Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
Targeted Therapy in Translational Cancer Research
TranslationalOncologyS E R I E S E D I T O R S
Robert C. Bast,Maurie Markmanand Ernest Hawk
Edited byApostolia-Maria Tsimberidou,Razelle Kurzrock andKenneth C. Anderson
Targeted Therapy inTranslational Cancer Research
Translational Oncology
SERIES EDITORS
ROBERT C. BAST, MDVice President for Translational ResearchThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
MAURIE MARKMAN, MDSenior Vice President for Clinical AffairsCancer Treatment Centers of America
Clinical Professor of MedicineDrexel University College of MedicinePhiladelphia, PA, USA
ERNEST HAWK, MD, MPHVice President, Division of OVP, Cancer Prevention and Population SciencesThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Targeted Therapy inTranslational CancerResearchEDITED BY
Apostolia-Maria Tsimberidou, MD, PhDDepartment of Investigational Cancer TherapeuticsThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Razelle Kurzrock, MDCenter for Personalized Cancer TherapyUC San Diego Moores Cancer CenterLa Jolla, CA, USA
Kenneth C. Anderson, MD, PhDLeBow Institute for Myeloma Therapeutics and Jerome Lipper Myeloma CenterDepartment of Medical Oncology, Dana-Farber Cancer InstituteHarvard Medical SchoolBoston, MA, USA
Copyright © 2016 by John Wiley & Sons, Inc. All rights reserved.
Published by John Wiley & Sons, Inc., Hoboken, New JerseyPublished simultaneously in Canada
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means,electronic, mechanical, photocopying, recording, scanning, or otherwise, except as permitted under Section 107 or 108 ofthe 1976 United States Copyright Act, without either the prior written permission of the Publisher, or authorizationthrough payment of the appropriate per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers,MA 01923, (978) 750-8400, fax (978) 750-4470, or on the web at www.copyright.com. Requests to the Publisher forpermission should be addressed to the Permissions Department, John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ07030, (201) 748-6011, fax (201) 748-6008, or online at http://www.wiley.com/go/permission.
The contents of this work are intended to further general scientific research, understanding, and discussion only and arenot intended and should not be relied upon as recommending or promoting a specific method, diagnosis, or treatment byhealth science practitioners for any particular patient. The publisher and the author make no representations or warrantieswith respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, includingwithout limitation any implied warranties of fitness for a particular purpose. In view of ongoing research, equipmentmodifications, changes in governmental regulations, and the constant flow of information relating to the use of medicines,equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert orinstructions for each medicine, equipment, or device for, among other things, any changes in the instructions or indicationof usage and for added warnings and precautions. Readers should consult with a specialist where appropriate. The fact thatan organization or Website is referred to in this work as a citation and/or a potential source of further information does notmean that the author or the publisher endorses the information the organization or Website may provide orrecommendations it may make. Further, readers should be aware that Internet Websites listed in this work may havechanged or disappeared between when this work was written and when it is read. No warranty may be created or extendedby any promotional statements for this work. Neither the publisher nor the author shall be liable for any damages arisingherefrom.
For general information on our other products and services or for technical support, please contact our Customer CareDepartment within the United States at (800) 762-2974, outside the United States at (317) 572-3993 or fax (317) 572-4002.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available inelectronic formats. For more information about Wiley products, visit our web site at www.wiley.com.
Library of Congress Cataloging-in-Publication Data:
Targeted therapy in translational cancer research / edited by Apostolia-Maria Tsimberidou, Razelle Kurzrock, Kenneth C.Anderson.
p. ; cm. – (Translational oncology)Includes bibliographical references and index.ISBN 978-1-118-46857-9 (cloth)I. Tsimberidou, Apostolia-Maria, editor. II. Kurzrock, Razelle, editor. III. Anderson, Kenneth C., editor. IV. Series:
Translational oncology (Series)[DNLM: 1. Molecular Targeted Therapy. 2. Neoplasms–drug therapy. 3. Immunotherapy. 4. Individualized
Medicine. 5. Translational Medical Research. QZ 267]RC271.I45616.99′4061–dc23
2015015320
Printed in Singapore by Markono Print Media Pte Ltd
10 9 8 7 6 5 4 3 2 1
Contents
List of Contributors, vii
Series Foreword, xiii
Foreword, xv
Preface, xvii
Part I Principles of Targeted Therapies
1 Toward Personalized Therapy for Cancer, 3Ashley M. Holder and Funda Meric-Bernstam
2 Combining Targeted Therapies, 14Jordi Rodon, Analia Azaro, Davis Torrejon,and Razelle Kurzrock
3 Principles of Targeted Immunotherapy, 27Susanne H. C. Baumeister and Glenn Dranoff
4 Cancer Stem Cell Principles, 39Allison C. Sharrow, Gabriel Ghiaur, and Richard J. Jones
5 The Tumor Microenvironment as a Target for TherapeuticIntervention, 47Hua Fang and Yves A. DeClerck
6 The Role of Angiogenesis in Cancer, 64Morgan Taylor, Robert L. Coleman, and Anil K. Sood
7 Epigenetics and Epigenetic Therapy of Cancer, 72Omotayo Fasan, Patrick Boland, Patricia Kropf, andJean-Pierre J. Issa
8 The Role of microRNAs in Cancer, 80Gianpiero Di Leva and Carlo M. Croce
9 Acute Myeloid Leukemia, 89Ofir Wolach and Richard M. Stone
10 Targeted and Functional Imaging, 101Jian Q. (Michael) Yu, Drew A. Torigian, and Abass Alavi
Part II Targeted Therapy in HematologicalMalignancies
11 Targeted Therapies in Chronic Myeloid Leukemia, 113Elias Jabbour and Jorge Cortes
12 Targeted Therapy for Acute Lymphoblastic Leukemia, 121Nitin Jain, Susan O’Brien, and Farhad Ravandi-Kashani
13 Chronic Lymphocytic Leukemia, 130Preetesh Jain and Susan O’Brien
14 Multiple Myeloma, 145Giada Bianchi and Kenneth C. Anderson
15 The Impact of Genomics on Targeted Therapy in MultipleMyeloma and Lymphomas, 157Jens G. Lohr and Birgit Knoechel
16 Targeted Therapy in Myelodysplastic Syndromes, 162Guillermo Montalban-Bravo andGuillermo Garcıa-Manero
17 Lymphoma and Targeted Therapies, 169Sonali M. Smith and Julie M. Vose
Part III Targeted Therapy in Solid Tumors
18 Targeted Therapy in Solid Tumors: Brain, 179Shiao-Pei Weathers, Barbara J. O’Brien,John F. de Groot, and W. K. Alfred Yung
19 Targeted Therapy for Breast Cancer, 190Harold J. Burstein
20 Targeted Therapy in Solid Tumors: Colorectal Cancer, 193Maen Abdelrahim, Scott Kopetz, and David Menter
21 Endometrial Cancer, 205Jessica L. Bowser, Russell R. Broaddus,Robert L. Coleman, and Shannon N. Westin
22 Targeted Therapy in Solid Tumors: Head and Neck, 216Marcus M. Monroe and Jeffrey N. Myers
23 Targeted Therapy in Solid Tumors: Lung Cancer, 224Saiama N. Waqar, Daniel Morgensztern,and Roy S. Herbst
24 Targeted Therapy in Melanoma, 231Keith T. Flaherty
25 Ovarian Cancer, 240Shannon N. Westin, Larissa A. Meyer, andRobert L. Coleman
26 Molecular Therapeutics: Pancreatic Cancer, 255David Fogelman, Milind Javle, and James Abbruzzese
27 Targeted Therapies for Pediatric Solid Tumors, 263Jasmine Quynh Dao and Patrick A. Zweidler-McKay
28 Prostate Cancer, 273William G. Nelson, Michael C. Haffner, andSrinivasan Yegnasubramanian
v
vi Contents
29 Renal Cell Carcinoma and Targeted Therapy, 287Benjamin A. Gartrell, Alexander C. Small,William K. Oh, and Matthew D. Galsky
30 Targeted Therapy in Solid Tumors: Sarcomas, 296Anthony P. Conley, Vinod Ravi, and Shreyaskumar Patel
Part IV Targeted Therapy for Specific MolecularAberrations
31 RAS-RAF-MEK Pathway: Aberrations and TherapeuticPossibilities, 305Javier Munoz and Filip Janku
32 The Phosphatidylinositol 3-Kinase Pathway in HumanMalignancies, 315Samuel J. Klempner, Thanh-Trang Vo, Andrea P. Myers,and Lewis C. Cantley
33 Current Status and Future Direction of PARP Inhibition inCancer Therapy, 325Saeed Rafii, Stan Kaye and Susana Banerjee
34 Targeting the c-Met Kinase, 341Chad Tang, M. Angelica Cortez, David Hong, andJames W. Welsh
35 KIT Kinase, 347Scott E. Woodman
36 TP53, 353Kensuke Kojima and Michael Andreeff
Part V Future Perspectives
37 Future Perspectives, 363Rabih Said and Apostolia-Maria Tsimberidou
Index, 371
List of Contributors
James Abbruzzese, MDDivision of Medical OncologyDuke Cancer InstituteDurham, NC, USA
Maen Abdelrahim, MD, PhDDepartment of Internal MedicineBaylor College of MedicineHouston, TX, USA
Abass Alavi, MD, MD(Hon.), PhD(Hon.), DSc(Hon.)Department of RadiologyHospital of the University of PennsylvaniaPhiladelphia, PA, USA
Kenneth C. Anderson, MD, PhDLeBow Institute for Myeloma Therapeutics and Jerome Lipper MyelomaCenterDepartment of Medical Oncology, Dana-Farber Cancer InstituteHarvard Medical SchoolBoston, MA, USA
Michael Andreeff, MD, PhDSection of Molecular Hematology and TherapyDepartment of LeukemiaThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Analia Azaro, MDEarly Clinical Drug Development GroupVall d’Hebron Institute of OncologyUniversitat Autonoma de BarcelonaBarcelona, Spain
Susana Banerjee, MBBS, MA, MRCP, PhDThe Royal Marsden HospitalLondon, UK
Robert C. Bast, MDThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Susanne H. C. Baumeister, MDDepartment of Pediatric OncologyDana-Farber Cancer InstituteBoston, MA
Division of Hematology-OncologyBoston Children’s HospitalHarvard Medical SchoolBoston, MA
Giada Bianchi, MDLeBow Institute for Myeloma Therapeutics and Jerome Lipper MyelomaCenterDepartment of Medical Oncology, Dana-Farber Cancer InstituteHarvard Medical SchoolBoston, MA, USA
Patrick Boland, MDDepartment of MedicineTemple University School of MedicinePhiladelphia, PA, USA
Jessica L. Bowser, PhDDepartment of PathologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Russell R. Broaddus, MD, PhDDepartment of PathologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Harold J. Burstein, MD, PhDDana-Farber Cancer InstituteBrigham and Women’s HospitalHarvard Medical SchoolBoston, MA, USA
Lewis C. Cantley, PhDMeyer Cancer Center at Weill Cornell Medical CollegeNew York, NY, USA
Robert L. Coleman, MDDepartment of Gynecologic Oncology and Reproductive MedicineCenter for RNA Interference and Non-Coding RNAThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Anthony P. Conley, MDDepartment of Sarcoma Medical OncologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Jorge Cortes, MDDepartment of LeukemiaThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
M. Angelica Cortez, PhDDepartment of Experimental Radiation OncologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
vii
viii List of Contributors
Carlo M. Croce, MDDepartment of Molecular Virology, Immunology and Medical GeneticsComprehensive Cancer CenterOhio State UniversityColumbus, OH, USA
Jasmine Quynh Dao, MDChildren’s Cancer HospitalThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
John F. de Groot, MDDepartment of Neuro-OncologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Yves A. DeClerck, MDDivision of Hematology-OncologyDepartment of Pediatrics and Department of Biochemistry and MolecularBiologyThe Saban Research Institute of Children’s Hospital Los AngelesLos Angeles, CA, USA
Department of MedicineCommittee on Clinical Pharmacology and PharmacogenomicsThe University of ChicagoChicago, IL, USA
Gianpiero Di Leva, PhDDepartment of Molecular Virology, Immunology and Medical GeneticsComprehensive Cancer CenterOhio State UniversityColumbus, OH, USA
Glenn Dranoff, MD, PhDDepartment of Medicine, Harvard Medical SchoolHuman Gene Transfer Laboratory Core, Dana-Farber Cancer InstituteBoston, MA, USA
Hua Fang, PhDDivision of Hematology-OncologyThe Saban Research Institute of Children’s Hospital Los AngelesLos Angeles, CA, USA
Department of MedicineCommittee on Clinical Pharmacology and PharmacogenomicsThe University of ChicagoChicago, IL, USA
Omotayo Fasan, MRCPDepartment of MedicineTemple University School of MedicinePhiladelphia, PA, USA
Department of Hematologic Oncology and Blood DisordersLevine Cancer InstituteCharlotte, NC, USA
Keith T. Flaherty, MDMassachusetts General Hospital Cancer CenterBoston, MA, USA
David Fogelman, MDDepartment of Gastrointestinal Medical OncologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Matthew D. Galsky, MDDivision of Hematology and Medical OncologyThe Tisch Cancer InstituteMount Sinai School of MedicineNew York, NY, USA
Guillermo Garcıa-Manero, MDDepartment of LeukemiaThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Benjamin A. Gartrell, MDDepartment of Medical OncologyMontefiore Medical CenterThe Albert Einstein College of MedicineBronx, NY, USA
Gabriel Ghiaur, MD, PhDThe Sidney Kimmel Comprehensive Cancer CenterThe Johns Hopkins University School of MedicineBaltimore, MD, USA
Michael C. Haffner, MDThe Sidney Kimmel Comprehensive Cancer Center and Brady UrologicalInstituteThe Johns Hopkins University School of MedicineBaltimore, MD, USA
Roy S. Herbst, MD, PhDDepartment of MedicineDivision of Medical OncologyYale Comprehensive Cancer CenterNew Haven, CT, USA
Ashley M. Holder, MDThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
David Hong, MDDepartment of Investigational Cancer TherapeuticsThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Jean-Pierre J. Issa, MDFels Institute for Cancer Research and Molecular BiologyTemple University School of MedicinePhiladelphia, USA
Elias Jabbour, MDDepartment of LeukemiaThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Nitin Jain, MDDepartment of LeukemiaThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Preetesh Jain, MD, DM, PhDDepartment of LeukemiaThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
List of Contributors ix
Filip Janku, MD, PhDDepartment of Investigational Cancer Therapeutics (Phase I Clinical TrialsProgram)Division of Cancer MedicineThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Milind Javle, MDDepartment of Gastrointestinal Medical OncologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Richard J. Jones, MDThe Sidney Kimmel Comprehensive Cancer CenterThe Johns Hopkins University School of MedicineBaltimore, MD, USA
Stan Kaye, MDThe Royal Marsden hospital and The Institute of Cancer ResearchLondon, UK
Samuel J. Klempner, MDDivision of Hematology/OncologyUniversity of California Irvine HealthOrange, CA, USA
Birgit Knoechel, MD, PhDBoston Children’s HospitalDana-Farber Cancer InstituteHarvard Medical SchoolBoston, MA, USA
Kensuke Kojima, MD, PhDSection of Molecular Hematology and TherapyDepartment of LeukemiaThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Scott Kopetz, MD, PhD, FACPDepartment of Gastrointestinal Medical OncologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Patricia Kropf, MDDepartment of MedicineTemple University School of MedicinePhiladelphia, PA, USA
Razelle Kurzrock, MDCenter for Personalized Cancer TherapyUC San Diego Moores Cancer CenterLa Jolla, CA, USA
Jens G. Lohr, MD, PhDDana-Farber Cancer InstituteBoston, MA, USA
Harvard Medical SchoolBoston, MA, USA
David Menter, PhDDepartment of Gastrointestinal Medical OncologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Funda Meric-Bernstam, MDDepartment of Investigational Cancer TherapeuticsInstitute for Personalized Cancer TherapyDepartment of Surgical OncologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Larissa A. Meyer, MD, MPHDepartment of Gynecologic Oncology and Reproductive MedicineThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Marcus M. Monroe, MDDepartment of OtolaryngologyUniversity of Utah School of MedicineSalt Lake City, UT
Guillermo Montalban-Bravo, MDDepartment of HematologyHospital Universitario La PazMadrid, Spain
Daniel Morgensztern, MDDepartment of MedicineDivision of Medical OncologyWashington University School of MedicineSt. Louis, MO, USA
Javier Munoz, MD, FACPDivision of Hematology/OncologyBanner MD Anderson Cancer CenterGilbert, AZ, USA
Andrea P. Myers, MD, PhDNovartis PharmaceuticalsCambridge, MA, USA
Jeffrey N. Myers, MD, PhDDepartment of Head and Neck SurgeryThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
William G. Nelson, MD, PhDThe Sidney Kimmel Comprehensive Cancer Center and Brady UrologicalInstituteThe Johns Hopkins University School of MedicineBaltimore, MD, USA
Barbara J. O’Brien, MDDepartment of Neuro-OncologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Susan O’Brien, MDDepartment of LeukemiaThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
William K. Oh, MDDivision of Hematology and Medical OncologyThe Tisch Cancer InstituteMount Sinai School of MedicineNew York, NY, USA
x List of Contributors
Shreyaskumar Patel, MDDepartment of Sarcoma Medical OncologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Saeed Rafii, MD, PhD, MRCPInstitute of Cancer SciencesThe University of Manchester and The Christie HospitalManchester, UK
Farhad Ravandi-Kashani, MDDepartment of LeukemiaThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Vinod Ravi, MDDepartment of Sarcoma Medical OncologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Jordi Rodon, MDEarly Clinical Drug Development GroupVall d’Hebron Institute of OncologyUniversitat Autonoma de BarcelonaBarcelona, Spain
Rabih Said, MD, MPHDepartment of Investigational Cancer TherapeuticsThe University of Texas MD Anderson Cancer CenterDepartment of Internal MedicineThe University of Texas Health Science CenterHouston, TX, USA
Allison C. Sharrow, PhDDepartment of PathologyJohns Hopkins University School of MedicineBaltimore, MD, USA
Department of Cancer Immunotherapeutics and Tumor ImmunologyBeckman Research InstituteCity of Hope Comprehensive Cancer CenterDuarte, CA, USA
Alexander C. Small, MDDivision of Hematology and Medical OncologyThe Tisch Cancer InstituteMount Sinai School of MedicineNew York, NY, USA
Sonali M. Smith, MDDepartment of MedicineThe University of ChicagoChicago, IL, USA
Anil K. Sood, MDDepartment of Gynecologic Oncology and Reproductive MedicineCenter for RNA Interference and Non-Coding RNADepartment of Cancer BiologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Richard M. Stone, MDDepartment of Medical OncologyDana-Farber Cancer InstituteBoston, MA, USA
Chad Tang, MDDepartment of Radiation OncologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Morgan Taylor, MDDepartment of Gynecologic Oncology and Reproductive MedicineThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Drew A. Torigian, MD, MA, FSARDepartment of RadiologyHospital of the University of PennsylvaniaPhiladelphia, PA, USA
Davis Torrejon, MDEarly Clinical Drug Development GroupVall d’Hebron Institute of OncologyUniversitat Autonoma de BarcelonaBarcelona, Spain
Apostolia-Maria Tsimberidou, MD, PhDDepartment of Investigational Cancer TherapeuticsThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Thanh-Trang Vo, PhDDepartment of Molecular Biology and BiochemistryUniversity of California IrvineIrvine, CA, USA
Julie M. Vose, MD, MBADivision of Hematology/OncologyUniversity of Nebraska Medical CenterOmaha, NE, USA
Saiama N. Waqar, MBBS, MSCIDepartment of MedicineDivision of Medical OncologyWashington University School of MedicineSt. Louis, MO, USA
Shiao-Pei Weathers, MDDepartment of Neuro-OncologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
James W. Welsh, MDDepartment of Radiation OncologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Shannon N. Westin, MD, MPHDepartment of Gynecologic Oncology and Reproductive MedicineThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Ofir Wolach, MDAdult Leukemia ProgramDepartment of Medical OncologyDana-Farber Cancer InstituteBoston, MA, USA
List of Contributors xi
Scott E. Woodman, MD, PhDDepartments of Melanoma Medical Oncology and Systems BiologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Srinivasan Yegnasubramanian, MD, PhDThe Sidney Kimmel Comprehensive Cancer Center and Brady UrologicalInstituteThe Johns Hopkins University School of MedicineBaltimore, MD, USA
Jian Q. (Michael) Yu, MD, FRCPCDepartment of Diagnostic ImagingFox Chase Cancer CenterPhiladelphia, PA, USA
W. K. Alfred Yung, MDDepartment of Neuro-OncologyThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Patrick A. Zweidler-McKay, MD, PhDChildren’s Cancer HospitalThe University of Texas MD Anderson Cancer CenterHouston, TX, USA
Series Foreword
While our knowledge of cancer at a cellular and molecular levelhas increased exponentially over the last decades, progress in theclinic has been more gradual, largely depending upon empirical tri-als using combinations of individually active anti-cancer drugs totreat the average patient. The challenge for the immediate future isto accelerate the pace of progress in clinical cancer care by enhanc-ing the bidirectional interaction between laboratory and clinic. Ournew understanding of human cancer biology and the heterogeneityof cancers at a molecular level must be used to identify novel targetsfor therapy, prevention, and detection focused on each individual.Barriers must be removed to facilitate the flow of targeted agents andfresh approaches from the laboratory to the clinic, while returningrelevant human specimens, images, and data from the clinic to thelaboratory for further analysis.
An Introduction to Translational Cancer Research providesa brief overview of current understanding of human cancer biol-ogy that is driving interests in targeted therapy and personalizedmanagement. Further development of molecular diagnostics shouldfacilitate earlier detection, more precise prognostication, and pre-diction of response across the spectrum of cancer development.Targeted therapy has already had a dramatic impact on severalforms of cancer and strategies are being developed to identifysmall groups of patients who would benefit from novel targeteddrugs in combination with each other or with more conventionalsurgery, radiotherapy, or chemotherapy. Development of personal-ized interventions—whether preventive or therapeutic in nature—will require multidisciplinary teams of investigators and the infras-tructure to match patient samples and agents in real time.
To accelerate translational cancer research, greater alignment willbe required between academic institutions, the National CancerInstitute, the Food and Drug Administration, foundations, pharma,and community oncologists. Ultimately, new approaches to preven-tion, detection, and therapy must be sustainable. In the long run,translational research and personalized management can reduce thecost of cancer care, which has escalated in recent years. More accu-rate and specific identification of at-risk members and risk strati-fication will be helpful to minimize the risks of over-diagnosis andover-treatment, while maximizing the benefits of screening, earlydetection, and preventive intervention. Patients who would bene-fit most can be identified and funds saved by avoiding treatment inthose whose cancers would not respond. Participation and educa-tion of community oncologists will be required, as will modifica-tion of practice patterns. For progress in the clinic to occur at anoptimal pace, leaders of translational teams must envision a clearpath to bring new concepts and new agents from the laboratory to
the clinic, to complete pharmaceutical or biological development, toobtain regulatory approval and to bring new strategies for detection,prevention, and treatment to patients in the community.
In a series of additional volumes regarding translational cancerresearch, several topics are explored in greater depth, includ-ing gene therapy by viral and non-viral vectors, biomarkers,immunotherapy, and this volume concerning targeted therapy.The purpose of these books has been not only to describe differentstrategies for controlling particular forms of cancer but also toidentify some of the barriers to translation using different reagentsor different strategies around common therapeutic or diagnosticmodalities. Potential barriers are many and include the need fora deeper understanding of science, methods to overcome thechallenge of tumor heterogeneity, the development of targeted ther-apies, the availability of patients with an appropriate phenotype andgenotype within a research center with the investigators, researchteams and infrastructure required for clinical/translational researchand the design of novel trials, adequate financial support, a viableconnection to diagnostic and pharmaceutical development, and astrategy for regulatory approval as well as for dissemination in thecommunity.
Targeted Therapy in Translational Cancer Research considersmany of these areas. Principles are beginning to emerge for iden-tifying therapeutic targets. A critical issue is how best to combinetherapies against different targets within the same cancer if we areto develop effective personalized care. Tumor initiating cells mustbe eliminated as well as their progeny. Not only the cancer cells buttumor vessels and microenvironment can be targeted. While a sepa-rate volume will consider immunotherapy, a chapter on principlesof immunotherapeutic targeting has been included, because of therapid progress in this area. A better understanding of epigeneticand miRNA regulation has suggested new approaches to targetedtherapy. The current status of targeted therapy for individual hema-tologic neoplasms and solid cancers has been reviewed extensively.As several major molecular targets and signaling pathways—TP53,PARP, Met, Kit, PI3K, and Ras/MAP—are important to cancersat multiple sites, chapters have also been devoted to strategiesfor their inhibition. Overall, this volume includes substantialperspective regarding the translational potential of targeted therapythat should provide useful information for investigators andclinicians.
Robert C. BastMaurie Markman
Ernest Hawk
xiii
Foreword
As a busy clinical oncologist/hematologist striving to be currentin preparing to see a new patient, or as a clinical investiga-tor trying to determine what might be the best new approachfor the patient with advanced refractory cancer, this volumetitled Targeted Therapy in Translational Cancer Research canbe of enormous help. In addition, for the young or experiencedbench scientist this offering gives both a basic background andan important grounding in the current field of clinical targetedtherapies.
The above comments should come as no surprise given the deepexperience of the editors and the contributors to this volume.
In a review of the parts of this volume there is excellent cover-age of (a) the principles of targeted therapies; (b) specific targetedtherapies in the hematologic and solid malignancies; and (c) cover-age of targeted therapies for specific molecular aberrations. As one
drills down into the well-written individual chapters there is excel-lent coverage of the current state of the art plus meaningful coverageof how the field is evolving.
It is gratifying to see chapters on functional imaging, the issueof combining targeted therapies, targeted immunotherapies, themicroenvironment, microRNAs, and tackling tough targets such asRas and TP53.
The chapter on specific organ types of cancer are both practicalfor us to catch up on the best treatments and yet comprehensiveenough to see how the treatments are evolving.
All in all, this is a wonderful volume to aid all of us in a practicaland deeper understanding of targeted therapies. Congratulations tothe contributors and editors of this special volume.
Daniel D. Von Hoff, MD, FACP
xv
Preface: Bench to Bedside and Back
In the last decade, emergent technologies have enhanced our under-standing of genomic, transcriptional, proteomic, epigenetic, andimmune mechanisms in carcinogenesis. This improved understand-ing has enabled the development of targeted cancer therapies andtransformed conventional treatment paradigms; it has provided theframework for the discovery of new targets, for validation of novelagents, for combination therapies predicated upon scientific ratio-nale, and for clinical trials that have already markedly improvedthe prognosis and outcome of patients with cancer. Excitingly, theimplication of immunomodulatory targets in carcinogenesis hasled to the development of new promising drugs based on the cen-tral principle that breaking tolerance using immune checkpointblockers can achieve durable responses. Moreover, although dis-tinct pathways are initially analyzed independently, interdepen-dent and compensatory mechanisms have derived innovative com-binations of targeted, immunomodulating, antiangiogenic, and/orchemotherapeutic agents, which have additive or synergistic cyto-toxicity and can overcome resistance to conventional therapies.Continued progress will require improved genomic classification ofthe various tumor types, delineation of the mechanisms of resistanceto treatment and disease progression, and improved understandingof metastasis, ultimately allowing for provision of therapies designedto target tumor heterogeneity early in the disease course.
This edition of Targeted Therapy in Translational CancerResearch for the Translational Oncology series provides a compre-hensive overview of recent developments in our understanding oftumor biology, elucidates the roles of targets and pathways involved
in carcinogenesis, and describes current state-of-the-art anticancertherapy, as well as the most promising areas of translational researchand targeted therapy. Basic principles of targeted therapy, includ-ing immunotherapy and the roles of cancer stem cells, the microen-vironment, angiogenesis, epigenetics, microRNAs, and functionalimaging in precision medicine, are highlighted. Major advances inthe therapeutic management of hematologic malignancies and solidtumors using conventional therapy, targeted therapy, immunother-apy, or novel treatment modalities are summarized. Importantly,advances in technology and bioinformatic analyses of complex datahave already allowed for improved characterization of tumor biol-ogy, function, and dynamic tumoral changes over time, therebyallowing for improved cancer diagnosis, prognosis, and therapy.
We are on the threshold of translating discoveries in cancer biol-ogy into unprecedented durable responses and improved clinicaloutcomes in the majority of patients with cancer. In this uniquetime in history, the discovery of novel therapeutic approaches tar-geting the molecular basis of cancer will allow for implementa-tion of precision medicine, with the promise of potentially cura-tive, well-tolerated therapies. Targeted Therapy in TranslationalCancer Research was written to increase the awareness and accessof basic and clinical researchers, caregivers, and patients alike tocutting-edge “bench to bedside and back” breakthroughs, whichhave transformed the diagnosis, prognosis, and treatment of cancer.
Apostolia-Maria TsimberidouKenneth C. Anderson
xvii
I PART I
Principles of TargetedTherapies
1 CHAPTER 1
Toward Personalized Therapy for CancerAshley M. Holder1 and Funda Meric-Bernstam2,3,41The University of Texas MD Anderson Cancer Center, Houston, TX, USA2Department of Investigational Cancer Therapeutics, Houston, TX, USA3Institute for Personalized Cancer Therapy, Houston, TX, USA4Department of Surgical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
Introduction
Personalized cancer care is grounded in the principle that thepatient’s genotype and molecular characterization of a tumor andits microenvironment can identify the most effective cancer man-agement for each patient while reducing toxicity. By tailoring ther-apy to a specific tumor, the approach of personalized cancer therapyis expected to save critical treatment time and healthcare costs byavoiding the selection of less beneficial therapies. Thus, the objec-tive of personalized cancer therapy is to harvest information aboutthe tumor—its DNA, RNA, proteins, and metabolism—within thecontext of the tumor microenvironment and the patient’s genotypeto inform treatment decisions. However, much work remains to bedone before this concept can be translated from the research envi-ronment to the clinical setting.
Several complementary components are necessary to achievepersonalized medicine throughout the continuum of cancer care(Figure 1.1). The first phase of personalized cancer care includes riskassessment, in order to identify patients at higher cancer risk, appro-priately modifying screening strategies and frequency, and offeringpreventive strategies. Once a cancer diagnosis is made, the care ofthe patient enters the second phase of personalized care—molecularcharacterization of the tumor to assess patient prognosis. Accord-ingly, patients at a high risk of recurrence can receive more inten-sive therapy, while patients at low risk may receive less toxic systemictherapy or may avoid additional therapy altogether.
The third phase in personalized care involves in-depth molecu-lar characterization of the tumor to identify potential therapeutictargets and to test for established and putative predictive markers,that is, markers predictive of response to specific therapies. Mark-ers predictive of adverse events can be used to select regimens withthe least toxicity. Early response to therapy may be monitored withpharmacodynamic markers of response.
Furthermore, as efficacy of treatment for recurrent diseaseimproves, a growing role for biomarkers is likely to develop inmonitoring early recurrence and providing a personalized programfor survivorship. Although currently standardized follow-up sched-ules based on cancer histology and stage exist for most cancertypes, more precise determination of expected prognosis (i.e., likeli-hood of recurrence) based on molecular subtype would personalize
cancer follow-up, including the frequency of follow-up visits and theneed for specialist follow-up. As many cancer treatments have long-term unintended effects, personalized survivorship programs canoffer more intensive screening for patients at higher risk of devel-oping these side effects.
Personalized Targeted Therapy
Principles of Molecular TherapeuticsEven in therapy-sensitive cancers such as breast cancer, only a sub-group of cancer patients achieve a pathologic complete responsewith currently available standard chemotherapy, underscoring theneed to develop novel targeted therapies.1, 2 Therefore, an importantcomponent of personalized therapy is the delivery of individualized“targeted” therapy, directed toward molecular aberrations in spe-cific tumors. The principle of molecular therapy is to target molec-ular differences between cancer cells and normal cells. To imple-ment molecular therapeutics, targets must first be identified usinggenomic and proteomic techniques. Notably, numerous differencesexist between cancer cells and normal cells; differentiating betweencancer “drivers” that play a key role in cancer progression and sur-vival and “passengers” that are present but not critical for cancermaintenance is a challenging but surmountable component, criticalto the success of targeted therapies. Extensive preclinical studies areneeded for functional characterization of the effect of specific genealterations on cancer initiation and progression and cancer cell sur-vival. The ideal target is usually differentially expressed or activatedin cancer cells conferring cell growth and survival advantage. Thus,target inhibition induces cancer cell cytostasis or, more preferably,cancer cell death.
Predictors of Response for Patient SelectionIn addition to the need for compelling therapeutic targets, drugsthat inhibit the identified targets, ideally through selective inhibi-tion, are necessary to minimize off-target toxicity. Biomarkers todetect the presence of the target within the tumor are employed toselect patients who will benefit from the targeted therapy. Often, thepresence of the target is pursued as a potential predictive marker;however, expression of the target itself may not be sufficient to
Targeted Therapy in Translational Cancer Research, First Edition. Edited by Apostolia-Maria Tsimberidou, Razelle Kurzrock and Kenneth C. Anderson.© 2016 John Wiley & Sons, Inc. Published 2016 by John Wiley & Sons, Inc.
3
4 PART I Principles of Targeted Therapies
Risk reduction
Early diagnosis
Avoid toxicity of adjuvant
therapy
More intensive adjuvant therapy
Least toxic therapy
Most efficacious
therapy
Personalized cancer riskassessment
Cancer preventioncancer genetics
Cancer epidemiologypathology
Surgeryradiationchemotherapytargeted therapyimmunomodulation
Predictive markers
Pathologymolecular diagnosticscancer geneticsinvestigational therapeutics
Immunologyoutcomes researchcomparative effectiveness
Prognostic markers
Clinical oncology pathologyradiology
Molecular diagnosticscancer geneticsimmunology
Assess treatment response
1
2
3
Personalized survivorship
program
Follow-upside effectssecond malignancies
Detect earlyrecurrence
Biomarker screening
Preventino for
high-risk patients
Low-risk for recurrence
Predictors of toxicity
High-risk forrecurrence
Intensive screeningfor high risk patients
Predictors ofresponse/resistance
Figure 1.1 The cancer care continuum for personalized medicine.
confer sensitivity to a therapy. For example, in colorectal cancerEGFR expression assessed by immunohistochemistry (IHC) is notconsidered to be a reproducible marker of sensitivity to cetux-imab.3 In contrast, patients with colorectal cancers bearing mutatedK-ras have reproducibly been shown not to benefit from cetuximab,whereas patients with tumors bearing wild-type K-ras do bene-fit.4–6 As this example demonstrates, predictive markers of response,sensitivity, and resistance must be carefully developed. Even still,most clinically approved targeted therapies have low rates of objec-tive tumor response in single agent therapy. Furthermore, predic-tive markers of response and clinical benefit remain elusive. Thus, itis important that extensive preclinical modeling to identify markersof response and resistance be performed early in drug development.For targeted therapies with strong rationale for predictive markers,trials can be conducted in patients selected for or enriched for cer-tain markers.
Pharmacodynamic Markers of ResponseEarly in drug development, pharmacodynamic markers of biolog-ical effect must be discovered to determine whether the putativetarget is inhibited by the novel therapeutic agent and to measurethe extent of target inhibition and downstream signaling inhibition.Biological inhibition of the target can be assessed in surrogate tis-sue samples, such as skin biopsies, hair follicles, peripheral bloodmononuclear cells, or platelets. However, ultimately there is value
added in determining the effect of the drug on tumor cells by obtain-ing pre-treatment and on-treatment biopsies.
Another important goal for molecular therapeutics is the devel-opment of early biomarkers of response. The traditional approachto assessing response in clinical trials has been to treat patients fortwo to three cycles and then evaluate treatment response with repeatimaging. However, with the implementation of targeted therapies,the discovery of pharmacodynamic markers of response that canassess response earlier would spare patients from unnecessary tox-icity, save the healthcare system the cost of administering ineffec-tive therapy, and facilitate the transfer to alternate therapeutic regi-mens without further disease progression. Through assessment ofbiomarkers pre-treatment and on-treatment with repeat biopsies,pharmacodynamic markers of response within the tumor can beexamined after only one cycle of therapy or even earlier; likewise,the biopsy assessment would permit correlation with radiographicresponse or clinical benefit on standard response assessment. Inaddition, an on-treatment biopsy can provide further informationabout adaptive responses to the current treatment. This insight canassist in planning future studies of rational combinatorial therapy.An area yet to be explored is the use of individual adaptive responsesto personalize combination therapies chosen.
Although obtaining pre- and on-treatment biopsies to assesspharmacodynamic markers of response is theoretically appeal-ing, this process presents several challenges. One barrier to early
CHAPTER 1 Toward Personalized Therapy for Cancer 5
assessment of treatment response is that measurement of target inhi-bition itself may be difficult. Pathway activation is often determinedthrough assessment of phosphorylation of downstream mediators,and phospho-specific residues are known to be relatively unstable.7, 8
The acquisition of a biopsy may also change the readout of the path-way and cell proliferation. Cold ischemia time and intratumoral het-erogeneity of the specimen can alter the measurable targets withinthe sample. There are no widely accepted approaches for quanti-tative assessment of downstream signaling; though IHC, reverse-phase protein array (RPPA), enzyme-linked immunosorbent assay(ELISA), and bead-based multiplex proteomics are all currently uti-lized, they have limitations. To minimize variability in assessmentof treatment response, researchers and clinicians must collaborateto optimize and standardize specimen collection and assay selectionfor each desired target within a tissue type.
Another valid concern is the significant cost added to clinicaltrials by pre- and on-treatment biopsies. Further, these biopsiesintroduce additional problems, such as biopsy quality and poten-tial increased morbidity. Despite the increasing number of early tri-als incorporating biopsies for pharmacodynamic assessment, only asmall fraction of phase I trials that included biomarkers made useof the biomarker results for dose selection.9 Some have proposedthat if the drug does not show preliminary evidence of antitumorefficacy in an early trial, the biopsies will be uninformative. How-ever, even if antitumor efficacy is not observed, pharmacodynamicassessment may serve other important roles, such as determiningwhether there was lack of or insufficient target inhibition. Theseresults could uncover the need to modify treatment dose or sched-ule. Furthermore, if there were inhibition of target but with inade-quate treatment response, the information gathered from the biopsycould suggest that the target may not be the primary driver in thattumor type or that there may be alternate resistance mechanismswithin the tumor.
Early Successes in Personalized TherapyDespite the challenges to biomarker selection, targeted therapydevelopment, and treatment response assessment, the field ofpersonalized cancer therapy has generated early successes incor-porating biomarkers and targeted therapies to transform cancertreatment.
Prognostic Stratification and Predictionof Chemotherapy Benefit in HormoneReceptor-Positive Breast Cancer
Several RNA-based prognosticators have recently been developed.Two of these commercially available multi-marker assays for breastcancer prognostication are notable as they are widely utilized. Intwo independent analyses of phase III clinical trials, one in node-negative and one in node-positive breast cancer with tamoxifen-alone control arms, the Oncotype Dx (Genomic Health) RT-PCR-based 21-gene recurrence score was shown to identify a group ofpatients with low recurrence scores, who do not appear to benefitfrom chemotherapy and a second group, with high scores, who dobenefit from chemotherapy. The role of chemotherapy in breast can-cer patients with hormone receptor-positive, node-negative, inter-mediate recurrence score tumors and hormone receptor-positive,node-positive, low and intermediate recurrence score tumors isbeing assessed prospectively in the TAILORx and RxPONDER stud-ies, respectively. In a non-randomized clinical setting, the Mam-moprint 70-gene signature was shown to be prognostic in node
negative and 1–3 node positive patients and to predict chemother-apy benefit in the high-risk group.10–12 The Mammoprint is beingprospective validated in the large adjuvant MINDACT (Microar-ray In Node-negative Disease May Avoid ChemoTherapy) clinicaltrial. For both Oncotype and Mammoprint assays, the discordancerates between the assay prediction and clinical-pathologic risk cat-egories are approximately 30%. Clinical utility studies demonstratethat assay use results in a change in treatment decision in 25–30% ofcases, most commonly from chemo-endocrine therapy to endocrinetherapy alone.13 The widespread use of these tools in clinical prac-tice suggests that clinicians not only are seeking prognostic toolsto assist in counseling patients and treatment planning but also arewilling to modify their clinical practice to incorporate new techno-logical adjuncts.
HER2-Targeted Therapy in Breast Cancer
Twenty percent of breast cancers display HER2 amplification, whichis associated with a poorer prognosis compared to those withoutHER2 overexpression.14, 15 However, treatment of these breast can-cers in the adjuvant setting with trastuzumab, a monoclonal anti-body targeting the extracellular domain of the protein encodedfor by HER2, has been shown to improve survival in both earlystage and metastatic breast cancers with HER2 amplification.16–19
This initial success was rapidly followed by development of addi-tional anti-HER2 therapies such as lapatinib, pertuzumab, andT-DM1. Even still, many HER2-positive tumors do not respond toHER2-targeted therapy, suggesting that additional biomarkers areneeded to predict intrinsic resistance and the emergence of acquiredresistance.
BRAF Inhibitors in BRAF Mutant Melanoma
B-Raf is a member of the Raf kinase family of serine–threoninekinases that activates the MAP/ERK signaling pathway. Mutationsin BRAF have been detected in 40–60% of melanomas.20, 21 Lessthan a decade after the development of the RAF inhibitor vemu-rafenib, a phase III randomized trial confirmed that the BRAFV600E mutation was a response-specific predictive biomarker fortreatment of melanoma with vemurafenib. Patients with therapy-naıve metastatic melanoma harboring the BRAF V600E mutationhad significantly longer progression free and overall survival whentreated with vemurafenib compared to standard chemotherapy.22
The rapid clinical development of B-Raf inhibitors exemplifies howmolecular identification of a driver aberration can be rapidly trans-lated into a clinically effective therapy. However, in spite of theimpressive response rates (48% for vemurafenib compared with5% for dacarbazine), responses were short-lived, demonstrating theneed for combinatorial therapy with other drugs or immunotherapyto obtain durable responses.
Strategies for Comprehensive MolecularCharacterization
With the advent of high-throughput technologies, interest in theutilization of multimarker technologies to assist in tumor molecu-lar classification and selection of optimal personalized therapy hasintensified. A brief summary of strategies commonly utilized forcomprehensive molecular characterization is provided below and inTable 1.1.
6 PART I Principles of Targeted Therapies
Table 1.1 Strategies for comprehensive molecular characterization.
Technology Detection target
Tissue
requirement Advantages Disadvantages
DNA
Hot spot mutation
testing
Single nucleotide
variations
Blood, fresh/frozen
tissue or FFPE
Minimal DNA required
Cost effective
High throughput
Limited to hot spot mutations
assayed
Targeted gene
sequencing
Mutations in
candidate genes
Fresh/frozen tissue
or FFPE
Complete sequencing of
open reading frames
Larger amount of tissue
required
Must differentiate germline
SNPs from somatic
mutations
Limited gene panel
Whole exome and
genome sequencing
Mutations Fresh or
high-quality
frozen tissue
Valuable target discovery
Comprehensive
FFPE not optimized
Predicting functional impact
of mutation
Expensive
DNA methylation
screening
Methylation Blood, fresh or
high-quality
frozen tissue
High throughput High-quality frozen material
required
RNA
Quantitative PCR Relative gene
expression
Blood, fresh or
high-quality
frozen tissue
Monitor treatment effect on
pathway expression
Quantitates relative to
housekeeping gene
Microarray-based gene
expression profiling
Relative mRNA or
miRNA expression
Blood, fresh or
high-quality
frozen tissue
Monitor treatment effect on
pathway expression
High throughput
Cost effective
Reproducibility of results due
to sample preparation and
type of platform
RNA sequencing Absolute RNA
abundance,
splicing variants,
mutations, fusions
Fresh/frozen tissue Monitor treatment effect on
pathways
High throughput
Base pair resolution
“Reads” are proxies for mRNA
abundance
Reproducibility monitoring
Need to reconstruct short
“reads”
Ribosome footprinting Quantitate
expression
Fresh/frozen tissue Information on protein
abundance regulation
Does not quantitate proteins
but only translation
efficiency
Protein
Stable isotopic labeling
with amino acids in
cell culture (SILAC)
Relative protein
concentration
Fresh/frozen tissue
or FFPE
High throughput
High accuracy and sensitivity
Isotopic labeling may not be
feasible
High-resolution tandem
mass spectrometry
Absolute protein
quantification
Blood, fresh/frozen
tissue or FFPE
Cost effective Dependent on calibration or
reference standards
RPPA Relative protein
expression and
activation
Blood, fresh/frozen
tissue or FFPE
Cost effective
High throughput
Proteins must have
high-quality antibodies
available
Immunohistochemistry Relative protein
expression and
activation
Fresh/frozen tissue
or FFPE
Tissue morphology
Intratumoral location
Proteins must have
high-quality antibodies
available
Low throughput
Larger amount of sample
required
Metabolomics Metabolite
expression and
pathway activation
Blood, urine, or
fresh/frozen
tissue
High throughput Sample harvest conditions can
alter results
CHAPTER 1 Toward Personalized Therapy for Cancer 7
Genomic ProfilingMuch of the effort in biomarker discovery for personalized cancertherapy has been directed at genomic markers, in part because oftargeted therapies entering the market with DNA-based predictivemarkers, such as BRAF V600E as a predictor of response to BRAFinhibitors, and also because of the recent advances allowing multi-plex genomic testing to be performed in a rapid, reproducible, andrelatively cost-effective manner.
Recently, several high-throughput genotyping methods havemoved into the Clinical Laboratory Improvement Amend-ments (CLIA) environment including the MassARRAY System(Sequenom), SNaPshot technology (Applied Biosystems), and ionsemiconductor sequencing (Ion Torrent Technology). Multiplexhot spot mutation testing, also referred to as high-throughputSNP genotyping, has many advantages: requiring minimal DNA,accommodating formalin-fixed paraffin-embedded (FFPE) tissue,processing multiple samples simultaneously, detecting mutationspresent in a small proportion (5%) of cells, and being relativelycost-effective. However, this technique is limited to evaluating onlythe hot spot mutations being assayed. Hot spot genotyping does nothave the capability to provide full coverage of all tumor suppressorgenes, to detect new mutations in known cancer-related genes, orto discover novel cancer-related genes.
In addition to high-throughput SNP genotyping, targetedsequencing has recently become available in the CLIA environ-ment. Target enrichment allows for selective capture of genomicregions of interest (usually exomes) and subsequent sequencing ofcancer-relevant genes, including actionable targets, and commonmutations. This technique has several benefits: complete sequenc-ing of genes, including tumor suppressor genes; directing analyticalresources to the most relevant genes in a select panel (e.g., 200–400);and accommodating FFPE tissue. The drawbacks of targeted exomesequencing are the larger amounts of tissue required, the need todifferentiate germline SNPs from somatic mutations, and the limita-tions of novel gene discovery resulting from the limited gene panel.Sequencing alone will also not capture other critical alterations suchas epigenetic changes.
As the cost for whole exome sequencing (WES) and wholegenome sequencing (WGS) has decreased, the utility of theseapproaches in personalized cancer therapy is now being explored.The advantage of these techniques is the comprehensive genomicanalysis of the tumor, yielding mutational, gene copy number, andrearrangement data. This complete examination can detect changesresulting in oncogene activation or tumor suppressor gene inactiva-tion, perhaps uncovering alterations in the exome or genome thatare essential for the maintenance of the malignant phenotype. Inaddition, the genomic data harvested from WES and WGS can aidin the development of novel targeted therapies, and assist in theselection of currently available treatments likely to be most effec-tive. However, the minimum quantity of DNA required is signifi-cantly greater than other genomic techniques, and WES/WGS anal-ysis of FFPE samples is only being optimized now. Even still, WESand WGS are prone to high rates of false positive and false nega-tive calls, especially in samples with low tumor cellularity, neces-sitating validation with additional technologies. Possible solutionsto these concerns are creating a standardized algorithm for callingsingle nucleotide variants (SNVs) and stringent standards to assessthe reliability of calls in the CLIA environment. Despite the devel-opment of tools to assist in calling SNVs, hurdles to incorporat-ing WES and WGS in personalized cancer therapy involve predict-ing the functional impact of every mutation and prioritizing each
mutation as a driver or passenger. In addition, the large amount ofdata generated from WES/WGS creates considerable challenges tothe capacity and security of information storage, as well as to thetimely turnaround of bioinformatic analysis.
Next-generation targeted sequencing and WES/WGS approachesalso have the advantage of providing information on DNA copynumber. Other high-throughput technologies being pursued toassess copy number alterations include comparative genomichybridization, single nucleotide polymorphism arrays, digital kary-otyping, and molecular inversion probes.23, 24
Epigenetic ProfilingGenomic technologies can detect genetic alterations that yieldresponse-predictive biomarkers; however, the frequency of muta-tions in some cancers is quite low. An alternative to mutationalanalysis is epigenetic or DNA methylation screening. Epigeneticprofiling of immortalized cancer cell lines can uncover associationsbetween methylated genes and therapeutic sensitivity. Inactivationof DNA mismatch repair genes can be assessed using epigenetictools that can then provide prognostic stratification for clinicalapplication, such as CpG island methylator phenotype (CIMP) incolorectal cancer.25 In addition, methylation screening can detectactivation of oncogenic signaling through the silencing of pathwaysignaling regulators. To detect mechanisms of resistance, methy-lation screening of a tumor pre- and post-treatment can identifyepigenetic changes following chemotherapy that may alter the anti-tumor efficacy of other agents, such as the use of the methylatingagent temozolomide based on the methylation status of the MGMTpromoter in glioblastoma.26, 27
Transcriptional ProfilingTo produce an individualized signature of a patient’s tumor, geneexpression profiling utilizes mRNA, microRNA, and non-codingRNA. This unique transcriptome can then be used for classificationof unique molecular subtypes, prognostic assessment, and to predicttherapeutic responsiveness of tumors. In addition, transcriptionalprofiling of cancers before and after neoadjuvant systemic therapycan provide crucial information about the effect of treatment on theregulation of pathways and biological processes, potentially reveal-ing new targets for therapy.28
Other technologies such as exon junction arrays and genometiling arrays use probes to the expected splice sites for each gene,thus allowing detection of splicing isoforms. As interest in massiveparallel sequencing of RNA (RNA-seq) has intensified, RNA-basedtechnologies are continuing to evolve rapidly. Compared with tra-ditional transcriptional profiling with microarray technology, RNA-seq has the ability to detect other abnormalities in the cancer tran-scriptome in addition to changes in RNA expression, includingalternative splicing, novel transcripts, and gene fusion.29 Further-more, RT-PCR-based multiplex assays, such as Oncotype Dx, thatassess expression of selected RNA panels are likely to have sustainedutility.
Proteomic ProfilingIHC is a well-validated tool to assess therapeutic biomarkers, suchas the estrogen receptor in breast cancer. However, IHC has limita-tions as a low-throughput technology requiring larger amounts ofsample and considerable clinical manpower and expense to processand interpret each biomarker of interest. An advantage of IHC isits visualization of the protein of interest within the tumor, provid-ing information about intratumoral location and tissue morphology.
8 PART I Principles of Targeted Therapies
The development of a multiplex method for IHC could transportthis worthwhile and validated tool into the realm of personalizedoncology.
Other assays, such as ELISA, and new-generation assays, such asbead-based multiplexed proteomic assays, can allow for assessmentof a panel of proteins but still present challenges regarding not onlylinear range and challenges in absolute quantitation but also scal-ability to a large sample set. Mass-spectrometry-based proteomicsremains a powerful discovery tool. RPPA is a protein array designedthat allows the measurement of protein expression levels in a largenumber of biological samples simultaneously in a quantitativemanner. Briefly, lysates from cell lines, tissue lysates, or biologicalfluids can be spotted onto reverse-phase protein microarrays andprobed with a panel of high-quality, monospecific antibodies. RPPAis a relatively cost-effective, high-throughput method to identifycancer subtypes, resistance biomarkers, and functional pathways.One drawback of RPPA is that it is limited to proteins for whichhigh-quality antibodies are available. Given that most biomarkersand drug targets are proteins, proteomics has an advantage overtranscriptional profiling for monitoring therapeutic response,discovering novel targets, and exposing mechanisms of pathwayresistance in a personalized manner. The proteomic signature canalso guide treatment selection by stratifying tumors into molecularcategories and allowing the clinical team to incorporate the mostefficacious therapy into the treatment plan. Further, RPPA is mainlyutilized through comparison of a sample with other samples in a set.Approaches to normalize expression of a sample compared to con-trols are needed to transition this approach from a discovery tool toa point-of-care assay. Large-scale validation of proteomic signaturesas well as proteomic platforms must be achieved before proteomicprofiles can enter wide-spread clinical use.
MetabolomicsMetabolomics utilizes mass spectrometry, nuclear magnetic reso-nance, and gas and liquid chromatography to reveal small moleculemetabolites and metabolic pathway alterations essential for themaintenance of the malignant phenotype. Metabolic screening canalso aid in the early detection of cancers, especially those for whichscreening is difficult, by assessing biomarkers not only in tumortissues but also in patients’ body fluids. Alterations in mitochon-drial metabolism, a characteristic of invasive cancer, can differen-tiate malignancy from normal tissues. In vivo metabolic screeningfor staging and monitoring cancer is achieved through PET imag-ing technology that capitalizes on the increased metabolic activityof malignant cells to expose residual disease and metastases. Alter-ations in imaging may also occur with different tumor subtypes. Forexample, mutations in the isocitrate dehydrogenase (IDH) genes arefrequently found in gliomas. This results in the production of anoncometabolite, 2-hydroxyglutarate (2-HG), which can be detectednoninvasively in gliomas with IDH mutations using magnetic reso-nance spectroscopy.30
Integrated Multi-analyte AnalysisWith increasing access to high-throughput technologies and theability to perform assays on smaller amount of tissues, the fron-tier of multi-analyte analysis is expanding to incorporate DNA,RNA, and protein data, to perform integrated analysis for bettermolecular classification of tumors, and to identify the most suit-able actionable targets. Although such a systems biology approachmay indeed be the future of personalized medicine, integrated anal-ysis of high-throughput sequencing is still in the pilot stages.31, 32
Greater emphasis on big data, and sharing of clinically annotatedhigh-throughput data, is likely to improve predictive algorithms.
A Personalized Approach to InvestigationalTherapy Selection
Increasingly, genomic characterization directs patients to specificclinical trials targeting the aberrant gene product or downstreamsignaling pathway. Routine comprehensive testing of patients withadvanced disease could facilitate faster delivery of effective therapiesto patients, while enriching for patients with matched aberrationsin targeted therapy trials and accelerating accrual in those trials.Comprehensive testing is of the greatest value to patients who areinterested in participating in clinical trials, are potentially eligiblefor therapeutic trials, and are able to access a menu of activelyaccruing targeted therapy trials. Unfortunately, most patientshave limited access to pathway-matched investigational targetedtherapies due to lack of relevant trials or lack of availability in earlyclinical trials.
Standardized molecular testing in the CLIA environment canfacilitate a variety of phase II clinical trial designs (Figure 1.2).In one commonly used approach, a biomarker can be used forpatient selection for treatment with an agent targeting that alter-ation or the downstream pathway activated by that alteration (Fig-ure 1.2a), or for randomization between standard of care or the tar-geted therapy (Figure 1.2b). Alternately, the biomarker can be usedfor prospective stratification, but all patients may be treated in asingle arm study with the targeted therapy (Figure 1.2c) or ran-domized to targeted therapy or standard therapy (Figure 1.2d). Inumbrella trials, patients are allocated to one of several treatmentsbased on their biomarker profile (Figure 1.2e). In adaptive trials,patients are initially randomly allocated but subsequently allocatedby biomarkers linked to therapeutic approaches discovered in theinitial part of the treatment (Figure 1.2f) (e.g., ISPY 2, BATTLE tri-als 20).33–38 One common strategy is to match patients to trials basedon a chosen molecular aberration, such as enrolling patients withPIK3CA-mutant breast cancer in trials that have the presence of aPIK3CA mutation as an eligibility criterion (Figure 1.2a). However,biomarker assessment can also enrich trials without biomarker eligi-bility criteria through the enrollment of patients with pathway aber-rations, for example, matching patients with PIK3CA mutationsin trials with agents targeting the PI3K pathway. Ultimately, thistechnique may enhance the clinical benefit achieved for enrolledpatients, as supported by a study in which matching treatment deliv-ered to patient tumor genotype improved response rate even in earlyclinical trials.39 However, it should be noted that the response andclinical benefit rates observed in this study may not be representa-tive of the general population due to small sample size and lack ofrandomization.
There is also increasing interest in treating individual patientsthrough off-label use of drugs targeting an aberration approvedfor another indication or compassionate use of agents in clinicaltrials that do not “fit” the patient characteristics. Cancer centersneed to determine ways to facilitate treatment of these patients on“N-of-1” studies (Figure 1.2g).40–42 “N-of-1” trials use clinicopatho-logic molecular characteristics to select an individualized therapyplan. The most significant challenge is in quantifying benefit of an“N-of-1” effort; the test is of the process and not the individualbiomarker/drug pairing. Clinical benefit has been suggested in “N-of-1” settings by measuring time-to-progression on the trial drugcompared to time-to-progression on the most recent treatment.43, 44
CHAPTER 1 Toward Personalized Therapy for Cancer 9
x
No x x
No x
X-t
arg
ete
d th
era
py
Y-t
arg
ete
d th
era
py
Z-t
arg
ete
d th
era
py
Sta
nd
ard
of p
ractice
(a)
Bio
mar
ker-
sele
cted
sin
gle
arm
tri
al(e
) B
iom
arke
r-se
lect
ed u
mb
rella
tri
als
(f) A
dap
tive
ran
do
miz
atio
n t
rial
s
(g)
N-o
f-1
tria
l
(b)
Bio
mar
ker-
sele
cted
ran
do
miz
ed t
rial
xX
-targ
ete
d t
hera
py
X-t
arg
ete
d th
era
py
Y-t
arg
ete
d th
era
py
Z-t
arg
ete
d th
era
py
X-t
arg
ete
d t
hera
py
X-t
arg
ete
d t
hera
py
X-t
arg
ete
d t
hera
py
X-t
arg
ete
d t
hera
py
Sta
ndard
of
pra
ctice
Sta
ndard
of
pra
ctice
Sta
ndard
of
pra
ctice
No x
Ob
jective
re
sp
on
se
ra
te,
pro
gre
ssio
n-f
ree
su
rviv
al
Ob
jective
re
sp
on
se
ra
te,
pro
gre
ssio
n-f
ree
su
rviv
al
Ob
jective
re
sp
on
se
ra
te,
pro
gre
ssio
n-f
ree
su
rviv
al
Obje
ctive
re
sp
on
se
ra
te,
pro
gre
ssio
n-f
ree
su
rviv
al
x
No x
(c)
Bio
mar
ker-
stra
tifie
d s
ing
le a
rm t
rial
Obje
ctive
re
sp
on
se
ra
te,
pro
gre
ssio
n-f
ree
su
rviv
al
Obje
ctive
re
sp
on
se
ra
te,
pro
gre
ssio
n-f
ree
su
rviv
al
(d)
Bio
mar
ker-
stra
tifie
d r
and
om
ized
tri
al
Sta
nd
ard
of p
ractice
Sta
ndard
of pra
ctice thera
py
Inte
gra
ted
bio
mark
er
analy
sis
Adaptive
random
ization
YZ
None
XYZ
None
X
Tim
e-t
o-p
rogre
ssio
n
vers
us la
st th
era
py
an
d r
esp
on
se
Bio
mark
er-
based o
ptim
al th
era
py
Fig
ure
1.2
Bio
mar
ker-
dri
ven
clin
ical
tria
ls.(
a,b
)B
iom
arke
r-se
lect
edsi
ng
lear
mtr
ials
may
del
iver
targ
eted
ther
apy
top
atie
nts
sele
cted
for
asp
ecifi
cg
eno
mic
alte
rati
on
or
oth
erb
iom
arke
ro
rm
ayra
nd
om
ize
pat
ien
tsw
ith
ab
iom
arke
rto
inve
stig
atio
nal
ther
apy
or
stan
dar
do
fca
reth
erap
y.(c
,d)
Pati
ents
may
be
stra
tifi
edb
ased
on
the
bio
mar
ker
stat
us,
and
pat
ien
tsw
ith
or
wit
ho
ut
the
bio
mar
ker
may
rece
ive
the
inve
stig
atio
nal
ther
apy
or
they
may
be
ran
do
miz
edb
etw
een
stan
dar
do
fca
reth
erap
yan
dth
ein
vest
igat
ion
alar
m.(
e)In
um
bre
llatr
ials
,mu
ltip
leb
iom
arke
rsm
ayb
e
sim
ult
aneo
usl
yas
sess
ed,a
nd
pat
ien
tsal
loca
ted
totr
eatm
ent
bas
edo
nb
iom
arke
rst
atu
s.(f
)In
adap
tive
tria
ls,i
nit
ially
pat
ien
tsca
nb
era
nd
om
lyal
loca
ted
bet
wee
ntr
eatm
ent
arm
s,b
ut
sub
seq
uen
tly
the
allo
cati
on
isad
apti
ve,b
ased
on
dis
ease
con
tro
lor
oth
ersh
ort
-ter
men
dp
oin
tsin
each
bio
mar
ker
sub
typ
efo
rea
chth
erap
y.(g
)In
N-o
f-1
tria
ls,p
atie
nts
are
giv
ena
ther
apeu
tic
reg
imen
asse
ssed
tom
atch
thei
rg
eno
mic
/mo
lecu
lar
pro
file
mo
stcl
ose
ly.
10 PART I Principles of Targeted Therapies
Collaborations with industry are needed to access novel inves-tigational drugs that may be of benefit in these trials. Biomarker-selected trials for rare alterations are especially challenging.Collaboration across institutions is necessary to enroll across a vari-ety of institutions, usually leveraging local testing for enrollment.The efficiency of such trials is increased with the increasing utiliza-tion of “basket trials,” trials that test the efficacy of agents either ina histology-independent manner, or by accruing a variety of tumortypes, with planned analysis in disease-specific cohorts.
With increasing multiplex testing, patients are frequently foundto have more than one actionable alteration. Novel strategies areneeded to rapidly test novel combination therapies, either target-ing more than one alteration at a time, or targeting one actionablegene and additional survival pathway shown to be associated withintrinsic or acquired resistance.
The design of clinical trials inherently poses significant chal-lenges not only to biomarker discovery but also to validating thebenefit of targeted therapies. Phase I trials often contain heavilypretreated patients, adding additional complexity to the challengesof tumor heterogeneity and molecular evolution. Phase II trialscan have small sample sizes with inadequate power for biomarkervalidation. In addition, to validate a biomarker, patients with andwithout the biomarker must be treated with a drug; however, thismethodology also raises ethical concerns about treating patientsin non-marker matched trials if strong rationale exists for thebiomarker’s predictive benefit. However, if genomic markers wereused at the onset during the development of a drug and thera-pies were proven effective, studies in populations with and withoutmarkers still would have been of importance to determine the pre-dictive value of a marker. Another challenge is that even in phaseIII trials of targeted therapies may yield few objective responders asmany novel treatments are cytostatic but not cytotoxic. One com-monly utilized strategy for discovery in phase I, II, and III trials is toselect patients for additional analysis through an unusual responderprotocol; this method focuses efforts on patients who fail to respondas predicted or who achieve notably better results on a treatmentregimen.
Challenges to Personalized Cancer Therapy
Although personalized therapy holds much promise, biomarker-based treatment is utilized in only a few cancer types at this time.Furthermore, no data has demonstrated that comprehensive molec-ular profiling provides added value for the patient or decreaseshealthcare costs. Before personalized therapy can be implemented,several challenges must be overcome; these obstacles are discussedbelow.
Tumor HeterogeneityFrom variation among the proportion of cancer cells having spe-cific mutations to variation among the types of mutations, tumorshave considerable heterogeneity. Presently available multiplex tech-nologies can detect mutations that exist in as few as 5% of a tumor’scells; in-depth sequencing may be able to detect even rarer tumors.It is currently not known whether a minimum proportion of tumorcells must contain a somatic mutation to observe an effect on tumorbiology, response to targeted therapies, or resistance to alternatepathway inhibitors. The relative tumor cellularity also influencesthe “percentage mutant” through the relative proportion of normalDNA in the total DNA analyzed. In select cases with low tumor cel-lularity, microdissection of tumor cells may be necessary prior to
genomic screening, resulting in markedly increased cost. Address-ing this issue is critical to determining the genomic sequencingcoverage-depth necessary to make clinical decisions.
Molecular EvolutionBiomarker assessment of archival tissue, typically the primarytumor specimen, often serves as the basis for patient treatment deci-sions. However, tumors evolve as the disease progresses, acquiringadditional mutations that provide a growth or survival advantageand selecting for a population of subclones. In a study of pancre-atic cancer metastases and matched primary tumors, sequencingrevealed that the clonal populations that resulted in distant metas-tases were also represented in the primary tumor; however, theseclones had genetically evolved from the parental non-metastaticclone.45 Furthermore, it is not yet known whether “founder muta-tions” that exist in parental clones or “progressor mutations” thatarise following clonal evolution are better therapeutic targets. It isalso unclear if the concordance of biomarkers between primary andrecurrent tumors differ by cancer tissue of origin. In breast can-cer, discordance in the standard of care markers—estrogen recep-tor, progesterone receptor, and human epidermal growth factorreceptor-2 (HER2)—between the primary tumor and metastaseshas been observed and is associated with poorer outcomes.46 Sim-ilarly, there is discordance in immunohistochemical markers ofphosphatidylinositol 3-kinase (PI3K) pathway activation, as well asin PIK3CA mutation status between primary and recurrent breastcancers.47, 48 Interestingly, the discordance observed is not onlyattributable to metastases acquiring additional aberrations but alsofrom loss of aberrations that had been detected in the primarytumor. In contrast to PIK3CA mutation status in breast cancer, ahigh concordance in K-Ras status between primary tumors andmatched liver metastases has been reported49 To inform the selec-tion of samples for biomarker assessment, additional studies areneeded to establish the concordance of key biomarkers among dif-ferent cancer tissues of origin and different metastatic sites.
Cancer cells adapt and acquire resistance through several mech-anisms upon prolonged treatment with targeted therapy. Onemethod is through loss of the target, as observed in a study ofbreast cancer patients treated with neoadjuvant trastuzumab-basedchemotherapy; on post-treatment biopsy, a third of the samplesfrom patients who did not have a complete pathologic responsenow displayed loss of the HER2 amplification that had been presentin their pretreatment biopsies.50 Another means by which cancersdevelop resistance is the acquisition of additional genomic aberra-tions. In lung cancer, a second mutation in EGFR (T790M) andMET amplification have been described as two mechanisms ofdrug resistance to the EGFR inhibitors erlotinib and gefitinib.51–53
Subpopulations of cells with MET amplification were identifiedeven prior to drug exposure, suggesting that drug treatment effec-tively selects for these subpopulations.54 To reveal mechanisms ofacquired drug resistance, sequential tumor biopsies and system-atic genetic and histological analyses were performed in 37 patientswith drug-resistant non-small cell lung cancers (NSCLC) harbor-ing EGFR mutations.55 Across sequential biopsies, every tumorretained the activating EGFR mutations; some developed knownmechanisms of resistance, including the EGFR T790M mutationand MET amplifications. Others displayed novel genetic changes,including EGFR amplification, PIK3CA mutations, and markersof epithelial-to-mesenchymal transition. Some tumors transformedinto small cell lung cancers (SCLC) that were sensitive to stan-dard SCLC treatments. Serial biopsies in three patients revealed that