biologic therapy of leukemia - m. kalaycio (humana, 2003) ww

CONTEMPORARY HEMATOLOGY

BIOLOGICTHERAPY OFLEUKEMIA

Edited by

Matt Kalaycio, MD

BIOLOGIC THERAPY OF LEUKEMIA

CONTEMPORARY HEMATOLOGY

Gary J. Schiller, MD, SERIES EDITOR

Biologic Therapy of Leukemia, edited by MATT KALAYCIO, 2003

Chronic Lymphocytic Leukemia: Molecular Genetics, Biology, Diagnosis,and Management, by GUY B. FAGUET, 2003

Modern Hematology: Biology and Clinical Management, by REINHOLD

MUNKER, ERHARD HILLER, AND RONALD PAQUETTE, 2000

Red Cell Transfusion: A Practical Guide, edited by MARION E. REID AND

SANDRA J. NANCE, 1998

HUMANA PRESSTOTOWA, NEW JERSEY

BIOLOGIC THERAPY

OF LEUKEMIA

Edited by

MATT KALAYCIO, MD

The Cleveland Clinic Foundation, Cleveland, OH

© 2003 Humana Press Inc.999 Riverview Drive, Suite 208Totowa, New Jersey 07512humanapress.comFor additional copies, pricing for bulk purchases, and/or information about other Humana titles,contact Humana at the above address or at any of the following numbers: Tel: 973-256-1699;Fax: 973-256-8341; E-mail: [email protected]; website at humanapress.com

All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmittedin any form or by any means, electronic, mechanical, photocopying, microfilming, recording, or other-wise without written permission from the Publisher.

All articles, comments, opinions, conclusions, or recommendations are those of the author(s), and do notnecessarily reflect the views of the publisher.

Due diligence has been taken by the publishers, editors, and authors of this book to ensure the accuracyof the information published and to describe generally accepted practices. The contributors herein havecarefully checked to ensure that the drug selections and dosages set forth in this text are accurate in accordwith the standards accepted at the time of publication. Notwithstanding, as new research, changes ingovernment regulations, and knowledge from clinical experience relating to drug therapy and drugreactions constantly occurs, the reader is advised to check the product information provided by themanufacturer of each drug for any change in dosages or for additional warnings and contraindications.This is of utmost importance when the recommended drug herein is a new or infrequently used drug. Itis the responsibility of the health care provider to ascertain the Food and Drug Administration status ofeach drug or device used in their clinical practice. The publisher, editors, and authors are not responsiblefor errors or omissions or for any consequences from the application of the information presented in thisbook and make no warranty, express or implied, with respect to the contents in this publication.

This publication is printed on acid-free paper. ∞ANSI Z39.48-1984 (American National Standards Institute)Permanence of Paper for Printed Library Materials.

Production Editor: Robin B. Weisberg.

Cover Illustration: From Fig. 5 in Chapter 4, “Drug Immunoconjugate Therapy of Acute Myeloid Leu-kemia” by Arthur E, Frankel, Bayard L. Powell, Eli Estey, and Martin S. Tallman.

Cover design by Patricia F. Cleary.

Authorization to photocopy items for internal or personal use, or the internal or personal use of specificclients, is granted by Humana Press Inc., provided that the base fee of US $20.00 per copy is paid directlyto the Copyright Clearance Center at 222 Rosewood Drive, Danvers, MA 01923. For those organizationsthat have been granted a photocopy license from the CCC, a separate system of payment has been arrangedand is acceptable to Humana Press Inc. The fee code for users of the Transactional Reporting Service is:[1-58829-071-9/03 $20.00].

Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1Library of Congress Cataloging-in-Publication Data

Biologic therapy of leukemia / edited by Matt Kalaycio. p. ; cm.-- Includes bibliographical references and index. ISBN 1-58829-071-9 (alk. paper) 1-59259-383-6 (e-book) 1. Leukemia—Immunotherapy. 2. Leukemia—Gene therapy. 3. Cytokines—Therapeutic use. 4. Biological products—Therapeutic use. I. Kalaycio, Matt. II. Series.

[DNLM: 1. Leukemia‚therapy. 2. Biological Therapy. 3. Cytokines-therapeutic use. 4. Gene Therapy. 5. Immunotherapy. 6. Oligonucleotides, Antisense—therapeutic use. WH 250 B615 2003]

RC643.B457 2003616.99'41906—dc21 2002192220

To Linda, Mollie, Jason, and Rachel

PREFACE

In the latter part of the 20th century, hematologists and medical oncologistswere trained to treat leukemia with systemic therapy that was cytotoxic to bothnormal and malignant cells. Some of these therapies, such as methotrexate andL-asparaginase, were developed within the context of known biologic patho-physiology, but most were developed in relative ignorance of biologic mecha-nisms and cannot therefore be considered “biologic.” The usual goal of treatmentwas to eliminate rapidly dividing, or malignant, cells with DNA-damaging agentsthat spared normal tissue only in a relative sense. The paradigm of systemic, non-specific therapy dominated oncologic thought at the time:

Leukemia is by very definition a wide-spread systemic disease at the time ofdiagnosis. For this reason systemic therapy which reaches simultaneouslyevery cell in the body is the most logical form of treatment and is probably theonly type which offers, theoretically, the possibility of complete cure. (1)

To an extent, the systemic, nonspecific treatment approach was successful andcertainly resulted in cures when before none were possible. However, thisapproach failed to cure the majority of patients with leukemia and is usuallyassociated with significant toxicity. No other way was known, and for a time, noother way seemed possible.

The frequent failure of nonspecific treatments, remarkable advances inmolecular biology, and well-timed serendipity, led to new approaches that arerevolutionizing the management of leukemia as we enter the 21st century. Incontrast to the treatments of the past, the new approaches can collectively beclassified as truly “biologic” therapies because they take advantage of the knownbiology of leukemia. Thus, treatment can often be directed at the leukemia,sparing normal tissues and causing less tissue damage. These new targeted treat-ments represent the beginning of a new age in leukemia therapeutics.

As exciting as these are, clinicians often find it difficult to access appropriatemedical information on these new treatments when faced with a patient who maybenefit from them. The advances are coming so often, and so quickly, that treat-ments are sometimes approved for use before the information that supports theirclaimed efficacy can be published in peer-reviewed literature. Large textbooksattempting to publish accurate and current information on leukemia are doomedto obsolescence before reaching print.

These practical concerns prompted the publication of this book. BiologicTherapy of Leukemia is devoted to these new biologic therapies and provides a

vii

viii Preface

rapidly accessible, authoritative source of practical information for cliniciansattempting to use these treatments for their patients.

Some of the treatments described in this text, such as interferon and all-transretinoic acid, have been available for some time and are well-described in themedical literature. However, that information is difficult to access when contrast-ing their efficacy with newer treatments, such as imatinib mesylate and arsenictrioxide, which are also described in this text. Other treatments, such as P-glyco-protein inhibitors and interleukins, have been dancing on the edges of clinicalpractice and may yet find their place based on emerging data. The graft vs leu-kemia effect has been better defined and promises to completely alter the wayallogeneic stem cell transplant is employed in the future. Finally, therapeuticapproaches that reverse failure of apoptosis, alter genetic codes, and modulateimmunologic mechanism are no longer mere theory, but are now being tested inthe clinic and warrant close attention by the oncologic community.

The authors and I hope that clinicians treating patients will find BiologicTherapy of Leukemia helpful. We all share the goal of eradicating leukemia andI believe the information contained in these pages moves us closer to that goal.I thank the contributors for their expertise and willingness to share it. I stand inawe of their knowledge and dedication.

Matt Kalaycio, MD

REFERENCE

1. Burchenal, J.H. Treatment of the leukemias. Semin Hematol 1966;3:122.

ix

CONTENTS

Preface ........................................................................................................... viiContributors .................................................................................................... xi

PART I: IMMUNOTHERAPY

1 Human Leukemia-Derived Dendritic Cells as Toolsfor Therapy .......................................................................................... 3

David Claxton

2 The Graft vs Leukemia Effect ...............................................................13Brian J. Bolwell

3 Unconjugated Monoclonal Antibodies .................................................29Matt Kalaycio

4 Drug Immunoconjugate Therapy of Acute MyeloidLeukemia ...........................................................................................43

Arthur E. Frankel, Bayard L. Powell, Eli Estey,and Martin S. Tallman

5 Radiolabeled Monoclonal Antibodies ..................................................59John M. Burke and Joseph G. Jurcic

PART II: CYTOKINES

6 Interferons..............................................................................................81Thomas Fischer

7 Interleukin-2 Treatment of Acute Leukemia ........................................93Peter Kabos and Gary J. Schiller

PART III: TARGETED THERAPEUTICS

8 Antisense Therapy ...............................................................................109Stanley R. Frankel

9 Signal Transduction Inhibitors ............................................................127Michael E. O’Dwyer and Brian J. Druker

10 P-Glycoprotein Inhibition in Acute Myeloid Leukemia ....................145Thomas R. Chauncey

11 Targeting the Apoptotic Machinery as a PotentialAntileukemic Strategy ....................................................................163

Benjamin M. F. Mow and Scott H. Kaufmann

PART IV: DIFFERENTIATION AGENTS

12 Arsenicals: Past, Present, and Future ................................................189Chadi Nabhan and Martin S. Tallman

13 All-Trans-Retinoic Acid in the Treatment of AcutePromyelocytic Leukemia ................................................................205

Pierre Fenaux and Laurent Degos

PART V: GENE THERAPY

14 Gene Therapy ......................................................................................225Paul J. Orchard and R. Scott McIvor

Index .............................................................................................................261

x Contents

xi

CONTRIBUTORS

BRIAN J. BOLWELL, MD • Bone Marrow Transplant Program, Departmentof Hematology and Medical Oncology, Cleveland Clinic Foundation,Cleveland, OH

JOHN M. BURKE, MD • Department of Medicine, Memorial Sloan-KetteringCancer Center, New York, NY

THOMAS R. CHAUNCEY, MD, PhD • University of Washington Schoolof Medicine, VA Pugent Sound Health Care System, Seattle, WA

DAVID CLAXTON, MD • Division of Hematology/Oncology, Penn State HersheyMedical Center, Hershey, PA

LAURENT DEGOS, MD, PhD • Institut d’Hematologie, Hôpital Saint Louis, Paris,France

BRIAN J. DRUKER, MD • Leukemia Center, Oregon Health & ScienceUniversity Cancer Institute, Portland, OR

ELI ESTEY, MD • Section of Acute Leukemia and Myelodysplastic Syndromes,MD Anderson Cancer Center, Houston, TX

PIERRE FENAUX, MD • Service des Maladies du Sang, CHU, Lille, FranceTHOMAS FISCHER, MD • Johannes Gutenberg University of Mainz, Mainz, GermanyARTHUR E. FRANKEL, MD • Department of Medicine, Wake Forest University

School of Medicine, Winston Salem, NCSTANLEY R. FRANKEL, MD, FACP • Medical Operations, Genta Incorporated,

Chicago, IL, and Department of Medicine, Greenbaum Cancer Center,University of Maryland, Baltimore, MD

JOSEPH G. JURCIC, MD • Leukemia Service, Memorial Sloan-Kettering CancerCenter, New York, NY

PETER KABOS, MD • Maxine Dunitz Neurosurgical Institute, Cedars-SinaiMedical Center, Los Angeles, CA.

MATT KALAYCIO, MD • Leukemia Program, Department of Hematologyand Medical Oncology, Cleveland Clinic Foundation, Cleveland, OH

SCOTT H. KAUFMANN, MD, PhD • Division of Oncology Research, Mayo Clinic,and Department of Molecular Pharmacology, Mayo Graduate School,Rochester, MN

R. SCOTT MCIVOR, MD, PhD • Department of Genetics, Cell Biology,and Development, Institute of Human Genetics, University of Minnesota,Minneapolis, MN

BENJAMIN M. F. MOW, MD • Division of Hematology, National UniversityHospital, Singapore

xii Contributors

CHADI NABHAN, MD • Division of Hematology/Oncology, NorthwesternUniversity Medical School, Chicago, IL

MICHAEL E. O’DWYER, MD • Leukemia Center, Oregon Health & ScienceUniversity Cancer Institute, Portland, OR

PAUL J. ORCHARD, MD • Department of Pediatrics, Institute of HumanGenetics, University of Minnesota, Minneapolis, MN

BAYARD L. POWELL, MD • Section of Hematology/Oncology, Wake ForestUniversity School of Medicine, Winston Salem, NC

GARY J. SCHILLER, MD • Division of Hematology-Oncology, UCLA Schoolof Medicine, Los Angeles, CA

MARTIN S. TALLMAN, MD • Division of Hematology/Oncology, NorthwesternUniversity Medical School, Chicago, IL

I IMMUNOTHERAPY

1. INTRODUCTION: IMMUNOTHERAPY OF LEUKEMIA

Excluding selected subsets of human leukemia, progress in the therapy ofthis group of disorders has been relatively modest since the late 1980s. In con-trast, the elucidation of the pathogenesis and mechanisms of progression ofthese diseases has been substantial. The development of specific agents tar-geted at selected molecular changes specific to leukemic cells offers great hopefor the therapy of these disorders. Even these targeted therapies, such as therecently developed tyrosine kinase inhibitor Imatinib Mesylate, may fail tocontrol advanced forms of the target disease (1). Additionally, for many humanleukemias, no such well-defined genetic background is available at present.Accordingly, efforts to advance the therapy for these diseases must focus onother approaches.

There is substantial experimental and clinical data suggesting the usefulness ofT-cell-mediated antigen-specific immunotherapy for leukemia. The central prin-ciple common to all these approaches is the development of effective cell-medi-ated immunity able to selectively or semiselectively target leukemic or malignantcells. Thus, these approaches ultimately yield an active immune process with thepotential for ongoing control of residual malignant cellular elements.

1 Human Leukemia-Derived DendriticCells as Tools for Therapy

David Claxton, MD

CONTENTS

INTRODUCTION: IMMUNOTHERAPY OF LEUKEMIA

DENDRITIC CELLS AS TOOLS FOR THERAPY

HUMAN DENDRITIC CELLS DERIVED FROM MYELOID

LEUKEMIAS

CLINICAL STUDIES USING LEUKEMIA-DERIVED DENDRITIC

CELLS

FUTURE STUDIES AND OPPORTUNITIES

REFERENCES

From: Biologic Therapy of LeukemiaEdited by: M. Kalaycio © Humana Press Inc., Totowa, NJ

3

Alloimmune antileukemic effects are well established, both experimentallyand clinically. Several investigators have described animal models that docu-ment the ability of allogeneic immune cells to cure transplantable experimentalleukemias. In human clinical studies, this effect is also clearly active. Thus, forboth matched sibling and unrelated donor hematopoietic stem cell transplants,the acquisition of an allodisparate T-cell population and the recognition of hostallo-antigens expressed on leukemias have been shown to be active in the ulti-mate control of chronic myelogenous leukemia (CML) and acute myeloidleukemia (AML). Relevant antigens targeted in clinical studies and identifiedas active in this phenomenon include the minor histocompatiblity antigenHA-1 (2). In certain cases, such antigens have been shown to be hematopoi-etic-specific polymorphic proteins (3). Thus, the graft vs hematopoietic effectsdeveloping after allogeneic stem cell transplantation control the normalhematopoietic and leukemic elements derived from the host simultaneously.

Autologous cellular immune activity directed toward leukemias has alsobeen described in experimental and clinical systems. Several murine systemshave been described that demonstrate the activity of autologous immunity(4–7) against transplantable congenic leukemia. Several authors have pub-lished clinical series suggesting therapeutic effects of systemic IL-2 with (8) orwithout (9–12) IL-2 activated cells for human leukemias. Antileukemic T-cellclones or lines have been described (13).

2. DENDRITIC CELLS AS TOOLS FOR THERAPY

Human dendritic cells represent the so-called “professional antigen-present-ing cells” responsible for initiating all antigen-specific cell-mediated immunityfrom naïve T-cell elements. The development of techniques for the ex vivo dif-ferentiation of dendritic cells from peripheral blood-derived monocytes or bonemarrow-derived CD34 cells has presented clinical investigators with the oppor-tunity to use these cells for the initiation of antitumor immunity in experimentaland clinical studies. Several clinical trials have been reported in which allo-geneic or autologous tumor lysates or antigen-specific peptides have beenpulsed into dendritic cells derived in one of these two ways (14–17). The devel-opment of tumor-specific autologous cellular cytotoxicity has been demon-strated in several human tumor systems, including human AML (18). Commonto all of these systems, however, is the use of normal autologous dendritic cellsbrought to a state of maturation that are suitably primed to acquire fresh antigenfrom the extracellular environment. These antigens then are provided by theinvestigator and effectively presented to autologous T-cells through the antigenprocessing and presenting machinery of the dendritic cells. In the followingpages, a similar system for the initiation of antileukemic cell-mediated immu-nity is described that, however, stems from the ability of the original leukemiccells to differentiate directly into effective antigen-presenting cells and to pre-

4 Claxton

sent the endogenously expressed leukemic antigens. It is our contention thatthis system may be the ideal one in which to study the therapeutic potency ofhuman dendritic cells for the initiation of antitumor cell-mediated activity.

3. HUMAN DENDRITIC CELLS DERIVED FROM MYELOID LEUKEMIAS

In 1994, two groups independently described cellular differentiation of cellswith the characteristics of myeloid dendritic cells from AML (19,20). Cellsderived from patients and cultured ex vivo in recombinant cytokines, includingGM-CSF and tumor necrosis factor-alpha (TNF-α), matured into cells with themorphology and activity of typical mature dendritic cells (19). These cells gen-erated brisk allogeneic mixed lymphocyte reactions, as would be expected ofmature dendritic cells.

In 1996, Choudhury et al. demonstrated that CML cells could differentiateinto cells with the morphology, phenotype, and allostimulatory function ofmature dendritic cells (21). Simultaneously, these cells were shown to bederived from the malignant clone by in situ hybridization to nucleic acidprobes specific for the Philadelphia chromosome breakpoints of the bcr-ablgenes. These leukemia-derived dendritic cells were able to stimulate autolo-gous preactivated T-cells to acquire cytotoxicity specific for the leukemic cloneand not for normal human leukocyte antigen (HLA)-matched targets or autolo-gous remission bone marrow cells. The cytotoxic activity demonstrated wasfound within the CD8 cellular compartment. There was no NK type cytotoxic-ity given the near absence of cytotoxicity for K562 cells. This was the firstreport of a leukemia-derived dendritic cell-inducing autologous leukemia-spe-cific cell-mediated cytotoxicity (21).

Since this publication, a total of 30 peer-reviewed publications describinghuman leukemia-derived dendritic cells have been identified by literaturesearch. These publications have examined the differentiation of both CML andAML into active antigen-presenting cells. Several authors have documentedthe clonal origin of the culture-derived dendritic cells (22–26). Most reportshave shown the acquisition of relatively high expression of the critical costim-ulatory molecules CD80 and 86, together with the mature dendritic cell markerCD83, on the leukemia-derived dendritic cells. Five independent groups havedemonstrated the acquisition of leukemia-specific cellular cytotoxicity inautologous T-cells stimulated by leukemia-derived dendritic cells in CML(22,27–30). Five independent investigative groups, including our own group,have also shown that AML may similarly differentiate into clonal dendriticcells capable of stimulating cytotoxic leukemia-specific T-cell activity(23,25,31–33). In the case of both anti-CML and anti-AML activities, theseautologous cytotoxic reactions have been shown to be major histocompatibilitycomplex (MHC) restricted as demonstrated by partial blocking of cytotoxicity

Chapter 1 / Dendritic Cells and Therapy 5

by HLA class 1- or class 2-specific monoclonal blocking antibodies (23,27).Thus, the cytotoxic activities identified have the characteristics expected fordendritic cell-initiated T-cell mediated responses.

Several questions arise from this large body of work. Fundamental ques-tions exist about the uncertainty concerning the details of the immune responseto leukemia-derived dendritic cells. Specifically, it is unclear what antigens arebeing responded to and which subsets of effector cells are participating. Cur-rently, there is little data to document which antigens may be responsible foreliciting responses to leukemia-derived dendritic cells; however, our currentwork has suggested that a limited number of target antigens may be sharedbetween patients. Yasukawa et al. have shown that CML-derived dendritic cellsmay specifically present a bcr-abl B3A2 junction-specific peptide fragment toCD4-positive T cells via an HLA-DRB1*0901 restricted mechanism (34). Onthe other hand, Bertazzoli et al. have shown that T-cells derived from CMLpatients respond less well to a B3A2 bcr-abl-specific peptide in a proliferativeassay than those from normal donors (35). Thus, the role of bcr-abl junction-derived antigens and the responses elicited by leukemia-derived dendritic cellsremain unresolved but dubious.

4. CLINICAL STUDIES USING LEUKEMIA-DERIVEDDENDRITIC CELLS

Fujii et al. described a single patient with CML who was treated with autol-ogous peripheral stem cell transplantation followed by a series of infusions ofleukemia-derived dendritic cells (22). The leukemia-derived dendritic cellswere generated in the cytokine combination GM-CSF, TNF-α, and IL-4. Afterreceiving four infusions of post-transplant-administered culture-derived autol-ogous leukemic dendritic cells, the patient developed a partial cytogeneticresponse because the total number of Philadelphia (Ph) chromosome-positivemetaphases in the bone marrow fell from 20 out of 20 to 7 out of 20 within sev-eral months.

We carried out a study using CML-derived dendritic cells to stimulate autol-ogous T-cells to develop a therapeutic cell line for adoptive immunotherapy.This study, which enrolled five patients and treated two, has been brieflyreported and described elsewhere (36). A schematic diagram of this study isshown in Fig. 1. Patients had CML cells collected by apheresis when they hadsubstantial numbers of circulating dendritic cells or after stimulation with GM-CSF. These cells were subsequently used to stimulate autologous T-cells preac-tivated with OKT3 and IL-2. After coculture, cells were cryopreserved inaliquots and subsequently administered to patients. A schematic view of thecell culture approach is given in Fig. 2.

Two days before cell infusion, modest doses of cyclophosphamide wereadministered to cytoreduce the disease. After receiving the activated lympho-

6 Claxton


Fig. 1. Schematic view of dendritic cell-activated lymphocyte therapy for CML.

Fig. 2. Culture procedures for the generation of autologous T-cells activated by leukemia-derived dendritic cells. * Culture A (DC) must be >20% CD86 + and >40% HLA-DR +.Culture B(AL) must be >60% CD3 +.

cytes, patients received intermittent subcutaneous low-dose IL-2. This wasgiven 5d on, 2d off and 5d on with each treatment of chemotherapy. The ther-apy was well tolerated. Mild fever and local skin reactions were seen in associa-tion with IL-2 therapy, as might be expected. Disease stabilized in both patients,and the second patient treated, who was in frank accelerated phase of CMLbefore initiation of therapy, reverted into a chronic phase. Marrow metaphases,which had been marked with trisomy 8 in addition to the Ph1 chromosome, tran-siently reverted toward simple Ph-positive metaphases. The previous thrombo-cytopenia in this patient was reversed, and the patient was well for the durationof the therapy, as shown in Fig. 3. Thus, this trial showed that it was feasible togenerate large numbers of active dendritic cells that had the characteristics ofprofessional antigen-presenting cells. Administration of these cells to patientswas not associated with adverse effects. Adverse effects seen with the programwere those expected from the concomitant chemotherapy or the IL-2 given.

5. FUTURE STUDIES AND OPPORTUNITIES

AMLs and myelodysplastic syndromes are often ultimately untreatable withavailable conventional therapies. Accordingly, the ability to generate potentdendritic cells from the majority of patients with advanced AML or myelodys-plastic syndrome presents great opportunities for study.

This group of diseases is attractive for study for several reasons. First, it isstraightforward to harvest large numbers of malignant cells from patients byvenipuncture or bone marrow collection. Second, subsets of patients who maybe successfully treated into remission but who have a poor prognosis with con-

8 Claxton

Fig. 3. Time course of patient 2 treated with DC-AL for accelerated-phase CML. The hori-zontal axis represents months (April through August). Leukocytosis and thrombocytopeniareverted to normal on protocol.

ventional chemotherapy administered in remission are easily identified. Thus,these diseases offer the opportunity for administration of immunotherapy inthe context of minimal residual disease, which is nonetheless known to portenda poor prognosis with conventional therapy. This is perhaps the optimal settingin which to test the activity of novel immunologic treatments.

Significant questions remain concerning the optimal mode of delivery ofdendritic cell immunotherapy to such patients, however. It is unclear whethervaccination with activated antigen-presenting cells via the subcutaneous, intra-dermal, or intravenous infusion routes or administration of ex-vivo activated T-cells after dendritic cells stimulation is the optimal approach for initiation ofdisease-specific immunity. Certainly the latter approach (adoptive transfer ofactivated T-cells), which was pursued in the CML protocol described, presentsconsiderable technical challenges. The logistic and regulatory challenges ofthe laboratory processing and culture of cells in a program of ex vivo immuneactivation are formidable.

We are currently pursuing clinical trials of dendritic cells as vaccination-basedimmunotherapy for the treatment of poor prognosis, newly diagnosed patientswith AML, and patients whose diseases relapsed but are believed to have a rea-sonable chance of successful remission induction. Patients will receive standardremission-induction chemotherapy and will receive dendritic cell vaccinationwhen they are recovering from consolidation therapy. Vaccination therapy willcontinue for a few months, and patients will ultimately have immune cell popula-tions assayed for antileukemic T-cell responses. Additionally, cutaneous cell-mediated T-cell responses to vaccinating dendritic cells will be followed.

An additional appealing arena for the study of therapeutic activity ofleukemia-derived dendritic cells is that of allogeneic stem cell transplantation.Although this therapy is active, for patients with poor-prognosis AML, diseaserelapse after transplantation is still frequent. Thus, the treatment of diseaserelapse after transplantation with chemotherapy followed by vaccination withdendritic cell-activated lymphocytes is of great importance. There are experi-mental data to suggest that T-cell vaccination after allogeneic transplantationmay lead to control of disease.

The development of novel therapies is often slower than might be initiallyexpected. At the time when such therapies are identified as potentially active ininitial experimental laboratory studies, the natural belief is that clinical proofof principle will follow rapidly. Monoclonal antibodies were identified approx20 yr ago, but the first monoclonal anticancer therapies to show antitumoractivity have only been available recently. Thus, it may be expected that initialfrustrations with the development of cellular anticancer therapies may beexpected to yield to success in the treatment of many human tumors. There iscertainly abundant experimental evidence for the potent cellular immune activ-ities that should be active in the control of human cancer.


The author believes that human myeloid leukemias and myelodysplasticsyndromes may offer a nearly ideal system in which to test the hypothesis thathuman malignancies may be controlled through dendritic cell or antigen-pre-senting cell therapy. Trials of several vaccine-based and adoptive immunother-apeutic approaches are urgently needed.

REFERENCES1. Druker BJ, Sawyers CL, Kantarjian H, et al. Activity of a specific inhibitor of the BCR-ABL

tyrosine kinase in the blast crisis of chronic myeloid leukemia and acute lymphoblasticleukemia with the Philadelphia chromosome [see comments] [erratum appears in N Engl JMed 2001;345(3):232]. N Engl J Med 2001;344:1038–1042.

2. Brossart P, Spahlinger B, Grunebach F, et al. Induction of minor histocompatibility antigenHA-1-specific cytotoxic T-cells for the treatment for leukemia after allogeneic stem celltransplantation [letter; comment]. Blood 1999;94:4374–4376.

3. Mutis T, Verdijk R, Schrama E, Esendam B, Brand A, Goulmy E. Feasibility of immunother-apy of relapsed leukemia with ex vivo-generated cytotoxic T lymphocytes specific forhematopoietic system-restricted minor histocompatibility antigens [see comments]. Blood1999;93:2336–2341.

4. Matulonis U, Dosiou C, Freeman G, et al. B7-1 is superior to B7-2 costimulation in theinduction and maintenance of T-cell-mediated antileukemia immunity. Further evidence thatB7-1 and B7-2 are functionally distinct. J Immunol 1996;156:1126–1131.

5. Enright H, McGlave PB. Biology and treatment of chronic myelogenous leukemia. Oncol-ogy (Huntington) 1997;11:1295–1300.

6. Runyon K, Lee K, Zuberek K, Collins M, Leonard JP, Dunussi-Joannopoulos K. The combi-nation of chemotherapy and systemic immunotherapy with soluble B7-immunoglobulin Gleads to cure of murine leukemia and lymphoma and demonstration of tumor-specific mem-ory responses. Blood 2001;97:2420–2426.

7. Dunussi-Joannopoulos K, Runyon K, Erickson J, Schaub RG, Hawley RG, Leonard JP. Vac-cines with interleukin-12-transduced acute myeloid leukemia cells elicit very potent thera-peutic and long-lasting protective immunity. Blood 1999;94:4263–4273.

8. Benyunes MC, Massumoto C, York A, Higuchi CM, Buckner CD, Thompson JA, et al. Inter-leukin-2 with or without lymphokine-activated killer cells as consolidative immunotherapyafter autologous bone marrow transplantation for acute myelogenous leukemia. Bone Mar-row Transplant 1993;12:159–163.

9. Toren A, Ackerstein A, Slavin S, Nagler A. Role of interleukin-2 in human hematologicalmalignancies [review]. Med Oncol 1995;12:177–186.

10. Messina C, Zambello R, Rossetti F, et al. Interleukin-2 before and/or after autologous bonemarrow transplantation for pediatric acute leukemia patients. Bone Marrow Transplant1996;17:729–735.

11. Mandelli F, Vignetti M, Tosti S, Andrizzi C, Foa R, Meloni G. Interleukin-2 treatment inacute myelogenous leukemia. Stem Cells 1993;11:263–268.

12. Fefer A, Benyunes M, Higuchi C, et al. Interleukin-2 +/– lymphocytes as consolidativeimmunotherapy after autologous bone marrow transplantation for hematologic malignan-cies. Acta Haematol 1993;89(Suppl 1):2–7.

13. Coleman S, Fisher J, Hoy T, Burnett AK, Lim SH. Autologous MHC-dependent leukemia-reactive T lymphocytes in a patient with CML. Leukemia 1996;10:483.

14. Tjoa BA, Erickson SJ, Bowes VA, et al. Follow-up evaluation of prostate cancer patients infusedwith autologous dendritic cells pulsed with PSMA peptides. Prostate 1997;32:272–278.

15. Nestle FO, Alijagic S, Gilliet M, et al. Vaccination of melanoma patients with peptide- ortumor lysate-pulsed dendritic cells [see comments]. Nat Med 1998;4:328–332.

10 Claxton

16. Kugler A, Stuhler G, Walden P, et al. Regression of human metastatic renal cell carcinomaafter vaccination with tumor cell-dendritic cell hybrids [see comments]. Nat Med2000;6:332–336.

17. Morse MA, Deng Y, Coleman D, et al. A Phase I study of active immunotherapy with carci-noembryonic antigen peptide (CAP-1)-pulsed, autologous human cultured dendritic cells inpatients with metastatic malignancies expressing carcinoembryonic antigen. Clin CancerRes 1999;5:1331–1338.

18. Fujii S, Fujimoto K, Shimizu K, et al. Presentation of tumor antigens by phagocytic dendriticcell clusters generated from human CD34+ hematopoietic progenitor cells: induction ofautologous cytotoxic T lymphocytes against leukemic cells in acute myelogenous leukemiapatients. Cancer Res 1999;59:2150–2158.

19. Santiago-Schwarz F, Coppock DL, Hindenburg AA, Kern J. Identification of a malignantcounterpart of the monocyte-dendritic cell progenitor in an acute myeloid leukemia. Blood1994;84:3054–3062.

20. Srivastava BL, Srivastava A, Srivastava MD. Phenotype, genotype and cytokine productionin acute leukemia involving progenitors of dendritic Langerhans’ cells. Leukemia Res1994;18:499–411.

21. Choudhury A, Gajewski JL, Liang JC, et al. Use of leukemic dendritic cells for the genera-tion of antileukemic cellular cytotoxicity against Philadelphia chromosome-positive chronicmyelogenous leukemia. Blood 1997;89:1133–1142.

22. Fujii S, Shimizu K, Fujimoto K, et al. Analysis of a chronic myelogenous leukemia patientvaccinated with leukemic dendritic cells following autologous peripheral blood stem celltransplantation. Jpn J Cancer Res 1999;90:1117–1129.

23. Woiciechowsky A, Regn S, Kolb HJ, Roskrow M. Leukemic dendritic cells generated in thepresence of FLT3 ligand have the capacity to stimulate an autologous leukemia-specificcytotoxic T-cell response from patients with acute myeloid leukemia. Leukemia2001;15:246–255.

24. Oehler L, Berer A, Kollars M, et al. Culture requirements for induction of dendritic cell dif-ferentiation in acute myeloid leukemia. Ann Haematol 2000;79:355–362.

25. Harrison BD, Adams JA, Briggs M, Brereton ML, Yin JA. Stimulation of autologous prolif-erative and cytotoxic T-cell responses by “leukemic dendritic cells” derived from blast cellsin acute myeloid leukemia. Blood 2001;97:2764–2771.

26. Cignetti A, Bryant E, Allione B, Vitale A, Foa R, Cheever MA. CD34(+) acute myeloid andlymphoid leukemic blasts can be induced to differentiate into dendritic cells. Blood1999;94:2048–2055.

27. Choudhury A, Toubert A, Sutaria S, Charron D, Champlin RE, Claxton DF. Humanleukemia-derived dendritic cells—ex-vivo development of specific antileukemic cytotoxic-ity. Crit Rev Immunol 1998;18:121–131.

28. Nieda M, Nicol A, Kikuchi A, et al. Dendritic cells stimulate the expansion of bcr-abl-spe-cific CD8+ T-cells with cytotoxic activity against leukemic cells from patients with chronicmyeloid leukemia. Blood 1998;91:977–983.

29. Eibl B, Ebner S, Duba C, et al. Dendritic cells generated from blood precursors of chronicmyelogenous leukemia patients carry the Philadelphia translocation and can induce a CML-specific primary cytotoxic T-cell response. Genes Chromosomes Cancer 1997;20:215–223.

30. Chen X, Regn S, Raffegerst S, Kolb HJ, Roskrow M. Interferon alpha in combination withGM-CSF induces the differentiation of leukaemic antigen-presenting cells that have thecapacity to stimulate a specific anti-leukaemic cytotoxic T-cell response from patients withchronic myeloid leukaemia. Br J Haematol 2000;111:596–607.

31. Choudhury A, Liang JC, Thomas EK, et al. Dendritic cells derived in vitro from acute myel-ogenous leukemia cells stimulate autologous, antileukemic T-cell responses. Blood1999;93:780–786.


32. Charbonnier A, Gaugler B, Sainty D, Lafage-Pochitaloff M, Olive D. Human acutemyeloblastic leukemia cells differentiate in vitro into mature dendritic cells and induce thedifferentiation of cytotoxic T, cells against autologous leukemias. Eur J Immunol1999;29:2567–2578.

33. Hagihara M, Shimakura Y, Tsuchiya T, et al. The efficient generation of CD83-positiveimmunocompetent dendritic cells from CD14-positive acute myelomonocytic or monocyticleukemia cells in vitro. Leukemia Res 2001;25:249–258.

34. Yasukawa M, Ohminami H, Kojima K, et al. HLA class II-restricted antigen presentation ofendogenous bcr-abl fusion protein by chronic myelogenous leukemia-derived dendritic cellsto CD4(+) T lymphocytes. Blood 2001;98:1498–1505.

35. Bertazzoli C, Marchesi E, Passoni L, et al. Differential recognition of a BCR/ABL peptideby lymphocytes from normal donors and chronic myeloid leukemia patients. Clin CancerRes 2000;6:1931–1935.

36. Choudhury A, Toubert A, Sutaria S, et al. Human leukemia-derived dendritic cells—ex vivodevelopment of specific antileukemic cytotoxicity. Critical Reviews in Immunology1998;18:121–131.

12 Claxton

1. INTRODUCTION

The original rationale of bone marrow transplantation (BMT) was solelybased on the concept of dose intensity. The logic was as follows: the ability todeliver anticancer therapy (chemotherapy and/or radiation therapy) is limitedby dose toxicities, primarily toxicity to normal bone marrow; tumors not sus-ceptible to repetitive doses of modest amounts of chemotherapy might be com-pletely obliterated with one extremely large dose of chemotherapy and/orradiation therapy; a consequence of one large dose of therapy is destruction ofnormal hematopoiesis, resulting in permanent aplasia; if normal matched mar-row were available for transplantation, then these “lethal” doses of chemother-apy could be administered to a patient, the tumor might be eradicated, and theinfusion of donor allogeneic bone marrow would restore normal hematopoiesisand save the patient from iatrogenic death. Clinical success with autologousBMT has shown validity of this theory of dose intensity. However, it hasbecome clear throughout the past 20 yr that powerful immunologic forces con-tribute to the potential for cure in allogeneic BMT (alloBMT). The immuno-logic reaction by which donor cells from the graft generate an anticancer effect

13


2 The Graft vs Leukemia Effect

Brian J. Bolwell, MD

CONTENTS

INTRODUCTION

THE RELATIONSHIP OF GRAFT VS HOST DISEASE

WITH THE GRAFT VS LEUKEMIA EFFECT

THE RELATIONSHIP OF T-CELL DEPLETION WITH

THE GVL VS LEUKEMIA EFFECT

DONOR LEUKOCYTE INFUSIONS

NONMYELOABLATIVE ALLOGENEIC TRANSPLANTATIONS

SUMMARY

REFERENCES

is known as the graft vs leukemia (GVL) or graft vs tumor (GVT) effect. Thischapter focuses on clinical aspects of the GVL effect: the relationship betweengraft vs host disease (GVHD) and the GVL effect, T-cell depletion and its rela-tionship to the GVL effect, donor leukocyte infusions (DLI) as a treatment ofdisease relapse after alloBMT, and current results of nonmyeloablative allo-geneic transplantation.

2. THE RELATIONSHIP OF GRAFT VS HOST DISEASE WITH THE GRAFT VS LEUKEMIA EFFECT

One of the major complications of alloBMT is GVHD. GVHD is animmunologic phenomenon occurring when immunocompetent donor cells per-ceive host tissues to be “foreign” and mount an immunologic attack againstthem. Acute GVHD usually occurs within the first 100 d after transplantationand affects the skin, liver, and gastrointestinal tract. Chronic GVHD occurs100 or more d after transplantation and is generally belived to be less of a ful-minate disorder; clinical manifestations can affect almost any organ but com-monly involve the liver, skin, eyes, bone marrow, mouth (sicca syndrome), andlungs. GVHD prophylaxis and treatment employ potent immunosuppressivetherapy directed toward reducing lymphocyte number and function. Unfortu-nately, the immunosuppressive therapy predisposes patients to opportunisticinfections. Therefore, both the development of GVHD and its treatment areclinically vexing problems associated with significant morbidity and mortality.Despite these toxicities, the development of GVHD may be beneficial, becauseGVHD is frequently associated with the GVL effect, resulting in a lower riskof disease relapse after transplantation.

Although it was originally postulated more than 40 yr ago that donorhematopoietic cells might generate an anticancer effect (1,2), the clinical rela-tionship of GVHD with leukemic relapse was not documented until the 1970sand early 1980s. Odom et al. described two children with acute lymphoblasticleukemia (ALL) who relapsed after alloBMT. When clinical GVHD devel-oped, the children subsequently developed a remission (3). Weiden et al. com-pared leukemic patients undergoing a syngeneic BMT with those receiving analloBMT and observed differences in the relative relapse rates for those withand without clinical GVHD (4). The patients with clinical GVHD had a relapserate 2.5 times less than patients without GVHD. Additionally, the relapse ratewas higher in syngeneic patients than in allogeneic transplantation recipientswho did not develop GVHD. This finding suggested that cells in the allogeneicgraft produced a GVL effect.

Subsequent studies by the Seattle transplantation group further defined therelationship between GVHD and leukemic relapse. One study of patients withleukemia receiving an alloBMT showed that clinical GVHD augmented the

14 Bolwell

GVT effect, because patients with clinically significant chronic GVHD had a27% risk of long-term relapse, compared with a 55% risk of relapse for patientswith subclinical GVHD (p = 0.0003) (5). Thus, clinical chronic GVHD wasstrongly associated with a long-term GVL effect. A second study of more than1200 recipients of alloBMT reported that patients with acute leukemia trans-planted when in relapse had a lower relapse rate if they subsequently developedeither acute or chronic GVHD (6). The conclusion was, again, that chronicGVHD leads to a durable antileukemic, or GVL, effect.

The International Bone Marrow Transplant Registry reported a landmarkanalysis of more than 2000 recipients of human leukocyte antigen (HLA)-iden-tical sibling BMTs, examining the relationship between GVHD and diseaserelapse. Decreased risk of relapse was observed in recipients of non-T-cell-depleted allografts with acute (relative risk 0.68, p = 0.03), chronic (relativerisk 0.43, p = 0.01), and both acute and chronic GVHD (relative risk 0.33, p =0.0001) when compared with recipients without GVHD (7). This large multi-institutional trial confirmed an unequivocal relationship of both acute andchronic GVHD with decreased leukemic relapse. Many trials have substanti-ated these findings, including specific associations with a GVT effect inleukemia, lymphoma, and myeloma (8–17).

In summary, abundant data emerged from 1978 to 1992 describing a strongrelationship between the development of clinical GVHD, both acute andchronic, with reduced risk of relapse following alloBMT. The logical conclu-sion of these observations was that cells in the donor graft, which resulted inGVHD, also led to a profound antitumor effect.

3. THE RELATIONSHIP OF T-CELL DEPLETION WITH THE GVL VS LEUKEMIA EFFECT

Although it was evident by 1990 that a relationship existed betweenGVHD and reduced risk of leukemic relapse, the development of GVHDitself was unfortunately associated with significant morbidity and mortality.Therapeutic options to treat and prevent GVHD were limited. Mortalityfrom GVHD could overshadow the risks of leukemic relapse for somepatients. Therefore, many centers began clinical trials of T-cell depletion inalloBMT to reduce the incidence and severity of GVHD. It was believed thatthe T-cells in the donor graft were, in part, responsible for the developmentof clinical GVHD. Removing these T-cells might reduce the risk of morbid-ity and mortality from clinical GVHD. Unfortunately, many studies subse-quently showed that T-cell depletion was associated with an increased risk ofleukemic relapse.

Several small studies of T-cell-depleted alloBMT showed a trend towardincreased risk of relapse after transplantation (18,19). Two large reports sub-

Chapter 2 / Graft vs Leukemia Effect 15

sequently demonstrated an association of an increased risk of disease relapsefollowing alloBMT in patients receiving T-cell-depleted bone marrow. Gold-man et al. described 405 patients with chronic myelogenous leukemia(CML) receiving alloBMTs when they were in the chronic phase. The prob-ability of relapse was higher for recipients of T-cell-depleted marrow com-pared with non-T-cell-depleted marrow (relative risk 5.4, p < 0.0001) (20).The International Bone Marrow Transplant Registry compared more than731 recipients of T-cell-depleted HLA-identical sibling BMT with 2480recipients of non-T-cell-depleted marrow. Although T-cell depletion didreduce the risk of acute and chronic GVHD leukemic relapse increased.Leukemic relapse was 2.75 times more likely after T-cell depletion forpatients with acute leukemia in first remission or patients with CML inchronic phase (p < 0.0001) (21). Thus, although T-cell depletion did reducethe risk of GVHD, leukemia-free survival was not enhanced because of theloss of the GVL effect.

Most of these studies used pan T-cell depletion, or removal of all T-cell sub-sets from the HLA-matched sibling donor graft. Subsequent reports have sug-gested that less-than-full T-cell depletion might reduce the risk of GVHD whileretaining a GVL effect. In particular, selective CD8+ T-cell depletion has beenreported to reduce the risk of GVHD without losing the GVL effect (22,23).Additionally, it has been suggested that T-cell depletion in recipients of unre-lated BMTs might reduce the risk of GVHD without losing the GVL effect(24). Strategies in which T-cell depletion is used to reduce the risk of acuteGVHD have also been described, but T cells are then subsequently infused(“add back”) to generate a GVL effect (25,26). However, no large multicentertrial has investigated these strategies and such data remains preliminary.

In summary, data concerning T-cell depletion demonstrates that manipu-lating the cellular composition of the allogeneic marrow graft can influencethe risk of leukemic relapse. These powerful data confirm that the cellsthemselves, specifically the donor T-cells, have the capacity to mount anantileukemic effect.

Given the relationship of clinical GVHD with the GVL effect and becauseT-cell depletion reduces the GVL effect, it is clear that T-cells are criticalmediators in the GVL effect. The precise cellular mechanism, however,remains unknown. In particular, a fundamental question is whether the GVLeffect is independent of the GVHD effect. Thus, is the GVL effect simplyimmunologic GVHD directed against alloantigens shared by host tissues andleukemia cells, or, alternatively, are there donor cells that specifically recog-nize tumor antigens and generate the GVL effect? The association of clinicalGVHD with the GVL effect would strongly imply that the GVL effect is sim-ply an alloantigen reaction directed against all host tissues, both normal andleukemic. However, abundant data exist suggesting that it is possible to sepa-

16 Bolwell

rate the GVL effect from the alloantigen GVHD reaction (27–33). Both naturalkiller cells and different T-cell subsets, may have a role in the GVL effect(34–35). An ongoing up-to-date elusive goal is to harness the GVL effect andminimize clinical GVHD toxicity.

Theoretically, if one accepts the hypothesis that the GVL effect operatesthrough mechanisms other than simple alloantigen recognition, one mightdecrease the morbidity and mortality of GVHD if a state of immune tolerancecould be obtained. Therefore, if both donor and recipient cells are present in agiven host, without clinical GVHD, then a state of immune tolerance wouldtheoretically exist and hopefully the donor cells might still be able to generatea GVL effect. This situation has been described clinically and is known asmixed hematopoietic chimerism. It is generally defined as the coexistence ofboth donor and recipient cells after alloBMT.

Mixed hematopoietic chimerism has long been known to exist after trans-plantation, with conflicting clinical implications. Although some authorshave found that the detection of mixed chimerism may be associated withincreased risk of relapse in certain disease states (36–38), others have foundthat mixed hematopoietic chimerism after alloBMT is common and is notnecessarily associated with an increased risk of disease relapse (39–44). Forexample, Huss and Deeg described mixed hematopoietic chimerism inpatients with aplastic anemia or CML undergoing alloBMT; the incidence ofrejection was higher (but not significantly) in patients with aplastic anemiawith mixed chimeras. Intriguingly, among patients with CML, both overallsurvival and relapse-free survival were superior in mixed as opposed to com-plete chimeras (45). The development of stable mixed chimerism is theoreti-cally attractive; however, in clinical practice, the majority of patientsundergoing either an ablative or a nonmyeloablative allogeneic transplanta-tion clinically either evolve into a fully chimeric state or experience diseaserelapse (46).

4. DONOR LEUKOCYTE INFUSIONS

The use of the GVL effect as adoptive immunotherapy was proven conclu-sively with results obtained from a treatment known as DLI. The use of DLIwas pioneered in patients who relapsed after alloBMT. The theory wasstraightforward: if a patient relapsed after receiving an ablative alloBMT, andif that patient also did not have overt clinical GVHD, then the infusion of addi-tional donor cells (DLI) might be sufficient to produce a cellular immunothera-peutic effect and result in clinical remission. Initially, small studiesinvestigated the use of donor buffy coat leukocytes for patients with CML whorelapsed after alloBMT and found that a combination of α interferon and DLIresulted in both clinical and cytogenetic remissions (47,48).


The largest series examining the efficacy of DLI was a survey of 25 NorthAmerican BMT programs regarding their use of DLIs (49). One-hundredforty patients who received DLI relapsed after alloBMT. Diseases includedCML (n = 56), acute myeloid leukemia (AML) (n = 46), ALL (n = 15),myelodysplastic syndrome (MDS) (n = 6), non-Hodgkin’s lymphoma (NHL)(n = 6), multiple myeloma (n = 5), Hodgkin’s disease (n = 2), and other(n = 4). Donor leukocytes were obtained by leukopheresis from nonprimeddonors in a median of leukopheresis sessions during a median 7-d period. Theleukocytes were not manipulated in vitro. The cell yield was variable, butmost centers obtained a mean mononuclear cell (MNC) dose of approx5 × 108 MNCs per kilogram. CML responded best to DLIs. Of the 55 evalu-able patients with CMI, 60% achieved a complete response to DLI. Patientswho relapsed in chronic phase had a 74% chance of achieving a remissionwith DLI, patients in accelerated phase had a 33% response rate, and onlyone of six patients in blast crisis achieved a remission. Responses were moremodest in other diseases. Fifteen percent of the patients with AML whorelapsed achieved a complete response, 18% of patients with ALL, 40% ofMDS, and 50% of myeloma. Median time to remission in patients with CMLwas 85 d, and 34 d for patients with AML. Sixty percent of evaluable patientsdeveloped acute GVHD and 61% developed chronic GVHD. The mediantime to development of acute GVHD was 32 d. Eighteen percent developedpancytopenia related to DLI at a median of 21 d after infusion. This pancy-topenia resolved without treatment in 13 patients, resolved with granulocytecolony-stimulation factor (G-CSF) treatment in 8, resolved after bone mar-row boost in 2, and did not resolve in 3. Importantly, there was a clear corre-lation of disease response with development of clinical GVHD. Of 45evaluable completely responding patients, 42 developed acute GVHD, and 36of 41 developed chronic GVHD. The correlation of acute and chronic GVHDwith complete remission was statistically significant (p < 0.0001). Of the 23patients who did not develop either acute or chronic GVHD, only 3 obtaineda complete response to DLI.

This landmark study by Collins et al. conclusively demonstrated that adop-tive immunotherapy with DLIs in a large series of patients has the potential tolead to clinical and cytogenetic remissions in several diseases, with CMLappearing to be the most amenable to this therapy. Correlation of clinicalresponse with GVHD was strong.

Long-term follow-up of this cohort of patients was recently published (50).Seventy-three patients achieved a complete remission after DLI, and long-termfollow-up was available for 66, with a median follow-up of 32 mo. The proba-bility of survival at 1, 2, and 3 yr was 83%, 71%, and 61%, respectively.Patients with CML had 1-, 2-, and 3-yr survival rates of 87%, 76%, and 73%;for other diseases, survival probability at 1 and 2 yr was 77% and 65%. This

18 Bolwell

follow-up study concluded that the majority of remissions achieved with DLIpersist for years. Additional data have confirmed that DLI-induced remissionsare durable (51).

Although adoptive immunotherapy with DLI has excellent efficacy inpatients with CML, its efficacy is somewhat more modest in patients with lym-phoid malignancies. Fewer than 50% of patients with ALL or multiplemyeloma have been reported to achieve complete responses with DLI (55,56).

The use of G-CSF (filgrastim) for the treatment of relapse after allogeneic BMThas been described as a potential alternative to DLI (57,58). The largest seriesreported 14 patients relapsing after allogeneic transplantation (n = 5 CML, n = 5AML, n = 2 MDS, n = 1 Chronic lymphocytic leukemia [CLL], n = 1 ALL). Fil-grastim was given at 5µg/kg subcutaneously for 21 d. Of the participant, 43%achieved a complete response. Most patients developed chronic GVHD.

In summary, infusions of donor leukocytes induce remission in the absenceof any other therapy, proving that donor hematopoietic cells have the capacityto generate a GVL or GVT effect and resulting in meaningful and potentiallydurable clinical remissions.

5. NONMYELOABLATIVE ALLOGENEIC TRANSPLANTATIONS

We now know that a major component of cure in alloBMT is the GVTeffect. Indeed, some chemotherapy-resistant malignancies are potentiallycured by alloBMT. In this setting, the GVL effect may be the most importantcontributor to cure. Therefore, if one were to hypothesize that a given groupof patients might be cured by the GVL effect but would probably not benefitby high doses of chemotherapy, then why should such patients receive high-dose chemotherapy? Instead, it would make more sense to significantlydecrease the intensity of the pretransplant conditioning regimen and simplydeliver enough immunosuppressive therapy to prohibit graft rejection. Onewould then infuse donor hematopoietic cells and rely entirely on the GVTeffect to generate a tumor response. This is the fundamental concept of non-myeloablative (“mini”) allogeneic transplantation. To summarize, the ratio-nale is straightforward:

1. Some malignancies will not be cured by high-dose chemotherapy.2. Some malignancies may be cured by the GVT effect.3. If so, a minimal BMT preparative regimen would be desirable to prevent graft

rejection and minimize toxicity.4. Once the donor cells engraft, a GVT effect will hopefully result, leading to a

clinical remission.

Mini-transplantations are attractive for several reasons. A significant reduc-tion in the ablative preparative regimen will generate less acute toxicity forpatients undergoing alloBMT. Regimen-related toxicity of the traditional abla-


tive BMT preparative regimen can be severe (57), and it has been suggestedthat gastrointestinal mucosal damage can result in a release in inflammatorycytokines and actually stimulate the production of acute GVHD (58). A mini-transplantation could avoid many of these regimen-related toxicities. Reduc-tion in treatment-related morbidity and mortality might also facilitatealloBMT in older patients and patients with concurrent medical illnesses whomight be otherwise ineligible for fully ablative transplantation regimens.Additionally, some mini-transplantation regimens allow the procedure to beperformed as an outpatient, which is certainly attractive for some patientsundergoing transplantation.

Early data have suggested that a mini-transplantation is feasible and ofteneffective. Slavin et al. reported data on 26 patients with a variety of disorderswho underwent a nonmyeloablative transplant using fludarabine, anti-T-lym-phocyte globulin, and moderate dose busulfan (8 mg/kg) (59). The patients thenreceived G-CSF mobilized peripheral blood progenitor cell (PBPC) allogeneictransplantation, with cyclosporine as the sole GVHD prophylactic agent. Of the26 patients, 17 achieved complete chimerism and the remainder partialchimerism. Fourteen patients did not experience GVHD; severe GVHD was thecause of death for four patients. With a median follow-up of 8 mo, 85% ofpatients were alive and 81% were disease free. The conclusion was that non-myeloablative allogeneic transplants were well tolerated and offered excitingpromise. Giralt et al. reported on 15 patients undergoing nonmyeloablative stemcell transplantation to treat refractory AML or MDS (60). The nonmyeloablativeregimen was not uniform. GVHD prophylaxis consisted of cyclosporine andmethylprednisolone. Acute GVHD occurred in only three patients. Bone mar-row chimerism (greater than 90% donor cells) occurred in 75 patients by d 30after infusion. The procedure was well tolerated, and again, the conclusion wasthat nonmyeloablative transplantation offered exciting promise for a generallyelderly (median age 59 yr; range 27–71 yr) patient population.

Although reported follow-up for most nonmyeloablative allogeneic trans-plantations is relatively brief, several series do have somewhat mature follow-up. The M.D. Anderson experience with mini-transplantations was recentlyreported (61). Seventy-eight patients received fludarabine and melphalan as apreparative regimen, and eight received cladribine and melphalan. The medianpatient age was 52 yr (22–70 yr range). Most patients had advanced hematologicmalignancies. The median percentage of donor cells at 1 mo in 75 patients was100%. The probability of grades 2–4 and 3–4 acute GVHD was 0.49 and 0.29,respectively. Disease-free survival at 1 yr was 57% for patients in first remissionand 49% for patients with more advanced disease. The conclusion was that dis-ease control can be achieved by nonmyeloablative alloBMT.

McSweeney et al. reported on 45 patients with hematologic malignanciesin HLA-identical sibling donors receiving low-dose total body irradiation

20 Bolwell

(200 cGy) and cyclosporine plus mycophenolate for GVHD prophylaxis (62).Of the eligible patients 53% had the transplantation performed entirely as anoutpatient. Nonfatal graft rejection occurred in 20% of patients, and fludara-bine was later added to this preparative regimen to control graft rejection.The incidence of grade 2–3 acute GVHD was 47%. With a median follow-upof 417 d, overall survival was 67%, nonrelapse mortality was 7%, and relapsemortality was 27%. This minimally ablative regimen was extremely well tol-erated and demonstrated significant potential efficacy for elderly patients inneed of alloBMT.

Possibly the prototypic experience of mini-allogeneic transplantations wasreported by Childs et al. using mini-transplantations for metastatic renal cellcarcinoma (63). Renal cell carcinoma is refractory to chemotherapy but occa-sionally responds to immunologic therapy such as IL-2. Because somepatients respond to immunologic therapy and because the GVT effect is poten-tially powerful immunologic therapy, the goal of this trial was to use the GVTeffect to treat metastatic renal cell carcinoma. Nineteen patients with refrac-tory metastatic renal cell carcinoma received a preparative regimen ofcyclophosphamide and fludarabine followed by infusion of peripheral bloodstem cell allograft from HLA-identical siblings or a sibling with a one HLAantigen mismatch. The median follow-up was 402 d. Of the 19 patients, 9 sur-vived, 2 died of transplant-related causes, and 8 died of progressive disease.Of the 19 patients, 53% showed disease regression. Of these patients, 30% hada complete response and 70% had a partial response. There was a dramaticcorrelation of development of disease response with the development of clini-cal GVHD. Prolonged tumor regression occurred in the majority of patientswith grade 2–4 acute GVHD (9 of 10 patients) and in a minority of those with-out acute GVHD (1 of 9, p = .005). The conclusion was that mini-allogeneicstem cell transplantation can lead to sustained tumor regression in patientswith refractory metastatic renal cell carcinoma and was strongly associatedwith the development of clinical GVHD. This group has also emphasized thedevelopment of full donor chimerism of T-cells as a requirement for the GVTresponse (64).

Early data concerning the use of nonmyeloablative alloBMT is exciting. Theinitial toxicity is diminished compared with a traditional ablative transplanta-tion. However, it is not certain whether mini-transplantation will be as effectiveas fully ablative transplantation in controlling disease relapse. A retrospectivestudy comparing ablative and nonmyeloablative patients with hematologicmalignancies showed that survival was actually decreased in nonmyeloablativerecipients (52% vs 28%), with the majority of deaths secondary to diseaserelapse (65).

From January 2000 through September 2001, 20 evaluable patients receiveda nonmyeloablative alloBMT using a uniform preparative regimen at the


Cleveland Clinic Foundation. The patient characteristics and clinical outcomesare shown in Table 1. All patients were treated with a nonmyeloablative prepara-tive regimen consisting of fludarabine (30 mg/m2/d for 3 d), followed by TBI(200 cGy). The patients received donor PBPCs the day after TBI completion. Themedian patient age was 52 yr (range 28–62, with seven patients older than 60).All patients initially experienced prompt hematopoietic engraftment with neu-trophil recovery by day +10 after transplantation, and most patients were treatedas outpatients. As shown in Table 1, 6 patients achieved either a completeresponse or an excellent partial response, 8 patients were alive with progressivedisease, and 6 died. The most common cause of death was chronic GVHD. Infec-tion complications have been common, especially cytomegalovirus viremia (66).

Lineage-specific chimerism analysis has shown a significant difference inthe kinetics of peripheral blood-nucleated cell chimerism and T-cell

22 Bolwell

Table 1Clinical Characteristics of Patients Who Received Nonmyeloablative Transplantations

at the Cleveland Clinic Foundation 2000–2001

Age at Disease Status at Current No. Transplantation, Yr Transplantation Disease Status

1 57 AML–CR 2 Dead (acute GVHD)2 51 CLL–Refractory PR3 62 NHL–Rel 2 CR4 62 AML–Rel 2 CR5 38 MM–PR 2 Dead (chronic GVHD)6 62 CML–Chronic CR7 62 MDS–RA Dead (chronic GVHD)8 48 RCC–Progressive Dead (progressive disease)9 44 MM–PR 3 Progressive disease10 61 Waldenstrom’s–Refractory PR11 52 MDS–Unclass Dead (cGVHD)12 52 MFB–Stable CR13 57 CML–Chronic Progressive disease14 45 AML–CR 2 Dead (progressive disease)15 48 RCC–Progressive Progressive disease16 60 MDS–RAEB Progressive disease17 52 CML–Chronic Progressive disease18 45 MM–PR 3 Progressive disease19 49 MM–CR 1 Progressive disease20 48 CLL–Rel 2 Progressive disease

AML = acute myeloid leukemia; GVHD = graft vs host disease; CLL = chronic lymphocyticleukemia; PR = partial remission; NHL = non-Hodgkin’s lymphoma; Rel = relapse; CR = com-plete remission; MM = multiple myeloma; MDS = myelodysplastic syndrome; RA = refractoryanemia; RCC = renal cell carcinoma; MFB = myelofibrosis; RAEB = refractory anemia withexcess blasts.

chimerism. The mean peripheral buffy coat donor chimerism, using all nucle-ated cells, is 95% donor cells by day +21. In contrast, the kinetics of T-cellchimerism are more variable. Most patients ultimately achieve 100% donor T-cell chimerism; however, some patients experience rapid T-cell chimerism byday +49, and others do not experience complete T-cell chimerism until day+200 or longer. Five patients never achieved 100% donor T-cell chimerism,and all five patients relapsed. Fifteen patients achieved 100% T-cell chimerism;nevertheless, 5 of these patients developed progressive disease, including 3with multiple myeloma. All patients who achieved a complete or an excellentpartial response achieved 100% T-cell chimerism.

Several conclusions can be drawn from this data. First, the preparative regi-men described is associated with limited early toxicity and reduced treatment-related mortality compared with an ablative allogeneic transplantation.Peripheral blood–nucleated cell chimerism develops rapidly. Complete T-cellchimerism appears to be a requirement for an ongoing disease response,although the development of complete T-cell chimerism does not guaranteeabsence of progressive disease. The leading cause of death in our small cohortof patients is chronic GVHD. Our experience demonstrates the feasibility ofminitransplantations even for an elderly population. Continued obstacles aredisease relapse and clinical GVHD.

6. SUMMARY

The author believes that the GVL effect is the most potent immunologic ther-apy ever described in man. The GVL effect associated with DLI can save patientswho are relapsing after alloBMT who would otherwise be incurable. The clinicaloutcome data of nonmyeloblative allogeneic transplantation, although prelimi-nary, demonstrate the exciting therapeutic promise of the GVL effect.

The single biggest clinical problem of the GVL effect is its almost universalassociation with clinical GVHD. GVHD remains the major cause of morbidityand mortality after alloBMT. Those who perform basic and clinical researchinvolving alloBMT have a simple and straightforward research goal: to sepa-rate the GVL effect from GVHD in a clinically meaningful way. To date, wehave been unable to achieve this goal. As we become more knowledgeableabout the biochemical nature of the GVL effect, graft engineering, and thecauses and treatments of clinical GVHD, our ability to maximize the GVLeffect and minimize the toxicity of GVHD will result in better and more pow-erful oncologic immunotherapy.

REFERENCES1. Barnes D, Corp M, Loutit J, Neal F. Treatment of murine leukaemia with x-rays and homol-

ogous bone marrow: II. Br J Haematol 1957;3:241–252.2. De Vries M, Vos O. Treatment of mouse lymphosarcoma by total-body X irradiation and by

injection of bone marrow and lymphnode cells. J Natl Cancer Inst 1958;21:1117–1129.


3. Odom L, Githens J, Morse H, et al. Remission of relapsed leukaemia during a graft-versus-host reaction. Lancet 1978;2:537–540.

4. Weiden P, Flournoy N, Thomas E, et al. Antileukemic effect of graft-versus-host disease inhuman recipients of allogeneic-marrow grafts. N Engl J Med 1979;300:1068–1073.

5. Sullivan K, Storb R, Buckner C, et al. Graft-versus-host disease as adoptive immunotherapyin patients with advanced hematologic neoplasms. N Engl J Med 1989;320:828–834.

6. Sullivan K, Weiden P, Storb R, et al. Influence of acute and chronic graft-versus-host diseaseon relapse and survival after bone marrow transplantation from HLA-identical siblings astreatment of acute and chronic leukemia. Blood 1989;73:1720–1728.

7. Horowitz M, Gale R, Sondel P, et al. Graft-versus-leukemia reactions after bone marrowtransplantation. Blood 1990;75:555–562.

8. Jones R, Ambinder R, Piantadosi S, Santos G. Evidence of graft-versus-lymphoma effectassociated with allogeneic bone marrow transplantation. Blood 1991;77:649–653.

9. Kersey J, Weisdorf D, Nesbit M, et al. Comparison of autologous and allogeneic bone mar-row transplantation for treatment of high-risk refractory acute lymphoblastic leukemia. NEngl J Med 1987;317:461–467.

10. Weisdorf D, Nesbit M, Ramsay N, et al. Allogeneic bone marrow transplantation for acutelymphoblastic leukemia in remission: prolonged survival associated with acute graft-versus-host disease. J Clin Oncol 1987;5:1348–1355.

11. Doney K, Fisher L, Appelbaum F, et al. Treatment of adult acute lymphoblastic leukemia withallogeneic bone marrow transplantation. Multivariate analysis of factors affecting acute graft-versus-host disease, relapse, and relapse-free survival. Bone Marrow Trans 1991;7:453–459.

12. Passweg J, Tiberghien P, Cahn J, et al. Graft-versus-leukemia effects in T lineage and B lin-eage acute lymphoblastic leukemia. Bone Marrow Trans 1998;21:153–158.

13. Uzunel M, Mattsson J, Jaksch M, Remberger M, Ringdén O. The significance of graft-ver-sus-host disease and pretransplantation minimal residual disease status to outcome after allo-geneic stem cell transplantation in patients with acute lymphoblastic leukemia. Blood2001;98:1982–1985.

14. Mendoza E, Territo M, Schiller G, Lill M, Kinkel L, Wolin M. Allogeneic bone marrowtransplantation for Hodgkin’s and non-Hodgkin’s lymphoma. Bone Marrow Trans1995;15:199–303.

15. Ratanatharathorn V, Uberti J, Karanes C, et al. Prospective comparative trial of autologousversus allogeneic bone marrow transplantation in patients with non-Hodgkin’s lymphoma.Blood 1994;84:1050–1055.

16. Björkstrand B, Ljungman P, Svensson H, et al. Allogeneic bone marrow transplantation versusautologous stem cell transplantation in multiple myeloma: a retrospective case-matched studyfrom the european group for blood and marrow transplantation. Blood 1996;88:4711–4718.

17. Tricot G, Vesole D, Jagannath S, Holton J, Munshi N, Barlogie B. Graft-versus-myelomaeffect: proof of principle. Blood 1996;87:1196–1198.

18. Maraninchi D, Blaise D, Rio B, et al. Impact of T-cell depletion on outcome of allogeneicbone-marrow transplantation for standard-risk leukaemias. Lancet 1987;2:175–178.

19. Mitsuyasu R, Champlin R, Gale R, et al. Treatment of donor bone marrow with monoclonalanti-T-cell antibody and complement for the prevention of graft-versus-host disease. Ann IntMed 1986;105:20–26.

20. Goldman J, Gale R, Horowitz M, et al. Bone marrow transplantation for chronic myeloge-nous leukemia in chronic phase. Ann Int Med 1988;108:806–814.

21. Marmont A, Horowitz M, Gale R, et al. T-cell depletion of HLA-identical transplants inleukemia. Blood 1991;78:2120–2130.

22. Champlin R, Ho W, Gajewski J, et al. Selective depletion of CD8+ T lymphocytes for pre-vention of graft-versus-host disease after allogeneic bone marrow transplantation. Blood1990;76:418–423.

24 Bolwell

23. Champlin R, Jansen J, Ho W, et al. Retention of graft-versus-leukemia using selective deple-tion of CD8-positive T lymphocytes for prevention of graft-versus-host disease followingbone marrow transplantation for chronic myelogenous leukemia. Trans Proc1991;23:1695–1696.

24. Drobyski W, Ash R, Casper J, et al. Effect of T-cell depletion as graft-versus-host disease prophy-laxis on engraftment, relapse, and disease-free survival in unrelated marrow transplantation forchronic myelogenous leukemia. Blood 1994;83:1980–1987.

25. Barrett A, Mavroudis D, Tisdale J, et al. T cell-depleted bone marrow transplantation and delayedT cell add-back to control acute GVHD and conserve a graft-versus-leukemia effect. Bone Mar-row Trans 1998;21:543–551.

26. Johnson B, Truitt R. Delayed infusion of immunocompetent donor cells after bone marrow trans-plantation breaks graft-versus-host tolerance and allows for persistent antileukemic reactivitywithout severe graft-versus-host disease. Blood 1995;85:3302–3312.

27. Nieda M, Nicol A, Kikuchi A, et al. Dendritic cells stimulate the expansion of bcr-abl specificCD8+ T cells with cytotoxic activity against leukemic cells from patients with chronic myeloidleukemia. Blood 1998;91:977–983.

28. Krijanovski O, Hill G, Cooke K, et al. Keratinocyte growth factor separates graft-versus-leukemiaeffects from graft-versus-host disease. Blood 1999;94:825–831.

29. Jiang Y, Barrett A, Goldman J, Mavroudis D. Association of natural killer cell immune recoverywith a graft-versus-leukemia effect independent of graft-versus-host disease following allogeneicbone marrow transplantation. Ann Hematol 1997;74:1–6.

30. Glass B, Uharek L, Zeis M, et al. Graft-versus-leukaemia activity can be predicted by naturalcytotoxicity against leukaemia cells. Br J Haematol 1996;93:412–420.

31. Tsukada N, Kobata T, Aizawa Y, Yagita H, Okumura K. Graft-versus-leukemia effect and graft-versus-host disease can be differentiated by cytotoxic mechanisms in a murine model of allo-geneic bone marrow transplantation. Blood 1999;93:2738–2747.

32. Weiss L, Weigensberg M, Morecki S, et al. Characterization of effector cells of graft vs leukemiafollowing allogeneic bone marrow transplantation in mice inoculated with murine B-cellleukemia. Cancer Immunol Immunother 1990;31:236–242.

33. Pan L, Teshima T, Hill G, et al. Granulocyte colony-stimulating factor-mobilized allogeneic stemcell transplantation maintains graft-versus-leukemia effects through a perforin-dependent path-way while preventing graft-versus-host disease. Blood 1999;93:4071–4078.

34. Hsieh M, Korngold R. Differential use of FasL- and perforin-mediated cytolytic mechanisms byT-cell subsets involved in graft-versus-myeloid leukemia responses. Blood 2000;96:1047–1055.

35. Jiang Y, Kanfer E, Macdonald D, et al. Graft-versus-leukaemia following allogeneic bone marrowtransplantation: emergence of cytoxic T lymphocytes reacting to host leukaemia cells. Bone Mar-row Trans 1991;8:253–258.

36. Bertheas M, Lafage M, Levy P, et al. Influence of mixed chimerism on the results of allogeneicbone marrow transplantation for leukemia. Blood 1991;78:3103–3106.

37. Mackinnon S, Barnett L, Heller G, O’Reilly R. Minimal residual disease is more common inpatients who have mixed T-cell chimerism after bone marrow transplantation for chronic myel-ogenous leukemia. Blood 1994;83:3409–3416.

38. Delage R, Soiffer R, Dear K, Ritz J. Clinical significance of bcr-abl gene rearrangement detectedby polymerase chain reaction after allogeneic bone marrow transplantation in chronic myeloge-nous leukemia. Blood 1991;78:2759–2767.

39. van Leeuwen J, van Tol M, Joosten A, et al. Persistence of host-type hematopoiesis after allo-geneic bone marrow transplantation for leukemia is significantly related to the recipient’s ageand/or the conditioning regimen, but it is not associated with an increased risk of relapse. Blood1994;83:3059–3067.

40. Petz L, Yam P, Wallace R, et al. Mixed hematopoietic chimerism following bone marrowtransplantation for hematologic malignancies. Blood 1987;70:1331–1337.


41. Fishleder A, Bolwell B, Lichtin A. Incidence of mixed chimerism using busulfan/cyclophos-phamide containing regimens in allogeneic gone marrow transplantation. Bone MarrowTrans 1992;9:293–297.

42. Ely P, Miller W. bcr/abl mRNA detection following bone marrow transplantation for chronicmyelogenous leukemia. Transplantation 1991;52:1023–1028.

43. Miyamura K, Tahara T, Tanimoto M, et al. Long persistent bcr-abl positive transcriptdetected by polymerase chain reaction after marrow transplant for chronic myelogenousleukemia without clinical relapse: a study of 64 patients. Blood 1993;81:1089–1093.

44. van Leeuwen J, van Tol M, Joosten A, et al. Mixed T-lymphoid chimerism after allogeneicbone marrow transplantation for hematologic malignancies of children is not correlated withrelapse. Blood 1993;82:1921–1928.

45. Huss R, Deeg H, Gooley T, et al. Effect of mixed chimerism on graft-versus-host disease,disease recurrence and survival after HLA-identical marrow transplantation for aplastic ane-mia or chronic myelogenous leukemia. Bone Marrow Trans 1996;18:767–776.

46. Theil K, Warshawsky I, Tubbs R, et al. Dynamics of T-Cell (CD3+) mixed chimerism afternon-myeloablative allogeneic peripheral blood stem cell transplant (PBSCT). Proc Am Soci-ety Hem 2001;98:189a.

47. Drobyski W, Keever C, Roth M, et al. Salvage immunotherapy using donor leukocyte infu-sions as treatment for relapsed chronic myelogenous leukemia after allogeneic bone marrowtransplantation: efficacy and toxicity of a defined T-cell dose. Blood 1993;82:2310–2318.

48. Kolb H, Mittermuller J, Clemm C, et al. Donor leukocyte transfusions for treatment of recur-rent chronic myelogenous leukemia in marrow transplant patients. Blood 1990;76:2462–2465.

49. Collins R, Shpilberg O, Drobyski R, et al. Donor leukocyte infusions in 140 patients withrelapsed malignancy after allogeneic bone marrow transplantation. J Clin Oncol1997;15:433–444.

50. Porter D, Collins R, Shpilberg O, et al. Long-term follow-up of patients who achieved com-plete remission after donor leukocyte infusions. Biol Blood Marrow Trans 1999;5:253–261.

51. Dazzi F, Szydlo R, Cross N, et al. Durability of responses following donor lymphocyte infu-sions for patients who relapse after allogeneic stem cell transplantation for chronic myeloidleukemia. Blood 2000;96:2712–2716.

52. Collins R, Goldstein S, Giralt S, et al. Donor leukocyte infusions in acute lymphocyticleukemia. Bone Marrow Trans 2000;26:511–516.

53. Salama M, Nevill T, Marcellus D, et al. Donor leukocyte infusions for multiple myeloma.Bone Marrow Trans 2000;26:1179–1184.

54. Lokhorst H, Schattenbert A, Cornelissen J, Thomas L, Verdonck L. Donor leukocyte infu-sions are effective in relapsed multiple myeloma after allogeneic bone marrow transplanta-tion. Blood 1997;90:4206–4211.

55. Giralt S, Escudier S, Kantarjian H, et al. Preliminary results of treatment with filgrastim Forrelapse of leukemia and myelodysplasia after allogeneic bone marrow transplantation. NEngl J Med 1993;329:757–760.

56. Bishop M, Tarantolo S, Pavletic Z, et al. Filgrastim as an alternative to donor leukocyte infu-sion for relapse after allogeneic stem-cell transplantation. J Clin Oncol 2000;18:2269–2272.

57. Bearman S, Appelbaum F, Buckner C, et al. Regimen-related toxicity in patients undergoingbone marrow transplantation. J Clin Oncol 1988;6:1562–1568.

58. Hill G, Crawford J, Cooke K, et al. Total body irradiation and acute graft-versus-host disease:the role of gastrointestinal damage and inflammatory cytokines. Blood 1997;90:3204–3213.

59. Slavin S, Nagler A, Naparstek E, et al. Nonmyeloablative stem cell transplantation and celltherapy as an alternative to conventional bone marrow transplantation with lethal cytoreduc-tion for the treatment of malignant and nonmalignant hematologic diseases. Blood1998;91:756–763.

26 Bolwell

60. Giralt S, Estey E, Albitar M, et al. Engraftment of allogeneic hematopoietic progenitor cellswith purine analog-containing chemotherapy: harnessing graft-versus-leukemia withoutmyeloablative therapy. Blood 1997;89:4531–4536.

61. Giralt S, Thall P, Khouri I, et al. Melphalan and purine analog-containing preparative regi-mens: reduced-intensity conditioning for patients with hematologic malignancies undergo-ing allogeneic progenitor cell transplantation. Blood 2001;97:631–637.

62. McSweeney P, Niederwieser D, Shizuru J, et al. Hematopoietic cell transplantation in olderpatients with hematologic malignancies: replacing high-dose cytotoxic therapy with graft-versus-tumor effects. Blood 2001;97:3390–3400.

63. Childs R, Chernoff A, Contentin N, et al. Regression of metastatic renal-cell carcinoma afternonmyeloablative allogeneic peripheral-blood stem-cell transplantation. N Engl J Med2000;343:750–758.

64. Childs R, Clave E, Contentin N, et al. Engraftment kinetics after nonmyeloablative allo-geneic peripheral blood stem cell transplantation: full donor T-cell chimerism precedesalloimmune responses. Blood 1999;94:3234–3241.

65. Couriel D, Giralt S, De Lima M, et al. Graft-versus-host disease (GVHD) after non-myeloab-lative (NMA) versus myeloablative (MA) conditioning regimens in fully matched siblingdonor hematopoietic stem cell transplants (HSCT). Proc Am Society Hem 2000;1996:480a.

66. Mossad S, Avery R, Longworth D, et al. Infectious complications within the first year afternonmyeloablative allogeneic peripheral blood stem cell transplantation. Bone Marrow Trans2001;28:491–495.


1. INTRODUCTION

Monoclonal antibodies (MAbs) have been studied as a treatment forleukemia for approximately 20 years but have only recently been used success-fully. The history of MAb development has been reviewed extensively (1–4).The purpose of this chapter is to review the clinical data supporting the use ofunconjugated MAbs in the treatment of leukemia.

Unconjugated MAbs are those that are not conjugated to a toxin or radioiso-tope. These antibodies typically begin as mouse antibodies to cell surface deter-minants and are humanized to reduce immunogenicity and avoid humanantimouse antibodies (HAMA). By definition, MAbs depend on immunologicmechanisms such as antibody-dependent cellular cytotoxicity (ADCC) to effectcell death. Other mechanisms, such as complement-mediated cellular toxicityand apoptosis, have also been described, but the true nature of their antileukemiceffects remain uncertain. However, MAbs are generally well tolerated and easyto administer. Of course, these features mean little if the antibodies are not effi-cacious in the treatment of leukemia. There are several unconjugated MAbsavailable for use or in the late stages of clinical development.

29


3 Unconjugated Monoclonal Antibodies

Matt Kalaycio, MD

CONTENTS

INTRODUCTION

HUM195RITUXIMAB

ALEMTUZUMAB (CAMPATH-1H)OTHER ANTIBODIES

CONCLUSIONS

REFERENCES

2. HUM195

CD33 is a surface glycoprotein of uncertain function that is expressed on90% of clonogenic myeloid blasts (5). There is no known expression of CD33outside hematopoietic tissue and, importantly, it is not expressed on pluripo-tential stem cells. Unlike some other antigens, however, CD33 is internalizedon binding, thus limiting exposure of any antigen-antibody complexes to thereticuloendothelial system. However, complement-mediated cytotoxicity hasbeen demonstrated in leukemia cell lines (6).

M195 was the first murine anti-CD33 antibody generated. In a dose-escalat-ing trial of M195 in patients with advanced acute myeloid leukemia (AML), 10patients were treated with M195 but no clinical responses were documented(7). However, the study demonstrated complete saturation of CD33 bindingsites with rapid modulation. The antibody was well tolerated, but HAMA weredetected in 67% of patients, which may have contributed to treatment failure.

To overcome HAMA, HuM195, a CDR-grafted (humanized) antibody, wasconstructed. In addition to having similar pharmacokinetics to M195, HuM195also demonstrated ADCC, probably due to the human IgG1 isotype framework(8). In another clinical trial, 13 patients with advanced AML were treated witha dose-escalating schedule. In this trial, however, no HAHA developed and onepatient experienced a reduction in marrow blasts at the highest dose level of 10mg/m2 (9). A second trial tested HuM195 at supersaturating doses. Tenpatients with advanced AML were treated at doses of 12, 24, or 36 mg/m2/d for4 d as well as a second cycle of treatment 2 wk later. Clinical responses wereseen in four patients, and one patient achieved a complete remission (10). Acytokine-release syndrome, characterized by fevers and rigors, was noted atthese higher dose levels. The investigators conducting the study theorized thata potential contributor to the low response rate was the lack of effector cells inthe setting of advanced leukemia.

To test the theory that HuM195 would work better in the setting of alower leukemic burden, a clinical trial was designed in which patients inremission of acute promyelocytic leukemia (APL) were treated withHuM195 after achieving complete remission with standard chemotherapy.APL is characterized by the reciprocal translocation of the PML oncogeneon chromosome 15 with the RARα gene on chromosome 17. The chimericPML/RARα gene can be monitored by polymerase chain reaction (PCR)techniques even in the setting of histologic remission. Of 22 patients in firstremission of APL but who were still PCR positive for PML/RARα, 11(50%) achieved molecular complete remission after treatment with HuM1953mg/m2 twice weekly for six doses (11). This study demonstrates proof ofprinciple that HuM195 is capable of eliminating minimal residual disease inpatients with APL.

30 Kalaycio

In a larger study of patients with advanced AML, HuM195 was adminis-tered to 49 patients with active leukemia. Patients were randomized to either a12-mg/m2 infusion or a 36-mg/m2 infusion administered weekly for 4 wk.Although no responses were seen in patients with more than 30% marrowblasts, the percentage of blasts decreased in 12 (60%) of 20 patients with initialblast counts between 5% and 30% and 2 patients (10%) achieved a completeremission (12). HuM195 was well tolerated, and only one patient experiencedinfusion-related toxicity that required therapy termination. The most commonside effects were self-limited fever, chills, and hypotension that usuallyrequired no specific therapy. The study demonstrates that HuM195 is a well-tolerated treatment with minimal, but observable, activity in patients withadvanced AML. Unfortunately, a subsequent prospective randomized trialcomparing standard chemotherapy combined with HuM195 versus standardchemotherapy alone in patients with relapsed refractory AML failed to show asignificant advantage for the combination treatment arm (13). The ultimaterole, if any, for HuM195 remains to be clarified by future clinical trials.

3. RITUXIMAB

CD20 is a B-lymphocyte surface antigen that does not internalize on bind-ing antibody (nonmodulating). Most B-cell lymphoproliferative disorders havehigh-level CD20 expression, but the function of CD20 is unknown. Rituximabis a chimeric anti-CD20 MAb that induces cytotoxicity when bound to surfaceCD20. Several potential mechanisms of action may explain rituximab’s thera-peutic activity, including ADCC, complement-mediated cytotoxicity, andapoptosis induction (14–17) Rituximab administered as single agent at a doseof 375 mg/m2 weekly for 4 wk results in an approx 50% response rate inrelapsed low-grade non-Hodgkin’s lymphoma (18). Rituximab is also well tol-erated in patients with lymphoma, with most side effects limited to an initialinfusion cytokine-release syndrome characterized by fever, chills, and, occa-sionally, hypotension. The excellent tolerability and remarkable activity of rit-uximab have prompted testing in nearly all CD20-positive lymphomas eitheralone or in combination with standard chemotherapy. Rituximab has also beentested against lymphoid leukemias with CD20 expression.

3.1. B-Cell Chronic Lymphocytic LeukemiaB-cell chronic lymphocytic leukemia (B-CLL) is a clonal neoplasm of mature

lymphocytes expressing a unique immunophenotype characterized by co-expression of CD5, CD19, and CD23. CD20 is also expressed but only at lowlevels (19). In early trials of rituximab for low-grade lymphomas, patients withsmall lymphocytic leukemia (SLL), the nodal equivalent of B-CLL, wereincluded. However, in contrast to the high response rates noted in follicular lym-

Chapter 3 / Unconjugated Monoclonal Antibodies 31

phomas with standard doses of rituximab, patients with relapsed SLL achievedremission rates of no more than 15% (18). When studies of patients withrelapsed B-CLL treated with standard doses of rituximab also demonstrated lowresponse rates (20,21), enthusiasm for rituximab as a treatment alternative for B-CLL/SLL waned further. In a study of 28 patients treated with rituximab 375mg/m2 weekly for four doses, only 7 patients (25%) achieved a partial remissionthat lasted a median of only 20 wk (22).

In addition to suggesting a limited role in the management of B-CLL/SLL,early studies of rituximab also revealed potentially serious, and previouslyundescribed, side effects. The early studies of rituximab given to patients withhigh white blood cell counts revealed a low incidence of a potentially fatal syn-drome characterized by rapid multisystem organ failure, severe coagulopathy,and tumor lysis syndrome (23–26). Patients with white blood cell countshigher than 100,000/µl are most at risk, and caution is warranted if rituximab isgiven to patients with high levels of circulating malignant cells (27). The whiteblood cell count should be reduced to a level lower than 100,000/µl wheneverpossible before treatment.

Cautious use of rituximab for patients with high white blood cell countsnotwithstanding, investigators at M.D. Anderson Cancer Center in Houston,Texas, sought to overcome the limited expression of CD20 on CLL lympho-cytes by escalating the dose of rituximab to maximally tolerated levels inpatients with advanced diseases (28). The investigators treated 50 patients withpreviously treated B-CLL (n = 40) or other low-grade lymphoid leukemias (n =10). Fifty-three percent of patients were refractory to fludarabine, and 43%were refractory to alkylating agents. In addition, 80% of the patients hadadvanced stage disease. All patients were treated with an initial intravenousdose of rituximab 375 mg/m2, which was followed by three weekly intra-venous infusions of a higher dose. The dose levels ranged from 500 to 2250mg/m2. Most toxicity was limited to the first dose and largely consisted of mildto moderate fever and chills. However, six patients (12%) experienced severetoxicity manifested as fever, chills, hypoxia, and hypotension. Importantly,only one of these patients had B-CLL. The other five patients had other lym-phoid leukemias, such as mantle cell leukemia. Thus, severe first-dose toxicitywas noted in 2% of patients with B-CLL, regardless of white blood cell countand 50% of patients with other lymphoid leukemias. There was no dose-limit-ing toxicity to subsequent higher doses of rituximab.

Overall, 40% of patients achieved a partial response (28). No patient achieveda complete remission. Importantly, a dose-response effect was noted when thepatients at various dose levels were analyzed (Fig. 1). The response rate of 80%at the highest dose levels, 75% if only patients with B-CLL are considered, is farhigher than was previously noted at standard doses of rituximab. Dose escala-tion, then, appears to overcome either the antibody resistance of the leukemic

32 Kalaycio

clone or the low-level expression of CD20. Unfortunately, the responses wereshort lived, with a median progression-free survival of 8 mo.

The improved efficacy of higher doses of rituximab was further corrobo-rated by a study reported by Byrd and colleagues, who designed a stepped-dose schedule of rituximab to reduce the incidence of severe first infusionreactions (29). The first dose of rituximab was 100 mg/m2. Subsequent cohortsof patients were then treated with either 250 or 375 mg/m2 intravenously threetimes a week, beginning 2 d after the initial 100 mg/m2 dose and continuing for12 doses or 4 wk. After the initial infusion, rituximab could be safely adminis-tered during 1 h. The stepped-dose approach lowered the incidence of infu-sion-related toxicity, with only two patients experiencing grade III/IV toxicity.Thirty-three patients with either B-CLL or SLL were treated, 28 of whom werepreviously treated. In addition to one complete response, 14 patients (42%)achieved a partial response (29). Importantly, of the six previously untreatedpatients receiving rituximab on this protocol, five achieved a partial remission.This study and the one reported by O’Brien et al. (28) clearly demonstrated theclinical efficacy of rituximab in the treatment of B-CLL (Fig. 1). The two trialsalso demonstrated the difficulty of achieving a complete remission with single-agent rituximab in this often heavily pretreated patient population. Rituximabgenerally reduces the white blood cell count, reduces splenomegaly, andimproves both anemia and thrombocytopenia. However, rituximab is lesseffective in clearing the bone marrow of malignant cells and in reducing thesize of enlarged lymph nodes. As a result, remissions are usually short lived.

The relatively modest benefit of rituximab in previously treated B-CLL andthe intriguing activity noted in the six untreated patients in Byrd’s study (29)


Fig. 1. Dose-response effect of rituximab for patients with chronic lymphocytic leukemia.Data from refs. 22, 28, and 29.

suggest a potential role for rituximab earlier in the natural history of the dis-ease. Furthermore, rituximab’s excellent tolerability profile makes it an idealagent with which to investigate combination therapies. Indeed, preliminaryresults of two clinical trials that combine fludarabine, cyclophosphamide, andrituximab for patients with B-CLL have recently been reported.

In the first study, 43 patients with previously treated B-CLL were treatedfirst with rituximab 375 mg/m2 on day 1, followed by fludarabine 25 mg/m2

and cyclophosphamide 250 mg/m2 on days 2 through 4 (30). The dose of ritux-imab was increased to 500 mg/m2, and chemotherapy was started on day 1 forthe subsequent five cycles of treatment. Although only 21% of these patientswere resistant to fludarabine, 70% of patients achieved a remission and 14%achieved a complete remission. The toxicity of the regimen was different fromwhat would normally be expected with the combination of fludarabine andcyclophosphamide alone in this patient population. However, 12% of patientsdid develop hypotension with the first infusion of rituximab, further suggestingcaution on initiation of treatment. The second study used the same chemobio-therapeutic regimen described. For this study, however, the patients had previ-ously untreated B-CLL. Of the 135 evaluable patients, the median age was 57yr and the median β2-microglobulin level was 4 mg%. However, only 39% ofthe patients were Rai stage III or IV at enrollment (31). The first cycle of ritux-imab was generally well tolerated, but 8% of patients experienced grade III–IVfever, chills, or blood pressure changes.

An unprecedented 63% of these patients achieved a complete remissionwith this combination therapy (31). The overall remission was an impressive95%, and with a median follow-up of approximately 1 yr, the progression-freesurvival is more than 22 mo and the median survival has not yet been reached.These results compare favorably with results obtained in previously untreatedpatients with B-CLL treated with fludarabine and cyclophosphamide alone.With the two-drug regimen, 47% of patients achieved a complete remission,with an overall response rate of 53% (32). Rituximab works synergisticallywith chemotherapy in the clinic, an observation borne out of experimentalobservations (33). If the promising results from this preliminary analysis areconfirmed in a larger patient, population or if the improved remission ratestranslate into improved survival when compared with standard chemotherapeu-tic regimens, rituximab may become an important component of initial therapyfor patients with B-CLL.

3.2. Hairy Cell LeukemiaRituximab has also been explored as a treatment for hairy cell leukemia,

another low-grade B-cell lymphoproliferative disorder characterized by high-level CD20 expression (34). In a pilot study of patients with previously treatedactive hairy cell leukemia, nine patients were treated with standard doses of rit-

34 Kalaycio

uximab 375 mg/m2 weekly for 8 wk. Although no patients with lym-phadenopathy responded, four patients (50%) achieved a remission, three ofwhich were complete (35). The patients were too few and follow-up was tooshort to accurately determine remission duration, but the clinical activity of rit-uximab was demonstrated. In another study of patients with hairy cellleukemia, who were previously treated 14 evaluable patients were treated withrituximab 375 mg/m2 weekly for 4 wk. A remission was achieved in threepatients (21%), and a partial remission was achieved in one patient, for anoverall remission rate of 29% (36). No patient relapsed within the 9-momedian follow-up. These small preliminary studies suggest that rituximab hasmodest activity against previously treated hairy cell leukemia when adminis-tered in standard doses. Whether results would improve in less heavily pre-treated patients or with higher doses requires additional study.

The cumulative experience suggests that rituximab is an important new ther-apeutic alternative in patients with low-grade B-cell leukemias. Whetheradministered alone or as part of a combination treatment regimen, rituximabincreases remission rates with little toxicity. Patients who are not heavily pre-treated and are without bulky lymphadenopathy are those most likely to derivebenefit. Whether rituximab can improve survival rates is currently uncertain,but clear palliative benefit can be obtained in properly selected patients forwhom other options are limited.

4. ALEMTUZUMAB (CAMPATH-1H)

Most normal and malignant T and B lymphocytes express high levels ofCD52. CD52 is a small nonmodulating surface antigen that induces ADCC andbinds complement when bound with antibody (37).

Investigators at Cambridge Pathology developed the first rat anti-CD52 anti-body, CAMPATH-1M. This MAb-induced complement-mediated cytotoxicityin vitro but had little clinical activity (38). A subsequent rat MAb, CAMPATH-1G, induced ADCC and could deplete lymphocytes from the circulation in vivo(39). Finally, a genetically reshaped human IgG1 CD52 MAb, CAMPATH-1Hor alemtuzumab, was developed that mediates ADCC and demonstrates clinicalactivity. Alemtuzumab is a powerful immunosuppressant and effectively elimi-nates circulating lymphocytes. These properties led to the use of alemtuzumabas prophylactic treatment for graft vs host disease (GVHD) and to promote stemcell engraftment after allogeneic bone marrow transplantation (41–43). Morerecently, alemtuzumab has been used to treat patients with lymphoid malignancy(24). Several small studies and case reports suggest clinical activity, but noneconvincingly demonstrated a role for alemtuzumab in the treatment of B-celllymphoma (40,44,45). In contrast to rituximab, the lack of a direct apoptoticmechanism of action and the need for effector cells probably explain alem-tuzumab’s failure to effectively treat lymphadenopathy (46).


4.1. T-Cell Prolymphocytic LeukemiaUnlike the experience in B-cell lymphoma, alemtuzumab consistently

induces high remission rates in patients with T-cell prolymphocytic leukemia(T-PLL). In one study, 15 patients with refractory T-PLL were treated withalemtuzumab 10-mg intravenous (IV) test dose, followed by 30 mg three timesa week as tolerated for 6 wk. A complete remission was achieved in ninepatients (60%), and two other patients had a partial remission, for an overallremission rate of 73% (47). The median duration of remission was 6 mo. Otherthan an acute infusion-related syndrome associated with the test dose charac-terized by fever and rigors, alemtuzumab was well tolerated. However, seriousinfections complicated two courses of treatment, and three patients developedlocal Herpes simplex infections. Hematologic toxicity was generally mild, buttwo patients developed severe bone marrow aplasia (47).

A more recent larger study confirms the effectiveness of alemtuzumab in thetreatment of T-PLL. Alemtuzumab induced a remission in 75% of 36 evaluablepatients with advanced refractory and 2 patients with untreated T-PLL treatedat a dose of 30 mg iv three times a week after an initial dose escalation (48).Patients were treated to their maximal responses, and 23 (60%) achieved acomplete remission. Importantly, the remission rate was higher for patientswith diseases limited to blood, marrow, and spleen, with lower response ratesnoted in those patients with hepatic and central nervous system involvement.This study also demonstrated that alemtuzumab can induce remissions inpatients in relapse after an initial alemtuzumab-induced remission in as manyas 42% of patients. Unfortunately, there was median survival of only 10 mo inthis population of heavily pretreated patients.

In patients with advanced T-PLL, alemtuzumab was reasonably well toler-ated. Tumor lysis syndrome was not observed, and toxicity was limited to thefirst infusion syndrome of fever, rigors, and nausea. Alemtuzumab did result inprolonged and profound lymphopenia that predisposed patients to significantinfections, such as cryptococcal meningitis, cytomegalovirus (CMV) infection,Pneumocystis pneumonia, and Legionella pneumonia (48).

Therefore, the accumulated evidence suggests that alemtuzumab is anextremely effective treatment for T-PLL. Because T-PLL is relatively resistantto chemotherapy including purine analogs (49), alemtuzumab should be con-sidered as part of the initial therapy for these patients.

4.2. B-CLLAlemtuzumab has been studied in two small studies of patients with B-CLL.

In the first, 29 patients with advanced, relapsed, or refractory CLL were treatedwith alemtuzumab 30 mg iv three times a week for as many as 12 wk. Circulat-ing lymphocytes were eliminated in 28 patients but only 42% of patients wereable to achieve a remission by standard criteria. Only two patients experienced

36 Kalaycio

regression of lymphadenopathy (50). The first infusion of alemtuzumab wastypically complicated by fever and rigors, but subsequent infusions were moreeasily tolerated. Grade IV neutropenia was noted in 20% of patients, but infec-tions secondary to profound lymphocytopenia and immunosuppression werethe most serious toxicities encountered. Fifty-two opportunistic infectionsoccurred, 28 of which were serious.

In a second small study, seven patients with advanced B-CLL or B-PLLwere treated with alemtuzumab administered subcutaneously. After a test doseof 10 mg, patients were treated with 30 mg three times a week. Four patientsresponded, but serious infections (such as reactivated CMV infections)occurred in five patients (51). The early experience with these two trials sug-gested the potential for significant clinical benefit from alemtuzumab but alsothe potential for serious infectious toxicity that prompted antimicrobial pro-phylaxis in subsequent clinical trials.

The results from a large study of alemtuzumab in patients with advanced B-CLL have recently been reported. Patients who had been exposed to alkylatingagents and were refractory to fludarabine were treated with alemtuzumab begin-ning with an initial dose of 3 mg, followed in 2 d with a dose of 10 mg, and esca-lated to a dose of 30 mg administered three times a week for 4 to 12 wk,depending on response. Of the 93 evaluable patients treated in this study, 33%achieved at least a partial remission (52). Another 59% had antitumor responses(reductions in lymphocyte count, improved blood cell counts, resolution of con-stitutional symptoms, etc.) but failed to meet criteria for partial remission. Similarto the experience with rituximab, patients with bulky lymphadenopathyresponded less well, as did older patients and those with β 2 microglobulin levelsgreater than 5. The median time to progressive disease in the patients in whom aremission was achieved was approx 9 mo. These results are remarkable, given thepopulation of patients treated. In fact, based largely on the data from this study,the Food and Drug Administration (FDA) of the United States approved alem-tuzumab in spring 2001 for use in patients with purine analog refractory CLL.

In addition to significant clinical activity, alemtuzumab was relatively welltolerated in the aforementioned study. Premedication with acetaminophen andantihistamines were given to lessen the severity of infusion-related side effects,but fever, rigors, nausea and vomiting, and dyspnea were still common mani-festations of the first infusion syndrome (Fig. 2) (52). Despite early prophy-laxis against infectious diseases with cotrimoxazole and famciclovir, 27% ofpatients experienced grade 3–5 infections, most of which were pneumonias.Hematologic toxicity was also noted, although difficulties emerged as theinvestigators tried to distinguish toxicity from the hematopoietic failureinduced by the leukemia before treatment. Grade 3–4 anemia occurred in 47%,neutropenia in 70%, and thrombocytopenia in 52% of patients. Within 4 mo,most patients improved their blood cell counts.


Given alemtuzumab’s effectiveness in advanced disease, investigators havebegun to explore alemtuzumab’s potential role in patients with less advancedCLL. In a small study, nine patients with previously untreated B-CLL weretreated with alemtuzumab 30 mg three times a week after an initial rapid doseescalation. The first five patients received iv infusions, but the next four weretreated by subcutaneous injection. Of the nine patients treated, seven (78%)achieved a complete remission in the bone marrow, but only three out of eightpatients with enlarged lymph nodes achieved remission (53). Therefore, onlythree patients achieved complete remission. Of interest, the median responseduration had not been reached with short follow-up, even for patients who onlyachieved a partial remission. Whether administered subcutaneously or intra-venously, the first dose of alemtuzumab caused fevers and rigors and onepatient experienced a CMV infection. Several other preliminary studies sug-gest high remission rates in previously untreated patients, but larger random-ized trials are needed to confirm any purported survival advantage comparedwith currently available standard therapies (54,55).

5. OTHER ANTIBODIES

Other antibodies are in development or have been evaluated for the treat-ment of B-CLL and other low-grade lymphoproliferative disorders. Two ofthese, Hu1D10 and T101, are unconjugated. Hu1D10 is a humanized IgG1κdirected against human leukocytic atigen (HLA)-polymorphism that is usuallyexpressed on both normal and malignant B-lymphocytes. Hu1D10 inducescytotoxicity by inducing apoptosis in the absence of effector cells and comple-ment through transmembrane signaling (56).

T101 is an anti-CD5 antibody that demonstrated few antileukemic effectswhen tested in patients with advanced CLL. Unfortunately, toxic pulmonary

38 Kalaycio

Fig. 2. Incidence of side effects after treatment with alemtuzumab in patients with advancedchronic lymphocytic leukemia (CLL). Data from ref. 52.

reactions and limited efficacy inhibited the clinical development of what wouldbe a more targeted approach to B-CLL therapy (57).

CONCLUSIONS

After a slow start, monoclonal antibodies are becoming important tools inleukemia treatment. Although many early studies were disappointing regardingremission rates, those studies provided a framework on which subsequent stud-ies capitalized. The available body of literature provides several general con-cepts with regard to the clinical application of unconjugated MAbs:

1. Aggressive disease, such as acute leukemia or advanced B-CLL, responds poorlyto unconjugated antibodies administered alone. However, less advanced diseaseresponds better, especially if antibody is administered with chemotherapy. Evenin the absence of remission, these antibodies effectively clear leukemia cellsfrom the blood that may improve blood counts, reduce transfusion require-ments, and lessen constitutional symptoms even in the absence of an objectivemeasurable remission.

2. Unconjugated antibodies are generally well tolerated but have the potential forserious side effects when administered in the setting of a large tumor burden.The antilymphocyte antibodies may also induce profound immunosuppressionwith risk for infections, especially in the case of alemtuzumab.

3. Purely biologic therapies, such as MAbs, are capable of significant antileukemiceffects. This observation alone is enough to warrant additional research into themechanisms by which these antibodies work and into the potential for combina-tion with other biologic therapies that may enhance those mechanisms.

The specific role MAbs will play in the management of leukemia remains tobe determined. For now, unconjugated MAbs have been clearly demonstratedto benefit select patients with advanced leukemias if only temporarily. How-ever, the accumulated evidence to date suggests that they will someday be usedas part of a treatment strategy, likely in combination, in early stage patients tomaximize response rates and improve survival if not cure the disease. Coupledwith the promise of other biologic therapies discussed in this book, the hope ofcure does not appear as unattainable as it did even in the early 1990s.

REFERENCES1. Levy R. A perspective on monoclonal antibody therapy: where we have been and where we

are going. Sem Hematol 2000;37:43–46.2. Maloney DG. Advances in immunotherapy of hematologic malignancies. Curr Opin Hema-

tol 1998;5:237–243.3. Foon KA, Schroff RW, Bunn PA, Jr. Clinical applications of monoclonal antibodies for

patients with leukemia and lymphoma. Prog Clin Biol Res 1986;211:265–284.4. Hainsworth JD. Monoclonal antibody therapy in lymphoid malignancies. Oncologist

2000;5:376–384.


5. Legrand O, Perrot JY, Baudard M, et al. The immunophenotype of 177 adults with acutemyeloid leukemia: proposal of a prognostic score. Blood 2000;96:870–877.

6. McGraw KJ, Rosenblum MG, Cheung L, Scheinberg DA. Characterization of murine andhumanized anti-CD33, gelonin immunotoxins reactive against myeloid leukemias. CancerImmunol Immunother 1994;39:367–374.

7. Scheinberg DA, Lovett D, Divgi CR, et al. A phase I trial of monoclonal antibody M195 inacute myelogenous leukemia: specific bone marrow targeting and internalization of radionu-clide. J Clin Oncol 1991;9:478–490.

8. Caron PC, Co MS, Bull MK, Avdalovic NM, Quenn C, Scheinberg DA. Biological andimmunological features of humanized M195 (anti-CD33) monoclonal antibodies. CancerRes 1992;52:6761–6767.

9. Caron PC, Jurcic JG, Scott AM, et al. A phase 1B trial of humanized monoclonal antibodyM195 (anti-CD33) in myeloid leukemia: specific targeting without immunogenicity. Blood1994;83:1760–1768.

10. Caron PC, Dumont L, Scheinberg DA. Supersaturating infusional humanized anti-CD33 mon-oclonal antibody HuM195 in myelogenous leukemia. Clin Cancer Res 1998;4:1421–1428.

11. Jurcic JG, DeBlasio T, Dumont L, Yao TJ, Scheinberg DA. Molecular remission inductionwith retinoic acid and anti-CD33 monoclonal antibody HuM195 in acute promyelocyticleukemia. Clin Cancer Res 2000;6:372–380.

12. Feldman E, Kalaycio M, Schulman P, et al. Humanized monoclonal anti-CD33 antibodyHuM195 in the treatment of relapsed/refractory acute myelogenous leukemia (AML): pre-liminary report of a phase II trial. Proc Am Soc Clin Oncol 1999;18:4a.

13. Feldman E, Stone RM, Brandwein J, et al. Phase III randomized trial of an anti-CD33 mono-clonal antibody (HUM195) in combination with chemotherapy compared to chemotherapyalone in adults with refractory of first-relapse acute myeloid leukemia. Proc Am Soc ClinOncol 2002;21:261a.

14. Hofmeister JK, Cooney D, Coggeshall KM. Clustered CD20 induced apoptosis: src-familykinase, the proximal regulator of tyrosine phosphorylation, calcium influx, and caspase 3-dependent apoptosis. Blood Cells Molecules, Dis 2000;26:133–143.

15. Golay J, Zaffaroni L, Vaccari T, et al. Biologic response of B lymphoma cells to anti-CD20monoclonal antibody rituximab in vitro: CD55 and CD59 regulate complement-mediatedcell lysis. Blood 2000;95:3900–3908.

16. Clynes RA, Towers TL, Presta LG, Ravetch JV. Inhibitory Fc receptors modulate in vivocytoxicity against tumor targets. Nature Med 2000;6:443–446.

17. Shan D, Ledbetter JA, Press OW. Signaling events involved in anti-CD20-induced apoptosisof malignant human B cells. Cancer Immunol Immunother 2000;48:673–683.

18. McLaughlin P, Hagemeister FB, Grillo-Lopez AJ. Rituximab in indolent lymphoma: the sin-gle-agent pivotal trial. Sem Oncol 1999;26:79–87.

19. Caligaris-Cappio F. B-chronic lymphocytic leukemia: a malignancy of anti-self B cells.Blood 1996;87:2615–2620.

20. Ladetto M, Bergui L, Ricca I, Campana S, Pileri A, Tarella C. Rituximab anti-CD20 mono-clonal antibody induces marked but transient reductions of peripheral blood lymphocytes inchronic lymphocytic leukaemia patients. Med Oncol 2000;17:203–210.

21. Winkler U, Jensen M, Manzke O, Schulz H, Diehl V, Engert A. Cytokine-release syndrome inpatients with B-cell chronic lymphocytic leukemia and high lymphocyte counts after treatmentwith an anti-CD20 monoclonal antibody (rituximab, IDEC-C2B8). Blood 1999;94:2217–2224.

22. Huhn D, von Schilling C, Wilhelm M, et al. Rituximab therapy of patients with B-cellchronic lymphocytic leukemia. Blood 2001;98:1326–1331.

23. Dillman RO. Infusion reactions associated with the therapeutic use of monoclonal antibodiesin the treatment of malignancy. Cancer Metastasis Rev 1999;18:465–471.

40 Kalaycio

24. Dyer MJ. The role of CAMPATH-1 antibodies in the treatment of lymphoid malignancies.Sem Oncol 1999;26:52–57.

25. Kunkel L, Wong A, Maneatis T, et al. Optimizing the use of rituximab for treatment of B-cellnon-Hodgkin’s lymphoma: a benefit-risk update. Sem Oncol 2000;27:53–61.

26. Yang H, Rosove MH, Figlin RA. Tumor lysis syndrome occurring after the administration ofrituximab in lymphoproliferative disorders: high-grade non-Hodgkin’s lymphoma andchronic lymphocytic leukemia. Am Hematol 1999;62:247–250.

27. Byrd JC, Waselenko JK, Maneatis TJ, et al. Rituximab therapy in hematologic malignancypatients with circulating blood tumor cells: association with increased infusion-related sideeffects and rapid blood tumor clearance. J Clin Oncol 1999;17:791–795.

28. O’Brien SM, Kantarjian H, Thomas DA, et al. Rituximab dose-escalation trial in chroniclymphocytic leukemia. J Clin Oncol 2001;19:2165–2170.

29. Byrd JC, Murphy T, Howard RS, et al. Rituximab using a thrice weekly dosing schedule inB-cell chronic lymphocytic leukemia and small lymphocytic lymphoma demonstrates clini-cal activity and acceptable toxicity. J Clin Oncol 2001;19:2153–2164.

30. Garcia-Manero G, O’Brien S, Cortes J, et al. Combination fludarabine, cyclophosphamide,and rituximab for previously treated patients with chronic lymphocytic leukemia. Blood2000;96:757a.

31. Wierda W, O’Brien S, Albitar M, et al. Combined fludarabine, cyclophosphamide, and ritux-imab achieves a high complete remission rate as initial treatment for chronic lymphocyticleukemia. Blood 2001;98:771a.

32. Flinn IW, Byrd JC, Morrison C, et al. Fludarabine and cyclophosphamide with filgrastimsupport in patients with previously untreated indolent lymphoid malignancies. Blood2000;96:71–75.

33. Golay J, Xiao YM, Di Gaetano N, Dastoli G, Rambaldi A, Introna M. Fludarabine synergiseswith anti CD20 monoclonal antibody rituximab in complement mediated cell lysis. Blood2000;96:339a.

34. Hagberg H. Chimeric monoclonal anti-CD20 antibody (rituximab)—an effective treatmentfor a patient with relapsing hairy cell leukaemia. Med Oncol 1999;16:221–222.

35. Thomas DA, O’Brien S, Cortes J, et al. Pilot study of rituximab in refractory or relapsedhairy cell leukemia. Blood 1999;94:705a.

36. Nieva J, Bethel K, Baker T, Saven A. Phase II study of rituximab in the treatment of cladrib-ine-failed patients with hairy cell leukemia. Blood 2001;98:364a–365a.

37. Domagala A, Kurpisz M. CD52 antigen—a review. Med Sci Monitor 2001;7:325–331.38. Hale G, Bright S, Chumbley G, et al. Removal of T-cells from bone marrow for transplanta-

tion: a monoclonal antibody fixes human complement. Blood 1983;62:873–882.39. Dyer MJS, Hale G, Hayhoe FHJ, Waldmann H. Effects of CAMPATH-1 antibodies in vivo

in patients with lymphoid malignancies. Blood 1989;73:1431–1439.40. Hale G, Dyer MJ, Clark MR, et al. Remission induction in non-Hodgkin lymphoma with

reshaped human monoclonal antibody CAMPATH-1H. Lancet 1988;2:1394–1399.41. Hale G, Zhang MJ, Bunjes D, et al. Improving the outcome of bone marrow transplantation

by using CD52 monoclonal antibodies to prevent graft-versus-host disease and graft rejec-tion. Blood 1998;92:4581–4590.

42. Kottaridis PD, Milligan DW, Chopra R, et al. In vivo CAMPATH-1H prevents graft-versus-host disease following nonmyeloablative stem cell transplantation. Blood 2000;96:2419–2425.

43. Cull GM, Haynes AP, Byrne JL, et al. Preliminary experience of allogeneic stem cell trans-plantation for lymphoproliferative disorders using BEAM-CAMPATH conditioning: aneffective regimen with low procedure-related toxicity. Br J Haematol 2000;108:754–760.

44. Pangalis GA, Dimopoulou MN, Angelopoulou MK, Tsekouras CH, Siakantaris MP. Cam-path-1H in B-chronic lymphocytic leukemia: report on a patient treated thrice in a 3 yearperiod. Med Oncol 2000;17:70–73.


45. Lundin J, Osterborg A, Brittinger G, et al. CAMPATH-1H monoclonal antibody in therapyfor previously treated low-grade non-Hodgkin’s lymphomas: a phase II multicenter study.European Study Group of CAMPATH-1H Treatment in Low-Grade Non-Hodgkin’s Lym-phoma. J Clin Oncol 1998;16:3257–3263.

46. Dyer MJS, Osterborg A. The use of therapeutic monoclonal antibodies in chronic lympho-cytic leukemia. In: Cheson BD, ed. Chronic Lymphoid Leukemias. New York: MarcelDekker; 2001:335–352.

47. Pawson R, Dyer MJS, Barge R, et al. Treatment of T-cell prolymphocytic leukemia withhuman anti-CD52 antibody. J Clin Oncol 1997;15:2667–2672.

48. Dearden CE, Matutes E, Cazin B, et al. High remission rate in T-cell prolymphocyticleukemia with CAMPATH-1H. Blood 2001;98:1721–1726.

49. Mercieca J, Matutes E, Dearden C, MacLennan K, Catovsky D. The role of pentostatin in thetreatment of T-cell malignancies: analysis of response rate in 145 patients according to dis-ease subtype. J Clin Oncol 1994;12:2588–2593.

50. Osterborg A, Dyer MJS, Bunjes D, et al. Phase II multicenter study of human CD52 anti-body in previously treated chronic lymphocytic leukemia. J Clin Oncol 1997;15:1567–1574.

51. Bowen AL, Zomas A, Emmett E, Matutes E, Dyer MDC. Subcutaneous CAMPATH-1H infludarabine-resistant/relapsed chronic lymphocytic and B-prolymphocytic leukaemia. Br JHaematol 1997;96:617–619.

52. Keating MJ, Flinn I, Jain V, et al. Therapeutic role of alemtuzumab (CAMPATH-1H) inpatients who have failed fludarabine: results of a large international study. Blood2002;99:3554–3561.

53. Osterborg A, Fassa AS, Anagnostopoulos A, Dyer MJS, Catovsky D, Mellstedt H. Human-ized CD52 monoclonal antibody CAMPATH-1H as first-line treatment in chronic lympho-cytic leukemia. Br J Haematol 1996;93:151–153.

54. O’Brien SM, Thomas DA, Cortes J, et al. CAMPATH-1H for minimal residual disease inchronic lymphocytic leukemia. J Clin Oncol 2001;21:284a.

55. Lundin J, Kimby E, Bjorkholm M, et al. Phase II trial of subcutaneous anti-CD52 mono-clonal antibody alemtuzuzmab (Campath-1H) as first line treatment for patients with B-cellchronic lymphocytic leukemia (B-CLL). Blood 2002;100:768–773.

56. Kostelny SA, Link BK, Tso JY, et al. Humanization and characterization of the anti-HLA-DR antibody 1D10. Int J Cancer 2001;93:556–565.

57. Foon KA, Schroff RW, Bunn PA, et al. Effects of monoclonal antibody therapy in patientswith chronic lymphocytic leukemia. Blood 1984;64:1085–1093.

42 Kalaycio

1. INTRODUCTION

Drug immunoconjugates are a novel class of cell surface receptor-targetedagents. They consist of monoclonal antibodies (MAbs) covalently linked tocytotoxic drugs. The first application of this class of compounds is in the treat-ment of patients with acute myeloid leukemia (AML). In this review, we exam-ine the biology of AML and identify the rationale for the use of the anti-CD33antibody in the selective targeting of this disease. We look at the structure andfunction of calicheamicin and how it came to be linked successfully to the anti-CD33 antibody. The phase I and II clinical data are discussed, with particularattention to toxicities and mechanisms of resistance. We relate the clinicalinformation with the prior observations on AML biology and antibody-calicheamicin conjugate pharmacology. Finally, we suggest strategies toreduce the normal tissue toxicities and broaden the clinical use of this newexciting therapy. Lessons learned with this agent may be useful in the develop-ment and application of future drug immunoconjugates.

43


4 Drug Immunoconjugate Therapy of Acute Myeloid Leukemia

Arthur E. Frankel, MD, Bayard L. Powell, MD,Eli Estey, MD, and Martin S. Tallman, MD

CONTENTS

INTRODUCTION

ACUTE MYELOID LEUKEMIA

GEMTUZUMAB OZOGAMICIN—STRUCTURE AND FUNCTION

GEMTUZUMAB OZOGAMICIN CLINICAL TRIALS

PHARMACOKINETICS AND IMMUNE RESPONSE

TOXICITIES

CLINICAL RESPONSES

ONGOING AND PLANNED CLINICAL STUDIES

REFERENCES

2. ACUTE MYELOID LEUKEMIA

Patients with AML have excess myeloblasts in their marrow. Theseleukemic blasts frequently have clonal cytogenetic and immunophenotypicabnormalities. The focus of current chemotherapies or immunotherapies hasbeen to deplete these abnormalities. However, recent experimental observa-tions suggest that in most leukemias (other than acute promyelocytic leukemia[APL]), the self-renewing long-term proliferating stem cells have a phenotypeand physiology that is distinct from the majority of blasts. Only rare (1/1000 to1/106) mononuclear cells from these patients survive and proliferate in liquidculture or immunocompromised mice (1). These leukemic stem cells expressCD34 and lack myeloid differentiation antigens, including CD33 and CD38(2,3). They proliferate slowly and overexpress CD123 death-associated proteinkinase (DAPK) and interferon regulatory factor 1 (IRF-1) (4,5). In some butnot all cases, they lack the Thy1 antigen (CD90) expressed on normal stemcells (6). Different genetic lesions in these bland lineage-negative mononuclearcells can produce different patterns of differentiated offspring (7).

Because cytotoxic chemotherapies and many cell surface-targeted com-pounds selectively kill mature myeloblasts, how are durable remissionsobtained? Two hypotheses are presented. There may be some, albeit reduced,chemosensitivity and differentiation antigen density on the primitive leukemicstem cells. Consequently, repeated high doses of drugs or targeted proteinsmay kill a fraction of the repopulating cells and permit prolonged leukemiaremission. Alternatively, cytoreductions of mature myeloblasts may remove acritical feedback loop. Differentiating myeloid cells may produce cytokinesthat help the leukemic stem cells to survive and grow (8).

Therapies for AML also lead to recovery of normal polyclonal hematopoiesis.How is this achieved when the drugs and immunoconjugates injure normalmyeloid progenitors as well as leukemic progenitors? Again, we postulate threenonexclusive theories. The leukemic progenitors may be more sensitive thannormal progenitors due to altered signal transduction or apoptotic pathways inthe malignant cells. Alternatively, there may be many more normal stem cellsthan leukemic stem cells. If chemotherapy or immunotherapy depletes an equalnumber (rather than an equal fraction) of normal and neoplastic stem cells, thenormal stem cells may have a numeric regrowth advantage. Finally, leukemicblasts may secrete factors that inhibit normal hematopoiesis (9). With depletionof leukemic blasts, normal hematopoiesis may occur.

The ongoing clinical studies with molecularly targeted compounds such asgemtuzumab ozogamicin (Mylotarg™, Wyeth, NY) may help to elucidate themechanism for leukemia control and normal hematopoietic reconstitution.Ancillary studies with the blood or marrow from these patients may be usefulin identifying leukemia stimulatory or normal marrow-suppressive molecules.

44 Frankel et al.

3. GEMTUZUMAB OZOGAMICIN—STRUCTURE AND FUNCTION

The mouse hybridoma producing the IgG1 MAb p67 was prepared by immu-nizing a mouse with AML blasts and screening hybridoma supernatants for reac-tivity to AML cells but not B cells (10). The p67 antibody reacted with myeloid,differentiation antigen CD33. This antigen was also recognized by anti-My9 andM195 (11,12). CD33 is absent from platelets, T-cells, erythrocytes, B-cells, andearly normal stem cells (LTC-IC) (13). The antigen is present on peripheral bloodmonocytes and marrow myeloblasts, promyelocytes, and weakly on metamyelo-cytes and granulocytes. CD33 is also present on the myeloid progenitors (CFU-GM and CFU-GEMM) and 80% to 90% of AML and blast crisis chronicmyelogenous leukemia (CML) blasts. Thus, the p67 anti-CD33 antibody selec-tively targets AML blasts but not early normal stem cells. In APL, the antibodybinds the entire leukemic stem cell pool (14). For other AMLs, as noted,leukemic stem cells may still be affected if they have some CD33 expression or ifthey are dependent on cytokines produced by CD33-positive leukemic blasts.

The p67 antibody was radiolabeled with 131I and used for imaging and ther-apy of patients with AML (15). The radioimmunoconjugate was selective tomarrow and leukemic compartments in patients but was rapidly dehalo-genated. As a result, there was poor tumor dosimetry and significant deposi-tion of free radioiodine in the thyroid, stomach, and kidneys. Preclinical workshowed that the radioiodinated antibody-retained antigen affinity (Ka = 4 ×1010 M–1) but that rapid internalization of the antibody occurred with 38%removal of surface-bound antibody in 4 h (16). The antibody–antigen endocy-tosis properties of the p67 antibody make it attractive for construction of adrug immunoconjugate.

Before further study, the complementarity-determining regions of p67 weregrafted onto a human IgG4 framework (Fig. 1) (17) to reduce the risk ofimmunogenicity in patients. The human IgG4 isotype was chosen because ithas fewer Fc-dependent effector functions and is thus considered less likely tocause unwanted clinical reactions. The humanized p67 retained the affinity ofthe murine antibody for p67. It was weakly immunogenic in monkeys. Anti-bodies formed were against antihuman Ig framework epitopes.

The selection of the proper drug for drug immunoconjugates is critical. Thedrug must be potent—able to kill a cell with fewer than 1000 molecules in thecell. It must be linkable to antibody so that it retains bioactivity when released.A class of drugs that meets these requirements is the calicheamicins isolatedfrom the Kerrville, Texas, soil organism, Micromonospora echinospora ssp.Calichensis (18). These small molecular weight compounds are sequence spe-cific, minor groove binding, deoxyribonucleic acid (DNA) damaging antibi-otics. Calicheamicin γ1I, the parent compound, is shown in Fig. 2. It consists of

Chapter 4 / Drug Immunoconjugates and AML 45

a methyl trisulfide portion (the trigger), a enediyne moiety that aromatizes to a1,4-dehydrobenzene-diradical (the warhead), and a sugar aromatic ring back-bone—4,6-didoxy-4[hydroxyamino]-β-D-glucopyranoside A ring, N–O glyco-sidic linkage to the thio sugar B ring, 4-ethylamino sugar E-ring attached to theA ring, a persubstituted dimethoxy aromatic ring attached to the B ring, and a3-O-methyl-α-L-rhamnopyranoside D-ring (the targeting region). Bioreduc-tive cleavage by reaction with cell glutathione leads to β-addition of the result-ing thiol to the enone form to form dihydrothiophene. This compoundundergoes cyclization to the 1,4-diyl6, also called the 1,4-dehydrobenzene-diradical (Fig. 3) (19). This reactive species abstracts hydrogen from thedeoxyribose backbone when DNA is present. Interestingly, the H abstractionon the deoxyribose ring and subsequent oxidative strand scission, as shown inFig. 4, only occurs at sequence-specific cleavage sites on DNA. In particular,the sequence TCCT is required. Figure 5 shows the cleavage that generatesdouble-strand breaks (20). The drug is active at subpicogram/mL concentra-tions and is approx 1000-fold more active than doxorubicin against murinetumors (21). The DNA damage leads to cell death. The unique chemical prop-

46 Frankel et al.

Fig. 1. Illustration of humanized antibody displaying heavy and light chains of humanimmunoglobulin IgG4 (gray) and mouse anti-CD33 CDR grafts (white). Disulfide bondsare shown in black.

erties of the molecule made it an excellent candidate for linkage to antibodiesand selective release inside cells.

Conjugation of calicheamicin to humanized p67 antibody so that the drugwould be released only intracellularly was accomplished by taking advantageof the methyltrisulfide trigger. Because the thiol-leaving group is not requiredfor drug action, the methyltrisulfide trigger was derivatized with a mercaptohy-drazide to yield the hydrazide analogue. A dimethyl group was employed nearthe disulfide to reduce spontaneous and premature disulfide reduction in thebloodstream (22). An N-acetyl group was attached to the amino sugar at theethyl nitrogen, based on empirical experiments that measured activity versustoxicity of conjugates (22). The antibody was attached using the bifunctionallinker AcBut. This linker attaches to lysines on the antibody and forms ahydrazone with the hydrazide of the NAc-γ-calicheamicin DMH (Fig. 6). Thishydrazone is hydrolyzed in the acidic environment of the intracellular endo-some/lysosomes through which the antibody is routed after internalization(Fig. 7). The anti-CD33 calicheamicin (previously called CMA-676; now withthe generic name gemtuzumab ozogamicin and tradename Mylotarg) is selec-tively toxic to CD33-bearing normal and malignant myeloid cells.


Fig. 2. Chemical structure of calicheamicin γ11. The important aspects of its structure are

the methyltrisulfide (trigger), the diyne-ene moiety that aromatizes to a 1,4-dehydrobene-diradical (the warhead), and a sugar-aromatic ring backbone (the 4,6-didoxy-4 [hydrox-yamino]-β-D-glucopyranoside A ring, N-O glycosidic linkage to the thio sugar B ring, andattached to the A ring—the 4-ethylamino sugar E-ring and attached to the B ring, through amethyl,diacetyl iodinated aromatic ring, the 3-O-methyl-α-L-rhamnopyranoside D-ring).

4. GEMTUZUMAB OZOGAMICIN CLINICAL TRIALS

Gemtuzumab ozogamicin has been studied for treatment of relapsed AML inone phase I clinical trial and three phase II clinical trials. The phase I study wasdone at the Fred Hutchinson Cancer Research Center and the City of HopeNational Medical Center (23). Forty patients were treated with one to three dosesof 0.25 mg/m2 to 9 mg/m2 gemtuzumab ozogamicin 2-h intravenous (iv) infu-sions with 2 or more weeks between doses. Premedications included aceta-minophen and diphenhydramine. The median patient age was 54 yr (range 24 to73); the group included 21 men and 19 women. Half of the patients were in firstrelapse, and the remainder in second or greater relapse. Fourteen patients had pre-viously received allogeneic bone marrow transplantations (BMT) and 4 hadreceived autologous BMTs. Complex cytogenetic abnormalities were present inthe blasts of 22 patients, and single abnormalities were present in 7 patients (2each with inv[16] and t[9;11] and 1 each with t[8;21], t[4;11], and t[6;11]). Ninepatients had previous myelodysplasia syndrome (MDS).

The phase II studies were multi-institutional, and the data from the threestudies have been pooled because treatment with gemtuzumab ozogamicin wasthe same in each study (24–26). The 201 study in the United States and Canada

48 Frankel et al.

Fig. 3. Bioreductive cleavage by reaction with cell glutathione leads to β-addition of theresulting thiol to the enone form to form dihydrothiophene. This compound undergoescyclization to form the 1,4-diyl6 also called the 1,4-dehydrobenzene-diradical. This is thebioactive intermediate.

included 13 centers, the 202 study in Europe included 19 centers, and the 203study in both the United States and Europe included 21 centers. Seventy-twopatients were evaluable from the 201 study, 62 patients from the 202 study, and54 patients from the 203 study. All patients had to be in first relapse AML andolder than 18 years. In the 203 study, the patients had to be older than 60 years.In the 201 and 202 studies, the duration of first remission (CR1) had to be 6 moor more. In the 203 study, the CR1 had to be 3 mo or more. All patients had tohave a white blood cell count (WBC) less than 30,000/µL and no prior MDS.Prior BMT was only permitted in the 202 study. All patients had to have CD33expression on leukemic blasts (staining intensity > 4 × unstained cells on ≥ 80%of cells). Patients were scheduled to receive two infusions of gemtuzumabozogamicin at 9 mg/m2 given 14 d apart. To date, 188 patients are evaluable.The median age was 60 yr (range 22 to 87), with 56% (105) men and 44% (83)women. Most of the patients were white (93%). The mean duration of CR1was 11.1 mo. Most had undergone postremission consolidation therapy(94%), which frequently included high-dose cytarabine (67%). Favorable cyto-


Fig. 4. Cleavage mechanism of the TCCT site at the 5′C. The carbon-centered radicalabstracts a 5′H from the deoxyribose sugar. The sugar is then attacked by dioxygen to forma peroxyl radical that, in the presence of thiol, forms an aldehyde and causes strand scission.

genetics at relapse were rare (4% in the patients who had cytogenetic studiesperformed at relapse).

5. PHARMACOKINETICS AND IMMUNE RESPONSE

Serum concentrations of the hP67.6 and calicheamicin component of gem-tuzumab ozogamicin and peripheral blast saturation were measured in thephase I and II studies (23). The half-life of hP67.6 was approx 67 h, and thepeak concentration was proportional to the dose. The free calicheamicin area

50 Frankel et al.

Fig. 5. The calicheamicin double-stranded DNA cleavage site. The cleavage is at the 3′-Tand 5′-A of TCCT/AGGA.


Fig. 6. Illustration showing the molecular structure of gemtuzumab ozogamicin. ThehP67.6 is the humanized anti-CD33 antibody shown as a circle, which is attached throughan ε amino group of lysine to an Ac-But acid labile linker and, finally, to calicheamicin 1γ1.

Fig. 7. Cell intoxication by gemtuzumab ozogamicin: (a) binding to cell surface CD33; (b)internalization to endosomes; (c) cleavage of acid labile bond and release of calicheamicin;(d) diffusion through endosome or lysosome membrane bilayer to cytosol; (e) reduction ofdisulfide followed by diradical formation; (f) insertion into DNA and double-strand DNAcleavage; and (g) programmed cell death.

under the concentration curve was 0.3% of the hP67.6 area under the concen-tration curve. Saturation of circulating blasts was observed at 4 mg/m2 orgreater.

Antibodies to gemtuzumab ozogamicin were measured in all of the clinicaltrials. In the phase I study, one patient developed antibody to the calicheam-icin-linker complex after the third dose and a second patient developed anti-bodies to the calicheamicin-linker during the second dose of a second courseof gemtuzumab ozogamicin. In the phase II studies, no humoral immuneresponse to gemtuzumab ozogamicin occurred in the 188 evaluable patients.

6. TOXICITIES

Three classes of side effects were observed in the clinical trials. An acuteinfusion reaction occurred in most patients with fever and chills, and less com-monly, hypotension, hypertension, diarrhea, abdominal pain, headaches, dysp-nea, nausea, vomiting, and asthenia. Thirty-eight percent of patients had one ormore of these as grade 3 or 4 events. Symptoms occurred usually within 8 to12 h. The symptoms resembled those seen with other antibody- and recombi-nant protein-based therapeutics (27).

Myelosuppression occurred in all patients, and 97% and 99% of patientsshowed grade 3 to 4 neutropenia and thrombopenia, respectively. Secondary com-plications of severe (grades 3 and 4) infections (27%) and bleeding (14%) wereseen. The median time to recovery of neutrophils to more than 500/µL was 42 dfrom the first dose of gemtuzumab ozogamicin for remission patients. The mediantime to recovery of platelets to 25,000/µL was 36 d and 75 d from the first dose ofgemtuzumab ozogamicin for patients achieving a complete remission (CR) orcomplete remission, except for incomplete platelet recovery (CRp), respectively.Transient myelosuppression was expected, because normal committed myeloidprecursors express CD33 (28). The duration of myelosuppression was not signifi-cantly different from that seen with high-dose cytarabine salvage regimens (29).

Liver injury is the unique and most clinically significant drug-related toxicityobserved in the preapproval clinical trials and in postapproval studies (24,30,31).Grade 3 or 4 elevations in liver transaminases (AST and ALT) or in total bilirubinoccurred in 16% and 26% of patients, respectively. These signs of liver injurywere delayed, with maximal abnormalities 1 to 2 wk after therapy. A fraction ofpatients (2% in the phase II trials and up to 11% of some combination therapystudies) also developed a veno-occlusive disease (VOD)-like syndrome with livertenderness and enlargement, jaundice, and fluid retention. The incidence of liverdamage was increased by previous bone marrow transplantation (both autologousand allogeneic). Co-administration of other cytotoxic drugs and cytokines (suchas interleukin-11) may also predispose patients to the liver lesion (31). Ultra-sound, computed tomography scan, and magnetic resonance imaging may showreversal of portal flow consistent with portal hypertension. Rarely have patients

52 Frankel et al.

had transjugular liver biopsies that show sinusoidal cell apoptosis, increasedextracellular matrix, and centrilobular necrosis and congestion. Because CD33 isnot expressed by hepatocytes or sinusoidal endothelium, the hypothesis has beenadvanced that gemtuzumab ozogamicin reacts with the fixed macrophages in theliver Kupffer cells. The dying or activated Kupffer cells may release cytokinesthat activate stellate cells (32). Activated stellate cells secrete collagen and otherextracellular matrix material, which can lead to constriction of sinusoids causingportal hypertension and centrilobular ischemia. In mild cases, only transami-nasemia is seen, but in severe cases, a VOD-like syndrome ensues. Noapproaches to date have been effective as prophylaxis or treatment.

Despite these various side effects, most patients tolerate gemtuzumabozogamicin well. Mucositis was minimal, and no evidence of cardiac or centralnervous system (CNS) toxicity was seen.

7. CLINICAL RESPONSES

Among the 40 patients treated in the phase I trial, there were three CRs andfive CRps (23). Remissions occurred only in patients receiving at least twodoses and in patients showing at least 80% saturation of circulating blasts.

Based on these encouraging results, the phase II trials were undertaken(24,33). The overall remission rate was 31% (n = 188), with 15% CRs and 16%CRps. In the 201 study, the remission rate was 35%, with 18% CRs and 17%CRps (n = 72). In the 202 study, the remission rate was 33%, with 15% CRs and18% CRps (n = 62). In the 203 study, the remission rate was 24%, with 9% CRsand 15% CRps (n = 54). The lower remission rate in the 203 study may havebeen due to the older age and shorter CR1 in these patients. When analyzed forprognostic factors, age and duration of CR1, but not cytogenetics, predictedresponse rate. Patients who were younger than 60 yr had a remission rate (CR +CRp) of 31% compared with 28% for patients equal to or older than 60 yr. Sim-ilarly, patients with CR1 ≥ 12 mo had a remission rate of 39% (32 of 82) versus25% (26 of 106) for those with CR1 < 12 mo. The combination of older age andshort CR1 was as expected. The remission rate for patients younger than 60 yrwith CR1 ≥ 12 months was 39% (16 of 41). The remission rate for patients olderthan 60 yr with CR1 for less than 6 mo was 10% (2 of 21). Most patients whoachieved remission went on to other therapies. Thus, analysis of overall and dis-ease-free survival is complicated. Nevertheless, a fraction of patients in remis-sion were alive (30%) and disease-free (25%) at 30 mo (24). The type ofremission—CR versus CRp—did not influence relapse-free or overall survival.The total (remission plus nonresponding) patient median overall survival was5.5 mo, which compares favorably with other salvage regimens (29). One of thepatients in the 202 study had a t(9;22)-positive FAB M5 AML and responded togemtuzumab ozogamicin (33). A three-log cytoreduction was documented byquantitative polymerase chain reaction (PCR).


Since Food and Drug Administration (FDA) approval, gemtuzumabozogamicin has been used for a several patients with relapsed or refractoryAML. Forty-three patients treated at Wake Forest, Cornell, and New York Med-ical Colleges who had early relapse of AML (< 6 mo), refractory AML, AML insecond or greater relapse, CML blast crisis, or AML after MDS received twodoses of gemtuzumab ozogamicin 9 mg/m2 on days 1 and 15 (34). Side effectswere transient and similar to those observed in the phase I/II trials and includedinfusion reactions (2%), myelosuppression (95%), and hepatic injury (21%).The overall remission rate was 14% in all patients, with 9% CR and 5% CRp.In patients with relapsed or refractory AML (excluding the CML patients), theremission rate was 17%. This is similar to the 17% remission rate found in thephase II studies for elderly patients with CR1 duration of 3 to 12 mo. The meanduration of gemtuzumab ozogamicin remissions was 4 mo. Only one patientdid not have normal cytogenetics—a CRp patient who had t(5;6).

Clinical resistance to gemtuzumab ozogamicin may occur by several phar-macologic barriers. The protein conjugate may fail to reach sanctuary sitessuch as the CNS, testes, or other extramedullary sites. This is not truly cellularresistance but an explanation for the refractory disease in these patients. In thephase I clinical trial, relapses after gemtuzumab ozogamicin remissions wereobserved in these sites (23). Systemic gemtuzumab ozogamicin treatment of apatient with CNS disease failed to clear blasts from the cerebrospinal fluid(CSF), and drug levels were unmeasurable in the CSF (35). Alternatively, theleukemic blasts may be resistant to gemtuzumab ozogamicin because of rapidefflux by a drug transporter or failure of CD33 internalization. Multidrug-resis-tant (MDR-1+) AML cell lines were not sensitive to gemtuzumab ozogamicin,but the addition of multidrug-resistant modifiers restored sensitivity (36).Bernstein and colleagues have recently shown a correlation between CD33internalization and sensitivity to gemtuzumab ozogamicin (unpublished data).The importance of any of these hypothetical resistance mechanisms on gem-tuzumab ozogamicin clinical activity is presently unknown.

8. ONGOING AND PLANNED CLINICAL STUDIES

Based on the results from the phase II studies, gemtuzumab ozogamicin wasapproved by the FDA in May 2000 for the treatment of CD33-positive relapsedAML in patients equal to or older than 60 yr (37). Since approval, additionalstudies with Mylotarg have been initiated to evaluate different schedules, dif-ferent side-effect prophylaxis regimens, combinations with cytotoxicchemotherapy, and new indications.

An alternative schedule with day 1 and 8 treatments will be tested. Becauseof the several-day half-life of gemtuzumab ozogamicin, there is significant cir-culating drug remaining at day 8. To avoid higher than necessary levels, the

54 Frankel et al.

second dose will be reduced to 6 mg/m2. This new schedule is more consistentwith the current AML treatment regimens and should shorten the duration ofmyelosuppression.

Anecdotal reports suggest that steroids block the acute infusion reactions. Cor-ticosteroid prophylaxis (2 mg dexamethasone the day before and the morning oftherapy) will be tested at Wake Forest and several other sites. Combination ther-apy with cytarabine, cytarabine/daunorubicin, mitoxantrone, idarubicin/cytara-bine, daunorubicin/cytarabine/6TG, or cytarabine/mitoxantrone/amifostine willbe tested in phase I/II trials. A Wyeth-Ayerst–sponsored phase I/II multicentertrial, protocol 205, examines gemtuzumab ozogamicin on days 1 and 8 withcytarabine continuous infusion on days 1 to 7. Older patients (> 60 years) withboth de novo and relapsed/refractory disease will be treated. Gemtuzumabozogamicin doses will be 6 mg/m2 and 4 mg/m2 initially. The initial cytarabinedose will be 100 mg/m2. Dose escalation will proceed to gemtuzumab ozogam-icin 9 mg/m2 and 6 mg/m2, along with cytarabine 200 mg/m2. Another Wyeth-Ayerst–sponsored phase I/II trial, protocol 206, treats relapsed/refractory patientsand those younger than 60 yr with de novo AML with gemtuzumab ozogamicinon day 4, daunorubicin on days 1 to 3, and cytarabine on days 1 to 7. There isdose escalation. Twelve patients have been treated to date. Other studies willexplore different combination regimens including (a) gemtuzumab ozogamicinon days 1 and 8, with five daily doses of cytarabine 1 gm/m2; (b) gemtuzumabozogamicin on days 7 and 14, with cytarabine 3 gm/m2 on days 1 to 5; (c) gem-tuzumab ozogamicin on days 1 and 14, with mitoxantrone; (d) gemtuzumabozogamicin on day 3 with cytarabine 2 gm/m2, mitoxantrone, and amifostine;and (e) gemtuzumab ozogamicin plus daunorubicin, cytarabine, and 6TG. Thus,soon we should have a wealth of information on the safety and efficacy of com-bining gemtuzumab ozogamicin with other cytotoxic drugs.

Gemtuzumab ozogamicin is also being tested as a single agent in severalnew disease states, including pediatric-relapsed AML, MDS, acute lympho-cytic leukemia (ALL), and elderly de novo AML. The Wyeth-Ayerst pediatricAML protocol 102 has accrued 28 patients, including 9 evaluable at 9 mg/m2

for two doses. Because of a transient grade 4 transaminase elevation, addi-tional patients are being accrued at 7.5-mg/m2 doses. The trial is ongoing, butthe response rate resembles that seen in adults. In the Wyeth-Ayerst multicen-ter protocol 207, intermediate and high-risk MDS patients receive a singledose or two doses of gemtuzumab ozogamicin followed by up to three addi-tional doses at monthly or greater intervals. Nine patients have been enrolled,and marrow blast cytoreductions have been seen. It is too early to assessresponse. Elderly patients with de novo AML will be treated as noted in proto-cols 205 and 206 and the ECOG study. In addition, these elderly patients withde novo AML are being treated at M.D. Anderson. Patients with normal cyto-genetics have had a 50% CR rate (Estey, unpublished data). However, at M.D.


Anderson, unlike the phase II trials in which older patients with poor-risk cyto-genetics had an overall remission rate of 30%, patients with complex cytoge-netic abnormalities or chromosome 5 or 7 changes have had a CR rate of 10%.Additional sites performing studies of single-agent gemtuzumab ozogamicinin elderly patients with AML include EORTC, Northwestern, Rush, and theMRC. Patients with CD33-positive ALL are being treated with single-agentMylotarg salvage protocols, and remissions have been observed. Patients withAML in remission are being randomized to receive Mylotarg as part of consol-idation therapy and as part of the pretransplantation or posttransplantation con-ditioning regimens. These trials should define new disease indications forgemtuzumab ozogamicin.

The next decade should see a better understanding of mechanisms of resis-tance and toxicity to this novel agent, as well as provide evidence for the opti-mal disease setting and combination regimens for patients with CD33-positivehematologic malignancies.

REFERENCES1. Lapidot T, Fajerman Y, Kollet O. Immune-deficient SCID and NOD/SCID mice models as

functional assays for studying normal and malignant human hematopoiesis. J Mol Med1997;75:664–673.

2. Brendel C, Neubauer A. Characteristics and analysis of normal and leukemic stem cells: cur-rent concepts and future directions. Leukemia 2000;14:1711–1717.

3. Bonnet D, Dick JE. Human acute myeloid leukemia is organized as a hierarchy that origi-nates from a primitive hematopoietic cell. Nat Med 1997;3:730–737.

4. Jordan CT, Upchurch D, Szilvassy SJ, et al. The interleukin-3 receptor alpha chain is a uniquemarker for human acute myelogenous leukemia stem cells. Leukemia 2000;14:1777–1784.

5. Guzman ML, Upchurch D, Grimes B, et al. Expression of tumor-suppressor genes interferonregulatory factor 1 and death-associated protein kinase in primitive acute myelogenousleukemia cells. Blood 2001;97:2177–2179.

6. Blair A, Hogge DE, Ailles LE, Lansdorp PM, Sutherland HJ. Lack of expression of Thy-1(CD90) on acute myeloid leukemia cells with long-term proliferative ability in vitro and invivo. Blood 1997;89:3104–3112.

7. Pereira DS, Dorrell C, Ito CY, et al. Retroviral transduction of TLS-ERG initiates a leuke-mogenic program in normal human hematopoietic cells. Proc Natl Acad Sci U S A1998;95:8239–8244.

8. Demetri GD, Griffin JD. Hemopoietins and leukemia. Hematol Oncol Clin North Am1989;3:535–553.

9. Lichtman MA. Interrupting the inhibition of normal hematopoiesis in myelogenousleukemia: a hypothetical approach to therapy. Stem Cells 2000;18:304–306.

10. Andrews RG, Torok-Storb B, Bernstein ID. Myeloid associated differentiation antigens onstem cells and their progeny identified by monoclonal antibodies. Blood 1983;62:124–131.

11. Griffin JD, Linch D, Sabbath K, Larcom P, Schlossman SF. A monoclonal antibody reactivewith normal and leukemic human myeloid progenitor cells. Leuk Res 1984;8:521–534.

12. Tanimoto M, Scheinberg DA, Cordon-Cardo C, Huie D, Clarkson BD, Old LJ. Restrictedexpression of an early myeloid and monocytic cell surface antigen defined by monoclonalantibody M195. Leukemia 1989;3:339–348.

56 Frankel et al.

13. Andrews RG, Takahashi M, Segal G, Powell JS, Bernstein ID, Singer JW. The L4F3 antigenis expressed by unipotent and multipotent colony-forming cells but not by their precursors.Blood 1986;68:1030–1035.

14. Turhan AG, Lemoine FM, Debert C, et al. Highly purified primitive hematopoietic stem cellsare PML-RARA negative and generate nonclonal progenitors in acute promyelocyticleukemia. Blood 1995;85:2154–2161.

15. Appelbaum FR, Matthews DC, Eary JF, et al. The use of radiolabeled anti-CD33 antibody toaugment marrow irradiation prior to marrow transplantation for acute myelogenousleukemia. Transplantation 1992;54:829–833.

16. van der Jagt RH, Badger CC, Appelbaum FR, et al. Localization of radiolabeled antimyeloidantibodies in a human acute leukemia xenograft tumor model. Cancer Res 1992;52:89–94.

17. Parren PW. Preparation of genetically engineered monoclonal antibodies for humanimmunotherapy. Hum Antibodies Hybridomas 1992;3:137–145.

18. Lee MD, Dunne TS, Chang CC, et al. Calicheamicins, a novel family of antitumor antibi-otics. 4. Structure elucidation of calicheamicins β1

Br, γ1Br, α2

I, α3I, β1

I, γ1I, and δ1

I. J AmChem Soc 1992;114:985–997.

19. De Voss JJ, Hangeland JJ, Townsend CA. Characterization of the in vitro cyclization chemistryof calicheamicin and its relation to DNA cleavage. J Am Chem Soc 1990;112:4554–4556.

20. Zein H, Poncin M, Nilakantan R, Ellestad GA. Calicheamicin γ1I and DNA: molecular

recognition process responsible for site-specificity. Science 1989;244:697–699.21. Zein N, Sinha A, McGahren WJ, Ellestad GA. Calicheamicin γ1

I: an antitumor antibiotic thatcleaves double-stranded DNA site specifically. Science 1988;240:1198–1201.

22. Hinman LM, Hamann PR, Wallace R, Menendez AT, Durr FE, Upeslacis J. Preparation andcharacterization of monoclonal antibody conjugates of the calicheamicins: a novel andpotent family of antitumor antibiotics. Cancer Res 1993;53:3336–3342.

23. Sievers EL, Appelbaum FR, Spielberger RT, et al. Selective ablation of acute myeloidleukemia using antibody-targeted chemotherapy: a phase I study of an anti-CD33 calicheam-icin immunoconjugate. Blood 1999;93:3678–3684.

24. Sievers EL, Larson RA, Stadtmauer EA, et al. Efficacy and safety of Mylotarg (gemtuzumabozogamicin, CMA-676) in patients with CD33-positive acute myeloid leukemia in firstrelapse. J Clin Oncol, in press.

25. Stadtmauer EA, Larson RA, Sievers EL, et al. An updated report of the efficacy and safety ofgemtuzumab ozogamicin in 188 patients with acute myeloid leukemia in first relapse. ProcAm Soc Clin Oncol 2001;20:301a.

26. Sievers EA, Larson RA, Estey E, Lowenberg B, Berger MS, Appelbaum FR. Update of pro-longed disease-free survival in patients with acute myeloid leukemia in first relapse treatedwith gemtuzumab ozogamicin followed by hematopoietic stem cell transplantation. Proc AmSoc Clin Oncol 2001;20:302a.

27. Byrd JC, Murphy T, Howard RS, et al. Rituximab using a thrice weekly dosing schedule inB-cell chronic lymphocytic leukemia and small lymphocytic lymphoma demonstrates clini-cal activity and acceptable toxicity. J Clin Oncol 2001;19:2153–2164.

28. Bernstein ID, Singer JW, Andrews RG, et al. Treatment of acute myeloid leukemia cells invitro with a monoclonal antibody recognizing a myeloid differentiation antigen allows nor-mal progenitor cells to be expressed. J Clin Invest 1987;79:1153–1159.

29. Karanes C, Kopecky KJ, Head DR, et al. A phase III comparison of high dose ARA-C(HIDAC) versus HIDAC plus mitoxantrone in the treatment of first relapsed or refractoryacute myeloid leukemia. Southwest Oncology Group Study. Leuk Res 1999;23:787–794.

30. Neumeister P, Eibl M, Zinke-Cerwenka W, Scarpatetti M, Sill H, Linkesch W. Hepatic veno-occlusive disease in two patients with relapsed acute myeloid leukemia treated with anti-CD33 calicheamicin (CMA-676) immunoconjugate. Ann Hematol 2001;80:119–120.


31. Giles F, Kantarjian H, Kornblau S, et al. Mylotarg (gemtuzumab ozogamicin) therapy isassociated with hepatic venoocclusive disease in patients who have not received stem celltransplantation. Cancer, in press.

32. DeLeve LD, McCuskey RS, Wang X, et al. Characterization of a reproducible rat model ofhepatic veno-occlusive disease. Hepatology 1999;29:1779–1791.

33. de Vetten MP, Jansen JH, Van der Reijden BA, Berger MS, Zijlmans JM, Lowenberg B. Mol-ecular remission of Philadelphia/bcr-abl-positive acute myeloid leukaemia after treatmentwith anti-CD33 calicheamicin conjugate (gemtuzumab ozogamicin, CMA-676). Br JHaematol 2000;111:277–279.

34. Roboz GJ, Knovich MA, Schuster MW, et al. Efficacy and safety of gemtuzumab ozogam-icin in patients with poor-prognosis acute myeloid leukemia, submitted.

35. Weinthal JA, Lenarsky C, Goldman S, et al. Central nervous system and plasma gem-tuzumab ozogamicin (Mylotarg, CMA-676) levels in a patient with medullary and extra-medullary relapsed acute myeloid leukemia: a case report. Blood 2000;96:219b.

36. Naito K, Takeshita A, Shigeno K, et al. Calicheamicin-conjugated humanized anti-CD33monoclonal antibody (gemtuzumab ozogamicin, CMA-676) shows cytocidal effect onCD33-positive leukemia cell lines, but is inactive on P-glycoprotein-expressing sublines.Leukemia 2000;14:1436–1443.

37. Niculescu-Duvaz I. Technology evaluation: gemtuzumab ozogamicin, Celltech group. CurrOpin Mol Ther 2000;2:691–696.

58 Frankel et al.

1. INTRODUCTION

By targeting therapy to specific cell types and disease sites, monoclonalantibodies (MAbs) offer the possibility of improved efficacy and decreasedtoxicity compared with conventional chemotherapy. Nonetheless, the opti-mistic view of the early 1980s that MAbs were “magic bullets” has now beenreplaced by a more realistic understanding of their therapeutic potential. Sincethe 1980s various strategies employing MAbs for the treatment of cancer haveevolved. Native MAbs can be used to focus an inflammatory response against atumor cell. The binding of a MAb to a target cell can result in complementactivation, thereby initiating several biologically important effects, including

59


5 Radiolabeled Monoclonal Antibodies

John M. Burke, MD

and Joseph G. Jurcic, MD

CONTENTS

INTRODUCTION

ANTIGENIC TARGETS

RADIOISOTOPE SELECTION

CONJUGATION OF RADIOISOTOPES TO ANTIBODIES

PHARMACOKINETICS

DOSIMETRY

RADIOIMMUNOTHERAPY FOR AML WITH β-PARTICLE

EMITTERS

ALPHA-PARTICLE IMMUNOTHERAPY FOR AMLRADIOIMMUNOTHERAPY FOR ADULT T-CELL

LEUKEMIA/LYMPHOMA: 90Y-ANTI-TAC

CONCLUSIONS

REFERENCES

This work was supported by the Lauri Strauss Leukemia Foundation.

the induction of chemotaxis for phagocytic cells and the production of themembrane attack complex that disrupts cell membrane integrity. Anotherimportant mechanism for tumor cell killing is antibody-dependent cell-medi-ated cytotoxicity (ADCC), in which an effector cell expressing an Fc receptorbinds to a cell-bound MAb and is triggered to kill the target cell. Examples ofantibodies with intrinsic immunologically mediated antitumor activity includethe chimeric anti-CD20 antibody rituximab (1) and the humanized anti-CD52antibody CAMPATH-1H for chronic lymphocytic leukemia (CLL) (2).

Because of the lack of potency of many unconjugated MAbs, investigatorshave studied another approach in which MAbs are used as delivery vehicles forcytotoxic agents. Chemotherapeutic agents such as doxorubicin, calicheam-icin, methotrexate, and vinca alkaloids have been conjugated to various MAbs.Toxins used clinically have been either bacterial products, such as diphtheriatoxin and Pseudomonas exotoxin A, or plant products, such as ricin, gelonin,pokeweed antiviral protein, and saponin. Significant antileukemic effects havebeen observed with the anti-CD33-calicheamicin conjugate gemtuzumabozogamicin for acute myeloid leukemia (AML) (3) and the anti-CD22-pseudomonas exotoxin construct BL22 for hairy cell leukemia (4).

In an alternative strategy, antibodies can be used to target radioisotopes directlyto sites of disease to increase the antitumor effects of native MAbs. Becauseradioisotopes emit particles capable of inducing lethal deoxyribonucleic acid(DNA) damage to cells lying within a fixed range, radioimmunoconjugates mayallow the killing of antigen-negative tumor variants or tumor cells not reached byMAbs. This approach has produced promising results in B-cell lymphomas (5,6).Leukemias are well suited to treatment with radioimmunotherapy for several rea-sons. First, because of their location in the blood, bone marrow, spleen, and lymphnodes, malignant cells are readily accessible to circulating MAbs. Second, targetantigens on leukemic blasts are well known and can be easily characterized inindividual patients by flow cytometry. Finally, leukemias are radiosensitivetumors. This chapter focuses on issues affecting MAb pharmacokinetics and thephysical properties of various radionuclides used clinically. We also review theresults of recent clinical trials in the radioimmunotherapy of leukemia.

2. ANTIGENIC TARGETS

Immunophenotypic characterization of the various stages and lineagesfound during hematopoietic differentiation provides the rationale for selectionof MAbs that bind selectively to neoplastic cells while sparing normal tissues(Table 1). Leukemia-associated antigens, however, are not tumor specific, norare they always stage or lineage specific. For example, CD10, found on B-cellacute lymphoblastic leukemia (ALL), is also expressed by mature B-cell lym-phomas and T-cell ALL. CD33, found on most AML cells, is expressed by nor-mal myeloid precursors.

60 Burke and Jurcic

The earliest myeloid progenitors express CD34; more committed progeni-tors acquire CD33 and HLA-DR. Most AML cells express CD33, CD13, andCD15. HLA-DR is typically found on all subtypes of AML except acutepromyelocytic leukemia (APL). Monocytic leukemias express antigens associ-ated with more mature granulocytes and monocytes, including CD11a/18,CD11c, CD14, and CD15. Cell-surface antigen expression in early phasechronic myelogenous leukemia (CML) resembles that of mature granulocytesbut is heterogeneous in blast crisis.

Nearly all neoplasms of B-cell origin express CD19 and HLA-DR. ALLs ofB-cell lineage are derived from the earliest stages of B-cell differentiation.Most express CD10 and CD34, and a small proportion express CD20. CD5 andCD19 are found on small lymphocytic lymphoma and CLL. Additionally, theyweakly express CD20. Most T-cell ALL and lymphoblastic lymphoma are themalignant counterpart of the earliest T-cells that express CD2, CD5, and CD7.

3. RADIOISOTOPE SELECTION

Radioisotopes have unstable nuclei and decay by emitting charged particles(either α particles or β particles) with or without photons (γ rays). Properties ofseveral radioisotopes that have been used in the treatment of leukemia arelisted in Table 2.

Chapter 5 / Radiolabeled Antibodies 61

Table 1Selected Target Antigens for Immunotherapy of Leukemia

Antigen Disease Antibody

CD5 ALL, CLL T101, Tp67CD7 ALL Tp41CD14 AML AML2-23CD15 AML PM81CD19 ALL, CLL Anti-B4CD20 CLL Tositumomab, rituximab, 2B8, 1F5CD22 HCL, CLL LL2, huLL2, RFB4CD25 ATL anti-TacCD33 AML, CML MY9, p67, M195, HuM195CD45 AML, MDS, ALL BC8CD52 CLL CAMPATH-1HHLA-DR CLL Lym-1, Hu1D10

ALL = acute lymphocytic leukemia; CLL = chronic lymphocytic leukemia;AML = acute myeloid leukemia; HCL = hairy cell leukemia; ATL = adult T-cellleukemia; CML = chronic myelogenous leukemia; MDS = myelodysplastic syn-drome.

To date, most clinical radioimmunotherapy trials have used β particle-emit-ting radionuclides. β particles are electrons that, compared with α particles, havea relatively long range (0.8–5 mm) and low linear energy transfer (~ 0.2keV/µm). Because of the distances they travel, β particles create a “field effect”that can destroy tumor cells to which the radioimmunoconjugate is not directlybound, including nearby antigen-negative tumor cells. This property makesradioimmunotherapy with β particle-emitters theoretically more useful for thetreatment of bulky disease and for myeloablation before bone marrow transplan-tation (BMT). On the other hand, β particle-emitting radioimmunoconjugatescan damage normal “bystander” cells, resulting in nonspecific tissue toxicity. βparticle emitters that have been used in the treatment of leukemias includeiodine-131 (131I), yttrium-90 (90Y), and rhenium-188 (188Re).

α particles are helium nuclei with a range of only a few cell diameters(0.05–0.08 mm) and a linear energy transfer of approx 100 keV/µm. Becauseof the high linear energy transfer, only one or two α particles traversing anucleus can destroy a cell, making α particles among the most potent cyto-toxic agents (7). The short range of α particles potentially results in decreasedirradiation of normal bystander cells and reduced toxicity. These propertiesmake α particles ideally suited for treating minimal residual disease, in whichthe goal of therapy is to kill individual tumor cells selectively, not bulky tumormasses. α particle emitters that have been investigated in the treatment ofleukemias include bismuth-212 (212Bi), bismuth-213 (213Bi), and actinium-225 (225Ac) (8).

γ rays are photons that can travel several centimeters in human tissue. Theyare often emitted together with α and β particles. γ emissions facilitate biodis-tribution and dosimetry studies because quantitative imaging is possible usinga γ camera. Treatment with large doses of isotopes such as 131I, associated withhigh-energy photon emissions, however, requires patient isolation, can result insignificant exposure to hospital staff, and poses a waste hazard. Although theuse of a pure β-emitting isotope such as 90Y can overcome the radiation safetyissues associated with γ irradiation, biodistribution studies require the adminis-

62 Burke and Jurcic

Table 2Characteristics of Selected Radioisotopes for the Treatment of Leukemia

Particle(s) Particulate Mean Range of α- or Isotope Emitted Half-Life Energy (keV) β-Particle Emission (mm)

Iodine-131 β, γ 8.0 d 970 0.8Rhenium-188 β, γ 17 h 2120 2.4Yttrium-90 β 64 h 2280 2.7Bismuth-213 α, γ 46 min 5982 0.05–0.08Actinium-225 α, γ 10.0 d 5935 0.05–0.08

tration of MAb trace-labeled with a second isotope, typically indium-111(111In), whose biodistribution is not identical to 90Y. Positron emission tomog-raphy (PET) imaging of MAbs trace-labeled with 86Y is one strategy that mayimprove radiation dosimetry estimates for 90Y-labeled antibodies.

4. CONJUGATION OF RADIOISOTOPES TO ANTIBODIES

Various methods can be used to conjugate radioisotopes to MAbs. Because131I binds to tyrosine residues, it can be conjugated directly to antibodies usingthe chloramine-T method. Tumor resistance due to internalization of the anti-gen-antibody complex, followed by rapid degradation of the radioconjugateand expulsion of isotope metabolites, represents a significant disadvantage totherapy with some 131I-labeled MAb constructs. This problem could poten-tially be overcome by the use of radiometals, which are retained by cells aftercatabolism (9) or by novel iodination methods, such as tyramine cellobiose,resulting in more stable radioimmunoconjugates (10).

The identification of suitable chelating agents for radiometals has been chal-lenging. Studies have shown that the use of the macrocyclic ligand 1,4,7,10-tetraazacyclododecane tetraacetic acid (DOTA) can result in stable 90Yimmunoconjugates and significantly reduce bone uptake of radioyttrium (11).Because DOTA has been shown to be immunogenic (12), several diethylenetri-amine pentaacetic acid (DTPA)-derived chelates have been evaluated. Althoughnone have held yttrium as well as DOTA, cyclohexyl-A (CHX-A)-DTPA wasfound to be a suitable chelate for 90Y (13). This chelate has also been used togenerate bismuth-containing radioimmunoconjugates for clinical use (14).Another radiometal, 188Re, has been directly labeled to the anti-CD66c antibodyBW250/183 using tris-(2-carboxyethyl) phosphine as a reducing agent (15).

Radiolabeling of MAbs can cause loss of biologic function, especially whenthey are labeled with 131I at high specific activities (16). This decrease inimmunoreactivity is related directly to the number of tyrosine residues in thehypervariable region of the MAb to which radioiodine attaches. Immunoreac-tivity of MAb fragments is lost at even lower specific activities because thereare fewer tyrosine residues in the constant region (17). Although complemen-tarity-determining regions account for less than one tenth of the entire MAbsequence, they typically contain 20% to 30% of the tyrosine residues in theMAb. In contrast, lysine residues, which bind ligands used for radiometalchelation, are more uniformly distributed. Therefore, radiometal constructsmay be more suitable for high specific activity labeling.

5. PHARMACOKINETICS

Factors such as variability in tumor burden and number of binding sites percell among patients, MAb specificity and binding avidity, immunoreactivity,


MAb internalization after binding, and immunogenicity contribute to the diffi-cult and poorly understood pharmacokinetics of radiolabeled MAbs. The num-ber of available antigen sites will alter antibody pharmacokinetics andbiodistribution. For example, in a dose-escalation trial of trace-labeled 131I-anti-CD33 MAb M195 for myeloid leukemias, superior targeting to sites ofdisease as determined by γ camera imaging was seen with a comparativelysmall dose (18). This may be explained in part by the relatively low number ofbinding sites (~ 10,000–20,000) on each leukemia cell.

Heterogenous antigen expression can provide a mechanism of resistance tothe cytotoxic effects of MAbs and account for toxicities associated with ther-apy. If the targeted antigen is lacking from a subset of tumor cells, residualtumor will remain after antibody therapy. Radioimmunotherapy offers a poten-tial solution to this problem because radiation kills cells within a given rangeregardless of whether they express the target antigen. On the other hand, targetantigen expression on normal tissues, including hematopoietic cells, can resultin myelosuppression and other significant side effects.

The influence of antigenic modulation on MAb-based treatments relates tospecific therapeutic applications. Tumors in which antigen-antibody complexesremain on the cell surface may be better suited to treatments dependent uponimmune-mediated cytotoxicity or delivery of radioisotopes with long-rangedemissions, such as 131I. Internalization of the antigen-antibody complex afterbinding can optimize delivery of some radioisotopes, such as short-ranged αparticle emitters. Antigen modulation can shorten the retention time of some131I-labeled MAbs due to catabolism of the radioimmunoconjugate.

Because most MAbs used clinically are derived from mice, they can gener-ate a human antimouse antibody (HAMA) response. HAMA has been impli-cated in poor therapeutic results by neutralizing MAb on repeated doses andMAb enhancing clearance. Usually no additional toxicities are seen; however,with large MAb doses, circulating immune complexes can lead to serum sick-ness. The use of chimeric and humanized MAbs remains the most promisingstrategy to avoid HAMA responses. For some humanized MAbs, however, aprolonged biologic half-life may result in nonspecific dose deposition and tox-icity when used to deliver radioisotopes or chemotherapeutic agents.

6. DOSIMETRY

In most radioimmunotherapy trials, biodistribution and dosimetry studiesare performed routinely. Serial γ camera imaging and measurements of plasma,urine, bone marrow, and tissue biopsy radioactivity are used to estimateabsorbed radiation doses to different organs and tumor sites. These techniquesare based on the Medical Internal Radiation Dose model (19). The validity ofthese predictions, however, is limited by the accuracy in measuring activityusing γ camera imaging and by the inability to visualize all sites of disease in

64 Burke and Jurcic

patients. Single-photon emission-computed tomography (SPECT) mayincrease the accuracy of planar scintigraphy, especially when used in conjunc-tion with computed tomography (20,21). Nevertheless, the quantitative valueof SPECT remains unknown. Based on dosimetric data, models, such as theone developed to simulate the distribution of the anti-CD33 MAb M195 (22),may provide information about radiation doses delivered to tissues not directlysampled and may also be used to estimate total tumor burden and tumor burdenin individual organs.

7. RADIOIMMUNOTHERAPY FOR AML WITH β-PARTICLE EMITTERS

Most clinical trials of radioimmunotherapy in leukemia have focused on theuse of β-emitting radionuclides. The anti-CD33 MAbs M195 and HuM195 havebeen studied extensively at Memorial Sloan-Kettering Cancer Center. At theFred Hutchinson Cancer Research Center, the anti-CD33 MAb p67 and the anti-CD45 MAb BC8 have been investigated. Recently, use of the anti-CD66 MAbBW 250/183 has been reported from the Ulm University Hospital. Results ofrecent trials of radioimmunotherapy in leukemia are summarized in Table 3.

7.1. 131I-M195 and 131I-HuM195M195 is a monoclonal IgG2a antibody directed at CD33, a cell-surface gly-

coprotein expressed on most myeloid leukemia cells (23,24). CD33 is alsoexpressed on normal myelomonocytic and erythroid progenitor cells but not onpluripotent stem cells, mature granulocytes, lymphoid cells, or nonhematopoi-etic cells (25,26).

An early phase I trial showed that trace-labeled 131I-M195 could rapidly andspecifically target known sites of leukemia in patients (18). Subsequently,elimination of large leukemic burdens occurred in patients treated with thera-peutically labeled 131I-M195. In an initial phase I trial, escalating doses of 131I-M195 (50 to 210 mCi/m2) were used to treat 24 patients with relapsed orrefractory myeloid leukemias. Whole-body γ camera imaging demonstratedrapid uptake of the radiolabeled antibody in the bone marrow, as well as uptakein the liver and spleen. The isotope remained at these sites for at least 3 d. Themaximum tolerated dose of 131I-M195 was not reached, although one patienthad grade 4 hepatic toxicity. At doses of 135 mCi/m2 or greater, profoundmyelosuppression occurred, allowing eight patients to proceed to either allo-geneic (n = 5) or autologous (n = 3) bone marrow transplantation (BMT). Of24 patients, 22 had decreases in bone marrow blasts; only two patients treatedat the two lowest dose levels did not respond. Three patients had completeremissions. HAMA developed in 37% of patients. Two patients who developedHAMA were retreated; their plasma 131I-M195 levels could not be maintained,and no therapeutic benefit was seen (27).


Table 3Recent Clinical Trials of Radiolabeled Antibodies for Leukemia

Radiolabeled Isotope No. of Antibody Disease Dose Patients Results Comments Reference

131I-M195 Advanced AML, 50–210 mCi/m2 24 CR in 3 of 8 patients 5 patients received autolo- 27MDS, blastic CML receiving BMT gous BMT; 3 received

allogeneic BMT

131I-M195 Relapsed APL 50 or 70 mCi/m2 7 2 of 7 patients with Used after retinoic acid 34minimal disease tran- in patients with relapsed siently became PCR diseasenegative

131I-M195, Advanced AML, 122–437 mCi 31 CR in 28 patients; how- Used with Bu/Cy before 28131I-HuM195 MDS, blastic CML ever, long-term DFS allogeneic BMT

in 3 patients

90Y-HuM195 Advanced AML 0.1–0.3 mCi/kg 19 13 patients had reduc- Higher doses result in 35tions in marrow blasts; prolonged myelosup-1 CR pression

213Bi-HuM195 Advanced AML, 0.28–1 mCi/kg 18 14 patients had reduc- First demonstration of 47CMML tions in marrow blasts; safety of α-particle

no CRs therapy

66

131I-p67 AML 110–330 mCi 9 3 of 4 patients treated Given with Cy/TBI before 36with therapeutic doses BMT. Many patients had relapsed unfavorable biodistribution

131I-BC8 Advanced AML, ALL 76–612 mCi 44 7 of 25 patients with Given with Cy/TBI before 40AML or MDS and 3 BMTof 9 patients with ALL had long-term DFS

131I-BC8 AML in first 101–263 mCi 24 18 patients with long- Given with Bu/Cy before 41remission term DFS allogeneic BMT

188Re-BW High-risk AML, MDS 11.1 GBq (mean) 36 45% DFS at median Given as part of preparative 42250/183 18 months regimen before BMT

90Y-anti-Tac ATL 5–15 mCi 18 7 patients had PRs; 2 6 patients developed HAMA 50had CRs

AML = acute myeloid leukemia; MDS = myelodysplastic syndrome; APL = acute promyelocytic leukemia; CML = chronic myelogenous leukemia;CMML = chronic myelomonocytic leukemia; ALL = acute lymphoblastic leukemia; ATL = adult T-cell leukemia; CR = complete remission; BMT = bonemarrow transplantation; PCR = polymerase chain reaction; DFS = disease-free survival; PR = partial remission; Bu = busulfan; Cy = cyclophosphamide;TBI = total body irradiation; HAMA = human antimouse antibodies.

67

Although encouraging, these trials served to illustrate several limitations ofM195 as a therapeutic agent for leukemia. Murine M195 was unable to medi-ate leukemia cell killing by ADCC in vitro, nor did it lead to the death ofleukemic cells in the presence of human complement (24). In addition, manypatients treated with M195 developed HAMA, which adversely affected thepharmacokinetics of M195 and precluded repeated treatments with it(18,27,28). Humanized M195 (HuM195) was engineered in an attempt to over-come these limitations. HuM195 was constructed by grafting complementar-ity-determining regions of M195 into a human IgG1 framework and backbone(29). HuM195 maintained the binding specificity of M195 and had increasedbinding avidity. Unlike M195, HuM195 induced rabbit complement-mediatedcytotoxicity against both HL60 cells and fibroblasts transfected with CD33genes, and it induced ADCC using human peripheral blood mononuclear cellsas effectors (30). A phase I trial conducted in patients with advanced myeloidleukemias showed that trace-labeled 131I-HuM195 had similar biodistributionand pharmacology to murine M195, without significant immunogenicity.Moreover, HuM195 proved to be a suitable carrier for radionuclides (31).Treatment with native HuM195 demonstrated activity against minimal residualdisease in patients with APL (32) and produced occasional complete remis-sions in patients with relapsed or refractory myeloid leukemias with low-bur-den disease (33). For these reasons, HuM195 was used in place of M195 inmore recent clinical trials.

131I-M195 and 131I-HuM195 were both studied as part of a preparative regi-men before allogeneic BMT, consisting of 131I-labeled antibody (122–437 mCiin 2–4 divided doses) followed by busulfan (16 mg/kg in divided doses over 4d) and cyclophosphamide (90–120 mg/kg total dose). Sixteen patients withrefractory or relapsed AML, 14 patients with CML in accelerated or blasticphases and 1 patient with myelodysplastic syndrome underwent transplanta-tion with this regimen. Nineteen patients received 131I-M195; 12 received 131I-HuM195. No significant toxicities were attributable to the addition of theradiolabeled antibody to the preparative regimen. No delays in engraftmentwere seen. Of the 31 patients, 28 achieved complete remission, and threepatients with refractory AML remained in remission for 4.5–8 yr after trans-plant (28). This study showed that radioimmunotherapy could potentially beused to intensify antileukemic therapy before stem cell transplantation.

The role of nonmyeloablative doses of 131I-M195 given in the minimalresidual disease setting was investigated in patients with APL. Seven patientswith relapsed disease were treated with all-trans retinoic acid until theyattained clinical complete remission. Effects of therapy on residual diseasewere monitored by reverse transcription-polymerase chain reaction (RT-PCR)amplification of PML/RAR-α mRNA. Of the seven patients, six had minimalresidual disease detectable by RT-PCR after retinoic acid induction. Patients

68 Burke and Jurcic

then received either 50 or 70 mCi/m2 of 131I-M195. Toxicity was limited tomyelosuppression, but no episodes of febrile neutropenia occurred. Twopatients transiently became RT-PCR negative. The median disease-free sur-vival was 8 mo, which compares favorably with the 3-mo median disease-freesurvival seen in patients with relapsed APL treated with retinoic acid alone.HAMA developed in five patients. Thus, 131I-M195 had activity against APL,but therapy was limited by significant myelosuppression and by the formationof HAMA (34).

7.2. 90Y-HuM195Early trials showed that myeloablative therapy with 131I-anti-CD33 con-

structs has several disadvantages. First, because of decreased immunoreactiv-ity when labeled at high specific activities, multiple infusions of 131I-M195 and131I-HuM195 are needed to deliver adequate radiation doses to the marrow forablation. Second, the 8-d half-life of 131I delays the time from treatment tostem cell or marrow infusion in patients undergoing transplantation. Addition-ally, patients receiving high doses of 131I-antibody constructs must be hospital-ized and isolated because of high-energy γ emissions. 90Y offers severaladvantages over 131I for myeloablation. The higher energy longer ranged βemissions of 90Y permit a lower effective dose than 131I. The absence of γemissions eliminates the need for radiation isolation and allows large doses tobe given safely in the outpatient setting. In addition, radiometals such as 90Yare retained within cells better than 131I after internalization of antigen–anti-body complexes (9). For these reasons, 90Y-labeled HuM195 was studied in arecent phase I trial.

Nineteen patients with relapsed or refractory AML were treated withescalating doses (0.1–0.3 mCi/kg) of 90Y-HuM195, given as a single infu-sion without marrow support. Biodistribution and dosimetry studies, per-formed by co-administering trace-labeled 111In-HuM195, demonstrated thatup to 560 cGy and 750 cGy were delivered to the marrow and spleen,respectively. Transient low-grade liver function abnormalities occurred in 11patients. Myelosuppression, lasting 9–62 d, was dose limiting, and the maxi-mum tolerated dose was 0.275 mCi/kg. Thirteen of the 19 patients hadreductions in bone marrow blasts. One patient who received 0.275 mCi/kgachieved a complete remission lasting 5 mo. All patients treated at the high-est dose level (0.3 mCi/kg) had hypocellular bone marrows without evidenceof leukemia on biopsies performed 2 and 4 wk after treatment (35). Theseresults suggest that 90Y-HuM195 may be useful to reduce leukemic burdensand ablate the marrow as part of a preparative regimen for BMT. Clinical tri-als investigating this strategy in the setting of autologous peripheral bloodprogenitor transplantation and nonmyeloablative allogeneic transplantationare currently underway.


7.3. 131I-p67The murine IgG1 anti-CD33 antibody p67 labeled with 131I was studied as

part of a preparative regimen before BMT (36). In a phase I trial, nine patientswith advanced AML received p67 labeled with trace doses of 131I. γ cameraimaging demonstrated initial specific uptake of 131I-p67 in the marrow in mostpatients. However, the half-life of the radiolabeled antibody was only 9–41 h;this relatively brief half-life contrasts with that of 131I-M195, in which the radi-olabel remained in the marrow for at least 3 d (18). Four of the nine patientshad “favorable biodistribution,” defined as greater uptake of 131I in the marrowand spleen greater than in nonhematopoietic organs. These four patients thenreceived therapeutic doses of 131I-p67 (110–230 mCi) followed by cyclophos-phamide (120 mg/kg) and total body irradiation (TBI) (12 Gy) followed byallogeneic BMT. The therapy was well tolerated, but three of the four patientseventually relapsed (37). Because of the unfavorable biodistribution in manypatients and the short residence time of 131I-p67 in the marrow, the investiga-tors have since focused on the 131I-labeled anti-CD45 antibody BC8 discussedin the following section.

7.4. 131I-BC8The murine IgG1 MAb BC8 is directed at CD45, a tyrosine phosphatase

expressed on virtually all leukocytes, including myeloid and lymphoid precur-sors in the bone marrow, mature lymphocytes in lymph nodes, and mostmyeloid and lymphoid leukemia cells. Unlike the anti-CD33 MAbs, BC8 doesnot internalize after binding to its antigen (38,39).

In a phase I trial, 44 patients with advanced acute leukemia or myelodyspla-sia received BC8 labeled with trace doses of 131I (5-10 mCi). Thirty-sevenpatients (84%) had favorable biodistribution of the radiolabeled antibody, withhigher radiation-absorbed doses to the marrow and spleen than to normalorgans. Thirty-four of these patients then received therapeutic doses of 131I-BC8 (76 to 612 mCi) in a dose-escalation trial. Patients then receivedcyclophosphamide (120 mg/kg) and TBI (12 Gy) followed by either allogeneicor autologous stem cell transplantation. The maximum tolerated dose was anestimated radiation-absorbed dose of 10.5 Gy to the liver. One of six patientstreated at the maximum tolerated dose had grade 3 hepatic toxicity. Based onaverage estimates of radiation-absorbed dose, the investigators calculated thatthe 131I-BC8 delivered an additional 24 Gy to the marrow and 50 Gy to thespleen in patients treated at the maximum tolerated dose. Of the 25 patientswith AML or myelodysplastic syndrome (MDS), 7 were alive and free of dis-ease at a median of 65 mo after transplantation. Of the nine patients with ALL,three were alive and free of disease at 19, 54, and 66 mo (40).

Based on these promising results, a phase I/II trial of 131I-BC8, togetherwith busulfan and cyclophosphamide, was begun in patients with AML in first

70 Burke and Jurcic

remission. After trace-labeled 131I-BC8, γ camera imaging showed that 90% ofpatients had favorable biodistribution. These patients were then treated withtherapeutic doses of 131I-BC8, delivering 3.5 Gy (four patients) or 5.25 Gy (allsubsequent patients) to the liver, half the maximum tolerated dose defined inthe phase I study. Toxicities attributable to the 131I-BC8 were minimal. In anencouraging preliminary report, 18 of 24 patients treated with therapeuticdoses were alive and disease free at a median of 42 mo after transplant (41).

7.5. 188Re-Anti-CD66The glycoprotein CD66c, also known as nonspecific cross-reacting antigen

(NCA), is expressed on myeloid cells (200,000 molecules per cell) but not onleukemia cells. BW 250/183 is a murine monoclonal IgG1 antibody directed atCD66 (15,42). The radiometal 188Re has a 17-h half-life and emits both β and γparticles. The β emissions of 188Re are well suited for therapy, and the γ emis-sions facilitate biodistribution and dosimetry studies.

In a pilot dosimetry trial, 12 patients undergoing BMT were treated with188Re-BW 250/183 followed by a standard preparative regimen. Most patientshad a favorable biodistribution. After administration of 9.7 GBq (260 mCi) of188Re-BW 250/183, the median bone marrow dose was 14 Gy (15,43). Subse-quently, 36 patients with high-risk AML or MDS were treated with 188Re-BW250/183. Patients then received one of three preparative regimens: TBI (12Gy) plus cyclophosphamide (120 mg/kg), busulfan (12.8 mg/kg) pluscyclophosphamide (120 mg/kg), or TBI (12 Gy) plus thiotepa (10 mg/kg) andcyclophosphamide (120 mg/kg). Patients receiving grafts from unrelateddonors or from mismatched related donors also received antithymocyte globu-lin to prevent graft rejection. Thirty-one patients received allogeneic grafts(mostly T-cell depleted), 1 received a syngeneic graft, and 4 received autolo-gous grafts. All patients had a favorable biodistribution of 188Re-BW 250/183.The mean therapeutic dose administered was 11.1 GBq (300 mCi) injected in1 or 2 fractions, and the median dose delivered to the bone marrow was 14.9Gy (range, 8.1–28 Gy). The administration of radiolabeled antibody did notresult in any significant additional toxicity beyond that associated with theconventional preparative regimen. Six patients (17%), however, developedrenal toxicity between 6 and 12 mo after the transplant, possibly due to radia-tion. Engraftment occurred in all patients and was not delayed. At a medianfollow-up of 18 mos, 9 of 35 evaluable patients had relapsed, 8 of whom sub-sequently died. Eight patients (22%) died from transplant-related toxicity. Dis-ease-free survival at a median of 18 mo was 45% and was significantly higherfor those undergoing transplantation in remission (67%) than for those trans-planted with overt leukemia (31%) (42). This study suggests that 188Re-BW250/183, like 90Y-HuM195 and 131I-BC8, may deliver significant doses ofradiation to the marrow.


8. ALPHA-PARTICLE IMMUNOTHERAPY FOR AML

8.1. Preclinical StudiesTo increase the antileukemic activity of native MAbs but avoid the nonspe-

cific cytotoxic effects of β-emitting radionuclides, targeted α particle therapyhas been investigated in several experimental systems. In one of the firstreports suggesting the feasibility of this approach, 212Bi conjugated to a tumor-specific MAb 103A was used against murine erythroleukemia. Targeting of theconstruct to neoplastic cells with the spleen was seen within 1 h after injection.When 212Bi-103A was injected 13 d after inoculation with leukemia, reduc-tions in splenomegaly and the absence of liver metastases were noted. Whengiven on day 8, no histologic evidence of leukemia developed (44). Subse-quently, administration of the α-emitter 212Bi conjugated to the anti-CD25MAb anti-Tac after inoculation of nude mice with a CD25-expressing lym-phoma cell line led to prolonged tumor-free survival and prevented the devel-opment of leukemia in some animals. Treatment of established tumor with212Bi-anti-Tac, however, failed to produce responses (45).

213Bi is an α-particle emitter with a 46-min half-life. It was prepared from a225Ac/213Bi generator and conjugated to HuM195 using the bifunctionalchelating agent 2-(4-isothiocyanatobenzyl (SCN)-CHX-A-DTPA (14,46).Intraperitoneal injections of up to 20 mCi/kg and intravenous (iv) injections of10 mCi/kg 213Bi-HuM195 could safely be given to mice. In vitro the applica-tion of bismuth-labeled HuM195 resulted in dose-dependent and specificactivity-dependent killing of CD33-positive HL60 cells. Approximately 50%of target cells were killed when only 2 bismuth atoms were bound to the cellsurface (14).

8.2. Phase I Trial of 213Bi-HuM195Based on these preclinical data, 18 patients with relapsed or refractory AML

or CML were treated with 213Bi-HuM195 (0.28-1 mCi/kg) in a phase I trial(47). Treatment with 213Bi-HuM195 was tolerated well, and dose-limiting tox-icity was not observed. Further dose escalation beyond 1 mCi/kg was limitedby the availability of 225Ac. Six patients had transient grade 1 or 2 abnormali-ties in liver function tests. Myelosuppression occurred in all patients and lasteda median of 14 d (range, 8–34 d). γ camera imaging showed that the majorityof the administered activity localized to the bone marrow, liver, and spleenwithin 5 to 10 min after injection. The absorbed dose ratios between these sitesand the whole body were 1000-fold greater than those seen with β-emittingconstructs in this antigen system (48). Thirteen of 15 evaluable patients (87%)had reductions in circulating blasts after therapy, and 14 of the 18 patients(78%) had reductions in the percentage of bone marrow blasts. However, nocomplete remissions were observed. Because of the nature of α particle radia-

72 Burke and Jurcic

tion, complete remission at 30 d after treatment would have required the indi-vidual targeting and killing of 99.9% of the leukemia cells. Because thepatients treated on this study had tumor burdens of up to 1012 cells, each withan average CD33 density of 10,000 per cell, roughly 1016 leukemic bindingsites were available to HuM195. Because approx 1 in 2700 molecules ofHuM195 carried the radiolabel at the specific activities injected, it was difficultto deliver one to two 213Bi atoms to every leukemia cell, even if optimal anti-body targeting were assumed (47). Therefore, treatment of overt leukemia with213Bi-HuM195 as a single agent would require extraordinarily high injectedactivities. Nevertheless, the short range and high linear energy of α particlesmake 213Bi-HuM195 ideally suited for the treatment of residual disease. Cur-rently, the use of 213Bi-HuM195 for the elimination of minimal disease afterpartial cytoreduction with cytarabine in patients with myeloid leukemias isunder investigation. This trial provides the rationale for continued investigationof α particle immunotherapy in a variety of cancers where small volume, min-imal residual, or micrometastatic disease is present.

8.3. Targeted α Particle Nano-GeneratorsMore recently, 225Ac has been conjugated to a variety of MAbs using the

bifunctional chelate SCN-DOTA. 225Ac has a 10-d half-life and decays by αemission through three atoms, each of which also emits an α particle. In vitro,225Ac coupled to internalizing MAbs specifically killed leukemia, lymphoma,breast, ovarian, neuroblastoma, and prostate cancer cells at doses 1000 timesless than 213Bi-containing radioimmunoconjugates. In xenograft models of dis-seminated human lymphoma and solid prostate carcinoma, single doses atnanocurie levels of tumor-specific constructs prolonged survival and cured asubstantial fraction of animals without toxicity (49). Therefore, in this strategy,225Ac-SCN-DOTA serves as an atomic nano-generator that delivers a cascadeof four α particles to the inside of a cancer cell by an internalizing antibody. Aphase I trial of 225Ac-HuM195 in advanced myeloid leukemias is planned.

9. RADIOIMMUNOTHERAPY FOR ADULT T-CELLLEUKEMIA/LYMPHOMA: 90Y-ANTI-TAC

Adult T-cell leukemia/lymphoma (ATL) is a malignancy of lymphocytescaused by infection with the human T-lymphotrophic virus type I. The diseaseis characterized by circulating malignant cells, diffuse lymphadenopathy, andhypercalcemia. In patients with ATL, each leukemic cell expresses10,000–35,000 molecules of the α subunit of the interleukin (IL)-2 receptor(also called CD25 or Tac). In contrast, normal resting lymphocytes do notexpress CD25. Anti-Tac is a murine monoclonal antibody that prevents theinteraction of IL-2 with CD25 (50).


In a phase I/II trial, 18 patients with ATL were treated with anti-Tac labeledwith 90Y. The first 9 patients were treated in a phase I dose-escalation trial with5-15 mCi of 90Y-anti-Tac, and the next 9 patients were treated in a phase II trialwith a uniform 10-mCi dose. Observed toxicities included grade 3 or 4 hemato-logic toxicity (12 patients), transient grade 3 hepatic toxicity (3 patients), andtransient proteinuria (1 patient). In addition, 1 patient with hypertension and dia-betes died of unexplained cardiac asystole 23 d after the administration of 90Y-anti-Tac. Six patients developed HAMA. Of 16 evaluable patients, 7 had partialremissions (mean duration, 9.2 mo) and 2 had complete remissions. One patientwho attained a complete remission developed MDS and died of secondary AML3 yr after receiving 90Y-anti-Tac. At autopsy, the patient had persistent ATL inthe skin. The other patient with a complete remission remained without evidenceof disease for more than 3 yr after the initiation of therapy (50).

10. CONCLUSIONS

Radiolabeled antibodies have the potential to improve the outcome ofpatients with leukemias. Early clinical studies have demonstrated that radiola-beled antibodies have activity in leukemias and can be given safely to patientswith advanced disease. Two applications of radioimmunotherapy appear mostpromising: (1) cytoreduction before BMT and (2) elimination of minimalresidual disease. Myeloablative doses of β-emitting radioimmunoconjugateshave already demonstrated significant activity when used in several transplan-tation preparative regimens (28,36,39,40). Subsequent studies have suggestedthat the use of radiometals, such as 90Y and 188Re, may overcome many of thedifficulties associated with 131I (35,42). Whether the use of radioimmunother-apy as part of BMT preparative regimens improves patient outcomes comparedwith standard preparative treatments remains unknown. Comparative trials areneeded to answer this question definitively.

The choice of an optimal radioimmunoconjugate can be dictated by specificclinical situations. Therapy with long-ranged β-emitters may be better forkilling bulky disease, but the use of shorter ranged α particle-emitting isotopescould potentially result in more efficient single cell killing with less nonspe-cific toxicity (47). These physical properties make targeted α particleimmunotherapy ideal for the treatment of minimal residual disease. Treatmentwith 225Ac constructs, which can deliver an in vivo generator of four α parti-cles directly to a tumor cell, could further increase the antitumor activity previ-ously seen with 213Bi-containing constructs (49).

REFERENCES1. O’Brien SM, Kantarjian H, Thomas DA, et al. Rituximab dose-escalation trial in chronic

lymphocytic leukemia. J Clin Oncol 2001;19:2165–2170.

74 Burke and Jurcic

2. Osterborg A, Dyer MJ, Bunjes D, et al. Phase II multicenter study of human CD52 antibodyin previously treated chronic lymphocytic leukemia. European Study Group of CAMPATH-1H Treatment in Chronic Lymphocytic Leukemia. J Clin Oncol 1997;15:1567–1574.

3. Sievers EL, Larson RA, Stadtmauer EA, et al. Efficacy and safety of gemtuzumab ozogam-icin in patients with CD33-positive acute myeloid leukemia in first relapse. J Clin Oncol2001;19:3244–3254.

4. Kreitman RJ, Wilson WH, Bergeron K, et al. Efficacy of the anti-CD22 recombinantimmunotoxin BL22 in chemotherapy-resistant hairy-cell leukemia. N Engl J Med2001;345:241–247.

5. Kaminski MS, Zelenetz AD, Press OW, et al. Pivotal study of iodine I 131 toxitumomab forchemotherapy-refractory low-grade or transformed low-grade B-cell non-Hodgkin’s lym-phomas. J Clin Oncol 2001;19:3918–3928.

6. Witzig TE, White CA, Wiseman GA, et al. Phase I/II trial of IDEC-Y2B8 radioimmunother-apy for treatment of relapsed or refractory CD20+ B-cell non-Hodgkin’s lymphoma. J ClinOncol 1999;17:3793–3803.

7. Macklis RM, Lin JY, Beresford B, Atcher RW, Hines JJ, Humm JL. Cellular kinetics,dosimetry, and radiobiology of alpha-particle radioimmunotherapy: induction of apoptosis.Radiat Res 1992;130:220–226.

8. McDevitt MR, Sgouros G, Finn RD, et al. Radioimmunotherapy with alpha-emittingnuclides. Eur J Nucl Med 1998;25:1341–1351.

9. Scheinberg DA, Strand M. Kinetic and catabolic considerations of monoclonal antibody tar-geting in erythroleukemic mice. Cancer Res 1983;43:265–272.

10. Ali SA, Warren SD, Richter KY, et al. Improving tumor retention of radioiodinated antibody:aryl carbohydrate adducts. Cancer Res 1990;50(suppl):783s–788s.

11. Deshpande SV, DeNardo SJ, Kukis DL, et al. Yttrium-90-labeled monoclonal antibody for ther-apy: labeling by a new macrocyclic bifunctional chelating agent. J Nucl Med 1990;31:473–479.

12. Kosmas C, Snook D, Gooden CS, et al. Development of humoral immune responses againsta macrocyclic chelating agent (DOTA) in cancer patients receiving radioimmunoconjugatesfor imaging and therapy. Cancer Res 1992;52:904–911.

13. Camera L, Kinuya S, Garmestani K, et al. Evaluation of the serum stability and in vivobiodistribution of CHX-DTPA and other ligands for yttrium labeling of monoclonal antibod-ies. J Nucl Med 1994;35:882–889.

14. Nikula TK, McDevitt MR, Finn RD, et al. Alpha-emitting bismuth cyclohexylbenzyl DTPAconstructs of recombinant humanized anti-CD33 antibodies: pharmacokinetics, bioactivity,toxicity and chemistry. J Nucl Med 1999;40:166–176.

15. Seitz U, Neumaier B, Glatting G, Kotzerke J, Bunjes D, Reske SN. Preparation and evalua-tion of the rhenium-188-labelled anti-NCA antigen monoclonal antibody BW 250/183 forradioimmunotherapy of leukaemia. Eur J Nucl Med 1999;26:1265–1273.

16. Larson SM. A tentative biological model for the localization of radiolabelled antibody intumor: the improtance of immunoreactivity. Nucl Med Biol 1986;13:393–399.

17. Nikula TK, Bocchia M, Curcio MJ, et al. Impact of the high tyrosine fraction in complemen-tarity-determining regions: measured and predicted effects of radioiodination on IgGimmunoreactivity. Molec Immunol 1995;32:865–872.

18. Scheinberg DA, Lovett D, Divgi DR, et al. A phase I trial of monoclonal antibody M195 inacute myelogenous leukemia: specific bone marrow targeting and internalization of radionu-clide. J Clin Oncol 1991;9:478–490.

19. Society of Nuclear Medicine. MIRD Primer for Absorbed Dose Calculations. Washington,DC: Society of Nuclear Medicine; 1988.

20. Koral KF, Zasadny KR, Kessler ML, et al. CT-SPECT fusion plus conjugate views for deter-mining dosimetry in iodine-131-monoclonal antibody of lymphoma. J Nucl Med1994;35:1714–1720.


21. Sgouros G, Chiu S, Pentlow KS, et al. Three-dimensional dosimetry for radioimmunother-apy treatment planning. J Nucl Med 1993;34:1595–1601.

22. Sgouros G, Graham MC, Divgi CR, Larson SM, Scheinberg DA. Modeling and dosimetry ofmonoclonal antibody M195 (anti-CD33) in acute myelogenous leukemia. J Nucl Med1993;34:422–430.

23. Scheinberg DA, Tanimoto M, McKenzie S, Strife A, Old LJ, Clarkson BD. Monoclonal anti-body M195: a diagnostic marker for acute myelogenous leukemia. Leukemia 1989;3:440–445.

24. Tanimoto M, Scheinberg DA, Cordon-Cardo C, Huie D, Clarkson BD, Old-LJ. Restrictedexpression of an early myeloid and monocytic cell surface antigen defined by monoclonalantibody M195. Leukemia 1989;3:339–348.

25. Andrews RG, Torok-Storb B, Bernstein ID. Myeloid-associated differentiation antigens onstem cells and their progeny identified by monoclonal antibodies. Blood 1983;62:124–132.

26. Griffin JD, Linch D, Sabbath K, Larcom P, Schlossman SF. A monoclonal antibody reactivewith normal and leukemic human myeloid progenitor cells. Leuk Res 1984;8:521–534.

27. Schwartz MA, Lovett DR, Redner A, et al. Dose-escalation trial of M195 labeled with iodine131 for cytoreduction and marrow ablation in relapsed or refractory myeloid leukemias. JClin Oncol 1993;11:294–303.

28. Jurcic JG, Caron PC, Nikula TK, et al. Radiolabeled anti-CD33 monoclonal antibody M195for myeloid leukemias. Cancer Res 1995;55:5908s–5910s.

29. Co MS, Avdalovic NM, Caron PC, Avdalovic MV, Scheinberg DA, Queen C. Chimeric andhumanized antibodies with specificity for the CD33 antigen. J Immunol 1992;148:1149–1154.

30. Caron PC, Co MS, Bull MK, Avdalovic NM, Queen C, Scheinberg DA. Biological andimmunological features of humanized M195 (anti-CD33) monoclonal antibodies. CancerRes 1992;52:6761–6767.

31. Caron PC, Jurcic JG, Scott AM, et al. A phase 1B trial of humanized monoclonal antibodyM195 (anti-CD33) in myeloid leukemia: specific targeting without immunogenicity. Blood1994;83:1760–1768.

32. Jurcic JG, DeBlasio T, Dumont L, Yao TJ, Scheinberg DA. Molecular remission inductionwith retinoic acid and anti-CD33 monoclonal antibody HuM195 in acute promyelocyticleukemia. Clin Cancer Res 2000;6:372–380.

33. Caron PC, Dumont L, Scheinberg DA. Supersaturating infusional humanized anti-CD33 mon-oclonal antibody HuM195 in myelogenous leukemia. Clin Cancer Res 1998;4:1421–1428.

34. Jurcic JG, Caron PC, Miller WH Jr, et al. Sequential targeted therapy for relapsed acutepromyelocytic leukemia with all-trans retinoic acid and anti-CD33 monoclonal antibodyM195. Leukemia 1995;9:244–248.

35. Jurcic JG, Divgi CR, McDevitt MR, et al. Potential for myeloablation with yttrium-90-HuM195 (anti-CD33) in myeloid leukemia [Abstract]. Proc Am Soc Clin Oncol 2000;19:8a.

36. Appelbaum FR, Matthews DC, Eary JF, et al. The use of radiolabeled anti-CD33 antibody toaugment marrow irradiation prior to marrow transplantation for acute myelogenousleukemia. Transplantation 1992;54:829–833.

37. Ruffner KL, Matthews DC. Current uses of monoclonal antibodies in the treatment of acuteleukemia. Semin Oncol 2000;27:531–539.

38. Matthews DC, Appelbaum FR, Eary JF, et al. Radiolabeled anti-CD45 monoclonal antibod-ies target lymphohematopoietic tissue in the macaque. Blood 1991;78:1864–1874.

39. Matthews DC, Appelbaum FR, Eary JF, et al. Development of a marrow transplant regimenfor acute leukemia using targeted hematopoietic irradiation delivered by 131I-labeled anti-CD45 antibody combined with cyclophosphamide and total body irradiation. Blood1995;85:1122–1131.

40. Matthews DC, Appelbaum FR, Eary JF, et al. Phase I study of (131)I-anti-CD45 antibodyplus cyclophosphamide and total body irradiation for advanced acute leukemia andmyelodysplastic syndrome. Blood 1999;94:1237–1247.

76 Burke and Jurcic

41. Matthews DC, Appelbaum FR, Eary JF, Mitchell D, Press OW, Bernstein ID. 131I-anti-CD45 antibody plus busulfan/cyclophosphamide in matched related transplants for AML infirst remission [abstract]. Blood 1996;88:142a.

42. Bunjes D, Buchmann I, Duncker C, et al. Rhenium 188-labeled anti-CD66 (a, b, c, e) mono-clonal antibody to intensify the conditioning regimen prior to stem cell transplantation forpatients with high-risk acute myeloid leukemia or myelodysplastic syndrome: results of aphase I-II study. Blood 2001;98:565–572.

43. Kotzerke J, Glatting G, Seitz U, et al. Radioimmunotherapy for the intensification of condi-tioning before stem cell transplantation: differences in dosimetry and biokinetics of 188Re-and 99mTc-labeled anti-NCA-95 MAbs. J Nucl Med 2000;41:531–537.

44. Hueneke RB, Pippin CG, Sguire RA, Brechbiel MW, Gansow OA, Strand M. Effective alpha-particle-mediated radioimmunotherapy of murine leukemia. Cancer Res 1992;52:6095–6100.

45. Hartmann F, Horak EM, Garmestani K, et al. Radioimmunotherapy of nude mice bearing ahuman interleukin 2 receptor α-expressing lymphoma utilizing the α-emitting radionuclide-conjugated monoclonal antibody 212Bi-anti-Tac. Cancer Res 1994;54:4362–4370.

46. McDevitt MR, Finn RD, Ma D, Larson SM, Scheinberg DA. Preparation of alpha-emitting213Bi-labeled antibody constructs for clinical use. J Nucl Med 1999;40:1722–1727.

47. Jurcic JG, Larson SM, Sqouros G, et al. Targeted α particle immunotherapy for myeloidleukemia. Blood 2002;100:1233–1239.

48. Sgouros G, Ballangrud AM, Jurcic JG, et al. Pharmacokinetics and dosimetry of an alpha-particle emitter labeled antibody: 213Bi-HuM195 (anti-CD33) in patients with leukemia. JNucl Med 1999;40:1935–1946.

49. McDevitt MR, Ma D, Lai LT, et al. Tumor therapy with targeted atomic nanogenerators. Sci-ence 2001;294:1537–1540.

50. Waldmann TA, White JD, Carrasquillo JA, et al. Radioimmunotherapy of interleukin-2R alpha-expressing adult T-cell leukemia with Yttrium-90-labeled anti-Tac. Blood 1995;86:4063–4075.


II CYTOKINES

1. INTRODUCTION

Interferons (IFNs) represent a family of pleiotropic proteins. They were firstdescribed as antiviral agents in 1957 by Issacs and Lindenmann (1). It waslater shown that they play an important role not only in antiviral control butalso in cellular proliferation control and in immune system modulation.According to the cellular origin, IFNs can be classified as leukocyte, fibroblast,or immune IFN (2). Leukocyte and fibroblast IFNs are also called type I IFNs,and immune IFN is often called type II IFN (2). IFNs were the first cytokinesused in clinical trials of patients with cancer. Results of pivotal clinical studiesusing IFN-α for treatment of chronic myelogenous leukemia (CML) werealready published in 1983 by Talpaz et al. (3). Further development of IFNs asantitumor agents developed rapidly, and today recombinant IFN-α is approvedworldwide in more than 40 countries for treatment of various malignancies andviral diseases (4).

2. IFN SUBTYPES AND BIOSYNTHESIS

IFN subtypes are summarized in Table 1. IFN-α is coded by more than 20closely related genes, which show an 80% to 85% homology in amino acidsequence (2). Biologic stimuli of IFN-α biosynthesis include viruses, bacteria,

81


6 Interferons

Thomas Fischer, MD

CONTENTS

INTRODUCTION

IFN SUBTYPES AND BIOSYNTHESIS

IFN RECEPTOR AND SIGNAL TRANSDUCTION

BIOLOGIC ACTIVITY

PHARMACOKINETICS AND TOXICITY

IFN-α THERAPY OF CMLREFERENCES

microplasma, and protozoa (5). In addition to these organisms, some cytokinesand growth factors, such as CSF-1, interleukin (IL)-1, IL-2, or tumor necrosisfactor (TNF), can also induce IFN-α synthesis.

IFN-β is coded by a single gene on chromosome 9 (2). At the deoxyribonu-cleic acid (DNA) level, there is only a 30% homology with IFN-α subtypes. Sim-ilar to IFN-α, biosynthesis of IFN-β is also induced by most microorganisms.

IFN-γ is coded also by a single gene located on chromosome 12 (2). IFN-γdoes not show homology with IFN-α or IFN-β. IFN-γ is synthesized by stimu-lated lymphocytes. Classical stimuli are alloantigens or mitogenes. In addition,NK cells are also capable of producing IFN-γ (5).

3. IFN RECEPTOR AND SIGNAL TRANSDUCTION

Because IFN-β and IFN-γ play no role in the biologic therapy of leukemia, thischapter focuses on only IFN-α receptors and signal transduction. Two subunits ofthe human IFN-α receptor have been identified: IFN-αR1 and IFN-αR2 (4,6). Themajor ligand-binding chain of the IFN-α receptor is IFN-αR2. This compound canbe expressed as a short form (IFN-αR2a) or as a soluble (IFN-αR2b) or large vari-ant (IFN-αR2c). The large-form IFN-αR2c is composed of 550 amino acids and isthe functional receptor in IFN-α signaling (4,6). IFN-αR1 is composed of 557amino acids and is critical in IFN-α signaling (7,8,9). Binding of IFN-α to theIFN-αR2 receptor chain results in oligomerization of receptor components. Thenext step in molecular events is activation of intracellular kinases, which areclosely associated with receptor chains (4,10–14). These kinases have been identi-fied as JAK1 and TYK2 kinases. Activation of JAK1 and TYK2 then results inphosphorylation of tyrosine residues on the cytoplasmic domains of the IFN-α

82 Fischer

Table 1Interferon Subtypes

IFN Type I IFN Type II

Subtypes IFN-α 1–22 IFN-γIFN-βIFN-ω

Genes 27 1

No. of amino acids 165–166 143

pH stability Stabile Labile

Cellular origin Leukocytes T lymphocytesFibroblasts NK cells

Receptor IFNα-R1 IFNγ-R1IFNα-R2c IFNγ-R2

receptor chains. This event then results in activation of specific docking sites forcytoplasmic signal transducer and activator of transcription (STAT) proteins.STAT proteins then dock to these specific IFN-α receptor chain sites. This enablesJAK kinases to also phosphorylate STAT proteins at specific tyrosine residues. Asa consequence of these events, homodimers and heterodimers of STAT proteinsare formed and translocate to the nucleus. The major STAT heterodimer in IFN-αsignaling is called IFN-stimulated gene factor 3 (ISGF3) and is composed ofSTAT1, STAT2, and the so called P48 protein, a member of the interferon regula-tory factor (IRF) family. ISGF3 binds of specific recognition sites within the pro-moter of IFN-stimulated genes (Fig. 1). These recognition sites for ISGF3 aretermed IFN-stimulated response elements (ISRE), which are necessary and suffi-cient to mediate IFN-α activation of transcription (15,16). In addition to ISGF3,STAT1 homodimers are also involved in signal transduction of IFN-α. STAT1homodimers do not bind to the ISRE sequence but to the palindromic γ-IFN acti-vatine site GAS sequence. The important role of JAK and STAT proteins in intra-cellular signaling has first been elucidated for IFN-α signal transduction.However, recently it has been shown that this is a common theme involved in sig-naling of many cytokines and growth factors.

4. BIOLOGIC ACTIVITY

IFNs play an important role in the defense of viral, bacterial, and parasiticinfections. In addition, they have antitumor activities. Recently, significantprogress has been made in the identification and characterization of a multi-

Chapter 6 / Interferons 83

Fig. 1. IFN-α signaling. Upon binding IFN-α to its receptor, the JAK/STAT signal trans-duction pathway is activated, which leads to enhanced transcription of IFN-induced genes.

tude of IFN-inducible proteins (5). The molecular targets for antiviral activitiesof IFN are viral replication, viral transcription, and viral translation. The IFN-inducible 2–5 oligoadenylate synthetase/RNase L system and the Mx-proteinare particularly important for these functions (2,5,17,18). Although the antivi-ral activity mechanisms of IFNs are understood, little is known about theirantitumor activity mechanisms. Basically, a direct antitumor effect can be sep-arated from more indirect activities. A selection of IFN-inducible and IFN-modified proteins that are believed to play an important role in direct IFNantitumor activity are summarized in Table 2 (19–23).

IFN-α exerts direct antiproliferative effects on several cell types. The fol-lowing mechanisms are believed to play an important role in this function:inhibition of cell-cycle transition, modulation of apoptosis, and induction ofIFN-inducible genes directly involved in growth control. In addition, IFN-α isprofoundly involved in the regulation of adhesion molecules and thereby medi-ated signals of hematopoietic progenitor cells.

The effects of IFN-α on cell-cycle transition involve modulation of the fol-lowing cell-cycle regulatory elements: IFN-α upregulates cyclin-dependentkinase (Cdk) inhibitors and inhibits expression of cell cyclins (21,22,23). As aconsequence of these events, cyclin-/Cdk-associated kinase activities are sup-pressed by IFN-α in the G1 phase of the cell cycle. This results in increasedexpression of the underphosphorylated retinoblastoma gene that ultimatelyleads to reversible arrest in the G1 phase of the cell cycle.

84 Fischer

Table 2Potential Mechanisms of Direct Antitumor IFN Activities

Protein Function Modification by IFN

PKR (= P68 Kinase) Translation control, potential Induction by IFN-α, -β, -γtumor suppressor gene

IRF-1 (IFN regulatory Transcription factor, potential Induction by IFN-α, -β, -γfactor-1) tumor suppressor gene

RNase L Endoribonuclease, potential Induction by IFN-α, -βtumor suppressor gene

RB (retinoblastoma Cell cycle regulation, tumor Modification of phosphory-gene) suppressor gene lation by IFN-α

Cyclins Cell cycle regulation Downregulation of expres-sion by IFN-α

Cyclin dependant Cell cycle regulation Downregulation of expres-kinases (Cdks) sion and inhibition of

kinase activity by IFN-αc-myc Transcription factor, Downregulation of c-myc

oncogene mRNA by IFN-α

IFN = interferon.

IFN-α also has modulatory functions in apoptosis. Depending on the cellu-lar context, IFN may either stimulate or inhibit apoptotic processes. Stimula-tion of apoptosis is involved in the growth inhibitory effects of IFN-α. One ofthe molecular mechanisms is enhancement of CD95L-induced apoptosis viaactivation of caspase 3 (24). Recently, it has also been found that IFN-α upreg-ulates phospholipid-scramblase I (PLSCRI) and thus may have a modulatoryfunction in apoptosis via regulation of this protein (25). IFN-α is significantlyinvolved in regulation of cell adhesion between hematopoietic progenitor cellsand bone marrow stromal cells or extracellular matrix components. Extracellu-lar matrix is produced by stromal cells, including endothelial cells, fibroblasts,osteoblasts, and macrophages and is composed of fibronectine, hyaluronicacid, integrins, and other components (26,27). IFN-α is able to modulatefibronectine- and integrin-dependent adhesion function in normal and malig-nant hematopoietic progenitor cells (28).

Indirect mechanisms involved in growth control of malignant hematopoiesisby IFN-α include activation of immune effector mechanisms and modulation ofcytokine synthesis at the level of bone marrow stroma. For hematopoiesis, afinely tuned balance of stem cell renewal and regulation of differentiation is aprerequisite (29). This involves the action of positive and negative regulatorycytokines at the level of hematopoietic cells and bone marrow stromal cells. Asan example, hematopoietic growth factors, such as IL-7, IL-11, stem cell, andgranulocyte-macrophase colony-stimulating factor (GM-CSF), are constantlyproduced by bone marrow stromal cells and have a stimulatory effect onhematopoiesis (30–33). IFN-α significantly inhibits the production of stimulatoryeffectors in the bone marrow microenvironment, such as GM-CSF or CSF (29).

IFN-α exerts pleiotropic functions on the immune system. It modulates thegrowth, differentiation, and function of various immunologic effector cells,and it regulates the ability of these effector cells to interact with infected ormalignant cells. However, the main mechanism that enhances the interactionbetween immune effector cells and their target cells is IFN-induced upregula-tion of major histocompatibility complex (MHC) class I antigen expressionand adhesion molecule expression. In a recent report, upregulation of MHCexpression was suggested to be involved in clearing malignant cells by T-cellimmunity (34). This study demonstrated that a strong correlation between thepresence of T cells specifically recognizing a peptide (PR1) from proteinase 3and clinical responses after IFN-α therapy could be observed in patients withCML. Proteinase 3 is a myeloid tissue–restricted antigen that is overexpressedin CML cells. It has been proposed that IFN-α may have induced clinicalremissions by facilitating expansion of autologous leukemia-reactive cytotoxicT-lymphocytes (CTLs). The molecular mechanism involved may be upregula-tion of MHC class I antigen expression or upregulation of tumor antigens inleukemic cells, thereby precipitating an immune response.


5. PHARMACOKINETICS AND TOXICITY

Upon subcutaneous injection of recombinant IFN-α, C-max in plasma isreached within 4 to 8 h (35). There are no significant differences in intravenous(iv), subcutaneous, or intramuscular (im) routes with regard to bioavailability.The T1/2 is approx 4 h. Elimination of IFN-α mainly occurs by proteolyticdegradation in the kidneys. Penetration into the central nervous system (CNS)is usually not observed when applying doses below 50 × 106 IU (35).

IFN-α side effects can be separated into acute and chronic toxicity (35).Acute toxicity is represented by a flu-like syndrome, characterized by fever,chills, bone pain, and headache. These side effects usually appear 2 to 4 happlication IFN-α after. Acute side effects can be attenuated significantly byprophylactic application of acetaminophen at a dose of 500 to 1000 mg orally(po) taken 2 h before and 2 h after subcutaneous injection of IFN. Side effectsare generally dose related but tend to disappear within days due to tachyphy-laxis. A less common side effect is gastrointestinal toxicity, such as nausea,vomiting, and diarrhea. Cardiovascular toxicity, such as hypotension, anginapectoris, or tachyarrhythmia, is also rarely observed. The chronic toxicity ofIFN generally manifests as fatigue, depression, polyneuropathy, and asthenia.Autoimmune phenomena, such as autoimmune hepatitis or hypothyroidism,may also be observed during prolonged IFN-α therapy. Occlusion of retinalveins or arteries and cotton wool spots are relatively rare events.

Side effects can be managed by interruption of IFN-α therapy and subse-quent dose reduction by 25% to 50%. However, several patients do not tolerateIFN-α in the long run.

6. IFN-α THERAPY OF CML

The pivotal study by Talpaz et al. showed that IFN-α induces completehematologic and cytogenetic remissions in CML (36). Subsequently, severalclinical trials demonstrated that IFN-α is an active drug in the treatment ofCML (37). IFN-α induces complete hematologic responses in 70% to 80% ofpatients in the chronic phase (CP) within a few weeks to months of therapy(37). IFN-α has also been shown to induce durable cytogenetic remissions.The rate of cytogenetic remissions reported varies widely and ranges from 0%to 44% (37).

Today, most centers use the well-established criteria of the Houston group(37,38) to define hematologic and cytogenetic remissions (Tables 3 and 4).

Based on the results of four prospective randomized studies and a meta-analy-sis, IFN-α is generally accepted as prolonging survival in patients with CML(39–43). Recently, a European meta-analysis (44) showed that patients achievinga complete cytogenetic response with IFN have an excellent survival rate: the 10-yr survival rate from first CCR is 72% (Table 5) and is related to the risk profile.

86 Fischer

Patients with low-risk Sokal and Euro scores have a 10-yr survival probability of89% and 81%, respectively (44). The median time to first CCR was 19 mo (44).

Recent evidence suggests that survival on IFN-α is better if treatment is ini-tiated within the first 6 months of diagnosis (45) and in patients with fewerthan 10% blasts in the peripheral blood (45). The Sokal score and the recentlyimproved Hasford score (http://www.Pharmacoepi.de) have recently beenshown to be strong predictors of survival on IFN-α (46).

Therapy of chronic phase CML can be initiated with hydroxyurea at a doseof 1 to 3 g daily. This results in a faster tumor debulking than using IFN-α. Inaddition, allopurinol is given at a dose of 300 to 600 mg per day. Once theWBC drops below 20,000/µl, IFN-α may be started and hydroxyurea plusallopurinol gradually tapered (38).

The optimal dose of IFN-α is still a matter of dabate. Usually, a dose of 5 ×106 IU/m2 is recommended (38). However, in the long run, the IFN-α dosemust be adjusted once the WBC decreases to below 2000/µl or the plateletcount to below 50,000/µl. The goal should be to decrease the WBC to a level of2000 to 4000/µl and to achieve a complete hematologic remission. For this


Table 3Definitions of Hematologic Remissions for Chronic Phase with CML Patients (37,38)

Complete hematologic response (CHR)Normal white blood cell count (WBC) (< 10.000/µl)Normal platelet count (< 450,000/µl)Normal differential (no immature cells)Disappearance of all symptoms and signs of CML

Partial hematologic response (PHR)Decrease in WBC to < 50% of the pretreatment level and < 20.000/µl orNormalization of the WBC but persistent splenomegaly or immature cells in the

differential

Table 4Definitions of Cytogenetic Remissions (37,38)*

Remission Definition

Complete cytogenetic remission (CCR) Absence of metaphases harboring the Ph+ chromosome

Partial cytogenetic remission (PCR) 1% to 34% Ph+ metaphasesMinor cytogenetic remission (mCR) 35% to 90% Ph+ metaphasesNo cytogenetic response (NCR) 91% to 100% Ph+ metaphases

* At least 20 metaphases should be evaluated. Complete and partial cytogenetic responses arereferred to as major cytogenetic responses (Ph+ < 35%).

goal, a maximal dose of IFN-α that does not induce grade III/IV toxicity is rec-ommended. Initially, IFN-α therapy may be started at a lower dose and thendose escalated. A possible schedule would be 3 × 106 IU/d for 1 wk. Subse-quently, the dose could be increased to 5 × 106 IU/d for 1 wk and then 5 × 106

IU/m2/d. This schedule has the advantage that tachyphylaxis is induced earlyand higher doses may be tolerated more easily. Initial side effects may be sup-pressed by giving 500 to 1000 mg acetaminophen po 2 hr before and 2 hr afteradministration of IFN-α at bedtime. Patients achieving a cytogenetic responseshould continue IFN-α therapy and minimal residual disease (MRD) should bemonitored by quantitative PCR regularly. Patients should be followed by bonemarrow cytogenetic analysis every 3 to 6 mo.

There is a significant difference in the risk of relapse in patients with rela-tively high bcr-abl transcript levels as compared to patients with low levels(47). Particularly, relapse was rare in patients with a bcr-abl/abl ratio below0.045%. This suggests that it is advisable to continue IFN-α therapy until lowMRD levels are observed (47). Once desired MRD level is achieved and IFN-αhas been given at least for 3 yr after documentation of a CCR, IFN-α may betapered gradually (38,47). Further treatment decisions may then be based onthe dynamics of MRD levels.

IFN-α may have deleterious effects on the outcome of subsequent bonemarrow transplantation (48). However, recent studies demonstrate that priortreatment with IFN-α does not adversely influence allogeneic BMT, providingIFN is discontinued at least 90 d before bone marrow transplantation or if onlya short course of IFN-α has been given (45,49).

6.1. IFN-α Treatment of Hairy Cell LeukemiaThe standard of care for hairy cell leukemia is induction therapy with cladrib-

ine (2 CDA) or pentostatin. Alternatively, IFN-α may be used as induction ther-apy if there is a history of infection, which puts the patients at high risk for

88 Fischer

Table 5Survival Estimates (Kaplan-Meier) for Patients in the

Chronic Phase of CML

Therapy Probability of Survival

Busulfan 32% at 5 yr Hydroxyurea 44% at 5 yr IFN-α 59% at 5 yr CCR on IFN-α 72% at 10 yr from CCR Unrelated allogeneic SCT 57% at 5 yr Related allogeneic SCT 58% at 7 yr Autologous PBSCT 66% at 4 yr

myelosuppression (50). Nucleoside analogs show a more rapid therapeutic onsetin comparison to IFN-α but are significantly more myelosuppressive. Therefore,IFN-α may be useful as a prephase therapy. In addition, IFN-α may be effectiveinˇ20 the event of relapse after cladribine or if there is primary resistance tocladribine. As for CML, the mode of action in the treatment of hairy cellleukemia is currently unknown. IFN-α has been reported to induce partial orcomplete remission in more than 80% of patients with hairy cell leukemia (50).The dose recommended for induction therapy is 2 to 3 × 106 IU IFN-α twice aweek. However, doses as low as 0.2 to 0.5 × 106 IU may be effective. The firstresponse is a decrease in spleen size, followed by an increase in platelets andhemoglobin. Later, there is normalization of granulocytes and monocytes. Main-tenance therapy is 2 to 3 × 106 IU IFN-α once a week (50).

REFERENCES1. Issacs A, Lindenmann J. The interferon. Proc Royal Soc London (Series B). J Virus Interfer-

ence 1957;259.2. Pestka S, Langer JA, Zoon KC, Samuel, CE. Interones and their actions. Ann Rev Biochem

1987;56:727–777.3. Talpaz M, McCredie KB, Mavligit GM, Gutterman JU. Leukocyte interferon-induced

myeloid cytoreduction in CML. 1983;62:689–692.4. Pfeffer LM, Dinarello CA, Herberman RB, et al. Biological properties of recombination α-

interferons: 40th anniversary of the discovery of interferons. Cancer Res 1998;58:2489–2499.5. Sen GC, Lengyel P. The interferon system. A bird’s eye view of its biochemistry. J Biol

Chem 1992;267:5017–5020.6. Pestka S. The interferon receptors. Semin Oncol 1997;24:9.7. Uzé G, Lutfalla G, Gresser I. Genetic transfer of a functional human interferon α receptor

into mouse cells; cloning and expression of its cDNA. Cell 1990;60:225–234.8. Müller U, Steinhoff U, Reis LF, et al. Functional role of type I and type II interferons in

antiviral defense. Science 1994;264:1918–1921.9. Hwang SY, Hertzog PJ, Holland KA, et al. A null mutation in the gene encoding a type I inter-

feron receptor component eliminates antiproliferative and antiviral responses to interferon αand β and alters macrophage responses. Proc Natl Acad Sci USA. 1995;92:11284–11288.

10. Stark GR, Kerr IM, Williams BRG, Silverman RH, Schreiber RD. How cells respond tointerferons. Ann Rev Biochem 1998;67:227–264.

11. Platanias LC, Fish EN. Signaling pathways activated by interferons. Exp Hematol1999;27:1583–1592.

12. Ransohoff RM. Cellular responses to interferons and other cytokines: the JAK-STAT para-digm. N Engl J Med 1998;338:616.

13. Ihle JN, witthuhn BA, Quelle FW, et al. Signaling by the cytokine receptor superfamily:JAKs and STATs. Trends Biochem Sci 1994;19(5):222.

14. Darnell JE Jr., Kerr IM, Stark GR. JAK-STAT pathways and transcriptional activation inresponse to IFNs and other extracellular signaling proteins. Science 1993;264:1415.

15. Williams BR. Transcriptional regulation of interferon-stimulated genes. Eur J Biochem1991;200(1):1–11.

16. Williams BRG. Signal transduction and transcriptional regulation of interferon-α-stimulatedgenes. J Interferon Res 1991;11(4):207–13.

17. Sen GC, Ransohoff RM. Interferon-induced antiviral actions and their regulation. Adv VirusRes 1993;42:57–102.


18. Staeheli P. Interferon-induced proteins and the antiviral state. Adv Virus Res 1990;38:147–200.

19. Ben Dori R, Resnitzki D, Kimchi A. Reduction in p53 synthesis during differentiation ofFriend-erythroleukemia cells. Correlation with the commitment to terminal cell division.FEBS Lett 1983;162:384–389.

20. Lengyel P. Tumor-suppressor genes: news about the interferon connection. Proc Natl AcadSci USA 1992;90:5893–5895.

21. Tiefenbrun N, Melamed D, Levy N, et al. Alpha interferon suppresses the cyclin D3 andcdc25A genes, leading to a reversible G0-like arrest. Mol Cell Biol 1996;16:3934–3944.

22. Yamada H, Ochi K, Nakada S, et al. Interferon modulates the messenger RNA of G1-con-trolling genes to suppress the G1-to-S transition in Daudi cells. Mol Cell Biochem1995;152:149–158.

23. Zhang K, Kumar R. Interferon-alpha inhibits cyclin E- and cyclin D1-dependent CDK-2kinase activity associated with RB protein and E2F in Daudi cells. Biochem Biophys ResCommun 1994;200:522–528.

24. Roth W, Wagenknecht B, Dichgans J, Weller M. Interferon-alpha enhances CD95L-inducedapoptosis of human malignant glioma cells. J Neuroimmunol 1998;87(1–2):121–129.

25. Zhou Q, Zhao J, Al-Zoghaibi F, et al. Transcription control of the human plasma membranephospholipid scramblase 1 gene is mediated by interferon-alpha. Blood 2000;95(8):2593–9.

26. Verfaillie CM, McCarthy JB, McGlave PB. Mechanisms underlying abnormal trafficking ofmalignant progenitors in chronic myelogenous leukemia. Decreased adhesion to stroma andfibronectin but increased adhesion to the basement membrane components laminin and col-lagen type IV. J Clin Invest 1992;90:232.

27. Simons PJ, Masinovsky B, Longenecker BM, et al. Vascular cell adhesion molecule-1expressed by bone marrow stromal cells mediates the binding of hematopoietic progenitorcells. Blood 1992;80:388.

28. Bhatia R, McCarthy JB, Verfaillie CM. Interferon-α restores normal β1-integrin mediatednegative regulation of chronic myelogenous leukemia progenitor proliferation. Blood1996;87:3883.

29. Aman J, Keller U, Derigs G, Mansour M, Huber C, Peschel C. Regulation of cytokine expres-sion by Interferon-α in human bone marrow stromal cells: inhibition of hematopoietic growthfactors and induction of Interleukin-1 receptor antagonist. Blood 1994;84:4142–4150.

30. Namen AE, Schmierer AE, March CJ, et al. B cell precusor growth-promoting activity.Purification and characterization of a growth factor active an lymphocyte precursors. J ExpMed 1988;167:988.

31. Paul SR, Bennett F, Calvette JA, et al. Molecular cloning or a cDNA encoding interleukin11, a stromal cell-derived lymphopoietic and hematopoietic cytokine. Proc Natl Acad SciUSA 1990;87:512.

32. Heinrich MC, Dooley DC, Freed AC, et al. Constitutive expression of steel factor gene byhuman stromal cells. Blood 1993;82:771.

33. Charbord P, Tamayo E, Sealand S, et al. Granulocyte-macrophage colony-stimulating factor(GM-CSF) in human long-term bone marrow cultures: endogenous production in the adher-ent layer and effect of exogenous GM-CSF in granulomonopoiesis. Blood 1991;78:1230.

34. Molldrem J, Lee P, Wang C, et al. Evidence that specific T lymphocytes may participate inthe elimination of chronic myelogenous leukemia Nature Med 2000;6:1018–1023.

35. Dorr RT. Interferon-alpha in malignant and viral diseases. A review. Drugs 1993;45:177–211.36. Talpaz M, Kantarjian HM, McCredie K, Trujillo JM, Keating MJ, Gutterman JU. Hemato-

logic remission and cytogenetic improvement by recombinant human interferon alpha A inCML. N Engl J Med 1986;314:1065–1069.

37. Silver RT, Woolf SH, Hehlmann R, et al. An evidence-based analysis of the effect of busul-fan, hydroyurea, interferon, and allogneic bone marrow transplantation in treating the

90 Fischer

chronic phase of chronic myeloid leukemia: developed for the American Society of Hema-tology. Blood 1999;94(5):1517–1536.

38. Kantarjian HM, O’Brien S, Anderlini P, Talpaz M. Treatment of chronic myelogenousleukemia: current status and investigational options. Blood 1996;87(8):3069–3081.

39. Hehlmann R, Heimpel H, Hasford J, and the German CML Study Group, et al. Randomizedcomparison of interferon-α with Busulfan and Hydroxyurea in chronic myelogenousleukemia. Blood 1994;84:4064–4077.

40. The Italian Cooperative Study Group on Chronic Myeloid Leukemia. Interferon-α-2a ascompared with conventional chemotherapy for the treatment of chronic myeloid leukemia. NEngl J Med 1994;330:820–825.

41. Allan NC, Richards SM, Shepherd PC. UK Medical Research Council randomized, multi-center trial of interferon-α n1 for chronic myeloid leukaemia: improved survival irrespectiveof cytogenetic response. The UK Medical Research Council’s Working Parties for Therapeu-tic Trials in Adult Leukaemia. Lancet 1995;345:1392–1397.

42. Ohnishi K, Ohno R, Tomonafa M, et al. A randomized trial comparing interferon-α withBusulfan for newly diagnosed chronic myelogenous leukemia in chronic phase. Blood1995;86:906–916.

43. Chronic Myeloid Leukemia Trialist’s Collaborative Group. Interferon-α versus chemother-apy for chronic myeloid leukemia: a meta-analysis of seven randomized trials. J Natl CancerInst 1997;89:1616–1620.

44. Bonifazi F, de Vivo A, Rosti G, European Study Group on Interferon in Chronic MyeloidLeukemia, et al. Chronic myeloid leukemia and interferon-alpha: a study of complete cyto-genetic responders. Blood 2001;98(10):3074–3081.

45. Hehlmann R, Hochhaus A, Berger U, Reiter A. Current trends in the management of chronicmyelogenous leukemia. Ann Hematol 2000;79:345–354.

46. Hasford J, Pfirrmann M, Hehlmann R, et al. A new prognostic score for the survival ofpatients with chronic myeloid leukemia treated with interferon-α. J Natl Cancer Inst1998;90:850–858.

47. Hochaus A, Reiter A, Saubele S, Cross NCP for the German CML Study Group and the UKMRC CML Study Group, et al. Molecular heterogeneity in complete cytogenetic respondersafter interferon-α therapy for chronic myelogenous leukemia: low levels of minimal residualdisease are associated with continuing remission. Blood 2000;95(1):62–66.

48. Beelen DW, Graeven U, Elmaagacli AH, et al. Prolonged administration of interferon-α inpatients with chronic-phase Philadelphia chromosome-positive chronic myelogenousleukemia before allogeineic bone marrow transplantation may adversely affect transplantoutcome. Blood 1995;85(10):2987–2990.

49. Giralt S, Szydlo R, Goldman JM, et al. Effect of short-term interferon therapy on the out-come of subsequent HLA-identical sibling bone marrow transplantation for chronic myel-ogenous leukemia: an analysis from the International Bone Marrow Transplant Registry.Blood 2000;95(2):410–415.

50. Pralle H, Dempke W. Kompendium Internistische Onkologie. In: Schmoll HJ, Höffken K,Possinger K, eds. Springer (Berlin, Heidelberg, New York); 1999:243–255.


1. INTRODUCTION

Significant progress has been made in chemotherapy treatment of patientswith acute leukemia, including effective strategies for both remission inductionand dose-intensive consolidation. Long-term survival, however, remains elu-sive. Despite a better understanding of disease biology, including pretreatmentcharacteristics and unique cytogenetic features, the prognosis for adult patientswith acute leukemia remains poor. In acute myeloid leukemia (AML), 60–80%of patients will initially achieve complete remission (CR) but the majority ofthem will relapse (1,2). Only 10%–30% of patients achieve prolonged leukemia-free survival (3,4,5). In the case of acute lymphocytic leukemia (ALL), 60–65%of children can achieve long-term survival but only 10–30% of adults survive 5yr free of recurrence (6,7).

Although induction chemotherapy for acute leukemia involves a cytoreductivestrategy, the main clinical investigation effort involves preventing recurrence.

93


7 Interleukin-2 Treatment of Acute Leukemia

Peter Kabos, MD and Gary J. Schiller, MD

CONTENTS

INTRODUCTION

ADOPTIVE IMMUNE THERAPY

DONOR LEUKOCYTE INFUSION

AUTOLOGOUS IMMUNE ACTIVATION

BIOLOGY OF INTERLEUKIN-2 AND PRECLINICAL STUDIES

IN ACUTE LEUKEMIA

TOXICITY AND BIOLOGIC EFFECT OF IL-2USE OF INTERLEUKIN-2 IN OTHER MALIGNANCIES

CLINICAL USE OF INTERLEUKIN-2 IN ACUTE LEUKEMIA

CONCLUSIONS

REFERENCES

Despite an improvement in remission duration, relapse remains the major causeof death for more than 50% of adult patients (5,6). Therefore, leukemia patientstypically receive regimens of dose-intensive chemotherapy with or without stemcell support to consolidate CR (6,8). Relapse of disease is presumably due to pro-liferation of leukemia cells that survive induction, consolidation, and evenpreparative conditioning. Alternative approaches to cytotoxic chemotherapy formaintenance of CR include nonspecific immune stimulation and adoptiveimmunotherapy through allogeneic transplantation (see the following section).

For decades there have been attempts to stimulate the host immune system tocontrol the population of leukemia cells. Initial approaches involved immuniza-tion with Corynebacterium parvum, bacille Calmette-Guerin, and transfusion ofnonirradiated blood. These modalities were unsuccessful. On the other hand,allogeneic transplantation offers a potentially curative mechanism of adoptiveimmune therapy, leading to a re-examination of the role of immunotherapy inthe management of acute leukemia (9).

Currently, allogeneic bone marrow transplantation (alloBMT) represents thebest method for an immune-mediated eradication of minimal residual disease. Ini-tially, alloBMT was viewed as supportive care used for restoring hematopoesisafter supralethal doses of chemoradiotherapy (10,11). It has been recognized,however, that high-dose chemoradiotherapy often does not eradicate leukemia andthat the allogeneic graft itself provides an additive antileukemia effect (12,13).There are several lines of evidence supporting the concept of graft vs leukemia(GVL) after alloBMT. First, patients with AML with acute and chronic graft vshost disease (GVHD) have a reduced risk of leukemia relapse (14–17). Second,the risk of relapse is higher after syngeneic BMT (18–20). Direct comparison ofallogeneic and autologous transplantation for acute leukemia suggests that theantileukemia effect of alloBMT relies on the immunoreactive action of the graft asmuch as the myeloablative preparative therapy itself (5). In addition, T-cell deple-tion of allotransplantations also significantly increases the risk of relapse (14,21).Allotransplantation, however, is associated with major therapeutic limitations. Ahistocompatible sibling donor is available to only a limited number of youngerpatients. Allogeneic transplantation also carries a higher potential for complica-tions mostly from regimen-related toxicity, infection, and especially GVHD,accounting for high treatment-related morbidity and mortality (22).

2. ADOPTIVE IMMUNE THERAPY

The mechanism of GVL in alloBMT has been the focus of research and isbelieved to be due to adoptive immunotherapy. The importance of adoptiveimmunotherapy was first demonstrated with the use of lymphokine-activatedkiller (LAK) cells in combination with interleukin (IL)-2 in patients withmetastatic cancer. This combination produced objective tumor regression inpatients with renal and colon cancers and CR in a patient with metastatic

94 Kabos and Schiller

melanoma (23,24). In these early studies, LAK cells were derived from periph-eral blood mononuclear cells and activated ex vivo with high-dose IL-2. Exvivo activated LAK cells were capable of lysing tumor cells in a process unre-lated to major histocompatibility proteins (25). After initial preclinical success,clinical efforts focused on the use of LAK cells and IL-2 in the treatment ofsolid tumors such as renal cell carcinoma and melanoma. Unfortunately, noadvantage has been clearly demonstrated for the administration of LAK cellsand IL-2 over administration of IL-2 alone in metastatic cancer (26). Recentstudies have offered an explanation for the limited therapeutic efficacy of LAKcells, given their inability to home exclusively to the tumor (27,28). A searchfor more potent killer cells has led to the discovery of tumor-infiltrating lym-phocytes (TILs) (29). With the phenotype of cytotoxic lymphocytes, these cellskill in a major histocompatability complex (MHC)-restricted manner and alsoappear to be more potent than LAK cells. More importantly, they are more effi-cient in homing to the tumor (30,31). Immunologic treatment with TILs can behighly successful, as demonstrated in patients with cytomegalovirus (CMV)infections after alloBMT (32,33). TIL therapy of malignancies has great poten-tial but depends on identifying a tumor-specific antigen. Currently, due to thelack of leukemia-specific antigens, nonspecifically acting donor leukocytes arebeing used in the setting of alloBMT as means of adoptive immunotherapy.

3. DONOR LEUKOCYTE INFUSION

Initially, donor leukocytes were used in combination with interferon-α to suc-cessfully re-induce remission in patients with relapsed chronic myelogenousleukemia (CML) after BMT (34). This success led to studies of donor-leukocyteinfusion in multicenter trials for patients with CML, as well as acute leukemiarelapsed after alloBMT (35,36). Based on these studies, donor leukocyte infu-sion (DLI) is most effective in patients with relapsed CML, in which 80% ofpatients achieve CR, whereas patients with more advanced acute leukemia havea much lower likelihood of response. In patients with AML who relapsed afteralloBMT, response rates range from 15% to 20%; few patients with ALLrespond (37–39). DLI is associated with major complications, including GVHDand pancytopenia. Predictors for developing GVHD include previous T-lympho-cyte–depleted transplantation and the concomitant use of interferon-α. Howeverimperfect, the fact that some patients with acute leukemia may be re-induced toenter remission after DLI strongly supports the concept of an immune-mediatedantileukemia effect (38,40–43).

4. AUTOLOGOUS IMMUNE ACTIVATION

A major limitation of autologous stem cell transplantation in the managementof acute leukemia is the absence of an immune effect. Autologous transplanta-tions, however, offer many advantages when compared with allogeneic transplan-

Chapter 7 / IL-2 Treatment of Leukemia 95

tations, allowing for the use of myeloablative preparative conditioning in a largerunselected population of patients without the risk of treatment-related morbidityand mortality. Autologous transplantation is also not limited by age or prognosticfeatures. Therefore, a combination of GVL effect with the advantages of autolo-gous transplantation would be desirable (44). Several drugs, includingcyclosporine (17,45), IL-1-α (46), and linomide (47) have been used in patientswith acute leukemia to boost the GVL effect after autologous transplantation.More recently, IL-2 has been identified as a candidate for immunotherapy withthe potential to mimic the immune effect seen in allogeneic transplantation(48–50). IL-2 has been the focus of extensive clinical investigations in acuteleukemia and has shown promise as a potential immunologic activator.

5. BIOLOGY OF INTERLEUKIN-2 AND PRECLINICAL STUDIES IN ACUTE LEUKEMIA

The IL-2 gene is located on chromosome 4 and encodes a 15-17 kDa glycopro-tein that is secreted in response to antigen activation primarily by CD4-positiveT-cells, playing a critical role in the immune response. IL-2 induces proliferationand activation of T-cells, B-cells, and natural killer (NK) cells, and also leads tothe generation of LAK cells. Furthermore, it stimulates the production of sec-ondary cytokines by IL-2-responsive cells, potentiating its biologic effect.

Preclinical leukemia studies should sensitivity to cytolysis and/or growthinhibition mediated by IL-2-activated effector cells in vitro (52–55). Theseexperiments also demonstrated that human primary leukemia blasts resistant toNK cells were sensitive to normal cytotoxic effectors activated by IL-2 (56). Inanimal models, IL-2 with or without LAK cells was capable of eradicatingmurine leukemias (57,58). LAK cells, normally decreased in the majority ofnewly diagnosed patients with acute leukemia, were also shown to stop the invivo growth of human leukemia cell lines (51).

Few leukemia blasts express a functional IL-2 receptor. Therefore, the pos-sibility of stimulating leukemia clones is limited, and, in most instances, IL-2does not induce proliferation of blasts in vitro (59,60). Furthermore, experi-ments on nude mice demonstrated that IL-2 inhibits the growth of leukemiacells (59). These encouraging preclinical results, together with the inverse cor-relation of the LAK cell activity and disease activity in acute leukemia, as wellas the limited clinical success of IL-2 in other metastatic cancers, formed thebasis for testing IL-2 in acute leukemia.

6. TOXICITY AND BIOLOGIC EFFECT OF IL-2

IL-2 is administered by continuous infusion, by bolus intravenous (iv) injec-tion, or subcutaneously. Its use is associated with dose-dependent toxicity. Mostpatients experience side effects that may require dose reduction or discontinua-


tion of treatment and are mainly related to the capillary-leak syndrome (61–63).More common side effects may include a fever, nausea, diarrhea, cutaneousrashes, pruritus, increases in liver and renal parameters, fluid retention, weightgain, hypotension, confusion, and cardiac arrhythmias (62). All patients showsome degree of hepatomegaly and/or splenomegaly and variable leukocytosis.The major side effects, although potentially lethal, usually disappear after dis-continuation of IL-2 treatment and their duration can be shortened by corticos-teroids. High-dose IL-2 (> 8 × 106/m2/d) treatment is usually administered underclose supervision in the hospital for no more than 4–5 d due to its toxicity. Theleukocyte differential count is affected by IL-2 and varies throughout the regi-men. IL-2 infusion is accompanied by neutrophilia and increased circulatingeosinophils. After discontinuation of IL-2, a rebound lymphocytosis, with anincrease in large granular lymphocytes, can be noted. These changes are mostlikely in response to the secondary release of cytokines triggered by IL-2.Changes in peripheral blood are accompanied by an increase in lymphocytesand eosinophils in the bone marrow (62). One of the major complications oflong-term high-dose IL-2 treatment is marked thrombopenia, primarily due tosequestration (64) and the antimegakaryocytic effects of LAK cells (65). Low-dose IL-2 treatment (0.5×106/m2/d), on the other hand, may be safely adminis-tered for weeks (66,68). As expected, the side effects are much less pronounced,but this regimen can also effectively increase the number of circulating cytotoxiccells (69). Patients with IL-2 maintenance therapy develop a persistent lympho-cytosis with an increase in large granular lymphocytes.

The infusion of high-dose IL-2 induces marked phenotypic and functionalmodifications of the immune markers in patients with acute leukemia (70–71),leading to an increase in the number of circulating and bone marrow CD3-pos-itive cells and an amplification of cytotoxic cells (CD16 positive, CD56 posi-tive), especially in the CD3-negative NK population. Expression of activationmarkers HLA-Dr and TAC on circulating and bone marrow lymphocytes canalso be observed. There is an increase in NK cell activity and IL-2-inducedLAK activity. LAK cells are activated not only in the peripheral blood but alsoin the bone marrow, with the potential to attack the site of minimal residualleukemia. The peak of biologic effects coincides with maximum lymphocyto-sis that occurs 2 to 3 d after IL-2 discontinuation. As mentioned, a secondaryeffect of IL-2 administration is the increase of circulating cytokines andgrowth factors (TNF, IFN, IL-3, IL-5, and GM-CSF). This increase incytokines can explain transient leukocytosis and eosinophilia, whereas theincrease in growth factors may potentially increase the susceptibility to cyto-toxic agents, as documented in some patients receiving IL-2 (72).

7. USE OF INTERLEUKIN-2 IN OTHER MALIGNANCIES

Before its use in hematologic malignancies, IL-2 was tested and shown to bebeneficial in other malignancies. IL-2 administered to patients with metastatic


melanoma and renal cell cancer induced CR in 1% to 5% and partial remissionwith shrinkage of tumor metastases in 5% to 15% of patients (73). Even moreencouraging were the initial reports of IL-2 use in combination with LAK cellsin renal cell carcinoma, with CR rates in up to 30% of patients (74–76). Largermulticenter studies show, however, a more modest response independent of theuse of LAK cells (77).

8. CLINICAL USE OF INTERLEUKIN-2 IN ACUTE LEUKEMIA

8.1. Interleukin-2 as a Single AgentThe first IL-2 clinical trials in patients with acute leukemia focused on

demonstrating an antileukemia effect and included patients with advanced dis-ease refractory to chemotherapy (63,69,78). These initial European studies alsoevaluated the toxicity of high-dose IL-2 regimens and documented the feasibil-ity of administration to patients with acute leukemia (57). An objective responsewas seen in 20% of patients (79). Complete remission with IL-2 as a singleagent was achieved only in 17% of patients with AML. In patients with ALL,IL-2 alone failed to induce CR but significantly reduced the peripheral blastcounts. An antitumor effect was also documented with lower doses of IL-2, andthese regimens produced significant immune stimulation as well (79). For somepatients who did not respond to single-agent treatment with IL-2, there was evi-dence of new sensitivity to previously ineffective chemotherapy (80).

8.2. Use of IL-2 with Other Therapeutic ModalitiesAlthough IL-2 had only modest efficacy as a single agent, results were suffi-

ciently encouraging to warrant further investigation. A setting for further IL-2clinical testing seemed to be in minimal residual leukemia present at the timeof first or second remission (81). A variety of postremission treatment strate-gies, such as myelosuppressive maintenance therapy, multiple and prolongedcycles of consolidation therapy, autologous BMT, and alloBMT achievedleukemia-free survival for 25–45% patients in first CR (82–84). The activity oflow-dose maintenance chemotherapy is debatable in adult ALL and is of nobenefit in AML. Therefore, clinical trials examined the use of IL-2 in remissionand in stem cell transplantation (Table 1). In a phase I trial, Robinson et al.tested the use of IL-2 after alloBMT in children with acute leukemia beyondfirst remission (85). These patients, in CR without active GVHD after unmodi-fied allogeneic matched-sibling BMT, received escalating doses of IL-2. IL-2was administered as a continuous infusion for 5 d (0.9, 2.0, 6.0 × 106 IU/m2/d).After 6 d of rest, the patients received IL-2 maintenance therapy (0.9 × 106

IU/m2/d) for 10 d by continuous infusion. At the time of publication, 10 of 17patients were in sustained CR 5 to 67 mo after transplantation.

In a pilot trial by Margolin et al., patients with poor prognosis leukemia andlymphoma received IL2-activated autologous bone marrow and peripheral


blood stem cell transplantation followed by IL-2 treatment (86). Patients firstreceived bone marrow or G-CSF-mobilized autologous peripheral blood stemcells that were exposed to IL-2 for 24 h ex vivo. This treatment was followedby low-dose IL-2 infusion until hematologic reconstitution and later by inter-mediate-dose continuous infusion for six 2-wk maintenance cycles at 1-mointervals. Of the 12 patients with acute leukemia treated, 2 with ALL were inCR at 38 and 43 mo, 1 with ALL died, and 3 of 9 patients with AML were inCR after 21, 46, and 53 mo.

In vivo IL-2 mobilization for the purpose of autologous peripheral bloodprogenitor cell (PBPC) transplantation was studied by Schiller et al. (87). IL-2was administered after recovery from high-dose cytarabine-based consolida-tion chemotherapy with G-CSF as mobilization of peripheral blood progenitorcells for 49 patients with AML in first remission. Forty-one patients receivedmyeloablative chemoradiotherapy followed by the infusion of these IL-2-mobilized autologous PBPC. IL-2 was administered after transplantation at the


Table 1Interleukin-2 in the Treatment of Acute Leukemia

Author Year Subjects Disease Therapy Results

Robinson 1996 17 children ALL in alloBMT fol- 10 of 17 et al. (85) CR1 lowed by patients in

IL-2 infusion CRMargolin 1999 12 adults AML and Ex vivo IL-2 CR in 2 of 3

et al. (86) ALL in treated PBSC, patients with CR1 IL-2 infusion ALL; 3 of 9

patients inAML

Schiller 2001 49 adults AML in IL-2 mobilized CR in 49% et al. (87) CR1 autologous with no

PBPC, SQ clear benefit IL-2 from IL-2

treatmentAttal et al. (46) 1995 60 adults ALL in aBMT followed No survival

CR1 by IL-2 benefit from infusion IL-2

treatmentBlaise et al. (88) 2000 130 adults AML and aBMT followed No survival

ALL in by IL-2 benefit from CR1 infusion IL-2

treatment

ALL = acute lymphocytic leukemia; alloBMT = allogenous bone marrow transplantation;BMT = autologous bone marrow transplantation; IL = interleukin; CR = complete remission;AML = acute myeloid leukemia; CR1 = first complete remission, PBSC = peripheral blood stemcells, PBPC = peripheral blood progenitor cells, SQ = subcutaneously.

same dose and schedule as given after consolidation/mobilization (3 × 106

units subcutaneously twice daily for 10 d). After a median follow-up fromremission of 15 mo, actuarial leukemia-free survival for all patients from timeof CR was 49%. The authors demonstrated the feasibility of autologous trans-plantation of IL-2-mobilized PBSC for an unselected population of adultpatients with AML in first CR, but there was no clear evidence of survival ben-efit in this phase II study.

In an attempt to simulate a GVL effect, Attal et al. studied the effect of IL-2after autologous BMT in patients with ALL in first CR (46). In their random-ized multicenter prospective study of 60 patients with ALL in first remissionundergoing autologous BMT, 30 patients were assigned to receive continuousIL-2 infusion for a total of five cycles every other week. The first 5-d cycle wasfollowed by four 2-d cycles. IL-2 was administered at a dose of 12 × 106

U/m2/d. The probability of continued remission 3 yr after autologous BMTwas similar for both groups, 29% for those treated with IL-2, and 27% forthose not receiving IL-2. IL-2 did not decrease the high relapse rate after autol-ogous BMT. Unfortunately, the authors chose to examine a disease state inwhich little GVL can be demonstrated and in which DLIs after unsuccessfulalloBMT are ineffective.

Blaise et al. tried to clarify the effect of IL-2 immunotherapy in acuteleukemia during first CR (88). They conducted a prospective multicenter ran-domized trial that included patients with both AML and ALL in first CR afterautologous BMT. Seventy-eight patients with AML and 52 patients with ALLwere randomized to receive five IL-2 cycles (12 × 106 IU/m2/d) after autologousBMT. Of 65 patients randomized into the study group, 38 (59%) started the IL-2treatment. A total of 13 patients started each of the five cycles, 7 of these patientsreceived more than 95% of the total scheduled dose. The data analysis wasbased on an intent to treat, with a median follow-up of 7 yr. No difference in out-come was observed between patients in the study and control groups. In thestudy group, relapse occurred in 66% of patients, compared with 55% ofpatients in the control group. Survival was 33% vs 43%, and leukemia-free sur-vival was 29% vs 36% in the study and control groups, respectively. Althoughthis study failed to show any benefit of IL-2 treatment in acute leukemia, the lowrate of compliance, partly due to toxicity, might have obscured any difference inoutcomes. Further trials must address both the timing of IL-2 administrationafter autologous BMT as well as the toxicity of the regimen.

9. CONCLUSIONS

After more than 10 yr of clinical experience with IL-2 in acute leukemia, itsplace in treatment has not been established. The biologic effects of IL-2, aswell as results of some preliminary clinical studies in the relapsed setting,point to the potential benefit of immunotherapy in the treatment of acute


leukemia. Unfortunately, no single dose or delivery system of IL-2 has beenshown to be more effective in improving survival or in alternating surrogatebiologic endpoints. Randomized clinical trials of IL-2 as a single agent, in con-junction with autologous or allogeneic transplantation of stem cells, have beendisappointing. None of the randomized trials showed benefit in long-term sur-vival when compared with standard regimens, although the Attal study was notoptimal. The Blaise study attempted to address a population best suited for animmune-mediated antineoplastic effect; however, small numbers would havemade detecting a small benefit impossible. IL-2 has, however, helped to bringattention to an immunologic approach in the treatment of acute leukemia. Abetter understanding of the biology of the disease and a combination of multi-ple immunotherapeutic agents may be needed to induce an immunologicallyactive effect and potentially improve survival in patients with acute leukemia.

REFERENCES1. Cassileth PA, Begg CB, Bennett JM, et al. A randomized study of the efficacy of consolida-

tion therapy in adult acute nonlymphocytic leukemia. Blood 1984;63:843–847.2. Gale RP, Foon KA. Therapy for acute myelogenous leukemia. Semin Hematol 1987;24:40–54.3. Dutcher JP, Wiernik PH, Markus S, et al. Intensive maintenance therapy improves survival in

adult acute nonlymphocytic leukemia: an eight-year follow-up. Leukemia 1988;2:413–419.4. Mandelli F, et al. GIMEMA experience in the treatment of adult acute myelogenous

leukemia. In: Gale RP, ed. Acute Myelogenous Leukemia: Progress and Controversies. NewYork: Wiley; 1990:273–291.

5. Zittoun RA, Mandelli F, Willemze R, et al. Autologous or allogeneic bone marrow transplan-tation compared with intensive chemotherapy in acute myelogenous leukemia. N Engl J Med1995;332:217–223.

6. Copelan EA, McGuire EA. The biology and treatment of acute lymphoblastic leukemia inadults. Blood 1995;85:1151–1168.

7. Pui C-H, Kane JR, Crist WM. Biology and treatment of infant leukemias. Leukemia1995;9:762–769.

8. Preti A, Kantarjian HM. Management of adult acute lymphocytic leukemia: present issuesand key challenges. J Clin Oncol 1994;12:1312–1322.

9. Foon KA, Smalley RV, Riggs CW, et al. The role of immunotherapy in acute myelogenousleukemia. Arch Intern Med 1983;143:1726–1731.

10. Thomas ED. Bone marrow transplantation for malignant disease. J Clin Oncol 1983;1:517–531.11. Thomas ED. The role of bone marrow transplantation for eradication of malignant disease.

Cancer 1982;49:1963–1969.12. Weiden PL, Flournoy N, Thomas ED, et al. Antileukemic effect of graft-versus-host disease

in human recipients of allogeneic marrow grafts. N Engl J Med 1979;300:1068–1073.13. Champlin R, Khouri I, Kornblau S, et al. Allogeneic hematopoetic transplantation as adop-

tive immunotherapy. Hematol Oncol Clin N Am 1999;13:1041–1057.14. Horowitz MM, Gale RP, Sondel PM, et al. Graft-versus-leukemia reactions after bone mar-

row transplantation. Blood 1990;75:555–562.15. Sullivan KM, Storb R, Buckner CD, et al. Graft-versus-host disease as adoptive immunother-

apy in patients with advanced hematologic neoplasms. N Engl J Med 1989;320:828–834.16. Weiden PL, Sullivan KM, Flournoy N, et al. for the Seattle Marrow Transplant Team.

Antileukemic effect of chronic graft-versus-host disease: contribution to improved survivalafter allogeneic marrow transplantation. N Engl J Med 1981;304:1529–1533.


17. Yeager A, Vogelsang G, Jones R, et al. Induction of cutaneous graft-versus-host disease byadministration of cyclosporine to patients undergoing autologous bone marrow transplanta-tion for acute myeloid leukemia. Blood 1992;79:3031–3035.

18. Fefer A, Cheever MA, Greeberg PD. Identical-twin (syngeneic) marrow transplantation forhematologic cancers. J Natl Cancer Inst 1986;76:1269–1273.

19. Gale RP, Champlin RE. How does bone marrow transplantation cure leukemia? Lancet1984;2:28–30.

20. Gale RP, Horowitz MM, Ash RC, et al. Identical-twin bone marrow transplants for leukemia.Ann Intern Med 1994;120:646–652.

21. Marmont AM, Horowitz MM, Gale RP, et al. T-cell depletion of HLA-identical transplantsin leukemia. Blood 1991;78:2120–2130.

22. Berman E, Little C, Gee T, et al. Reasons that patients with acute myelogenous leukemia donot undergo allogeneic bone marrow transplantation. N Engl J Med 1992;326:156–160.

23. Rosenberg SA, Lotze JT, Muul LM, et al. Observations on the systemic administration ofautologous lymphokine-activated killer cells and recombinant interleukin-2 to patients withmetastatic cancer. N Engl J Med 1985;313:1485–1492.

24. Rosenberg SA, Packard BS, Aebersold PM, et al. Use of tumor-infiltrating lymphocytes andinterleukin-2 in the immunotherapy of patients with metastatic melanoma: a special report.N Engl J Med 1988;319:1676–1680.

25. Rosenstein M, Yron I, Kaufman Y, et al. Lymphokine activated killer cells: lysis of fresh syn-geneic NK-resistant murine tumor cells by lymphocytes cultured in interleukin-2. CancerRes 1984;44:1946–1953.

26. Rosenberg SA, Lotze MT, Muul LM, et al. A progress report on the treatment of 157 patientswith advanced cancer using lymphokine activated killer cells and interleukin-2 or high doseinterleukin-2 alone. N Engl J Med 1987;316:889–897.

27. Foa R, Fierro MT. Cesano, A. et al. Defective lymphokine-activated killer cell generationand activity in acute leukemia patients with active disease. Blood 1991;78:1041–1046.

28. Carson WE, Caligiuri MA. Cellular adoptive immunotherapy with natural killer cells. In:Morstyn G and Sheridan W, eds. Cell Therapy: Stem Cell Transplantation, Gene Therapy,and Cellular Immunotherapy. Cambridge: Cambridge University Press; 1996:451–474.

29. Rosenberg SA, Spiess P, Lafreniere R. A new approach to the adoptive immunotherapy ofcancer with tumor-infiltrating lymphocytes. Science 1986;223:1318–1321.

30. Fisher B, Packard BS, Read EJ, et al. Tumor localization of adoptively transferred indium-111 labeled tumor infiltrating lymphocytes in patients with metastatic melanoma. J ClinOncol 1989;7:250–261.

31. Griffith KD, Read EJ, Carrasquillo CS, et al. In vivo distribution of adoptively transferredindium-111 labeled tumor infiltrating lymphocytes and peripheral blood lymphocytes inpatients with metastatic melanoma. J Natl Cancer Inst 1989;81:1709–1717.

32. Riddell SR, Greenberg PD. Principles for adoptive T cell therapy of human viral disease.Ann Rev Immunol 1995;13:545–586.

33. Riddell SR, Watanabe KS, Goodrich JM, et al. Restoration of viral immunity in immunodefi-cient humans by the adoptive transfer of T cell clones. Science 1992;257:238–241.

34. Kolb HJ, Mittermuller J, Clemm CH, et al. Donor leukocyte transfusions for treatment ofrecurrent chronic myelogenous leukemia in marrow transplant patients. Blood1990;76:2462–2465.

35. Collins RH, Shpilberg O, Drobyski WR, et al. Donor leukocyte infusions in 140 patientswith relapsed malignancy after allogeneic bone marrow transplantation. J Clin Oncol1997;15:433–444.

36. Soiffer RJ. Donor lymphocyte infusions: the door is open. J Clin Oncol 1997;15:416–417.37. Giralt SA, Kolb HJ. Donor lymphocyte infusion. Curr Opin Oncol 1996;8:96–102.38. Kolb HJ, Schattenberg A, Goldman JM, et al. Graft-versus-leukemia effect of donor lympho-

cyte transfusions in marrow grafted patients. Blood 1995;86:2041–2050.


39. van Rhee F, Lin F, Cullis JO, et al. Relapse of chronic myeloid leukemia after allogeneicbone marrow transplant: the case for giving donor leukocyte transfusions before the onset ofhematologic relapse. Blood 1994;83:3377–3383.

40. Antin JH. Graft-versus-leukemia: no longer an epiphenomenon. Blood 1993;82:2273–2277.41. Cullis JO, Jiang YZ, Schwarer AP, et al. Donor leukocyte infusions for chronic myeloid

leukemia in relapse after allogeneic bone marrow transplantation. Blood 1992;79:1379–1381.42. Drobyski WR, Keever CA, Roth MS, et al. Salvage immunotherapy using donor leukocyte

infusions as treatment for relapsed chronic myelogenous leukemia after allogeneic bone mar-row transplantation: efficacy and toxicity of a defined T-cell dose. Blood 1993;82:2310–2318.

43. Mackinnon S, Papadopoulos EB, Carabasi MH, et al. Adoptive immunotherapy evaluatingescalating doses of donor leukocytes for relapse of chronic myeloid leukemia after bonemarrow transplantation: separation of graft-versus-leukemia responses from graft-versus-host disease. Blood 1995;86:1261–1268.

44. Klingemann HG. Introducing graft-versus-leukemia into autologous stem cell transplanta-tion. J Hematother 1995;4:261–267.

45. Jones R, Vogelsang G, Hess A, et al. Induction of graft-versus-host disease after autologoustransplantation. Lancet 1991;9:754–757.

46. Klingemann H, Grigg A, Wilkie-boyd K, et al. Treatment with recombinant interferon a-2bearly after bone marrow transplantation in patients at high risk for relapse. Blood1991;78:3306–3311.

47. Uckun FM, Kersey JH, Haake R, et al. Pretransplantation burden of leukemic progenitorcells as a predictor of relapse after bone marrow transplantation for acute lymphoblasticleukemia. N Engl J Med 1993;329:1296–1301.

48. Attal M, Blaise D, Marit G, et al. Consolidation treatment of adult acute lymphoblasticleukemia: a prospective, randomized trial comparing allogeneic versus autologous bonemarrow transplantation and testing the impact of recombinant interleukin-2 after autologousbone marrow transplantation. Blood 1995;86:1619–1628.

49. Fefer A. Graft-versus-tumor responses: adoptive cellular therapy in bone marrow transplan-tation. In: Forman SJ, Blume KG, Thomas ED, eds. Bone Marrow Transplantation. Boston;Blackwell; 1994:231–241.

50. Fefer A, Robinson N, Benyunes MC. Interleukin-2 therapy after autologous stem cell trans-plant for acute myelogenous leukemia in first remission. In: DickeK, Keating A, eds. Autolo-gous Blood and Marrow Transplantation: Proceedings of the Ninth International Symposium.Arlington, TX; Carden Jennings, 1999:46–53.

51. Margolin KA, Wright C, Forman SJ. Autologous bone marrow purging by in situ IL-2 acti-vation of endogenous killer cells. Leukemia 1997;9:723–728.

52. Dawson MM, Johnston D, Taylor GM, et al. Lymphokine activated killing of fresh humanleukemias. Leuk Res 1986;10:683–688.

53. Lista P, Fierro MT, Liao X-S, et al. Lymphokine-activated killer (LAK) cells inhibit theclonogenic growth of human laukemic stem cells. Eur J Haematol 1989;42:425–430.

54. Lotzova E, Savary CA, Herberman RB. Inhibition of clonogenic growth of fresh leukemiacells by unstimulated and IL-2 stimulated NK cells of normal donors. Leuk Res1987;11:1059–1066.

55. Oshimi K, Oshimi Y, Akutsu M, et al. Cytotoxicity of interleukin-2-activated lymphocytesfor leukemia and lymphoma cells. Blood 1986;68:938–948.

56. Fierro MT, Biao XS, Lusso P, et al. In vitro and in vivo susceptibility of human leukemiccells to lymphokine activated killer activity. Leukemia 1988;2:50–54.

57. Lim SH, Newland AC, Kelsey S, et al. Continuous intravenous infusion of high-dose recom-binant interleukin-2 for acute myeloid leukaemia: a phase II study. Cancer ImmunolImmunother 1992;34:337–342.

58. Toze CL, Barnett MJ, Klingemann HG. Response of therapy-related myelodysplasia to low-dose interleukin-2. Leukemia 1993;7:463–465.


59. Foa R, Caretto P, Fierro MT, et al. Interleukin 2 does not promote the in vitro and in vivoproliferation and growth of human acute leukaemia cells of myeloid and lymphoid origin. BrJ Haematol 1990;75:34–40.

60. Touw I, Groot-Loonen J, Broeders L, et al. Recombinant hematopoietic growth factors fail toinduce a proliferative response in precursor B acute lymphoblastic leukemia. Leukemia1989;3:356–362.

61. Blaise D, Viens P, Olive D, et al. Hematologic and immunologic effects of the systemicadministration of recombinant Interleukin-2 after autologous bone marrow transplantation.Blood 1990;76:1092–1097.

62. Foa R. Interleukin 2 in the management of acute leukemia. Br J Haematol 1996;92:1–8.63. Maranichi D, Blaise D, Viens P, et al. High-dose recombinant interleukin 2 and acute

myeloid leukemias in relapse. Blood 1991;78:2182–2187.64. Paciucci PA, Mandeli J, Oleksowiz L, et al. Thrombocytopenia during immunotherapy with

interleukin-2 by constant infusion. Am J Med 1990;89:308–312.65. Guarini A, Sanavio F, Novarino A, et al. Thrombocytopenia in acute leukaemia patients

treated with IL2: cytolytic effect of LAK cells on megakaryocytic progenitors. Br J Haema-tol 1991;79:451–456.

66. Blaise D, Olive D, Stoppa AM, et al. Hematologic and immunologic effects of the systemicadministration of recombinant Interleukin-2 after autologous bone marrow transplantation.Blood 1990;76:1092–1097.

67. Meloni G, Foa R, Vignetti M, et al. Interleukin-2 may induce prolonged remissions inadvanced acute myelogenous leukemia. Blood 1994;84:2158–2163.

68. Soiffer RJ, Murray C, Gonin R, et al. Effect of low dose interleukin-2 on disease relapseafter T-cell depleted allogeneic bone marrow transplantation. Blood 1994;84:964–968.

69. Soiffer RJ, Murray C, Cochran K, et al. Clinical and immunologic effects of prolonged infu-sion of low-dose recombinant interleukin-2 after autologous and T-cell-depleted allogeneicbone marrow transplantation. Blood 1992;79:517–526.

70. Foa R, Guarini A, Gillio Tos A, et al. Peripheral blood and bone marrow immunophenotypicand functional modifications induced in acute leukemia patients treated with interleukin 2: evi-dence of in vivo lymphokine activated killer cell generation. Cancer Res 1991;51:964–968.

71. Gottlieb DJ, Prentice HG, Heslop HE, et al. Effects of recombinant interleukin-2 administra-tion on cytotoxic function following high-dose chemo-radiotherapy for hematological malig-nancy. Blood 1989;74:2335–2342.

72. Foa R, Meloni G, Tosti S, et al. Treatment of acute myeloid leukaemia patients with recom-binant interleukin 2: a pilot study. Br J Haematol 1991;77:491–496.

73. Rosenberg SA. Perspectives on the use of IL2 in cancer treatment. Cancer J 1997;3:52–56.74. Belldegrun A, Figlin R, Haas GP, deKernion JB. Immunotherapy for metastatic renal cell

carcinoma. World J Urol 1991;9:157–165.75. Ilson DH, Motzer RJ, Kradin RL, et al. A phase II trial of IL-2 and interferon alpha-2a in

patients with advanced renal cell carcinoma. J Clin Oncol 1992;10:1124–1130.76. Sleifter DT, Janssen RA, Butler J, et al. Phase II study of subcutaneous IL-2 in unselected

patients with advanced renal cell cancer in an outpatient basis. J Clin Oncol1992;10:1119–1123.

77. McCabe MS, Stablein D, Hawkins MJ. The modified group C experience-phase III random-ized trials of IL-2 vs IL-2/LAK in advanced renal cell carcinoma and advanced melanoma.Proc Am Soc Clin Oncol 1991;10:213.

78. Adler A, Chervenick PA, Whiteside TL, et al. Interleukin 2 induction of lymphokine-acti-vated killer (LAK) activity in the peripheral blood and bone marrow of acute leukemiapatients. I. Feasibility of LAK generation in adult patients with active disease and in remis-sion. Blood 1988;71:709–716.

79. Blaise D, Maranichi D. Interleukin 2 in the treatment of acute leukemia. Leukemia Res1998;22:1165–1170.


80. Foa R. Does interleukin-2 have a role in the management of acute leukemia? J Clin Oncol1993;11:1817–1825.

81. Macdonald D, Jiang Y, Gordon A, et al. Recombinant interleukin-2 for acute myeloidleukemia in first complete remission: a pilot study. Leuk Res 1990;1:967–973.

82. Applebaum FR, Dahlberg S, Thomas ED, et al. Bone marrow transplantation or chemotherapyafter remission induction for adults with acute nonlymphoblastic leukemia. 1984;101:581–588.

83. Mayer RJ, Davis RB, Schiffer CA, et al. Intensive postremission chemotherapy in adultswith acute myeloid leukemia. New Engl J Med 1994;331:896–903.

84. Schiller GJ, Nimer SD, Territo MC, et al. Bone marrow transplantation versus high-dosecytarabine-based consolidation chemotherapy for acute myelogenous leukemia in firstremission. J Clin Oncol 1992;10:41–46.

85. Robinson N, Sanders JE, Benyunes MC, et al. Phase I trial of interleukin-2 after unmodifiedHLA-matched sibling bone marrow transplantation for children with acute leukemia. Blood1996;87:1249–1254.

86. Margolin KA, Van Besien K, Wright C, et al. Interleukin-2-activated autologous bone mar-row and peripheral blood stem cells in the treatment of acute leukemia and lymphoma. BiolBlood Marrow Transplant 1999;5:36–45.

87. Schiller G, Wong S, Lowe T, et al. Transplantation of IL-2 mobilized autologous peripheralblood progenitor cells for adults with acute myelogenous leukemia in first remission.Leukemia 2001;15:757–763.

88. Blaise D, Attal M, Reiffers J, et al. Randomized study of recombinant interleukin-2 afterautologous bone marrow transplantation for acute leukemia in first complete remission. EurCytokine Netw 2000;11:91–98.


III TARGETED THERAPEUTICS

1. INTRODUCTION

The goal of 21st-century leukemia treatment is to discover and validate tar-gets for new active and specific therapies. Improved understanding of the mol-ecular biology of leukemia has identified a series of target genes whose proteinproducts are integral to the pathogenesis of leukemia. Much of this insight hasinitially been garnered from cytogenetic observations and clinical correlations.These sentinel discoveries have been bolstered by the availability of genomicand proteomic screening.

Most of the conventional antileukemic agents have nonspecific cytotoxiceffects on deoxyribonucleic acid (DNA) polymerase or protein synthesis.Nucleic acids can be designed to target specific genes. Antisense oligonu-cleotides are 15–25 bases (mers) of DNA designed to target by Watson-Crickbase pairing a unique sequence of messenger ribonucleic acid (mRNA) andlead to degradation of that specific message to eradicate the production of thecoded protein. Although other strategies using nucleic acids are being tested(1), the use of antisense oligonucleotides has progressed to the clinic and is thefocus of this chapter.

Although preclinical studies have identified multiple potential targets, bcl-2has emerged as the predominant target for advanced-stage large clinicalleukemia studies. This chapter briefly reviews the basic components of anti-

109


8 Antisense Therapy

Stanley R. Frankel, MD, FACP

CONTENTS

INTRODUCTION

FUNDAMENTAL COMPONENTS OF ANTISENSE

OLIGONUCLEOTIDES

TARGETS FOR ANTISENSE THERAPY

CONCLUSIONS

REFERENCES

sense oligonucleotide therapy and then focuses on preclinical and clinicalleukemia studies.

2. FUNDAMENTAL COMPONENTS OF ANTISENSEOLIGONUCLEOTIDES

During the last two decades, advances in chemical formulation and genesequencing have made available antisense reagents for clinical use. The effi-cacy of such compounds depends on their ability to enter leukemic cells and toeffectively target the mRNA of interest and on the clinical effect of reductionof the target protein.

Antisense therapy involves the administration of synthetic oligonucleotidesthat are complementary to specific mRNA transcripts. Antisense targets themRNA of interest by Watson-Crick base pairing, thereby providing specificityand avidity.

Antisense oligonucleotides must be incorporated into cells to be effective.Although they may be active at nanomolar to micromolar concentrations,uptake varies with the cell type. Intracellular uptake occurs by pinocytosis.

Inside the cell, the antisense oligodeoxynucleotide hybridizes to the specificmolecule target mRNA to form a DNA/RNA duplex. Ribonuclease H (RNaseH) recognizes the DNA/RNA duplex, cleaves the mRNA strand, and rendersthe message nontranslatable. The specific targeted mRNA fragments are subse-quently destroyed by ribonucleases. As a consequence, levels of the target pro-tein are diminished, dependent on the effectiveness of the antisense and thehalf-life of the mRNA and protein.

2.1. Antisense ClassesAntisense compounds may differ in their size, ionic charge, and structural

relationship to the natural nucleic acid target. Antisense molecules are gener-ally ionically charged and relatively large. First-generation modifications ofthe oligo to stabilize them against nuclease digestion of the phosphodiesterlinkage included phosphoramidates, methyl phosphonates, phosphorothioates,phosphorodiamidate morpholinos, and α-oligonucleotides. The most commonbackbone replaces one of the oxygen atoms in the phosphodiester linkage witha sulfur atom to form a phosphorothioate. (1,2) Phosphorothioate (PS)oligonucleotides (oligos) contain at least one of the nonbridging oxygens ofthe internucleotide phosphodiester linkages replaced with sulfur. PS oligosinhibit gene expression by hybridization arrest (i.e., interference with the pro-cessing of mRNA by hybridization), followed by cleavage of the mRNA byRNase-H. These oligos are polyanionic and, as a result, can bind to several fac-tors that may produce nonspecific effects. Critical to their function is resistanceto nuclease digestion.

110 Frankel

2.2. Mechanism of Antisense ActionAntisense oligos must reach the target cell and remain in the cell at a concen-

tration sufficient to trigger their biologic effects. Intracellular uptake of oligos isdependent on physical variable, time, and concentration. At concentrationsbelow 1µ mol/L, uptake of PS oligodeoxynucleotides (ODN) is predominantlyvia a receptor-like mechanism, while at higher concentrations a fluid-phaseendocytosis mechanism predominates (3). ODN have been reported to beendocytosed via clathrin-coated pits on the cytoplasmic membrane. Severalclasses of receptor-like binding proteins have also been described (3). Theintracellular oligos may escape the endosome/lysosome compartment and thenmigrate rapidly into the nucleus. The mechanism is one of diffusion, with sub-sequent trapping of the oligo (4).

At the intracellular target, the oligo specifically binds to the sequence ofinterest by Watson-Crick base pairing. Affinity must be sufficient so that thetarget mRNA is inactivated. This inactivation can be due to either sequestrationof the message or, more likely, by cleavage of the mRNA when in the het-eroduplex configuration with the antisense oligo. A ubiquitous enzyme, RNaseH, cleaves only the RNA strand and leaves the antisense oligo intact to catalyt-ically bind to another strand of specific mRNA. Other mechanisms may alsoplay a role in the antisense effect (5).

3. TARGETS FOR ANTISENSE THERAPY

Antisense design requires a known sequence of a target mRNA. Initial cyto-genetic observations of translocations in leukemia directed early efforts at genemapping and sequence identification. Genes located at translocation break-points were believed to play a role in the pathogenesis of the disease andbecame reasonable therapeutic targets. As signaling pathways are unraveled inleukemia, additional targets for therapy with antisense approaches emerge. Ifthe protein encoded by the target gene (such as Bcl-2) is important in tumorcell survival, progression, or resistance to treatment, then antisense administra-tion may be beneficial.

3.1. BCR-ABLThe seminal discovery of the Philadelphia (Ph) chromosome by Nowell and

Hungerford offered the first rational target for antisense therapy. The subse-quent discovery, cloning, and sequencing of BCR-ABL that is the sine qua nonfor chronic myelogenous leukemia (CML) was an obviously attractive targetfor antisense development.

Antisense therapy directed against the bcr-abl fusion product or the individ-ual components has been investigated both ex vivo and in vivo (5–8). Therehave been conflicting reports regarding the efficacy and specificity of these

Chapter 8 / Antisense Therapy 111

approaches, and none has led to broad clinical investigation (1). The in vitroresults have been provocative, with early exploration of ex vivo purging strate-gies for autologous transplantation (9–11). There is some modulation of adher-ence and signaling in response to antisense treatment, which may bemodulated by the use of α interferon (12,13). However, the effects of the oli-gos may not be sequence specific (14). The design of antisense directed againstbcr-abl has been considerably more problematic than was initially assumed(8). In addition to drug delivery and targeting, the relatively long half-life ofthe fusion protein may limit the potential benefit of this approach (15).

3.2. C-mybDespite the initial attraction of a translocation-specific sequence, proteins that

played a more central role in neoplastic cell survival and growth based on theiroverexpression were also rational targets for antisense therapy. MYB protein is anuclear-binding protein that controls the G1/S transition. C-myb also functionsas a transactivator of other cellular control genes (16,17). There may be rela-tively increased sensitivity to downregulation of myb in leukemic cells. Myb’sability to transactivate genes required for cell proliferation and growth underliesits importance for normal hematopoietic cell development and suggests that per-turbation of its function may play a role in leukemogenesis (18,19). A modestsurvival benefit was shown when c-myb PS antisense ODN were tested in ahuman leukemia (K562) SCID mouse model (20). These chimeric mice weretreated with PS-modified antisense ODNs for c-myb. Animals treated with anti-sense c-myb oligos survived 3.5-fold longer than control animals. There was adecreased incidence of central nervous system (CNS) and ovarian infiltration inthe antisense treated animals. Although it may be difficult to deliver antisensetherapy in vivo, this approach may be feasible for ex vivo purging of bone mar-row. Constant infusion of even low doses of antisense ODNs was useful. Sup-pression of the target gene was clearly demonstrated in a melanoma model (21).A phase I trial of the c-myb PS at doses up to 2 mg/kg/d was tolerable, withsome stabilization of a patient with blast-crisis CML. This trial was terminatedprematurely because the drug was not available (22).

Because of the difficulty and expense of synthesizing large quantities ofclinical material, an initial approach was to use the agents in vitro to purgeautologous marrow of tumor cells in CML (23). Allograft-ineligible patientswith CML had CD34+ marrow cells purged with ODN for either 24 (n = 19) or72 (n = 5) h. After purging, Myb mRNA levels declined substantially inapproximately 50% of patients. Analysis of bcr/abl expression in long-termculture-initiating cells suggested that purging had been accomplished at aprimitive cell level in more than 50% of patients and was ODN dependent.Day-100 cytogenetics were evaluated in 13 surviving patients who engraftedand had evaluable metaphases. There were two complete cytogenetic remis-

112 Frankel

sions and three major cytogenetic responses, and eight patients remained 100%Ph-positive.

3.3. c-mycFunctional c-myc may be required for the development of the CML pheno-

type after bcr-abl activation. Antisense to c-myc had an antiproliferative effectagainst K562 and primary CML-blast crisis cells in a SCID mouse model,although there was a marked synergistic antiproliferative effect when antisenseto bcr-abl was also administered (24). A PS antisense to c-myc suppressedgrowth of HL-60 cells, likely by an apoptotic response. However, c-mycexpression rebounded at the end of exposure (25). Although phosphoroamidateconstructs were more effective than a PS oligos in vitro, there was therapeuticequivalency in vitro (26). A 15 mer targeted against the translation initiationcodon site of c-myc was not active in vitro unless placed in a fusogenic lyso-some (26). This compound inhibited 70% of HL-60 growth at a concentrationof 2.5 µm. A triple helix-forming oligo with a PS backbone targeting c-myc P2promoter has also been shown to have sequence specific activity. CEMleukemia cells accumulated in S-phase, and apoptosis was induced (27).

3.4. p53Another target for modulation was p53. A 20-mer PS was used. Sixteen

patients with either refractory acute myeloid leukemia (AML) (n = 6) oradvanced myelodysplastic syndrome (MDS) (n = 10) were treated with 1.2–6mg/kg/day for 10 d by continuous intravenous (iv) infusion. Although leukemiccell growth in vitro was inhibited when compared with pretreatment samples,there were no clinical complete responses (28). A purging strategy was also ini-tiated. Bone marrow cells were incubated with 10 µm OL(1)p53 for 36 h beforetransplantation. Although safety was established based on reasonable time toplatelet and neutrophil engraftment, no efficacy correlates were reported (29).

3.5. bcl-23.5.1. BCL-2: ANTI-APOPTOTIC MECHANISMS

Much of the understanding concerning the molecular regulation of pro-grammed cell death originates with the B-cell lymphoma/leukemia associatedgene 2 (bcl-2) gene family. The best characterized members of this family, Bcl-2 and Bax, act by differential homodimerization and heterodimerization. Bcl-2homodimers act as repressors of apoptosis, whereas Bcl-2/Bax heterodimersact as promoters (30). These effects are more dependent on the balancebetween Bcl-2 and Bax than on Bcl-2 quantity alone (31). During the last 5 yr,many new members of the Bcl-2 family have been characterized, including thedeath promoters BCL-xS (34), BAD (35), BIK (34), BAK (35), and the apop-tosis suppressors BCL-xL (32) and BFL-1 (36).


BCL-2 was originally identified because of its association with the t(13,17)chromosomal translocation present in most follicular B-cell non-Hodgkin’slymphomas (37,38). With this translocation, the Bcl-2 gene is moved from itsnormal chromosomal location at 18q21 into juxtaposition with theimmunoglobulin heavy chain (IgH) locus at 14q32, (resulting in deregulationof the Bcl-2/IgH fusion gene. As a result, Bcl-2 protein is overexpressed.

Bcl-2 protein either forms selective membrane pores or more likely controlspre-existing permeability transition pores (37). Bcl-2 is able to stabilize themitochondrial transmembrane potential (39) and antagonize Bax-inducedrelease of cytochrome c into the cytoplasm (40). Immunoelectron microscopyhas shown the association of Bcl-2 with the mitochondrial membrane (41).Functional analysis has shown that Bcl-2 acts upstream of caspases, preventingtheir activation (42). Moreover, earlier studies on mitochondria showed highertransmembrane potential in tumor cells relative to normal cells (43). Bcl-2 caninhibit the nuclear import of wt p53 after DNA damage (44).

Cancers are characterized by dysregulation of proliferation, differentiation,and balance between cell survival and cell death, with the balance typicallyshifted toward prolonged survival. Such survival plays a major role in the resis-tance to cytotoxic agents exhibited by these refractory leukemias and is relatedto abrogation of apoptosis pathways that would normally be activated by keyantileukemic drugs, including ara-C and anthracyclines (45,46).

One approach to augment net drug cytotoxicity relates to modulation of thepathways, culminating in apoptosis. Apoptosis is a normal physiologic processby which cellular life span is controlled and specific cells are programmed forelimination. Activation of apoptosis pathways may serve as a common finalpathway for many cytotoxic agents, irrespective of the primary mechanism ofaction. In fact, induction of apoptosis in response to DNA strand breakage mayat least in part explain the cytotoxic activity of cell cycle-active antileukemicnucleoside drugs (ara-C) against noncycling cells, in addition to the cycling cellpopulation (47). The induction of DNA strand breaks and thus the activation ofapoptosis pathways is likely to be a common mechanistic motif for drugs thatinteract with topoisomerases, including anthracyclines and epipodophyllotoxins.

Taking these distinctive findings together, it is apparent that BCL-2 and itsfamily members are convergence points for diverse determinants of normal andmalignant hematopoietic cell survival. As such, Bcl-2 presents a pivotal molec-ular target for therapy of the acute leukemias. The ability to downregulate netBcl-2 activity could translate into an increase in the antileukemic activity ofagents that exert at least some of their cytotoxic effects by inducing apoptosis.

3.5.2. BCL-2 EXPRESSION IN LEUKEMIA

Bcl-2 is upregulated in chronic lymphocytic leukemia (CLL) and othermajor tumor types and is believed to be responsible for maintaining the viabil-

114 Frankel

ity of cancerous cells and contributing to chemotherapy and radiotherapy resis-tance (38,49–51).

3.5.2.1. BCL-2 Expression and Dysregulation in CLL. B-cell CLL (B-CLL) represents a neoplastic disorder caused primarily by defective programmedcell death (PCD) as opposed to increased cell proliferation (52). Overexpressionof Bcl-2 in CLL allows for accumulation of CLL cells in GO (53,54). There is anaccumulation of long-lived neoplastic B cells expressing Bcl-2 protein. Defectsin the PCD pathway also contribute to chemoresistance (53).

Expression of a stable amount of Bcl-2 protein blocks apoptosis and pro-longs tumor survival. Bcl-2 is universally expressed in CLL cells (55).Increased Bcl-2 protein expression is observed in CLL cells (56). Relative lev-els of Bcl-2 oncoprotein represent one of the key determinants of the sensitiv-ity of lymphocytic cells to killing by essentially all drugs currently availablefor the treatment of cancer (57,58). Studies in CLL cells demonstrate a correla-tion between drug-induced apoptosis and the ratio of endogenous levels of Bcl-2 to Bax proteins (59,60). Similarly, the Bcl-2/Bax ratio has been correlatedwith disease status and is elevated in patients with progressive disease (61) orresistant disease (62). Higher Bcl-2 expression levels were found more often inpatients with progressive vs stable CLL (63). By immunoblot analysis, 80% ofpatients at stage C (p = 0.019) expressed high levels of Bcl2 (64).

The clinical relevance of Bcl-2 as a target for CLL therapy has been vali-dated by the finding that high levels of expression of Bcl-2 is an adverse fea-ture for survival in previously untreated patients with CLL (65). Severallymphokines and growth factors have also been reported to upregulate Bcl-2expression in CLL, suggesting that this increase in Bcl-2 expression may playa role in the delay of fludarabine-induced apoptosis (66). In a multivariateanalysis based on measurement of levels of expression of apoptotic controlproteins, Bcl-2 levels emerge as the most important protein in predicting sur-vival (67).

Reduction of Bcl-2 expression by antisense therapy sensitized cells tochemotherapy-induced apoptosis (60,68,71) supporting study of this approachin vivo.

3.5.2.2. BCL-2 Expression and Dysregulation in Acute Leukemia. Inrecent clinical studies, abnormal Bcl-2 expression was proven to be predictiveof poor response to treatment and adverse clinical outcome in patients withAML, although the data for acute lymphoblastic leukemia are not as robust(69–81). Bcl-2 levels may be higher in relapsed AML than at initial diagnosis(82). Most investigators have shown that the Bcl-2 expression is higher in theCD34+ population of AML blasts (83,84). However, further subset analysishas attributed this to the CD13+CD33+ subsets of CD34 cells and did not showa correlation between bcl-2 expression and prognosis that was independent of


cytogenetic group (85). As more complicated analyses have been performed,the significance of bcl-2 levels has been questioned (86). Possibly, the Bcl-2/Bax ratio might be more important, or in some studies, the bax level alone(87–91). Small samples of pediatric AML cases have not shown a correlationbetween Bcl-2 expression and outcome (92). The persistence of bcl-2 bearingblasts after chemotherapy treatment suggests an association of Bcl-2 expres-sion with multidrug resistance (93). Interactions with wt1 have also beenreported (79). The expression of Bcl-2 protein by flow cytometry correlatedwith that of CD34 and P170 and with the percentage of blast cells in patientswith MDS. Two-color analyses demonstrated that CD34 and Bcl-2 were usu-ally expressed in the same cells. No significant correlation was found withcytogenetic abnormalities. Expression of anti-apoptotic proteins was associ-ated with decreased survival. Consequently, Bcl-2 proteins expression waswell correlated with the International Prognostic Scoring System (IPSS) (94).

3.5.3. G3139 THERAPY IN CANCER

G3139 (Genasense®, oblimersen sodium, Genta Inc, Berkeley Heights, NJ) isan 18-mer PS antisense ODN directed against the first six codons of the openreading frame of Bcl-2. G3139 studies in xenograft models have shown markedenhancement of the efficacy of standard cytotoxic chemotherapy and rituximabin several cancers, including non-Hodgkin’s lymphoma, melanoma, breast can-cer, gastric cancer, and nonsmall-cell lung cancer (95). G3139 clinical activityhas been reported in lymphoma, melanoma, and prostate cancer (96–98).

In animals, after iv or subcutaneous (sc) injection, G3139 distributes rapidlyto highly perfused organs, especially lung and bone marrow. Oligos are gener-ally excreted unchanged, predominantly by the kidney (99). G3139 biodistrib-ution studies in vivo have demonstrated high tissue/plasma ratios, particularlyin the kidney and liver, but also significant distribution to the bone marrow andspleen (99). In addition, in vitro and in vivo studies showed both biologic andantitumor activity with submicromolar concentrations (e.g., ~ 170 mM).

The first human study of G3139 was performed in patients with non-Hodgkin’s lympohoma. Twenty-one patients received G3139 administered bycontinuous sc infusion (96,100). Thrombocytopenia, infusion site reactions,and fatigue were dose limiting in 2 patients treated at a level of 5.3 mg/kg/day.However, the tolerance to treatment in this study may have been closely linkedto the prolonged (2-wk) infusion schedule given by the sc route, and otherstudies have easily escalated the Genasense doses to 7 mg/kg/d even whengiven intravenously in combination with cytotoxic chemotherapy (101).Although the administered drug dose was quite low in most patients, i.e., sub-stantially below doses now known to be both safe and optimally effective withrespect to Bcl-2 downregulation, several major responses were observed.

116 Frankel

3.5.1.1. G3139 in CLL. Reduction of Bcl-2 expression by antisense therapysensitized cells to chemotherapy-induced apoptosis (60,68), supporting studyof this approach in vivo. In vitro, CLL cells are triggered to undergo apoptosismore vigorously when treated with Bcl-2 antisense than they are by fludara-bine, corticosteroids, or rituximab (102). In addition, fresh CLL cells obtainedfrom patients and studied ex-vivo have shown downregulation of Bcl-2 proteinafter antisense exposure and increased killing by fludarabine (103). Theenhanced antitumor activity of fludarabine in fresh CLL cells is consistent withthe major enhancement of cytotoxicity by cytosine arabinoside and otherantimetabolites (68). Treatment with antisense oligos with the same sequenceas oblimersen sodium also decreased Bcl-2 levels in CLL cell lines and in cellstaken from patients. In this study, antisense oligos directed against Bcl-2 pro-tein also significantly increased killing of CLL cells when combined with flu-darabine and enhanced the immunologic attack from lymphokine-activatedkiller (LAK) cells (103a).

A phase I trial of G3139 in patients with refractory or relapsed CLL estab-lished that the maximum tolerated dose (MTD) for cycle 1 in CLL asmonotherapy is 3 mg/kg/d, although patients could be safely escalated to 4mg/kg/d in subsequent cycles. Tumor lysis and transient clinical benefit,including decreased circulating CLL cells, softening and shrinkage of lymphnodes, and splenomegaly reduction, were reported in some patients (104).Patients treated with either G3139 at 7 or 5 mg/kg/d experienced high fever,hypotension, and hypoglycemia. Back pain requiring narcotics for pain controlwas also observed. This suggests that patients with lymphoma or CLL aremore sensitive to the treatment side effects of oblimersen sodium comparedwith patients with solid tumors, most likely due to cytokine release from thetarget tumor cells (108). A 3-mg/kg/d dose appears to be well tolerated eitheras a single agent or combined with fludarabine and cyclophosphamide. Aphase II study of this dose level is ongoing in CLL patients. At least twopatients have had partial responses to G3139 monotherapy (104a).

A randomized phase III trial of fludarabine/cyclophosphamide plus G3139is underway in patients with refractory or relapsed CLL who have previouslyreceived fludarabine therapy. Patients receive fludarabine 25 mg/m2/d withcyclophosphamide 250 mg/m2 for 3 d. Patients randomized to receive G3139begin a 7-d infusion of 3 mg/kg/d for 4 d before administration of the cytotoxicagents. Cotrimoxazole, allopurinol, acyclovir, and filgrastim are used prophy-lactically. This trial tests the hypothesis that downregulation of Bcl-2 byG3139 will sensitize CLL cells to undergo apoptosis when treated with cyto-toxic chemotherapy.

3.5.1.2. G3139 in AML. Downregulation of Bcl-2 may lower the apoptoticthreshold and restore chemosensitivity in chemoresistant leukemic cells (105).


In vitro experiments had suggested that antisense to Bcl-2 would be effectivein the treatment of AML (106–110). Preclinical analysis using liposomal Bcl-2antisense oligos in myeloid leukemia HL-60 cells, HL-60-doxorubicin-resis-tant cells, and CD34+ blast cells from patients with primary AML have estab-lished Bcl-2 as a critical target for antisense strategies in AML (110).

The combination of G3139 and gemtuzumab ozogamicin has been exploredin HL60 AML cells. HL-60 cells were cultured for 72 h with Mylotarg™(Wyeth Laboratories, Philadelphia, PA) (1.25–10 ng/mL), with 20 µM G3139using the streptolysin O transfection method, or with both. Mylotarg decreasedthe cell number in a dose-dependent manner (40%, 60%, and 71% growth inhi-bition at 1.25, 5, and 10 ng/mL). This growth-inhibitory effect was due to a pro-nounced G2M cell-cycle block arrest (more than 95% at 5 and 10 ng/mL) andmodest induction of apoptosis as determined by Annexin V positivity and DNAfragmentation assay (10.2% and 15.9%, respectively, at 10 ng/mL). G3139alone did not decrease cell numbers at 72 hrs under these conditions. Bcl-2 pro-tein was downregulated by 50% at 72 h of G3139 treatment. However, in combi-nation with Mylotarg, Genasense-enhanced cell death (Annexin V, 20.3% ([+]);sub-G1, 45.4 ([+]) cells) resulting in further decrease in cell number (69%, 75%,and 84% growth inhibition at 2.5, 5, and 10 ng/mL, respectively). The G2M cell-cycle block was not affected by Bcl-2 AS. The number of cells with DNA frag-mentation (sub G1) increased two- to threefold. These data demonstrated thatG3139 downregulates Bcl-2 in AML cells and enhances Mylotarg-induced celldeath by lowering the apoptotic threshold (111).

At Ohio State University, a phase I/II study was initiated under sponsorship ofthe National Cancer Institute to evaluate a constant intravenous infusion dose ofG3139 with escalating doses of fludarabine, Ara-C, and G-CSF (FLAG) forrefractory or relapsed acute leukemia (112). In this dose-escalation study, G3139(4 or 7 mg/kg/d in successive cohorts) is administered by continuous iv infusionon days 1–10, with fludarabine (starting at 15 mg/m2) and Ara-C (starting at 1000mg/m2) given daily on days 6–10 and escalated in successive cohorts.

Twenty patients were enrolled on this study (13 women and 7 men). Themedian age was 56 yr. Seventeen patients had AML, 5 with primary refractorydisease, 8 in first relapse, and 4 in subsequent relapses. Three patients hadALL, 2 with refractory Philadelphia chromosome-positive disease, and onewith t(5;14)(q31;q32) relapsed disease and hypereosinophilia. Of the 20patients, 9 received high-dose cytarabine (HiDAC) with previous treatments, 1had autologous stem cell transplantation, and 1 matched unrelated donor(MUD) stem cell transplantation. The median time to relapse from the initialtreatment for relapsed patients was 7 mo (range 3 to 21 mo). The median num-ber of previous treatments was two.

Of the 20 patients, 9 (45%) had disease response, 6 (5 with AML, 1 withALL) with complete remission and 3 (2 with AML and 1 with ALL) with no

118 Frankel

evidence of disease but failure to recover normal neutrophil and/or plateletcounts, or to remain in remission for 30 d or more (incomplete remission). Twoof the 20 patients were removed from protocol therapy due to rapidly risingblast counts before receiving chemotherapy.

Median time for neutrophil recovery from start of chemotherapy (i.e., day 6)was 23 d (range 8 to 38 d); median time for platelet recovery (≥ 50,000) was 39d (range 21 to 56 d).

The pharmacokinetic behavior of G3139 was similar to previous reports insolid tumor patients. The mean Css level for the 4-mg/kg dose was 3.19 ± 1.29µg/mL (range 1.59–5.69 µg/mL), which is significantly lower than the Css of5.47 ± 2.16 µg/mL (range 2.67–8.38 µg/mL) for the 7-mg/kg dose (p = 0.023).When normalized to dose, the Css values were 0.78 ± 0.33 and 0.78 ± 0.30µg/mL for the 4- and 7-mg/kg doses, respectively. These results indicated thatthe Css levels were proportional, and the pharmacokinetics linear for the twodoses (i.e., 4 and 7 mg/kg) administered. The linearity in pharmacokineticswas also reflected by dose-dependent differences in area under the curve(AUC) values (p < 0.05) and similar total clearance (4.35 ± 1.85 L/h at 4 mg/kgand 3.89 ± 1.48 L/h at 7 mg/kg, p > 0.5).

Bcl-2 mRNA levels were downregulated in 75% of evaluable patients. Nounanticipated adverse events were noted. The safety data, coupled with the ini-tial 50% response rate—including patients with refractory acute leukemia andprevious high dose cytrabine (HDAC)—support further development of G3139in combination regimens for refractory or relapsed leukemia.

These results prompted initiation of an ongoing phase II trial of G3139 plusgemtuzumab ozogamicin in elderly patients with relapsed AML. Objectivesare to determine the complete and overall response rates, duration of response,and safety in this population. G3139 has also been granted designation as anorphan drug from the Food and Drug Administration for the treatment ofpatients with AML. A pilot study of G3139 combined with daunorubicin andcytarabine in patients older than 60 yr of age with newly diagnosed AMLisunderway with plans to proceed to a phase III trial.

3.5.1.3. G3139 in CML. G3139 has been evaluated in xenograft model forPhiladelphia chromosome positive leukemia (113). Imatinib mesylate (STI-571, Gleevec™, Novartis Pharma AG, Basel) is an inhibitor of the Abelsonkinase constitutively activated in the BCR-ABL fusion gene product. In thisstudy, nude mice were transplanted with imatinib mesylate-resistant BCR-ABL-transformed TF-1 cells and then treated with placebo (n = 5) or G3139 7mg/kg/d ip for 14 d (n = 5). All of the untreated mice died within 10 wk, whilethe majority of mice treated with 3139 survived longer than 6 mo and demon-strated reduced tumor volume), with 3 of 5 G3139-treated mice demonstratingcomplete tumor regression. In addition, cells harvested from mice treated with


G3139 (7 mg/kg/d for 7 d) were more sensitive to further treatment with ima-tinib mesylate, daunorubicin, cytarabine, or etoposide, with each combinationshowing additive or synergistic activity in the induction of apoptosis. WhenBcl-2 was analyzed in the transformed cells, there was overexpression noted inthe mitochondrial fraction. This is consistent with previous reports that BCR-ABL may induce Bcl-2 expression (114).

These findings support the ongoing evaluation of G3139, both alone and incombination with other chemotherapeutic agents in the treatment of myeloidleukemias. The CALGB has recently initiated a trial of G3139 in patients withchronic phase CML treated with imatinib mesylate who have not had a com-plete hematologic response or major cytogenetic response.

4. CONCLUSIONS

Antisense oligonucleotides have been used for more than a decade to down-regulate gene expression. PS structures, such as G3139, have moved forwardinto clinical testing. These molecules successfully inhibit gene expression.Their clinical success will depend heavily on the appropriateness of the target.Preclinical activity and early clinical results suggest that antisense therapyusing Bcl-2 as a target is highly promising with relatively minimal toxicity.Downregulation of Bcl-2 may greatly enhance the activity of other types ofstandard therapy: chemotherapy, monoclonal antibodies, or radiation therapy.Other targets will require further study as structure-activity relationships anddrug delivery and stability are perfected.

REFERENCES1. Gewirtz AM, Sokol DL, Ratajczak MZ. Nucleic acid therapeutics: state of the art and future

prospects. Blood 1998;92(3):712–736.2. Cook PD. Medicinal chemistry of antisense oligonucleotides. In: Crooke ST, ed. Antisense

Drug Technology: Principles, Strategies, and Applications. New York: Marcel Dekker, Inc.,2001:29–56.

3. Crooke ST. Molecular mechanisms of action of antisense drugs. Biochim Biophys Acta1999;1489(1):31–44.

4. Wang H, Prasad G, Buolamwini JK, Zhang R. Antisense anticancer oligonucleotide thera-peutics. Curr Cancer Drug Targets 2001;1(3):177–196.

5. de Fabritiis P, Skorski T, De Propris MS, et al. Effect of bcr-abl oligodeoxynucleotides on theclonogenic growth of chronic myelogenous leukaemia cells. Leukemia 1997;11(6):811–819.

6. Skorski T, Perrotti D, Nieborowska-Skorska M, Gryaznov S, Calabretta B. Antileukemiaeffect of c-myc N3′→P5′ phosphoramidate antisense oligonucleotides in vivo. Proc NatlAcad Sci USA 1997;94(8):3966–3971.

7. Skorski T, Nieborowska-Skorska M, Wlodarski P, Zon G, Iozzo RV, Calabretta B. Anti-sense oligodeoxynucleotide combination therapy of primary chronic myelogenousleukemia blast crisis in SCID mice. Blood 1996;88(3):1005–1012.

8. Clark RE. Antisense therapeutics in chronic myeloid leukaemia: the promise, the progressand the problems. Leukemia 2000;14(3):347–355.

120 Frankel

9. Sun J, Wu B, Liu Q, et al. [Autologous bone marrow transplantation for patients withchronic myelogenous leukemia after in vitro purging of the graft with bcr/abl antisenseoligodeoxynucleotides]. Zhonghua Xue Ye Xue Za Zhi 2000;21(8):400–402.

10. Clark RE, Grzybowski J, Broughton CM, et al. Clinical use of streptolysin-O to facilitateantisense oligodeoxyribonucleotide delivery for purging autografts in chronic myeloidleukaemia. Bone Marrow Transplant 1999;23(12):1303–1308.

11. de Fabritiis P, Petti MC, Montefusco E, et al. BCR-ABL antisense oligodeoxynucleotide invitro purging and autologous bone marrow transplantation for patients with chronic myel-ogenous leukemia in advanced phase. Blood 1998;91(9):3156–3162.

12. Bellucci R, Sala R, De Propris MS, Cordone I, de Fabritiis P. Interferon-alpha and bcr-abl anti-sense oligodeoxynucleotides in combination enhance the antileukemic effect and the adher-ence of CML progenitors to preformed stroma. Leuk Lymphoma 1999;35(5–6):471–481.

13. Bhatia R, Verfaillie CM. Inhibition of BCR-ABL expression with antisense oligodeoxynu-cleotides restores beta 1 integrin-mediated adhesion and proliferation inhibition in chronicmyelogenous leukemia hematopoietic progenitors. Blood 1998;91(9):3414–3422.

14. Smetsers TF, Linders EH, van de Locht LT, de Witte TM, Mensink EJ. An antisense Bcr-Abl phosphodiester-tailed methylphosphonate oligonucleotide reduces the growth ofchronic myeloid leukaemia patient cells by a non-antisense mechanism. Br J Haematol1997;96(2):377–381.

15. Spiller DG, Giles RV, Broughton CM, et al. The influence of target protein half-life on theeffectiveness of antisense oligonucleotide analog-mediated biologic responses. AntisenseNucleic Acid Drug Dev 1998;8(4):281–293.

16. Oh IH, Reddy EP. The myb gene family in cell growth, differentiation and apoptosis. Onco-gene 1999;18(19):3017–3033.

17. Weston K. Reassessing the role of C-MYB in tumorigenesis. Oncogene 1999;18(19):3034–3038.

18. Ness SA. Myb binding proteins: regulators and cohorts in transformation. Oncogene1999;18(19):3039–3046.

19. Gewirtz AM. Myb targeted therapeutics for the treatment of human malignancies. Onco-gene 1999;18(19):3056–3062.

20. Ratajczak MZ, Kant JA, Luger SM, et al. In vivo treatment of human leukemia in a scidmouse model with c-myb antisense oligodeoxynucleotides. Proc Natl Acad Sci U S A1992;89(24):11823–11827.

21. Hijiya N, Zhang J, Ratajczak MZ, Kant JA, DeRiel K, Herlyn M, et al. Biologic and thera-peutic significance of MYB expression in human melanoma. Proc Natl Acad Sci USA1994;91(10):4499–4503.

22. Gewirtz AM. Nucleic acid therapeutics for the treatment of human leukemia. In: CrookeST, ed. Antisense Drug Technology: Principles, Strategies, and Applications. New York:Marcel Dekker; 2001:761–771.

23. Luger SM, O’Brien SG, Ratajczak J, et al. Oligodeoxynucleotide-mediated inhibition of c-mybgene expression in autografted bone marrow: a pilot study. Blood 2002;99(4):1150–1158.

24. Skorski T, Nieborowska-Skorska M, Wlodarski P, Zon G, Iozzo RV, Calabretta B. Anti-sense oligodeoxynucleotide combination therapy of primary chronic myelogenousleukemia blast crisis in SCID mice. Blood 1996;88(3):1005–1012.

25. Kimura S, Maekawa T, Hirakawa K, Murakami A, Abe T. Alterations of c-myc expressionby antisense oligodeoxynucleotides enhance the induction of apoptosis in HL-60 cells.Cancer Res 1995;55(6):1379–1384.

26. Kondoh M, Matsuyama T, Suzuki R, et al. Growth inhibition of human leukemia HL-60cells by an antisense phosphodiester oligonucleotide encapsulated into fusogenic lipo-somes. Biol Pharm Bull 2000;23(8):1011–1013.


27. McGuffie EM, Pacheco D, Carbone GM, Catapano CV. Antigene and antiproliferativeeffects of a c-myc-targeting phosphorothioate triple helix-forming oligonucleotide inhuman leukemia cells. Cancer Res 2000;60(14):3790–3799.

28. Bishop MR, Iversen PL, Bayever E, et al. Phase I trial of an antisense oligonucleotideOL(1)p53 in hematologic malignancies. J Clin Oncol 1996;14(4):1320–1326.

29. Bishop MR, Jackson JD, Tarantolo SR, et al. Ex vivo treatment of bone marrow with phos-phorothioate oligonucleotide OL(1)p53 for autologous transplantation in acute myeloge-nous leukemia and myelodysplastic syndrome. J Hematother 1997;6(5):441–446.

30. Oltvai ZN, Korsmeyer SJ. Checkpoints of dueling dimers foil death wishes [comment].Cell 1994;79(2):189–192.

31. McDonnell TJ, Beham A, Sarkiss M, Andersen MM, Lo P. Importance of the Bcl-2 familyin cell death regulation. Experientia 1996;52(10–11):1008–1017.

32. Boise LH, Gonzalez-Garcia M, Postema CE, et al. bcl-x, a bcl-2-related gene that functionsas a dominant regulator of apoptotic cell death. Cell 1993;74(4):597–608.

33. Adams JM, Cory S. The Bcl-2 protein family: arbiters of cell survival. Science1998;281(5381):1322–1326.

34. Boyd JM, Gallo GJ, Elangovan B, et al. Bik, a novel death-inducing protein shares a dis-tinct sequence motif with Bcl-2 family proteins and interacts with viral and cellular sur-vival-promoting proteins. Oncogene 1995;11(9):1921–1928.

35. Chittenden T, Harrington EA, O’Connor R, et al. Induction of apoptosis by the Bcl-2homologue Bak. Nature 1995;374(6524):733–736.

36. Choi SS, Park IC, Yun JW, Sung YC, Hong SI, Shin HS. A novel Bcl-2 related gene, Bfl-1,is overexpressed in stomach cancer and preferentially expressed in bone marrow. Oncogene1995;11(9):1693–1698.

37. Gross A, McDonnell JM, Korsmeyer SJ. BCL-2 family members and the mitochondria inapoptosis. Genet Dev 1999;13(15):1899–1911.

38. McDonnell TJ, Korsmeyer SJ. Progression from lymphoid hyperplasia to high-grademalignant lymphoma in mice transgenic for the t(14;18). Nature 1991;349(6306):254–256.

39. Zamzami N, Susin SA, Marchetti P, et al. Mitochondrial control of nuclear apoptosis [com-ment]. J Exp Med 1996;183(4):1533–1544.

40. Rosse T, Olivier R, Monney L, et al. Bcl-2 prolongs cell survival after Bax-induced releaseof cytochrome c [comment]. Nature 1998;391(6666):496–499.

41. Krajewski S, Tanaka S, Takayama S, Schibler MJ, Fenton W, Reed JC. Investigation of thesubcellular distribution of the bcl-2 oncoprotein: residence in the nuclear envelope, endoplas-mic reticulum, and outer mitochondrial membranes. Cancer Res 1993;53(19):4701–4714.

42. Perry DK, Smyth MJ, Wang H-G, et al. Bcl-2 acts upstream of the PARP protease and pre-vents its activation [abstract]. Cell Death Differ 1997;4:29–33.

43. Chen LB. Mitochondrial membrane potential in living cells. Ann Rev Cell Biol 1988;4:155–181.

44. Beham A, Marin MC, Fernandez A, et al. Bcl-2 inhibits p53 nuclear import following DNAdamage. Oncogene 1997;15(23):2767–2772.

45. Herr I, Debatin KM. Cellular stress response and apoptosis in cancer therapy. Blood2001;98(9):2603–2614.

46. Reed JC. Dysregulation of apoptosis in cancer. J Clin Oncol 1999;17(9):2941–2953.47. Johnstone RW, Ruefli AA, Lowe SW. Apoptosis: a link between cancer genetics and

chemotherapy. Cell 2002;108(2):153–164.48. Ben-Ezra JM, Kornstein MJ, Grimes MM, Krystal G. Small cell carcinomas of the lung

express the Bcl-2 protein. Am J Pathol 1994;145(5):1036–1040.49. Tsujimoto Y, Cossman J, Jaffe E, Croce CM. Involvement of the bcl-2 gene in human fol-

licular lymphoma. Science 1985;228(4706):1440–1443.50. Miyashita T, Reed JC. Bcl-2 oncoprotein blocks chemotherapy-induced apoptosis in a

human leukemia cell line. Blood 1993;81(1):151–157.

122 Frankel

51. Reed JC. Bcl-2 and the regulation of programmed cell death. J Cell Biol 1994;124(1–2):1–6.52. Bentley DP, Pepper CJ. The apoptotic pathway: a target for therapy in chronic lymphocytic

leukemia. Hematol Oncol 2000;18(3):87–98.53. Caligaris-Cappio F, Gottardi D, Alfarano A, et al. The nature of the B lymphocyte in B-

chronic lymphocytic leukemia. Blood Cells 1993;19(3):601–613.54. Reed JC. Molecular biology of chronic lymphocytic leukemia. Semin Oncol 1998;25(1):

11–18.55. Consoli U, Santonoeito A, Stagno F, et al. Multidrug resistance mechanisms in chronic

lymphocytic leukaemia. Br J Haematol 2002;116(4):774–780.56. Klein A, Miera O, Bauer O, Golfier S, Schriever F. Chemosensitivity of B cell chronic lym-

phocytic leukemia and correlated expression of proteins regulating apoptosis, cell cycle andDNA repair. Leukemia 2000;14(1):40–46.

57. Gottardi D, Alfarano A, De Leo AM, et al. In leukaemic CD5+ B cells the expression ofBCL-2 gene family is shifted toward protection from apoptosis. Br J Haematol1996;94(4):612–618.

58. Reed JC, Kitada S, Takayama S, Miyashita T. Regulation of chemoresistance by the bcl-2oncoprotein in non-Hodgkin’s lymphoma and lymphocytic leukemia cell lines. Ann Oncol1994;5 Suppl 1:61:95–5:61–65.

59. Thomas A, El Rouby S, Reed JC, et al. Drug-induced apoptosis in B-cell chronic lympho-cytic leukemia: relationship between p53 gene mutation and bcl-2/bax proteins in drugresistance. Oncogene 1996;12(5):1055–1062.

60. Pepper C, Thomas A, Hoy T, Cotter F, Bentley P. Antisense-mediated suppression of Bcl-2highlights its pivotal role in failed apoptosis in B-cell chronic lymphocytic leukaemia. Br JHaematol 1999;107(3):611–615.

61. Molica S, Dattilo A, Giulino C, Levato D, Levato L. Increased bcl-2/bax ratio in B-cellchronic lymphocytic leukemia is associated with a progressive pattern of disease [letter].Haematologica 1998;83(12):1122–1124.

62. McConkey DJ, Chandra J, Wright S, et al. Apoptosis sensitivity in chronic lymphocyticleukemia is determined by endogenous endonuclease content and relative expression ofBCL-2 and BAX. J Immunol 1996;156(7):2624–2630.

63. Marschitz I, Tinhofer I, Hittmair A, Egle A, Kos M, Greil R. Analysis of Bcl-2 proteinexpression in chronic lymphocytic leukemia. A comparison of three semiquantitation tech-niques. Am J Clin Pathol 2000;113(2):219–229.

64. Aviram A, Rabizadeh E, Zimra Y, et al. Expression of bcl-2 and bax in cells isolated fromB-chronic lymphocytic leukemia patients at different stages of the disease. Eur J Haematol2000;64(2):80–84.

65. Robertson LE, Plunkett W, McConnell K, Keating MJ, McDonnell TJ. Bcl-2 expression inchronic lymphocytic leukemia and its correlation with the induction of apoptosis and clini-cal outcome. Leukemia 1996;10(3):456–459.

66. Konig A, Menzel T, Lynen S, et al. Basic fibroblast growth factor (bFGF) upregulates theexpression of bcl-2 in B cell chronic lymphocytic leukemia cell lines resulting in delayingapoptosis. Leukemia 1997;11(2):258–265.

67. Faderl S, Keating MJ, Do KA, et al. Expression profile of 11 proteins and their prognosticsignificance in patients with chronic lymphocytic leukemia (CLL). Leukemia 2002;16(6):1045–1052.

68. Kitada S, Takayama S, De Riel K, Tanaka S, Reed JC. Reversal of chemoresistance of lym-phoma cells by antisense-mediated reduction of bcl-2 gene expression. Antisense Res Dev1994;4(2):71–79.

69. Bincoletto C, Saad ST, da Silva ES, Queiroz ML. Haematopoietic response and bcl-2expression in patients with acute myeloid leukaemia. Eur J Haematol 1999;62(1):38–42.

70. Porwit-MacDonald A, Ivory K, Wilkinson S, Wheatley K, Wong L, Janossy G. Bcl-2 pro-tein expression in normal human bone marrow precursors and in acute myelogenousleukemia. Leukemia 1995;9(7):1191–1198.


71. Bradbury DA, Zhu YM, Russell NH. Bcl-2 expression in acute myeloblastic leukaemia:relationship with autonomous growth and CD34 antigen expression. Leuk Lymphoma1997;24(3–4):221–228.

72. Tothova E, Fricova M, Stecova N, Kafkova A, Elbertova A. High expression of Bcl-2 pro-tein in acute myeloid leukemia cells is associated with poor response to chemotherapy.Neoplasma 2002;49(3):141–144.

73. Yue B, Chen Y, Yu D, Xiang Z. Study on the relationship between the Bcl-2/Bax ratio andthe growth types of leukemic cells and drug resistance in acute myelogenous leukemia. JTongji Med Univ 1998;18(2):101–104.

74. Maung ZT, MacLean FR, Reid MM, et al. The relationship between bcl-2 expression andresponse to chemotherapy in acute leukaemia. Br J Haematol 1994;88(1):105–109.

75. Wuchter C, Karawajew L, Ruppert V, et al. Clinical significance of CD95, Bcl-2 and Baxexpression and CD95 function in adult de novo acute myeloid leukemia in context of P-gly-coprotein function, maturation stage, and cytogenetics. Leukemia 1999;13(12):1943–1953.

76. Wuchter C, Karawajew L, Ruppert V, et al. Constitutive expression levels of CD95 and Bcl-2 as well as CD95 function and spontaneous apoptosis in vitro do not predict the responseto induction chemotherapy and relapse rate in childhood acute lymphoblastic leukaemia. BrJ Haematol 2000;110(1):154–160.

77. Campos L, Rouault JP, Sabido O, et al. High expression of bcl-2 protein in acute myeloidleukemia cells is associated with poor response to chemotherapy. Blood 1993;81(11):3091–3096.

78. Karakas T, Maurer U, Weidmann E, Miething CC, Hoelzer D, Bergmann L. High expres-sion of bcl-2 mRNA as a determinant of poor prognosis in acute myeloid leukemia. AnnOncol 1998;9(2):159–165.

79. Karakas T, Miething CC, Maurer U, et al. The coexpression of the apoptosis-related genesbcl-2 and wt1 in predicting survival in adult acute myeloid leukemia. Leukemia2002;16(5):846–854.

80. Campos L, Sabido O, Viallet A, Vasselon C, Guyotat D. Expression of apoptosis-control-ling proteins in acute leukemia cells. Leuk Lymphoma 1999;33(5–6):499–509.

81. Prokop A, Wieder T, Sturm I, et al. Relapse in childhood acute lymphoblastic leukemia isassociated with a decrease of the Bax/Bcl-2 ratio and loss of spontaneous caspase-3 pro-cessing in vivo. Leukemia 2000;14(9):1606–1613.

82. Bensi L, Longo R, Vecchi A, et al. Bcl-2 oncoprotein expression in acute myeloidleukemia. Haematologica 1995;80(2):98–102.

83. Bradbury DA, Russell NH. Comparative quantitative expression of bcl-2 by normal andleukaemic myeloid cells. Br J Haematol 1995;91(2):374–379.

84. Suarez L, Vidriales B, Garcia-Larana J, et al. Multiparametric analysis of apoptotic andmulti-drug resistance phenotype in elderly patients with acute myeloid leukemia accordingto the blast cell maturation stage. Haematologica 2001;86(12):1287–1295.

85. Andreeff M, Jiang S, Zhang X, et al. Expression of Bcl-2-related genes in normal and AMLprogenitors: changes induced by chemotherapy and retinoic acid. Leukemia 1999;13(11):1881–1892.

86. Schaich M, Illmer T, Seitz G, et al. The prognostic value of Bcl-XL gene expression forremission induction is influenced by cytogenetics in adult acute myeloid leukemia. Haema-tologica 2001;86(5):470–477.

87. Ong YL, McMullin MF, Bailie KE, Lappin TR, Jones FG, Irvine AE. High bax expression is agood prognostic indicator in acute myeloid leukaemia. Br J Haematol 2000;111(1):182–189.

88. Kornblau SM, Thall PF, Estrov Z, et al. The prognostic impact of BCL2 protein expression inacute myelogenous leukemia varies with cytogenetics. Clin Cancer Res 1999;5(7):1758–1766.

89. Kornblau SM, Vu HT, Ruvolo P, et al. BAX and PKCalpha modulate the prognostic impact ofBCL2 expression in acute myelogenous leukemia. Clin Cancer Res 2000;6(4):1401–1409.

124 Frankel

90. Kohler T, Leiblein S, Borchert S, et al. Absolute levels of MDR-1, MRP, and BCL-2MRNA and tumor remission in acute leukemia. Adv Exp Med Biol 1999;457(10):177–185.

91. Kohler T, Schill C, Deininger MW, et al. High Bad and Bax mRNA expression correlatewith negative outcome in acute myeloid leukemia (AML). Leukemia 2002;16(1):22–29.

92. Naumovski L, Martinovsky G, Wong C, et al. BCL-2 expression does not not correlate withpatient outcome in pediatric acute myelogenous leukemia. Leuk Res 1998;22(1):81–87.

93. Ahmed N, Sammons J, Hassan HT. Bcl-2 protein in human myeloid leukaemia cells and itsdown-regulation during chemotherapy-induced apoptosis. Oncol Rep 1999;6(2):403–407.

94. Boudard D, Vasselon C, Bertheas MF, et al. Expression and prognostic significance of Bcl-2 family proteins in myelodysplastic syndromes. Am J Hematol 2002;70(2):115–125.

95. Auer R, Corbo M, Fegan C, Frankel S, Cotter F. Bcl-2 antisense (Genasense™) inducesapoptosis and potentiates activity of both cytotoxic chemotherapy and rituximab in primaryCLL cells [abstract 3358]. Blood 2001;98(11):808a.

96. Waters JS, Webb A, Cunningham D, et al. Phase I clinical and pharmacokinetic study ofbcl-2 antisense oligonucleotide therapy in patients with non-Hodgkin’s lymphoma [com-ment]. J Clin Oncol 2000;18(9):1812–1823.

97. de Bono J, Rowinsky E, Kuhn J, et al. Phase I pharmacokinetic (PK) and pharmacody-namic (PD) trial of bcl-2 antisense (Genasense) and Docetaxel (D) in hormone refractoryprostate cancer. Proc Am Soc Clin Oncol 2001;20; Abstract 474.

98. Jansen B, Wacheck V, Heere-Ress E, et al. Chemosensitisation of malignant melanoma byBCL2 antisense therapy. Lancet 2000;356(9243):1728–1733.

99. Raynaud FI, Orr RM, Goddard PM, et al. Pharmacokinetics of G3139, a phosphorothioateoligodeoxynucleotide antisense to bcl-2, after intravenous administration or continuoussubcutaneous infusion to mice. J Pharmacol Exp Ther 1997;281(1):420–427.

100. Webb A, Cunningham D, Cotter F, et al. BCL-2 antisense therapy in patients with non-Hodgkin lymphoma. Lancet 1997;349(9059):1137–1141.

101. Klasa RJ, Gillum AM, Klem RE, Frankel SR. Oblimersen Bcl-2 antisense: facilitatingapoptosis in anticancer treatment. Antisense Nucleic Acid Drug Dev 2002;12(3):193–213.

102. Czuczman MS, Fallon A, Mohr A, et al. Phase II Study of rituximab plus fludarabine inpatients (pts) with low-grade lymphoma (LGL): final report [abstract]. Blood 2001;98:601a.

103. Vu UE, Dickinson J, Wang P, et al. Differentially expressed genes in B-chronic lympho-cytic leukemia: evidence that manipulation of bcl-2 gene leads to increased killing of CLLcells [abstract]. Blood 2000;96:714a

103a. Vu UE, Pavletic ZS, Wang X, Joshi SS. Increased cytotoxicity against B-chronic lympho-cytic leukemia by cellular manipulations: potentials for therapeutic use. Leuk Lymphoma2000;39(5-6):573–582.

104. O’Brien S, Giles F, Kanti R, et al. Bcl-2 antisense (Genasense) as monotherapy for refrac-tory chronic lymphocytic leukemia. Blood 2001;98(11):772a.

104a. Rai KR, O’Brien S, Cunningham C, et al. Genasense (Bcl-2 Antisense) Montherapy inpatients with relapsed or refractory chronic lymphocytic leukemia: Phase 1 and 2 results.Blood 2002;100[11],384a.

105. Cotter FE. Antisense therapy of hematologic malignancies. Semin Hematol 1999;36(4Suppl 6):9–14.

106. Campos L, Sabido O, Rouault JP, Guyotat D. Effects of BCL-2 antisense oligodeoxynu-cleotides on in vitro proliferation and survival of normal marrow progenitors and leukemiccells. Blood 1994;84(2):595–600.

107. Lin Y, Lu L, Hu J. [Inhibition of Bcl-2 expression in HL60 cells by incubation with anti-sense phosphorothioate oligodeoxynucleotides]. Zhonghua Yi Xue Yi Chuan Xue Za Zhi1999;16(1):12–15.

108. Reed JC, Stein C, Subasinghe C, et al. Antisense-mediated inhibition of BCL2 protoonco-gene expression and leukemic cell growth and survival: comparisons of phosphodiester andphosphorothioate oligodeoxynucleotides. Cancer Res 1990;50(20):6565–6570.


109. Keith FJ, Bradbury DA, Zhu YM, Russell NH. Inhibition of bcl-2 with antisense oligonu-cleotides induces apoptosis and increases the sensitivity of AML blasts to Ara-C. Leukemia1995;9(1):131–138.

110. Konopleva M, Tari AM, Estrov Z, et al. Liposomal Bcl-2 antisense oligonucleotides enhanceproliferation, sensitize acute myeloid leukemia to cytosine-arabinoside, and induce apopto-sis independent of other antiapoptotic proteins. Blood 2000;95(12):3929–3938.

111. Khodadoust M, Konopleva M, Leysath C, Frankel S, Giles F, Andreeff M. Bcl-2 antisenseoligodeoxynucleotide (Genasense) enhances gemtuzumab ozogamicin (Mylotarg)-inducedcytotoxicity in acute myeloid leukemia. Blood 2001;98(11):102a [Abstract 429].

112. Marcucci G, Byrd JC, Dai G, et al. Phase I and pharmacodynamic studies of G3139, a bcl-2 antisense oligonucleotide, in combination with chemotherapy in refractory or relapsedacute leukemia. Blood 2002;2002–2006.

113. Tauchi T, Nakajima A, Sumi M, Shimamoto T, Sashida G, Ohyashiki K. G3139 (Bcl-2 anti-sense oligonucleotide) is active against Gleevec-resistant bcr-abl-positive cells. Proc AmAssoc Cancer Res 2002;43:abstract 4702.

114. Sanchez-Garcia I, Grutz G. Tumorigenic activity of the BCR-ABL oncogene is mediatedby BCL2. Proc Natl Acad Sci USA 1995;92(12):5287–5291.

126 Frankel

1. INTRODUCTION

The development of the first successful leukemia treatments owed much moreto empiric observation than rational drug design in an era when the biology ofleukemia was poorly understood. Although cytotoxic chemotherapeutic drugshave played and continue to play an essential role in cancer management, theirrelative lack of specificity, numerous toxicities, and frequency of resistance havelimited this approach. Recent efforts have focused on identifying the biologicbasis of leukemia, expecting that agents that more precisely target leukemia canbe developed to maximize responses while minimizing toxicity. This approachrequires the identification of appropriate targets, and the necessary tools toundertake this process have only recently become available.

Although any protein in a leukemic cell could be considered a target, thischapter focuses on molecular pathogenetic targets; that is, oncogene productsthat are clearly responsible for the molecular pathogenesis of the disease. Thisincludes proteins that are mutated or aberrantly expressed, with that eventbeing critical to the malignant process. Advances in the fields of cytogenetics,molecular genetics, and biochemistry during the past 50 yr have greatly

127


9 Signal Transduction Inhibitors

Michael E. O’Dwyer, MD

and Brian J. Druker, MD

CONTENTS

INTRODUCTION

IDENTIFYING THE TARGETS

VALIDATING THE TARGETS

HITTING THE TARGET

TRANSLATING THE SUCCESS OF STI571 TO OTHER MOLECULAR TARGETS

REFERENCES

advanced the understanding of leukemia pathogenesis and helped to identifycandidate targets. We now know that oncogenic activation can result from sev-eral different mechanisms, including chromosomal translocations, activatingmutations, loss of function mutations, and deletions. This chapter shows howthe application of this knowledge has aided in the development of rational mol-ecularly based treatment approaches, particularly the BCR-ABL tyrosinekinase inhibitor, STI571 (Gleevec™, imatinib mesylate, Novartis, Basel,Switzerland).

2. IDENTIFYING THE TARGETS

2.1. Chromosomal TranslocationsThe identification of the Philadelphia (Ph) chromosome as the first consis-

tent chromosomal abnormality in a human malignancy and the subsequent dis-covery that many types of leukemia and lymphoma are associated withrecurrent chromosomal translocations suggested that these chromosomalabnormalities had a role in the pathogenesis of these diseases (1). Numerousrecurrent chromosomal translocations have since been identified, and the mol-ecular consequences of some of these are now well characterized (2). Chromo-somal translocations result in cellular transformation by one of two principalmechanisms. In the first, juxtaposition to a transcriptionally active region ofthe genome leads to increased expression of the target gene. This is commonlyseen in lymphoid malignancies, in which target genes are translocated to theimmunoglobulin or T-cell receptor loci in the case of B- and T-cell malignan-cies, respectively. In follicular lymphomas, the t(14;18) leads to increasedexpression of the anti-apoptotic protein Bcl-2 as a result of the juxtaposition ofthe bcl-2 gene on chromosome 18 with the immunoglobulin heavy chain locuson chromosome 14 (3,4). A similar mechanism operates in both Burkitt’s lym-phoma as a consequence of t(8;14) and mantle cell lymphoma as a conse-quence of t(11;14), with overexpression of the c-myc and cyclin D1 genes,respectively (5–7). The second mechanism, which is more common inleukemia, results in the generation of chimeric oncogenes, which alter the phe-notype of the affected cell.

The production of chimeric oncogenes involving transcription factors can dis-rupt the normal function of these factors, leading to altered cellular differentia-tion. In acute promyelocytic leukemia (APL) associated with the t(15;17)translocation, the normal response of the retinoic acid receptor-α (RAR-α) to itsligand, retinoic acid (RA), is disrupted by the fusion protein PML-RAR-α,resulting in a blockade in myeloid differentiation. However, pharmacologicdoses of RA (1 M) are capable of relieving this repression, restoring normal cel-lular differentiation (8–11). RA therapy may also lead to degradation of PML-RAR-α by a proteolytic pathway, contributing to its ability to restore normal

128 O’Dwyer and Druker

differentiation (12). The therapeutic efficacy of RA was first demonstrated in astudy from China that showed that all-trans retinoic acid (ATRA) produced dif-ferentiation and remission in patients with APL with less toxicity than conven-tional chemotherapy (13). This was the first example of a small moleculesuccessfully targeting the underlying molecular lesion in a form of leukemia.However, this was an empiric discovery, because the molecular basis of APLhad not yet been elucidated. Further experience with ATRA in APL showed thatthe majority of patients relapsed if treated with ATRA alone. Subsequent clinicaltrials have now refined the treatment of this disease, and using a combination ofchemotherapy and ATRA, high rates of durable remission are seen.

Disruption of the transcription factor known as core binding factor (CBF)occurs in several different types of leukemia. CBF is a heterodimeric proteinconsisting of acute myeloid leukemia 1 (AML1; also known as CBFa2), thedeoxyribonucleic acid (DNA)-binding component, and CBFβ, which binds toAML1 and stabilizes its binding to DNA (14). Twelve percent of cases of acutemyeloid leukemia (AML) have a translocation between chromosomes 8 and 21,t(8;21) (15). This translocation generates a dominant negative AML1/ETOfusion protein, which interferes with the normal trans-activating function ofAML1 (16). The AML1 protein regulates the expression of genes crucial in nor-mal hematopoietic development, differentiation, and function, including thegenes for myeloperoxidase, neutrophil elastase, interleukin (IL) 3 and GM-CSF.The ETO fusion partner binds to the nuclear receptor corepressor complex (N-CoR), which then recruits Sin3 and histone deacetylase. This complex inhibitsthe expression of normal AML1-responsive genes, disrupting hematopoiesis(17). A related form of AML associated with inv(16) also interferes with normalAML1-dependent transcriptional regulation. In this case, the fusion proteinCBFβ-MYH11 functions as a dominant repressor, again recruiting a corepressorwith histone deacetylase activity (18). CBFβ-MYH11 transcripts have beendetected in up to 10% of patients with newly diagnosed AML (19). Finally, themost frequent chromosomal translocation seen in childhood acute lymphoblasticleukemia (ALL), t(12;21), accounting for approx 25% of cases of commonALL, results in the generation of a TEL-AML1 chimeric protein. TEL-AML1contributes to leukemogenesis through the recruitment of a nuclear corepressorcomplex with histone deacetylase activity (20).

Clearly, the AML1 transcription factor would be an ideal target for anATRA-like agent that restored normal transcriptional activity to this protein. Inthe absence of such a specific agent, a shared theme between t(8;21), inv(16),and t(12;21), as well as t(15;17) in APL, is that of transcriptional repressionrelated to excessive histone deacetylase activity. Thus, histone deacetylaseinhibitors might be expected to be a general class of agents that could find usein these leukemias. Clinical trials of such agents, including sodium phenylbu-tyrate and trichostatin A, among others, are in progress.

Chapter 9 / Signal Transduction Inhibitors 129

Another common consequence of translocation events is activation of atyrosine kinase. In translocations involving genes encoding tyrosine kinases,the partner gene usually encodes a dimerization motif, which leads to constitu-tive activation of the tyrosine kinase. Chronic myelogenous leukemia (CML) isthe classic example. The Ph chromosome, a shortened chromosome 22, wasthe first consistent chromosomal abnormality identified in a human malignancy(1). With the development of improved chromosomal-banding techniques inthe early 1970s, it became apparent that the Ph chromosome was the result of areciprocal translocation between the long arms of chromosomes 9 and 22,t(9;22)(q34;q11) (21). The molecular consequences of this translocation weresubsequently shown to be the juxtaposition of the c-Abl oncogene from chro-mosome 9 with sequences from chromosome 22, the breakpoint cluster region(BCR), giving rise to a fusion BCR-ABL gene (22). The size of the proteingenerated by the fusion gene varies depending on where the breakpoint occursin the BCR region. A 210kDa fusion protein (p210) is seen in approx 95% ofpatients with CML and up to 20% of adult patients with ALL. A 185kDa fusionprotein (p185) is seen in 10% of adults with ALL and is the predominant BCR-ABL fusion protein in Ph-positive pediatric patients with ALL, although only5% of pediatric patients with ALL are Ph positive. The product of this fusiongene is a constitutively active tyrosine kinase with markedly enhanced enzy-matic activity compared with the ABL kinase. This enhanced tyrosine kinaseactivity is critical for its transforming activity (23). Other examples of translo-cations involving genes encoding tyrosine kinases include t(5;12), producing aTEL-PDGFR fusion tyrosine kinase associated with some cases of chronicmyelomonocytic leukemia (CMML), and t(2;5), producing a NPM-ALKfusion tyrosine kinase associated with T-cell anaplastic large-cell lymphoma(ALCL) (24,25).

2.2. Activating MutationsActivating mutations in cytokine receptors, specifically receptor tyrosine

kinases, is being increasingly recognized as a means of cellular transformationin hematopoietic malignancies. Activating mutations of two members of theclass III receptor tyrosine kinase family, c-kit and Flt3, have been documentedin leukemic cells from patients with AML. An internal tandem duplication(ITD) of the juxtamembrane-coding region of the Flt3 gene on chromosome 13has been reported in up to 30% of adult patients with AML and up to 16.5% ofpediatric patients with AML (26–29). More recently, another 7% of patientswith AML and 3% of patients with myelodysplasia had a point mutation in theactivation loop of the Flt3 kinase domain (30). Consequently, mutant Flt3receptors dimerize in the absence of ligand with constitutive tyrosine kinaseactivation, resulting in growth factor independence. Clinically, the presence ofFlt3-ITD is associated with poor prognosis, with affected patients having lower


rates of complete remission (CR) after induction chemotherapy and higherrelapse rates. Activating mutations in c-kit are less common and occur predom-inantly in patients with CBF-related AML (31,32). These could be important“second-hits” in the development of these subtypes of leukemia (see followingsection) and are the targets of ongoing clinical trials with specific kinaseinhibitors.

3. VALIDATING THE TARGETS

Any cell protein that is involved in a growth or survival pathway could beconsidered a target for a therapeutic agent. Agents that target these proteinsand pathways may be of use and would be expected to result in incrementaladvances in the treatment of cancer. However, to recapitulate the quantumimprovements seen with agents such as ATRA and STI571, the target shouldbe an oncogene that has clearly been shown to be responsible for malignanttransformation. Rendering cell lines growth factor independent after retroviralexpression of an oncogene is supportive evidence of a role for an oncogene inmalignant transformation. However, the strongest evidence comes from animalmodels in which knock-in (translocations or activating mutations) or knockout(deletions or loss of function mutations) strategies can recapitulate the humandisease. In the case of CML, experiments in transgenic mice and murine recip-ients of BCR-ABL transduced hematopoietic stem cells demonstrate thatexpression of BCR-ABL alone can induce leukemia (33,34). Similarly, retrovi-ral insertion of the NPM-ALK gene into murine hematopoietic cells followedby transplantation into lethally irradiated recipients causes lymphoid malig-nancy in mice (35). Finally, transfection of the murine IL-3-dependent cellline, 32D, with a mutant Flt3 gene, results in growth factor independence, andtransplantation of these cells into syngeneic mice leads to the rapid develop-ment of leukemia (36). Whether hematopoietic progenitors expressing a simi-lar Flt3 mutant would rapidly develop leukemia must be determined.

The situation may be more complex for other oncogenes. Transgenic miceexpressing PML-RAR-α develop APL, although with relatively long latency(37). However, by adding the reciprocal translocation partner, RAR-α-PML,disease latency is significantly shortened. Experiments in mice expressingchimeric CBF proteins also indicate that these abnormalities alone may beinsufficient to induce leukemia. When AML1 was replaced by AML1-ETO inmice using a knock-in strategy, a block in hematopoiesis was seen but mouseembryos died in midgestation due to development of severe central nervoussystem (CNS) hemorrhages (38). To overcome this embryonic lethality, subse-quent experiments employed a model in which the AML1-ETO expression wasinducible under the control of a tetracycline-responsive element. Despite highexpression of AML1-ETO in the bone marrow cells of these mice, no leukemia


was seen during 24 mo of observation, although a partial block of myeloid dif-ferentiation was seen (39). Although mice expressing AML1-ETO do notdevelop leukemia, they are at a much higher risk of developing leukemia afterexposure to alkylating agents as compared with control mice (40). Similarly,mice expressing CBFβ-MYH11 did not develop leukemia but after exposure tolow-dose alkylating agents, had a high rate of leukemic transformation,whereas no cases of leukemia were seen in similarly treated control mice (41).This suggests that secondary mutations cooperate with AML1-ETO to induceleukemia and that AML1-ETO–induced leukemia may be a multistep process.

In the case of Burkitt’s lymphoma, transgenic mice overexpressing c-mycdevelop lymphomas (42). However, it is likely that other abnormalities cooper-ate with c-myc to induce lymphoma. Up to 80% of cases of Burkitt’s lym-phoma exhibit mutations in the p53 tumor suppressor gene, and whenwild-type p53 is expressed in a Burkitt’s lymphoma cell expressing a mutatedform of p53, rapid cell death by apoptosis ensues (43). In fact, enforcedexpression of c-myc causes apoptosis as well as cellular proliferation (44,45).Therefore, it may only be with the development of secondary abnormalitiesthat circumvent apoptosis, such as Bcl-2 overexpression or p53 mutations, thatc-myc is capable of malignant transformation.

4. HITTING THE TARGET

For maximal utility, the identification of crucial early events in malignantprogression is the first step in the successful development of a targeted therapy.An equally important issue is the selection of patients for clinical trials basedon the presence of the appropriate target. Finally, the normal cellular functionof the target will determine the toxicity of a targeted agent, and an ideal targetwould be dispensable for normal cellular function. Currently, kinase inhibitorsare easier to develop than agents that target transcription factors. This is in partdue to kinases having well-defined hydrophobic ATP-binding pockets to whichinhibitors can easily be targeted, as opposed to transcription factors that oftenhave broad flat hydrophilic-binding surfaces that do not present useful struc-tures for disrupting protein–protein interactions.

4.1. Tyrosine Kinase InhibitorsBecause deregulated tyrosine kinase activity is involved in the pathogenesis

and disease progression of CML, CMML associated with t(5;12), ALCL asso-ciated with t(2;5), and AML associated with ITDs of Flt3, these diseases areobvious choices for the development of specific signal transduction inhibitors.The following discussion focuses on the development of STI571 for the treat-ment of CML, because this serves as a paradigm for the development of similartherapies.


4.1.1. CHRONIC MYELOGENOUS LEUKEMIA

CML accounts for 20% of all cases of leukemia, with an annual incidence of1 to 1.5 cases per 100,000. The median age of onset is approx 60 yr, but all agegroups are affected (46). Three clinical phases are recognized: a chronic phaselasting 4–6 yr, an accelerated phase lasting 6–18 mo, and a blast phase lasting3–6 mo. The chronic phase is characterized by a massive proliferation of matur-ing myeloid cells (white cells and frequently platelets), which, as the diseaseprogresses, lose their capacity to differentiate until eventually the disease termi-nates in an acute leukemia, termed blast crisis. The accelerated phase is an inter-mediate phase characterized by increasing myeloid immaturity, systemicsymptoms, and refractoriness to therapy. The only therapy known to cure CMLis allogeneic stem cell transplantation, but because most patients are too old orlack suitable donors for this procedure, less than one third of patients are candi-dates for this treatment. The 5–10-yr survival after allogeneic stem cell trans-plantation is 65%, with significant procedural-related morbidity and mortality.Oral chemotherapy agents, such as hydroxyurea or busulfan, can control bloodcounts in most patients in chronic phase but do not delay the onset of blast crisis(47). Patients treated with interferon-α live an average of 2 yr longer thanpatients treated with chemotherapy (48). The best survival advantage is seen inthose patients who achieve a major cytogenetic response (a reduction in the per-centage of marrow metaphases containing the Ph chromosome to less than35%), although this occurs in less than one third of patients. Moreover, as manyas 20% of patients discontinue interferon-α therapy due to intolerable toxicity.The addition of subcutaneous ara-C to interferon-α has increased response ratesbut at the cost of increased toxicity (49). These shortcomings provided the impe-tus for the development of a more effective, less toxic therapy for CML.

4.1.2. BCR-ABL: THE IDEAL TARGET

BCR-ABL affects numerous downstream signaling pathways, which leadsto increased cellular proliferation, decreased adhesion, inhibition of apoptosis,and possibly genetic instability (50). However, because all of these events aredependent on the tyrosine kinase activity of the fusion protein, it is clear thatinhibition of the enzymatic activity of BCR-ABL should be an effective treat-ment for CML because BCR-ABL is present in the majority of patients withCML, is the causative abnormality of the disease, and its kinase activity isessential for transformation. Moreover, because Abl knock-out mice are viable,it is likely that ABL kinase activity would be dispensable for normal cellularfunction, suggesting that an ABL kinase inhibitor would have a relativelyselective effect on BCR-ABL-expressing cells (51).

4.1.3. DEVELOPING AN INHIBITOR OF THE BCR-ABL TYROSINE KINASE

Tyrosine kinases, such as BCR-ABL, catalyze the transfer of phosphate fromadenosine triphosphate (ATP) to selected tyrosine residues on substrate proteins.


With their tyrosine residues in the phosphorylated form, substrate proteinsassume conformational changes leading to association with other downstreameffectors, propagating signal transduction. Tyrosine kinases have a vital role incell growth, differentiation, and survival. Because all protein kinases use ATP asa phosphate donor and as there is a high degree of conservation among kinasedomains, particularly in the ATP-binding sites, it was believed that inhibitors ofATP binding would lack sufficient target specificity to be clinically useful. Thiswas the case with the first tyrosine kinase inhibitors identified, such as her-bimycin-A, which were all natural plant derivatives. However, in 1988, Yaish etal. published a series of compounds known as tyrphostins that demonstrated thatspecific tyrosine kinase inhibitors could be developed (52). Near the same time,scientists at Ciba Geigy (now Novartis) were performing high throughoutscreens of chemical libraries searching for compounds with kinase inhibitoryactivity. They eventually identified a lead compound with kinase inhibitory ofthe 2-phenylaminopyrimidine class. Although of low potency and poor speci-ficity, this lead compound served as a base from which a series of related com-pounds were synthesized. By analyzing the relationship between structure andactivity, this series of compounds was optimized to inhibit several targets (53).One series of compounds, optimized against the platelet-derived growth factorreceptor (PDGF-R) proved to be equally active against the ABL tyrosine kinase.STI571 (formerly CGP57148, now Gleevec; imatinib mesylate) emerged as thelead compound for clinical development based on its superior in vitro selectivityagainst CML cells and its drug-like properties, including pharmacokinetics andformulation properties (53).

4.1.3.1. Preclinical Studies. Experiments in our laboratory showed thatSTI571 was a potent and selective inhibitor of the ABL tyrosine kinases,including BCR-ABL (54). The concentration (IC50) of STI571 that resulted ina 50% reduction in substrate phosphorylation and cellular tyrosine phosphory-lation induced by BCR-ABL was 0.025 µM and 0.25 µM, respectively. Theonly other tyrosine kinase found to be inhibited by STI571, besides ABL andthe PDGF-R, was c-kit. STI571 specifically inhibited the proliferation ofmyeloid cell lines containing BCR-ABL. In addition, colony-forming assaysfrom patients with CML showed a marked decrease (92–98%) in the numberof BCR-ABL colonies formed with no inhibition of normal colony formationwhen grown in the presence of 1 µM STI571. Similar results were reportedelsewhere (55). Long-term marrow culture experiments showed that prolongedexposure to STI571 produced a sustained inhibitory effect on CML progeni-tors, with little toxicity to normal progenitors (56). Subsequent experimentsshowed that p185- and p210-expressing cells were equally sensitive to STI571(57,58). Dose-dependent inhibition of tumor growth was seen in BCR-ABL-inoculated mice treated with STI571, but a daily dosing failed to eradicate thetumors (54). Gambacorti and colleagues subsequently showed that three


times/day oral dosing of STI571 effectively eradicated BCR-ABL-containingtumors in nude mice (59). Because the half-life of STI571 in mice is approx4 h, it seemed that continuous exposure to STI571 would be required for opti-mal antileukemic effects.

4.1.3.2. Clinical Trials of STI571 in CML. Based on the efficacy ofSTI571 in several preclinical models and an acceptable animal toxicology pro-file, a phase I clinical trial with STI571 started in June 1998. This was a dose-escalation study designed to establish the maximum tolerated dose (MTD),with clinical efficacy being a secondary endpoint. Patients were enrolled in 14successive dose cohorts ranging from 25 to 1000 mg of STI571. Patients wereeligible if they were in the chronic phase of CML and had failed therapy withinterferon-α. STI571 was administered as a once daily oral therapy, and noother cytoreductive agents were allowed. Once doses of 300 mg or greaterwere reached, 53 of 54 patients achieved a complete hematologic response(60). Responses were typically seen within the first 3 wk of therapy and havebeen maintained in 96% of patients, with a median follow-up of 310 d. At thisdose level (≥ 300 mg), major cytogenetic responses were seen in 31% ofpatients, while 13% achieved a complete cytogenetic response. Side effectshave been minimal, with no dose-limiting toxicities encountered. Grade 2 and3 myelosuppression was observed at a dose ≥ 300 mg in 21% and 8% ofpatients, respectively. Myelosuppression is likely consistent with a therapeuticeffect because the Ph-positive clone contributes the majority of hematopoiesisin these patients. Pharmacokinetic studies showed that the half-life of STI571is 13–16 h, which is sufficient to permit once-daily dosing. Although the fol-low-up on this group of patients is relatively short (median 1 yr), these dataindicate that an ABL-specific tyrosine kinase inhibitor has significant activityin CML, even in interferon refractory patients.

Given the effectiveness of STI571 in patients in chronic phase who had failedinterferon, the phase I studies were expanded to include patients with CML inmyeloid and lymphoid blast crisis and patients with relapsed or refractory Phchromosome-positive ALL. Patients have been treated with daily doses of 300 to1000 mg of STI571. Twenty-one of 38 (55%) patients with myeloid blast crisisresponded to therapy, defined by a decrease in percentage of marrow blasts toless than 15%. Eight of 38 (21%) patients had marrow blasts cleared to < 5%(61). Seven of 38 (18%) of the patients in myeloid blast crisis have remained inremission on STI571, with follow-up ranging from 101 to 349 d. Fourteen of 20(70%) patients with lymphoid phenotype disease, CML in lymphoid blast crisis,or Ph-positive ALL responded, with 11 of 20 (55%) clearing their marrows to <5% blasts. Unfortunately, all but one of the lymphoid phenotype patientsrelapsed between days 42 and 123. Thus, STI571 has remarkable single-agentactivity in CML blast crisis and Ph-positive ALL, but responses tend not to bedurable. However, these studies demonstrate that in the majority of cases, the


leukemic clone in BCR-ABL–positive acute leukemias, including CML blastcrisis, remains at least partially dependent on BCR-ABL kinase activity for sur-vival. In late 1999, phase II studies were initiated to evaluate the safety and effi-cacy of STI571 in larger cohorts of patients. Patients in all phases of the disease(chronic phase [having failed interferon], accelerated, and blast crisis) wereenrolled in studies at 27 institutions in six countries. Evaluation of response andpharmacokinetic data from the phase I study indicated that doses of 400 to 600mg should be optimal for phase II testing (62). Between December 1999 andMay 2000, 532 patients in chronic phase who were refractory to or intolerant ofinterferon-α were started on treatment with STI571 at a 400-mg daily dose. Inrecently updated results, after a median exposure of 18 mo, 60% and 41% ofpatients achieved major and complete cytogenetic responses, respectively. Only8% of patients discontinued treatment due to disease progression, with only 2%of all patients stopping therapy due to adverse events (63). Of 235 treatedpatients in accelerated phase 51% achieved a complete hematologic response(CHR) with or without peripheral blood recovery (neutrophils > 1.0 × 109/L andplatelets > 100 × 109/L). Eighteen percent achieved a complete cytogeneticresponse (64). Again, these results were achieved without substantial toxicity,though not surprising, up to 30% of patients experienced grade 3/4 myelosup-pression in this study. Nevertheless, only 2% of patients developed febrile neu-tropenia. Of 260 treated patients in myeloid blast crisis 52% had some form ofresponse, with 24% clearing their marrows to less than 5% blasts (65). Majorand complete cytogenetic responses were seen in 15% and 7% of patients,respectively. Toxicity was comparable to that seen in the accelerated-phasestudy. Median survival was 6.9 mo, and 20% of patients are projected to be aliveat 18 mo. Historically, patients treated with chemotherapy for myeloid blast cri-sis have had a median survival of approx 3 mo.

4.1.4. DOSE SELECTION

From the dose-finding study, complete hematologic responses occurred inalmost all patients in chronic phase treated at doses of 300 mg and higher andcytogenetic responses were seen once this dose level was reached. In addition,pharmacokinetic data showed that this dose level achieved in vivo concentrationsapproaching the predicted in vitro IC90 for cellular proliferation of 1 µM (62).Finally, an analysis of white blood and platelet count responses over time sug-gested that doses of 400 to 600 mg were on the plateau of a dose-response curve,indicating that this dose range would be an efficacious dose for phase II testing.However, in the dose-escalation study, an MTD of STI571 was never reached(60). Although traditional drug development uses dose escalation until an MTD isestablished, with molecularly targeted therapies, this may not be an appropriateendpoint. A more appropriate endpoint may be the dose that achieves moleculartarget inhibition. Therefore, in the case of STI571 and CML, an optimal dose


should approximate that which achieves maximal BCR-ABL kinase inhibition.An analysis of BCR-ABL kinase inhibition, assaying for decreases in phosphory-lation of the BCR-ABL substrate, Crkl, has suggested that a plateau in inhibitionis seen with more than 250 mg (60). Additional experiments are being conductedto determine the percentage of kinase activity that is inhibited at these dose levels(66). CML lends itself to this kind of molecular monitoring because tumor cellsare easily accessible and the kinase itself or its substrates can be monitored forinhibition. These types of assays are clearly more problematic for solid tumors,but are necessary to determine the penetration of these types of agents into solidtumors. In the absence of specific assays, even information about intracellulardrug levels in tumor samples would be a useful surrogate. This type of dataregarding maximal kinase inhibition could be particularly useful in explainingresponse variability and could also be useful in individualizing therapy.

4.1.5. FUTURE DIRECTIONS IN CML THERAPY

The clinical data presented here demonstrate that STI571 is employed tooptimum effect in CML when used early, before disease progression. An ongo-ing phase III randomized study is comparing STI571 with interferon and ara-Cin newly diagnosed patients. The results of this study, when combined withmore mature data from the phase II studies, will help to determine the place ofSTI571 in future CML treatment algorithms. It is tempting to speculate that asBCR-ABL may be the sole oncogenic abnormality driving proliferation inearly stage disease, STI571 alone may be sufficient therapy in some patientswith CML. However, as additional genetic abnormalities accumulate with dis-ease progression, CML cells may no longer be solely dependent on BCR-ABLfor survival. Thus, in blast crisis patients, therapy with STI571 alone is clearlyinsufficient for most patients. Here, the paradigm of APL and ATRA may beparticularly instructive. It is unlikely that PML-RAR-α is the sole molecularabnormality that causes APL, but it is clearly one of the critical pathogeneticevents. Targeting PML-RAR-α with ATRA yields a high response rate, butmost patients relapse on single-agent therapy. However, combinations ofATRA with chemotherapy lead to high rate of cure. Similarly, in blast crisis, itis unlikely that BCR-ABL is the sole oncogenic abnormality, yet it remainscritical to the survival of the leukemic clone. Thus, combinations of STI571with other antileukemic agents for blast crisis patients could lead to a signifi-cantly improved prognosis. This paradigm is likely to apply to most leukemiasand lymphomas where multistep disease pathogenesis is the rule.

4.1.6. THE PROBLEM OF RESISTANCE AND THE RATIONALE

FOR COMBINATION THERAPY

Despite the high initial response rates in patients in blast crisis manypatients relapse and not all respond to STI571. One of the most useful catego-


rizations of relapse/resistance mechanisms has been separation of patients withpersistent inhibition of the BCR-ABL kinase and those with reactivation of theBCR-ABL kinase. Patients with persistent inhibition of the BCR-ABL kinaseare predicted to have additional molecular abnormalities other than BCR-ABLdriving the growth and survival of the malignant clone. In contrast, patientswith persistent BCR-ABL kinase activity or reactivation of the kinase arebelieved to have resistance mechanisms that either prevent STI571 from reach-ing the target or render the target insensitive to BCR-ABL. In the former cate-gory are mechanisms such as drug efflux or protein binding of STI571. In thelatter category would be mutations of the BCR-ABL kinase that render BCR-ABL insensitive to STI571 or amplification of the BCR-ABL protein (67–70).

In patients who relapse after an initial response to STI571, the majority ofthese patients have reactivation of the BCR-ABL kinase (71). No changes in druglevels have been observed, and leukemic cells from these patients have decreasedcellular sensitivity to STI571. This suggests that resistance is due to intrinsic cel-lular properties rather than protein binding of drug or drug metabolism. Interest-ingly, half of these patients have developed point mutations in the ABL kinase thatrender the kinase variably less sensitive to STI571 (71). At least one of the pointmutations is at a site predicted to be a contact site between STI571 and the ABLkinase based on the crystal structure (71,72). Several others are at residues adja-cent to contact points, whereas others are in the kinase activation loop (73,74).Finally, approximately one third of patients who relapse after an initial responsehave BCR-ABL amplification (71). BCR-ABL mutation and amplification havenot been commonly seen in patients with de novo STI571 resistance, and ongoingstudies are aimed at identifying mechanisms of resistance in these patients.

To circumvent resistance, the combination of STI571 with other activeantileukemic agents seems desirable. We have shown that inhibition of BCR-ABL by STI571 can reverse the intrinsic drug resistance seen in CML cells andthat combinations with drugs such as daunorubicin, ara-C, and interferon-α areassociated with additive or even synergistic effects in vitro, providing a strongrationale for combination studies (75). Similar studies have been performedwith etoposide and ara-C (76). Combination studies with low-dose interferonand ara-C are currently underway for patients in chronic phase, while combi-nations of STI571 with high-dose chemotherapy regimens (vincristine,daunorubicin, and prednisone and high-dose ara-C) are also in progress forpatients in lymphoid and myeloid blast crisis, respectively. As mechanisms ofCML disease progression and relapse become apparent, it is hoped that agentsthat target these specific abnormalities could also be developed.

4.1.7. OTHER THERAPEUTIC TARGETS FOR STI571

Although STI571 was tested as a treatment for BCR-ABL-associatedleukemias, its original target was the PDGF-R, and it was subsequently shown to


inhibit the c-kit tyrosine kinase. Thus, STI571 should also have activity in dis-eases associated with constitutive activation of these kinases. In the case of c-kit,activating mutations are associated with a gastrointestinal stromal tumor (GIST)(77,78), which is highly refractory to chemotherapy. Results from an ongoingphase I study using STI571 to treat patients with GIST have shown responserates close to 60% (79,80). A particularly interesting finding from this study wasthat activating mutations of c-kit correlated with response, whereas patientsexpressing wild-type c-kit had a significantly lower response rate (79). This sug-gests that other tumors where c-kit is expressed but not activated by mutationmay also be less likely to respond to STI571. Tumors that express c-kit includegerm cell tumors, small-cell lung cancer (SCLC), AML, neuroblastoma,melanoma, ovarian cancer, and myeloma. The exception to this might be thefraction of patients with AML who express c-kit-activating point mutations. Inany case, studies are ongoing with STI571 to determine whether tumors thatexpress c-kit will respond to STI571.

The majority of cases of systemic mastocytosis have a mutation of asparticacid 816 to valine (D816V) in the kinase domain of c-kit, resulting in activa-tion of c-kit. Unfortunately, the kinase activity of the D816V mutant isoformwas recently shown to be resistant to STI571 (81). Thus, STI571 is unlikely tobe useful in this disorder.

With respect to PDGF-R as a target, patients with CMML with a (5;12)translocation resulting in expression of the constitutively active Tel-PDGF-Rfusion protein tyrosine kinase are an ideal target. STI571 has shown in vitroinhibition of leukemic cell lines expressing Tel-PDGF-R, and this finding hasbeen corroborated in patient studies where remarkable clinical benefits withSTI571 have been observed (57,82). Glioblastomas, the most common braintumor and a highly chemotherapy- and radiation-resistant tumor, are associatedwith an autocrine growth loop involving PDGF and its receptor. STI571inhibits the growth of glioblastoma cells injected into the brains of nude mice,suggesting that this agent could have potential as therapy for this currentlyincurable disease (83). Numerous other malignancies have also been reportedto have autocrine activation of PDGF-R, including non-small cell lung, breast,and prostate cancers and several sarcomas; however, the data supporting a rolefor PDGF-R activation in these diseases are less compelling (84). Neverthe-less, clinical trials with STI571 in these diseases could be envisioned to testthis hypothesis. Unlike the case with BCR-ABL (and possibly Tel-PDGF-R),however, it is unlikely that a defect in a single protein kinase is responsible formalignant transformation in most of the aforementioned tumors; therefore, it isunreasonable to expect results as dramatic as those seen in the treatment ofCML when using STI571 alone for these other indications. Greater efficacymay be expected when the kinase inhibitor is used in combination withchemotherapy or even other molecularly targeted therapies. Finally, PDGF-R


activation may have a role in several fibrotic disorders, such as myelofibrosis,pulmonary fibrosis, and hepatic fibrosis (85). Given the acceptable toxicityprofile, an exploration of the activity of STI571 in these disorders may also bewarranted.

5. TRANSLATING THE SUCCESS OF STI571 TO OTHER MOLECULAR TARGETS

The clinical trials with STI571 are a dramatic demonstration of the potentialof targeting molecular pathogenetic events in a malignancy. In applying thisparadigm to other malignancies, it is important to recognize that BCR-ABLand CML have several features that were critical to the success of this agent.As noted, BCR-ABL tyrosine kinase activity has clearly been demonstrated tobe critical to the pathogenesis of CML. Thus, not only was the target of STI571known but it also was directed against the critical event in the development ofCML. Another important feature that the results demonstrate, as with mostmalignancies, is that treatment of early stage disease yields better results.Specifically, the rate and durability of responses have been notably superior inpatients in chronic phase as opposed to those in blast phase. Therefore, toreproduce the success of STI571 in other malignancies, it is imperative to iden-tify the critical early events in malignant progression. It is equally importantthat selection for clinical trials is limited to those patients whose malignanciesexpress the appropriate target. In clinical trials using STI571, this was clearlyfeasible because patients with activation of BCR-ABL were easily identifiableby the presence of the Ph chromosome. In this regard, as reagents to analyzemolecular endpoints are developed, these same reagents should be useful inidentifying appropriate candidates for treatment with a specific agent. With thecombination of a critical pathogenetic target that is easily identifiable early inthe course of the disease and an agent that targets this abnormality, remarkableresults can be achieved. The obvious goal is to identify these early patho-genetic events in each malignancy and to develop agents that specifically targetthese abnormalities.

ACKNOWLEDGMENTS

BJD is funded by grants from the NCI, a Specialized Center of ResearchAward from The Leukemia and Lymphoma Society, a Clinical Scientist Awardfrom the Burroughs Wellcome Fund, and the T. J. Martell Foundation.

REFERENCES1. Nowell PC, Hungerford DA. A minute chromosome in human chronic granulocytic leukemia.

Science 1960;132:1497–1501.2. Rowley JD. The role of chromosome translocations in leukemogenesis. Semin Hematol

1999;36:59–72.


3. Tsujimoto Y, Finger LR, Yunis J, Nowell PC, Croce CM. Cloning of the chromosome break-point of neoplastic B cells with the t(14;18) chromosome translocation. Science 1984;226:1097–1099.

4. Reed JC, Tsujimoto Y, Epstein SF, et al. Regulation of bcl-2 gene expression in lymphoidcell lines containing normal #18 or t(14;18) chromosomes. Oncogene Res 1989;4:271–282.

5. Dalla-Favera R, Bregni M, Erikson J, Patterson D, Gallo RC, Croce CM. Human c-myc oncgene is located on the region of chromosome 8 that is translocated in Burkitt lymphomacells. Proc Natl Acad Sci U S A 1982;79:7824–7827.

6. Adams JM, Gerondakis S, Webb E, Corcoran LM, Cory S. Cellular myc oncogene is alteredby chromosome translocation to an immunoglobulin locus in murine plasmacytomas and isrearranged similarly in human Burkitt lymphomas. Proc Natl Acad Sci U S A 1983;80:1982–1986.

7. Bosch F, Jares P, Campo E, et al. PRAD-1/cyclin D1 gene overexpression in chronic lym-phoproliferative disorders: a highly specific marker of mantle cell lymphoma. Blood1994;84:2726–2732.

8. Guidez F, Ivins S, Zhu J, Soderstrom M, Waxman S, Zelent A. Reduced retinoic acid-sensitiv-ities of nuclear receptor corepressor binding to PML- and PLZF-RARalpha underlie molecu-lar pathogenesis and treatment of acute promyelocytic leukemia. Blood 1998;91:2634–2642.

9. Collins SJ. Acute promyelocytic leukemia: relieving repression induces remission. Blood1998;91:2631–2633.

10. Lin RJ, Nagy L, Inoue S, Shao W, Miller WH Jr., Evans RM. Role of the histone deacetylasecomplex in acute promyelocytic leukaemia. Nature 1998;391:811–814.

11. He LZ, Guidez F, Tribioli C, et al. Distinct interactions of PML-RARalpha and PLZF-RAR-alpha with co-repressors determine differential responses to RA in APL. Nat Genet1998;18:126–135.

12. Yoshida H, Kitamura K, Tanaka K, et al. Accelerated degradation of PML-retinoic acidreceptor alpha (PML-RARα) oncoprotein by all-trans-retinoic acid in acute promyelocyticleukemia: possible role of the proteasome pathway. Cancer Res 1996;56:2945–2948.

13. Huang ME, Ye YC, Chen SR, et al. Use of all-trans retinoic acid in the treatment of acutepromyelocytic leukemia. Blood 1988;72:567–572.

14. Liu P, Tarle SA, Hajra A, et al. Fusion between transcription factor CBF beta/PEBP2 betaand a myosin heavy chain in acute myeloid leukemia. Science 1993;261:1041–1044.

15. Langabeer SE, Walker H, Rogers JR, et al. Incidence of AML1/ETO fusion transcripts inpatients entered into the MRC AML trials. MRC Adult Leukaemia Working Party. Br JHaematol 1997;99:925–928.

16. Downing JR, Higuchi M, Lenny N, Yeoh AE. Alterations of the AML1 transcription factor inhuman leukemia. Semin Cell Dev Biol 2000;11:347–360.

17. Wang J, Hoshino T, Redner RL, Kajigaya S, Liu JM. ETO, fusion partner in t(8;21) acutemyeloid leukemia, represses transcription by interaction with the human N-CoR/mSin3/HDAC1 complex. Proc Natl Acad Sci U S A 1998;95:10860–10865.

18. Lutterbach B, Hou Y, Durst KL, Hiebert SW. The inv(16) encodes an acute myeloidleukemia 1 transcriptional corepressor. Proc Natl Acad Sci U S A 1999;96:12822–12827.

19. Langabeer SE, Walker H, Gale RE, et al. Frequency of CBF beta/MYH11 fusion transcriptsin patients entered into the U.K. MRC AML trials. The MRC Adult Leukaemia WorkingParty. Br J Haematol 1997;96:736–739.

20. Guidez F, Petrie K, Ford AM, et al. Recruitment of the nuclear receptor corepressor N-CoRby the TEL moiety of the childhood leukemia-associated TEL-AML1 oncoprotein. Blood2000;96:2557–2561.

21. Rowley JD. Letter: a new consistent chromosomal abnormality in chronic myelogenousleukemia identified by quinacrine fluorescence and Giemsa staining. Nature 1973;243:290–293.

22. Shtivelman E, Lifshitz B, Gale RP, Canaani E. Fused transcript of abl and bcr genes inchronic myelogenous leukaemia. Nature 1985;315:550–554.


23. Lugo TG, Pendergast AM, Muller AJ, Witte ON. Tyrosine kinase activity and transformationpotency of bcr-abl oncogene products. Science 1990;247:1079–1082.

24. Golub TR, Barker GF, Lovett M, Gilliland DG. Fusion of PDGF receptor beta to a novel ets-like gene, tel, in chronic myelomonocytic leukemia with t(5;12) chromosomal translocation.Cell 1994;77:307–316.

25. Morris SW, Kirstein MN, Valentine MB, et al. Fusion of a kinase gene, ALK, to a nucleolarprotein gene, NPM, in non-Hodgkin’s lymphoma. Science 1994;263:1281–1284.

26. Nakao M, Yokota S, Iwai T, et al. Internal tandem duplication of the flt3 gene found in acutemyeloid leukemia. Leukemia 1996;10:1911–1918.

27. Rombouts WJ, Blokland I, Lowenberg B, Ploemacher RE. Biological characteristics andprognosis of adult acute myeloid leukemia with internal tandem duplications in the Flt3gene. Leukemia 2000;14:675–683.

28. Abu-Duhier FM, Goodeve AC, Wilson GA, et al. FLT3 internal tandem duplication muta-tions in adult acute myeloid leukaemia define a high-risk group. Br J Haematol2000;111:190–195.

29. Meshinchi S, Woods WG, Stirewalt DL, et al. Prevalence and prognostic significance of Flt3internal tandem duplication in pediatric acute myeloid leukemia. Blood 2001;97:89–94.

30. Yamamoto Y, Kiyoi H, Nakano Y, et al. Activating mutation of D835 within the activationloop of FLT3 in human hematologic malignancies. Blood 2001;97:2434–2439.

31. Gari M, Goodeve A, Wilson G, et al. c-kit proto-oncogene exon 8 in-frame deletion plusinsertion mutations in acute myeloid leukaemia. Br J Haematol 1999;105:894–900.

32. Beghini A, Peterlongo P, Ripamonti CB, et al. C-kit mutations in core binding factorleukemias. Blood 2000;95:726–727.

33. Daley GQ, Van Etten RA, Baltimore D. Induction of chronic myelogenous leukemia in miceby the P210bcr/abl gene of the Philadelphia chromosome. Science 1990;247:824–830.

34. Heisterkamp N, Jenster G, ten Hoeve J, Zovich D, Pattengale PK, Groffen J. Acuteleukaemia in bcr/abl transgenic mice. Nature 1990;344:251–253.

35. Kuefer MU, Look AT, Pulford K, et al. Retrovirus-mediated gene transfer of NPM-ALKcauses lymphoid malignancy in mice. Blood 1997;90:2901–2910.

36. Mizuki M, Fenski R, Halfter H, et al. Flt3 mutations from patients with acute myeloidleukemia induce transformation of 32D cells mediated by the Ras and STAT5 pathways.Blood 2000;96:3907–3914.

37. Merghoub T, Gurrieri C, Piazza F, Pandolfi PP. Modeling acute promyelocytic leukemia inthe mouse: new insights in the pathogenesis of human leukemias. Blood Cells Mol Dis2001;27:231–248.

38. Okuda T, Cai Z, Yang S, et al. Expression of a knocked-in AML1-ETO leukemia geneinhibits the establishment of normal definitive hematopoiesis and directly generates dysplas-tic hematopoietic progenitors. Blood 1998;91:3134–3143.

39. Rhoades KL, Hetherington CJ, Harakawa N, et al. Analysis of the role of AML1-ETO inleukemogenesis, using an inducible transgenic mouse model. Blood 2000;96:2108–2115.

40. Higuchi M, O’Brien D, Lenny N, Yang S, Cai Z, Downing JR. Expression of AML1-ETOimmortalizes myeloid progenitors and cooperates with secondary mutations to induce granu-locytic sarcoma/acute myeloid leukemia. Blood 2000;96:222a.

41. Castilla LH, Garrett L, Adya N, et al. The fusion gene Cbfβ-MYH11 blocks myeloid differenti-ation and predisposes mice to acute myelomonocytic leukaemia. Nat Genet 1999; 23:144–146.

42. Adams JM, Harris AW, Pinkert CA, et al. The c-myc oncogene driven by immunoglobulinenhancers induces lymphoid malignancy in transgenic mice. Nature 1985; 318:533–538.

43. Ramqvist T, Magnusson KP, Wang Y, Szekely L, Klein G, Wiman KG. Wild-type p53induces apoptosis in a Burkitt lymphoma (BL) line that carries mutant p53. Oncogene 1993;8:1495–1500.

44. Milner AE, Grand RJ, Waters CM, Gregory CD. Apoptosis in Burkitt lymphoma cells is dri-ven by c-myc. Oncogene 1993; 8:3385–3391.


45. Gavioli R, Frisan T, Vertuani S, Bornkamm GW, Masucci MG. c-myc overexpression acti-vates alternative pathways for intracellular proteolysis in lymphoma cells. Nat Cell Biol2001; 3:283–288.

46. O’Dwyer ME, Druker BJ. Chronic myelogenous leukaemia—new therapeutic principles. JIntern Med 2001; 250:3–9.

47. Silver RT, Woolf SH, Hehlmann R, et al. An evidence-based analysis of the effect of busul-fan, hydroxyurea, interferon, and allogeneic bone marrow transplantation in treating thechronic phase of chronic myeloid leukemia: developed for the American Society of Hema-tology. Blood 1999; 94:1517–1536.

48. Chronic Myeloid Leukemia Trialists’ Collaborative Group. Interferon alfa versus chemother-apy for chronic myeloid leukemia: a meta-analysis of seven randomized trials. J Natl CancerInst 1997;89:1616–1620.

49. Guilhot F, Chastang C, Michallet M, et al. Interferon alfa-2b combined with cytarabine ver-sus interferon alone in chronic myelogenous leukemia. French Chronic Myeloid LeukemiaStudy Group. N Engl J Med 1997;337:223–229.

50. Deininger MW, Goldman JM, Melo JV. The molecular biology of chronic myeloid leukemia.Blood 2000;96:3343–3356.

51. Tybulewicz VL, Crawford CE, Jackson PK, Bronson RT, Mulligan RC. Neonatal lethalityand lymphopenia in mice with a homozygous disruption of the c-abl proto-oncogene. Cell1991;65:1153–1163.

52. Yaish P, Gazit A, Gilon C, Levitzki A. Blocking of EGF-dependent cell proliferation by EGFreceptor kinase inhibitors. Science 1988;242:933–935.

53. Druker BJ, Lydon NB. Lessons learned from the development of an abl tyrosine kinaseinhibitor for chronic myelogenous leukemia. J Clin Invest 2000;105:3–7.

54. Druker BJ, Tamura S, Buchdunger E, et al. Effects of a selective inhibitor of the Abl tyrosinekinase on the growth of Bcr-Abl positive cells. Nat Med 1996;2:561–566.

55. Deininger MW, Goldman JM, Lydon N, Melo JV. The tyrosine kinase inhibitor CGP57148Bselectively inhibits the growth of BCR-ABL-positive cells. Blood 1997;90:3691–3698.

56. Kasper B, Fruehauf S, Schiedlmeier B, Buchdunger E, Ho AD, Zeller WJ. Favorable thera-peutic index of a p210(BCR-ABL)-specific tyrosine kinase inhibitor; activity on lineage-committed and primitive chronic myelogenous leukemia progenitors. Cancer ChemotherPharmacol 1999;44(5):433–438.

57. Carroll M, Ohno-Jones S, Tamura S, et al. CGP 57148, a tyrosine kinase inhibitor, inhibitsthe growth of cells expressing BCR-ABL, TEL-ABL, and TEL-PDGFR fusion proteins.Blood 1997;90:4947–4952.

58. Beran M, Cao X, Estrov Z, et al. Selective inhibition of cell proliferation and BCR-ABLphosphorylation in acute lymphoblastic leukemia cells expressing Mr 190,000 BCR-ABLprotein by a tyrosine kinase inhibitor (CGP-57148). Clin Cancer Res 1998;4:1661–1672.

59. le Coutre P, Mologni L, Cleris L, et al. In vivo eradication of human BCR/ABL-positiveleukemia cells with an ABL kinase inhibitor. J Natl Cancer Inst 1999;91:163–168.

60. Druker BJ, Talpaz M, Resta DJ, et al. Efficacy and safety of a specific inhibitor of theBCR-ABL tyrosine kinase in chronic myeloid leukemia. N Engl J Med 2001;344:1031–1037.

61. Druker BJ, Sawyers CL, Kantarjian H, et al. Activity of a specific inhibitor of the BCR-ABLtyrosine kinase in the blast crisis of chronic myeloid leukemia and acute lymphoblasticleukemia with the Philadelphia chromosome. N Engl J Med 2001;344:1038–1042.

62. Peng B, Hayes M, Druker BJ, et al. Clinical pharmacokinetics and pharmacodynamics ofST1571 in a phase I trial in chronic myelogenous leukemia (CML) patients [abstract 3468].Proc Am Assoc Cancer Res 2000;41.

63. Kantarjian H, Sawyers C, Hochhaus A, et al. Gleevec (imatinib mesylate) induced hemato-logic and cytogenetic responses confirmed and expanded in patient’s with chronic myeloidleukemia (CML)—a phase II study update. Blood 2001;98:845a.


64. Talpaz M, Silver RT, Druker B, et al. Gleevec (formerly ST1571): an active drug in patientswith Ph+ chronic myeloid leukemia in accelerated phase—updated results of a phase IIstudy. Blood 2001;96:845a.

65. Sawyers C, Hochhaus A, Feldman E, et al. Gleevec/Glivic (imatinib mesylate, STI571) inpatients with chronic myeloid leukemia (CML) in myeloid blast crisis: updated results of aphase II study. Blood 2001;98:845a.

66. Karamlou K, Lucas L, Druker B. Identification of molecular endpoints as a guide for clinicaldecision making in STI571-treated chronic myelogenous leukemia patients. Blood 2000;96:98a.

67. Mahon FX, Deininger MW, Schultheis B, et al. Selection and characterization of BCR-ABLpositive cell lines with differential sensitivity to the tyrosine kinase inhibitor STI571: diversemechanisms of resistance. Blood 2000;96:1070–1079.

68. le Coutre P, Tassi E, Varella-Garcia M, et al. Induction of resistance to the Abelson inhibitorSTI571 in human leukemic cells through gene amplification. Blood 2000;95:1758–1766.

69. Weisberg E, Griffin JD. Mechanism of resistance to the ABL tyrosine kinase inhibitorSTI571 in BCR/ABL-transformed hematopoietic cell lines. Blood 2000;95:3498–3505.

70. Gambacorti-Passerini C, Barni R, le Coutre P, et al. Role of alphal acid glycoprotein in the invivo resistance of human BCR-ABL(+) leukemic cells to the abl inhibitor STI571. J NatlCancer Inst 2000;92:1641–1650.

71. Gorre ME, Mohammed M, Ellwood K, et al. Clinical resistance to STI-571 cancer therapycaused by BCR-ABL gene mutation or amplification. Science 2001;293:876–880.

72. Schindler T, Bornmann W, Pellicena P, Miller WT, Clarkson B, Kuriyan J. Structural mecha-nism for STI-571 inhibition of abelson tyrosine kinase. Science 2000;289:1938–1942.

73. Hochhaus A, Kreil S, Corbin A, et al. Roots of clinical resistance to STI-571 cancer therapy.Science 2001;293:2163.

74. Barthe C, Cony-Makhoul P, Melo JV, Mahon JR. Roots of clinical resistance to STI-571 can-cer therapy. Science 2001;293:2163.

75. Thiesing JT, Ohno-Jones S, Kolibaba KS, Druker BJ. Efficacy of STI571, an abl tyrosinekinase inhibitor, in conjunction with other antileukemic agents against bcr-abl-positive cells.Blood 2000;96:3195–3199.

76. Fang G, Kim CN, Perkins CL, et al. CGP57148B (STI-571) induces differentiation andapoptosis and sensitizes Bcr-Abl-positive human leukemia cells to apoptosis due toantileukemic drugs. Blood 2000;96:2246–2253.

77. Hirota S, Isozaki K, Moriyama Y, et al. Gain-of-function mutations of c-kit in human gas-trointestinal stromal tumors. Science 1998;279:577–580.

78. Lux ML, Rubin BP, Biase TL, et al. KIT extracellular and kinase domain mutations in gas-trointestinal stromal tumors. Am J Pathol 2000;156:791–795.

79. Blanke CD, von Mehren M, Joensuu H, et al. Evaluation of the safety and efficacy of an oralmolecularly-targeted therapy, STI571, in patients with unresectable or metastatic gastrointestinalstromal tumors (GISTS) expressing c-KIT (CD117). Proc Am Soc Clin Oncol 2001;20:1a.

80. Van Oosterom AT, Judson I, Verweij J, et al. STI571, an active drug in metastatic gastrointestinalstromal tumors (GIST), an EORTC Phase I study. Proc Am Soc Clin Oncol 2001;20:1a.

81. Heinrich MC, Wait CL, Yee KWH, Griffith DJ. STI571 inhibits the kinase activity of wildtype and juxtamembrane c-kit mutants but not the exon 17 D816V mutation associated withmastocytosis. Blood 2000;96:173b.

82. Apperley JF, Schultheis B, Chase A, et al. Chronic myeloproliferative diseases with t(5;12) anda PDGFRB fusion gene: complete cytogenetic remissions on STI571. Blood 2001;98:726a.

83. Kilic T, Alberta JA, Zdunek PR, et al. Intracranial inhibition of platelet-derived growth fac-tor-mediated glioblastoma cell growth by an orally active kinase inhibitor of the 2-pheny-laminopyrimidine class. Cancer Res 2000;60:5143–5150.

84. Kolibaba KS, Druker BJ. Protein tyrosine kinases and cancer. Biochimica Biophysica Acta1997;1333:F217–248.

85. Ostman A, Heldin CH. Involvement of platelet-derived growth factor in disease: develop-ment of specific antagonists. Adv Cancer Res 2001;80:1–38.


1. INTRODUCTION

The expression of P-glycoprotein (Pgp) in patients with acute myeloidleukemia (AML) and acute lymphoblastic leukemia (ALL) is almost invariablyassociated with a poor prognosis. In multiple clinical series, Pgp has beenassociated with a poor response to chemotherapy, early relapse, decreasedremission duration, and shorter overall survival. Expression of Pgp is typicallyhigher in patients with more advanced disease, such as those who haverelapsed or progressed after initial response. Pgp shares significant structuralhomology with other efflux pumps comprising the large superfamily of adeno-sine triphosphate (ATP)-binding cassette (ABC) transporters. Immunologicdetection of the drug efflux pumps, Pgp, and multidrug resistance-associatedprotein 1 (MRP1) typically correlate with their respective functional drugresistance assays. Other markers of resistance described in AML include lung-resistance protein (LRP), bcl-2, and breast cancer resistance protein (BCRP),although their pathophysiology and clinical relevance is less clear and method-ology for their quantification are not as well standardized.

Preclinical studies have shown that small molecules capable of reversingefflux can restore drug sensitivity in resistant tumor models. Initial clinicalstudies were limited by both potency and specificity of the reverser, whereas

145


10 P-Glycoprotein Inhibition in Acute Myeloid Leukemia

Thomas R. Chauncey, MD, PhD

CONTENTS

INTRODUCTION

BACKGROUND

PROGNOSIS OF DRUG RESISTANCE PROFILES

CLINICAL TRIALS OF P-GLYCOPROTEIN REVERSAL

CONCLUSION

REFERENCES

later studies with more effective agents have, in many instances, been limitedby pharmacokinetic interactions exacerbating the clinical toxicities ofchemotherapy. Although one large randomized study has demonstrated aproven survival advantage without increased toxicity using cyclosporine,inconsistent results with other modulators raise doubt about the use and overallstrategy of using drug efflux blockers in patients with established Pgp overex-pression. Many of these patients have additional resistance mechanisms andachieving meaningful clinical responses will likely require more complex clin-ical strategies. Preventing or delaying development of drug resistance inchemosensitive patients remains another therapeutic strategy for evaluation.

2. BACKGROUND

Primary chemotherapy for patients with AML leads to high completeresponse rates in younger patients with significantly lower rates in older patientsas well as those with leukemia secondary to antecedent myelodysplastic syn-dromes (MDS) or previous chemoradiotherapy. For patients who achieve com-plete remission (CR), response durability is also significantly lower in olderpatients or those with a clinically apparent secondary onset (1–7). AML featuresin older patients and those with previous MDS include an increased incidence ofcytogenetic abnormalities associated with a poor prognosis, surface phenotypessuggestive of a more primitive etiology, and increased incidence of inherent bio-logic resistance. Of the cellular mechanisms of drug resistance described forpatients with acute leukemia, the best characterized resistance profile is the phe-notype of multidrug resistance (MDR) mediated by Pgp. Pgp expression consis-tently emerges as a clinically significant marker of resistance in patients withAML (8–15), although biologic resistance can be attributed to several differentmechanisms. Although the role of Pgp as a marker for resistance in adult ALLhas not been as rigorously evaluated, there is increasing evidence that Pgp resis-tance plays an important role in clinical outcome (16–18).

Pgp confers cross-resistance to a variety of mechanistically and structurallyunrelated cytotoxic drugs, such as anthracyclines, taxanes, Vinca alkaloids, andepipodophyllotoxins (8,9,19). All anthracyclines are subject to Pgp-mediatedresistance, despite evidence that idarubicin has greater cellular retention and isless susceptible to Pgp-mediated efflux (20,21). Pgp is a member of the ABCgene superfamily, which is conserved throughout species evolution. Commonto ABC transporters is a heterodimeric transmembrane glycoprotein complex,each with six transmembrane domains and an ATP binding site with conservedsequences known as the Walker A and Walker B domains (19,22). ABC trans-porters are involved in a range of transport functions from the epithelial trans-porter mutated in cystic fibrosis (CTFR), the canalicular multispecific organicanion (cMOAT) and canalicular bile acid (cBAT) transporters of liver, mono-

146 Chauncey

cyte secretion of IL-1β (ABC1), antigen processing in T-cells (TAP), choles-terol and phospholipid efflux (ABCA1), and non-Pgp-mediated anthracyclinetransport (BCRP), also known as the mitoxantrone resistance gene (MXR),ABCP, and ABCG2. Current ABC nomenclature has been reviewed (19,22,23)and is available on the Web (24). The phylogenetic conservation of ABC trans-porters suggests a broad role in protection from naturally occurring toxic com-pounds (xenobiotics). In mammalian species, the systemic elimination ofxenobiotics by the liver and kidney, the integrity of the blood–brain barrier, theisolation of germ cells from the systemic circulation, fetal isolation frommaternal circulation, and protection of the stem cell compartment are all medi-ated in part by ABC transporters.

To standardize assays, consensus recommendations have outlined criteria fordetermining Pgp in clinical samples (25), encouraging both immunophenotypicassessment as well as functional assays. Anthracyclines and, alternatively, fluo-rescent dyes such as rhodamine 123 (Rh123), Di(OC)2, or calcein are used todetermine functional efflux (11,12,26), calculated as the decrease in cellular flu-orescence compared with baseline assessment. Efflux measurements and pheno-typic determinations on different cell populations can be compared using theKolmogorov-Smirnov (KS) statistic. Expression profiling for resistance genes isan intriguing research tool (27) that may provide future clinical use.

Multidrug-resistance associated protein (MRP) is also a member of the ABCsuperfamily and is capable of efflux and intracellular sequestration in conjunc-tion with glutathione conjugation or cotransport (26,28). MRP describes agroup of transporters, of which MRP1 is the only member implicated in drugtransport. The substrate specificity of MRP1 is similar but more limited thanPgp, and its normal physiologic role may be detoxification of intracellular oxi-dants. It has been suggested that the location of MRP1 genes on chromosome16 may contribute to the favorable prognosis found in patients with AML withinv(16) abnormalities. MRP1 is typically assessed using flow cytometry. Func-tional assays using fluorescent dyes in the presence or absence of reversers,such as cyclosporine and probenecid (26), or after glutathione depletion (11),can specifically assess MRP-mediated efflux.

Lung-resistance protein was initially identified in a lung cancer cell line dur-ing in vitro selection for drug resistance (29–32). It has significant homologywith rodent vault proteins, which are subcellular organelles likely involved innuclear-cytoplasmic transport. Enforced expression of LRP in transfectionexperiments is not sufficient to confer resistance, suggesting that other cofac-tors or posttranslational assembly is necessary for biologic function (29). Ofinterest, LRP expression was increased in patients with relapsed AML afterresponse to induction therapy that included cyclosporine to overcome Pgpresistance (32), suggesting that modulating Pgp-mediated resistance may resultin selection or upregulation of LRP as a secondary resistance mechanism after

Chapter 10 / Pgp Inhibition in AML 147

Pgp reversal. LRP evaluation can be assessed with flow cytometry, immunocy-tochemistry, and reverse transcriptase-polymerase chain reaction (RT-PCR),although given the subcellular cytoplasmic localization, permeabilizing agentsmust be used with conventional flow cytometric techniques. Furthermore, post-translational regulation can make immunologic detection inconsistent. Thesetechniques have been compared on clinical samples and cell lines, and the rela-tive advantages and highlights of each have been described (33).

The BCRP/MXR/ABCP/ABCG2 transporter was initially described inbreast cancer cell lines and represents a “half-transporter,” consisting of sixtransmembrane domains and an ATP binding site. It is believed that the func-tional complex involves formation of a homodimer or heterodimeric partneringwith another subunit (19,34,35). BCRP is characterized by high affinity formitoxantrone and topoisomerase I inhibitors, as well as a resistance spectrumthat includes other anthracyclines but unlike Pgp does not include Vinca alka-loids or taxanes (19,36). BCRP has been assessed using RT-PCR, monoclonalantibodies, and selective efflux inhibitors, although optimal evaluation of clini-cal samples has not been determined.

With the availability of many small molecules capable of reversing Pgp,several clinical trials have been designed and executed to overcome this spe-cific resistance. Despite initial promise from encouraging preclinical data,many of these trials have been unsuccessful, attributable to toxicities arisingfrom the pharmacokinetic effects of Pgp modulation on the cytotoxic anti-cancer drugs.

3. PROGNOSIS OF DRUG RESISTANCE PROFILES

Studies from different investigators using both flow cytometry and func-tional efflux assays have shown that clinical specimens expressing Pgp doworse than patients who are Pgp negative (8–15,26), with a progressiveincrease in Pgp expression with advancing age and significant correlation ofPgp expression with decreasing remission rates and increasing incidence ofresistant disease (11,12,14). The increase in Pgp expression at relapse may alsoaccount for differences in daunorubicin sensitivity in patients with acutepromyelocytic leukemia (APL) at presentation compared with first relapse(37). Most studies confirm strong correlation between phenotypic Pgp expres-sion and functional efflux, although the Southwest Oncology Group (SWOG)has also shown a small but consistent blast population with discordancebetween functional and structural profiles, with some cells showing effluxresistant to cyclosporine and no phenotypic evidence of Pgp expression (11), aphenomena also suggested by other investigators (38,39). This profile may rep-resent expression of alternate transporters; however, the prognostic impact ofthis resistance profile is not yet characterized.

148 Chauncey

In the same SWOG studies, MRP1 was expressed at relatively low levelsand decreased with advancing age, although neither MRP1 nor LRP detectionprovided any clinically significant prognostic value (11). Although some stud-ies confirm MRP1’s lack of prognostic use (28,40), others have shown thatMRP1 detection is predictive of outcome and adds prognostic value to Pgp forboth CR rates, relapse-free survival, and overall survival for patients express-ing both phenotypes (26). These discrepancies may be attributable to the dif-ferent size populations tested or to methodologic differences with SWOGassessing MRP-specific efflux after glutathione depletion and others usingefflux inhibition by probenecid and alternate fluorescent substrates.

In several series, LRP expression in AML has been more predictive thanPgp for remission rates, resistant disease, and overall survival (20,32,40),although this has not been a consistent finding (9,11,41). These discrepantfindings may be attributable to variation in technique (33).

Studies evaluating BCRP in patients with AML show a range of expression,generally poor correlation with Pgp expression (38,42,43) and inconsistentprognostic value (43–46). Inasmuch as the resistance spectrum of anticancerdrugs for BCRP is similar to Pgp, this may represent other clinically signifi-cant transporters in patients with acute leukemia as well as other malignancies,and BCRP may account for some of the non-Pgp-cyclosporine-resistant effluxdescribed (11).

Many investigations have suggested that coexpression of drug resistancemarkers is significantly more predictive of clinical outcomes than expressionof a single mechanism (9,11,47–49). These discordant findings can again beattributed not only to assay technique but also to selection of populations stud-ied as well as the biologic heterogeneity of acute leukemia.

Patients with adult ALL have not been as well characterized but have beenshown to have a relatively high incidence of phenotypic Pgp expression(16–18) with variable prognostic implications. A recent study demonstratedhigh Pgp expression along with strong concordance of phenotypic Pgp withincreased functional efflux measured by Di(OC)2 (16). Of interest, this studyconfirmed the discrepant finding of immunologically undetectable Pgp in cellswith cyclosporine-reversible efflux, suggesting alternate membrane trans-porters as previously described in clinical AML samples.

4. CLINICAL TRIALS OF P-GLYCOPROTEIN REVERSAL

Encouraged by substantial preclinical evidence that small molecules canovercome drug resistance in cell culture, rodent xenografts, and transgenicmurine models, clinical trials have tested whether Pgp efflux reversers canimprove outcome in patients with AML with high Pgp expression. Initial trialsin several malignancies, using drugs such as verapamil and quinine showed


limited efficacy because toxicities from these compounds did not allow ade-quate serum levels to reverse Pgp. Confirming this unfavorable therapeuticindex, a prospective comparative trial of quinine with a mitoxantrone andcytarabine regimen in patients with high-risk AML, ALL, and transformedmyelodysplasia and myeloproliferative disorders failed to demonstrate a con-sistent clinical advantage, although encouraging trends in Pgp-expressingcases were observed (50). Similarly, attempts with quinine and verapamiltogether have also been unsuccessful in patients with non-Hodgkin’s lym-phoma (NHL) (51).

Because many Pgp blockers affect the pharmacokinetics of anticancer drugexcretion by the kidney and liver and in some instances compete for hepaticmetabolic pathways such as the P450 system, it has typically been necessary toreduce the chemotherapy dose during concurrent therapy with Pgp blockers toachieve a comparative regimen of equivalent clinical toxicity. These interac-tions may be further complicated by evidence that some Pgp substrates canactivate the hepatic and intestinal orphan nuclear receptor, SXR, which in turncan induce expression of both P450 enzymes and Pgp, thereby enhancing inac-tivation and excretion of the cytotoxic compound (52). To accommodate thesepharmacokinetic effects, many phase I and phase II trials were designed toestablish regimens equitoxic to similar regimens without the Pgp-revertingdrug (8,9,53–61) (Table 1).

Of the drugs initially tested, cyclosporine and later its analog, PSC833,showed the highest Pgp-reversing activity and the most promising therapeuticindices. Using cyclosporine, a phase I–II study in patients with high-risk AMLthat included relapsed and refractory disease demonstrated clinical efficacywith a response rate greater than expected based on historical reference, remis-sion “inversions” with durations of clinical remission longer than previousremissions, and absence or decrease of MDR1 expression in patients withleukemia at later relapse (61). Based on the encouraging results from this pilotstudy, a comparative phase III study in an identical population was developedand recently completed. The SWOG study used cyclosporine at high doses (16mg/kg/d) concurrent with daunorubicin by continuous infusion (45 mg/m2 × 3)after 5 d of high-dose ara-C (3 gm/M2/d), while the control arm received thesame regimen without cyclosporine. This study is significant because it showedadvantages for the addition of cyclosporine in complete response rate, decreasedincidence of resistant disease, and importantly improvement in both disease-free and overall survival that are durable beyond 2 yr (62). Although thedaunomycin dose was not reduced on the cyclosporine arm, toxicities wereequivalent, except for reversible hyperbilirubinemia in those patients receivingcyclosporine. Of note, an identical regimen in patients with myeloid blast crisisof chronic myelogenous leukemia (CML) did not show any benefit for theaddition of cyclosporine (63).

150 Chauncey

Results of the SWOG study contrast similar trials using cyclosporine inhigh-risk AML conducted by the Medical Research Council (MRC) (64) andEastern Cooperative Oncology Group (ECOG) (65) treated with dauno-mycin, etoposide, and cytarabine in conventional pulse doses, which showedno benefit from the addition of cyclosporine in adults (Table 2). Another sin-gle-arm study in children with high-risk AML by the Children’s CooperativeGroup/Pediatric Oncology Group (CCG/POG) (57) using mitoxantrone,etoposide, and cyclosporine showed no benefit compared with historicalcontrols.

Differences in trial design, study population, and dosing could account forthese conflicting results. The MRC study used a lower cyclosporine dose thatdoes not consistently lead to serum levels adequate for reversing Pgp, whereasthe MRC, ECOG, and CCG/POG studies also gave cytotoxic drugs by conven-tional pulse administration, leading to greater peak pharmacokinetic effectsand likely exacerbating clinical toxicity. The MRC, ECOG, and CCG/POGstudies also used etoposide, which is likely a poorer Pgp substrate (62,66) anda less effective antileukemic agent for primary AML therapy (2), yet one that issignificantly modulated by the pharmacokinetic effects of cyclosporine andPSC833. For primary AML therapy, etoposide probably contributes relativelymore to toxicity than efficacy, and this discrepancy is exacerbated by the phar-macokinetic effects of cyclosporine and PSC833.


Table 1Phase I–II Trials of Pgp Modulation in Acute Myeloid Leukemia

Institution/Author (Reference) Phase Size Indication Modulator Regimen

List (61) I–II 42 Rel/ref CSA HIDAC-D (civi)M.D. Anderson/Kornblau (60) I 110 Rel/ref PSC833 MECALGB 9420/Lee (59) I 110 ≥60 PSC833 ADEAdvani (58) II 37 Rel/ref PSC833 MECPOG 9222/Dahl (57) II 66 Rel/ref (≤20) CSA MESWOG 9617/Chauncey (56) I 31 >55 PSC833 MEVisani (55) I 23 >60 PSC833 DACALGB 9621/Kolitz (54) I 398 <60 PSC833 ADEDorr (53) I–II 43 Rel/ref PSC833 HIDAC-D (civi)SWOG 9918/Chauncey (79) II 16 >55 PSC833 D (civi) A

Rel/ref = relapsed/refractory; CSA = cyclosporine; HIDAC = high-dose cytarabine; ME =mitoxantrone, etoposide; ADE = cytarabine, daunorubicin, etoposide; MEC = mitoxantrone,etoposide, cytarabine; DA = daunorubicin, cytarabine; CALGB = Cancer and Leukemia GroupB; POG = Pediatric Oncology Group; SWOG = Southwest Oncology Group; civi = continuousintravenous infusion.

Preclinical investigations and small clinical trials offer rationale for daunoru-bicin administration by continuous infusion. Small studies evaluating intracellu-lar daunorubicin accumulation show greater blast retention when equivalentdoses are administered over a 24-h period compared with a 10-min infusion(67). Furthermore, in Pgp-expressing cell lines, in vitro cytotoxicity studies per-formed with Pgp sensitizers show that lower total anticancer drug exposures arerequired to induce equivalent cytotoxicity if given as an extended exposure(68–70). A recent ex vivo study showed increased daunorubicin blast retentionwhen administered by continuous infusion in the presence of the Pgp reverser,PSC833 (71). In the SWOG study (62), full-dose daunorubicin without doseattenuation in the cyclosporine arm was presumably achievable, because theexcess toxicities seen in other studies were more directly attributable to the peakpharmacodynamic effects of bolus infusions. Peak daunorubicin levels are bluntedwith the use of continuous infusion (53,67,72,73), leading to lower peak concen-trations albeit higher steady-state levels. In this study, higher steady-state levelsin the presence of cyclosporine appear critical to the clinical advantages seen,because increasing daunorubicin steady-state levels show strong correlationwith improvement in CR, relapse-free survival, and overall survival, only ifcyclosporine was administered. These preclinical and clinical studies suggestthat protracted exposure of daunorubicin in the presence of Pgp reverters may be

152 Chauncey

Table 2Comparative Trials of Pgp Modulation in Acute Myeloid Leukemia

Institution/Author (Ref) Size Indication Modulator Regimen Outcome

ECOG/Tallman (65) 38 Rel/ref CSA MEC No effectMRC AML-R/Liu Yin (64) 235 Rel/ref CSA ADE/HIDAC-DE No effectSWOG 9126/List (62) 226 Rel/ref CSA HIDAC-D (civi) Improved

RFS, OSSWOG 9032/List (63) 82 CML-BP CSA HIDAC-D (civi) No effectCALGB 9720/Baer (77) 126 ≥60 PSC833 DA Early closure

(toxicity)ECOG 2995/Greenberg (78) 127 Rel/ref PSC833 MEC Early closure

(efficacy)HOVON, MRC, C302 (9) 428 >60 PSC833 DA No effectC301 (9) 256 Rel/ref PSC833 MEC No effect

ECOG = Eastern Cooperative Oncology Group; rel/ref = relapsed/refractory; CSA =cyclosporine; MEC = mitoxantrone, etoposide, cytarabine; MRC = Medical Research Council;ADE = cytarabine, daunorubicin, etoposide; HIDAC = high-dose cytarabine; SWOG = South-west Oncology Group; RFS = relapse-free survival; OS = overall survival; CML-BP = chronicmyelogenous leukemia in blast phase; CALGB = Cancer and Leukemia Group B; DA =daunorubicin, cytarabine; HOVON = Stiching Haemato-Oncologie voor Volwassenen Nederland.

not only more efficacious but also less toxic. Furthermore, it is clear from otherinvestigations evaluating steady-state daunorubicin levels that not all patientsshow modulation of daunorubicin pharmacokinetics with cyclosporines, soattenuating daunorubicin doses to accommodate pharmacokinetic effects mayresult in lower drug exposures for many patients (53,60). Despite the increase intoxicities seen in entire cohorts without dose attenuation when cyclosporine iscombined with daunorubicin given by conventional bolus administration, somepatients will have lower serum steady-state levels than expected with fulldaunorubicin doses in the absence of cyclosporines. The results of SWOG 9126also raise issues about whether previous exposure to high-dose cytarabine beforecontinuous infusion daunorubicin with cyclosporine represents a critical phar-macologic interaction.

PSC833 (valspodar) is a cyclosporine analog with significantly more invitro Pgp-reverting activity than cyclosporine and a comparative lack ofimmunosuppressive and nephrotoxic effects. PSC833 retains cyclosporine’spharmacokinetic effects with inhibition of the metabolism and excretion ofanticancer drugs, such as anthracyclines and epipodophyllotoxins (74–76).Several phase I–II studies with PSC833 were performed in patients with AMLwith high Pgp expression (Table 1). For older patients with newly diagnosedAML as well as those with relapsed and refractory disease with concurrentPSC833, dose reduction of anticancer therapy was necessary to obtain regi-mens of equivalent toxicity (53–56,58–60). Dose reduction ratios were similarfrom study to study, with the absolute reduction dependent on the populationstudied and the relative dose intensity of each regimen tested.

Unfortunately, prospective phase III comparative studies of these regimenshave not shown improvement in clinical outcome. In older patients with newlydiagnosed AML, one study was stopped early due to excessive toxicity in thePSC833 group, despite attenuated doses of daunorubicin (77), whereas otherstudies evaluating patients with relapsed and refractory AML show no con-vincing efficacy for PSC833 (78). To mitigate the excessive toxicities seenwith PSC833 and daunomycin by conventional bolus administration, the strat-egy of daunorubicin by continuous infusion with PSC833 was investigated in aphase II study in older patients with AML, but this too was limited by signifi-cant regimen-related toxicity and closed before completing its targeted accrual(79). Patients in the SWOG study experienced the expected toxicities fromanthracycline dose escalation, including severe mucositis, pneumonia, and sep-sis. Reviewing all randomized PSC833 AML trials now complete, pharmacoki-netic effects have typically been excessive and/or dose attenuation has limitedpotential efficacy, despite occasional subset analysis showing trends towardimproved outcomes with PSC833 in patients with Pgp-expressing blasts (77).

The use of Pgp-reversing compounds with significant pharmacokineticeffects on concomitant anticancer therapy makes clinical trials both difficult to


perform and difficult to analyze. The successful SWOG study withcyclosporine used a schedule and consolidation schema distinct from otherregimens tested. Whether continuous infusion daunomycin with cyclosporinewas particularly advantageous, perhaps acting in synergy with high-dosecytarabine in the relapsed/refractory population, or whether cyclosporine andnot PSC833 affects other transporters or has other biologic effects leading toimproved outcome, are critical speculative issues without current resolution.Cyclosporine may have activity in a broader range of ABC transporters thanother current Pgp reversers and may also affect angiogenesis or activeparacrine growth factors (62). Current SWOG trials are planned to evaluate thesafety and efficacy of cyclosporine with continuous infusion daunorubicin inolder patients with AML.

Alternatively, Pgp reversers without pharmacokinetic effects have been devel-oped and are being tested in clinical trials (80,81). Zosuquidar (LY335979) is aneffective and selective Pgp inhibitor that is 500–1500 times more potent thancyclosporine on in vitro testing (80,82,83). In contrast to other Pgp reverterstested, it has negligible in vivo pharmacokinetic interactions with anthracy-clines, making it attractive for use in conjunction with daunorubicin in patientswith AML. In a phase II trial of patients that included relapsed and refractoryAML, newly diagnosed secondary AML and RAEB-t, a daunorubicin andcytarabine regimen, along with zosuquidar showed predominant dose-limitingtoxicities of reversible ataxia, confusion, and agitation. Of interest, reversibleataxia was also seen in PSC833 studies associated with high peak concentrations(53,55,84). The response rates, including 38% CR and 23% CR with incompleteplatelet recovery, in this high-risk population warrants further investigation. Exvivo studies confirm that leukemia samples exposed to in vivo zosuquidar showdecreased Rh123 efflux activity consistent with the drug’s anti-Pgp efficacy(82). Strategies to evaluate zosuquidar in newly diagnosed AML in olderpatients are being developed. Biricodar (VX-710) is another compound withoutapparent pharmacokinetic effects on doxorubicin when tested in solid tumors(81). It is a particularly attractive compound because it has demonstrated revers-ing activity for both Pgp- and MRP1-mediated transport.

Another potential strategy involves the use of gemtuzumab ozogamicin(Mylotarg; Wyeth Pharmaceuticals, Collegeville, PA) in conjunction with Pgpblockade. Gemtuzumab ozogamicin is a monoclonal antibody to the CD33epitope linked to calicheamicin, which is an extremely potent naturallyoccurring compound inducing deoxyribonucleic acid (DNA) cleavage. Bydesign, this agent directly targets CD33-expressing cells, with relative speci-ficity for AML and other myeloid progenitors. Response to gemtuzumabozogamicin has been shown to correlate both with expression and function ofPgp, as well as inducible apoptosis in ex vivo samples (85–87). Other evi-dence shows that cyclosporine can increase the proportion of leukemic speci-

154 Chauncey

mens undergoing apoptosis. Together, these results suggest that use of gem-tuzumab ozogamicin in conjunction with Pgp sensitizers will lead toincreased leukemia-directed cytotoxicity (86,88). Limitations to gemtuzumabozogamicin efficacy include an apparent increased hepatotoxicity, especiallyin patients who have previously undergone dose-intensive therapy with allo-geneic or autologous stem cell transplantation (89). Although this toxicitycan be significant in a minority of patients treated with gemtuzumab ozogam-icin, it remains unclear whether systemic Pgp blockade will exacerbate thisphenomena. It is intriguing to speculate that the somewhat limited therapeuticindex of single agent gemtuzumab ozogamicin may improve if used with con-current Pgp blockade and anecdotal reports indicate the potential tolerabilityand efficacy of this approach (89) with larger comparative trials currentlybeing developed.

Alternatively, targeting populations with established Pgp expression maynot be the most efficacious approach for any Pgp-reversing agent. Despite evi-dence that Pgp-mediated resistance emerges through clonal expansion of spon-taneous mutations (90), other data show a variety of cellular stressors,including chemotherapy, are capable of inducing MDR1 expression (91–95).Pgp expression may represent a marker for evolution or selection of othermechanisms of drug resistance (96), whether induced or selected by previouschemotherapy (97) or as an evolution of the leukemic process. Studies demon-strating the frequent occurrence and additive prognostic value of multipleresistance markers are in accord with this concept. Moreover, preclinical datasuggest that Pgp reversers can suppress emergence of resistance (90,98,99),whereas other studies show that Pgp expression can occur rapidly after expo-sure to cytotoxic therapy. One study demonstrated increased Pgp expressionand function within hours of ex vivo exposure to cytotoxic agents (94) andanother in patients with metastatic sarcoma showed evidence that in vivo Pgpgene expression is rapidly inducible within minutes of treatment (95). Thereare conflicting data on this issue, with a recent small series of patients withAML followed sequentially from diagnosis to relapse showing no Pgp expres-sion increase as assessed by phenotype, functional efflux, and clonal homozy-gosity for those patients with informative Pgp polymorphisms (100).Hopefully, these disparate results will be further investigated in larger clinicalseries. Using Pgp-blocking agents in patients without Pgp expression or otherresistance mechanisms is an alternate strategy that is currently being pursuedin clinical trials such as CALGB 19808 (54). However, because patients with-out Pgp expression have a better prognosis, determining the efficacy of thisapproach will require longer follow-up and must account for the distinct prog-nostic categories of better risk patients.

New insight into Pgp function suggests roles in cellular physiology beyondxenobiotic efflux. The Pgp apparatus may serve broad anti-apoptotic functions


that interfere with caspase-dependent apoptosis, perhaps in concert with mem-brane signal transduction (101,102). In many tissues, ABC transporters play acentral role in lipid transport as well as membrane sphingomyelin polarity andcontent, the latter known to affect caspase-dependent apoptotic signaling. Evi-dence from both cell lines and ex vivo leukemia samples (103) suggests thatthe anti-apoptotic effects of Pgp may be mediated through the sphingomyelin-ceramide pathway, an effect reversible with exogenous sphingomyelin given invitro. Pgp can lower membrane sphingomyelin content, making less availablefor ceramide production, and can also enhance sphingomyelin redistribution,thereby limiting TNF-α-driven apoptosis, an effect antagonized by PSC833(104). Pgp can inhibit caspase-3-dependent apoptosis initiated by stimuli suchas Fas ligation or ultraviolet irradiation, an effect blocked by the Pgp reverser,verapamil (105), whereas PSC833 has been shown to augment the apoptosisinduced by serum and growth factor withdrawal (103). Although blocking Pgpwith currently available pharmaceuticals may not achieve the clinicalresponses eagerly anticipated in clinical trials, future trial design using drugcombinations effecting caspase-independent pathways may be enhanced byconcurrent Pgp blockade. With a better understanding of Pgp’s role in cellularapoptotic processes, therapeutic strategies can be designed to accommodatethese complex interrelationships.

5. CONCLUSION

Of the better characterized markers of drug resistance in AML, phenotypicand functional Pgp expression predicts outcome more consistently than LRPand MRP1, although the prognostic utility of these and other resistance mark-ers will improve and change with larger studies and more consistent methodol-ogy. Although preclinical experiments have shown that some resistantphenotypes can be overcome with reversing agents, biologic resistance is acomplex interaction of multiple cellular alterations, with mutations leading todrug resistance also conferring other biologic advantages. As such, overcom-ing clinical resistance in leukemia therapy may require not only targeted resis-tance modifiers but also a more complete biologic and clinical understandingof the leukemic process.

REFERENCES1. Estey EH. Therapeutic options for acute myelogenous leukemia. Cancer 2001;92:1059–1073.2. Goldstone AH, Burnett AK, Wheatley K, et al. Attempts to improve treatment outcomes in

acute myeloid leukemia (AML) in older patients: the results of the United Kingdom Med-ical Research Council AML11 trial. Blood 2001;98:1302–1311.

3. Hiddeman W, Kern W, Schoch C, et al. Management of acute myeloid leukemia in elderlypatients. J Clin Oncol 1999;17:3569–3576.

156 Chauncey

4. Lowenberg B, Downing JR, Burnett A. Acute myeloid leukemia. N Engl J Med1999;341:1051–1062.

5. Wheatley K, Burnett AK, Goldstone AH, et al. A simple, robust, validated and highly pre-dictive index for the determination of risk-directed therapy in acute myeloid leukaemiaderived from the MRC AML 10 trial. Brit J Haematol 1999;107:69–79.

6. Bloomfield CD, Lawrence D, Byrd JC, et al. Frequency of prolonged remission durationafter high-dose cytarabine intensification in acute myeloid leukemia varies by cytogeneticsubtype. Cancer Res 1998;58:4173–4179.

7. Mayer RJ, Davis RB, Schiffer CA, et al. Intensive postremission chemotherapy in adultswith acute myeloid leukemia. N Engl J Med 1994;331:896–903.

8. Chauncey TR. Drug resistance mechanisms in acute leukemia. Curr Opin Oncol2001;13:21–26.

9. Sonneveld P, List AF. Chemotherapy resistance in acute myeloid leukaemia. Best Pract ResClin Haematol 2001;14:211–233.

10. van der Kolk DM, de Vries EG, van Putten WJ, et al. P-glycoprotein and multidrug resis-tance protein activities in relation to treatment outcome in acute myeloid leukemia. ClinCancer Res 2000;6:3205–3214.

11. Leith CP, Kopecky KJ, Chen I-M, et al. Frequency and clinical significance of the expres-sion of the multidrug resistance proteins MDR1/P-glycoprotein, MRP1, and LRP in acutemyeloid leukemia. Blood 1999;94:1086–1099.

12. Leith CP, Kopecky KJ, Godwin J, et al. Acute myeloid leukemia in the elderly: assessmentof multidrug resistance (MDR1) and cytogenetics distinguishes biologic subgroups withremarkably distinct responses to standard chemotherapy. A Southwest Oncology Groupstudy. Blood 1997;89:3323–3329.

13. van den Heuvel-Eibrink MM, van der Holt B, te Boekhorst PA, et al. MDR 1 expression isan independent prognostic factor for response and survival in de novo acute myeloidleukaemia. B J Haematol 1997;99:76–83.

14. Willman CL. The prognostic significance of the expression and function of multidrug resis-tance transporter proteins in acute myeloid leukemia: studies of the Southwest OncologyGroup Leukemia Research Program. Semin Hematol 1997;34(4 suppl 5):25–33.

15. Campos L, Guyotat D, Archimbaund E, et al. Clinical significance of multidrug resistanceP-glycoprotein expression on acute nonlymphoblastic leukemia cells at diagnosis. Blood1992;79:473–476.

16. Gazitt Y, Lee D-Y, Wang ME, Boldt DH. Functional MDR1 expression in adult acute lym-phoblastic leukemia (ALL) [abstract]. Blood 1998;92(suppl 1):676a.

17. Wattel E, Lepelley P, Merlat A, et al. Expression of the multidrug resistance P glycoproteinin newly diagnosed adult acute lymphoblastic leukemia: absence of correlation withresponse to treatment. Leukemia 1995;9:1870–1874.

18. Savignano C, Geromin A, Michieli M, et al. The expression of the multidrug resistance relatedglycoprotein in adult acute lymphoblastic leukemia. Haematologica 1993;78:261–263.

19. Litman T, Druley TE, Stein WD, Bates SE. From MDR to MXR: new understanding ofmultidrug resistance systems, their properties and clinical significance. Cell Mol Life Sci2001;58:931–959.

20. Schroder JK, Kasimir-Bauer S, Seeber S, Scheulen ME. In vitro effect of multidrug resis-tance modifiers on idarubicinol efflux in blasts of acute myeloid leukemia. J Cancer ResClin Oncol 2000;126:111–116.

21. Fukushima T, Yamashita T, Yoshio N, et al. Effect of PSC 833 on the cytotoxicity of idaru-bicin and idarubicinol in multidrug-resistant K562 cells. Leukemia Res 1999;23:37–42.

22. Klein I, Sarkadi B, Varadi A. An inventory of the human ABC proteins. Biochim BiophysActa 1999;1461:237–262.


23. Muller M, Jansen PLM. Molecular aspects of hepatobiliary transport. Am J Physiol1997;272:G1285–G1303.

24. Available at: http://www.med.rug.nl/mdl/humanabc.htm. 25. Beck WT, Grogan TM, Willman CL, et al. Methods to detect P-glycoprotein-associated

multidrug resistance in patients’ tumors: consensus recommendations. Cancer Res1996;56:3010–3020.

26. Legrand O, Simonin G, Beauchamp-Nicqud A, Zittoun R, Marie J. Simultaneous activity ofMRP1 and Pgp is correlated with in vitro resistance to daunorubicin and with in vivo resis-tance in adult acute myeloid leukemia. Blood 1999;94:1046–1056.

27. Kudoh K, Ramanna M, Ravatn R, et al. Monitoring the expression profiles of doxorubicin-induced and doxorubicin-resistant cancer cells by cDNA microarray. Cancer Res2000;60:4161–4166.

28. Filipits M, Suchomel RW, Zöchbauer S. Multidrug resistance-associated protein in acutemyeloid leukemia: no impact on treatment outcome. Clin Cancer Res 1997;3:1419–1425.

29. Dalton WS, Scheper RJ. Lung resistance-related protein: determining its role in multidrugresistance [editorial]. J Natl Cancer Inst 1999;91:1604–1605.

30. Filipits M, Pohl G, Stranzl T, et al. Expression of the lung resistance protein predicts pooroutcome in de novo acute myeloid leukemia. Blood 1998;5:1508–1513.

31. Kickhoefer VA, Rajavel KS, Scheffer GL, Dalton WS, Scheper RJ, Rome LH. Vaults areup-regulated in multidrug-resistant cancer cell lines. J Biol Chem 1998;273:8971–8974.

32. List AF, Spier CS, Grogan TM, et al. Overexpression of the major vault transporter proteinlung-resistance protein predicts treatment outcome in acute myeloid leukemia. Blood1996;87:2464–2469.

33. Legrand O, Simonin G, Zittoun R, Marie J-P. Lung resistance protein (LRP) gene expres-sion in adult acute myeloid leukemia: a critical evaluation by three techniques. Leukemia1998;12:1367–1374.

34. Ross DD, Yang W, Abruzzo LV, et al. Atypical multidrug resistance: breast cancer resis-tance protein messenger RNA expression in mitoxantrone-selected cell lines. J Natl CancerInst 1999;91:429–433.

35. Doyle LA, Yang W, Abruzzo LV, et al. A multidrug resistance transporter from humanMCF-7 breast cancer cells. Proc Natl Acad Sci USA 1998;95:15,665–15,670.

36. Honjo Y, Hryeyna CA, Yan Q-W, et al. Acquired mutations in the MXR/BCRP/ABCP genealter substrate specificity in MXR/BCRP/ABCP-overexpressing cells. Cancer Res2001;61:6635–6639.

37. Michieli M, Damiani D, Ermacora A, Geromin A, Michelutti A, Masolini P. P-glycoprotein(PGP), lung resistance-related protein (LRP) and multidrug resistance-associated protein(MRP) expression in acute promyelocytic leukaemia. Br J Haematol 2000;108:703–709.

38. Ross DD. Novel mechanisms of drug resistance in leukemia. Leukemia 2000;14:467–473.39. Merlin JL, Guerci A-P, Marchal S, et al. Influence of SDZ-PSC833 on daunorubicin intra-

cellular accumulation in bone marrow specimens from patients with acute myeloidleukemia. Br J Haematol 1998;103:480–487.

40. Filipits M, Stranzl T, Pohl G, Heinzl H, Jager U, Geissler K. Drug resistance factors inacute myeloid leukemia: a comparative analysis. Leukemia 2000;14:68–76.

41. Pallis M, Turzanski J, Harrison G, et al. Use of standardized flow cytometric determinantsof multidrug resistance to analyse response to remission induction chemotherapy inpatients with acute myeloblastic leukaemia. Br J Haematol 1999;104:307–312.

42. Ross DD, Karp JE, Chen TT, Doyle LA. Expression of breast cancer resistance protein inblast cells from patients with acute leukemia. Blood 2000;96:365–368.

43. van den Heuvel-Eibrink MM, Wiemer EA, Prins A, et al. Increased expression of the breastcancer resistance protein (BCRP) in relapsed or refractory acute myeloid leukemia (AML).Leukemia 2002;16:833–839.

158 Chauncey

44. Abbott BL, Colapietro AM, Barnes Y, et al. Low levels of ABCG2 expression in adult AMLblast samples. Blood 2002;15(100):4594–4601.

45. Suvannasankha A, Minderman H, O’Loughlin KL, et al. Breast cancer resistance protein(BCRP) expression in acute myeloid leukemia (AML) [abstract]. Blood 2001;98(suppl1):307a.

46. van der Kolk DM, Vellenga E, Scheffer GL, et al. Expression and activity of breast cancerresistance protein (BCRP) in de novo and relapsed acute myeloid leukemia. Blood2002;99:3763–3770.

47. Borg AG, Burgess R, Green LM, Scheper RJ, Liu Yin JA. P-glycoprotein and multidrug resis-tance-associated protein, but not lung resistance protein, lower the intracellular daunorubicinaccumulation in acute myeloid leukaemic cells. Br J Haematol 2000;108:48–54.

48. Kasimir-Bauer S, Ottinger H, Meusers P, et al. In acute myeloid leukemia, coexpression ofat least two proteins including P-glycoprotein, the multidrug resistance-related protein bcl-2, mutant p53, and heat-shock protein 27, is predictive of the response to inductionchemotherapy. Exp Hematol 1998;26:1111–1117.

49. Goasguen JE, Lamy T, Bergeron C, et al. Multifactorial drug-resistance phenomenon in acuteleukemias: impact of P170-MDR1, LRP56 protein, glutathione-transferases and metalloth-ionein systems on clinical outcome. Leukemia Lymphoma 1996;23:567–576.

50. Solary E, Witz B, Caillot D, et al. Combination of quinine as a potential reversing agentwith mitoxantrone and cytarabine for the treatment of acute leukemias: a randomized mul-ticenter study. Blood 1996;88:1198–1205.

51. Gaynor ER, Unger JM, Miller TP, et al. Infusional CHOP chemotherapy (CVAD) with orwithout chemosensitizers offers no advantage over standard CHOP therapy in the treatmentof lymphoma: a Southwest Oncology Group study. J Clin Oncol 2001;19:750–755.

52. Synold TW, Dussault I, Forman BM. The orphan nuclear receptor SXR coordinately regu-lates drug metabolism and efflux. Nat Med 2001;7:584–590.

53. Dorr R, Karanes C, Spier C, et al. Phase I/II study of the P-glycoprotein modulator PSC833 in patients with acute myeloid leukemia. J Clin Oncol 2001;19:1589–1599.

54. Kolitz JE, George SL, Dodge RK, et al. Dose escalation studies of ara-C (A), daunorubicin(D) and etoposide (E) with and without multidrug resistance (MDR) modulation with PSC-833 (P) in untreated adult patients with acute myeloid leukemia (AML) < 60 years: finalinduction results of CALGB 9621 [abstract]. Blood 2001;98(suppl 1):461a.

55. Visani G, Milligan D, Leoni F, et al. Combined action of PSC 833 (Valspodar), a novelMDR reversing agent, with mitoxantrone, etoposide and cytarabine in poor-prognosis acutemyeloid leukemia. Leukemia 2001;15:764–771.

56. Chauncey TR, Rankin C, Anderson JE, et al. A phase I study of induction chemotherapy forolder patients with newly diagnosed acute myeloid leukemia (AML) using mitoxantrone,etoposide, and the MDR modulator PSC 833: a Southwest Oncology Group Study (9617).Leukemia Res 2000;24:567–574.

57. Dahl GV, Lacayo NJ, Brophy N, et al. Mitoxantrone, etoposide, and cyclosporine therapy inpediatric patients with recurrent or refractory acute myeloid leukemia. J Clin Oncol2000;18:1867–1875.

58. Advani R, Saba HI, Tallman MS, et al. Treatment of refractory and relapsed acute myeloge-nous leukemia with combination chemotherapy plus the multidrug resistance modulatorPSC833 (valspodar). Blood 1999;93:787–795.

59. Lee EJ, George SL, Caligiuri M, et al. Parallel phase I studies of daunomycin given withcytarabine and etoposide with or without the multidrug resistance modulator PSC-833 inpreviously untreated patients 60 years of age or older with acute myelogenous leukemia:results of Cancer and Leukemia Group B Study 9420. J Clin Oncol 1999;17:2831–2839.

60. Kornblau SM, Estey E, Madden T, et al. Phase I study of mitoxantrone plus etoposide withmultidrug blockade by SDZ PSC-833 in relapsed or refractory acute myelogenousleukemia. J Clin Oncol 1997;15:1796–1802.


61. List AF, Spier CM, Greer J, et al. Phase I/II trial of cyclosporine as a chemotherapy-resis-tance modifier in acute leukemia. J Clin Oncol 1993;11:1652–1660.

62. List AF, Kopecky KJ, Willman CL, et al. Benefit of cyclosporine modulation of drug resis-tance in patients with poor-risk acute myeloid leukemia: a Southwest Oncology Groupstudy. Blood 2001;98:3212–3220.

63. List AF, Kopecky KJ, Willman CL, et al. Cyclosporine inhibition of P-glycoprotein inchronic myeloid leukemia blast phase. Blood 2002;100:1910–1912.

64. Liu Yin JA, Wheatley K, Rees JKH, et al. Comparison of “sequential” versus “standard”chemotherapy as re-induction treatment, with or without cyclosporine, inrefractory/relapsed acute myeloid leukaemia (AML): results of the UK Medical ResearchCouncil AML-R trial. Br J Haematol 2001;113:713–726.

65. Tallman MS, Lee E, Sikic BI, et al. Mitoxantrone, etoposide and cytarabine pluscyclosporine in patients with relapsed or refractory acute myeloid leukemia: an EasternCooperative Oncology Group pilot study. Cancer 1999;85:358–367.

66. Politi PM, Arnold ST, Felsted RL, Sinha BK. P-glycoprotein-independent mechanism ofresistance to VP-16 in multidrug-resistance tumor cell lines: pharmacokinetic and pho-toaffinity labeling studies. Mol Pharmacol 1990;37:790–796.

67. Paul C, Tidefelt U, Liliemark J, Peterson C. Increasing the accumulation of daunorubicin inhuman leukemic cells by prolonging the infusion time. Leukemia Res 1989;13:191–196.

68. Toffoli G, Turniotto L, Gigante M, et al. Doxorubicin, vincristine, and actinomycin-D, butnot teniposide, require long-lasting uninterrupted verapamil pressure to overcome drugresistance in multidrug-resistant cells. Cancer Detect Prev 1993;17:425–432.

69. Slate D, Michelson S. Drug resistance-reversal strategies: comparison of experimental datawith model predictions. J Natl Cancer Inst 1991;83:1574–1580.

70. Cass CE, Janowska-Wieczorek A, Lynch MA, et al. Effect of duration of exposure to vera-pamil on vincristine activity against multidrug-resistant human leukemic cell line. CancerRes 1989;49:5798–5804.

71. Tidefelt U, Liliemark J, Gruber A, et al. P-Glycoprotein inhibitor valspodar (PSC 833)increases the intracellular concentrations of daunorubicin in vivo in patients with P-glyco-protein-positive acute myeloid leukemia. J Clin Oncol 2000;18:1837–1844.

72. Lewis JP, Meyers FJ, Tanaka L. Daunomycin administered by continuous intravenous infu-sion is effective in the treatment of acute nonlymphocytic leukaemia. Br J Haematol1985;61:261–265.

73. DeGregorio MW, Carrera CJ, Klock JC, et al. Cellular and plasma kinetics of daunorubicingiven by two methods of administration in a patient with acute leukemia. Cancer Treat Rep1982;66:2085–2088.

74. Friche E, Jensen PB, Nissen NI. Comparison of cyclosporine A and SDZ PSC833 as mul-tidrug-resistance modulators in a daunorubicin-resistant Ehrlich ascites tumor. CancerChemother Pharmacol 1992;30:235–237.

75. Boesch D, Gaveriaux C, Jachez B, et al. In vivo circumvention of P-glycoprotein-mediatedmultidrug resistance of tumor cells with SDZ PSC 833. Cancer Res 1991;51:4226–4233.

76. Twentyman PR, Bleehen NM. Resistance modification by PSC-833, a novel non-immuno-suppressive cyclosporin A. Eur J Cancer 1991;27:1639–1642.

77. Baer MR, George SL, Dodge RK, et al. Phase 3 study of the multidrug resistance modula-tor PSC-833 in previously untreated patients 60 years of age and older with acute myeloidleukemia: Cancer and Leukemia Group B Study 9720. Blood 2002;100:1224–1232.

78. Greenberg P, Advani R, Tallman M, et al. Treatment of refractory/relapsed AML withPSC833 plus mitoxantrone, etoposide, cytarabine (PSC-MEC) vs MEC: randomized phaseIII trial (E2995) [abstract]. Blood 1999;94(suppl 1):383a.

79. Chauncey TR, Gundacker H, List AF, et al. personal communication

160 Chauncey

80. Dantzig AH, Shepard RL, Cao J, et al. Reversal of P-glycoprotein-mediated multidrugresistance by a potent cyclopropyldibenzosuberane modulator, LY335979. Cancer Res1996;56:4171–4179.

81. Peck RA, Hewett J, Harding MW, et al. Phase I and pharmacokinetic study of the novelMDR1 and MRP1 inhibitor Biricodar administered alone and in combination with doxoru-bicin. J Clin Oncol 2001;19:3130–3141.

82. Gerrard G, Ganeshaguru K, Baker R, et al. A phase I study of a Pgp inhibitor Zosuquidar(LY 335979), given by short intravenous infusion in combination with daunorubicin andcytarabine in AML/MDS patients [abstract]. Blood 2001;98(suppl 1):177b.

83. Cripe LD, Tallman M, Karanes C, et al. A phase II trial of Zosuquidar (LY335979), a mod-ulator of P-glycoprotein (Pgp) activity, plus daunorubicin and high-dose cytarabine inpatients with newly diagnosed secondary acute myeloid leukemia (AML), a refractory ane-mia with excess blasts in transformation (RAEB-t) or relapsed/refractory AML [abstract].Blood 2001;98(suppl 1):595a.

84. Patnaik A, Warner E, Michael M, et al. Phase I dose-finding and pharmacokinetic study ofpaclitaxel and carboplatin with oral valspodar in patients with advanced solid tumors. JClin Oncol 2000;18:3677–3689.

85. Bross PF, Beitz J, Chen G, et al. Approval summary: gemtuzumab ozogamicin in relapsedacute myeloid leukemia. Clin Cancer Res 2001;7:1490–1496.

86. Naito K, Takeshita A, Shigeno K, et al. Calicheamicin-conjugated humanized anti-CD33monoclonal antibody (gemtuzumab ozogamicin, CMA-676) shows cytocidal effect onCD33-positive leukemia cell lines, but is inactive on P-glycoprotein-expressing sublines.Leukemia 2000;14:1436–1443.

87. Sievers EL, Appelbaum FR, Spielberger RT, et al. Selective ablation of acute myeloidleukemia using antibody-targeted chemotherapy: a phase I study of an anti-CD33calicheamicin immunoconjugate. Blood 1999;93:3678–3684.

88. Linenberger ML, Hong T, Flowers D, et al. Multidrug-resistance phenotype and clinicalresponses to gemtuzumab ozogamicin. Blood 2001;98:988–994.

89. Giles FJ, Kantarjian HM, Kornblau SM, et al. Mylotarg (gemtuzumab ozogamicin) therapyis associated with hepatic venoocclusive disease in patients who have not received stem celltransplantation. Cancer 2001;92:406–413.

90. Beketic-Oreskovic L, Durán G, Chen G, Dumontet C, Sikic B. Decreased mutation rate forcellular resistance to doxorubicin and suppression of mdr1 gene activation by thecyclosporin PSC 833. J Natl Cancer Inst 1995;87:1593–1600.

91. Chaudhary PM, Roninson IB. Induction of multidrug resistance in human cells by transientexposure to different chemotherapeutic drugs. J Natl Cancer Inst 1993;85:632–639.

92. Chin KV, Tanaka S, Darlington G, Pastan I, Gottesman MM. Heat shock and arseniteincrease expression of the multidrug resistance (MDR1) gene in human renal carcinomacells. J Biol Chem 1990;265:221–226.

93. Mickley LA, Bates SE, Richert ND, et al. Modulation of the expression of a multidrugresistance gene (mdr-1/P-glycoprotein) by differentiating agents. J Biol Chem1989;264:18,031–18,040.

94. Hu XF, Slater A, Kantharidis P, et al. Altered multidrug resistance phenotype caused by anthra-cycline analogues and cytosine arabinoside in myeloid leukemia. Blood 1999;93:4086–4095.

95. Abolhoda A, Wilson AE, Ross H, Danenberg PV, Burt M, Scotto KW. Rapid activation ofMDR1 gene expression in human metastatic sarcoma after in vivo exposure to doxorubicin.Clin Cancer Res 1999;5:3352–3356.

96. Broxterman HJ, Sonneveld P, Pieters R, et al. Do P-glycoprotein and major vault protein(MVP/LRP) expression correlate with in vitro daunorubicin resistance in acute myeloidleukemia? Leukemia 1999;13:258–265.


97. Nakayuma M, Wada M, Harada T, et al. Hypomethylation status of CpG sites at the pro-moter region and overexpression of the human MDR1 gene in acute myeloid leukemias.Blood 1998;92:4296–4307.

98. Abbaszadegan MR, Foley NE, Gleason-Guzman MC, Dalton WS. Resistance to thechemosensitizer verapamil in a multi-drug-resistant (MDR) human multiple myeloma cellline. Int J Cancer 1996;66:506–514.

99. Futscher BW, Foley NE, Gleason-Guzman MC, Meltzer PS, Sullivan DM, Dalton WS. Ver-apamil suppresses the emergence of P-glycoprotein-mediated multi-drug resistance. Int JCancer 1996;66:520–525.

100. van den Heuvel-Eibrink MM, Wiemer EA, de Boevere MJ, et al. MDR1 gene-related clonalselection and P-glycoprotein function and expression in relapsed or refractory acutemyeloid leukemia. Blood 2001;97:3605–3611.

101. Johnstone RW, Cretnewy E, Smyth MJ. P-glycoprotein protects leukemia cells against cas-pase-dependent, but not caspase-independent, cell death. Blood 1999;93:1075–1085.

102. Johnstone RW, Ruefli AA, Tainton KM, Smyth MJ. A role for P-glycoprotein in regulatingcell death. Leukemia Lymphoma 2000;38:1–11.

103. Pallis M, Russell N. P-glycoprotein plays a drug-efflux-independent role in augmentingcell survival in acute myeloblastic leukemia and is associated with modulation of a sphin-gomyelin-ceramide apoptotic pathway. Blood 2000;95:2897–2904.

104. Bezombes C, Maestre N, Laurent G, et al. Restoration of TNF-alpha-induced ceramidegeneration and apoptosis in resistant human leukemia KG1a cells by the P-glycoproteinblocker PSC833. FASEB J 1998;12:101–109.

105. Ruefli AA, Smyth MJ, Johnstone RW. HMBA induces activation of a caspase-independentcell death pathway to overcome P-glycoprotein-mediated multidrug resistance. Blood2000;95:2378–2385.

162 Chauncey

1. INTRODUCTION

Apoptosis is a morphologically and biochemically distinctive cell deathprocess that occurs under several physiologic and pathologic conditions (1,2).Morphologically, apoptosis is characterized by plasma membrane blebbing fol-lowed by chromatin condensation and disassembly of the cell into multiplemembrane-enclosed fragments, which are then engulfed by neighboring cells orprofessional phagocytes (2). These biochemical changes are believed to reflect,at least in part, the action of caspases, unique intracellular proteases that digestcritical polypeptides required for cellular integrity and survival (3,4).

Studies performed since the late 1980s, have demonstrated that virtually allof the agents currently used to treat acute leukemia induce apoptosis in suscep-tible cells (5,6). Conversely, there has also been a growing recognition that theprocess of leukemogenesis might involve changes that inhibit apoptosis (6,7).

163


11 Targeting the Apoptotic Machineryas a Potential Antileukemic Strategy

Benjamin M. F. Mow, MD

and Scott H. Kaufmann, MD, PHD

CONTENTS

INTRODUCTION

MULTIPLE ROLES FOR CASPASES DURING APOPTOSIS

REGULATION OF APOPTOTIC PATHWAYS

ALTERED APOPTOTIC PATHWAYS IN LEUKEMIA

INDUCTION OF APOPTOSIS BY ANTILEUKEMIC THERAPY

NEW AGENTS THAT DIRECTLY TARGET APOPTOTIC PATHWAYS

OR THEIR REGULATION

SUMMARY

REFERENCES

In the following sections, the current understanding of apoptotic pathways isoutlined, some of the ways in which these pathways are altered during theprocess of leukemogenesis are described, and the current status of agentsdesigned to reverse some of these alterations is reviewed.

2. MULTIPLE ROLES FOR CASPASES DURING APOPTOSIS

As mentioned above, many of the biochemical and morphological featuresof apoptotic cells reflect the selective proteolytic cleavage of a subset of cellu-lar polypeptides (3,8). For example, cleavage of the inhibitor of caspase-acti-vated deoxyribonuclease (ICAD) liberates CAD (9), an endonuclease thatcontributes to the internucleosomal deoxyribonucleic acid (DNA) degradationand chromatin condensation observed in apoptotic cells (10). Likewise, cleav-age of lamins, filamentous polypeptides that form a structural meshwork insidethe inner nuclear membrane, facilitates apoptotic nuclear fragmentation (11).

These cleavages result from the action of caspases, intracellular cysteine pro-teases that cleave next to aspartate residues (3,4). Of the 11 known human cas-pases, 5 (caspases 3, 6, 8, 9, and 10) have increasingly well-defined roles inapoptosis. Each is synthesized as a relatively inactive precursor containing aprodomain, a large subunit, and a small subunit. Activation of these zymogenscommonly involves proteolytic cleavage at multiple aspartate residues, includingone between the large and small subunits and another between the prodomain andthe large subunit. After a series of conformational changes, the activated enzymesultimately consist of two large and two small subunits (12). The fact that activa-tion usually involves digestion at potential caspase cleavage sites opens the possi-bility that caspases participate in intracellular proteolytic cascades (4).

Like other enzyme cascades, the caspase proteolytic cascades consist ofupstream and downstream participants with discrete roles. Caspases 3 and 6,the major downstream or “effector” caspases, are responsible for most of thecleavages that disassemble the cell. Caspases 8, 9, and 10, the major upstreamor “initiator” caspases, are responsible for transducing various signals into pro-teolytic activity. These initiator caspases are activated as a consequence of sig-naling through at least two distinct pathways (Fig. 1).

2.1. The Death Receptor PathwaySignaling through the death receptor (DR) pathway (also called the extrinsic

pathway) begins when certain death-inducing cytokines bind to their specializedcell surface receptors (13,14). For example, Fas ligand (FasL) expressed on thesurface of cytotoxic lymphocytes binds to Fas (also known as CD95 or Apol-1), areceptor on the surface of target cells. As a consequence of this ligand-receptorinteraction, the cytoplasmic domain of Fas undergoes an apparent conformationalchange and then binds the adaptor molecule FADD (Fas-associated protein with

164 Mow and Kaufmann

death domain). After interacting with Fas, FADD not only oligomerizes (15) butalso binds procaspases 8 and 10, which have low but detectable enzymatic activity(16–18). These interactions cause juxtaposition of multiple zymogens and result inapparent autocatalytic liberation of active caspases 8 and 10 (16–18). Other DRssignal similarly, although the adaptor molecules differ in some cases (13).

In some lymphoid cells (so-called “type I cells”), the caspase 8 and/or 10generated by DR ligation directly activates procaspase 3 (19). In other cells(“type II cells”), however, caspase 8 or 10 is insufficient to activate procaspase-3. Instead, the cytoplasmic protein Bid is cleaved to produce a fragment thatactivates the mitochondrial pathway (4).

2.2. The Mitochondrial PathwayThe mitochondrial pathway (also called the intrinsic pathway) is activated

by the selective release of a subset of mitochondrial polypeptides into thecytoplasm (4). Several models have been advanced to explain the altered local-ization of these polypeptides (20,21). Some research stresses the importanceof a phenomenon known as mitochondrial permeability transition, whichinvolves the opening of a pore in the mitochondrial membranes (22), whereasother research focuses on the insertion of two polypeptides, Bax and Bak, inthe outer mitochondrial membrane, where they oligomerize to form pores thatmediate cytochrome c release (23–26). Although Bax and Bak are constitu-tively expressed in many cells, these polypeptides are ordinarily found in thecytoplasm or loosely associated with mitochondria. Recent evidence (27) sug-gests that their insertion into the mitochondrial outer membrane is induced by“BH3-only” polypeptides, a group of small proapoptotic polypeptides thatincludes the Bid fragment generated by caspase-8 during DR signaling in typeII cells (25,26); Bad, which is dephosphorylated and activated upon with-drawal of interleukin-induced survival signals (28); Bim, which is displacedfrom microtubules in response to microtubule-directed drugs (29); andPUMA, which is synthesized in a p53-dependent manner in response to DNAdamage (30).

As a result of Bax and Bak insertion in the outer mitochondrial membrane,several mitochondrial polypeptides are released (31). The most widely studied isthe electron transporter cytochrome c, which binds to a cytoplasmic scaffoldingprotein Apaf-1 (apoptotic protease-activating factor-1), causing an adenosinetriphosphate (ATP)- or dATP-dependent conformational change that allowsApaf-1 to bind and activate procaspase 9 (32). Activated caspase 9 in turncleaves procaspase 3 to yield active caspase 3. Other mitochondrial polypeptidesthat are released include second mitochondrial activator of caspases (SMAC)and HtRA2, which facilitate caspase activation by binding to caspase inhibitors(see Subheading 3.2.2), as well as endonuclease G, which also enters thenucleus and participates in chromatin digestion in some apoptotic cells (31).

Chapter 11 / Apoptosis in Leukemia Treatment 165

3. REGULATION OF APOPTOTIC PATHWAYS

Because of the lethal consequences of inadvertent caspase activation, it is notsurprising that the caspase activation pathways are highly regulated. Several fac-tors that affect the activation of one or both pathways have been identified.

3.1. Death Receptor Pathway RegulationSome DR pathway regulation occurs at the DR level. First, DR expression

varies according to cell lineage and stage of development (14). DR4 and DR5,for example, are expressed by leukemia cells (33) and other neoplastic cells butnot by many normal cells, except for thymocytes (14). These observations pro-vide at least a partial explanation for the selectivity of TNFα-related apoptosis-inducing ligand (TRAIL) for neoplastic cells (34,35). Moreover, DRexpression can be induced by certain treatments, particularly DNA damagingagents that induce DR5 synthesis (36). Second, DR signaling is regulated byphosphorylation. Protein kinase C activation, for example, inhibits DR recruit-ment of FADD after treatment with FasL or TRAIL (37). Although the mecha-nistic basis for these observations is incompletely understood, it has beensuggested that protein kinase Cα is the responsible isoform (38). Third, DRsignaling can be modulated by the expression of decoy receptors, cell surface,or secreted proteins that bind ligand but do not signal (13).


Fig. 1. (Facing page.) Two major pathways of caspase activation. In the death-receptor path-way (left), binding of specialized cell surface receptors such as Fas to its extracellular ligandFasL induces recruitment of the adaptor molecule FADD followed by procaspase 8 to cytoplas-mic “death domains” of the receptors. Assembly of this macromolecular complex results inactivation of caspase 8, which can then activate procaspase 3 directly or cleave the proapoptoticBcl-2 family member Bid to activate the mitochondrial pathway (box). In the mitochondrialpathway (right), multiple polypeptides are released from mitochondria. Once in the cytoplasm,cytochrome c acts as a cofactor for assembly of a macromolecule complex (“apoptosome”)containing Apaf-1 and procaspase 9, which exhibits caspase 9 activity. Although procaspase-9can be activated without cleavage, some or all of the caspase 9 is typically cleaved upon activa-tion. This cleaved caspase 9 reacts with XIAP to form an enzymatically inactive complex. Asimilar reaction (not shown) inhibits active caspase 3. SMAC/Diablo, which is also releasedfrom mitochondria, reacts with XIAP, liberating the active caspases and facilitating down-stream cleavages. Other mitochondrial polypeptides, including endonuclease G and apoptosisinducing factor (AIF) also participate in apoptotic events in some cell types (31).

Events leading to cytochrome c release are depicted in greater detail in the box. SomeBH3 only Bcl-2 family members, e.g., the fragment of Bid resulting from caspase-8-medi-ated cleavage (tBid), facilitate insertion of Bax and Bak in the mitochondrial outer mem-brane, where they participate in cytochrome c release (39), possibly by forming pores.Antiapoptotic Bcl-2 family members, including Bcl-2, Bcl-xL and the adenovirus 19 kilo-dalton E1B protein, inhibit Bax/Bak mitochondrial membrane insertion and cytochrome crelease, most likely by sequestering BH3 only polypeptides (27).

Interactions between FADD and initiator caspases are also regulated. Manycells express Flice-like inhibitory protein (FLIP) (3) which contains a caspase-8-like prodomain but lacks an active site cysteine. Because FLIP competitivelyinhibits procaspase 8 binding or activation, FLIP overexpression causes resis-tance to various DR ligands.

Finally, expression of procaspases 8 and 10 can vary widely. In particular,procaspase 10 is expressed at much higher levels in primary lymphocytes thanin lymphoid leukemia cell lines (17). Although complete absence of both ofthese procaspases results in inability to activate DR pathways (18), it ispresently unclear how diminished expression of one or the other of thesezymogens affects sensitivity to DR ligands.

3.2. Regulation of the Mitochondrial PathwayThe mitochondrial pathway is regulated both upstream of cytochrome c

release and downstream of caspase 9 activation. Two important polypeptidefamilies, the Bcl-2 family and the inhibitor of apoptosis protein (IAP) family,participate in these events.

3.2.1. REGULATION OF CYTOCHROME C RELEASE

BY BCL-2 FAMILY MEMBERS

As indicated, insertion of Bax and Bak into the mitochondrial outer mem-brane is postulated to play a critical role in cytochrome c release (39). Theeffects of these polypeptides are antagonized by antiapoptotic Bcl-2 familymembers (40,41). The founding member of this family, Bcl-2, was originallyidentified as a polypeptide that is expressed at high levels when the corre-sponding gene is juxtaposed to the immunoglobulin heavy chain promoter as aconsequence of the t(14;18) translocation found in indolent B-cell lymphomas.Subsequent analysis demonstrated that Bcl-2 diminishes the induction of apop-tosis by several different stimuli (40,41). Several structural homologs, includ-ing Bcl-xL and Mcl-1, have similar effects. These effects were traced, at leastin part, to the ability of these polypeptides to inhibit mitochondrial cytochromec release (42). Subsequent studies have demonstrated that these antiapoptoticBcl-2 family members inhibit Bax and Bak insertion into the outer mitochon-drial membrane (23,43), perhaps by binding and sequestering BH3 only familymembers such as Bid, Bad, Bim, and PUMA (27).

The balance between proapoptotic and antiapoptotic Bcl-2 family members isregulated by transcriptional mechanisms as well as posttranslational modifica-tions. The proapoptotic Bcl-2 family members Bax, Noxa, and PUMA are all syn-thesized in a p53-dependent manner after DNA damage or other cellular stresses(44). Conversely, expression of the antiapoptotic family members Bcl-xL (45) andMcl-1 (46) is regulated by Stat3 (signal transducer and activator of transcription3), a transcription factor that is activated by antiapoptotic cytokines (47).


Activities of certain Bcl-2 family members are also modulated by phospho-rylation (42). For example, the antiapoptotic function of Bcl-2 is enhanced byprotein kinase Cα- (PKCα) or ERK (extracellular signal-regulated kinase)-mediated phosphorylation on 70Ser (48,49). Mcl-1 is also phosphorylated in anERK-dependent manner (50). Other kinases affect other Bcl-2 family mem-bers. Protein kinase A-mediated phosphorylation of BAD on 155Ser inactivatesits proapoptotic BH3 domain, whereas cytokine-stimulated Akt-mediated phos-phorylation on 112Ser and 136Ser contributes to cytoplasmic sequestration ofBAD by 14-3-3 proteins (42).

3.2.2. REGULATION OF CASPASE ACTIVATION BY INHIBITOR

OF APOPTOSIS PROTEIN FAMILY MEMBERS

The presence of IAPs within cells conveys a second level of regulation(51,52). Multiple members of this gene family, including X-linked IAP (XIAP),cellular IAP1 (cIAP1), cIAP2, and neuronal apoptosis inhibitory protein(NAIP), are expressed in various tissues. These polypeptides all contain multiplefinger-like baculovirus inhibitor repeat (BIR) domains that bind to the surfacesof certain caspases and allow sequences between the BIRs to block the activesites of target enzymes (53). Consistent with these observations, XIAP, cIAP1,and cIAP2 directly inhibit the activities of active caspases 3, 7, and 9 (51,52). Inaddition, these IAPs contain domains that transfer ubiquitin to target caspases,thereby marking them for proteasome-mediated degradation (54,55).

It is currently believed that IAPs act as a cellular buffer for small amounts ofcaspases that are inadvertently activated (56). The capacity of this buffer is regu-lated on at least two levels. First, IAP expression is regulated. Transcription of thegenes encoding XIAP, cIAP1, and cIAP2, for example, is stimulated by nuclearfactor-κB (NFκB) (3), a transcription factor implicated in antiapoptotic signaling(57). Second, the ability of XIAP, cIAP1, and cIAP2 to bind to active caspases isregulated. In particular, two small polypeptides that are released from mitochon-dria concomitant with cytochrome c, SMAC and HtRA2, bind to the BIR regionsof IAPs and inhibit their ability to interact with caspases (58,59).

4. ALTERED APOPTOTIC PATHWAYS IN LEUKEMIA

4.1. Chronic Lymphocytic LeukemiaThe potential role of altered apoptosis in the pathogenesis of leukemia was

first demonstrated for chronic lymphocytic leukemia (CLL), a disorder charac-terized by gradual accumulation of morphologically mature lymphocytes. Itwas recognized decades ago that CLL cells had a low proliferative index com-pared with acute leukemias. The demonstration that 70% of CLL specimenscontain elevated levels of Bcl-2 (60), a polypeptide now known to inhibit mito-chondrial release of cytochrome c (42) and SMAC (61), provided an explana-


tion for these observations. Although chromosomal rearrangements that juxta-pose the Bcl-2 gene to the immunoglobulin light or heavy chain promotershave been described in rare cases of CLL, most CLL specimens lack theserearrangements. Instead, Bcl-2 overexpression reflects signaling initiated byinterleukin (IL)-4 (62) as well as Bcl-2 gene methylation (60).

In addition to Bcl-2, Mcl-1 and the antiapoptotic Bcl-2 binding protein Bag-1 are also overexpressed in a substantial portion of CLL cases (63). Moreover,p53 is mutated in a subset of CLL cases (64), providing another mechanism forthe inhibition of apoptotic pathways that could otherwise be activated by DNAdamage or cellular stress. Collectively, these alterations not only contribute tothe accumulation of CLL, cells but also render CLL cells somewhat moreresistant to chemotherapy than they might be otherwise.

4.2. Large Granular Lymphocyte LeukemiaLarge granular lymphocyte (LGL) leukemia is a rare chronic lymphoprolifer-

ative disorder characterized by accumulation of large granular lymphocytes, typ-ically of T-cell or natural killer (NK) cell origin (65). Symptoms are oftenrelated to neutropenia and anemia. The LGL leukemia cells typically containconstitutively activated STAT3 and elevated levels of the antiapoptotic Bcl-2family member Mcl-1 (46). These cells are resistant to Fas-induced apoptosis(66). In addition, they synthesize and release FasL, which has been implicated inarrest of erythroid maturation (67) and induction of granulocyte apoptosis. Thus,resistance of the LGL leukemia cells to apoptosis plays an important role in notonly their expansion, but also their ability to suppress normal hematopoiesis.

4.3. Chronic Myeloid DisordersBased in part on analogy to CLL, several groups have tested the hypothesis

that apoptosis might be altered in chronic myelogenous leukemia (CML). In 95%of CML cases, a characteristic t(9;22) chromosomal translocation juxtaposes the5′ end of the BCR gene with the 3′ end of the ABL gene, resulting in expression ofa unique constitutively active 210 kDa kinase, p210BCR/ABL (68). Transfection ofcomplementary DNA (cDNA) encoding this kinase into cytokine-dependentmurine 32D myeloid cells inhibits drug-induced apoptosis (69). Conversely,downregulation of BCR/ABL using antisense oligonucleotides enhances the rateof spontaneous or drug-induced apoptosis in CML cell lines (70) and clinicalsamples (71). Subsequent studies have demonstrated that p210BCR/ABL activatesthe phosphatidylinositol-3 kinase/Akt pathway as well as transcription mediatedby STAT5 and NFκB (68,72,73). Collectively, these signals cause phosphoryla-tion (inhibition) of the proapoptotic protein Bad as well as enhanced expressionof the antiapoptotic proteins Bcl-xL (74) and XAIP (75). These events arebelieved to inhibit cytochrome c release from the mitochondria and activation ofthe caspase cascade.


Emerging evidence suggests that apoptosis might also be deranged in otherchronic myeloproliferative syndromes. Neutrophils from patients with chronicneutrophilic leukemia, a rare BCR/ABL-negative chronic myeloproliferativesyndrome characterized by accumulation of mature neutrophils (76), alsoexhibit diminished apoptosis (77). Moreover, erythroid progenitors frompatients with polycythemia vera, a stem cell disorder characterized by expan-sion of the red cell compartment, contain elevated levels of Bcl-xL (78). Themechanistic bases for both of these observations remain to be established.

4.4. Acute Myeloid LeukemiaCurrent understanding suggests that development of the neoplastic phenotype

in solid tumors requires at least two changes, one that enhances proliferation andanother that inhibits apoptosis (7). Whether this is also true in acute leukemias isless clear. Perhaps the best studied example involves acute promyelocyticleukemia (APL), a disorder in which an almost universal t(15;17) chromosomaltranslocation brings together the promyelocytic leukemia gene PML, whichencodes a nuclear ubiquitin ligase, and the retinoic acid receptor gene RARα (79).Soon after its discovery, the PML-RARα fusion protein was shown to renderleukemia cell lines resistant to the induction of apoptosis (80). Subsequent analy-sis (reviewed in [81]) has shown that PML-RARα acts as a dominant negativeinhibitor of both RARα and PML. It is believed to competitively inhibit bindingof RARα to cognate DNA sequences, thereby inhibiting RA-induced expressionof genes involved in terminal differentiation of myeloid precursors. In addition,PML-RARα heterodimerizes with and inhibits wildtype PML, a polypeptide thatappears to be somehow required for induction of apoptosis by several agents (82).

It has been suggested that sensitivity to spontaneous or drug-induced apop-tosis is also altered in poor risk acute myeloid leukemia (AML) (83,84). Theseclaims are based on experiments in which AML samples are incubated underserum-free conditions or with drug in vitro. It is important to recognize that thebehavior of AML cells might be modified by the artificial conditions employedin these experiments. In particular, contact of blasts with stroma-derived cells(85) or a stromal cell line (86) inhibits spontaneous and drug-induced apopto-sis. These observations raise the possibility that resistance to the induction ofapoptosis might be underestimated in studies that expose cells to cytokine-freemedium or drugs in the absence of stroma. Nonetheless, the available data sug-gest that sensitivity of AML samples to induction of apoptosis varies (83,84).

The biochemical basis for this variation remains to be determined. Severalstudies have suggested that Bcl-2 overexpression is associated with a poorprognosis in AML (87–89). Another study has reported that Mcl-1 is selec-tively elevated in approx 50% of AML cases at the time of relapse comparedwith initial diagnosis (90). Elevated XIAP expression has also been identified


as a negative prognostic factor in AML (91). Finally, expression of messengerribonucleic acid (mRNA) encoding several apoptotic regulators is reportedlyaltered in AML samples with isolated trisomy 8, a harbinger of poor outcome,compared with other leukemias (92). Further studies examining additionalalterations in apoptotic pathways in AML are warranted.

4.5. Acute Lymphocytic LeukemiaThere are also several examples of altered apoptotic pathways in acute lym-

phocytic leukemia (ALL). First, approx 30% of adult patients with ALLexpress p190BCR/ABL, a chimeric BCR/ABL kinase with higher constitutiveenzymatic activity and stronger transforming ability than p210BCR/ABL. Trans-fection of cytokine-dependent murine hematopoietic cells with cDNA encodingp190BCR/ABL results in cytokine-independent proliferation and inhibition of apop-tosis (93). Interestingly, a mutant p190BCR/ABL kinase that suppresses apoptosisbut does not stimulate proliferation has been identified. The poor transformingability of this kinase suggests that suppression of apoptosis and stimulation ofproliferation are both required for transformation by p190BCR/ABL (93).

Apoptosis is also inhibited in at least one other ALL variant (reviewed in [94] ).The t(17;19) translocation observed in pro-B lymphocyte ALL results in the for-mation of a chimeric protein containing domains from E2A, a bHLH (basichelix-loop-helix) transcription factor, and HLF (hepatocyte leukemia factor), abasic leucine zipper (bZIP) transcription factor (95). Expression of this chimericpolypeptide inhibits p53-induced apoptosis and abrogates the interleukin-3dependence of normal pro-B lymphocytes. Conversely, dominant-negative inhi-bition of E2A-HLF induces apoptosis, suggesting that the chimeric proteinincreases the number of immature lymphoid cells by preventing or delaying theirdeath. It has been proposed that E2A-HLF does this by competing with thehuman homolog of the Caenorhabditis elegans transcription factor CES-2 andactivating antiapoptotic target genes that are normally repressed by CES-2-likeproteins (96). The E2A-HLF-induced apoptotic delay is believed to allow accu-mulation of additional mutations that contribute to leukemic transformation (94).

5. INDUCTION OF APOPTOSIS BY ANTILEUKEMIC THERAPY

In view of the emerging data suggesting that regulation of apoptotic path-ways might be altered in many leukemias, it is not surprising that virtually allantileukemic agents currently in use induce apoptosis in susceptible cells invitro (5). Apoptosis has also been detected in circulating blasts after initiationof induction chemotherapy (97,97a), providing support for the potential impor-tance of apoptosis in the clinical setting. That antineoplastic agents kill cells bytwo different mechanisms, that is, their primary effects on cellular metabolismand their ability to activate intrinsic cell death pathways, has led to extensivediscussion of two different questions.


5.1. Which Apoptotic Pathway(s) Are Triggered by Antileukemic Agents?

Because the answer to this question has implications for potential mecha-nisms of resistance, multiple studies have examined the pathways activated byvarious agents. Studies in fibroblasts or thymocytes from mice with targeteddeletions of various DR components have suggested that many of theantileukemic agents activate the mitochondrial pathway (5,98). Whether thisactivation is a primary event or occurs downstream of caspase 8 activation intype II cells has been more difficult to establish. Because caspase 8 can be acti-vated downstream of caspase 6 when the mitochondrial pathway is activated(99), caspase 8 activation by itself does not establish that caspase 8 is the initi-ating caspase after a particular treatment. Instead, further studies using domi-nant negative mutants of caspase 8, caspase 8-deficient cells, or inhibitors ofDR signaling are required to establish caspase 8 as an initiator caspase. Resultsof these types of studies have provided evidence that etoposide and topotecanare among the agents that can initiate apoptosis by inducing caspase 8 activa-tion in at least some cell types (100,101).

5.2. Does Resistance to Apoptosis Translate Into Drug Resistance?Much of the current interest in apoptotic pathways in the chemotherapy

community stems from the hope that better understanding of these pathwayswill provide new insight into drug resistance. Whether this will prove to be thecase, however, remains to be established. Some investigators argue that even ifapoptotic pathways are inhibited, cells treated with various chemotherapeuticagents will still die as a result of the effects of the primary lesions (e.g., DNAstrand breaks). Others argue that the primary lesions are compatible with pro-longed survival and possible repair in the absence of caspase activation.

There are relatively few studies that address this issue satisfactorily. Studiesfrom Nunez and colleagues revealed that expression of Bcl-xL in FL5.12 cellsnot only delayed apoptosis but also allowed an increased number of cells tosurvive treatment with etoposide or methotrexate and ultimately repopulate aculture in vitro (102). Forced overexpression of Bcl-2 has likewise beenreported to enhance colony formation after leukemia cell lines are treated withvarious agents (103), although this has not been a universal finding (104). Thelong-term protective effects of antiapoptotic Bcl-2 family members presum-ably reflect the ability of these polypeptides to inhibit mitochondrial release ofseveral proapoptotic factors in addition to cytochrome c.

In contrast, none of the papers describing antiapoptotic effects of IAPs(105–107) has demonstrated any long-term protective effect using repopula-tion or colony-forming assays. Moreover, current understanding (outlinedabove) suggests that IAPs allow cytochrome c release and caspase 9 cleavagebefore they act (108). Because other proapoptotic polypeptides, including AIF


and endonuclease G (31), are released along with cytochrome c, it is more dif-ficult to envision how IAP overexpression might influence long-term survivalof cells exposed to proapoptotic stimuli.

6. NEW AGENTS THAT DIRECTLY TARGET APOPTOTICPATHWAYS OR THEIR REGULATION

The improved understanding of apoptotic pathways has led to studies thatattempt to alter apoptotic pathways for therapeutic benefit. In the case ofleukemias, the intent would be to selectively activate the apoptotic machineryin neoplastic cells or to reverse the antiapoptotic changes that occur during theprocess of leukemogenesis. These therapies are reviewed in the context of theapoptotic pathways and antiapoptotic changes described. Because of spacelimitations, the following examples are illustrative rather than comprehensive.

6.1. Activation of Death Receptor PathwaysAlthough TNF-α, the founding member of the DR ligand family, kills a num-

ber of cancer and leukemia cell lines, the septic shock-like syndrome caused bythis agent limits its ability to be administered systemically (109). Accordingly,TNF-α is reserved for local infusions into isolated perfused limbs.

Several observations have given rise to optimism that the TNF-α homologTRAIL might be a more suitable death ligand for systemic treatment of cancerand leukemia. First, TRAIL is used by interferon (IFN) γ-stimulated lympho-cytes and NK cells to kill their targets, which include transformed cells (110).Second, the limited expression of DR4 and DR5 in normal cells (14) raises thepossibility that the TRAIL might exhibit some selectivity for solid tumors andleukemia. Consistent with this hypothesis, two recombinant versions ofTRAIL have been shown to kill neoplastic cell lines, including leukemia lines,in vitro, and shrink tumors in animals (34,35). Although some investigatorshave reported that TRAIL is toxic to human hepatocytes in primary culture,these results are not been reproducible (111), and early human clinical trials ofTRAIL are anticipated to begin soon.

Because hepatocytes do not express DR5, an alternative approach thatwould avoid the potential hepatotoxicity of TRAIL (if it exists) would be selec-tive crosslinking of DR5 with a suitable antibody. Toward this end, a mono-clonal antibody that binds DR5 but not DR4, TRAIL decoy receptors, Fas, orTNF-α receptors has been developed (112). Like recombinant human TRAIL,this antibody kills malignant hepatocytes and certain leukemia cell lines butspares normal human hepatocytes.

The effects of all-trans-retinoic acid (ATRA) on APL cells provide an exampleof the potential importance of TRAIL in antileukemic therapy. Although ATRAhas long been viewed as an agent that restores RARα-mediated differentiation in


APL, recent studies indicate that ATRA also upregulates TRAIL expression in anAPL cell line and clinical APL samples in vitro (113). Importantly, soluble TRAILreceptors abolish ATRA toxicity in culture, suggesting that the toxicity reflectsparacrine effects of the ATRA-induced TRAIL. The observation that DNA damag-ing agents upregulate DR5 (36), coupled with these new results, provides a conve-nient explanation for the efficacy of ATRA in combination with an anthracyclineor other DNA-damaging agent to bring about durable remissions in APL (114).

DNA damaging agents, including etoposide and cytarabine, also upregulateDR5 and potentiate the effects of TRAIL in other AML lines (115). If TRAILexhibits antileukemic activity, the effects of these combinations certainly needbe tested.

6.2. Reversing Inhibition of the Mitochondrial Pathway6.2.1. INHIBITING BCL-2 OVEREXPRESSION WITH ANTISENSE

OLIGONUCLEOTIDES

The observation that elevated Bcl-2 expression is associated with a poor prog-nosis in AML and ALL (87–89) has led to studies examining the effects of Bcl-2downregulation. Several studies have reported that cytarabine-induced apoptosiscan be enhanced in vitro by treating leukemia cells with Bcl-2 antisenseoligonucleotides (116,117). Antisense oligonucleotides are short synthetic sin-gle-stranded DNA molecules (15–25 bases) that enter cells, bind to complemen-tary sequences within target mRNA, and either prevent translation of the targetmessage or induce its RNase H-mediated degradation (118,119). Because ofbound proteins and conformational factors, not all portions of the mRNAsequence are equally effective targets of antisense oligonucleotides. A series ofstudies (reviewed in [119] ) has demonstrated that an 18-base phosphorothioateoligonucleotide complementary to the first 18 nucleotides of the Bcl-2 codingsequence (now called G3139 or Genasense™, Genta Corp., Berkeley Heights,NJ) is particularly effective at downregulating Bcl-2 in lymphoma cells contain-ing the t(14;18) chromosomal translocation and inhibiting engraftment of thesecells when administered to nude mice for 14 d after cell inoculation.

Based on these results, a phase I study of G3139 was completed in patientswith relapsed non-Hodgkin’s lymphoma (NHL) (119,120). Twenty-onepatients were treated with escalating oligonucleotide doses as a 14-d continu-ous infusion. Inflammatory reactions at the injection site were noted in allpatients, and dose-limiting toxicities were thrombocytopenia, hypotension,fever, and asthenia. Examination of paired lymph node, peripheral blood, orbone marrow samples revealed that G3139 treatment resulted in decreased Bcl-2 levels in 58% of cases. Objective responses were noted in 14% of patients,with one complete response that was sustained for ≥ 36 months. It should benoted, however, that this trial examined G3139 alone.


As indicated in Section 3.2.1., Bcl-2 inhibits the ability of proapoptoticstimuli to induce mitochondrial cytochrome c release. Accordingly, even inneoplasms in which Bcl-2 overexpression is not a primary oncogenic event,Bcl-2 downregulation would be predicted to enhance the induction of apopto-sis by other agents. Based on this rationale, G3139 is currently undergoingextensive clinical testing in combination with other antineoplastic agents(121). One current strategy involves combining G3139 with FLAG fludara-bine, ara-C, and G-CSF), a popular regimen for the treatment of relapsed andrefractory AML (122). For CLL, G3139 is being combined with fludarabineplus cyclophosphamide. Data supporting the safety and improved efficacy ofthis approach are awaited with interest.

6.2.2. BISPECIFIC ANTISENSE OLIGONUCLEOTIDES

Antisense oligonucleotides provide the potential to selectively downregulatea single mRNA. This selectivity is both a strength and a potential weakness.Although some studies analyzing Bcl-2 family members in intermediate gradeNHL have suggested a strong correlation between Bcl-2 overexpression andprognosis, others have suggested that Bcl-xL might also contribute (123). Bcl-xL would not be downregulated by G3139 because of differences in mRNAsequence.

To circumvent this problem, a Bcl-xL/Bcl-2-bispecific antisense moleculethat targets a region of sequence identity between the respective mRNAs hasbeen created (124). The resulting downregulation of both Bcl-2 and Bcl-xL isassociated with induction of apoptosis in vitro and in vivo as well as inhibitionof variety of tumors grown as xenografts in nude mice (125). Further preclini-cal and clinical studies of this unique molecule are awaited with interest.

6.2.3. SMALL MOLECULE INHIBITORS OF BCL-2 HOMOLOG SYNTHESIS

Current understanding suggests that expression of antiapoptotic Bcl-2 fam-ily members is, at least in part, a regulated process. As indicated in Section3.2.1., expression of Mcl-1 and Bcl-xL is stimulated by activated STAT3 andSTAT5 transcription factors (45,46,126). Additional studies have shown thatexpression of Bcl-2, Mcl-1, and Bcl-xL is also regulated by signaling throughthe raf/Mek1/ERK pathway (127–129). These observations suggest severalstrategies for downregulating expression of antiapoptotic Bcl-2 family mem-bers using small molecule inhibitors.

6.2.3.1. Inhibition of Signal Transducer and Activator of TranscriptionActivation. During the course of normal cytokine-mediated signaling, Januskinases (Jaks) bound to the cytoplasmic domains of cytokine-bound receptorsare activated. The Jaks in turn phosphorylate cytosolic STAT proteins, whichthen dimerize, enter nuclei, and activate transcription of target genes (47). Inaddition, STAT5 is directly phosphorylated and activated by BCR/ABL


kinases (reviewed in [73,126]). These observations suggest at least two strate-gies for downregulating antiapoptotic Bcl-2 family members.

AG490, a small-molecule Jak2 inhibitor, downregulates expression of Bcl-xL in multiple myeloma cells (45) and Mcl-1 in LGL leukemia cells in vitro(46). In addition, AG490 has demonstrated antileukemic activity in severecombined immunodeficient (SCID) mice bearing human leukemia xenografts(130). Unfortunately, low potency and poor solubility limit the efficacy ofAG490 in vivo. Nonetheless, these observations provide a starting point for thedevelopment of more potent and useful Jak inhibitors.

STI571 is a potent and somewhat selective BCR/ABL kinase inhibitor. Asmight be expected, this drug inhibits STAT5 phosphorylation and Bcl-xLexpression in CML cells (126). In addition, STI571 inhibits other antiapoptoticsignaling pathways downstream of BCR/ABL. Collectively, these effects con-tribute to the induction of apoptosis by STI571 as well as its ability to sensitizecells to other apoptosis inducing agents (see Chapter 9).

6.2.3.2. Inhibition of the Ras/Mek1/ERK Pathway. Constitutive ERK acti-vation has been observed in approx 50% of human specimens with acuteleukemia (131). Recent studies have shown that treatment of AML cell lines withone of two Mek1 inhibitors, PD98059 or PD184352 (now known as CI-1040),results in decreased levels of the antiapoptotic Bcl-2 family members Bcl-xL andMcl-1 followed by spontaneous induction of apoptosis in vitro (132). In addition,PD98059 inhibits proliferation of leukemic progenitors (but not normal myeloidprogenitors) in colony-forming assays (132). Because CI-1040 is well toleratedin phase I trials in patients with solid tumors, there is ample rationale for furtherpreclinical and possible clinical studies of this agent in acute leukemia.

6.2.3.3. Flavopiridol. Although the mechanism of its proapoptotic effectsremains to be more fully elucidated, the kinase inhibitor flavopiridol representsanother example of a small molecule that downregulates antiapoptotic Bcl-2family members. Flavopiridol is a semisynthetic flavonoid that was originallyidentified as an inhibitor of cyclin-dependent kinases (133), the enzymes thatenable passage of cells from one cell cycle stage to another (134). Althoughflavopiridol was believed to be cytostatic, the initial report that it inducedapoptosis in HL-60 human leukemia cells (135) was quickly followed by thedemonstration that this agent kills several leukemia and lymphoma-derivedcell lines as well as clinical CLL isolates (133). In a CLL cell line and in CLLsamples, flavopiridol decreases expression of the antiapoptotic proteins Bcl-2,Mcl-1, and/or XIAP (136,137). One potential explanation for these findings isthe flavopiridol-induced inhibition of cyclin-dependent kinase 9 and the conse-quent inhibition of RNA polymerase II, which depends on cyclin-dependentkinase 9-mediated phosphorylation for activation (138). According to thisview, polypeptides such as Bcl-2 and Mcl-1 that are synthesized in a regulated


fashion have short-lived messages. When RNA polymerase II cannot function,mRNA synthesis stops, and the messages for several of the antiapoptotic pro-teins turnover, quickly leading to decreased polypeptide levels (138).

Based on these antiproliferative and proapoptotic effects in preclinical mod-els, flavopiridol is currently undergoing clinical testing. Phase I clinical trialshave examined three schedules: a 72-h infusion, a 24-h infusion, and daily 1-hinfusions for 5 consecutive d (133,139). The daily schedule is particularly ger-mane to the treatment of hematologic malignancies because it generates peakconcentrations in the range previously observed to induce apoptosis in humanleukemia cell lines and CLL cells in vitro. In a phase I study performed at theNational Cancer Institute, dose-limiting toxicities of neutropenia, diarrhea, andfatigue were observed on this schedule.

As might be expected of an agent that inhibits cell cycle progression,flavopiridol has important effects on the action of other anticancer drugs(140). In particular, when flavopiridol is administered before or concomitantwith proliferation-dependent agents such as cytarabine, it inhibits their activity. Onother hand, if cytarabine is delayed until cells that survive flavopiridol treatmenthave resumed cycling, the cytotoxic effects of flavopiridol and cytarabine can besynergistic (140). Efforts to harness this sequence-dependent synergy in a clinicaltrial for patients with relapsed/refractory acute leukemia are currently under way.

6.2.4. INHIBITION OF BCL-2 FUNCTION

Rather than inhibiting the synthesis of Bcl-2 family members, an alternativeapproach would be the inhibition of Bcl-2 function. Two different strategieshave been explored.

The first strategy involves inhibition of Bcl-2 phosphorylation. As indicatedin Section 3.2.1., current understanding suggests that Bcl-2 is a more potentinhibitor of apoptosis when phosphorylated on Ser70. Because PKCα has beenimplicated in this phosphorylation (48), PKCα inhibition would be expected tofacilitate activation of the mitochondrial pathway. Although several small-mol-ecule PKC inhibitors have been identified (141), these agents often lack speci-ficity because of homologies between the various PKC isoforms as well assimilarities between PKC and other kinases. Of interest, therefore, is ISIS-3521. This antisense oligonucleotide directed against PKCα mRNA is cur-rently being evaluated for potential anticancer activity as a single agent and incombination with various drugs (142). Although PKCα undoubtedly has manytargets in cancer cells, it will be interesting to see whether Bcl-2 phosphoryla-tion is inhibited at therapeutic concentrations of this agent.

Wang et al. took a different approach in targeting Bcl-2 function (143). Cur-rent understanding suggests that Bcl-2 inhibits apoptosis by binding toproapoptotic family members (27). These protein-protein interactions involve ahydrophobic pocket on the Bcl-2 surface. Starting with the predicted structure


of Bcl-2, Wang et al. simulated the potential binding of a library of compoundsto the critical protein-protein interaction domain. Of 28 compounds identifiedas potential ligands for this pocket, 1 (HA14-1) demonstrated binding to Bcl-2in vitro and the ability to induce apoptosis in HL-60 human acute leukemiacells, which intrinsically express high Bcl-2 levels (143). This is undoubtedlyone of many small molecules that will be identified or designed to inhibit vari-ous antiapoptotic Bcl-2 family members.

6.2.5. INHIBITION OF INHIBITOR OF APOPTOSIS PROTEIN SYNTHESIS

OR FUNCTION

As indicated in Section 4.4., elevated XIAP expression has been associatedwith a poor prognosis in AML (91). In light of these results, approaches forinhibiting XIAP synthesis or function are potentially germane to antileukemictherapy.

Conceptually, the most straightforward way to inhibit XIAP synthesiswould be the use of a suitable antisense oligonucleotide. This approach sensi-tizes non-small cell lung cancer cells to radiation (144) and ovarian cancercells to cisplatin-induced apoptosis (145). It remains to be determined whethersimilar molecules will sensitize AML cells to antileukemic agents.

Other approaches for downregulating XIAP expression also are feasible. Asindicated in Section 3.2.2., XIAP expression is enhanced by NFκB signaling.Additional studies have demonstrated that NFκB is constitutively activated in alarge percentage of primary AML samples (146). Moreover, this NFκB signal-ing can be inhibited in vitro by NFκB decoy oligonucleotides, i.e., double-stranded oligonucleotides that contain the NFκB binding site and competeactivated NFκB away from genomic sequences (147). Although these decoyoligonucleotides sensitize leukemic blasts to cytarabine in vitro (146), furtherpreclinical study is required to determine whether similar effects will beobserved in vivo and whether XIAP downregulation plays a role in this process.

An alternative approach involves the inhibition of XIAP function by syn-thetic peptides and peptidomimetics. The proapoptotic mitochondrial proteinSMAC binds XIAP and inhibits its interaction with cleaved caspase-9 (108).The N-terminal tetrapeptide on SMAC, ala-val-pro-ile (AVPI), is believed toplay a critical role in this process (148). Several investigators are currentlyintroducing this peptide sequence into cells and examining its effect on drug-induced apoptosis. To the extent that XIAP contributes to drug resistance inleukemia, it is predicted that AVPI peptides will increase drug-induced apopto-sis. If these predictions are confirmed, there will be considerable interest in thedesign and synthesis of peptide-like molecules that might have similar effects.

7. SUMMARY

Biochemical analyses have identified several polypeptides that participate inor regulate the apoptotic process. There is emerging evidence that apoptotic


pathways are inhibited, at least to some extent, in leukemia cells. On the otherhand, as indicated by examples in the preceding section, several experimentaltreatments can alter the apoptotic machinery and enhance the induction of apop-tosis by antileukemic drugs in vitro. Current efforts are directed toward perform-ing suitable preclinical studies and clinical trials to determine whether any ofthese approaches can improve therapeutic outcomes in vivo. Results of thesestudies will determine whether the current interest in dysregulation of apoptoticpathways as a potential mechanism of clinical drug resistance is well founded.

8. ACKNOWLEDGMENTS

The authors thank Deb Strauss for secretarial assistance and Joya Chandra,Wei Meng, David Steensma, Ruben Mesa, Michael Heldebrant, and ChristinaArnt for helpful conversations. Work in the Kaufmann laboratory is support byR01 CA69008.

REFERENCES1. Thompson CB. Apoptosis in the pathogenesis and treatment of disease. Science

1995;267:1456–1462.2. Wyllie AH, Kerr JF, Currie AR. Cell death: the significance of apoptosis. Int Rev Cytol

1980;68:251–306.3. Earnshaw WC, Martins LM, Kaufmann SH. Mammalian caspases: structure, activation,

substrates, and functions during apoptosis. Annu Rev Biochem 1999;68:383–424.4. Budihardjo I, Oliver H, Lutter M, Luo X, Wang X. Biochemical pathways of caspase acti-

vation during apoptosis. Annu Rev Cell Dev Biol 1999;15:269–290.5. Kaufmann SH, Earnshaw WC. Induction of apoptosis by cancer chemotherapy. Exp Cell

Res 2000;256:42–49.6. Herr I, Debatin K-M. Cellular stress response and apoptosis in cancer therapy. Blood

2001;98:2603–2614.7. Evan GI, Vousden KH. Proliferation, cell cycle and apoptosis in cancer. Nature

2001;411:342–348.8. Stroh C, Schulze-Osthoff K. Death by a thousand cuts: an ever increasing list of caspase

substrates. Cell Death Differ 1998;5:997–1000.9. Enari M, Sakahira H, Yokoyama H, Okawa K, Iwamatsu A, Nagata S. A caspase-activated

DNase that degrades DNA during apoptosis, and its inhibitor ICAD. Nature 1998;391:43–50.10. Samejima K, Tone S, Earnshaw WC. CAD/DFF40 Nuclease is dispensable for high molec-

ular weight DNA cleavage and stage I chromatin condensation in apoptosis. J Biol Chem2001;276:45,427–45,432.

11. Rao L, Perez D, White E. Lamin proteolysis facilitates nuclear events during apoptosis. JCell Biol 1996;135:1441–1455.

12. Chai J, Wu Q, Shiozaki E, Srinivasula SM, Alnemri ES, Shi Y. Crystal structure of a procas-pase-7 zymogen: mechanisms of activation and substrate binding. Cell 2001;107:399–407.

13. Ashkenazi A. Targeting death and decoy receptors of the tumour-necrosis factor super-family. Nature Reviews Cancer 2002;2:420–430.

14. Schulze-Osthoff K, Ferrari D, Los M, Wesselborg S, Peter ME. Apoptosis signaling bydeath receptors. Eur J Biochem 1998;254:439–459.

15. Siegel RM, Martin DA, Zheng L, et al. Death-effector filaments: novel cytoplasmic struc-tures that recruit caspases and trigger apoptosis. J Cell Biol 1998;141:1243–1253.


16. Salvesen GS, Dixit VM. Caspase activation: the induced-proximity model. Proc Natl AcadSci USA 1999;96:10964–10967.

17. Wang J, Chun HJ, Wong W, Spencer DM, Lenardo MJ. Caspase-10 is an initiator caspase indeath receptor signaling. Proc Natl Acad Sci USA 2001;98:13884–13888.

18. Kischkel FC, Lawrence DA, Tinel A, et al. Death receptor recruitment of endogenous caspase-10 and apoptosis initiation in the absence of caspase-8. J Biol Chem 2001;276:46639–46646.

19. Scaffidi C, Fulda S, Srinivasan A, et al. Two CD95 (APO-1/Fas) signaling pathways. EmboJ 1998;17:1675–1687.

20. Green DR, Reed JC. Mitochondria and apoptosis. Science 1998;281:1309–1312.21. Desagher S, Martinou JC. Mitochondria as the central control point of apoptosis. Trends

Cell Biol 2000;10:369–377.22. Zamzami N, Kroemer G. The mitochondrion in apoptosis: how Pandora’s box opens. Nat

Rev Mol Cell Biol 2001;2:67–71.23. Antonsson B, Montessuit S, Sanchez B, Martinou JC. Bax is present as a high molecular

weight oligomer/complex in the mitochondrial membrane of apoptotic cells. J Biol Chem2001;276:11615–11623.

24. Eskes R, Desagher S, Antonsson B, Martinou JC. Bid induces the oligomerization andinsertion of Bax into the outer mitochondrial membrane. Mol Cell Biol 2000;20:929–935.

25. Korsmeyer SJ, Wei MC, Saito M, Weiler S, Oh KJ, Schlesinger PH. Pro-apoptotic cascadeactivates BID, which oligomerizes BAK or BAX into pores that result in the release ofcytochrome c. Cell Death Differ 2000;7:1166–1173.

26. Wei MC, Lindsten T, Mootha VK, et al. tBID, a membrane-targeted death ligand, oligomer-izes BAK to release cytochrome c. Genes Dev 2000;14:2060–2071.

27. Cheng EH, Wei MC, Weiler S, et al. BCL-2, BCL-X(L) Sequester BH3 domain-only mole-cules preventing Bax- and Bak-mediated mitochondrial apoptosis. Mol Cell 2001;8:705–711.

28. Gajewski TF, Thompson CB. Apoptosis meets signal transduction: elimination of a badinfluence. Cell 1996;87:589–592.

29. Bouillet P, Metcalf D, Huang DC, et al. Proapoptotic Bcl-2 relative Bim required for certainapoptotic responses, leukocyte homeostasis, and to preclude autoimmunity. Science1999;286:1735–1738.

30. Yu J, Zhang L, Hwang PM, Kinzler KW, Vogelstein B. PUMA induces the rapid apoptosisof colorectal cancer cells. Mol Cell 2001;7:673–682.

31. Kaufmann SH, Hengartner M. Programmed cell death: alive and well in the new mille-nium. Trends Cell Biol 2001;11:526–534.

32. Rodriguez J, Lazebnik Y. Caspase-9 and APAF-1 form an active holoenzyme. Genes Dev1999;13:3179–3184.

33. Wuchter C, Krappmann D, Cai Z, et al. In vitro susceptibility to TRAIL-induced apoptosisof acute leukemia cells in the context of TRAIL receptor gene expression and constitutiveNF-kappa B activity. Leukemia 2001;15:921–928.

34. Ashkenazi A, Pai RC, Fong S, et al. Safety and antitumor activity of recombinant solubleApo2 ligand. J Clin Invest 1999;104:155–162.

35. Walczak H, Miller RE, Ariail K, et al. Tumoricidal activity of tumor necrosis factor-relatedapoptosis-inducing ligand in vivo. Nat Med 1999;5:157–163.

36. Wu GS, Burns TF, McDonald ER 3rd, et al. KILLER/DR5 is a DNA damage-induciblep53-regulated death receptor gene. Nat Genet 1997;17:141–143.

37. Meng XW, Heldebrant MP, Kaufmann SH. Phorbol-12-myristate 13-acetate inhibits deathreceptor-mediated apoptosis in Jurkat cells by disrupting recruitment of Fas-associatedpolypeptide with death domain. J Biol Chem 2002;277:3776–3783.

38. Gomez-Angelats M, Cidlowski JA. Protein kinase C regulates FADD recruitment anddeath-inducing signaling complex formation in Fas/CD95-induced apoptosis. J Biol Chem2001;276:44944–44952.


39. Wei MC, Zong WX, Cheng EH, et al. Proapoptotic BAX and BAK: a requisite gateway tomitochondrial dysfunction and death. Science 2001;292:727–730.

40. Yang E, Korsmeyer SJ. Molecular thanatopsis: a discourse on the Bcl-2 family and celldeath. Blood 1996;88:386–401.

41. Reed JC. Bcl-2 family proteins. Oncogene 1998;17:3225–3236.42. Gross A, McDonnell JM, Korsmeyer SJ. BCL-2 family members and the mitochondria in

apoptosis. Genes Dev 1999;13:1899–1911.43. Perez D, White E. TNF-alpha signals apoptosis through a bid-dependent conformational

change in Bax that is inhibited by E1B 19K. Mol Cell 2000;6:53–63.44. Shen Y, White E. p53-dependent apoptosis pathways. Adv Cancer Res 2001;82:55–84.45. Catlett-Falcone R, Landowski TH, Oshiro MM, et al. Constitutive activation of Stat3 signaling

confers resistance to apoptosis in human U266 myeloma cells. Immunity 1999;10:105–115.46. Epling-Burnette PK, Liu JH, Catlett-Falcone R, et al. Inhibition of STAT3 signaling leads

to apoptosis of leukemic large granular lymphocytes and decreased Mcl-1 expression. JClin Invest 2001;107:351–362.

47. Turkson J, Jove R. STAT proteins: novel molecular targets for cancer drug discovery. Onco-gene 2000;19:6613–6626.

48. Ruvolo PP, Deng X, Carr BK, May WS. A functional role for mitochondrial protein kinaseCalpha in Bcl2 phosphorylation and suppression of apoptosis. J Biol Chem 1998;273:25436–25442.

49. Deng X, Ruvolo P, Carr B, May WS Jr. Survival function of ERK1/2 as IL-3-activated,staurosporine-resistant Bcl2 kinases. Proc Natl Acad Sci USA 2000;97:1578–1583.

50. Domina AM, Smith JH, Craig RW. Myeloid cell leukemia 1 is phosphorylated through twodistinct pathways, one associated with extracellular signal-regulated kinase activation andthe other with G2/M accumulation or protein phosphatase 1/2A inhibition. J Biol Chem2000;275:21688–21694.

51. Deveraux QL, Reed JC. IAP family proteins—suppressors of apoptosis. Genes Dev1999;13:239–252.

52. Salvesen GS, Duckett CS. IAP proteins: blocking the road to death’s door. Molecular andCell Biol. 2002;3:401–410.

53. Goyal L. Cell death inhibition: keeping caspases in check. Cell 2001;104:805–808.54. Yang Y, Fang S, Jensen JP, Weissman AM, Ashwell JD. Ubiquitin protein ligase activity of

IAPs and their degradation in proteasomes in response to apoptotic stimuli. Science2000;288:874–877.

55. Suzuki Y, Nakabayashi Y, Takahashi R. Ubiquitin-protein ligase activity of X-linked inhibitorof apoptosis protein promotes proteasomal degradation of caspase-3 and enhances its anti-apoptotic effect in Fas-induced cell death. Proc Natl Acad Sci USA 2001;98:8662–8667.

56. Miller LK. An exegesis of IAPs: salvation and surprises from BIR motifs. Trends Cell Biol1999;9:323–328.

57. Van Antwerp DJ, Martin SJ, Verma IM, Green DR. Inhibition of TNF-induced apoptosis byNF-kappa B. Trends Cell Biol 1998;8:107–111.

58. Ekert PG, Silke J, Hawkins CJ, Verhagen AM, Vaux DL. DIABLO promotes apoptosis byremoving MIHA/XIAP from processed caspase 9. J Cell Biol 2001;152:483–490.

59. Verhagen AM, Silke J, Ekert PG, et al. HtrA2 promotes cell death through its serine pro-tease activity and its ability to antagonize inhibitor of apoptosis proteins. J Chem Biol2002;277:445–454.

60. Hanada M, Delia D, Aiello A, Stadtmauer E, Reed JC. Bcl-2 Gene hypomethylation andhigh-level expression in B-cell chronic lymphocytic leukemia. Blood 1993;82:1820–1828.

61. Adrain C, Creagh EM, Martin SJ. Apoptosis-associated release of Smac/DIABLO from mito-chondria requires active caspases and is blocked by Bcl-2. EMBO J 2001;20:6627–6636.

62. Panayoitidis P, Ganeshaguru K, Jabbar SA, Hoffbrand AV. Interleukin-4 inhibits apoptoticcell death and loss of the Bcl-2 protein in B-chronic lymphocytic leukaemia cells in vitro.Br J Haematol 1993;85:439–445.


63. Kitada S, Andersen J, Akar S, et al. Expression of apoptosis-regulating proteins in chroniclymphocytic leukemia: correlations with in vitro and in vivo chemoresponses. Blood1998;91:3379–3389.

64. Bannerji R, Byrd JC. Update on the biology of chronic lymphocytic leukemia. Curr OpinCell Biol 2000;12:22–29.

65. Lamy T, Loughran TP Jr. Current concepts: large granular lymphocyte leukemia. Blood Rev1999;13:230–234.

66. Lamy T, Liu JH, Landowski TH, Dalton WS, Loughran TP Jr. Dysregulation ofCD95/CD95 ligand-apoptotic pathway in CD3(+) large granular lymphocyte leukemia.Blood 1998;92:4771–4777.

67. De Maria R, Zeuner A, Eramo A, et al. Negative regulation of erythropoiesis by caspase-mediated cleavage of GATA-1. Nature 1999;401:489–493.

68. Deininger MW, Goldman JM, Melo JV. The molecular biology of chronic myeloidleukemia. Blood 2000;96:3343–3356.

69. Laneuville P, Timm M, Hudson AT. BCR/ABL expression in 32D c13 (G) cells inhibitsapoptosis induced by protein tyrosine kinase inhibitors. Cancer Res 1994;54:1360–1366.

70. Smetsers TF, Skorski T, van de Locht LT, et al. Antisense Bcr-Abl oligonucleotides induceapoptosis in the Philadelphia chromosome-positive cell line BV173. Leukemia1994;8:129–140.

71. Bedi A, Zehnbauer BA, Barber JP, Sharkis SJ, Jones RJ. Inhibition of apoptosis by Bcr-Ablin chronic myeloid leukemia. Blood 1994;83:2038–2044.

72. Faderl S, Talpaz M, Estrov Z, O’Brien S, Kurzrock R, Kantarjian HM. The biology ofchronic myeloid leukemia. N Engl J Med 1999;341:164–172.

73. Sawyers CL. Chronic myeloid leukemia. N Engl J Med 1999;340:1330–1340.74. Amarante-Mendes GP, Naekyung Kim C, Liu L, et al. Bcr-Abl exerts its antiapoptotic

effect against diverse apoptotic stimuli through blockage of mitochondrial release ofcytochrome C and activation of caspase-3. Blood 1998;91:1700–1705.

75. Fang G, Kim CN, Perkins CL, et al. CGP57148B (STI-571) induces differentiation andapoptosis and sensitizes Bcr-Abl-positive human leukemia cells to apoptosis due toantileukemic drugs. Blood 2000;96:2246–2253.

76. Elliott MA, Dewald GW, Tefferi A, Hanson CA. Chronic neutrophilic leukemia (CNL): aclinical, pathologic, and cytogenetic study. Leukemia 2001;15:35–40.

77. Hara K, Abe Y, Hirase N, et al. Apoptosis resistance of mature neutrophils in a case ofchronic neutrophilic leukaemia. Eur J Haematol 2001;66:70–71.

78. Silva M, Richard C, Benito A, Sanz C, Olalla I, Fernandez-Luna JL. Expression of Bcl-xin erythroid precursors from patients with polycythemia vera. N Engl J Med1998;338:564–571.

79. Melnick A, Licht JD. Deconstructing a disease: RARalpha, its fusion partners, and theirroles in the pathogenesis of acute promyelocytic leukemia. Blood 1999;93:3167–3215.

80. Grignani F, Ferrucci PF, Testa U, et al. The acute promyelocytic leukemia-specific PML-RAR alpha fusion protein inhibits differentiation and promotes survival of myeloid precur-sor cells. Cell 1993;74:423–431.

81. Wang ZG, Delva L, Gaboli M, et al. Role of PML in cell growth and the retinoic acid path-way. Science 1998;279:1547–1551.

82. Wang ZG, Ruggero D, Ronchetti S, et al. PML is essential for multiple apoptotic pathways.Nat Genet 1998;20:266–272.

83. Los M, Herr I, Friesen C, Fulda S, Schulze-Osthoff K, Debatin KM. Cross-resistance ofCD95- and drug-induced apoptosis as a consequence of deficient activation of caspases(ICE/Ced-3 proteases). Blood 1997;90:3118–3129.

84. Smith BD, Bambach BJ, Vala MS, et al. Inhibited apoptosis and drug resistance in acutemyeloid leukaemia. Br J Haematol 1998;102:1042–1049.

85. Bendall LJ, Daniel A, Kortlepel K, Gottlieb DJ. Bone marrow adherent layers inhibit apop-tosis of acute myeloid leukemia cells. Exp Hematol 1994;22:1252–1260.


86. Garrido SM, Appelbaum FR, Willman CL, Banker DE. Acute myeloid leukemia cells areprotected from spontaneous and drug-induced apoptosis by direct contact with a humanbone marrow stromal cell line (HS-5). Exp Hematol 2001;29:448–457.

87. Campos L, Rouault JP, Sabido O, et al. High expression of bcl-2 protein in acute myeloidleukemia cells is associated with poor response to chemotherapy. Blood 1993;81:3091–3096.

88. Maung ZT, Maclean FR, Reid MM, et al. The relationship between bcl-2 expression andresponse to chemotherapy in acute leukaemia. Br J Haematol 1994;88:105–109.

89. Karakas T, Maurer U, Weidmann E, Miething CC, Hoelzer D, Bergmann L. High expres-sion of bcl-2 mRNA as a determinant of poor prognosis in acute myeloid leukemia. AnnOncol 1998;9:159–165.

90. Kaufmann SH, Karp JE, Svingen PA, et al. Elevated expression of the apoptotic regulatorMcl-1 at the time of leukemic relapse. Blood 1998;91:991–1000.

91. Tamm I, Kornblau SM, Segall H, et al. Expression and prognostic significance of IAP-fam-ily genes in human cancers and myeloid leukemias. Clin Cancer Res 2000;6:1796–1803.

92. Virtaneva K, Wright FA, Tanner SM, et al. Expression profiling reveals fundamental bio-logical differences in acute myeloid leukemia with isolated trisomy 8 and normal cytoge-netics. Proc Natl Acad Sci USA 2001;98:1124–1129.

93. Cortez D, Kadlec L, Pendergast AM. Structural and signaling requirements for BCR-ABL-mediated transformation and inhibition of apoptosis. Mol Cell Biol 1995;15:5531–5541.

94. Look AT. Oncogenic transcription factors in the human acute leukemias. Science1997;278:1059–1064.

95. Inaba T, Roberts WM, Shapiro LH, et al. Fusion of the leucine zipper gene HLF to the E2Agene in human acute B- lineage leukemia. Science 1992;257:531–534.

96. Metzstein MM, Hengartner MO, Tsung N, Ellis RE, Horvitz HR. Transcriptional regulatorof programmed cell death encoded by Caenorhabditis elegans gene ces-2. Nature1996;382:545–547.

97. Li X, Gong J, Feldman E, Seiter K, Traganos F, Darzynkiewicz Z. Apoptotic cell death dur-ing treatment of leukemias. Leukemia Lymphoma 1994;13:65–70.

97a. Stahnke K, Fulda S, Friesen C, Staub G, Debatin K-M. Activation of apoptosis pathways inperipheral blood lymphocytes by in vivo chemotherapy. Blood 2001;98:3066–3073.

98. Villunger A, Strasser A. Does “death receptor” signaling play a role in tumorigenesis andcancer therapy? Oncol Res 1998;10:541–550.

99. Slee EA, Harte MT, Kluck RM, et al. Ordering the cytochrome C-initiated caspase cascade:hierarchical activation of caspases-2, -3, -6, -7, -8, and -10 in a Caspase-9-dependent man-ner. J Cell Biol 1999;144:281–292.

100. Micheau O, Solary E, Hammann A, Dimanche-Boitrel M-T. Fas ligand-independent,FADD-mediated activation of the Fas death pathway by anticancer drugs. J Biol Chem1999;274:7987–7992.

101. Shao RG, Cao CX, Nieves-Neira W, Dimanche-Boitrel MT, Solary E, Pommier Y. Activa-tion of the Fas pathway independently of Fas ligand during apoptosis induced by camp-tothecin in p53 mutant human colon carcinoma cells. Oncogene 2001;20:1852–1859.

102. Simonian PL, Grillot DA, Nunez G. Bcl-2 and Bcl-XL can differentially block chemother-apy-induced cell death. Blood 1997;90:1208–1216.

103. Miyashita T, Reed JC. Bcl-2 oncoprotein blocks chemotherapy-induced apoptosis in ahuman leukemia cell line. Blood 1993;81:151–157.

104. Yin DX, Schimke RT. Bcl-2 expression delays drug-induced apoptosis but does not increaseclonogenic survival after drug treatment in HeLa cells. Cancer Res 1995;55:4922–4928.

105. Uren AG, Pakusch M, Hawkins CJ, Puls KL, Vaux DL. Cloning and expression of apopto-sis inhibitory protein homologs that function to inhibit apoptosis and/or bind tumor necro-sis factor receptor-associated factors. Proc Natl Acad Sci USA 1996;93:4974–4978.

106. Duckett CS, Nava VE, Gedrich RW, et al. A conserved family of cellular genes related tothe baculovirus IAP gene and encoding apoptosis inhibitors. EMBO J 1996;15:2685–2694.


107. Liston P, Roy N, Tamai K, et al. Suppression of apoptosis in mammalian cells by NAIP anda related family of IAP genes. Nature 1996;379:349–353.

108. Srinivasula SM, Hegde R, Saleh A, et al. A conserved XIAP-interaction motif in Caspase-9and Smac/DIABLO regulates caspase activity and apoptosis. Nature 2001;410:112–116.

109. Lejeune FJ, Ruegg C, Lienard D. Clinical applications of TNF-alpha in cancer. Curr OpinImmunol 1998;10:573–580.

110. Walczak H, Krammer PH. The CD95 (APO-1/Fas) and the TRAIL (APO-2L) apoptosissystems. Exp Cell Res 2000;256:58–66.

111. Gores GJ, Kaufmann SH. Is TRAIL hepatotoxic? Hepatology 2001;34:3–6.112. Ichikawa K, Liu W, Zhao L, et al. Tumoricidal activity of a novel anti-human DR5 mono-

clonal antibody without hepatocyte cytotoxicity. Nat Med 2001;7:954–960.113. Altucci L, Rossin A, Raffelsberger W, Reitmair A, Chomienne C, Gronemeyer H. Retinoic

acid-induced apoptosis in leukemia cells is mediated by paracrine action of tumor-selectivedeath ligand TRAIL. Nat Med 2001;7:680–686.

114. Fenaux P, Chastang C, Chevret S, et al. A randomized comparison of all transretinoic acid(ATRA) followed by chemotherapy and ATRA plus chemotherapy and the role of mainte-nance therapy in newly diagnosed acute promyelocytic leukemia. The European APLGroup. Blood 1999;94:1192–1200.

115. Wen J, Ramadevi N, Nguyen D, Perkins C, Worthington E, Bhalla K. Antileukemic drugsincrease death receptor 5 levels and enhance Apo-2L- induced apoptosis of human acuteleukemia cells. Blood 2000;96:3900–3906.

116. Reed JC, Stein C, Subasinghe C, et al. Antisense-mediated inhibition of BCL2 protoonco-gene expression and leukemic cell growth and survival: comparisons of phosphodiester andphosphorothioate oligodeoxynucleotides. Cancer Res 1990;50:6565–6570.

117. Campos L, Sabido O, Rouault JP, Guyotat D. Effects of BCL-2 antisense oligodeoxynu-cleotides on in vitro proliferation and survival of normal marrow progenitors and leukemiccells. Blood 1994;84:595–600.

118. Gewirtz AM, Sokol DL, Ratajczak MZ. Nucleic acid therapeutics: state of the art and futureprospects. Blood 1998;92:712–736.

119. Cotter FE. Antisense therapy of hematologic malignancies. Semin Hematol 1999;36:9–14.120. Waters JS, Webb A, Cunningham D, et al. Phase I clinical and pharmacokinetic study of

bcl-2 antisense oligonucleotide therapy in patients with non-Hodgkin’s lymphoma. J ClinOncol 2000;18:1812–1823.

121. Banerjee D. Genosense (Genta Inc.). Current Opinion in Investigational New Drugs2001;2:574–580.

122. Estey E, Thall P, Andreeff M, et al. Use of granulocyte colony-stimulating factor before,during, and after fludarabine plus cytarabine induction therapy of newly diagnosedacute myelogenous leukemia or myelodysplastic syndromes: comparison with fludara-bine plus cytarabine without granulocyte colony-stimulating factor. J Clin Oncol1994;12:671–678.

123. Bairey O, Zimra Y, Shaklai M, Okaon E, Rabizadeh E. Bcl-2, Bcl-X, Bax, and Bak expres-sion in short- and long-lived patients with diffuse large B-cell lymphomas. Clin Cancer Res1999;5:2860–2866.

124. Zangemeister-Wittke U, Leech SH, Olie RA, et al. A novel bispecific antisense oligonu-cleotide inhibiting both bcl-2 and bcl-xL expression efficiently induces apoptosis in tumorcells. Clin Cancer Res 2000;6:2547–2555.

125. Gautschi O, Tschopp S, Olie RA, et al. Activity of a novel bcl-2/bcl-xL-bispecific antisenseoligonucleotide against tumors of diverse histologic origins. J Natl Cancer Inst 2001;93:463–471.

126. Horita M, Andreu EJ, Benito A, et al. Blockade of the Bcr-Abl kinase activity induces apopto-sis of chronic myelogenous leukemia cells by suppressing signal transducer and activator oftranscription 5-dependent expression of Bcl-xL. J Exp Med 2000;191:977–984.


127. Townsend KJ, Trust JL, Traupman MA, Eastman A, Craig RW. Expression of the antiapoptoticMCL1 gene product is regulated by a mitogen activated protein kinase-mediated pathway trig-gered through microtubule disruption and protein kinase C. Oncogene 1998;17:1223–1234.

128. Boucher MJ, Morisset J, Vachon PH, Reed JC, Laine J, Rivard N. MEK/ERK signalingpathway regulates the expression of Bcl-2, Bcl-X(L), and Mcl-1 and promotes survival ofhuman pancreatic cancer cells. J Cell Biochem 2000;79:355–369.

129. Bell MP, Huntoon CJ, Graham D, McKean DJ. The analysis of costimulatory receptor sig-naling cascades in normal T lymphocytes using in vitro gene transfer and reporter geneanalysis. Nat Med. 2001;7:1155–1158.

130. Meydan N, Grunberger T, Dadi H, et al. Inhibition of acute lymphoblastic leukaemia by aJak-2 inhibitor. Nature 1996;379:645–648.

131. Towatari M, Iida H, Tanimoto M, Iwata H, Hamaguchi M, Saito H. Constitutive activation ofmitogen-activated protein kinase pathway in acute leukemia cells. Leukemia 1997;11:479–484.

132. Milella M, Kornblau SM, Estrov Z, et al. Therapeutic targeting of the MEK/MAPK signaltransduction module in acute myeloid leukemia. J Clin Invest 2001;108:851–859.

133. Senderowicz AM. Development of cyclin-dependent kinase modulators as novel therapeu-tic approaches for hematological malignancies. Leukemia 2001;15:1–9.

134. Sherr CJ. The Pezcoller lecture: cancer cell cycles revisited. Cancer Res 2000;60:3689–3695.135. Bible KC, Kaufmann SH. Flavopiridol (NSC 649890, L86-8275): a cytotoxic flavone that

induces cell death in human lung carcinoma cells. Cancer Res 1996;56:4856–4861.136. Konig A, Schwartz GK, Mohammad RM, A1-Katib A, Gabrilove JL. The novel cyclin-

dependent kinase inhibitor flavopiridol downregulates Bcl-2 and induces growth arrest andapoptosis in chronic B-cell leukemia lines. Blood 1997;90:4307–4312.

137. Kitada S, Zapata JM, Andreeff M, Reed JC. Protein kinase inhibitors flavopiridol and 7-hydroxy-staurosporine down-regulate antiapoptosis proteins in B-cell chronic lymphocyticleukemia. Blood 2000;96:393–397.

138. Lam LT, Pickeral OK, Peng AC, et al. Genomic-scale measurement of mRNA turnover andthe mechanisms of action of the anti-cancer drug flavopiridol. Genome Biol2001;2:0041.1–0041.11.

139. Senderowicz AM, Sausville EA. Preclinical and clinical development of cyclin-dependentkinase modulators. J Natl Cancer Inst 2000;92:376–387.

140. Bible KC, Kaufmann SH. Cytotoxic synergy between flavopiridol (NSC 649890, L86-8275) and various antineoplastic agents: the importance of sequence of administration.Cancer Res 1997;57:3375–3380.

141. Goekjian PG, Jirousek MR. Protein kinase C inhibitors as novel anticancer drugs. Exp OpinInvest Drugs 2001;10:2117–2140.

142. Li K, Zhang J, Sirois P, Hu Y. Technology evaluation: ISIS-3521. Curr Opin Mol Ther1999;1:393–398.

143. Wang JL, Liu D, Zhang ZJ, et al. Structure-based discovery of an organic compound thatbinds Bcl-2 protein and induces apoptosis of tumor cells. Proc Natl Acad Sci USA2000;97:7124–7129.

144. Holcik M, Yeh C, Korneluk RG, Chow T. Translational upregulation of X-linked inhibitorof apoptosis (XIAP) increases resistance to radiation induced cell death. Oncogene2000;19:4174–4177.

145. Li J, Feng Q, Kim JM, et al. Human ovarian cancer and cisplatin resistance: possible role ofinhibitor of apoptosis proteins. Endocrinology 2001;142:370–380.

146. Guzman ML, Neering SJ, Upchurch D, et al. Nuclear factor-kappaB is constitutively acti-vated in primitive human acute myelogenous leukemia cells. Blood 2001;98:2301–2307.

147. Romano MF, Lamberti A, Bisogni R, et al. Enhancement of cytosine arabinoside-inducedapoptosis in human myeloblastic leukemia cells by NF-kappa B/Rel-specific decoyoligodeoxynucleotides. Gene Ther 2000;7:1234–1237.

148. Chai J, Du C, Wu JW, Kyin S, Wang X, Shi Y. Structural and biochemical basis of apoptoticactivation by Smac/DIABLO. Nature 2000;406:855–862.


IV DIFFERENTIATION AGENTS

1. INTRODUCTION

Despite its historical reputation as a toxin and a poison, arsenic trioxide hasbeen used therapeutically in a variety of diseases as early as the 18th century.The compound known as Fowler’s solution contained a potassium bicarbonate-based arsenic trioxide (ATO) that was in use until the early 20th century (1,2).The fascination with arsenic was enhanced when Paul Ehrlich developed sal-varsan, a formulation that contained organic arsenic used to treat syphilis (3).However, clinicians’ interests in arsenic were abandoned as data accumulatedon its toxic effects (1,4).

The ATO activity in chronic myelogenous leukemia (CML) was reported in1938 by Forkner and Scott (5). However, arsenic lost favor as a therapy for

189


12 ArsenicalsPast, Present, and Future

Chadi Nabhan, MD

and Martin S. Tallman, MD

CONTENTS

INTRODUCTION

MECHANISMS OF ACTION

APOPTOSIS AND MITOCHONDRIAL DAMAGE

EFFECT ON PML/RAR-αEFFECT ON ANGIOGENESIS

ARSENIC TRIOXIDE IN ACUTE PROMYELOCYTIC LEUKEMIA

ARSENIC IN OTHER HEMATOLOGIC MALIGNANCIES

ARSENIC TRIOXIDE IN MULTIPLE MYELOMA

ARSENIC TRIOXIDE IN CHRONIC MYELOGENOUS LEUKEMIA

ARSENIC TRIOXIDE IN OTHER HEMATOLOGIC MALIGNANCIES

ARSENIC IN SOLID TUMORS

ARSENIC TRIOXIDE-ASSOCIATED TOXICITIES

CONCLUSION

REFERENCES

CML when radiation was introduced for the treatment of leukemia. Clinicaluse of arsenic as anticancer therapy ceased in the United States during the1970s, but recent reports of remarkable activity in acute promyelocyticleukemia (APL) and other in vitro data in many different cell lines, redirectedthe attention toward arsenic and has rekindled its therapeutic potential (6,7).This chapter focuses the proposed mechanisms of action of ATO, its role inAPL, and its potential future directions in other disorders.

2. MECHANISMS OF ACTION

Understanding how certain agents work permits one to exploit potential syn-ergism when combined with other therapies. Limited information is availableabout the specific mechanism(s) of action of ATO. Most in vitro studies wereperformed in the NB4 (APL) cell line. Several mechanisms have been pro-posed, including apoptosis induction in several cell lines, partial cellular differ-entiation, inhibition of angiogenesis, degradation of specific APL fusiontranscripts, antiproliferation, and mitochondrial injury (Table 1).

3. APOPTOSIS AND MITOCHONDRIAL DAMAGE

Neoplasia generally occurs when the balance between cell proliferation andcell death is disrupted. In most malignancies, apoptotic pathways are per-turbed. Among these, upregulation of antiapoptotic genes, such as bcl-2, playsan important role. Caspase cascade activation and deactivation frequentlydetermine the final outcome of cells (8,9). Other contributing factors includep53 mutations, deletion of retinoblastoma (RB) gene, and cyclin D2 overex-pression. It has been proposed that ATO exerts its activity through affectingone or more of these pathways (10,11).

Several studies have been conducted in an APL cell line (NB4) showing thatATO enhances apoptosis by downregulating bcl-2 protein expression (12,13). Thisactivity is independent of promyelocytic leukemia/retinoic acid receptor-alpha(PML/RAR-α) expression. The apoptotic activity of ATO occurs at a concentra-tion of 1–2 µmol/L. ATO also activates the caspase cascade in APL cells, whichpromotes proteolysis of intracellular proteins, including poly-ADP-ribose-poly-merase (PARP), completing the apoptotic process (10,14).

190 Nabhan and Tallman

Table 1Potential Mechanisms of Action of Arsenic Trioxide

Apoptosis inductionCaspase cascade activationMitochondrial injuryRelocalization of PML/RARα transcript productAntiangiogenesis effect

Mitochondria are an integral part of the cell machinery and play an impor-tant role in apoptosis. This is, in part related to the function of mitochondria asthe major site of activity of Bax (proapoptotic) and bcl-2 (antiapoptotic) genes,with the balance between both genes contributing to the fate of the cell (15). Inpathologic conditions, apoptotic signals trigger progressive permeabilizationof the mitochondrial membrane through the permeability transition pore com-plex (PTPC). ATO induces such a process, allowing the release of intramito-chondrial compounds such as cytochrome C and other apoptosis-inducingfactors (16,17).

Another mechanism by which ATO induces mitochondrial injury is therelease of reactive oxygen species (ROS) that generally causes loss of mito-chondrial membrane potential (18). Cells that contain higher levels of glu-tathione (GSH) are resistant to ATO-induced apoptosis, and manipulation ofGSH levels has an effect on ATO activity (18,19). This might explain whyNB4-APL cell lines with relatively low levels of reduced GSH, GSH-peroxi-dase, and catalase are sensitive to ATO. Furthermore, ATO inhibits GSH-per-oxidase and increases cellular hydrogen peroxide content (18–20). Additiveeffects and potential synergism have been explored by combining ATO withagents that reduce GSH levels and allows for more ROS generation. Dai andcolleagues showed that by increasing the intracellular levels of hydrogen per-oxide with ascorbic acid, the ATO-induced apoptotic activity in lymphoma celllines increases (19). Gartenhaus and Evens tested a similar concept in a multi-ple myeloma in vitro model to enhance ATO activity (21,22). In addition,recent data suggest that glutathione depletion could render cells that are resis-tant to ATO to become sensitive to that agent (23). These data encourage futurecombination regimens with ATO and other pharmacologic compounds thatdirectly affect the mitochondria or can perturb the redox system.

4. EFFECT ON PML/RAR-αEarly studies demonstrated that the characteristic chromosomal translocation

t(15;17) in the leukemic cells of patients with APL results in the creation offusion gene with a protein product that arrests myeloid cell maturation andinhibits differentiation (24). The remarkable activity of all-trans-retinoic acid(ATRA) in APL is attributed to induction of differentiation through interferencewith the transcript gene (25). ATO also has activity in cellular differentiation inthe bone marrow and peripheral blood of patients with APL (26). This was con-firmed by demonstrating that leukemic cells treated with ATO lose their primi-tive markers that are present before therapy and that after therapy, these cellsexpress markers of mature myeloid cells. On the molecular level, Andre and col-leagues demonstrated that ATO induces the relocalization of PML andPML/RAR-α onto the nuclear body from the cytoplasm in NB4 cell lines and

Chapter 12 / Arsenicals 191

also induces the degradation of these proteins (27). This observation was alsoseen in non-APL myeloid cell lines (28). This relocalization inhibits the activityof PML and promotes maturation. However, many studies revealed that someATO effects are independent of both PML and RAR-α. Shao et al. studied ATOactivity on cell lines that did not express an intact PML/RAR-α fusion proteinand were resistant to ATRA (29). In that study, ATO inhibited cell growth andinduced apoptosis despite absence of the chimeric gene product.

5. EFFECT ON ANGIOGENESIS

Interrupting the blood supply to tumors can inhibit their growth. The forma-tion of new vessels supplying tumor cells has been proposed as a potentialmechanism by which malignant cells grow. Lew et al. studied the effect ofATO on experimental solid tumor system, showing that it produced preferen-tial vascular shutdown and eventually necrosis (30). Proposed mechanisms bywhich ATO exerts this antiangiogenic effect include activation of endothelialcells, upregulation of endothelial cell adhesion molecules, apoptosis ofendothelial cells, and inhibition of vascular endothelial growth factor (VEGF)production (31). Although the proposal that ATO interrupts the vicious circlecreated by leukemic cells producing VEGF and stimulating endothelial cellproduction of many cytokines is attractive, the exact mechanisms of ATO-related antiangiogenesis properties remain under investigation (10).

6. ARSENIC TRIOXIDE IN ACUTE PROMYELOCYTIC LEUKEMIA

Initial reports describing the ATO activity in APL (Table 2) first came fromChina when investigators from the Harbin Medical University observed that 21of 32 patients treated with an ATO-containing compound achieved completeremission (CR) and a 5-yr overall survival of 50% (32). An updated report fromthe same group showed a CR rate in 42 previously treated patients of 52% and aCR rate in 30 untreated patients of 73% (33). An initial dose-escalating studyconducted at Memorial Sloan-Kettering Cancer Center showed remarkableresults (26). In that study, 12 patients with relapsed APL were treated with ATOat doses ranging from 0.06 to 0.2 mg/kg/d until leukemic cells were eliminatedfrom the bone marrow. Of the 12 patients, 11 achieved CR. Furthermore, 8 ofthese 11 patients who were initially positive for the PML/RAR fusion transcriptby reverse transcription-polymerase chain reaction (RT-PCR) tested negativeafter two courses of therapy. The median duration of treatment in the 11 patientsachieving CR was 33 d at a median daily dose of 0.16 mg/kg/d. These resultswere validated in a multicenter trial conducted in the United States and included40 patients with previously treated APL, all of whom had been previouslyexposed to ATRA (34). Thirty-four patients (85%) achieved CR. Of the 40


patients, 2 had resistant disease and 3 died early. Analysis of both the pilot studyand the multicenter trial revealed that the overall survival for all patients was66% at 18 months, with a relapse-free survival of 50% (Fig. 1). These dataclearly demonstrate that ATO is highly effective in patients with relapsed APL.The role of ATO in patients with previously untreated APL is currently beingaddressed in a North American Intergroup Trial in which previously untreatedpatients will receive induction with ATRA and chemotherapy (daunorubicin andcytarabine). Subsequently, all patients in CR are randomized to either twocourses of ATO or not before receiving two courses of consolidation with dau-nourubicin and 1 wk of ATRA.


Table 2Patients With Relapsed and Refractory Acute Promyelocytic Leukemia Achieving

Complete Remission After One Course of Arsenic Trioxide Therapy

Reference N Complete Remission % Complete Remission

Zhang, 1996 (33) 42 22 52Niu, 1999 (65) 47 40 85

25 24 96Soignet, 1998 (26) 12 11 92Soignet, 2000 (77) 40 35 85

Fig. 1. Relapse-free survival in patients with relapsed APL treated with ATO in the US mul-ticenter trial. Reproduced with permission from ref. 34.

There are data to suggest that ATO and ATRA may be synergistic (35,36).ATO decreases ATRA-induced differentiation in APL cell lines but is syner-gistic with ATRA in inducing differentiation in resistant cells. Conversely,ATO-induced apoptosis is less with ATRA pretreatment in NB4 cells (35,37).The synergistic activity was shown in a transgenic mouse model when Lalle-mand-Breitenback demonstrated that mice treated with combined ATRA/ATOhave longer CR than these treated with either agent alone (Fig. 2) (36). Theseobservations prompted studies to exploit potential synergism in humans. In apreliminary report, Dombret and colleagues treated six patients who relapsedafter combination ATRA/chemotherapy with combined ATO/ATRA, whereasfour patients with similar characteristics were treated with ATO alone (38). Nodifferences were observed in toxicity or response. Despite the enhanced activ-ity of ATRA and ATO combined, it remains to be determined if this approachimproves survival in patients who are in relapse or are untreated.

Some investigators explored administering low-dose ATO to reduce toxic-ity with maintaining efficacy. Shen and colleagues reported on 20 cases ofrelapsed APL treated with ATO at 0.08 mg/kg/d intravenously for 28 d,showing 80% CR rate with overall survival and relapse-free survival at61.55% ± 15.79% and 49.11% ± 15.09% at 2 yr, respectively (39). Cardiactoxicity and gastrointestinal side effects were observed less often in the low-dose cohort compared with historical controls. The efficacy was comparable,suggesting that this approach should be studied further in a prospective clin-ical trial.


Fig. 2. Survival curve of leukemic mice left untreated (▲▲) or treated with retinoic acid for28 d (●), arsenic (■■), or both (bold line). Reproduced with permission from ref. 36.

7. ARSENIC IN OTHER HEMATOLOGIC MALIGNANCIES

Because ATO may induce apoptosis independent of PML/RAR-α, it isplausible that ATO could have a broader role in hematologic malignancies. Invitro data in non-APL cell lines complement the known potential mechanismsproposed in APL cells. Huang and colleagues showed that ATO induces cas-pase activation and apoptosis in myeloid non-APL cell lines (14). This alsooccurred in multiple myeloma and in cells from a human megakaryocytic cellline (40–42). As in APL cell lines, angiogenesis is also inhibited by ATO inother non-APL lines, and this is believed to occur through an antiproliferativeeffect on human vascular endothelial cells by inhibiting the production ofVEGF (31). Increased angiogenesis and VEGF production have recently beendescribed in patients with APL (43). In that regard, ATO resembles ATRAbecause they both inhibit angiogenesis, allowing for exploring synergisticeffects of combination regimens. These in vitro data formed the basis for sev-eral phase II clinical trials exploring the efficacy of ATO in patients withhematologic malignancies.

8. ARSENIC TRIOXIDE IN MULTIPLE MYELOMA

The evidence that ATO induces apoptosis in multiple myeloma (MM) celllines (44) and that this activity is not inhibited by interleukin (IL)-6, aknown growth factor in MM, prompted investigators to explore the activityof this agent in MM (45–47). Munshi and colleagues studied the efficacy ofATO in patients with refractory MM. In that study, eligible patients hadrelapsed or resistant MM, at least one previous cycle of high-dose therapywith autologous stem cell transplant, and normal renal, liver, and cardiacfunctions (48,49). Patients in that study received ATO at 0.15 mg/kg/d for 60d. Treatment was continued for an additional 30 d in responding patients. Ofnine evaluable patients, six had chromosome 13 abnormalities, five hadadvanced disease, extensive bone marrow involvement, and seven had two ormore courses of high-dose therapy. Preliminary results showed that of thefour patients who completed more than 30-d infusion, two had more than a50% reduction in their paraprotein, one had stable disease, and one pro-gressed (49). Of the five patients who were treated for less than 30 d, twohad stable disease and three progressed. Although the clinical responses inthat trial were modest, the patient population was heavily pretreated, withthe majority failing more than one autologous stem cell transplantation.Because the toxicities reported in that study were minimal, this agent islikely to be explored in combination regimens in MM and can even be con-sidered for incorporation into front-line therapy in well-designed clinical tri-als, particularly in high-risk patients (48).


9. ARSENIC TRIOXIDE IN CHRONIC MYELOGENOUS LEUKEMIA

Puccetti and colleagues studied the apoptotic effects of ATO in severalleukemic cell lines expressing BCR-ABL fusion protein, a pathognomonic fea-ture of CML (50). This study demonstrated that ATO induces apoptosis in cellstransfected with this protein compared with cells that do not and that it inhibitsthe proliferation of leukemic blasts that are Philadelphia chromosome (Ph)positive without affecting peripheral progenitor cells. ATO had minimal apop-totic activity in cells that did not express Ph (Fig. 3). This observation stronglysuggests that clinical applications of ATO in Ph+ diseases are warranted.

10. ARSENIC TRIOXIDE IN OTHER HEMATOLOGIC MALIGNANCIES

Other studies are being conducted and are currently under way in manyhematologic malignancies. Investigators from China reported potential activityof ATO in 27 patients with malignant lymphoma (14 with Hodgkin’s disease, 9with histiocytosarcoma, 2 with plasmacytoma, and 2 with lympholeukosar-coma), with a remission rate of 48% (51). Although a decade has passed bysince these data were reported, the understanding of how ATO exerts its activi-ties in many different cell lines and the outstanding results in APL renewed the


Fig. 3. The apoptotic effects of ATO on Ph+ cell lines expressing p185 (BCR-Abl) andp210 (BCR-Abl) with respect to Ph- lymphoblastic cell lines. All cells were treated with 2µM of ATO (+ represents treatment, – represents no treatment). Reproduced with permis-sion from ref. 50.

interest in this agent. Recently, Japanese study revealed that ATO inducesapoptosis in a caspase-dependent manner in human-T-leukemia virus cell lines(52). Many US institutions are evaluating ATO in refractory non-Hodgkin’slymphoma (NHL), acute myeloid leukemia (AML), MM, myelodysplasticsyndromes (MDSs) (53,54), and Hodgkin’s disease. Of significance, a phase IItrial combining ATO and α-interferon (IFN) in patients with relapsed and/orrefractory adult T-cell leukemia was recently reported (55). In that study, eightpatients were treated, three with single-agent ATO at 0.15 mg/kg and five witha combination of ATO at 0.15 mg/kg and IFN at 9 million units/d for a maxi-mum duration of 56 d. Median duration of therapy was 20 d, and none of thepatients was able to achieve 56 d of therapy, because of either toxicity (threepatients) or progression (four patients). The results of this study showed oneCR, three PRs, and four progressions. At the time of this report, one patientwas still alive at 20 mo of follow-up. Murgo summarized the current ongoingtrails (56,57). The results on these trials are currently not published.

11. ARSENIC IN SOLID TUMORS

The interest in ATO as an innovative approach in solid tumors was encour-aged by the impressive results in APL and the promising preclinical data. Zhouand colleagues studied the effect of ATO in androgen-dependent and indepen-dent prostate cancer cell lines and showed apoptosis induction (58). Thisobservation is important because it shows that ATO may have activity in theearly stages of prostate cancer and potentially can be used earlier. Tradition-ally, most chemotherapeutic agents have been used as salvage regimens inprostate cancer when patients fail androgen blockade therapy and their diseasebecomes androgen independent. The ATO activity in patients with advancedprostate cancer is currently being studied in the context of a clinical trial, theresults of which will be available soon. In that study, patients will receive ATOat 0.2 mg/kg/d on days 1–5 and 8–12 of a 4-wk cycle (56).

Because renal cell carcinoma is highly vascular, investigators at MemorialSloan-Kettering Cancer Center are exploring the activity of ATO in advanceddisease. The antiangiogenesis properties of ATO might be effective in that set-ting. Patients with advanced renal cell carcinoma will receive ATO at 0.3mg/kg/d on days 1–5 of a 4-wk schedule (56).

Other studies are being conducted in transitional cell carcinoma of the blad-der based on preclinical data of ATO-induced cytotoxicity in these cell lines(59). Zheng and colleagues showed that ATO induces cell death in humanpapilloma virus-infected cervical cell lines (60). The efficacy of ATO inadvanced cervical cancer is currently being tested (56). Some other reports ofactivity were published in abstract form on ATO in melanoma cell lines, ovar-ian cancer, and breast and lung cancer cell lines (61,62).


12. ARSENIC TRIOXIDE-ASSOCIATED TOXICITIES

The most prominent adverse events with ATO in APL have included weightgain and fluid retention, leukocytosis, the APL differentiation syndrome, and pro-longation of the QTc interval on the electrocardiogram (Table 3). Peripheral neu-ropathy, hyperglycemia, and cutaneous reactions have also been described (37,63).

Among the 52 patients treated on the combined pilot and US multicenter tri-als, nausea was observed in 75%, cough in 65%, fatigue in 63%, headache in60%, emesis in 58%, tachycardia in 55%, diarrhea in 53%, hypokalemia in50%, and skin rash in 43% (34). The APL differentiation syndrome is reminis-cent of the retinoic acid syndrome manifested as fever, rash, peripheral edema,pulmonary infiltrates, and pleural and pericardial effusions (64). In general,similar to the therapy for the retinoic acid syndrome, early administration ofdexamethasone is effective when ATO is continued. In addition, hyperleukocy-tosis developed in 55% of patients in some reports, ranging from 11,900 µL to167,000 µL (65,66).

Prolongation of QTc interval on the electrocardiogram and in some situa-tions a suggestion of sudden cardiac death has been reported (67,68). It isimportant that potassium and magnesium levels be kept within normal limitsand that medications known to prolong QTc interval are avoided if at all possi-ble. Among 19 patients with several hematologic malignancies treated withinstitutionally prepared ATO, 3 cases of torsade de pointes were reported (69).In this report, the torsade de pointes occurred early in each case and in onlyone case was the QTc interval reported. In the first case, serum potassium was3.3 meq/L while the magnesium was 1.7 mg/dL. The second case had evenlower potassium value at 2.4 meq/L with a magnesium level at 1.7 mg/dL. Thethird case was the only one with normal electrolyte levels. The calcium valueswere not reported. A second observation was recently reported by Westerveltand colleagues providing details of another institutionally prepared ATO in


Table 3Adverse Effects Observed with ATO in APL Patients

Dose limitingWeight gain and fluid retentionNeuropathy

Most commonSkin reactionHyperglycemiaLeukocytosisProlonged QTc interval

patients with relapsed or refractory APL, of whom three individuals experi-enced sudden cardiac death (68). It is important to note, however, that thesethree patients were all critically ill and intubated, and one had low electrolytevalues, making it difficult to conclude that these fatal events are actually attrib-utable to ATO and not other confounding issues.

Prolongation of QTc interval has been studied extensively in patients withAPL treated with ATO. One-thousand thirty-nine electrocardiograms from 99patients receiving between 5 and 60 d of ATO were studied, and prolongationof QTc interval (defined as greater than 415 msec in men and greater than 479msec in women) was observed in 69% of patients (70). All patients, however,were asymptomatic and the QTc interval returned to baseline between cycles.It has been recommended that ATO be held if the QTc interval is prolonged to500 msec and restarted when the QTc falls to less than 460 msec off the drug.Because some reports showed less cardiac toxicity with low-dose ATO withoutcompromising efficacy, it is crucial to investigate this approach (39).

The mechanism of QTc prolongation induced by ATO is not entirely clear.Some suggest that it could be related to neuropathic injury and an effect on thecardiac sympathetic system (71,72). It needs to be determined if there is a directeffect of ATO on the myocardium. These studies are currently under way.

Limited data exist on ATO in previously untreated patients with APL. Niu andcolleagues reported a CR rate of 72.7% at a median of 35 d, but severe hepato-toxicity occurred in two patients, resulting in fulminant hepatic failure (65). Thespecific cause for the severe hepatic toxicity has not been established.

In some countries, arsenicals have been considered a serious public healthproblem because chronic exposure can cause many health hazards (4). Arsenic ispresent in high concentrations in well water in many parts of the western UnitedStates, South America, and Taiwan (1). Widespread use of arsenic-containingherbicides and pesticides, its use to promote growth of poultry, and its industrialuse caused this compound to be ubiquitous to a certain degree (73). Conse-quently, it is estimated that the average daily human intake of arsenic is about300 µg, almost all of which is ingested with food and water (74). Despite thisenvironmental association, arsenic has never been shown to be a carcinogen(75). The reported cases of possible association with secondary malignancieswere anecdotal and were obtained from experiments in rodent models (76).

13. CONCLUSION

ATO has emerged as the most active agent in patients with relapsed/refrac-tory APL and is clearly the treatment of choice in patients who are resistant toretinoids. This remarkable activity, together with laboratory data suggestingthe possible role in other malignancies, provides the foundation to explorepotential activity of this agent. Mechanisms of action are still being studied,


but abundant evidence exists that arsenic induces apoptosis, causes mitochon-drial damage, and inhibits angiogenesis. Understanding how this compoundacts will promote the investigation of combination therapies that may enhancethe activity. Arsenic is well tolerated, with few serious side effects, allowingfor this agent to emerge as a part of future combination therapies in manymalignancies. It is possible that combining ATRA with arsenic will increasethe likelihood of cure in patients with APL.

REFERENCES1. Antman KH. Introduction: the history of arsenic trioxide in cancer therapy. Oncologist

2001;6:1,2.2. Kwong YL, Todd D. Delicious poison: arsenic trioxide for the treatment of leukemia. Blood

1997;89:3487,3488.3. Aronson SM. Arsenic and old myths. Rhode Island Med 1994;77:233,234.4. Gallagher RE. Arsenic—new life for an old potion. N Engl J Med 1998;339:1389–1391.5. Forkner C, Scott T. Arsenic as a therapeutic agent in chronic myelogenous leukemia. JAMA

1938;97:3–5.6. Sun H, Ma L, Hu X, Zhang T. Treatment of acute promyelocytic leukemia by Ailing-1 ther-

apy with use of syndrome differentiation of traditional chinese medicine. Chin J Comb TradChin Med West Med 1992;12:170–171.

7. Shen ZX, Chen GQ, Ni JH, et al. Use of arsenic trioxide (As2O3) in the treatment of acutepromyelocytic leukemia (APL): II. Clinical efficacy and pharmacokinetics in relapsedpatients. Blood 1997;89:3354–3360.

8. Cryns V, Yuan J. Proteases to die for. Genes Dev 1998;12:1551–1570.9. Akao Y, Yamada H, Nakagawa Y. Arsenic-induced apoptosis in malignant cells in vitro.

Leukemia Lymphoma 2000;37:53–63.10. Waxman S, Anderson KC. History of the development of arsenic derivatives in cancer ther-

apy. Oncologist 2001;6:3–10.11. Kitamura K, Minami Y, Yamamoto K, et al. Involvement of CD95-independent caspase 8

activation in arsenic trioxide-induced apoptosis. Leukemia 2000;14:1743–1750.12. Chen GQ, Zhu J, Shi XG, et al. In vitro studies on cellular and molecular mechanisms of

arsenic trioxide (As2O3) in the treatment of acute promyelocytic leukemia: As2O3 inducesNB4 cell apoptosis with downregulation of Bcl-2 expression and modulation of PML-RARalpha/PML proteins. Blood 1996;88:1052–1061.

13. Wang ZG, Rivi R, Delva L, et al. Arsenic trioxide and melarsoprol induce programmed celldeath in myeloid leukemia cell lines and function in a PML and PML-RARalpha indepen-dent manner. Blood 1998;92:1497–1504.

14. Huang XJ, Wiernik PH, Klein RS, Gallagher RE. Arsenic trioxide induces apoptosis ofmyeloid leukemia cells by activation of caspases. Med Oncol 1999;16:58–64.

15. Crompton M. Mitochondrial intermembrane junctional complexes and their role in celldeath. J Physiol 2000;529(pt 1). 11–21.

16. Cai X, Shen YL, Zhu Q, et al. Arsenic trioxide-induced apoptosis and differentiation areassociated respectively with mitochondrial transmembrane potential collapse and retinoicacid signaling pathways in acute promyelocytic leukemia. Leukemia 2000;14:262–270.

17. Larochette N, Decaudin D, Jacotot E, et al. Arsenite induces apoptosis via a direct effect onthe mitochondrial permeability transition pore. Exp Cell Res 1999;249:413–421.

18. Jing Y, Dai J, Chalmers-Redman RM, Tatton WG, Waxman S. Arsenic trioxide selectivelyinduces acute promyelocytic leukemia cell apoptosis via a hydrogen peroxide-dependentpathway. Blood 1999;94:2102–2111.


19. Dai J, Weinberg RS, Waxman S, Jing Y. Malignant cells can be sensitized to undergo growthinhibition and apoptosis by arsenic trioxide through modulation of the glutathione redox sys-tem. Blood 1999;93:268–277.

20. Yang CH, Kuo ML, Chen JC, Chen YC. Arsenic trioxide sensitivity is associated with lowlevel of glutathione in cancer cells. Br J Cancer 1999;81:796–799.

21. Gartenhaus R, Prachand S, Paniaqua M, Li Y, Gordon L. Arsenic trioxide cytotoxicity insteroid and chemotherapy-resistant myeloma cell lines: enhancement of apoptosis by manip-ulatioon of cellular redox state. Clin Cancer Res, 2002:8(2);566–572.

22. Evens A, Prachard S, Li Y, Paniaqua M, Gordon L, Gartenhaus R. In vitro potentiation ofarsenic trioxide cytotoxicity through novel combination therapy with Imexon for multiplemyeloma [abstract 1580]. Blood 2001;98:375a.

23. Davison K, Miller W. Gluthathione depletion restores arsenic-sensitivity to arsenic-resistantAPL cells [abstract 4222]. Proc AACR 2001;2:786.

24. de The H, Lavau C, Marchio A, Chomienne C, Degos L, Dejean A. The PML-RAR alphafusion mRNA generated by the t(15;17) translocation in acute promyelocytic leukemiaencodes a functionally altered RAR. Cell 1991;66:675–684.

25. Huang ME, Ye YC, Chen SR, et al. Use of all-trans retinoic acid in the treatment of acutepromyelocytic leukemia. Blood 1988;72:567–572.

26. Soignet SL, Maslak P, Wang ZG, et al. Complete remission after treatment of acute promye-locytic leukemia with arsenic trioxide. N Engl J Med 1998;339:1341–1348.

27. Andre C, Guillemin MC, Zhu J, et al. The PML and PML/RARalpha domains: from autoimmu-nity to molecular oncology and from retinoic acid to arsenic. Exp Cell Res 1996;229:253–260.

28. Zhu J, Koken MH, Quignon F, et al. Arsenic-induced PML targeting onto nuclear bodies:implications for the treatment of acute promyelocytic leukemia. Proc Natl Acad Sci USA1997;94:3978–3983.

29. Shao W, Fanelli M, Ferrara FF, et al. Arsenic trioxide as an inducer of apoptosis and loss ofPML/RAR alpha protein in acute promyelocytic leukemia cells. J Natl Cancer Inst1998;90:124–133.

30. Lew YS, Brown SL, Griffin RJ, Song CW, Kim JH. Arsenic trioxide causes selective necro-sis in solid murine tumors by vascular shutdown. Cancer Res 1999;59:6033–6037.

31. Roboz GJ, Dias S, Lam G, et al. Arsenic trioxide induces dose- and time-dependent apopto-sis of endothelium and may exert an antileukemic effect via inhibition of angiogenesis.Blood 2000;96:1525–1530.

32. Sun H, Ma L, Hu X. Treatment of acute promyelocytic leukemia by ailing-1 therapy with useof syndrom differentiation of traditional chinese medicine. Chin J Comb Trad Chin MedWest Med 1992;12:170–171.

33. Zhang P, Wang S, Hu X. Arsenic trioxide treated 72 cases of acute promyelocytic leukemia.Chin J Hematol 1996;17:58–62.

34. Soignet SL, Frankel SR, Douer D, et al. United States multicenter study of arsenic trioxide inrelapsed acute promyelocytic leukemia. J Clin Oncol 2001;19:3852–3860.

35. Jing Y, Wang L, Xia L, et al. Combined effect of all-trans retinoic acid and arsenic trioxide inacute promyelocytic leukemia cells in vitro and in vivo. Blood 2001;97:264–269.

36. Lallemand-Breitenbach V, Guillemin MC, Janin A, et al. Retinoic acid and arsenic synergizeto eradicate leukemic cells in a mouse model of acute promyelocytic leukemia. J Exp Med1999;189:1043–1052.

37. Tallman M. Arseni trioxide: its role in acute promyelocytic leukemia and potential in otherhematologic malignancies. Blood Rev 2001;15:133–142.

38. Dombret H, Rousselot P, Hermine O. Treatment of relapsing acute promyelocytic leukemiawith arsenic trioxide with or without ATRA: the French AS98 Study [abstract 3578]. Blood2000;96:828a.


39. Shen Y, Shen ZX, Yan H, et al. Studies on the clinical efficacy and pharmacokinetics of low-dose arsenic trioxide in the treatment of relapsed acute promyelocytic leukemia: a compari-son with conventional dosage. Leukemia 2001;15:735–741.

40. Zhang W, Ohnishi K, Shigeno K, et al. The induction of apoptosis and cell cycle arrest byarsenic trioxide in lymphoid neoplasms. Leukemia 1998;12:1383–1391.

41. Lu M, Levin J. Sulpice E, et al. Effect of arsenic trioxide on viability, proliferation, andapoptosis in human megakaryocytic leukemia cell lines. Exp Hematol 1999;27:845–852.

42. Alemany M, Levin J. The effects of arsenic trioxide (As2O3) on human megakaryocyticleukemia cell lines. With a comparison of its effects on other cell lineages. Leukemia Lym-phoma 2000;38:153–163.

43. Kini AR, Peterson LA, Tallman MS, Lingen MW. Angiogenesis in acute promyelocyticleukemia: induction by vascular endothelial growth factor and inhibition by all-trans retinoicacid. Blood 2001;97:3919–3924.

44. Bajenova RPO, Feinman R. Arsenic compounds induce apoptosis in multiple myeloma: roleof NFKb-RIP pathway [abstract 1383]. Blood 1999;94:309a.

45. Tang B, Bajenova O, Siegel-Feinman R. Arsenic compounds induce apoptosis in multiplemyeloma (MM), activate pro-caspase-3 but do not affect BCL2 family members. Blood1998;92:638a.

46. Rousselot P, Labaume S, Marolleau JP, et al. Arsenic trioxide and melarsoprol induce apop-tosis in plasma cell lines and in plasma cells from myeloma patients. Cancer Res1999;59:1041–1048.

47. Park WH, Seol JG, Kim ES, et al. Arsenic trioxide-mediated growth inhibition in MC/CARmyeloma cells via cell cycle arrest in association with induction of cyclin-dependent kinaseinhibitor, p21, and apoptosis. Cancer Res 2000;60:3065–3071.

48. Munshi NC. Arsenic trioxide: an emerging therapy for multiple myeloma. Oncologist2001;6:17–21.

49. Munshi N, Desikan R, Zangari M. Marked anti-tumor effect of arsenic trioxide in high-riskrefractory multiple myeloma. Blood 1999;94:123a.

50. Puccetti E, Guller S, Orleth A, et al. BCR-ABL mediates arsenic trioxide-induced apoptosisindependently of its aberrant kinase activity. Cancer Res 2000;60:3409–3413.

51. Li YS, Zhang TD, Li CW, et al. [Traditional Chinese and Western medicine in the treatmentof 27 patients with malignant lymphoma]. Zhonghua Zhong Liu Za Zhi 1988;10:61–62.

52. Kenji I, Ryuji I, Atae U, Kimiharu U, Shuichi H. Arsenic trioxide induces apoptosis inHTLV-1 infected T-cell lines and fresh adult T-cell leukemia cells through caspase-depen-dent pathway [abstract 1474]. Blood 2000;96;341a.

53. Dutcher J, Wiernik P, Novik Y. Major hematologic response in a patient with myelodysplasia(MDS) to arsenic trioxide [abstract 4856]. Blood 2000;96;2606.

54. Donelli A, Chiodino C, Panissidi T, Roncaglia R, Torelli G. Might arsenic trioxide be useful inthe treatment of advanced myelodysplastic syndromes? Haematologica 2000;85:1002–1003.

55. Hermine O, Dombert H, Poupon J, Arnulf B, Damaj F. Phase II trial of arsenic trioxide andcombination of alpha interferon and arsenic in patients with relapsed/refractory adult T-cellleukemia (ATL) [abstract 1478]. Blood 2000;96;342a.

56. Murgo AJ. Clinical trials of arsenic trioxide in hematologic and solid tumors: overview ofthe National Cancer Institute Cooperative Research and Development Studies. Oncologist2001;6:22–28.

57. Murgo AJ, McBee WL, Cheson BD. Clinical trials referral resource. Clinical trials witharsenic trioxide. Oncology (Huntingt) 2000;14:206, 211, 215–216 passim.

58. Zhou D-C, Gao M, Ferrari A. Arsenic trioxide is an equipotent inducer of apoptosis in andro-gen-dependent and androgen-independent LNCaP prostate cancer cell sublines [abstract3999]. Proc AACR 1998;39:588.

59. Yang C, Wang T, Chen Y. Cytotoxicity of arsenic trioxide in cancer cell lines [abstract 1552].Proc AACR 1998;39:227.


60. Zheng J, Deng YP, Lin C, Fu M, Xiao PG, Wu M. Arsenic trioxide induces apoptosis ofHPV16 DNA-immortalized human cervical epithelial cells and selectively inhibits viral geneexpression. Int J Cancer 1999;82:286–292.

61. Islam M, Kirkwood J. Arsenic trioxide induces apoptosis of human melanoma cell lines in-vitro [abstract 1435]. Proc ASCO 2001;360a.

62. Zeng L. Arsenic trioxide-induced apoptosis in breast and lung cancer cell lines [abstract2350]. Proc AACR 2001;437.

63. Che-Pin L, Huang MJ, Chang IY, Lin WY, Sheu YT. Retinoic acid syndrome induced byarsenic trioxide in treating recurrent all-trans retinoic acid resistant acute promyelocyticleukemia. Leukemia Lymphoma 2000;38:195–198.

64. Tallman MS, Andersen JW, Schiffer CA, et al. Clinical description of 44 patients with acutepromyelocytic leukemia who developed the retinoic acid syndrome. Blood 2000;95:90–95.

65. Niu C, Yan H, Yu T, et al. Studies on treatment of acute promyelocytic leukemia with arsenictrioxide: remission induction, follow-up, and molecular monitoring in 11 newly diagnosedand 47 relapsed acute promyelocytic leukemia patients. Blood 1999;94:3315–3324.

66. Camacho LH, Soignet SL, Chanel S, et al. Leukocytosis and the retinoic acid syndrome inpatients with acute promyelocytic leukemia treated with arsenic trioxide. J Clin Oncol2000;18:2620–2625.

67. Ohnishi K, Yoshida H, Shigeno K, et al. Prolongation of the QT interval and ventriculartachycardia in patients treated with arsenic trioxide for acute promyelocytic leukemia. AnnIntern Med 2000;133:881–885.

68. Westervelt P, Brown RA, Adkins DR, et al. Sudden death among patients with acute promye-locytic leukemia treated with arsenic trioxide. Blood 2001;98:266–271.

69. Unnikrishnan D, Dutcher JP, Varshneya N, et al. Torsades de pointes in 3 patients withleukemia treated with arsenic trioxide. Blood 2001;97:1514–1516.

70. Barbey JT, Soignet S. Prolongation of the QT interval and ventricular tachycardia in patientstreated with arsenic trioxide for acute promyelocytic leukemia. Ann Intern Med2001;135:842.

71. Roden DM, Lazzara R, Rosen M, Schwartz PJ, Towbin J, Vincent GM. Multiple mecha-nisms in the long-QT syndrome. Current knowledge, gaps, and future directions. The SADSFoundation Task Force on LQTS. Circulation 1996;94:1996–2012.

72. Viskin S. Long QT syndromes and torsade de pointes. Lancet 1999;354:1625–1633.73. Klaassen C. Heavy metals and heavy-metal antagonists. In: Hardman G, Limbrid, eds.

Goodman & Gilman’s The Pharmacological Basis of Therapeutics. New York: McGraw-Hill; 1996:1649–1672.

74. Landrigan PJ. Arsenic—state of the art. Am J Ind Med 1981;2:5–14.75. Novick SC, Warrell RP Jr. Arsenicals in hematologic cancers. Semin Oncol 2000;27:495–501.76. Chan P, Huff J. Arsenic carcinogenesis in animals and in humans: mechanistic, experimenta,

and epidemiological evidence. J Environ Sci Health 1997;C15:83–122.77. Soignet S, Frankel S, Tallman M, Dour D, Scheinberg D. Arsenic trioxide (ATO) in relapsed

acute promyelocytic leukemia (APL): the combined results and follow-up from the US pilotand multicenter trials [abstract 3575]. Blood 2000;96:827a.


1. INTRODUCTION

All-trans-retinoic acid (ATRA) can induce complete remission (CR) in mostpatients with acute promyelocytic leukemia (APL) through in vivo differentia-tion of APL blasts. However, it cannot eliminate the leukemic clone and must

205


13 All-Trans-Retinoic Acid in the Treatment of Acute Promyelocytic Leukemia

Pierre Fenaux, MD, PhD

and Laurent Degos, MD, PhD

CONTENTS

INTRODUCTION

BACKGROUND: RESULTS OF CHEMOTHERAPY ALONE IN

ACUTE PROMYELOCYTIC LEUKEMIA

FIRST RESULTS OBTAINED WITH ALL-TRANS-RETINOIC ACID

ALONE IN ACUTE PROMYELOCYTIC LEUKEMIA

ALL-TRANS-RETINOIC ACID FOLLOWED BY INTENSIVE

CHEMOTHERAPY IN NEWLY DIAGNOSED ACUTE

PROMYELOCYTIC LEUKEMIA

CURRENT ISSUES CONCERNING TREATMENT COMBINING

ALL-TRANS-RETINOIC ACID AND CHEMOTHERAPY IN

NEWLY DIAGNOSED ACUTE PROMYELOCYTIC LEUKEMIA

MAINTENANCE TREATMENT IN ACUTE PROMYELOCYTIC

LEUKEMIA

PROGNOSTIC FACTORS IN PATIENTS TREATED

WITH ALL-TRANS-RETINOIC ACID AND CHEMOTHERAPY

TREATMENT OF RELAPSING ACUTE PROMYELOCYTIC LEUKEMIA

ALL-TRANS-RETINOIC ACID SYNDROME AND OTHER SIDE

EFFECTS

REFERENCES

be used in combination with anthracycline-based chemotherapy. It has nowbeen demonstrated that the combination of ATRA and chemotherapy gave bet-ter survival than chemotherapy alone in newly diagnosed APL, due to fewerrelapses and to a slightly higher CR rate. It is also probable that maintenancetreatment with ATRA, and possibly with low-dose chemotherapy, can furtherreduce the incidence of relapse. Overall, more than 90% newly diagnosedpatients with APL can achieve CR, and approx 75% can be cured by the com-bination of ATRA and chemotherapy. ATRA syndrome remains the majorcombination of ATRA treatment, which should be prevented by addition ofchemotherapy and/or dexamethasone in case of raising white blood cell(WBC) counts.

Until the late 1980s, intensive cytoreductive chemotherapy, usually combin-ing an anthracycline and cytosine arabinoside (AraC) was the only effectivetreatment for APL, likewise for other types of acute myeloid leukemia (AML).The demonstration that ATRA could differentiate APL blasts both in vitro and invivo has greatly improved the therapeutic approach for APL. ATRA and otherretinoids, on the other hand, have limited efficacy in other types of leukemias.

2. BACKGROUND: RESULTS OF CHEMOTHERAPY ALONE IN ACUTE PROMYELOCYTIC LEUKEMIA

APL is a specific type of AML, characterized by the morphology of blastcells (abnormal promyelocyte) by t(15;17) translocation, which fuses the PMLgene on chromosome 15 to the retinoid acid receptor (RAR) α on chromosome17 and by a coagulopathy combining disseminated intravascular coagulation(DIC) and fibrinolysis (1,2).

Using anthracycline AraC regimens, CR rates of only 50% to 60% had gen-erally been reported in the 1970s, but results subsequently improved, and CRrates of 70% to 80% were reported in the 1980s (3–5). Failure to achieve CR inearly reports was mainly due to central nervous system (CNS) bleeding duringthe first days of treatment or to sepsis during the phase of aplasia, whereasresistant leukemia was generally seen in less than 10% of the patients, proba-bly reflecting the high sensitivity of APL cells to anthracyclines. Of note is thatit remained unclear if anthracycline-AraC combinations were superior toanthracyclines alone if the latter were given at high dose, e.g., at least 300mg/m2 during induction for daunorubicin. Other anthracyclines, includingzorubicin or idarubicin, were at least as effective as daunorubicin in APL,whereas Amsa was probably less effective than anthracyclines (6).

Significant coagulopathy, present at diagnosis in 80% of APL cases, is wors-ened (or triggered in the remaining patients) by the onset of chemotherapy.Intensive platelet support during chemotherapy is crucial in the management ofAPL coagulopathy and clearly reduces the incidence of hemorrhagic deaths.

206 Fenaux and Degos

Intensive platelet support is especially important in patients with hyperleuko-cytosis, who have an increased risk of early death due to bleeding. By contrast,the role of heparin and antifibrinolytic agents is unproven (5,7).

The optimal postinduction chemotherapy remains controversial in APL. InAML as a whole, it is proved that intensive consolidation chemotherapy gener-ally improves outcome, by comparison to milder consolidation courses fol-lowed by prolonged maintenance therapy. However, in APL, two studies havesuggested that prolonged maintenance chemotherapy with 6 mercaptopurineand methotrexate can prolong remissions when compared with shorter consoli-dation regimens (8,9).

Age older than 50 yr (9), hyperleukocytosis at diagnosis (5), microgranularAPL variant, severe bleeding at diagnosis, and or major thrombocytopeniawere associated with a higher risk of early death in newly diagnosed patientswith APL treated with chemotherapy alone. Shorter remissions were seen inpatients with hyperleukocytosis (3) and in patients with microgranular APLvariant (9).

Thus, anthracycline-AraC regimens with sufficient anthracycline dosage,associated with intensive platelet support during induction, yielded CR in 75%to 80% of newly diagnosed patients with APL, with a risk of early death due tobleeding of approx 10% to 15%. With anthracycline-based consolidation andpossibly maintenance chemotherapy, median CR duration ranged from 11 to25 mo so that only 35% to 45% of the patients could be cured by chemother-apy alone. Patients with high leucocyte counts had a particularly poor progno-sis with chemotherapy alone, because their CR rates were only 50% to 60%and their risk of relapse was high.

3. FIRST RESULTS OBTAINED WITH ALL-TRANS-RETINOICACID ALONE IN ACUTE PROMYELOCYTIC LEUKEMIA

In the first reports of ATRA therapy (10–13), CR rates of approx 90% werereported in newly diagnosed and first relapse APL, generally with a 45mg/m2/d dose of ATRA. The presence of Auer rods in neutrophils, the absenceof aplasia, and the study of X chromosome-linked polymorphisms showed thatresponse was not obtained by cytotoxicity but by differentiation of APL blastsinto neutrophils, leading to progressive replacement of leukemichematopoiesis by normal polyclonal hematopoiesis (10–12,14,15). Recentfindings from our group also showed a correlation between in vitro differentia-tion of APL blasts by ATRA and with clinical results obtained in vivo with thisdrug (16). Rapid improvement of coagulopathy, instead of the initial worsen-ing observed with conventional chemotherapy, was also seen.

These experiences, however, showed two major drawbacks to ATRA treat-ment. The first, mainly in newly diagnosed APL, was a rapid rise in leukocytes

Chapter 13 / ATRA in APL Treatment 207

in one-third to one-half of the patients, accompanied by clinical signs of“ATRA syndrome,” which proved fatal in some patients (11,17,18). Additionof intensive anthracycline-AraC chemotherapy reduced leukocyte counts andallowed most patients to enter CR (17). High-dose dexamethasone also had afavorable effect on ATRA syndrome. ATRA therapy was also associated withdevelopment of resistance to this drug: patients who achieved CR with ATRAand received either ATRA alone or low-dose chemotherapy for maintenancetherapy generally relapsed within a few months of CR achievement (11,19).These findings led clinicians to administer a treatment that combined ATRAand intensive chemotherapy in APL.

4. ALL-TRANS-RETINOIC ACID FOLLOWED BY INTENSIVE CHEMOTHERAPY IN NEWLY

DIAGNOSED ACUTE PROMYELOCYTIC LEUKEMIA

Nonrandomized studies and two randomized trials have demonstrated thesuperiority of combined treatment with ATRA and intensive chemotherapyover intensive chemotherapy alone in newly diagnosed APL.

4.1. Nonrandomized StudiesThese studies are summarized in Table 1. In the first study, 26 newly diag-

nosed cases of APL treated with ATRA until CR, followed by three courses ofdaunorubicin-AraC were compared with a historical control group treated bychemotherapy alone (17,20). ATRA followed by chemotherapy slightly (butnot significantly) improved the CR rate and sharply reduced the number ofearly relapses occurring within 18 mo of CR achievement, whereas the numberof late relapses was similar to that seen after chemotherapy alone (20). Thoseresults have been largely confirmed (Table 1).

3.2. Randomized StudiesA European trial (APL 91) comparing chemotherapy alone (three courses of

daunorubicin and AraC) and ATRA followed by the same chemotherapy innewly diagnosed APL was started in 1991. In the ATRA group, the firstchemotherapy course was rapidly added to ATRA if WBC were greater than5000 mm3 at diagnosis or increased during treatment. The trial was prema-turely stopped after 18 mo, because event-free survival (EFS) was significantlybetter in the ATRA group (20,21). The last interim analysis, performed 73 moafter closing date of the study, confirmed the significantly higher actuarialEFS, relapse rate, and survival rate in the ATRA group (Table 1) (22). Theseresults confirmed that the combination of ATRA and chemotherapy reducedthe incidence of early relapses without increasing the overall incidence of laterrelapses by comparison with chemotherapy alone. Of note, however, was the


Table 1Published Experience Comparing All-Trans-Retinoic Acid Followed by Intensive Chemotherapy

(and Chemotherapy Alone), in Newly Diagnosed Acute Promyelocytic Leukemia

Complete Comparison with Chemotherapy Alone

No. of Remission Type of Results of Authors Patients Rate, % Follow-up Comparison Chemotherapy p

Fenaux et al. 26 96 Historical (29 cases) Complete remission (CR) NS(1992) (17) rate: 76%

EFS: 62% at 4 yr EFS: 28% at 4 yr <0.01DFI: 70% at 4 yr DFI: 42% at 4 yr <0.05survival: 77% at 4 yr survival: 40% at 4 yr <0.01

Warrell et al. 49 85 Median CR duration Historical (80 cases) Median CR duration 14 mo 0.008(1994) (25) not reached (3+– Median survival 17 mo 0.0009

38+mo) median survival not reached

Kanamaru et al. 109 89 Historical (64 cases) CR rate: 70% 0.004(1995) (47) EFS at 23 mo: 75% EFS at 23 mo: 48% 0.0007

DFS at 23 mo: 81% DFS at 23 mo: 65% 0.07Fenaux et al. 54 91 EFS: 63% at 4 yr Randomized (47 cases) CR rate: 81% NS

(1993, 1994, Relapse: 31% at 4 yr* EFS: 17% at 4 yr* 10–4

and 2000) Relapse: 78% at 4 yr 0.03(20,21,22)(successive updates) Survival: 76% at 4 yr* Survival: 49% at 4 yr*

Tallman et al. 172 72 DFS: 67% at 3 yr Randomized (174 cases) CR rate: 69% NS(1997) (23) Survival: 77% at 3 yr DFS: 32% at 3 yr <0.001

Survival: 50% at 3 yr <0.001

EFS=Event-free survival; DFI = disease-free interval; DFS = disease-free survival.

* Results updated at the reference date of January 1, 2000.

209

occurrence of two late relapses (at 58 and 74 mo, respectively) in patientstreated with ATRA combined with chemotherapy, whereas no relapse was seenbeyond 43 mo after chemotherapy alone. It is therefore important to determine,with longer follow-up of other published APL series, if a few late relapses canindeed occur after ATRA combined with chemotherapy and whether mainte-nance treatment can reduce or eliminate them.

A US intergroup study randomized ATRA followed by chemotherapy andchemotherapy alone in newly diagnosed APL, and patients who achieved CRwere further randomized between no maintenance and ATRA maintenance (seefollowing). The CR rate was similar in patients who received and did notreceive ATRA for induction, but the incidence of relapse was significantlylower in patients who received ATRA during induction, compared with thosewho received chemotherapy alone, and this translated into survival differences(23) (Table 1).

Recently, it has been proved that the addition of ATRA to chemotherapy cansomewhat increase CR rates compared with chemotherapy alone. Indeed, tworecently closed European trials of ATRA combined with chemotherapy innewly diagnosed APL, which overall included approx 1000 patients, showedCR rates of 92% and 93%, respectively (20,24). Thus, with better knowledgeof the use of ATRA (and especially of the prophylaxis of its major side effect,the ATRA syndrome) high CR rates above 90% can be achieved by combiningthis drug with chemotherapy in newly diagnosed APL. By contrast, CR ratesabove 80% have rarely been reported in newly diagnosed APL treated withchemotherapy alone.

Finally, it was shown that treatment with ATRA led to significantly fewerplatelet and red blood cell (RBC) transfusions, less days with fever and withantibiotics, shorter hospital stay, and overall lower cost than treatment withchemotherapy (25).

5. CURRENT ISSUES CONCERNING TREATMENT COMBININGALL-TRANS-RETINOIC ACID AND CHEMOTHERAPY

IN NEWLY DIAGNOSED ACUTE PROMYELOCYTIC LEUKEMIA

5.1. Duration and Dosing of All-Trans-Retinoic Acid During Induction Treatment

Most centers administer ATRA until achievement of hematologic CR, whichusually occurs after 40 to 60 d if ATRA is used alone and after fewer than 30 dif chemotherapy is combined with ATRA from the onset. A British MedicalResearch Council randomized study compared a short course (5 d) of ATRAfollowed by chemotherapy to a long course of ATRA (until CR achievement)combined with the same chemotherapy in newly diagnosed APL. A better out-come was seen in the latter group, demonstrating that longer ATRA adminis-


tration during induction is required (26). It is, however, unknown whether dis-continuation of ATRA before CR achievement (e.g., after 15 to 20 d of treat-ment) is sufficient to achieve full activity of the drug, particularly in reductionof the incidence of relapses.

Although ATRA is generally used at a dose of 45 mg/m2/d, it has beenshown that lower doses, i.e., 25 mg/m2/d, gave similar CR results (27). Theselower doses are often applied in children when severe headache due to ATRAdevelops (28). It is not certain, however, if the additive or synergistic effectobtained with ATRA at 45 mg/m2/d and chemotherapy on reducing the inci-dence of relapses in APL would be similar with lower doses of ATRA.

5.2. Scheduling of All-Trans-Retinoic Acid and Chemotherapy in Acute Promyelocytic Leukemia

The original combinations of ATRA and chemotherapy in APL used ATRAalone until CR achievement, followed by chemotherapy. The European group(APL 93 trial) randomized newly diagnosed APL patients with WBC counts≤ 5000 mm3 between ATRA followed by chemotherapy (ATRA→chemother-apy) and ATRA plus chemotherapy (ATRA + chemotherapy, where chemother-apy was started on day 3 of ATRA treatment). The CR rate was similar in thetwo groups, but relapses at 2 yr were significantly less frequent in the ATRA +chemotherapy group (69). This suggested that the “additive” or “synergistic”effect of ATRA and chemotherapy on reducing the incidence of relapse in APLwas optimal when the two treatment modalities were administered together.Also of note was that early addition of chemotherapy to ATRA reduced theincidence of ATRA syndrome (see following section).

5.3. Role of AraC in Acute Promyelocytic Leukemia ChemotherapyFor treatment with chemotherapy alone, it is unclear whether it is useful to

combine AraC to anthracyclines in APL treatment, when chemotherapy iscombined to ATRA. Recently, Sanz et al. (29) treated 100 cases of newly diag-nosed APL with ATRA and four successive courses of chemotherapy withidarubicin or mitoxantrone alone, followed by maintenance treatment withintermittent ATRA and low-dose continuous chemotherapy. The CR rate was89%, and the 2-yr EFS was 79%. After the end of chemotherapy, 93% of thepatients were polymerase chain reaction (PCR) negative. No mortality and lim-ited morbidity were seen during consolidation chemotherapy, whereas a 5%mortality was observed by the European APL group after consolidationchemotherapy with two daunorubicin-AraC courses (69). Updated results ofthis study, with longer follow up, did not show a increased incidence of laterelapses (30). Estey et al. (31) also obtained favorable results in APL usingATRA followed by idarubicin without AraC in 43 patients. The European APLgroup is currently randomizing newly diagnosed patients with APL between


treatment with ATRA + daunorubicin and treatment with ATRA + daunoru-bicin + AraC.

6. MAINTENANCE TREATMENT IN ACUTE PROMYELOCYTIC LEUKEMIA

Two randomized studies strongly support that maintenance treatment has abeneficial role in newly diagnosed APL treated with ATRA and consolidationchemotherapy. The US intergroup randomized patients who received ATRAfollowed by three daunorubicin-AraC courses to continuous maintenance withATRA (45 mg/m2/d) during 1 yr or no maintenance (23). The incidence ofrelapse was significantly lower in patients who received maintenance ATRA(10 of 46 cases) than in patients who received no maintenance (21 of 54patients). Liver toxicity of the treatment was, however, relatively important.Patients who had received no ATRA during induction therapy also benefitedfrom ATRA maintenance. In addition, patients who received chemotherapyalone followed by ATRA maintenance had a similar outcome to patients whoreceived ATRA followed by chemotherapy but no maintenance ATRA (23).

The European APL group (APL 93 trial) randomized patients who hadachieved CR with a combination of ATRA and chemotherapy (ATRA→chemotherapy or ATRA + chemotherapy) between no maintenance, mainte-nance with ATRA (45 mg/m2/d) 15 d every 3 mo, continuous low-dosechemotherapy with 6 mercaptopurine (6MP) and methotrexate, or both during2 yr, using a 2-by-2 factorial design (69). The rationale for intermittent ratherthan continuous ATRA for maintenance was based on pharmakokinetic studiesshowing a progressive decrease of serum peak levels of ATRA after a fewweeks of treatment, due to hypercatabolism of the drug (32–34).

This hypercatabolism was reversible a few weeks after drug discontinua-tion (34). The rationale for low-dose maintenance chemotherapy was basedon two nonrandomized studies. The incidence of relapse after 2 yr was 25%in patients who received no maintenance ATRA vs 13% in those whoreceived maintenance ATRA (p = 0.02) and 27% in patients who received nomaintenance chemotherapy vs 11% in those who received maintenancechemotherapy (p = 0.0003). Furthermore, an additive effect of intermittentATRA and low-dose chemotherapy in reducing the risk of relapse was seen.Regarding survival, the effect was significant for maintenance withchemotherapy but was only borderline for ATRA, at least with the currentfollow up (20). In addition, patients with WBC counts higher than5000/mm3 who remain at higher risk of relapse after ATRA and intensivechemotherapy benefitted particularly from maintenance with bothchemotherapy and ATRA. Finally, liver toxicity (and other toxicities) weremoderate with intermittent ATRA.


7. PROGNOSTIC FACTORS IN PATIENTS TREATED WITH ALL-TRANS-RETINOIC ACID AND CHEMOTHERAPY

Although a large improvement in outcome has been observed with the com-bination of ATRA and chemotherapy, not all patients with newly diagnosedAPL achieve CR and some patients still relapse.

7.1. Prognostic Complete Remission Achievement FactorsFive percent to 10% of patients with APL fail to achieve CR with ATRA and

chemotherapy, almost exclusively due to early death, because resistance toATRA is exceptional (probably less than 1/500) in cytogenetically (t[15;17]) ormolecularly (PML-RARα rearrangement) confirmed APL (20,24). Majorcauses of early death include CNS bleeding, ATRA syndrome (see followingsection), and sepsis, the latter usually occurring during the chemotherapy-induced aplasia phase. High WBC counts and older age remain the major riskfactors for early death in APL (69).

7.2. Prognostic Relapse FactorsUsing a combination of ATRA and chemotherapy followed by maintenance

treatment, approx 10% to 15% of patients with APL still relapse. Both pretreat-ment factors and evaluation of minimal residual disease (MRD) by reversetranscription (RT)-PCR analysis of PML-RARα rearrangement contribute toevaluation of the relapse risk.

7.2.1. PRETREATMENT FACTORS

Pretreatment factors include high WBC counts in all reports (Table 1) (5),even if, as seen in previous section, results of the European APL group suggestthat maintenance treatment may particularly reduce the risk of relapse in thispopulation (20). Most of the other risk factors described, including morpho-logic M3 variant, low platelet count, CD34 or CD2 expression on blast surface,presence of the short isoform (S isoform or bcr3) and possibly the rare bcr2 (orV) breakpoint in PML-RARα, as compared to bcr1 (or L breakpoint), are gen-erally correlated to high WBC counts (35,36). On the other hand, a meta-analysis of Gimema and Pethema protocols in APL, which both includedmaintenance with ATRA and low-dose chemotherapy, showed that WBC andplatelet counts had independent prognostic value for relapse. Disease-free sur-vival (DFS) at 4 yr was 98% in patients with platelets > 50,000/mm3, 90% inpatients with platelets < 50,000/mm3 but WBC < 10,000/mm3, and only 68%in patients with platelets < 50,000/mm3 and WBC > 10,000/mm3 (4).

Prognostic factors independent of the WBC count include CD13 and CD56(37) expression, associated with an increased risk of relapse in some studiesand importance of in vitro differentiation of APL blasts with ATRA (16),


whereas cytogenetic abnormalities in addition to t(15;17) do not confer ahigher relapse risk (3,30,38).

7.2.2. MONITORING MINIMAL RESIDUAL DISEASE

A good correlation has been found in APL between RT-PCR findings andthe risk of relapse. Patients with detectable PML-RARα fusion mRNA by RT-PCR at the end of consolidation and perhaps, more importantly, patients withpositive findings after a phase of negative results are at high risk of relapse(16,24,39,40). RT-PCR results should however always be interpreted carefully.First, for unknown reasons, PML-RARα fusion transcript appears to be easilydegraded in bone marrow samples, often making results uninterpretable. Inaddition, despite consensus meetings (41), there may be a certain interlabora-tory variation in the sensitivity of RT-PCR, especially because some laborato-ries perform one-round PCR and others two-round PCR. Thus, comparisonbetween successive samples rather than interpretation of one given sample isadvised. Quantitative PCR techniques, especially with Taqman probes, allowbetter analysis of an increase or decrease of fusion PML-RARα mRNA in suc-cessive samples and should improve clinical interpretation of results and thera-peutic decisions (42).

7.3. Extramedullary RelapsesA relatively large number of cases of extramedullary relapses have been

reported in APL treated with ATRA and chemotherapy: 13 of the 97 relapses ina Gimema group study (43), 3 of the 75 relapses in the European APL 93 trial(69), and 3 of the 28 relapses in an MRC study (43,44). Extramedullary relapsein these studies mainly included the CNS (in 14 of the 21 cases) and less oftenthe skin or other organs. CNS relapse in 10 of the 14 cases was associated withmorphologic or cytogenetic evidence of marrow relapse.

It is uncertain, however, if there is a real increase in the incidence ofextramedullary relapses during the last few years, which could be attributablein some way to the advent of ATRA.

8. TREATMENT OF RELAPSING ACUTE PROMYELOCYTIC LEUKEMIA

8.1. Treatment or Retreatment with All-Trans-Retinoic AcidMost (85% to 90%) of the patients previously treated with chemotherapy

alone who are in first relapse can achieve a second CR with ATRA (19,25,45).However, because ATRA is now administered during front-line therapy, mostpatients in relapse have already received ATRA. Relapses occurring shortly afterATRA discontinuation are usually resistant to ATRA, even at higher doses(25,46). Pharmacokinetic studies have indeed shown that prolonged use of


ATRA is associated with hypercatabolism of the drug through cytochrome p450mechanisms, with dramatic reduction of the serum peak levels of ATRA (33).

Later relapses, however, may respond to ATRA. Indeed, in the APL 91European trial, all 10 patients initially treated with ATRA who relapsed andwere retreated with ATRA achieved a second CR (38). All those relapsesoccurred more than 6 mo after CR achievement (i.e., more than 6 mo afterATRA discontinuation), and pharmokinetic studies indeed suggest that thehypercatabolism of ATRA induced by treatment with this drug is reversibleafter a few weeks of ATRA discontinuation (34,46). Another reason for thosegood responses could be that in some of those patients, chemotherapy wasrapidly added to ATRA, generally due to increasing WBC counts, and thispossibly overcame partial resistance to ATRA. Recently, the European APLgroup completed a study of ATRA followed by intensive chemotherapy withetoposide, mitoxantrone, and AraC (EMA) in 50 patients with APL in firstrelapse, 39 of whom had received ATRA during their first-line treatmentsMedian duration of first CR was 17 mo (minimum duration 6 mo). Of the 39patients previously treated with ATRA, 13 received ATRA alone for salvageand they all achieved CR. Twenty-six also received early chemotherapy, due tohigh WBC counts at relapse or increasing WBC counts with ATRA, and 24achieved CR (47).

The long-term outcome of patients initially treated with ATRA who relapseand achieve a second CR after retreatment with ATRA, with or withoutchemotherapy, is not known precisely due to the small number of publishedcases with sufficient follow-up. Long-term results of the APL 91 trial suggestthat allogeneic and possibly autologous stem cell transplantation can durablysalvage a proportion of those cases (22). In our recent study in relapsing APL,however, allogeneic bone marrow transplantation (BMT) was associated with ahigh rate of toxic mortality, probably due to the intensity of consolidationchemotherapy (EMA protocol) before the procedure (47). By contrast, 17 ofthe 22 patients who were autografted were still in second CR, with a 3-yr DFSof 77%. Nine of them had already reached a second CR longer that their firstCR. PML-RARα transcript was no longer detectable in eight of the ninepatients tested (73). Finally, in a few anecdotal cases, we observed a secondCR that was longer than the first CR, using maintenance treatment combiningintermittent ATRA and continuous 6MP and MTX.

8.2. Other RetinoidsAs maintenance treatment with ATRA is used by more and more groups, a

large proportion of first relapses now concern patients who are receiving orhave recently received ATRA and will probably show resistance to this drug. Inaddition, in some patients, resistance to ATRA is not due to hypercatabolism ofthe drug but is irreversible and due to the occurrence of point mutations in the


DNA binding domain of the RARα gene (48). Finally, patients in second orthird relapse generally show in vivo and in vitro ATRA resistance.

In ATRA resistance preliminary studies suggest that liposomal ATRA maystill induce some responses, possibly because this mode of ATRA administra-tion does not induce hypercatabolism of the drug and reduction of its plasmalevels (49). Favorable preliminary results with 9cis retinoic acid, a retinoid thatinteracts with both RXR and RAR receptors and does not induce its own catab-olism to the same extent as ATRA, have been published, but these responsesoccurred in patients who were possibly not ATRA resistant (50). Am80, a syn-thetic retinoid more potent than ATRA as an in vitro differentiation inducer,induced CR in 58% of a cohort of patients with APL who were relapsing (51).However, because all those patients had discontinued ATRA for at least 18 mo,it is unclear whether Am80 was superior to ATRA in this situation and whetherit was able to overcome ATRA resistance.

8.3. Arsenic Trioxide and Other ApproachesIn fact, the advent of arsenic trioxide (ATO) (As2O3) has modified the out-

come of patients refractory to ATRA (see Chapter 12), and this drug is cur-rently tested earlier in APL treatment. Other new approaches are currently onlyat an early study stage; they include inhibitors of histone deacetylase (anenzyme associated with transcriptional repression), particularly sodiumphenylbutyrate, used with success in animal models and in a few patients withAPL resistant to ATRA (52), and anti-CD33 monoclonal antibodies (gem-tuzumab ozogamicin) (53).

9. ALL-TRANS-RETINOIC ACID SYNDROME AND OTHER SIDE EFFECTS

ATRA therapy is usually well tolerated and its side effects moderate, withthe exception of ATRA syndrome.

9.1. All-Trans-Retinoic-Acid Syndrome9.1.1. INCIDENCE AND CLINICAL SIGNS

A precise description of clinical symptoms of ATRA syndrome was madeby Frankel et al. (18) and two large series of cases of ATRA syndrome haverecently been published (54,55). Clinical signs of ATRA syndrome combinefever, respiratory distress, weight gain, lower extremity edema, pleural or peri-cardial effusions, hypotension, and occasionally renal failure. These signs arepreceded by increasing WBC counts in the majority of cases, but some patientsdevelop symptoms at normal WBC counts (25). Some cases of ATRA syn-drome can also occur upon recovery from aplasia in patients who havereceived early chemotherapy and are still receiving ATRA (54).


The ATRA syndrome occurred in 6% to 27% of the patients in previousreports, and mortality of the syndrome ranged from 8% to 15% (18,21,23,56,57).In the European APL 93 trial, patients who survived ATRA syndrome had ahigher risk of bone marrow relapse than patients who had no ATRA syndrome(32% vs 15% at 2 yr) (54). In another study, the occurrence of ATRA syn-drome was associated with a possibly higher risk of subsequent extramedullaryrelapse (58). Finally, ATRA syndrome has not been reported during mainte-nance treatment with ATRA.

9.1.2. PATHOPHYSIOLOGY OF HYPERLEUKOCYTOSIS

AND ALL-TRANS-RETINOIC ACID SYNDROME

The pathophysiology of hyperleukocytosis and ATRA syndrome is still notcompletely understood. Clinical signs of the ATRA syndrome are reminiscentof those observed in endotoxic shock and in the adult respiratory distress syn-drome (ARDS), and a possible stimulatory effect of ATRA on cytokine expres-sion by APL cells has been envisaged: induction of IL-1b and G-CSF secretionby APL cells under ATRA may contribute to hyperleukocytosis in vivo; secre-tion of IL-1b, IL6, TNFα, and IL-8, which are involved in leucocyte activationand adherence, could have a pathogenetic role in ATRA syndrome (59). It hasalso been shown that ATRA-induced aggregation of NB4 cells (an APL cellline), a process mediated by LFA1 and ICAM2 adhesion molecules andreversed by addition of methylprednisolone (60). These findings suggest thatmodification of the adhesive properties of APL cells by ATRA could play arole in ATRA syndrome.

9.1.3. PROPHYLAXIS AND TREATMENT

Prevention and early treatment of ATRA syndrome are required to improveits outcome and are currently attempted by two approaches. One of them,mainly used by the European and Japanese groups (17,21,56), consists ofadding chemotherapy from the onset of ATRA in patients presenting withhigh WBC counts (WBC > 5000/mm3 in the European trial or > 3000/mm3 inthe Japanese trials) or when increases in the WBC are seen with ATRA. Thisapproach has been associated with a low incidence of fatal ATRA syndromebut leads to early chemotherapy administration in many patients. However,intensive chemotherapy, if not administered early, would have to be adminis-tered later as consolidation treatment. In addition, results of the EuropeanAPL 93 trial suggest that early onset of chemotherapy (ATRA+chemother-apy) reduces the incidence of relapse compared with ATRA followed bychemotherapy (ATRA→chemotherapy). Finally, in the APL 93 trial, therewere fewer cases of ATRA syndrome in patients who received ATRA +chemotherapy (11%) compared with those treated by ATRA→chemotherapy(18%) (19).


The usual US approach is to prevent ATRA syndrome by high-dose intravenous(iv) corticosteroids (dexamethasone, 10 mg iv twice daily for 3 or more d) as soonas the first symptoms occur. This attitude proved effective in the US intergroupstudy, for both preventing the ATRA syndrome and reducing its mortality (23).

In fact, there is consensus that patients presenting with high WBC counts(e.g., > 15,000 to 20,000/mm3) will often develop severe ATRA syndrome withATRA alone and require chemotherapy and iv dexamethasone from the onsetof treatment. Some of these patients even have symptoms analogous to those ofthe ATRA syndrome at diagnosis (61).

9.2. Coagulopathy and ThrombosisNo worsening of the bleeding tendency is observed in patients with APL

undergoing ATRA therapy. In the European APL 91 trial, median time to dis-appearance of significant coagulopathy was 6 d after chemotherapy alone and3 d in the ATRA group (p = 0.001) (21). ATRA therapy may be especiallyimportant in reducing the severity of the bleeding tendency in patients withhyperleukocytic APL, a population still at relatively high risk of early deathwith chemotherapy alone.

On the other hand, treatment with ATRA may lead to a transient period ofhypercoagulability, which could explain the few well-documented cases ofthromboembolic events in patients with APL treated with ATRA (62).

8.3. Other All-Trans-Retinoic Acid Side EffectsATRA side effects include dryness of lips and mucosae, isolated fever in the

absence of other signs of ATRA syndrome (or infection), increases in transam-inases and triglycerides (which never required treatment discontinuation in ourexperience), and headache due to intracranial hypertension, which may besevere in children and associated with signs of pseudotumor cerebri (28).Lower ATRA doses (25 mg/m2/d) can reduce this side effect in children (28).

Other side effects, including bone marrow necrosis, hypercalcemia (63), ery-thema nodosum (64), marked basophilia, severe myositis (65), Sweet syndrome(66), Fournier’s gangrene (necrotizing fasciitis of the penis and scrotum) (67),thrombocytosis (68), and necrotizing vasculitis, have rarely been reported withATRA treatment.

REFERENCES1. Bennett JM, Catowsky D, Danier MT, Flandrin G, Galton D, Gralnick M, Sultan C. Propos-

als for the classification of the acute leukemias. Br J Haematol 1976;33:451.2. Bennett JM, Catowsky D, Danier MT, et al. A variant form of hypergranular promyelocytic

leukemia (M3). Ann Intern Med 1980;92:280.3. De Botton S, Chevret S, Sanz M, et al. Additional chromosomal abnormalities in patients

with acute promyelocytic leukemia (APL) do not confer poor prognosis: results of APL 93trial. Br J Haematol 2000;111:801–806.


4. Sanz MA, Martin G, Rayon C, et al. A modified AIDA protocol with anthracycline-basedconsolidation results in high antileukemic efficacy and reduced toxicity in newly diagnosedPML/RARalpha-positive acute promyelocytic leukemia. Blood 1999;94:3015–3021.

5. Fenaux P, Pollet JP, Vandenbossche L, Dupriez B, Jouet JP, Bauters F. Treatment of acutepromyelocytic leukemia: a report on 70 cases. Leukemia Lymphoma 1991;4:237–248.

6. Fenaux P, Tertian G, Castaigne S. A randomized trial of amsacrine and rubidazone in 39patients with acute promyelocytic leukemia. J Clin Oncol 1991;9:1556–1561.

7. Tallman MS. The thrombophilic state in acute promyelocytic leukemia. Semin ThrombHemostasis 1999;25:209–215.

8. Kantarjian H, Keating M, Walter R, et al. Role of maintenance chemotherapy in acutepromyelocytic leukemia. Cancer 1987;59:1258–1263.

9. Marty M, Ganem G, Fischer J. Leucémie aiguë promyélocytaire: étude rétrospective de 119malades traités par daunorubicine. Nouv Rev Fr Hematol 1984;24:371–378.

10. Huang M, Yu-Chen Y, Shu-Rong C, et al. Use of all trans retinoic acid in the treatment ofacute promyelocytic leukemia. Blood 1988;72:567.

11. Castaigne S, Chomienne C, Daniel MT, Berger R, Fenaux P, Degos L. All trans retinoic acidas a differentiating therapy for acute promyelocytic leukemias. I Clinical results. Blood1990;76:1704.

12. Warrel RP, Frankel SR, Miller W, et al. Differentiation therapy of acute promyelocyticleukemia with tretinoin (all trans retinoic acid). N Engl J Med 1991;324:1385.

13. Chen ZX, Xue YQ, Zhang R, et al. A clinical and experimental study on all trans retinoicacid treated acute promyelocytic leukemia patients. Blood 1991;78:1413.

14. Chomienne C, Ballerini P, Balitrand N, et al. All trans retinoic acid in promyelocyticleukemias. II. In vitro studies structure function relationship. Blood 1990;76:1710.

15. Elliott S, Taylor K, White S, et al. Proof of differentiating mode of action of all-trans retinoicacid in acute promyelocytic leukemia using X-linked clonal analysis. Blood 1991;79:1916.

16. Cassinat B, Chevret S, Zassadowski F, for the European APL Group, et al. In vitro all-transretinoic acid sensitivity of acute promyelocytic leukemia blasts: a novel indicator of poorpatient outcome. Blood 2001;98(9):2862–2864.

17. Fenaux P, Castaigne S, Dombret H, et al. All-transretinoic acid followed by intensivechemotherapy gives a high complete remission rate and may prolong remissions in newly diag-nosed acute promyelocytic leukemia: a pilot study on 26 cases. Blood 1992;80:2176–2181.

18. Frankel SR, Eardley A, Lauwers G, Weiss M, Warrell R. The “retinoic acid syndrome” inacute promyelocytic leukemia. Ann Intern Med 1992;117:292.

19. De Botton S, Coiteux V, Chevret S, et al. Early onset of chemotherapy can reduce the inci-dence of ATRA syndrome in newly diagnosed acute promyelocytic leukemia (APL). Resultsof a randomized study. Blood 2001;98(11):766a.

20. Fenaux P, Wattel E, Archimbaud E, et al. Prolonged follow up confirms that all transretinoicacid (ATRA) followed by chemotherapy reduces the risk of relapse in newly diagnosed acutepromyelocytic leukemia (APL). Blood 1994;84:666–667.

21. Fenaux P, Le Deley MC, Castaigne S, et al. Effect of all transretinoic acid in newly diag-nosed acute promyelocytic leukemia. Results of a multicenter randomized trial. Blood1993;82:3241–3249.

22. Fenaux P, Chastang C, Chevret S, et al. A randomized comparison of all transretinoic acid(ATRA) followed by chemotherapy and ATRA plus chemotherapy and the role of mainte-nance therapy in newly diagnosed acute promyelocytic leukemia. Blood 1999;94:1192–1200.

23. Fenaux P, Chevret S, Guerci A, et al. Long term follow up confirms the benefit of all tran-sretinoic acid in acute promyelocytic leukemia. Leukemia 2000;14:1371–1377.

24. Tallman MS, Andersenn JW, Schiffer CA, et al. All trans-retinoic acid in acute promyelo-cytic leukemia. N Engl J Med 1997;337:1021–1028.

25. Mandelli F, Diverio D, Avvisati G, et al. Molecular remission in PML/RARalpha-positiveacute promyelocytic leukemia by combined all-trans retinoic acid and idarubicin (AIDA)therapy. Blood 1997;1014–1021.


26. Warrell RP, Maslak P, Eardley A, Heller G, Miller WH, Frankel SR. Treatment of acutepromyelocytic leukemia with all-trans retinoic acid: an update of the New York experience.Leukemia 1994;8:926–933.

27. Burnett AK, Grimwade S, Solomon E, Wheatley K, Goldstone AH. Presenting white bloodcell count and kinetics of molecular remission predict prognosis in acute promyelocyticleukemia treated with all-trans retinoic acid: result of the randomized MRC Trial. Blood1999;93:4131–4143.

28. Castaigne S, Lefebvre P, Chomienne C, et al. Effectiveness and pharmacokinetics of low-dose all-trans retinoic acid (25 mg/m2) in acute promyelocytic leukemia. Blood1993;82:3560–3563.

29. Mahmoud HH, Hurwitz CA, Roberts WM, Santana VM, Ribeiro RC, Krance RA. Tretinointoxicity in children with acute promyelocytic leukaemia. Lancet 1993;342:1394–1395.

30. Sanz MA, Martin G, Barragan E, PETHEMA group, et al. Risk-adapted treatment of acutepromyelocytic leukemia: results of the Spanish PETHEMA trials using ATRA and anthracy-cline-based chemotherapy. Blood 2001;98(11):765a.

31. Estey E, Thall PF, Pierce S, Kantarjian H, Keating M. Treatment of newly diagnosed acutepromyelocytic leukemia without cytarabine. J Clin Oncol 1997;15:483–490.

32. Cornic M, Delva L, Guidez F, Balitrand N, Degos L, Chomienne C. Induction of retinoicacid-binding protein in normal and malignant human myeloid cells by retinoic acid in acutepromyelocytic leukemia patients. Cancer Res 1992;52:3329–3334.

33. Muindi JRF, Frankel SR, Huseltion C, et al. Clinical pharmacology of oral all-trans retinoicacid in patients with acute promyelocytic leukemia. Cancer Res 1992;52:2138–2142.

34. Adamson PC, Bailey J, Pluda J, et al. Pharmacokinetics of all-trans-retinoic acid adminis-tered on an intermittent schedule. J Clin Oncol 1995;13:1238–1241.

35. Claston DF, Reading CL, Nagarajan L, et al. Correlation of CD2 expression with PML genebreakpoints in patients with acute promyelocytic leukemia. Blood 1992;80:582–586.

36. Paietta E, Andersen J, Gallagher R, et al. The immunophenotype of acute promyelocyticleukemia (APL): an ECOG study. Leukemia 1994;8:1108–1112.

37. Murray CK, Estey E, Paietta E, et al. CD 56 expression in acute promyelocytic leukemia: apossible indicator of poor treatment outcome? J Clin Oncol 1999;17:293–297.

38. Gallagher RE, Willman CL, Slack JL, et al. Association of PML-RARalpha fusion mRNAtype with pretreatment hematologic characteristics but not treatment outcome in acutepromyelocytic leukemia: an intergroup molecular study. Blood 1997;90:1656–1663.

39. Biondi A, Rambaldi A, Pandolfi PP, et al. Molecular monitoring of the myl/retinoic acidreceptor alpha fusion gene in acute promyelocytic leukemia by polymerase chain reaction.Blood 1992;80:492–497.

40. Diverio D, Rossi V, Avvisati G, et al. Early detection of relapse by prospective reverse tran-scriptase-polymerase chain reaction analysis of the PML/RARalpha fusion gene in patientswith acute promyelocytic leukemia enrolled in the GIMENA-AIEOP multicenter AIDA trial.GIMENA-AIEOP multicenter AIDA Trial. Blood 1998;92:784–789.

41. Bolufer P, Barragan E, Sanz MA, et al. Preliminary experience in external quality control of RT-PCR PML-RAR alpha detection in promyelocytic leukemia. Leukemia 1998;12:2024–2028.

42. Cassinat B, Zassadowski F, Balitrand N, et al. Quantitation of minimal residual disease inacute promyelocytic leukemia patients with t(15;7) translocations using real-time RT-PCR.Leukemia 2000;14:324–328.

43. Liso V, Specchia G, Pogliani EM, et al. Extramedullary involvement in patients with acutepromyelocytic leukemia: a report of seven cases. Cancer 1998;83:1522–1528.

44. Evans GD, Grimwade DJ. Extramedullary disease in acute promyelocytic leukemia.Leukemia Lymphoma 1999;33:219–229.

45. Frankel SR, Eardley A, Heller G, et al. All-trans retinoic acid for acute promyelocyticleukemia. Results of the New York study. Ann Int Med 1994;120:279–286.


46. Delva L, Cornic M, Balitrand N, et al. Resistance to all-trans retinoic acid (ATRA) therapyin relapsing acute promyelocytic leukemia: study of in vitro ATRA sensitivity and cellularretinoic acid binding protein levels in leukemic cells. Blood 1993;82:2175–2181.

47. Thomas X, Dombret H, Cordonnier C, et al. Treatment of relapsing acute promyelocyticleukemia by all-trans retinoic acid therapy followed by timed sequential chemotherapy andstem cell transplantation. Leukemia, 2000;14:1006–1013.

48. Ding W, Li YP, Nobile LM, et al. Leukemic cellular retinoic acid resistance and missense muta-tions in the PML-RARalpha fusion gene after relapse of acute promyelocytic leukemia fromtreatment with all-trans retinoic acid and intensive chemotherapy. Blood 1998;92:1172–1183.

49. Estey EH, Giles FJ, Kantarjian H, et al. Molecular remissions induced by liposomal-encap-sulated all-trans retinoic acid in newly diagnosed promyelocytic leukemia. Blood1999;94:2230–2235.

50. Soignet SL, Benedetti F, Fleischauer A, et al. Clinical study of 9-cis retinoic acid (LGD1057) in acute promyelocytic leukemia. Leukemia 1998;12:1518–1521.

51. Takeuchi M, Yano T, Omoto E, et al. Relapsed acute promyelocytic leukemia previouslytreated with all-trans retinoic acid: clinical experience with a new synthetic retinoid, AM-80.Leukemia Lymphoma 1998;31:441–451.

52. Warrell RP, Jr, He LZ, Richon V, Calleja E, Pandolfi PP. Therapeutic targeting of transcrip-tion in acute promyelocytic leukemia by use of an inhibitor of histone deacetylase. J NatlCancer Inst 1998;90:1621–1625.

53. Estey E, Giles F, Cortes J, et al. Gemtuzumab ozogamycin (“Mylotarg”) in untreated acutepromyelocytic leukemia (APL). Blood 2001;98(11):766a.

54. De Botton S, Dombret H, Sanz M, et al. Incidence, clinical features, and outcome of all-trans-retinoic acid syndrome in 413 cases of newly diagnosed acute promyelocytic leukemia.Blood 1998;92:2712–2718.

55. Tallman MS, Andersen JW, Schiffer CA, et al. Clinical description of 44 patients with acutepromyelocytic leukemia who developed the retinoic acid syndrome. Blood 2000;95:90–95.

56. Kanamaru A, Takemoto Y, Tanimoto M, et al. All-trans retinoic acid for the treatment ofnewly diagnosed acute promyelocytic leukemia. Blood 1995;85:1202–1206.

57. Vahdat L, Maslak P, Miller WH, Early mortality and the retinoic acid syndrome in acutepromyelocytic leukemia: impact of leukocytosis, low-dose chemotherapy, PMN/RAR-alphaisoform, and CD13 expression in patients treated with all-trans retinoic acid. Blood1994;84:3843–3849.

58. Ko BS, Tang JL, Chen YC, et al. Extramedullary relapse after all-trans retinoic acid treat-ment in acute promyelocytic leukemia—the occurrence of retinoic acid syndrome is a riskfactor. Leukemia 1999;13:1406–1408.

59. Dubois C, Schlageter MH, de Gentile A, et al. Modulation of II-6 and II-1b and G-CSF secre-tion by all trans retinoic acid in acute promyelocytic leukemia. Leukemia 1994;8:1750–1757.

60. Larson RS, Brown DC, Sklar LA. Retinoic acid induces aggregation of the acute promyelo-cytic leukemia cell line NB-4 by utilization of LFA-1 and ICAM-2. Blood 1997;90:2747–2756.

61. Stadler M, Ganser A, Hoelzer D. Acute promyelocytic leukemia. N Engl J Med1994;330:140–141.

62. Forjaz De Lacerda J, Alves Do Carmo J, Lurdes Guerra M, Geraldes J, Forjaz De LacerdaJM. Multiple thrombosis in acute promyelocytic leukaemia after tretinoin. Lancet1993;342:114.

63. Akiyama H, Nakamura N, Nagasaka S, Sakamaki H, Onozawa Y. Hyper-calcaemia due toall-trans retinoic acid. Lancet 1992;i:308,309.

64. Hakimian D, Tallman MS, Zugerman C, Caro WA. Erythema nodosum associated with all-tran-sretinoic acid in the treatment of acute promyelocytic leukemia. Leukemia 1993;7:758–759.

65. Miranda N, Oliveira P, Frade MJ, Melo J, Marques MS, Parreira A. Myositis with tretinoin.Lancet 1994;344:1096.


66. Arun B, Berberian B, Azumi N, Frankel SR, Luksenburg H, Freter C. Sweet’s syndrome dur-ing treatment with all-trans retinoic acid in a patient with acute promyelocytic leukemia.Leukemia Lymphoma 1998;31:613–615.

67. Levy V, Jeffarbey J, Aouad K, Zittoun R. Fournier’s gangrene during induction treatment ofacute promyelocytic leukemia, a case report. Ann Hematol 1998;76:91–92.

68. Kentos A, Le Moine F, Crenier L, et al. All-trans retinoic acid induced thrombocytosis in apatient with acute promyelocytic leukaemia. Br J Haematol 1997;97:685–692.


V GENE THERAPY

1. INTRODUCTION

Gene therapy may be defined as the introduction of nucleic acids into a cellwith the intention of altering function to achieve a therapeutic benefit (1). Genetherapy approaches can be divided into the manipulation of reproductive cells tomaintain a genetic modification in future generations (germ-line gene therapy) orgene transfer into more differentiated tissues, such as cells of the lung, liver, orbrain, or hematopoietic cells (somatic gene therapy). Clinical gene therapy proto-cols for cancer are of this latter category. Cells can be genetically manipulated exvivo, which has the clear advantage of facilitating control over the gene transferprocedure, and allows testing of the cell product before the administration of thecells therapeutically. However, protocols have been devised to target malignantcells in vivo, with the intention of correcting genetic abnormalities or to result intheir eradication. The Journal of Gene Medicine Web site currently lists 596 clin-ical trials initiated since 1989, including 376 (63%) that are related to the treat-ment of malignancies, with 2389 patients enrolled in cancer-related gene therapyprotocols (2). A total of 46 (8%) of all gene therapy trials focus on the treatmentof leukemia or lymphoid malignancies. A review of these approaches and othersbeing considered for therapy can be categorized into several therapeutic strategies(Table 1). However, before discussing these applications, an overview of the cur-rent means by which gene transfer and expression can be achieved is warranted.

225


14 Gene Therapy

Paul J. Orchard, MD

and R. Scott McIvor, PhD

CONTENTS

INTRODUCTION

METHODS TO ACHIEVE GENE TRANSFER AND EXPRESSION

GENE THERAPY FOR LEUKEMIA/LYMPHOMA

FUTURE CHALLENGES IN GENE THERAPY

OF LEUKEMIA/LYMPHOMA

REFERENCES

2. METHODS TO ACHIEVE GENE TRANSFER AND EXPRESSION

2.1. Physical Methods of Gene TransferThe transfer of genetic material by physical means can be accomplished by

using deoxyribonucleic acid (DNA) alone or with the assistance of carrierssuch as liposomes. These methods have recently been reviewed (3). The use ofnaked DNA techniques to obtain expression has shown promise, especially intissues such as skeletal (4–7) and cardiac (8–11) muscle, and has also beeneffective in achieving gene transfer in endothelial cells (12) and skin (13–15).Carrier vehicles such as cationic liposomes can increase the association of neg-atively charged DNA and the cell membrane, a strategy that has been used inseveral target cells (16–18), including hematopoietic and leukemic cells(19–24). A potential advantage of using liposomes is the opportunity to modifythe liposome to enhance delivery to a specific cell target (25–27). In addition,there is great flexibility in the size of the gene that can be physically trans-ferred (28). Electroporation is another methodology that has been used toincrease the gene transfer efficacy of naked DNA, and has been used in vacci-nation experiments in a murine model of multiple myeloma (29,30). The “genegun” achieves transfer of genetic material by depositing DNA onto gold ortungsten microprojectiles that are accelerated to high velocity for introductioninto the nucleus of the recipient cell (31). Unfortunately, physical gene transferhas not yet achieved the level of gene transfer efficiency in hematopoietic cellsobserved with viral vectors (24,32) but has the advantage of minimizingimmunogenicity and infectious risks that are a concern with viruses (1).

2.2. Viral Vectors as Gene Transfer VehiclesViruses have, throughout millions of years, evolved remarkably efficient

means of introducing and expressing DNA or ribonucleic acid (RNA) in

226 Orchard and McIvor

Table 1Gene Therapy Approaches for Leukemia and Lymphoma

1) Marking studies to investigate leukemia biology2) Augment immune responsiveness

• Modify leukemia/lymphoma cells; increase immunogenicity, enhance vaccination• Modify effector cells; express cytokines, engineer specificity

3) Chemoprotection; modification of stem cells to provide resistance to myelosuppres-sive therapy

4) Repair or inactivation of genetic abnormalities• Tumor suppressor genes; repair lost function• Oncogene downregulation; inhibit expression of genes contributing to malignancy

5) Prodrug/prosuicide genes; engineer T-cells to provide graft vs leukemia, allowingselective depletion if graft vs host disease observed

eukaryotic cells. The biologic processes developed by viruses to achieve genetransfer can be harnessed through modification of the viral genome to achieveexpression of particular genes of interest. The usefulness of a particular viralvector for gene therapy depends on several factors, including (1) the ability tomodify the virus to accommodate the gene of interest while retaining thecapacity of the virus to achieve efficient transduction; (2) the ability to designvectors to remain replication incompetent, because in most cases an activeinfection is to be avoided in the recipient; (3) the immunologic response gener-ated to particular viruses, because these may prove significant and potentiallylife threatening; (4) the ability to generate reproducibly high concentrations ofvirus required for efficient gene transfer; and (5) the necessity of long-termexpression to realize a therapeutic benefit (33–35). Characteristics of several ofthe viral vectors currently being used for gene therapy are listed in Table 2. Theapplicability of a viral vector for a given disorder is therefore related to thegoal of therapy, taking into account the characteristics of the gene to be deliv-ered, the susceptibility of different target cells to viral transduction, the prolif-erative state of the target cell, and the duration of expression necessary forsuccessful intervention (35). Although a large number of viral agents are beingdeveloped for clinical gene transfer applications, for the purposes of this dis-cussion the focus is on the vectors that have received the majority of attentionin the field of gene therapy for hematologic malignancy.

2.2.1. MURINE RETROVIRAL VECTORS

Retroviral vectors, particularly those based on Moloney murine leukemiaviruses (MoMLV) have distinct advantages that have resulted in their wideclinical use, including the ability to transduce a broad range of target cells,achieving stable integration into the target cell genome, relatively efficientgene transfer, and, in most cases, high-level expression of the transgene. In

Chapter 14 / Gene Therapy 227

Table 2Viral Vectors for Gene Therapy

RNA/ Mitosis Transgene Titer Type of Virus DNA Integration Required Size (virions/mL)

Murine retrovirus ssRNA Yes Yes 10 kb 106–107

Lentivirus ssRNA Yes No 10 kb 106–107

Adenovirus dsDNA No No 37 kb 1011

Adenoassociated virus ssDNA Yes No 4.9 kb 1010

(AAV)Herpes Simple Virus dsDNA No No 30–40 kb 109

(HSV-1)

ss = single strand; RNA = ribonucleic acid; ds = double strand; DNA = deoxyribonucleicacid.

addition, their safety record has been encouraging (36). Retroviruses containtwo identical single strands of RNA, as well as viral replication enzymeswithin a protein core and a membrane envelope. Upon infection, the viral RNAis reverse transcribed into DNA, which integrates randomly into the host cellgenome. Long terminal repeats (LTRs) located at the 5′ and 3′ ends of thegenome regulate viral replication and expression of the gag, pol, and env genes(Fig. 1). In MoMLV, the gag and pol proteins are alternatively translated fromthe same message. The gag gene provides the structural proteins necessary forthe virus, including the capsid and nucleoprotein complex. The pol geneencodes reverse transcriptase, the enzyme that generates DNA from viralRNA, and the integrase enzyme important in providing stable integration of theviral sequences into the host genome. The env gene products are viral envelopeglycoproteins that determine receptor binding and cell type specificity. To gen-erate retroviral vectors, the gag, pol, and env gene products are provided intrans, meaning that these genes need not be present in the retroviral genome to


Fig. 1. Murine leukemia virus and constructs for gene therapy. The wild-type retrovirus 5′retroviral long terminal repeat (LTR) contains the promoter/enhancer elements for transcrip-tion (arrow). The packaging sequence (ψ) is necessary for the recognition of the viral RNAfor encapsidation into the virus particle. The gag/pol gene, which encodes the structuralproteins of the virus (gag) and the reverse transcriptase and integrase enzymes (pol) aretranslated from a single gene. The envelope (env) gene is expressed primarily through splic-ing, using the slice donor (SD) and splice acceptor (SA) sequences (A). In the commontypes of retroviral constructs used for gene therapy, the gag/pol and env genes are removedand replaced by a transgene that is either expressed from the promoter/enhancer of the LTRusing the splicing mechanisms of the virus (B) or without splicing (C). Commonly, an inter-nal promoter is used, which may provide expression of a second gene, such as the neomycinphosphotransferase gene.

generate a retroviral particle. Sequences required “in cis” (i.e., within the virusgenome) include those necessary for expression, replication, and packaging,including the packaging signal. Retroviral vector constructs are commonlydesigned in which the gag, pol, and env genes have been substituted by thegene(s) of interest. Beginning in 1983 the development of “packaging” celllines provided stable expression of the retroviral proteins, facilitating the gen-eration of replication incompetent retroviruses and providing a practical meansof generating high titer virus (37,38). Packaging lines have been designed toprovide the expression of the necessary gag, pol, and env genes while minimiz-ing potential production of replication competent virus. Because these genesare not present in the vector, the retroviral replicative cycle is interrupted (Fig.2). Many packaging cell lines have been generated that are capable of produc-ing high titer virus (106–107 virions/mL). The substitution of envelope proteinsfrom different viruses is termed pseudotyping, and has been used as a means toalter the capacity of the virus to transduce specific cell types. For instance, theecotropic envelope gene of MoMLV mediates gene transfer into rat and murinecells (39,40), while vectors incorporating the amphotropic envelope are capa-ble of transducing several host cell types, including rodent and human cells(41,42). Retrovirus produced from packaging cell lines using the gibbon apeleukemia virus (GALV) envelope gene are efficient in transducing human andsimian cells, but do not transduce murine cells (41,43–45). Pseudotyping canalso be carried out using envelope protein from viruses other than retroviruses,including the vesicular stomatitis virus G glycoprotein (VSV-G), which hasabundant membrane receptors (46–48).

Retroviral constructs based on MoMLV vectors were used for the first genetherapy trial, in which a retrovirus containing the adenosine deaminase (ADA)gene was used to transduce lymphocytes from 2 patients with severe combinedimmunodeficiency (SCID) associated with ADA deficiency in 1990 (49). Inaddition, encouraging results with a MoMLV-based vector was reported in thetreatment of X-linked SCID by ex vivo transduction of hematopoietic stem cells(50). Retroviruses remain the most common vehicles for gene therapy; in the 46gene therapy trials that targeted leukemia or lymphoma, 32 (70%) used retrovi-ral vectors (51). No significant adverse events have been described thus far.Insertional mutagenesis remains the primary concern in the use of retroviral vec-tors, because random integration could inactivate a tumor suppressor gene oractivate a proto-oncogene (1). In addition, despite the engineering of retroviralpackaging lines to minimize recombination events, the potential remains for thegeneration of replication-competent virus. The development of lymphoma inrhesus macaques after the administration of cells transduced with a retroviralvector contaminated with replication-competent virus is a clear reminder of thetheoretic risk of retroviral vectors (52–54). As MoMLV viruses are inactivated inhuman serum, their use for providing gene transfer in vivo is limited (55). How-


ever, the viruses are not adversely affected by cerebral spinal fluid (56), and theclinical use of those viruses in brain tumors has been explored (57–59).

2.2.2. LENTIVIRAL VECTORS

Lentiviruses, including the human immunodeficiency virus (HIV), are retro-viruses containing additional regulatory sequences (tat and rev) not present insimple murine retroviruses, as well as accessory sequences (vpr, vif, vpu, and


Fig. 2. Use of replication-incompetent retrovirus to achieve gene transfer. Retroviral con-structs are introduced into the packaging cell line, where clones are identified that have sta-ble expression of the integrated construct, often by resistance to an antibiotic such as theneomycin. The packaging line contains the gag/pol and env genes that have previously beenintroduced and, using these proteins, the RNA produced from the construct is packaged intoviral particles that are released from the cells by budding. The virus binds to specific recep-tors on the target cell, and it is subsequently internalized, uncoated, and dsDNA producedfrom the viral RNA using the reverse transcriptase that is a gene product of the pol gene.The dsDNA is transported into the nucleus of an actively dividing cell, where stable integra-tion is achieved. Expression of the gene of interest follows.

nef) important in the lentivirus replication cycle (35,60). There is understand-able concern regarding the potential for recombination events resulting in theproduction of replication-competent virus derived from HIV, thus far limitingto the implementation of these viruses clinically. However, the potential oflentiviruses to transduce nondividing cells is an important advantage for manygene therapy applications (61–64). Specifically, lentiviral vectors have greatpotential in the transduction of human hematopoietic stem cells. There is evi-dence that human stem cells can be transduced with these vectors in theabsence of stimulatory cytokines required for gene transfer with murine retro-viruses (65–68). To minimize the possibility of generating infectious virus andto extend the range of cells that can be transduced, lentivirus vectors have, inmost cases, been pseudotyped with VSV-G envelope (67,69,70). In a compara-tive study of acute lymphocytic leukemia (ALL) cell lines, lentivirus vectorscompared favorably with MoMLV vectors in achieving high transduction effi-ciency (24). There also is evidence that lentiviruses may facilitate in vivo trans-duction, a limitation of murine retroviruses (35,71). Alternative lentiviralconstructs make use of HIV-2, which is less pathogenic in humans (72,73), orother nonhuman lentiviruses, such as the simian immunodeficiency virus (SIV)(74,75), feline immunodeficiency virus (FIV) (62,76), or the equine infectiousanemia virus (77,78). As in the murine leukemia viruses, the size of the trans-gene that can be efficiently packaged within a lentivirus is limited; Kumar et al.demonstrated that as the insert size was increased to 12 kb, the titer of a VSV-G pseudotyped lentivirus decreased to less than 1% of what was observed incontrols, likely due to difficulties in achieving efficient packaging of the virus(69). Nonetheless, despite these concerns, lentiviruses present an alternativefor achieving enhanced gene transfer efficiency in nonproliferating cells andfor in vivo gene therapy applications.

2.2.3. ADENOVIRAL VECTORS

Adenoviruses are double-stranded DNA viruses consisting of genes encod-ing 11 viral proteins (35). The adenoviral genome is large (36 kb) and isflanked by inverted terminal repeats (ITRs) (79). Adenoviruses do not integrateinto the genomes of recipient cells but rather express their gene products fromepisomal vector genomes, resulting in transient expression (80). High titers (upto 1012 virions/mL) can be achieved, and subsequently efficient gene transferhas been observed (81). Because many individuals have previously beenexposed to adenoviruses, those with intact immunologic function have thepotential to generate a significant immune response (82,83). First-generationadenoviral vectors, in which the E1 gene has been replaced by the gene ofinterest, are more likely to generate an immune response directed against trans-duced cells in a clinical setting (84). Later designs, in which additional aden-oviral genes have been removed from the vector, are less immunogenic


(85–88). In the third-generation adenoviral vectors, the entire adenoviralgenome has been removed, except for the ITRs and the packaging signal. This“gutless” strategy has provided ample space for large transgenes (theoreticallyup to 37 kb); hence the term “high-capacity” adenoviral vectors (82,89).Although these vectors may be successful in diminishing the inflammatoryresponse to transduced cells (90), the most significant challenge is their depen-dence on the presence of a helper virus to provide the adenoviral proteins forvector packaging, and the elimination of this helper virus from vector prepara-tions (91). The recent death of an individual being treated for ornithine tran-scarbamylase deficiency has raised considerable concern regarding the use ofadenoviral vectors, and for gene therapy in general (34,92,93). In this phase Iclinical study, a vector depleted of the E1 and E4 genes was administered intra-hepatically to patients. Fevers, myalgias, nausea, and vomiting were com-monly observed within 48 h of treatment (94). The final patient on the trialdied from respiratory insufficiency that developed after the administration ofthe virus; marrow aplasia was observed as well (95).

Adenoviruses have generated considerable interest for gene therapy applica-tions due to their ability to transduce cells in vivo; clinical trials have beendesigned to test their use in muscle (96), liver (94,97,98), pulmonary (99–101),and endothelial tissues (85,102), as well as the central nervous system (CNS)(103,104). Several investigators have described a limited capability of aden-oviruses to transduce human hematopoietic cells (105,106), but recent techniquessuch as modification of the adenoviral fiber proteins may enhance binding tohematopoietic targets (107). Myeloid and lymphoid leukemia cells may provemore amenable to transduction than undifferentiated cells (108–110). Additionalstrategies to enhance the efficacy of adenovirus-mediated gene transfer includethe use of agents such as poly-L-lysine (105), polycations, and lipid complexes(111). For example, poly-L-lysines bind well to heparan sulfates, which areexpressed at high levels in multiple myeloma, and thus high-efficiency genetransfer can be obtained in myeloma cells using this strategy (112). Adenovirus-mediated gene transfer to enhance expression of CD154 (which binds to CD40on T cells) has been tested clinically in B-lineage malignancies, with the intent ofenhancing immune surveillance directed against the tumor, with reported clinicalresponses (113). Vaccination strategies with adenoviruses in the treatment oflymphoma and multiple myeloma have also been reported (113–118).

2.2.4. ADENOASSOCIATED VIRUS

Adenoassociated viruses (AAVs) are parvoviruses that contain a 4.7-kb sin-gle-stranded DNA genome with two major open reading frames, Rep and Cap,flanked by 5′ and 3′ ITRs (35). At least six serotypes have been described in pri-mates, but none has been shown to be pathogenic, a clear advantage for the useof AAV as a vehicle for gene therapy (119). An interesting characteristic of wild-type AAV is site-specific integration into a region of the human chromosome 19


(120,121). However, site-specific integration is not observed in AAV vectorsdesigned for gene transfer if the Rep gene is not included in the vector(122,123). AAV replicates only when the cell is co-infected with helper aden-ovirus or herpesvirus (124). In the absence of these viruses, AAV is uncoatedwithin the nucleus and integrates into the host cell genome, where it is latentuntil activated by helper virus infection. Most AAV vectors designed for genetransfer and expression have been derived from serotype 2 (125). The size of thetransgene that can be packaged into AAV vectors that have the Rep and Capgenes completely is limited to 4.9 kb (126), although new approaches toincrease the size of the gene of interest are being tested (127). AAV vector gen-eration is somewhat cumbersome, requiring cotransfection of vector and AAVhelper (Cap and Rep) plasmids, along with adenovirus helper plasmid or co-infection with adenovirus. This latter approach requires removal of contaminat-ing adenovirus by physical methods such as heat inactivation and cesiumchloride density gradients (125). This technique, though effective, has for themost part been replaced by 2 and 3 plasmid transfection protocols (128). Severalpackaging cell lines expressing Cap and Rep proteins have been established, butthe titer generated from these lines has not improved upon that achievable bytransient transfection. AAV has been tested extensively for delivery and expres-sion of genes in terminally differentiated and nonproliferating cells such asbrain, liver, and muscle (129–133), including ongoing trials for hemophilia Btargeting muscle (134,135). Transduction of human and murine hematopoieticcells has been reported, with variable degrees of success (136–141). Of interest,a related virus, parvovirus B19, has been used as a vector to transduce humanerythroid precursors, achieving superior gene transfer efficiency in comparisonwith AAV-2 (142). A final yet-unanswered question relating to AAV success as agene transfer tool, especially in vivo, is the immunologic response to the virus,because a majority of individuals are seropositive to AAV type 2 (143).

2.2.5. HERPES SIMPLEX VIRUS VECTORS

The Herpes simplex virus type 1 (HSV-1) is a large and complex virus, witha genome of 152 kb consisting of double-stranded DNA (144). The biology ofthe virus and its means of transduction have recently been reviewed (145). Thevirus has the potential to be an important tool in gene therapy based on its abil-ity to transduce nondividing cells and to accommodate large transgenes, and itscapacity to transduce a broad range of host cells (146). HSV-1 is capable ofinfecting and producing a latent infection in nondividing cells of the peripheralnervous system and CNS, which has generated great interest in their use todeliver and express genes in these tissues (145,147–149). A phase I trial for thetreatment of malignant glioma using an oncolytic HSV-1 virus was completedwith minimal toxicity (150). For hematologic applications, the transduction ofhuman chronic lymphocytic leukemia (CLL) cells has been demonstrated withHSV-1 virus (151), and the success of this approach may relate to the high lev-


els of herpes virus entry mediator (Hve) A present on human CLL cells (152).Interestingly, the CLL cells were shown to be relatively resistant to the cyto-pathic effects of the virus, and this was believed to be related to the antiapop-totic effect of bcl-2 expression in these cells. The resistance of hematopoieticprogenitors to HSV-1 oncolytic viruses has led to the investigation of HSV-1virus as a biologic purging agent for solid tumor contamination in marrow orperipheral blood stem cell populations (153).

2.3. Summary: Gene Transfer MethodsPhysical means of gene transfer for the purposes of gene therapy in

leukemia/lymphoma have the potential of achieving: (1) introduction andexpression of large or complex genes, (2) in vivo gene transfer and expression,and (3) minimized potential for recombinant events and immunologicresponses present with the use of viral vectors. The low incidence of genetransfer remains the primary disadvantage of physical methods, althoughimproving methodology will likely increase gene transfer efficacy and allowthe transfer of DNA to specific cell targets. The murine retroviruses have beenthe primary means of achieving gene transfer and expression in clinical investi-gations to date, primarily because: (1) packaging and viral production isstraightforward, (2) efficient gene transfer is observed in many cell types, (3)they provide stable integration into the host cell, and (4) no significant toxicityor immunologic response has yet been demonstrated. Their use is limited pri-marily to ex vivo transduction applications targeting proliferating cell popula-tions. In contrast, lentiviruses, commonly based on the HIV-1 virus, can beused in vivo and to transduce cells not actively dividing. However, safety con-cerns have slowed their integration into clinical gene therapy protocols. Aden-oviral vectors can also be used for in vivo applications, but adenoviral vectorsare capable of generating a significant inflammatory response, and geneexpression is expected to be transient. AAV provides another option for in vivogene transfer and transduction of nonproliferating cells. However, packagingremains cumbersome, and the size of the gene that can be incorporated intothese vectors is limited. The development of HSV-1-based vector systemsholds promise, primarily due to their large capacity and ability to transducenonproliferating cells. Despite limitations in the use of viral vectors for clinicalapplications, continuing progress in their development suggests that they willcontinue to be the primary means of gene delivery in the near future.

3. GENE THERAPY FOR LEUKEMIA/LYMPHOMA

3.1. Marking StudiesAlthough not therapeutic in intent, gene transfer to investigate homing of cells

to a particular location or tumor, or to study the persistence of a population of


cells is of great historic and practical importance in gene therapy (Fig. 3). Thefirst example of clinical gene transfer for the purpose of marking transduced cellswas carried out in patients with solid malignancies by Rosenberg et al., whointroduced the neomycin phosphotransferase gene (NeoR) into tumor-infiltratinglymphocytes (TIL), providing proof that retroviral-mediated gene transfer couldbe accomplished in a clinical setting without toxicity (154). Of the 46 gene ther-apy studies that list leukemia or lymphoma in the primary eligibility criteria forinclusion, 19 (41%) have used molecular techniques to “mark” cells of interest.All of these studies have used replication-incompetent murine retroviruses con-taining a gene not ordinarily present in human hematopoietic cells, most com-monly the NeoR gene. Transduction of a stem cell source, leukemic blasts, orimmunologically active cells is performed ex vivo, and the cells are re-infused forsubsequent testing. A common strategy that has been employed is the marking ofa portion of autologous marrow (typically 10%–30%) before the infusion of themarrow as a rescue in an autologous transplantation setting for acute myeloidleukemia (AML) (155), chronic myelogenous leukemia (CML) (156), follicularlymphoma (157), or multiple myeloma (158,159). Identification of marked cells


Fig. 3. Gene therapy strategies for leukemia/lymphoma. There are currently 46 clinical genetherapy trials for leukemia or lymphoma. Of these, the primary focus of specific trialsincludes marking (n = 19), enhancing the immune response to leukemia/lymphoma by vacci-nation strategies (n = 14), or engineering of effector cells such as T cells (n = 2), applicationsto correct the genetic abnormalities of leukemia (n = 2), providing resistance of hematopoi-etic precursors to facilitate additional chemotherapy (n = 3) or the use of genes providing thecapacity for negative selection such as HSV-tk (n = 6).

is accomplished using molecular analysis such as a southern blot or the poly-merase chain reaction (PCR), which is more sensitive and may provide informa-tion if only a small proportion of the cells are marked. In addition, cellsexpressing the NeoR gene can be selected in the presence of the antibiotic, pro-viding the potential for purification of the transduced cells after their infusion.Using these methods, in an important series of investigations Brenner et al. docu-mented that in children with AML who relapsed after autologous transplantation,several had blasts marked with the NeoR gene, proving that residual marrow dis-ease contributed to the relapse of these patients (155). The rate of gene transferinto hematopoietic progenitors was 2%–15% with the techniques used, and, insome cases, cells containing the gene could be identified 5 yr after marking, sug-gesting that a stem cell population was successfully transduced (160). Similarly,Deisseroth used a NeoR retroviral vector to transduce CD34 selected cells anddocumented that in the case of autologous transplantation for CML, the NeoRgene could be identified in residual leukemia (156). These studies suggest that“purging” of marrow was ineffective in these situations. Questions such as therelative contributions of marrow and peripheral blood stem cells to hematologicrecovery have also been addressed using gene therapy techniques. Dunbarreported a study in which two distinct NeoR-containing retroviruses (LNL6 andG1Na [161]) were used in an autologous setting, one to transduce marrow andthe other to transduce peripheral blood stem cells from the same patient. Thesevectors could be distinguished molecularly, allowing the determination of the rel-ative number of marrow and nucleated blood cells marked with each vector.Although the overall gene transfer frequency was low, studies suggested superiorpersistence of marked peripheral blood cells (58). This study design using twomarker vectors has the capacity to ask other complex questions, such as compar-isons of purging techniques (162). Investigators have demonstrated the ability togenetically mark marrow from patients with myeloma as well as acute leukemia,although low frequencies of gene transfer have remained an obstacle (159,163).In addition to labeling leukemia cells and CD34+ cells, protocols have beendesigned to follow effector cells infused for the purposes of immunotherapy.Brenner and Heslop developed a procedure to generate donor T cells specific forB cells expressing Epstein-Barr Virus (EBV) proteins, and marked these cellswith an NeoR retrovirus to follow the presence of these cells in vivo as treatmentfor EBV lymphoproliferative disease, documenting persistence of marked T cellsfor months after infusion (164). Marking is also to be performed for EBV-specificT cells generated for the treatment of EBV-positive Hodgkin’s disease undergo-ing transplantation or receiving immunotherapy (165). Techniques to provideautologous EBV-specific cells are being developed to facilitate the infusion ofEBV-specific T cells for individuals who do not have access to donor cells or forrecipients of solid organ transplantations with EBV disease (166,167). Finally, aprocedure in which an adenoviral vector is used to transduce dendritic cells with


the LMP2A gene has been designed to enhance the immune response to EBVantigens, generating specific cytotoxic T cells (168). It has been proposed withinthis same protocol to mark the EBV-specific T cells produced after interaction tothe engineered dendritic cells with an NeoR gene, using a murine retrovirus tofollow their persistance and location. Therefore, the design of this protocol usestwo independent viral vectors, one an adenovirus and the other a retrovirus, toachieve its aims. As stated, marker genes have primarily been nonfunctionalgenes such as NeoR, although the β-galactosidase gene has also been tested(169). However, due to concerns about the potential immunogenicity associatedwith the expression of these nonhuman proteins and rejection of the marked cells,the next generation of vectors to be used for marking studies contain nontran-scribed sequences that can be evaluated molecularly through procedures such asquantitative PCR without the production of gene products that may proveimmunogeneic (170).

3.2. Modification of the Immune ResponseThe documentation of the graft vs leukemia (GVL) effect in decreasing

relapse after allogeneic transplantation and the success of donor lymphocyteinfusion (DLI) in producing sustained remission after leukemic relapse are clearindicators of the power of the immune system to control leukemia. Malignantcells may evade immune surveillance through several mechanisms, includingactively suppressing the immune response (171–174), alteration of theCD95/Fas system (175) and downregulation of major histocompatibility com-plex (MHC) class I or II expression (176). The concept of exploiting theimmunologic response to provide therapy for leukemia is not new. Mathé in1965 treated children with ALL in remission with vaccinations consisting ofleukemic blasts and BCG, documenting prolonged DFS in 8 of 20 patients,while all 10 controls relapsed (177). Our current ability to transfer and expressgenes has led to myriad possibilities to manipulate the immune system as ameans of therapy. Of these, two primary approaches can be defined: (1) geneticmodification of the malignant cell, or a cell to be used with it as a vaccine toaugment the immunogenicity of the leukemia, and (2) to engineer an increasedcapacity of an immunologically active cell population (effector cells) to eradi-cate cancerous cells otherwise inefficiently identified or killed. Theseapproaches have the common goal of eliminating residual leukemic cells usingan immunologic approach, but are distinct and are discussed independently.

3.2.1. ENGINEERING OF LEUKEMIA CELLS TO INCREASE

AN IMMUNOLOGIC RESPONSE

These applications have generally had the goal of designing a more effectivetumor vaccine than was previously possible, and, with few exceptions, this isperformed by modification of cells ex vivo. More than a decade ago it was


shown that the endogenous expression of immunologically active cytokinescan decrease tumorigenicity and enhance immunogenicity in several murinemalignancies (178–181). Since that time, numerous cytokines, human leuko-cyte antigen (HLA) molecules, costimulatory proteins, and integrins have beentested and shown to have value in this role. A detailed description of theseapproaches is beyond the scope of this discussion, and previous reviews havewell documented the issues involved (1,182–185). Several investigators havedescribed increased survival of mice vaccinated with leukemia or lymphomacells engineered to express cytokine genes, such as interleukin (IL)-2 and gran-ulocyte macrophage colony stimulating factor (GM-CSF) (186–190). It ishypothesized that the production of high levels of a cytokine directly stimula-tory to cytotoxic T cells or antigen-presenting cells in the immediate environ-ment surrounding the malignant cell may prove more effective in initiating ageneralized immune response than systemic administration, which can onlyhope to achieve modest serum levels. The transfer of the human IL-2 genestimulates an increased immune response to murine myeloma (189). Clini-cally, an adenoviral vector encoding the IL-2 gene was used to transduce anintracranial plasmacytoma in a patient with myeloma. Although gene transferwas documented, no apparent clinical response was observed (191). Brennerhas tested the combination of human CD40 ligand (hCD40L) to increase theexpression of costimulatory factors on the malignant cells and enhanceimmune recognition and IL-2 in CLL blasts and lymphoma, and has shownthat the expression of both molecules provides increased antitumor immunitythan either alone (114,118). Vaccination of animals with AML blasts engi-neered to express IL-12, another cytokine stimulatory to T cells, providesincreased survival in mice with previously established leukemia (192,193).

The use of agents that can enhance the processing and presentation of anti-gens, most notably GM-CSF, may also be effective at establishing specificcytotoxic T cells directed against the malignant cell. In a murine model ofAML, vaccinations of leukemic cells engineered to produce GM-CSF by retro-viral-mediated gene transfer were superior to vaccine strategies in which theB7.2, IL-4, or TNF-α genes were used (194). In a murine Philadelphia chro-mosome positive (Ph+) ALL model, GM-CSF expression was effective in vac-cination experiments but CD80 (B7.1) and GM-CSF coexpression providedthe greatest protection, against the administration of wildtype leukemia (195).An interesting approach to GM-CSF production in the local environment forthe purposes of leukemia vaccinations uses the K562 cell line engineered toexpress high GM-CSF levels (196). The development of a cell line producingGM-CSF that can be used instead of transducing autologous cells has thepotential to make vaccinations much less complex. In addition, because theK562 line does not express class I or II MHC molecules, the likelihood that thecells will be readily rejected is diminished.


A means of providing increased immunologic recognition of B-cell malig-nancies, which express immunoglobulin idiotype molecules on the cell mem-brane, is the development of a T-cell response to idiotype antigens usingvaccinations. This concept has merit without the use of gene transfer tech-niques, because vaccinations of idiotype-specific proteins and keyhole limpethemocyanin (KLH) administered to 16 patients with follicular non-Hodgkin’slymphoma (NHL) resulted in an increase in the cytotoxic T-cell precursor(CTLp) frequency all patients in vitro, and eight sustained responses werenoted (197). However, B-cell neoplasms do not present antigens well, whichlimits the potency of a T-cell response (113). These cells can be engineered toexpress costimulatory molecules such as B7.1 and B7.2 or CD40L (CD154) bygene transfer, making them much more likely to generate a T-cell response.This strategy was employed in a clinical trial in which an adenoviral vectorwas used to express CD154 in CLL cells ex vivo (117). The infusion of engi-neered cells was in general well tolerated, although fevers, elevated transami-nases, and arthralgias were observed. The elevation of plasma cytokine levels,including TNF and IL-12, was observed, and an increase in the number ofleukemia-specific T-cells was reported, suggesting that this line of interventionmay prove important for B-cell malignancies.

An alternative mechanism by which an enhanced antitumor response may beobtained is by increasing the processing and presentation of tumor-associatedantigens by cells other than the malignant population. The primary cell target ofthis approach is the dendritic cell, which was first described in murine tissues bySteinman in the early 1970s (198,199). T-cells can be “educated” to respond topeptide fragments derived from antigens bound to MHC class I (intracellular orendogenous antigens) or class II (exogenous antigens) molecules (200). Thedendritic cell has a greater density of MHC, adhesion, and costimulatory mole-cules on the cell surface than other cells capable of presenting antigen, makingthem well suited for interacting with T-cells toward the generation of specificcellular responses (201). As techniques for the generation of dendritic cells fromperipheral blood have been developed, they have become a primary focus in theengineering of T-cell responses (38,202). To facilitate antigen presentation andincrease antileukemia or myeloma T-cell responses, several potential antigensources can be used, including synthetic peptides, proteins derived from tumorcell extracts, or RNA derived from tumor cells (203–209). The use of aBCR/ABL-derived peptide and dendritic cells markedly increases the number ofT-cells directed against murine CML, and vaccination with peptide-pulsed den-dritic cells had protective effects against the administration of wildtype leukemia(210). It is possible to obtain dendritic cells directly from myeloid leukemiacells, which appear to present antigens of leukemic origin and increase the spe-cific T-cell response (206). Westermann reported similar findings for dendriticcells derived from patients with CML and that the transduction of dendritic cells


with the IL-7 gene increased the T-cell response to CML cells (205). To increaseantigen specificity, the desired antigen can be produced within the dendritic cellusing viral vectors. The engineering of dendritic cells to express LMP2A anti-gen with an adenoviral vector results in an enhanced T-cell response, which mayprove a viable means of increasing the response to Hodgkin’s lymphoma (168).

3.2.2. GENETIC MODIFICATION OF EFFECTOR CELLS

An alternative approach to the manipulation of malignant cells or antigen-presenting cells is the direct modification of the effector cell population toincrease the recognition, function, or killing of effector cells. Clinical trialswere initiated in which TIL were genetically modified to express the TNFgene, with the intention of killing the malignant cells through the local pro-duction of providing TNF at the tumor site, resulting in increased control ofthe malignancy (211,212). The endogenous expression of IL-2 in a humannatural killer (NK) cell line or donor-derived human NK cells increases theability to kill tumor targets (213,214). This approach was also tested by trans-ducing murine T-cells with a retroviral vector containing the IL-2 cDNA.T-cells expressing IL-2 grew independently and maintained antigen speci-ficity (215). Similarly, transduction of hematopoietic precursors has alsobeen reported with IL-2 vectors, and animals receiving transduced cells haveimproved survival after a leukemia challenge (216). A drawback to thisapproach, however, is the potential for autocrine stimulation and uncontrolledproliferation of cells transduced with a stimulatory cytokine, which wouldlikely require coupling the expression of the gene designed to enhance func-tion with a “suicide gene” such as the herpes simplex virus thymidine kinase(HSV-tk) gene. Another means of responding to this concern uses constructsdesigned with alternative cytokine genes to stimulate T-cells; Minamotodesigned a fusion gene in which the extracellular moiety of the erythropoietinreceptor was combined with genes encoding signaling domains of the IL-2receptor. The presence of erythropoietin in the environment was sufficient todrive T-cell proliferation (217).

In contrast to providing a nonspecific stimulus to T-cells, the capacity to gen-erate effector cell specificity through gene transfer has also been explored. Dem-bic et al. documented that it is possible to transfer the α and β T-cell receptorgenes from a T-cell with the desired specificity into an otherwise naive T-celland thereby generate specificity (218). However, the ability to recognize antigenafter the transfer of T-cell receptors is MHC restricted, because the antigen ispresented in association with the MHC molecule, which is a significant draw-back to this approach (219). Chimeric genes have been described in which thevariable regions of antibodies were fused to T-cell receptors to generate speci-ficity, because this would not require antigen to be expressed in association withMHC molecules and, therefore, cells could be engineered in a non-MHC-


restricted fashion (219,220). The ability to isolate genes encoding the variableregions of antibodies and combine them into a single gene with maintained anti-gen specificity allowed further modification of this strategy (221,222). Thesesingle-chain Fv (scFv) genes can be combined with genes encoding the trans-membrane and intracytoplasmic domains of a protein that signal T-cells to pro-liferate and become cytotoxic, such as the ζ chain of the T-cell receptor or γchain of the Fc receptor (223,224) (Fig. 4). The expression of these chimericgenes have been shown to markedly increase the killing of several malignantcells bearing a target antigen and signaling through the intracytoplasmic ζ or γchain results in activation and cytokine expression (223–228). A clinical trial isunderway in which an anti-CD20 scFv/ζ plasmid is transfected into autologousT-cells expanded in the presence of antigen before infusion (229). This methodfor producing specific effector cells has great promise, but issues such as pro-longed function of expanded cells and the potential for rejection of cells withchimeric antibody/T-cell receptor proteins remain to be addressed.


Fig. 4. Engineering of T-cell specificity. The variable regions of the light and heavy chain(VL and HV) antibody genes are combined with a flexible linker, and joined to the trans-membrane and cytoplasmic portions of a gene that can provide signaling within a T-cell,such as the zeta chain of the T-cell receptor. This construct can be introduced into the effec-tor cell by various gene transfer methods. The protein is anchored in the cell membrane withthe single chain Fv region (scFv) external to the cell where it can bind the desired antigen ofleukemia or lymphoma cells. Binding results in killing of the cell expressing the antigen andactivation of the effector cell through signaling pathways.

3.3. Gene Transfer to Provide Drug ResistanceA primary limitation of the amount of chemotherapy that can be administered is

prolonged myelosuppression. Altering the cellular metabolism of chemotherapeu-tic agents in normal hematopoietic precursors may provide relative resistance tothese toxic agents, potentially allowing chemotherapy dose escalation and subse-quently greater killing of the malignant cells. Expression of a dihydrofolate reduc-tase (DHFR) gene modified to alter its sensitivity to methotrexate providesmethotrexate resistance in vivo (230–232). A clinical trial for CML was designedusing a vector containing both antisense BCR/ABL sequences and a resistantDHFR gene to allow the in vivo selection of transduced hematopoietic precursors(233). The administration of methotrexate after transplantation could thereforeresult in the selected expansion of precursors expressing antisense BCR/ABLsequences. The human multidrug resistance 1 (MDR1) gene has also been a focusof this type of experiment, primarily to provide protection of the marrow in associ-ation with the administration of agents such as paclitaxel, daunomycin, and etopo-side (234). Gene transfer of the MDR1 gene into hematopoietic precursors toprovide resistance of the marrow to chemotherapy used as treatment of solidtumors has been performed, with documentation of transgene persistence to 1 yrafter transplantation (235). In this report several patients had a slight increase inthe number of circulating cells containing the transgene after the administration oforal etoposide, suggesting an in vivo selection effect. Similarly, Moscow describeda protocol for patients with metastatic breast cancer in which CD34+ cells wereexposed to an MDR1- or NeoR-containing retrovirus. After transplantation sub-strates of the MDR1 gene, including doxorubicin, vinblastine and paclitaxel wereadministered. In several patients, soon after transplantation the presence of MDR1could not be detected by PCR, although after therapy it was detectable; in contrast,there was no increase in the ability to detect NeoR-marked cells (236). This sug-gests that in vivo selection strategies may be possible and that this approachdeserves further study. This type of protocol is also being explored for individualswith NHL (237). One potential concern, however, was raised by Bunting et al.,who described the development of a myeloproliferative state in murine recipientsof MDR1 transduced ex vivo-expanded precursor cells (238). It is unclear if thisobservation was a result of the culture conditions or the vector, but it was not seenin experiments in which a control DHFR virus was used (238). A final drug-resis-tance gene that may be useful clinically is cytidine deaminase, which catalyzes thedeamination and therefore inactivation of cytosine nucleoside analogues, such ascytosine arabinoside, and could also be used to provide protection of hematopoi-etic precursors expressing this gene (239,240).

3.4. Strategies to Interfere with the Malignant GenotypeMalignant transformation may result from the loss of function of tumor sup-

pressor gene function or from acquired expression of oncogenes. The technical


difficulties associated with correcting these defects on a molecular basis isdaunting, because in either circumstance virtually all of the cells with anabnormal genotype would require successful genetic modification to expect apositive benefit of this type of therapeutic intervention. In addition, becausepatients with leukemia or lymphoma have widely disseminated disease proto-cols using gene transfer for leukemia therapy would require high gene-transferefficiency in vivo, which is clearly beyond our current capability. Nevertheless,because p53 deletions have been described in hematopoietic malignancies(241–243) and expression of the p53 gene by retroviral-mediated gene transferin a model of murine myeloid leukemia has been shown to inhibit the virulenceof the leukemia (244), this approach may theoretically be possible. Anotherscenario in which gene transfer of the p53 gene could be used is ex vivo purg-ing of leukemia or solid malignancies (245,246).

With respect to reversing malignant transformation by targeting oncogenes,specifically inactivating the causative gene within the genome may prove moredifficult than other approaches, such as using antisense methods to inhibit tran-scription of gene products, the destruction of the RNA message after transcrip-tion, or blocking the action of the translated gene. There is evidence thatantisense oligonucleotides designed to target antiapoptotic genes such as Bcl-2can inhibit proliferation and increase the sensitivity of AML blasts to cytosinearabinoside (21). Antisense therapies are currently being employed in the treat-ment of follicular lymphoma by targeting Bcl-2, and could also prove a futuretarget of gene therapy (247,248). Another means of using antisense strategiesin leukemia treatment is to inhibit the action of MDR1 genes, leading toincreased sensitivity of AML blasts to chemotherapy (249). It has previouslybeen mentioned that the BCR/ABL gene has been a target of antisenseapproaches (250). In addition, expression of a BCR deletion mutant gene deliv-ered by an adenoviral vector has proven capable of inhibiting proliferation andinducing apoptosis in BCR/ABL-positive cell lines (251). The dependence ofmalignant cells on a specific growth factor may also be a target of gene ther-apy. The proliferation of an IL-6 dependent human myeloma line was inhibitedthrough the production of an IL-6 antagonist by another cell line engineeredwith an adenoviral vector (252).

The use of ribozymes in gene therapy for hematopoietic disorders also mer-its discussion. Ribozymes are RNA molecules that bind specifically to otherRNA sequences and have the potential to cleave or alter their targets (253).Several ribozymes exist, but the hammerhead ribozyme has received signifi-cant attention for therapeutic applications (254). The hammerhead ribozyme issmall (30–40 nucleotides) and consists of three motifs: the first and third aredesigned to specifically bind to an RNA template by Watson-Crick pairing,while the second domain is catalytic, usually for a GUC site. Hammerheadribozymes have been designed and tested to cleave MDR1 sequences to inhibit


drug resistance in AML cell lines (255), AML RTO fusion transcripts (256),and PML/RARα transcripts in acute promyleocytic leukemia (APL) (257,258).In addition, Kuwabara and colleagues have tested the potential of ribozymes orrelated molecules termed maxizymes in targeting the BCR/ABL RNA tem-plate and have demonstrated high activity as well as specificity for the fusiontranscript (254,259,260). This group also documented that animals receivingCML cells transduced with a murine retroviral construct containing the max-izyme survived, while all controls died (261). Despite the encouraging natureof these reports, delivery of these molecules in vivo remains a significant hur-dle to the implementation of these strategies.

3.5. Use of “Suicide Genes” to Provide Negative SelectionGenes that alter function to facilitate specific killing of engineered cells

have been termed “suicide genes.” Most commonly these gene products pro-vide an enzymatic function that converts a relatively nontoxic “prodrug” intoan agent that is toxic to the cell. The best tested of these is the HSV-tk gene,which converts acyclovir or ganciclovir (GCV) into a monophosphate that canbe converted by the cell into the triphosphate form, which is toxic to activelydividing cells. The HSV-tk gene has been used in clinical trials in which thevirus is used to transduce intracranial tumors or localized solid tumors in vivo(58,103,262–265). In the treatment of hematopoietic malignancies, however,the HSV-tk gene has been used to enhance the safety of cellular immunother-apy (266,267). Although allogeneic transplantation remains an important ther-apeutic modality for hematologic malignancies, in part due to the immunologicprocess termed the GVL effect, the morbidity and mortality associated withgraft vs host disease (GVHD) remains high (268,269). One approach proposedto address these issues is the expression of a gene such as HSV-tk providingthe potential for negative selection in donor T-cells, either in association withthe initial transplant to assist in engraftment or with DLI administered afterrelapse (270,271). There have been reports suggesting that this is feasible inmurine models (272,273) and that the approach may prove advantageous in aclinical setting (271,274). However, it remains unclear if HSV-tk is the optimalT-cell suicide gene, because standard cytomegalovirus (CMV) prophylaxis andtreatment regimens rely on acyclovir or GCV; in addition, a cell-mediatedresponse to T-cells expressing the HSV-tk/hygromycin fusion gene has beenidentified in patients with AIDS (275). This suggests that the expression ofHSV-tk will be immunogenic and may lead to eradication of HSV-tk+ cells.Alternative genes for negative selection include cytosine deaminase, whichfacilitates the conversion of the substrate 5-fluorocytosine (5-FC) to 5-fluo-rouracil (5-FU) (276,277). Another gene undergoing testing as a means ofeliminating cells in vivo is a chimeric gene constructed to include the nervegrowth factor (NGFR) extracellular and transmembrane domains and an intra-


cellular domain derived from Fas; when exposed to a chemical dimerizer, cellsexpressing this gene are induced to undergo apoptosis (278,279). The potentialuse of other suicide genes is being explored, including the Escherichia colinitroreductase, which converts CB1954 (a mustard prodrug) to an intermedi-ate, which can be activated into an alyklating agent, and the cytochrome p450enzyme to enhance hydroxylation of cyclophosphamide (280,281).

4. FUTURE CHALLENGES IN GENE THERAPY OF LEUKEMIA/LYMPHOMA

There has been healthy skepticism regarding the current and potential useful-ness of gene therapy in the treatment of malignancies. Although many gene ther-apy protocols have enrolled patients, it has in most cases been difficult toprovide clear demonstration of a therapeutic effect. Primary limitations haveincluded the difficulty in obtaining efficient gene transfer, especially into nondi-viding cells, and the ability to maintain adequate expression of a desired productonce gene transfer is achieved. In some circumstances this has been complicatedby the use of genes producing proteins that are immunogenic. Other concernsrelate to safety issues; although the murine retroviruses used in many trials havebeen well tolerated, the death of an individual after administration of cells trans-duced with an adenoviral vector have raised concern in investigators in the field,regulatory groups, and the public. It is clear that protocol design needs to incor-porate clear means of defining outcomes to further develop the field.

Nevertheless, there is reason to be optimistic regarding the future of genetherapy. Continuing progress is being made in achieving more efficient genetransfer and expression using both physical and viral strategies. Exciting devel-opments have been described in the use of gene therapy for immunodeficien-cies and hemophilia. There has been evidence of persistence of gene modifiedcells in patients with leukemia, as well as in other disorders. Indeed, evidenceof a therapeutic effect has been suggested in the treatment of GVHD after theinfusion of HSV-tk-expressing donor lymphocytes. Further development andrefinement of gene transfer as a therapeutic tool seems inevitable but willrequire continuing evaluation for safety and efficacy.

REFERENCES1. Rubanyi GM. The future of human gene therapy. Mol Aspects Med 2001;22:113–142.2. Available at: http://www.wiley.co.uk/genetherapy/clinical/. Accessed January 02.3. Nishikawa M, Huang L. Nonviral vectors in the new millennium: delivery barriers in gene

transfer. Hum Gene Ther 2001;12:861–870.4. Wolff JA. Naked DNA transport and expression in mammalian cells. Neuromus Disord

1997;7:314–318.5. Zhang G, Budker V, Williams P, Subbotin V, Wolff JA. Efficient expression of naked dna deliv-

ered intraarterially to limb muscles of nonhuman primates. Hum Gene Ther 2001;12:427–438.


6. Hartikka J, Sukhu L, Buchner C, et al. Electroporation-facilitated delivery of plasmid DNAin skeletal muscle: plasmid dependence of muscle damage and effect of poloxamer 188.Mol Ther 2001;4:407–415.

7. Wolff JA, Williams P, Acsadi G, Jiao S, Jani A, Chong W. Conditions affecting direct genetransfer into rodent muscle in vivo. Biotechniques 1991;11:474–485.

8. Aoki M, Morishita R, Muraishi A, et al. Efficient in vivo gene transfer into the heart in therat myocardial infarction model using the HVJ (Hemagglutinating Virus of Japan)—lipo-some method. J Mol Cell Cardiol 1997;29:949–959.

9. Losordo DW, Vale PR, Symes JF, et al. Gene therapy for myocardial angiogenesis: initialclinical results with direct myocardial injection of phVEGF165 as sole therapy for myocar-dial ischemia. Circulation 1998;98:2800–2804.

10. Losordo DW, Vale PR, Isner JM. Gene therapy for myocardial angiogenesis. Am Heart J1999;138:S132–S141.

11. Lathi KG, Vale PR, Losordo DW, et al. Gene therapy with vascular endothelial growth fac-tor for inoperable coronary artery disease: anesthetic management and results. AnesthAnalg 2001;92:19–25.

12. Isner JM. Arterial gene transfer of naked DNA for therapeutic angiogenesis: early clinicalresults. Adv Drug Deliv Rev 1998;30:185–197.

13. Drabick JJ, Glasspool-Malone J, King A, Malone RW. Cutaneous transfection and immuneresponses to intradermal nucleic acid vaccination are significantly enhanced by in vivoelectropermeabilization. Mol Ther 2001;3:249–255.

14. Chesnoy S, Huang L. Enhanced cutaneous gene delivery following intradermal injection ofnaked DNA in a high ionic strength solution. Mol Ther 2002;5:57–62.

15. Hengge UR, Walker PS, Vogel JC. Expression of naked DNA in human, pig, and mouseskin. J Clin Invest 1996;97:2911–2916.

16. Simoes S, Pires P, Duzgunes N, Pedrosa de Lima MC. Cationic liposomes as gene transfer vec-tors: barriers to successful application in gene therapy. Curr Opin Mol Ther 1999;1:147–157.

17. Coutelle C, Williamson R. Liposomes and viruses for gene therapy of cystic fibrosis. JAerosol Med 1996;9:79–88.

18. Chonn A, Cullis PR. Recent advances in liposomal drug-delivery systems. Curr OpinBiotechnol 1995;6:698–708.

19. Mologni L, Marchesi E, Nielsen PE, Gambacorti-Passerini C. Inhibition of promyelocyticleukemia (PML)/retinoic acid receptor-alpha and PML expression in acute promyelocyticleukemia cells by anti-PML peptide nucleic acid. Cancer Res 2001;61:5468–5473.

20. Kitajima I, Hanyu N, Soejima Y, et al. Efficient transfer of synthetic ribozymes into cellsusing hemagglutinating virus of Japan (HVJ)-cationic liposomes. Application forribozymes that target human t-cell leukemia virus type I tax/rex mRNA. J Biol Chem1997;272:27,099–27,106.

21. Konopleva M, Tari AM, Estrov Z, et al. Liposomal Bcl-2 antisense oligonucleotidesenhance proliferation, sensitize acute myeloid leukemia to cytosine-arabinoside, andinduce apoptosis independent of other antiapoptotic proteins. Blood 2000;95:3929–3938.

22. Tari AM, Tucker SD, Deisseroth A, Lopez-Berestein G. Liposomal delivery of methylphos-phonate antisense oligodeoxynucleotides in chronic myelogenous leukemia. Blood1994;84:601–607.

23. Wu Y, Yu L, McMahon R, Rossi JJ, Forman SJ, Snyder DS. Inhibition of bcr-abl oncogeneexpression by novel deoxyribozymes (DNAzymes). Hum Gene Ther 1999;10:2847–2857.

24. Mascarenhas L, Stripecke R, Case SS, Xu D, Weinberg KI, Kohn DB. Gene delivery tohuman B-precursor acute lymphoblastic leukemia cells. Blood 1998;92:3537–3545.

25. Nabel EG, Plautz G, Nabel GJ. Site-specific gene expression in vivo by direct gene transferinto the arterial wall. Science 1990;249:1285–1258.

26. Dass CR, Walker TL, Burton MA, Decruz EE. Enhanced anticancer therapy mediated byspecialized liposomes. J Pharm Pharmacol 1997;49:972–925.


27. Ledley FD. Nonviral gene therapy: the promise of genes as pharmaceutical products. HumGene Ther 1995;6:1129–1144.

28. Verma IM, Somia N. Gene therapy—promises, problems and prospects. Nature 1997;389:239–242.

29. Xiang J, Qi Y, Chen Y. Inhibition of established tumor growth in syngeneic mice by localinoculation of engineered mouse myeloma cells secreting a recombinant fusion proteinRM4/TNF. Cancer Gene Ther 1997;4:353–358.

30. Laning J, Kawasaki H, Tanaka E, Luo Y, Dorf ME. Inhibition of in vivo tumor growth bythe beta chemokine, TCA3. J Immunol 1994;153:4625–4635.

31. Turner JG, Tan J, Crucian BE, et al. Broadened clinical utility of gene gun-mediated, gran-ulocyte-macrophage colony-stimulating factor cDNA-based tumor cell vaccines as demon-strated with a mouse myeloma model. Hum Gene Ther 1998;9:1121–1130.

32. Wright MJ, Wightman LM, Lilley C, et al. In vivo myocardial gene transfer: optimization,evaluation and direct comparison of gene transfer vectors. Basic Res Cardiol 2001;96:227–236.

33. Somia N, Verma IM. Gene therapy: trials and tribulations. Nat Rev Genet 2000;1:91–99.34. SoRelle R. Human gene therapy: science under fire. Circulation 2000;101:E9023–E9024.35. Pfeifer A, Verma IM. Gene therapy: promises and problems. Annu Rev Genomics Hum

Genet 2001;2:177–211.36. Brenner MK. Gene transfer and the treatment of haematological malignancy. J Intern Med

2001;249:345–358.37. Watanabe S, Temin HM. Construction of a helper cell line for avian reticuloendotheliosis

virus cloning vectors. Mol Cell Biol 1983;3:2241–2249.38. Mann R, Mulligan RC, Baltimore D. Construction of a retrovirus packaging mutant and its

use to produce helper-free defective retrovirus. Cell 1983;33:153–159.39. Wang H, Kavanaugh MP, North RA, Kabat D. Cell-surface receptor for ecotropic murine

retroviruses is a basic amino-acid transporter. Nature 1991;352:729–731.40. Kim JW, Closs EI, Albritton LM, Cunningham JM. Transport of cationic amino acids by

the mouse ecotropic retrovirus receptor. Nature 1991;352:725–728.41. Miller DG, Miller AD. A family of retroviruses that utilize related phosphate transporters

for cell entry. J Virol 1994;68:8270–8276.42. Kavanaugh MP, Miller DG, Zhang W, et al. Cell-surface receptors for gibbon ape leukemia

virus and amphotropic murine retrovirus are inducible sodium-dependent phosphate sym-porters. Proc Natl Acad Sci USA 1994;91:7071–7075.

43. Bauer TR, Jr., Miller AD, Hickstein DD. Improved transfer of the leukocyte integrin CD18subunit into hematopoietic cell lines by using retroviral vectors having a gibbon apeleukemia virus envelope. Blood 1995;86:2379–2387.

44. Lam JS, Reeves ME, Cowherd R, Rosenberg SA, Hwu P. Improved gene transfer intohuman lymphocytes using retroviruses with the gibbon ape leukemia virus envelope. HumGene Ther 1996;7:1415–1422.

45. Porter CD, Collins MK, Tailor CS, et al. Comparison of efficiency of infection of human genetherapy target cells via four different retroviral receptors. Hum Gene Ther 1996;7:913–919.

46. Gallardo HF, Tan C, Ory D, Sadelain M. Recombinant retroviruses pseudotyped with thevesicular stomatitis virus G glycoprotein mediate both stable gene transfer and pseudo-transduction in human peripheral blood lymphocytes. Blood 1997;90:952–957.

47. Thomsen S, Vogt B, von Laer D, et al. Lack of functional Pit-1 and Pit-2 expression onhematopoietic stem cell lines. Acta Haematol 1998;99:148–155.

48. Arai T, Matsumoto K, Saitoh K, et al. A new system for stringent, high-titer vesicular stom-atitis virus G protein-pseudotyped retrovirus vector induction by introduction of Crerecombinase into stable prepackaging cell lines. J Virol 1998;72:1115–1121.

49. Blaese RM, Culver KW, Miller AD, et al. T lymphocyte-directed gene therapy for ADA-SCID: initial trial results after 4 years. Science 1995;270:475–480.


50. Cavazzana-Calvo M, Hacein-Bey S, de Saint Basile G, et al. Gene therapy of human severecombined immunodeficiency (SCID)-X1 disease. Science 2000;288:669–672.

51. Available at: http://www4.od.nih.gov/oba/rdna.htm. Accessed.52. Vanin EF, Kaloss M, Broscius C, Nienhuis AW. Characterization of replication-competent

retroviruses from nonhuman primates with virus-induced T-cell lymphomas and observa-tions regarding the mechanism of oncogenesis. J Virol 1994;68:4241–42450.

53. Purcell DF, Broscius CM, Vanin EF, Buckler CE, Nienhuis AW, Martin MA. An array ofmurine leukemia virus-related elements is transmitted and expressed in a primate recipientof retroviral gene transfer. J Virol 1996;70:887–897.

54. Donahue RE, Kessler SW, Bodine D, et al. Helper virus induced T cell lymphoma in non-human primates after retroviral mediated gene transfer. J Exp Med 1992;176:1125–1135.

55. Cosset FL, Takeuchi Y, Battini JL, Weiss RA, Collins MK. High-titer packaging cells pro-ducing recombinant retroviruses resistant to human serum. J Virol 1995;69:7430–7436.

56. Shimizu K, Miyao Y, Tamura M, et al. Infectious retrovirus is inactivated by serum but notby cerebrospinal fluid or fluid from tumor bed in patients with malignant glioma. Jpn JCancer Res 1995;86:1010–1013.

57. Rainov NG, Kramm CM, Banning U, et al. Immune response induced by retrovirus-medi-ated HSV-tk/GCV pharmacogene therapy in patients with glioblastoma multiforme. GeneTher 2000;7:1853–1858.

58. Rainov NG. A phase III clinical evaluation of herpes simplex virus type 1 thymidine kinaseand ganciclovir gene therapy as an adjuvant to surgical resection and radiation in adultswith previously untreated glioblastoma multiforme. Hum Gene Ther 2000;11:2389–2401.

59. Culver KW, Ram Z, Wallbridge S, Ishii H, Oldfield EH, Blaese RM. In vivo gene transferwith retroviral vector-producer cells for treatment of experimental brain tumors. Science1992;256:1550–1552.

60. Buchschacher GL, Jr., Wong-Staal F. Development of lentiviral vectors for gene therapy forhuman disease. Blood 2000;95:2499–2504.

61. Naldini L, Blomer U, Gage FH, Trono D, Verma IM. Efficient transfer, integration, and sus-tained long-term expression of the transgene in adult rat brains injected with a lentiviralvector. Proc Natl Acad Sci USA 1996;93:11382–11388.

62. Naldini L. Lentiviruses as gene transfer agents for delivery to non-dividing cells. Curr OpinBiotechnol 1998;9:457–463.

63. Lewis P, Hensel M, Emerman M. Human immunodeficiency virus infection of cellsarrested in the cell cycle. Embo J 1992;11:3053–3058.

64. Lewis PF, Emerman M. Passage through mitosis is required for oncoretroviruses but not forthe human immunodeficiency virus. J Virol 1994;68:510–516.

65. Douglas JL, Lin WY, Panis ML, Veres G. Efficient human immunodeficiency virus-basedvector transduction of unstimulated human mobilized peripheral blood CD34+ cells in theSCID- hu Thy/Liv model of human T cell lymphopoiesis. Hum Gene Ther 2001;12:401–413.

66. Guenechea G, Gan OI, Inamitsu T, et al. Transduction of human CD34+ CD38– bone mar-row and cord blood-derived SCID-repopulating cells with third-generation lentiviral vec-tors. Mol Ther 2000;1:566–573.

67. Akkina RK, Walton RM, Chen ML, Li QX, Planelles V, Chen IS. High-efficiency genetransfer into CD34+ cells with a human immunodeficiency virus type 1-based retroviralvector pseudotyped with vesicular stomatitis virus envelope glycoprotein G. J Virol1996;70:2581–2585.

68. Gatlin J, Padgett A, Melkus MW, Kelly PF, Garcia JV. Long-term engraftment of nonobesediabetic/severe combined immunodeficient mice with human CD34+ cells transduced by aself-inactivating human immunodeficiency virus type 1 vector. Hum Gene Ther 2001;12:1079–1089.

69. Kumar M, Keller B, Makalou N, Sutton RE. Systematic determination of the packaginglimit of lentiviral vectors. Hum Gene Ther 2001;12:1893–1905.


70. Farson D, Witt R, McGuinness R, et al. A new-generation stable inducible packaging cellline for lentiviral vectors. Hum Gene Ther 2001;12:981–997.

71. Diaz RM, Bateman A, Emiliusen L, et al. A lentiviral vector expressing a fusogenic glyco-protein for cancer gene therapy. Gene Ther 2000;7:1656–1663.

72. Corbeau P, Kraus G, Wong-Staal F. Transduction of human macrophages using a stableHIV-1/HIV-2-derived gene delivery system. Gene Ther 1998;5:99–104.

73. Sadaie MR, Zamani M, Whang S, Sistron N, Arya SK. Towards developing HIV-2lentivirus-based retroviral vectors for gene therapy: dual gene expression in the context ofHIV-2 LTR and Tat. J Med Virol 1998;54:118–128.

74. Pandya S, Boris-Lawrie K, Leung NJ, Akkina R, Planelles V. Development of an Rev-inde-pendent, minimal simian immunodeficiency virus-derived vector system. Hum Gene Ther2001;12:847–857.

75. White SM, Renda M, Nam NY, et al. Lentivirus vectors using human and simian immunod-eficiency virus elements. J Virol 1999;73:2832–2840.

76. Browning MT, Schmidt RD, Lew KA, Rizvi TA. Primate and feline lentivirus vector RNApackaging and propagation by heterologous lentivirus virions. J Virol 2001;75:5129–5140.

77. Yamada K, Olsen JC, Patel M, Rao KW, Walsh CE. Functional correction of fanconi anemiagroup C hematopoietic cells by the use of a novel lentiviral vector. Mol Ther 2001;3:485–490.

78. Olsen JC. Gene transfer vectors derived from equine infectious anemia virus. Gene Ther1998;5:1481–1487.

79. Kochanek S. High-capacity adenoviral vectors for gene transfer and somatic gene therapy.Hum Gene Ther 1999;10:2451–2459.

80. Harui A, Suzuki S, Kochanek S, Mitani K. Frequency and stability of chromosomal integra-tion of adenovirus vectors. J Virol 1999;73:6141–6146.

81. Yeh P, Perricaudet M. Advances in adenoviral vectors: from genetic engineering to theirbiology. FASEB J 1997;11:615–623.

82. Morsy MA, Gu M, Motzel S, et al. An adenoviral vector deleted for all viral codingsequences results in enhanced safety and extended expression of a leptin transgene. ProcNatl Acad Sci USA 1998;95:7866–7871.

83. Simon RH, Engelhardt JF, Yang Y, et al. Adenovirus-mediated transfer of the CFTR gene tolung of nonhuman primates: toxicity study. Hum Gene Ther 1993;4:771–780.

84. Molnar-Kimber KL, Sterman DH, Chang M, et al. Impact of preexisting and inducedhumoral and cellular immune responses in an adenovirus-based gene therapy phase I clini-cal trial for localized mesothelioma. Hum Gene Ther 1998;9:2121–2133.

85. Qian HS, Channon K, Neplioueva V, et al. Improved adenoviral vector for vascular gene ther-apy: beneficial effects on vascular function and inflammation. Circ Res 2001;88:911–917.

86. Wang Q, Finer MH. Second-generation adenovirus vectors. Nat Med 1996;2:714–716.87. Torrent C, Jullien C, Klatzmann D, Perricaudet M, Yeh P. Transgene amplification and per-

sistence after delivery of retroviral vector and packaging functions with E1/E4-deleted ade-noviruses. Cancer Gene Ther 2000;7:1135–1144.

88. Yeh P, Dedieu JF, Orsini C, Vigne E, Denefle P, Perricaudet M. Efficient dual transcomple-mentation of adenovirus E1 and E4 regions from a 293-derived cell line expressing a mini-mal E4 functional unit. J Virol 1996;70:559–565.

89. Parks RJ, Chen L, Anton M, Sankar U, Rudnicki MA, Graham FL. A helper-dependentadenovirus vector system: removal of helper virus by Cre-mediated excision of the viralpackaging signal. Proc Natl Acad Sci USA 1996;93:13565–13570.

90. Schiedner G, Morral N, Parks RJ, et al. Genomic DNA transfer with a high-capacity aden-ovirus vector results in improved in vivo gene expression and decreased toxicity. Nat Genet1998;18:180–183.

91. Kafri T, Morgan D, Krahl T, Sarvetnick N, Sherman L, Verma I. Cellular immune responseto adenoviral vector infected cells does not require de novo viral gene expression: implica-tions for gene therapy. Proc Natl Acad Sci USA 1998;95:11377–11382.


92. Marshall E. Gene therapy death prompts review of adenovirus vector. Science 1999;286:2244–2245.

93. Lehrman S. Virus treatment questioned after gene therapy death. Nature 1999;401:517–518.94. Raper SE, Yudkoff M, Chirmule N, et al. A pilot study of in vivo liver-directed gene trans-

fer with an adenoviral vector in partial ornithine transcarbamylase deficiency. Hum GeneTher 2002;13:163–175.

95. Marshall E. Gene therapy on trial. Science 2000;288:951–957.96. Stedman H, Wilson JM, Finke R, Kleckner AL, Mendell J. Phase I clinical trial utilizing

gene therapy for limb girdle muscular dystrophy: alpha-, beta-, gamma-, or delta- sarcogly-can gene delivered with intramuscular instillations of adeno-associated vectors. Hum GeneTher 2000;11:777–790.

97. Habib NA, Hodgson HJ, Lemoine N, Pignatelli M. A phase I/II study of hepatic artery infu-sion with wtp53-CMV-Ad in metastatic malignant liver tumors. Hum Gene Ther 1999;10:2019–2034.

98. Sung MW, Yeh HC, Thung SN, et al. Intratumoral adenovirus-mediated suicide gene trans-fer for hepatic metastases from colorectal adenocarcinoma: results of a phase I clinical trial.Mol Ther 2001;4:182–191.

99. Wilson JM, Engelhardt JF, Grossman M, Simon RH, Yang Y. Gene therapy of cystic fibrosislung disease using E1 deleted adenoviruses: a phase I trial. Hum Gene Ther 1994;5:501–519.

100. Boucher RC, Knowles MR, Johnson LG, et al. Gene therapy for cystic fibrosis using E1-deleted adenovirus: a phase I trial in the nasal cavity. The University of North Carolina atChapel Hill. Hum Gene Ther 1994;5:615–639.

101. Perricone MA, Morris JE, Pavelka K, et al. Aerosol and lobar administration of a recombi-nant adenovirus to individuals with cystic fibrosis. II. Transfection efficiency in airwayepithelium. Hum Gene Ther 2001;12:1383–1394.

102. Cohen ZR, Duvdevani R, Nass D, Hadani M, Ram Z. Intraarterial delivery of genetic vec-tors for the treatment of malignant brain tumors. Isr Med Assoc J 2001;3:117–120.

103. Sandmair AM, Loimas S, Puranen P, et al. Thymidine kinase gene therapy for humanmalignant glioma, using replication-deficient retroviruses or adenoviruses. Hum Gene Ther2000;11:2197–2205.

104. Adachi Y, Tamiya T, Ichikawa T, et al. Experimental gene therapy for brain tumors usingadenovirus-mediated transfer of cytosine deaminase gene and uracil phosphoribosyltrans-ferase gene with 5-fluorocytosine. Hum Gene Ther 2000;11:77–89.

105. Schwarzenberger P, Huang W, Oliver P, Osidipe T, Theodossiou C, Kolls JK. Poly-L-lysine-based molecular conjugate vectors: a high efficiency gene transfer system for humanprogenitor and leukemia cells. Am J Med Sci 2001;321:129–136.

106. Blagosklonny MV, el-Deiry WS. In vitro evaluation of a p53-expressing adenovirus as ananti-cancer drug. Int J Cancer 1996;67:386–392.

107. Yotnda P, Onishi H, Heslop HE, et al. Efficient infection of primitive hematopoietic stemcells by modified adenovirus. Gene Ther 2001;8:930–937.

108. Meeker TC, Lay LT, Wroblewski JM, Turturro F, Li Z, Seth P. Adenoviral vectors effi-ciently target cell lines derived from selected lymphocytic malignancies, including anaplas-tic large cell lymphoma and Hodgkin’s disease. Clin Cancer Res 1997;3:357–364.

109. Medina DJ, Sheay W, Goodell L, et al. Adenovirus-mediated cytotoxicity of chronic lym-phocytic leukemia cells. Blood 1999;94:3499–3508.

110. Garcia-Castro J, Segovia JC, Garcia-Sanchez F, et al. Selective transduction of murinemyelomonocytic leukemia cells (WEHI- 3B) with regular and RGD-adenoviral vectors.Mol Ther 2001;3:70–77.

111. Clark PR, Stopeck AT, Brailey JL, et al. Polycations and cationic lipids enhance adenovirustransduction and transgene expression in tumor cells. Cancer Gene Ther 1999;6:437–446.

112. Gonzalez R, Vereecque R, Wickham TJ, et al. Transduction of bone marrow cells by theAdZ.F(pK7) modified adenovirus demonstrates preferential gene transfer in myeloma cells.Hum Gene Ther 1999;10:2709–2717.


113. Kipps TJ, Chu P, Wierda WG. Immunogenetic therapy for B-cell malignancies. SeminOncol 2000;27:104–109.

114. Takahashi S, Yotnda P, Rousseau RF, et al. Transgenic expression of CD40L and inter-leukin-2 induces an autologous antitumor immune response in patients with non-Hodgkin’slymphoma. Cancer Gene Ther 2001;8:378–387.

115. Wong CP, Levy R. Recombinant adenovirus vaccine encoding a chimeric T-cell antigen recep-tor induces protective immunity against a T-cell lymphoma. Cancer Res 2000;60:2689–2695.

116. Dotti G, Savoldo B, Takahashi S, et al. Adenovector-induced expression of human-CD40-ligand (hCD40L) by multiple myeloma cells. A model for immunotherapy. Exp Hematol2001;29:952–961.

117. Wierda WG, Cantwell MJ, Woods SJ, Rassenti LZ, Prussak CE, Kipps TJ. CD40-ligand(CD154) gene therapy for chronic lymphocytic leukemia. Blood 2000;96:2917–2924.

118. Takahashi S, Rousseau RF, Yotnda P, et al. Autologous antileukemic immune responseinduced by chronic lymphocytic leukemia B cells expressing the CD40 ligand and inter-leukin 2 transgenes. Hum Gene Ther 2001;12:659–670.

119. Xiao X, Li J, McCown TJ, Samulski RJ. Gene transfer by adeno-associated virus vectorsinto the central nervous system. Exp Neurol 1997;144:113–124.

120. Kotin RM, Menninger JC, Ward DC, Berns KI. Mapping and direct visualization of aregion-specific viral DNA integration site on chromosome 19q13-qter. Genomics1991;10:831–834.

121. Kotin RM, Siniscalco M, Samulski RJ, et al. Site-specific integration by adeno-associatedvirus. Proc Natl Acad Sci USA 1990;87:2211–2215.

122. Ponnazhagan S, Erikson D, Kearns WG, et al. Lack of site-specific integration of therecombinant adeno-associated virus 2 genomes in human cells. Hum Gene Ther1997;8:275–284.

123. Kearns WG, Afione SA, Fulmer SB, et al. Recombinant adeno-associated virus (AAV-CFTR) vectors do not integrate in a site-specific fashion in an immortalized epithelial cellline. Gene Ther 1996;3:748–755.

124. Siegl G, Bates RC, Berns KI, et al. Characteristics and taxonomy of Parvoviridae. Intervi-rology 1985;23:61–73.

125. Muzyczka N. Use of adeno-associated virus as a general transduction vector for mam-malian cells. Curr Top Microbiol Immunol 1992;158:97–129.

126. Dong JY, Fan PD, Frizzell RA. Quantitative analysis of the packaging capacity of recombi-nant adeno-associated virus. Hum Gene Ther 1996;7:2101–2112.

127. Duan D, Sharma P, Yang J, et al. Circular intermediates of recombinant adeno-associatedvirus have defined structural characteristics responsible for long-term episomal persistencein muscle tissue. J Virol 1998;72:8568–8577.

128. Xiao X, Li J, Samulski RJ. Production of high-titer recombinant adeno-associated virusvectors in the absence of helper adenovirus. J Virol 1998;72:2224–2232.

129. Duan D, Yue Y, Engelhardt JF. Expanding AAV packaging capacity with trans-splicing oroverlapping vectors: a quantitative comparison. Mol Ther 2001;4:383–391.

130. Duan D, Yan Z, Yue Y, Ding W, Engelhardt JF. Enhancement of muscle gene delivery withpseudotyped adeno-associated virus type 5 correlates with myoblast differentiation. J Virol2001;75:7662–7671.

131. Xiao W, Chirmule N, Berta SC, McCullough B, Gao G, Wilson JM. Gene therapy vectorsbased on adeno-associated virus type 1. J Virol 1999;73:3994–4003.

132. Bjorklund A, Kirik D, Rosenblad C, Georgievska B, Lundberg C, Mandel RJ. Towards aneuroprotective gene therapy for Parkinson’s disease: use of adenovirus, AAV andlentivirus vectors for gene transfer of GDNF to the nigrostriatal system in the rat Parkinsonmodel. Brain Res 2000;886:82–98.

133. Hartigan-O’Connor D, Chamberlain JS. Developments in gene therapy for muscular dys-trophy. Microsc Res Tech 2000;48:223–238.


134. Kay MA, Manno CS, Ragni MV, et al. Evidence for gene transfer and expression of factorIX in haemophilia B patients treated with an AAV vector. Nat Genet 2000;24:257–261.

135. High KA. Gene therapy: a 2001 perspective. Haemophilia 2001;7(suppl 1):23–27.136. Ponnazhagan S, Mukherjee P, Wang XS, et al. Adeno-associated virus type 2-mediated

transduction in primary human bone marrow-derived CD34+ hematopoietic progenitorcells: donor variation and correlation of transgene expression with cellular differentiation. JVirol 1997;71:8262–8267.

137. Ponnazhagan S, Yoder MC, Srivastava A. Adeno-associated virus type 2-mediated trans-duction of murine hematopoietic cells with long-term repopulating ability and sustainedexpression of a human globin gene in vivo. J Virol 1997;71:3098–3104.

138. Goodman S, Xiao X, Donahue RE, et al. Recombinant adeno-associated virus-mediatedgene transfer into hematopoietic progenitor cells. Blood 1994;84:1492–1500.

139. Anderson R, Macdonald I, Corbett T, Hacking G, Lowdell MW, Prentice HG. Constructionand biological characterization of an interleukin-12 fusion protein (Flexi-12): delivery to acutemyeloid leukemic blasts using adeno-associated virus. Hum Gene Ther 1997;8:1125–1135.

140. Luo F, Zhou SZ, Cooper S, et al. Adeno-associated virus 2-mediated gene transfer andfunctional expression of the human granulocyte-macrophage colony-stimulating factor.Exp Hematol 1995;23:1261–1267.

141. Omori F, Messner HA, Ye C, et al. Nontargeted stable integration of recombinant adeno-associated virus into human leukemia and lymphoma cell lines as evaluated by fluorescencein situ hybridization. Hum Gene Ther 1999;10:537–543.

142. Ponnazhagan S, Weigel KA, Raikwar SP, Mukherjee P, Yoder MC, Srivastava A. Recombi-nant human parvovirus B19 vectors: erythroid cell-specific delivery and expression oftransduced genes. J Virol 1998;72:5224–5230.

143. Rabinowitz JE, Samulski RJ. Building a better vector: the manipulation of AAV virions.Virology 2000;278:301–308.

144. Robbins PD, Ghivizzani SC. Viral vectors for gene therapy. Pharmacol Ther 1998;80:35–47.145. Burton EA, Wechuck JB, Wendell SK, Goins WF, Fink DJ, Glorioso JC. Multiple applica-

tions for replication-defective herpes simplex virus vectors. Stem Cells 2001;19:358–377.146. Ozuer A, Wechuck JB, Goins WF, Wolfe D, Glorioso JC, Ataai MM. Effect of genetic back-

ground and culture conditions on the production of herpesvirus-based gene therapy vectors.Biotechnol Bioeng 2002;77:685–692.

147. Carlezon WA, Jr., Nestler EJ, Neve RL. Herpes simplex virus-mediated gene transfer as atool for neuropsychiatric research. Crit Rev Neurobiol 2000;14:47–67.

148. Wood MJ, Byrnes AP, Pfaff DW, Rabkin SD, Charlton HM. Inflammatory effects of genetransfer into the CNS with defective HSV-1 vectors. Gene Ther 1994;1:283–291.

149. Maguir-Zeis KA, Bowers WJ, Federoff HJ. HSV vector-mediated gene delivery to the cen-tral nervous system. Curr Opin Mol Ther 2001;3:482–490.

150. Markert JM, Medlock MD, Rabkin SD, et al. Conditionally replicating herpes simplexvirus mutant, G207 for the treatment of malignant glioma: results of a phase I trial. GeneTher 2000;7:867–874.

151. Tolba KA, Bowers WJ, Hilchey SP, et al. Development of herpes simplex virus-1 amplicon-based immunotherapy for chronic lymphocytic leukemia. Blood 2001;98:287–295.

152. Eling DJ, Johnson PA, Sharma S, Tufaro F, Kipps TJ. Chronic lymphocytic leukemia Bcells are highly sensitive to infection by herpes simplex virus-1 via herpesvirus-entry-medi-ator A. Gene Ther 2000;7:1210–1216.

153. Wu A, Mazumder A, Martuza RL, et al. Biological purging of breast cancer cells using anattenuated replication-competent herpes simplex virus in human hematopoietic stem celltransplantation. Cancer Res 2001;61:3009–3015.

154. Rosenberg SA, Aebersold P, Cornetta K, et al. Gene transfer into humans—immunotherapyof patients with advanced melanoma, using tumor-infiltrating lymphocytes modified byretroviral gene transduction. N Engl J Med 1990;323:570–578.


155. Brenner MK, Rill DR, Moen RC, et al. Gene-marking to trace origin of relapse after autol-ogous bone-marrow transplantation. Lancet 1993;341:85–86.

156. Deisseroth AB, Zu Z, Claxton D, et al. Genetic marking shows that Ph+ cells present inautologous transplants of chronic myelogenous leukemia (CML) contribute to relapse afterautologous bone marrow in CML. Blood 1994;83:3068–3076.

157. Bachier CR, Giles RE, Ellerson D, et al. Hematopoietic retroviral gene marking in patientswith follicular non-Hodgkin’s lymphoma. Leukemia Lymphoma 1999;32:279–288.

158. Dunbar CE, Cottler-Fox M, O’Shaughnessy JA, et al. Retrovirally marked CD34-enrichedperipheral blood and bone marrow cells contribute to long-term engraftment after autolo-gous transplantation. Blood 1995;85:3048–3057.

159. Stewart AK, Sutherland DR, Nanji S, et al. Engraftment of gene-marked hematopoieticprogenitors in myeloma patients after transplant of autologous long-term marrow cultures.Hum Gene Ther 1999;10:1953–1964.

160. Brenner MK. Hematological malignancies. FASEB J 1997;11:640–648.161. Miller AR, Skotzko MJ, Rhoades K, et al. Simultaneous use of two retroviral vectors in

human gene marking trials: feasibility and potential applications. Hum Gene Ther1992;3:619–624.

162. Heslop HE, Rooney CM, Brenner MK. Gene-marking and haemopoietic stem-cell trans-plantation. Blood Rev 1995;9:220–225.

163. Cornetta K, Srour EF, Moore A, et al. Retroviral gene transfer in autologous bone marrowtransplantation for adult acute leukemia. Hum Gene Ther 1996;7:1323–1329.

164. Rooney CM, Smith CA, Ng CY, et al. Use of gene-modified virus-specific T-lymphocytesto control Epstein-Barr-virus-related lymphoproliferation. Lancet 1995;345:9–13.

165. Heslop H, Rooney C, Brenner M, et al. Administration of neomycin resistance gene-marked EBV-specific cytotoxic T-lymphocytes as therapy for patients receiving a bonemarrow transplant for relapsed EBV-positive Hodgkin disease. Hum Gene Ther 2000;11:1465–1475.

166. Savoldo B, Cubbage ML, Durett AG, et al. Generation of EBV-Specific CD4(+) cytotoxic Tcells from virus naive individuals. J Immunol 2002;168:909–918.

167. Liu Z, Savoldo B, Huls H, et al. Epstein-Barr virus (EBV)-specific cytotoxic T-lympho-cytes for the prevention and treatment of EBV-associated post-transplant lymphomas.Recent Results Cancer Res 2002;159:123–133.

168. Gahn B, Siller-Lopez F, Pirooz AD, et al. Adenoviral gene transfer into dendritic cells effi-ciently amplifies the immune response to LMP2A antigen: a potential treatment strategy forEpstein-Barr virus–positive Hodgkin’s lymphoma. Int J Cancer 2001;93:706–713.

169. Bagnis C, Chabannon C, Gravis G, et al. Transient detection of β-galactosidase activity inhematopoietic cells, following reinjection of retrovirally marked autologous blood progeni-tors in patients with breast or ovarian cancer receiving high- dose chemotherapy. ExpHematol 2002;30:108–115.

170. Heim DA, Hanazono Y, Giri N, et al. Introduction of a xenogeneic gene via hematopoieticstem cells leads to specific tolerance in a rhesus monkey model. Mol Ther 2000;1:533–544.

171. Finke JH, Zea AH, Stanley J, et al. Loss of T-cell receptor zeta chain and p56lck in T-cellsinfiltrating human renal cell carcinoma. Cancer Res 1993;53:5613–5616.

172. Isomaki P, Panesar M, Annenkov A, et al. Prolonged exposure of T cells to TNF down-reg-ulates TCR zeta and expression of the TCR/CD3 complex at the cell surface. J Immunol2001;166:5495–5507.

173. Horiguchi S, Petersson M, Nakazawa T, et al. Primary chemically induced tumors induceprofound immunosuppression concomitant with apoptosis and alterations in signal trans-duction in T cells and NK cells. Cancer Res 1999;59:2950–2956.

174. Mizoguchi H, O’Shea JJ, Longo DL, Loeffler CM, McVicar DW, Ochoa AC. Alterations insignal transduction molecules in T lymphocytes from tumor-bearing mice. Science1992;258:1795–1798.


175. Nagata S. Fas ligand and immune evasion. Nat Med 1996;2:1306–1307.176. Garrido F, Cabrera T, Concha A, Glew S, Ruiz-Cabello F, Stern PL. Natural history of HLA

expression during tumour development. Immunol Today 1993;14:491–499.177. Mathe G, Amiel JL, Schwarzenberg L, et al. Active immunotherapy for acute lymphoblas-

tic leukaemia. Lancet 1969;1:697–699.178. Dranoff G, Jaffee E, Lazenby A, et al. Vaccination with irradiated tumor cells engineered to

secrete murine granulocyte-macrophage colony-stimulating factor stimulates potent, spe-cific, and long-lasting anti-tumor immunity. Proc Natl Acad Sci USA 1993;90:3539–3543.

179. Tepper RI, Pattengale PK, Leder P. Murine interleukin-4 displays potent anti-tumor activityin vivo. Cell 1989;57:503–512.

180. Fearon ER, Pardoll DM, Itaya T, et al. Interleukin-2 production by tumor cells bypasses Thelper function in the generation of an antitumor response. Cell 1990;60:397–403.

181. Gansbacher B, Bannerji R, Daniels B, Zier K, Cronin K, Gilboa E. Retroviral vector-medi-ated gamma-interferon gene transfer into tumor cells generates potent and long lasting anti-tumor immunity. Cancer Res 1990;50:7820–7825.

182. Pardoll DM. Cancer vaccines. Immunol Today 1993;14:310–316.183. Forni G, Parmiani G, Guarini A, Foa R. Gene transfer in tumor therapy [review]. Ann Oncol

1994;5:789–794.184. Forni G, Giovarelli M, Cavallo F, et al. Cytokine-induced tumor immunogenicity: from

exogenous cytokines to gene therapy [review]. J Immunother 1993;14:253–257.185. Dunussi-Joannopoulos K, Weinstein HJ, Arceci RJ, Croop JM. Gene therapy with B7.1 and

GM-CSF vaccines in a murine AML model. J Pediatr Hematol Oncol 1997;19:536–540.186. Cignetti A, Guarini A, Carbone A, et al. Transduction of the IL2 gene into human acute

leukemia cells: induction of tumor rejection without modifying cell proliferation and IL2receptor expression. J Natl Cancer Inst 1994;86:785–791.

187. Orchard PJ, Katsanis E, Boyer M, May C, McIvor RS, Blazar BR. Interleukin-2 secretion bytransduced and unselected BDL-2 lymphoma results in increased survival in mice with pre-viously established disseminated disease. Cancer Biother Radiopharm 1996;11:155–164.

188. Dilloo D, Bacon K, Holden W, et al. Combined chemokine and cytokine gene transferenhances antitumor immunity. Nat Med 1996;2:1090–1095.

189. Kopantzev E, Roschke V, Rudikoff S. Interleukin-2-mediated modulation of plasma celltumor growth in a model of multiple myeloma. Hum Gene Ther 1998;9:13–19.

190. Hsieh CL, Pang VF, Chen DS, Hwang LH. Regression of established mouse leukemia byGM-CSF-transduced tumor vaccine: implications for cytotoxic T lymphocyte responsesand tumor burdens. Hum Gene Ther 1997;8:1843–1854.

191. Stewart AK, Schimmer AD, Bailey DJ, et al. In vivo adenoviral-mediated gene transfer ofinterleukin-2 in cutaneous plasmacytoma. Blood 1998;91:1095–1097.

192. Dunussi-Joannopoulos K, Leonard JP. Interleukin-12 gene therapy vaccines: directing theimmune system against minimal residual leukemia. Leukemia Lymphoma 2001;41:483–492.

193. Dunussi-Joannopoulos K, Runyon K, Erickson J, Schaub RG, Hawley RG, Leonard JP.Vaccines with interleukin-12-transduced acute myeloid leukemia cells elicit very potenttherapeutic and long-lasting protective immunity. Blood 1999;94:4263–4273.

194. Dunussi-Joannopoulos K, Dranoff G, Weinstein HJ, Ferrara JL, Bierer BE, Croop JM.Gene immunotherapy in murine acute myeloid leukemia: granulocyte-macrophage colony-stimulating factor tumor cell vaccines elicit more potent antitumor immunity comparedwith B7 family and other cytokine vaccines. Blood 1998;91:222–230.

195. Stripecke R, Skelton DC, Pattengale PK, Shimada H, Kohn DB. Combination of CD80 andgranulocyte-macrophage colony-stimulating factor coexpression by a leukemia cell vac-cine: preclinical studies in a murine model recapitulating Philadelphia chromosome-posi-tive acute lymphoblastic leukemia. Hum Gene Ther 1999;10:2109–2122.

196. Borrello I, Sotomayor EM, Cooke S, Levitsky HI. A universal granulocyte-macrophagecolony-stimulating factor-producing bystander cell line for use in the formulation of autol-ogous tumor cell- based vaccines. Hum Gene Ther 1999;10:1983–1991.


197. Nelson EL, Li X, Hsu FJ, et al. Tumor-specific, cytotoxic T-lymphocyte response after idio-type vaccination for B-cell, non-Hodgkin’s lymphoma. Blood 1996;88:580–589.

198. Steinman RM, Lustig DS, Cohn ZA. Identification of a novel cell type in peripheral lym-phoid organs of mice. 3. Functional properties in vivo. J Exp Med 1974;139:1431–1445.

199. Steinman RM, Adams JC, Cohn ZA. Identification of a novel cell type in peripheral lym-phoid organs of mice. IV. Identification and distribution in mouse spleen. J Exp Med1975;141:804–820.

200. Baggers J, Ratzinger G, Young JW. Dendritic cells as immunologic adjuvants for the treat-ment of cancer. J Clin Oncol 2000;18:3879–3882.

201. Ribas A, Butterfield LH, Economou JS. Genetic immunotherapy for cancer. Oncologist2000;5:87–98.

202. Movassagh M, Baillou C, Cosset FL, Klatzmann D, Guigon M, Lemoine FM. High level ofretrovirus-mediated gene transfer into dendritic cells derived from cord blood and mobi-lized peripheral blood CD34+ cells. Hum Gene Ther 1999;10:175–187.

203. Hsu FJ, Benike C, Fagnoni F, et al. Vaccination of patients with B-cell lymphoma usingautologous antigen- pulsed dendritic cells. Nat Med 1996;2:52–58.

204. Mitchell DA, Nair SK. RNA-transfected dendritic cells in cancer immunotherapy. J ClinInvest 2000;106:1065–1069.

205. Westermann J, Kopp J, Korner I, et al. Bcr/abl+ autologous dendritic cells for vaccinationin chronic myeloid leukemia. Bone Marrow Transplant 2000;25(suppl 2):S46–S49.

206. Choudhury BA, Liang JC, Thomas EK, et al. Dendritic cells derived in vitro from acutemyelogenous leukemia cells stimulate autologous, antileukemic T-cell responses. Blood1999;93:780–786.

207. Paczesny S, Beranger S, Salzmann JL, Klatzmann D, Colombo BM. Protection of miceagainst leukemia after vaccination with bone marrow-derived dendritic cells loaded withapoptotic leukemia cells. Cancer Res 2001;61:2386–2389.

208. Shedlock DJ, Weiner DB. DNA vaccination: antigen presentation and the induction ofimmunity. J Leukoc Biol 2000;68:793–806.

209. Reichardt VL, Okada CY, Liso A, et al. Idiotype vaccination using dendritic cells afterautologous peripheral blood stem cell transplantation for multiple myeloma—a feasibilitystudy. Blood 1999;93:2411–2419.

210. He L, Feng H, Raymond A, et al. Dendritic-cell-peptide immunization provides immuno-protection against bcr-abl-positive leukemia in mice. Cancer Immunol Immunother2001;50:31–40.

211. Rosenberg SA, Anderson WF, Blaese M, et al. The development of gene therapy for thetreatment of cancer. Ann Surg 1993;218:455–463; discussion 463–464.

212. Hwu P, Yannelli J, Kriegler M, et al. Functional and molecular characterization of tumor-infiltrating lymphocytes transduced with tumor necrosis factor-alpha cDNA for the genetherapy of cancer in humans. J Immunol 1993;150:4104–4115.

213. Miller J, Tessemer-Tuck J, Blake N, et al. Endogenous IL-2 production by natural killercells maintains cytotoxic and proliferative capacity following retroviral mediated genetransfer. Exp Hematol 1997;25:1140–1148.

214. Korbelik M, Sun J. Cancer treatment by photodynamic therapy combined with adoptiveimmunotherapy using genetically altered natural killer cell line. Int J Cancer2001;93:269–274.

215. Treisman J, Hwu P, Minamoto S, et al. Interleukin-2-transduced lymphocytes grow in anautocrine fashion and remain responsive to antigen. Blood 1995;85:139–145.

216. Tam YK, Klingemann HG. Antileukemic effect of interleukin-2-transduced murine bonemarrow after autologous transplantation. Biol Blood Marrow Transplant 1999;5:231–242.

217. Minamoto S, Treisman J, Hankins WD, Sugamura K, Rosenberg SA. Acquired erythropoi-etin responsiveness of interleukin-2-dependent T lymphocytes retrovirally transduced withgenes encoding chimeric erythropoietin/interleukin-2 receptors. Blood 1995;86:2281–2287.


218. Dembic Z, Haas W, Weiss S, et al. Transfer of specificity by murine alpha and beta T-cellreceptor genes. Nature 1986;320:232–238.

219. Gross G, Gorochov G, Waks T, Eshhar Z. Generation of effector T cells expressingchimeric T cell receptor with antibody type-specificity. Transplant Proc 1989;21:127–130.

220. Gross G, Waks T, Eshhar Z. Expression of immunoglobulin-T-cell receptor chimeric mole-cules as functional receptors with antibody-type specificity. Proc Natl Acad Sci USA1989;86:10024–10028.

221. Bird RE, Hardman KD, Jacobson JW, et al. Single-chain antigen-binding proteins. Science1988;242:423–426.

222. Eshhar Z, Waks T, Gross G, Schindler DG. Specific activation and targeting of cytotoxiclymphocytes through chimeric single chains consisting of antibody-binding domains andthe gamma or zeta subunits of the immunoglobulin and T-cell receptors. Proc Natl Acad SciUSA 1993;90:720–724.

223. Hwu P, Shafer GE, Treisman J, et al. Lysis of ovarian cancer cells by human lymphocytesredirected with a chimeric gene composed of an antibody variable region and the Fc recep-tor gamma chain. J Exp Med 1993;178:361–366.

224. Hwu P, Yang JC, Cowherd R, et al. In vivo antitumor activity of T cells redirected withchimeric antibody/T-cell receptor genes. Cancer Res 1995;55:3369–3373.

225. Moritz D, Wels W, Mattern J, Groner B. Cytotoxic T lymphocytes with a grafted recognitionspecificity for ERBB2-expressing tumor cells. Proc Natl Acad Sci USA 1994;91:4318–4322.

226. Altenschmidt U, Klundt E, Groner B. Adoptive transfer of in vitro-targeted, activated Tlymphocytes results in total tumor regression. J Immunol 1997;159:5509–5515.

227. Hwu P, Rosenberg SA. The genetic modification of T cells for cancer therapy: an overviewof laboratory and clinical trials. Cancer Detect Prev 1994;18:43–50.

228. Hombach A, Schneider C, Sent D, et al. An entirely humanized CD3 zeta-chain signalingreceptor that directs peripheral blood t cells to specific lysis of carcinoembryonic antigen-positive tumor cells. Int J Cancer 2000;88:115–120.

229. Jensen MC, Clarke P, Tan G, et al. Human T lymphocyte genetic modification with nakedDNA. Mol Ther 2000;1:49–55.

230. McIvor RS. Drug-resistant dihydrofolate reductases: generation, expression and therapeuticapplication. Bone Marrow Transplant 1996;18(suppl 3):S50–S54.

231. James RI, May C, Vagt MD, Studebaker R, McIvor RS. Transgenic mice expressing thetyr22 variant of murine DHFR: protection of transgenic marrow transplant recipients fromlethal doses of methotrexate. Exp Hematol 1997;25:1286–1295.

232. McIvor RS, Simonsen CC. Isolation and characterization of a variant dihydrofolate reduc-tase cDNA from methotrexate-resistant murine L5178Y cells. Nucleic Acids Res1990;18:7025–7032.

233. Zhao RC, McIvor RS, Griffin JD, Verfaillie CM. Gene therapy for chronic myelogenousleukemia (CML): a retroviral vector that renders hematopoietic progenitors methotrexate-resistant and CML progenitors functionally normal and nontumorigenic in vivo. Blood1997;90:4687–4698.

234. Kondratov RV, Komarov PG, Becker Y, Ewenson A, Gudkov AV. Small molecules that dra-matically alter multidrug resistance phenotype by modulating the substrate specificity of P-glycoprotein. Proc Natl Acad Sci USA 2001;98:14078–14083.

235. Abonour R, Williams DA, Einhorn L, et al. Efficient retrovirus-mediated transfer of themultidrug resistance 1 gene into autologous human long-term repopulating hematopoieticstem cells. Nat Med 2000;6:652–658.

236. Moscow JA, Huang H, Carter C, et al. Engraftment of MDR1 and NeoR gene-transducedhematopoietic cells after breast cancer chemotherapy. Blood 1999;94:52–61.

237. Devereux S, Corney C, Macdonald C, et al. Feasibility of multidrug resistance (MDR-1)gene transfer in patients undergoing high-dose therapy and peripheral blood stem cell trans-plantation for lymphoma. Gene Ther 1998;5:403–408.


238. Bunting KD, Galipeau J, Topham D, Benaim E, Sorrentino BP. Transduction of murinebone marrow cells with an MDR1 vector enables ex vivo stem cell expansion, but theseexpanded grafts cause a myeloproliferative syndrome in transplanted mice. Blood1998;92:2269–2279.

239. Beausejour CM, Eliopoulos N, Momparler L, Le NL, Momparler RL. Selection of drug-resistant transduced cells with cytosine nucleoside analogs using the human cytidine deam-inase gene. Cancer Gene Ther 2001;8:669–676.

240. Eliopoulos N, Bovenzi V, Le NL, et al. Retroviral transfer and long-term expression ofhuman cytidine deaminase cDNA in hematopoietic cells following transplantation in mice.Gene Ther 1998;5:1545–1551.

241. Newcomb EW. P53 gene mutations in lymphoid diseases and their possible relevance todrug resistance. Leukemia Lymphoma 1995;17:211–221.

242. Tsai T, Davalath S, Rankin C, et al. Tumor suppressor gene alteration in adult acute lym-phoblastic leukemia (ALL). Analysis of retinoblastoma (Rb) and p53 gene expression inlymphoblasts of patients with de novo, relapsed, or refractory ALL treated in SouthwestOncology Group studies. Leukemia 1996;10:1901–1910.

243. Zhou M, Gu L, Yeager AM, Findley HW. Sensitivity to Fas-mediated apoptosis in pediatricacute lymphoblastic leukemia is associated with a mutant p53 phenotype and absence ofBcl-2 expression. Leukemia 1998;12:1756–1763.

244. Bossi G, Scardigli R, Musiani P, et al. Development of a murine orthotopic model ofleukemia: evaluation of TP53 gene therapy efficacy. Cancer Gene Ther 2000;7:135–143.

245. Scardigli R, Bossi G, Blandino G, Crescenzi M, Soddu S, Sacchi A. Expression of exoge-nous wt-p53 does not affect normal hematopoiesis: implications for bone marrow purging.Gene Ther 1997;4:1371–1378.

246. Hirai M, LaFace D, Robinson S, et al. Ex vivo purging by adenoviral p53 gene therapy doesnot affect NOD-SCID repopulating activity of human CD34+ cells. Cancer Gene Ther2001;8:936–947.

247. Cotter FE, Johnson P, Hall P, et al. Antisense oligonucleotides suppress B-cell lymphomagrowth in a SCID-hu mouse model. Oncogene 1994;9:3049–3055.

248. Cotter FE, Waters J, Cunningham D. Human Bcl-2 antisense therapy for lymphomas.Biochim Biophys Acta 1999;1489:97–106.

249. Motomura S, Motoji T, Takanashi M, et al. Inhibition of P-glycoprotein and recovery ofdrug sensitivity of human acute leukemic blast cells by multidrug resistance gene (mdr1)antisense oligonucleotides. Blood 1998;91:3163–3171.

250. Zhao RC, Jiang Y, Verfaillie CM. A model of human p210(bcr/ABL)-mediated chronicmyelogenous leukemia by transduction of primary normal human CD34(+) cells with aBCR/ABL-containing retroviral vector. Blood 2001;97:2406–2412.

251. Wang Y, Liu J, Wu Y, et al. Expression of a truncated first exon BCR sequence in chronicmyelogenous leukemia cells blocks cell growth and induces cell death. Cancer Res2001;61:138–144.

252. Saggio I, Ciapponi L, Savino R, Ciliberto G, Perricaudet M. Adenovirus-mediated genetransfer of a human IL-6 antagonist. Gene Ther 1997;4:839–845.

253. Phylactou LA. Ribozyme and peptide-nucleic acid-based gene therapy. Adv Drug DelivRev 2000;44:97–108.

254. Kuwabara T, Warashina M, Tanabe T, Tani K, Asano S, Taira K. Comparison of the specifici-ties and catalytic activities of hammerhead ribozymes and DNA enzymes with respect to thecleavage of BCR-ABL chimeric L6 (b2a2) mRNA. Nucleic Acids Res 1997;25:3074–3081.

255. Matsushita H, Kizaki M, Kobayashi H, et al. Restoration of retinoid sensitivity by MDR1ribozymes in retinoic acid- resistant myeloid leukemic cells. Blood 1998;91:2452–2458.

256. Kozu T, Sugio Y, Fukuyama T, Sueoka E, Otsuka T, Akagi K. Ribonuclease H attack ofleukaemic fused transcripts AML1-MTG8 (ETO) by DNA/RNA chimeric hammerheadribozymes. Genes Cells 2000;5:637–647.


257. Suwanai H, Matsushita H, Kobayashi H, Ikeda Y, Kizaki M. A novel therapeutic technologyof specific RNA inhibition for acute promyelocytic leukemia: improved design of max-izymes against PML/RARalpha mRNA. Int J Oncol 2002;20:127–130.

258. Nason-Burchenal K, Allopenna J, Begue A, Stehelin D, Dmitrovsky E, Martin P. Targetingof PML/RARalpha is lethal to retinoic acid-resistant promyelocytic leukemia cells. Blood1998;92:1758–1767.

259. Kuwabara T, Hamada M, Warashina M, Taira K. Allosterically controlled single-chainedmaxizymes with extremely high and specific activity. Biomacromolecules 2001;2:788–799.

260. Kuwabara T, Warashina M, Tanabe T, Tani K, Asano S, Taira K. A novel allostericallytrans-activated ribozyme, the maxizyme, with exceptional specificity in vitro and in vivo.Mol Cell 1998;2:617–627.

261. Tanabe T, Kuwabara T, Warashina M, Tani K, Taira K, Asano S. Oncogene inactivation in amouse model. Nature 2000;406:473–474.

262. Zwacka RM, Dunlop MG. Gene therapy for colon cancer. Hematol Oncol Clin North Am1998;12:595–615.

263. Kunishige I, Samejima Y, Shiki Y, et al. Suicide gene therapy for human uterine adenocarci-noma cells using herpes simplex virus thymidine kinase. Gynecol Oncol 1999;72:16–25.

264. Yamamoto S, Suzuki S, Hoshino A, Akimoto M, Shimada T. Herpes simplex virus thymi-dine kinase/ganciclovir-mediated killing of tumor cells induces tumor-specific cytotoxic tcells in mice. Cancer Gene Ther 1997;4:91–96.

265. Klatzmann D. Gene therapy for metastatic malignant melanoma: evaluation of tolerance tointratumoral injection of cells producing recombinant retroviruses carrying the herpes sim-plex virus type 1 thymidine kinase gene, to be followed by ganciclovir administration. HumGene Ther 1996;7:255–667.

266. Tiberghien P, Reynolds CW, Keller J, et al. Ganciclovir treatment of herpes simplex thymi-dine kinase-transduced primary T lymphocytes: an approach for specific in vivo donor T-cell depletion after bone marrow transplantation? Blood 1994;84:1333–1341.

267. Bordignon C, Bonini C, Verzeletti S, et al. Transfer of the hsv-tk gene into donor peripheralblood lymphocytes for in vivo modulation of donor anti-tumor immunity after allogeneicbone marrow transplantation. Hum Gene Ther 1995;6:813–819.

268. Gale RP, Horowitz MM. Graft-versus-leukemia in bone marrow transplantation. The Advi-sory Committee of the International Bone Marrow Transplant Registry. Bone MarrowTransplant 1990;6(suppl 1):94–97.

269. Horowitz MM, Gale RP, Sondel PM, et al. Graft-versus-leukemia reactions after bone mar-row transplantation. Blood 1990;75:555–562.

270. Tiberghien P, Cahn JY, Brion A, et al. Use of donor t-lymphocytes expressing herpes-sim-plex thymidine kinase in allogeneic bone marrow transplantation—a phase i–ii study. HumGene Ther 1997;8:615–624.

271. Bonini C, Ferrari G, Verzeletti S, et al. Hsv-tk gene transfer into donor lymphocytes forcontrol of allogeneic graft-versus-leukemia. Science 1997;276:1719–1724.

272. Cohen JL, Boyer O, Salomon B, et al. Prevention of graft-versus-host disease in mice usinga suicide gene expressed in T lymphocytes. Blood 1997;89:4636–4645.

273. Contassot E, Ferrand C, Angonin R, et al. Ganciclovir-sensitive acute graft-versus-host dis-ease in mice receiving herpes simplex virus-thymidine kinase-expressing donor T cells in abone marrow transplantation setting. Transplantation 2000;69:503–508.

274. Tiberghien P. Use of suicide gene-expressing donor T-cells to control alloreactivity afterhaematopoietic stem cell transplantation. J Intern Med 2001;249:369–377.

275. Riddell SR, Elliott M, Lewinsohn DA, et al. T-cell mediated rejection of gene-modified hiv-specific cytotoxic t lymphocytes in hiv-infected patients. Nat Med 1996;2:216–223.

276. Haberkorn U, Oberdorfer F, Gebert J, et al. Monitoring gene therapy with cytosine deami-nase: in vitro studies using tritiated-5-fluorocytosine. J Nucl Med 1996;37:87–94.


277. Mullen CA, Kilstrup M, Blaese RM. Transfer of the bacterial gene for cytosine deaminaseto mammalian cells confers lethal sensitivity to 5-fluorocytosine: a negative selection sys-tem. Proc Natl Acad Sci USA 1992;89:33–37.

278. Amara JF, Courage NL, Gilman M. Cell surface tagging and a suicide mechanism in a sin-gle chimeric human protein. Human Gene Ther 1999;10:2651–2655.

279. Thomis DC, Marktel S, Bonini C, et al. A Fas-based suicide switch in human T cells for thetreatment of graft- versus-host disease. Blood 2001;97:1249–1257.

280. Beltinger C, Uckert W, Debatin KM. Suicide gene therapy for pediatric tumors. J Mol Med2001;78:598–612.

281. Greco O, Dachs GU. Gene directed enzyme/prodrug therapy of cancer: historical appraisaland future prospectives. J Cell Physiol 2001;187:22–36.


Index 261

261

INDEX

AAAV, see Adeno-associated virusABC transporters, see Multidrug-

resistance associated protein;P-glycoprotein

Activating mutations, cytokinereceptors, 130, 131

Acute lymphocytic leukemia (ALL),apoptosis dysregulation, 172drug resistance relationship, 173,

174interleukin-2 therapy, see

Interleukin-2relapse and prognosis, 93, 94

Acute myeloid leukemia (AML),apoptosis dysregulation, 171, 172Bcl-2,

expression, 115, 116G3139 trials, 117–119

chemotherapy, 146chromosomal translocations, 129dendritic cell characteristics, 5, 6gemtuzumab ozogamicin therapy,

see Gemtuzumab ozogamicinHuM195 trials, 30, 31interleukin-2 therapy, see

Interleukin-2markers, 44P-glycoprotein reversal trials,

biricodar, 154cyclosporine, 150–152, 154gemtuzumab ozogamicin com-

bination therapy, 154, 155prospects, 156PSC833, 150–153, 155

quinine, 149, 150verapamil, 149, 150zosuquidar, 154

prognosis, 146radioimmunotherapy,

alpha-particle emitters, 72, 73beta-particle emitters, 65–71

relapse and prognosis, 93, 94Acute promyelocytic leukemia

(APL),all-trans retinoic acid therapy, see

All-trans retinoic acidarsenic trioxide trials and all-trans retinoic acid combina-tion therapy, 192–194, 200chemotherapy response, 206,207

chromosomal translocations, 128,244

HuM195 trials, 30prognosis, 148

Adeno-associated virus (AAV), genetherapy vectors, 232, 233

Adenovirus, gene therapy vectors,231, 232

Adult T-cell leukemia/lymphoma(ATL), radioimmunotherapy,73, 74

AG490, apoptosis induction in leuke-mia management, 177

Alemtuzumab,CD52 targeting, 35development, 35leukemia trials,

B-cell chronic lymphocyticleukemia, 36–38

262 Index

T-cell prolymphocytic leuke-mia, 36

ALL, see Acute lymphocytic leuke-mia

All-trans retinoic acid (ATRA),acute promyelocytic leukemiatreatment,

arsenic trioxide combinationtherapy, 192–194, 200, 216

chemotherapy combinationtherapy,

AraC role, 211, 212dosing and duration during

induction treatment, 210,211

maintenance treatment, 212nonrandomized studies, 208overview, 205, 206prognostic factors,

complete remission achieve-ment factors, 213

extramedullary relapse, 214minimal residual disease

monitoring, 214pretreatment factors, 213,

214randomized studies, 208, 210scheduling, 211

coagulopathy and thrombosis in-duction, 218

monotherapy response, 207, 208rationale, 128, 129relapse management, 214, 215side effects, 218syndrome,

clinical signs, 216, 217hyperleukocytosis pathophysi-

ology, 217incidence, 216, 217prevention, 217, 218

treatment, 217, 218synthetic retinoids, 215, 216TRAIL induction, 174, 175

AML, see Acute myeloid leukemiaAML1-ETO,

chromosomal translocation, 129tumorigenesis role, 131, 132

Angiogenesis, arsenic trioxide inhibi-tion, 192, 195

Antisense therapy,bispecific antisense oligonucle-

otides in apoptosis targeting,176

DNA modifications, 110mechanisms of action, 110, 111oligonucleotide features, 109, 110ribonuclease H conjugation, 110ribozyme utilization, 243, 244targets,

Bcl-2 and G3139 targeting,113–120

BCR-ABL, 111, 112, 243Myb, 112, 113Myc, 113p53, 113

APL, see Acute promyelocytic leuke-mia

Apoptosis,arsenic trioxide induction, 190,

191Bcl-2,

anti-apoptotic mechanisms,113, 114

cytochrome c release regula-tion, 168, 169, 173

caspases,regulation by inhibitor of

apoptosis proteins, 169role in apoptosis, 164

death receptor pathway,

Index 263

activation, 174, 175overview, 164, 165regulation, 167, 168

drug resistance relationship, 173, 174leukemia dysregulation,

acute lymphocytic leukemia,172

acute myeloid leukemia, 171,172

chronic lymphocytic leukemia,169, 170

chronic myelogenous leukemia,170, 171

large granular lymphocyte leu-kemia, 170

mitochondrial pathway,inhibition reversal in therapy,

175–179overview, 165regulation, 168, 169

morphological features, 163therapeutic induction,

AG490, 177all-trans retinoic acid induction

of TRAIL, 174, 175Bcl-2 antisense targeting, 175,

176bispecific antisense oligonucle-

otides, 176caspase activation, 173CI-1040, 177DR5 activation, 174flavopiridol, 177, 178inhibitor of apoptosis protein

inhibitors, 179PD98059, 177prospects, 179, 180protein kinase C inhibitors, 178STI571, 177TRAIL therapy, 174

Arsenic trioxide (ATO),acute promyelocytic leukemia

trials and all-trans retinoicacid combination therapy,192–194, 200, 216

chronic myelogenous leukemiatrials, 196

history of use, 189, 190mechanisms of action,

angiogenesis inhibition, 192, 195apoptosis induction, 190, 191mitochondrial injury, 191PML/RAR-α transcript product

relocalization, 191, 192miscellaneous hematologic malig-

nancy trials, 196, 197multiple myeloma trials, 195solid tumor trials, 197toxicity, 198, 199

ATL, see Adult T-cell leukemia/lymphoma

ATO, see Arsenic trioxideATRA, see All-trans retinoic acid

B

BC8, acute myeloid leukemiaradioimmunotherapy, 70, 71

B-cell chronic lymphocytic leukemia(B-CLL),

alemtuzumab trials, 36–38rituximab trials, 31–34

Bcl-2,anti-apoptotic mechanisms, 113,

114cytochrome c release regulation,

168, 169, 173expression in leukemias,

acute myeloid leukemia, 115,116

chromosomal translocations, 128

264 Index

chronic lymphocytic leukemia,114, 115

G3139,acute myeloid leukemia trials,

117–119chronic lymphocytic leukemia

trials, 117chronic myelogenous leukemia

trials, 119, 120combination therapy, 176non-Hodgkin’s lymphoma tri-

als, 116, 175pharmacology, 116

inhibition for apoptosis inductionin therapy,

AG490, 177antisense targeting, 175, 176bispecific antisense oligonucle-

otides, 176CI-1040, 177flavopiridol, 177, 178PD98059, 177protein kinase C inhibitors, 178STI571, 177

B-CLL, see B-cell chronic lympho-cytic leukemia

BCR-ABL,antisense targeting, 111, 112, 243chromosomal translocation, 130STI571 targeting in chronic myel-

ogenous leukemia,applications for other molecular

targets and cancers, 140clinical trials, 135, 136combination therapy, 138development, 133, 134dose selection, 136, 137preclinical studies, 134, 135prospects, 137rationale, 133

resistance, 137, 138specificity of receptor inhibi-

tion, 138–140tyrosine kinase inhibition, 128

tumorigenesis role, 131tyrosine kinase activation, 130

BCRP,drug transport, 148prognostic value, 149

Biricodar, P-glycoprotein reversaland clinical trials, 154

BMT, see Bone marrow transplanta-tion

Bone marrow transplantation (BMT),autologous transplantation, 95, 96graft vs host disease, 14, 15graft vs leukemia effect, see Graft

vs leukemia effectinterleukin-2 combination therapy

in acute leukemia, 98–100nonmyeloablative allogeneic

transplantation, 19–23rationale in cancer therapy, 13

Burkitt’s lymphoma, gene mutations,132

CCAMPATH-1H, see AlemtuzumabCaspases,

apoptosis role, 164arsenic trioxide activation, 190regulation by inhibitor of

apoptosis proteins, 169therapeutic activation, 173

CBF, see Core binding factorCD5 antibody, see T101CD20 antibody, see Rituximab,CD25, adult T-cell leukemia/lym-

phoma radioimmunotherapytargeting, 73, 74

CD33 antibody, see HuM195; p67

Index 265

CD45 antibody, see BC8CD52 antibody, see AlemtuzumabCD66, acute myeloid leukemia

radioimmunotherapy targeting, 71Chronic lymphocytic leukemia (CLL),

alemtuzumab, 36–38apoptosis dysregulation, 169, 170Bcl-2,

expression, 114, 115G3139 trials, 117

rituximab, 31–34Chronic myelogenous leukemia (CML),

apoptosis dysregulation, 170, 171arsenic trioxide trials, 196chromosomal translocation, see

BCR-ABLdendritic cell,

characteristics, 5, 6clinical trials, 6, 8prospects, 8–10

epidemiology, 133G3139 trials, 119, 120interferon-α therapy, 86–88phases, 133prognosis, 133STI571 therapy, see STI571,

CI-1040, apoptosis induction in leu-kemia management, 177

C-kit,activating mutations, 130, 131STI571 inhibition of receptor ty-

rosine kinase, 139CLL, see Chronic lymphocytic leu-

kemiaCML, see Chronic myelogenous

leukemiaCore binding factor (CBF),

components, 129therapeutic targeting, 129, 131,

132

Cyclin D1, chromosomal transloca-tion and overexpression, 128

Cyclosporine, P-glycoprotein rever-sal and clinical trials, 150–152,154

DDC, see Dendritic cellDendritic cell (DC),

ex vivo differentiation, 4leukemia therapy,

chronic myelogenous leukemiatrials, 6, 8

prospects, 8–10rationale, 3, 4

myeloid leukemia cell characteris-tics, 5, 6

DLI, see Donor leukocyte infusionDonor leukocyte infusion (DLI),

clinical trials in leukemia, 18, 19,95

gene therapy utilization, 244graft vs host disease, 18graft vs leukemia effect, 17–19

DR5, activation, 174

FFlavopiridol, apoptosis induction in

leukemia management, 177,178

Flt3, activating mutations, 130, 131

G

G3139acute myeloid leukemia trials,

117–119Bcl-2 targeting, 116chronic lymphocytic leukemia

trials, 117chronic myelogenous leukemia

trials, 119, 120

266 Index

combination therapy, 176non-Hodgkin’s lymphomatrials, 116, 175pharmacology, 116

Gemtuzumab ozogamicin,acute myeloid leukemia treatment

rationale and response, 44,53, 54

clinical trials, 48–50, 54–56design and structure, 45–47immune response, 52P-glycoprotein reversal combina-tion therapy, 154, 155pharmacokinetics, 50, 52toxicity, 52, 53

Gene therapy, see also Antisensetherapy,

clinical trials, 225definition, 225drug resistance gene transfer to

hematopoietic precursors,242

effector cell genetic modification,cytokine expression, 240T-cell receptor genes, 240, 241

leukemia cell genetic modification,antigen expression, 239, 240costimulatory molecule expres-

sion, 239cytokine gene expression, 238rationale, 237

malignant transformation interfer-ence, 242–244

marking studies, 234–237prospects, 245suicide genes and negative selec-

tion, 244, 245vectors,

adeno-associated virus, 232,233

adenovirus, 231, 232

criteria for successful virusvectors, 227

herpes simplex virus, 233, 234lentivirus, 230, 231Moloney murine leukemia

virus, 227–230physical gene transfer, 226,

234prospects, 234

GM-CSF, see Granulocyte-macroph-age colony-stimulating factor

Graft vs leukemia (GVL) effect,definition, 13, 14donor leukocyte infusion, 17–19graft vs host disease,

gene therapy utilization in pre-vention, 244

relationship, 14, 15, 23lymphocyte-activated killer cells,

94, 95T-cell depletion effects in bone

marrow transplantation,15–17

Granulocyte-macrophage colony-stimulating factor (GM-CSF),leukemia cell engineering forexpression, 238

GVL effect, see Graft vs leukemiaeffect

HHairy cell leukemia,

interferon-α therapy, 88, 89rituximab trials, 34, 35

Herpes simplex virus (HSV), genetherapy vectors, 233, 234

HSV, see Herpes simplex virusHu1D10, leukemia therapy, 38HuM195

CD33 targeting, 30leukemia trials,

Index 267

acute myeloid leukemia, 30, 31acute promyelocytic leukemia,

30radioimmunotherapy trials of

acute myeloid leukemia,65, 68, 69, 72, 73

IIAPs, see Inhibitor of apoptosis pro-

teinsIL-2, see Interleukin-2Inhibitor of apoptosis proteins

(IAPs),caspase regulation, 169inhibitors in apoptosis induction,

179Interferons,

antiviral activity, 83, 84biosynthesis, 81, 82classification, 81interferon-α therapy,

antitumor mechanisms, 84, 85chronic myelogenous leukemia

trials, 86–88hairy cell leukemia trials, 88, 89pharmacokinetics, 86toxicity, 86

receptor and signaling, 82, 83subtypes, 81, 82

Interleukin-2 (IL-2),acute leukemia trials,

outcomes, 100, 101single-agent trials, 98stem cell transplantation combi-

nation, 98–100function, 96gene, 96immune marker effects, 97leukemia cell engineering for

expression, 238

lymphocyte-activated killer celleffects, 96

solid tumor management, 97, 98toxicity, 96, 97

LLarge granular lymphocyte leukemia,

apoptosis dysregulation, 170Lentivirus, gene therapy vectors,

230, 231LRP, see Lung-resistance proteinLung-resistance protein (LRP),

assays, 148prognostic value, 147

MMM, see Multiple myelomaMoloney murine leukemia virus,

gene therapy vectors, 227–230Monoclonal antibody therapy,

alemtuzumab, 35–38gemtuzumab ozogamicin,

acute myeloid leukemia treat-ment rationale andresponse, 44, 53, 54

clinical trials, 48–50, 54–56design and structure, 45–47immune response, 52pharmacokinetics, 50, 52toxicity, 52, 53

Hu1D10, 38HuM195, 30, 31human antimouse antibody

response development, 29,30, 54

prospects for unconjugated anti-body therapy, 39

radiolabeled antibodies,acute myeloid leukemia

radioimmunotherapy,alpha-particle emitters, 72, 73

268 Index

beta-particle emitters, 65–71adult T-cell leukemia/lym-

phomaradioimmunotherapy, 73,74

antigenic targets, 60, 61conjugation methods, 63dosimetry, 64, 65mechanism of action, 60pharmacokinetics, 63, 64prospects, 74radioisotope selection, 61–63

rationale, 29, 43, 59, 60rituximab, 31–35T101, 38, 39

MRP, see Multidrug-resistance asso-ciated protein

Multidrug-resistance associated pro-tein (MRP),

ABC transporter, 147genetic engineering of hematopoi-

etic precursors, 242prognostic value, 147, 149

Multiple myeloma (MM), arsenictrioxide trials, 195

Myb, antisense targeting, 112, 113Myc,

antisense targeting, 113chromosomal translocation and

overexpression, 128

NNHL, see Non-Hodgkin’s lymphomaNon-Hodgkin’s lymphoma (NHL),

G3139 trials, 116, 175

Pp53,

antisense targeting, 113gene therapy, 243

p67, acute myeloid leukemiaradioimmunotherapy, 70

PD98059, apoptosis induction inleukemia management, 177

PDGF, see Platelet-derived growthfactor

P-glycoprotein (PgP),ABC transporter, 146, 147assays, 147prognostic value, 145, 146, 148, 149resistance, 155reversal trials in acute myeloid

leukemia,biricodar, 154cyclosporine, 150–152, 154gemtuzumab ozogamicin com-

bination therapy, 154, 155prospects, 156PSC833, 150–153, 155quinine, 149, 150verapamil, 149, 150zosuquidar, 154

therapeutic targeting rationale,145, 146

PgP, see P-glycoproteinPKC, see Protein kinase CPlatelet-derived growth factor (PDGF),

STI571 inhibition of receptortyrosine kinase, 138–140

PML/RARα, transcript productrelocalization with arsenic tri-oxide, 191, 192

Protein kinase C (14K), inhibitors inapoptosis induction, 178

PSC833, P-glycoprotein reversal andclinical trials, 150–153, 155

QQuinine, P-glycoprotein reversal and

clinical trials, 149, 150

Index 269

RRadiolabeled antibodies, see Mono-

clonal antibody therapyRituximab,

CD20 targeting, 31leukemia trials,

B-cell chronic lymphocyticleukemia, 31–34

hairy cell leukemia, 34, 35

SSTI571,

apoptosis induction in leukemiamanagement, 177

applications for other moleculartargets and cancers, 140

chronic myelogenous leukemiamanagement,

clinical trials, 135, 136combination therapy, 138dose selection, 136, 137preclinical studies, 134, 135prospects, 137rationale, 133resistance, 137, 138

development, 133, 134specificity of receptor inhibition,

138–140tyrosine kinase inhibition, 128

Suicide gene, negative selection, 244,245

TT101, leukemia therapy, 38, 39T-cell,

depletion effects in bone marrowtransplantation, 15–17

genetic modification for leukemiatherapy, 240, 241

T-cell prolymphocytic leukemia,alemtuzumab trials, 36

TRAIL,antileukemia therapy, 174death receptor pathway,

activation, 174, 175overview, 164, 165regulation, 167, 168

VVerapamil, P–glycoprotein reversal

and clinical trials, 149, 150

ZZosuquidar, P-glycoprotein reversal

and clinical trials, 154

270 Index

biologic therapy of leukemia - m. kalaycio (humana, 2003) ww

Documents