescued the cell-cycle defects observed in these mouse embryo fibroblasts. At the molecular level, the Ser-477/Thr-479 phosphorylated AKT displayed increased the association with stress-activated protein kinase-interacting 1 and mammalian target of rapamycin (mTOR) complexes, but it did not alter the association with phosphatases.12 Further, Liu et al12 suggested that Ser-477/ Thr-479 phosphorylation could lock AKT in an active conformation, a scenario that may also be observed if the carboxy-terminus of AKT is deleted. It is worth noting that Ser-477 phosphorylation can still occur in a cell cycle–independent manner, albeit not by Cdk2, but by the mTORC2/Rictor complex following the stimulation of insulin or during DNA damage by the related DNA damage–dependent protein kinase. Further studies are required to understand the differential role of distinct kinases in AKT Ser-477/Thr-479 phosphorylation and its compartmentalization in cells. What is the advantage of the cell cycle–specific regulation of AKT activity? Is it an alternate mechanism to regulate the activation of AKT in cells to suppress growth-promoting signals? One can envisage a regulatory feedback loop wherein Cdk2 regulates cellcycle progression by acting on other substrates but yet it also controls the AKT signaling network interacting with it and regulating AKT activity in a temporal manner; in turn, activated AKT would respond by transmitting growth signals that trigger re-entry into the cell cycle. Indeed, the deletion of cyclin A2 in mouse embryonic stem cells impaired the phosphorylation-AKT 240 Cancer Control

Ser-477/Thr-479 and caused elevated apoptosis.12 By the same token, it is possible to envisage the outcome when this regulation is lost in cancer cells. Liu et al12 also assessed this possibility and observed hyperphosphorylation of AKT at Ser-477 in certain cancers. For example, a positive correlation was observed between Ser-477/Thr-479 phosphorylated AKT and AKT Ser473 phosphorylation in patients with breast cancer. However, by contrast to Ser-473 phosphorylation, high levels of Ser-477/Thr-479 phosphorylation occurred at a relatively higher rate in the earlier developmental stages of breast cancer.12 Whether this is indicative of a subset of rapidly cycling breast cells predisposed to overcome cell-cycle checkpoints, develop genomic instability, and become cancerous remains to be seen. If so, then, as Liu et al12 have suggested, this phosphorylation could be utilized as a biomarker to detect early-stage breast cancer. With the identification of the novel mode of AKT phosphorylation mediated by the Cdk2/cyclin A2 complex, additional avenues to tackle tumor development became apparent. Cdk2, a crucial regulator of the cell division cycle, is overactive in many types of cancers and numerous inhibitors are now available to block its activity in cancer cells. One such inhibitor is seliciclib (CYC202, roscovitine), which inhibits CDK2, CDK7, and CDK9. It has been evaluated in several phase 1 and 2 studies and has shown early signs of anticancer activity.18 However, Cdk2 inhibitors may be not be effective in blocking cell cycle–regulated AKT activation, because cancer cells may quickly adapt and employ mTORC2 pathways to compensate for the loss of Cdk2/ cyclin A activity. Therefore, direct AKT inhibitors may be the key to target AKT that is being activated by multiple kinases. One such drug that appears to hold promise is an oral allosteric inhibitor of AKT, MK-2206, that is undergoing phase 2 clinical trials (NCT01283035).19 MK-2206 binds to the pleckstrin-homology domain of AKT and inhibits its activity in a non–adenosine triphosphate competitive manner by causing a change in the conformation of AKT, thus preventing its localization to the plasma membrane. Although phosphorylation-AKT Ser-477/Thr-479 has been carefully examined for its cell cycle–dependent modulation, how Cdk2/cyclin A accomplishes this is not clear. Most Cdk2 targets reside in the nucleus, so whether this phosphorylation predominantly occurs on nuclear AKT is not known. It is worth noting that the phosphorylation was reduced in insulin-stimulated cells treated with the PI3K inhibitor, LY29002, indicating that the plasma membrane localization of AKT or the PI3K signaling has a role.12 However, the loss of phosphorylation-AKT Ser-477/ Thr-479 by LY29002 could be an indirect effect caused due to cell-cycle arrest. Further, the relative amounts July 2014, Vol. 21, No. 3

of phosphorylation-AKT Ser-477 as compared with Ser-473 must be determined, an answer that will be critical in order to evaluate the contribution of this novel phosphorylation in the overall AKT activation. In conclusion, normal cells as well as cancer cells appear to utilize the enzymatic activity of a variety of kinases to maintain optimal AKT activity. Cdk2/ cyclin A2–mediated phosphorylation-AKT adds a new twist in this ongoing saga. However, unlike other AKT-interacting kinases (see Table4,8,10-17), Cdk2 is predominantly functional in the nucleus and is involved in the regulation of the G1 to S phase progression of the cell cycle. Whether this will provide a new mode of tackling AKT activation as a therapeutic strategy remains to be seen. References 1. Manning BD, Cantley LC. AKT/PKB signaling: navigating downstream. Cell. 2007;129(7):1261-1274. 2. Bellacosa A, Kumar CC, Di Cristofano A, et al. Activation of AKT kinases in cancer: implications for therapeutic targeting. Adv Cancer Res. 2005;94:29-86. 3. Vivanco I, Sawyers CL. The phosphatidylinositol 3-kinase AKT pathway in human cancer. Nat Rev Cancer. 2002;2(7):489-501. 4. Mahajan K, Coppola D, Challa S, et al. Ack1 mediated AKT/PKB tyrosine 176 phosphorylation regulates its activation. PLoS One. 2010;5(3):e9646. 5. Mahajan K, Mahajan NP. PI3K-independent AKT activation in cancers: a treasure trove for novel therapeutics. J Cell Physiol. 2012;227(9):3178-3184. 6. Franke TF, Yang SI, Chan TO, et al. The protein kinase encoded by the Akt proto-oncogene is a target of the PDGF-activated phosphatidylinositol 3-kinase. Cell. 1995;81(5):727-736. 7. Burgering BM, Coffer PJ. Protein kinase B (c-Akt) in phosphatidylinositol-3-OH kinase signal transduction. Nature. 1995;376(6541):599-602. 8. Stephens L, Anderson K, Stokoe D, et al. Protein kinase B kinases that mediate phosphatidylinositol 3,4,5-trisphosphate-dependent activation of protein kinase B. Science. 1998;279(5351):710-714. 9. Stokoe D, Stephens LR, Copeland T, et al. Dual role of phosphatidylinositol-3,4,5-trisphosphate in the activation of protein kinase B. Science. 1997;277(5325):567-570. 10. Sarbassov DD, Guertin DA, Ali SM, et al. Phosphorylation and regulation of Akt/PKB by the rictor-mTOR complex. Science. 2005;307(5712):1098-1101. 11. Joung SM, Park ZY, Rani S, et al. Akt contributes to activation of the TRIF-dependent signaling pathways of TLRs by interacting with TANK-binding kinase 1. J Immunol. 2011;186(1):499-507. 12. Liu P, Begley M, Michowski W, et al. Cell-cycle-regulated activation of Akt kinase by phosphorylation at its carboxyl terminus. Nature. 2014;508(7497): 541-545. 13. Guo JP, Coppola D, Cheng JQ. IKBKE protein activates Akt independent of phosphatidylinositol 3-kinase/PDK1/mTORC2 and the pleckstrin homology domain to sustain malignant transformation. J Biol Chem. 2011;286(43): 37389-37398. 14. Xie X, Zhang D, Zhao B, et al. IkappaB kinase epsilon and TANK-binding kinase 1 activate AKT by direct phosphorylation. Proc Natl Acad Sci USA. 2011; 108(16):6474-6479. 15. Mahajan K, Coppola D, Chen YA, et al. Ack1 tyrosine kinase activation correlates with pancreatic cancer progression. Am J Pathol. 2012;180(4): 1386-1393. 16. Chen R, Kim O, Yang J, et al. Regulation of Akt/PKB activation by tyrosine phosphorylation. J Biol Chem. 2001;276(34):31858-31862. 17. Zheng Y, Peng M, Wang Z, et al. Protein tyrosine kinase 6 directly phosphorylates AKT and promotes AKT activation in response to epidermal growth factor. Mol Cell Biol. 2010;30(17):4280-4292. 18. Gallorini M, Cataldi A, di Giacomo V. Cyclin-dependent kinase modulators and cancer therapy. BioDrugs. 2012;26(6):377-391. 19. Hirai H, Sootome H, Nakatsuru Y, et al. MK-2206, an allosteric Akt inhibitor, enhances antitumor efficacy by standard chemotherapeutic agents or molecular targeted drugs in vitro and in vivo. Mol Cancer Ther. 2010;9(7): 1956-1967.

July 2014, Vol. 21, No. 3

Cancer Control 241

Special Report

Novel Pancreatic Cancer Vaccines Could Unleash the Army Within Gregory M. Springett, MD, PhD

Background: Despite recent progress with novel chemotherapy regimens, pancreatic ductal adenocarcinoma remains the fourth leading cause of cancer death in the United States. Innovative approaches to treatment of this disease are needed to accelerate progress. Methods: A review was conducted of the results of 2 pancreatic cancer vaccine programs with results that have shown promise in early-phase clinical trials. Results: In a phase 2 trial, a cell-based allogeneic pancreatic cancer vaccine exploiting the hyperacute rejection response targeted against alpha-1,3 galactosyl epitopes (algenpantucel-L) has shown improvement in disease-free and overall survival rates in the adjuvant setting compared with a historical control. This vaccine has advanced to ongoing phase 3 trials. Compared with GVAX alone, a second whole-cell vaccine employing GM-CSF–expressing pancreatic cancer cells (GVAX) to enhance the antigen presentation in a priming phase followed by a Listeria-based vaccine targeting mesothelin in a boost phase improved survival rates. This vaccine platform is undergoing additional phase 2 testing. Conclusions: Allogenic whole-cell pancreatic adenocarcinoma vaccines show promise in early-phase trials and have the potential to improve survival rates by unleashing antitumor immunity.

Introduction Pancreatic ductal adenocarcinoma (PDA) remains near the top of the list of leading causes of death from cancer. With 46,420 new cases and 39,590 deaths, the rate of fatality remains above 85%.1 The apparent intrinsic resistance of PDA is highlighted by the long list of conventional and targeted chemotherapy agents that have failed to produce clinically significant improvements in survival rates among patients with this malignancy. In the 15 years between the initial approval of gemcitabine for PDA and the successful phase 3 trials of the multiagent regimens 5-fluorouracil/folinic acid/oxaliplatin/irinotecan and gemcitabine/nab-paclitaxel, overall survival (OS) rates for patients with advanced disease have improved from 6 months to 11 months.2,3 Five-year survival rates for patients with early-stage PDA who undergo resection From the Gastrointestinal Tumor Program, Experimental Therapeutics Program, and Drug Discovery Program at the H. Lee Moffitt Cancer Center & Research Institute, Tampa, Florida. Submitted April 23, 2014; accepted April 28, 2014. Address correspondence to Gregory M. Springett, MD, PhD, Gastrointestinal Tumor Program, Moffitt Cancer Center, 12902 Magnolia Drive, WCB-GI PROG, Tampa, FL 33612. E-mail: Gregory.Springett@ Moffitt.org No significant relationship exists between the author and the companies/organizations whose products or services may be referenced in this article. The author has disclosed that this article discusses unlabeled/unapproved uses of the drugs algenpantucel-L, GVAX, and CRS-207 for the treatment of pancreatic cancer. 242 Cancer Control

followed by adjuvant therapy have improved from approximately 10% to 20%.4 Given this slow rate of progress, it is understandable that a desire exists for investigating a different approach to this disease. Encouraging results with 2 new pancreatic cancer vaccine therapies give hope that immunotherapy can achieve a leap in progress.5,6

Amenable to Immunotherapy? Several clinical observations suggest that the enhancement of antitumor immunity in PDA may provide a clinical benefit. The accumulation of CD8+ T cells in human PDA correlates with improved survival rates.7-9 Numerous tumor-associated antigens (TAAs) have been identified in PDA. Among the most promising are mesothelin, mucin 1 (MUC1), and Kirsten rat sarcoma (Kras). Antibodies to TAAs are found in the serum of patients with PDA, and the presence of these antibodies correlates with survival.10 In addition, the presence of PDA is associated with immunosuppression characterized by elevated levels of CD4+ CD25+ Foxp3+ regulatory T cells and CD11b+ CD14– CD33+ myeloid-derived suppressor cells, which downregulate antitumor immune responses. In animal models of pancreatic cancer, these cells may represent 50% of the leukocytes infiltrating the tumor.11 The elevation of myeloid-derived suppressor cells is an independent prognostic factor in PDA.12 Finally, several vaccine strategies have shown efficacy in preclinical animal models.13 July 2014, Vol. 21, No. 3

What Is Required for an Effective Antitumor Immune Response? The sequence of events leading to tumor rejection has been organized into a conceptual framework called the cancer-immunity cycle.14 The cytotoxic T-lymphocyte (CTL)–mediated elimination of tumor cells is an antigen-dependent process. Ideal TAAs are present on tumor cells alone and absent — or at least expressed at reduced levels — in normal cells. Whole exome sequencing of patient-derived tumor cells can identify a complete set of mutated genes and their respectively abnormal protein products.15 As this technology is applied to additional tumor types, the research has become clear that no 2 patients with similar cancer diagnoses have an identical set of mutations.15 The antigen specificity of tumor-infiltrating lymphocytes capable of mediating tumor rejection has been compared with the unique library of mutations in a given patient’s tumor for some cancers. Researchers are finding that the TAAs to which CTLs respond are peptides derived from these mutated proteins. In particular, a subset of mutant peptides, which are capable of binding to the major histocompatibility complex (MHC) with high affinity for antigen presentation, are relevant mutations for antitumor immunity. Genome projects that sequence pancreatic cancer have identified an average of 25 to 45 mutations per patient, ranging from 1 to 116.15,16 In addition, the expression of nonmutated antigens, such as mesothelin, telomerase, survivin, MUC1, human epidermal growth factor receptor 2, and carcinoembryonic antigen, are upregulated. Most of these mutant and overexpressed proteins are intracellular and released upon tumor cell necrosis. These released proteins undergo capture, phagocytosis, or receptor-mediated endocytosis by antigen-presenting cells (APCs) that can present on MHC classes I and II. Peptides that are 9 to 10 amino acids in length are generated from these proteins and then bind to the MHC and are presented to T cells. Dendritic cells are the most versatile APCs and are potent activators of T cells. Tissue dendritic cells migrate to lymph nodes following antigen uptake and processing where they interact with T cells to prime an immune response. The activation of tumor-directed T cells requires a bipartite signal. The T-cell receptor binds to antigen associated with the MHC presented by APCs. CD28 on T cells also binds B7-1 on APCs, thus generating a costimulatory signal. These 2 signaling events lead to the activation of the T cell. The presence of T-cell receptor/antigen–MHC binding in the absence of CD28/ B7-1 binding results in T-cell anergy and tolerance of the tumor. In the presence of the bipartite signal, activated CTLs then migrate to tumor sites, infiltrate the tumor, and kill the tumor cells. The lysis of tumor cells by CTLs then releases more TAA that amplifies the signal. The cycle repeats, thus generating a July 2014, Vol. 21, No. 3

positive feedback loop. In principle, the process can continue until every tumor cell is eliminated, resulting in complete regression of the tumor.

Algenpantucel-L Immunotherapy The initial priming step of the cancer-immunity cycle requires tumor cell lysis to release a mixture of TAAs to APCs in an environment of immunostimulatory cytokines. Even when this occurs, APCs may fail to phagocytose and process released TAAs. Algenpantucel-L is designed to harness the activity of hyperacute graft rejection to enhance these initial steps, thus leading to better immune priming (Fig 1). Hyperacute graft rejection after xenotransplantation results in lysis of foreign cells within minutes. The antigen that triggers this response is galactosylalpha-1,3-galactose (alpha gal) on the cell surface glycoproteins of mammalian cells, with the exception of humans and Old World primates. The absence of this antigen on human cells is due to the inactivation of the GGTA1 gene for alpha-1,3-galactosyltransferase about 20 million years ago during the evolution of primates.17 The human pseudogene contains a 2 base-pair frameshift mutation. Pre-existing high titer antibodies to alpha gal represent 1% to 2% of all circulating antibodies.17 Binding of these antibodies to alpha gal on nonhuman transplanted cells activates complement-mediated lysis and antibody-dependent cell-mediated cytotoxicity.17 Anti–alpha gal antibody bound to autologous tumor cells modified by transfection with the GGTA1 gene to express alpha gal targets those tumor cells for opsonization by APCs via the antibody Fc-gamma receptor. This allows the APCs to phagocytose the entire tumor cell with their complete library of TAAs. These APCs then migrate to regional lymph nodes and process and present TAA peptides to CD8+ cytotoxic T cells in association with MHC class I as well as CD4+ T cells in association with MHC class II. The algenpantucel-L vaccine consists of 2 human pancreatic cancer cell lines (HAPa-1 and HAPa-2) modified to express alpha gal by retroviral transduction of the murine GGTA1 gene.5 These 2 cell lines were selected for their representation of known TAAs for pancreatic cancer (eg, mesothelin, carcinoembryonic antigen). The engineered cells are irradiated and administered as intradermal injections. In a phase 1 study, algenpantucel-L was administered to 7 participants. No dose-limiting toxicities were seen.5 A phase 2 study was conducted in 70 patients in the adjuvant setting in which the primary end point was 12-month disease-free survival (DFS).5 Beginning 6 weeks after R0 or R1 resection of PDA, vaccination with either 100 million or 300 million cells commenced. Chemotherapy with gemcitabine and chemoradiation with fluorouracil as a radiosensitizer was given according to Cancer Control 243

the RTOG-9704 standard. Vaccine injections were given every 2 weeks during chemotherapy and chemoradiation for up to 14 vaccinations. Restaging computed tomography scans were obtained at the end of treatment and then every 3 months for 1 year, then every 6 months for 2 years, and then yearly thereafter. The 12-month DFS rate was 62% for the entire cohort. There appeared to be a dose effect because the 12-month DFS rate for the cohort receiving 300 million cells was 81% compared with 51% for those receiving 100 million cells. The 12-month OS rate for the entire cohort was 86% (96% at a dose of 300 million cells and 79% at a dose of 100 million cells). This compared favorably with the 12-month OS rate of the RTOG-9704 historical control (69%), despite a larger proportion of patients with node-positive disease in the algenpantucel-L study (81%) compared with RTOG-9704 (68%).5 At 3 years, the DFS and OS rates were 26% and 39%, respectively. The most frequent adverse event related to the vaccine was injection site reaction (grades 1/2) and was seen in 51% of participants. Grade 3 events related to the vaccine were lymphopenia (6%), injection site

Lentiviral GGTA1 vector

reaction (3%), and leukopenia (3%). No grade 4 events were seen. With regard to immune parameters, 90% of patients showed increases in anti–alpha gal antibodies with elevated titers for more than 200 days. Antimesothelin and anti-carcinoembryonic antigen antibodies were also detected. Elevation in antimesothelin antibodies correlated with OS.18 Based on these results, a randomized phase 3 trial at more than 70 centers was initiated in 2010 and enrolled 722 patients in 2013. Survival analysis is in progress (NCT01072981). A second phase 3 study called the Pancreatic Immunotherapy with Algenpantucel-L for Locally Advanced Non-Resectable trial has also been initiated and will evaluate the activity of algenpantucel-L combined with standard chemotherapy and chemoradiation in 280 patients with borderline resectable and locally advanced PDA (NCT01836432).

GVAX + CRS-207 Another whole-cell vaccine platform that has recently shown promise is the sequential 2-vaccine program with GVAX and CRS-207 (Fig 2). GVAX, like algenpan-

Anti–alpha gal complement fixation Alpha gal

PDA cell targeted Active CTL

Cell lysis

HAPa-1 and HAPa-2

TAA release

Inactive CD8+ T cell

Antigen presentation on MHC classes I/II

APC

Fig 1.— Algenpantucel-L–mediated tumor immunity. Alpha gal = galactosyl-alpha-1,3-galactose, APC = antigen-presenting cell, CTL = cytotoxic T-lymphocyte, MHC = major histocompatibility complex, PDA = pancreatic ductal adenocarcinoma, TAA = tumor-associated antigen. 244 Cancer Control

July 2014, Vol. 21, No. 3

tucel-L, is an irradiated allogeneic vaccine composed of 2 human PDA cell lines (Panc 10.05 and Panc 6.03) that have been modified by transfection of a plasmid containing the human GM-CSF gene.19 When they are intradermally injected, these cells secrete high levels of granulocyte–macrophage colony-stimulating factor (GM-CSF) at the vaccination site. In animal models, GM-CSF is the most potent cytokine at attracting APCs and promoting their differentiation. APCs from tumor-bearing hosts show reduced antigen-presenting activity. When treated with GM-CSF, antigen-presenting activity is rescued and these cells migrate to regional lymph nodes and activate CD4+ and CD8+ T cells. In a phase 1 study, 14 patients were treated with escalating doses of GVAX.19 No dose-limiting toxicities were seen. Delayed-type hypersensitivity responses to injection of irradiated autologous tumor cells was seen in 3 patients treated with at least 100 million cells. Although DFS was not a primary end point in the study, DFS longer than 25 months was noted in these 3 patients. The results from that trial led to a phase 2 study in 60 patients with R0 and R1 resected PDA.20 Patients received 500 million vaccine cells starting 8 to 10 weeks following surgery. Fluorouracil-based

chemoradiation was then given. In patients who remained free of disease, 3 additional vaccine treatments were given 1 month apart. The primary end point was DFS, with OS and induction of antitumor immune response as secondary end points. The rate of DFS at 12 months was 67.4% and the median DFS rate was 17.3 months. OS at 12 months was 85% with a median OS rate of 24.8 months. Postvaccination induction of antimesothelin CD8+ T cells in HLA-A1+ and HLA-A2+ patients correlated with DFS. CRS-207 is a genetically engineered strain of Listeria monocytogenes. L. monocytogenes is a gram-positive bacterium that is an intracellular pathogen capable of cell-to-cell spread by virtue of the actA virulence gene and invasion of nonphagocytic cells via the inlB gene. CRS-207 is a live-attenuated strain with deletions of both of these virulence genes. It has also been engineered to express mesothelin. Therefore, CRS-207 is able to directly deliver the TAA mesothelin to the intracellular compartment of APCs for processing and presentation on MHC classes I and II. These APCs can then activate effector T cells. L. monocytogenes also induces an inflammatory cytokine response that further recruits APCs.

Active CTL

PDA cell targeted Human GM‐CSF gene

GM­‐CSF

Panc 10.05 Panc 6.03

TAA

Antigen presentation on MHC class I/II

Inactive CD8+ T cell APC

PRIME

CRS­‐207

BOOST

Fig 2.— GVAX + CRS-207–mediated tumor immunity. CTL = cytotoxic T-lymphocyte, GM-CSF = granulocyte–macrophage colony-stimulating factor, MHC = major histocompatibility complex, PDA = pancreatic ductal adenocarcinoma, TAA = tumor-associated antigen. July 2014, Vol. 21, No. 3

Cancer Control 245

CRS-207 was evaluated in a phase 1 study of escalating doses in 17 patients with progressed mesothelioma, PDA, non–small-cell lung cancer, and ovarian cancer.21 The maximum tolerated dose was 1 × 109 CFU given for up to 4 doses. Evidence was suggestive of antimesothelin T-cell responses. In the study, 6 of 17 patients survived for at least 15 months, and 3 of these participants had PDA.21 However, it is worth noting that some of these patients were participants in prior GVAX trials, a fact suggestive of the synergy between the 2 vaccines, with GVAX priming an immune response and CRS-207 later boosting that response. Therefore, a phase 2 study of 90 patients with PDA and progressive disease or intolerant of chemotherapy was conducted.6 Participants were randomized 1:2 to GVAX alone for 6 doses every 3 weeks or 2 doses of GVAX followed by 4 doses of CRS-207. In prior studies of GVAX, large numbers of immunosuppressive regulatory T cells were seen at vaccine sites.19 Because cyclophosphamide treatment reduces the number and activity of regulatory T cells,22 low-dose cyclophosphamide was given during this study as an immune modulator prior to GVAX.6 The primary end point was OS. With a median follow-up of 7.8 months, the OS rate in those assigned to the GVAX arm was 3.9 months; for those assigned to the GVAX + CRS-207 arm, the OS rate was 6.1 months (hazard ratio [HR] 0.54; P = .011).6 In patients who received at least 3 doses of the vaccine, the median OS rates were 4.6 months for the GVAX arm and 9.7 months for the GVAX + CRS-207 arm (HR 0.44; P = .0074). Based on these promising results, a phase 2b trial was initiated. Study researchers intend to enroll 240 patients with metastatic PDA in the second line or greater to GVAX in combination with CRS-207, CRS-207 alone, or chemotherapy alone (NCT02004262). These 2 vaccine platforms have accomplished proof of principle that cell-based vaccines for pancreatic cancer can induce immune responses to relevant TAAs. These immune responses appear to correlate with rates of DFS and OS. Definitive proof of clinically significant effectiveness will depend on the results of ongoing randomized trials. If such efficacy is demonstrated, then efforts to augment the benefit by combining the vaccines with immune checkpoint inhibition can be expected to unleash the antitumor army within.

4. Antoniou G, Kountourakis P, Papadimitriou K, et al. Adjuvant therapy for resectable pancreatic adenocarcinoma: review of the current treatment approaches and future directions. Cancer Treat Rev. 2014;40(1):78-85. 5. 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(1):94-101. 6. Le DT, Wang-Gillam A, Picozziet V Jr, al. A phase 2, randomized trial of GVAX pancreas and CRS-207 immunotherapy versus GVAX alone in patients with metastatic pancreatic adenocarcinoma: updated results. J Clin Oncol. 2014;32(3 suppl):177. 7. Bazhin AV, Shevchenko I, Umansky V, et al. Two immune faces of pancreatic adenocarcinoma: possible implication for immunotherapy. Cancer Immunol Immunother. 2014;63(1):59-65. 8. Fukunaga A, Miyamoto M, Cho Y, et al. CD8+ tumor-infiltrating lymphocytes together with CD4+ tumor-infiltrating lymphocytes and dendritic cells improve the prognosis of patients with pancreatic adenocarcinoma. Pancreas. 2004;28(1):e26-e31. 9. 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(2 pt 1):498-504. 10. Heller A, Zörnig I, Müller T, et al. Immunogenicity of SEREX-identified antigens and disease outcome in pancreatic cancer. Cancer Immunol Immunother. 2010;59(9):1389-1400. 11. Shevchenko I, Karakhanova S, Soltek S, et al. Low-dose gemcitabine depletes regulatory T cells and improves survival in the orthotopic Panc02 model of pancreatic cancer. Int J Cancer. 2013;133(1):98-107. 12. Gabitass RF, Annels NE, Stocken DD, et al. Elevated myeloid-derived suppressor cells in pancreatic, esophageal and gastric cancer are an independent prognostic factor and are associated with significant elevation of the Th2 cytokine interleukin-13. Cancer Immunol Immunother. 2011;60(10):1419-1430. 13. Greten TF, Jaffee EM. Cancer vaccines. J Clin Oncol. 1999;17(3): 1047-1060. 14. Chen DS, Mellman I. Oncology meets immunology: the cancer-immunity cycle. Immunity. 2013;39(1):1-10. 15. Vogelstein B, Papadopoulos N, Velculescu VE, et al. Cancer genome landscapes. Science. 2013;339(6127):1546-1558. 16. Biankin AV, Waddell N, Kassahn KS, et al. Pancreatic cancer genomes reveal aberrations in axon guidance pathway genes. Nature. 2012;491(7424):399-405. 17. Galili U. The alpha-gal epitope and the anti-Gal antibody in xenotransplantation and in cancer immunotherapy. Immunol Cell Biol. 2005;83(6): 674-686. 18. Rossi GR, Hardacre JM, Mulcahy MF, et al. Effect of algenpantucel-L immunotherapy for pancreatic cancer on anti-mesothelin antibody (Ab) titers and correlation with improved overall survival. J Clin Oncol. 2013;31 (15 suppl):3007. 19. Jaffee EM, Hruban RH, Biedrzycki B, et al. Novel allogeneic granulocyte-macrophage colony-stimulating factor-secreting tumor vaccine for pancreatic cancer: a phase I trial of safety and immune activation. J Clin Oncol. 2001;19(1):145-156. 20. 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(2):328-335. 21. Le DT, Brockstedt DG, Nir-Paz R, et al. A live-attenuated Listeria vaccine (ANZ-100) and a live-attenuated Listeria vaccine expressing mesothelin (CRS207) for advanced cancers: phase I studies of safety and immune induction. Clin Cancer Res. 2012;18(3):858-868. 22. Le DT, Jaffee EM. Regulatory T-cell modulation using cyclophosphamide in vaccine approaches: a current perspective. Cancer Res. 2012;72(14): 3439-3444.

References 1. Howlader N, Noone AM, Krapcho M, et al, eds. SEER Cancer Statistics Review, 1975-2011. Bethesda, MD: National Cancer Institute. Published November 2013, modified April 2014. http://seer.cancer.gov/csr/1975_2011. Accessed April 28, 2014. 2. Von Hoff DD, Ervin T, Arena FP, et al. Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine. N Engl J Med. 2013;369(18): 1691-1703. 3. Conroy T, Desseigne F, Ychou Met al; Groupe Tumeurs Digestives of Unicancer; PRODIGE Intergroup. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med. 2011;364(19):1817-1825. 246 Cancer Control

July 2014, Vol. 21, No. 3

Pathology Report

Histopathological and Immunophenotypical Features of Intestinal-Type Adenocarcinoma of the Gallbladder and its Precursors Yan You, MD, Katherine Bui, Marilyn M. Bui, MD, PhD, Mokenge Malafa, MD, and Domenico Coppola, MD

Background: Intestinal-type adenocarcinoma of the gallbladder is an unusual malignancy associated with low- and high-grade intraepithelial neoplasms. The literature on the clinicopathologic characteristics of the precursor lesions of gallbladder cancer is limited, due in part to the variability in its definition and terminology. Methods: Here we report one case of intestinal-type adenocarcinoma of the gallbladder with distinctive morphology and associated precursor lesions. All of the hematoxylin and eosin stained slides were reviewed. Immunostains were performed using the avidin–biotin complex method for CK20, CK7, CDX2, MUC1, MUC2, and MUC-5AC. We also reviewed the literature discussing the current terminology from the World Health Organization for these lesions. Results: A 70-year-old man presented with epigastric abdominal pain and bloating. Computed tomography demonstrated a large heterogeneous gallbladder mass. Macroscopically, the gallbladder was 7.5 × 5.5 × 4.5 cm with smooth serosa. The lumen was occupied by a 5.0 × 4.5 × 3.0 cm irregular friable exophytic mass. The remaining mucosa had a tan brown to pink color with granular/papillary excrescences of up to 0.7 cm in thickness. Histologically, the tubulopapillary adenoma was lined by pseudostratified columnar epithelium with low and extensive high-grade dysplasia. Goblet cell and cystic dilatation were present in some glands. Immunohistochemistry showed that the intestinal type was positive for CK20, CK7, and CDX2, focally positive for MUC1/2, and negative for MUC-5AC. Conclusion: This case showed the complete spectrum of the progression of intestinal-type intracholecystic papillary-tubular neoplasms of the gallbladder.

Introduction Gallbladder carcinoma is a relatively uncommon neoplasm that has geographical and ethnic variations in its incidence. In the United States, it is more common in American Indians and Hispanic Americans than in Caucasians or African Americans.1,2 There is a female predominance, with the female-to-male ratio being 3–4:1.3 Most patients diagnosed with gallbladder carcinoma are in the sixth or seventh decade of life.4 Important risk factors for the disease include genetic backgrounds, gallstones, and abnormal choledochopancreatic junctions.4 The signs and From the Department of Pathology (YY) at the Peking Union Medical College Hospital, Beijing, China, the University of South Florida Honors College (KB), Tampa, Florida, and the Departments of Anatomic Pathology (MMB, DC) and Gastroenterology (MM, DC) at the H. Lee Moffitt Cancer Center & Research Institute, Tampa, Florida. Submitted January 8, 2014; accepted April 30, 2014. Address correspondence to Domenico Coppola, MD, H. Lee Moffitt Cancer Center, 12902 Magnolia Drive, Tampa, FL 33647. E-mail: [email protected] No significant relationships exist between the authors and the companies/organizations whose products or services may be referenced in this article. July 2014, Vol. 21, No. 3

symptoms are not specific, often resembling those of chronic cholecystitis. Right upper-quadrant pain is common. If the tumor is located in the gallbladder neck or its duct, then obstructive jaundice may present clinically. Gallbladder carcinoma usually forms an infiltrating, grey-white mass. Some carcinomas cause diffuse thickness of the gallbladder wall, while some present as polypoid or adenomatous in appearance. Most cases are not detectable on gross examination.5 Histologically, most gallbladder carcinomas exhibit pancreatobiliary-type epithelium, while others are composed of intestinal-, gastric foveolar-, or gastric pyloric-type epithelium. According to the most recent World Health Organization (WHO) classification, the precursor lesions of gallbladder epithelial tumors are adenoma, biliary intraepithelial neoplasia, intracystic or intraductal papillary neoplasm, and mucinous cystic neoplasm.4 Here we report one case of intestinal-type adenocarcinoma of the gallbladder with distinctive morphology, and we will discuss the histopathological and immunophenotypical features as well as the precursors and prognosis of the disease. Cancer Control 247

Materials and Methods A 70-year-old man presented with epigastric abdominal pain and bloating. He was subsequently diagnosed and treated with intestinal-type adenocarcinoma of the gallbladder. His clinical, radiological, and pathological data were retrospectively reviewed following the research guidelines of the University of South Florida and the H. Lee Moffitt Cancer Center in Tampa, Florida. The tissue was processed according to the guidelines of the College of American Pathologists. The hematoxylin and eosin stain and immunohistochemical (IHC) studies were performed at the Moffitt Cancer Center, and the IHC staining was carried out with the Discovery XT System (Ventana Medical Systems, Tucson, Arizona) as per the manufacturer’s protocol.

Results

Clinical Information Serum tumor markers, including total bilirubin, alkaline phosphatase, and carbohydrate antigen 19-9, were negative or within reference range. Computed tomog-

raphy demonstrated a large, heterogeneous gallbladder mass. Cholecystectomy and lymphadenectomy were performed. Gross Examination of the Tumor Macroscopically, the enlarged gallbladder was 7.5 × 5.5 × 4.5 cm in size with smooth serosa. The cystic duct had a 0.1-cm luminal diameter at its opening. The lumen was occupied by a friable exophytic mass 5.0 × 4.5 × 3.0 cm in size that was irregularly pink to tan in color. The rest of the mucosa had a tan brown to pink color with granular/papillary excrescences that were up to 0.7 cm in thickness. In addition, the lumen also contained a 3.0 × 2.5 × 2.5 cm irregular frambesiform calculus that was green-brown in color. Histology and Immunohistochemical Studies Histologically, the exophytic mass had a tubulopapillary adenomatous appearance with a large base attached to the mucosa (Fig 1A). The exophytic lesion was covered by a pseudostratified columnar epithe-

A

B

C

D

E

F

Fig 1. — (A) Tubulopapillary adenomatous appearance of the tumor. H&E stain, ×12.5. (B) The lesion has low- to high-grade dysplasia and focal invasion. H&E stain, ×40. (C) Surrounding microadenomatous satellite. H&E stain, ×12.5. (D) CK7 IHC stain was positive in the cytoplasm of the tumor. CK7 stain, ×100. (E) CDX2 IHC stain was positive in the nuclei of the tumor. CDX2 stain, ×100. (F) MUC1 IHC stain was positive in the cytoplasm of the tumor. MUC1 stain, ×100. H&E = hematoxylin and eosin, IHC = immunohistochemical. 248 Cancer Control

July 2014, Vol. 21, No. 3

lium with nuclear crowding with low- and extensive high-grade dysplasia. The neoplastic cells showed enlarged hyperchromatic and cigar-shaped nuclei, dense cytoplasmic chromophilia, and an increased nucleocytoplasmic rate. Invasion into the perimuscular connective tissue was also identified (Fig 1B). Goblet cell and cystic dilatation were present in some glands. In the vicinity of the main mass, several small tubulartype adenomas were present that had micropapillary architecture and showed low-grade epithelial dysplasia (Fig 1C). IHC confirmed the intestinal-type epithelium, which was positive for CK20, CK7 (Fig 1D), and CDX2 (Fig 1E), focally positive for MUC1 (Fig 1F), MUC2 and negative for MUC-5AC. No regional lymph node metastasis was observed.

Discussion Although microscopic foci of an intestinal differentiation often exist in cases of gallbladder adenocarcinoma, epithelial neoplasms composed wholly or predominantly of cells with an intestinal phenotype are unusual in the gallbladder. Two morphological variants of invasive adenocarcinoma of the intestinal type have been previously described.6 The first variant consists of glands predominantly lined with goblet cells, absorptive columnar cells, and a variable number of neuroendocrine and Paneth cells. The second variant is composed of branching tubular glands and papillary components, closely resembling colonic adenocarcinoma, and contains fewer goblet cells than the first type.7 Both types express antibodies to CDX2, MUC2, CEA, and CK20.4 The histological and immunophenotypical features of this case are consistent with the colonic variant of intestinal adenocarcinoma of the gallbladder. This carcinoma arose in a tubulopapillary adenoma and progressed to low-grade dysplasia, then highgrade dysplasia, and eventually invasion. Surrounding microadenomatous satellites were also present. These features are consistent with the proposed theory of the adenoma–carcinoma sequence of carcinogenesis, similar to what has been demonstrated in the large intestine.8 Adenomas of the gallbladder are an uncommon, benign epithelial neoplasm with a low incidence that ranges from 0.14% to 1.1% in different series.8,9 Most are single, small in size (< 2 cm), and are associated with cholelithiasis.10,11 Those measuring 1 cm or more are considered to be more frequently associated with cancer,12 and they are classified as being either tubular, papillary, or tubulopapillary according to the growth pattern and divided into pyloric, intestinal, foveolar, and biliary types.4,10 Several studies have demonstrated the association between adenoma and adenocarcinoma, and the reported malignant transformation rate ranges from 5% to 23.5%.8,9,13-15 However, July 2014, Vol. 21, No. 3

no definitive evidence of genetic alterations currently supports the relationship between adenoma and adenocarcinoma of the gallbladder.16,17 In addition, few researchers have found that gallbladder adenomas frequently present with mutations in the RAS/RAF/ MAPK pathway compared with gallbladder carcinomas, suggesting that adenomas and gallbladder carcinomas arise through different molecular pathways.18,19 In 2010, the WHO introduced intracystic papillary neoplasms into the classification of neoplasms of the gallbladder, a term that included pancreatobiliary and intestinal phenotypes.4 These lesions were previously designated as being papillary adenoma or noninvasive papillary carcinoma.4 Although intracystic papillary neoplasms have more mitotic figures, greater architectural complexity, and cytological atypia than conventional adenomas, no specific and/or quantitative criteria distinguish them.4 Therefore, Adsay et al20 proposed the term intracholecystic papillary-tubular neoplasms (ICPNs) of the gallbladder, a term that parallels the neoplasms present in the pancreas and ampullae (intraductal papillary mucinous neoplasms, intraductal tubulopapillary neoplasms, and biliary intraductal papillary neoplasms). Classifying a lesion as an ICPN requires that the lesion be an exophytic (papillary or polypoid) intramucosal gallbladder mass that measures 1.0 cm or more and is composed of preinvasive neoplastic (dysplasia) cells that form a compact lesion distinct from neighboring mucosa.20 This definition includes adenomas and intracystic papillary neoplasms and is recognized by the latest WHO classification. Our case showed the complete spectrum of the progression of intestinal-type ICPN with focal invasive carcinoma. In the study by Adsay et al,20 invasive carcinoma arose in 55% of ICPNs at the time of diagnosis; by contrast, 6.4% of all invasive carcinomas of the gallbladder contained an ICPN component. These results indicate that ICPNs are precursors of gallbladder carcinogenesis; however, it is worth noting that they are rare lesions.20 In addition, ICPNs had a significantly better prognosis than pancreatobiliary-type carcinomas of the gallbladder, even when they harbored invasive carcinoma. The median survival rates were 35 months for ICPNs with invasive carcinoma versus 9 months for pancreatobiliary-type carcinomas, a survival difference that was independent of tumor size and stage.20 These data suggest that ICPN-associated invasive carcinomas may have a distinctive biological nature.20 The literature on the clinicopathological characteristics of intestinal-type tumoral intraepithelial neoplasms is fairly limited. Fewer than 50 cases have been reported.13,20-24 Recently, 2 large series describing adenomas or ICPNs showed that the histology of intestinal-type lesions are similar to colonic adenomas or intestinal-type intraductal papillary mucinous neoCancer Control 249

plasms.20,21 Immunohistochemically, these lesions are 100% CK20+, 75% to 78% CDX2+, 33% to 50% MUC2+, 25% MUC1+, and 0% MUC-5AC+. High-grade dysplasia/carcinoma in situ was recognized in 46% (13/28) of intestinal adenomas. Invasive carcinoma was seen in 3.5% (1/28) of intestinal adenomas compared with 60% (6/10) of ICPNs, which may be due to the difference of definition. No deaths were observed.20,21

Conclusions Invasive adenocarcinoma that arises from a tumoral intraepithelial neoplasm appears to have a better overall clinical outcome than pancreatobiliary-type adenocarcinomas unaccompanied by ICPNs and presents with unique biological properties. The unified concept of adenomatous precursor lesions will help us better understand the nature of gallbladder neoplasms using the current terminology. Further studies are warranted, including those that focus on predictive factors, pathogenesis, and molecular genetic alterations.

Hum Pathol. 1999;30(1):21-25. 19. Pai RK, Mojtahed K, Pai RK. Mutations in the RAS/RAF/MAP kinase pathway commonly occur in gallbladder adenomas but are uncommon in gallbladder adenocarcinomas. Appl Immunohistochem Mol Morphol. 2011; 19(2):133-140. 20. Adsay V, Jang KT, Roa JC, et al. Intracholecystic papillary-tubular neoplasms (ICPN) of the gallbladder (neoplastic polyps, adenomas, and papillary neoplasms that are ≥1.0 cm): clinicopathologic and immunohistochemical analysis of 123 cases. Am J Surg Pathol. 2012;36(9):1279-1301. 21. Albores-Saavedra J, Chablé-Montero F, González-Romo MA, et al. Adenomas of the gallbladder. Morphologic features, expression of gastric and intestinal mucins, and incidence of high-grade dysplasia/carcinoma in situ and invasive carcinoma. Hum Pathol. 2012;43(9):1506-1513. 22. Petrović B, Petrović A, Živković V, et al. Intestinal type of villous adenoma of gallbladder. Acta Medica Medianae. 2010;49(3):50-54. 23. Toledo C, Matus CE, Barraza X, et al. Expression of HER2 and bradykinin B1 receptors in precursor lesions of gallbladder carcinoma. World J Gastroenterol. 2012;18(11):1208-1215. 24. Chang HJ, Jee CD, Kim WH. Mutation and altered expression of betacatenin during gallbladder carcinogenesis. Am J Surg Pathol. 2002;26(6): 758-766.

References 1. Shaffer EA. Gallstone disease: epidemiology of gallbladder stone disease. Best Pract Res Clin Gastroenterol. 2006;20(6):981-996. 2. Giang TH, Ngoc TT, Hassell LA. Carcinoma involving the gallbladder: a retrospective review of 23 cases - pitfalls in diagnosis of gallbladder carcinoma. Diagn Pathol. 2012;7:10. 3. Randi G, Franceschi S, La Vecchia C. Gallbladder cancer worldwide: geographical distribution and risk factors. Int J Cancer. 2006;118(7):1591-1602. 4. Albores-Saavedra J, Adsay NV, Crawford JM, et al. Carcinoma of the gallbladder and extrahepatic bile ducts. In: Bosman FT, Carneiro F, Hruban RH, et al, eds. WHO Classification of Tumors of the Digestive System. 4th ed. Lyon: IARC Press; 2010:266-274. 5. Roa I, Araya JC, Villaseca M, et al. Gallbladder cancer in a high risk area: morphological features and spread patterns. Hepatogastroenterology. 1999;46(27):1540-1546. 6. Albores-Saavedra J, Cruz-Ortiz H, Alcantara-Vazquez A, et al. Unusual types of gallbladder carcinoma. A report of 16 cases. Arch Pathol Lab Med. 1981;105(6):287-293. 7. Albores-Saavedra J, Nadji M, Henson DE. Intestinal-type adenocarcinoma of the gallbladder. A clinicopathologic study of seven cases. Am J Surg Pathol. 1986;10(1):19-25. 8. Kozuka S, Tsubone N, Yasui A, et al. Relation of adenoma to carcinoma in the gallbladder. Cancer. 1982;50(10):2226-2234. 9. Roa I, de Aretxabala X, Morgan R, et al. Clinicopathological features of gallbladder polyps and adenomas [in Spanish]. Rev Med Chil. 2004;132(6): 673-679. 10. Henson DE, Klimstra DS. Tumors of the Gallbladder, Extrahepatic Bile Ducts and Ampulla of Vater (Atlas of Tumor Pathology). 3rd ed., fascicle 27. Washington, DC: American Registry of Pathology; 2000:21-35. 11. Murakata LA, Albores-Saavedra J. Benign and malignant tumors of the gallbladder and extrahepatic bile ducts. In: Odze R, Goldblum J, Crawford J, eds. Surgical Pathology of the GI Tract, Liver, Biliary Tract, and Pancreas. Philadelphia: Saunders; 2004:639-672. 12. Goldin RD, Roa JC. Gallbladder cancer: a morphological and molecular update. Histopathology. 2009;55(2):218-229. 13. Lee SH, Lee DS, You IY, et al. Histopathologic analysis of adenoma and adenoma-related lesions of the gallbladder [in Korean]. Korean J Gastroenterol. 2010;55(2):119-126. 14. Nakajo S, Yamamoto M, Tahara E. Morphometrical analysis of gallbladder adenoma and adenocarcinoma with reference to histogenesis and adenoma-carcinoma sequence. Virchows Arch A Pathol Anat Histopathol. 1990;417(1):49-56. 15. Roa I, de Aretxabala X, Araya JC, et al. Preneoplastic lesions in gallbladder cancer. J Surg Oncol. 2006;93(8):615-623. 16. Kim YT, Kim J, Jang YH, et al. Genetic alterations in gallbladder adenoma, dysplasia and carcinoma. Cancer Lett. 2001;169(1):59-68. 17. Albores-Saavedra J, Tuck M, McLaren BK, et al. Papillary carcinomas of the gallbladder: analysis of noninvasive and invasive types. Arch Pathol Lab Med. 2005;129(7):905-909. 18. Wistuba II, Miquel JF, Gazdar AF, et al. Gallbladder adenomas have molecular abnormalities different from those present in gallbladder carcinomas. 250 Cancer Control

July 2014, Vol. 21, No. 3

Case Report

Follicular Lymphoma With Progression to Diffuse Large B-Cell Lymphoma and Concurrent CD5-Negative Mantle Cell Lymphoma-3 Entities in a Lymph Node Janese A. Trimaldi, MD, Jeremy W. Bowers, MD, Celeste Bello, MD, and Elizabeth M. Sagatys, MD

Summary: A 68-year-old woman with a history of follicular lymphoma had pathological findings of grade 3B follicular lymphoma, mantle cell lymphoma (MCL), and diffuse large B-cell lymphoma (DLBCL) identified in 1 lymph node. The DLBCL appeared to be a transformation of the follicular lymphoma. The nodules were diffusely and strongly positive for CD20, BCL6, and BCL2. CD43 highlighted smaller lymphocytes in a fraction of the nodules. BCL1 staining was variable with a mixture of nodular and mantle zone patterns. The diffuse areas showed weaker positivity for CD10, BCL2, and BCL6. CD3 and CD5 highlighted intermixed T cells. The Ki-67 proliferative index was overall estimated to be 60%. Fluorescent in situ hybridization performed on the lymph node was positive for CCND1/IGH. The patterns of BCL1 and BCL6 staining demonstrated 2 separate populations of neoplastic B lymphocytes.

by small lymphocytic lymphoma (SLL)/chronic lymphocytic leukemia (CLL) and classic HL, or, more commonly, a composite of SLL/CLL and DLBCL.5 The clonality of the second lymphoma in Richter syndrome has been the focus of many studies because of the insight it provides into lymphomagenesis. The aggressive component of Richter syndrome may be derived from the original neoplastic clone or derived from a second unrelated neoplastic clone.6,7 Low-grade NHLs transform into high-grade neoplasms at variable frequency, but this phenomenon usually represents an evolution of the same clonal process.8-10 On occasion, 2 distinct diseases can arise from a single clone.10,11 A number of so-called “biphenotypic B-cell neoplasms” with 2 phenotypically unrelated malignant populations arising in a patient either synchronous or metasynchronous have been described.12,13

Background

Clinical History A 68-year-old woman noticed an enlarging mass on the right side of her neck. Upon presentation, she was asymptomatic. A surgical excision revealed a 3.4-cm lymph node, which was diagnosed as grade 1 follicular lymphoma. She was followed with close observation. Increasing lymphadenopathy was identified approximately 22 months later, and excisional biopsy of a left axillary lymph node was performed at that time. Tissue sections showed histological evidence of grade 3 follicular lymphoma, DLBCL, and focal MCL. Following the diagnosis of CL, she received 6 cycles with cyclophosphamide, doxorubicin, vincristine, prednisone, and rituximab. The patient did well for approximately 15 months before she developed shortness of breath and was found to have a pleural effusion in the right lung. Imaging also revealed extensive lymphadenopathy involving the mediastinum, hilar, and right retrocrural lymph nodes. A significant pleural thickening in the right pleural area and a soft-tissue mass in the right pleura were also identified on imaging. Flow cytometry was performed on the effusion and confirmed the presence of B-cell lympho-

Composite lymphomas (CLs) are an uncommon type of lymphoid neoplasm defined as the coexistence of 2 morphologically and phenotypically distinct types of lymphoid neoplasms occurring in a single site.1 The identified combinations are varied and can include any combination of Hodgkin lymphoma (HL) with non-Hodgkin lymphoma (NHL) or 2 morphologically distinct types of NHL.2-4 One of the most classic, well-recognized CL is Richter syndrome, represented From the Department of Pathology and Cell Biology (JAT), University of South Florida Morsani College of Medicine, Tampa, Florida, the Department of Pathology (JWB), Bayfront Medical Center, St Petersburg, Florida, and the Departments of Malignant Hematology (CB) and Hematopathology and Laboratory Medicine (EMS) at the H. Lee Moffitt Cancer Center & Research Institute, Tampa, Florida. Submitted May 2, 2014; accepted May 7, 2014. Address correspondence to Elizabeth M. Sagatys, MD, Department of Hematopathology and Laboratory Medicine, Moffitt Cancer Center, 12902 Magnolia Drive, MCC-LAB, Tampa, FL 33612. E-mail: [email protected] No significant relationships exist between the authors and the companies/organizations whose products or services may be referenced in this article. July 2014, Vol. 21, No. 3

Case Report

Cancer Control 251

ma. The patient subsequently developed night sweats and lost 25 pounds. She received 2 cycles of bendamustine and rituximab but had persistent disease. She continued to require numerous paracenteses and was started on salvage therapy. She died approximately 6 months following the onset of pleural effusions. Pathological Findings Review of the hematoxylin and eosin stained sections of the left axillary lymph node biopsy showed that the majority of the lymph node architecture was effaced by a nodular and diffuse infiltrate (Fig 1). The nodular areas were predominantly composed of intermediate- to large-sized lymphocytes with vesicular chromatin, compatible with centroblasts. Intermixed centrocytes were rare. Diffuse areas of centroblasts were also seen (Fig 2). Increased mitotic figures were noted. In addition, nodules of smaller lymphocytes with hyperchromatic chromatin and slightly irregular nuclear contours were found. Immunohistochemical stains performed on the lymph node showed that CD20 was strongly and diffusely positive in the majority of the lymphocytes. BCL2 was also positive, with stronger staining in the

A

nodular regions. CD43 highlighted mostly smaller lymphocytes in some of the nodules. CD10 and BCL6 highlighted the expanded follicles, with weaker staining in the large cells of the more diffuse areas. BCL1 had variable staining, with a mantle zone pattern in some areas, and a more nodular pattern in others, highlighting the small irregular lymphocytes. BCL6 and BCL1 appeared to be staining 2 separate populations in the nodules (Fig 3). The Ki-67 proliferative index was estimated at approximately 60%. CD3 and CD5 highlighted intermixed T cells. CD5 was negative in the B-cell population. Fluorescent in situ hybridization studies of the axillary lymph node identified a translocation involving chromosomes 11q13 and 14q32.3 (CCNH/IGH fusion), confirming the presence of MCL. Overall, the lymph node showed DLBCL and grade 3b follicular lymphoma, each comprising approximately 40% of the node; MCL comprised approximately 20% of the node.

Discussion A review of the literature for coexisting follicular and MCLs revealed approximately 12 reported cases.11,13-20 Most cases described the 2 lymphomas adjacent to each other. One case reported an intermixed pattern of MCL and follicular lymphoma in a patient who had a poor outcome.20 In addition to follicular lymphoma, MCL has been found to coexist with CLL/SLL, plasma cell dyscrasias, and HL.3,15,21,22 MCL is a mature B-cell neoplasm expressing the pan B-cell markers CD19, CD20, CD22, and CD79a, along with the aberrant expression of CD5. MCL usually has a more aggressive clinical course than other small B-cell NHLs. An increased proliferation index usually indicates a more aggressive clinical course in MCL. Our case had a proliferative index of 60%, which is considered high, and the patient had a poor outcome. Our case exhibited an aberrant phenotype in the MCL, lacking CD5, which has been previously re-

B Fig 1. — (A) Low power view of nodular “area.” (B) Diffuse areas of lymph node. Hematoxylin and eosin stain, ×40. 252 Cancer Control

Fig 2. — High power view of area of diffuse large B-cell lymphoma. Hematoxylin and eosin stain, ×400. July 2014, Vol. 21, No. 3

A BCL1

B BCL6

C CD20

D Ki-67

Fig 3. — (A) Immunohistochemical stains on the lymph node showing the areas of BCL1-positive mantle cell lymphoma and (B) BCL6-positive diffuse large B-cell lymphoma and grade 3 follicular lymphoma. (C) CD20 is diffusely and strongly positive in all areas of the lymph node. (D) The Ki-67 proliferation index was approximately 60% overall with no clear difference in the different areas.

ported.23,24 Aberrant phenotypes (CD5 negative, CD10 positive, BCL6 positive) may be associated with blastoid and pleomorphic variants of MCL.23 One review of 25 CD5 negative cases of MCL showed a lymphocytic variant in 20 cases and a blastoid variant in 5 cases.25 Our case displayed a lymphocytic morphology, and CD10 and BCL6 were negative in the MCL in our case. Follicular lymphoma usually expresses germinal center cell markers CD10 and BCL6 and antiapoptotic gene BCL2. Follicular lymphoma can be traditionally separated from MCL through the demonstration of a BCL2/IGH rearrangement and a lack of CCND1/IGH rearrangement. MCL colonizing the follicle center will usually express BCL2 by immunohistochemistry, so it is important to perform CCND1 (BCL1) immunohistochemistry, fluorescent in situ hybridization, or both in cases where any concern exists for MCL. A long-term study by Montoto et al26 showed that, at 10 years, follicular lymphoma transforms to DLBCL in approximately 28% of cases. Advanced-stage and high-risk Follicular Lymphoma International Prognostic Index and International Prognostic Index scores at diagnosis July 2014, Vol. 21, No. 3

correlate with an increased risk of transformation.26 Our case demonstrated transformation to DLBCL from the originally diagnosed grade 1 follicular lymphoma in approximately 2 years. CLs are theorized to represent a more aggressive phase of disease and may have a worse prognosis. Our case had an aggressive clinical course and a poor prognosis following the identification of the CL. As we move toward more personalized medicine, it will be important to recognize these cases, because targeted therapy may improve outcomes. We must also recognize those cases that may require more aggressive therapy. In cases in which clinical findings do not match pathological findings, additional biopsies, including excisional lymph node biopsies, may be necessary to identify areas of transformation or possible CL. References 1. Kim H, Hendrickson R, Dorfman RF. Composite lymphoma. Cancer. 1977;40(3):959-976. 2. Dargent JL, Lespagnard L, Meiers I, et al. Composite follicular lymphoma and nodular lymphocyte predominance Hodgkin’s disease. Virchows Arch. 2005;447(4):778-780. Cancer Control 253

3. Caleo A, Sanchez-Aguilera A, Rodriguez S, et al. Composite Hodgkin lymphoma and mantle cell lymphoma: two clonally unrelated tumors. Am J Surg Pathol. 2003;27(12):1577-1580. 4. Kim H. Composite lymphoma and related disorders. Am J Clin Pathol. 1993;99(4):445-451. 5. Trump DL, Mann RB, Phelps R, et al. Richter’s syndrome: diffuse histiocytic lymphoma in patients with chronic lymphocytic leukemia. A report of five cases and review of the literature. Am J Med. 1980;68(4):539-548. 6. Ohno T, Smir BN, Weisenburger DD, et al. Origin of the Hodgkin/ReedSternberg cells in chronic lymphocytic leukemia with “Hodgkin’s transformation.” Blood. 1998;91(5):1757-1761. 7. Giles FJ, O’Brien SM, Keating MJ. Chronic lymphocytic leukemia in (Richter’s) transformation. Semin Oncol. 1998;25(1):117-125. 8. De Jong D, Voetdijk BM, Beverstock GC, et al. Activation of the c-myc oncogene in a precursor-B-cell blast crisis of follicular lymphoma, presenting as composite lymphoma. N Engl J Med. 1988;318(21):1373-1378. 9. Zelenetz AD, Chen TT, Levy R. Histologic transformation of follicular lymphoma to diffuse lymphoma represents tumor progression by a single malignant B cell. J Exp Med. 1991;173(1):197-207. 10. Dogan A, Du MQ, Aiello A, et al. Follicular lymphomas contain a clonally linked but phenotypically distinct neoplastic B-cell population in the interfollicular zone. Blood. 1998;91(12):4708-4714. 11. Tsang P, Pan L, Cesarman E, et al. A distinctive composite lymphoma consisting of clonally related mantle cell lymphoma and follicle center cell lymphoma. Hum Pathol. 1999;30(8):988-992. 12. Steinhoff M, Hummel M, Assaf C, et al. Cutaneous T cell lymphoma and classic Hodgkin lymphoma of the B cell type within a single lymph node: composite lymphoma. J Clin Pathol. 2004;57(3):329-331. 13. Zamo A, Zanotti R, Lestani M, et al. Molecular characterization of composite mantle cell and follicular lymphoma. Virchows Arch. 2006;448(5): 639-643. 14. Ilgenfritz RB, Le Tourneau A, Arborio M, et al. Composite mantle cell and follicular lymphoma. A case report. Hum Pathol. 2009;40(2):259-263. 15. Fend F, Quintanilla-Martinez L, Kumar S, et al. Composite low grade B-cell lymphomas with two immunophenotypically distinct cell populations are true biclonal lymphomas. A molecular analysis using laser capture microdissection. Am J Pathol. 1999;154(6):1857-1866. 16. Roullet MR, Martinez D, Ma L, et al. Coexisting follicular and mantle cell lymphoma with each having an in situ component: A novel, curious, and complex consultation case of coincidental, composite, colonizing lymphoma. Am J Clin Pathol. 2010;133(4):584-591. 17. Wang S, Tzankov A, Xu-Monette ZY, et al. Clonally related composite follicular lymphoma and mantle cell lymphoma with clinicopathologic features and biological implications. Hum Pathol. 2013;44(12):2658-2667. 18. Korac P, Horvat T, Kardum Paro MM, et al. Follicular and mantle cell lymphoma characteristics present simultaneously in the same lymph node. Appl Immunohistochem Mol Morphol. 2013;21(6):572-576. 19. Matsuoka A, Tsushima T, Tanibuchi M, et al. Composite lymphoma cosisting of mantle cell lymphoma and follicular lymphoma. Rinsho Ketsueki. 2013;54(11):2056-2061. 20. Liu Y, Li P, Guo Y, et al. A unique composite follicular lymphoma and mantle cell lymphoma with a mixed cell pattern and aggressive course. Am J Clin Pathol. 2014;141(5):737-741. 21. Yamaguchi M, Ohno T, Miyata E, et al. Analysis of clonal relationship using single-cell polymerase chain reaction in a patient with concomitant mantle cell lymphoma and multiple myeloma. Int J Hematol. 2001;73(3):383-385. 22. Tinguely M, Rosenquist R, Sundstrom C, et al. Analysis of a clonally related mantle cell and Hodgkin lymphoma indicates Epstein-Barr virus infection of a Hodgkin/Reed-Sternberg cell precursor in a germinal center. Am J Surg Pathol. 2003;27(11):1483-1488. 23. Morice WG, Hodnefield JM, Kurtin PJ, et al. An unusual case of leukemic mantle cell lymphoma with a blastoid component showing loss of CD5 and aberrant expression of CD10. Am J Clin Pathol. 2004;122(1):122-127. 24. Bell ND, King JA, Kusyk C, et al. CD5 negative diffuse mantle cell lymphoma with splenomegaly and bone marrow involvement. South Med J. 1998;91(6):584-587. 25. Liu Z, Dong HY, Gorczyca W, et al. CD5- mantle cell lymphoma. Am J Clin Pathol. 2002;118(2):216-224. 26. Montoto S, Davies AJ, Matthews J, et al. Risk and clinical implications of transformation of follicular lymphoma to diffuse large B-cell lymphoma. J Clin Oncol. 2007;25(17):2426-2433.

254 Cancer Control





July 2014, Vol. 21, No. 3

FACULTY POSITION: MEDICAL ONCOLOGIST Moffitt Cancer Center’s Department of Gastrointestinal Oncology invites applications for a Junior-level faculty appointment. The successful candidate will have an interest in clinical duties and research activities focused on neuroendocrine oncology and will place and refer patients for clinical trials. Moffitt Cancer Center has one of the largest neuroendocrine tumor clinics in the country. Successful candidates must hold an MD, be ABIM board certified/eligible, and fellowship trained in medical oncology. Experience in a clinical multidisciplinary academic setting is preferred. Florida medical license or eligibility is required. For inquiries about the position, contact Jonathan Strosberg, MD, Department of Gastrointestinal Oncology, at [email protected] or 813-745-3636. To apply, visit our Web page at MOFFITT.org/careers and refer to requisition number 11756.

FACULTY POSITION: SENIOR MEDICAL ONCOLOGIST Moffitt Cancer Center’s Department of Gastrointestinal Oncology invites applications for a Senior-level medical oncologist. The successful candidate will have an interest in clinical duties and research activities focused on colorectal cancer. The candidate will place and refer patients for clinical trials. Successful candidates must hold an MD, be ABIM board certified/eligible, and fellowship trained in medical oncology. Experience in a clinical, multidisciplinary academic setting is preferred. Florida medical license or eligibility is required. For inquiries about the position, contact Mokenge Malafa, MD, Department of Gastrointestinal Oncology, at [email protected] or 813-745-1432. To apply, visit our Web page at MOFFITT.org/careers and refer to requisition number 11758.

FACULTY POSITIONS: ANATOMIC SURGICAL PATHOLOGISTS The Moffitt Cancer Center is seeking to fill surgical pathologist positions in its Department of Anatomic Pathology. The successful candidates must hold an MD or MD/PhD and have an interest in genitourinary, gastrointestinal, breast, dermatological, or cytopathology. Those with additional expertise in pathology informatics are strongly encouraged to apply. The candidate will also be a member of an organ-specific multidisciplinary team of physicians and scientists, and he or she will have opportunities to join established basic research programs, with excellent opportunities for translational research in areas of related interest. Opportunities also exist for participation in molecular and early cancer detection studies. Candidates will diagnose and report specific cancers, confer and interact with physicians, medical students, residents, and fellows, and participate in pathology tumor board meetings. In addition to a division-specific focus, the positions will include responsibility in general surgical pathology. An established basic or translational research focus is highly desirable. Successful applicants should be board certified/eligible in anatomic pathology. Experience in an academic, multidisciplinary clinical setting is preferred. Licensure or eligibility for licensure in Florida is required. For inquiries about the position, contact Anthony Magliocco, MD, Department of Anatomic Pathology, at [email protected] or 813-745-3741. To apply, visit our Web page at MOFFITT.org/careers and refer to requisition number 12798. The H. Lee Moffitt Cancer Center & Research Institute, a rapidly growing NCI-designated Comprehensive Cancer Center, is committed to education through a wide range of residency and fellowship programs. The Cancer Center is composed of a large ambulatory care facility, a 206-bed hospital, with a 36-bed blood and marrow transplant program, 15 state-of-the-art operating suites, a 30-bed intensive care unit, a high-volume screening program, and a basic science research facility. The Moffitt Research Institute is composed of approximately 150 principal investigators, 58 laboratories, and 306,000 square feet of research space. The Moffitt Cancer Center is affiliated with the University of South Florida. Primary and secondary university appointments are available as applicable. Academic rank is commensurate with qualifications and experience.

Moffitt Cancer Center provides a tobacco-free work environment. It is an equal opportunity, affirmative action employer and a drug-free workplace.

July 2014, Vol. 21, No. 3

Cancer Control 255

Contact and General Information

Cancer Control: Journal of the Moffitt Cancer Center is published by H. Lee Moffitt Cancer Center & Research Institute and is included in Index Medicus®/ MEDLINE® and EMBASE®/Excerpta Medica, Thomson Reuters Science Citation Index Expanded (SciSearch®) and Journal Citation Reports/Science Edition. Cancer Control currently has an impact factor of approximately 3.6. This peer-reviewed journal contains articles on the spectrum of actions and approaches needed to reduce the impact of human malignancy. Cancer Control is sent at no charge to approximately 15,000* medical professionals, including oncologists in all subspecialties; selected primary care physicians; medical researchers who specialize in oncology; and others who have a professional interest in cancer control.

To Cancel a Subscription:

* This includes approximately 1,000 oncologists from over

To Inquire About Reprints:

To Access the Journal Online:

For information on obtaining reprints and pricing information, please contact the Editorial Coordinator.

Most issues and supplements of Cancer Control are available at MOFFITT.org/ccj.

To Submit an Article for Publication:

85 countries.

To Change Your Mailing Address: Please provide your old address and your new address to: Veronica Nemeth, Editorial Coordinator Cancer Control Journal Moffitt Cancer Center MBC-JRNL 12902 Magnolia Drive Tampa, FL 33612 E-mail: [email protected] Fax: 813-449-8680 Phone: 813-745-1348

Please send your cancellation request, including your name and address, to the Editorial Coordinator, as listed to the left.

To Order an Individual, Institutional, or International Subscription: For information on subscription rates, please contact the Editorial Coordinator.

To Request Permission to Reprint Copyrighted Materials: To request permission to reprint tables, figures, or materials copyrighted by Cancer Control, please contact the Editorial Coordinator.

The editor welcomes submission of manuscripts pertaining to all phases of oncology care for possible publication in Cancer Control. Articles are subject to editorial evaluation and peer review. Author guidelines are available online at MOFFITT.org/ccj.

Follow us on Twitter @MoffittResearch #MoffittCCJ

Celebrating Over 20 Years of Publishing YOUR OPINION COUNTS! Access our readership survey at Moffitt.org/ccj

256 Cancer Control

July 2014, Vol. 21, No. 3

Save the Date!

presents

Advances in the Management of Multiple Myeloma March 6–7, 2015 St. Petersburg Beach, Florida Course Directors: Melissa Alsina, MD, Rachid Baz, MD, and Kenneth H. Shain, MD, PhD Moffitt Cancer Center, Tampa, Florida Conference Overview: Advances in the Management of Multiple Myeloma conference is designed to foster the exchange of the most recent advances in the biology and treatment of multiple myeloma. National and international leading experts in the field will present in a format promoting discussion and interaction with participants. Target Audience: This education program is directed toward hematologists, medical and surgical oncologists, and BMT physicians who diagnose, treat, and manage patients with multiple myeloma. Oncology fellows, nurses, and physician assistants who are interested in the diagnosis, care, and treatment of multiple myeloma are also invited to attend.

To be added to the conference mailing list, contact: Moffitt Cancer Center | Marsha Moyer, MBA | 813-745-2286 | [email protected]

S AV E T H E D A T E MOFFITT CANCER CENTER PRESENTS

NEW FRONTIERS IN UROLOGIC ONCOLOGY Saturday, September 20, 2014 Marriott Waterside | Tampa, FL CONFERENCE DIRECTOR: Julio Pow-Sang, MD This one-day conference designed for urologists, medical oncologists, radiation oncologists, and researchers will provide the newest research and clinical advances in urologic oncology focusing on prostate, bladder, kidney and testicular/penile cancers.

CONTACT: [email protected] PROVIDED BY