New Agents for the Treatment of Acute Lymphoblastic Leukemia
Vaskar Saha Pamela Kearns ●
Editors
New Agents for the Treatment of Acute Lymphoblastic Leukemia
Editors Vaskar Saha Paediatric and Adolescent Oncology School of Cancer and Enabling Sciences Manchester Academic Health Science Centre The University of Manchester Manchester M20 4BX
[email protected]
Pamela Kearns School of Cancer Sciences University of Birmingham Edgbaston, Birmingham B15 2TT
[email protected]
ISBN 978-1-4419-8458-6 e-ISBN 978-1-4419-8459-3 DOI 10.1007/978-1-4419-8459-3 Springer New York Dordrecht Heidelberg London Library of Congress Control Number: 2011925850 © Springer Science+Business Media, LLC 2011 All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights. While the advice and information in this book are believed to be true and accurate at the date of going to press, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made. The publisher makes no warranty, express or implied, with respect to the material contained herein. Printed on acid-free paper Springer is part of Springer Science+Business Media (www.springer.com)
Brief Overview
Over 80% of children with acute lymphoblastic leukemia (ALL) are cured by modern chemotherapeutic regimens. These are the results of carefully conducted randomised clinical trials. Mainly, current protocols have progressively intensified therapy for all children and now use risk stratification to intensify treatment for high-risk subgroups. As a result, therapy-related toxicity now outweighs disease recurrence as a determinant of outcome. The prognosis for those who relapse or do not respond to therapy remains poor. Further intensification is unlikely to benefit those who fail therapy and may increase toxicity for those who do. Perhaps, the most intense of therapies is allogeneic bone marrow transplantation. However, disease recurrence is the most common cause of transplantation failure in these patients. The drugs that we use for ALL are now over 30 years old. To improve outcome further and to decrease toxicity, we need new drugs. These are exciting times. At long last, new agents that can potentially be used in the ALL armamentarium are increasing rapidly. These include both conventional cytotoxics and targeted therapy. However, there are a number of problems that need to be solved. How does one evaluate the effect of a new drug in a disease where over 80% of patients are cured from what is otherwise a fatal disease? What are acceptable surrogate markers of response to treatment given that therapeutic failure can occur over a number of years? With shrinking number of patients suitable for phase I and II trials, what new models of trial design are required to obtain answers in the quickest possible time? What cellular and animal models will we accept as most predictive for the clinical effect of the drug in question? These are some of the questions that need to be answered before we can incorporate new agents into frontline therapy for patients with ALL.
v
Acknowledgements
This work was funded in part by a programme grant from Cancer Research UK to Vaskar Saha. The editors thank Ms Charlotte O’Horo for editorial assistance.
vii
Contents
1 The Need for New Agents........................................................................ Tim Eden 2 Identifying Targets for New Therapies in Children with Acute Lymphoblastic Leukemia..................................................... Shekhar Krishnan, Ashish Masurekar, and Vaskar Saha
1
25
3 Preclinical Evaluation.............................................................................. Barbara Szymanska, Hernan Carol, and Richard B. Lock
39
4 Design of Early-Phase Trials................................................................... James A. Whitlock and Terzah M. Horton
61
5 Strategies for Trial Design and Analyses............................................... Maria Grazia Valsecchi and Paola De Lorenzo
83
6 An Overview on Animal Models of ALL............................................... 105 Michael A. Batey and Josef H. Vormoor 7 Targeting Bcl-2 Family Proteins in Childhood Leukemia.................... 117 Guy Makin and Caroline Dive 8 Targeting Leukemia Stem Cells and Stem Cell Pathways in ALL...................................................................................... 143 Clare Pridans and Brian J.P. Huntly 9 Nucleoside Analogues.............................................................................. 167 Pamela Kearns and Vaskar Saha 10 FLT3 Inhibitors as Therapeutic Agents in MLL Rearranged Acute Lymphoblastic Leukemia........................................ 189 Ronald W. Stam and Rob Pieters
ix
x
Contents
11 The Role of Tyrosine Kinase Inhibitors in the Treatment of ALL......................................................................... 203 S. Wilson 12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia........................................................................ 221 Arend von Stackelberg 13 Therapeutic Utility of Proteasome Inhibitors for Acute Leukemia................................................................................. 273 Joya Chandra and Claudia P. Miller 14 Targeting Epigenetic Pathways in ALL................................................. 299 Pamela Kearns 15 Incorporating New Therapies into Frontline Protocols....................... 311 Paul S. Gaynon and Theresa M. Harned Index.................................................................................................................. 329
Contributors
Michael A. Batey Northern Institute for Cancer Research, Newcastle University, Paul O’ Gorman Building, Medical Science, Framlington Place, Newcastle upon Tyne NE2 4HH, UK Hernan Carol Leukemia Biology Program, Children’s Cancer Institute Australia for Medical Research, Lawy Cancer Research Center, UNSW PO Box 81, Randwick, NSW 2031, Australia; University of New South Wales, Sydney, NSW 2031, Australia Joya Chandra Department of Pediatrics Research, Children’s Cancer Hospital at M.D. Anderson, University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030, USA
[email protected] Caroline Dive Clinical and Experimental Pharmacology, Paterson Institute for Cancer Research, University of Manchester, Manchester, UK; School of Cancer and Enabling Sciences, University of Manchester, Manchester, UK
[email protected] Tim Eden 5 South Gillsland Road, Edinburgh, EH10 5DE
[email protected] Paul S. Gaynon Therapeutic Advances in Childhood Leukemia Consortium, Institute for Pediatric Clinical Research, Children’s Hospital Los Angeles, University of Southern California, Los Angeles, CA, USA
[email protected]
xi
xii
Contributors
Theresa M. Harned Therapeutic Advances in Childhood Leukemia Consortium, Institute for Pediatric Clinical Research, Children’s Hospital Los Angeles, University of Southern California, Los Angeles, CA, USA Terzah M. Horton Department of Pediatrics, Division of Hematology/Oncology, Baylor College of Medicine, 1102 Bates, Suite 750, USA
[email protected] Brian J.P. Huntly Department of Haematology, University of Cambridge, Cambridge Institute for Medical Research, Hill Road, Cambridge CB2 OXY, UK
[email protected] Pamela Kearns School of Cancer Sciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK
[email protected] Shekhar Krishnan Cancer Research UK Children’s Cancer Group, School of Cancer and Enabling Sciences, Manchester Academic Health Sciences Centre, The Unversity of Manchester, Central Manchester and Manchester Children’s NHS Trust, Christie Hospital NHS Foundation Trust, Manchester M20 4BX, UK
[email protected] Richard B. Lock Leukaemia Biology Program, Children’s Cancer Institute Australia for Medical Research, Lowy Cancer Research Centre, UNSW PO Box 81, Randwick, NSW 2031, Australia; University of New South Wales, Sydney, NSW 2031, Australia
[email protected] Paola De Lorenzo Centre of Biostatistics for Clinical Epidemiology, Department of Clinical Medicine and Prevention, University of Milano-Bicocca, Via Cadore, 48, 20052 Monza (MI), Italy Guy Makin Clinical and Experimental Pharmacology, Paterson Institute for Cancer Research, University of Manchester, Manchester, UK; School of Cancer and Enabling Sciences, University of Manchester, Manchester, UK
[email protected]
Contributors
Ashish Masurekar Cancer Research UK Children’s Cancer Group, School of Cancer and Enabling Sciences, Manchester Academic Health Sciences Centre, The Unversity of Manchester, Central Manchester and Manchester Children’s NHS Trust, Christie Hospital NHS Foundation Trust, Manchester M20 4BX, UK
[email protected] Claudia P. Miller Department of Developmental Neurobiology, St. Jude Children’s Research Hospital, 262 Danny Thomas Place, 38105, Memphis, TN, USA
[email protected] Rob Pieters Erasmus MC – Sophia Children’s Hospital Pediatric Oncology/Hematology, Dr. Molewaterplein 60, 3000 CB, Rotterdam, the Netherlands
[email protected] Clare Pridans Department of Haematology, University of Cambridge, Cambridge Institute for Medical Research, Hill Road, Cambridge CB2 OXY, UK Vaskar Saha Cancer Research UK Children’s Cancer Group, School of Cancer and Enabling Sciences, Manchester Academic Health Sciences Centre, The Unversity of Manchester, Central Manchester and Manchester Children’s NHS Trust, Christie Hospital NHS Foundation Trust, Manchester M20 4BX, UK
[email protected] Arend von Stackelberg Pädiatrische Onkologie/Hämatologie, Charité, OHC, Augustenburger Platz 13353, Berlin, Germany
[email protected] Ronald W. Stam Erasmus MC – Sophia Children’s Hospital Pediatric Oncology/Hematology, Dr. Molewaterplein 60, 3000 CB, Rotterdam, The Netherlands
[email protected] Barbara Szymanska Leukemia Biology Program, Children’s Cancer Institute Australia for Medical Research, Lawy Cancer Research Center, UNSW PO Box 81, Randwick, NSW 2031, Australia; University of New South Wales, Sydney, NSW 2031, Australia
[email protected]
xiii
xiv
Maria Grazia Valsecchi Centre of Biostatistics for Clinical Epidemiology, Department of Clinical Medicine and Prevention, University of Milano-Bicocca, Via Cadore, 48, 20052 Monza (MI), Italy
[email protected] Josef H. Vormoor Northern Institute for Cancer Research, Newcastle University, Sir James Spence Institute, 4th floor, Royal Victoria Infirmary, Newcastle Upon Tyne, NE1 4LP, UK
[email protected] James A. Whitlock Division of Haematology/Oncology, The Hospital for Sick Children, 555 University Avenue, Toronto, ON M5G 1X8, Canada
[email protected] S. Wilson School of Cancer Sciences, University of Birmingham, Birmingham, West Midlands, UK
[email protected]
Contributors
Glossary
ACTH Adrenocorticotrophic hormone ADA Adenosine deaminase ADCC Antibody-dependent cellular cytotoxicity ADCP Antibody-dependent cell-mediated phagocytosis ALL Acute lymphoblastic leukemia Allo-SCT Allogeneic stem cell transplantation AML Acute myeloid leukemia ANC Absolute neutrophil count APCs Antigen-presenting cells APML Acute promyelocytic leukemia AS Asparagine synthetase ATM Ataxia telangiectasia mutated ATP Adenosine triphosphate ATRA All-trans retinoic acid AUC Area under the curve BCLL B-lymphocytic leukemia BCP B-cell precursor BFM Berlin-Frankfurt-Münster BH BCl-2 homology BITE Bispecific T-cell engaging antibodies C Constant CALLA Common acute lymphoblastic leukemia antigen CCG Children’s Cancer Group CDC Complement-dependent cytotoxicity CDDP Cisdiamminedichloridoplatinum CDR Complementarity-determining regions CHOP Cyclophosphamide, doxorubicin, vincristine, prednisone CI Confidence intervals CI50 Inhibitory concentration for 50% effect CLL Chronic lymphocytic leukemia CML Chronic myeloid leukemia CNS Central nervous system xv
xvi
COG CR CRM CRp CSF CTLs
Glossary
Children’s Oncology Group Complete remission/response Continual reassessment method Complete response with insufficient platelet recovery Cerebrospinal fluid Cytotoxic T lymphocytes
DCs Dendritic cells dCK deoxyCytidine kinase dGuo deoxyGuanosine dHPLC Denaturing high performance liquid chromatography DLBCL Diffuse large B-cell lymphoma DLT Dose-limiting toxicities DNR Daunorubicin EC50 Effective concentration for 50% effect EFS Event-free survival time EGFP Enhanced green fluorescent protein EMEA European Medicine’s Agency EMSA Electrophoretic mobility shift assay ER Endoplasmic reticular ERK Extracellular signal-related kinase ESFT Ewing sarcoma family tumours FDA Food and Drug Administration FL FLT3 ligand FLK2 Fetal liver kinase 2 FLT3 Fms-like tyrosine kinase 3 FMS Macrophage colony-stimulating factor receptor GCP Good clinical practice GEP Gene expression profiling GFP Green fluorescent protein GMTZ Gemtuzumab ozogamicin GR Glucocorticoid receptors GSK3a Glycogen synthase kinase 3 isoforms a GSK3b Glycogen synthase kinase 3 isoforms b GvHD Graft-versus-host-disease HAMA Human anti-mouse antibodies HATs Histone acetylases HDAC Histone deacetylase HDACi Histone deacetylase inhibitors HSC Haematopoietic stem cell HSCT Haematopoietic stem cell transplant ICH ICN
International Conference on Harmonization Intracellular NOTCH
Glossary
xvii
IFNa Interferon alpha IGF-1 Insulin-like growth factor 1 IL2rg IL-2 receptor common g chain INR International normalised ratio ITAM Immunoreceptor tyrosine-based activation motif ITD Internal tandem duplication ITIMs Immunoreceptor tyrosine-based inhibitory motifs ITT Intention to treat IV Intravenous JAK JM JNK KDR
Janus kinase Juxtamembrane Jun amino terminal kinase Kinase insertion domain receptor
Ki Inhibitory concentration KIR Killer Ig-like receptor KIT Steel factor receptor LAK Lymphokine-activated killer cells LASNASE l-Asparaginase LPD Lymphoproliferative diseases LSC Leukemia stem cell MAC Membrane attack complex MAPK Mitogen-activated protein kinase MCC Maleimidomethyl-cyclohexane-carboxylate MCR Maintained complete response MDS Myelodysplastic syndrome MEF Mouse embryonic fibroblast MEK MAPK kinase MHC Major histocompatibility complex MIMP Mitochondrial inner membrane potential miRNA microRNA MLL Mixed lineage leukemia MMRC Multiple myeloma research consortium MOMP Mitochondrial outer membrane potential MRC Medical Research Council MRD Minimal residual disease MTD Maximum tolerated dose MTOR Mammalian target of rapamycin MTT Methyl thiazolyl diphenyltetrazolium bromide NCI NHL NK NLS
National Cancer Institute Non-Hodgkin lymphoma Natural killer Nuclear localizing sequences
xviii
Glossary
NNT NOD NSG
Number needed to treat Non-obese diabetic NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ immunodeficient mice
OMM
Outer mitochondrial membrane
PCR Polymerase chain reaction PD Progressive disease PDGFR Platelet-derived growth factor receptor PGP P-glycoprotein PMAS Prospective meta-analysis strategy PNP Purine Nucleoside Phosphorylase PP Per protocol PPTP Paediatric Preclinical Testing Program PR Partial remission/response PTL Parthenolide PTT Partial thrombo-plastin time ROS RPIID RTK
Reactive oxygen species Recommended phase II dose Receptor tyrosine kinase
SC Stem cell SCID Severe-combined immunodeficient SCT Stem cell transplantation SD Standard deviation SD Stable disease SiRNA Small interfering RNA Smac Second mitochondrial derived activator of caspases STK1 Stem cell tyrosine kinase 1 TK Tyrosine kinases TKD Tyrosine kinase domain TM Transmembrane TNF Tumour necrosis factor TPMT Thiopurine methyltransferase TRAIL TNF related apoptosis inducing ligand TRK Tropomyosin-related kinase UPR
Unfolded protein response
V Variable VEGFR Vascular endothelial growth factor receptor VOD Veno-occlusive disease VPA Valproic acid WBC 6MP
White blood cell 6-Mercaptopurine
Chapter 1
The Need for New Agents Tim Eden
1.1 Introduction Since the 1960s, we have progressed from little expectation of survival for children with acute lymphoblastic leukemia (ALL) to 80% 5-year event-free survival and probable long-term cure. Therapy is long and toxic in terms of the physical, emotional and psychological impact on our patients and their families. We can try to alleviate wherever possible those side effects but must not do so at the expense of significantly worsened survival. The balance to be achieved between efficacy and toxicity must be quantified and assessed. How that has been achieved is one of the medical success stories of the twentieth century. It is useful to explore the pathway along which the early pioneers made their progress towards finding potential curative therapy, what mistakes we have made along the way and how we can improve therapy further, to achieve 100% cure rates. Can we improve on the way in which we use our currently available cytotoxics, or do we need totally new approaches for some or all ALL patients?
1.2 Historical Background Following the first description in 1827 by Velpeau [1] of a patient with undoubtedly chronic leukemia, it was not until 1845 that Virchow [2] and Bennett [3] and Craigie [4] independently identified “white blood” as a distinct entity. Virchow [5] subsequently first used the term “leukemia” but it was only in 1857 that Friedreich [6] described what was clearly acute as against chronic leukemia [1], whilst 21 years later Neumann [7] reported on the first case of acute myeloid leukemia. Diagnosis was initially by light microscopy. With the development of distinguishing stains,
T. Eden (*) 5 South Gillsland Road, Edinburgh, EH10 5DE e-mail:
[email protected]
V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3_1, © Springer Science+Business Media, LLC 2011
1
2
T. Eden
more precise cytological classification was established. The application of cytogenetics and immunophenotyping has increasingly enabled more precise and definitive diagnosis of subtypes [8, 9]. ALL is now recognised to be a heterogeneous group of different entities with different biological characteristics and resistance patterns [9]. Advances in treatment lagged far behind the ability to make the diagnosis of acute leukemia. Between 1865, when arsenicals were first utilised [10], and the 1940s, when work commenced on folate and its antagonists, no progress was made for acute leukemia, although splenic radiation and subsequent use of radionuclides were utilised to produce at least some palliation in chronic myelogenous leukemia and polycythemia vera [11, 12]. In 1947, Farber et al. [13], realising that folate could possibly accelerate acute leukemia in children, utilised an antagonist aminopterin synthesised by Dr Subba Row of Lederle [14]. This induced remissions, some of which lasted for a few months. This heralded in a concentrated period of research into potential antileukemic agents. The anti-leukemic therapeutic value of adrenal corticosteroids was reported initially with the newly identified adrenocorticotrophic hormone (ACTH) in 1949 [15] but was rapidly replaced by the synthetic prednisolone. Methotrexate (4-amino-N10-methyl-folate analogue) replaced aminopterin as a more useful and tolerable agent [16] subsequently available for oral, systemic (IV and IM) plus intrathecal injection. Research into purine metabolism led to the production of a purine analogue, 6-mercaptopurine [17], which also induced remissions in acute leukemia especially if combined sequentially with, or alongside, prednisolone and methotrexate [16]. However, remissions were short-lived, and a generally pessimistic attitude to the potential for cure persisted. In 1959, the research yielded a new alkylating agent, cyclophosphamide, an analogue of nitrogen mustard. The latter had produced such a degree of myelosuppression when used to “gas” soldiers in the 1914–1918 war that its potential as an anti-leukemic agent had been queried and tested [18], but it carried with it excessive toxicity. Efficacy in ALL therapy was reported for cyclophosphamide by Fernbach et al. [19] with tolerable toxicity. In the same year, an alkaloid vincristine, extracted from the periwinkle plant, was shown to induce remission in ALL. Its mode of action was recognised to be quite different to the antimetabolites and alkylators. Subsequently, we know that its cytotoxicity results from an interaction with and disruption of microtubules especially those of the mitotic spindle apparatus [20]. The initial use of single-agent therapy that induced relatively short-term remissions was followed by an era in which researchers started to use combinations of drugs, e.g. prednisone, methotrexate and 6-mercaptopurine. A number of reports in the early 1960s suggested 3–5% 5-year survival rates with such approaches [21–23]. Krivit et al. [23] identified the need for ongoing treatment post-remission induction. Frei et al. reported on National Cancer Institute [USA] studies using combination therapy [24], but still long-term cure appeared elusive [24]. Many paediatricians felt that children could not be cured and that only palliation was justified. Luckily, some “pioneers” persisted. A group of investigators at the St Jude Children’s Research Hospital in Memphis led by Don Pinkel identified the major reasons for failure: primary and secondary
1 The Need for New Agents
3
resistance, “isolated” extramedullary disease, especially within the central nervous system (CNS), which subsequently led to systemic failure and therapy-related toxicities that put restraints on investigators and induced “breaks” in therapy. Finally and possibly most significantly, the pessimism that shrouded the treatment of acute leukemia – “if you can’t cure why put patients through the ordeal of such toxic therapy” – was not helpful in finding solutions through collaboration of researchers. In such an environment, dedication and commitment was required by anyone continuing to try to find cures. Studies at St Jude started soon after the hospital opened in 1962. Critically, the group identified the need for four treatment phases: induction of remission, intensification or consolidation, CNS-directed therapy and prolonged continuation therapy. They designated their approach as “total therapy” [25]. Early on, they demonstrated the risk of infection especially the emergence of significant immunosuppression-induced pneumocystis carinii pneumonia [25, 26], but reduction of chemotherapy dosage to lessen immunosuppression led to more relapses [27] and was counterproductive. The second and most spectacular breakthrough was the use of effective CNSdirected therapy based on the premise that a majority of children with ALL (50–80% at least) had microscopic extramedullary spread of their ALL and a small minority (3–5%) had overt CNS disease at diagnosis. The application of CNS irradiation early on in treatment was tested in a randomised trial, and 5-year survival rose rapidly from 20% [26] to 50% [28]. This heralded in a new era, and optimism began to replace pessimism. From little expectation of cure in the 1960s, 75–80% overall cure rates are now reported, at least from resource-rich countries [29], so we need to look at what are the key elements and essential drugs. It is important to note as well that significant deviation from the “total therapy” strategy has almost uniformly been associated with decreased survival [30] and reversion to it, subsequent improvement [31]. What appears to be needed in ALL is an almost continuous sustained treatment.
1.3 Induction Therapy: Can We Improve Efficacy? 1.3.1 Basic Template All of the essential agents currently used in ALL therapy were in place by 1970, in particular the standard induction drugs: vincristine, prednisone, l-asparaginase and the anthracyclines daunorubicin or doxorubicin. The template for successful induction (remission rates of 94–96%) includes 4–6 weeks of oral steroid therapy, weekly vincristine injections intravenously (four to six doses), l-asparaginase therapy (dose and schedule, product dependent) and intrathecal methotrexate or triple intrathecal therapy. The use of an anthracycline has been tested but is now normally restricted to those with a higher risk of relapse.
4
T. Eden
1.3.2 Steroids Prednisone (or prednisolone) has been the steroid of choice since the 1960s, sometimes given as a 7-day prephase to assess tumour response, which is a moderately good predictor of primary resistance and a potential modulator of cell lysis [32]. Following the report by Balis et al. [33] regarding greater CNS penetrance by dexamethasone, the Dutch Childhood Leukemia Study Group demonstrated a significant improvement in survival in their ALL VI trial using dexamethasone compared with the previous most similar ALL study [34] that used prednisolone. Two large randomised trials have demonstrated significant benefits for dexamethasone. Both trials have shown significant reduction in CNS relapse rates and overall improvement in event-free survival [35, 36] for standard risk patients treated with dexamethasone during induction and in continuation therapy, compared with the use of prednisone [35] or prednisolone [36]. Mitchell et al. [36] reported benefit across all risk groups. In a small randomised trial, Igarashi et al. [37] reported no benefit for dexamethasone when their comparative prednisolone dose was 60 mg/ m2/day compared with 40 mg/m2/day used in the other two trials. The dexamethasone dose used in the US, UK and Dutch studies were comparable at 6–6.5 mg/m2/day. Thus, it remains somewhat unclear whether the benefit of type of steroid is merely a matter of dosage or whether the reported lower clearance, larger volume of distribution, longer half-life and greater CNS penetrance of dexamethasone do all truly confer major benefits [33, 38]. Two smaller non-randomised studies raised anxiety about short-term induction morbidity and mortality regarding the use of dexamethasone especially in combination with anthracyclines during induction. However, no significant difference in morbidity or mortality was seen between the randomised arms of either of the UK or US trials [35, 36]. Corticosteroids bind intracellularly to glucocorticoid receptors (GR) and can induce apoptosis [39] through a series of potential pathways and interactions not yet fully defined but almost certainly involve differential regulation of BCL-2 gene family members [40]. More potent cytotoxicity has been reported in in vitro cell line studies with dexamethasone than in prednisolone. Prolonged and continuous exposures (greater than 48–60 h) are required for apoptosis to occur [41]. Reported differences in lineage sensitivity, with B cell lines (especially precursor B cells) being more sensitive than T cells to steroids may indeed relate to variation of receptor numbers in the different lineages [42]. Reduction of GR numbers in mutated cell lines is associated with steroid resistance. Using an MTT assay, Kaspers et al. [43] showed a median 16-fold greater anti-leukemic activity with dexamethasone compared with prednisolone (with a considerably lower median half-maximal lethal concentration 50). However, for studies incorporating a stromal support for the leukemic blasts, in vitro, the difference between the two steroids was reduced to a sixfold greater cytotoxicity with dexamethasone [42]. Overall, there does appear to be a degree of correlation between low glucocorticoid receptor levels and high risk features, as defined by infancy, older age (>10 years) and overt CNS disease. Most patients with precursor B cell ALL have higher receptor levels than those with T and mature B cell ALL [44].
1 The Need for New Agents
5
Gene expression profiling has facilitated identification of unique leukemia associated markers, which can be monitored for therapeutic response, but it has also identified differentially expressed genes in drug-sensitive and resistant ALL. In steroid-resistant ALL, identified very early in those given a 7-day prephase of steroids, for example, there appears to be an over-expression of the anti-apoptotic MCL1 gene. Wei et al. [45] using a database of drug-associated gene expression profiles identified that the mTOR inhibitor rapamycin profile matched the signature of steroid sensitivity. This agent could sensitise blasts to steroid-induced apoptosis through modulation of MCL1 within primarily resistant cell lines and in vivo transgenic mice. In addition, several transcription genes were under-expressed. The potential to modulate glycolytic pathways and increase steroid responsiveness has emerged from these studies. Most of the acute toxicity of steroids, hyperglycaemia, behavioural change, myopathy, weight gain and immunosuppression do appear to recede with time, off treatment. A worrying feature of modern-era higher-dose steroid usage has been the development of osteonecrosis (avascular necrosis). For those under 10 years of age, the risk appears low at about 1% but increases to greater than 15% in those over 16 years of age in the UK and US steroid trials [35, 36]. A recent review of the latest UK data confirms, in ALL 2003 trial, rates of 1, 13 and 16%, respectively, for those aged <10 years, 10–15 years and >16 years with no sex difference (Vora A, personal communication 2011). In ALL 97/99, there was no overall difference in incidence of avascular necrosis by type of steroid used. This risk has led to the suggestion of a need to either cap steroid dosage in teenagers and young adults or adopt an intermittent strategy, for example, of alternative weeks on and off steroids during induction. No level-one evidence exists to support either of these strategies at present, in terms of reduced toxicity or equi-efficacy. Steroids are such a key component of induction and intensification that there is a great need to ensure optimal dosing and scheduling with early recognition of any steroid resistance for which modulation may be clinically possible, in the future. The long-term sequelae of steroids may be less if adopting the strategy pioneered by the BFM Group to remove all steroid and vincristine pulses from maintenance therapy where induction, consolidation and intensification therapy is so intensive in the first 6 months of treatment [46].
1.3.3 Vincristine This drug is the mainstay of many cancer treatments, not least ALL. Single-drug studies using a dose of 2 mg/m2/week (or 0.06 mg/kg/week), given weekly to relapsing patients in the 1960s induced remissions in 50–60% of patients [47]. Following a bolus intravenous injection, there is considerable inter and intrapatient variation of clearance, volume of distribution and elimination half-life. There is rapid cellular uptake (half life of approximately 5–8 min) and tissue binding, but a long terminal elimination phase of about 14 h so that low levels persist in the plasma. Cerebrospinal fluid (CSF) levels in adults are 20–30-fold lower than
6
T. Eden
c oncurrent plasma levels. There appears to be poor correlation between vincristine exposure and the troublesome side effect of neuropathy [48]. However, in 2008, Lönnerholm et al. [49] reported that in a series of 86 ALL patients, there was a trend for more relapses in those with higher median total plasma clearance (on day 1 of treatment) and lower area under the curve (AUC) for plasma concentration. This was only significant in the standard risk group of their series where the relative risk of relapse was significantly increased for patients with clearance values above the median (RR 5.2, p = 0.036) and AUC values below the median (RR 5.8, p = 0.025). They believed that vincristine was more likely to be important in standard risk patients who may receive less further intensification, and so getting the dosing right was more crucial. Resistance to vincristine has historically been related to decreased retention and cellular accumulation, related to the level of expression of P-glycoprotein and, consequently, the extent of cellular efflux of the drug. From Lönnerholm et al.’s work [49], it is obvious that other genes or polymorphic variants play a part in the interpatient variation of handling intracellular levels of vincristine. Gene expression profiling has been used to screen leukemia cell lines and more importantly, individual patient-sourced blast cells to identify gene expression patterns that define subsets of ALL and those related to in vitro drug sensitivity/resistance and consequentially potential outcome of therapy [50, 51]. Holleman et al. [51] specifically looked at in vitro sensitivity to steroids, vincristine, l-asparaginase and daunomycin. There appeared to be a little overlap between the significant genes for each drug’s sensitivity/resistance profile. This clearly emphasises the value of multiagent chemotherapy and also that there is no single solution to resistance. The same team of international collaborators, in a further study, identified a subset of patients with a poor outcome who had a phenotype that was discordant for sensitivity to asparaginase and vincristine [118]. In both the studies, the genes identified had, by and large, not been identified before as resistance genes but were predictive for outcome. Cario et al. [52] extended our understanding by linking a set of 54 genes expressed in diagnostic blasts, which predicted for persistent minimal residual disease post-induction and which appeared to be related to proliferation and apoptotic pathway impairment. Increasingly, such studies are also identifying potential novel therapeutic targets. For example, the addition into therapy of inhibitors of enzymes encoded by overexpressed resistance-associated genes within blasts. Of considerable interest is where germ line polymorphisms might influence gene expression within ALL cells, particularly where the polymorphism lies in gene regulator regions [53].
1.3.4 l-Asparaginase In 1922, high levels of the enzyme l-asparaginase were identified in guinea pig serum, but it was not until 1953 that it was shown that this serum destroyed lymphoma cell lines. It took another 8 years to identify asparaginase as the active agent
1 The Need for New Agents
7
[54]. Early leukemia researchers used guinea pig serum to treat ALL and indeed considered harvesting it for more widespread usage even from the larger related animal, the agouti. Luckily for these animals, l-asparaginase was extracted from bacteria, most notably from Escherichia coli [55] and subsequently from the plant pathogen Erwinia carotovora [56] (subsequently reclassified as chrysanthemi). In early studies, a single dose of E. coli l-asparaginase was shown to achieve remission in 25–60% of patients with relapsed ALL, with a median remission duration of 122 days. However, a number of different products have been produced and marketed with varying degrees of purification, different physical and chemical characteristics, especially their half-lives, and elimination (presumed to be through the reticuloendothelial system). Essentially, there are now in use two forms of native E. coli asparaginase and one of Erwinia, and both forms of E. coli l-asparaginase have been pegylated [57]. This involves polymerisation of polyethylene glycol to l-asparaginase. PEG Asparaginase has two advantages over the native drug. Firstly, pegylating E. coli l-Asparaginase protects its catalytic moiety from degradation and increases the half-life of the product from 26–30 h (native E. coli l-Asparaginase) to 5–7 days (PEG Asparaginase) [119]. Secondly, pegylation prevents interaction between the highly immunogenic antigenic sites of this native bacterial enzyme and the immune cells. As a result of the aforementioned two reasons, fewer doses of the pegylated product are needed in comparison to the native bacterial enzyme to achieve a comparabale clinical response (for example, 1,000 international units of PEG Asparaginase every 14 days compared with 10,000 international units of Erwinase every second day). Clinically, the incidence of anaphylaxis has fallen from around 25% to less than 1% following the switch to PEG Asparaginase [120]. Although the exact impact of “silent neutralising anti-l-Asparaginase antibodies” following PEG Asparaginase needs to be determined, there is evidence from one randomised trial showing that contrary to the native E. coli l-Asparaginase, the silent neutralising anti-l-Asparaginase antibodies that develop following PEG Asparaginase are of low titre and are not associated with low l-Asparaginase activity [120]. In order to reduce the risk of antibody production during induction and consolidation [121], when asparaginase is mostly used, some trials now use the pegylated E. coli product de novo, but some only use it in relapsed ALL. For efficacy it has been recognised that adequate plasma levels of enzyme to fully deplete asparagine need to be greater than 100 units/litre and ideally maintained for 3–4 weeks during induction [57]. We have naively assumed that asparagine depletion is the mechanism of its cytotoxic effect on blast cells. However, intracellular asparagine and glutamine are not normally depleted [58]. Furthermore, it has long been postulated that lymphoblasts are deficient in asparagine synthetase (AS) and hence their vulnerability to deficiency of the non-essential amino acid asparagine. However, some cells with AS remain sensitive to l-asparaginase. Additionally, it has been reported that mesenchymal stem cells that form the microenvironment in which marrow blasts survive and grow have AS levels, on average, 20 times higher than leukemic blasts [59]. There are some differences in the expression of AS levels between B-lineage cells (low) and T-lineage cells (higher), which have been correlated with decreased sensitivity to asparaginase, but clearly the picture is very
8
T. Eden
complex, and we do not really understand the mechanism of action of its resistance to this drug even after 40 years of use. Suffice to say that no matter how it works enzyme levels do require to be high and efficacy does appear to correlate with effective and prolonged asparagine depletion. Apoptosis may not be induced by this drug [57], but suboptimal asparagine depletion has been shown to affect survival. There has been some controversy about the dosage required; 1,000 international units of pegylated E.coli asparaginase is effective in depleting asparagine for 14–16 days given either intramuscularly or intravenously, but 2,500 units is more effective in depleting CSF asparagine [122, 123]. The enzyme itself does not cross the blood– brain barrier; nevertheless, the CSF can be depleted of asparagine. Monitoring of enzyme levels and antibody production has recently uncovered an interesting story. In a UK series of patients, nearly 80% attained high levels of enzyme during induction with doses of 1,000 international units of pegylated asparaginase (Oncaspar ò, Medac GmbH, Germany), but the remainder did not. About 5% had early inactivation and never recovered, but a similar number had low enzyme levels in induction but post-remission achieved good levels. The early inactivation was clearly not due to antibody production [60], but later inactivation was. The early inactivation by blast cells opens up the need to monitor both asparaginase levels, especially in highrisk patients where the problem principally lies, and the potential to overcome the problem of this unique resistance mechanism. Combined with a better understanding of the role of mesenchymal stem cells in protecting blast cells, we can clearly see that there is much still to learn about optimising asparaginase delivery. The potential to create recombinant asparaginase has led to early clinical trials. Finally, another intriguing piece of evidence was reported by Yang et al. [61]. Asparaginase induces hypoalbuminemia, which in turn is associated with lower dexamethasone clearance. This may enhance the effect of dexamethasone resulting in a degree of synergy. However, those who develop inhibiting antibodies to asparaginase or whose blasts produce a cleaving enzyme may have faster dexamethasone clearance. Once again, the potential complexity and interactions between drugs within combination therapy are emphasised. There is no doubt that we still have a lot to learn about the three key drugs universally used in ALL induction and how we can optimise their efficacy, let alone understand more accurately how they work and the nature of the resistance mechanisms that develop within blasts in response to exposure to these drugs.
1.3.5 Anthracyclines Anthracyclines produce their cytotoxicity via a number of intracellular interactions including free radical formation, inhibition of topoisomerase II, DNA intercalation, disturbance of helicase function and effects on signal transduction [62]. Their cytotoxicity appears to be concentration- and exposure-time-dependent. Klumper et al. [63] reported for childhood ALL leukemic blasts IC50 values of 0.117 mg/mL for daunorubicin and 0.347 mg/mL for doxorubicin. They have played a key role in
1 The Need for New Agents
9
AML, but in ALL their potential toxicity has led to a limitation on usage for standard-risk patients and reservation of their use in remission induction to higher risk patients. The largest single randomised trial for anthracyclines was the UK Medical Research Council trial VIII (1981–1984) [64]. Six hundred and thirty children were randomised to either receive or not receive two doses of daunorubicin (45 mg/m2/ dose) on the first 2 days of induction. The trial yielded a major improvement in disease-free survival (from 40 to 55%) in UK, which had been falling behind the results reported, for that era, from USA and Germany. This improvement was not due to the daunorubicin, which although improving early marrow clearance [65] and disease-free survival for those achieving remission was associated with more induction and early remission deaths. Long-term overall survival showed no benefit for inclusion of an anthracycline. This study demonstrated the degree of myelosuppression induced by anthracyclines. As confirmed in the ALL-BFM 90 study [46], survival can be improved even without anthracyclines. UKALL VIII, a trial without intensification or consolidation, comes from an historical era but clearly showed what could be achieved with a three drug induction and a sustained compliant approach to therapy. The results matched the results of the derivative CSG 162 protocol from USA but fell short of what is now possible [66]. Even then, it lagged behind those of the three BFM studies reported by Riehm et al. [67]. The BFM studies were once again demonstrating the real value of consolidation and intensification in effective therapy. The late side effects of anthracycline therapy, most specifically cardiotoxicity [68] and the potential of any topoisomerase 2 inhibitor to induce second malignancies [69], have pushed investigators towards reducing total dose exposure to anthracyclines in ALL to less than 200 mg/m2, although it remains unclear whether there is a true threshold effect for either toxicity. The real long-term benefit of concurrent cardioprotectants and/or liposomal anthracyclines also remains unclear.
1.3.6 Summary of Induction Challenges The three most essential induction drugs for the majority of ALL patients are now well defined: vincristine, steroids and l-asparaginase. Identification of those with resistance to all or any single agent would improve early disease control especially for high-risk patients if therapy is suitably adjusted or altered. Anthracyclines probably should be reserved for higher-risk patients and those with slow early response to the standard three agents. Whenever extra agents or a pulsed therapeutic approach has been substituted for the standard approach, survival has decreased [70]. It is doubtful if new agents will improve remission induction for the majority of patients. The exceptions are the 5–6% of cases with potentially primary refractory disease identified by molecular genetics or cytogenetics (e.g. BCR-ABL positive ALL, hypodiploidy) or by slow early response. Traditionally, early response has been measured by peripheral blood and/or bone marrow clearance of blasts [65] but
10
T. Eden
is now greatly enhanced by minimal residual disease monitoring by either molecular [71] or flow-cytometric [72] methods. The clinical parameters of age (<1 year to >10 years), gender, initial white cell count, etc. and standard marrow examination numerically failed to identify the majority of those who relapsed. This is where gene expression profiling for resistance patterns could play a role. Not only can those at risk be potentially identified more accurately by the new technology but it is also able to identify potential therapeutic targets (vide infra). Even in relapse, clinical remission can be achieved with vincristine, steroids and asparaginase in a high percentage of cases, but it is the depth of remission that is important in de novo and relapse cases, and which we can now measure more accurately. Getting induction right for each patient is critical for survival.
1.4 Consolidation and Intensification Even with no evidence of peripheral or marrow blasts on morphological examination at the end of induction, it is likely that there remains a tumour burden even as high as 108 blasts. The St Jude team emphasised, in 1968, the importance of consolidation [25]. It is not clear what combinations work best; more of the same for favourable groups [73], high-dose methotrexate and mercaptopurine, reinduction with steroids, vincristine and high-dose asparaginase or an augmented programme interspersing standard drugs with intravenous methotrexate during periods of myelosuppression [46, 74, 75]. What is more certain is that early post-remission treatment, just like induction, needs to be almost continuous, sustained and without gaps. The MRC UKALL X protocol [76] introduced a randomisation between no post-induction consolidation (as in its predecessor UKALL VIII), an early pulse (week 5), a late pulse (week 20) or both (week 5 + week 20). Those receiving two pulses of therapy had a significantly improved overall 10-year event-free survival (all risk groups) of 60% (54% in UKALL VIII and 56% with no pulses). UKALL XI [77], its successor, used the two pulses and randomised for a third but without further event-free survival improvement (10-year event-free survival 60%) [66, 77]. These results were not as favourable as reported from many other groups [29]. Interestingly, overall survival for the three studies showed an improving trend (UKALL VIII, 65%, X 71%, XI 79%) due to better post-relapse rescue. The problem with the UKALL X pulses was that the five drug combination of cytarabine, etoposide, thioguanine, daunorubicin, prednisolone, vincristine and intrathecal methotrexate required a 3–4 week marrow recovery time. We presume that it also allowed some residual blast recovery. The third module used in UKALL XI (in a randomised fashion) more closely resembled the BFM form of intensification using a much less myelosuppressive combination of an 8-week schedule of more vincristine, asparaginase, intrathecal methotrexate, intravenous cyclophosphamide and cytarabine and oral thioguanine. Much less recovery time was required, and it is this form of intensification that is now more widely used because it provides a sustained cytotoxic attack on residual blasts using drugs with different mechanisms
1 The Need for New Agents
11
of action but without excessive myelosuppression. Since the change in therapy to such an approach, event-free and overall survival in UK has once again reached comparable levels of those reported by other groups [36]. Choice of what agents are optimal during consolidation and intensification may need to be adjusted to subgroups of patients with particular biological characteristics or in the future potentially completely individualised.
1.4.1 Sensitivity/Resistance Patterns It has long been recognised that some subtypes of ALL have differential sensitivity or resistance to all or some cytotoxics. For example, hyperdiploid ALL (25% of childhood ALL) appears to be very sensitive to methotrexate with blast cells accumulating high intracellular levels of methotrexate and its active polyglutamates. Most hyperdiploid ALL blasts carry three or more copies of chromosome 21 including the folate transporter gene for cellular influx of methotrexate and some related compounds. Conversely, B-cell lineage blasts with ETV6-RUNX1 or E2APBX1 fusion genes and T-lineage ALL appear to accumulate much lower levels of polyglutamates [78]. Kager et al. [78] using oligonucleotide microarrays on diagnostic blasts showed lower expression of the reduced folate carrier (SLC19A1) in pre-B ALL with E2A-PBX1, high expression of the methotrexate efflux transporter ABCG2 in ETV6-RUNX1 ALL and reduced expression of FBGS, which catalyses the formation of polyglutamates in T-lineage ALL. As a result of these characteristics, surely, therapy should be modified by incorporating agents to which these blasts maybe more sensitive. Both the St Jude Total Therapy XIII protocol and the ALL-BFM 90 protocol used high-dose systemic methotrexate in consolidation and yielded 5-year event-free survival for pre-B ALL of 89.5 ± 7.3% and 93 ± 6%, respectively [79, 46]. Previous evidence had suggested less favourable outcome for this form of ALL when treated with standard antimetabolite therapy [80]. It is important to carefully schedule both hydration and leucoverin rescue (not too early or at too high a dose) following methotrexate; otherwise, its benefit can be reduced or even negated as reported by Skärby et al. [81] and observed in MRC UKALL XI [77]. ETV6-RUNX1 leukemia blasts appear to be selectively sensitive to l-asparaginase, whilst BCR-ABL and iAMP21 cells appear to be much less so. In the latter two forms of ALL, it has recently been identified that these blasts more commonly overexpress proteases, especially asparaginyl endopeptidase, which can cleave E. coli l-asparaginase and inactivate it [60]. The novel finding of AML1 gene amplification in ALL was reported by Niini et al. [82] and subsequently defined as representing multiple copies of the gene on a duplicated chromosome 21, dup(21) in near-diploid blast cell karyotypes of childhood ALL cases and in adult ALL [83]. This is now classified as iAMP21. The importance of this identification was that these patients (2% of cases) had a median age of 9 (higher than the childhood peak) median lower white cell counts (3.9) than other patients (12.4) amongst 1,630 childhood ALL patients screened [84]. All were
12
T. Eden
of pre-B or precursor B-cell lineage. In the trial, ALL 97, they had all been treated as standard risk, but their 5-year event-free and overall survival were very low at 29 and 71%, respectively, compared with 78 and 87% for all other patients, and relapses were early. Having identified this “new high risk” group, in the following ALL trial 2003, these patients were treated on a high-risk strategy with four drug induction, two delayed intensifications and higher-dose systemic methotrexate. To date, none have relapsed. What component of therapy made the difference is difficult to define. iAMP21 blast cells are associated with a deletion of the reduced folate carrier, suggesting decreased sensitivity to methotrexate. In ALL 2003, highrisk patients received intravenous methotrexate, and this may have contributed to the improved results. Other groups using the strategy now adopted by the UK researchers did not report such an adverse outcome with iAMP21 leukemia. This emphasises that prognosis is ultimately therapy related, but we cannot always define which part works for whom with a great deal of precision. Leukemias with BCR-ABL fusion genes and those with MLL gene rearrangements are discussed later as particular challenges for which truly new approaches are being explored. However, in the context of sensitivity, those with t(4;11) in infancy and adult ALL appear to be more sensitive to high-dose cytarabine possibly related to an over-expression of hENT1 gene, another cell membrane transporter. There is no doubt that gene expression profiling of leukemic cells has greatly improved our understanding of disease response and the ability to predict outcome. It has also provided information about new potential targets for therapy especially in resistant disease. The fact whether it will lead in the future to individualisation of treatment remains unclear. Accuracy of predicting relapse in T ALL and hyperdiploid (>50 chromosomes) ALL was emphasised by Yeoh et al. [50]. Since hyperdiploid ALL has such an overall good prognosis, predicting those with gene patterns that suggest a less good outcome is a major breakthrough.
1.5 Continuation Therapy The evidence from the earliest attempts to cure ALL demonstrated that, for reasons we still do not understand well, this disease requires continuation or “maintenance” therapy to prevent relapse. Some researchers have successfully truncated treatment following intensive initial therapy [85], but the definition of to whom this is safe is not currently possible. We routinely advocate 2–3 years of ongoing treatment with oral methotrexate (weekly) and a daily thiopurine, usually 6-mercaptopurine, both of which were originally discovered nearly 60 years ago. Adjustment of doses to maintain neutrophil counts in the range 0.5–1.5 × 109/L and platelets >75 × 109/L is advocated. We do not know how long continuation therapy really needs to be administered. Lennard et al. [86] reported higher relapse rates in children with lower red cells thioguanine nucleotide concentration and emphasised the importance of thiopurine methyltransferase (TPMT) in determining the cytotoxic effects of 6-mercaptopurine.
1 The Need for New Agents
13
Correlation between better survival and those with intermediate or low TPMT activity was reported by Relling et al. [87]. TPMT is subject to considerable genetic polymorphism. A very small minority of the population are homozygous for a mutant “null” allele, who are incredibly sensitive to mercaptopurine and can only be treated with very much reduced dosages. Stanulla et al. [88] reported from a series of 814 patients, treated on the ALL-BFM 2000 study, that those heterozygous for allelic variants of TPMT with low enzyme activity (55 patients) had a lower rate of minimal residual disease positivity (9.1%) after a 4-week cycle of 6-mercaptopurine (no previous exposure) than those with homozygous wild-type (755) alleles (22.8%). This amounted to a 2.9-fold reduction in risk of relapse for the wild-type heterozygotes. The threshold for MRD negativity was 10−4 using quantitative detection with allele-specific oligonucleotide-PCR methodology. This implies that dose modification on the basis of TPMT genotype might be extremely beneficial for the majority of patients who are in fact wild-type homozygotes and that 6MP may be a very useful component of early consolidation as it was in this study [88]. 6-thioguanine does not require to be metabolised to be activated, and leads to higher intracellular nucleotide levels, and in the CSF. Three recent trials have been randomised between 6-mercaptopurine and 6-thioguanine as the form of antimetabolite during consolidation and continuation therapy. 6-thioguanine at 60 mg/m2/ day proved to be too toxic to platelets [89], but even at 40 mg/m2/day profound thrombocytopenia, high rates of hepatic veno-occlusive disease (VOD) and increased rates of remission death were reported [89, 90]. In both these studies, a superior anti-leukemic response was noted, but this did not outweigh the toxic morbidity and mortality. The third study reported less toxicity but no advantage for 6-thioguanine over 6-mercaptopurine [91]. Low levels of TPMT may contribute to thioguanine-induced VOD but do not predict accurately enough. In the UK study (ALL 97/99) [90], all patients received a short course of thioguanine during intensification with a low level of toxicity especially of VOD. 6-Mercaptopurine, therefore, remains the purine of choice in ALL ongoing treatment, but in view of its increased anti-leukemic effect, short-term exposure to thioguanine in consolidation may be beneficial. Although pulses of vincristine and prednisone were included in continuation therapy, and demonstrated to be beneficial in an era before intensification [92], most therapy groups now omit them [46]. Indeed, Conter et al. [93] reported no benefit for six pulses of dexamethasone and vincristine given in the early phase of continuing therapy compared with no such pulses. A proposed trial randomising patients to receive or not receive such pulses is planned to start in UK shortly. The impact of blast genotype on sensitivity to methotrexate has already been covered in the Section 1.4. Since methotrexate is a mainstay of continuation therapy, similar considerations must be made with reference to its effectiveness in specific ALL subtypes during remission. What we really do not fully understand is how such long-term relatively lowdosage continuation therapy works. Is it indeed by direct blast cell cytotoxicity? The study by Stanulla et al. [88] confirms that adequate 6MP therapy can lead to reduced MRD. However, the role of the niche where blasts survive and proliferate
14
T. Eden
within the bone marrow and the role of mesenchymal/stroma cells in supporting their survival has already been alluded to [59]. Alternatively, could continuation therapy work by immune modulation? We just do not know, and when you do not know the mechanisms of action, it is often very difficult to improve effectiveness. Consequently, exploration of the role of the stroma and of immunomodulation would be advantageous. These may well be two areas where new agents are required in the future.
1.6 CNS-Directed Therapy The use of CNS irradiation and intensive therapy to provide both systemic and targeted extramedullary disease control, pioneered by the St Jude team, produced a dramatic increase in survival [28]. On the downside, cranio-spinal and subsequently cranial irradiation (24 and 18 Gy) were identified as causing significant growth retardation and impairment of intellectual development. Overt CNS disease at diagnosis is relatively rare (<5%) and probably is the only clear indication for cranial irradiation in de novo ALL, although some researchers now claim that highdose methotrexate and intensive intrathecal therapy can even cure overt CNS ALL. This remains controversial, but radiotherapy can be replaced by intrathecal methotrexate for most patients [94]. The risk factors that appear to increase CNS relapse are high white cell count and profound thrombocytopenia at diagnosis, T-lineage ALL and CSF blast cells at diagnosis (probably including low levels introduced by traumatic lumbar puncture) [95, 96]. Paradoxically, as marrow relapses have been reduced with modern era intensive therapy, CNS disease represents a greater proportion of relapses. Most trials now rely on intrathecal methotrexate, given as a prolonged course of injections during induction, consolidation intensification and during the first year of continuation therapy. How many doses of intrathecal therapy is adequate has not been clearly defined in the context of complex systemic treatment regimens that contain CNS “active” agents including high-dose steroids, especially dexamethasone, more effective asparaginase therapy through its asparagine depletion, effective thiopurine therapy and of course high-dose systemic methotrexate and/or “Capizzi” type approaches [35, 36, 90]. The protocol “M” of the BFM strategy has been proved to be very effective in this context [97]. Most series now report overall low CNS relapse rates of 3% or less. Triple intrathecal therapy (hydrocortisone, cytarabine and methotrexate) was pioneered by the Pediatric Oncology Group in USA. The CCG 1952 Study [98] demonstrated improved CNS leukemia control with triple versus single methotrexate injections but strangely increased marrow and testicular relapses. This emphasises the need for a truly systemic control approach without too much focus on “isolated” extramedullary disease sites. However, more research into the mechanisms of how blast cells escape from their marrow environment, the process of metastatic disease progression and how we can interfere with that may be a fruitful approach in the development of therapy for resistant and/or refractory disease.
1 The Need for New Agents
15
1.7 Subtype Definition and Treatment Challenges ALL is clearly a heterogeneous group of diseases, not only defined by immunophenotyping into the broad categories of B-cell lineage and T-cell ALL but also by cytogenetics and molecular genetics. Genetic studies have defined the leukemias with aberrant expression of proto-oncogenes, those with fusion genes resulting in chromosomal translocation (usually encoding for transcription factors or kinases) and chromosomal replication (e.g. hyperdiploidy with more than 50 chromosomes) [9]. Looking at all childhood ALL (from infancy onwards), 3% of most series have BCR-ABL; 2–3% have MLL rearrangements (especially in infancy), most commonly t(4;11), t(11;19) and 1% have hypodiploidy (less than 45 chromosomes), all of which have an adverse prognostic significance when treated with the conventional therapy delineated above. With older patients beyond childhood, there is a decrease in relative proportion of favourable subtypes, e.g. ETV6-RUNX1 and hyperdiploidy (>50) and an increase in BCR-ABL (25%), MLL rearrangements (10%) and hypodiploidy (2%) [9, 99]. There is a higher incidence of T-cell ALL in teenagers and adults and an increased expression of certain HOX genes. The change in subtype incidence goes some way to explain worsened outcome with increasing age. However, there have also been a series of reports demonstrating that teenagers and young adults treated on paediatric type protocols fare better than if treated on “adult” protocols. The biggest difference between the protocols in all of the reports are the higher dosages of vincristine, asparaginase and steroids and less cyclophosphamide and cytosine in the early stages of childhood therapy [100–105]. These findings have influenced policy decisions to change the upper age limit for childhood trial protocols (up to 25 or even older) with a subsequent reduction in relapse rates. However, poorer tolerance of therapy with increasing age, along with developing co-morbidities and decreased adherence to therapy by physicians and patients as a result of real or perceived increased toxicity also contribute to suboptimal outcome. The story of iAMP21 ALL has already been spelt out where outcome is clearly treatment protocol related. BCR-ABL acute ALL (and CML) treatment has been revolutionised by the addition of an inhibitor of the BCR-ABL tyrosine kinase. The BCR-ABL gene fusion product is a kinase affecting interacting signalling pathways that control cell proliferation, survival and self-renewal and in particular RAS [106]. The first inhibitor introduced into clinical practice, imatinib mesylate, does appear to improve remission duration considerably but resistance appears to develop at least in some. It is currently being tested in both European-wide and American paediatric trials. Alternative inhibitors are also now in trial alongside standard ALL treatment to see whether resistance can be overcome (Dasatinib and Nilotrinib). In ALL with MLL gene rearrangements most commonly seen during childhood in the first year of life (70–80% of cases at least), the selective greater sensitivity to cytarabine and the confusing monocytoid blast morphology (in many) have led to a mixed treatment strategy using elements of ALL and AML therapy with apparent improved outcome although still inferior to older children [107]. Many of these patients have blasts with over-expression of another receptor kinase FLT3.
16
T. Eden
A number of inhibitors are now being tested in clinical trials, which are discussed elsewhere in this book. FLT3 is also over-expressed in some hyperdiploid blasts. Controversy still exists as to whether T-cell ALL carries increased risk, independent of the association with high initial white cell counts and bulky extramedullary disease (mediastinal, nodal and CNS disease). Modern intensive therapy appears to be negating the risks previously observed even in adult patients. Mature B-cell ALL now treated with pulsed Burkitt’s lymphoma type therapy is associated with very high cure rates [108, 109]. A majority of T-lineage ALL blasts carry mutations of NOTCH pathway genes especially NOTCH1 which is involved in cell regulation through a network of responder genes including MYC [110, 111]. NOTCH1 mutations sufficient to alter signalling may play a crucial role in initiating T-cell ALL [112]. This pathway is being exploited for potential new therapeutic targets, e.g. gamma-secretase inhibitors (the enzyme is essential for NOTCH1 signalling [112]). However, these inhibitors have marked toxicity profiles on other organs, e.g. on gastrointestinal stem cells, which may limit usefulness [113]. This is a cautionary tale regarding the targeting of cell regulation pathways. Ballerini et al. [114] have recently reported on two interesting new features identified in T-cell ALL of childhood; the significantly adverse survival for those with TLX3 gene expression and especially when combined with expression of the fusion gene NUP214-ABL1 found in 3/18 of the TLX3 positive cases. This information is not only of prognostic significance but also points the way towards a need for alternative therapeutic approaches for such patients. Aberrant methylation (widespread hypomethylation, regional hypermethylation and increased cellular capacity for methylation) is a common feature in human neoplasia, including lymphoid malignancies. Researchers are beginning to explore ways to interfere with these abnormalities, e.g. use of decitabine, but it is too early to predict whether the approach will be valuable. There has also been a focus on anti-angiogenic agents, monoclonal antibodies (e.g. anti-CD20) inhibitors of signal transduction and, as alluded to earlier, increasing interest in the role of stromal cells. To date, again, it is premature to report real clinical success for any such approaches. Other authors cover the new agents in greater depth in their chapters. The question for this text is whether new agents are required at all and for what purpose and indeed for what form of ALL.
1.8 Conclusions The aim of this chapter is to review leukemia treatment and explore whether new agents are required and so what can we conclude. • Survival dramatically improved when the “total body therapy” concept was adopted. • The most critical feature has been to use the most effective induction drugs: vincristine, steroids and l-asparaginase with anthracyclines reserved for higherrisk patients.
1 The Need for New Agents
17
• However, recent evidence has suggested that we can deliver these standard drugs somewhat more effectively; higher-dose prednisone or dexamethasone; consider individual patient handling of vincristine with potential dose modification; and above all, recognise those patients who cleave l-asparaginase or develop antibodies to it. If we do not, we are impairing achievement of remission, increasing risk of relapse and giving ineffective potentially painful injections. • New forms of l-asparaginase, certainly pegylated, and probably recombinant forms are needed in routine practice. • The need for post-induction consolidation and intensification without any significant gaps has been demonstrated (almost continuous, not pulsed therapy, which itself may induce prolonged myelosuppression). • The first 6 months of therapy is crucial for survival, and the ability to identify minimal residual disease requiring intensification of therapy is the success story of the 1990s. • What is less clear is how this partial or complete resistance can be overcome, e.g. manipulation of the glycolytic pathway to overcome steroid resistance, selective use of specific agents for ALL subtypes, e.g. methotrexate for hyperdiploid patients or an increase in dosage where accumulation of polyglutamates is low (e.g. T-cell and E2A-PBX1 ALL). • Identification of subtypes with blast gene expression profiles that offer new targets, and hence new agents, can be developed to overcome resistance, e.g. BCR-ABL and FLT3 expressors with tyrosine kinase inhibitors and FLT3 inhibitors respectively, although how we deliver these alongside standard therapy is not yet fully defined. • Some of the identified targeted therapies, e.g. gamma secretase inhibitors, have had no totally predictable toxicity profiles. • All of these points to the potential to deliver very personalised ALL therapy for those at high risk. Stratification by MRD status at weeks 5 and 12/13 has ushered in such an era. • We have to be only too aware, however, that MRD measurement does not cure anyone but makes us aware of patients at low and high risk of relapse. It is how we respond to that with our therapy that matters. • What we have not yet fully exploited are identified individual polymorphisms in crucial metabolising genes, e.g. TPMT, which offer the opportunity to exploit slow or fast metabolism [115]. • So, yes, we need new agents for primary refractory/resistant disease, for optimal delivery of standard drugs and for matching both host genomics and “blast” characteristics to optimise cytotoxicity. • Above all, we need to understand the development of resistance, the role of stromal cells in protecting blasts and the mechanisms involved in extramedullary migration. • The era when we substantially decrease intensity and duration of therapy for most ALL patients is not yet here, but we must continue to strive to do so without loss of what we have achieved over the last four decades.
18
T. Eden
• Finally, we must remember that the great improvement survival has been restricted to about 20% of the children who acquire ALL worldwide. The remainder receive little supportive care and certainly no curative therapy. We must all strive to reach out and provide them with the hope of cure [116, 117].
References 1. Velpeau A. Sur la resorption du pusaet sur l’alteration du sang dans les maladies clinique de persection nenemant. Premier observation. Rev Med. 1827; 2:216. 2. Virchow R. Weisses blut. Notiz Geb Natur Heilk. 1845; 36:152–6. 3. Bennett JH. Case of hypertrophy of the spleen and liver in which death took place from suppuration of the blood. Edinburgh Med Surg J. 1845; 64:413–423. 4. Craigie D. Case of disease of the spleen in which death took place in consequence of the presence of purulent matter in the blood. Edinburgh Med Surg J. 1845; 64:400–413. 5. Virchow R. Die leukämie. In Virchow R (ed) Gesammelte abhandlungen zur wissenschaft lichen medizin. Frankfurt Meidinger. 1856; 190–211. 6. Friedreich N. Ein neuer fall von leukämie. Virchow’s Arch Pathol Anat. 1857; 12:37–58. 7. Neumann E. Ueber myelogene leukämie. Berl Klin Wochenschr. 1878; 15:69–72. 8. Piller GJ. Leukemia – A brief historical review from ancient times to 1950. British Journal of Haematology. 2001; 12:282–292. 9. Pui C-H, Robison L, Look AT. Acute lymphoblastic leukemia. Lancet. 2008; 371:1030–43. 10. Lissauer H. Zwei fälle von leucaemie. Berl Klin Wochenschr. 1865; 2:403–404. 11. Senn N. The therapeutic value of the Roentgen ray in the treatment of pseudo leukemia. N Y Med J. 1903; 77:665–668. 12. Lawrence JH. Nuclear physics and therapy: preliminary report on a new method for the treatment of leukemia and polycythemia. Radiology. 1940; 35:51–60. 13. Farber S, Diamond LK, Mercer RD et al. Temporary remissions in acute leukemia in children produced by the folic acid antagonist, 4-amino-pteroyl glutamic acid (aminopterin). N Engl J Med. 1948; 238:787–793. 14. Seeger DR, Smith JM, Hultquist ME. Antagonist for pteroylglutamic acid. J Am Chem Soc. 1947; 69:2567. 15. Farber S. The effect of ACTH in acute leukemia in childhood. In Mote JR (ed) First Clinical ACTH Conference New York. Blakiston. 1950; 325. 16. Farber S, Toch R, Seers EM et al. Advances in chemotherapy of cancer in man. Adv Cancer Res. 1956; 4:1–71. 17. Burchenal JH, Murphy ML, Ellison RR et al. Clinical evaluation of a new antimetabolite, 6 mercaptopurine, in treatment of leukemia and allied diseases. Blood. 1953; 8:965–999. 18. Goodman LS, Wintrobe MW, Dameshek W et al. Nitrogen mustard therapy. Use of methylbis (beta-chloroethyl) amine hydrochloride and tris (beta-chloroethyl) amine hydrochloride for Hodgkin’s disease, lymphosarcoma, leukemia, and certain allied and miscellaneous disorders. JAMA. 1946; 132:126–132. 19. Fernbach DJ, Sutow WW, Thurman WG et al. Clinical evaluation of cyclophosphamide. A new agent for the treatment of children with acute leukemia. JAMA. 1962; 182:30–37. 20. Rowinsky EK, Donehover RC. Antimicrotubule agents. In: Chabner BA, Longo DL (eds) Cancer chemotherapy and biotherapy: principles and practice. JB Lippincot Company, Philadelphia. 1996; 263–293. 21. Burchenal JH, Murphy ML. Long term survivors in acute leukemia. Cancer Res. 1965; 25:1491–1494. 22. Zuelzer WW. Implications of long-term survival in acute stem cell leukemia treated with composite cyclic therapy. Blood. 1964; 24:477–494.
1 The Need for New Agents
19
23. Krivit W, Gilchrist G, Beatty E. The need for chemotherapy after prolonged complete remission in acute leukemia of childhood. J Pediatr. 1970; 76:138–141. 24. Frei E III, Karon M, Levin RH et al. The effectiveness of combinations of antileukemia agents in inducing and maintaining remission in children with acute leukemia. Blood. 1965; 26:642–656. 25. George P, Hernandez K, Hustu O et al. A study of ‘total therapy’ of acute leukemia in children. J Pediatr. 1968; 72:399–408. 26. Pinkel D. Five-year follow up of ‘total therapy’ of childhood lymphocytic leukemia. JAMA. 1971; 216:648–652. 27. Pinkel D, Hernandez K, Borella L et al. Drug dosage and remission duration in childhood lymphocytic leukemia. Cancer. 1971; 27:247–256. 28. Aur RJA, Simone JV, Hustu HO et al. A comparative study of central nervous system irradiation and intensive chemotherapy early in remission of childhood acute lymphocytic leukemia. Cancer. 1972; 29:381–391. 29. Schrappe M, Camitta B, Pui C-H et al. Spotlight on long term results of pediatric ALL clinical trials from 12 study groups worldwide. Leukemia. 2000; 14:2193–2195. 30. Medical Research Council. The treatment of acute lymphoblastic leukemia in childhood leukemia UK ALL III. The effects of added cytosine arabinoside and/or asparaginase and a comparison of continuous or discontinuous mercaptopurine in regimens for standard-risk ALL. Medical and Pediatric Oncology. 1982; 10:501–510. 31. Medical Research Council UK ALL Trials 1972–84. Improvement in treatment for children with acute lymphoblastic leukemia. Report to the Council by the Working Party on Leukemia in Childhood. Lancet. 1986; 1:408–411. 32. Schrappe M, Reiter A, Riehm H. Cyto reduction and prognosis in childhood acute lymphoblastic leukemia. Journal of Clinical Oncology. 1996; 14:2403–2406. 33. Balis FM, Lester CM, Chrousos GP et al. Differences in cerebro-spinal fluid penetration of cortico-steroids: possible relationship to the prevention of meningeal leukemia. Journal of Clinical Oncology. 1987; 5:202–207. 34. Veerman AJP, Hählen K, Kamps WA et al. High cure rate with a moderately intensive treatment regimen in non-high risk childhood acute lymphoblastic leukemia: results of protocol ALL VI from the Dutch Childhood Leukemia Study Group. Journal of Clinical Oncology. 1996; 14:911–918. 35. Bostrom BC, Sensel MR, Sather HN et al. Dexamethasone versus prednisone and daily oral versus weekly intravenous mercaptopurine for patients with standard-risk acute lymphoblastic leukemia: a report from the Children’s Cancer Group. Blood. 2003; 101: 3809–3817. 36. Mitchell CD, Richards SM, Kinsey SE et al. Benefit of dexamethasone compared with prednisolone for childhood acute lymphoblastic leukemia: results of the UK Medical Research Council ALL 97 randomised trial. Br J Haematol. 2005; 128:734–745. 37. Igarashi S, Manabe A, Ohara A et al. No advantage of dexamethasone over prednisolone for the outcome of standard- and intermediate-risk childhood acute lymphoblastic leukemia in the Toyko Children’s Cancer Study Group L 95 – 14 protocol. J Clin Oncol. 2005; 23:6489–98. 38. Richter O, Ern B, Reinhardt D, Becker B. Pharmacokinetics of dexamethasone in children. Pediatric Pharmacol. 1983; 3:329–337. 39. Thompson EB. Apoptosis and steroid hormones. Mol Endocrinol. 1994; 8:665–673. 40. Laane E, Panaretakis T, Pokrovskaja K et al. Dexamethasone induced apoptosis in acute lymphoblastic leukemia involves differential regulation of Bcl-2 family members. Haematologica. 2007; 92 (11):1460–1469. 41. Wood AC, Elvin P, Hickman JA. Induction of apoptosis by anti-cancer drugs with disparate modes of action: kinetics of cell death and changes in C-MYC expression. Br J Cancer. 1995; 71:937–941. 42. Ito C, Evans WE, McNinch L et al. Comparative cytotoxicity of dexamethasone and prenisolone in childhood acute lymphoblastic leukemia. J Clin Oncol. 1996; 14:2370–2376.
20
T. Eden
43. Kaspers GJ, Veerman AJ, Popp-Snijders C et al. Comparison of the anti-leukemic activity in vitro of dexamethasone and prednisolone in childhood acute lymphoblastic leukemia. Med Pediatr Oncol. 1996; 27:114–121. 44. Quddus FF, Leventhal BG, Boyett JM et al. Glucocorticoid receptors in immunological subtypes of childhood acute lymphoblastic leukemia cells: a pediatric oncology group study. Cancer Res. 1985; 45:6482–6486. 45. Wei G, Twomey D, Lamb J et al. Gene expression based chemical genomics identifies rapamycin as a modulator of MCL1 and glucocorticoid resistance. Cancer Cell. 2006; 10:331–342. 46. Schrappe M, Reiter A, Ludwig WD. Improved outcome in childhood acute lymphoblastic leukemia despite reduced use of anthracyclines and cranial radiotherapy: results of trial ALL – BFM90. Blood. 2000; 95:3310–3322. 47. Karon M, Freireich E, Frei E. The role of vincristine in the treatment of childhood acute leukemia. Clin Pharmacol Ther. 1966; 7:332–339. 48. Crom WR, Siebold SN, Syold T et al. Pharmacokinetics of vincristine in children and adolescents with acute lymphoblastic leukemia. J Pediatr. 1994; 125:642–649. 49. Lönnerholm G, Frost BM, Abrahamsson J et al. Vincristine pharmacokinetics is related to clinical outcome in children with standard-risk acute lymphoblastic leukemia. British Journal of Haematology. 2008; 142:616–621. 50. Yeoh E-J, Ross ME, Shurtleff SA et al. Classification, subtype discovery and prediction of outcome in pediatric acute lymphoblastic leukemia by gene expression profiling. Cancer Cell. 2002; 1:133–143. 51. Holleman A, Cheok MH, den Boer ML et al. Gene expression patterns in drug resistant acute lymphoblastic leukemia cells and response to treatment. N Engl J Med. 2004; 351: 533–542. 52. Cario G, Stanulla M, Fine BM et al. Distinct gene expression profiles determine molecular treatment response in childhood acute lymphoblastic leukemia. Blood. 2005; 105:821–826. 53. Cheok MH, Evans WE. Acute lymphoblastic leukemia: a model for the pharmacogenomics of cancer therapy. Nature Reviews Cancer. 2006; 6:117–129. 54. Becker FF, Broome JD. L-asparaginase: inhibition of early mitoses in regenerating rat liver. Science. 1967; 156:1602–1603. 55. Mashburn LT, Wriston JC. Tumour inhibitory effect from Escherichia coli. Archives of Biochemistry and Biophysics. 1964; 105:450–452. 56. Wade HE, Elsworth R, Herbert E et al. A new L-asparaginase with anti-tumour activity? Lancet. 1968; 2:776–777. 57. Pinheiro JPV, Boos J. The best way to use asparaginase in childhood acute lymphoblastic leukemia still to be defined. British Journal of Haematology. 2004; 125:117–127. 58. Appel IM, Kazemier KM, Boos J et al. Pharmacokinetic, pharmacodynamic and intracellular effects of PEG-asparaginase in newly diagnosed childhood acute lymphoblastic leukemia: results from a single agent window study. Leukemia. 2008;22:1665–1679. 59. Iwamoto S, Mihara K, Downing JR. Mesenchymal cells regulate the response of acute lymphoblastic leukemia cells to asparaginase. The Journal of Clinical Investigation. 2007; 117:1049–1057. 60. Patel N, Krishnan S, Offman MN et al. A dyad of lymphoblastic lysosomal cysteine proteases degrade the key anti-leukemic drug L-asparaginase. Journal of Clinical Investigation. 2009; 119(7):1964–1973. 61. Yang L, Panetta JC, Cai X et al. Asparaginase may influence dexamethasone pharmacokinetics in acute lymphoblastic leukemia. Journal of Clinical Oncology. 2008; 26(12):1932–1939. 62. Doroshow JH. Anthracyclines and anthracenediones. In: Chabner BA, Longo DL (eds) Chemotherapy and biotherapy: principles and practice. JB Lippincott Company, Philadelphia. 1996; 409–434. 63. Klumper E, Pieters R, Veerman AP et al. In vitro cellular drug resistance in children with relapsed/ refractory acute lymphoblastic leukemia. Blood. 1995; 86:3861–3868.
1 The Need for New Agents
21
64. Eden OB, Lilleyman JS, Richards S et al. Results of Medical Research Council Childhood Leukemia Trial UK ALL VIII. British Journal of Haematology. 1991; 78:187–196. 65. Lilleyman JS, Gibson BS, Stevens RF et al. Clearance of marrow infiltration after one week therapy for childhood lymphoblastic leukemia: clinical importance and the effect of daunorubicin. Br J Haematol. 1997; 97:603–606. 66. Eden OB, Harrison G, Richards S et al. Long term follow up of the United Kingdom Medical Research Council protocols for childhood acute lymphoblastic leukemia, 1980–1997. Leukemia. 2000; 14(12):2307–2320. 67. Riehm H, Gadner H, Henze G et al. Acute lymphoblastic leukemia. Treatment results in 3 BFM studies (1970–1981). In: Murphy SB, Gilbert JR (eds) Leukemia Research: Advances in cell biology and treatment. Elsevier Biomedical, New York. 1998; 251–263. 68. Lipschultz SE, Colan SD, Gelber RD et al. Late cardiac effects of doxorubicin therapy for acute lymphoblastic leukemia in childhood. N Engl J Med. 1991; 324:808–815. 69. Ng A, Taylor GM, Eden OB. Treatment-related leukemia – a clinical and scientific challenge. Cancer Treatment Reviews. 2000; 6:377–391. 70. Eden OB, Lilleyman JS, Shaw MP et al. MRC Leukemia Trial VIII compared with trials II – VII: lessons for future management. Haematology, Blood Transfusion. 1987; 30:448–455. 71. Van Dongen JJ, Seriu T, Panzer-Grumayer ER et al. Prognostic value of minimal residual disease in acute lymphoblastic leukemia in childhood. Lancet. 1998; 352:1731–1738. 72. Coustan-Smith E, Sancho J, Behm FG et al. Prognostic importance of measuring early clearance of leukemic cells by flow cytometry in childhood acute lymphoblastic leukemia. Blood. 2002; 100:52–58. 73. Loh ML, Goldwasser MA, Silverman LB et al. Prospective analysis of TEL/ AML1-positive patients treated on Dana-Farber Cancer Institute Consortium Protocol 95 – 01. Blood. 2006; 107:4508–4513. 74. Pui C-H, Evans WE. Treatment of acute lymphoblastic leukemia. N Engl J Med. 2006; 354(2):166–178. 75. Nachman JB, Sather HN, Sensel MG et al. Augmented post-induction therapy for children with high-risk acute lymphoblastic leukemia and a slow response to initial therapy. N Engl J Med. 1998; 338:1663–1671. 76. Chessells JM, Bailey CC, Richards SM. Intensification of treatment and survival in all children with lymphoblastic leukemia: results of the UK MRC Trial UKALL X. Lancet. 1995; 345:143–148. 77. Hill FGH, Richards SM, Gibson B et al. Successful treatment without cranial radiotherapy of children receiving intensified chemotherapy for acute lymphoblastic leukemia. Results of the risk stratified randomised CNS treatment trial MRC UKALL XI. Br J Haematol. 2004; 124 (1):33–46. 78. Kager L, Cheok M, Yang W et al. Folate pathway gene expression differs in subtypes of acute lymphoblastic leukemia and influences methotrexate pharmacodynamics. The Journal of Clinical Investigation. 2005; 115(1):110–117. 79. Pui C-H, Relling MV, Downing JR. Acute lymphoblastic leukemia. N Engl J Med. 2004; 350:1535–1548. 80. Raimondi SC, Behm FG, Robertson PK et al. Cytogenetics of pro-B cell acute lymphoblastic leukemia with emphasis on prognostic implications of the t (1;19). J Clin Oncol. 1990; 8:1380–1388. 81. Skärby TV, Anderson H, Heldrup J et al. High leucovorin doses during high-dose methotrexate treatment may reduce the cure rate in childhood lymphoblastic leukemia. Leukemia. 2006; 20:1955–1962. 82. Niini T, Kanerva J, Vettenranta K et al. AML1 gene amplification: a novel finding in childhood ALL. Haematologica. 2000; 85:362–366. 83. Harewood L, Robinson H, Harris R et al. Amplification of AMLI on a duplicated chromosome 21 in acute lymphoblastic leukemia: a study of 20 cases. Leukemia. 2003; 17:547–553.
22
T. Eden
84. Moorman AV, Richards SM, Robinson HM et al. Prognosis of children with acute lymphoblastic leukemia (ALL) and intrachromosomal amplification of chromosome 21 (iAMP21). Blood. 2007; 109:2327–2330. 85. Toyoda Y, Manabe A, Tsuchida M et al. Six months of maintenance chemotherapy after intensified treatment for acute lymphoblastic leukemia of childhood. J Clin Oncol. 2000; 18:1508–1516. 86. Lennard L, Lilleyman JS, Van Loon J et al. Genetic variation in response to 6 – mercaptopurine for childhood acute leukemia. Lancet. 1990; 336:225–229. 87. Relling MV, Hancock ML, Boyett JM et al. Prognostic importance of 6 –mercaptopurine dose intensity in acute lymphoblastic leukemia. Blood. 1999; 93:2817–2823. 88. Stanulla M, Schaeffëler E, Flohr T et al. Thiopurine methyltransferase (TPMT) genotype and early treatment response to mercaptopurine in childhood acute lymphoblastic leukemia. JAMA. 2005; 293(12):1485–1489. 89. Stork LC, Sather H, Hutchinson RJ et al. Comparison of mercaptopurine (MP) with thioguanine (TG) and I/T methotrexate (ITM) with I/T “triples” (ITT) in children with standard-risk ALL: results CCG – 1952. Blood. 2002; 100:156a 90. Vora A, Mitchell CD, Lennard L et al. Toxicity and efficacy of 6-thioguanine versus 6-mercaptopurine in childhood lymphoblastic leukemia: a randomised trial. Lancet. 2006; 368:1339–1348. 91. Harms DO, Gobel U, Spaar HJ et al. Thioguanine offers no advantage of mercaptopurine in maintenance treatment of childhood ALL: results of randomised trial COALL-92. Blood. 2003; 102:2736–2740. 92. ALL Collaborative Group. Duration and intensity of maintenance chemotherapy in acute lymphoblastic leukemia: overview of 42 trials involving 12,000 randomised children. Lancet. 1996; 346:1783–1788. 93. Conter V, Valsecchi MG, Silvestri D et al. Pulses of vincristine and dexamethasone in addition to intensive chemotherapy for children with intermediate-risk acute lymphoblastic leukemia: a multicentre randomised trial. Lancet. 2007; 369:123–131. 94. Clarke M, Gaynon P, Hann I et al. CNS-directed therapy for childhood acute lymphoblastic leukemia: childhood ALL Collaborative Group overview of 43 randomised trials. J Clin Oncol. 2003; 21:1798–1809. 95. Gajjar A, Harison PL, Sandlund JT et al. Traumatic lumbar puncture at diagnosis adversely effects outcome in childhood acute lymphoblastic leukemia. Blood. 2000; 96:3381–3384. 96. Te Loo DM, Kamps WA, Van der Does-van den Berg AV et al. Prognostic significance of blasts in the cerebrospinal fluid without pleocytosis or a traumatic lumbar puncture in children with acute lymphoblastic leukemia: the experience of the Dutch Childhood Oncology Group. J Clin Oncol. 2006; 24:2332–2336. 97. Schrappe M, Reiter A, Zimmerman M et al. Long term results of four consecutive trials in childhood ALL performed by the ALL-BFM Study Group from 1981 – 1995. Leukemia. 2000; 14:2205–2222. 98. Matloub Y, Lindemulder S, Gaynon PS et al. Intrathecal triple therapy decreases central nervous system relapse but fails to improve event free survival when compared with intrathecal methotrexate: results of the Children’s Cancer Group (CCG) 1952 study for standard-risk acute lymphoblastic leukemia, reported by the Children’s Oncology Group. Blood. 2006; 108:1165–1173. 99. Chessells JM, Hall E, Prentice HG et al. The impact of age on outcome in lymphoblastic leukemia; MRC UKALL X and Xa compared: a report from the MRC Paediatric and Adult Working Parties. Leukemia. 2004; 12:463–473. 100. Nachman J, Sather HN, Buckley JD et al. Young adults 16 – 21 years of age at diagnosis entered on Children’s Cancer Group acute lymphoblastic leukemia and acute myeloblastic leukemia protocols. Cancer. 1993; 71:3377–3385. 101. Boissel N, Auclerc M-F, Lhėritier V et al. Should adolescents with acute lymphoblastic leukemia be treated as old children or young adults? Comparison of the French FRALLE-93 and LALA-94 trials. J Clin Oncol. 2003; 21:774–780.
1 The Need for New Agents
23
102. De Bont JM, Holt B, Dekker Am et al. Significant difference in outcome for adolescents with acute lymphoblastic leukemia treated on pediatric versus adult protocols in the Netherlands. Leukemia. 2004; 18:2032–2035. 103. Ramanujachar R, Richards S, Hann I et al. Adolescents with acute lymphoblastic leukemia: outcome on UK National Paediatric (ALL97) and Adult (UKALL XII/ E2993) trials. Pediatric Blood and Cancer. 2007; 48:254–261. 104. Hallböök H, Gustafsson E, Smedmyr B et al. Treatment outcome in young adults and children > 10 years of age with acute lymphoblastic leukemia in Sweden: a comparison between a pediatric protocol and an adult protocol. Cancer. 2006; 107: 1551–1561. 105. Barry E, De Angelo DJ, Neuberg D et al. Favourable outcome for adolescents with acute lymphoblastic leukemia treated on Dana-Farber Cancer Institute Acute Lymphoblastic Leukemia Consortium protocols. J Clin Oncol. 2007; 25:813–819. 106. Ren R. Mechanisms of BCR-ABL in the pathogenesis of chronic myelogenous leukemia. Nat Rev Cancer. 2005; 5:172–183. 107. Pieters R, Schrappe M, De Lorenzo P et al. A treatment protocol for infants younger than 1 year with acute lymphoblastic leukemia (Interfant-99): a observational study and a multicentre randomised trial. Lancet. 2007; 370:240–250. 108. Patte C, Auperin A, Michon J et al. The Societe Francaise d’Oncologie Pediatrique LMB89 protocol: highly effective multiagent chemotherapy tailored to the tumor burden and initial response in 561 unselected children with B cell lymphomas and L3 leukemia. Blood. 2001; 97:3370–3379. 109. Reiter A, Schrappe M, Ludwig WD et al. Intensive ALL-type therapy without local radiotherapy provides a 90% event free survival for children with T cell lymphoblastic lymphoma: a BFM group report. Blood. 2000; 95:416–421. 110. Weng AP, Ferrando AL, Lee W et al. Activating mutations of NOTCH I in human T cell acute lymphoblastic leukemia. Science. 2004; 306:269–271. 111. Palomero T, Lim WK, Odom DT et al. Notch I directly regulates C-MYC and activates a feed-forward-loop transcriptional network promoting leukemic cell growth. Proc Natl Acad Sci. 2006; 103:18261–18266. 112. Armstrong SA, Look AT. Molecular genetics of acute lymphoblastic leukemia. J Clin Oncol. 2005; 23:6306–6315. 113. Wong GT, Manfra D, Poulet FM et al. Chronic treatment with the gamma-secretase inhibitor LY-411.575 inhibits beta-amyloid peptide production and alters lymphopoiesis and intestinal cell differentiation. J Bio Chem. 2004; 279:12876–12882. 114. Ballerini P, Landman-Parker J, Cayuela JM et al. Impact of genotype on survival of children with T cell acute lymphoblastic leukemia treated according to French protocol FRALL-93: the effect of TLX3/ HOX11L gene expression on outcome. Haematologica. 2008; 93(11): 1658–1668. 115. Rocha JCC, Cheng C, Lui W et al. Pharmacogenetics of outcome in children with acute lymphoblastic leukemia. Blood. 2005; 105(12):4752–4758. 116. Eden T. Translation of cure for acute lymphoblastic leukemia to all children. Br J Haem. 2002; 118:945–951 117. Eden T, Pui C-H, Schrappe M et al. All children have the right to full access to treatment for cancer. Lancet. 2004; 364:1121–1122. 118. Lugthart S, Cheok MH, den Boer ML et al. Identification of genes associated with chemotherapy cross resistance and treatment response in childhood acute lymphoblastic leukemia. Cancer Cell. 2005; 7:375–386. 119. Asselin BL, Whitin JC, Coppola DJ. Comparative pharmacokinetic studies of three asparaginase preparations. J Clin Oncol. 1993; 11:1780–1786. 120. Avramis VI, Sense S, Periclou AP et al. A randomized comparison of native Escherichia coli asparaginase and polyethylene glycol conjugated asparaginase for treatment of children with newly diagnosed standard-risk acute lymphoblastic leukemia: a Children’s Cancer Group study. Blood. 2002; 99:1986–1994.
24
T. Eden
121. Woo MH, Hak LJ, Storm MC et al. Anti-asparaginase antibodies following E. coli asparaginase therapy in pediatric acute lymphoblastic leukemia. Leukemia. 1998; 12:1527–1533. 122. Hawkins DS, Park JR, Thomson BG et al. Asparaginase pharmacokinetics after intensive polyethylene glycol-conjugated l-asparaginase therapy for children with relapsed acute lymphoblastic leukemia. Clin Cancer Res. 2004; 10:5335–5341. 123. Appel IM, Pinheiro JP, den Boer ML et al. Lack of asparagine depletion in the cerebrospinal fluid after one intravenous dose of PEG-asparaginase: a window study at initial diagnosis of childhood ALL. Leukemia. 2003; 17:2254–2256.
Chapter 2
Identifying Targets for New Therapies in Children with Acute Lymphoblastic Leukemia Shekhar Krishnan, Ashish Masurekar, and Vaskar Saha
Summary For those of us who look after children with acute lymphoblastic leukemia (ALL), these are heady times. Cure rates on current therapeutic regimens are now approaching 90% [1, 2]. Therapy is almost entirely chemotherapy-based with very few patients now receiving irradiation [3]. Why then in this group of patients should we be looking for new agents? The obvious one is that we are reaching the limits of what can be achieved with combination chemotherapy [4]. In a sense we have been lucky. Almost all of the earliest chemotherapeutic agents proved effective in childhood ALL. Children tolerate combination chemotherapy better than adults. This has allowed us to gradually intensify therapy in all groups and in particular those at a higher risk of relapse. This risk-stratified approach to intensification has proven to be highly effective [5–11]. One problem we now face is the high cost of cure. Treatment-related mortality and morbidity [12] is almost balancing out the relative risk of relapse. Allogeneic stem cell transplant (allo-SCT), the ultimate in treatment intensity, cannot cure patients unless disease burden is first reduced using chemotherapy [13, 14]. Thus, intensification of therapy is unlikely to improve outcome any further. We therefore need new drugs not only to cure those currently failed by therapy but also to decrease the morbidity of current treatment. At present most protocols use ten or more drugs over a period of 2–3 years to treat children with ALL. The cost of treatment and supportive care is prohibitive for countries with restricted resources. This includes the most heavily populated parts of the world. Thus, the remarkable success rates seen in developed countries are yet to be translated globally [15]. To provide a solution for all children with ALL we need shorter,
S. Krishnan (*) Cancer Research UK Children’s Cancer Group, School of Cancer and Enabling Sciences, Manchester Academic Health Sciences Centre, The University of Manchester, Central Manchester and Manchester Children’s NHS Trust, Christie Hospital NHS Foundation Trust, Manchester M20 4BX, UK e-mail:
[email protected] V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3_2, © Springer Science+Business Media, LLC 2011
25
26
S. Krishnan et al.
cheaper therapeutic strategies. Finally, childhood ALL is a paradigm for successful cancer therapy. In terms of modern biology, it is one of the most heavily investigated. In a sense, having resolved the therapeutic dilemma we now have the luxury of dissecting out the mechanisms of cure and resistance. It is likely that the biological mechanisms that regulate the variations in the therapeutic response and side effects are common to more than one tumour type. Thus, the mechanisms identified are likely to have wider application in the treatment of cancer.
2.1 Understanding Disease Biology In the following sections, we pose key questions, the solutions to which we believe are fundamental in advancing and refining therapy in childhood ALL.
2.2 Can We Further Optimise Current Therapy? Biologically, childhood ALL is a heterogeneous disease. Cytogenetic analysis demonstrates this and outcome clearly varies by genetic subtype. However, even within a cytogenetic subset, there is disparity. A small proportion of those with hyperdiploidy relapse. Similarly, a small proportion of those with Philadelphia chromosome positive (Ph+) ALL respond well to chemotherapy. As hyperdiploidy is more common, relapse in this group poses a bigger clinical problem. Biological heterogeneity is also reflected in the therapy used. Drugs used predominantly affect nucleic acid integrity (intercalating agents, epipodophyllotoxins), synthesis (anti-metabolites), replication (mitotic spindle poisons) and transcription (steroids). Notable in this armamentarium is the drug l-asparaginase, which exerts a unique cytotoxic effect specific to this cancer and is a pivotal drug in the treatment of childhood ALL. The most sensitive predictor of outcome has proven to be the early response to therapy, measured either by a decrease in circulating blasts, percentage residual blasts in the marrow or molecular level of disease during initial therapy. Thus, the heterogeneity of disease has serendipitously been tackled by the use of multi-targeted therapy effective against the most frequent subtypes of childhood ALL. This accords with the Goldie–Coldman hypothesis [16] which avers that drug resistant clones are less likely to evolve in tumours treated with the most effective combination chemotherapy. However, while this sweeping approach may stochastically benefit the majority, a proportion of patients will necessarily be over treated and some will not receive appropriate therapy. Further fine-tuning is still possible with current chemotherapeutic agents. For example, there is evidence to suggest that patients with ETV6-RUNX1 ALL are more sensitive to asparaginase [17] and that overexpression of the folate reductase carrier gene through duplication of chromosome 21 renders patients with hyperdiploid ALL more sensitive to methotrexate [18]. Thus, protocols could be adapted to increase exposure of the specific drug in each cytogenetic category [19]. The problem we face is that we are still unsure about the precise
2 Identifying Targets for New Therapies in Children
27
mechanisms of actions of drugs and the consequences of their interactions. This limits our ability to predict recurrences. A better understanding of the biological processes is now ushering in an era of individualised therapy. An exemplar of this is the ABL tyrosine kinase inhibitors in Ph+ ALL [20].
2.3 What Are the Origins of Relapse? Relapsed ALL is broadly risk-stratified by the duration of first remission. Those who relapse early are often incurable, even with allo-SCT. In contrast those who relapse late, off therapy, have survival rates of over 70% with conventional chemotherapy [21–23]. Genome-wide analysis has recently shown that in almost 90% of cases, disease recurrence is due to a sub-clone present at original diagnosis [24]. This observation is supported by results of recent xenotransplantation experiments performed in more permissive immunodeficient mice recipients. In these studies, lymphoblast populations designated mature by immunophenotypic criteria also appear to possess stem cell properties, suggesting that “stemness” in ALL is more widely prevalent than previously recognised [25]. Gene expression profiling (GEP) has also been used to investigate relapsed disease [26–28]. GEP analyses suggest that transcriptional signatures differ between diagnostic and relapse blasts in early but not late relapses [28]. This suggests that early relapses occur as the result of a sub-clone already present at original diagnosis. Intriguingly, GEP analyses suggest that this clone is highly proliferative and thus the mechanisms by which it resists chemotherapy and allo-SCT remain to be elucidated. In contrast, in late relapses, there are at least two possibilities. There is evidence to suggest that these relapses are derived from the same ancestral clone that gave rise to the original leukemia [29, 30]. In essence, this is a second leukemia but as result of its origin, it retains the chemosensitivity of the original disease. Thus, these patients respond well to chemotherapy. Within this group of later relapses, we know that there are patients who show a slower clearance of disease. These patients often require allo-SCT to sustain remission. As discussed later, these differences may be accounted for by germline polymorphisms in genes regulating drug metabolism.
2.4 Why Do Relapses Occur at Extramedullary Sites? A conundrum in relapsed ALL is the high incidence of recurrence at extramedullary sites such as the central nervous system (CNS). Such relapses tend to occur late and can be either isolated or combined with a bone marrow relapse. Curiously, the outcome of combined or isolated extramedullary relapses is better than isolated marrow relapse. This is puzzling as in most cases of isolated extramedullary disease it is possible to detect low levels of marrow involvement using molecular techniques [31]. Why and how do lymphoblasts enter extramedullary compartments? The clue
28
S. Krishnan et al.
lies perhaps in the observation of a striking dichotomy in CNS disease incidence between diagnosis and relapse. While CNS disease is a rare feature of de novo ALL, it is seen in around 30% of ALL relapses [19]. It could thus be argued that this phenomenon is selected for by chemotherapy [32]. One possibility is that residual leukemic cells, protected by interactions with the marrow microenvironment, proceed to breach endothelial-matrix barriers and infiltrate extramedullary niches. It is likely that within these sanctuary niches, cells are protected from chemotherapy-induced cell death and give rise to extramedullary recurrences.
2.5 How Do We Account for the Heterogeneity in Treatment Response? Heterogeneity in response to any single drug is commonly observed in patients with ALL. There is evidence to suggest that this is considerably influenced by host genome polymorphisms that regulate drug handling. Polymorphisms relating to increased drug clearance may be responsible for a slower clearance of disease [33–35]. Leukemic blasts may also contribute to the variations in therapeutic response. An example is the enzyme thiopurine S-methyltransferase (TPMT). The metabolism of thiopurines is regulated by TPMT. Lower levels of the enzyme are associated with higher toxicity and better outcomes. TPMT is located on chromosome 6p. Duplication of this region in the somatic genome can result in high levels of the enzyme in lymphoblasts, which are then able to clear the drug more rapidly [36]. Thus, the tolerated dose of thiopurine may be insufficient to kill blast cells in such cases. More recently, lymphoblasts have been shown to produce proteases capable of inactivating asparaginase [37].
2.6 What Is the Role of the Tumour Microenvironment? Host-tumour interactions are clearly an important component of the spectrum of mechanisms leading to therapeutic failure. Mesenchymal stem cells and haematopoietic stem cell (HSC) niches may provide a protective marrow microenvironment for leukemic cells [38]. As remarked earlier, intrinsically chemosensitive ALL blasts that have the ability to migrate to HSC niches may weather the chemotherapy storm under the umbrella of the microenvironment and re-emerge to cause disease recurrence. These patients may respond to allo-SCT, where ablative conditioning creates an empty marrow niche, thus removing the protective microenvironment. The niche is then colonised by donor HSCs that are presumably able to outcompete residual leukemic cells. Some leukemic cells are capable of modifying the HSC niche to their own advantage, creating their own microenvironment and displacing normal haematopoietic progenitors [39]. Clearly such disease is likely to be incurable with conventional chemotherapy and
2 Identifying Targets for New Therapies in Children
29
allo-SCT. As illustrated by the success of tyrosine kinase inhibitors in Ph+ ALL, such leukemias require targeted therapy. In this context, it is entirely plausible that the same signalling mechanisms that facilitate long-term survival of blasts also facilitate disease progression and extramedullary spread. If so, targeting these survival pathways may prevent disease recurrence.
2.7 How Do We Discover Novel Biological Targets? Transcriptional profiling has been shown to be predictive of in vitro chemosensitivity [40], as well as the rapidity of response to therapy [41, 42]. However overexpressed genes often lie at the end of a regulatory cascade and it is difficult to ascertain which if any of these genes are directly responsible for therapeutic failure. Additionally, global profiling, even if multi-omic and integrated, does not intrinsically have the resolution to detect expression signatures of minor sub-clones that later account for relapse. Yet, despite these limitations, microarray platforms are already aiding discovery of potential adverse prognostic markers amenable to therapeutic targeting. Aberrant kinase activity has recently been identified as a recurring feature of high-risk disease. Detailed analyses by a number of groups [43–46] show that as a result of either a somatic translocation or deletion, some patients overexpress the cytokine receptor, cytokine receptor-like factor 2 (CRLF2). CRLF2 overexpressing patients have a significantly worse outcome [43]. GEP analyses shows that these patients have an expression signature similar to that seen in Ph+ ALL [47, 48], which includes the adverse-risk Ikaros deletion [43]. More importantly CRLF2 overexpression is associated with somatic activating Janus family kinase (JAK) mutations [49]. These studies suggest that Ikaros, JAK and CRLF2 aberrations cooperate in leukemogenesis and targeting the JAK-STAT signalling mechanism may be an attractive therapeutic strategy.
2.8 ALL: A Conceptual Model for Treatment Figure 2.1 is a proposed conceptual model that integrates lymphoblast characteristics and reciprocal host-tumour interactions to establish a biological and therapeutic paradigm in childhood ALL. Using this model, disease may be categorised into three broad groups: Group 1. Highly proliferative, stroma-independent and exquisitely chemosensitive blasts. Group 2. Intrinsically chemosensitive blasts that evade chemotherapy by interacting with the host microenvironment; ablative allo-SCT here removes stromal chemoprotection and is curative. Inhibitors capable of disrupting adhesive tumour-stroma interactions (e.g., CXCR4 antagonists) may also have a role here [50].
30
S. Krishnan et al.
Fig. 2.1 An integrated disease paradigm for therapy in childhood ALL. See text for details
Group 3. B lasts that evade chemotherapy by establishing stroma-independent tumour niches; ablative allo-SCT is ineffective in this group and strategies that target key survival mechanisms are required, exemplified by the ABL tyrosine kinase inhibitors in Ph+ ALL. Powerfully intersecting with this model are host germline polymorphisms that determine drug disposal and tolerance.
2.9 New Molecular and Cellular Treatment Targets As remarked previously, with the notable exception of asparaginase, all cytotoxic agents including steroids essentially target nuclear mechanisms. The cell nucleus continues to be a focus of drug targeting and many compounds in this class have entered early clinical trials. This includes a number of new nucleoside analogues, the aurora kinase inhibitors that target mitotic spindles [51] and the inhibitors of histone deacetylases [52] and DNA methyltransferases [53, 54] that target the dysregulated transcriptional programme in ALL blasts. The chapters in this monograph discuss a number of alternative approaches targeting cellular processes and molecules identified in Fig. 2.2. The alternative approaches may be loosely categorised as below. Not all targets have well identified pre-clinical agents in phase I trials.
2 Identifying Targets for New Therapies in Children
31
Fig. 2.2 A schematic representation of molecular and cellular targets of therapy in acute lymphoblastic leukemia. The lymphoblast nucleus is the principal therapeutic target. Tyrosine kinase inhibitors suppress constitutively activated receptor (e.g., FLT3) or downstream cytoplasmic tyrosine kinases (JAK, SRC, ABL1). Leukemic cell apoptosis is enhanced by suppressing mTOR kinase activity (Rapamycin and analogues) or by NFkB-mediated proteasome inhibition. Nonclassical (lysosomal) death pathways may also be triggered by BCL2 family antagonists or by antibody ligation of surface molecules. Monoclonal antibodies typically mediate leukemic cell clearance by activating immune effector mechanisms or by disrupting stromal adhesion. Targeting activated Notch signalling is a potential strategy in T-lineage disease. Asparaginase is unique in its cytotoxicity, selectively perturbing blast cell protein synthesis through substrate depletion. Response to therapy is strongly influenced by host germline polymorphisms governing drug disposal and tolerance. ABL Abelson murine leukemia viral oncogene homolog tyrosine kinase 1; BCL2 B-cell lymphoma 2 family of antiapoptotic molecules; CD cluster of differentiation antigens; HDAC histone deacetylase; JAK Janus family tyrosine kinases; FLT3 FMS-like tyrosine kinase receptor 3; mTOR mammalian target of rapamycin; NFkB nuclear factor kappa-light-chain enhancer of activated B cells complex of proteins; PI3K phosphoinositide-3-kinase; RTK receptor tyrosine kinase; SRC sarcoma protooncogene family of tyrosine kinases; STAT signal transducer and activator of transcription family of proteins. Stars indicate targeting agents in clinical trials in other diseases but not yet in ALL. Additional details in text
2.9.1 Steroid-Sensitising Adjuvants Steroid resistance may be overcome by antagonists of the mammalian target of rapamycin (mTOR) kinase or by pro-apoptotic small molecules. mTOR inhibitors have shown promise as steroid-sensitising agents, operating through down-regulation of the antiapoptotic BCL2 family molecule, MCL1 [55]. The BCL2 antagonist
32
S. Krishnan et al.
Obatoclax too is able to restore steroid sensitivity but appears to do this by activating autophagic necroptosis and thus bypassing a block in mitochondrial apoptosis [56].
2.9.2 Monoclonal Antibodies to Surface Molecules Surface molecules on the lymphoblast plasma membrane may be targeted using a number of naked and conjugated antibodies. These antibodies typically mediate blast clearance by binding to cognate proteins and activating cellular or non-cellular immune effectors. Alternatively, these antibodies disrupt the function of target molecules (as in the case of antibodies to integrins) or trigger alternative cell death mechanisms (see below).
2.9.3 Kinase Inhibitors As highlighted earlier, dysregulated kinase activity is consistently noted in high-risk ALL. We do not fully understand the mechanisms and molecules responsible for kinase survival signalling in pre-B lymphoblasts. This activation may be constitutive as in the case of activating JAK mutations. Alternatively, aberrant activation may be triggered by homotypic or heterotypic adhesion and maintained by paracrine or autocrine mechanisms [57]. Inhibitors targeting key activated kinases, including the receptor tyrosine kinase, FLT3 and the cytoplasmic kinases, SRC [58] and LYN [59], are now in clinical trials. Similarly, JAK and SYK inhibitors have entered trials for autoimmune and inflammatory diseases but have not yet been tested in childhood ALL. Phosphoinositide-3-kinase (PI3 kinase) inhibitors are gradually entering clinical trials but as this is a large family of kinases, further work is required to clarify the pertinent isoforms in childhood ALL.
2.9.4 Alternative Cell Death Pathway Triggers Exploring non-apoptotic cell-death mechanisms as a therapeutic strategy is in its nascence but holds promise. For instance in mature B-cell neoplasms, type II antibodies directed against surface CD20 molecules trigger cell death by destabilising lysosomes, leading to intracellular lysosomal leak and caspase-independent cell death [60]. Our observations suggest that aberrant lysosome trafficking is a feature of ALL blast cells and targeting lysosomal hydrolases is an approach that merits investigation. Proteasome inhibitors may similarly operate through both apoptotic and non-apoptotic cell death triggers and are discussed in detail in a later chapter.
2 Identifying Targets for New Therapies in Children
33
2.9.5 Others A number of alternative approaches also have the potential to be successful. Relapses in precursor T-lineage ALL (T-ALL) are an especial therapeutic challenge. More than 50% of T-ALL harbour activating NOTCH1 mutations. The enzyme g-secretase catalyses the activating cleavage of the NOTCH1 receptor and g-secretase inhibitors have entered clinical trials. g-Secretase is also required for the maturation of the intestinal mucosa and thus gut toxicity is dose limiting, though there is evidence from a murine model that this may be overcome by the concomitant use of steroids [61].
2.10 Concluding Remarks After decades marked by a dearth of new agents, the recently invigorated drug pipeline is an exciting development in the treatment of childhood ALL. But this presents its own challenges. How do we integrate these new agents within contemporary treatment protocols? How do we optimally investigate these drugs in clinical trials? How do we examine their specific effects in the context of multi-agent chemotherapy? And importantly, how do we make these agents available to resource-constrained populations? There are no easy answers as yet and radical unorthodox approaches are probably necessary. The issue of germline polymorphisms has not been addressed in this chapter or discussed elsewhere in the monograph. These polymorphisms are likely key determinants in eventual treatment response and drug toxicity. Suffice to say, we just do not know enough about the different pathways responsible for the degradation of drugs used in childhood ALL as we have not had the tools to investigate this in detail. With the advent of cheaper germline whole genome sequencing, this is set to change. So, to end from where we started, these are heady times for those of us who look after children with ALL. Among the first to show that a cancer can be cured, we as a community can now proceed to demonstrate how an understanding of the biology of the disease can be harnessed to individualise therapy. This will not only lead to more cures and less toxicity but hopefully cheaper and simpler treatment options that can be applied globally. Acknowledgement This work was supported by a programme grant from Cancer Research UK (VS, AM) and a fellowship from the University of Manchester (SK).
References 1. Conter V, Bartram CR, Valsecchi MG, Schrauder A, Panzer-Grumayer R, Moricke A, et al. Molecular response to treatment redefines all prognostic factors in children and adolescents with B-cell precursor acute lymphoblastic leukemia: results in 3184 patients of the AIEOPBFM ALL 2000 study. Blood. 2010 Feb 12;115(16):3206–14.
34
S. Krishnan et al.
2. Pui CH, Campana D, Pei D, Bowman WP, Sandlund JT, Kaste SC, et al. Treating childhood acute lymphoblastic leukemia without cranial irradiation. N Engl J Med. 2009 Jun 25; 360(26):2730–41. 3. Pui CH. Prophylactic cranial irradiation: going, going, gone. Lancet Oncol. 2009 Oct;10(10): 932–3. 4. Saha V. Simplifying treatment for children with ALL. Lancet. 2007 Jan 13;369(9556):82–3. 5. Silverman LB, Stevenson KE, O’Brien JE, Asselin BL, Barr RD, Clavell L, et al. Long-term results of Dana-Farber Cancer Institute ALL Consortium protocols for children with newly diagnosed acute lymphoblastic leukemia (1985-2000). Leukemia. 2010 Feb;24(2):320–34. 6. Conter V, Arico M, Basso G, Biondi A, Barisone E, Messina C, et al. Long-term results of the Italian Association of Pediatric Hematology and Oncology (AIEOP) Studies 82, 87, 88, 91 and 95 for childhood acute lymphoblastic leukemia. Leukemia. 2010 Feb;24(2):255–64. 7. Gaynon PS, Angiolillo AL, Carroll WL, Nachman JB, Trigg ME, Sather HN, et al. Long-term results of the children’s cancer group studies for childhood acute lymphoblastic leukemia 1983-2002: a Children’s Oncology Group Report. Leukemia. 2010 Feb;24(2):285–97. 8. Salzer WL, Devidas M, Carroll WL, Winick N, Pullen J, Hunger SP, et al. Long-term results of the pediatric oncology group studies for childhood acute lymphoblastic leukemia 1984– 2001: a report from the children’s oncology group. Leukemia. 2010 Feb;24(2):355–70. 9. Moricke A, Zimmermann M, Reiter A, Henze G, Schrauder A, Gadner H, et al. Long-term results of five consecutive trials in childhood acute lymphoblastic leukemia performed by the ALL-BFM study group from 1981 to 2000. Leukemia. 2010 Feb;24(2):265–84. 10. Mitchell C, Richards S, Harrison CJ, Eden T. Long-term follow-up of the United Kingdom medical research council protocols for childhood acute lymphoblastic Leukemia, 1980–2001. Leukemia. 2010 Feb;24(2):406–18. 11. Henze G. Early postinduction intensification therapy is essential in childhood acute lymphoblastic leukemia. Nature Clinical Practice. 2008 Sep;5(9):502–3. 12. Schmiegelow K, Al-Modhwahi I, Andersen MK, Behrendtz M, Forestier E, Hasle H, et al. Methotrexate/6-mercaptopurine maintenance therapy influences the risk of a second malignant neoplasm after childhood acute lymphoblastic leukemia: results from the NOPHO ALL92 study. Blood. 2009 Jun 11;113(24):6077–84. 13. Bader P, Kreyenberg H, Henze GH, Eckert C, Reising M, Willasch A, et al. Prognostic value of minimal residual disease quantification before allogeneic stem-cell transplantation in relapsed childhood acute lymphoblastic leukemia: the ALL-REZ BFM Study Group. J Clin Oncol. 2009 Jan 20;27(3):377–84. 14. Knechtli CJ, Goulden NJ, Hancock JP, Grandage VL, Harris EL, Garland RJ, et al. Minimal residual disease status before allogeneic bone marrow transplantation is an important determinant of successful outcome for children and adolescents with acute lymphoblastic leukemia. Blood. 1998 Dec 1;92(11):4072–9. 15. Ribeiro RC, Pui C-H. Saving the Children -- Improving Childhood Cancer Treatment in Developing Countries. N Engl J Med. 2005 May 26, 2005;352(21):2158–60. 16. Goldie JH, Coldman AJ. The genetic origin of drug resistance in neoplasms: implications for systemic therapy. Cancer Res. 1984 Sep;44(9):3643–53. 17. Ramakers-van Woerden NL, Pieters R, Loonen AH, Hubeek I, van Drunen E, Beverloo HB, et al. TEL/AML1 gene fusion is related to in vitro drug sensitivity for L-asparaginase in childhood acute lymphoblastic leukemia. Blood. 2000 Aug 1;96(3):1094–9. 18. Kager L, Cheok M, Yang W, Zaza G, Cheng Q, Panetta JC, et al. Folate pathway gene expression differs in subtypes of acute lymphoblastic leukemia and influences methotrexate pharmacodynamics. The Journal of Clinical Investigation. 2005 Jan;115(1): 110–7. 19. Krishnan S, Wade R, Moorman AV, Mitchell C, Kinsey SE, Eden TO, et al. Temporal changes in the incidence and pattern of central nervous system relapses in children with acute lymphoblastic Leukemia treated on four consecutive Medical Research Council trials, 1985–2001. Leukemia. 2010 Feb;24(2):450–9.
2 Identifying Targets for New Therapies in Children
35
20. Schultz KR, Bowman WP, Aledo A, Slayton WB, Sather H, Devidas M, et al. Improved early event-free survival with imatinib in Philadelphia chromosome-positive acute lymphoblastic leukemia: a children’s oncology group study. J Clin Oncol. 2009 Nov 1;27(31):5175–81. 21. Roy A, Cargill A, Love S, Moorman AV, Stoneham S, Lim A, et al. Outcome after first relapse in childhood acute lymphoblastic Leukemia - lessons from the United Kingdom R2 trial. Br J Haematol. 2005 Jul;130(1):67–75. 22. Ko RH, Ji L, Barnette P, Bostrom B, Hutchinson R, Raetz E, et al. Outcome of patients treated for relapsed or refractory acute lymphoblastic leukemia: a Therapeutic Advances in Childhood Leukemia Consortium study. J Clin Oncol. 2010 Feb 1;28(4):648–54. 23. Einsiedel HG, von Stackelberg A, Hartmann R, Fengler R, Schrappe M, Janka-Schaub G, et al. Long-term outcome in children with relapsed ALL by risk-stratified salvage therapy: results of trial acute lymphoblastic leukemia-relapse study of the Berlin-Frankfurt-Munster Group 87. J Clin Oncol. 2005 Nov 1;23(31):7942–50. 24. Mullighan CG, Phillips LA, Su X, Ma J, Miller CB, Shurtleff SA, et al. Genomic analysis of the clonal origins of relapsed acute lymphoblastic leukemia. Science. 2008 Nov 28;322(5906):1377–80. 25. le Viseur C, Hotfilder M, Bomken S, Wilson K, Rottgers S, Schrauder A, et al. In childhood acute lymphoblastic leukemia, blasts at different stages of immunophenotypic maturation have stem cell properties. Cancer Cell. 2008 Jul 8;14(1):47–58. 26. Bhojwani D, Kang H, Moskowitz NP, Min DJ, Lee H, Potter JW, et al. Biologic pathways associated with relapse in childhood acute lymphoblastic leukemia: a Children’s Oncology Group study. Blood. 2006 Jul 15;108(2):711–7. 27. Kirschner-Schwabe R, Lottaz C, Todling J, Rhein P, Karawajew L, Eckert C, et al. Expression of late cell cycle genes and an increased proliferative capacity characterize very early relapse of childhood acute lymphoblastic leukemia. Clin Cancer Res. 2006 Aug 1;12(15):4553–61. 28. Staal FJ, de Ridder D, Szczepanski T, Schonewille T, van der Linden EC, van Wering ER, et al. Genome-wide expression analysis of paired diagnosis-relapse samples in ALL indicates involvement of pathways related to DNA replication, cell cycle and DNA repair, independent of immune phenotype. Leukemia. 2010 Mar;24(3):491–9. 29. Ford AM, Fasching K, Panzer-Grumayer ER, Koenig M, Haas OA, Greaves MF. Origins of “late” relapse in childhood acute lymphoblastic leukemia with ETV6-RUNX1 fusion genes. Blood. 2001 Aug 1;98(3):558–64. 30. Zuna J, Ford AM, Peham M, Patel N, Saha V, Eckert C, et al. TEL deletion analysis supports a novel view of relapse in childhood acute lymphoblastic leukemia. Clin Cancer Res. 2004 Aug 15;10(16):5355–60. 31. Hagedorn N, Acquaviva C, Fronkova E, von Stackelberg A, Barth A, zur Stadt U, et al. Submicroscopic bone marrow involvement in isolated extramedullary relapses in childhood acute lymphoblastic leukemia: a more precise definition of “isolated” and its possible clinical implications, a collaborative study of the Resistant Disease Committee of the International BFM study group. Blood. 2007 Dec 1;110(12):4022–9. 32. Raanani P, Shpilberg O, Ben-Bassat I. Extramedullary disease and targeted therapies for hematological malignancies--is the association real? Ann Oncol. 2007 Jan;18(1):7–12. 33. Trevino LR, Shimasaki N, Yang W, Panetta JC, Cheng C, Pei D, et al. Germline genetic variation in an organic anion transporter polypeptide associated with methotrexate pharmacokinetics and clinical effects. J Clin Oncol. 2009 Dec 10;27(35):5972–8. 34. Trevino LR, Yang W, French D, Hunger SP, Carroll WL, Devidas M, et al. Germline genomic variants associated with childhood acute lymphoblastic leukemia. Nat Genet. 2009 Sep;41(9):1001–5. 35. Yang JJ, Cheng C, Yang W, Pei D, Cao X, Fan Y, et al. Genome-wide interrogation of germline genetic variation associated with treatment response in childhood acute lymphoblastic leukemia. Jama. 2009 Jan 28;301(4):393–403. 36. Cheng Q, Yang W, Raimondi SC, Pui CH, Relling MV, Evans WE. Karyotypic abnormalities create discordance of germline genotype and cancer cell phenotypes. Nat Genet. 2005 Aug;37(8):878–82.
36
S. Krishnan et al.
37. Patel N, Krishnan S, Offman MN, Krol M, Moss CX, Leighton C, et al. A dyad of lymphoblastic lysosomal cysteine proteases degrades the antileukemic drug L-asparaginase. The Journal of Clinical Investigation. 2009 Jul;119(7):1964–73. 38. Iwamoto S, Mihara K, Downing JR, Pui CH, Campana D. Mesenchymal cells regulate the response of acute lymphoblastic leukemia cells to asparaginase. J Clin Invest. 2007 Apr;117(4):1049–57. 39. Colmone A, Amorim M, Pontier AL, Wang S, Jablonski E, Sipkins DA. Leukemic cells create bone marrow niches that disrupt the behavior of normal hematopoietic progenitor cells. Science. 2008 Dec 19;322(5909):1861–5. 40. Holleman A, Cheok MH, den Boer ML, Yang W, Veerman AJ, Kazemier KM, et al. Geneexpression patterns in drug-resistant acute lymphoblastic leukemia cells and response to treatment. N Engl J Med. 2004 Aug 5;351(6):533–42. 41. Cario G, Stanulla M, Fine BM, Teuffel O, Neuhoff NV, Schrauder A, et al. Distinct gene expression profiles determine molecular treatment response in childhood acute lymphoblastic leukemia. Blood. 2005 Jan 15;105(2):821–6. 42. Kang H, Chen IM, Wilson CS, Bedrick EJ, Harvey RC, Atlas SR, et al. Gene expression classifiers for relapse-free survival and minimal residual disease improve risk classification and outcome prediction in pediatric B-precursor acute lymphoblastic leukemia. Blood. 2010 Feb 18;115(7):1394–405. 43. Harvey RC, Mullighan CG, Chen IM, Wharton W, Mikhail FM, Carroll AJ, et al. Rearrangement of CRLF2 is associated with mutation of JAK kinases, alteration of IKZF1, Hispanic/Latino ethnicity, and a poor outcome in pediatric B-progenitor acute lymphoblastic leukemia. Blood. 2010 Feb 4;115(26):5312–21. 44. Yoda A, Yoda Y, Chiaretti S, Bar-Natan M, Mani K, Rodig SJ, et al. Functional screening identifies CRLF2 in precursor B-cell acute lymphoblastic leukemia. Proc Natl Acad Sci USA. 2010 Jan 5;107(1):252–7. 45. Mullighan CG, Collins-Underwood JR, Phillips LA, Loudin MG, Liu W, Zhang J, et al. Rearrangement of CRLF2 in B-progenitor- and Down syndrome-associated acute lymphoblastic leukemia. Nat Genet. 2009 Nov;41(11):1243–6. 4 6. Russell LJ, Capasso M, Vater I, Akasaka T, Bernard OA, Calasanz MJ, et al. Deregulated expression of cytokine receptor gene, CRLF2, is involved in lymphoid transformation in B-cell precursor acute lymphoblastic leukemia. Blood. 2009 Sep 24;114(13):2688–98. 47. Mullighan CG, Su X, Zhang J, Radtke I, Phillips LA, Miller CB, et al. Deletion of IKZF1 and prognosis in acute lymphoblastic leukemia. N Engl J Med. 2009 Jan 29;360(5):470–80. 48. Den Boer ML, van Slegtenhorst M, De Menezes RX, Cheok MH, Buijs-Gladdines JG, Peters ST, et al. A subtype of childhood acute lymphoblastic Leukemia with poor treatment outcome: a genome-wide classification study. Lancet Oncol. 2009 Feb;10(2):125–34. 49. Mullighan CG, Zhang J, Harvey RC, Collins-Underwood JR, Schulman BA, Phillips LA, et al. JAK mutations in high-risk childhood acute lymphoblastic leukemia. Proc Natl Acad Sci U S A. 2009 Jun 9;106(23):9414–8. 50. Nervi B, Ramirez P, Rettig MP, Uy GL, Holt MS, Ritchey JK, et al. Chemosensitization of acute myeloid leukemia (AML) following mobilization by the CXCR4 antagonist AMD3100. Blood. 2009 Jun 11;113(24):6206–14. 51. Moore AS, Blagg J, Linardopoulos S, Pearson AD. Aurora kinase inhibitors: novel small molecules with promising activity in acute myeloid and Philadelphia-positive leukemias. Leukemia. 2010 Feb 11;24(4):671–8. 52. Yanez L, Bermudez A, Richard C, Bureo E, Iriondo A. Successful induction therapy with decitabine in refractory childhood acute lymphoblastic leukemia. Leukemia. 2009 Jul;23(7):1342–3. 53. Sayin DB, Kurekci E, Karabulut HG, Ezer U, Bokesoy I. DNA methyltransferase expression differs with proliferation in childhood acute lymphoblastic leukemia. Mol Biol Rep. 2009 Sep 10;37(5):2471–6.
2 Identifying Targets for New Therapies in Children
37
54. Horton TM, Thompson PA, Berg SL, Adamson PC, Ingle AM, Dolan ME, et al. Phase I pharmacokinetic and pharmacodynamic study of temozolomide in pediatric patients with refractory or recurrent leukemia: a Children’s Oncology Group Study. J Clin Oncol. 2007 Nov 1;25(31):4922–8. 55. Wei G, Twomey D, Lamb J, Schlis K, Agarwal J, Stam RW, et al. Gene expression-based chemical genomics identifies rapamycin as a modulator of MCL1 and glucocorticoid resistance. Cancer Cell. 2006 Oct;10(4):331–42. 56. Bonapace L, Bornhauser BC, Schmitz M, Cario G, Ziegler U, Niggli FK, et al. Induction of autophagy-dependent necroptosis is required for childhood acute lymphoblastic leukemia cells to overcome glucocorticoid resistance. J Clin Invest. 2010 Mar 1;120(4):1310–23. 57. Harwood NE, Batista FD. Early events in B cell activation. Annu Rev Immunol. 2010 Apr 23;28:185–210. 58. Li S. Src-family kinases in the development and therapy of Philadelphia chromosome-positive chronic myeloid leukemia and acute lymphoblastic leukemia. Leuk Lymphoma. 2008 Jan;49(1):19–26. 59. Kantarjian H, Coutre PL, Cortes J, Pinilla-Ibarz J, Nagler A, Hochhaus A, et al. Phase 1 study of INNO-406, a dual Abl/Lyn kinase inhibitor, in Philadelphia chromosome-positive leukemias after imatinib resistance or intolerance. Cancer. 2010 Mar 22;116(11):2665–72. 60. Ivanov A, Beers SA, Walshe CA, Honeychurch J, Alduaij W, Cox KL, et al. Monoclonal antibodies directed to CD20 and HLA-DR can elicit homotypic adhesion followed by lysosome-mediated cell death in human lymphoma and leukemia cells. J Clin Invest. 2009 Aug;119(8):2143–59. 61. Real PJ, Tosello V, Palomero T, Castillo M, Hernando E, de Stanchina E, et al. Gammasecretase inhibitors reverse glucocorticoid resistance in T cell acute lymphoblastic leukemia. Nat Med. 2009 Jan;15(1):50–8.
Chapter 3
Preclinical Evaluation Barbara Szymanska, Hernan Carol*, and Richard B. Lock
3.1 Introduction In contrast to paediatric disease, ALL in adults is not only relatively rare comprising approximately 20% of all adult leukemias, but is also much more resistant to currently available treatments [1]. At present, the first complete remission rates (80–90%) attained in adult ALL approach those achieved in paediatric cases [2]. Nevertheless, only 30–40% of adult patients achieve long-term disease-free survival, and new chemotherapeutic agents and novel approaches to treatment must be developed for adults with ALL in order to improve the relatively low cure rates. Although the incidence of paediatric ALL is close to twofold higher than that in adults [3], the high cure rates achieved with standard therapy limit the number of paediatric patients available for clinical trials of novel drugs. Therefore, validated and predictive preclinical models of ALL are likely to become essential to prioritise new drugs for clinical trials if a significant improvement in survival and a reduction in the long-term side effects of conventional therapy are to be achieved. It is an open question whether preclinical testing in ALL will, in fact, predict how a drug performs in the clinic. In the past, preclinical models have been criticised for overestimating a drug’s efficacy, leading to disappointing clinical activity. It is timely, therefore, to critically review the attributes of currently available experimental models, to attempt to understand the reasons for past failures, and to summarise contemporary efforts aimed at the systematic preclinical evaluation of new drugs for the treatment of ALL. While the challenges to develop better treatments for both
* Barbara Szymanska and Hernan Carol contributed equally to this chapter. R.B. Lock (*) Leukemia Biology Program, Children’s Cancer Institute Australia for Medical Research, Lowy Cancer Research Centre, UNSW PO Box 81, Randwick, NSW 2031, Australia and University of New South Wales, Sydney, NSW 2031, Australia e-mail:
[email protected] V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3_3, © Springer Science+Business Media, LLC 2011
39
40
B. Szymanska et al.
paediatric and adult ALL remain considerable, validated and predictive preclinical models are likely to play an essential role in meeting these challenges.
3.2 Historical Perspective Initial anti-cancer drug screening and discovery programs were more empirical in nature, much smaller in scale and proceeded slower compared to contemporary projects. They utilised poorly characterised tumour models, which typically involved mice bearing rapidly growing murine leukemias (e.g. L1210 and P388) against which randomly produced molecules with insufficiently defined modes of action were tested [4, 5]. The foundation for progressing with development of a particular anti-cancer drug was based on its ability to inhibit tumour cell proliferation in vitro and/or in vivo without prior delineation of its molecular or cellular targets. Over the past four decades, the substantial progress made in our understanding of mechanisms of tumour development, progression, and resistance to therapy has led to a new era of rational drug development, in which molecules are screened, or designed, for interaction with specific targets [6, 7]. This new dawn has resulted in an unprecedented expansion in the number of new therapeutic agents with potential clinical activity.
3.2.1 Success and Failure: l-Asparaginase and Cisplatin Prior to the initiation of the anti-cancer screening program by the National Cancer Institute (NCI, USA) in 1955, the development of chemotherapeutic agents was often fortuitous, as illustrated by the steps that guided the discovery of l-asparaginase (l-ASNase) and cisplatin [cisplatinum, cis-diamminedichloridoplatinum(II), CDDP]. In 1953, while investigating immunotherapy for lymphoma, John Kidd used guinea pig serum as a source of complement. He fortuitously discovered that the growth of subcutaneously transplanted lymphoma cell lines in mice and rats was inhibited by intraperitoneal administration of guinea pig serum alone [8]. This effect was species-specific, as treatment with sera from other sources, such as rabbit or horse, did not have the same effect. Furthermore, the effect of guinea pig serum was confined to lymphomas, with mammary carcinomas and fibrosarcomas unaffected [8]. From this original observation it took another 8 years for the active component of the serum to be identified as l-ASNase [9]. During the same period it was reported that murine leukemia cell lines required l-asparagine to sustain their growth in vitro [10]. Subsequently, l-ASNase isolated from Escherichia coli was shown to be as effective as guinea pig serum, which provided a more readily available source of the enzyme [11]. Data from the first clinical trials using partially purified enzyme from guinea pig serum [12] or E. coli derived l-ASNase [13, 14] were published 5–6 years after its identification. Following successful clinical trials
3 Preclinical Evaluation
41
l-ASNase was incorporated into paediatric ALL protocols, which contributed to a remarkable improvement in outcome [15]. Cisplatin and other platinum based compounds are examples of drugs that are not effective in ALL, although they are a main component of many solid tumour treatment regimens. They were discovered in the early 1960s by Rosenberg et al. [16] as a fortuitous observation of the alteration of the growth pattern of bacterial cultures. In vitro anti-cancer activity was later described against the Sarcoma 180 and the L1210 (leukemia) cell lines [17], but clinical trials with cisplatin reported high levels of renal and gastrointestinal toxicity. Cisplatin was tested in the 1970s in children with cancer not responding to conventional chemotherapeutic drugs [18, 19]. No responses in children with ALL were observed, in contrast to some individuals with solid tumours. Retrospective experimental evaluation of cisplatin has confirmed little evidence for activity against paediatric ALL (see Sect. 3.5 and [20]).
3.2.2 The NCI In Vivo and In Vitro Panels Following reports that demonstrated correlations between the efficacy of compounds against transplanted tumours and their clinical activity (reviewed in [21]), the NCI (USA) initiated a large-scale anti-cancer drug screening program in the mid 1950s, which was based on syngeneic transplantable murine leukemias. Originally only rodent models were used and the screening program included the L1210 leukemia cell line. Later, another murine leukemia cell line (P388) was included [22]. It was not until 1975 that the P388 model was selected as the initial screen, followed by a panel of transplantable tumours. In the 1980s it became evident that these early models had a limited ability to predict activity in humans, and they predominantly identified the DNA-damaging classes of anti-cancer agents [23]. Activity against a range of tumour histiotypes, rather than specific to a certain cancer, was also considered of high priority, possibly in the hope of identifying the anticancer “magic bullet”. With the view of improving the outcome of drugs with potential activity for solid tumour treatment, early in the 90s this pre-screening was substituted by a panel of 60 cell lines cultured in vitro (the NCI60 panel) [24]. For a comprehensive recent review on the gestation of and a recount of the main findings by the NCI60 human tumour cell line drug screening see Shoemaker [25]. While of obvious utility, the validity of this screening program has been questioned mainly due to the fact that the in vitro assay used is based on inhibition of cell growth and not on cell kill, which is unlikely to differentiate between cytostatic and cytotoxic drugs [26]. As a consequence of the need to summarise the output of complex experiments and large sets of data the “COMPARE” format of reporting was created for the NCI60 panel [27]. In this way a unique profile of activity can be generated for each drug. It was also discovered that drugs with similar mechanisms of action or similar targets showed similar COMPARE profiles [28], thereby enabling preliminary classification of new drugs with unknown mechanisms of action [29].
42
B. Szymanska et al.
3.3 Evaluation Criteria for Robust Preclinical Models Currently, the pathway of anti-cancer drug development is lengthy and costly, taking on average 8–10 years from the initial hypothesis to completion of Phase III clinical trials and to gaining regulatory approval, with an estimated cost of US$0.8–2.0 billion per drug [30, 31]. Furthermore, the number of anti-cancer agents that fail in the clinic far outweighs those that prove effective. During 1991–2000, the attrition rate of potential anti-cancer drugs that entered clinical testing was very high, with only 5% of agents gaining US FDA approval [32]. While the reasons for drug failure in the clinic are multifactorial, the increased use of predictive preclinical evaluation measures may significantly reduce this alarming rate of attrition. The value of any preclinical model will ultimately depend on its ability to accurately predict clinical response, and models, by definition, only approximate the clinical setting. Nevertheless, we consider that the following criteria should be among those desired when selecting a model system in which to evaluate any new anti-leukaemic drug: 1. Homologous cancer biology. Does the model accurately reflect the biology of human cancer and recapitulate the molecular and cellular events associated with the particular tumour histiotype? 2. Heterogeneity. ALL is biologically, genetically, phenotypically, and clinically heterogenous disease; a relevant model should reflect this heterogeneity. 3. Penetrance/take rate/reproducibility. Most, or preferably all, animals in an experimental cohort should develop cancer in a timely and reproducible fashion. 4. Disease monitoring. A model should be amenable to routine, reliable, reproducible, and cost effective methods of monitoring disease progression and drug responses. 5. Stringent endpoints for determining the effects of therapy should be clearly defined. 6. Responses to established drugs should reflect those encountered in patients. 7. Pharmacokinetics and pharmacodynamics. Differences in these parameters between the selected species/strain and humans should be accommodated, since increased host tolerance leads to overestimation of drug efficacy, and viceversa. The following section briefly compares attributes of some of the models available for preclinical drug testing.
3.4 Preclinical Models of ALL Several experimental models of ALL are available for preclinical testing of new drugs, and they have been extensively analysed and compared in several articles and reviews [33–35]. Each model has advantages and limitations, and suffers from unknown predictive power of clinical response. Two recent meta-analyses evaluated
3 Preclinical Evaluation
43
the predictive accuracy of preclinical testing: Johnson et al. [36] analysed the predictive power of xenograft models retrospectively for a set of 39 compounds which had both Phase II clinical trials and in vivo xenotransplantation data. It was found that those compounds that exhibited activity in at least one third of the xenograft models were more likely to have positive Phase II clinical trial outcomes. A similar study of results obtained by the Canadian NCI [37] found that xenograft models were predictive of Phase II clinical trial outcome in non-small-cell lung cancer and ovarian cancer but not in colon and breast cancer. A similar analysis has yet to be carried out for ALL.
3.4.1 In Vitro Experimental Models Cell lines established from ALL patient biopsy specimens have become essential tools to study many aspects of leukemia, and it is difficult to deny their relevance to cancer research [38]. Nevertheless, while cell lines are useful for a multitude of studies, it is well known that leukemia cells selected for autonomous in vitro growth have sustained additional genetic abnormalities not present in the primary disease state, p53 and Bax mutations being obvious examples [39, 40]. Moreover, it is difficult to reconcile dynamic in vivo drug exposures with the more static situation in vitro, although adjustments to in vitro culture conditions to achieve higher correlations with the results observed in vivo have been proposed [41]. The greatest utility for ALL cell lines in preclinical drug testing may lie in their use for the rapid identification of synergistic drug combinations, which can then be further validated using in vivo models.
3.4.2 In Vivo Experimental Models Autochthonous (“originating in the place where found”) models of cancer include spontaneously occurring neoplasms or tumours induced using a variety of viral or chemical carcinogens in rodents and other species of animals [42–44]. In general, these models have proven valuable in studies of carcinogen-induce tumorigenesis [45] and chemoprevention [43], although for the most part they have been replaced by genetically engineered models. With the advances in technologies that allow manipulation of the genome it has become almost routine to model human cancers via transgenic or knockout animals (e.g. via aberrant activation of an oncogene, or conditional knockout of a tumour suppressor gene) [33, 46]. Such models have contributed significantly to our understanding of cancer [47]. These models are thus well suited for testing targeted therapies and “proof-of-principle” research, in which novel agents are directed at specific molecular targets within a well-characterised biochemical pathway. Furthermore, these genetically engineered strains are immunocompetent and can therefore be used for preclinical evaluation of immunomodulating agents.
44
B. Szymanska et al.
Transgenic mouse models of haematological malignancies have also been generated, such as Em-myc mice, which develop aggressive leukemias and lymphomas when the c-myc gene is placed under the control of the immunoglobulin heavy chain enhancer (Em) [48, 49]. These types of models are useful for preclinical drug testing in leukemia, and deficiencies associated with variable penetrance and rates of cancer development can be overcome by transplantation of primary tumours into syngeneic wild-type littermates [50]. In a novel approach, several transgenic zebrafish models of ALL have been developed [51–54]. The work that led to the establishment of the transgenic model of ALL in zebrafish was pioneered by Langenau et al. [53] who over-expressed mouse c-myc fused to enhanced green fluorescent protein (EGFP) under the control of the rag2 zebrafish promoter. The transparency of zebrafish tissues allows direct imaging of cancer progression, in real time, making this model useful for studying tumorigenesis. The zebrafish has been widely utilised for assessing drug toxicity. There is some evidence of similar toxicity profiles to those of mammalian species [55] and it could be adapted for preclinical testing of chemotherapeutics. The development of human cancer xenograft models in immune-deficient mice represents a milestone in the progression towards more clinically relevant experimental models of cancer. The discovery of the nude (nu/nu), athymic mouse strain [56] and the subsequent successful propagation of human cancer cells in these mice [57] were instrumental in this process. Nude mice lack the Foxn1 gene and are unable to generate functional T cells [58]. However, in contrast to solid tumours, human ALL only engrafted with relatively low efficiency in nude mice even when injected intraperitoneally following irradiation [59], and residual immunity negatively affected tumour take rates [60, 61]. The discovery of the severe-combined immunodeficient (SCID) mouse strain, which lacks functional lymphocytes of both T and B lineage [62, 63], dramatically improved xenograft models of human ALL [64, 65]. Leukemia cell lines and patient biopsy samples inoculated intravenously into SCID mice manifested as systemic disease with human cells detectable in peripheral blood, spleen, and bone marrow [64]. Despite the success with using SCID mice, this strain does retain some ability to reject xenografts via their natural killer (NK), NK-T cells and macrophages, and complement system activity [66]. Immunosuppression with irradiation or drugs can improve take rates of tumours but only transiently as immunity recovers with time, and this can undermine the validity of drug efficacy studies [66, 67]. A more severely immunocompromised strain was generated by crossing SCID mice with the non-obese diabetic (NOD) strain to generate NOD/SCID [67, 68]. NOD/SCIDs have lower NK and macrophage activity than SCIDs, and they also have impaired complement system activation due to lack of C5. Human spleen and peripheral blood mononuclear cells engraft more efficiently in NOD/SCID than SCID mice [68]. Additional strains of immune-deficient mice have been generated with the aim of further improving engraftment of human haematopoietic cells, including NOD/SCID/b2mnull [69, 70] Rag2null/γcnull [71] and NOD/SCID/γcnull [72], all of which appear to be more deficient in NK cell activity than NOD/SCID.
3 Preclinical Evaluation
45
Initially, ALL xenografts were established by subcutaneous inoculation of cell lines or patient samples [61, 73–75]. While ectopic xenograft models have the advantage of relatively fast growth and accessibility, the site of tumour growth represents neither the primary nor the metastatic site of ALL, nor does it provide the appropriate microenvironmental niche for ALL cells, and therefore it may have little resemblance to the clinical disease. Most of these problems appear to be overcome by the development of orthotopic xenotransplantation models, which involve intravenous inoculation, leukemia cell homing to the bone marrow, and eventual infiltration of haematolymphoid organs [76, 77]. Orthotopic ALL xenograft models utilising patient biopsy specimens established in NOD/SCID mice fulfil most of the desirable criteria of preclinical models listed in Sect. 3.3, including that relating to disease heterogeneity when panels of xenografts are used [76, 77]. Moreover, monitoring of ALL progression and response to therapy is straightforward, and endpoints for determining drug efficacy can be clearly defined [76–78]. Nevertheless, limitations of these models include: intraspecies differences in the bone marrow microenvironment; that immune-based therapies can only be partially evaluated in immune-deficient hosts; and that the pharmacokinetics of a particular drug may differ between mice and humans.
3.5 Application of Preclinical Models of ALL for Drug Testing With the recent explosion in the number of new drugs that have the potential to show clinical activity against ALL, a central purpose of preclinical testing should be to prioritise those entities for the limited number of clinical trials that can be carried out. The aim would be to eventually incorporate a new drug into existing therapies to further improve their efficacy, minimise the likelihood of relapse, and limit toxicity. The following recommendations may assist in the process of ranking new drugs: 1. Stringent criteria, analogous to the clinical setting, should be developed to assess anti-leukaemic efficacy, and not be limited to in vitro cell killing or delay of disease progression in vivo. 2. Single-agent efficacy should be observed across several experimental models that represent the heterogeneity of the disease. Where a single ALL subtype is targeted, multiple models representative of that subtype should also be tested. 3. Single-agent activity should be observed over a broad concentration range (³5fold), and not only at the maximum tolerated dose (MTD) for in vivo studies, and the concentration × time (C × T) of drug exposure at which in vitro or in vivo anti-leukaemic effects are observed should reflect those achievable in the plasma of humans. 4. Only novel drugs with ³50% complete response (CR) rates across several models should be considered for further testing [79].
46
B. Szymanska et al.
5. While cell lines studies are a very relevant component of the drug development process, their role in preclinical testing may be more suitable to rapid screening of lead compounds and in vitro identification of synergistic drug combinations for subsequent in vivo verification. 6. The properties of in vitro and in vivo synergistic anti-leukaemic activity, but not normal tissue toxicity, with established drugs used in the treatment of ALL are highly desirable. The following sections review several of the novel drugs that have been subjected to preclinical testing using various ALL models, as single agents and in combination with established drugs. Each section culminates in a description of contemporary approaches to systematically prioritise new drugs for clinical trials in ALL.
3.5.1 Preclinical Testing of Drugs as Single Agents Over the past two decades, diverse in vivo models of leukemia have been used to highlight the efficacy of novel chemotherapeutic agents. A large proportion of these models involved leukemia cell lines engrafted either subcutaneously [74, 75, 80, 81], intraperitoneally [82], or intravenously [74, 83–87] into SCID mice, or occasionally nude mice [81]. Using these models the efficacy of a long list of drugs has been demonstrated, including: desoxyepothilone B [80], nucleoside analogues [82], topoisomerase I inhibitors [85, 86], a proteasome inhibitor [83], aminophylline [84], flavopiridol [74], laminin-derived peptides [75], and monoclonal antibodies against CD47 [87]. In vivo ALL models based on mouse cell lines inoculated intraperitoneally into BALB/c mice [88] or human cell lines transplanted intravenously into SCID mice, were also used to demonstrate the efficacy of a number of immunotoxin conjugates, which included anti-CD19-pokeweed antiviral protein [88–90], anti-CD19-genistein [91], and anti-CD7-pokeweed antiviral protein [92]. Recent studies more frequently used primary ALL patient samples, rather than cell lines, engrafted into NOD/SCID mice [93–96]. Treatment with the thalidomide analogue CC-4047 reduced tumour volume in mice engrafted with primary human ALL cells as a subcutaneous model, and decreased dissemination into the spleen of a systemic model [93]. Treatment with an anti-FLT3 monoclonal antibody increased the event-free survival (EFS) of mice engrafted with an ALL cell line, as well as decreased the engraftment of patient biopsy samples in an orthotopic model of the disease [94]. Transgenic mouse models have also been used to test various chemotherapeutic agents. The farnesyl transferase inhibitor SCH66336 and a FLT3 tyrosine kinase inhibitor demonstrated single agent activity in Bcr/Abl-positive P190 transgenic mice and transgenic mice with a constitutively active FLT3 internal tandem duplication, respectively [97, 98]. Em-ret transgenic mice, which constitutively express activated RET tyrosine kinase in B-lineage precursors and develop ALL between 3 and 8 months of life, responded to the mTOR inhibitor rapamycin [99].
3 Preclinical Evaluation
47
Taken together, the results cited above give the impression that a sufficient number of new drugs have been identified to successfully treat and cure all patients with ALL. This is clearly not the case, possibly due to the necessity to interpret the results in the context of the recommendations set out in the previous section. In fact, based on the attrition rate of drugs that make it to clinical trials, <10% of the drugs that showed activity in preclinical models, will produce results in the clinic good enough for them to be advanced. This success rate clearly needs improvement, particularly for the treatment of paediatric ALL in which the bar is already set at an extremely high level. 3.5.1.1 The Paediatric Preclinical Testing Program One example of a systematic approach to prioritise new drugs for clinical trials in children with cancer is the Pediatric Preclinical Testing Program (PPTP), which was established by the NCI in 2003 [78]. The principal aim of the PPTP is to use stringent objective response criteria to assess the anti-cancer activity of new drugs against panels of molecularly characterised xenografts and cell lines that are representative of the most common paediatric malignancies [78, 100]. The PPTP ALL component consists of a panel of ten xenografts for in vivo experimentation that are representative of heterogeneous disease subtypes, and a panel of representative cell lines for in vitro studies [78]. For a comprehensive description of the methodology used see Houghton et al. [78]. A schematic representation of the types of response curves used by PPTP to generate the objective response measures is provided in Fig. 3.1.
Fig. 3.1 Schematic representation of six classifications of ALL xenograft response defined by the Pediatric Preclinical Testing Program (PPTP). Each line represents the theoretical proportions of human CD45+ cells in murine peripheral blood over time, and is colour-coded according to each objective response score. For additional details see [78]
48
B. Szymanska et al.
In the case of the ALL xenografts, levels of engraftment are followed by flow cytometry to assess the percentage of human CD45+ cells in the murine peripheral blood until a level of 25% (event) is reached. Weekly samples are analysed, and the EFS of individual mice is measured. Also, individual animals are scored based on Objective Response measures, modelled after the clinical setting (Fig. 3.1). Thus, if there is no containment of leukemia progression, or only a moderate delay, a score of progressive disease (PD) or stable disease (SD) is assigned. Alternatively, a definite regression in leukaemic burden constitutes an Objective Response, resulting in scores of partial response (PR), complete response (CR), and maintained complete response (MCR). Drugs to be tested are selected on a consultative basis, although individual drugs are tested in a coded fashion, with the identity of drugs only being unblinded once testing is complete. As an initial triage each drug is tested at its MTD in NOD/SCID mice, and only those drugs that exhibit ³50% objective response rates across the ALL xenograft panel are considered for additional testing. Using this methodology over 20 novel drugs and 4 established drugs have been tested, including ABT-263 [101], alvespamicin (17-DMAG) [102], aplidin [103], AZD2171 [104], AZD6244, bortezomib [105], dasatinib [106], ispinesib [107], lapatinib [108], MLN8237, rapamicin [109], SAR3419, sorafenib, sunitinib [110], vorinostat (SAHA) [111], SCH717454 (19D12) [112], and the 4 established drugs vincristine, cyclophosphamide, cisplatin and topotecan [20, 78]. Results of this testing are summarised in Fig. 3.2 and Table 3.1. A “COMPARE like” format has been adopted for graphic display to summarise results and make them immediately comparable between different drugs [78]. Also a “heat map” format is used to display the Objective Response measure (from green indicating not effective, PD, to red indicating very effective, MCR) for each drug across all xenografts tested [78]. Figure 3.2 shows a collated “heat map” with the results of the ALL xenograft testing. The results reported for the four established drugs used as validation confirmed that the xenografts performed as expected, in that they responded very well to vincristine and cyclophosphamide, but not to cisplatin. The results with topotecan were not all together surprising, since this drug has shown some activity against ALL in clinical trials [113, 114]. Moreover, the evaluation of cisplatin has allowed us to establish a threshold of activity under which a drug is deemed not useful for any additional testing. Figure 3.2 and Table 3.1 also indicate that only 5 out of 16 new drugs elicited ³50% objective response rate across the panel of xenografts, a success rate of less than one in three using these stringent criteria. In contrast, a greater proportion (8/16) of drugs significantly prolonged mouse EFS, suggesting that measurements of progression delay may overestimate anti-leukaemic activity leading to failure in the clinic. It is worth noting that cisplatin produced significant growth delays in 50% of the xenografts tested, while it did not generate a single objective response. A similar case is seen for alvespamicin and lapatinib. Other results that stress the difference in relevance of the two methods of assessment are
3 Preclinical Evaluation
49
Fig. 3.2 “Heat map” representation of the objective response measures for 16 experimental drugs (blue shading) and 4 established drugs (grey shading) tested against the PPTP ALL xenograft panel since 2005. The designation of each xenograft is indicated above the heat map for B-cell precursor (yellow) or T-lineage (blue) ALL. The median objective response score of each xenograft is indicated as: Progressive Disease, green; Stable Disease, grey; Partial Response, yellow; Complete Response, orange; Maintained Complete Response (MCR), red; Technical Failure, black. The column on the right indicates the number of xenografts that achieved an Objective Response (OR) (PR, CR or MCR) versus the total number tested (N)
those of dasatinib. This drug only generated an objective response in the xenograft carrying a BCR-ABL translocation (ALL-4), against which the drug is selectively active [115], while two additional xenografts experienced a significant growth delay LGD. The success rate of less than one in three suggests that: (1) active drugs identified in this fashion have real potential to show clinical activity against ALL; and (2) inactive drugs may be prevented from entering clinical trials and thereby not delay the evaluation of potentially active drugs. The first possibility has the opportunity of being tested since MLN8237, an Aurora A kinase inhibitor tested by PPTP in 2007 (Fig. 3.2 and Table 3.1), was fast-tracked by the Children’s Oncology Group and the NCI into a Phase I/II clinical trial for relapsed/refractory paediatric ALL that was opened to accrual in August 2008. Those novel drugs that show ³50% objective response rates are prioritised for testing over a range of doses, as well as in combination with established drugs, consistent with the recommendations in Sect. 3.5.
Table 3.1 Comparison of evaluation methods for drugs tested as single agents by the PPTP against ALL xenografts Response ratea Drug Target Event-free survival (EFS) OR measure References Established drugs Vincristine Microtubules 100.0 100.0 [78] Cyclophosphamide DNA 100.0 100.0 [78] Cisplatin DNA 50.0 0.0 [20] Topotecan DNA 100.0 87.5 Unpublishedb Investigational drugs Bortezomib Proteasome 57.0 57.0 [105] Alvespamycin HSP90 83.0 0.0 [102] Dasatinib BCR/ABL and Src family tyrosine kinase (TK) 43.0 14.5 [106] AZD2171 VEGF receptor TK 0.0 0.0 [104] Ispinesib Kinesin spindle protein 100.0 66.7 [107] Sunitinib Multiple receptor TK 37.5 12.5 [110] Rapamycin mTOR 62.5 25.0 [109] Lapatinib EGFR/HER2 100.0 0.0 [108] ABT-263 BCl-2 family 82.5 50.0 [101] 19D12 IGF-I receptor 25.0 0.0 [112] Vorinostat Histone deacetylase 0.0 0.0 [111] AZD6244 MEK 1/2 0.0 0.0 Unpublished b MLN8237 Aurora A kinase 100.0 100.0 Unpublishedb Aplidin Unknown 37.5 12.5 [103] Sorafenib Protein kinase 12.5 0.0 Unpublishedb SAR3419 CD19 87.5 62.5 Unpublishedb a Comparison of response rates expressed as the percentage of xenografts versus total number of xenografts tested that exhibited either a significant increase in mouse EFS compared to vehicle controls, or an objective response (PR, CR, or MCR). Numbers in bold face denote drugs with ³ 50% response rates b Copies of conference presentations available at http://pptp.stjude.org/index.php
50 B. Szymanska et al.
3 Preclinical Evaluation
51
3.5.2 Preclinical Testing of Drugs in Combinations While the methodology for detecting synergy in vitro is well established [116, 117], and recently reviewed in [118], synergistic drug effects are more difficult to assess in vivo and have been the subject of several excellent articles [119–121]. Therapeutic synergy at its most stringent definition is “a therapeutic effect achieved with a tolerated regimen of a combination treatment that exceeds the optimal effect achieved at any tolerated dose of monotherapy associated with the same drugs used in the combination” [122]. It would be desirable to test drugs in vivo over a range of doses as well as at different ratios, but this is seldom practical. For example, to test a combination of three drugs at three different doses, the matrix of different experimental groups is not only too large to handle in practice, but results would be extremely hard to analyse and extract reliable conclusions. Moreover, this problem is only secondary to the fact that for most drugs there is scarce or no information on how the pharmacological parameters in the human compare to those of the species in which the in vivo model is set up (discussed below in Sect. 3.6). Therefore, various compromises have to be made to optimise workload, costs and feasibility. A meeting in 2003 tried to address these issues for cancer preclinical drug combination testing in general [123]. Its main recommendations were that the selection criteria for clinical evaluation of drug combinations included: (a) significant synergy in vitro; (b) synergy across a spectrum of tumour types; (c) activity against primary human tumour cells; (d) reasonable mechanism of action; and (e) in vivo evidence of efficacy. A brief review of the literature reveals numerous examples of preclinical in vivo synergistic interactions between novel and established drugs in various experimental models of ALL [50, 81, 96, 124–127]. In many instances these studies suffer from the same limitations cited above for single agent studies, in that it is not possible to test the generality of the effect due to limitations in resources, but more often than not the combination is compared to equal (rather than optimal, see definition of synergy above) doses of each single agent when tested for therapeutic synergy. Due to in vivo combination experiments being particularly “resource heavy”, it would appear appropriate to return to simpler systems, such as in vitro testing of cell lines or rational design of combinations based on known mechanisms of action or molecular targets affected, in order to prioritise subsequent in vivo experiments. The use of animal models for final verification of synergy would then appear appropriate, using the most stringent definitions of synergy, and with combinations tested against several in vivo models. A question frequently encountered is that if the aim of the PPTP is to eventually use novel drugs in the clinic in combination with established drugs, and the first phase of prioritising novel drugs is as single agents, then isn’t there a distinct possibility of eliminating certain drugs from further testing that on their own are inactive but in combination may be highly synergistic? This question raises interesting philosophical and ethical questions, probably the most important of which is what preclinical evidence would be required to convince paediatric and adult oncologists to administer this “inactive” drug in good faith that the preclinical synergism will be replicated in humans?
52
B. Szymanska et al.
For a program with the magnitude of throughput such as the PPTP it is simply not feasible to test every new drug in combination in vivo. Only the most promising single agents are pursued in stringent second stage multiple dosing and combination studies, to identify in a preclinical setting the most effective combinations of a novel active drug with established drugs which, by definition, have demonstrated clinical efficacy. Therefore, in terms of the PPTP the above question remains unanswered. However, this paradigm is likely to change with the advent of target-based drug design. As we move towards more refined and specific targeting of aberrant cellular processes, a drug may only be an efficient cytotoxic in combination with another drug. One or more alternative or rescue pathways may have to be simultaneously blocked or inhibited (the so called “concentric model”), or cytotoxic reinforcement could result from pharmacologic intervention of several components linked in one pathway (the “linear model”) [128]. Therefore, there is a risk that we could dismiss valuable drugs after testing them as single agents, which could be active in combinations. One such example is the g-secretase inhibitor Compound E, which targets NOTCH1 activation, a very promising target for T-lineage ALL [129]. While Compound E produced unacceptably high gastrointestinal toxicity when used as a single agent, when combined with dexamethasone it resulted in better than expected efficacy, both by means of reducing toxicity in enterocytes and possibly through improving glucocorticoid receptor function in resistant cells. A second example is the therapeutic potential of a novel BH3-mimetic ABT737, which binds competitively to anti-apoptotic Bcl-2 proteins (Bcl-2, Bcl-xL and Bcl-w) [6]. While we have shown that the orally available analogue of ABT-737 for clinical use, ABT-263, exhibits good activity against a panel of paediatric ALL xenografts [101], its single agent efficacy is less impressive than the established drugs vincristine and cyclophosphamide [78] (Fig. 3.2 and Table 3.1). Nevertheless, we have also shown that ABT-737 augments the in vivo efficacy of an “inductiontype” regimen consisting of vincristine, dexamethasone, and l-ASNase against multiple chemoresistant xenografts derived from patients with fatal disease [124]. Moreover, true therapeutic in vivo synergy (as defined above) has also been demonstrated against paediatric ALL xenografts when ABT-737 was combined with topotecan and L-ASNase [130]. Therefore, we should be careful not to throw the baby out with the bath water, and each of these molecularly targeted therapeutic strategies should be considered on its own merit, and tested in a wider range of preclinical models prior to translation to the clinic.
3.6 Influence of Pharmacokinetics on In Vivo Preclinical Drug Testing Retrospective analysis of pharmacokinetic and pharmacodynamic parameters from preclinical and clinical data supports the intuitive notion that these are crucial in determining efficacy. The distribution of a certain drug in different compartments
3 Preclinical Evaluation
53
and organs, as well as metabolic turnover, in experimental animals could be very different to that in human. Adjustments to treatment regimens in experimental models to simulate the peak plasma concentration and area under the curve of a certain drug in humans are critical if the continued rate of failures in the clinic is to be reduced (see Sect. 3.3). In general terms, the MTD for a drug is higher for mice than humans, which could in part explain the perennial problem of preclinical models overestimating the efficacy of anti-cancer drugs [131]. An essential component of the PPTP described above is accurate modelling of drug pharmacokinetics in experimental mice compared with available information on any tested drug in humans. An interesting example of this is that the only drug tested to date by the PPTP that was not carried out at the drug’s MTD was topotecan, which was based on extensive prior knowledge of the drug’s pharmacokinetics in the paediatric cancer population [131]. Nevertheless, topotecan showed good in vivo efficacy, consistent with recent data from clinical trials [113, 114].
3.7 Conclusions and Future Challenges There is no doubt that the potential for development of new anti-leukaemic drugs has increased dramatically over the past two decades with advances in drug design, high-throughput screening methods, and our understanding of cancer biology. Nevertheless, a major limiting step in the drug approval pipeline lies in the identification of drugs that will perform as well in clinical trials as they did in preclinical evaluation. Even mammoth systematic preclinical screening programs such as that undertaken by PPTP are experiments in themselves, and their predictive power will only be understood in the fullness of time. In the foreseeable future different challenges loom in the preclinical testing arena for paediatric and adult ALL. For paediatric patients, the major challenge appears to be how to salvage the small proportion of patients who experience early and multiple relapses, a population that experiences the worst outcome. As new drugs are identified by systematic preclinical screening efforts and moved into the clinical trials pipeline, it will be essential to develop the molecular tools to identify which patients are likely to respond to one particular novel drug and not another, and thereby significantly advance the science of individualised medicine. We believe that this is a realistic goal in the paediatric oncology field. The major challenge for preclinical drug evaluation in the adult ALL field is to identify new drugs that can contribute to improving the cure rate to approaching that of paediatric ALL. Perhaps unique initiatives such as the PPTP should be expanded into the adult oncology field, since it cannot be assumed that the same drugs that are shown to be active against paediatric ALL will exhibit equal efficacy against adult disease. Nevertheless, novel drugs identified to be active against highrisk or relapsed paediatric ALL should also be evaluated systematically against preclinical models of adult ALL, thereby contributing to the pipeline for early phase clinical trials in adult ALL.
54
B. Szymanska et al.
A challenge common to both paediatric and adult ALL will be how best to combine these novel active drugs with established drugs in the clinic. Notwithstanding the technical difficulties associated with preclinical evaluation of novel drugs in combinations that were discussed above, this is an area that requires continued development, re-evaluation, and consultation with our clinician colleagues. Acknowledgements This work was supported by NO1CM42216 from the NCI (USA), by Children’s Cancer Institute Australia for Medical Research, and by the University of Southern California Childrens Hospital Los Angeles Institute for Pediatric Clinical Research. Children’s Cancer Institute Australia for Medical Research is affiliated with the University of New South Wales and Sydney Children’s Hospital. The authors would like to acknowledge technical assistance from Ingrid Boehm, Mila Dolotin, and Clare Boland, as well as helpful discussions with Professor Peter Houghton and Chris Morton (St Jude Children’s Research Hospital, Memphis, TN, USA) and Dr Malcolm Smith (Cancer Therapy Evaluation Program, National Cancer Institute, Bethesda, MD, USA).
References 1. Pulte D, Gondos A, Brenner H. Improvement in survival in younger patients with acute lymphoblastic leukemia from the 1980s to the early 21st century. Blood. 2009;113:1408–1411. 2. Larson S, Stock W. Progress in the treatment of adults with acute lymphoblastic leukemia. Curr Opin Hematol. 2008;15:400–407. 3. Jabbour EJ, Faderl S, Kantarjian HM. Adult acute lymphoblastic leukemia. Mayo Clin Proc. 2005;80:1517–1527. 4. Alberts DS, van Daalen Wetters T. The effect of phenobarbital on cyclophosphamide antitumor activity. Cancer Res. 1976;36:2785–2789. 5. Goldin A, Venditti JM, Kline I. Evaluation of antileukemic agents employing advanced leukemia L-1210 in mice. V. Cancer Res. 1962;(Suppl):157–220. 6. Oltersdorf T, Elmore SW, Shoemaker AR, et al. An inhibitor of Bcl-2 family proteins induces regression of solid tumours. Nature. 2005;435:677–681. 7. Buchdunger E, Cioffi CL, Law N, et al. Abl protein-tyrosine kinase inhibitor STI571 inhibits in vitro signal transduction mediated by c-kit and platelet-derived growth factor receptors. J Pharmacol Exp Ther. 2000;295:139–145. 8. Kidd JG. Regression of transplanted lymphomas induced in vivo by means of normal guinea pig serum. I. Course of transplanted cancers of various kinds in mice and rats given guinea pig serum, horse serum, or rabbit serum. J Exp Med. 1953;98:565–582. 9. Broome JD. Evidence that the L-asparaginase activity of guinea pig serum is responsible for its antilymphoma effects. Nature. 1961;191:1114–1115. 10. Haley EE, Fischer GA, Welch AD. The requirement for L-asparagine of mouse leukemia cells L5178Y in culture. Cancer Res. 1961;21:532–536. 11. Mashburn LT, Wriston JC, Jr. Tumor inhibitory effect of L-asparaginase from Escherichia coli. Arch Biochem Biophys. 1964;105:450–452. 12. Dolowy WC, Henson D, Cornet J, et al. Toxic and antineoplastic effects of L-asparaginase. Study of mice with lymphoma and normal monkeys and report on a child with leukemia. Cancer. 1966;19:1813–1819. 13. Hill JM, Roberts J, Loeb E, et al. L-asparaginase therapy for leukemia and other malignant neoplasms. Remission in human leukemia. JAMA. 1967;202:882–888. 14. Oettgen HF, Old LJ, Boyse EA, et al. Inhibition of leukemias in man by L-asparaginase. Cancer Res. 1967;27:2619–2631. 15. Muller HJ, Boos J. Use of L-asparaginase in childhood ALL. Crit Rev Oncol/Hematol. 1998; 28:97–113.
3 Preclinical Evaluation
55
16. Rosenberg B, Renshaw E, Vancamp L, et al. Platinum-induced filamentous growth in Escherichia coli. J Bacteriol. 1967;93:716–721. 17. Rosenberg B, VanCamp L, Trosko JE, et al. Platinum compounds: a new class of potent antitumour agents. Nature. 1969;222:385–386. 18. Nitschke R, Starling KA, Vats T, et al. Cis-diamminedichloroplatinum (NSC-119875) in childhood malignancies: a Southwest Oncology Group study. Med Pediatr Oncol. 1978; 4:127–132. 19. Vietti TJ, Nitschke R, Starling KA, et al. Evaluation of cis-dichlorodiammineplatinum(II) in children with advanced malignant diseases: Southwest Oncology Group Studies. Cancer Treat Rep. 1979;63:1611–1614. 20. Tajbakhsh M, Houghton PJ, Morton CL, et al. Initial testing of cisplatin by the pediatric preclinical testing program. Pediatr Blood Cancer. 2008;50:992–1000. 21. Gellhorn A, Hirschberg E. Investigation of diverse systems for cancer chemotherapy screening. I. Summary of results and general correlations. Cancer Res Suppl. 1955;3:1–13. 22. Goldin A, Venditti JM, Kline I, et al. Evaluation of antileukemic agents employing advanced leukemia L1210 in mice II. Cancer Res. 1960;20:382–448. 23. Alley M, Hollingshead M, Dykes D, et al. Human tumor xenograft models in NCI drug development. In: Teicher, BA, Andrews, PA, editors. Anticancer drug development guide: preclinical screening, clinical trials, and approval. 2nd ed. Humana, Totowa (NJ). 2004;pp. 125–152. 24. Monks A, Scudiero D, Skehan P, et al. Feasibility of a high-flux anticancer drug screen using a diverse panel of cultured human tumor cell lines. J Natl Cancer Inst. 1991;83:757–766. 25. Shoemaker RH. The NCI60 human tumour cell line anticancer drug screen. Nat Rev Cancer. 2006;6:813–823. 26. Brown JM. NCI’s anticancer drug screening program may not be selecting for clinically active compounds. Oncol Res. 1997;9:213–215. 27. Paull KD, Shoemaker RH, Hodes L, et al. Display and analysis of patterns of differential activity of drugs against human tumor cell lines: Development of mean graph and COMPARE algorithm. J Natl Cancer Inst. 1989;81:1088–1092. 28. Keskin O, Bahar I, Jernigan RL, et al. Characterization of anticancer agents by their growth inhibitory activity and relationships to mechanism of action and structure. Anticancer Drug Des. 2000;15:79–98. 29. Boyd MR, Paull KD. Some practical considerations and applications of the national cancer institute in vitro anticancer drug discovery screen. Drug Dev Res. 1995;34:91–109. 30. DiMasi JA, Hansen RW, Grabowski HG. The price of innovation: new estimates of drug development costs. J Health Econ. 2003;22:151–185. 31. Kelloff GJ, Sigman CC. New science-based endpoints to accelerate oncology drug development. Eur J Cancer. 2005;41:491–501. 32. Kola I, Landis J. Can the pharmaceutical industry reduce attrition rates? Nat Rev Drug Discov. 2004;3:711–715. 33. Carver BS, Pandolfi PP. Mouse modeling in oncologic preclinical and translational research. Clin Cancer Res. 2006;12:5305–5311. 34. Lee EM, Bachmann PS, Lock RB. Xenograft models for the preclinical evaluation of new therapies in acute leukemia. Leuk Lymphoma. 2007;48:659–668. 35. Kerbel RS. What is the optimal rodent model for anti-tumor drug testing? Cancer Metastasis Rev. 1998;17:301–304. 36. Johnson JI, Decker S, Zaharevitz D, et al. Relationships between drug activity in NCI preclinical in vitro and in vivo models and early clinical trials. Br J Cancer. 2001;84:1424–1431. 37. Voskoglou-Nomikos T, Pater JL, Seymour L. Clinical predictive value of the in vitro cell line, human xenograft, and mouse allograft preclinical cancer models. Clin Cancer Res. 2003; 9:4227–4239. 38. Zoli W, Ricotti L, Tesei A, et al. In vitro preclinical models for a rational design of chemotherapy combinations in human tumors. Crit Rev Oncol/Hematol. 2001;37:69–82. 39. Drexler HG, Fombonne S, Matsuo Y, et al. p53 Alterations in human leukemia-lymphoma cell lines: in vitro artifact or prerequisite for cell immortalization? Leukemia. 2000; 14:198–206.
56
B. Szymanska et al.
40. Meijerink JP, Mensink EJ, Wang K, et al. Hematopoietic malignancies demonstrate loss-of-function mutations of BAX. Blood. 1998;91:2991–2997. 41. Inaba M, Tashiro T, Sato S, et al. In vitro-in vivo correlation in anticancer drug sensitivity test using AUC-based concentrations and collagen gel droplet-embedded culture. Oncology. 1996; 53:250–257. 42. Bernard N, Devevey L, Jacquemont C, et al. A new model of pre-B acute lymphoblastic leukemia chemically induced in rats. Exp Hematol. 2005;33:1130–1139. 43. Corpet DE, Pierre F. How good are rodent models of carcinogenesis in predicting efficacy in humans? A systematic review and meta-analysis of colon chemoprevention in rats, mice and men. Eur J Cancer. 2005;41:1911–1922. 44. Cotter SM, Essex M. Animal model: feline acute lymphoblastic leukemia and aplastic anemia. Am J Pathol. 1977;87:265–268. 45. Maronpot RR, Flake G, Huff J. Relevance of animal carcinogenesis findings to human cancer predictions and prevention. Toxicol Pathol. 2004;32 Suppl 1:40–48. 46. Lewandoski M. Conditional control of gene expression in the mouse. Nat Rev Genet. 2001; 2:743–755. 47. Macleod KF, Jacks T. Insights into cancer from transgenic mouse models. J Pathol. 1999; 187:43–60. 48. Adams JM, Harris AW, Langdon WY, et al. c-myc-Induced lymphomagenesis in transgenic mice and the role of the Pvt-1 locus in lymphoid neoplasia. Curr Top Microbiol Immunol. 1986;132:1–8. 49. Langdon WY, Harris AW, Cory S, et al. The c-myc oncogene perturbs B lymphocyte development in E-mu-myc transgenic mice. Cell. 1986;47:11–18. 50. Mason KD, Vandenberg CJ, Scott CL, et al. In vivo efficacy of the Bcl-2 antagonist ABT-737 against aggressive Myc-driven lymphomas. Proc Natl Acad Sci USA. 2008;105:17961–17966. 51. Chen J, Jette C, Kanki JP, et al. NOTCH1-induced T-cell leukemia in transgenic zebrafish. Leukemia. 2007;21:462–471. 52. Feng H, Langenau DM, Madge JA, et al. Heat-shock induction of T-cell lymphoma/Leukemia in conditional Cre/lox-regulated transgenic zebrafish. Br J Haematol. 2007;138:169–175. 53. Langenau DM, Traver D, Ferrando AA, et al. Myc-induced T cell leukemia in transgenic zebrafish. Science. 2003;299:887–890. 54. Sabaawy HE, Azuma M, Embree LJ, et al. ETV6-RUNX1 transgenic zebrafish model of precursor B cell acute lymphoblastic leukemia. Proc Natl Acad Sci USA. 2006;103:15166–15171. 55. McGrath P, Li C-Q. Zebrafish: a predictive model for assessing drug-induced toxicity. Drug Discov Today. 2008;13:394–401. 56. Flanagan SP. ‘Nude’, a new hairless gene with pleiotropic effects in the mouse. Genet Res. 1966;8:295–309. 57. Rygaard J, Povlsen CO. Heterotransplantation of a human malignant tumour to “Nude” mice. Acta Pathol Microbiol Scand. 1969;77:758–760. 58. Segre JA, Nemhauser JL, Taylor BA, et al. Positional cloning of the nude locus: genetic, physical, and transcription maps of the region and mutations in the mouse and rat. Genomics. 1995;28:549–559. 59. Luo Y, Hara H, Haruta Y, et al. Establishment of ascitic tumor of human pre-B acute lymphoblastic leukemia in nonconditioned nude mice. Cancer Res. 1989;49:706–710. 60. Cavallo F, Riccardi C, Forni M, et al. Growth and dissemination of human malignant lymphoblasts in immunosuppressed nu/nu mice. Nat Immun Cell Growth Regul. 1991;10:256–264. 61. Caretto P, Forni M, d’Orazi G, et al. Xenotransplantation in immunosuppressed nude mice of human solid tumors and acute leukemias directly from patients or in vitro cell lines. Ric Clin Lab. 1989;19:231–243. 62. Bosma GC, Custer RP, Bosma MJ. A severe combined immunodeficiency mutation in the mouse. Nature. 1983;301:527–530. 63. McCune JM, Namikawa R, Kaneshima H, et al. The SCID-hu mouse: murine model for the analysis of human hematolymphoid differentiation and function. Science. 1988; 241:1632–1639.
3 Preclinical Evaluation
57
64. Kamel-Reid S, Letarte M, Sirard C, et al. A model of human acute lymphoblastic leukemia in immune-deficient SCID mice. Science. 1989;246:1597–1600. 65. Uckun FM. Severe combined immunodeficient mouse models of human leukemia. Blood. 1996;88:1135–1146. 66. Bosma MJ, Carroll AM. The SCID mouse mutant: definition, characterization, and potential uses. Annu Rev Immunol. 1991;9:323–350. 67. Greiner DL, Hesselton RA, Shultz LD. SCID mouse models of human stem cell engraftment. Stem Cells. 1998;16:166–177. 68. Greiner DL, Shultz LD, Yates J, et al. Improved engraftment of human spleen cells in NOD/ LtSz-scid/scid mice as compared with C.B-17-scid/scid mice. Am J Pathol. 1995;146:888–902. 69. Christianson SW, Greiner DL, Hesselton RA, et al. Enhanced human CD4+ T cell engraftment in beta2-microglobulin-deficient NOD-scid mice. J Immunol. 1997;158:3578–3586. 70. Kollet O, Peled A, Byk T, et al. beta2 Microglobulin-deficient (B2m(null)) NOD/SCID mice are excellent recipients for studying human stem cell function. Blood. 2000;95:3102–3105. 71. Goldman JP, Blundell MP, Lopes L, et al. Enhanced human cell engraftment in mice deficient in RAG2 and the common cytokine receptor gamma chain. Br J Haematol. 1998;103:335–342. 72. Ito M, Hiramatsu H, Kobayashi K, et al. NOD/SCID/gamma(c)(null) mouse: an excellent recipient mouse model for engraftment of human cells. Blood. 2002;100:3175–3182. 73. Wissink EHJ, Verbrugge I, Vink SR, et al. TRAIL enhances efficacy of radiotherapy in a p53 mutant, Bcl-2 overexpressing lymphoid malignancy. Radiother Oncol. 2006;80:214–222. 74. Arguello F, Alexander M, Sterry JA, et al. Flavopiridol induces apoptosis of normal lymphoid cells, causes immunosuppression, and has potent antitumor activity In vivo against human leukemia and lymphoma xenografts. Blood. 1998;91:2482–2490. 75. Yoshida N, Ishii E, Nomizu M, et al. The laminin-derived peptide YIGSR (Tyr-Ile-Gly-SerArg) inhibits human pre-B leukaemic cell growth and dissemination to organs in SCID mice. Br J Cancer. 1999;80:1898–1904. 76. Liem NL, Papa RA, Milross CG, et al. Characterization of childhood acute lymphoblastic leukemia xenograft models for the preclinical evaluation of new therapies. Blood. 2004; 103:3905–3914. 77. Lock RB, Liem NL, Papa RA. Preclinical testing of antileukemic drugs using an in vivo model of systemic disease. Methods Mol Med. 2005;111:323–334. 78. Houghton PJ, Morton CL, Tucker C, et al. The pediatric preclinical testing program: Description of models and early testing results. Pediatr Blood Cancer. 2006;49:928–940. 79. Gaynon PS. Childhood acute lymphoblastic leukemia and relapse. Br J Haematol. 2005; 131:579–587. 80. Chou TC, Zhang XG, Harris CR, et al. Desoxyepothilone B is curative against human tumor xenografts that are refractory to paclitaxel. Proc Natl Acad Sci USA. 1998;95: 15798–15802. 81. Yang J, Ikezoe T, Nishioka C, et al. AZD1152, a novel and selective aurora B kinase inhibitor, induces growth arrest, apoptosis, and sensitization for tubulin depolymerizing agent or topoisomerase II inhibitor in human acute leukemia cells in vitro and in vivo. Blood. 2007; 110:2034–2040. 82. Gourdeau H, Bibeau L, Ouellet F, et al. Comparative study of a novel nucleoside analogue (Troxatyl, troxacitabine, BCH-4556) and AraC against leukemic human tumor xenografts expressing high or low cytidine deaminase activity. Cancer Chemother Pharmacol. 2001; 47:236–240. 83. Miller CP, Ban K, Dujka ME, et al. NPI-0052, a novel proteasome inhibitor, induces caspase-8 and ROS-dependent apoptosis alone and in combination with HDAC inhibitors in leukemia cells. Blood. 2007;110:267–277. 84. Myers DE, Chandan-Langlie M, Chelstrom LM, et al. In vitro and in vivo anti-leukemic efficacy of cyclic AMP modulating agents against human leukemic B-cell precursors. Leuk Lymphoma. 1996;22:259–264. 85. Tomkinson B, Bendele R, Giles FJ, et al. OSI-211, a novel liposomal topoisomerase I inhibitor, is active in SCID mouse models of human AML and ALL. Leuk Res. 2003;27:1039–1050.
58
B. Szymanska et al.
86. Uckun FM, Stewart CF, Reaman G, et al. In vitro and in vivo activity of topotecan against human B-lineage acute lymphoblastic leukemia cells. Blood. 1995;85:2817–2828. 87. Uno S, Kinoshita Y, Azuma Y, et al. Antitumor activity of a monoclonal antibody against CD47 in xenograft models of human leukemia. Oncol Rep. 2007;17:1189–1194. 88. Uckun FM, Chelstrom LM, Irvin JD, et al. In vivo efficacy of B43 (anti-CD19)-pokeweed antiviral protein immunotoxin against BCL-1 murine B-cell leukemia. Blood. 1992;79: 2649–2661. 89. Jansen B, Uckun FM, Jaszcz WB, et al. Establishment of a human t(4;11) leukemia in severe combined immunodeficient mice and successful treatment using anti-CD19 (B43)-pokeweed antiviral protein immunotoxin. Cancer Res. 1992;52:406–412. 90. Uckun FM, Manivel C, Arthur D, et al. In vivo efficacy of B43 (anti-CD19)-pokeweed antiviral protein immunotoxin against human pre-B cell acute lymphoblastic leukemia in mice with severe combined immunodeficiency. Blood. 1992;79:2201–2214. 91. Uckun FM, Evans WE, Forsyth CJ, et al. Biotherapy of B-cell precursor leukemia by targeting genistein to CD19-associated tyrosine kinases. Science. 1995;267:886–891. 92. Waurzyniak B, Schneider EA, Tumer N, et al. In vivo toxicity, pharmacokinetics, and antileukemic activity of TXU (anti-CD7)-pokeweed antiviral protein immunotoxin. Clin Cancer Res. 1997;3:881–890. 93. Shalapour S, Zelmer A, Pfau M, et al. The thalidomide analogue, CC-4047, induces apoptosis signaling and growth arrest in childhood acute lymphoblastic leukemia cells in vitro and in vivo. Clin Cancer Res. 2006;12:5526–5532. 94. Piloto O, Nguyen B, Huso D, et al. IMC-EB10, an anti-FLT3 monoclonal antibody, prolongs survival and reduces nonobese diabetic/severe combined immunodeficient engraftment of some acute lymphoblastic leukemia cell lines and primary leukemic samples. Cancer Res. 2006;66:4843–4851. 95. Juarez J, Dela Pena A, Baraz R, et al. CXCR4 antagonists mobilize childhood acute lymphoblastic leukemia cells into the peripheral blood and inhibit engraftment. leukemia. 2007; 21:1249–1257. 96. Teachey DT, Sheen C, Hall J, et al. mTOR inhibitors are synergistic with methotrexate: an effective combination to treat acute lymphoblastic leukemia. Blood. 2008;112:2020–2023. 97. Lee BH, Williams IR, Anastasiadou E, et al. FLT3 internal tandem duplication mutations induce myeloproliferative or lymphoid disease in a transgenic mouse model. Oncogene. 2005;24:7882–7892. 98. Reichert A, Heisterkamp N, Daley GQ, et al. Treatment of Bcr/Abl-positive acute lymphoblastic leukemia in P190 transgenic mice with the farnesyl transferase inhibitor SCH66336. Blood. 2001;97:1399–1403. 99. Brown VI, Fang J, Alcorn K, et al. Rapamycin is active against B-precursor leukemia in vitro and in vivo, an effect that is modulated by IL-7-mediated signaling. Proc Natl Acad Sci USA. 2003;100:15113–15118. 100. Neale G, Su X, Morton CL, et al. Molecular characterization of the pediatric preclinical testing panel. Clin Cancer Res. 2008;14:4572–4583. 101. Lock R, Carol H, Houghton PJ, et al. Initial testing (stage 1) of the BH3 mimetic ABT-263 by the pediatric preclinical testing program. Pediatr Blood Cancer. 2008;50:1181–1189. 102. Smith MA, Morton CL, Phelps DA, et al. Stage 1 testing and pharmacodynamic evaluation of the HSP90 inhibitor alvespimycin (17-DMAG, KOS-1022) by the pediatric preclinical testing program. Pediatr Blood Cancer. 2008;51:34–41. 103. Morton CL, Houghton PJ, Gorlick R, et al. Initial testing of Aplidin by the pediatric preclinical testing program. Pediatr Blood Cancer. 2009;53(3):509–512. 104. Maris JM, Courtright J, Houghton PJ, et al. Initial testing of the VEGFR inhibitor AZD2171 by the pediatric preclinical testing program. Pediatr Blood Cancer. 2008;50:581–587. 105. Houghton PJ, Morton CL, Kolb EA, et al. Initial testing (stage 1) of the proteasome inhibitor bortezomib by the pediatric preclinical testing program. Pediatr Blood Cancer. 2007; 50:37–45.
3 Preclinical Evaluation
59
106. Kolb EA, Gorlick R, Houghton PJ, et al. Initial testing of dasatinib by the pediatric preclinical testing program. Pediatr Blood Cancer. 2008;50:1198–1206. 107. Carol H, Lock R, Houghton PJ, et al. Initial testing (stage 1) of the kinesin spindle protein inhibitor Ispinesib by the pediatric preclinical testing program. Pediatr Blood Cancer. 2009; 53(7):1255–1263. 108. Gorlick R, Kolb EA, Houghton PJ, et al. Initial Testing (stage 1) of Lapatinib by the pediatric preclinical testing program. Pediatr Blood Cancer. 2009;53(4):594–598. 109. Houghton PJ, Morton CL, Kolb EA, et al. Initial testing (stage 1) of the mTOR inhibitor rapamycin by the pediatric preclinical testing program. Pediatr Blood Cancer. 2007;50: 799–805. 110. Maris JM, Courtright J, Houghton PJ, et al. Initial testing (stage 1) of sunitinib by the pediatric preclinical testing program. Pediatr Blood Cancer. 2008;51:42–48. 111. Keshelava N, Houghton PJ, Morton CL, et al. Initial testing (stage 1) of Vorinostat (SAHA) by the pediatric preclinical testing program. Pediatr Blood Cancer. 2009;53(3):505–508. 112. Kolb EA, Gorlick R, Houghton PJ, et al. Initial testing (stage 1) of a monoclonal antibody (SCH 717454) against the IGF-1 receptor by the pediatric preclinical testing program. Pediatr Blood Cancer. 2008;50:1190–1197. 113. Furman WL, Stewart CF, Kirstein M, et al. Protracted intermittent schedule of topotecan in children with refractory acute leukemia: a pediatric oncology group study. J Clin Oncol. 2002;20:1617–1624. 114. Hijiya N, Stewart CF, Zhou Y, et al. Phase II study of topotecan in combination with dexamethasone, asparaginase, and vincristine in pediatric patients with acute lymphoblastic leukemia in first relapse. Cancer. 2008;112:1983–1991. 115. Lombardo LJ, Lee FY, Chen P, et al. Discovery of N-(2-chloro-6-methyl- phenyl)-2-(6-(4(2-hydroxyethyl)- piperazin-1-yl)-2-methylpyrimidin-4- ylamino)thiazole-5-carboxamide (BMS-354825), a dual Src/Abl kinase inhibitor with potent antitumor activity in preclinical assays. J Med Chem. 2004;47:6658–6661. 116. Barenbaum MC. Synergy, additivism and antagonism in immunosuppression. A critical review. Clin Exp Immunol. 1977;28:1–18. 117. Chou TC, Talalay P. Quantitative analysis of dose-effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul. 1984;22:27–55. 118. Chou T-C. Preclinical versus clinical drug combination studies. Leuk Lymphoma. 2008; 49:2059–2080. 119. Carter SK. Planning combined therapy – the interaction of experimental and clinical studies. Cancer Chemother Rep. 1974;2 4:3–11. 120. Corbett TH, Griswold DP, Jr., Wolpert MK, et al. Design and evaluation of combination chemotherapy trials in experimental animal tumor systems. Cancer Treat Rep. 1979; 63:799–801. 121. Hanfelt JJ. Statistical approaches to experimental design and data analysis of in vivo studies. Breast Cancer Res Treat. 1997;46:279–302. 122. Rose WC, Wild R. Therapeutic synergy of oral taxane BMS-275183 and cetuximab versus human tumor xenografts. Clin Cancer Res. 2004;10:7413–7417. 123. Gitler MS, Monks A, Sausville EA. Preclinical models for defining efficacy of drug combinations: mapping the road to the clinic. Mol Cancer Ther. 2003;2:929–932. 124. Kang MH, Kang YH, Szymanska B, et al. Activity of vincristine, L-ASP, and dexamethasone against acute lymphoblastic leukemia is enhanced by the BH3-mimetic ABT-737 in vitro and in vivo. Blood. 2007;110:2057–2066. 125. Kuroda J, Kimura S, Segawa H, et al. The third-generation bisphosphonate zoledronate synergistically augments the anti-Ph+ leukemia activity of imatinib mesylate. Blood. 2003; 102:2229–2235. 126. Messinger Y, Yanishevski Y, Avramis VI, et al. Treatment of human B-cell precursor leukemia in SCID mice using a combination of the investigational biotherapeutic agent B43-PAP with cytosine arabinoside. Clin Cancer Res. 1996;2:1533–1542.
60
B. Szymanska et al.
127. Uckun FM, Zheng Y, Cetkovic-Cvrlje M, et al. In vivo pharmacokinetic features, toxicity profile, and chemosensitizing activity of alpha-cyano-beta-hydroxy-beta- methyl-N(2,5-dibromophenyl)propenamide (LFM-A13), a novel antileukemic agent targeting Bruton’s tyrosine kinase. Clin Cancer Res. 2002;8:1224–1233. 128. Decker S, Sausville EA. Preclinical Modeling of Combination Treatments: Fantasy or Requirement? Ann NY Acad Sci. 2005;1059:61–69. 129. Real PJ, Tosello V, Palomero T, et al. [gamma]-Secretase inhibitors reverse glucocorticoid resistance in T cell acute lymphoblastic leukemia. Nat Med. 2009;15:50–58. 130. Szymanska B, Barber N, Wilczynska-Kalak U, et al. Targeting the apoptotic machinery of pediatric acute lymphoblastic leukemia in vivo using the BH3-mimetic ABT-737. Proc Am Assoc Cancer Res. 2008;49:2990. 131. Zamboni WC, Stewart CF, Thompson J, et al. Relationship between topotecan systemic exposure and tumor response in human neuroblastoma xenografts. J Natl Cancer Inst. 1998; 90:505–511.
Chapter 4
Design of Early-Phase Trials James A. Whitlock and Terzah M. Horton
4.1 Introduction Children with newly diagnosed ALL obtain a complete remission (CR) following treatment with standard agents in greater than 98% of cases [1], contributing to 5-year event-free survival of 71–88% in children with high-risk and standard-risk B-precursor ALL, respectively, enrolled on recent Children’s Oncology Group (COG) trials [2]. In contrast, less than 75% of children with ALL in early first relapse (variously defined as occurring less than 12 months from end therapy or less than 36 months from initial diagnosis) following contemporary COG frontline therapies attain a second CR with standard induction agents [3, 4]; children in second relapse fare even more poorly, with reinduction rates consistently around 40% following various combinations of standard agents [5]. Failure to attain remission following relapse is one of several factors associated with a survival rate of only 20% following marrow relapse off COG frontline therapies [6, 7]. Low rates of reinduction in children with early relapse of ALL and poor EFS after attaining a second or subsequent remission provide compelling justification for the investigation of new agents in this setting. Such investigations must be conducted in a manner that allows the accurate assessment of toxicities and efficacy while addressing the unwillingness of many pediatric oncologists to withhold standard chemotherapeutic agents from children in first relapse of ALL, due to their effectiveness in attaining a second remission in many cases.
J.A. Whitlock (*) Division of Haematology/Oncology, The Hospital for Sick Children, 555 University Avenue, Toronto, ON M5G 1X8, Canada e-mail:
[email protected] V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3_4, © Springer Science+Business Media, LLC 2011
61
62
J.A. Whitlock and T.M. Horton
4.2 Preclinical Evaluation of Novel Agents in Childhood Leukemias The clinical development of new anticancer agents can be expedited by welldesigned preclinical studies that may predict optimal dosing schedules, anticipate toxicities, and identify synergistic or antagonistic combinations with other anticancer agents. The National Cancer Institute of USA implemented the Pediatric Preclinical Testing Program (PPTP) with the objective of identifying agents with significant activity in panels of pediatric preclinical models as a mechanism for prioritizing agents for advancement to clinical trials for children with selected malignancies [8]. The PPTP has evaluated over 15 new agents for activity against specific childhood malignancies (http://pptp.nchresearch.org/). A number of new agents, including the proteasome inhibitor bortezomib [9], the mTOR inhibitor rapamycin [10], the tyrosine kinase inhibitor dasatinib [11], and the BH3 mimetic ABT-263 [12] have shown promising activity against childhood ALL cell lines in this preclinical model, supporting their prioritization for evaluation in clinical trials of relapsed childhood ALL.
4.3 Phase I Trials and Trial Design Issues 4.3.1 Conventional Phase I Trial Design Preclinical evaluation of a novel agent in adults is typically followed by singleagent phase I evaluation in adults to identify dose-limiting toxicities (DLTs) and a maximum tolerated dose (MTD) for subsequent phase II studies. These toxicityseeking studies are a prerequisite for disease-specific efficacy studies at the recommended phase II dose (RPIID). A novel agent needs pediatric testing, both as a single agent and in combination with other chemotherapeutic agents, in an effort to identify the proper context for incorporation of the new agent into future treatment regimens. The starting dose in such first-in-human trials is typically derived from an increment of the lethal dose in mice; this pharmacologically guided approach minimizes the risk of significant toxicity at the starting dose [13]. While more efficient than empirically selected starting doses, the pharmacologically guided approach typically generates a starting dose that is still well below both the MTD and the therapeutic dose in humans [14]. The most widely used clinical trial design in phase I studies is the 3 + 3 trial design, a variation of the original phase I up-and-down methodology [15]. In this approach, three subjects are evaluated at a specified dose level; if none experiences a DLT, the dose is escalated and an additional three subjects are evaluated at the next higher dose level. In the event that one of three subjects at a dose level suffers a DLT, up to three additional subjects are evaluated at that dose level. If two or more subjects suffer DLT in a cohort of up to six patients, the MTD is deemed to have been
4 Design of Early-Phase Trials
63
exceeded. The MTD is defined as that dose level at which 0–1 subject experiences a DLT, when two or more subjects experience a DLT at the next higher dose level. The success of this study design in limiting the number of patients exposed to a potentially toxic dose of a new agent is mitigated by the inherent delays encountered in moving from one dose level to the next while awaiting the collection of toxicity data upon completion of a dose level [15].
4.3.2 Conventional Pediatric Phase I Trial Design A similar approach to that described above has traditionally been utilized in pediatric cancer drug development; however, given the relative rarity of pediatric malignancies and the challenges in developing a drug primarily for a pediatric indication, early-phase studies in children are often deferred until a given drug is known to have commercial viability and thus expected availability, based on demonstrated activity in adult phase II trials. A typical approach in a pediatric single-agent phase I trial of a novel cytotoxic agent is to initiate the study at 75–80% of the established adult MTD, with subsequent incremental dose escalations to toxicity [16]. This follow-on approach to phase I testing in children has the potential benefit of decreasing the number of pediatric subjects exposed to subtherapeutic doses of the new agent, and thus improving the risk–benefit ratio. While discussion continues regarding the relative benefits of phase I oncology trials in children and their ethical justification [17], the necessity of conducting such trials as a basis for subsequent phase II trials is generally accepted by the pediatric oncology field and is based on the observation that young children, in particular, may demonstrate altered metabolism and pharmacokinetics of chemotherapeutic agents. Thus, the pediatric MTD of a cytotoxic agent may be less than, the same as, or greater than the adult MTD of that agent [14]. The desire to minimize the number of subjects exposed to subtherapeutic doses of a new agent while maintaining an appropriate balance of risk and benefit [18], together with the limited number of pediatric patients available for phase I trial participation, the lengthy time required for their completion, and the increasing number of new agents, particularly molecularly targeted agents that are biologically relevant to only a small proportion of the pediatric cancer population, have spurred the development of several modifications of the traditional 3 + 3 phase I trial design [19–23]. These so-called accelerated designs include the continual reassessment method (CRM), accelerated titration designs, and the rolling six design.
4.3.3 Continual Reassessment Method of Phase I Trial Design The conventional 3 + 3 phase I trial design has been criticized as being developed largely ad hoc and without intrinsic statistical validity [15]. In an effort to identify an approach that reduces the number of patients exposed to subtherapeutic doses and thus has improved ethical equipoise, O’Quigley et al. developed a Bayesian
64
J.A. Whitlock and T.M. Horton
dose-finding method, which they termed the CRM [19]. The CRM replaces the evaluation of a predetermined number of patients at each dose level with a method that incorporates a continuous reassessment of the risk of toxicity in each accrued subject, thus allowing a more informed decision in regard to dose escalation with each subsequent subject. The purported benefits of the CRM include treating fewer patients at subtherapeutic doses and a better estimate of the true risk of toxicity at the MTD. While the performance of this method in published simulations has been encouraging, potential disadvantages of the CRM include an increased duration of time for completion of trials and the possibility of increased toxicity compared with standard designs. Despite modifications introduced into the original design of the CRM to address these and other concerns by restricting dose escalation and assigning more than one subject per group [20, 21], the CRM has been slow to be accepted by clinical investigators and statisticians. Additional refinements, such as incorporating evaluation of both toxicity and response using a Bayesian method in a dose-finding weighted design, have been proposed [23].
4.3.4 Accelerated Titration Designs The conventional 3 + 3 approach to phase I trial design does not permit dose escalation for subsequent courses in the same patient – that is, intrapatient dose escalation. Disallowing intrapatient dose escalation is intended to minimize problems with assessment and attribution of toxicities that may occur when a single subject receives multiple doses of a new agent at different dose levels, thus rendering that subject’s data potentially nonevaluative. Titration designs involve dose escalation within patients until the desired biologic outcome is obtained – in the case of phase I oncology trials, the “desired” outcome being toxicity. The use of intrapatient dose escalation can reduce both the number of participants and the amount of time required to complete a trial. Although accelerated designs that do not incorporate intrapatient dose escalations provide most of the advantages of accelerated titration designs, with little or no increase in risk compared with the standard 3 + 3 design, they do not provide as great a reduction in the number of undertreated patients. Titration designs provide patients accrued early in the trial or those who have an especially high individual tolerance for the drug a greater opportunity for therapeutic benefit [22]. This approach, like other modifications of the conventional phase I design, has been slow to be adopted by both clinical investigators and clinical trial statisticians.
4.3.5 Rolling Six Design Based on the remarkable safety of conventional pediatric phase I trial designs, with a reported toxic death rate of less than 1% [24, 25], Skolnik et al. proposed an alternative trial design with the objective of shortening the time to trial completion
4 Design of Early-Phase Trials
65
while continuing to maintain an adequate safety profile [26]. The so-called rolling six design allows for the concurrent accrual of 2–6 subjects at a dose level; a decision as to which dose level to enroll a subject is based on the number of patients currently enrolled and evaluative, the number of subjects experiencing DLTs to date, and the number of subjects still at risk of developing a DLT at the time of new patient entry. In a simulated comparison of the 3 + 3 and rolling six methods in a typical pediatric phase I clinical trial scenario, the mean ± standard deviation (SD) time to complete a pediatric phase I study was 294 ± 75 days for the rolling six design versus 350 ± 84 days for the 3 + 3 design (assuming an average interpatient enrollment interval of 10 days), with only a small increase in the total number of subjects enrolled when using the rolling six design (n = 3), and no difference in the distribution of DLTs between designs. The improved performance of the rolling six design is attributable to a significant decrease in the number of times a study is suspended to accrual is compared with the 3 + 3 method.
4.3.6 Obstacles to Conventional Phase I Trials in Childhood ALL The traditional single-agent phase I dose-escalation approach can be problematic when applied to childhood ALL. Chemotherapeutic regimens incorporating standard drugs are reasonably successful in reinducing remissions in relapsed diagnosed childhood ALL, with various combinations of standard agents resulting in remission reinduction rates of 71–95% following first marrow relapse [27–37]. Given that attainment of second remission is a virtual prerequisite for cure in recurrent childhood ALL, and given the overall response rate of less than 10% in pediatric phase I trials [14, 24], many families and treating physicians are understandably reluctant to subject a child with first or subsequent relapse of ALL to treatment with a novel single agent, in lieu of familiar treatments with higher anticipated response rates. Accrual of pediatric ALL patients (or lack thereof) to single-agent phase I trials in the COG and its predecessors reflects these concerns. These limitations have led to the development of a new approach to the evaluation of new agents in first marrow relapse of childhood ALL in the COG – the platform approach.
4.3.7 Platform Approach to Phase I Evaluation of New Agents in Childhood ALL As an increasing proportion of childhood ALL agents are cured of their disease with intensified therapies, the diminishing proportion of children who suffer relapse have increasingly resistant disease. Patients suffering an early marrow relapse (defined in the COG as relapse occurring less than 36 months from diagnosis) have a particularly dismal prognosis [6, 7]. Recent molecular data have supported the hypothesis
66
J.A. Whitlock and T.M. Horton
that early and late relapsing populations of ALL are biologically distinct [38]. The relative chemoresistance associated with early marrow relapse has led to a preference for allogeneic stem cell transplantation in second or subsequent remission as the treatment of choice, including unrelated donors as the stem cell source, although most controlled studies show no compelling benefit of allo-SCT in patients with early relapse who do not have a matched sibling donor [3, 39, 40]. Analyses of end-induction minimal residual disease (MRD) provide compelling evidence that patients who achieve a state of MRD negativity prior to allo-SCT fare better than those who demonstrate persistent MRD [41, 42]. One strategy to improve these poor outcomes has been to attempt to improve the “depth” of second remission (i.e., attain lower levels of MRD) through intensification of reinduction regimens, such as the six-drug regimen employed in CCG-1941. Unfortunately, the intensified reinduction strategy in CCG-1941 failed to improve reinduction rates in early marrow relapse while contributing to an increased incidence of infectious deaths and other treatment-related toxicities [3]. The COG subsequently adopted an alternative approach to improving the depth of second remission by employing three intensive, successive reinduction treatment blocks in the AALL01P2 study [4]. The three blocks were derived from combinations that were previously shown to be effective in recurrent ALL (Fig. 4.1). This approach proved to have acceptable toxicity, with a toxic death rate of less than 5% following amendment of the original protocol to replace dexamethasone with prednisone and idarubicin with doxorubicin in Block 1. The postamendment Block 1 therapy is based on the POG-9310 reinduction regimen, consisting of weekly vincristine, weekly PEG-asparaginase, daily oral
COG AALL01P2 “Triple Re-Induction” Block 1:
Block 2:
Block 3:
VPPD
CTX/ETOP -> HD MTX
HD Ara-C/Asp
/wkxx44 VCR 1.5 mg/m22 /wk PDN 40 mg2/m/ddays 1-29 PEG-ASP 2500 IU/m2/wk x 4 DOX 60 mg/m2 day 1 ITT (Imatinib for Ph+)
CTX 440 mg/m22xx55 ETOP 100 mg/m2 x 5 (Imatinib for Ph+) - then: MTX 5 g/m2 over 24 hr with leucovorin rescue
Ara -C 3 g/m2 x 8 L-asp 6000 IU/m2 x 2 (Imatinib for Ph+)
Raetz, JCO 26.3971, 2008
Fig. 4.1 COG AALL01P2 “triple re-induction”
4 Design of Early-Phase Trials
67
prednisone for 4 weeks, and doxorubicin on Day 1 [31]. Block 2 consists of (1) a 5-day course of cyclophosphamide and etoposide beginning on Day 1, based on a 40% CR rate attained with ifosfamide and etoposide in children with refractory ALL [43], with replacement of ifosfamide by equitoxic dosing with the less nephrotoxic cyclophosphamide and (2) an intravenous high-dose methotrexate on Day 22 at 5 gm/m2 over 24 h with leucovorin rescue as employed in the ALL-BFM 90 protocol [44]. Block 3 is the Capizzi II regimen of sequential high-dose Ara-C and l-asparaginase, which demonstrated a 45% CR rate in advanced ALL [45]. Reinduction rates in AALL01P2 were promising, with second remission attained in 96% of late marrow relapses (LR; defined as relapse ³36 months from diagnosis) and 68% of early marrow relapses (ER; defined as relapse <36 months from diagnosis). ER patients could be further divided into a very early relapse subset (defined as relapse <18 months from diagnosis) with only a 45% CR2 rate, and a subset of those relapsing 18–36 months from diagnosis, with a 79% CR2 rate [4]. The validity of the three-block approach in improving the depth of second remission was demonstrated with the finding that MRD status at the end of Block 1 had prognostic significance in both ER and LR patients. Furthermore, patients with MRD positivity at the end of Block 1 who subsequently became MRD negative fared better than those who had persistent MRD at the completion of the three blocks of therapy. Thus, MRD status at end induction can serve as an additional measure of response, in addition to CR2 rate based upon conventional morphology. In addition to demonstrating that the three-block approach to reinduction has reasonable efficacy, this protocol had acceptable treatment-related toxicities, suggesting that the three-block regimen could serve as a platform for evaluating the addition of promising new agents in future trials. The COG is currently testing this approach through its investigation of epratuzumab, an unconjugated monoclonal antibody against an epitope of CD22, in ADVL04P2, a phase I/II study in which epratuzumab is administered concurrently in Block 1 with the same multiagent chemotherapy regimen utilized in AALL01P2. Early results indicate that this approach is feasible and is associated with effective targeting of leukemic cells by epratuzumab in most subjects [46]. The AALL01P2 study was designed solely as a reinduction protocol for patients with first relapse of ALL; treatment with additional chemotherapy or allogeneic SCT after patients completed the three blocks of protocol therapy and entered second remission was at the discretion of the treating physician. However, an advantage of the three-block approach in AALL01P2 is that the approximately 4 months required to deliver the three blocks of therapy is a similar time frame to that required for the identification and workup of unrelated donors for patients who subsequently undergo unrelated SCT. Thus, an additional benefit of the three-block approach developed in AALL01P2 is to provide disease control for a sufficient period of time so as to allow patients in second remission to proceed to SCT. Prior to AALL01P2, CCG ALL relapse studies generally suffered from poor accrual, in turn adversely affecting their conduct and leading in some instances to premature study closure [3]. While the AALL01P2 study did not meet its target accrual rate of 110 patients per year, it did achieve its scientific objectives by accruing
68
J.A. Whitlock and T.M. Horton
145 patients between January 2003 and December 2005. Differences in study design and eligibility preclude a direct comparison of accrual data for AALL01P2 and earlier CCG ALL relapse studies; however, the inclusion of a platform, or backbone, of familiar drugs with proven efficacy in ALL reinduction appears to have assuaged the concerns of treating physicians who have previously been reluctant to enroll patients into COG relapsed ALL studies.
4.3.8 Window Approach to Evaluation of New Agents in Childhood ALL The platform approach to the evaluation of a new agent in relapsed childhood ALL, as described above, has the advantage of evaluating a novel agent while providing conventional reinduction therapy for patients with relapsed disease, thus avoiding the significant ethical and accrual problems experienced with single-agent trials in first relapse ALL. However, a potential disadvantage of this approach is the need to assess, in at least a preliminary manner, the pharmacokinetics and activity of a new agent as a single agent, if that agent has not been previously administered to children or has not been previously utilized against a specific disease (in this case ALL). This limitation can be addressed through the use of an upfront single-agent window design, in which the new agent is given prior to the administration of the combination of new-agent and conventional combination therapy. Upfront phase II window designs have previously been employed in the evaluation of novel therapies in newly diagnosed high-risk childhood sarcomas [47], and in the evaluation of different doses of standard agents such as methotrexate or asparaginase in newly diagnosed high-risk childhood ALL [48, 49]. In order to obtain single-agent pharmacokinetic data and preliminary singleagent efficacy data for epratuzumab in children with ALL, an upfront single-agent window was incorporated into the design of the COG ADVL04P2 study of epratuzumab. Patients enrolling in Part A of the study received four doses of epratuzumab twice weekly during a 14-day so-called “reduction” phase, followed by four weekly doses of epratuzumab administered in combination with Block 1 chemotherapeutic agents. To minimize the risk of patient endangerment in the event of progressive disease during the reduction (window) phase of the study, strict removal or “bailout” criteria were included in the study design that specified that patients with an increase in the WBC to greater than 100 × 109/L, or with the development of symptoms attributable to a rapidly rising absolute blast count, were to abandon the window phase and proceed immediately to Block 1 conventional chemotherapy in combination with epratuzumab [46]. Although limited dose escalation of epratuzumab was a part of the original study design of ADVL04P2, the lack of toxicities associated with epratuzumab administration made a traditional phase I dose escalation to toxicity approach impractical. Instead, the dose and schedule of epratuzumab were based on results obtained in adult single-agent studies, and the objectives of Part A of the ADVL04P2 study were devised to establish the feasibility of this
4 Design of Early-Phase Trials
69
novel chemoimmunotherapy regimen and to perform epratuzumab PK studies. Results of PK studies from both adult trials and the initial portion (Part A) of ADVL04P2 led to schedule modifications in the subsequent phase II (Part B) portion of the study.
4.3.9 Phase I End Points: Determination of Dose-Limiting Toxicities The conventional end point in a phase I study is the selection of a RPIID based on the identification of a MTD, which in turn is based on the achievement of a prespecified threshold of DLTs. The identification of DLTs, the MTD, and the RPIID for a new agent administered as a single agent is relatively straightforward. Much more challenging are the evaluation and attribution of DLTs in the context of combination chemotherapy for relapsed ALL. The incorporation of new agents into established chemotherapy regimens presents significant challenges in the identification and attribution of toxicities. Epratuzumab, as utilized in the COG ADVL04P2 study, was in many ways the ideal type of agent with which to pilot a platform approach; unconjugated monoclonal antibodies such as epratuzumab or rituximab are not predicted to have significant additive toxicities when combined with the four-drug regimen of cytotoxic chemotherapy in Block 1 of AALL01P2. Rituximab was safely combined with standard multiagent regimens for both newly diagnosed and recurrent nonHodgkin’s lymphoma in COG trials ANHL01P1 and ANHL0121, respectively. In contrast to the ease with which unconjugated antibodies have been combined with conventional cytotoxic chemotherapy in childhood lymphoid malignancies, the experience with combining a conjugated antibody with conventional chemotherapy in relapsed childhood acute myeloid leukemia (AML) has been less benign. Gemtuzumab ozogamicin (GMTZ; Mylotarg®) is a humanized anti-CD33 antibody linked to calicheamicin, a highly potent chemotherapeutic agent with significant toxicities when administered as a single agent in patients with relapsed AML [50]. Thus, it was not surprising that GMTZ administered in combination with conventional cytotoxic chemotherapy for AML proved to be associated with significant toxicities. COG AAML00P2, a pilot study for relapsed childhood AML, combined GMTZ with each of two commonly used relapsed childhood AML salvage chemotherapy regimens – high-dose cytarabine administered with either l-asparaginase (Capizzi II) or mitoxantrone. The starting dose of GMTZ 3 mg/m2 proved to have acceptable toxicities when administered in combination with cytarabine and mitoxantrone; however, the same starting dose of GMTZ in combination with the Capizzi II regimen led to DLTs and a dose de-escalation of GMTZ to 2 mg/m2 for this combination [51]. Many novel agents of interest in relapsed ALL similarly carry risks of significant toxicities when administered as either a single agent or in combination. A platform approach similar to that in AALL01P2 was successfully used in a multicenter phase
70
J.A. Whitlock and T.M. Horton
I trial of bortezomib for marrow relapse of childhood ALL [52] and will be applied in upcoming studies evaluating bortezomib and temsirolimus in relapsed childhood ALL in the COG.
4.4 Approaches to the Determination of DLTs in Multiagent Phase I Trials Toxicities experienced by patients receiving Block 1 chemotherapy in AALL01P2 (vincristine, prednisone, l-asparaginase, and doxorubicin) are summarized in Table 4.1. Although patients and their clinicians are willing to accept significant toxicities to obtain remission, these rates of toxicity associated with the reinduction platform can easily obfuscate toxicities attributable to the novel agent. As shown in Table 4.1, the total number of Grade 3 and 4 toxicities (excluding febrile neutropenia) totaled 120%, with 67% of patients experiencing at least one severe or lifethreatening nonhematologic toxicity. This toxicity rate clearly exceeds the conventional rate of acceptable severe toxicities for single agents tested in dose-finding clinical trials, which is traditionally set at <2/6 toxicities or 33% (see above). It is likely that the toxicities associated with the reinduction platform, which would be unrelated to the new agent, would inappropriately limit new-agent dose escalation. A common definition of nonhematologic DLT is listed in Table 4.2. In many cases, there would be no dose of a new, potentially efficacious agent that would meet conventional DLT rates in the context of a reinduction platform such as AALL01P2. For example, if the backbone toxicity rate is 20%, the chance of finding no acceptable dose for a new agent is 20%. However, it the backbone toxicity rate increases to 24% or greater, there would be no acceptable dose of an experimental agent in over half the clinical trials testing the new agent (Fig. 4.2). Since the backbone toxicity rate will have significant impact on dose finding, alternative DLT definitions have been developed that will enable defining an appropriate dose of a new agent without either underdosing due to the background toxicity, or overdosing resulting in an unacceptable increase in morbidity.
Table 4.1 Nonhematologic Grade 3 and 4 toxicities that occurred in more than 5% of the 109 patients receiving VPLD in COG AALL01P2 Hypertension 7% Elevated ALT 13% Low fibrinogen 32% Febrile neutropenia 58% Pancreatitis 7% Hyperglycemia 11% Stomatitis 6% Hypocalcemia 8% Hemorrhage 6% Hypokalemia 7% Elevated bilirubin 6% Elevated lipase 8% Hypoalbuminemia 9%
4 Design of Early-Phase Trials
71
Table 4.2 Conventional nonhematologic DLT definition Although the DLT definitions used by different clinical trial consortia vary, many are similar to the one used in USA by the Children’s Oncology Group (COG) Nonhematologic dose-limiting toxicity: Any Grade 3 or greater nonhematological toxicity attributable to the investigational drug Common exclusions Grade 3 and 4 nausea and vomiting Grade 3 transaminase (AST/ALT) elevation that returns to Grade £1 or baseline prior to the next treatment cycle Grade 3 fever or infection Electrolyte disturbances due to tumor lysis syndrome Alopecia
Chance of rejecting all doses
100%
80%
60%
40%
20%
0% 0%
20%
40%
60%
80%
100%
Background toxicity rate Fig. 4.2 Probability of rejecting all doses of a new agent X when agent X is added to a backbone chemotherapy regimen but adds negligible toxicity to the combination, as a function of background toxicity rate. Figure courtesy of Rich Sposto
4.4.1 Approaches to Defining Novel Agent Nonhematologic DLT 1. Exclusion approach with tolerance of an increased DLT rate: Toxicities that are common in the backbone chemotherapy can be excluded from the DLT definition. For example, a study could exclude all toxicities common to the backbone regimen (i.e., all toxicities in Table 4.1) and all other tolerable Grade 3 and 4 toxicities, such as resolving metabolic abnormalities and resolving transaminase elevations. The DLT rate with these exclusions would decrease to 33%. The rate of toxicities would be considered dose-limiting only if it exceeded an increased DLT threshold, such as 50–55%.
72
J.A. Whitlock and T.M. Horton
Although this is an exact approach, there are at least three challenges with this approach. First, it does not exclude those toxicities that occur at a low frequency (1–5%) in the reinduction platform. Although these toxicities are uncommon as individual events, they sum to an additional 33% toxicity. Second, this approach does not account for the ability of the new agent to increase the frequency of a known toxicity, since these toxicities would be excluded from the DLT definition. Third, it does not take into account that a significant increase in the frequency of some of the common toxicities seen with the backbone regimen (i.e., pancreatitis or hemorrhage) would be considered intolerable. 2. Functional DLT approach: With some exceptions (nausea, vomiting, anorexia, fatigue, infection, fever, resolving metabolic abnormalities, and hypofibrinogenemia), Grade 4 (life-threatening) toxicities would be considered intolerable. Many Grade 3 toxicities, however, can be tolerated if they do not impact the ability to provide timely delivery of the reinduction platform. An alternative approach to defining new-agent DLTs is to include life-threatening toxicities, with the few exception listed above, as DLTs. However, this approach would only consider a Grade 3 toxicity as a DLT if it was of sufficient severity to substantially delay administration of the platform regimen for >7 (or 14) days. Although limited information is available on toxicity resolution in AALL01P2, most Grade 3 toxicities (with the exception of neutropenic fever) resolved within 7 days. Using this approach to defining a DLT, the baseline toxicity rate in the AALL01P2 study would decrease to 9–16%. This approach has several advantages over approach 1 and is being tested in two active COG clinical trials. The advantages of this approach include (a) being more tolerant of the reporting bias between historical phase II and III studies, and phase I studies including new agents, (b) being able to detect increases in Grade 3 and 4 toxicities due to the new agent if these toxicities are already present in the reinduction platform, and (c) being tolerant of toxicities that do not delay administration of the reinduction platform, which contains efficacious agents. Reporting bias is common in phase II and III trials. The incidence of Grade 3 and 4 toxicities is often substantially underreported in historical control studies with reinduction agents, yet these same toxicities are vigilantly reported in phase 1 trials that include new agents. Since centers are more likely to report all severe toxicities when careful monitoring is mandated, there is likely to be a substantial increase in the perceived rate of severe toxicities using a reinduction platform. However, since severe toxicities count as DLTs only if they delay delivery of the reinduction platform, this method would compensate for increases in event reporting. A functional DLT definition also takes into account the ability of new agents to exacerbate toxicities associated with the reinduction platform. This is an important advantage. As illustrated in the relapsed AML clinical trial AAML00P2, which incorporated gemtuzumab ozogamicin into a platform Capizzi II regimen, the addition of an experimental agent to a reinduction platform can be associated with significant toxicities (see above). It is difficult to know a priori which toxicities might be exacerbated by a new agent. This approach to DLT definition can account for the potential of a new agent to potentiate toxicity as well as enhance efficacy.
4 Design of Early-Phase Trials
73
Since it is unknown if any new agent will improve efficacy, it is important not to delay administration of the reinduction platform, which is known to be effective for relapsed ALL. Hence, this definition, which considers Grade 3 toxicities DLTs only if they delay the administration of Block 2 chemotherapy, puts an emphasis on timely delivery of therapy with known efficacy. Development of a “functional” DLT algorithm for use in high-risk ALL clinical trials Using the functional DLT definition, the following algorithm was developed for evaluating DLTs when a novel agent is combined with a reinduction platform: First, tolerable toxicities are defined as exclusion sets: Exclusion set A: Acceptable Grade 3 or greater nonheme toxicities expected to occur with the reinduction platform alone at a frequency of 7–10% or greater. Exclusion set B: Acceptable Grade 3 or greater nonheme toxicities expected to occur with the novel agent alone at a frequency of 10% or greater. Exclusion set C: Grade 3 toxicities that are acceptable and result in delay/omission of subsequent backbone chemotherapy for ³7 days and occur with a frequency of ³10% (e.g., Grade 3 or 4 infections). Next, DLTs are defined as nonexcluded Grade 4 toxicities and nonexcluded Grade 3 toxicities that delay administration of the reinduction platform by more than 7 days: DLT definition: (a) Any Grade 4, nonheme toxicity possibly, probably or definitely related to study drug, with the exception of exclusion set A and B, and (b) Any Grade 3 or 4 nonheme toxicity that results in omission/delay of the subsequent course of chemotherapy for >7 days, with the exception of exclusion set C. As an additional safety mechanism, the two COG trials incorporating this DLT definition for high-risk ALL patients also include a stopping rule to monitor for the potential increased incidence of severe toxicities or increased toxic death rate above that expected from the reinduction platform. This provides another safety measure, halting the study if unanticipated severe toxicities occur. Functional DLT definition in practice With the goal of improving efficacy for children with high-risk ALL, two clinical trials have used the above functional DLT definition for incorporation of new agents into reinduction platforms: 1. FLT3 inhibition for infants with ALL: A current phase III COG study (AALL0631) incorporates the FLT3 inhibitor lestaurtinib into a multiagent chemotherapy platform for very-high-risk infants. A similar historical trial (POG-9407) assessed severe toxicities with the same backbone regimen. Those toxicities with a frequency of >10% included stomatitis (24%), transaminase elevations (17%), diarrhea (17%), neurologic toxicities (10%), constitutional symptoms, transient laboratory abnormalities (electrolytes, coagulation abnormalities and liver function tests),
74
J.A. Whitlock and T.M. Horton
transient blood pressure perturbations, skin toxicity, and tumor lysis syndrome. These toxicities were designated as Exclusion Set A. Lestaurtinib toxicities have been previously assessed in adults and in pediatric neuroblastoma study NANT N2007-001. Toxicities associated with lestaurtinib included vomiting, anorexia, nausea, and transient transaminase elevations. These were designated as Exclusion Set B. In the POG-9407 study, most severe toxicities resolved within 7 days, with the exception of febrile neutropenia and infections. These were designated as Exclusion Set C. These exclusion sets were combined to develop the following DLT definition: Nonhematologic DLT: • A ny Grade 4 nonhematologic toxicity that occurs after the first dose of lestaurtinib and is at least possibly related to lestaurtinib, with the following exceptions: –– Febrile neutropenia or infection. –– Constitutional symptoms (e.g., fatigue, fever, weight change). –– Metabolic/laboratory abnormalities that resolve to £Grade 2 within: • 14 days, for ALT/SGPT, AST/SGOT, alkaline phosphatase. • 7 days, for amylase, lipase, total bilirubin. • 48 h for all other laboratory abnormalities. –– Coagulation abnormalities (INR, PTT, or fibrinogen) that resolve to £Grade 2 within 48 h. • Any Grade 3 nonhematologic toxicity that occurs after the first dose of lestaurtinib, is at least possibly attributable to lestaurtinib, and results in omission or delay of the beginning of the subsequent course of chemotherapy for greater than 7 days, with the following exceptions: –– Febrile neutropenia or infection. –– Mucositis or diaper area skin breakdown must result in omission or delay of the beginning of the subsequent course of chemotherapy for greater than 14 days to be considered a DLT. Hematologic DLT: No hematologic toxicity will be considered a DLT while a patient is receiving scheduled doses of lestaurtinib. After completion of a lestaurtinib course, persistent Grade 3 or greater neutrophils and/or platelets that result in a greater than 21-day delay in the start of the following course of chemotherapy will be considered a DLT, unless the delay in neutrophil or platelet recovery is due to another clearly identifiable factor such as relapse or myelosuppressive infection. 2. Proteasome inhibition for children with relapsed ALL: The proteasome inhibitor bortezomib is being tested in a COG pilot phase II clinical trial (AALL07P1) that uses the same reinduction platform as the historical control trial AALL01P2 (9).
4 Design of Early-Phase Trials
75
Both nonhematologic and hematologic toxicities for the historical control study have been well characterized (Table 4.1), including the average time to ANC and platelet recovery. Although severe toxicities in the historical control trial were quite frequent, few toxicities (with the exception of infection) resulted in delay in therapy administration beyond 7 days. The DLT definition for AALL07P1 is summarized below: Nonhematologic DLT • Any Grade 4 nonhematologic toxicity that occurs after the first dose of bortezomib and is at least possibly attributable to bortezomib, with the following exceptions: • Any Grade 3 or 4 nonhematologic toxicity that occurs after the first dose of bortezomib is at least possibly attributable to bortezomib and results in omission or delay of the beginning of the subsequent course of chemotherapy for greater than 7 days, with the exception of the following. • Fever or infection • Gastrointestinal symptoms (anorexia, nausea, vomiting, dehydration, mucositis) • Constitutional symptoms (fatigue, anorexia, malaise) • Hypofibrinogenemia • Metabolic/laboratory abnormalities that resolve to
76
J.A. Whitlock and T.M. Horton
at the completion of the first course of therapy. In both examples, the new agent would be dose reduced if four or more patients of the first 12 experience prolonged ANC or platelet recovery.
4.5 Phase II Trials and Trial Design Issues 4.5.1 Single-Arm Phase II Design Following identification of the MTD and RPIID of a novel agent administered singly or in combination, the activity of the new agent or combination against a specific type of tumor must be assessed in a phase II trial. The most widely used phase II trial design is the two-stage design developed by Simon [53]. The twostage design seeks to optimize the study population by minimizing the sample size required to identify agents with low activity. A typical single-arm phase II study enrolls 14–60 patients [54]. Promising results in a single-arm phase II trial may lead to further investigation of the novel agent or combination against standard therapy in a randomized phase III study. However, interpretation of uncontrolled phase II studies is fraught with risks. Overt or covert bias in subject selection frequently occurs in uncontrolled phase II studies. The danger of overinterpreting the results of uncontrolled single-arm phase II studies of novel regimens is perhaps best illustrated by multiple reports of phase II studies in the 1980s of so-called “third-generation” chemotherapy regimens for adult non-Hodgkin’s lymphomas, which reported higher response rates in single-arm phase II trials than those historically seen with the standard regimen of CHOP. A subsequent randomized phase III study that compared three of these regimens with CHOP determined that none offered superior outcomes to CHOP, but were each associated with worse toxicity profiles [55] Estey and Thall refer to the confounding variables inherent in separate single-arm phase II studies as “trial effects,” in contrast to “treatment effects” [54].
4.5.2 Randomized Phase II Designs The inherent bias frequently seen in uncontrolled phase II trials is best controlled through randomization. Randomization is a required element in phase III trial design, when a novel therapy is compared to a standard therapy. However, randomization can also be applied to phase II trials, when the desired outcome is to select one of two (or more) experimental regimens to take forward to a phase III trial. The statistical expectations, and thus the number of required subjects, in a randomized phase II trial are less than those of a phase III trial. As described by Steinberg and Venzon, “The arms can be conducted as if they were independent phase II trials, but the final results will be considered jointly for ranking the therapies or choosing one
4 Design of Early-Phase Trials
77
for further study. The purpose of this type of design is to simultaneously evaluate the efficacy of the therapies but not to formally compare them with the strict power and significance level of a phase III trial. This relaxation of requirements is permitted because one is merely making a rational choice of one arm and is not establishing statistically that the arm selected is superior [56].” However, even with the smaller sample size required of a randomized phase II trial design, the number of patients required may exceed those available in a reasonable period of time for relatively uncommon situations such as relapsed childhood ALL. Thus, increasing the efficiency of the randomized phase II design has been proposed through statistical modifications that allow for early selection of a “winner” [56]. Another approach to improving phase II study design is the phase II screening trial design, which restricts the sample size through adjustment of the false-positive error rates (alpha or type I error) and false-negative error rates (beta or type II error) [57]. This approach can more efficiently identify both highly active and inactive regimens, thus allowing the more rapid conduct of studies and the more efficient evaluation of novel agents. However, investigators must resist a temptation to directly compare the results of the two arms (either two experimental arms or an experimental arm and a standard control arm), as this violates the intent of the study design and can lead to selection of an inferior treatment. Other limitations in the application of the phase II screening design include the necessity of having two or more new agents/regimens simultaneously available for phase II testing following completion of phase I testing, and the very real obstacle of convincing pharmaceutical sponsors to provide an investigational agent for inclusion in a “pick-the-winner” drug development strategy [57]. Finally, it should be noted that none of these modifications of the traditional phase II single-arm trial are sufficient substitutes for a properly designed phase III trial.
4.5.3 Phase II End Points in Relapsed Childhood ALL The conventional phase II study end point is response rate; in relapsed acute leukemia trials, this typically equates to rate of CR, which remains the most clinically relevant benchmark for assessment of efficacy. CR is typically defined as a reduction in leukemic blasts to less than 5% as assessed by light microscopic evaluation of marrow morphology, together with recovery of normal hematopoiesis as measured by absolute neutrophil count and untransfused platelet count [58]. The more recent, less restrictive measure of complete response with insufficient platelet recovery (CRp) is increasingly utilized in relapse trials for both childhood AML [50] and childhood ALL [59]. Other response assessments that have not been validated include partial remission (defined as a ³50% decrease in marrow blasts with normalization of peripheral counts) [60] and partial remission-cytolytic (defined as complete disappearance of circulating blasts and achievement of at least 50% reduction from baseline in bone marrow blast count) [46].
78
J.A. Whitlock and T.M. Horton
End-reinduction MRD can reflect a drug’s activity against acute leukemia. MRD status may be determined by molecular [61] or flow cytometry [62] methods. Both retrospective [41] and prospective [4] analyses of MRD levels in relapsed childhood ALL have demonstrated prognostic significance. However, further studies that validate end-induction MRD as a surrogate for clinical response will be required before this measure is accepted by regulatory agents such as the Food and Drug Administration [60]. For early relapse patients who will undergo allogeneic SCT, the ability to not only attain remission but also remain in remission until the time of transplant is a pragmatic measure of successful therapy, which may be objectively captured by short-term event-free survival rates. In COG ADVL04P2, 4-month EFS serves as an objective end point, together with CR2 rate and MRD burden after Block 1, in the evaluation of new agents using this platform approach. The 4-month EFS end point allows determination of the contribution of a new agent to short-term disease control with the three-block approach of AALL01P2, without the contaminating influences of subsequent allo-SCT-related toxicities. While time to transplantation may have clear clinical relevance in assessing the efficacy of a new agent or new treatment for relapsed leukemia, it is unlikely to be meaningful if used as the sole measure of benefit [60]. In summary, the efficient and effective evaluation of new agents in the treatment of ALL requires study designs that acknowledge the limitations of traditional single-agent approaches to therapy evaluation in this population, while providing accurate assessment of the safety and the efficacy of novel agents. Study designs that accomplish these goals have been developed and implemented in recent and ongoing clinical trials.
References 1. Schultz KR, Pullen DJ, Sather HN, et al. Risk- and response-based classification of childhood B-precursor acute lymphoblastic leukemia: a combined analysis of prognostic markers from the Pediatric Oncology Group (POG) and Children’s Cancer Group (CCG). Blood. 2007;109:926–935. 2. Gaynon PS, Camitta BC, Matloub Y, et al. Outcomes for B-precursor patients in Legacy Children’s Cancer Group (CCG) and Pediatric Oncology Group (POG) studies in childhood acute lymphoblastic leukemia (ALL): A Children’s Oncology Group (COG) report. ASH Annual Meeting Abstracts. 2007;110:847. 3. Gaynon PS, Harris RE, Altman AJ, et al. Bone marrow transplantation versus prolonged intensive chemotherapy for children with acute lymphoblastic leukemia and an initial bone marrow relapse within 12 months of the completion of primary therapy: Children’s Oncology Group study CCG-1941. J Clin Oncol. 2006;24:3150–3156. 4. Raetz EA, Borowitz MJ, Devidas M, et al. Reinduction Platform for Children With First Marrow Relapse of Acute Lymphoblastic Leukemia: A Children’s Oncology Group Study. J Clin Oncol. 2008;26:3971–3978. 5. Gaynon PS. Childhood acute lymphoblastic Leukemia and relapse. Br J Haematol. 2005;131:579–587. 6. Gaynon PS, Qu RP, Chappell RJ, et al. Survival after relapse in childhood acute lymphoblastic leukemia: impact of site and time to first relapse--the Children’s Cancer Group Experience. Cancer. 1998;82:1387–1395.
4 Design of Early-Phase Trials
79
7. Nguyen K, Devidas M, Cheng SC, et al. Factors influencing survival after relapse from acute lymphoblastic leukemia: a Children’s Oncology Group study. Leukemia. 2008;22: 2142–2150. 8. Houghton PJ, Morton CL, Tucker C, et al. The pediatric preclinical testing program: description of models and early testing results. Pediatr Blood Cancer. 2007;49:928–940. 9. Houghton PJ, Morton CL, Kolb EA, et al. Initial testing (stage 1) of the proteasome inhibitor bortezomib by the pediatric preclinical testing program. Pediatr Blood Cancer. 2008;50:37–45. 10. Houghton PJ, Morton CL, Kolb EA, et al. Initial testing (stage 1) of the mTOR inhibitor rapamycin by the pediatric preclinical testing program. Pediatr Blood Cancer. 2008; 50:799–805. 11. Kolb EA, Gorlick R, Houghton PJ, et al. Initial testing of dasatinib by the pediatric preclinical testing program. Pediatr Blood Cancer. 2008;50:1198–1206. 12. Lock R, Carol H, Houghton PJ, et al. Initial testing (stage 1) of the BH3 mimetic ABT-263 by the pediatric preclinical testing program. Pediatr Blood Cancer. 2008;50:1181–1189. 13. Collins JM, Grieshaber CK, Chabner BA. Pharmacologically guided phase I clinical trials based upon preclinical drug development. J Natl Cancer Inst. 1990;82:1321–1326. 14. Smith M, Bernstein M, Bleyer WA, et al. Conduct of phase I trials in children with cancer. J Clin Oncol. 1998;16:966–978. 15. Storer BE. Design and analysis of phase I clinical trials. Biometrics. 1989;45:925–937. 16. Marsoni S, Ungerleider RS, Hurson SB, Simon RM, Hammershaimb LD. Tolerance to antineoplastic agents in children and adults. Cancer Treat Rep. 1985;69:1263–1269. 17. Ross L. Phase I research and the meaning of direct benefit. J Pediatr. 2006;149:S20-24. 18. Ratain MJ, Mick R, Schilsky RL, Siegler M. Statistical and ethical issues in the design and conduct of phase I and II clinical trials of new anticancer agents. J Natl Cancer Inst. 1993;85:1637–1643. 19. O’Quigley J, Pepe M, Fisher L. Continual reassessment method: a practical design for phase 1 clinical trials in cancer. Biometrics. 1990;46:33–48. 20. Goodman SN, Zahurak ML, Piantadosi S. Some practical improvements in the continual reassessment method for phase I studies. Stat Med. 1995;14:1149–1161. 21. Moller S. An extension of the continual reassessment methods using a preliminary up-anddown design in a dose finding study in cancer patients, in order to investigate a greater range of doses. Stat Med. 1995;14:911–922; discussion 923. 22. Simon R, Freidlin B, Rubinstein L, Arbuck SG, Collins J, Christian MC. Accelerated titration designs for phase I clinical trials in oncology. J Natl Cancer Inst. 1997;89:1138–1147. 23. Loke YC, Tan SB, Cai Y, Machin D. A Bayesian dose finding design for dual endpoint phase I trials. Stat Med. 2006;25:3–22. 24. Shah S, Weitman S, Langevin AM, Bernstein M, Furman W, Pratt C. Phase I therapy trials in children with cancer. J Pediatr Hematol Oncol. 1998;20:431–438. 25. Lee DP, Skolnik JM, Adamson PC. Pediatric phase I trials in oncology: an analysis of study conduct efficiency. J Clin Oncol. 2005;23:8431–8441. 26. Skolnik JM, Barrett JS, Jayaraman B, Patel D, Adamson PC. Shortening the timeline of pediatric phase I trials: the rolling six design. J Clin Oncol. 2008;26:190–195. 27. Reaman GH, Ladisch S, Echelberger C, Poplack DG. Improved treatment results in the management of single and multiple relapses of acute lymphoblastic leukemia. Cancer. 1980;45:3090–3094. 28. Buchanan GR, Rivera GK, Boyett JM, Chauvenet AR, Crist WM, Vietti TJ. Reinduction therapy in 297 children with acute lymphoblastic leukemia in first bone marrow relapse: a Pediatric Oncology Group Study. Blood. 1988;72:1286–1292. 29. Feig SA, Ames MM, Sather HN, et al. Comparison of idarubicin to daunomycin in a randomized multidrug treatment of childhood acute lymphoblastic leukemia at first bone marrow relapse: a report from the Children’s Cancer Group. Med Pediatr Oncol. 1996;27:505–514. 30. Giona F, Testi AM, Rondelli R, et al. ALL R-87 protocol in the treatment of children with acute lymphoblastic leukemia in early bone marrow relapse. Br J Haematol. 1997;99:671–677.
80
J.A. Whitlock and T.M. Horton
31. Abshire TC, Pollock BH, Billett AL, Bradley P, Buchanan GR. Weekly polyethylene glycol conjugated L-asparaginase compared with biweekly dosing produces superior induction remission rates in childhood relapsed acute lymphoblastic leukemia: a Pediatric Oncology Group Study. Blood. 2000;96:1709–1715. 32. Lawson SE, Harrison G, Richards S, et al. The UK experience in treating relapsed childhood acute lymphoblastic leukemia: a report on the medical research council UKALLR1 study. Br J Haematol. 2000;108:531–543. 33. Leahey AM, Bunin NJ, Belasco JB, Meek R, Scher C, Lange BJ. Novel multiagent chemotherapy for bone marrow relapse of pediatric acute lymphoblastic leukemia. Med Pediatr Oncol. 2000;34:313–318. 34. Testi AM, Del Giudice I, Arcese W, et al. A single high dose of idarubicin combined with high-dose ARA-C for treatment of first relapse in childhood ‘high-risk’ acute lymphoblastic leukemia: a study of the AIEOP group. Br J Haematol. 2002;118:741–747. 35. Thomson B, Park JR, Felgenhauer J, et al. Toxicity and efficacy of intensive chemotherapy for children with acute lymphoblastic leukemia (ALL) after first bone marrow or extramedullary relapse. Pediatr Blood Cancer. 2004;43:571–579. 36. Einsiedel HG, von Stackelberg A, Hartmann R, et al. Long-term outcome in children with relapsed ALL by risk-stratified salvage therapy: results of trial acute lymphoblastic leukemiarelapse study of the Berlin-Frankfurt-Munster Group 87. J Clin Oncol. 2005;23:7942–7950. 37. Roy A, Cargill A, Love S, et al. Outcome after first relapse in childhood acute lymphoblastic leukemia - lessons from the United Kingdom R2 trial. Br J Haematol. 2005;130:67–75. 38. Bhojwani D, Kang H, Moskowitz NP, et al. Biologic pathways associated with relapse in childhood acute lymphoblastic leukemia: a Children’s Oncology Group study. Blood. 2006;108:711–717. 39. Wheeler K, Richards S, Bailey C, Chessells J. Comparison of bone marrow transplant and chemotherapy for relapsed childhood acute lymphoblastic leukemia: the MRC UKALL X experience. Medical Research Council Working Party on Childhood Leukemia. Br J Haematol. 1998;101:94–103. 40. Harrison G, Richards S, Lawson S, et al. Comparison of allogeneic transplant versus chemotherapy for relapsed childhood acute lymphoblastic leukemia in the MRC UKALL R1 trial. MRC Childhood Leukemia Working Party. Ann Oncol. 2000;11:999–1006. 41. Eckert C, Biondi A, Seeger K, et al. Prognostic value of minimal residual disease in relapsed childhood acute lymphoblastic leukemia. Lancet. 2001;358:1239–1241. 42. Coustan-Smith E, Gajjar A, Hijiya N, et al. Clinical significance of minimal residual disease in childhood acute lymphoblastic leukemia after first relapse. Leukemia. 2004;18:499–504. 43. Crooks GM, Sato JK. Ifosfamide and etoposide in recurrent childhood acute lymphoblastic leukemia. J Pediatr Hematol Oncol. 1995;17:34–38. 44. von Stackelberg A, Hartmann R, Buhrer C, et al. High-dose compared with intermediate-dose methotrexate in children with a first relapse of acute lymphoblastic leukemia. Blood. 2008;111:2573–2580. 45. Wells RJ, Feusner J, Devney R, et al. Sequential high-dose cytosine arabinoside-asparaginase treatment in advanced childhood leukemia. J Clin Oncol. 1985;3:998–1004. 46. Raetz EA, Cairo MS, Borowitz MJ, et al. Chemoimmunotherapy Reinduction With Epratuzumab in Children With Acute Lymphoblastic Leukemia in Marrow Relapse: A Children’s Oncology Group Pilot Study. J Clin Oncol. 2008;26:3756–3762. 47. Lager JJ, Lyden ER, Anderson JR, Pappo AS, Meyer WH, Breitfeld PP. Pooled analysis of phase II window studies in children with contemporary high-risk metastatic rhabdomyosarcoma: a report from the Soft Tissue Sarcoma Committee of the Children’s Oncology Group. J Clin Oncol. 2006;24:3415–3422. 48. Niemeyer CM, Gelber RD, Tarbell NJ, et al. Low-dose versus high-dose methotrexate during remission induction in childhood acute lymphoblastic leukemia (Protocol 81-01 update). Blood. 1991;78:2514–2519. 49. Schorin M, Blattner S, Gelber R, et al. Treatment of childhood acute lymphoblastic leukemia: results of Dana- Farber Cancer Institute/Children’s Hospital Acute Lymphoblastic Leukemia Consortium Protocol 85-01. J Clin Oncol. 1994;12:740–747.
4 Design of Early-Phase Trials
81
50. Sievers EL, Larson RA, Stadtmauer EA, et al. Efficacy and safety of gemtuzumab ozogamicin in patients with CD33-positive acute myeloid leukemia in first relapse. J Clin Oncol. 2001;19:3244–3254. 51. Aplenc R, Alonzo TA, Gerbing RB, et al. Safety and efficacy of gemtuzumab ozogamicin in combination with chemotherapy for pediatric acute myeloid leukemia: a report from the Children’s Oncology Group. J Clin Oncol. 2008;26:2390–3295. 52. Messinger YH, Gaynon PS, Raetz E, et al. Remarkable Activity of Bortezomib Combined with Chemotherapy in a Phase I Study of Relapsed Childhood Acute Lymphoblastic Leukemia (ALL). A Report from the Therapeutic Advances in Childhood Leukemia (TACL) Consortium. ASH Annual Meeting Abstracts. 2008;112:1919. 53. Simon R. Optimal two-stage designs for phase II clinical trials. Control Clin Trials. 1989;10:1–10. 54. Estey EH, Thall PF. New designs for phase 2 clinical trials. Blood. 2003;102:442–448. 55. Fisher RI, Gaynor ER, Dahlberg S, et al. Comparison of a standard regimen (CHOP) with three intensive chemotherapy regimens for advanced non-Hodgkin’s lymphoma. N Engl J Med. 1993;328:1002–1006. 56. Steinberg SM, Venzon DJ. Early selection in a randomized phase II clinical trial. Stat Med. 2002;21:1711–1726. 57. Rubinstein LV, Korn EL, Freidlin B, Hunsberger S, Ivy SP, Smith MA. Design issues of randomized phase II trials and a proposal for phase II screening trials. J Clin Oncol. 2005;23:7199–7206. 58. Cheson BD, Bennett JM, Kopecky KJ, et al. Revised recommendations of the International Working Group for Diagnosis, Standardization of Response Criteria, Treatment Outcomes, and Reporting Standards for Therapeutic Trials in Acute Myeloid Leukemia. J Clin Oncol. 2003;21:4642–4649. 59. Jeha S, Gaynon PS, Razzouk BI, et al. Phase II study of clofarabine in pediatric patients with refractory or relapsed acute lymphoblastic leukemia. J Clin Oncol. 2006;24:1917–1923. 60. Appelbaum FR, Rosenblum D, Arceci RJ, et al. End points to establish the efficacy of new agents in the treatment of acute leukemia. Blood. 2007;109:1810–1816. 61. Cazzaniga G, Biondi A. Molecular monitoring of childhood acute lymphoblastic leukemia using antigen receptor gene rearrangements and quantitative polymerase chain reaction technology. Haematologica. 2005;90:382–390. 62. Campana D, Coustan-Smith E. Minimal residual disease studies by flow cytometry in acute leukemia. Acta Haematol. 2004;112:8–15.
Chapter 5
Strategies for Trial Design and Analyses Maria Grazia Valsecchi and Paola De Lorenzo
5.1 Experimental Designs in Clinical Research Well-designed experimental research is a necessary basis for therapeutic development and clinical care decisions. An experiment may be defined as a series of carefully collected observations made under conditions that are controlled or arranged by the investigator [1]. Its counterpart in clinical research is the clinical trial. Although sharing essential features with the classic experimental studies, clinical trials differ markedly by their own nature of studies involving human subjects. Consider first that a clinical trial must 1. Examine valuable and feasible research questions (e.g. determine the effectiveness of a treatment in a specific disease) 2. Be based on a rigorous methodology, suitable to answer the specific research question being asked 3. Respect ethical constraints, adherence to which minimizes risks to the safety and well-being of individuals Peculiarities of studies on human subjects arise from all of these three aspects, which are in fact closely interrelated. For instance, the number and characteristics of observational units included in the study cannot be chosen or enlarged as much as scientifically desirable. This constraint is especially present in targeted therapies that tend to be directed to rare subtypes of cancers, both in paediatric and adult oncology. In addition, human responses to interventions are usually found to be considerably more variable or delayed in time than responses obtained from fully controlled experiments in laboratories. Such constraints are to be taken into account right from the design of the study, but typically influence also its conduct, that is, the execution of the experiment. Both are crucial phases regulated by rigorous
M.G. Valsecchi (*) Centre of Biostatistics for Clinical Epidemiology, Department of Clinical Medicine and Prevention, University of Milano-Bicocca, Via Cadore, 48, 20052 Monza (MI), Italy e-mail:
[email protected] V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3_5, © Springer Science+Business Media, LLC 2011
83
84
M.G. Valsecchi and P. De Lorenzo
procedures aimed at ensuring that valid conclusions can be drawn from the study. Procedures are set up with two major aims: minimizing biases (i.e. systematic errors that may deviate the estimate of interest from its “true” unknown value) and maximizing precision (i.e. accuracy of estimates, through control of random errors with appropriate sample size). Design and conduct are followed by the analysis and interpretation of results: the latter depends heavily on the former and cannot reliably fix major faults that occurred earlier in the study. Clinical trials are traditionally classified into four phases, from Phase I to Phase IV, following the stages of pharmacological development of new drugs for human use. Phase I is the earliest stage carried out to investigate pharmacokinetics and safety of the new product to establish the most tolerable dose. Phase II trials study the safety and the activity profiles on a larger scale and typically lead to Phase III trials, in which the new treatment is compared to an alternative (an established “standard of care” or placebo) with the aim to demonstrate improved efficacy or increased safety with equal efficacy. Phase IV trials are performed, usually on an observational basis, after the new intervention has received regulatory approval, and aim at obtaining additional information about its risks (e.g. rare but serious side effects, long-term side effects), benefits (long-term outcome) and optimal use. Of note, in these definitions, the new product’s beneficial property is referred to as either activity or efficacy, with two different meanings. The former relates to disease control, while the latter refers to clinical outcome. Examples are the impact of a drug/therapeutic strategy on response, either morphological or molecular (activity), as compared to the impact on event-free survival (EFS) (efficacy, see also Sect. 5.4). The Phase I–IV classification has recently been criticized because it does not provide a general terminology able to include the development of medical products that do not necessarily fit that of traditional pharmaceuticals (e.g. targeted agents). Piantadosi [2] proposed a more flexible and general approach classifying studies on the basis of their principal aims: (a) Early-development studies (translational or treatment-mechanism testing studies), (b) Middle-development studies (assessing treatment tolerability), (c) Late-development studies (including comparative studies as well as expanded safety or post-marketing studies). The traditional classification is easily recognized into this new “descriptive” one. Early-development studies include translational trials. The name reflects their purpose: connecting laboratory findings to clinic and vice versa, in a circular fashion that enables laboratory research to incorporate clinical evidence. A more traditional early phase study is the one focussed on understanding the treatment mechanism (i.e. bioavailability of the new compound) and the dose-safety profile, in a context guided by a biological model. Clinical outcome of the new treatment is formally introduced in Middle-development studies, where its relationship to safety and activity is studied. This stage is particularly important when various new compounds/interventions are available and it is necessary to select out the more promising ones. This may be the case in cancer research and in fact oncology Phase II trials are a typical example. Given the developmental purpose, they usually involve a broader definition of activity, for instance utilizing biomarkers or an early
5 Strategies for Trial Design and Analyses
85
response (e.g. complete remission rate in leukemia). In specific circumstances, Middle-development studies can comprise a randomized question, when there are two or more new promising treatments and the primary objective is to select the one to be considered for a successive comparative trial. So-called Phase II–III trials are an example (see Sect. 5.5). Comparative trials are the most important among Latedevelopment studies and are focussed on questions of comparison of treatment efficacy (as such they correspond to Phase III trials, according to the classical trials’ categories). These represent the gold standard for gathering evidence in medical research as they are designed to minimize bias through control over treatment assignment, systematic errors and ascertainment of end points and are completed by predefined analysis [3]. Replication of trials and consolidation of their results with meta-analyses allow to further increase credibility of conclusions. A key feature in comparative trials is equipoise [4]. Equipoise is defined as the state in which any rational, informed person (patient, clinician or researcher) does not favour a treatment over the other(s) being studied. This condition of genuine uncertainty is fundamental in clinical research comprising a randomization procedure. In the following sections, we use either the traditional or the descriptive classification, depending on the context, for the sake of clarity.
5.2 Peculiarities of Studies in Lymphoblastic Leukemia Starting from the 1940s, when lymphoblastic leukemia was almost invariably a fatal disease, clinical trials have addressed questions about the role of various combinations -and intensities of chemotherapic agents and of radiotherapy. Long-term survival from acute lymphoblastic leukemia (ALL) is nowadays in the order of 80 and 40% in children and adults, respectively. Despite the improvements achieved in the last decades, a pressing need for further research on therapeutic approaches remains, even in childhood ALL, basically in three different directions. The first line of research deals with sub-populations of patients who present features related to a high risk of failing current frontline or relapse therapies. An example of a high-risk population is that of children with ALL who have persistence of disease, either at molecular or at morphological level, after induction and consolidation phases in frontline therapy. Typically, these high-risk sub-populations are small and make a challenging target for trialists that need adequate sample size to provide evidence on treatment effect [5]. These are, however, sub-populations where there is room for marked improvements and where new promising drugs could be tested. In childhood ALL, for instance, patients with persistence of disease have a dismal prognosis even with allogeneic transplantation. They are eligible for Phase II or Phase III studies that include novel drugs that have already gone through early phases of clinical development in adults. Characterization of rare sub-populations results also from the development of targeted therapies, directed to patients with specific genetic abnormalities. An example in childhood ALL is the subgroup of Ph+ patients, accounting for only
86
M.G. Valsecchi and P. De Lorenzo
3% of the ALL population, who are candidates for treatment with imatinib or second-generation tyrosine kinase inhibitors. Going on Phase II–III studies are aiming to evaluate if the use of these new drugs can markedly improve the historical 50% 4-year EFS of Ph+ ALL children. The second direction of research is that of Phase III studies that aim at modest, but clinically important outcome improvements (efficacy), with more efficient definitions of therapeutic strategies that use existing chemotherapeutic agents. These studies are generally powered to detect less than 10% absolute increase (or difference) in EFS in the sub-population of patients, usually of relevant size, who have a relatively good prognosis. In childhood ALL, these studies typically address so-called intermediate risk patients, accounting for approximately 50% of the ALL population, who have a 4-year EFS of 70–80%. These studies may ask a randomized question on intensification or on the inclusion of new formulations of existing drugs such as, for example, the pegylated l-asparaginase instead of the native product. The third direction of research deals with the optimization of treatment by testing strategies that are at least as efficacious as those currently used but that carry a lower burden of toxicity and complications. Typically, this direction generates studies of non-inferiority, targeted to sub-populations of patients with good clinical outcome. As an example, in childhood ALL, these studies would target patients with approximately 90% 4-year EFS, for whom the attempt is to deintensify certain therapy elements with known short- or long-term side effects, without compromising the EFS outcome. This third line of clinical research is related to frontline treatments, while the two previous lines may apply also to the context of relapsed patients. The latter are also the candidate population for Phase I studies. The three directions of clinical research illustrated here have different implications on sample size. One arm Phase II studies on activity and safety of a novel drug typically include a few tens of patients (see for instance, [2, 6]). In randomized trials, which compare the novel therapeutic approach to a control arm, i.e. the best current standard, sample size depends strongly on the target difference the trial is powered to identify, and marginally on the defined level of type I and II error. The smaller the difference, the greater the number of participants needed for enrolment (Table 5.1). For studies on patients with poor outcome, where experimental therapeutic strategies with new drugs are applied in the hope of marked improvements, the sample size needed may be in the order of a few hundreds. It becomes much higher in patients where only relatively modest improvements can be expected and very large in non-inferiority trials if tolerability is set to be around a −3 or −4% absolute margin for patients who have a good outcome with standard treatment. In any of these contexts, the sample size needed is difficult to achieve in a reasonable time frame considering either that high-risk leukemia may represent a small subset or that lymphoblastic leukemia is not a common cancer overall. For this reason, mainly in childhood leukemia, many countries have developed national study groups that design and run multicentre clinical trials. On top of this, they have
5 Strategies for Trial Design and Analyses
87
Table 5.1 Number of subjects needed in different scenarios of superiority and non-inferiority trials with two parallel groups of equal allocation and comparison of event-free survival (EFS) based on log-rank test (one-sided test, type I error 0.025 and 0.80 power, see [36]) Standard treatment, Standard treatment, 4-year EFS (%) Δ in 4-year EFS (%) Total number of subjects (events) 30 181 (109) +20 30 +15 307 (192) 75 +10 508 (102) 75 +5 2,186 (492) 90 −4 2,077 (249) 90 −3 3,554 (409) A positive open triangle indicates the target improvement and a negative open triangle indicates the target non-inferiority margin that the trial aims at detection. The number of subjects to be enrolled in each treatment group is half the total number reported in the last column.
developed international collaborations to address therapeutic questions in trials for rare subgroups [7] or in large trials powered to detect modest differences [8]. International cooperation has sometimes needed an innovative approach to study design or study conduct, as discussed in Sect. 5.7. Even with national and international cooperation, research in small sub-populations may be difficult. Research institutions and medicine regulatory agencies have issued guidelines on how to run clinical trials in small populations [9, 10]. They propose that there are no special methods for the design and analysis of these trials but accept that less conventional and/or less commonly seen methodological approaches may be adopted if they help to improve the study. In the regulatory approval process, deviations from standard randomized controlled trials should only be considered when completely unavoidable and would need to be justified. As stated in the EMEA guidelines [10]: Approaches outlined in this document for situations where large studies are not feasible should not be interpreted as a general paradigm change in drug development. The methods described here to increase the efficiency of the design and analyses are also applicable for studies in large populations but are often not used because of increased complexity. The general principle can be applied to the following scenarios: (1) when randomized controlled trials are feasible even though the interpretation will be less clear compared to typically sized Phase III trials. This may be improved by special trial designs and/or refined statistical approaches; (2) when randomized controlled trials will be severely underpowered. However, controlled studies with low statistical power in case of an important treatment effect may be preferable to no controlled studies; (3) when randomized controlled trials are not feasible and only case series (with external control groups) or even only anecdotal case reports are available. Here, alternative approaches are required. Such compromise positions will usually be at the cost of increased uncertainty concerning the reliability of the results and hence the reliability of the effectiveness, safety and risk–benefit of the product. Additional follow-up data, post-approval, will be necessary. We have to consider the trade-off between relatively small amounts of high-quality evidence (for example, small randomized trials) and relatively larger amounts of lower-quality evidence (for example, large uncontrolled case series). This will always have to be judged on a case-by-case basis.
Section 5.5 revises general principles of non-standard study design.
88
M.G. Valsecchi and P. De Lorenzo
5.3 Studies of Superiority and Non-Inferiority When an established effective therapy already exists, there are obvious strong ethical and clinical reasons for taking this, rather than placebo, as a control arm in an experimental trial on a new promising treatment. Basically, there are two questions for which the clinicians want an answer: 1 . Is the new treatment better than the established therapy (superiority design) 2. Is the new treatment as good as the established therapy, yet preferable on other grounds such as higher safety, more convenient route of administration or posology, cost (non-inferiority design) A new treatment may be the result of adding a novel drug to the established therapy or of replacing another drug in that therapy. The first case, often called “add-on” randomized clinical trial, drives naturally to a superiority design. The second case, often called “head-to-head” trial, may lead to a non-inferiority design as well as to a superiority design. Of note, both the “add-on” and “head-to-head” trials, while having an active control, may include placebo. For example, one would add the new drug or the placebo preparation on top of the established treatment. In the “head-tohead” trial, each of the two drugs being compared would be given together with the placebo corresponding to the other drug. In the context of complex chemotherapeutic schedules used for leukemia treatment, these two types of trial do not describe all possible situations. First of all, an “add-on” trial might be designed as a “difference” trial, which adopts a more prudent approach to testing. This study, indeed, answers the question whether the new treatment is different from the established therapy with a two-sided test of hypothesis rather than one-sided as in the superiority design. It leaves open the possibility that the effect of the new treatment goes in the unexpected, but not impossible, direction of a worst rather than a better outcome. This prudent approach is recommended especially in the context of comparing two different treatment strategies, where the experimental arm does not involve a novel promising drug with regard to the standard control arm but, for instance, uses the same drugs with increased intensity. The relevance of the two-sided approach is in accounting at the design stage for the possibility that a greater intensity could also harm the patients rather than improve their outcome. Technically, a two-sided hypothesis testing with type-I error a = 0.05 can be reconciled to a one-sided approach with a = 0.025, and on the contrary a one-sided test with a = 0.05 corresponds to a two-sided test with a = 0.10. In the following, for simplicity, we do not further discuss the “difference” trial, having in mind that this is an extension of the superiority trial and overcoming the definitions by basing our reasoning and interpretation of results on confidence intervals (CI) rather than hypothesis testing. Non-inferiority “head-to-head” trials are often viewed with suspicion as they may be used to promote drugs of less therapeutic value [11]. However, non-inferiority designs can be useful in the context of intensive chemotherapeutic regimens, for addressing questions on deintensification. These questions are very important, as discussed in Sect. 5.2, for subgroups of patients that reach a very good outcome but at a too high price in terms of complications and toxicity.
5 Strategies for Trial Design and Analyses
89
E
D
C B A
I NEW TREATMENT WORST
0
S NEW TREATMENT BETTER
Fig. 5.1 Results of different trials represented in terms of confidence intervals for treatment effect for superiority and non-inferiority designs
Superiority and non-inferiority can be described graphically provided that we can employ a single primary outcome measure to describe both the benefits and harms of the experimental treatment (see Fig. 5.1). This representation requires an understanding of CI and how they can be used to conclude not only on the statistical significance but also on the clinical relevance of trial results. Suppose the difference between treatments is expressed as the absolute difference in the 4-year EFS so that the null hypothesis of no difference is represented by zero. Other summary statistics may be used, for example hazard ratios or odds ratios for proportions, and in that case the null hypothesis would be represented by the value one. Of interest is the position of the confidence interval relative to this “no difference” value, as the CI represents the range of values for the true difference that are plausible in the light of the trial data. Values to the right of the zero vertical line correspond to a better outcome on the new treatment while those on the left to a worst outcome (i.e. control is better). There are two other important vertical lines in Fig. 5.1. One is the inferiority boundary I (to the left of zero) to the left of which the new treatment is importantly inferior to established effective therapy. The superiority line S is such that, to its right, the new treatment is importantly superior to established effective therapy. The definition of these boundaries, either I or S, must be done at the design stage based on what is important to patients. As discussed in Sect. 5.2, it is the most influential determinant of sample size. For superiority trials, the S boundary should reflect the minimal improvement that is judged to be worth for switching to the new treatment in future patients. For non-inferiority trials, the selection of the margin I reflects the maximum disadvantage that is acceptable in view of advantages
90
M.G. Valsecchi and P. De Lorenzo
in other aspects of the new treatment. This choice can be difficult, both on clinical and ethical grounds, and while there are guidelines for it [12], it should always be carefully justified in the specific context of the trial. Usually, boundaries I and S are not symmetric, and I is nearer to the null line because, due to the effort that goes into the development of effective treatments, the choice of I should ensure that any loss of clinically important benefit could be detected. For sake of completeness, we underline here that a non-inferiority trial is not the same as an equivalence trial. The latter refers to a study designed to confirm the absence of a meaningful difference between treatments, in both directions of benefit or harm. For equivalence to be shown, the extremes of the CI should lie between I and S. Equivalence designs make mostly sense in the area of bioequivalence studies. By looking at Fig. 5.1, we can comment on various situations. Trial results of type A, B, C are likely to occur in situations where a non inferiority trial was planned. In this setting, trial C would be “positive”, demonstrating non-inferiority of the new treatment compared to the control. This is because the lower end of the CI is superior to I. On the contrary, A and B would be “negative” trials but with different clinical relevance, in that A shows more clearly the inferiority of the new treatment. Suppose that result D is unexpectedly obtained in a planned non-inferiority trial, can the objective of the comparison be switched to interpret the trial as a superiority experiment? Technically speaking, this is feasible by giving p-values for the superiority test under the Intention To Treat analysis (ITT, see Sect. 5.8), but the size of the benefit might not be relevant for clinical use, and this is not a matter of statistical reasoning but rather of clinical judgement. Going back to Fig. 5.1, trial results of the type D and E are those we would like to obtain when designing a superiority study. Trial D rejects the null hypothesis of no difference, but only results of trial E are both statistically significant and convincing in showing an important clinical benefit. This level of evidence on treatment effect, with the lower end of the CI that is above a convincing margin of superiority, is more commonly achieved as a result of meta-analyses rather than from single trials. Trial C, if designed as a superiority trial, would be “negative” as it would not reject the null hypothesis. This means that we are in the absence of proof of a difference and, strictly speaking, this does not constitute a proof of absence of difference. Yet, much debate has been done on the possibility to interpret this trial as one proving non-inferiority of the new treatment as compared with the control, and regulatory bodies have issued guidelines on this aspect [13]. Switching the objective from superiority to non-inferiority can be much more seriously misleading than the opposite switching as, for instance, a poorly conducted or an underpowered superiority trial may easily lead to a non-significant result, respectively, because treatment effect is diluted or it has wide confidence intervals. For this reason, and also because this is the type of switching that has practical relevance in the registration process, it should be made under very controlled conditions. Regulatory bodies recognize this difficulty as shown by the following considerations taken from the EMEA guidelines [13]: A trial to show equivalence or non-inferiority must show a high degree of consistency with protocolled plans, if it is to be reliable. Deviations from the inclusions criteria, from the intended treatment regimen, from the schedule, manner and precision of taking measurements,
5 Strategies for Trial Design and Analyses
91
and so on, all tend to reduce the sensitivity of the trial and to make conclusions of ‘no difference’ more likely, even when deviations are of an unsystematic or random nature. The size of the bias associated with these and other departures from the protocol is generally unknown and may render such a trial uninterpretable. Failure to show a difference between two treatments can also arise when both treatments are inefficacious, perhaps as a result of being inappropriately administered. This problem does not affect superiority trials to the same extent because the demonstration of a difference is itself validation of the sensitivity of the trial. For these reasons, switching from superiority to non-inferiority is likely to carry with it a lesser degree of confidence in the conclusions.
As a final consideration, trial design requires that experts in various fields (clinical, biological, pharmacological, biostatistical) cooperate to formulate in advance, as clearly and precisely as possible, based on up-to-date knowledge, the objectives and hypothesis of the trial. We think that anyone involved in research may convey that a trial where the results support a sound pre-specified hypothesis is more persuasive than one where the results are “surprising”. Thus, unplanned testing for non-inferiority in trials designed for superiority should be discouraged. Study design that incorporates both non-inferiority and superiority testing are possible, although they can be useful and applicable in selected contexts [14].
5.4 Activity, Efficacy and Surrogate End Points Early clinical trials of a novel oncology agent or combination regimen usually examine the biological activity of the therapy with end points that track the progress or extent of disease. Traditionally, they measure the objective response rate relatively early in time. This approach may change in the future since response may not be appropriate for evaluating treatments without direct cytotoxic effects. Results on activity are useful to assess whether the intervention is sufficiently promising to justify the conduct of phase III trials. Here, the goal is to assess clinical efficacy, through outcome measures that unequivocally reflect tangible benefits to patients. These studies must provide direct evidence about a “definitive” clinically meaningful outcome, which, in oncology, is essentially duration of survival. However, in situations where effective second-line treatments will be applied if patients do poorly after frontline therapies, survival is not the best outcome for primary therapies. A more specific clinical end point is time to frontline treatment failure, which is not confounded by the secondary treatment. In leukemia, this is typically the EFS time, i.e. the time to one of the following events, whichever occurs first: relapse, death and, according to study design, also resistance and secondary malignant neoplasm. The pressure to accelerate approval of novel drugs or the attempt to shorten the time to trial results has generated a growing interest on the use of end points on activity as surrogate end points for efficacy [15, 16]. However, this use is only justified in Middle-Late Development studies if the surrogate end point is convincingly associated with the relevant clinical outcome (EFS). In leukemia, the recent spread of reliable methods to measure the presence of disease at molecular level (minimal residual disease – MRD) has created new perspectives in trial design. The potential
92
M.G. Valsecchi and P. De Lorenzo
gain in shortening duration of clinical trials is evident, as MRD can be measured early to detect response, while EFS needs a long follow-up, especially if the event rate is initially low. The gain in efficiency may come from the fact that a novel drug given early during treatment may potentially affect MRD levels quite markedly even in a population that is expected to have a relatively good long-term outcome. However, deciding that efficacy of treatment can be assessed in terms of response (or MRD levels) requires that response is properly validated as a surrogate end point. This is a difficult task, and examples of definitive validation of putative surrogate end points are rare in oncology and usually require a meta-analysis of several trials [17, 18]. Validation of a surrogate requires that (i) the candidate end point is a prognostic factor for an established clinical end point (EFS or survival) and (ii) it captures fully the treatment effect on the clinical end point. While the first condition is easily met, since, in practice, candidate surrogates are often selected because of their strong correlation with clinical outcome, the second condition is much more problematic to prove. Indeed, it does not only mean that treatment must affect both the potential surrogate and the clinical outcome but also requires that all mechanisms of action of treatment are reflected in the effect on the surrogate end point [19, 20]. Figure 5.2a illustrates the behaviour of a perfect surrogate. Here, the experimental treatment shows a marked effect on response, with an absolute increase of 20% on the 40% baseline level in the control arm. Responders (and non-responders) have exactly the same failure rate whether they were treated with the experimental or standard therapy because here response captures fully the effect of treatment on clinical outcome. So, with a 10% long-term failure rate in responders and 40% in non-responders (or correspondingly a 90 or 60% 4-year
Standard treatment 100 pts.
Experimental treatment 100 pts.
ENDPOINT Response
RELATIVE EFFECT OF EXPERIMENTAL TREATMENT YES
NO
YES
NO
40
60
60
40
High activity PERFECT SURROGATE: Same failure rate in responders/non responders
Panel (a) 4-year Failure Rate No. failures
10%
40%
10%
40%
4
24
6
16
28%
Modest clinical benefit
22% Panel (b)
4-year Failure Rate No. failures
10%
40%
15%
50%
4
24
9
20
28%
29%
NO SURROGACY: Higher failure rate in responders/non responders No clinical benefit
Fig. 5.2 Two different scenarios in surrogacy: perfect surrogacy, Panel (a) and no surrogacy, Panel (b)
5 Strategies for Trial Design and Analyses
93
EFS in responders and non-responders, respectively), the effect on clinical outcome is that failure rate changes from 28% in the standard arm to 22% in the experimental arm, with an absolute gain of 6%. Even if simplified, this is not an unrealistic scenario and shows very clearly that a high activity on a valid surrogate may translate into a modest effect on clinical outcome. In other words, even if early response is a perfect surrogate, a fairly large increase in response rate is required to translate into detectable EFS or survival gain [17, 18]. The situation becomes more complicated if we do not have a valid surrogate as we simulated in (b). Here, the same benefit on response (20% increase) translates into no clinical benefit (28% vs. 29% long-term failure rate) because the higher migration of patients to the responders group, in the experimental treatment arm, carries with it a higher long-term failure rate both in this group and in the one of non-responders (which has now lost the patients at better prognosis). Unfortunately, failure to translate a better response into a better survival is not uncommon but, in spite of this, claims of surrogacy are still made without proper validation [21]. Failure to establish surrogacy can occur in many ways, as discussed by Fleming and DeMets [22]: Although there are many explanations for this failure, such as the existence of causal pathways of the disease process that are not mediated through the surrogate end point and that might be influenced differently by the intervention, the most plausible explanation is usually that the intervention has unintended mechanisms of action that are independent of the disease process. These unintended mechanisms can readily cause the effect on the true clinical outcome to be inconsistent with what would have been expected solely on the basis of evaluation of surrogate end points. These mechanisms are insidious because they are often unanticipated and unrecognized.
This occurs, for instance, in cases where toxicity that may negatively impact on long-term mortality obscures the effect seen earlier on response. In leukemia, the MRD has not yet been formally validated as a surrogate end point, while its prognostic value has extensively been shown and its use as an indicator of activity has solid grounds. In extreme situations, such as those of multiresistant patients, the achievement and maintenance of negativity at the molecular level for a given period may per se be an end point of clinical interest, if it is related to a condition of patient benefit. It may be an end point also in studies which aim at impacting on molecular remission as an intermediate step for optimizing further treatment. For instance, this is the case of studies on patients who, based on their high risk of relapse, are candidates for bone marrow transplantation. Here, an experimental treatment that increases the rate of MRD negativity increases the chance of success of transplant or even allows to spare patients from undergoing a transplant. As a final consideration, even taking into account the difficulties in running trials in small populations, EMEA makes the following point on the use of surrogate: “Also it has to be pointed out that surrogate markers cannot serve as final proof of clinical efficacy or long-term benefit. If they are intended to be the basis for regulatory review and approval then, unless they are properly validated, there should be a predetermined plan to supplement such studies with further evidence to support clinical benefit, safety and risk/benefit assessment” [10].
94
M.G. Valsecchi and P. De Lorenzo
5.5 Design Strategies 5.5.1 Adaptive Designs The need to speed up the process of drugs development, thus ensuring that optimal treatments are promptly adopted and resources efficiently allocated, has been recognized for a long time and traditionally addressed with so-called “group sequential” designs. Such designs allow the incorporation of evidence emerging from accumulating data into the ongoing trial through interim analyses. This essentially alters the sampling plan avoiding the inflation of the pre-specified type-I error (see Sect. 5.8). “Group sequential” designs are sequential sampling methods with the fundamental feature of being fully set in advance: various scenarios of future interim findings are considered to decide upon successive trial execution. A broader range of design modifications, for instance sample size re-estimation, discontinuation of treatment arms, changes in the primary measure of treatment response, is allowed by adaptive designs. Adaptive designs make use of known interim findings to modify the sampling plan and are, therefore, data-driven. For this reason, they raise concern about the unintended consequence of introducing bias and imprecision in the estimates. In addition, the adoption of an adaptive design for a Phase III trial could be seen as a main contradiction to the confirmatory nature of such Late-development studies, especially if these are pivotal trials aiming at approval of a new product. Adaptive approaches are justified by the need to cope with difficult experimental situations, where the paucity of available knowledge determines uncertainty about crucial aspects of the study. These difficulties must be acknowledged in advance so that any modification of the trial design, for example sample size re-estimation, is anticipated and justified in the study protocol. Adaptive designs cannot in fact be used as a remedy for poor trial planning but actually require more upfront planning both statistically and in managing logistics and implementation. In short, a proper adaptive design must control potential sources of bias, first of all by ensuring that the type-I error is fully preserved and valid estimates and confidence intervals for the treatment effect will be available as a summary of the totality of evidence. Additional measures to be taken operationally, in order to maintain the integrity of the trial, include the careful disclosure of interim analyses. In practice, access to the available data and related results must be limited (it is recommended that clinical team be isolated from both), and communication of interim analysis decisions is to be preferred over communication of interim results. A general concern about monitoring designs is heterogeneity in treatment effects estimated from the different stages of the trial. This may be due to random variation, but in practice it cannot be excluded that a calendar time effect be present, for instance because of a learning process or a change in patients’ important prognostic factors occurring over time. Heterogeneity in stage results can be detected by applying a heterogeneity test, although its statistical power is generally limited. Should it indicate that a discrepancy exists, further investigation into possible causes is recommended [23].
5 Strategies for Trial Design and Analyses
95
Traditional Design
Treatment options A B C Control Phase II
Phase III
Time - gap
Time
Stage 2 (confirming)
Stage 1 (learning)
A B C Control Treatment selection
Phase II/III
Adaptive Seamless Design Fig. 5.3 Comparison of adaptive seamless and traditional design
An interesting example of adaptive design is so-called adaptive “seamless” design, in which two separate phases of drug development that can be conducted as separate trials are in fact combined into a single one, both operationally and inferentially. Data gathered from the first stage (learning stage, e.g. dose selection) are used to adapt trial plan of the second, confirming stage, and data from both are used to run the final analysis and answer study questions. This adaptive “seamless” design does not simply select out the most promising treatment option, but actually allows continuation of the preferred arm without the need to start a new trial (see Fig. 5.3). Application of this type of study requires that the end point for the learning phase be immediately available for continuation with the confirming stage and that it be adopted as primary end point also to answer the study question [24].
5.5.2 Bayesian Designs Bayesian statistics is a methodology used both in the design and in the analysis stage of a trial that provides a coherent method for learning from evidence as it accumulates. This is performed by combining prior information with accumulating evidence through a mathematical model, and leads to the so-called “posterior distribution” of the clinical end point of interest. The Bayesian approach may be useful especially when good prior information on clinical use of a device exists so that incorporating it in the design and analysis
96
M.G. Valsecchi and P. De Lorenzo
of the future trial may result in a smaller-sized or shorter-duration study. Yet, it can be used also when prior information is only partially (or not) informative, since it can provide flexible methods for interim analyses and modifications of ongoing trials (e.g. changes to sample size or changes in the randomization scheme). Criticism has been raised about adoption of a Bayesian approach, as the use of prior knowledge is controversial when it is based on subjective evaluations rather than on empirical evidence. Moreover, implementation of Bayesian methods can be more difficult than the routinely frequentist counterparts: for instance, they usually require sensitivity analyses to check robustness of models and priors [25]. This is particularly critical in cancer Phase I trials, where the main goal is to estimate the maximum dose level of a new agent associated with a predefined acceptable level of toxicity. Preliminary or preclinical studies can provide information about the likelihood of toxicity at the starting dose, but the way this is incorporated into an appropriate “prior distribution” relies on specific and complex methods [26]. The Bayesian approach for trial design usually does not determine sample size in advance; instead, it may specify a particular criterion to stop the trial. Appropriate stopping criteria may be based on a specific amount of information about the clinical measure of interest (e.g. a sufficiently narrow credible interval, as defined below) or an appropriately high probability for a pre-specified hypothesis. At any point during a Bayesian clinical trial, it is possible to update the computation of the expected additional number of observations needed to meet the stopping criterion. As the sample size is not explicitly part of the stopping criterion, the trial can be terminated whenever enough information has been collected to answer the trial questions. When sizing a Bayesian trial, it is recommended that the minimum sample size according to safety and effectiveness end points is pre-specified. It is also wise to anticipate a minimum level of information from the current trial needed to verify model assumptions and appropriateness of prior information used. Pre-specified interim analyses can be part of a Bayesian trial. Various methods are available. For instance, investigators may wish to calculate the probability that, given the data observed up to a certain point, the trial will be positive (reject the null hypothesis) with further recruitment and follow-up. This question may be answered to by using the Bayesian predictive distribution. This tool is used to predict whether future data will result in a posterior probability of a positive result that is sufficiently high to suggest that the trial may be stopped early for efficacy or, conversely, sufficiently low, so to indicate that the trial may be stopped early for futility (see also Sect. 5.8). This appealing feature makes Bayesian monitoring useful also for trials planned within a classical frequentist approach [27]. In fact, the criticisms against Bayesian methods appear to be less relevant in the context of clinical trial monitoring, even from a non-Bayesian point of view. For instance, the problem related to the choice of a prior distribution is not so crucial anymore. Since the purpose is to prevent premature termination of the trial, the preference hardly goes to an optimistic prior distribution, but rather to a distribution that reduces the chance of wrongly claiming that the results have already become conclusive. The final analyses within the Bayesian approach include hypothesis testing and interval estimates. For Bayesian hypothesis testing, the posterior distribution can be
5 Strategies for Trial Design and Analyses
97
used to calculate the probability that a particular hypothesis regarding the end point is true, given the observed data. Interval estimates in the Bayesian context are called credible intervals: a 95% credible interval is the interval with 95% posterior probability of including the “true” unknown treatment effect. Bayesian methods can be useful for the evaluation of treatments for rare cancers, which are made difficult by the problem of small numbers and limited statistical power. A Bayesian approach suggests basing the evaluation of trial results on a posterior distribution that combines previous knowledge with data expected from the planned trial itself [28]. This is done already at the design stage, to increase robustness of information from and to provide justification for the small trial. Also, in such a type of application, the key step is summarizing the information available before the trial.
5.6 Study Conduct: Protocol and Good Clinical Practice Principles The protocol is the written scientific plan for a clinical trial. It addresses all aspects of the research, and as such, it serves the investigators involved in the trial as well as third parties, including ethics committees and institutional review boards, regulatory authorities, peer-reviewed journals. In addition, it is widely recognized that knowledge of planned clinical research is in the interest of the scientific community at large, and registration of new protocols in specific databases has recently become a legal requirement (e.g. in Europe, registration in EudraCT). Table 5.2 illustrates its structure and content. The protocol is a plan for clinical research; therefore, it may be that during trial conduct investigators need to face unexpected events that give rise to protocol deviations. Some of these may introduce bias or inaccuracy in estimates, while others may have only a negligible effect on the integrity of the trial. Piantadosi [2] suggests that the impact of deviations should be evaluated in terms of both design validity (i.e. the trial is suitably planned and executed) and biological validity (i.e. the trial is based on an appropriate and feasible question and provides valid biological evidence): for instance, minor flaws in design can be disregarded if biological validity is maintained. The International Conference on Harmonization (ICH) guideline on Good Clinical Practice (GCP) provides a common standard for the mutual acceptance of clinical data by regulatory authorities in the European Union, Japan and USA [29]. Its major aims are the protection of rights and safety of human subjects involved in clinical research, as well as the production of reliable results. Therefore, the GCP principles apply to investigations that have an impact on safety and well-being of patients, irrespective of whether they will be part of a regulatory approval submission. Indeed, compliance with the ICH guideline is usually not a legal requirement but is the standard that is expected when producing evidence in clinical research, through either commercial or academic trials. On the contrary, academic trials differentiate
98
M.G. Valsecchi and P. De Lorenzo
Table 5.2 Protocol structure and content Section Content Title page Study name and title Date and number of current version Principal investigator and members of the Study Committee (e.g. protocol statistician, contact persons at participating institutions/national groups) Sponsor details Trial registration number(s) Outline Schematic description of the treatment plan Introduction and background Description of the scientific background and rationale for the study Report of available historical data Aims Identification of the primary aim, corresponding to the main question addressed Secondary aim(s) on additional issues Study design Type of study (phase and design) Patients’ registration Patients’ eligibility criteria (inclusions and exclusions) Criteria to identify extent/severity of the disease (patients’ stratification, if any) Randomization procedure (if any) Treatment plan Details of therapy, including dosages and schedule, modifications due to toxicity/side effects, information on study drugs Management of toxicity Supportive care Required observations Evaluation of primary and secondary end points and toxicity Serial measurements of interest Statistical considerations Summary of end point definitions Description of expected accrual, computation of sample size, power analysis Methods for interim and final analyses Guidelines for stopping rules Organization, logistics Procedures for trial conduct Data collection, management Property of data and publication of results Data and Safety Monitoring Committee (DSMC) Data collection forms Appendices Informed consent Add-on studies
from trials promoted by the pharmaceutical industry because they are usually concerned with public health questions regarding “best practice” for patients care and because of greater financial constraints. So, while adherence to GCP principles is not questioned, the efficiency and relevance of operating procedures adopted to follow the rules are to be considered, to avoid disincentives to clinical research especially in the academic context. The crucial point is that the conduct of the trial should be fit for purpose that is proportionate to the risk involved in the intervention.
5 Strategies for Trial Design and Analyses
99
These considerations apply also to so-called “observational studies”, which evaluate various aspects related to the wide use of a medical product in clinical practice, for instance the large-scale safety profile and pharmacoeconomy issues. Although participation into such studies do not mean additional risks to patients, it is required that the basic GCP principles are met and reflected into the plan of the study as documented by the protocol.
5.7 International Clinical Trials The design of international studies has received much attention in the last decade, especially in cancer research, where the advances in biosciences and technology have increased the ability to understand and model biological mechanisms. In fact, this resulted in a rapid identification of many biomarkers by which it is possible to define (possibly small) subgroups patients who are most likely to benefit from new therapeutic strategies. Clinical trials are, therefore, more and more concerned with small populations and rare subgroups of common cancers. In an attempt to overcome their intrinsic limitation due to the small numbers, clinical institutions and national groups joined into international networks, thus to enlarge the potential recruitment basis. Basic requirement for a standard international clinical trial is that participants agree on a common treatment protocol. A more flexible approach that can be adopted refers to a prospective meta-analysis strategy (PMAS): the international trial is designed to ask the same randomized question within treatment protocols that share the clinical rationale, but may vary in specific components [30]. For instance, in childhood ALL, national cooperative groups across the world have developed standard treatment protocols that yield comparable outcomes by applying different combinations of similar chemotherapy elements. Therefore, the clinical interest often converges to the same treatment element(s) that may be investigated by asking the same randomized question. The PMAS approach allows running a large trial instead of many small similar trials. It provides evidence on the effect of the experimental treatment element(s) in different settings, borrowing the principles of meta-analysis and avoiding its limitations, because the study is prospectively designed and monitored as an ordinary randomized clinical trial. Heterogeneity in the results of the different protocols involved needs to be carefully considered in the planning phase and controlled in the statistical analysis. A potential pitfall of the method is indeed the risk that a too high heterogeneity in groups’ outcomes jeopardizes the joint estimation of the treatment effect in the common randomized question. As compared to a standard clinical trial, an international clinical trial requires increased operational efforts in managing logistics and implementation, also to comply with the many requirements set by regulatory authorities, for example in the European Union, the Directive 2001/20/EC [31]. In the recent experience in international trials in childhood leukemia, the logistics was efficiently managed by
100
M.G. Valsecchi and P. De Lorenzo
maintaining the intrinsic two-level structure of the international trial itself. In fact, the trial management can be organized allowing that (1) at the national level, a clinical Chairperson and a Data Centre be responsible for coordinating the trial conduct and data management within their network of clinical centres. National Chairpersons and Data Centre representatives are members of the Study Committee. (2) at the international level, a clinical Trial Chairperson and a Trial Data Centre (jointly referred to as the Coordination Unit) be responsible for coordinating the national levels on behalf of the Study Committee. Operationally, study conduct based on this structure can be further eased by resorting to the Internet, which provides a worldwide, easy-accessed infrastructure. Many trial operations can be carried out through a devoted Web site, which is set up to ensure quality standards both in data confidentiality through separation of demography data from other patient’s data and data security by means of procedures including encryption of data traffic over the Internet and password-protected access to the server [32].
5.8 Statistical Analyses This section overviews the crucial aspects of statistical analysis in clinical research. Proper planning of analysis, consistent with the aims of the study, is essential to avoid data-driven analysis, which can easily produce misleading conclusions. For this reason, the analyses are to be specified in the protocol (see Sect. 5.6) by defining the subjects’ populations and the methods that will be applied to answer the study questions. In order to identify the populations to be analysed, clear definitions of eligibility criteria, deviations (e.g. patients who switch therapy or are non-compliant with therapy plan) and dropouts (e.g. patients who discontinue treatment) are to be given, together with the approach that will be taken to account for them in the analyses. As far as the primary end point is concerned, three approaches can be taken: the analysis by “intention to treat” (ITT), “per protocol” (PP) or “as treated” (TP). With the ITT analyses, the randomized treatments (for simplicity, standard vs. experimental) are compared by contrasting the patients in the treatment group they have been assigned to, irrespective of deviations and drop-outs. This method protects against systematic errors and guarantees validity of statistical tests and generalizability of results and is, therefore, the gold standard. PP and TP analyses select retrospectively patients to be included in the analyses, by evaluating the adherence to protocol: PP is the stricter one, since it considers only the subset of patients who actually complied with the treatment they were assigned to (according to a given definition of compliance). TP, instead, evaluates patients who received a given treatment, irrespective of the randomized assignment (thus including patients who switched treatment arm). In a superiority trial, ITT analysis is the main analysis, while TP analysis is usually added to evaluate the robustness of trial results in the presence of deviations. If the extent of non-adherence to protocol exceeds a reasonable level, which is regarded as compatible with clinical practice, the scientific
5 Strategies for Trial Design and Analyses
101
integrity of the protocol itself may be questioned. In a non-inferiority trial, the ITT and the PP analysis have equal importance and should, therefore, give similar results to ensure robustness of conclusions [33]. Frequently, the protocol comprises intermediate analyses of the primary end point and/or of safety, which are performed with pre-specified timing and modalities during the course of the study. The interim analyses are mainly motivated by ethical concerns and aim at providing quantitative considerations for decision making about the future of a clinical trial that is, whether it may be continued, either with or without modifications, or stopped. Reasons for stopping include toxicity (treatment(s) proved to be unacceptably too toxic), efficacy (one of the treatments shows an unexpectedly marked superiority) and futility (given the accumulated results, it is unlikely that the superiority of the experimental treatment can ever be demonstrated). In practice, for most studies, the trial progress is reviewed by an independent Data and Safety Monitoring Committee (DSMC). Although monitoring does not rest on purely statistical grounds, interim analyses provide major information. There are two major statistical strategies. The classical “frequentist” approach uses group sequential designs that provide evidence based on the accumulated data available at pre-specified time points (so-called “looks” at data, usually not exceeding the number of 2 or 3). At each look, a relatively stringent significance level is used so that the overall type I error is maintained at the desired level (e.g. 5%). For instance, the O’Brien/Fleming rule uses extremely conservative criteria at the beginning of the study, when differences are much more likely to be spurious. Other methods, for example the Pocock rule, use a constant significance level at every analysis, but this is maintained at a lower level than the overall one. The alternative approach uses a “Bayesian” strategy that combines the external or prior evidence/beliefs with the observed data from the ongoing trial. The trial evidence is balanced against the prior evidence and should it outweigh the latter, it suggests that early termination can be considered. Although the formalization of the prior evidence is challenging, this approach has the advantage to incorporate an evaluation as to whether results to date are sufficiently conclusive to alter the management and therapy of future patients [27]. The final analysis on the primary outcome as specified in the protocol is conducted to estimate and test hypothesis about treatment effects in the eligible population. It is common practice to perform also subgroup analyses, that is, to evaluate treatment effect in patients with given baseline characteristic(s), to investigate whether results observed overall vary across specific sub-populations or, in other words, whether they show heterogeneity. Heterogeneity can take two forms, qualitative and quantitative. Qualitative interactions are very rare in clinical trials and arise when treatment turns out to be beneficial for some patients but harmful for others. Such a finding is rarely observed because patients, who according to the underlying scientific model could be harmed by treatment, are excluded a priori from eligibility. Quantitative interactions are observed when the experimental treatment is always beneficial to patients but to a different extent depending on patients’ characteristic(s) and are more common. Subgroup analyses should be listed and motivated in the study protocol, to avoid data-driven definition of subgroups that adds to critical limitations they are per se
102
M.G. Valsecchi and P. De Lorenzo
subject to: (i) high probability of false-positive findings, because of multiplicity of hypothesis testing (the adoption of more stringent criteria for significance is therefore recommended); (ii) high false-negative rate, as tests do not have sufficient power (small numbers in subgroups preclude conclusive results). From the methodological point of view, the most appropriate statistical approach to heterogeneity is the test of interaction between treatment and the target baseline covariate. Yet, failure to identify significant interactions, if due to lack of power, does not rule out heterogeneity. Often, this issue is dealt with, with proper power, only in metaanalysis studies. Appropriately conducted subgroups analysis can actually convey important information to the scientific community. For this reason, guidelines have recently been proposed for the reporting of subgroup analyses, with the aim to promote clear, uniform and complete reporting [34]. In general, it is strongly recommended that quality standards internationally established for reporting results of both randomized trials (CONSORT) and observational studies (STROBE) are fulfilled [35].
References 1. Stebbing LS 1961. Philosophy and the Physicist. Midlesex, UK: Penguin. 2. Piantadosi S. 2005. Clinical Trials: A Methodologic Perspective, 2nd Edition. New York: Wiley. Chapter 6 3. Byar, D.P. On combining information: historical controls, overviews and comprehensive cohort studies. Recent Results in Cancer Research 1978; 111:95–8. 4. Freedman, B. Equipoise and the ethics of clinical research. New England Journal of Medicine 1987; 317:141–145. 5. Maitournam A, Simon R. On the efficiency of targeted clinical trials. Statistics in Medicine 2005; 24(3):329–39. 6. Pocock SJ 1996. Clinical Trials: A Practical Approach. New York: Wiley. 7. Pieters R, Schrappe M, De Lorenzo P, Hann I, De Rossi G, Felice M, Hovi L, LeBlanc T, Szczepanski T, Ferster A, Janka G, Rubnitz J, Silverman L, Stary J, Campbell M, Li CK, Mann G, Suppiah R, Biondi A, Vora A, Valsecchi MG. A treatment protocol for infants younger than 1 year with acute lymphoblastic leukemia (Interfant-99): an observational study and a multicentre randomised trial. Lancet 2007; 370(9583):240–50. 8. Conter V, Valsecchi MG, Silvestri D, Campbell M, Dibar E, Magyarosy E, Gadner H, Stary J, Benoit Y, Zimmermann M, Reiter A, Riehm H, Masera G, Schrappe M. Pulses of Vincristine and Dexamethasone in addition to intensive chemotherapy for children with intermediate-risk acute lymphoblastic leukemia: a multicentre randomised trial. Lancet 2007; 369(9556):123–31. 9. The National Academy of Science 2001. Small Clinical Trials: Issues and Challenges. Washington DC, USA: National Academy Press. 1 0. European Agency for Evaluation of Medicinal Products, Committee for medicinal products for human use. Guideline on clinical trials in small populations. CHMP/ EWP/83561/2005, 2006 (http://www.emea.europa.eu/pdfs/human/ewp/8356105en.pdf accessed on March, 20 2009). 11. Garattini S, Bertelè V. Non-inferiority trials are unethical because they disregard patients’ interests. Lancet 2007; 370: 1875–77. 12. European Agency for Evaluation of Medicinal Products, Committee for medicinal products for human use. Guideline on the choice of the non-inferiority margin. EMEA/CPMP/
5 Strategies for Trial Design and Analyses
103
EWP/2158/99, 2005 (http://www.emea.europa.eu/pdfs/human/ewp/215899en.pdf accessed on March, 20 2009). 13. European Agency for Evaluation of Medicinal Products, Committee for medicinal products for human use. Points To Consider On Switching Between Superiority And Non-Inferiority. EMEA/CPMP/EWP/482/99, 2000 (http:// www.emea.europa.eu/pdfs/human/ewp/048299en. pdf accessed on March, 20 2009). 14. Freidlin B, Korn EL, George SL, Gray R. Randomized clinical trial design for assessing noninferiority when superiority is expected. Journal of Clinical Oncology 2007; 25(31):5019–23. 15. Pazdur R. Response rates, survival, and chemotherapy trials. Journal of the National Cancer Institute 2000; 92(19):1552–3. 16. Fleming TR. Surrogate end points and FDA’s accelerated approval process. Health Aff 2005; 24:67–78. 17. Buyse M, Thirion P, Carlson RW, Burzykowski T, Molenberghs G, Piedbois P. Relation between tumour response to first-line chemotherapy and survival in advanced colorectal cancer: a meta-analysis. Meta-Analysis Group in Cancer. Lancet 2000; 356(9227):373–8. 18. Bruzzi P, Del Mastro L, Sormani MP, Bastholt L, Danova M, Focan C, Fountzilas G, Paul J, Rosso R, and Venturini M. Objective response to chemotherapy as a potential surrogate end point of survival in metastatic breast cancer patients. Journal of Clinical Oncology 2005; 23:5117–25. 19. Prentice RL. Surrogate endpoints in clinical trials: definition and operational criteria. Statistics in Medicine 1989; 8:431–40. 20. De Gruttola VG, Clax P, DeMets DL, Downing GJ, Ellenberg SS, Friedman L, Gail MH, Prentice R, Wittes J, Zeger SL. Considerations in the evaluation of surrogate endpoints in clinical trials. Summary of a National Institutes of Health workshop. Control Clin Trials 2001: 22(5); 485–502. 21. George SL. Response rate as an endpoint in clinical trials. Journal of the National Cancer Institute 2007; 99(2):98–9. 22. Fleming, TR, DeMets, DL. Surrogate End Points in Clinical Trials: Are We Being Misled? Ann Intern Med 1996; 125(7):605–13. 23. European Agency for Evaluation of Medicinal Products, Committee for medicinal products for human use. Reflection paper on methodological issues in confirmatory clinical trials with flexible design and analysis plan. CHMP/EWP/2459/02, 2006 (http://www.emea.europa.eu/ pdfs/human/ewp/245902en.pdf accessed on March, 20 2009). 24. Thall PF. A review of phase 2–3 clinical trial designs. Lifetime Data Anal. 2008; 14(1):37–53. 25. U.S. Department of Health and Human Services, Food and Drug Administration, Division of Biostatistics. Guidance for the Use of Bayesian Statistics in Medical Device Clinical Trials (http://www.fda.gov/cdrh/osb/guidance/1601.pdf, accessed on March, 20 2009). 26. Tighiouart M, Rogatko A, Babb JS. Dose escalation with overdose control. Statistics in Medicine 2005; 24:2183–96. 27. Fayers PM, Ashby D, Parmar MK. Tutorial in biostatistics: Bayesian data monitoring in clinical trials. Statistics in Medicine 1997; 16:1413–30. 28. Tan SB, Dear KBG, Bruzzi P, Machin D. Strategy for randomised clinical trials in rare cancers. BMJ 2003; 327:47–9. 29. European Agency for Evaluation of Medicinal Products, Committee for medicinal products for human use. Note for Guidance on Good Clinical Practice (ICH Topic E6). CPMP/ ICH/135/95, 2002 (http://www.emea.europa.eu/pdfs/human/ewp/8356105en.pdf accessed on March, 20 2009). 30. Valsecchi MG, Masera G. A new challenge in clinical research in childhood ALL: The prospective meta-analysis strategy for intergroup collaboration. Ann Oncol 1996; 7:1005–8. 31. Anon Directive 2001/20/EC of the European Parliament and of the Council of 4 April 2001 on the approximation of the laws, regulations and administrative provisions of the Member States relating to the implementation of good clinical practice in the conduct of clinical trials on medicinal products for human use. Off J 2001; 44:34–44 (http://eur-lex.europa.eu/LexUriServ/ LexUriServ.do?uri=OJ:L:2001:121:0034:0044:EN:PDF, accessed on March 20, 2009).
104
M.G. Valsecchi and P. De Lorenzo
32. Valsecchi MG, Silvestri D, Covezzoli A, De Lorenzo P. Web-based international studies in limited populations of paediatric leukemia. Pediatr Blood Cancer 2008; 50:270–273. 33. European Agency for Evaluation of Medicinal Products, Committee for Medicinal Products for Human Use. Note for Guidance on Statistical Principles for Clinical Trials (ICH Topic E9). CPMP/ICH/363/96, 1998 (http://www.emea.europa.eu/pdfs/human/ewp/8356105en.pdf). 34. Wang R, Lagakos SW, Ware JH, Hunter DJ, Drazen JM. Statistics in medicine – reporting of subgroup analyses in clinical trials. NEJM 2007; 357(21): 2189–94. 35. The Equator Network – Enhancing the quality and transparency of health research. http:// www.equator-network.org/ (accessed on March 20, 2009). 36. Machin D, Campbell MJ, Fayers PM. 1997. Sample size tables for clinical studies, 2nd Edition. New York: Wiley-Blackwell.
Chapter 6
An Overview on Animal Models of ALL Michael A. Batey and Josef H. Vormoor
6.1 In Vivo Modelling in Preclinical Research The aim of in vivo preclinical study programmes can be essentially twofold. Firstly, to develop models which are physiologically representative of the human condition and to use these to improve our understanding of the clinical disease. Secondly, as part of a drug development method portfolio we need to establish whether a novel agent has significant anti-tumour efficacy in these clinically relevant models at plasma drug concentrations which are achievable without showing unreasonable toxicity in animals or patients. Currently, the most common preclinical models in use for haematological diseases are genetically engineered mouse strains and human tumour xenografts grown in immunocompromised mice [1]. Transgenic, knock-out and knock-in mouse models are relatively recent models that can mirror defined stages of human cancer development. Currently, the value of these models in predicting the clinical situation, both in terms of drug development, and in disease progression remains undefined. The choice of genetically engineered model must be carefully made to avoid over-interpretation of any results obtained [2]. Ultimately, all new therapies will need to be tested on human cells. Xenograft models have the advantage of showing a higher degree of consistency and predictability for disease development. Properly used and interpreted human tumour xenografts grown in immunodeficient mice are useful, though not entirely predictive of disease behaviour in patients, and will continue to allow us to make inroads into our understanding of leukemias, and to make important choices in preclinical drug development [2]. This review will focus on the role and development of human tumour xenograft models.
J.H. Vormoor (*) Northern Institute for Cancer Research, Newcastle University, Sir James Spence Institute, 4th floor, Royal Victoria Infirmary, Newcastle Upon Tyne, NE1 4LP, UK e-mail:
[email protected] V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3_6, © Springer Science+Business Media, LLC 2011
105
106
M.A. Batey and J.H. Vormoor
6.2 The In Vivo Xenograft Model The choice of a xenograft model is usually between two main sub types, ectopic and orthotopic. Ectopic (subcutaneous) models can help to identify preclinically relevant compounds, but as they are physiologically unrepresentative of the actual disease in most cases, with tumours growing as a defined mass on the flank of an animal, the information obtained from these xenografts has to be viewed with a degree of caution [3]. It is for example, rare to find metastatic involvement when using subcutaneous models [4] and the physical structure of the tumour may be dissimilar to the clinical example, leading to differences in tumour growth and pharmacology. Orthotopic models, where the implanted cells are grown in their respective clinically relevant host tissues, allow for the development of a more representative tumour microenvironment, and potentially a more faithful pattern of disease spread and metastasis. Recent animal models of acute lymphoblastic leukemia (ALL) have focused on orthotopic implantation of patient samples and cell lines, in an attempt to provide an optimal microenvironment for these cells leading to the closest simulation of clinical pathology.
6.3 The Severe Combined Immunodeficient Mouse The development of immunocompromised animals such as athymic mice (nu/nu) [5] enabled the use of human tumour explants and cell lines for xenotransplantation. Athymic mice lack functional T cells, however residual immunity in these animals limited the engraftment of human leukemia cells [6] and other models were required. It was not until the development more immunocompromised mouse strains such as the bg/nu/xid (BNX) mouse and the severe combined immunodeficient (SCID) mouse [7, 8], that leukemia cells were shown to reliably engraft in an in vivo model system [9]. The SCID mouse harbours a rare recessive mutation on chromosome 16 which is responsible for deficient activity of the protein kinase, DNA activated catalytic polypeptide (Prkdc), an enzyme involved in DNA repair. Since V(D)J recombination does not occur, the humoral and cellular immune systems fail to mature. This DNA repair defect also renders mice carrying the scid mutation twice as sensitive to ionising radiation when compared to wild type mice [10]. SCID mice have intact thymic stroma, but are severely deficient in functional T or B lymphocytes, and have no ability to activate the complement system, while the NK function is unaffected [11]. They cannot therefore efficiently fight infections, and do not reject many foreign tumours or transplants [7]. They are prone to opportunistic infection if housed conventionally, but in a pathogen free environment will survive normally for 12–24 months. Roughly 10% of these animals will develop a T-cell lymphoma over the course of their life caused by insertional mutagenesis by murine retroviruses [12]. A minority of older SCID mice become “leaky” producing functional B and T cells [10]. The BNX mouse has three important mutations.
6 An Overview on Animal Models of ALL
107
The nude (nu) mutation renders the animal athymic, the beige mutation (bg) reduces the number of NK cells and the xid mutation limits the number of lymphokine activated killer cells (LAK), which are important in host response to foreign cells. Much of the early work in this field was performed in these strains [13]. Once it had been shown that normal human hematopoietic cells could be engrafted in BNX [9] or SCID [14] mice it was postulated that these animals may provide a model for systemic development of human leukemia. SCID mice injected intravenously with leukemia patient samples and cell lines were shown to develop orthotopic systemic leukemia resembling human disease. The human leukemic cells cause infiltration of the murine bone marrow, spleen, liver, and CNS and circulation of leukemic blasts in the peripheral blood. Human cells can be monitored in the mice by flow cytometry using human specific antibodies such as those against CD44+ [15] (a cell surface glycoprotein involved in lymphocyte activation and haematopoiesis [16]) or CD45+ [17] (a pan haematopoietic marker).
6.4 The SCID Model in ALL Although it has been shown that engraftment of patient samples in SCID mice is not a significant indicator of patient outcome [18], the SCID model has been used to investigate several subtypes of ALL, including those carrying the t(4;11) translocation [19] and those with the E2A-PBX1 fusion transcript positive t(1;19) translocation [20]. In many cases, particular features of patient disease and higher engraftment levels in the mice have been shown to correspond with clinical observations. The utility of the SCID model for ALL has been found to be limited to the more aggressive patient samples as the residual immunity of these animals compromises their ability to engraft more benign disease [12]. In one large study of 681 patient bone marrow biopsy samples, only 15.3% successfully engrafted in SCID mice [18]. In attempts to overcome this, various methods have been used including irradiated mice, human cytokine stimulation, and high doses of human bone marrow (BM) cells. The implantation of cells into human foetal organs such as thymus and liver, or human foetal bone transplanted in SCID mice, the SCID-hu model, has been shown to enable the initiation of multilineage engraftment of purified CD34+ cells (a cell surface marker for undifferentiated pluripotent haematopoietic stem cells) in this microenvironment [21, 22]. Although studies using this approach have provided a great deal of information, it is not without its difficulties, primarily the need to obtain human foetal tissue and the ethical concerns surrounding this, as well as the need for the initial surgical procedure [23]. Work undertaken in the laboratory of John E. Dick based on the principles of bone marrow transplantation in humans and mice [13] showed that intravenous transplantation into irradiated recipient animals could result in human cells repopulating the mouse hematopoietic tissues [9, 24, 25].
108
M.A. Batey and J.H. Vormoor
6.5 The NOD/SCID Mouse To find better recipient mouse strains for human transplantation, further immune deficient animals created by homologous recombination were tested [10]. Shultz et al. [26] developed NOD/SCID mice by backcrossing the SCID strain with the NOD/LtSz strain. The NOD (non obese diabetic) mice have defects in the complement pathway and macrophage function [26] rendering the new mice with less residual immunity than the SCID strain. NOD/SCID mice also do not show autoimmune diabetes as they lack B and T cells due to the SCID mutation and a lower level of NK activity than the SCID animals. This lowering of immunity allows the NOD/ SCID mice to display higher levels of engraftment than SCID mice with both normal and leukemic human cells and cell lines, and enables engraftment with lower cell doses, as well as the engraftment of purified CD34+ cells and their subpopulations [27–29]. Interestingly ALL cells (unlike normal haematopoietic cells or myeloid leukemias) can engraft non-irradiated immune deficient mice such as the SCID [30] and the NOD/SCID strain [31], as well as irradiated mice, and indeed that total body irradiation may impair ALL engraftment in these animals [31]. It has been shown in NOD/SCID models using childhood ALL xenografts that responses can be obtained to conventional therapeutics such as vincristine and dexamethasone which mirror the observations following treatment in the original patients from whom the cells were derived [32]. These observations provide evidence that this model can afford an accurate representation of the clinical disease in preclinical setting and as such suggest potential for this type of model system in the more rapid transfer of interesting new agents into clinical trial. Further work investigating the biological characteristics of ALL patient samples at diagnosis and relapse in a similar NOD/SCID model [32, 33] showed that the immunophenotype of cells recovered from the mouse were essentially unaltered compared to the primary patient material, and that in some cases the pattern of clonal variation observed following tertiary transplantation in mice reflected that shown in patient bone marrow samples following clinical relapse. It has also been shown that following engraftment, implanted cells can be detected in peripheral murine blood, providing the basis for a continuous system of assessment of therapeutic efficacy and engraftment in NOD/ SCID mice [1]. These improvements in recipient mouse models made it possible to identify a cell type capable of initiating a multilineage human graft, the SCID repopulating cell (SRC) [34–36]. This system was found to be more robust than the SCID-hu model, generating many fold higher numbers of SRC cells.
6.6 Further Improvements in Recipient Animals It has been shown that the use of anti-CD122 antibody directed against the IL-2Rb chain (which targets several haematopoietic cell populations including NK cells and macrophages) could lead to improved engraftment of NOD/SCID mice [37]. More recently improved NOD/SCID mouse strains have been developed which
6 An Overview on Animal Models of ALL
109
lack both NK and macrophage activity, the NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ more commonly known as the NSG strain [38, 39], and the NOD.Cg-Prkdcscid Il2rgtm1Sug/ Jic or NOG strain [40]. The further depletion of immunity in these animals promises more efficient engraftment of leukemic cells than in conventional NOD/SCID mice and recent reports indicate that NSG mice are more susceptible to engraftment following injection of human leukemic cells [41]. Another interesting mouse strain is the Rag2(−/−)gamma(c)(−/−) which have a stable phenotype devoid of all T lymphocyte, B lymphocyte and NK cell function. These animals have been shown to engraft human tumour cells more efficiently than NSG/SCID mice under the same conditions [42]. The availability of these improved xenograft models is promising rapid progress for our understanding of the underlying biology in childhood ALL. Presently, many questions remain unanswered including the existence of the leukemic stem cell (LSC).
6.7 The Leukemic Stem Cell The sequential expression of immunophenotypic markers of B cell development has been well defined [43]. Many (such as CD10, CD22 and CD38) are present during normal precursor development, while others can be used to define the differing stages of this process. CD34 is expressed on haematopoietic stem cells while CD19 is a transcriptional target of PAX5, forming one of the earliest, B lymphoid-restricted markers. As such these cell surface proteins are useful tools for the identification of various leukemic blast subpopulations [44]. Although the idea of a malignant cancer stem cell was first proposed several decades ago [13], the first real evidence for these came from xenotransplantation studies in the early 1990s in acute myeloid leukemia [45]. Several subsequent studies suggested that a similar stem cell hierarchy could be applied to childhood ALL [46, 47]. It has been shown that in TEL/AML1-positive ALL, transduction of TEL/AML1 into human cord blood gives an abnormal CD34+CD19+CD38− candidate pre-leukemic cell population [48]. Only these cells had the ability to engraft primary and secondary NOD/SCID mice at levels between 0.5 and 8% human cells in murine bone marrow. Transplantation of CD34+CD19+CD38+ cells showed a much lower engraftment at a lower frequency and these cells were unable to engraft in secondary mice. These data suggested that the situation in ALL would be similar to that in acute myeloid leukemia (AML) and in normal haematopoiesis, with a subpopulation of LSC maintaining the malignancy. This is a central question regarding the LSC in childhood ALL, whether the malignant clone is maintained by a rare population of stem cells, or whether the majority of blasts’ cells have an unlimited proliferative capacity. Lymphoid malignancies may show differences to myeloid leukemia’s as mature B and T lymphocytes retain the ability to clonally expand [49]. In vitro studies have shown that in certain childhood ALL cases which could be maintained on stromal cell cultures, a high percentage of blast cells (up to 20%), are clonogenic [50].
110
M.A. Batey and J.H. Vormoor
It has recently been suggested that many of the current xenotransplantation and limiting dilution protocols underestimate the frequency of LSC [51] and that microenvironmental interactions play a critical role in the engraftment of these cells. There may be a LSC population with an impaired homing ability, but which retain the capacity to self renew, given the correct microenvironment. By directly transplanting FACs sorted cells intrafemorally into NOD/SCID mice, it has been shown that cells with limited homing abilities can indeed engraft [52], and that this procedure leads to an approximately tenfold more sensitive stem cell assay when compared to intravenous injection. Recent data utilising this direct intrafemoral injection technique in NSG mice has informed a new stem cell model for childhood ALL [44]. Each sorted blast population (CD34+CD19−, CD34+CD19+ and CD34−CD19+) was found to be able to transfer the leukemia onto primary, secondary and tertiary recipients. Furthermore, each sorted population also gave rise to all other populations present in the original leukemia, reconstituting the complete ALL phenotype. Using gene expression studies it was shown that the populations not only differed in terms of their cell surface markers, but at the level of RNA expression, showing that the different populations do indeed represent blasts at different levels of B cell maturation. These findings have been supported by further recent studies [53, 54], and therefore, evidence is accumulating that at least in some lymphoid leukemia’s blasts with self renewing capacity may exist at much higher frequency than was previously thought and that most, if not all, ALL blasts may possess properties currently associated with stem cells. The stem cell debate continues to rage [44, 55, 56], however it seems clear that these developments will require the re-evaluation of the accepted wisdom from previous xenograft models. It has been recently shown that in models of AML, human cells coated with anti-CD38 can be cleared in immune deficient mice [57]. Studies which have previously used anti-CD38 labelling in animal model systems of leukemia will have to be re-examined in light of these findings. It may be that certain current xenograft model systems underestimate the frequency of human stem cells and that the inability of a cell to engraft in a particular model does not prove that the cell lacks stemness.
6.8 Animal Models of ALL in Preclinical Testing Several ALL cell line models have been used in SCID mice for therapeutic efficacy studies. Agents tested in this way in ALL include topoisomerase I inhibitors [58], nucleoside and cAMP analogues [59, 60], epothilones [61] and tumour microenvironment targeting agents [62]. Xenograft models in SCID mice have also been used for the evaluation of immunotherapy [63, 64] and therapeutic efficacy has been demonstrated for novel agents using these models. As an example of current development, two drugs which are currently undergoing evaluation in the clinic in ALL are the novel mTOR inhibitor CCI-779 and the humanised anti-CD33 antibody conjugated with calicheamicin, gemtuzumab
6 An Overview on Animal Models of ALL
111
ozogamicin (Mylotarg). Both of these compounds gave promising results in preclinical models of ALL in immunocompromised mice [65, 66]. Preliminary results with mylotarg (Gemtuzumab Ozogamicin) in a clinical setting have shown some promise [67], and the results from the ongoing phase II clinical trial with CTI-779 will demonstrate whether the preclinical efficacy shown with this agent correlates into patient benefit. In the light of the more recent developments in animal modelling of ALL it may be that by applying our new protocols we can improve the clinical relevance of these preclinical tests, and provide a more efficient translation for novel therapies from the preclinical to the clinical setting. Certainly in our continued and improving use of animal models we should always be aware of their limitations. Progress in developing models of human leukemia has been limited at least in part by a lack of understanding of the microenvironmental factors necessary for successful engraftment. As previously mentioned the recent discovery that in some cases antibody labelling can impede engraftment of disease [57], should lead to caution in the design and interpretation of our preclinical experimental systems. Human cells in a xeno-environment do not necessarily faithfully mirror the patient situation and we need to appreciate this in our work with these models. In a drug development context, we must be aware of the differences in physiology, metabolism and pharmacokinetics between mouse and man. Although it has been shown that efficacy demonstrated in animal models of ALL can be predictive of clinical response, this is not always the case [32]. Immunocompromised mouse strains such as the NSG mouse, although proven to be good recipient mice for human leukemic cell engraftment, have defects in DNA repair which render them more sensitive to DNA damaging agents and will limit their use in the study of some novel chemotherapeutics or radiation therapy. The Rag2(−/−)gamma(c)(−/−) mouse strain has been shown to be much more resistant to ionising radiation [68] and may prove to be a useful tool in the study of ALL therapy.
6.9 Conclusions The recent data modifying the childhood ALL stem cell model [44, 69] and the realisation that historical models of ALL engraftment were biased against several populations of cells will require a reassessment of our understanding of the biology of this disease. We are only at the early stages of understanding how these cells self renew and propagate leukemia. Mouse strains such as the NSG mouse and techniques such as direct intrafemoral injection give us the ability to examine all cells involved in ALL and closely mimic human tumour pharmacology and microenvironment. By utilising the animal strains which continue to undergo development, we should be able to attain more representative animal models of ALL, which help to account for all blasts with LSC like behaviour. In this way we have access to the toolkit to examine both the pathology of ALL, and the most appropriate preclinical model for the testing of efficacy against this disease.
112
M.A. Batey and J.H. Vormoor
References 1. Lee EM, Bachmann PS, Lock RB. Xenograft models for the preclinical evaluation of new therapies in acute leukemia. Leukemia and Lymphoma, 2007. 48(4):659–668. 2. Sausville E, Burger AM. Contributions of human tumour xenografts to anticancer drug development. Cancer Research, 2006. 66:3351–3354. 3. Kerbel RS. Human tumour xenografts as predictive preclinical models for anticancer drug activity in humans: better than commonly perceived – but they can be improved. Cancer Biology Therapy, 2003. 2:134–139. 4. Khanna C, Hunter K. Modelling metastasis in vivo. Carcinogenesis, 2005. 26:513–523. 5. Lozzio BB, Machado EA, Lozzio CB, Lair S. Hereditary asplenic-athymic mice; transplantation of human myelogenous leukemic cells. Journal of Experimental Medicine, 1976. 143:225–231. 6. Cavallo F, Riccardi C, Forni M, Pericle F, Bosco MC, Giovarelli M, Soleti A, Forni G. Growth and dissemination of human malignant lymphoblast’s in immunosuppressed nu/nu mice. Natural Immunity and Cell Growth Regulation, 1991. 10:256–264. 7. Bosma GC, Custer RP, Bosma, MJ. A severe combined immunodeficiency mutation in the mouse. Nature, 1983. 301:527–530. 8. McCune JM, Mamikawa R, Kaneshima H, Schultz LD, Lieberman M, Weissman IL. The SCID-hu mouse: murine model for the analysis of human hematolymphoid differentiation and function. Science, 1988. 241:1632–1639. 9. Kamel-Reid S, Dick JE. Engraftment of immune deficient mice with human haematopoietic stem cells. Science, 1988. 242:1706–1709. 10. Lapidot T, Fajerman Y, Kollet O. Immune-deficient SCID and NOD/SCID mice models as functional assays for studying normal and malignant human hematopoiesis. Journal of Molecular Medicine, 1997. 75:664–673. 11. Dorshkind K, Pollack SB, Bosma MJ, Phillips RA. Natural killer (NK) cells are present in mice with severe combined immunodeficiency. Journal of Immunology, 1985. 134:3798–3801. 12. Leblond V, Autran B, Cesbron J-Y. The SCID mouse mutant: definition and potential use as a model for immune and hematological disorders. Hematology and Cell Therapy, 1997. 39:213–221. 13. Dick JE. Stem cell concepts renew cancer research. Blood, 2008. 112:4793–4807. 14. Mosier DE, Gulizia RJ, Baird SM, Wilson DB. Transfer of a functional human immune system to mice with severe combined immunodeficiency. Nature, 1988. 335:256–259. 15. Kamel-Reid S, Letarte M, Sirard C, Doedens M, Grunberger T, Fulop G, Freedman MH, Phillips RA, Dick JE. A model of human acute lymphoblastic leukemia in immune deficient SCID mice. Science, 1989. 246:1597–1600. 16. Ponta H, Sherman L, Herrlich PA. CD44: from adhesion molecules to signalling regulators. Nature Reviews in Molecular and Cellular Biology, 2003. 4:33–45. 17. Ballen KK, Valinski H, Greiner D, Shultz LD, Becker PS, Hsieh CC, Stewart FM, Quesenberry PJ. Variables to predict engraftment of umbilical cord blood into immunodeficient mice: usefulness of the non-obese diabetic severe combined immunodeficient assay. British Journal of Haematology, 2001. 114:211–218. 18. Uckun FM, Sather HN, Waurzyniak BJ, Sensel MG, Chelstrom L, Ek O. Prognostic significance of B-lineage leukemic cell growth in SCID mice: a Children’s Cancer group Study. Leukemia and Lymphoma, 1998. 30:503–514. 19. Uckun FM, Downing JR, Chelstrom LM, Gunther R, Ryan M, Simon J, Carroll AJ, TuelAhlgren L, Crist WM. Human t(4;11)(q21;q23) acute lymphoblastic leukemia in mice with severe combined immunodeficiency. Blood, 1994. 84:859–865. 20. Waurzyniak BJ, Heerema N, Sensel MG, Gaynon PS, Kraft P, Sather HN, Chelstrom L, Reaman GH, Uckun FM. Distinct in vivo engraftment and growth patterns of t(1;19)+/E2APBX1+ and t(9;22)+/BCR-ABL+ human leukemia cells in SCID mice. Leukemia and Lymphoma, 1998. 32:77–87.
6 An Overview on Animal Models of ALL
113
21. Baum CM, Weissman IL, Tsukamoto AS, Buckle A-S, Peault B. Isolation of a candidate human haematopoietic stem-cell population. Procedures of the National Academy of Science USA, 1992. 89:2804–2808. 22. Fraser CC, Kaneshima H, Hansteen G, Kilpatrick M, Hoffman R, Chen BP. Human allogenic stem cell maintenance and differentiation in a long term multilineage SCID-hu graft. Blood, 1995. 86:1680–1693. 23. Dick J.E. Normal and leukemic human stem cells assayed in SCID mice. Seminars in Immunology, 1991. 8:197–206. 24. Lapidot T, Pfulmio F, Doedens M, Murdoch B, Williams DE, Dick JE. Cytokine stimulation of multilineage haematopoiesis from immature human cells engrafted in SCID mice. Science, 1992. 255:1137–1141. 25. Vormoor J, Lapidot T, Pfulmio F, Risdon G, Patterson B, Broxmeyer HE, Dick JE. Immature human cord blood progenitors engraft and proliferate to high levels in severe combined immunodeficient mice. Blood, 1994. 83:2489–2497. 26. Shultz LD, Schweitzer PA, Christianson SW, Gott B, Schweiltzer IB, Tennent B, McKenna S, Mobraaten L, Rajan TV, Greiner DL, Leiter E. Multiple defects in innate and adaptive immunological function in NOD/LtSz-scid mice. Journal of Immunology, 1995. 154:180–191. 27. Dick JE. Normal and leukemic human stem cells assayed in SCID mice. Seminars in Immunology, 1996. 8:197–206. 28. Greiner DL, Hesselton RA, Shultz LD. SCID mouse models of human stem cell engraftment. Stem Cells, 1998. 16:166–177. 29. Lowary P, Shultz LD, Queensberry P. Multiple immune defects in the NOD/SCID mouse facilitate human hematopoietetic engraftment. Blood, 1994. 84 (Suppl 1): 346a. 30. Baersch G, Mollers T, Hotte A, Dockhorn-Dworniczak B, Rube C, Ritter J, Jurgens H, Vormoor J. Good engraftment of B-cell precursor ALL in NOD-SCID mice. Klinische Podiatry, 1997. 209:178–185. 31. Speigel A, Kollet O, Peled A, Abel L, Nagler A, Bielorai B, Rechavi G,Vormoor J, Lapidot T. Unique SDF-1 induced activation of human precursor-B ALL cells as a result of altered CXCR4 expression and signalling. Blood, 2004. 103:2900–2907. 32. Liem NL, Papa RA, Milross CG, Schmid MA, Tajbakhsh M, Choi S, Ramirez CD, Rice AM, Haber M, Norris MD, MacKenzie KL, Lock RB. Characterization of childhood acute lymphoblastic leukemia xenograft models for the preclinical evaluation of new therapies. Blood, 2004. 103:3905–3914. 33. Lock RB, Liem N, Farnsworth ML, Milross CG, Xue C, Tajbakhsh M, Haber M, Norris MD, Marshall GM, Rice AM. The nonobese diabetic/severe combined immunodeficient (NOD/ SCID) mouse model of childhood acute lymphoblastic leukemia reveals intrinsic differences in biologic characteristics at diagnosis and relapse. Blood, 2002. 99:4100–4108. 34. Conneally E, Cashamn J, Petzer A, Eaves C. Expansion in vitro of transplantable human cord blood stem cells demonstrated using a quantitative assay of their lympho-myeloid repoopulating activity in nonobese diabetic-scid/scid mice. Proceedings of the National Academy of Science USA, 1997. 94:9836–9841. 35. Larochelle A, Vormoor J, Hanenberg H, Wang JC, Bhatia M, Lapidot T, Moritz T, Murdoch B, Xiao XL, Kato I, Williams DA, Dick JE. Identification of primitive human hematopoietic cells capable of repopulating NOD/SCID mouse bone marrow: implications for gene therapy. Nature Medicine, 1996. 2:1329–1337. 36. Wang JC, Doedens M, Dick JE. Primitive human haematopoietic cells are enriched in cord blood compared with adult bone marrow or mobilzed peripheral cord blood as measured by the quantitative in vivo SCID-repopulating cell assay. Blood, 1997. 89:3919–3924. 37. McKenzie JL, Gan OI, Doedens M, Dick JE. Human short-term repopulating stem cells are efficiently detected following intrafemoral transplantation into NOD/SCID recipients depleted of CD122+ cells. Blood, 2005. 106:1259–1261. 38. Ohbo K, Suda T, Hashiyama M, Mantani A, Ikebe M, Miyakawa K, Moriyama M, Nakamura M, Katsuki M, Takahashi K, Yamamura K, Sugamura K. Modulation of hematopoiesis in mice with a truncated mutant of the interleukin-2 receptor gamma chain. Blood, 1996. 87:956–967.
114
M.A. Batey and J.H. Vormoor
39. Shultz LD, Lyons BL, Burzenski LM, Gott B, Chen X, Chaleff S, Kotb M, Gillies SD, King M, Mangada J, Greiner DL, Handgretinger R. Human lymphoid and myeloid cell development in NOD/LtSz-scid IL2R gamma null mice engrafted with mobilized human hemopoietic stem cells. Journal of Immunology, 2005. 174:6477–6489. 40. Ito M, Hiramatsu H, Kobayashi K, Suzue K, Kawahata M, Hioki K, Ueyama Y, Koyanagi Y, Sugamura K, Tsuji K, Heike T, Nakahata T. NOD/SCID/ gamma (c) null mouse: an excellent recipient mouse model for engraftment of human cells Blood, 2002. 100:3175–3182. 41. Agliano A, Martines-Padura I, Mancuso P, Marighetti P, Rabascio C, Pruneri G, Shultz LD and Bertolini F. Human acute leukemia cells injected in NOD/LtSz/IL-2R gamma null mice generate a faster and more efficient disease compared to other NOD/SCID related strains. International Journal of Cancer, 2008. 123:2222–2227. 42. Goldman JP, Blundell MP, Lopes L, Kinnon C, Di Santo JP, Thrasher AJ. Enhanced human cell engraftment in mice deficient in RAG2 and the common cytokine receptor gamma chain. British Journal of Haematology. 1998. 103:335–342. 43. Hystad ME, Myklebust JH, Bo TH, Sivertsen EA, Rian E, Forfang L, Munthe E, Rosenwald A, Chiorazzi M, Jonassen I, Staudt LM, Smeland EB. Characterization of early stages of human B cell development by gene expression profiling. Journal of Immunology, 2007. 179:3662–3671. 44. le Viseur C, Hotfilder M, Bomken S, Wilson K, Rottgers S, Schrauder A, Rosemann A, Irving J, Stam R W, Shultz LD, Harbott J, Jurgens H, Schrappe M, Pieters R, Vormoor J. In childhood acute lymphoblastic leukemia, blasts at different stages of immunophenotypic maturation have stem cell properties. Cancer Cell, 2008. 14:47–58. 45. Bonnet D, Dick JE. Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nature Medicine, 1997. 3:730–737. 46. Cobaleda C, Gutierrez-Cianca N, Perez-Losada J, Flores T, Garcia-Sanz R, Gonzalez M, Sanchez-Garcia I. A primitive hematopoietic cell is the target for the leukemic transformation in human philadelphia-positive acute lymphoblastic leukemia. Blood, 2000(95):1007–1013. 47. Cox CV, Evely RS, Oakhill A, Pamphilon DH, Goulden NJ, Blair A. Characterization of acute lymphoblastic leukemia progenitor cells. Blood, 2004. 104:2919–2925. 48. Hong D., Gupta, R., Ancliff, P., Atzberger, A., Brown, J., Soneji, S., Green, J., Colman, S., Piacibello, W., Buckle, V., Tsuzuki, S., Greaves, M., Enver, T. Initiating and cancer propagating cells in ETV6-RUNX1 associated childhood leukemia. Science, 2008. 319:336–339. 49. Vormoor J, Identifying the Acute Lymphoblastic Leukemia Stem Cell. 2007, ASCO. 50. Nishigaki H, Ito C, Manabe A, Kumagai M, Coustan-Smith E, Yanishevski Y, Behm FG, Raimondi SC, Pui CH, Campana D. Prevalence and growth characteristics of malignant stem cells in B-lineage acute lymphoblastic leukemia. Blood, 1997. 89:3735–3744. 51. Somervaille TC, Cleary ML. Identification and characterization of leukemia stem cells in murine MLL-AF9 acute myeloid leukemia. Cancer Cell, 2006. 10:253–254. 52. Mazurier F Doedens M, Gan OI, Dick JE. Rapid myeloerthroid repopulation after intrafemoral transplantation of NOD-SCID mice reveals a new class of human stem cells. Nature Medicine, 2003. 9:959–963. 53. Kong Y, Yoshida S, Saito Y, Doi T, Nagatoshi Y, Fukata M, Saito N, Yang SM, Iwamoto C, Okamura J, Liu KY, Huang XJ, Lu DP, Shultz LD, Harada M, Ishikawa F. CD34+CD38+CD19+ as well as CD34+CD38-CD19+ cells are leukemia-initiating cells with self renewal capacity in human B-precursor ALL. Leukemia and Lymphoma, 2008. 22:1207–1213. 54. Morisot S, Wayne AS, Bohana-Hashtan O, Kaplan IM, Hildreth R, Brown P, Stetler-Stevenson M, Civin CI. Leukemia Stem Cells (LSCs) Are Frequent in Childhood Precursor B Acute Lymphoblastic Leukemia (ALL). Blood (ASH Annual Meeting Abstracts), 2008. 112:1354. 55. Cox CV, Diamanti P, Evely RS, Kearns PR, Blair A. Expression of CD133 on leukemia initiating cells in childhood ALL. Blood, 2009. 113:3287–3296. 56. Dick JE. Looking ahead in cancer stem cell research. Nature Biotechnology, 2009. 27:44–46. 57. Taussig DC, Miraki-Moud F, Anjos-Alfonso F, Pearce DJ, Allen K, Ridler C, Lillington D, Oakervee H, Cavenagh J, Agrawal SG, Lister TA, Gribben JG, Bonnet D. Anti-CD38 antibody mediated clearance of human repopulating cells masks the heterogeniety of leukemia-initiating cells Blood, 2008. 112:568–575.
6 An Overview on Animal Models of ALL
115
58. Tomkinson B, Bendele R, Giles FJ, Brown E, Gray A, Hart K, LeRay JD, Meyer D, Pelanne M, Emerson DL. OSI-211, a novel liposomal topoisomerase I inhibitor, is active in SCID mouse models of human AML and ALL. Leukemia Research, 2003. 27:1039–1050. 59. Gourdeau H, Bibeau L,Ouellet F,Custeau D,Bernier L, Bowlin T,. Comparative study of a novel nucleoside analogue (Troxatyl, troxacitabine, BCH-4556) and Ara-C against leukemic human tumour xenografts expressing high or low cytidine deaminase activity. Cancer Chemotherapy and Pharmacology, 2001. 47:236–240. 60. Myers DE, Chandan-Langlie M, Chelstrom LM, Uckun FM. In vitro and in vivo anti-leukemic efficacy of cyclic AMP modulating agents against human leukemic B cell precursors. Leukemia and Lymphoma, 1996. 22:259–264. 61. Chou T-C, Zhang X-G, Harris CR, Kuduk SD, Balog A, Savin KA, Bertino JR, Danishefsky SJ. Desoxyepothilone B is curative against human tumour xenografts that are refractory to paclitaxel. Procedures of the National Academy of Sciences USA, 1998. 95:15798–15802. 62. Yoshida N, Ishii E, Nomizu M, Yamada Y, Mohri S, Kinukawa N, Matsuzaki A, Oshima K, Hara T, Miyazaki S. The laminin-derived peptide YIGSR (Tyr–Ile–Gly–Ser–Arg) inhibits human pre-B leukaemic cell growth and dissemination to organs in SCID mice. British Journal of Cancer, 1999. 80:1898–1904. 63. Uckun FM, Evans WE, Forsyth CJ, Waddick KG, Ahlgren LT, Chelstrom LM, Burkhardt A, Bolen J, Myers DE. Biotherapy of B-cell precursor leukemia by targeting genistein to CD-19 associated tyrosine kinases. Science, 1995. 267:886–891. 64. Waddick KG, Myers DE, Gunther R, Chelstrom LM, Cahandan-Langlie M, Irvin JD, Tumer N, Uckun FM. In vitro and in vivo antileukemic activity of B43-pokeweed antiviral protein against radiation-resistant human B-cell precursor leukemia cells. Blood, 1995. 86:4228–4233. 65. Golay J, Di Gaetano N, Amico D, Cittera E, Barbui AM, Giavazzi R, Biondi A, Rambaldi A, Introna M. Gemtuzumab ozogamicin (Mylotarg) has therapeutic activity against CD33+ acute lymphoblastic leukemias in vitro and in vivo. British Journal of Haematology, 2005. 128:310–317. 66. Teachey DT, Obzut DA, Cooperman J, Fang J, Carroll M, Choi JK, Houghton PJ, Brown VI, Grupp SA. The mTOR inhibitor CCI-779 induces apoptosis and inhibits growth in preclinical models of primary adult human ALL. Blood, 2006. 107:1149–1155. 67. Cheung K-C, Wong L-G, Yeung, Y-M. Treatment of CD33 positive refractory acute lymphoblastic leukemia with Mylotarg. Leukemia and Lymphoma, 2008. 49:596–597. 68. Chicha L, Tussiwand R, Traggiai E, Mazzucchelli L, Bronz L, Piffaretti J-C, Lanzavecchia A, Manz MG. Human Adaptive Immune System rag2 -/-gamma(c) -/- mice. Annals of the New York Academy of Sciences, 2005. 1044:234–243. 69. Kong D, Gupta R, Ancliff P, Atzberger A, Brown J, Soneji S, Green J, Colman S, Piacibello W, Buckle V, Tsuzuki S, Greaves M, Enver T. Initiating and cancer propagating cells in TELAML1-associated childhood leukemia. Science, 2008. 319:336–339.
Chapter 7
Targeting Bcl-2 Family Proteins in Childhood Leukemia Guy Makin and Caroline Dive
7.1 Apoptosis and Cancer Recent years have seen an enormous increase in our understanding of the regulation of apoptosis. It is now clear that apoptosis is a fundamental biological process that is essential for the prevention of tumour development [1]. In order to protect the organism from the effects of unrestrained proliferation, normal cells are only able to maintain division when in receipt of the correct environmental cues, and in the absence of these, proliferation is restrained by apoptosis, thus ensuring that cell division can only occur in the right place and at the right time. Elegant experimental models have demonstrated that driving proliferation in the context of a functional apoptotic pathway leads to cell death, whilst the same proliferative stimulus (in this case c-myc), once apoptosis has been disabled by over-expression of the antiapoptotic Bcl-2 family protein Bcl-xL, leads to unrestrained proliferation, invasion and metastasis [2]. Apoptosis is regulated by two pathways, the extrinsic or death receptor pathway, and the intrinsic or mitochondrial pathway (Fig. 7.1). In the extrinsic pathway, cell surface death receptors are activated by binding of their cognate ligands (Fas, TNFa, TRAIL), leading to the activation, via intracellular adaptor domains, of caspase 8. In some cell types, sufficient caspase 8 can be activated by this pathway to activate downstream effector caspases and trigger apoptosis, but in others the mitochondrial pathway must be recruited via the BH3-only protein Bid to ensure sufficient caspase activation for cell death [3]. The mitochondrial pathway of apoptosis is regulated by the Bcl-2 family of proteins. The interaction between the members of this family controls the release of
G. Makin (*) Clinical and Experimental Pharmacology, Paterson Institute for Cancer Research, University of Manchester, Manchester, UK and School of Cancer and Enabling Sciences, University of Manchester, Manchester, UK e-mail:
[email protected]
V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3_7, © Springer Science+Business Media, LLC 2011
117
118
G. Makin and C. Dive
Extrinsic pathway Death receptor ligation (Fas)
Intrinsic pathway Drug-induced damage signals
Bad, Noxa, Puma
DISC Pro-caspase-8
Bid, Bim
Bcl-2 Bcl-xL
Bid
Bax
Bak Bax
tBid
Cytochrome c
Smac
IAPS
Apoptosome Apaf-1,dATP pro-caspase 9
Effector caspases (3,6.7)
Substrates
Cell death
Fig. 7.1 The two pathways of apoptosis. In the extrinsic pathway, ligation of cell surface death receptors by their cognate ligands leads to the formation of the death induced signalling complex (DISC) and the activation of caspase 8. Caspase 8 is then able to activate effector caspases. Cleavage of bid by caspase 8 is also able to recruit the intrinsic pathway. In the intrinsic pathway, drug-induced damage is recognised and results in the activation of both BH3-only and multi-domain pro-apoptotic Bcl-2 family proteins, which can be prevented by the multi-domain anti-apoptotic proteins. Once Bak and Bax are activated then cytochrome c and smac are released from the mitochondrial intermembrane space, where cytochrome c allows the formation of the apoptosome and the activation of caspase 9, whilst smac prevents inhibitor of apoptosis proteins (IAPs) from blocking caspase activation
apoptogenic factors from the mitochondrial intermembrane space into the cytosol. There, they are able to, via the formation of a macromolecular structure known as the apoptosome, activate caspase 9 and downstream effector caspases [4]. Bcl-2 proteins fall into three groups. Bcl-2, Bcl-xL, Bcl-w and Mcl-1 are multi-domain anti-apoptotic proteins containing four Bcl-2 homology (BH) domains and function at membranes, especially at the mitochondrial outer membrane. Bak and Bax are multi-domain pro-apoptotic proteins containing
7 Targeting Bcl-2 Family Proteins in Childhood Leukemia
119
three BH domains and can be cytosolic or mitochondrial in location, but seem to be functional only at membranes; both proteins require activation. Either Bak or Bax is essential for cell death via the mitochondrial pathway [5]. The third group of Bcl-2 family proteins are also pro-apoptotic but contain only a single BH3 domain. These proteins, which include Bid, Bim, Bad and Noxa, regulate the interactions between the multi-domain proteins (Fig. 7.2). BH3-only proteins are of two different types. Bid and Bim are “activators” and are able to induce apoptosis in response to a range of signals, including cytotoxic drug damage, by directly activating Bax and Bak, leading to their oligomerisation and the release of apoptogenic factors from the mitochondrial intermembrane space [6, 7]. In this scenario, the anti-apoptotic Bcl-2 family proteins can prevent apoptosis, despite the activation of activator BH3-only proteins, by sequestering these proteins via binding between the amphipathic a helix of the BH3-only protein and the hydrophobic groove formed by the BH1, BH2 and BH3 domains of Bcl-2/Bcl-xL/Bcl-W/Mcl-1. In tumour cells, activator BH3 only proteins are more likely to be sequestered than in normal cells, and thus, these malignant cells are dependent upon the anti-apoptotic proteins (and can be considered to be primed for death) [8]. The other BH3-only proteins are unable to induce apoptosis directly. This group, which includes Bad and Noxa, compete for the BH3 binding groove of the anti-apoptotic Bcl-2 family proteins and thus are able to displace activator BH3-only proteins from Bcl-2 type proteins. These BH3-only proteins are referred to as “sensitisers”. The interaction between sensitiser BH3-only proteins and anti-apoptotic proteins is selective. Bad binds very well to Bcl-2 but very poorly to Mcl-1, whilst Noxa binds poorly to Bcl-2 but very well to Mcl-1. Peptides derived from these “sensitiser” BH3-only proteins have the same selective binding characteristics as the proteins and can be used in a technique referred to as BH3 profiling, to determine the dependence of a particular cell type on an anti-apoptotic protein for survival [6]. Using this technique, primary samples from patients with CLL demonstrated uniform dependence upon Bcl-2 [9]. In diffuse large B-cell lymphoma (DLCBL), only about 50% of tumours express high levels of Bcl-2, and thus, BH3 profiling was able to reveal different types of apoptotic block. In a panel of 18 DLCBL cell lines, three different levels of apoptotic block could be detected. Induction of loss of mitochondrial outer membrane potential (MOMP) after exposure of isolated mitochondria to sensitiser BH3 peptides indicated a proportion of DLCBL lines to be primed for apoptosis. Those that were dependent upon Bcl-2 alone could be distinguished from those dependent upon Bcl-2 and Mcl-1 by the difference in their response to Bad peptides that are less effective at inducing loss of MOMP in cells that are dependent on Mcl-1. Some DLCBL lines showed minimal response to sensitiser BH3 peptides, but loss of MOMP was induced by activator BH3 peptides (Bim and Bid), showing that in these cells the block to apoptosis is at the level of activation of BH3-only proteins. The final group of cell lines showed poor response to both sensitiser and activator BH3 peptides, indicating a block at the effector level [8]. This technique
120
G. Makin and C. Dive
Activator BH3only
Sensitizer BH3only
Bad Bid Puma Bim
Noxa
Bmf
Bik
Pro-apoptotic multidomain
Anti-apoptotic multidomain
Bax
Bcl-2
Bcl-XL
Bak
Bcl-w
Mcl-1
Cell death
Fig. 7.2 A model of the interactions between Bcl-2 family proteins. The pro-apoptotic multi-domain proteins Bax and Bak are required for death. Activator BH3-only proteins such as Bid and Bim can be sequestered by the anti-apoptotic multi-domain proteins, thus preventing activation of Bax and Bak. Sensitiser BH3-only proteins are able to displace activator proteins from the anti-apoptotic proteins, releasing them to activate Bax and Bak. Different sensitiser BH3-only proteins bind with different affinities to different anti-apoptotic proteins
7 Targeting Bcl-2 Family Proteins in Childhood Leukemia
121
predicted both the differences in protein expression of Bcl-2 family proteins across these DLCBL cell lines and their response to the Bcl-2/Bcl-xL smallmolecule inhibitor ABT-737 (see below)
7.2 Bcl-2 Family Proteins in Haematological Malignancy Bcl-2 is over-expressed in a number of haematological malignancies, and this has been correlated with advanced disease and poor prognosis after treatment. Experimental systems have shown directly that anti-apoptotic Bcl-2 family proteins are important both in the maintenance of leukemia and in its response to cytotoxic therapy. The Em-myc transgenic mouse develops a spontaneous B-cell lymphoma, often with a prolonged latent period [10], whilst mice that also express human Bcl-2 develop an immature lymphoblastic leukemia within a few days of birth [11]. Triply transgenic mice in which expression of human Bcl-2 can be repressed by doxycycline also develop this rapid lymphoblastic leukemia with elevated peripheral white blood cell counts. Repression of Bcl-2 in these mice by feeding them doxycycline leads to normalisation of the peripheral white count within 10 days of treatment, followed by clearance of the bone marrow, return of normal haemopoiesis, and normalisation of splenic size within 2 weeks of starting doxycycline. In addition, the doxycycline-treated mice had significantly prolonged survival in comparison with the untreated controls. In keeping with the roles of activating and sensitising BH3-only proteins discussed earlier, mitochondria isolated from leukemic cells in the triply transgenic mice underwent loss of MOMP after exposure to peptides derived from the sensitiser BH3only protein Bad [12]. A major advantage of the transgenic mouse model is the ability to evaluate the response of manipulated cells to cytotoxic therapy both in vitro, and through transplantation into syngeneic mice, in vivo. Isolated primary lymphoma cells from peripheral lymph nodes of Em-myc transgenic mice can be readily infected with a retroviral vector over-expressing human Bcl-2 within 4 days of culture. In vitro over-expression of Bcl-2 resulted in almost complete resistance to apoptosis induced by doxorubicin, maphosphamide or docetaxel in a short-term viability assay. This differential effect between Bcl-2 over-expressing and control cells was lost with time due to the strong selective pressure for acquired drug resistance in culture. As a result, Bcl-2 over-expressing cells had no survival advantage over control cells in clonogenic assay. However once Bcl-2 over-expressing lymphoma cells were transplanted into syngeneic mice, a dramatic effect on tumour response to doxorubicin, cyclophosphamide or docetaxel could be seen, with Bcl-2 over-expressing tumour-bearing mice having a highly significantly worse survival than those with normal Bcl-2 levels [13]. Thus, in transgenic mouse models Bcl-2 is of vital importance in leukemia, both for the maintenance of disease and in survival in conferring pleiotropic drug resistance. Hence, therapeutic strategies targeting Bcl-2 are a natural progression in haematological malignancy.
122
G. Makin and C. Dive
7.3 Targeting Bcl-2 in Haematological Malignancy 7.3.1 Bcl-2 Antisense: Oblimersen (G3139) Antisense oligonucleotides to Bcl-2 were one of the earliest therapeutic strategies to be developed to target the apoptotic pathway, and indeed antisense Bcl-2 has undergone extensive clinical evaluation, especially in malignant melanoma, and also in a wide range of haematologic malignancies. As early as 1994, phosphorothioate antisense oligonucleotides to Bcl-2 were shown to specifically downregulate Bcl-2 protein and to inhibit the growth of 697 leukemia cells in a concentration-dependent manner [14]. Bcl-2 antisense oligonucleotides also showed specific effects against Bcl-2-expressing AML cell lines and primary AML cells in culture and sensitised AML cells to daunorubicin and cytarabine [15]. This promising early work led to the development of G3139 (Genasense/oblimersen), an 18-mer phosporothioate oligonucleotide designed to bind to the first six codons of human Bcl-2 mRNA. In phase 1 study in patients with relapsed or refractory acute leukemia (17 AML, 3 ALL), oblimersen was given as a continuous infusion for 10 days at doses of 4 mg/kg/day and 7 mg/kg/day in combination with escalating doses of fludarabine and cytarabine (FLAG) on days 6–10. Toxicity was no greater than anticipated for FLAG alone, and dose-limiting toxicity was not achieved. Bcl-2 mRNA in bone marrow at day 5 was downregulated in 9 of 12 patients. Six patients showed complete response to this combination. Thus, Bcl-2 antisense can be safely delivered with FLAG chemotherapy to patients with refractory or relapsed acute leukemia and produces a significant response rate [16]. In poor-risk older patients with untreated AML, the combination of 7 mg/kg/day oblimersen as a continuous infusion for 10 days with cytarabine (days 4–10) and daunorubicin (days 4–6) resulted in CR in 14 of 29 patients. Steady-state plasma concentrations of oblimersen were achieved within 24 h and remained stable until the end of the infusion. After 72 h of oblimersen, Bcl-2 mRNA and protein levels were significantly decreased in the bone marrow of patients who achieved CR in comparison with those who did not. Toxicity of the combination of oblimersen and cytotoxics was no greater than with cytotoxics alone [17]. Significant levels of oblimersen could be detected in bone marrow and blood mononuclear cells after between 72 and 120 h of continuous oblimersen infusion. Four of six patients who had intracellular oblimersen concentrations >5 pmol/mg protein showed downregulation of Bcl-2 mRNA in bone marrow [18]. Oblimersen has potent activity as a single agent against imatinib-resistant Bcr-Abl-positive AML cell lines grown as subcutaneous xenografts in nude mice and is able to sensitise these cells to apoptosis induced by imatinib, cytarabine, daunorubicin and etoposide [19]. The combination of imatinib and oblimersen against imatinib resistant CML was studied in the Cancer and Leukemia Group B (CALGB) study 10107. Oblimersen was given as a continuous intravenous infusion over 10 days every 3 weeks, with daily imatinib, at doses of 4 mg/kg/day oblimersen and 600 mg daily imatinib, 7 mg/kg/day oblimersen and 600 mg daily imatinib and
7 Targeting Bcl-2 Family Proteins in Childhood Leukemia
123
7 mg/kg/day oblimersen and 800 mg daily imatinib. Twenty-one patients were treated without significant toxicity. Six patients achieved complete haematological response, with one also achieving complete cytogenetic response. However, the aim of the study was to achieve a reduction of more than 30% in the number of Bcr-Abl-positive metaphase cells in 20% of patients, and thus the combination was not deemed to be a success [20]. Antisense oligonucleotides to Bcl-2 specifically inhibit the expression of Bcl-2 mRNA in chronic lymphocytic leukemia cell lines in vitro and are able to decrease Bcl-2 protein levels and induce apoptosis in primary leukemic cells as well as sensitising CLL cells to dexamethasone, fludarabine, chlorambucil and rituximab [21–23]. In phase I–II study in 40 patients with relapsed or refractory CLL at doses from 3 to 7 mg/kg/day as a 5-day continuous infusion, administration of oblimersen was limited by a cytokine release syndrome needing dose reduction from 7 mg/kg/ day down to 3 mg/kg/day. In the phase II phase of this study at 3 mg/kg/day as a continuous infusion for 5 days, there was little toxicity, and two partial responses were seen in 26 evaluable patients, lasting from 2 to 6 months. Thirteen patients had stable disease for more than 2 months, and 11 patients had progressive disease by the end of the second cycle of oblimersen [24]. In a large multicentre international phase III study, the addition of oblimersen to fludarabine and cyclophosphamide was of benefit. Two hundred and forty-one patients with CLL who had received at least one previous course of fludarabine containing chemotherapy were randomised to 28-day cycles of fludarabine 25 mg/m2/day with cyclophosphamide 250 mg/m2/ day for 3 days, with or without oblimersen 3 mg/kg/day as a 7-day continuous infusion beginning 4 days before chemotherapy. Twenty patients (17%) in the oblimersen group achieved CR or nodular PR (nPR), compared to eight patients (6%) receiving chemotherapy alone (and this difference was statistically significant). Maximum benefit from the addition of oblimersen was seen in those patients who had a partial or better response to fludarabine for at least 6 months. In this group, the CR/nPR rate was 25% with oblimersen compared to 6% without. At a minimum of 24 months of follow-up, 5 of 20 patients receiving oblimersen had relapsed compared to six of eight in the chemotherapy-only group, and median duration of response was 20 months in the chemotherapy-only group and greater than 31 months in the oblimersen group, which difference was highly significant [25]. In a phase II study in patients with relapsed multiple myeloma in combination with thalidomide and dexamethasone, oblimersen showed encouraging efficiency. Thirtythree patients received oblimersen as a 7-day continuous infusion at doses from 5–7 mg/kg/day, in combination with 4 days of dexamethasone and continuous thalidomide. Cycles were given at 21-day intervals for three cycles and then every 35 days with a reduced dose of dexamethasone. Patients with continued clinical response could receive up to 2 years of therapy. A median of eight cycles of therapy per patient was given. The most frequent toxicities were myelosuppression, which was manageable, and peripheral neuropathy, which required dose reduction of the thalidomide. Seventythree percent of patients had documented responses, including 2 CRs, 4 near CRs and 12 PRs. Median duration of response was 13 months, suggesting that this combination is safe and effective in patients with refractory and relapsed multiple myeloma [26].
124
G. Makin and C. Dive
Bcl-2 is expressed in most EBV-associated lymphoproliferative disorders. Oblimersen is able to downregulate Bcl-2 protein, inhibit proliferation and induce apoptosis in EBV immortalised lymphoblastoid cell lines (LCL) in vitro. Although oblimersen was able to prevent the establishment of tumours by EBV-immortalised LCLs in SCID mice when given on the first day after tumour inoculation, it was not able to prevent the growth of established tumours, although it did significantly prolong the survival of these mice in comparison with nonsense oligonucleotide treatment [27]. When the humanised monoclonal anti-CD20 antibody rituximab was combined with oblimersen, there was significant enhancement of antiproliferative effect, and enhanced apoptosis in EBV transformed LCLs in vitro. In the SCID mouse model of human lymphoproliferative disease, the combination of rituximab and oblimersen led to significantly prolonged survival in comparison with either therapy alone, and the majority of animals receiving this combination had no detectable tumour [28]. Pre-clinical experiments in B-cell NHL cell lines have demonstrated that combining oblimersen with the proteasome inhibitor bortezomib enhances its growth inhibitory effect both in vitro and against xenografts in SCID mice. The combination of oblimersen followed by bortezomib and cyclophosphamide was statistically superior to other combinations and led to a number of complete remissions [29]. In a phase II study in B-cell NHL, significant responses were seen in combination with rituximab. Forty-eight patients with recurrent or refractory B-cell NHL, of whom 70% had received prior rituximab, were treated with oblimersen at 3 mg/kg/day as a 7-day continuous infusion for alternate weeks in a 6-week block, in combination with 375 mg/m2 of rituximab given on days 3, 8, 22, 25, 29, 36. Of the 42 evaluable patients, ten achieved CR and eight achieved PR, giving an overall response rate of 42%. The combination was well tolerated with myelosuppression, the main toxicity seen. Thus, oblimersen can be safely administered in combination with rituximab and bortezomib and offers potential therapeutic benefit in B-cell NHL [30]. Infant acute leukemias often contain the MLL gene rearrangement, most commonly caused by the (4:11) translocation. This has been associated with cytotoxic drug resistance and is correlated with poor prognosis [31]. In vitro data suggest that oblimersen is effective in reducing Bcl-2 mRNA and protein levels in t(4:11) MLL-rearranged cell lines and not only induces apoptosis by itself but also sensitises these cells to doxorubicin, 6-thioguanine, etoposide and cytarabine [32]. Oblimersen has completed phase 1 study in combination with doxorubicin and cyclophosphamide in children with relapsed solid tumours. Thirty-seven patients were enrolled, but eight were not evaluable for toxicity. The range of diagnoses included Ewing sarcoma, osteosarcoma, Wilms tumour and neuroblastoma. Oblimersen was given as a continuous 7-day infusion at doses from 3 to 7 mg/kg/ day. Doxorubicin 30 mg/m2/day and cyclophosphamide 500 mg/m2/day were given on days 5 and 6. The main toxicities were haematological, and these could be managed by dose reduction of the doxorubicin and cyclophosphamide. The MTD was 7 mg/kg/day oblimersen in combination with these agents. One patient with Ewing sarcoma had a partial response, and eight patients had prolonged stable
7 Targeting Bcl-2 Family Proteins in Childhood Leukemia
125
disease, receiving a median of seven courses. Seventy-three percent of evaluable patients had a reduction in Bcl-2 protein level in PBMCs by day 5 or 6 of the oblimersen infusion [33].
7.4 Bcl-2 Small-Molecule Inhibitors 7.4.1 HA14-1 HA14-1 is a small non-peptide organic ligand that binds to the hydrophobic surface pocket of Bcl-2 that is required to mediate the protein–protein interactions between Bcl-2 family proteins. HA14-1 was discovered using a computer screening process that models the ligand structure needed to bind this pocket. HA14-1 is able to induce apoptosis in HL60 AML cells at micromolar concentrations in a caspasedependent fashion that involves loss of mitochondrial membrane potential [34] and activation of the mitochondrial pathway of apoptosis. HA14-1 is able to induce apoptosis in a range of haematopoietic cell lines, including ALL, AML, CML, histiocytic lymphoma and multiple myeloma. Sensitivity to HA14-1 across this panel of cell lines correlated with expression of Bcl-2; cell lines showing the lowest Bcl-2 expression were the least sensitive to HA14-1. HA14-1 also induced apoptosis in primary AML blasts derived from two separate patients and was relatively less toxic to normal bone marrow progenitor cells. HA14-1 was also able to sensitise lymphoblastic leukemia cells to cytarabineinduced apoptosis [35]. HA14-1 induced apoptosis in primary CLL cells with an EC50 of less than 50 mM in 26 of 36 patient samples, as well as in CD19-positive B cells from patients with mantle cell lymphoma and splenic marginal zone lymphoma. HA14-1 was also able to induce apoptosis in CLL cells with p53 mutations or loss of ATM [36]. In multiple myeloma cell lines, HA14-1 is able to enhance apoptosis induced by the proteasome inhibitor bortezomib (Velcade®), which has been one of the most promising novel agents against this traditionally very chemo-resistant malignancy. This enhancement of apoptosis by HA14-1 was only seen when multiple myeloma cell lines were exposed to bortezomib for 10 h before exposure to HA14-1 and not when the two agents were given simultaneously, nor when HA14-1 was given before bortezomib, and could not be blocked by IL-6, well documented as a survival factor for MM cells [37]. Similar interactions between these agents were seen in dexamethasone-resistant multiple myeloma cells and between HA14-1 and an alternative proteasome inhibitor MG132. Formal analysis demonstrated that this interaction was synergistic. The combination of bortezomib and HA14-1 produced a dramatic increase in all markers of activation of the mitochondrial pathway of apoptosis, as well as in reactive oxygen species (ROS), which was abrogated by co-treatment of cells with the free radical scavenger L-NAC, which also reduced the loss of mitochondrial membrane potential. The combination of bortezomib and
126
G. Makin and C. Dive
HA14-1 produced a marked activation of JNK, which was blocked by the JNK inhibitor SP600125, which also significantly reduced bortezomib/HA14-1-induced apoptosis, implying that both ROS and JNK activation play a role in bortezomib/ HA14-1-induced apoptosis in MM cell lines. Constitutive mitogen-activated protein kinase (MAPK) signalling is frequent in primary AML blasts and promotes AML blast survival [38]. In AML cell lines with high constitutive MAPK activity, the simultaneous inhibition of the MAPK pathway with the MEK inhibitor PD184352 and treatment with HA14-1 resulted in a dramatic decrease in cell viability that was not observed in an AML cell line without constitutive MAPK activation. The combination increased loss of mitochondrial membrane potential and caspase activation, and inhibition of MAPK specifically inhibits the expression of survivin in AML cells with constitutive MAPK activity. The interaction between HA14-1 and the MEK inhibitor in AML cells was synergistic in three different fixed ratio combinations. Simultaneous exposure to PD184352 significantly enhanced the colony-inhibiting effect of HA14-1 against primary AML samples [39]. Macroautophagy (usually referred to as autophagy) is a tightly regulated lysosome dependent pathway that results in the formation of autophagosomes around cellular organelles, and their subsequent degradation. It is usually regarded as a cellular survival mechanism under conditions of nutrient deprivation. However, there is increasing evidence that autophagy may represent an important mode of cell death distinct from apoptosis in certain situations [40]. Anti-apoptotic Bcl-2 family proteins have been suggested to inhibit autophagy as well as apoptosis, through their interaction with the protein Beclin 1 [41]. However, data are also beginning to emerge to suggest that inhibition of steroid-induced apoptosis in lymphocytes by Bcl-2 over-expression leads to induction of autophagy instead and may be an important resistance mechanism [42]. Inhibition of Bcl-2 by HA14-1 induces both apoptosis and autophagy in L1210 murine leukemia cells, and inhibition of autophagy with the PI3K inhibitor wortmannin enhanced the induction of apoptosis, whilst blocking apoptosis with the caspase inhibitor zDEVD-fmk increased autophagy [43]. The induction of autophagy in L1210 cells by the novel anticancer agents XK469 or SH80 does not prevent the rapid induction of apoptosis by HA14-1, so at least in this model pre-existing autophagy does not reduce the efficacy of Bcl-2 targeting therapeutics [44].
7.5 Obatoclax (GX15-070) Obatoclax is a hydrophobic cycloprodigiosin-derived BH3 mimetic identified by chemical library screening. Unlike ABT-737, obatoclax is able to inhibit the binding of BH3 peptides to a wide range of Bcl-2 family proteins, including Mcl-1 and A1 [45]. The hydrophobic nature of the compound means that assays done in aqueous phase tend to underestimate the binding activity of obatoclax. An in silico docking algorithm predicts that obatoclax binds in a hydrophobic pocket at one end
7 Targeting Bcl-2 Family Proteins in Childhood Leukemia
127
of the Bcl-2 BH3 binding groove, which would be in close proximity to a lipid bilayer in membrane integrated Bcl-2, which would thus facilitate the entry of obatoclax to the BH3 binding groove in vivo [46]. Pre-treatment of isolated mitochondria with obatoclax in nanomolar concentrations prevented the formation of dimers between Bak and Mcl-1 and the ability of obatoclax to induce apoptosis is dependent upon the presence of Bax or Bak. The ability of obatoclax to inhibit the binding of Mcl-1 to Bak/Bax allows it to function in concert with ABT-737, and the combination of the two agents was able to induce apoptosis in Mcl-1 over-expressing KB oral carcinoma cells, to the same extent as ABT-737 and Mcl-1 siRNA. Obatoclax is also able to overcome Mcl-1-mediated resistance to the proteasome inhibitor bortezomib (Velcade®) in multiple myeloma cells [46]. CLL is the most common leukemia in the developed world and is characterised by the presence of large numbers of CD5-positive B cells, which tend to express high levels of anti-apoptotic Bcl-2 family proteins that contribute to relative resistance to cytotoxic chemotherapy. Obatoclax induces apoptosis in B cells derived from 9 of 11 CLL patients with an EC50 of 1.7 mM, and showed additive toxicity with fludarabine and chlorambucil in cells derived from five of these nine patients [36]. In a separate study of cells derived from 20 CLL patients, a 40-h treatment with obatoclax induced apoptosis with the same EC50, and all patients showed response to obatoclax, despite the presence of known high-risk genetic variables/genotypes such as deletions of chromosome 11q and 17p [47]. In these CLL cells, obatoclax induces the dissociation of Mcl-1 from Bak and the release of Bim from Bcl-2, and this leads to the loss of mitochondrial membrane potential within 3 h of exposure to obatoclax. CLL cell lines transfected with mutant Bcl-2 that mimics the phosphorylation of Bcl-2 on serine 70 were more resistant to obatoclax than wildtype cells, and obatoclax was less able to release Bim from this phosphorylated form of Bcl-2. Blocking the phosphorylation of Bcl-2 on serine 70 with the ERK inhibitor PD98059 enhanced the toxicity of obatoclax in CLL cells, and indeed, the two agents were highly synergistic in combination. Increasing the amount of Bcl-2 phosphorylated at serine 70 with okadaic acid treatment significantly decreased the cytotoxicity of obatoclax. Obatoclax was synergistic with bortezomib in these CLL cells, and this synergy was increased further by reducing pBCL-2 (ser70) levels by pre-treatment with PD98059 [47]. In 15 of 16 multiple myeloma cells, submicromolar concentrations of obatoclax induced significant reductions in cell viability with a mean EC50 of 246 nM. Obatoclax was also effective against dexamethasone- and melphalanresistant MM cell lines. The presence of IGF-1 did not affect cellular response to obatoclax, and IL-6 had only modest protective effect. Importantly, obatoclax had minimal toxicity against bone marrow stem cells. In MM cell lines, obatoclax causes the dissociation of Mcl-1 from Bak and induces apoptosis via the mitochondrial pathway. Obatoclax is also able to induce apoptosis at nanomolar in primary MM cells from 8 of 14 patients without toxicity to normal peripheral blood lymphocytes. Obatoclax was synergistic with dexamethasone, and additive with melphalan and bortezomib, when given before the proteasome inhibitor [48].
128
G. Makin and C. Dive
Micromolar concentrations of obatoclax inhibit the growth of AML cell lines by the induction of apoptosis, as well as by mitotic arrest. As in MM cells, obatoclax causes the dissociation of Bak from Mcl-1 but in addition efficiently caused the release of Bim from Bcl-2 and Mcl-1. Obatoclax was synergistic with ABT-737 in the ABT-737-resistant AML cell line OCI-AML3 and also synergised with cytarabine in a schedule-independent fashion. Micromolar concentrations of obatoclax were also able to induce apoptosis in primary AML samples and reduce their colonyforming ability without affecting that of normal bone marrow [49]. Obatoclax has now completed early-phase clinical trial in adults against CLL and refractory haematologic malignancies including AML, ALL, CML, CLL and myelodysplastic syndrome. In advanced CLL previously treated with a median of 4 chemotherapy regimes including fludarabine, 26 patients received a total of 74 cycles of obatoclax thrice weekly. Doses were 3.5–14 mg/m2 as a 1 h infusion or 20–40 mg/m2 as a 3 h infusion. The maximum tolerated dose was 28 mg/m2 over 3 h every 3 weeks. Dose-limiting toxicities were all neurological and included somnolence and euphoric mood at the time of the infusion. Biological effects of obatoclax were observed in 12 of 16 patients with serial samples, where activation of Bax and Bak in peripheral blood mononuclear cells persisted beyond the end of the infusion. A dose-related increase in circulating oligonucleasomal DNA–histone complexes was also observed, and 18 of 26 patients had a reduction in lymphocyte count. One patient had a PR, and sustained improvements in haemoglobin and platelet levels were seen across all dose levels. The biological effect and lack of conventional toxicity of obatoclax were very encouraging for future study [50]. In the second Phase I study, 44 patients with a range of diseases, which were mostly refractory AML (n = 25) or myelodysplastic syndrome (n = 14), received 24 h infusions of 7–40 mg/m2 of obatoclax every second week. After dose escalation was complete, patients received 20–28 mg/m2 over 24 h every week, and then 20–28 mg/m2 over 24 h for 2, 3, or 4 days every second week. Three hundred and six infusions were given in total with a median of 5 per patient (range 1–35). As with the other study, CNS effects were the most common with somnolence, dizziness, fatigue, euphoric mood and gait disturbance being observed in 34–43% of patients; patients reported feeling drunk. As before, these symptoms occurred during the infusion and ceased once the infusion was complete. Little haematologic toxicity was observed, and in general the drug was well tolerated. One patient with AML had CR, and 3 of 14 patients with MDS had haematologic improvement [51].
7.6 ABT-737/263 Technological advances have allowed a high-throughput NMR-based method [52] to be used to screen a chemical library to identify molecules that bind to the BH3binding pocket of Bcl-xL [53]. Modification of the compound 4¢-Fluoro-biphenyl4-carboxylic acid identified by this screen resulted in the formation of ABT-737. ABT-737 binds with high affinity to Bcl-2, Bcl-xL and Bcl-w (Ki 1nM), but does
7 Targeting Bcl-2 Family Proteins in Childhood Leukemia
129
not bind to Mcl-1 or A1. In biological assays, ABT-737 functioned like a Bad BH3 protein; by itself, it was unable to cause the release of cytochrome c from isolated mitochondria, but it was able to prevent either Bcl-2 or Bcl-xL from inhibiting the release of cytochrome c by a Bid protein, and this function was dependent upon the presence of either Bax or Bak. Thus, ABT-737 does not activate Bak or Bax by itself, but rather prevents Bcl-2 or Bcl-xL from sequestering the activating BH3only proteins like Bid. ABT-737 disrupts a Bcl-2 family protein–protein interaction and displaces a GFP-tagged BH3-only protein from Bcl-xL in intact cells. ABT-737 is synergistic with both cytotoxic chemotherapy (etoposide, doxorubicin, cisplatin, paclitaxel) and radiotherapy in a range of cell types. As a single agent, ABT-7373 was potent against both small-cell lung cancer (SCLC) and lymphoid cell lines. ABT-737 was cytotoxic against t(14:18) containing lymphoma cell lines and primary patient-derived follicular lymphoma cells and significantly improved survival in a xenograft model of follicular lymphoma. Potent activity of ABT-737 with IC50 <1 mM was observed against 13/22 SCLC cell lines, and ABT-737 was able to cause regression of established SCLC xenografts. ABT-737 is not orally bioavailable and has low aqueous solubility, which makes the formulation of intravenous preparations difficult. In addition, ABT-737 is most efficacious in animal studies when given by daily intraperitoneal dose for 2 weeks. Thus, an analogue with oral bioavailability but similar properties in terms of Bcl-2 and Bcl-xL binding was sought. ABT-263 is an analogue of ABT-737 with modifications at three key sites affecting charge balance, metabolism and oral absorption. ABT-263 has similar binding affinity for Bcl-2, Bcl-xL and Bcl-w to ABT-737, with Ki <1 nM, and like ABT-737 binds less well to Mcl-1 and A1. ABT-263 reverses the protective effect of Bcl-2 or Bcl-xL over-expression in FL5.12 (murine IL3-dependent pro-B lymphoid) cells in which apoptosis has been induced by IL-3 withdrawal, and this was attenuated by ZVAD, the pan caspase inhibitor. In co-immunoprecipitation studies, ABT263 induces a dose-dependent decrease in Bim:Bcl-xL and Bim:Bcl-2 interactions and disrupts the interaction between Bcl-xL and Bcl-xS in a mammalian two-hybrid system. Like ABT-737, ABT-263 was potent at inducing cell death in Mcl-1 knockout mouse embryonic fibroblasts (MEFS) but ineffective against Bcl-xL knockout MEFS, and was also, like ABT-737, unable to kill Bax/Bak knockout MEFS. ABT-263 was able to induce mitochondrial cytochrome c release and a time-and-dose-dependent increase in caspase 3 cleavage in SCLC H146 cells. Oral dosing of ABT-263 was able to achieve similar exposures to those with intraperitoneal dosing of ABT-737. Like ABT-737 ABT-263 was most potent against SCLC and haematologic tumours; 32% of SCLC and 48% of haematologic tumour cell lines showed EC50 values <1 mM. When given orally at 100 mg/kg/day for 21 days ABT-263 induced rapid tumour regression in both SCLS and ALL xenograft models. ABT-263 (100 mg/kg/day for 17 days) was highly effective in combination with a single 10 mg/kg rituximab dose against a DoHH2 B cell lymphoma xenograft and in combination with modified R-CHOP against a mantle cell lymphoma model. In a multiple myeloma xenograft model in which ABT-263 was ineffective as a single agent, it was still able to significantly enhance the effects of bortezomib. Like ABT-737, ABT-263 also rapidly
130
G. Makin and C. Dive
induces apoptosis in circulating platelets and at the plasma levels that were needed for antitumour effect in mice platelet counts in dogs were roughly halved [54]. As these preliminary studies had suggested that ABT-737/263 had promising single-agent activity in vitro against cell lines or primary tumour-derived samples from leukemia, lymphoma and multiple myeloma, many more detailed singletumour type studies have been forthcoming.
7.7 Lymphoma In the Em-myc transgenic mouse model of Burkitt lymphoma, ABT-737 (75 mg/kg/day for 14 days) had no effect as a single agent on mouse survival or tumour burden and had no effect on cells derived from these tumours in vitro. However, ABT-737 treatment significantly prolonged survival of mice transplanted with lymphomas derived from bitransgenic mice in which Em-myc and Bcl-2 were expressed together (Em-myc/bcl-2), when compared with vehicle alone [55]. This resistance to ABT-737 correlated with far higher protein levels of Mcl-1 in the Em-mycderived tumours than in the Em-myc/Bcl-2-derived tumours. Em-myc/Bcl-2derived lymphomas were very resistant to a single dose of intraperitoneal cyclophosphamide in comparison with tumours without Bcl-2 over-expression. However, the combination of low-dose (50 mg/kg or 175 mg/m2) cyclophosphamide with ABT-737 (75 mg/kg/day for 14 days) was highly effective, with mice bearing two of three tumours showing effective cure, and none of the 14 of 18 mice surviving to day 150 having any evidence of lymphoma macroscopically, microscopically or by molecular testing. Over-expression of Bcl-2 is found in almost all follicular lymphomas as a result of the t(14:18) translocation, and in 20% of diffuse B-cell lymphomas because of gene amplification. The potency of ABT-737 against lymphoid cell lines is very variable, with CI50 values ranging from 0.03 mM against the mantle cell line HBL-2 through 0.42 mM against the t(14:18) bearing large B-cell lymphoma line RL to 5.65 mM against the cutaneous T-cell line H9 [56]. Combinations of ABT737 with the novel proteasome inhibitor bortezomib, when either both drugs were given simultaneously or if ABT-737 was given 24 h before bortezomib, were significantly more effective than either drug alone against the large B-cell line RL, and more effective than combinations between ABT-737 and doxorubicin, etoposide and rituximab. A similar effective combination between ABT-737 and bortezomib was observed in the mantle cell lymphoma cell line HBL-2. This synergistic interaction between proteasome inhibitors and ABT-737 reflected an increase in the induction of apoptosis as measured by both loss of mitochondrial membrane potential and increase in staining with the carbanocyanine nucleic-acid stain yo-pro-1, in both RL and HBL-2 cells. The combination of ABT-737 and proteasome inhibition resulted in reduction of Mcl-1 protein levels and increase in Noxa protein levels in both cell lines. This enhanced induction of apoptosis by the combination of ABT-737 and bortezomib was also seen in primary cells from patients with CLL and DLBCL, and did not show enhanced toxicity against peripheral blood
7 Targeting Bcl-2 Family Proteins in Childhood Leukemia
131
mononuclear cells from healthy donors. The combination of ABT-737 and bortezomib was also significantly better than either agent alone in preventing the growth of HBL-2 xenografts in SCID mice. The MTOR inhibitor rapamycin is capable of inducing G1 arrest in DLBCL cell lines and is also able to induce apoptosis, whilst ABT-263 induces apoptosis in DLBCL cell lines within 48 h of exposure at nM concentrations. Rapamycin and ABT-263 were synergistic in cell killing in the DLBCL line DoHH-2, and their effects were additive in another DLBCL line and in a mantle cell lymphoma line. ABT-263 alone was effective at inhibiting growth of DLBCL and mantle cell lymphoma cells as xenografts in SCID mice, but the combination with rapamycin was significantly more effective at inhibiting tumour growth, and inducing apoptosis in both lymphoma types, and in DLBCL xenografts led to tumour regression [57]. The introduction of the monoclonal anti-CD20 antibody rituximab has been a significant advance in the treatment of B-cell non-Hodgkin’s lymphoma (B-NHL), especially in adults. Rituximab is particularly effective in combination with cytotoxic chemotherapy and RCHOP (Rituximab, Cyclophosphamide, Vincristine, Prednisolone) is now the standard therapy for these patients [58]. However, resistance to rituximab eventually develops, even in the face of continued CD20 expression. Rituximab induces apoptosis in lymphoma cell lines via the mitochondrial pathway and Bcl-xL is able to prevent this both in vitro and in a NOD/SCID mouse xenograft model [59]. Intrinsic resistance of B-NHL cell lines to rituximab correlated with high expression of anti-apoptotic Bcl-2 family proteins, Bcl-2, Bcl-xL and Mcl-1, and ABT-737 was able to sensitise those cell lines expressing high Bcl-2 and Bcl-xL (Sc-1), but not those expressing high levels of Mcl-1, against which ABT-737 is relatively ineffective, to rituximab-induced apoptosis. The poor binding of ABT-737 to Mcl-1 and A1 means that expression of either or both of these anti-apoptotic Bcl-2 family proteins is a potential resistance mechanism for tumour cells. However, Mcl-1 is a very short-lived protein, and cyclindependent kinase (CDK) inhibitors, such as roscovitine and flavopiridol, acting through inhibition of CDK9, can down-regulate Mcl-1 and are potent inducers of apoptosis in haematopoietic cells [60]. Co-treatment of U937 histiocytic lymphoma cells with roscovitine (at a dose sufficient to cause transcriptional down-regulation of Mcl-1) and ABT-737 leads to a dramatic increase in ABT-737-induced apoptosis. Similar synergistic interactions between roscovitine and ABT-737 were also seen in HL-60 myeloid leukemia cells and the T lymphoblastoid Jurkat line, and in primary AML blasts isolated from three patients with AML [61].
7.8 Leukemia 7.8.1 Acute Lymphoblastic Leukemia In ALL, Bcl-xL over-expression and high levels of Bcl-2 in proportion to Bax have both been associated with poor prognosis. High expression of Bcl-2 has also been associated with poor response to induction therapy.
132
G. Makin and C. Dive
BH3 profiling of mitochondria derived from four paediatric ALL cell lines showed varying degrees of Bcl-2-dependance, and exposure to ABT-737 for 24 h showed a 2-log range of response to this agent, which correlated tightly with the response of isolated mitochondria to the Bad BH3 peptide. All six primary ALL samples tested also showed response to ABT-737 in the nanomolar range and both primary samples in which there was sufficient material to perform BH3 profiling showed a Bcl-2-dependent profile [62]. ABT-737 showed synergy with l-asparaginase in seven different ALL cell lines using the DIMSCAN cytotoxicity assay, even in ALL cell lines (CEM and NALM-6) that were not sensitive to either ABT-737 or l-asparaginase as single agents. Synergy was also seen between ABT-737 and vincristine in six of seven ALL cell lines, and with dexamethasone in four of seven cell lines [63]. The mechanism of synergy involved an increase in mitochondrial membrane depolarisation, increased cytochrome c release and increased activation of caspases. There was no synergy between ABT-737 and l-asparaginase in normal lymphocytes. The combination of vincristine, dexamethasone and l-asparaginase (VXL) in combination with ABT737 was very effective against paediatric ALL samples established as xenografts in NOD/SCID mice, and the combination was significantly more effective than either ABT-737 or VXL alone. Resistance to ABT-737 in ALL cell lines did not correlate with any particular expression pattern of Bcl-2 family proteins, and indeed, the most sensitive line (COG-LL-319) had the highest basal expression of Mcl-1. However, in the two most resistant ALL cell lines levels of Mcl-1 increased with increasing exposure to ABT737, whilst in the two most sensitive cell lines Mcl-1 levels decreased over the same time course. Fenretinide (4HPR) treatment induced a concentration-dependent decrease in Mcl-1 levels in the two most resistant cell lines, which was dependent upon phosphorylation of Mcl-1 by the JNK pathway and upon the production of ROS by fenretinide. Fenretinide was synergistic with ABT-737 in all seven ALL cell lines but showed no cytotoxicity against non-proliferating normal lymphocytes, and this synergy was due to enhanced mitochondrial apoptosis [64].
7.8.2 Acute Myeloid Leukemia The advent of the Bcr/Abl tyrosine kinase inhibitor imatinib has been an considerable therapeutic advance for patients with myeloid and lymphobastic leukemia containing the t(9,22) Philadelphia chromosome. Imatinib blocks the proliferation driven by Bcr/Abl in these leukemic cells and induces apoptosis by the mitochondrial pathway that can be blocked by the over-expression of Bcl-2 and Bcl-xL. Treatment of K562 Bcr/Abl-positive leukemia cell lines with imatinib results in the up-regulation of Bim at the RNA level and the accumulation of hypophosphorylated forms of Bim. Suppression of Bim expression in these cells by RNAi resulted in reduction in response to imatinib, and the degree of Bim suppression correlated with the degree of resistance to imatinib-induced apoptosis. Imatinib still leads to
7 Targeting Bcl-2 Family Proteins in Childhood Leukemia
133
increases in Bim levels in these suppressed clones, as well as to dephosphorylation of Bad. Double knockdown experiments with fetal liver-derived haematopoetic cells transfected with Bcr/Abl demonstrate that loss of both Bim and Bad results in almost complete resistance to imatinib. ABT-737 was able to increase imatinibinduced apoptosis in K562 Bcr/Abl-positive leukemia cells, but also able to dramatically sensitise both Bim suppressed and Bcl-2 over-expressing K562 cells, as well as restoring sensitivity to imatinib in Bim/Bad double knockout Bcr/Abl transformed murine fetal liver haematopoietic cells [65]. Activating mutations of the receptor tyrosine kinase FLT3 are common in AML and are associated with resistance to conventional chemotherapy [66, 67]. Smallmolecule inhibitors of FLT3 are highly effective at dephosporylating the FLT3 receptor but produce heterogenous cytotoxicity against primary AML blasts and in the clinic [68, 69]. Expression of high levels of Bcl-2 by primary AML blasts has been suggested to account for their lack of response to FLT3 inhibitors. Overexpression of either Bcl-2 or Bcl-xL in a cell line model of FLT3-positive AML confers resistance to small-molecule inhibitors of FLT3, and this can be overcome by simultaneous treatment with ABT-737 [70]. ABT-737 was potent at inducing apoptosis as a single agent in four AML cell lines and was significantly more effective against primary AML blasts from 13 patients in comparison with normal peripheral blood mononuclear cells. Synergy between ABT-737 and FLT3 inhibitors was also seen in these AML cell lines and in primary AML blasts carrying activating FLT3 mutations. MAP kinase signalling is frequently activated in primary AML blasts and seems to predict for poor survival [71]. The multikinase inhibitor sorafenib, which inhibits the MAP kinase pathway target Raf, is very effective against FLT3-positive AML [72]. However, 70% of AML does not carry FLT3 mutations, and in these AML cells sorafenib is able to inhibit the phosphorylation of Erk and MEK1/2 and induces apoptosis according to baseline level of Erk activation, such that AML cells with constitutive Erk activation are more sensitive to sorafenib [73], and similar results are seen with primary AML blasts. Sorafenib up-regulates Bim protein levels, and repression of Bim in AML cells lead to resistance to this agent. Sorafenib was strongly synergistic with cytarabine and pre-treatment with sorafenib followed by a 48-h exposure to ABT-737 was also strongly synergistic.
7.8.3 Chronic Myeloid Leukemia Despite the immediate therapeutic benefit of the Bcr/Abl inhibitor imatinib in treating CML, a variety of resistance mechanisms are implicated in leukemic survival and account for disease progression. As most of these pathways converge on the mitochondrial pathway of apoptosis, direct regulation of this is an attractive therapeutic option in this disease. ABT-737 is potent at inducing apoptosis in both CML cell lines and in primary leukemic cells from untreated CML patients and did not have toxicity against activated lymphocytes [74]. There was no correlation between
134
G. Makin and C. Dive
expression patterns of Bcl-2 family proteins and sensitivity to ABT-737 in CML cell lines. Expression of p-glycoprotein increased resistance to ABT-737. ABT-737 was also able to significantly improve the survival of NOD/SCID mice bearing BV173 CML xenografts. ABT-737 was able to restore sensitivity to imatinib in imatinib-resistant K562 cells with over-expression of Bcl-2 family proteins, expression of p-glycoprotein and over-expression of Bcr/Abl. Although ABT-737 itself was able to kill CML cell lines bearing mutated Bcr/Abl, it was unable to sensitise these cells to imatinib. However, these CML cells were very sensitive to the combination of ABT-737 and the novel candidate alkaloid homoharringtonine.
7.8.4 Chronic Lymphocytic Leukemia Primary CLL cells are very sensitive to apoptosis-induced by ABT-737; 24 primary CLL samples showed a mean EC50 to 48 h exposure to ABT-737 of 4.5 nM. Bcl-2 and Bim levels in primary CLL cells are remarkably consistent and higher than in PBMCs. Mcl-1 and Bid do not seem to play a role in the response of these cells to ABT-737. Mitochondria isolated from CLL cells showed increased outer membrane permeability when exposed to peptides derived from the BH3 domains of Bad, Puma and Bmf, but not to Noxa or BNIP3, showing that CLL mitochondria are dependent upon Bcl-2 for the maintenance of outer membrane integrity [9]. Bim is consistently sequestered by Bcl-2 in CLL cells, and it is the release of Bim from Bcl-2 following ABT-737 treatment that leads to Bak and Bax activation and loss of outer membrane integrity. Exposure of CLL cells to ABT-737 results in a rapid (within 1 h) change in the conformation of the multi-domain pro-apoptotic Bcl-2 family proteins Bax and Bak, translocation of Bax to the mitochondria and the formation of oligomers of Bax, followed by the characteristic ultrastructural changes of apoptosis (chromatin condensation and nucleolar segregation) with 2 h of exposure. However, in addition to the classical nuclear changes of apoptosis, CLL cells treated with ABT-737 showed mitochondrial matrix swelling and discontinuity of the outer mitochondrial membrane (OMM), which are changes normally associated with necrosis, but not apoptosis [75]. These OMM changes seem to be due to the rapid induction of mitochondrial inner membrane permeability (MIMP), in a manner which is independent of both caspase activity and of the opening of the permeability transition pore (PTP). Similar changes in response to ABT-737 were observed in a variety of human B- cell malignancies, but not in the T lymphoblastic Jurkat cell line.
7.9 Myeloma The disease is classically very resistant to conventional cytotoxic therapies and MM cells express high levels of Bcl-2 and Bcl-xL, which seem to contribute to their resistance to therapy, making ABT-737 an attractive therapeutic option for this disease.
7 Targeting Bcl-2 Family Proteins in Childhood Leukemia
135
MM cell lines are sensitive to ABT-737 at micromolar doses, and exposure to ABT-737 leads to a dose- and time-dependent increase in apoptosis. Primary MM cells from the bone marrow of 12 patients were also sensitive to ABT-737 in similar doses. Potent synergy was observed when ABT-737 was combined with dexamethasone, but not with doxorubicin or with the proteasome inhibitor Velcade® [76]. ABT-737 is capable of inducing apoptosis in primary MM cells even after multiple prior therapies, and in cells derived from patients resistant to bortezomib, thalidomide and dexamethasone. ABT-737-induced apoptosis in MM cells is not blocked by either IL-6 or IGF-1, both of which are protective against chemotherapyinduced apoptosis in vivo. In this study, the combinations of ABT-737 and dexamethasone and ABT-737 and bortezomib showed additivity activity, but ABT-737 was synergistic with melphalan [77]. In contrast, others have reported a variable protective effect of IL-6 on ABT737-induced apoptosis in MM cell lines, although there were MM cell lines in which no protective effect of IL-6 was observed. Co-culture of MM cell lines with patient-derived stromal cells had a similar range of effects. Patient-derived MM cells in this study fell into two clear groups with 4 of 15 patient samples being highly sensitive to ABT-737 at sub micromolar doses, whilst the remaining 11 patient samples did not seem to respond to ABT-737 even after prolonged exposure. These authors demonstrated a synergistic interaction between ABT-737 and dexamethasone in MM cells, as well as between ABT-737 and melphalan. ABT-737 was also able to induce dose-dependent regression of a sub-cutaneous xenograft mouse model of MM and induced cleavage of caspase 3 within these tumours [78].
7.10 Paediatrics ABT-263 has been evaluated against the Pediatric Preclinical Testing Program. In high-throughput growth inhibition studies against rhabdomyosarcoma, rhabdoid tumour, ESFT, neuroblastoma, glioma, T- and B-ALL, AML, ALCL and NHL cell lines, 9 of 23 tumour lines had an IC50 to ABT-263 of <1 mM. In 44 xenograft models in immunocompromised mice, which also included Wilms tumour, medulloblastoma, ependymoma and osteosarcoma, but not AML, ALCL or NHL, ABT-263 significantly prolonged EFS relative to controls in 26% of solid tumour cell lines and five of six ALL xenografts. No objective responses were seen in any solid tumour xenografts, whilst three ALL xenografts showed objective responses, included two maintained responses in T-ALL [79].
7.11 Conclusion The regulation of apoptosis by the Bcl-2 family of proteins is clearly of major importance to the development of leukemia and lymphoid malignancies, and in addition, the anti-apoptotic Bcl-2 family proteins have a significant role in drug resistance.
136
G. Makin and C. Dive
This has meant that therapeutic strategies targeting these proteins have been extensively evaluated, in vitro, in xenograft and in some cases in vivo, against the broad family of haematological malignancies. Promising results have been obtained with a variety of approaches as described above. Significant numbers of early-phase clinical trials with oblimersen have now been completed, and promising results in NHL have led to an ongoing NCI phase 3 study comparing R-CHOP with and without oblimersen in DLBCL. Early clinical results with obatoclax have been promising and further assessment is ongoing; there are currently 14 studies of obatoclax, both alone and in combination, open in USA, including a Children’s Oncology Group phase 1 study of obatoclax in combination with vincristine and doxorubicin in relapsed solid tumours and leukemias. Although its clinical development is at an earlier stage, the in vitro efficacy of ABT-737 has lead to considerable interest in the clinical use of the orally bioavailable form, ABT-263. There are currently 13 clinical studies of ABT-263 registered in USA (http://clinicaltrials.gov, accessed 30 October 2009), against both solid tumours and haematological malignancies, both as a single agent (against SCLC, CLL and NHL), and in combination with gemcitabine, docetaxel, etoposide/cisplatin, carboplatin/pacletaxel and fludarabine/ cyclophosphamide/rituximab. The development of targeted therapeutics against anti-apoptotic Bcl-2 family proteins offers much promise for the future. It is likely that the most appropriate use of these, and further novel small-molecule inhibitors of Bcl-2 that are in development, will be in combination with conventional cytotoxic agents, as a rational strategy either to overcome resistance or to allow similar antitumour effects to be achieved at lower doses of cytotoxic agents, to reduce both short- and long-term toxicity. Pre-clinical studies with ABT-737 have already led to the development of in vitro techniques that predict the response of tumour-derived cells to these inhibitors. It does not require too great a leap of faith to envisage a future in which tumour cellderived mitochondria can be profiled in vitro to allow rational selection of one or more Bcl-2 family inhibitors, to sensitise the patient’s tumour cells to existing cytotoxic-based regimes.
References 1. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell. 2000 Jan 7;100(1):57–70. 2. Pelengaris S, Khan M, Evan GI. Suppression of Myc-induced apoptosis in beta cells exposes multiple oncogenic properties of Myc and triggers carcinogenic progression. Cell. 2002 May 3;109(3):321–34. 3. Scaffidi C, Fulda S, Srinivasan A, Friesen C, Li F, Tomaselli KJ, et al. Two CD95 (APO-1/ Fas) signaling pathways. The EMBO journal. 1998 Mar 16;17(6):1675–87. 4. Danial NN, Korsmeyer SJ. Cell death: critical control points. Cell. 2004 Jan 23;116(2):205–19. 5. Wei MC, Zong WX, Cheng EH, Lindsten T, Panoutsakopoulou V, Ross AJ, et al. Proapoptotic BAX and BAK: a requisite gateway to mitochondrial dysfunction and death. Science (New York, NY). 2001 Apr 27;292(5517):727–30. 6. Certo M, Del Gaizo Moore V, Nishino M, Wei G, Korsmeyer S, Armstrong SA, et al. Mitochondria primed by death signals determine cellular addiction to antiapoptotic BCL-2 family members. Cancer cell. 2006 May;9(5):351–65.
7 Targeting Bcl-2 Family Proteins in Childhood Leukemia
137
7. Cartron PF, Gallenne T, Bougras G, Gautier F, Manero F, Vusio P, et al. The first alpha helix of Bax plays a necessary role in its ligand-induced activation by the BH3-only proteins Bid and PUMA. Molecular cell. 2004 Dec 3;16(5):807–18. 8. Deng J, Carlson N, Takeyama K, Dal Cin P, Shipp M, Letai A. BH3 profiling identifies three distinct classes of apoptotic blocks to predict response to ABT-737 and conventional chemotherapeutic agents. Cancer cell. 2007 Aug;12(2):171–85. 9. Del Gaizo Moore V, Brown JR, Certo M, Love TM, Novina CD, Letai A. Chronic lymphocytic leukemia requires BCL2 to sequester prodeath BIM, explaining sensitivity to BCL2 antagonist ABT-737. The Journal of clinical investigation. 2007 Jan;117(1):112–21. 10. Adams JM, Harris AW, Pinkert CA, Corcoran LM, Alexander WS, Cory S, et al. The c-myc oncogene driven by immunoglobulin enhancers induces lymphoid malignancy in transgenic mice. Nature. 1985 Dec 12–18;318(6046):533–8. 11. Strasser A, Harris AW, Bath ML, Cory S. Novel primitive lymphoid tumours induced in transgenic mice by cooperation between myc and bcl-2. Nature. 1990 Nov 22;348(6299): 331–3. 12. Letai A, Sorcinelli MD, Beard C, Korsmeyer SJ. Antiapoptotic BCL-2 is required for maintenance of a model leukemia. Cancer cell. 2004 Sep;6(3):241–9. 13. Schmitt CA, Rosenthal CT, Lowe SW. Genetic analysis of chemoresistance in primary murine lymphomas. Nature medicine. 2000 Sep;6(9):1029–35. 14. Reed JC, Stein C, Subasinghe C, Haldar S, Croce CM, Yum S, et al. Antisense-mediated inhibition of BCL2 protooncogene expression and leukemic cell growth and survival: comparisons of phosphodiester and phosphorothioate oligodeoxynucleotides. Cancer research. 1990 Oct 15;50(20):6565–70. 15. Campos L, Sabido O, Rouault JP, Guyotat D. Effects of BCL-2 antisense oligodeoxynucleotides on in vitro proliferation and survival of normal marrow progenitors and leukemic cells. Blood. 1994 Jul 15;84(2):595–600. 16. Marcucci G, Byrd JC, Dai G, Klisovic MI, Kourlas PJ, Young DC, et al. Phase 1 and pharmacodynamic studies of G3139, a Bcl-2 antisense oligonucleotide, in combination with chemotherapy in refractory or relapsed acute leukemia. Blood. 2003 Jan 15;101(2):425–32. 17. Marcucci G, Stock W, Dai G, Klisovic RB, Liu S, Klisovic MI, et al. Phase I study of oblimersen sodium, an antisense to Bcl-2, in untreated older patients with acute myeloid leukemia: pharmacokinetics, pharmacodynamics, and clinical activity. J Clin Oncol. 2005 May 20;23(15):3404–11. 18. Dai G, Chan KK, Liu S, Hoyt D, Whitman S, Klisovic M, et al. Cellular uptake and intracellular levels of the bcl-2 antisense g3139 in cultured cells and treated patients with acute myeloid leukemia. Clin Cancer Res. 2005 Apr 15;11(8):2998–3008. 19. Tauchi T, Sumi M, Nakajima A, Sashida G, Shimamoto T, Ohyashiki K. BCL-2 antisense oligonucleotide genasense is active against imatinib-resistant BCR-ABL-positive cells. Clin Cancer Res. 2003 Sep 15;9(11):4267–73. 20. Wetzler M, Donohue KA, Odenike OM, Feldman EJ, Hurd DD, Stone RM, et al. Feasibility of administering oblimersen (G3139; Genasense) with imatinib mesylate in patients with imatinib resistant chronic myeloid leukemia – Cancer and leukemia group B study 10107. Leukemia & lymphoma. 2008 Jul;49(7):1274–8. 21. Pepper C, Thomas A, Hoy T, Cotter F, Bentley P. Antisense-mediated suppression of Bcl-2 highlights its pivotal role in failed apoptosis in B-cell chronic lymphocytic leukemia. Br J Haematol. 1999;107:611–5. 22. Pepper C, Hooper K, Thomas A, Hoy T, Bentley P. Bcl-2 antisense oligonucleotides enhance the cytotoxicity of chlorambucil in B-cell chronic lymphocytic leukemia cells. Leukemia & lymphoma. 2001 Jul;42(3):491–8. 23. Vu UE, Pavletic ZS, Wang X, Joshi SS. Increased cytotoxicity against B-chronic lymphocytic leukemia by cellular manipulations: potentials for therapeutic use. Leukemia & lymphoma. 2000 Nov;39(5–6):573–82. 24. O’Brien SM, Cunningham CC, Golenkov AK, Turkina AG, Novick SC, Rai KR. Phase I to II multicenter study of oblimersen sodium, a Bcl-2 antisense oligonucleotide, in patients with advanced chronic lymphocytic leukemia. J Clin Oncol. 2005 Oct 20;23(30):7697–702.
138
G. Makin and C. Dive
25. O’Brien S, Moore JO, Boyd TE, Larratt LM, Skotnicki A, Koziner B, et al. Randomized phase III trial of fludarabine plus cyclophosphamide with or without oblimersen sodium (Bcl-2 antisense) in patients with relapsed or refractory chronic lymphocytic leukemia. J Clin Oncol. 2007 Mar 20;25(9):1114–20. 26. Badros AZ, Goloubeva O, Rapoport AP, Ratterree B, Gahres N, Meisenberg B, et al. Phase II study of G3139, a Bcl-2 antisense oligonucleotide, in combination with dexamethasone and thalidomide in relapsed multiple myeloma patients. J Clin Oncol. 2005 Jun 20;23(18):4089–99. 27. Guinness ME, Kenney JL, Reiss M, Lacy J. Bcl-2 antisense oligodeoxynucleotide therapy of Epstein-Barr virus-associated lymphoproliferative disease in severe combined immunodeficient mice. Cancer research. 2000 Oct 1;60(19):5354–8. 28. Loomis R, Carbone R, Reiss M, Lacy J. Bcl-2 antisense (G3139, Genasense) enhances the in vitro and in vivo response of Epstein-Barr virus-associated lymphoproliferative disease to rituximab. Clin Cancer Res. 2003 May;9(5):1931–9. 29. O’Connor OA, Smith EA, Toner LE, Teruya-Feldstein J, Frankel S, Rolfe M, et al. The combination of the proteasome inhibitor bortezomib and the bcl-2 antisense molecule oblimersen sensitizes human B-cell lymphomas to cyclophosphamide. Clin Cancer Res. 2006 May 1;12(9):2902–11. 30. Pro B, Leber B, Smith M, Fayad L, Romaguera J, Hagemeister F, et al. Phase II multicenter study of oblimersen sodium, a Bcl-2 antisense oligonucleotide, in combination with rituximab in patients with recurrent B-cell non-Hodgkin lymphoma. British journal of haematology. 2008 Nov;143(3):355–60. 31. Pieters R, Loonen AH, Huismans DR, et al. Relation between age, immunophenotype and in vitro drug resistance in 395 children with acute lymphoblastic leukemia-implications for the treatment of infants. Leukemia. 1998: 12:1344–1348. 32. Robinson BW, Behling KC, Gupta M, Zhang AY, Moore JS, Bantly AD, et al. Abundant antiapoptotic BCL-2 is a molecular target in leukemias with t(4;11) translocation. British journal of haematology. 2008 Jun;141(6):827–39. 33. Rheingold SR, Hogarty MD, Blaney SM, Zwiebel JA, Sauk-Schubert C, Chandula R, et al. Phase I Trial of G3139, a bcl-2 antisense oligonucleotide, combined with doxorubicin and cyclophosphamide in children with relapsed solid tumors: a Children’s Oncology Group Study. J Clin Oncol. 2007 Apr 20;25(12):1512–8. 34. Wang JL, Liu D, Zhang ZJ, Shan S, Han X, Srinivasula SM, et al. Structure-based discovery of an organic compound that binds Bcl-2 protein and induces apoptosis of tumor cells. Proceedings of the National Academy of Sciences of the United States of America. 2000 Jun 20;97(13):7124–9. 35. Lickliter JD, Wood NJ, Johnson L, McHugh G, Tan J, Wood F, et al. HA14-1 selectively induces apoptosis in Bcl-2-overexpressing leukemia/lymphoma cells, and enhances cytarabineinduced cell death. Leukemia. 2003 Nov;17(11):2074–80. 36. Campas C, Cosialls AM, Barragan M, Iglesias-Serret D, Santidrian AF, Coll-Mulet L, et al. Bcl-2 inhibitors induce apoptosis in chronic lymphocytic leukemia cells. Experimental hematology. 2006 Dec;34(12):1663–9. 37. Pei XY, Dai Y, Grant S. The proteasome inhibitor bortezomib promotes mitochondrial injury and apoptosis induced by the small molecule Bcl-2 inhibitor HA14-1 in multiple myeloma cells. Leukemia. 2003 Oct;17(10):2036–45. 38. Milella M, Kornblau SM, Estrov Z, Carter BZ, Lapillonne H, Harris D, et al. Therapeutic targeting of the MEK/MAPK signal transduction module in acute myeloid leukemia. The Journal of clinical investigation. 2001 Sep;108(6):851–9. 39. Milella M, Estrov Z, Kornblau SM, Carter BZ, Konopleva M, Tari A, et al. Synergistic induction of apoptosis by simultaneous disruption of the Bcl-2 and MEK/MAPK pathways in acute myelogenous leukemia. Blood. 2002 May 1;99(9):3461–4. 40. Corcelle EA, Puustinen P, Jaattela M. Apoptosis and autophagy: Targeting autophagy signalling in cancer cells – ‘trick or treats’? The FEBS journal. 2009 Nov;276(21):6084–96. 41. Pattingre S, Tassa A, Qu X, Garuti R, Liang XH, Mizushima N, et al. Bcl-2 antiapoptotic proteins inhibit Beclin 1-dependent autophagy. Cell. 2005 Sep 23;122(6):927–39.
7 Targeting Bcl-2 Family Proteins in Childhood Leukemia
139
42. Swerdlow S, McColl K, Rong Y, Lam M, Gupta A, Distelhorst CW. Apoptosis inhibition by Bcl-2 gives way to autophagy in glucocorticoid-treated lymphocytes. Autophagy. 2008 Jul 1; 4(5):612–20. 43. Kessel D, Reiners JJ, Jr. Initiation of apoptosis and autophagy by the Bcl-2 antagonist HA14-1. Cancer letters. 2007 May 8;249(2):294–9. 44. Kessel D, Reiners JJ, Jr., Hazeldine ST, Polin L, Horwitz JP. The role of autophagy in the death of L1210 leukemia cells initiated by the new antitumor agents, XK469 and SH80. Molecular cancer therapeutics. 2007 Jan;6(1):370–9. 45. Zhai D, Jin C, Satterthwait AC, Reed JC. Comparison of chemical inhibitors of antiapoptotic Bcl-2-family proteins. Cell death and differentiation. 2006 Aug;13(8):1419–21. 46. Nguyen M, Marcellus RC, Roulston A, Watson M, Serfass L, Murthy Madiraju SR, et al. Small molecule obatoclax (GX15-070) antagonizes MCL-1 and overcomes MCL-1-mediated resistance to apoptosis. Proceedings of the National Academy of Sciences of the United States of America. 2007 Dec 4;104(49):19512–7. 47. Perez-Galan P, Roue G, Lopez-Guerra M, Nguyen M, Villamor N, Montserrat E, et al. BCL-2 phosphorylation modulates sensitivity to the BH3 mimetic GX15-070 (Obatoclax) and reduces its synergistic interaction with bortezomib in chronic lymphocytic leukemia cells. Leukemia. 2008 Sep;22(9):1712–20. 48. Trudel S, Stewart AK, Li Z, Shu Y, Liang SB, Trieu Y, et al. The Bcl-2 family protein inhibitor, ABT-737, has substantial antimyeloma activity and shows synergistic effect with dexamethasone and melphalan. Clin Cancer Res. 2007 Jan 15;13(2 Pt 1):621–9. 49. Konopleva M, Watt J, Contractor R, Tsao T, Harris D, Estrov Z, et al. Mechanisms of antileukemic activity of the novel Bcl-2 homology domain-3 mimetic GX15-070 (obatoclax). Cancer research. 2008 May 1;68(9):3413–20. 50. O’Brien SM, Claxton DF, Crump M, Faderl S, Kipps T, Keating MJ, et al. Phase I study of obatoclax mesylate (GX15-070), a small molecule pan-Bcl-2 family antagonist, in patients with advanced chronic lymphocytic leukemia. Blood. 2009 Jan 8;113(2):299–305. 51. Schimmer AD, O’Brien S, Kantarjian H, Brandwein J, Cheson BD, Minden MD, et al. A phase I study of the pan bcl-2 family inhibitor obatoclax mesylate in patients with advanced hematologic malignancies. Clin Cancer Res. 2008 Dec 15;14(24):8295–301. 52. Shuker SB, Hajduk PJ, Meadows RP, Fesik SW. Discovering high-affinity ligands for proteins: SAR by NMR. Science (New York, NY). 1996 Nov 29;274(5292):1531–4. 53. Oltersdorf T, Elmore SW, Shoemaker AR, Armstrong RC, Augeri DJ, Belli BA, et al. An inhibitor of Bcl-2 family proteins induces regression of solid tumours. Nature. 2005 Jun 2;435(7042):677–81. 54. Tse C, Shoemaker AR, Adickes J, Anderson MG, Chen J, Jin S, et al. ABT-263: a potent and orally bioavailable Bcl-2 family inhibitor. Cancer research. 2008 May 1;68(9):3421–8. 55. Mason KD, Vandenberg CJ, Scott CL, Wei AH, Cory S, Huang DC, et al. In vivo efficacy of the Bcl-2 antagonist ABT-737 against aggressive Myc-driven lymphomas. Proceedings of the National Academy of Sciences of the United States of America. 2008 Nov 18;105(46): 17961–6. 56. Paoluzzi L, Gonen M, Bhagat G, Furman RR, Gardner JR, Scotto L, et al. The BH3-only mimetic ABT-737 synergizes the antineoplastic activity of proteasome inhibitors in lymphoid malignancies. Blood. 2008 Oct 1;112(7):2906–16. 57. Ackler S, Xiao Y, Mitten MJ, Foster K, Oleksijew A, Refici M, et al. ABT-263 and rapamycin act cooperatively to kill lymphoma cells in vitro and in vivo. Molecular cancer therapeutics. 2008 Oct;7(10):3265–74. 58. Coiffier B, Pfreundschuh M, Stahel R, Vose J, Zinzani PL. Aggressive lymphoma: improving treatment outcome with rituximab. Anti-cancer drugs. 2002 Nov;13 Suppl 2:S43–50. 59. Stolz C, Hess G, Hahnel PS, Grabellus F, Hoffarth S, Schmid KW, et al. Targeting Bcl-2 family proteins modulates the sensitivity of B-cell lymphoma to rituximab-induced apoptosis. Blood. 2008 Oct 15;112(8):3312–21. 60. Chen R, Keating MJ, Gandhi V, Plunkett W. Transcription inhibition by flavopiridol: mechanism of chronic lymphocytic leukemia cell death. Blood. 2005 Oct 1;106(7):2513–9.
140
G. Makin and C. Dive
61. Chen S, Dai Y, Harada H, Dent P, Grant S. Mcl-1 down-regulation potentiates ABT-737 lethality by cooperatively inducing Bak activation and Bax translocation. Cancer research. 2007 Jan 15;67(2):782–91. 62. Del Gaizo Moore V, Schlis KD, Sallan SE, Armstrong SA, Letai A. BCL-2 dependence and ABT-737 sensitivity in acute lymphoblastic leukemia. Blood. 2008 Feb 15;111(4):2300–9. 63. Kang MH, Kang YH, Szymanska B, Wilczynska-Kalak U, Sheard MA, Harned TM, et al. Activity of vincristine, L-ASP, and dexamethasone against acute lymphoblastic leukemia is enhanced by the BH3-mimetic ABT-737 in vitro and in vivo. Blood. 2007 Sep 15;110(6):2057–66. 64. Kang MH, Wan Z, Kang YH, Sposto R, Reynolds CP. Mechanism of synergy of N-(4-hydroxyphenyl)retinamide and ABT-737 in acute lymphoblastic leukemia cell lines: Mcl-1 inactivation. Journal of the National Cancer Institute. 2008 Apr 16;100(8):580–95. 65. Kuroda J, Puthalakath H, Cragg MS, Kelly PN, Bouillet P, Huang DC, et al. Bim and Bad mediate imatinib-induced killing of Bcr/Abl+ leukemic cells, and resistance due to their loss is overcome by a BH3 mimetic. Proceedings of the National Academy of Sciences of the United States of America. 2006 Oct 3;103(40):14907–12. 66. Schnittger S, Schoch C, Dugas M, Kern W, Staib P, Wuchter C, et al. Analysis of FLT3 length mutations in 1003 patients with acute myeloid leukemia: correlation to cytogenetics, FAB subtype, and prognosis in the AMLCG study and usefulness as a marker for the detection of minimal residual disease. Blood. 2002 Jul 1;100(1):59–66. 67. Thiede C, Steudel C, Mohr B, Schaich M, Schakel U, Platzbecker U, et al. Analysis of FLT3-activating mutations in 979 patients with acute myelogenous leukemia: association with FAB subtypes and identification of subgroups with poor prognosis. Blood. 2002 Jun 15;99(12):4326–35. 68. Knapper S, Mills KI, Gilkes AF, Austin SJ, Walsh V, Burnett AK. The effects of lestaurtinib (CEP701) and PKC412 on primary AML blasts: the induction of cytotoxicity varies with dependence on FLT3 signaling in both FLT3-mutated and wild-type cases. Blood. 2006 Nov 15;108(10):3494–503. 69. Knapper S, Burnett AK, Littlewood T, Kell WJ, Agrawal S, Chopra R, et al. A phase 2 trial of the FLT3 inhibitor lestaurtinib (CEP701) as first-line treatment for older patients with acute myeloid leukemia not considered fit for intensive chemotherapy. Blood. 2006 Nov 15;108(10):3262–70. 70. Kohl TM, Hellinger C, Ahmed F, Buske C, Hiddemann W, Bohlander SK, et al. BH3 mimetic ABT-737 neutralizes resistance to FLT3 inhibitor treatment mediated by FLT3-independent expression of BCL2 in primary AML blasts. Leukemia. 2007 Aug;21(8):1763–72. 71. Kornblau SM, Womble M, Qiu YH, Jackson CE, Chen W, Konopleva M, et al. Simultaneous activation of multiple signal transduction pathways confers poor prognosis in acute myelogenous leukemia. Blood. 2006 Oct 1;108(7):2358–65. 72. Auclair D, Miller D, Yatsula V, Pickett W, Carter C, Chang Y, et al. Antitumor activity of sorafenib in FLT3-driven leukemic cells. Leukemia. 2007 Mar;21(3):439–45. 73. Zhang W, Konopleva M, Ruvolo VR, McQueen T, Evans RL, Bornmann WG, et al. Sorafenib induces apoptosis of AML cells via Bim-mediated activation of the intrinsic apoptotic pathway. Leukemia. 2008 Apr;22(4):808–18. 74. Kuroda J, Kimura S, Andreeff M, Ashihara E, Kamitsuji Y, Yokota A, et al. ABT-737 is a useful component of combinatory chemotherapies for chronic myeloid leukemias with diverse drug-resistance mechanisms. British journal of haematology. 2008 Jan;140(2):181–90. 75. Vogler M, Dinsdale D, Sun XM, Young KW, Butterworth M, Nicotera P, et al. A novel paradigm for rapid ABT-737-induced apoptosis involving outer mitochondrial membrane rupture in primary leukemia and lymphoma cells. Cell death and differentiation. 2008 May;15(5): 820–30. 76. Kline MP, Rajkumar SV, Timm MM, Kimlinger TK, Haug JL, Lust JA, et al. ABT-737, an inhibitor of Bcl-2 family proteins, is a potent inducer of apoptosis in multiple myeloma cells. Leukemia. 2007 Jul;21(7):1549–60.
7 Targeting Bcl-2 Family Proteins in Childhood Leukemia
141
77. Chauhan D, Velankar M, Brahmandam M, Hideshima T, Podar K, Richardson P, et al. A novel Bcl-2/Bcl-X(L)/Bcl-w inhibitor ABT-737 as therapy in multiple myeloma. Oncogene. 2007 Apr 5;26(16):2374–80. 78. Trudel S, Li ZH, Rauw J, Tiedemann RE, Wen XY, Stewart AK. Preclinical studies of the pan-Bcl inhibitor obatoclax (GX015-070) in multiple myeloma. Blood. 2007 Jun 15;109(12):5430–8. 79. Lock R, Carol H, Houghton PJ, Morton CL, Kolb EA, Gorlick R, et al. Initial testing (stage 1) of the BH3 mimetic ABT-263 by the pediatric preclinical testing program. Pediatric blood & cancer. 2008 Jun;50(6):1181–9.
Chapter 8
Targeting Leukemia Stem Cells and Stem Cell Pathways in ALL Clare Pridans and Brian J.P. Huntly
8.1 Introduction The development of cancer is characterised by the accumulation of multiple somatic mutations [1]. These mutations individually alter the properties of the cell and, in concert, lead to the development of the full malignant phenotype. Recent technological advances have increasingly allowed us to identify causative mutations, and we are now able to model their effects in cell-based and animal models, such that their molecular mechanisms are being unravelled. However, the consequences of these mutations for the normal cell biology of the organ system involved are poorly understood. Recently, the cancer stem cell hypothesis has received a great deal of attention [2]. This hypothesis is not new, having been proposed by eminent pathologists such as Virchow over 150 years ago [3], and basically proposes that the continued growth and propagation of a tumour is wholly dependent upon an often small subpopulation, termed the cancer stem cells or cancer-initiating cells. Although contentious for some malignancies [4, 5], the existence of the cancer stem cell is well accepted for haematological malignancies. It was here that the existence of the first cancer stem cell was demonstrated, in acute myeloid leukemia (AML) by John Dick and colleagues in Toronto over a decade ago [6, 7]. Subsequently, cancer stem cells have also been demonstrated in a number of solid organ tumours such as breast, CNS , prostate, colon and pancreas tumours (detailed further in [8]), suggesting that many malignancies are dependent upon such a compartment. There are obvious clinical implications for the existence of a hierarchical organisation of malignancies with a cancer stem cell at its apex. Many malignancies demonstrate significant tumour reduction following initial therapy. However, the majority of malignancies relapse, often with resistant disease, causing the death of most
B.J.P. Huntly (*) Department of Haematology, University of Cambridge, Cambridge Institute for Medical Research, Hills Road, Cambridge CB2 0XY, UK e-mail:
[email protected] V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3_8, © Springer Science+Business Media, LLC 2011
143
144
C. Pridans and B.J.P. Huntly
patients. This suggests that current combination therapies spare the cancer stem cell compartment and that it is this compartment that forms the reservoir for subsequent relapse and resistance. This defines the cancer stem cell as the critical target in the therapy of malignant disease, and a greater knowledge of the biology of cancer stem cells should facilitate the development of selective or specific therapies and improve treatment outcomes in the treatment of malignant disease. The aim of this chapter is to review the evidence for the existence of the leukemia stem cell (LSC) in acute lymphoblastic leukemia (ALL), to describe what is known about its biology and to identify how it might serve as a target for therapeutic intervention. We describe the available data to identify the surface phenotype(s) of the ALL LSC and then discuss potential molecular and cellular mechanisms that may be specifically targeted, either presenting available data or speculating on potential therapeutics extrapolated from the treatment of other malignancies. Unfortunately, few investigators have yet specifically looked at targeting the LSC compartment in ALL and often results obtained in targeting bulk populations are presented. In addition, in our choice of nomenclature although the term cancer- or leukemia-initiating cell is in fact more accurate and appropriate, we have chosen to use the term LSC due to its widespread usage. However, this does not suggest that we always equate a LSC as the transformed equivalent of a normal haematopoietic stem cell (HSC) and, as is demonstrated below, we provide ample evidence that LSC may be generated from transformation of progenitor populations.
8.2 Characteristics That Might Be Targeted in Leukemia Stem Cells In common with all other cells that sit at the apex of a tissue hierarchy, such as the HSC, the LSC has relatively few cell-fate choices (see Fig. 8.1). The cell may be quiescent or dividing. If it does divide it may differentiate, self-renew or undergo both processes through asymmetric division. In addition, at any point, it may undergo apoptosis. The balance of these decisions is determined by an integration of cell intrinsic factors and extrinsic cues, resulting from the interaction of the LSC with its niche. Oncogenic mutations affect all of these critical cell processes [1] identifying them as potential targets. However, to avoid unacceptable collateral damage, the therapy needs to be targeted, specifically or at least selectively, towards the LSC while relatively sparing the normal tissue counterpart (the HSC). As we describe, the LSC may also be physically targeted through its surface phenotype, particularly where this differs from the phenotype of the normal HSC. We discuss the available data to identify potential candidate pathways and therapeutics below. However, it is only now that data on the biology of LSC is becoming available, and it is hoped that a greater understanding of these critical cell populations and how they differ from normal HSC will lead to improvements in therapy in the short term.
8 Targeting Leukemia Stem Cells and Stem Cell Pathways in ALL
145
Fig. 8.1 Targeting Leukemia Stem Cells (LSC). Similarly to the normal haematopoietic stem cell, the LSC has a limited number of potential cell fate decisions. It may either quiesce, self-renew or differentiate or it may apoptose. These cell fate decisions are controlled by cell-intrinsic pathways and extrinsic cues delivered from the leukemia stem cell niche, are aberrantly regulated in ALL and may be specifically targeted. How these pathways and the aberrant surface phenotype of the ALL LSC might be targeted is discussed further in the text
8.3 Identification and Cellular Characteristics of the LSC in ALL As previously discussed, to fulfil the definition, a true leukemia stem cell or leukemiainitiating cell must have long-term self-renewal and be able to regenerate the exact diversity and phenotype of the original tumour. No in vitro system exists to demonstrate these characteristics and, at present, the only true measure of the self-renewal capabilities of a LSC is to engraft irradiated immunodeficient murine recipients and regenerate leukemia in vivo. Various strains of mice have been used as recipients in this assay, usually involving a background of NOD/SCID (Non-obese diabetic/severe combined immunodeficiency) mutation, sometimes in combination with other mutations such as those affecting the b2 microglobulin (b2M), RAG or the IL-2 receptor common g chain (IL2rgnull) [9]. All of these mouse strains are deficient in endogenous B- and T-cell activity but have varying degrees of NK dysfunction. However, a recent report has demonstrated that in AML, NOD/SCID mice may still be able to immunologically clear cells bound by commercially available CD38+ antibodies in an Fc-mediated fashion, but that this does not occur in the more immunocompromised NOD/SCID/IL2rgnull strain [10]. Therefore, the choice of immunocompromised recipient mouse strain, the differing choice of markers used to define LSC populations and perhaps even the commercial antibodies to these markers make direct comparisons between studies difficult and may explain at least some of the discrepancies in the following data presented and summarised in Table 8.1 and Fig. 8.2.
NOD/SCID (NK depleted)
le Viseur et al. [17]
Cox et al. [18]
NOD/SCID
46XX, 46XY (3), complex, del 6q, t(11;14) Complex (2), t(11;14), +8del 9p, del9p
Hyperdiploid (2), dic(9;20) – high risk
t(4;11) (5), t(11;19), relapse, relapse with dup(21q) – high-risk
7 B-ALL
2 B-ALL
ETV6-RUNX1, (2) – std-risk
2 B-ALL
7 T-ALL 5 T-ALL a Authors did not determine CD34 or CD38 status in these experiments
NOD/SCID/IL2rgnull
MLL rearrangements (2) – high-risk
NOD/SCID/IL2r null
Kong et al. [16]
3 B-ALL
Karyotype and risk (patient numbers) Ph1-ALL (7) – high-risk 46XX (2), 46XY – std-risk Complex, t(12;21), del(17p) – mixed 46XX (2), t(12;21) (2) – std-risk ETV6-RUNX1 (3) – std-risk P210 BCR-ABL1 (5) – high-risk P190 BCR-ABL1 (4) – high-risk ETV6-RUNX1 (4) – std-risk
Table 8.1 Leukemia stem cells (LSC) phenotypes in xenotransplantation studies References Xenotransplantation model Patient number Cobaleda et al. [12] NOD/SCID 7 B-ALL Cox et al. [13] NOD/SCID 3 B-ALL 3 B-ALL 4 B-ALL Castor et al. [14]a NOD/SCID 3 B-ALL 5 B-ALL 4 B-ALL Hong et al. [15] NOD/SCID 4 B-ALL
LSC/LIC phenotype CD34+CD38− CD34+CD10− CD34+CD10− CD34+CD19− CD19+ CD19− CD19+ CD34+ CD38−/lowCD19+ CD34+CD38+CD19+ CD34+CD38−CD19+ CD34+CD19+ and CD34−CD19+ CD34+CD19+, CD34−CD19+, and CD34+CD19− CD19+CD20−/low and CD19+CD20high CD34+CD4− CD34+CD7−
146 C. Pridans and B.J.P. Huntly
8 Targeting Leukemia Stem Cells and Stem Cell Pathways in ALL
147
Fig. 8.2 Critical LSC pathways and potential therapeutic interventions in ALL. Pathways which are aberrantly regulated and which may be critical for ALL LSC function are shown in this figure. Activating mutations or overexpressed genes are shown in red, inactivating mutations/deletions/ downregulated genes are shown in blue, with potential therapeutics shown in green. Please refer to the text for a more detailed description of the molecular pathways
Whilst the LSC was described for AML over a decade ago [6, 7], the identity of the LSC and the origins of malignant transformation in ALL remain less clear. Almost 10 years ago, it was postulated by Greaves that a potential explanation for the survival disparity between adult and childhood ALL was a differing target cell for transformation, with adult ALL arising in HSC and childhood ALL arising in committed B and T lymphocytes (reviewed in [11]). Although prescient, this is obviously an oversimplification and for the purpose of this review, the evidence for and the identity of the LSC in B-ALL and T-ALL are addressed individually, regardless of patient age, although this is mentioned where known. However, it is likely that differences in LSC ontogeny may be reflected in the biological aggressiveness or “risk” category of ALL as determined by such characteristics as the presence of specific chromosomal translocations (rearranging fusion genes such as
148
C. Pridans and B.J.P. Huntly
Fig. 8.3 The LSC hierarchy is heterogeneous in B-ALL. Panel A represents a linear schema of B-lymphoid ontogeny, with procession of differentiation from the HSC compartment to a mature B-cell through a number of lineage restricted B-cell progenitors. Panels B and C represent a cumulative summary of the LSC hierarchy in standard risk and high-risk ALL respectively. In standard-risk B-ALL, the phenotype of the ALL LSC is usually described as CD34+/CD19+ but may extend into the CD34-/CD19+ compartment. There appears to be more complexity in the surface phenotype of the ALL LSC in high-risk B-ALL where the initiating cell has been described to be either CD34 positive or negative or CD19 positive or negative (and combinations thereof), suggesting a larger and more complex LSC compartment in high-risk disease. In addition, there may also be a degree of plasticity within this compartment. Recent reports demonstrate that CD34 negative LSC can regenerate leukemias with CD34 positive populations and CD19 positive LSC can regenerate leukemias with CD19 negative populations, in contrast to the relationship of these antigens to differentiation in normal B-cell ontology (please see text and Table 1 for further details)
ETV6-RUNX1, BCR-ABL or MLL-AF4) or DNA ploidy and where available this information is be presented. Several authors have attempted to identify the LSC in B-ALL, sometimes with conflicting results. The first description of an LSC in ALL was from Cobaleda et al. in 2000, and recapitulated the situation described for AML. The authors demonstrated that CD34+/CD38− cells from Ph+ ALL, a high-risk ALL category, were able to transfer disease to NOD/SCID mice, while the CD34− and CD38+ fractions were not able to do so [12]. A later study by Cox et al. published in 2004 corroborated these findings in children with a variety of high- and standard-risk B-ALL. They demonstrated that the ALL cells capable of long-term proliferation in NOD/ SCID mice were CD34+ but further demonstrated that these cells lacked CD10 and CD19 expression [13]. However, this result contrasts with those of Castor et al. [14]
8 Targeting Leukemia Stem Cells and Stem Cell Pathways in ALL
149
who demonstrated that, in NOD/SCID and NOD/SCID/b2M mice, the LSC was CD19+ in all cases (the presence or absence of CD34 was not ascertained in these transplant experiments). Their cohort of ten patients was of mixed age, involving both adults and children, and all patients had gene rearrangements, either TELAML1 (alternately known as ETV6-RUNX1 and standard-risk disease) or a BCRABL fusion (generating either the P190 or P210 protein). Interestingly, whilst the LSC was uniformly CD19+ and the CD19− compartment did not transfer the disease or contain the recurrent chromosomal translocation products for ETV6-RUNX1 or BCR-ABL P190 patients, the CD34+/CD19− cells in BCR-ABL P210 patients did carry the BCR-ABL gene fusion as demonstrated by FISH. In four children carrying the t(12;21) translocation that rearranges ETV6-RUNX1, the positivity of the LSC for CD19 was corroborated by Hong et al. [15] following transplantation into NOD/ SCID mice. In addition, the same authors described a “preleukemic” stem cell population with the same phenotype (CD34+/CD38−/lo/CD19+) in the peripheral blood of an unaffected monochorionic twin, suggesting that the ETV6-RUNX1 fusion is an early event bestowing the preleukemic cell with altered self-renewal and survival properties. A recent report published by Kong et al. has further attempted to clarify the identity of the LSC in three high-risk B-ALL cases, two of which had Mixed Lineage Leukemia (MLL) fusions, this time following transplantation into NOD/ SCID/IL2rgnull mice. They purified three cellular populations; CD34+CD38+CD19+, CD34+CD38−CD19+ and CD34+CD38−CD19− and demonstrated that the fractions that were capable of initiating B-ALL were CD34+/CD19+ but could be either CD38 positive or negative [16]. They went on to demonstrate that the CD34+CD38−CD19− population exhibited normal haemopoietic reconstitution as the mice did not develop leukemia when transplanted with these cells, but they did not test the ability of the CD34− population to transfer the disease [16]. In an attempt to address some of these inconsistencies, le Viseur et al. developed a more sensitive and robust functional assay to illustrate self-renewal properties in postulated LSC cell populations. Their experimental design was to transplant whole BM and fractionated populations from children with ALL into NOD/SCID mice directly by intrafemoral injection to improve the efficiency of engraftment [17]. The recipient mice also had additional immunodeficiencies, in that either NK cells were depleted using an anti-CD122 monoclonal antibody or the NOD/SCID/ IL2rgnull strain was used. Engrafting leukemias were then transplanted into secondary, tertiary and quaternary recipients to assess long-term self-renewal. They found that the majority of leukemias from high-risk patients (86%) engrafted, whilst only 33% of leukemias engrafted from standard-risk patients, again detailing heterogeneity in LSC activity between biologically different leukemias. Phenotypic analysis of the engrafted cells was consistent with the original diagnosis of the patients and remained stable from primary to quaternary transplants. In order to determine which cell populations contained the LSC, CD34+CD19−, CD34+CD19+ and CD34−CD19+ cells were purified from patients or engrafted mice and further transplanted. With cells isolated from high-risk patients, all three populations were able to transfer leukemia into mice. However, using cells from standard-risk patients, it
150
C. Pridans and B.J.P. Huntly
was only the CD19+ populations that had this ability, which was, surprisingly, independent of CD34 expression [17]. Interestingly, in high-risk patients, all three compartments were able to re-establish the complete immunophenotype of the original leukemia (i.e. to regenerate not only themselves but also other compartments). This not only challenges the perceived differentiation hierarchy, with CD34 positivity, a marker of immature cells, and CD19 positivity, a marker of maturing cells, but also suggests a degree of plasticity within the leukemic stem and progenitor hierarchy. This is presented in Fig. 8.2 along with a proposed summary of LSC ontogeny for standard- and high-risk ALL. In contrast to B-ALL, the identity of the LSC in T-ALL remains poorly studied. In the only transplantation study, Cox et al. [18] investigated the origins of T-ALL in children. They demonstrated that following transplantation into NOD/SCID mice, the LSC demonstrated a primitive phenotype (CD34+CD4− and CD34+CD7−); however, as this work reflects only ten patient samples, it should be considered preliminary. Together, these findings demonstrate that there appears to be considerable heterogeneity in the surface phenotype of ALL LSC. Whilst this may in part represent the technical issues mentioned above, it is likely to also reflect the biology of the LSC compartment. Although, more information is required to develop universal cell surface markers with which to target ALL LSC, certain markers such as CD19 are attractive candidates. As can be seen from Table 8.1 and the LSC schematic in Fig. 8.2, the majority of cells with leukemia-initiating potential are CD19 positive. This marker is also absent from normal HSC and may prove the basis of future therapeutics directed against ALL LSC. Recent therapeutic successes in haematological malignancies with monoclonal antibodies and immunoconjugate therapy such as therapy with Rituximab (Mabthera®) in CD20+ lymphoproliferative disorders make targeting these differences an attractive therapeutic strategy. In addition, Gemtuzumab ozogamicin (GO, Myelotarg®), an antiCD33 antibody conjugated to the toxin calicheamicin, is already in widespread clinical usage in AML therapy. Interestingly, in the report of le Viseur [17], CD20+ cells from a small number of standard and high-risk patients were demonstrated to have LSC properties upon transplantation, and trials of Rituximab, in combination with conventional chemotherapy, are already underway in ALL [19].
8.4 Critical Stem Cell Pathways in the Induction and Maintenance of ALL The molecular biology and pathogenesis of ALL are covered in subsequent chapters in greater detail. However, in this section we specifically discuss those aberrant pathways that are implicated in stem cell function and that may provide specific molecular targets for ALL LSC. Specific therapeutics, where available or in development, are discussed as is any preliminary data with these agents, although in most cases this has been tested only against bulk disease rather than directed specifically at ALL LSC.
8 Targeting Leukemia Stem Cells and Stem Cell Pathways in ALL
151
8.4.1 Self-Renewal in ALL LSC Self-renewal is a prerequisite of cancer development and maintenance [1] and is a defining characteristic of both malignant and normal stem cells. There are many similarities in the pathways implicated in the self-renewal of normal HSC and LSC, and these pathways include the clustered HOX genes, the WNT/b-Catenin pathway, the PTEN/AKT/FOXO axis, the RB/p53 tumour suppressor network, BMI1 and polycomb group proteins and the NOTCH and Hedgehog pathways (reviewed in [2] and [3]). However, the finding that these pathways are aberrantly activated or that genes in the pathways are frequently mutated in leukemia offers hope that they may actually function differently in leukemic and normal stem cells and may thus be selectively targeted. Many of these pathways are dysregulated in ALL (reviewed in [20]) and may be associated with aberrant self-renewal in ALL LSC. We specifically discuss evidence for aberrant activity of the clustered and non-clustered HOX genes, abnormalities of beta helix-loop-helix (bHLH) transcriptional complexes, abnormalities of the PTEN pathway, NOTCH mutations and abnormalities of the RB/TP53 tumour suppressor network in ALL and how these might be targeted.
8.4.1.1 Clustered and Non-Clustered HOX Genes Although not as commonly differentially expressed as in AML, clustered HOX genes are frequently dysregulated in ALL [21–23] and are critically required for developmental haematopoiesis and HSC self-renewal [24]. It is likely that HOX genes are also critical for LSC self-renewal as murine bone marrow lacking either Hoxa9 or -a7 could not be transformed by MLL fusion proteins [25]. Clustered HOX genes are transcription factors and whilst their critical downstream mediators remain poorly understood, there is gathering evidence linking HOX gene dysregulation with a number of recurrent chromosomal translocations in ALL. The most common of these are the rearrangements of the MLL gene at chromosome 11q23. These are relatively common in both childhood (present in around 10% of all childhood cases and up to 80% of infant ALL) and in adult ALL (around 10%) [20, 21]. Common partner genes include AF4, ENL and AF9, although around 70 different partners have been reported to date. The prognostic significance of MLLfusion ALL appears to be mixed with the MLL-AF4 rearrangement, generated by the t(4;11) chromosomal translocation, associated with a dismal outcome, while in one series all 11 patients with T-ALL and MLL-ENL rearrangements were long-term survivors [21]. MLL fusions contain the N terminal portion of MLL fused to the C terminus of their partner gene and have a dominant gain-of-function effect enhancing their transcriptional activity [26]. This in turn would be predicted to upregulate critical effector genes such as the HOX A cluster, facilitating selfrenewal in the LSC. Recently, using a chemical screen approach, inhibitors of Glycogen synthase kinase 3 isoforms a and b (GSK3a and b) have shown efficacy
152
C. Pridans and B.J.P. Huntly
against leukemias associated with MLL rearrangements in in vitro and in vivo mouse models through a mechanism that involves destabilisation of the cyclindependent kinase inhibitor p27Kip1 [27]. Importantly, inhibition occurs at a level previously reported to promote expansion of normal HSC in vitro, suggesting that this therapy would not be toxic to the HSC itself [28]. Another drug originally developed as a non-ATP competitor of GSK3b, TDZD-8, has also recently shown efficacy against leukemia stem and progenitor cells from a number of haematological malignancies, including ALL, whilst relatively sparing normal stem and progenitor cells. The mechanism appears to be related to oxidative stress [29] and inhibition of the PKC and FLT3 signalling pathways. Rearrangement of the AF10 gene also occurs with the AP3 clathrin protein coding gene CALM through the t(10;11) chromosomal translocation. This rearrangement was first identified in the U937 cell line and may be associated with a number of different malignant haematological phenotypes. However, it is most commonly associated with T-ALL, and CALM-AF10 fusion transcripts were identified in 9% of a recent series of 131 T-ALL cases [30]. CALM-AF10associated leukemias also over-express HOX genes, and in murine models the LSC has characteristics of both myeloid and lymphoid cells, perhaps explaining its promiscuous phenotype [31, 32]. T-ALL may also be associated with chromosomal translocations that bring the non-clustered HOX11(TLX1) and HOX11L2(TLX3) genes or the HOX A cluster under the transcriptional regulation of the TCR loci. Unlike CALM-AF10 and MLL rearrangements, these translocations do not generate fusion genes but lead to the over-expression of HOX11, HOX11L2 or HOX A genes. HOX11 and HOX11L2 are also developmentally important genes [33] and gene expression analysis demonstrates similarities within patients with TCR HOX11, HOX11L2 or HOX A cluster rearrangements and also between these patients and patients with MLL fusions suggesting overlapping programmes of self-renewal [23]. Rearrangement of another non-clustered HOX gene, PBX1, occurs in B-precursor ALL. The t(1;19) fuses the transactivation domain of the bHLH transcription factor E2A (TCP3) to the C terminal portion of PBX1. PBX1 is an important HOX gene cofactor, and it is likely that HOX gene function is dysregulated by the E2A-PBX1 fusion [34]. PBX1 is also critical for the regulation of HSC self-renewal by maintaining quiescence in the HSC [35]; therefore, E2A-PBX1 may also facilitate selfrenewal in ALL LSC. In addition, mice deficient in E2 demonstrate a significant defect in lymphoid development [36]; therefore, it is likely that the E2A-PBX1 fusion protein not only affects self-renewal but also blocks lymphoid differentiation (see later). 8.4.1.2 NOTCH Mutations An early indication that the NOTCH pathway was involved in T-ALL pathogenesis came with the description of chromosomal translocations that generated a truncated and activated form of NOTCH1 [37]. Subsequently, enforced Notch1 signalling
8 Targeting Leukemia Stem Cells and Stem Cell Pathways in ALL
153
was demonstrated to generate T-ALL in murine models [38, 39]. The NOTCH1 protein is generated as a pro-protein, which, following furin-like proteolysis, migrates to the cell membrane as a heterodimer consisting of extracellular and intracellular NOTCH (ICN) components. Following ligand-induced activation, a series of further cleavages occurs, the last step catalysed by the enzyme g-secretase, to generate the ICN protein. This, in turn, translocates to the nucleus, forms a complex with the CSL protein and mastermind co-factors and initiates transcription of the NOTCH expression programme [40]. More recently, it has been reported that more than 50% of patients with T-ALL carry somatic activating point mutations of NOTCH1 [41]. These mutations either affect the heterodimerisation domain leading to increased cleavage and liberation of ICN or truncate the protein, resulting in loss of the PEST domain that targets the intracellular protein for degradation by the proteosome. Both mutations independently lead to increased activation of the NOTCH signalling pathway and may occur on the same allele. The increase in Notch pathway signalling appears to both mandate T-cell differentiation and augment self-renewal properties in primitive haematopoietic progenitors, resulting in T-ALL. A similar heterodimerisation domain mutation in the Caenorhabditis elegans Notch homologue GLP-1 has been described and leads to a massive proliferation of germ cells [42]. Inhibitors to g-secretase already exist due to the involvement of g-secretase in the production of amyloidogenic peptides in Alzheimer’s patients and have demonstrated efficacy in murine models of NOTCH-related T-ALL [43]. Trials are currently underway with g-secretase inhibitors such as MK-0752 in patients with relapsed T-ALL and while worries of toxicity exist from trials in Alzheimer’s patients [44], early reports suggest some response with acceptable side effects [45]. However, further data are required to document the effects of NOTCH inhibition on T-ALL LSC. 8.4.1.3 PTEN Pathway Quiescence and cell cycle entry are very tightly controlled in normal HSC. This control facilitates self-renewal and prevents depletion of the stem cell pool. Central to this cell cycle control appears to be the PI3K-AKT-FOXO pathway, mediated through effectors such as p21 [46–48]. Recent reports have demonstrated that constitutive activation of the PI3K pathway through targeted deletion of PTEN, the major lipid phosphatase attenuating PI3K signalling, results both in HSC depletion and in the development of acute leukemia in mice [46]. Following deletion of PTEN, HSC numbers were seen to transiently rise, with increased entry into cell cycle. However, this was followed by both HSC exhaustion and decreased stem cell function. In addition, the mice also demonstrated an initial myeloproliferative phase that was quickly followed by an acute leukemia that was transplantable to secondary recipients [46]. Inhibition of this pathway with rapamycin not only restored normal cell cycle characteristics, normal number and reconstitution function to Pten−/− HSC but also prevented the development of leukemia. In addition, in transplant experiments rapamycin also depleted the number of leukemia-initiating
154
C. Pridans and B.J.P. Huntly
cells, and, importantly, rapamycin was shown to prolong survival even when administered to mice with established leukemia. Recent reports have demonstrated that NOTCH1 signalling represses expression of PTEN and that PTEN mutations occur in 17% of T-ALL patients and mediate resistance to g-secretase inhibitors [49]. Additionally, treatment of PTEN-deficient T-ALL cells with inhibitors of AKT signalling demonstrated strong anti-leukemic effects in vitro [49], and over half of ALL blasts analysed demonstrated apoptosis in response to rapamycin therapy in vitro [50]. This data suggests that LSC in T-ALL may be effectively targeted with combined PI3-AKT and NOTCH1 inhibition and that mTOR inhibition may be beneficial in targeting other forms of ALL LSC. 8.4.1.4 Dysregulation of bHLH Proteins and Their Binding Partners bHLH proteins are transcription factors that form heterodimers with other bHLH proteins to regulate programmes of gene expression. Several bHLH proteins (MYC, MYC-N, SCL/TAL1 and LYL1) or their binding partners (such as the LIM domain proteins LMO1 and LMO2) are mutated or dysregulated in ALL, particularly T-ALL (reviewed in [21]). They are implicated in a number of processes including HSC induction, proliferation and differentiation [51–53] and appear to function as both transcriptional activators and repressors. MYC binds to canonical E-box DNA sequences (5¢-CACGTTG-3¢) as a heterodimer with MAX and this complex alters transcription at numerous loci, usually functioning as a transcriptional activator [54]. MAX may also bind to MAD, inhibiting MYC function. The major mechanism of MYC dysregulation in mature B-cell ALL and Burkitt lymphoma is over-expression due to translocation to the immunoglobulin genes. In T-ALL, MYC may also be over-expressed due to translocation under the control of the TCR promoter but is also often over-expressed downstream of aberrant NOTCH1, LMO 1/2, TAL1 and HOX11 and HOX11L2. In addition the paralogous family member N-MYC is often over-expressed in T-ALL with over-expression of LYL1 [21]. High levels of MYC allow for an increase in MYC:MAX complexes relative to MAD:MAX complexes and facilitate transcription of the MYC programme. MYC target genes are numerous and depend upon the cellular context. These include activation of genes in the protein biogenesis pathway (such as ribosomal genes and the translational machinery), metabolism (such as glycolytic gene expression and mitochondrial biogenesis) and other transcription factors and cell cycle proteins (such as E2F, BMI1 and D Cyclins) [48]. Repressed targets include differentiation genes and CDK inhibitors such as p21, p27, p15 and p18 [51, 54]. Although MYC inhibitors do not yet exist, the central role of MYC as an effector of many oncogenic lesions in T-ALL and other malignancies make it a critical target and therapeutic focus with agents in drug development. Its pleiotropic effects on normal cell growth have suggested that such targeting might lead to unacceptable toxicity, however a recent report suggests that targeting MYC may be efficacious with acceptable side-effects [55] further adding to its attraction as a therapeutic target.
8 Targeting Leukemia Stem Cells and Stem Cell Pathways in ALL
155
TAL1 (also known as SCL) is a master regulator of haematopoietic development and is required for the generation of HSC and all blood lineages. However, it is not expressed in mature T-cells [56, 57]. The paralogous bHLH transcription factor LYL1 is also required for HSC function as well as differentiation into the B-cell lineage [58]. TAL1 and LYL1 over-expression may also result from their translocation to the proximity of control elements of the TCR genes; however, other mechanisms account for the majority of cases with TAL1 over-expression [59]. In addition to its effects on self-renewal [60], TAL1 blocks T-cell differentiation (see below). TAL1 also interacts with the LIM-only domain gene LMO 2, and both LMO 2 and its paralogue LMO 1 are over-expressed by translocation to the TCR enhancer loci in a subgroup of T-ALL [61] or by deletion of a negative regulatory element. LMO 2 forms a high order complex with TAL1, which appears necessary for HSC and subsequent haematopoietic development, as homozygous disruption of Lmo 2 in mice leads to embryonic lethality and Lmo 2−/− cells do not contribute to haematopoiesis in chimaeric mice, a phenotype very similar to the Tal1 knockout. Therefore, their effects in normal HSC biology and frequent mutation or overexpression in ALL make the TAL1/LMO2 complex an attractive target for ALL LSC therapy. 8.4.1.5 Abnormalities of the RB/TP53 Tumour Suppressor Network Although the retinoblastoma gene (RB) and the TP53 gene encoding the p53 t ranscription factor are themselves rarely altered by mutation in ALL, components of their pathways which critically control cell cycle progression and apoptosis frequently are. RB controls entry into the cell cycle, by inhibiting the E2F family of transcription factors from initiating S phase. This inhibition is relieved upon mitogenic signalling through phosphorylation of RB by cyclin D-Cdks and cyclin E-Cdk2 complexes. These complexes are in turn inhibited by the INK4 proteins (p16INK4a, p15INK4b, p18INK4c and p19INK4d), preventing RB phosphorylation and S phase progression [62]. Functional inactivation of this pathway is common in ALL, with the deletion or epigenetic silencing of p16INK4a and p15INK4b occurring in most cases of childhood T-ALL, and some cases of B-ALL and adult T-ALL [63, 64]. The tumour suppressor TP53 becomes activated by a large number of cellular insults, such as DNA damage, hypoxia and oncogenic cellular proliferation. This activation leads to either apoptosis or the arrest of cell cycle, dependent upon cellular context. Activation of TP53 is controlled by HDM2, a protein that binds to TP53 and induces its degradation. HDM2 is in turn controlled by the p14ARF (p19Arf in mice) tumour suppressor [62]. Deletion or epigenetic silencing of p14ARF is a frequent event in ALL [65], and over-expression of HDM2 or silencing of the major transcriptional target of TP53, p21CIP1, occurs in roughly half of patients with ALL [66]. The tumour suppressor genes p14ARF and p16INK4a are transcribed from the same genomic locus (also called CDKN2a) by alternative splicing and along with their neighbour p15INK4b are frequently involved in homozygous deletions in ALL
156
C. Pridans and B.J.P. Huntly
suggesting a collaboration between the RB and p53 pathways in the pathogenesis of ALL. In mouse models, genetic inhibition of the Rb and p53 pathways has been associated with an alteration in HSC self-renewal [48, 67, 68]. Moreover, the major polycomb PRC2 complex member BMI1 has been demonstrated to be critical for HSC self-renewal [69] and this effect appears to be dependent upon the ability of BMI1 to repress expression of p14ARF and p16INK4a, as deletion of p19Arf and p16Ink4a in mice substantially rescues the loss of Bmi1 [70]. A recent report using mice compound deficient in p53, p16INK4a and p19Arf demonstrates that abrogation of these pathways can confer self-renewal properties onto non-stem cell populations such as early progenitors [71], perhaps mirroring the transformation of lymphoid progenitors in some forms of ALL discussed earlier. It is likely that the interruption of these critical pathways in ALL contributes to the self-renewal of the ALL LSC and that targeting these pathways may in turn specifically target the LSC in ALL. Therapeutic targeting of the RB and p53 pathways is, therefore, an attractive option [72], particularly where the TP53 protein is intact as is often the case in ALL. A number of agents such as nutlins and other small molecules that inhibit the TP53-HDM2 interaction and flavopiridol, which is a more specific cyclin-dependent kinase inhibitor that blocks RB phosphorylation and cell cycle progression, are available and have demonstrated efficacy against ALL and other haematological malignancies [73, 74]. Many of the pathways described in this section are likely to be critical for the function of LSC in ALL and interruption/interference of these pathways may yield therapeutic dividends. However, multiple mutations have been described in individual cases of ALL [21, 75], and it is likely that these therapeutics will need to be combined with conventional therapies which also target non-tumourigenic tumour cells. In addition, therapeutic efficacy must also be achieved with acceptable side effects for normal HSC function and differences between the self-renewal requirements of ALL LSC and HSC must be further identified and exploited.
8.4.2 Differentiation Defects in ALL LSC Another cardinal feature of cancer is a block in differentiation [1], which in ALL facilitates the accumulation of lymphoid blasts. This block likely reflects repression of gene programmes associated with differentiation and is mediated through loss-of-function mutations to master regulators of lymphoid differentiation, such as PAX5 and IKAROS. Alternative mechanisms are via repression of these differentiation programmes by leukemia-associated fusion oncogenes such as E2APBX1, ETV6-RUNX1 and PAX5 fusions or the continued aberrant expression of transcription factors such as TAL1 and LMO2 whose inactivation is required for differentiation [59]. PAX5 is a master regulator of B-cell development and is dysregulated in ALL both by rare chromosomal translocations [76, 77] or, more commonly, through mutation or deletion in up to 32% of cases of B-progenitor ALL [75]. The Ikaros family of transcription factors are required for both T and
8 Targeting Leukemia Stem Cells and Stem Cell Pathways in ALL
157
B lymphocyte development, and mutations of the IKAROS 1 gene IKZF1, particularly partial or complete deletions leading to a loss of function, are commonly seen in ALL [78]. This is particularly prevalent in Ph+ ALL where 84% of patients demonstrated such a deletion [78]. Other chromosomal translocations associated with ALL are also thought to antagonise differentiation pathways, although this occurs via poorly understood mechanisms [59]. As discussed previously, E2A is commonly fused to PBX1 in paediatric pre-B cell ALL cases and rarely to HLF in B-progenitor ALL [79]. E2A is required for normal lymphoid development [36], and its dysregulation through formation of the fusion protein presumably contributes to the differentiation block. Similarly, AML1 is critical for normal T and B-cell development [80, 81], and it is presumed that the ETV6-RUNX1 fusion protein represses many of the genes involved in these differentiation programmes [82]. In addition, for normal T-cell differentiation to proceed, genetic programmes induced by class I bHLH E47 homo or E47/HEB heterodimers are required. The TAL1 protein and its interacting protein LMO2, whose expressions are upregulated in many T-ALL cases as discussed previously, appear to antagonise these programmes and block differentiation. Having identified the effects of these mutations on differentiation it is theoretically possible to alleviate the blockage and facilitate differentiation of the leukemic blasts. In fact, the paradigm of differentiation therapy exists in the treatment of acute leukemias, such as acute promyelocytic leukemia (APML) associated with PML-RARA gene rearrangements, where pharmacological doses of all-trans retinoic acid (ATRA) override the differentiation block and induce myeloid differentiation [83]. In contrast, no current therapeutics is available to target the molecular lesions responsible for the differentiation block in ALL. However, future modulation with highly specific histone deacetylase inhibitors, peptidomimetics of PAX5 or IKZF1, or small molecular inhibition of or systemic siRNA delivery against leukemia-associated fusion genes or the TAL1/LMO2 complex may allow this.
8.4.3 Aberrant Cell Signalling in ALL LSC Aberrant intracellular signalling is a common finding in acute leukemias and alters proliferation, differentiation and survival via the constitutive engagement of pathways downstream of normal tyrosine kinase and cytokine receptors. In addition, in mouse models these mutant signalling pathways collaborate with mutations altering the function of critical transcription factors to generate the acute leukemia phenotype [84]. There is particular interest in the mutations that generate aberrant signalling, as these often arise in proteins with enzymatic function such as tyrosine kinases, which are in turn amenable to specific or selective inhibition with small molecules. With regard to abnormal signalling in ALL LSC, we specifically discuss targeting the BCR-ABL1 and NUP214-ABL1 gene rearrangement products, mutations of the tyrosine kinase domain of the receptor tyrosine kinase FLT3 (FLT3-TKD) and mutations of Janus Kinase (JAK) isoforms in the JAK/STAT signalling pathway.
158
C. Pridans and B.J.P. Huntly
8.4.3.1 BCR-ABL1 and NUP214-ABL1 Signalling The incidence of BCR-ABL1 rearrangements, associated with the Philadelphia chromosome and t(9;22) chromosomal translocation increases with patient’s age from around 5% in children to around one third of all adult cases of ALL [21]. BCR-ABL1 encodes a chimaeric and constitutively active tyrosine kinase, which alters many cellular processes such as proliferation, apoptosis and cell motility [85], and clinically Philadelphia-positive ALL (Ph+ ALL) is associated with a dismal outcome. It is an indication for early stem cell transplantation, and without this therapeutic option it is seldom associated with long-term survival. In addition, episomal rearrangement of ABL1 with the nucleoporin protein NUP214 has been described in around 5% of T-ALL patients [86]. Recently, the treatment of chronic myeloid leukemia (CML), the archetypal disease associated with BCR-ABL1 rearrangement, has been revolutionised by the introduction of imatinib a selective ABL1 kinase inhibitor. Ph+ ALL patients and cell lines that carry the NUP214-ABL1 rearrangement are sensitive to the tyrosine kinase inhibitors imatinib, nilotinib and dasatinib [87–89], suggesting their efficacy in ALL. Unfortunately, their clinical effect is short-lived as single agents, with the development of resistance, commonly through kinase domain mutations and other mechanisms. However, they are currently in trials in combination with conventional agents with promising initial results [90, 91]. In CML, the quiescent LSC is inherently resistant to tyrosine kinase inhibitors through unknown mechanisms [92–94]. The situation in Ph+ ALL is less clear; however, murine models suggest that BCR-ABL1 has different requirements for the induction of B-ALL, where it collaborates with Src kinases, than for the induction of CML, where this requirement is not necessary [95]. Taking advantage of this requirement, a further murine model has suggested that dual inhibition of both BCR-ABL1 and Src kinases with an agent such as dasatinib leads to significant reduction in leukemia-initiating cells, although they do persist throughout therapy [96]. Together, these data suggest that dual ABL/Src kinase inhibitors may more effectively target ALL LSC. In CML, the role of second-generation kinase inhibitors such as nilotinib and dasatinib is additionally being investigated both as upfront therapy and as salvage therapy for those who fail to respond to imatinib. Transformation by BCR-ABL1 is also mediated by its ability to enhance the expression and activity of specific RNA-binding proteins such as hnRNP 1 [97]. hnRNP 1 in turn associates with target transcripts such as the SET mRNA. SET is a potent inhibitor of the tumour suppressor PP2A, a major cellular serine/threonine phosphatase involved in the negative regulation of signalling initiated by multiple protein kinases [98]. PP2A activity is also reduced in Ph+ ALL and CML patients, and restoration of PP2A function resulted in growth suppression, enhanced apoptosis, restored differentiation, impaired clonogenic potential and decreased in vivo leukemogenesis in BCR-ABL1+ cells [98]. This identifies modulation of PP2A as a potential target in BCR-ABL1 positive leukemias and modulators such as FTY720, which is already in phase III trials with an acceptable side-effect profile for multiple sclerosis, have demonstrated significant preclinical efficacy [99, 100].
8 Targeting Leukemia Stem Cells and Stem Cell Pathways in ALL
159
One of the reports also suggests Ph+ ALL LSCs are preferentially targeted, since FTY720 induced apoptosis in CD34+/CD19+ progenitors whilst sparing normal CD34+ cells [99]. FTY720 will soon enter clinical trials.
8.4.3.2 FLT3-TKD Mutant Signalling The receptor tyrosine kinase FLT3 is important for both normal and malignant haematopoiesis. It is expressed in both lymphoid and myeloid lineages, and in bone marrow it is restricted to early progenitors [101]. Mice with targeted disruption of Flt3 demonstrate multiple abnormalities of development of the B and T-cell lineages, and FLT3 ligand (FL) can expand CD34+ cells in vitro and in vivo. In addition, mutations and over-expression of FLT3 are common in acute leukemias. Whilst internal tandem duplications of the juxtamembrane domain predominate in AML, the FLT3 mutations that occur in paediatric ALL mainly affect the kinase domain (TKD) directly, again constitutively activating the receptor [102]. These mutations mainly occur in MLL rearranged cases (15% of which also have a FLT3 TKD mutation) or in hyperdiploid disease [102, 103]. In addition, in MLL cases lacking a FLT3 TKD the majority significantly over-express FLT3 [102], making inhibition of FLT3 signalling with selective inhibitors an attractive proposition. In vitro, ALL blasts have been demonstrated to be sensitive to FLT3 inhibitors such as PKC412 [102, 104] and FLT3 inhibitors have shown efficacy in early trials in AML patients [105] where they are currently in Phase III trials along with combination chemotherapy. This has provided a rationale for their use in ALL and trials are currently planned [106]. 8.4.3.3 JAK Mutations and Aberrant Signalling JAK receptor associated tyrosine kinases are important signalling intermediaries downstream of type I cytokine receptors and engage a number of critical signalling pathways such as the STAT, RAS-RAF and PI3K pathways in HSC and progenitor cells and multiple mature haematopoietic lineages [107]. There are four family members, namely, JAK1, 2, 3 and TYK1, and mice deficient in the various forms demonstrate a number of haematopoietic and immunological deficiencies [108]. Moreover, STAT5, a major downstream signalling intermediary has also been associated with self-renewal properties of HSC [109, 110]. Recently, mutations of JAK1 and JAK2 have been described in ALL [111–113], and rearrangement of JAK2 with the TEL transcription factor as a result of t(9;12) is also a rare event associated with ALL [114]. Mutations in JAK1 more commonly occur in adult T-ALL (where they were present in up to 18% of cases) but also occur less frequently in childhood T-ALL and adult B-ALL [111], whereas mutations of JAK2 have been described in ALL associated with Down syndrome [113]. Fortuitously, selective JAK inhibitors are available and in development and are currently in
160
C. Pridans and B.J.P. Huntly
clinical trials in Ph+ negative myeloproliferative disorders associated with the JAK2 V617F mutation [115]. It is likely that they will also have utility in ALL cases associated with JAK mutations.
8.4.4 Apoptotic Defects in ALL LSC Evasion of normal apoptotic mechanisms is a common feature of all cancer types [1]. Apoptosis is mediated by one of two pathways, the extrinsic pathway linked to ligand binding of external death receptors such as FAS and the intrinsic pathway, whose induction relies on the balance of pro- and anti-apoptotic proteins of the BCL2 family and their subsequent effects on the outer mitochondrial membrane [116]. The decreased levels of apoptosis observed in many cancers are often regulated by over-expression or activation of BCL2 anti-apoptotic proteins family members such as BCL2, MCL-1, BCL-XL and BCL-w [117]. Glucocorticoids have long been known to induce apoptosis in lymphoid malignancies including ALL [118], and recently, ALL cell lines and patient samples have been frequently demonstrated to be dependent upon BCL2 for survival [117]. Anti-apoptotic members of the BCL2 family bind to and sequester pro-apoptotic BCL2 family members such as BAX and BAK and prevent them from altering the properties of the outer mitochondrial membrane and inducing apoptosis. A third group of BCL2 family proteins, the BH3-only proteins antagonise BAX/BAK sequestration and facilitate this reaction. Recently, small-molecule mimetics of BH3-only proteins have been developed to antagonise this interaction and facilitate apoptosis. These agents are covered in greater depth in later chapters, but agents such as ABT-737 have shown some efficacy in ALL and may work at the level of the LSC [117]. The NF-kB pathway has been shown to be activated in cancer, where it mediates growth and proliferation signals, tumour invasion, metastasis and particularly evasion of apoptosis [119]. NF-kB is also constitutively active in ALL [120], and this pathway also appears to be directly downstream of specific molecular aberrations in ALL, such as NOTCH1 [121] and TAL1 [122] mutations in T-ALL and BCR-ABL in Ph+ ALL [123]. ALL cell lines and patient samples are sensitive to inhibition of the pathway with IKK and proteasome inhibitors [121]. In AML, the NF-kB pathway is also activated, and this activation status has been demonstrated to differ in primary AML CD34+ cells and normal CD34+ cells [124]. It is likely, therefore, that activation of the NF-kB pathway also occurs in ALL LSC and that the differing requirements of leukemic and normal stem cells may be taken advantage of. This has been shown in AML where the ability of the proteosome inhibitor MG-132, a dominant negative inhibitor of NF-kB (IkB) and the sesquiterpine lactone Parthenolide (PTL), which also inhibits NF-kB and produces oxidative stress, generate apoptosis in CD34+ AML cells but not in normal CD34+ cells [124, 125]. PTL also preferentially induces apoptosis in ALL CD34+/CD10− cells [126] and clinical trials in refractory leukemias with an orally bioavailable analogue CL-1 are currently underway.
8 Targeting Leukemia Stem Cells and Stem Cell Pathways in ALL
161
8.5 Conclusions ALL is a truly heterogeneous disease. It occurs across all age groups with a bimodal distribution, and although it is morphologically similar across this spectrum, there are huge differences in the molecular biology and clinical behaviour between childhood and adult cases. The study of ALL LSC biology is in its infancy; however, the available data would suggest that this heterogeneity extends into the LSC compartment and may explain some of the clinical differences seen. A growing knowledge and understanding of the identity and biology of this critical population of cells will not only shed light on the heterogeneity of this disease but will also facilitate therapies directed specifically against the LSC in ALL. This chapter details what is known of the cellular and molecular biology of the LSC compartment in ALL and outlines processes and pathways that might be targeted for therapeutic gain. However, much work is still required, with the ultimate goal to bring rates of longterm cure in both children and adults into line.
References 1. Hanahan, D. and R. Weinberg, The hallmarks of cancer. Cell, 2000. 100:57–70. 2. Reya, T., et al., Stem cells, cancer, and cancer stem cells. Nature, 2001. 414:105–111. 3. Huntly, B.J. and D.G. Gilliland, Leukemia stem cells and the evolution of cancer-stem-cell research. Nat Rev Cancer, 2005. 5(4):311–21. 4. Kelly, P.N., et al., Tumor growth need not be driven by rare cancer stem cells. Science, 2007. 317(5836):337. 5. Hill, R.P., Identifying cancer stem cells in solid tumors: case not proven. Cancer Res, 2006. 66(4):1891–5; discussion 1890. 6. Lapidot, T., et al., A cell initiating human acute myeloid leukemia after transplantation into SCID mice. Nature, 1994. 367(6464):645–8. 7. Bonnet, D. and J.E. Dick, Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med, 1997. 3(7):730–7. 8. Kvinlaug, B.T. and B.J. Huntly, Targeting cancer stem cells. Expert Opin Ther Targets, 2007. 11(7):915–27. 9. Pearson, T., D.L. Greiner, and L.D. Shultz, Creation of “humanized” mice to study human immunity. Curr Protoc Immunol, 2008. Chapter 15:Unit 15 21. 10. Taussig, D.C., et al., Anti-CD38 antibody-mediated clearance of human repopulating cells masks the heterogeneity of leukemia-initiating cells. Blood, 2008. 112(3):568–75. 11. Greaves, M., Molecular genetics, natural history and the demise of childhood leukemia. Eur J Cancer, 1999. 35(14):1941–53. 12. Cobaleda, C., et al., A primitive hematopoietic cell is the target for the leukemic transformation in human philadelphia-positive acute lymphoblastic leukemia. Blood, 2000. 95(3):1007–13. 13. Cox, C.V., et al., Characterization of acute lymphoblastic leukemia progenitor cells. Blood, 2004. 104(9):2919–25. 14. Castor, A., et al., Distinct patterns of hematopoietic stem cell involvement in acute lymphoblastic leukemia. Nat Med, 2005. 11(6):630–7. 15. Hong, D., et al., Initiating and cancer-propagating cells in ETV6-RUNX1 associated childhood leukemia. Science, 2008. 319(5861):336–9. 16. Kong, Y., et al., CD34+CD38+CD19+ as well as CD34+CD38-CD19+ cells are leukemia-initiating cells with self-renewal capacity in human B-precursor ALL. Leukemia, 2008. 22(6):1207–13.
162
C. Pridans and B.J.P. Huntly
17. le Viseur, C., et al., In childhood acute lymphoblastic leukemia, blasts at different stages of immunophenotypic maturation have stem cell properties. Cancer Cell, 2008. 14(1):47–58. 18. Cox, C.V., et al., Characterization of a progenitor cell population in childhood T-cell acute lymphoblastic leukemia. Blood, 2007. 109(2):674–82. 19. Gokbuget, N. and D. Hoelzer, Novel antibody-based therapy for acute lymphoblastic leukemia. Best Pract Res Clin Haematol, 2006. 19(4):701–13. 20. Pui, C.H., M.V. Relling, and J.R. Downing, Acute lymphoblastic leukemia. N Engl J Med, 2004. 350(15):1535–48. 21. Armstrong, S.A. and A.T. Look, Molecular genetics of acute lymphoblastic leukemia. J Clin Oncol, 2005. 23(26):6306–15. 22. Armstrong, S.A., et al., MLL translocations specify a distinct gene expression profile that distinguishes a unique leukemia. Nat Genet, 2002. 30(1):41–7. 23. Soulier, J., et al., HOXA genes are included in genetic and biologic networks defining human acute T-cell leukemia (T-ALL). Blood, 2005. 106(1):274–86. 24. Owens, B.M. and R.G. Hawley, HOX and non-HOX homeobox genes in leukemic hematopoiesis. Stem Cells, 2002. 20(5):364–79. 25. Ayton, P.M. and M.L. Cleary, Transformation of myeloid progenitors by MLL oncoproteins is dependent on Hoxa7 and Hoxa9. Genes Dev, 2003. 17(18):2298–307. 26. Ernst, P., J. Wang, and S.J. Korsmeyer, The role of MLL in hematopoiesis and leukemia. Curr Opin Hematol, 2002. 9(4):282–7. 27. Wang, Z., et al., Glycogen synthase kinase 3 in MLL leukemia maintenance and targeted therapy. Nature, 2008. 455(7217):1205–9. 28. Sato, N., et al., Maintenance of pluripotency in human and mouse embryonic stem cells through activation of Wnt signaling by a pharmacological GSK-3-specific inhibitor. Nat Med, 2004. 10(1):55–63. 29. Guzman, M.L., et al., Rapid and selective death of leukemia stem and progenitor cells induced by the compound 4-benzyl, 2-methyl, 1,2,4-thiadiazolidine, 3,5 dione (TDZD-8). Blood, 2007. 110(13):4436–44. 30. Asnafi, V., et al., CALM-AF10 is a common fusion transcript in T-ALL and is specific to the TCRgammadelta lineage. Blood, 2003. 102(3):1000–6. 31. Okada, Y., et al., Leukaemic transformation by CALM-AF10 involves upregulation of Hoxa5 by hDOT1L. Nat Cell Biol, 2006. 8(9):1017–24. 32. Deshpande, A.J., et al., Acute myeloid leukemia is propagated by a leukemic stem cell with lymphoid characteristics in a mouse model of CALM/AF10-positive leukemia. Cancer Cell, 2006. 10(5):363–74. 33. Roberts, C.W., J.R. Shutter, and S.J. Korsmeyer, Hox11 controls the genesis of the spleen. Nature, 1994. 368(6473):747–9. 34. Chang, C.P., I. de Vivo, and M.L. Cleary, The Hox cooperativity motif of the chimeric oncoprotein E2a-Pbx1 is necessary and sufficient for oncogenesis. Mol Cell Biol, 1997. 17(1):81–8. 35. Ficara, F., et al., Pbx1 regulates self-renewal of long-term hematopoietic stem cells by maintaining their quiescence. Cell Stem Cell, 2008. 2(5):484–96. 36. Sigvardsson, M., M. O’Riordan, and R. Grosschedl, EBF and E47 collaborate to induce expression of the endogenous immunoglobulin surrogate light chain genes. Immunity, 1997. 7(1):25–36. 37. Ellisen, L.W., et al., TAN-1, the human homolog of the Drosophila notch gene, is broken by chromosomal translocations in T lymphoblastic neoplasms. Cell, 1991. 66(4):649–61. 38. Aster, J.C., et al., Essential roles for ankyrin repeat and transactivation domains in induction of T-cell leukemia by notch1. Mol Cell Biol, 2000. 20(20):7505–15. 39. Hoemann, C.D., et al., Two distinct Notch1 mutant alleles are involved in the induction of T-cell leukemia in c-myc transgenic mice. Mol Cell Biol, 2000. 20(11):3831–42. 40. Aster, J.C., W.S. Pear, and S.C. Blacklow, Notch signaling in leukemia. Annu Rev Pathol, 2008. 3:587–613. 41. Weng, A.P., et al., Activating mutations of NOTCH1 in human T cell acute lymphoblastic leukemia. Science, 2004. 306(5694):269–71.
8 Targeting Leukemia Stem Cells and Stem Cell Pathways in ALL
163
42. Berry, L.W., B. Westlund, and T. Schedl, Germ-line tumor formation caused by activation of glp-1, a Caenorhabditis elegans member of the Notch family of receptors. Development, 1997. 124(4):925–36. 43. Kindler, T., et al., K-RasG12D-induced T-cell lymphoblastic lymphoma/leukemias harbor Notch1 mutations and are sensitive to gamma-secretase inhibitors. Blood, 2008. 112(8): 3373–82. 44. Wolfe, M.S., Inhibition and modulation of gamma-secretase for Alzheimer’s disease. Neurotherapeutics, 2008. 5(3):391–8. 45. DeAngelo, D.J., et al., Aphase I clinical trial of the notch inhibitor MK-0752 in patients with T-cell acute lymphoblastic leukemia/lymphoma (T-ALL) and other leukemias. J Clin Oncol, 2006. 24(18S):6585. 46. Yilmaz, O.H., et al., Pten dependence distinguishes haematopoietic stem cells from leukemia-initiating cells. Nature, 2006. 441(7092):475–82. 47. Tothova, Z., et al., FoxOs Are Critical Mediators of Hematopoietic Stem Cell Resistance to Physiologic Oxidative Stress. Cell, 2007. 128(2):325–339. 48. Cheng, T., et al., Hematopoietic stem cell quiescence maintained by p21cip1/waf1. Science, 2000. 287(5459):1804–8. 49. Palomero, T., et al., Mutational loss of PTEN induces resistance to NOTCH1 inhibition in T-cell leukemia. Nat Med, 2007. 13(10):1203–10. 50. Avellino, R., et al., Rapamycin stimulates apoptosis of childhood acute lymphoblastic leukemia cells. Blood, 2005. 106(4):1400–6. 51. Wilson, A., et al., c-Myc controls the balance between hematopoietic stem cell self-renewal and differentiation. Genes Dev, 2004. 18(22):2747–63. 52. Begley, C.G. and A.R. Green, The SCL gene: from case report to critical hematopoietic regulator. Blood, 1999. 93(9):2760–70. 53. Loose, M., G. Swiers, and R. Patient, Transcriptional networks regulating hematopoietic cell fate decisions. Curr Opin Hematol, 2007. 14(4):307–14. 54. Eilers, M. and R.N. Eisenman, Myc’s broad reach. Genes Dev, 2008. 22(20):2755–66. 55. Soucek, L., et al., Modelling Myc inhibition as a cancer therapy. Nature, 2008. 455(7213):679–83. 56. Shivdasani, R.A., E.L. Mayer, and S.H. Orkin, Absence of blood formation in mice lacking the T-cell leukemia oncoprotein tal-1/SCL. Nature, 1995. 373:432–434. 57. Robb, L., et al., Absence of yolk sac hematopoiesis from mice with a targeted disruption of the scl gene. Proc Natl Adad Sci USA, 1995. 92:7075–7079. 58. Capron, C., et al., The SCL relative LYL-1 is required for fetal and adult hematopoietic stem cell function and B-cell differentiation. Blood, 2006. 107(12):4678–86. 59. O’Neil, J. and A.T. Look, Mechanisms of transcription factor deregulation in lymphoid cell transformation. Oncogene, 2007. 26(47):6838–49. 60. Reynaud, D., et al., SCL/TAL1 expression level regulates human hematopoietic stem cell self-renewal and engraftment. Blood, 2005. 106(7):2318–28. 61. Boehm, T., et al., The rhombotin family of cysteine-rich LIM-domain oncogenes: distinct members are involved in T-cell translocations to human chromosomes 11p15 and chromosome 11p13. Proc Natl Acad Sci USA, 1991. 88:4367–4371. 62. Sherr, C.J. and F. McCormick, The RB and p53 pathways in cancer. Cancer Cell, 2002. 2(2):103–12. 63. Omura-Minamisawa, M., et al., Universal inactivation of both p16 and p15 but not downstream components is an essential event in the pathogenesis of T-cell acute lymphoblastic leukemia. Clin Cancer Res, 2000. 6(4):1219–28. 64. Stock, W., et al., Cell cycle regulatory gene abnormalities are important determinants of leukemogenesis and disease biology in adult acute lymphoblastic leukemia. Blood, 2000. 95(7):2364–71. 65. Calero Moreno, T.M., et al., Deletion of the Ink4-locus (the p16ink4a, p14ARF and p15ink4b genes) predicts relapse in children with ALL treated according to the Nordic protocols NOPHO-86 and NOPHO-92. Leukemia, 2002. 16(10):2037–45. 66. Sherr, C.J., The INK4a/ARF network in tumour suppression. Nat Rev Mol Cell Biol, 2001. 2(10):731–7.
164
C. Pridans and B.J.P. Huntly
67. Cheng, T., et al., Stem cell repopulation efficiency but not pool size is governed by p27(kip1). Nat Med, 2000. 6(11):1235–40. 68. Yuan, Y., et al., In vivo self-renewing divisions of haematopoietic stem cells are increased in the absence of the early G1-phase inhibitor, p18INK4C. Nat Cell Biol, 2004. 6(5):436–42. 69. Lessard, J. and G. Sauvageau, Bmi-1 determines the proliferative capacity of normal and leukaemic stem cells. Nature, 2003. 423:255–260. 70. Oguro, H., et al., Differential impact of Ink4a and Arf on hematopoietic stem cells and their bone marrow microenvironment in Bmi1-deficient mice. J Exp Med, 2006. 203(10):2247–53. 71. Akala, O.O., et al., Long-term haematopoietic reconstitution by Trp53-/-p16Ink4a-/p19Arf-/- multipotent progenitors. Nature, 2008. 453(7192):228–32. 72. Chene, P., Inhibiting the p53-MDM2 interaction: an important target for cancer therapy. Nat Rev Cancer, 2003. 3(2):102–9. 73. Secchiero, P., et al., The MDM2 inhibitor Nutlins as an innovative therapeutic tool for the treatment of haematological malignancies. Curr Pharm Des, 2008. 14(21):2100–10. 74. Karp, J.E., et al., Phase I and pharmacokinetic study of flavopiridol followed by 1-beta-Darabinofuranosylcytosine and mitoxantrone in relapsed and refractory adult acute leukemias. Clin Cancer Res, 2005. 11(23):8403–12. 75. Mullighan, C.G., et al., Genome-wide analysis of genetic alterations in acute lymphoblastic leukemia. Nature, 2007. 446(7137):758–64. 76. Bousquet, M., et al., A novel PAX5-ELN fusion protein identified in B-cell acute lymphoblastic leukemia acts as a dominant negative on wild-type PAX5. Blood, 2007. 109(8):3417–23. 77. Fazio, G., et al., PAX5/TEL acts as a transcriptional repressor causing down-modulation of CD19, enhances migration to CXCL12, and confers survival advantage in pre-BI cells. Cancer Res, 2008. 68(1):181–9. 78. Mullighan, C.G., et al., BCR-ABL1 lymphoblastic leukemia is characterized by the deletion of Ikaros. Nature, 2008. 453(7191):110–4. 79. Hunger, S.P., et al., Two types of genomic rearrangements create alternative E2A-HLF fusion proteins in t(17;19)-ALL. Blood, 1994. 83(10):2970–7. 80. Ichikawa, M., et al., AML-1 is required for megakaryocytic maturation and lymphocytic differentiation, but not for maintenance of hematopoietic stem cells in adult hematopoiesis. Nat Med, 2004. 10(3):299–304. 81. Growney, J.D., et al., Loss of Runx1 perturbs adult hematopoiesis and is associated with a myeloproliferative phenotype. Blood, 2005. 106(2):494–504. 82. Hiebert, S.W., et al., The t(12;21) translocation converts AML-1B from an activator to a repressor of transcription. Mol Cell Biol, 1996. 16(4):1349–55. 83. Tenen, D.G., Disruption of differentiation in human cancer: AML shows the way. Nat Rev Cancer, 2003. 3(2):89–101. 84. Kelly, L. and D. Gilliland, Genetics of myeloid leukemias. Annual Review of Genomics and Human Genetics, 2002. 3:179–98. 85. Deininger, M., J. Goldman, and J. Melo, The molecular biology of chronic myeloid leukemia. Blood, 2000. 96:3343–3356. 86. Graux, C., et al., Fusion of NUP214 to ABL1 on amplified episomes in T-cell acute lymphoblastic leukemia. Nat Genet, 2004. 36(10):1084–9. 87. Talpaz, M., et al., Dasatinib in imatinib-resistant Philadelphia chromosome-positive leukemias. N Engl J Med, 2006. 354(24):2531–41. 88. Kantarjian, H., et al., Nilotinib in imatinib-resistant CML and Philadelphia chromosomepositive ALL. N Engl J Med, 2006. 354(24):2542–51. 89. Quintas-Cardama, A., et al., Activity of tyrosine kinase inhibitors against human NUP214ABL1-positive T cell malignancies. Leukemia, 2008. 22(6):1117–24. 90. Towatari, M., et al., Combination of intensive chemotherapy and imatinib can rapidly induce high-quality complete remission for a majority of patients with newly diagnosed BCR-ABLpositive acute lymphoblastic leukemia. Blood, 2004. 104(12):3507–12. 91. Yanada, M., et al., High complete remission rate and promising outcome by combination of imatinib and chemotherapy for newly diagnosed BCR-ABL-positive acute lymphoblastic
8 Targeting Leukemia Stem Cells and Stem Cell Pathways in ALL
165
leukemia: a phase II study by the Japan Adult Leukemia Study Group. J Clin Oncol, 2006. 24(3):460–6. 92. Graham, S.M., et al., Primitive, quiescent, Philadelphia-positive stem cells from patients with chronic myeloid leukemia are insensitive to STI571 in vitro. Blood, 2002. 99(1):319–25. 93. Jorgensen, H.G., et al., Nilotinib exerts equipotent antiproliferative effects to imatinib and does not induce apoptosis in CD34+ CML cells. Blood, 2007. 109(9):4016–9. 94. Copland, M., et al., Dasatinib (BMS-354825) targets an earlier progenitor population than imatinib in primary CML but does not eliminate the quiescent fraction. Blood, 2006. 107(11):4532–9. 95. Hu, Y., et al., Requirement of Src kinases Lyn, Hck and Fgr for BCR-ABL1-induced B-lymphoblastic leukemia but not chronic myeloid leukemia. Nat Genet, 2004. 36(5):453–61. 96. Hu, Y., et al., Targeting multiple kinase pathways in leukemic progenitors and stem cells is essential for improved treatment of Ph+ leukemia in mice. Proc Natl Acad Sci U S A, 2006. 103(45):16870–5. 97. Perrotti, D., et al., BCR-ABL suppresses C/EBPalpha expression through inhibitory action of hnRNP E2. Nat Genet, 2002. 30(1):48–58. 98. Neviani, P., et al., The tumor suppressor PP2A is functionally inactivated in blast crisis CML through the inhibitory activity of the BCR/ABL-regulated SET protein. Cancer Cell, 2005. 8(5):355–68. 99. Neviani, P., et al., FTY720, a new alternative for treating blast crisis chronic myelogenous leukemia and Philadelphia chromosome-positive acute lymphocytic leukemia. J Clin Invest, 2007. 117(9):2408–21. 100. Liu, Q., et al., FTY720 demonstrates promising preclinical activity for chronic lymphocytic leukemia and lymphoblastic leukemia/lymphoma. Blood, 2008. 111(1):275–84. 101. Gilliland, D.G. and J.D. Griffin, The roles of FLT3 in hematopoiesis and leukemia. Blood, 2002. 100(5):1532–42. 102. Armstrong, S.A., et al., Inhibition of FLT3 in MLL. Validation of a therapeutic target identified by gene expression based classification. Cancer Cell, 2003. 3(2):173–83. 103. Armstrong, S.A., et al., FLT3 mutations in childhood acute lymphoblastic leukemia. Blood, 2004. 103(9):3544–6. 104. Stam, R.W., et al., Targeting FLT3 in primary MLL-gene-rearranged infant acute lymphoblastic leukemia. Blood, 2005. 106(7):2484–90. 105. Knapper, S., et al., A phase 2 trial of the FLT3 inhibitor lestaurtinib (CEP701) as first-line treatment for older patients with acute myeloid leukemia not considered fit for intensive chemotherapy. Blood, 2006. 108(10):3262–70. 106. Stubbs, M.C. and S.A. Armstrong, FLT3 as a therapeutic target in childhood acute leukemia. Curr Drug Targets, 2007. 8(6):703–14. 107. Baker, S.J., S.G. Rane, and E.P. Reddy, Hematopoietic cytokine receptor signaling. Oncogene, 2007. 26(47):6724–37. 108. Ward, A.C., I. Touw, and A. Yoshimura, The Jak-Stat pathway in normal and perturbed hematopoiesis. Blood, 2000. 95(1):19–29. 109. Kato, Y., et al., Selective activation of STAT5 unveils its role in stem cell self-renewal in normal and leukemic hematopoiesis. J Exp Med, 2005. 202(1):169–79. 110. Schuringa, J.J., et al., Constitutive Activation of STAT5A Promotes Human Hematopoietic Stem Cell Self-Renewal and Erythroid Differentiation. J Exp Med, 2004. 200(5):623–35. 111. Flex, E., et al., Somatically acquired JAK1 mutations in adult acute lymphoblastic leukemia. J Exp Med, 2008. 205(4):751–8. 112. Jeong, E.G., et al., Somatic mutations of JAK1 and JAK3 in acute leukemias and solid cancers. Clin Cancer Res, 2008. 14(12):3716–21. 113. Bercovich, D., et al., Mutations of JAK2 in acute lymphoblastic leukemias associated with Down’s syndrome. Lancet, 2008. 372(9648):1484–92. 114. Lacronique, V., et al., A TEL-JAK2 fusion protein with constitutive kinase activity in human leukemia. Science, 1997. 278:1309–1312.
166
C. Pridans and B.J.P. Huntly
115. Levine, R.L. and D.G. Gilliland, Myeloproliferative disorders. Blood, 2008. 112(6):2190–8. 1 16. Adams, J.M. and S. Cory, The Bcl-2 apoptotic switch in cancer development and therapy. Oncogene, 2007. 26(9):1324–37. 117. Del Gaizo Moore, V., et al., BCL-2 dependence and ABT-737 sensitivity in acute lymphoblastic leukemia. Blood, 2008. 111(4):2300–9. 118. Schmidt, S., et al., Identification of glucocorticoid-response genes in children with acute lymphoblastic leukemia. Blood, 2006. 107(5):2061–9. 119. Basseres, D.S. and A.S. Baldwin, Nuclear factor-kappaB and inhibitor of kappaB kinase pathways in oncogenic initiation and progression. Oncogene, 2006. 25(51):6817–30. 120. Weston, V.J., et al., Apoptotic resistance to ionizing radiation in pediatric B-precursor acute lymphoblastic leukemia frequently involves increased NF-kappaB survival pathway signaling. Blood, 2004. 104(5):1465–73. 121. Vilimas, T., et al., Targeting the NF-kappaB signaling pathway in Notch1-induced T-cell leukemia. Nat Med, 2007. 13(1):70–7. 122. Chang, P.Y., et al., NFKB1 is a direct target of the TAL1 oncoprotein in human T leukemia cells. Cancer Res, 2006. 66(12):6008–13. 123. Munzert, G., et al., Constitutive NF-kappab/Rel activation in philadelphia chromosome positive (Ph+) acute lymphoblastic leukemia (ALL). Leuk Lymphoma, 2004. 45(6):1181–4. 124. Guzman, M.L., et al., Nuclear factor-kappaB is constitutively activated in primitive human acute myelogenous leukemia cells. Blood, 2001. 98(8):2301–7. 125. Guzman, M.L., et al., The sesquiterpene lactone parthenolide induces apoptosis of human acute myelogenous leukemia stem and progenitor cells. Blood, 2005. 105(11):4163–9. 126. Guzman, M.L., et al., An orally bioavailable parthenolide analog selectively eradicates acute myelogenous leukemia stem and progenitor cells. Blood, 2007. 110(13):4427–35.
Chapter 9
Nucleoside Analogues Pamela Kearns and Vaskar Saha
9.1 Clofarabine 9.1.1 Background Clofarabine (CL-F-ara-A, 2 chloro-2¢-flouro-deoxy-9-b-d-arabinofuranosyladenine), a new generation nucleoside analogue, is one of two new compounds that have been granted marketing approval for acute lymphoblastic leukemia in the paediatric age group, both by the US FDA and by the European Medicines Agency (EMA). It was the first new drug for paediatric leukemia to be approved in more than a decade and the only one to receive approval for paediatric use before adult use. The early-phase clinical trials in the paediatric age group generated the pivotal evidence of clofarabine’s efficacy in patients with highly refractory lymphoid leukemias. The approval granted was necessarily restricted to this population based on the data submitted. The marketing approval is for the use of clofarabine as single agent in children with acute lymphocytic leukemia (ALL) in second or higher relapse. There is an increasing interest in the use of clofarabine in combination regimes in less heavily pretreated children with ALL, particularly primary refractory disease. An increasing number of studies are exploring the benefit of clofarabine in this setting and will determine the real future role for this promising new agent in standard practice.
9.1.2 Pharmacology The structure of clofarabine was specifically designed to retain resistance to deamination by adenosine deaminase (ADA) (Fig. 9.1). In common with other nucleoside analogues, i.e. cladribine and fludarabine, clofarabine is transported into the cell by
P. Kearns (*) School of Cancer Sciences, University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK e-mail:
[email protected] V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3_9, © Springer Science+Business Media, LLC 2011
167
168
P. Kearns and V. Saha
Fig. 9.1 Nucleoside analogues have been in use in leukemia chemotherapy for many decades. They include the pyrimidine analogue, cytarabine and the purine analogues, thiopurines and fludarabine. The chemical structures of the drugs are shown in Figure 9.1. In general the compounds are prodrugs which enter the cell. The metabolites are phosphorylated and the triphosphate nucleotides responsible for their activity are incorporated into nucleic acid synthesis during replication, repair or transcription.
active nucleoside transporters and subsequently undergoes intracellular sequential phosphorylation by deoxycytidine kinase (dCK) to the triphosphate moiety. In vitro studies demonstrate that the phosphorylated form inhibits DNA polymerases and ribonucleotide reductase, depleting deoxynucleotide triphosphate pools and inhibiting DNA synthesis, thereby resulting in loss of cell viability [1–5]. In addition, a direct effect of clofarabine triphosphate on the integrity of mitochondria has been demonstrated in primary chronic lymphocytic leukemia (CLL) cells, resulting in the release of the pro-apoptotic mitochondrial proteins cytochrome c and apoptosis-inducible factor and activation of the Apaf-1-mediated response [6]. Subsequently,
9 Nucleoside Analogues
169
in vitro and in vivo studies using the T-ALL cell line CCRF-CEM demonstrated clofarabine-induced apoptosis correlated with a decrease in the anti-apoptotic proteins Bcl-XL and MCL-l, dephosphorylation of Akt and its downstream effectors Bad and FKHRL1 and an increase in the population of cells in G1/S and early S phase [7]. There is evidence to suggest a further potential mechanism of clofarabine activity via inhibition of DNA repair. DNA repair responses induced following exposure to alkylating agents can be reduced by pretreatment with clofarabine in a dose-dependent manner and correlate with the accumulation of intracellular clofarabine triphosphate [8]. Interestingly, a significant radiosensitising effect was shown in tumour cell lines pretreated with clofarabine. An early response to the induction of DNA doublestrand breaks is the phosphorylation of the histone variant H2AX at the break site. Phosphorylated H2AX (g-H2AX) form foci at DNA double-strand breaks induced by ionising radiation [9]. In in vitro and in vivo studies, low doses of clofarabine were shown to prolong the presence of g-H2AX foci and reduced clonogenic survival following irradiation. At higher doses, clofarabine increased the DNA doublestrand breaks, suggesting interference with DNA-damage repair mechanisms [10]. A better understanding of the multiple mechanisms of action of clofarabine will provide essential information in both planning effective combination therapies for clinical development and minimising potentially harmful treatment interactions.
9.1.3 Pharmacokinetics Animal models show that clofarabine is extensively distributed following intravenous administration, with a steady-state volume of distribution (Vdss) of approximately 1.4–2.6 L/kg in mice, 3.3–3.6 L/kg in rats and 0.9–1.2 L/kg in dogs [11]. Clofarabine displays non-linear pharmacokinetics with three phases of clearance and half-lives of 0.3, 1.3 and 12.8 h. Notably, it is predominantly renally excreted with 80% recovery in the urine and 10% in faeces. Rodent studies showed that clofarabine is rapidly absorbed with a mean absorption time of less than 2 h and a 57.5% bioavailability. Over 80% of clofarabine was recovered in the plasma with around 13% binding to plasma proteins [12]. The pharmacokinetics and pharmacodynamics of clofarabine has been examined in the course of several early-phase clinical trials [13–16]. Gandhi et al. reported plasma pharmacology data from 25 patients participating in phase I studies who received clofarabine doses ranging from 4 to 55 mg/m2/day for 5 days as a 1-h intravenous infusion. The results indicated a linear dose-relationship with plasma clofarabine concentration, which has been supported by pharmacokinetic studies in subsequent early-phase clinical trials. A wide interpatient variability in the end of infusion peak levels of clofarabine is observed. In addition, intracellular clofarabine triphosphate levels measured at the end of the first clofarabine infusion suggested dose proportionality but a wide interpatient variability in the concentrations of clofarabine triphosphate. At the maximum tolerated dose, the clofarabine trip’hosphate concentration was a median 19 mM (range, 3–52 mM). As would have
170
P. Kearns and V. Saha
been predicted from preclinical data, the clofarabine triphosphate concentration was high compared to the endogenous level of dATP, resulting in incorporation of the analogue into DNA. The accumulation of triphosphate was associated with decrease in DNA synthesis. In phase II studies, the relationship between the pharmacokinetics of clofarabine triphosphate accumulation and clinical response at the MTD was explored, revealing an accumulation advantage of the cytotoxic triphosphate in leukemia cells of responders [13, 15]. Population pharmacokinetics of clofarabine and its metabolite 6-ketoclofarabine have been modelled in adult and paediatric patients with haematological malignancies or solid tumours. Clofarabine pharmacokinetics were influenced by age, weight and estimated creatinine clearance, but not gender, race or disease type. Notably, a difference in clofarabine pharmacokinetics was observed between adults and children. In the paediatric population, the estimated creatinine clearance was considerably higher, the total systemic clearance and volume of distribution at steady-state lower and the b half-life shorter. This suggests that adults have higher exposure than children given a similar dose standardised to body surface area. Renal function was important to clofarabine clearance. As the estimated creatinine clearance decreased, exposure increased due to reduced total systemic clearance. It is suggested that dose reduction would be required in patients with moderate or severe estimated creatinine clearance (30–60 mL/min/1.73 m2 or <30 mL/ min/1.73 m2 respectively) [14, 16]. The utility of clofarabine in central nervous system (CNS) disease has been a point of debate. The pharmacokinetics and cerebrospinal fluid (CSF) penetration of intravenous clofarabine was explored in a non-human primate model and demonstrated only modest levels in the CSF. The median CSF penetration was 5% (range, 3–26%); however, the concentration levels were similar to cytotoxic concentrations in vitro, and therefore, further investigation of the potential use of clofarabine in the context of CNS disease would be of interest [17].
9.1.4 Clinical Trials Clofarabine’s clinical development was initiated in 1998 by the M.D. Anderson Centre, University of Texas. The first early phase studies in haematological malignancies administered clofarabine intravenously over 1 h for 5 consecutive days. Thirty-two patients were recruited to the phase I study for acute leukemia. The dose-limiting toxicity was hepatic, with two of four patients experiencing reversible grade 3 hepatotoxicity at the 55 mg/m2/day dose level. The recommended dose taken forward for phase II studies in adult acute leukemias was 40 mg/m2/ day. Of note, this was considerably higher than the recommended dose in adult solid tumour studies (2 mg/m2/day) where the dose-limiting toxicity was myelosuppression [18]. There was promising evidence of anti-leukemia activity in these early studies but no evidence of activity in solid tumours. The adult phase I study for patients
9 Nucleoside Analogues
171
with acute leukemia showed an overall response rate of 16% (two complete responses (CR), one acute myeloid leukemia (AML) and one ALL) and three patients with complete response without platelet recovery (CRp) [18]. This was sufficient activity to promote further development of clofarabine in phase II studies in adults. Kantarjian et al. reported the results of their phase II study using the adult recommended dose of 40 mg/m2/day intravenously over 1 h for 5 days, every 3–6 weeks. Sixty-two patients were recruited with relapsed and refractory acute myeloid leukemia (AML; n = 31), myelodysplastic syndrome (MDS; n = 8), chronic myeloid leukemia in blastic phase (CMLBP; n = 11) and acute lymphocytic leukemia (ALL; n = 12). The overall response rate was 48%, including 20 patients (32%) who achieved CR, 1 patient with a partial response (PR) and 9 (15%) achieving a CRp. Of note, only 2 (17%) of 12 with ALL achieved a CR or CRp. The toxicity profile was similar to the phase I studies, with reversible liver dysfunction in 15–25%. Liver transaminases, and in some cases bilirubin levels, generally increased at day 5 and resolved around day 10–15. Other non-haematological toxicities observed included skin rashes, erythrodysesthesia and mucositis [19]. There was an interesting correlation observed between the accumulation of clofarabine triphosphate in blasts and response. The median intracellular triphosphate level in responders was 18 mM compared with 10 mM in non-responders. There was a 1.8fold increase following the second clofarabine infusion in responders only [18]. In the paediatric phase 1 study for acute leukemias, the starting dose of clofarabine was 11.25 mg/m2/day. Clofarabine was infused intravenously over 1 h each day for 5 days. Six dose levels between 11.25 and 70 mg/m2/day were studied in 25 patients, 17 with ALL and 8 with AML. The maximum tolerated dose was at the 52 mg/m2 dose level. The dose-limiting toxicities occurred at 70 mg/m2 and were consistent with the adult studies, including reversible hepatotoxicity. In addition, skin rash was reported as a dose-limiting toxicity. It was noted in the paediatric study that some patients experienced irritability during the infusion that was mitigated by increasing the infusion time from 1 to 2 h. The recommended schedule for paediatric phase II studies was, therefore, 52 mg/m2/day for 5 days administered as a 2-h intravenous infusion. There was evidence of responses in the phase I study, with five patients achieving CR (four ALL and one AML) and three PR, giving an overall response rate of 32%. As in the adult study, end of infusion intracellular clofarabine triphosphate levels were measured, where possible. The levels in leukemia blasts varied between 6 and 19 mM and were associated with complete and sustained inhibition of DNA synthesis [20]. The recommended schedule (52 mg/m2/day for 5 days administered as a 2 h intravenous infusion every 2–6 weeks) was taken forward in two phase II, open-label, multicentre studies, one in USA [21] and one in Europe [22]. Both studies recruited paediatric patients with refractory or relapsed ALL and reported very similar results. The US study (CLO-212) recruited 61 paediatric patients, median age was 12 years (range, 1–20 years) and the European study (BIOV-111) recruited 74 patients, 65 evaluable for response, median age was 10 years (range 0.5–23 years). The overall response rates were 30 and 26%, respectively. Notably, the US study responders included six partial responses (PR) compared to one PR in the European study. This relative resistance may have been a reflection of the slightly
172
P. Kearns and V. Saha
more heavily pretreated population in the US study (median 2 (range 1–5) vs. 3 (range 2–6) regimens). The toxicity profile was similar in both studies and consistent with adult data. The most common adverse events included febrile neutropenia and reversible hepatotoxicity. There were durable remissions in responders with patients proceeding to hematopoietic stem-cell transplantation after clofarabine [21, 22]. The early-phase studies confirm clofarabine as active as a single agent in paediatric patients with multiple relapsed or refractory ALL. These data contributed to the marketing approval for clofarabine in USA and Europe. The marketing indication clearly reflects the patient population participating in the phase I and II studies; however, this is not the area of greatest clinical need for children with ALL and further studies exploring the use of clofarabine in the context of multi-agent chemotherapy protocols in less heavily pretreated patient populations are needed. Several early-phase trials combining clofarabine with cytarabine or cyclophosphamide and etoposide have been reported. As an inhibitor of ribonucleotide reductase, it has been proposed that clofarabine could modulate cytarabine triphosphate (ara-CTP) accumulation, thereby increasing the anti-leukemic activity of cytarabine. In vitro studies using a myeloid leukemia cell line demonstrated that clofarabine added prior to cytarabine increased ara-CTP accumulation. Exposure of the cells to clofarabine followed 4 h later by cytarabine resulted in accumulation of 248 mM ara-CTP at 3 h compared to 86 mM with cytarabine alone. There was also depletion in intracellular dATP and dGTP levels following clofarabine exposure. Clonogenic assays revealed synergistic cytotoxicity when clofarabine and cytarabine were combined in this schedule [23]. The result of clinical studies combining clofarabine and cytarabine also supported the potential enhanced therapeutic effect. In 32 adult patients with relapsed acute leukemia (25 AML, 2 ALL, 4 high-risk MDS and 1 CMLBP), clofarabine 40 mg/m2/ day was given as a 1-h intravenous infusion for 5 days (days 2 through 6) and cytarabine 1 g/m2/day as a 2-h intravenous infusion for 5 days (days 1 through 5). On days 2–5, the cytarabine infusion was commenced 4 h after the clofarabine infusion. An overall response rate of 38% was reported (seven CR and five CRp). No responses occurred in the three patients with ALL and CML. The combination regime was well tolerated with predominantly grade 2 or less adverse events, including transient liver enzymes abnormalities, nausea/vomiting, diarrhoea, skin rashes, mucositis, and palmoplantar erythrodysesthesias. The associated pharmacology studies confirmed that plasma clofarabine levels were associated with clofarabine triphosphate accumulation, which resulted in an increase in ara-CTP in the leukemic blasts [24]. Cyclophosphamide is an alkylating agent that is frequently used in the treatment of several malignancies including leukemias. Cyclophosphamide induces DNA damage via interstrand cross-link formations. DNA repair mechanisms are simultaneously activated potentially reducing the cytotoxic effect and is, therefore, a potential mechanism of resistance to cyclophosphamide. In vitro studies combining clofarabine with cyclophosphamide in primary leukemic cells have demonstrated synergistic cytotoxicity and inhibition of DNA repair [8]. Based on this potentially therapeutically advantageous interaction, Karp et al. studied the effect of combining clofarabine and cyclophosphamide in patients with relapsed acute leukemias. Eighteen patients were treated with cyclophosphamide 200 mg/m2 on day 1 and cyclophosphamide with
9 Nucleoside Analogues
173
clofarabine on day 2. Two dose levels of clofarabine were evaluated 20 mg/m2 for 6 patients and a dose de-escalation to 10 mg/m2 for a further 12 patients. The overall response rates were 50% (two CR and one PR) at the clofarabine dose 20 mg/m2 and 30% (three CR and one PR) at the clofarabine dose 10 mg/m2. Prolonged marrow aplasia was the dose-limiting toxicity at clofarabine dose 20 mg/m2. Of the six patients treated at the clofarabine dose 20 mg/m2, four experienced marrow aplasia for longer than 42 days. At the lower clofarabine dose 10 mg/m2, prolonged marrow aplasia was seen in 3 of 12 patients treated, 2 recovering with leukemia at days 55 and 58. In 13 patients, peripheral blood blasts were available for analysis. DNA damage was measured by flow cytometric measurement of g-H2AX. In 12 of 13 (92%) of patient samples, a significant elevation of g-H2AX was observed on day 2 following treatment of clofarabine and cyclophosphamide compared with cyclophosphamide alone on day 1, supporting the argument for a clofarabine-augmented, cyclophosphamide-mediated DNA damage effect [25]. The effect was observed in both AML and ALL blasts and warrants further exploration. In the paediatric age group, the combination of clofarabine with etoposide and cyclophosphamide in ALL has been explored. In a Phase I study of clofarabine with etoposide and cyclophosphamide for children with relapsed/refractory acute leukemias, all three drugs were administered for 5 consecutive days in induction and 4 consecutive days in consolidation, for a maximum of eight cycles. Twenty-five patients (20 ALL and 5 AML) were enrolled in five cohorts. The maximum tolerated dose was not reached in the phase I study and the recommended doses to be taken forward in phase II studies of clofarabine, cyclophosphamide and etoposide were 40, 440 and 100 mg/m2/day, respectively. The overall response rate was 64%, including CR in 9/20 ALL and 1/5 AML and CRp in six patients (2/20 ALL and 4/5 AML). Nine of the sixteen responding patients proceeded to hematopoietic stem cell transplantation. The study has now expanded to recruit the phase II cohort evaluating the combination in ALL patients only. An unexpected level of hepatotoxicity has caused some concern in this study [26]. Four of eight patients developed severe hepatotoxicity (three veno-occlusive disease and one hyperbilirubinaemia). Potential contributory factors included severe concurrent infection in three cases and previous haematopoietic stem cell transplant within 12 months of entry to the study in three cases. Nevertheless, a correlation with the drug combination cannot be completely excluded. As a consequence, the entry criteria to the ongoing study were amended to exclude patients with prior haematopoietic stem cell transplant or a history of viral hepatitis, cirrhosis or hyperbilirubinaemia [26]. A further study evaluating this combination in paediatric patients with advanced ALL was reported by Locatelli et al. The study enrolled 25 paediatric patients with either refractory or multiple relapsed ALL. The drug dose schedule differed slightly from the Hijiya et al. schedule above. Patients received clofarabine 40 mg/m2, cyclophosphamide 400 mg/m2 and etoposide 150 mg/m2 daily for 5 consecutive days. Similar responses were reported with an overall remission rate of 56% (13 CR and 1 CRp). In 7 of the 13 (54%) patients achieving CR, remissions were of sufficient duration to allow patients to receive allogeneic haematopoietic stem cell transplantation. The most common adverse events were febrile neutropenia, mucositis and reversible liver toxicity; no case of liver veno-occlusive disease was
174
P. Kearns and V. Saha
reported [27]. The two reported studies suggest that the combination regimen of clofarabine, cyclophosphamide and etoposide is reasonably tolerated in the paediatric population with some caution regarding factors that might exacerbate the risk of hepatotoxicity. The combination can induce durable clinical response in a relevant proportion of children with refractory, multiple relapsed ALL. The specific contribution of clofarabine to the observed therapeutic efficacy in these combination studies needs to be considered. In future studies, it will be important to evaluate this in a randomised setting. Equally, the evaluation of the pharmacology of drug interactions is essential to help elucidate both the mechanisms of enhanced anti-leukemia activity, as well as to understand the additional toxicities associated with combination therapies. Ideally, these studies should be conducted in less heavily pretreated patient populations.
9.2 Nelarabine 9.2.1 Background Nelarabine is an arabinosyl nucleoside analogue, similar to fludarabine (2-fluoroara-A) and cytarabine (ara-C). Ara-G was first synthesised in 1964 [28] Ara-G’s limited solubility restricted its clinical application. Renewed interest in its therapeutic potential came in the 1970s from studies carried out in patients with purine nucleoside phosphorylase (PNP) deficiency. PNP catabolises deoxyguanosine (dGuo), leading to increased dGTP levels. This is associated with a profound deficiency of T-cells, but not B or myeloid cells. Models of PNP deficiency showed that ara-G was specifically toxic to T-lymphocytes and T-lymphoblastoid cells. B-cells not only were less sensitive to ara-G but also did not accumulate high levels of GTP [29]. This suggested that ara-G could be useful in the treatment of T-cell malignancies. As enzymatic synthesis of purine arabinonucleosides became more straightforward, the purine analogue 2-amino-6-methoxypurine was found to be eight times more soluble than ara-G [30] and is among the many compounds derived by Gertrude Elion for which she received the 1988 Nobel Prize in medicine [31]. This compound, nelarabine [32], is demethoxylated by ADA to ara-G in the body.
9.2.2 Pharmacology Similar to other arabinosyl derivatives, ara-G is S-phase-specific and enters the cell via nucleoside transporters [33]. Once within the cell, ara-G is phosphorylated either by dCK or dGuo kinase to ara-GMP, which is the rate-limiting step [34]. The final compound ara-GTP is actively taken up by DNA strands during the S phase of the cell cycle. DNA elongation is terminated and cell death by apoptosis is the
9 Nucleoside Analogues
175
end result [35]. Among other possible mechanisms for ara-GTP accumulation is impaired function of ribonucleotide reductase due to an imbalance in deoxynucleoside biphosphate pools, leading to impaired DNA synthesis [36].
9.2.3 Pharmacokinetics When nelarabine is administered intravenously, Cmax is reached near the end of the infusion [37]. Both Cmax and AUC are dose proportional. Ninety-four per cent of nelarabine is converted to ara-G within 1 h, though there is a wide patient variability thought to be related to polymorphisms in ADA. Neither age nor gender appears to influence the pharmacokinetics of nelarabine. The elimination half-life of ara-G is however considerably shorter in children (2.1 h) than in adults (3 h). As ara-G is partially excreted in the urine, this difference in t½ has been attributed to diminished creatinine clearance in adult patients [37]. There is a dose-dependent linear increase in intracellular levels of ara-G in both T and B lymphoblasts. However, the levels of ara-GTP attained in T-lymphoblasts are significantly higher than those achieved in other cells. Furthermore, high levels of intracellular ara-GTP also positively correlated with a clinical response to nelarabine. Thus, not only is ara-G 15–250 times more toxic but it also accumulates more rapidly in T-lymphoblasts.
9.2.4 Clinical Trials Phase I Though reported at times as two separate trials, a single limited centre phase I trial was carried out in 93 patients [38]. Nelarabine was administered intravenously over 0.75–2 h daily for 5 days. Dose escalation from 5 to 90 mg/kg/day was carried out. The data obtained from this study provided the pharmacokinetic data presented previously. Toxicity was observed at 75 mg/kg/day. Subsequently, the milligram/kilogram dosing regimen was converted to dose per square metre of body surface area. A dose of 1.2 g/m²/day was chosen as considered to be equivalent to 30 mg/kg/day for an adult and 40 mg/kg/day for a child. There were 39 patients, 26 children and 13 adults, with T-ALL. There were 9 CR and 12 PR in this group or an overall response of 54%. Efficacy was noted at all dose ranges tested.
9.2.5 Phase II The Children’s Oncology Group (COG) conducted a phase II study of nelarabine in relapsed T-cell disease [39]. The age of patients ranged from 0.6 to 21.7 years. The dose of 1.2 g/m²/day for 5 days defined in the phase I trial proved to be too toxic.
176
P. Kearns and V. Saha
Subsequently, the dose was reduced to 650 mg/m²/day and then 400 mg/m²/day. Of the 136 children with relapsed T-cell disease that could be evaluated in this trial, there were 35 CR and 10 PR. Thus, the overall response rate was 33%. Among the 33 children with a first relapse of T-ALL, there were 16 CR and 2 PR or an overall response rate of 55%. A subsequent adult trial was performed in 39 patients with T-cell disease [40]. These patients received an infusion of nelarabine, at 1.5 g/m²/day on days 1, 3, 5, and were eligible for two additional courses if they achieved CR. Ten (26%) achieved CR, while a response was seen in 16 (41%). There are two more reports of phase II studies in adults. Using a similar dosing schedule, Thompson et al. [41] obtained a response in four of nine T-cell patients, of them two achieved CR. The GMALL group, also using the 1.5 g/m²/ day on alternate days, obtained a CR rate of 76% in 47 patients with T-ALL [42].
9.2.6 Other Trials Fludarabine, another purine analogue, also inhibits ribonucleoside reductase. Given prior to nelarabine, it could potentially increase the intracellular accumulation of ara-GTP. This combination was given to a mixed group of 13 patients, in them 9 had indolent disease [43]. There were seven responders. In a subsequent study, 35 patients with indolent leukemia, of whom 70% had fludarabine refractory disease, received nelarabine daily, nelarabine on alternate days or nelarabine along with fludarabine [44]. The responses were 20, 15 and 63%, respectively. Although the administration of fludarabine was expected to increase the intracellular GTP, there was no significant difference in overall AUC before and after fludarabine. Nor did fludarabine affect the linear accumulation and elimination of ara-G. However, as reported earlier, the median peak intracellular ara-GTP level was significantly higher in responders. This suggests that the combination of fludarabine and nelarabine may indeed be effective in refractory or indolent leukemias. The activity of nelarabine in other T-cell lymphomas has also been assessed. Nineteen adult patients (median age 58 years) with cutaneous or peripheral T-cell lymphoma were treated with 1.5 g/m2/day on days 1, 3 and 5 [45]. Considerable neurotoxicity was observed, and there were only two partial remissions.
9.2.7 Toxicities The main dose-limiting toxicity of nelarabine is neurological. This appears to be dose-dependent. The mechanism for this is unknown [46]. Both nelarabine and ara-G have excellent CSF penetration, and toxicity is either related to accumulation of ara-GTP and/or its effects on signal transmission. Central neurotoxicity is most often apparent as somnolence, dizziness, confusion and ataxia. Peripheral neurotoxicity is associated with paraesthesias, weakness and pain in the extremity [39, 47]. Neurotoxicity is more common in adults and those with prior toxicity to vincristine. There is little haematological toxicity. In the COG study, of 29 patients
9 Nucleoside Analogues
177
with <25% blasts in the bone marrow, two had grade 4 neutropenia and two grade 4 thrombocytopenia. Though other toxicities have been described, these are rare and difficult to separate as distinct from the disease process.
9.2.8 Discussion On current modern ALL protocols, over 90% of children with T-ALL are being cured. The prognosis for those who relapse is, however, dismal [48]. Many patients, who relapse, do not achieve CR2. Often in those that do, remissions are shallow and short-lived. As presented here, nelarabine is clearly an effective agent against T-cell disease and a step forward in the right direction. Though the response rates seen with nelarabine as a single agent are encouraging, remissions achieved have been short-lived. Multi-agent chemotherapy is the key to cure in childhood ALL and the question is what drugs are best used along with nelarabine and during which stage of therapy. Clearly, the response rate is linked to intracellular ara-GTP levels, and there is evidence to suggest increased efficacy when used along with fludarabine. The COG have initiated a study of nelarabine along with augmented BFM chemotherapy in intermediate risk T-cell ALL. Another approach being considered by the I-BFM group is to incorporate nelarabine along with the main relapse protocol for T-cell disease. Thus, to maximise our understanding of nelarabine, we need to evaluate its efficacy in the context of multidrug therapy, as well as the outcome of patients who have been transplanted after receiving the drug. Ideally, then the drug can be used in patients with T-cell disease to prevent relapses. Though nelarabine is active against other leukemia cells, at the moment the focus is very much on the treatment of T-cell leukemia and NHL.
9.3 Forodesine 9.3.1 Background Forodesine was designed to be a potent transition-state inhibitor of the enzyme PNP. The concept that this pathway might be an effective target in the treatment of leukemias arose from the study of a rare clinical condition of PNP deficiency. Children with PNP deficiency present with a severe form of combined immunodeficiency and have impaired T-cells but apparently normal B-cell function [49]. It was, therefore, postulated that selective inhibition of PNP could target T-cells and may prove a useful therapeutic strategy for the treatment of T-cell leukemia and lymphoma. Early clinical studies of the selective PNP inhibitor forodesine have supported this concept, and its use in acute lymphoblastic leukemia is under clinical investigation.
178
P. Kearns and V. Saha
9.3.2 Pharmacology PNP is a purine-metabolising enzyme that catalyses the phosphorolysis of purine nucleosides to their base equivalent. Under normal physiologic conditions, PNP catalyses the phosphorolysis of 2¢ dGuo to the guanine nucleobase and 2¢ deoxyribose-1-phosphate. When PNP is inhibited or deficient, dCK converts unmetabolized dGuo into 2¢ deoxyguanosine triphosphate (dGTP). Consequently, dGTP accumulates and blocks DNA synthesis through disrupting the equilibrium of the deoxynucleotide pool, leading to apoptosis. Near-complete inhibition of PNP is required for this to occur. In patients with PNP deficiency, high plasma levels of dGuo are observed [50]. Interestingly, although PNP is present in all mammalian cells, it is the T-cells that appear to be most influenced by the deficiency. The genetic deficiency of PNP is associated with a gradual specific loss of T-cell function after birth, but DNA synthesis in other lymphoid and non-lymphoid cells is spared [51]. This specificity was thought to be due to inherently high levels of dGuo phosphorylation and slow catabolism of dGTP in T-cells. Attempts to pharmacologically increase plasma dGuo by intravenous infusions were unsuccessful due its rapid degradation by PNP, present in major organs including liver, spleen and kidney, in circulating lymphocytes and erythrocytes. The inhibition of PNP became the principal target for therapeutic development. There are several known drugs with PNP inhibitory activity (including acyclovir, 6-mercaptopurine and allopurinol), and over 30 potential inhibitors of PNP were patented by 1998; however, none were successful therapeutically due to inadequate levels of PNP inhibition at clinically achievable concentrations. In 1995, the transition state structure of PNP was resolved, allowing the possibility of targeting of the transition state of the enzyme. Schramm et al. [52] used this knowledge to develop immucillin-H or BCX-1777 (now known as forodesine) as an analogue of the PNP transition-state substrate. Forodesine is now the most potent PNP inhibitor in clinical development. In vitro studies show forodesine is imported into the cell via the equilibrative nucleoside transporters (ENT1 and ENT2), whereas dGuo is taken up via concentrative nucleoside transporters (CNTs) [53]. This rationally designed, transition-state analogue inhibitor demonstrated effective inhibition of PNP at picomolar concentrations, resulting in elevated dGuo levels in plasma and increased dGTP concentrations in leukemia cells of different lineages [54–56]. Kicska et al. reported inhibition of in vitro growth of malignant T cell lines in the presence of dGuo without affecting a B-cell leukemia line or a colon carcinoma cell line. Activated human peripheral blood T lymphocytes were also sensitive to inhibition by forodesine. Forodesine showed an anti-proliferative effect on T lymphocytes stimulated in vitro with anti-CD3 antibody and interleukin-2 but no effect on unstimulated cells. Cytotoxicity in leukemia cells was associated with accumulation of intracellular dGTP and the subsequent failure to synthesise cellular DNA. A correlation between the degree of T-cell inhibition and the level of dCK activity was observed [54].
9 Nucleoside Analogues
179
The activity of forodesine has been evaluated in a number of animal models. Forodesine is an effective inhibitor of PNP from various species including human, mouse, rat, monkey and dog, with IC50s in the range of 0.48 to 1.57 nM. It was shown to be a 10–100-fold more potent inhibitor of human lymphocyte proliferation than other known PNP inhibitors. Forodesine was effective in elevating dGuo in mice similar to the levels observed in PNP-deficient patients [55]. Primate studies were undertaken to establish an appropriate starting dose for first-in-man studies, based not only on safety and pharmacokinetic parameters but also on pharmacodynamic end points, specifically plasma concentration of dGou [57]. Intravenous and oral administration of forodesine induced a rapid elevation of plasma dGuo up to 1.5 mM. The plasma level of dGuo level did not increase with increased intravenous doses, but the duration of elevation of dGou was increased. There was poor oral bioavailability in primates (8.2%), but in contrast to the intravenous dose response, increasing the oral dose of forodesine did increase dGuo accumulation. The reason for the difference is not certain. The results suggested that oral forodesine was absorbed slowly producing a sustained low concentration inducing prolonged increased plasma concentrations of dGuo. Both oral and intravenous administration induced a rapid rise in dGuo. Oral dosing at 4.4 mg/kg induced an equivalent increase in plasma dGou levels as an intravenous dose of 0.88 mg/kg, but the oral route produced a more prolonged accumulation [57]. Considerable variability in bioavailability and dGuo response to forodesine has been observed [58]. In addition, dGuo accumulation is to some extent determined by the source of dGuo. As dGuo is derived from degradation of DNA as a result of cell turnover, interindividual variability, disease state and even patient age may impact on the level of dGuo accumulation following PNP inhibition. These factors may need to be considered in the interpretation of results from clinical studies. In spite of the initial focus on T-cell diseases, recent studies have suggested that the cytotoxic effect of forodesine may extend to B-cell lineage leukemias. Primary B-Cell Chronic Lymphocytic Leukemia (B-CLL) cells, incubated with forodesine and dGuo demonstrated intracellular accumulation of dGTP without alteration in levels of other deoxynucleotides [59]. The accumulation of dGTP correlated with induction of apoptosis measured by caspase activation, changes in mitochondrial membrane potential and PARP cleavage. High expression levels in ZAP-70, CD38 or deletions of p53 or ATM are known to be associated with treatment resistance in B-CLL. Forodesine can effectively induce apoptosis in primary CLL irrespective of the presence of these poor prognosis genetic markers. An associated downregulation of MCL-1 and upregulation of BIM and P73 are observed. These data support forodesine acting through activation of the mitochondrial apoptosis pathway and bypassing the DNA damage/P53/ATM pathway [60]. The picomolar potency for PNP inhibition correlating with accumulation of dGuo and cytotoxic activity in vitro and in animal studies has together provided the rationale for the clinical investigation of forodesine in T-cell malignancies.
180
P. Kearns and V. Saha
The in vitro results in B-CLL have also stimulated interest in evaluating its potential use in the treatment of B-lineage haematological malignancies.
9.3.3 Early-Phase Clinical Trials Several clinical trials are ongoing evaluating the dosing schedule and toxicity profile of forodesine in haematological malignancies with early reports emerging at the ASCO and ASH Annual Meetings. Gandhi et al. [61] published the findings of the first proof of principle trial of forodesine in 2005. The trial evaluated forodesine in patients with advanced T-cell malignancies and examined the pharmacokinetics and pharmacodynamic end points. Five patients were treated with a 30-min intravenous infusion of forodesine 40 mg/m2 as a single dose on day 1 and every 12 h on days 2–5. Three patients had T-cell prolymphocytic leukemia (T PLL) and two T-ALL. A median peak level of forodesine of 5.4 mM was reached at the end of infusion, which would be sufficient to achieve effective PNP inhibition. This was substantiated by the observed increase in plasma dGuo concentrations in all patients. The twice-daily dosing schedule was selected based on the primate pharmacokinetic data that showed a term elimination rate (t½) of 3 h; however, a longer median t½ was measured in the clinical trial (median 10 h), suggesting a once daily schedule may be sufficient. Intracellular dGTP accumulated in four of five patients, by 10- to 60-fold and correlated with evidence of anti-leukemia activity in these four patients, but no objective responses were observed. In one patient, intrapatient dose escalations to 60 and 90 mg/m2 were achieved. Although dGuo levels increased accordingly, there was no associated increase in dGTP accumulation. The level of anti-leukemia activity observed included minor reduction in peripheral lymphadenopathy and reduction in peripheral white cell count and peripheral blast clearance. Forodesine was well tolerated in this study. The most common toxicity observed was grade 3–4 neutropenia. A maximum tolerated dose was not established because the trial was discontinued due to the low level of efficacy, and it was thought that efficacy may be improved by optimisation of the schedule of administration. The trial did confirm that forodesine could inhibit PNP in patients with T-cell disease and induce the predicted phamacodynamic effects [61]. It is postulated that the accumulation of dGTP following PNP inhibition is dependent on the level of dCK activity. B-CLL blasts have intrinsically high levels of dCK activity and, therefore, may be sensitive to PNP inhibition [62]. Preclinical studies have also shown effective forodesine-induced cytotoxicity in B-CLL cells, clinical trials evaluating forodesine in CLL patients with primary resistance to fludarabine-based therapy or with progressive disease are on-going. In a phase II trial of forodesine in B-CLL, eight patients received oral forodesine (200 mg/day) for up to 24 weeks with only mild side effects. Participating patients had a median lymphocyte count of 35.9 × 109/L and a median serum b2
9 Nucleoside Analogues
181
microglobulin of 6.45 mg/L. Six had Rai stage III–IV and were previously heavily treated. There was some evidence of transient anti-leukemia activity but no objective responses. This may be attributed to not achieving the target plasma level of forodesine (2 mM) in any patient. The PNP inhibition ranged from 57 to 89%, and the intracellular dGTP increase was disappointingly low (mean increase from 6 to 10 mM). In vitro studies with 2 mM forodesine resulted in accumulation of levels of dGTP in the range of 40–250 mM, which was associated with an in increase in apoptosis. It is clear that the dosing schedules need to be optimised to achieve higher steady-state levels of forodesine, allowing full evaluation of the potential efficacy of forodesine when the pharmacodynamic end points are achieved [63].
9.3.4 Conclusions The rationale and preclinical data for the therapeutic use of forodesine in haematological malignancies has been compelling; however, the results of the clinical trials to date are conflicting. The target pharmacodynamic end points as measured by plasma dGuo and intracellular dGTP are clearly achievable in the clinical setting. The ideal schedule to ensure the effects are sufficiently sustained to induce clinical responses is yet to be established. The potential role for forodesine in paediatric ALL is currently under evaluation. A European phase I trial investigating the safety and efficacy of several dosing schedules of oral and intravenous forodesine in paediatric patients with relapsed and refractory T and B lineage haematological malignancies is ongoing. It is hoped that the trial will provide guidance on the best direction of development of this interesting new agent.
9.4 Gemcitabine 9.4.1 Background Gemcitabine (2¢,2¢-difluoro-2¢-deoxycytidine) is a deoxycytidine analogue first synthesised at Eli Lilly. The hydrogens of 2-carbon of cytosine are replaced with the two fluorines in the so-called geminal configuration giving the compound its name. Originally synthesised and investigated as an antiviral agent, it was found to have activity against a broad range of solid and haematological cancer cell lines as well as xenograft models. Today, gemcitabine is used, either as a single agent or in combination, in the treatment of patients with pancreatic, non-small cell lung, ovarian and other cancers. This chapter explores the experience of using gemcitabine in haematological malignancies, particularly in acute lymphoblastic leukemias.
182
P. Kearns and V. Saha
9.4.2 Pharmacology Like most nucleoside analogues used in chemotherapy, gemcitabine is a prodrug. It is hydrophilic and is actively transported into the cell by nucleoside transporter proteins. Once inside the cell, gemcitabine is phosphorylated by dCK. Both the di and bi phosphate compounds are metabolically active. The compound is deaminated by cytidine deaminase to difluorodeoxyuridine. The latter compound is metabolically inert and excreted mainly by the kidney. Gemcitabine triphosphate is incorporated into DNA strands during replication. Unlike other nucleoside analogues where this results in chain termination, in the case of gemcitabine triphosphate another nucleotide is added before termination. As a result of this so-called “masked” chain termination, not only are DNA polymerases inhibited but proof reading by exonucleases is also impaired. 2-Carbon is the primary site of discrimination for both ribonucleotide reductase and DNA polymerase. Ribonucleotide reductase is the enzyme that catalyses the conversion of ribonucleotides to deoxynucleotides. This is the ratelimiting step in DNA synthesis. Though the biphosphates of cladribine, fludarabine and clofarabine all inhibit ribonucleotide reductase, in all cases this inhibition is reversed once the drug or its metabolites are eliminated. Gemcitabine on the contrary irreversibly inactivated ribonucleotide reductase. This results in the decrease in dCTP production and increases gemcitabine triphosphate incorporation in to DNA strands in a process termed “self-potentiation”. At low cellular concentrations (<50 mmol/L), gemcitabine triphosphate is eliminated with monophasic linear kinetic with a t½ of 2–6 h. At higher intracellular concentration, elimination becomes biphasic with a longer t½. The drug appears to be equally effective against confluent cells as well as cells in log phase growth, thus its effect is not confined to S phase and this may relate to its intracellular retention. Gemcitabine triphosphate is also incorporated into RNA and has been shown to inhibit the action of topoisomerase I by stabilising topoisomerase I cleavage complexes. How these contribute to cytotoxicity is unclear. We also do not know the downstream pathways that lead to cell death, though there is some evidence that suggest that caspase activation of apoptosis is integral to gemcitabine-induced cell death.
9.4.3 Pharmacokinetics In vitro studies carried out initially in cell lines and subsequently extended to primary cells showed that intracellular accumulation of gemcitabine triphosphate is saturated at 10–20 mmol/L of exogenous gemcitabine. Moreover, at these levels, gemcitabine appears to act as a substrate inhibitor for dCK. Thus, if gemcitabine levels rise beyond 20 mmol/L, the concentrations of intracellular gemcitabine triphosphate declines. Pharmacokinetic studies suggest that when infused intravenously at a rate of 10 mg/m2/min, then a level of 20 mmol/L of gemcitabine is obtained. This is termed as the fixed dose rate (FDR) of gemcitabine. At this infusion rate, continuous increases in intracellular gemcitabine triphosphate can be detected for up to 18 h. For the FDR delivery, generally 1,000 or 1,500 mg/m2 is
9 Nucleoside Analogues
183
infused over 100 or 150 min, respectively. In contrast, a phase I dose escalation trial showed that the MTD of gemcitabine was 2,200 mg/m2/week when given as a 30-min infusion. This is the most frequently used schedule for gemcitabine and is termed as the standard dose schedule (STD). A recent study used a crossover method to study pharmacokinetics of both methods. There was wide interpatient variability, but FDR gave higher intracellular levels. Both methods also gave rise to an increase in gemcitabine triphosphate concentrations on subsequent infusions as a result of self-potentiation. Thus, the FDR infusion rate may vary with time, tumour and individual. It is also unclear which method provides the best clinical benefit. Other factors that influence the efficacy of gemcitabine are polymorphisms in the genes involved in its metabolism including cytidine deaminase and the promoters of dCK and riboxynucleotide reductase.
9.4.4 Clinical Trials Phase I Initial phase I studies with gemcitabine were conducted using 30-min infusions given at weekly intervals for 3 weeks followed by a 1-week rest. The drug was well tolerated. The dose-limiting toxicity was found to be myelosuppression. The first phase I study in adult patients with leukemia used a FDR scheduling with doses ranging from 1,200 to 6,400 mg/m2/dose on a weekly schedule. The MTD was 4,800 mg/m2 with the DLT being bullous skin ulcerations and mucositis. Fourteen patients under 21 years of age with relapsed or refractory ALL (seven) or AML (seven) were also recruited to a similar phase I study. The MTD was 3,600 mg/m2/ week with hepatotoxicity being the DLT. One child with ALL cleared marrow disease without platelet recovery.
9.4.5 Clinical Trials Phase II The dose of 3,600 mg/m2/week infused at 10 mg/m2/min weekly for 3 weeks was then taken further into a phase II trial in children with relapsed ALL or AML. Of 30 evaluable children, only 1 of 20 patients with ALL achieved a remission. Most children experienced grade 3 or 4 haematological or hepatic toxicity. Another study recruited four ALL and two AML patients aged between 3 and 18 years. They were given gemcitabine as a 30-min infusion at a dose of 1,200 mg/ m2/weekly for 3 weeks. While hepatotoxicity was less of a problem, none of the patients showed a response. Thus, the conclusion is that gemcitabine as a single agent does not warrant further investigation in childhood ALL. Given the unique mechanisms of action of gemcitabine, it may potentiate other nucleoside analogues or topoisomerase inhibitors. A number of phase I trials have been carried out using gemcitabine in combination with mitoxantrone, irinotecan and fludarabine, primarily in AML patients. These have been well tolerated, and 50% of the patients given the mitoxantrone–gemcitabine combination showed a response.
184
P. Kearns and V. Saha
9.4.6 Discussion The clinical trials with gemcitabine highlight the problems of introducing new drugs in regimens for childhood ALL. The success rate is high with conventional therapy, and thus, a new drug must have high activity against ALL blasts. Evaluation of this is becoming more and more difficult as the only patients who can be evaluated in phase I and phase II trials are those who relapse after multimodal therapy and where we can assume that the cells are highly drug-resistant. Though we do not use singleagent therapy in childhood ALL, we believe that each agent used shows a specific cytotoxicity. Thus, our current criteria for evaluation require a new drug to show activity as a single agent and tolerable toxicity. Unfortunately, gemcitabine fails on both counts. A combination of vinorelbine, topotecan, thiotepa, dexamethasone and gemcitabine was evaluated in 28 paediatric patients with relapsed or refractory leukemia. Gemcitabine was given at the previous reported dose for the phase II study. Seven of fourteen ALL patients showed a response. In an adult population, three of eight ALL patients also showed a response on this combination. It is tenable that gemcitabine could have a role of potentiating the effect of drugs that are already in use. The issue really is that with such a combination, it is difficult to assess the contribution of gemcitabine to the response rate. The problem with all such studies is that as only a few patients are available for phase I and II studies, it is difficult to do a randomised study. Given these problems, it is unlikely that further multicentre trials of gemcitabine in childhood ALL will be carried out in future.
References 1. Xie KC, Plunkett W. Deoxynucleotide pool depletion and sustained inhibition of ribonucleotide reductase and DNA synthesis after treatment of human lymphoblastoid cells with 2-chloro-9-(2-deoxy-2-fluoro-beta-D-arabinofuranosyl) adenine. Cancer Res. 1996 Jul 1;56(13):3030–7. 2. Xie C, Plunkett W. Metabolism and actions of 2-chloro-9-(2-deoxy-2-fluoro-beta-Darabinofuranosyl)-adenine in human lymphoblastoid cells. Cancer Res. 1995 Jul 1;55(13):2847–52. 3. Lotfi K, Mansson E, Spasokoukotskaja T, Pettersson B, Liliemark J, Peterson C, et al. Biochemical pharmacology and resistance to 2-chloro-2¢-arabino-fluoro-2¢-deoxyadenosine, a novel analogue of cladribine in human leukemic cells. Clin Cancer Res. 1999 Sep;5(9):2438–44. 4. Parker WB, Shaddix SC, Rose LM, Shewach DS, Hertel LW, Secrist JA, III, et al. Comparison of the mechanism of cytotoxicity of 2-chloro-9-(2-deoxy-2- fluoro-beta-D-arabinofuranosyl) adenine, 2-chloro-9-(2-deoxy-2-fluoro- beta-D-ribofuranosyl)adenine, and 2-chloro-9-(2deoxy-2,2-difluoro- beta-D-ribofuranosyl)adenine in CEM cells. Mol Pharmacol. 1999 Mar;55(3):515–20. 5. Zhenchuk A, Lotfi K, Juliusson G, Albertioni F. Mechanisms of anti-cancer action and pharmacology of clofarabine. Biochem Pharmacol. 2009 Dec 1;78(11):1351–9. 6. Genini D, Adachi S, Chao Q, Rose DW, Carrera CJ, Cottam HB, et al. Deoxyadenosine analogs induce programmed cell death in chronic lymphocytic leukemia cells by damaging the DNA and by directly affecting the mitochondria. Blood. 2000 Nov 15;96(10):3537–43. 7. Takahashi T, Shimizu M, Akinaga S. Mechanisms of the apoptotic activity of Cl-F-araA in a human T-ALL cell line, CCRF-CEM. Cancer Chemother Pharmacol. 2002 Sep;50(3):193–201.
9 Nucleoside Analogues
185
8. Yamauchi T, Nowak BJ, Keating MJ, Plunkett W. DNA repair initiated in chronic lymphocytic leukemia lymphocytes by 4-hydroperoxycyclophosphamide is inhibited by fludarabine and clofarabine. Clin Cancer Res. 2001 Nov;7(11):3580–9. 9. Rogakou EP, Pilch DR, Orr AH, Ivanova VS, Bonner WM. DNA double-stranded breaks induce histone H2AX phosphorylation on serine 139. J Biol Chem. 1998 Mar 6;273(10):5858–68. 10. Cariveau MJ, Stackhouse M, Cui XL, Tiwari K, Waud W, Secrist JA, III, et al. Clofarabine acts as radiosensitizer in vitro and in vivo by interfering with DNA damage response. Int J Radiat Oncol Biol Phys. 2008 Jan 1;70(1):213–20. 11. Bonate PL, Arthaud L, Cantrell WR, Jr., Stephenson K, Secrist JA, III, Weitman S. Discovery and development of clofarabine: a nucleoside analogue for treating cancer. Nat Rev Drug Discov. 2006 Oct;5(10):855–63. 12. Qian M, Wang X, Shanmuganathan K, Chu CK, Gallo JM. Pharmacokinetics of the anticancer agent 2-chloro-9-(2-deoxy-2-fluoro-beta-D-arabinofuranosyl)adenine in rats. Cancer Chemother Pharmacol. 1994;33(6):484–8. 13. Gandhi V, Kantarjian H, Faderl S, Bonate P, Du M, Ayres M, et al. Pharmacokinetics and pharmacodynamics of plasma clofarabine and cellular clofarabine triphosphate in patients with acute leukemias. Clin Cancer Res. 2003 Dec 15;9(17):6335–42. 14. Bonate PL, Craig A, Gaynon P, Gandhi V, Jeha S, Kadota R, et al. Population pharmacokinetics of clofarabine, a second-generation nucleoside analog, in pediatric patients with acute leukemia. J Clin Pharmacol. 2004 Nov;44(11):1309–22. 15. Cooper T, Kantarjian H, Plunkett W, Gandhi V. Clofarabine in adult acute leukemias: clinical success and pharmacokinetics. Nucleosides Nucleotides Nucleic Acids. 2004 Oct;23(8-9):1417–23. 16. Bonate PL, Arthaud L, Stuhler J, Yerino P, Press RJ, Rose JQ. The distribution, metabolism, and elimination of clofarabine in rats. Drug Metab Dispos. 2005 Jun;33(6):739–48. 17. Berg SL, Bonate PL, Nuchtern JG, Dauser R, McGuffey L, Bernacky B, et al. Plasma and cerebrospinal fluid pharmacokinetics of clofarabine in nonhuman primates. Clin Cancer Res. 2005 Aug 15;11(16):5981–3. 18. Kantarjian HM, Gandhi V, Kozuch P, Faderl S, Giles F, Cortes J, et al. Phase I clinical and pharmacology study of clofarabine in patients with solid and hematologic cancers. J Clin Oncol. 2003 Mar 15;21(6):1167–73. 19. Kantarjian H, Gandhi V, Cortes J, Verstovsek S, Du M, Garcia-Manero G, et al. Phase 2 clinical and pharmacologic study of clofarabine in patients with refractory or relapsed acute leukemia. Blood. 2003 Oct 1;102(7):2379–86. 20. Jeha S, Gandhi V, Chan KW, McDonald L, Ramirez I, Madden R, et al. Clofarabine, a novel nucleoside analog, is active in pediatric patients with advanced leukemia. Blood. 2004 Feb 1;103(3):784–9. 21. Jeha S, Gaynon PS, Razzouk BI, Franklin J, Kadota R, Shen V, et al. Phase II study of clofarabine in pediatric patients with refractory or relapsed acute lymphoblastic leukemia. J Clin Oncol. 2006 Apr 20;24(12):1917–23. 22. Kearns P, Michel G, Nelken B, Joel S, Al-Ghazaly E, Beishuizen A, et al. A European Phase II Study (BIOV-111) of Clofarabine (Evoltra(R)) in Refractory and Relapsed Childhood Acute Lymphoblastic Leukemia: Final Results. ASH Annual Meeting Abstracts. 2007 November 16, 2007;110(11):11. 23. Cooper T, Ayres M, Nowak B, Gandhi V. Biochemical modulation of cytarabine triphosphate by clofarabine. Cancer Chemother Pharmacol. 2005 Apr;55(4):361–8. 24. Faderl S, Gandhi V, O’Brien S, Bonate P, Cortes J, Estey E, et al. Results of a phase 1-2 study of clofarabine in combination with cytarabine (ara-C) in relapsed and refractory acute leukemias. Blood. 2005 Feb 1;105(3):940–7. 25. Karp JE, Ricklis RM, Balakrishnan K, Briel J, Greer J, Gore SD, et al. A phase 1 clinical-laboratory study of clofarabine followed by cyclophosphamide for adults with refractory acute leukemias. Blood. 2007 Sep 15;110(6):1762–9. 26. Hijiya N, Gaynon P, Barry E, Silverman L, Thomson B, Chu R, et al. A multi-center phase I study of clofarabine, etoposide and cyclophosphamide in combination in pediatric patients with refractory or relapsed acute leukemia. Leukemia. 2009 Dec;23(12):2259–64.
186
P. Kearns and V. Saha
27. Locatelli F, Testi AM, Bernardo ME, Rizzari C, Bertaina A, Merli P, et al. Clofarabine, cyclophosphamide and etoposide as single-course re-induction therapy for children with refractory/multiple relapsed acute lymphoblastic leukemia. Br J Haematol. 2009 Nov;147(3): 371–8. 28. Reist EJ, Goodman L. Synthesis of 9-Beta-D-Arabinofuranosylguanine. Biochemistry. 1964 Jan;3:15–8. 29. Fairbanks LD, Taddeo A, Duley JA, Simmonds HA. Mechanisms of deoxyguanosine lymphotoxicity. Human thymocytes, but not peripheral blood lymphocytes accumulate deoxy-GTP in conditions simulating purine nucleoside phosphorylase deficiency. J Immunol. 1990 Jan 15;144(2):485–91. 30. Lambe CU, Averett DR, Paff MT, Reardon JE, Wilson JG, Krenitsky TA. 2-Amino-6methoxypurine arabinoside: an agent for T-cell malignancies. Cancer Res. 1995 Aug 1;55(15):3352–6. 31. Elion GB. The purine path to chemotherapy. Science (New York, NY). 1989 Apr 7;244(4900):41–7. 32. Kurtzberg J. The long and winding road of the clinical development of Nelarabine. Leuk Lymphoma. 2007 Jan;48(1):1–2. 33. Prus KL, Averett DR, Zimmerman TP. Transport and metabolism of 9-beta-D-arabinofuranosylguanine in a human T-lymphoblastoid cell line: nitrobenzylthioinosine-sensitive and -insensitive influx. Cancer Res. 1990 Mar 15;50(6):1817–21. 34. Rodriguez CO, Jr., Mitchell BS, Ayres M, Eriksson S, Gandhi V. Arabinosylguanine is phosphorylated by both cytoplasmic deoxycytidine kinase and mitochondrial deoxyguanosine kinase. Cancer Res. 2002 Jun 1;62(11):3100–5. 35. Rodriguez CO, Jr., Stellrecht CM, Gandhi V. Mechanisms for T-cell selective cytotoxicity of arabinosylguanine. Blood. 2003 Sep 1;102(5):1842–8. 36. Gudas LJ, Ullman B, Cohen A, Martin DW, Jr. Deoxyguanosine toxicity in a mouse T lymphoma: relationship to purine nucleoside phosphorylase-associated immune dysfunction. Cell. 1978 Jul;14(3):531–8. 37. Kisor DF, Plunkett W, Kurtzberg J, Mitchell B, Hodge JP, Ernst T, et al. Pharmacokinetics of nelarabine and 9-beta-D-arabinofuranosyl guanine in pediatric and adult patients during a phase I study of nelarabine for the treatment of refractory hematologic malignancies. J Clin Oncol. 2000 Mar;18(5):995–1003. 38. Kurtzberg J, Ernst TJ, Keating MJ, Gandhi V, Hodge JP, Kisor DF, et al. Phase I study of 506U78 administered on a consecutive 5-day schedule in children and adults with refractory hematologic malignancies. J Clin Oncol. 2005 May 20;23(15):3396–403. 39. Berg SL, Blaney SM, Devidas M, Lampkin TA, Murgo A, Bernstein M, et al. Phase II study of nelarabine (compound 506U78) in children and young adults with refractory T-cell malignancies: a report from the Children’s Oncology Group. J Clin Oncol. 2005 May 20;23(15):3376–82. 40. DeAngelo DJ, Yu D, Johnson JL, Coutre SE, Stone RM, Stopeck AT, et al. Nelarabine induces complete remissions in adults with relapsed or refractory T-lineage acute lymphoblastic leukemia or lymphoblastic lymphoma: Cancer and Leukemia Group B study 19801. Blood. 2007 Jun 15;109(12):5136–42. 41. Thompson MA, Pro B, Sarris A, Hagemeister FB, Goy A, Bleyer A, et al. Results of a Phase II Study of 506U78 (Nelarabine) in Refractory Indolent B-Cell or Peripheral T-Cell Lymphoma. ASH Annual Meeting Abstracts. 2005 November 16, 2005;106(11):2681. 42. Goekbuget N, Arnold R, Atta J, Bruck P, Hermann S, Horst H, et al. Compound GW506U78 Has High Single-Drug Activity and Good Feasibility in Heavily Pretreated Relapsed T-Lymphoblastic Leukemia (T-ALL) and T-Lymphoblastic Lymphoma (T-LBL) and Offers the Option for Cure with Stem Cell Transplantation (SCT). ASH Annual Meeting Abstracts. 2005 November 16, 2005;106(11):150. 43. Gandhi V, Plunkett W, Weller S, Du M, Ayres M, Rodriguez CO, Jr., et al. Evaluation of the combination of nelarabine and fludarabine in leukemias: clinical response, pharmacokinetics, and pharmacodynamics in leukemia cells. J Clin Oncol. 2001 Apr 15;19(8):2142–52.
9 Nucleoside Analogues
187
44. Gandhi V, Tam C, O’Brien S, Jewell RC, Rodriguez CO, Jr., Lerner S, et al. Phase I trial of nelarabine in indolent leukemias. J Clin Oncol. 2008 Mar 1;26(7):1098–105. 45. Czuczman MS, Porcu P, Johnson J, Niedzwiecki D, Kelly M, Hsi ED, et al. Results of a phase II study of 506U78 in cutaneous T-cell lymphoma and peripheral T-cell lymphoma: CALGB 59901. Leuk Lymphoma. 2007 Jan;48(1):97–103. 46. Cohen MH, Johnson JR, Massie T, Sridhara R, McGuinn WD, Jr., Abraham S, et al. Approval summary: nelarabine for the treatment of T-cell lymphoblastic leukemia/lymphoma. Clin Cancer Res. 2006 Sep 15;12(18):5329–35. 47. Cohen MH, Johnson JR, Justice R, Pazdur R. FDA drug approval summary: nelarabine (Arranon) for the treatment of T-cell lymphoblastic leukemia/lymphoma. Oncologist. 2008 Jun;13(6):709-14. 48. Harned TM, Gaynon P. Relapsed acute lymphoblastic leukemia: current status and future opportunities. Current oncology reports. 2008 Nov;10(6):453–8. 49. Giblett ER. ADA and PNP deficiencies: how it all began. Ann N Y Acad Sci. 1985;451:1–8. 50. carson DK, J. Seegmiller. J.E. Lymphospecific toxicity in adenosine deamminase deficiency and purine nucleoside phosphorylase deficiency:possible role of nucleoside kinase(s). Proc Natl Acad Sci U S A. 1977;74(12):5677–81. 51. stoop JWz, B.J. Hendrick G.F. van Heukelom L.H. Staal G.F. de Bree P.K. Wadman S.K. Ballieux R.F. Purine nucleoside phosphorylase deficiency associated with selective cellular immunodeficiency N Engl J Med. 1977;296(12):651–5. 52. Schramm VL. Development of transition state analogues of purine nucleoside phosphorylase as anti-T-cell agents. Biochim Biophys Acta. 2002 Jul 18;1587(2-3):107–17. 53. Huang M, Wang Y, Gu J, Yang J, Noel K, Mitchell BS, et al. Determinants of sensitivity of human T-cell leukemia CCRF-CEM cells to immucillin-H. Leuk Res. 2008 Aug;32(8):1268–78. 54. Kicska GA, Long L, Horig H, Fairchild C, Tyler PC, Furneaux RH, et al. Immucillin H, a powerful transition-state analog inhibitor of purine nucleoside phosphorylase, selectively inhibits human T lymphocytes. Proc Natl Acad Sci U S A. 2001 Apr 10;98(8):4593–8. 55. Bantia S, Ananth SL, Parker CD, Horn LL, Upshaw R. Mechanism of inhibition of T-acute lymphoblastic leukemia cells by PNP inhibitor – BCX-1777. Int Immunopharmacol. 2003 Jun;3(6):879–87. 56. Gandhi V, Balakrishnan K. Pharmacology and mechanism of action of forodesine, a T-cell targeted agent. Semin Oncol. 2007 Dec;34(6 Suppl 5):S8–12. 57. Kilpatrick JM, Morris PE, Serota DG, Jr., Phillips D, Moore DR, Bennett JC, et al. Intravenous and oral pharmacokinetic study of BCX-1777, a novel purine nucleoside phosphorylase transition-state inhibitor. In vivo effects on blood 2¢-deoxyguanosine in primates. Int Immunopharmacol. 2003 Apr;3(4):541–8. 58. Bantia S, Miller PJ, Parker CD, Ananth SL, Horn LL, Kilpatrick JM, et al. Purine nucleoside phosphorylase inhibitor BCX-1777 (Immucillin-H) – a novel potent and orally active immunosuppressive agent. Int Immunopharmacol. 2001 Jun;1(6):1199–210. 59. Balakrishnan K, Nimmanapalli R, Ravandi F, Keating MJ, Gandhi V. Forodesine, an inhibitor of purine nucleoside phosphorylase, induces apoptosis in chronic lymphocytic leukemia cells. Blood. 2006 Oct 1;108(7):2392–8. 60. Alonso R, Lopez-Guerra M, Upshaw R, Bantia S, Smal C, Bontemps F, et al. Forodesine has high antitumor activity in chronic lymphocytic leukemia and activates p53-independent mitochondrial apoptosis by induction of p73 and BIM. Blood. 2009 Aug 20;114(8):1563–75. 61. Gandhi V, Kilpatrick JM, Plunkett W, Ayres M, Harman L, Du M, et al. A proof-of-principle pharmacokinetic, pharmacodynamic, and clinical study with purine nucleoside phosphorylase inhibitor immucillin-H (BCX-1777, forodesine). Blood. 2005 Dec 15;106(13):4253–60. 62. Jacobsson B, Albertioni F, Eriksson S. Deoxynucleoside anabolic enzyme levels in acute myelocytic leukemia and chronic lymphocytic leukemia cells. Cancer Lett. 2001 Apr 26;165(2):195–200. 63. Balakrishnan K, Verma D, O’Brien S, Kilpatrick JM, Chen Y, Tyler BF, et al. Phase II and pharmacodynamic study of oral forodesine in patients with advanced and/or fludarabinetreated chronic lymphocytic leukemia. Blood. Aug 12;116(6):886–92.
Chapter 10
FLT3 Inhibitors as Therapeutic Agents in MLL Rearranged Acute Lymphoblastic Leukemia Ronald W. Stam and Rob Pieters
10.1 FLT3 Regulation in Normal Hematopoietic Cells Fms-like tyrosine kinase 3 represents a membrane-bound receptor tyrosine kinase (RTK) encoded by the FLT3 (a.k.a. STK1 or FLK2) gene on chromosome 13q12. Like the steel factor receptor (KIT), platelet-derived growth factor receptor (PDGFR), and macrophage colony-stimulating factor receptor (FMS), FLT3 belongs to the class III family of RTKs. It contains an extracellular ligand binding region composed of five immunoglobulin-like domains, a transmembrane (TM) domain, a juxtamembrane (JM) domain, and two separated intracellular tyrosine kinase domains (TKDs). Class III RTKs constitute a family of proteins that fulfill important functions in the proliferation, differentiation and survival of hematopoietic cells [1]. FLT3 expression is mainly restricted to early lymphoid progenitors and to some extent in early myeloid progenitors [2], indicating that FLT3 plays an essential role in early hematopoietic development. This was emphasized by studies in mouse models, demonstrating that Flt3 defective mice develop specific deficiencies in primitive lymphoid and myeloid progenitors [3]. Due to the autoinhibitory conformation of the JM domain, FLT3 resides at the cell membrane in an inactive state, displaying only minimal intracellular kinase activity [4]. Activation of FLT3 normally occurs through binding of the FLT3 ligand (FLT3L), a homodimeric transmembrane protein that either functions in its membrane-bound state, or as a soluble growth factor when it undergoes proteolytic cleavage and becomes detached and secreted from the cell surface [5, 6]. FLT3L is abundantly expressed in most human tissues, including hematopoietic organs such as the bone marrow and peripheral blood, suggesting that the presence of the FLT3 receptor, rather than the availability of FLT3L, dictates tissue specificity of FLT3 activation. Binding of dimeric FLT3L to the extracellular regions of monomeric FLT3 receptors rapidly induces receptor homodimerization, presumably
R. Pieters (*) Erasmus MC – Sophia Children’s Hospital Pediatric Oncology/Hematology, Dr. Molewaterplein 60, 3000 CB, Rotterdam, the Netherlands e-mail:
[email protected] V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3_10, © Springer Science+Business Media, LLC 2011
189
190
R.W. Stam and R. Pieters
stabilizing the ligand–receptor complex. Subsequently, FLT3 dimerization readily promotes autophosphorylation (i.e., activation) of the intracellular TKDs, thereby activating several downstream signaling pathways that stimulate and maintain hematopoietic cell proliferation and survival [7–9].
10.2 Aberrant FLT3 Activation Whereas, normal FLT3 regulation requires binding of FLT3L, in several hematopoietic malignancies, including acute myeloid leukemia (AML) and certain subtypes of acute lymphoblastic leukemia (ALL), genetic abnormalities have been described that constitutively activate the FLT3 receptor in the absence of its ligand. Genetic lesions inducing ligand-independent FLT3 activation basically fall into two categories. The most common type of activating FLT3 alterations are internal tandem duplications (ITDs) of the JM domain, first described in 1996 by Nakao et al. [10]. These ITDs essentially are in-frame inserted copies of small stretches of within the JM domain coding sequence. Although a wide variety of ITDs have been reported, varying both in length and exact position between patients, this type of alteration appears to be restricted to the JM domain of the FLT3 receptor. FLT3-ITDs disrupt and abolish the autoinhibitory activity of the JM domain, which results in receptor dimerization and subsequent autophosphorylation (activation) without the requirement of FLT3L [11]. The second type of activating FLT3 lesions are missense point mutations, small insertions or deletions affecting amino acids crucial for kinase activity within the second TKD. Analogous to point mutations described in the corresponding TKD in KIT [12], these mutations alter the conformation of the so-called activation loop of the receptor from an “inactive” to an “active” (or open) state in which the catalytic kinase domain becomes accessible. As a result this allows autophosphorylation and thus activation of the FLT3 receptor, again in the absence of its ligand [13]. Although most studies on aberrantly activated FLT3 mainly focus on FLT3-ITDs and mutations within the FLT3 activation loop, several attempts have been made, including efforts from our laboratory, to screen other parts of the FLT3 gene for the presence of additional genetic abnormalities. By using denaturing High Performance Liquid Chromatography (dHPLC) to study extended parts, or in our case the entire FLT3 gene, a number of additional FLT3 mutations have been identified [14, 15]. Unfortunately, the potential of these mutations to activate FLT3 in a ligandindependent manner in most cases remain uncertain. For example, we discovered point mutations in the TM domain as well as in the extracellular region of FLT3, but due to limited sample availability, we were unable to confirm ligand-independent FLT3 activation. However, a recent study provided evidence that some of these rare mutations outside the JM domain and the TKDs do have the potential to constitutively activate FLT3 [16]. Moreover, Schittenhelm et al. [17], recently discovered a FLT3 missense mutation within the first TKD that, like well-established mutations in the second TKD, also resulted in ligand-independent FLT3 phosphorylation.
10 FLT3 Inhibitors as Therapeutic Agents in MLL
191
Together these observations indicate that the full repertoire of activating FLT3 mutations is not per se limited to the JM and second TKD.
10.3 Transforming Potential of Mutated FLT3 Thus, mutations in FLT3 in acute leukemias, particularly in JM domain (FLT3-ITDs) and those affecting the TKDs, result in ligand-independent FLT3 activation. In turn this results in constitutive signals triggering several transduction cascades like the STAT5a, RAS/MAPK, and PI3-K/Akt pathways [18, 19], presumably stimulating leukemic cell survival and proliferation. Moreover, FLT3-ITDs in AML cells were shown to block myeloid differentiation by suppression of C/EBPa [20, 21], and to induce increased production of reactive oxygen species (ROSs) resulting in DNA damage [22]. Together, these findings raise the question whether aberrantly activated FLT3 in acute leukemias initiates malignant transformation, or whether it represents a secondary event in leukemogenesis that favors the maintenance of leukemic cells. A number of studies, have demonstrated that constitutive FLT3 activation in fact exhibits pronounced transforming potential when introduced in normal hematopoietic cells. For example, transfection of interleukin-3 dependent murine myeloid (32D) or lymphoid (Ba/F3) cells with constructs expressing constitutively activated FLT3, induced growth factor independent growth and resistance to apoptosis [23–25]. Moreover, transplantation of 32D or Ba/F3 cells expressing constitutively activating FLT3 mutants into syngeneic mice rapidly led to leukemogenesis [23, 25]. However, studies in paired presentation and relapse samples from AML patients showed contradictory evidence. Activating FLT3 mutations present at diagnosis were occasionally lost at relapse, while other patients that were negative for FLT3 activation at presentation sometimes acquire activating mutations during relapse [26, 27]. So, aberrantly activated FLT3 is capable of malignant transformation, but in AML patients FLT3 aberrations are more likely to behave as secondary events rather than initiators of leukemic transformation. Importantly, this clearly establishes a role for abnormal FLT3 activation in human leukemias, provides important insights in leukemia development and presents itself as a target for therapeutic intervention.
10.4 Activating FLT3 Mutations in AML and ALL In AML, FLT3 is by far the most frequently mutated gene. Numerous studies screening primary AML samples from adults for the presence of FLT3-ITDs reported this type of activating FLT3 mutation to occur with incidences ranging from 20 to 35% [7, 28]. In pediatric AML, however, the frequency of FLT3-ITDs appears to be somewhat lower with a prevalence of ~15% [29]. In both adult and pediatric AML, TKD loop activation mutations occur in approximately 7% of the cases [30–32].
192
R.W. Stam and R. Pieters
FLT3 mutations tended to be more prevalent in patients with FAB-M5 (monoblastic or monocytic) AML, and with normal karyotypes [32]. Furthermore, multiple studies have shown that among AML patients, the presence of FLT3-ITDs is associated with increased risk of relapse and reduced probability of survival. The clinical and prognostic relevance of FLT3-TKD mutations, however, remains less obvious [7, 28]. In ALL the incidence of activating FLT3 mutations is significantly lower and seem to arise preferentially in patients with hyperdiploid (>50 chromosomes) ALL, and in those with a translocation of the mixed lineage leukemia (MLL) gene. The overall prevalence of FLT3-TKD mutations in ALL is about 3–5% [13, 33, 34], whereas FLT3-ITDs are rare [35, 36]. Despite a favorable outcome associated with hyperdiploid ALL, about 8–25% of these patients carry activating FLT3-TKD mutations [33–36], or small deletions within the JM domain [37]. Given the relatively high frequency of activating FLT3 mutations and yet a superior clinical outcome of ALL patients with hyperdiploidy, aberrant FLT3 activation seems unlikely to provide these cells with an effective growth advantage. In contrast, ALL characterized by MLL translocations represents a highly unfavorable ALL subtype, most frequently diagnosed in infants (i.e., children <1 year of age). In MLL rearranged ALL, FLT3-TKD mutations occur in 3–20% of the cases [34, 35, 38, 39]. Whether activating FLT3 mutations are of clinical importance in these children remains to be confirmed.
10.5 Constitutive Activation of Wild-Type FLT3 in MLL Rearranged ALL Interestingly, apart from activating FLT3-TKD mutations, another mechanism leading to constitutive FLT3 activation appeared to be at work in MLL rearranged ALL cells. Comparing gene expression profiles between MLL rearranged ALL, conventional ALL and AML revealed that the FLT3 gene is consistently highly expressed in primary MLL rearranged ALL cells [40]. Shortly after this observation, we demonstrated that ALL cell lines carrying translocations of the MLL gene, and expressing high levels of wild-type FLT3, display levels of FLT3 phosphorylation (activation) comparable to that of AML cells carrying mutated FLT3 [38]. Using RT-PCR analysis in an independent cohort of patient samples we confirmed that FLT3 is significantly overexpressed in MLL rearranged ALL when compared to other, more favorable types of ALL [39]. Furthermore, we and others demonstrated that similar to the cell line observations, high-level expression of wild-type FLT3 in primary MLL rearranged ALL, was associated with constitutive FLT3 phosphorylation (activation) in the absence of mutations [39, 41]. Thus constitutively activated FLT3, either as a consequence of FLT3-TKD mutations or as a result of high-level expression of wild-type FLT3, appears to be present in the majority of MLL rearranged ALL cases. Taken together these finding indicate that in addition to a large group of AML patients, activated FLT3 may also be a valuable therapeutic target in ALL patients carrying translocations of the MLL gene.
10 FLT3 Inhibitors as Therapeutic Agents in MLL
193
However, the remarkably high levels of FLT3 expression in MLL rearranged ALL cells may be related to the hematopoietic cell type in which this leukemia originates. As described above, the highest FLT3 expression among normal hematopoietic cells is found during early B-cell development. MLL rearranged ALL cells typically display immunophenotypes that resemble highly immature pro-B cells, suggesting that the abundant expression of FLT3 in these cells may simply mirror normal expression patterns throughout B-cell development. If so, receptor activation due to high-level FLT3 expression may not as such be classified as a leukemia-specific abnormality, and therefore may not be considered to be a genuine therapeutic target, although still a useful target. Nevertheless, MLL rearranged ALL represents a highly aggressive type of leukemia that is notoriously characterized by cellular drug resistance and extremely high white blood cell counts at disease presentation. These characteristics imply that this leukemia strongly benefits from factors that effectively stimulate cell proliferation and survival, possibly partly realized by activated FLT3 signaling. In line with these suggestions, we recently found that among MLL rearranged ALL patients, the level of FLT3 expression is of prognostic relevance, with 1-year event free survival (EFS) rates of 36% for patients expressing high FLT3 levels compared to 71% for patients expressing lower levels of FLT3 [42]. Yet, these observations should be interpreted with caution as this analysis included only a small number of patients with relatively short follow-up periods. On the other hand, these data are in concordance with data published by Ozeki et al., [43] who showed that high-level expression of wild-type FLT3 is also an unfavorable prognostic factor for overall survival in AML.
10.6 FLT3 Inhibitors in AML The identification of activating FLT3 mutations and their clinical relevance in AML has led to the identification of various small molecule compounds that inhibit aberrant FLT3 signaling. To date, the most well characterized and established FLT3 inhibitors include midostaurin (PKC412; Novartis) [44], lestaurtinib (CEP-701; Cephalon) [45], and sunitinib (SU11248; Pfizer) [46]. The latter compound was preceded by semaxanib (SU5614; Pfizer). Although semaxanib initially showed promising results as a FLT3 kinase inhibitor in a variety of AML cell line models [47], the compound appeared to have poor pharmacologic properties, showed only modest clinical activity and induced severe toxicities in patients with refractory AML and Myelodysplastic Syndrome (MDS) [48]. Based on these and other findings, further studies investigating the use of semaxanib as a FLT3 inhibitor was abandoned and the compound was replaced by the rationally designed small molecule sunitinib [49]. Despite the FLT3-inhibiting activities of midostaurin, lestaurtinib, and sunitinib, none of these molecules were specifically designed as FLT3 inhibitors. Midostaurin was originally identified as a protein kinase C (PKC) inhibitor, and subsequently was shown also to inhibit various kinases like KIT, PDGFR, kinase insertion
194
R.W. Stam and R. Pieters
domain receptor (KDR), and vascular endothelial growth factor receptor 2 (VEGFR2) [50]. Before the FLT3 inhibiting activity of lestaurtinib became apparent, the compound was shown to inhibit tropomyosin-related Kinase (TRK) [51]. In contrast to midostaurin and lestaurtinib, sunitinib was designed to inhibit FLT3, but not specifically. The rationale behind the development of sunitinib was to design a small molecule capable of inhibiting a broad range of RTKs, including VEGFRs, PDGFRs, KIT, and FLT3 [49]. Nevertheless, all three compounds have shown significant potency against aberrantly activated FLT3 in both AML cell lines as well as primary AML patient cells, albeit with varying selectivity [45, 52, 53]. For example, lestaurtinib and midostaurin caused inhibition of FLT3 phosphorylation in primary AML cells, but in order to provoke cytotoxic responses, higher concentrations of midostaurin were required [52]. Lestaurtinib appeared to be more effective against primary AML cells carrying FLT3-ITDs as compared to cells carrying FLT3-TKD point mutations [54, 55], whereas midostaurin was shown to be more potent in FLT3-TKD bearing cells [56]. Moreover, while variable sensitivity of different FLT-TKD mutations have been reported [57], midostaurin was shown to be equally effective in cells carrying an array of distinct mutations in the FLT3-TKD [58].
10.7 Clinical Evaluation of FLT3 Inhibitors in AML The promising effects of FLT3 inhibitors on primary AML blasts in vitro, and the successful evaluation of these compounds in vivo in mouse models [44–46], justified the initiation of clinical trials determining the tolerance and efficacy of single-agent FLT3 inhibitor therapies in relapsed or refractory AML patients. The first clinical study involved a phase I/II trial designed to assess the hematologic effects of lestaurtinib monotherapy [59]. For this, fourteen relapsed or refractory adult AML patients were orally treated with 60 mg lestaurtinib twice a day, at two different study centers. All patients were characterized by constitutively FLT3 activation due to FLT3-ITDs, except for one patient who carried a FLT3-TKD mutation. Five patients demonstrated clinical responses, with four patients showing peripheral blast reductions, decreasing blast percentages from initially 27–94% to <5% after treatment. In one patient a bone marrow response was observed, showing a 25% decrease in bone marrow blasts. In addition, FLT3 phosphorylation in plasma samples from responding patients was markedly reduced, confirming the correlation between clinical responses and lestaurtinib-induced FLT3 inhibition [59]. A second phase II trial using lestaurtinib monotherapy in untreated elderly AML patients, further demonstrated the in vivo potential of this drug [60]. In total, 29 patients were treated twice daily for 8 weeks with initially 60 mg, escalating to 80 mg lestaurtinib, regardless of the presence of activating FLT3 mutations. In fact, only two patients carried a FLT3-ITD, and three patients harbored a FLT3-TKD mutations. Among the five patients carrying FLT3 mutations, three showed pronounced responses defined as transient blast reductions in the bone marrow and
10 FLT3 Inhibitors as Therapeutic Agents in MLL
195
peripheral blood. Interestingly, clinical responses were also observed in five patients carrying wild-type FLT3. Again, the observed clinical responses appeared to correlate with sustained FLT3 inhibition and in vitro lestaurtinib sensitivity as determined in mononuclear bone marrow cells from patients before treatment [60]. In both clinical trials, lestaurtinib was well tolerated and observed drug-related toxicities were minimal. Shortly after the publication of the first clinical results using lestaurtinib, similar results were reported using the FLT3 inhibitor midostaurin [61]. Nineteen adult patients with relapsed or refractory AML and one patient with high-grade MDS were orally treated with 75 mg midostaurin 3 times a day as a single agent for at least 2 weeks. Among these patients, 18 carried a FLT3-ITD and two had a FLT3TKD mutation. As in the first clinical lestaurtinib study, these patients were heavily pretreated and were not considered for further treatments with standard AML chemotherapies. In 14 patients, a reduction in peripheral blasts was observed, by at least 50%. Seven of these patients reached a greater than 2-log peripheral blast reduction, and most of these patient also displayed varying bone marrow responses. One individual ostensibly achieved a complete response. Interestingly, this latter patient was one of the two patients with a FLT3-TKD mutation. The impressive response in this patient possibly reflects the superior effects of midostaurin on FLT3-TKD mutated cells over cells carrying FLT3-ITDs in vitro [56], whereas lestaurtinib appeared to preferentially target ITD-induced FLT3 activation [54] (as described above). However, the response in the only other patient with a FLT3TKD mutation, representing the only MDS patient in this study, was difficult to evaluate as midostaurin administration had to be stopped at day 9 due to respiratory complications. Approximately 2 weeks later, this patient died from progressive pulmonary failure, which was not believed to be drug related. In addition, two other patients developed fatal pulmonary events, which in both cases was more likely to be drug related. Apart from these unfortunate events, midostaurin was generally well tolerated and side effects were mostly limited to nausea and vomiting. Finally, in most of the responding patients constitutive FLT3 phosphorylation was inhibited, indicating that clinical response was associated with midostaurin specific FLT3 inhibition [61]. Other FLT3 inhibitors for which preclinical testing revealed potent activity towards mutated FLT3 are sorafenib (BAY-43-9006; Onyx Pharmaceuticals) and tandutinib (MLN518; Millenium Pharmaceuticals Inc.). The latter compound has been evaluated in dose escalating phase I clinical trial in which eventually 5 out of 40 refractory AML or high-risk MDS patients appeared evaluable for assessment of antileukemic activity. Two of these patients showed reductions in peripheral and bone marrow blasts [62]. Taken together, the relatively successful results from the above described clinical trials show that FLT3 inhibitors like lestaurtinib and midostaurin exhibit clinical activity, at least in AML patients characterized by constitutively activated FLT3 as a result of mutations. Moreover, the data suggest that FLT3 inhibition represents a promising therapeutic strategy, that should now be tested in combination with conventional chemotherapies currently used in the treatment of AML. In fact, several
196
R.W. Stam and R. Pieters
in vitro studies already demonstrated synergy of FLT3 inhibitor in combined with drugs like cytarabine and daunorubicin, provided that drug administration is correctly sequenced to prevent antagonistic effects [53, 63–65].
10.8 FLT3 Inhibitors as Therapeutics for MLL Rearranged ALL Given the success of in vitro and in vivo inhibition of mutated FLT3, the obvious question surfaced whether FLT3 inhibitors would also affect leukemic cells displaying constitutive FLT3 activation that results from wild-type FLT3 overexpression as observed in MLL rearranged ALL. Although the number of studies addressing this question remains small, the results are indeed encouraging. Cell line studies showed that midostaurin induced leukemic cell death in MLL rearranged ALL cells displaying high-level expression of wild-type FLT3 in a dose-dependent manner. Interestingly, the cytotoxic response in these cells was comparable to that observed in FLT3-ITD positive AML cells and was accompanied by complete inhibition of FLT3 phosphorylation. In contrast, ALL cells that lacked significantly elevated FLT3 levels did not show FLT3 phosphorylation and were hardly affected by midostaurin exposure [38]. Data from our laboratory confirmed these findings in primary samples from children with MLL rearranged ALL. By determining the amount of FLT3 phosphorylation in these samples, we showed that high-level FLT3 expression was associated with significant receptor phosphoprylation and thus activation, which in all cases was completely abolished by midostaurin. In turn, inhibition of FLT3 activation in these cells by midostaurin resulted in dose-dependent leukemic cell death. In contrast, in ALL cells expressing low levels of FLT3, no phosphorylation was detected and exposure to midostaurin induced only limited amounts of apoptosis, and in some of these cases not at all [39]. Independent from our observations, similar data was obtained with the FLT3 inhibitor lestaurtinib. Like midostaurin, lestaurtinib effectively inhibited FLT3 phosphorylation and selectively killed pediatric ALL cells expressing high levels of wild-type FLT3, whereas ALL cells that showed low FLT3 expression barely responded [41]. Together these in vitro studies demonstrate that constitutively activated FLT3 as a result of wild-type overexpression can indeed be targeted by FLT3 inhibitors, and that the observed cytotoxic responses of these compounds is associated with the actual inhibition of aberrant FLT3 signaling. However, relative to Lestaurtinib, higher concentrations of midostaurin are required to induce comparable amounts of cell death in primary MLL rearranged ALL cells [39, 41]. This latter may reflect differences in selectivity between both drugs as lestaurtinib is less selective and consequently more effective in inducing cytotoxicity [66]. To assess the antileukemic effects of midostaurin in vivo, Armstrong et al. established a xenograft mouse model in which leukemia burden could be quantitated by bioluminescent imaging. In control mice injected with human MLL rearranged ALL cells, the leukemia rapidly progressed and in only 3 weeks after injection replaced the normal bone marrow and infiltrated other organs. In similar mice
10 FLT3 Inhibitors as Therapeutic Agents in MLL
197
orally treated with midostaurin however, engraftment of the leukemia failed, and the leukemia burden during the 3 weeks of treatment never exceeded the amount of detectable leukemic cells as observed at time of injection. In contrast, midostaurin had no antileukemic effect on engrafted ALL cells lacking constitutive FLT3 activation. Thus, this study demonstrated that FLT3 inhibitors also effectively and selectively target FLT3 activated by overexpression in MLL rearranged ALL cells in vivo [38]. Based upon these findings, two independent clinical trials have now been designed to determine the clinical activity of FLT3 inhibitors in pediatric ALL. One mainly European trial will study the efficacy of midostaurin in children with relapsed MLL rearranged ALL and relapsed AML. The second study shall investigate the clinical potential of lestaurtinib added to combination chemotherapy in infant ALL patients in the USA. The anticipated results from these studies should reveal whether or not inhibition of FLT3 may become a welcome addition in the therapeutic repertoire against difficult to treat, high-risk leukemias such as MLL rearranged ALL.
10.9 Future Perspectives Taken together, small molecule FLT3 inhibitors seem well on their way to become implemented in common AML treatment regimes, and in the near future may prove to be suitable for the treatment of certain high-risk subtypes of ALL as well. However, recent insights may have uncovered an unfortunate plot twist in the so far feel-good story of FLT3 inhibitors. Just as the initial success of the tyrosine kinase inhibitor Imatinib in patients with BCR-ABL positive chronic myeloid leukemia (CML) was hampered by the development of acquired resistance in patients treated with this drug [67, 68], a similar fate may lie in store for FLT3 inhibitors, at least for some of them. Several studies suggested the occurrence of primary resistance, but more importantly also on secondary acquired resistance to FLT3 inhibitors (reviewed in ref. [69]). Although still poorly understood, some of the resistance mechanisms may include inhibitor-induced production of less active metabolites, strong binding of the inhibitor to for example alpha-1 acid glycoprotein and the subsequent reduction of free drug levels, and the induction of resistance mutations [70]. Other possible mechanisms may involve the activation of compensatory signaling pathways [71], inhibitor-induced upregulation of antiapoptotic proteins like BCL2 [72] or MCL-1 [73], or elevated expression of the FLT3 receptor itself [60]. Finally, the bone marrow microenvironment in which at least part of the leukemic cell population resides, may play an important role in decreasing drug efficacy [74], presumably by the secretion of survival stimulating growth factors. Second generation compounds that may override resistance comparable to the development of dasatinib to overcome imatinib resistance [75], have been identified [76]. Nevertheless, inhibition of aberrantly activated FLT3 remains an attractive therapeutic strategy, and if activity of FLT3 inhibitors can be shown in subgroups
198
R.W. Stam and R. Pieters
of childhood ALL, studies needs to be done to explore their value when combined with regular chemotherapy. The future of FLT3 inhibitors will definitely not be in monotherapy.
References 1. O. Rosnet, and D. Birnbaum, Hematopoietic receptors of class III receptor-type tyrosine kinases. Crit Rev Oncog. 1993; 4:595–613. 2. O. Rosnet, H.J. Buhring, S. Marchetto, I. Rappold, C. Lavagna, D. Sainty, C. Arnoulet, C. Chabannon, L. Kanz, C. Hannum, and D. Birnbaum, Human FLT3/FLK2 receptor tyrosine kinase is expressed at the surface of normal and malignant hematopoietic cells. Leukemia. 1996; 10:238–48. 3. K. Mackarehtschian, J.D. Hardin, K.A. Moore, S. Boast, S.P. Goff, and I.R. Lemischka, Targeted disruption of the flk2/flt3 gene leads to deficiencies in primitive hematopoietic progenitors. Immunity. 1995; 3:147–61. 4. J. Griffith, J. Black, C. Faerman, L. Swenson, M. Wynn, F. Lu, J. Lippke, and K. Saxena, The structural basis for autoinhibition of FLT3 by the juxtamembrane domain. Mol Cell. 2004; 13:169–78. 5. C. Hannum, J. Culpepper, D. Campbell, T. McClanahan, S. Zurawski, J.F. Bazan, R. Kastelein, S. Hudak, J. Wagner, J. Mattson, and et al., Ligand for FLT3/FLK2 receptor tyrosine kinase regulates growth of haematopoietic stem cells and is encoded by variant RNAs. Nature. 1994; 168:643–8. 6. S.N. Savvides, T. Boone, and P. Andrew Karplus, Flt3 ligand structure and unexpected commonalities of helical bundles and cystine knots. Nat Struct Biol. 2000; 7:486–91. 7. D.L. Stirewalt, and J.P. Radich, The role of FLT3 in haematopoietic malignancies. Nat Rev Cancer. 2003; 3:650–65. 8. Y. Kikushige, G. Yoshimoto, T. Miyamoto, T. Iino, Y. Mori, H. Iwasaki, H. Niiro, K. Takenaka, K. Nagafuji, M. Harada, F. Ishikawa, and K. Akashi, Human Flt3 is expressed at the hemato poietic stem cell and the granulocyte/macrophage progenitor stages to maintain cell survival. J Immunol. 2008; 180:7358–67. 9. K.C. Weisel, S. Yildirim, E. Schweikle, L. Kanz, and R. Mohle, Regulation of FLT3 and its ligand in normal hematopoietic progenitor cells. Ann Hematol. 2009; 88:203–11. 10. M. Nakao, S. Yokota, T. Iwai, H. Kaneko, S. Horiike, K. Kashima, Y. Sonoda, T. Fujimoto, and S. Misawa, Internal tandem duplication of the flt3 gene found in acute myeloid leukemia. Leukemia. 1996; 10:1911–8. 11. H. Kiyoi, R. Ohno, R. Ueda, H. Saito, and T. Naoe, Mechanism of constitutive activation of FLT3 with internal tandem duplication in the juxtamembrane domain. Oncogene. 2002; 21:2555–63. 12. T. Furitsu, T. Tsujimura, T. Tono, H. Ikeda, H. Kitayama, U. Koshimizu, H. Sugahara, J.H. Butterfield, L.K. Ashman, Y. Kanayama, and et al., Identification of mutations in the coding sequence of the proto-oncogene c-kit in a human mast cell leukemia cell line causing ligand-independent activation of c-kit product. J Clin Invest . 1993; 92:1736–44. 13. Y. Yamamoto, H. Kiyoi, Y. Nakano, R. Suzuki, Y. Kodera, S. Miyawaki, N. Asou, K. Kuriyama, F. Yagasaki, C. Shimazaki, H. Akiyama, K. Saito, M. Nishimura, T. Motoji, K. Shinagawa, A. Takeshita, H. Saito, R. Ueda, R. Ohno, and T. Naoe, Activating mutation of D835 within the activation loop of FLT3 in human hematologic malignancies. Blood. 2001; 97:2434–9. 14. M. Bianchini, E. Ottaviani, T. Grafone, B. Giannini, S. Soverini, C. Terragna, M. Amabile, P.P. Piccaluga, M. Malagola, M. Rondoni, C. Bosi, M. Baccarani, and G. Martinelli, Rapid detection of Flt3 mutations in acute myeloid leukemia patients by denaturing HPLC. Clin Chem. 2003; 49:1642–50.
10 FLT3 Inhibitors as Therapeutic Agents in MLL
199
15. R.W. Stam, M.L. den Boer, P. Schneider, M. Meier, H.B. Beverloo, and R. Pieters, D-HPLC analysis of the entire FLT3 gene in MLL rearranged and hyperdiploid acute lymphoblastic leukemia. Haematologica. 2007; 92:1565–8. 16. S. Frohling, C. Scholl, R.L. Levine, M. Loriaux, T.J. Boggon, O.A. Bernard, R. Berger, H. Dohner, K. Dohner, B.L. Ebert, S. Teckie, T.R. Golub, J. Jiang, M.M. Schittenhelm, B.H. Lee, J.D. Griffin, R.M. Stone, M.C. Heinrich, M.W. Deininger, B.J. Druker, and D.G. Gilliland, Identification of driver and passenger mutations of FLT3 by high-throughput DNA sequence analysis and functional assessment of candidate alleles. Cancer Cell. 2007; 12:501–13. 17. M.M. Schittenhelm, K.W. Yee, J.W. Tyner, L. McGreevey, A.D. Haley, A. Town, D.J. Griffith, T. Bainbridge, R.M. Braziel, A.M. O’Farrell, J.M. Cherrington, and M.C. Heinrich, FLT3 K663Q is a novel AML-associated oncogenic kinase: Determination of biochemical properties and sensitivity to Sunitinib (SU11248). Leukemia. 2006; 20:2008–14. 18. C. Choudhary, C. Muller-Tidow, W.E. Berdel, and H. Serve, Signal transduction of oncogenic Flt3. Int J Hematol. 2005; 82:93–9. 19. D. Small, FLT3 mutations: biology and treatment. Hematology. Am Soc Hematol Educ Program. 2006;178–84. 20. R. Zheng, A.D. Friedman, M. Levis, L. Li, E.G. Weir, and D. Small, Internal tandem duplication mutation of FLT3 blocks myeloid differentiation through suppression of C/EBPalpha expression. Blood. 2004; 103:1883–90. 21. H.S. Radomska, D.S. Basseres, R. Zheng, P. Zhang, T. Dayaram, Y. Yamamoto, D.W. Sternberg, N. Lokker, N.A. Giese, S.K. Bohlander, S. Schnittger, M.H. Delmotte, R.J. Davis, D. Small, W. Hiddemann, D.G. Gilliland, and D.G. Tenen, Block of C/EBP alpha function by phosphorylation in acute myeloid leukemia with FLT3 activating mutations. J Exp Med. 2006; 203:371–81. 22. A. Sallmyr, J. Fan, K. Datta, K.T. Kim, D. Grosu, P. Shapiro, D. Small, and F. Rassool, Internal tandem duplication of FLT3 (FLT3/ITD) induces increased ROS production, DNA damage, and misrepair: implications for poor prognosis in AML. Blood. 2008; 111:3173–82. 23. M. Mizuki, R. Fenski, H. Halfter, I. Matsumura, R. Schmidt, C. Muller, W. Gruning, K. KratzAlbers, S. Serve, C. Steur, T. Buchner, J. Kienast, Y. Kanakura, W.E. Berdel, and H. Serve, Flt3 mutations from patients with acute myeloid leukemia induce transformation of 32D cells mediated by the Ras and STAT5 pathways. Blood. 2000; 96:3907–14. 24. K. Spiekermann, K. Bagrintseva, R. Schwab, K. Schmieja, and W. Hiddemann, Overexpression and constitutive activation of FLT3 induces STAT5 activation in primary acute myeloid leukemia blast cells. Clin Cancer Res. 2003; 9:2140–50. 25. K.F. Tse, G. Mukherjee, and D. Small, Constitutive activation of FLT3 stimulates multiple intracellular signal transducers and results in transformation. Leukemia. 2000; 14:1766–76. 26. P.D. Kottaridis, R.E. Gale, S.E. Langabeer, M.E. Frew, D.T. Bowen, and D.C. Linch, Studies of FLT3 mutations in paired presentation and relapse samples from patients with acute myeloid leukemia: implications for the role of FLT3 mutations in leukemogenesis, minimal residual disease detection, and possible therapy with FLT3 inhibitors. Blood. 2002; 100:2393–8. 27. L.Y. Shih, C.F. Huang, J.H. Wu, P.N. Wang, T.L. Lin, P. Dunn, M.C. Chou, M.C. Kuo, and C.C. Tang, Heterogeneous patterns of FLT3 Asp(835) mutations in relapsed de novo acute myeloid leukemia: a comparative analysis of 120 paired diagnostic and relapse bone marrow samples. Clin Cancer Res. 2004; 10:1326–32. 28. D.G. Gilliland, and J.D. Griffin, The roles of FLT3 in hematopoiesis and leukemia. Blood. 2002; 100:1532–42. 29. S. Meshinchi, T.A. Alonzo, D.L. Stirewalt, M. Zwaan, M. Zimmerman, D. Reinhardt, G.J. Kaspers, N.A. Heerema, R. Gerbing, B.J. Lange, and J.P. Radich, Clinical implications of FLT3 mutations in pediatric AML. Blood. 2006; 108:3654–61. 30. F.M. Abu-Duhier, A.C. Goodeve, G.A. Wilson, R.S. Care, I.R. Peake, and J.T. Reilly, Identification of novel FLT-3 Asp835 mutations in adult acute myeloid leukemia. Br J Haematol. 2001; 113:983–8. 31. S. Meshinchi, D.L. Stirewalt, T.A. Alonzo, Q. Zhang, D.A. Sweetser, W.G. Woods, I.D. Bernstein, R.J. Arceci, and J.P. Radich, Activating mutations of RTK/ras signal transduction pathway in pediatric acute myeloid leukemia. Blood. 2003; 102:1474–9.
200
R.W. Stam and R. Pieters
3 2. C. Thiede, C. Steudel, B. Mohr, M. Schaich, U. Schakel, U. Platzbecker, M. Wermke, M. Bornhauser, M. Ritter, A. Neubauer, G. Ehninger, and T. Illmer, Analysis of FLT3activating mutations in 979 patients with acute myelogenous leukemia: association with FAB subtypes and identification of subgroups with poor prognosis. Blood. 2002; 99:4326–35. 33. M. Braoudaki, M. Karpusas, K. Katsibardi, C. Papathanassiou, K. Karamolegou, and F. Tzortzatou-Stathopoulou, Frequency of FLT3 mutations in childhood acute lymphoblastic leukemia. Med Oncol. 2008; 26:460–2. 34. T. Taketani, T. Taki, K. Sugita, Y. Furuichi, E. Ishii, R. Hanada, M. Tsuchida, K. Ida, and Y. Hayashi, FLT3 mutations in the activation loop of tyrosine kinase domain are frequently found in infant ALL with MLL rearrangements and pediatric ALL with hyperdiploidy. Blood. 2004; 103:1085–8. 35. A. Andersson, K. Paulsson, H. Lilljebjorn, C. Lassen, B. Strombeck, J. Heldrup, M. Behrendtz, B. Johansson, and T. Fioretos, FLT3 mutations in a 10 year consecutive series of 177 childhood acute leukemias and their impact on global gene expression patterns. Genes Chromosomes Cancer. 2008; 47:64–70. 36. K. Paulsson, A. Horvat, B. Strombeck, F. Nilsson, J. Heldrup, M. Behrendtz, E. Forestier, A. Andersson, T. Fioretos, and B. Johansson, Mutations of FLT3, NRAS, KRAS, and PTPN11 are frequent and possibly mutually exclusive in high hyperdiploid childhood acute lymphoblastic leukemia. Genes Chromosomes Cancer. 2008; 47:26–33. 37. S.A. Armstrong, M.E. Mabon, L.B. Silverman, A. Li, J.G. Gribben, E.A. Fox, S.E. Sallan, and S.J. Korsmeyer, FLT3 mutations in childhood acute lymphoblastic leukemia. Blood. 2004; 103:3544–6. 38. S.A. Armstrong, A.L. Kung, M.E. Mabon, L.B. Silverman, R.W. Stam, M.L. Den Boer, R. Pieters, J.H. Kersey, S.E. Sallan, J.A. Fletcher, T.R. Golub, J.D. Griffin, and S.J. Korsmeyer, Inhibition of FLT3 in MLL. Validation of a therapeutic target identified by gene expression based classification. Cancer Cell. 2003; 3:173–83. 39. R.W. Stam, M.L. den Boer, P. Schneider, P. Nollau, M. Horstmann, H.B. Beverloo, E. van der Voort, M.G. Valsecchi, P. de Lorenzo, S.E. Sallan, S.A. Armstrong, and R. Pieters, Targeting FLT3 in primary MLL-gene-rearranged infant acute lymphoblastic leukemia. Blood. 2005; 106:2484–90. 40. S.A. Armstrong, J.E. Staunton, L.B. Silverman, R. Pieters, M.L. den Boer, M.D. Minden, S.E. Sallan, E.S. Lander, T.R. Golub, and S.J. Korsmeyer, MLL translocations specify a distinct gene expression profile that distinguishes a unique leukemia. Nat Genet. 2002; 30:41–7. 41. P. Brown, M. Levis, S. Shurtleff, D. Campana, J. Downing, and D. Small, FLT3 inhibition selectively kills childhood acute lymphoblastic leukemia cells with high levels of FLT3 expression. Blood. 2005; 105:812–20. 42. R.W. Stam, P. Schneider, P. de Lorenzo, M.G. Valsecchi, M.L. den Boer, and R. Pieters, Prognostic significance of high-level FLT3 expression in MLL-rearranged infant acute lymphoblastic leukemia. Blood. 2007; 110:2774–5. 43. K. Ozeki, H. Kiyoi, Y. Hirose, M. Iwai, M. Ninomiya, Y. Kodera, S. Miyawaki, K. Kuriyama, C. Shimazaki, H. Akiyama, M. Nishimura, T. Motoji, K. Shinagawa, A. Takeshita, R. Ueda, R. Ohno, N. Emi, and T. Naoe, Biologic and clinical significance of the FLT3 transcript level in acute myeloid leukemia. Blood. 2004; 103:1901–8. 44. E. Weisberg, C. Boulton, L.M. Kelly, P. Manley, D. Fabbro, T. Meyer, D.G. Gilliland, and J.D. Griffin, Inhibition of mutant FLT3 receptors in leukemia cells by the small molecule tyrosine kinase inhibitor PKC412. Cancer Cell. 2002; 1:433–43. 45. M. Levis, J. Allebach, K.F. Tse, R. Zheng, B.R. Baldwin, B.D. Smith, S. Jones-Bolin, B. Ruggeri, C. Dionne, and D. Small, A FLT3-targeted tyrosine kinase inhibitor is cytotoxic to leukemia cells in vitro and in vivo. Blood. 2002; 99:3885–91. 46. A.M. O’Farrell, T.J. Abrams, H.A. Yuen, T.J. Ngai, S.G. Louie, K.W. Yee, L.M. Wong, W. Hong, L.B. Lee, A. Town, B.D. Smolich, W.C. Manning, L.J. Murray, M.C. Heinrich, and J.M. Cherrington, SU11248 is a novel FLT3 tyrosine kinase inhibitor with potent activity in vitro and in vivo. Blood. 2003; 101:3597–605. 47. K. Spiekermann, R.J. Dirschinger, R. Schwab, K. Bagrintseva, F. Faber, C. Buske, S. Schnittger, L.M. Kelly, D.G. Gilliland, and W. Hiddemann, The protein tyrosine kinase
10 FLT3 Inhibitors as Therapeutic Agents in MLL
201
inhibitor SU5614 inhibits FLT3 and induces growth arrest and apoptosis in AML-derived cell lines expressing a constitutively activated FLT3. Blood. 2003; 101:1494–504. 48. F.J. Giles, A.T. Stopeck, L.R. Silverman, J.E. Lancet, M.A. Cooper, A.L. Hannah, J.M. Cherrington, A.M. O’Farrell, H.A. Yuen, S.G. Louie, W. Hong, J.E. Cortes, S. Verstovsek, M. Albitar, S.M. O’Brien, H.M. Kantarjian, and J.E. Karp, SU5416, a small molecule tyrosine kinase receptor inhibitor, has biologic activity in patients with refractory acute myeloid leukemia or myelodysplastic syndromes. Blood. 2003; 102:795–801. 49. L.Q. Chow, and S.G. Eckhardt, Sunitinib: from rational design to clinical efficacy. J Clin Oncol. 2007; 25:884–96. 50. D. Fabbro, S. Ruetz, S. Bodis, M. Pruschy, K. Csermak, A. Man, P. Campochiaro, J. Wood, T. O’Reilly, and T. Meyer, PKC412--a protein kinase inhibitor with a broad therapeutic potential. Anticancer Drug Des. 2000; 15:17–28. 51. D.J. George, C.A. Dionne, J. Jani, T. Angeles, C. Murakata, J. Lamb, and J.T. Isaacs, Sustained in vivo regression of Dunning H rat prostate cancers treated with combinations of androgen ablation and Trk tyrosine kinase inhibitors, CEP-751 (KT-6587) or CEP-701 (KT-5555). Cancer Res. 1999; 59:2395–401. 52. S. Knapper, K.I. Mills, A.F. Gilkes, S.J. Austin, V. Walsh, and A.K. Burnett, The effects of lestaurtinib (CEP701) and PKC412 on primary AML blasts: the induction of cytotoxicity varies with dependence on FLT3 signaling in both FLT3-mutated and wild-type cases. Blood. 2006; 108:3494–503. 53. K.W. Yee, M. Schittenhelm, A.M. O’Farrell, A.R. Town, L. McGreevey, T. Bainbridge, J.M. Cherrington, and M.C. Heinrich, Synergistic effect of SU11248 with cytarabine or daunorubicin on FLT3 ITD-positive leukemic cells. Blood. 2004; 104:4202–9. 54. P. Brown, S. Meshinchi, M. Levis, T.A. Alonzo, R. Gerbing, B. Lange, R. Arceci, and D. Small, Pediatric AML primary samples with FLT3/ITD mutations are preferentially killed by FLT3 inhibition. Blood. 2004; 104:1841–9. 55. A.J. Mead, R.E. Gale, P.D. Kottaridis, S. Matsuda, A. Khwaja, and D.C. Linch, Acute myeloid leukemia blast cells with a tyrosine kinase domain mutation of FLT3 are less sensitive to lestaurtinib than those with a FLT3 internal tandem duplication. Br J Haematol. 2008; 141:454–60. 56. R. Grundler, C. Thiede, C. Miething, C. Steudel, C. Peschel, and J. Duyster, Sensitivity toward tyrosine kinase inhibitors varies between different activating mutations of the FLT3 receptor. Blood. 2003; 102:646–51. 57. J.J. Clark, J. Cools, D.P. Curley, J.C. Yu, N.A. Lokker, N.A. Giese, and D.G. Gilliland, Variable sensitivity of FLT3 activation loop mutations to the small molecule tyrosine kinase inhibitor MLN518. Blood. 2004; 104:2867–72. 58. E.V. Barry, J.J. Clark, J. Cools, J. Roesel, and D.G. Gilliland, Uniform sensitivity of FLT3 activation loop mutants to the tyrosine kinase inhibitor midostaurin. Blood. 2007; 110:4476-9. 59. B.D. Smith, M. Levis, M. Beran, F. Giles, H. Kantarjian, K. Berg, K.M. Murphy, T. Dauses, J. Allebach, and D. Small, Single-agent CEP-701, a novel FLT3 inhibitor, shows biologic and clinical activity in patients with relapsed or refractory acute myeloid leukemia. Blood. 2004; 103:3669–76. 60. S. Knapper, A.K. Burnett, T. Littlewood, W.J. Kell, S. Agrawal, R. Chopra, R. Clark, M.J. Levis, and D. Small, A phase 2 trial of the FLT3 inhibitor lestaurtinib (CEP701) as firstline treatment for older patients with acute myeloid leukemia not considered fit for intensive chemotherapy. Blood. 2006; 108:3262–70. 61. R.M. Stone, D.J. DeAngelo, V. Klimek, I. Galinsky, E. Estey, S.D. Nimer, W. Grandin, D. Lebwohl, Y. Wang, P. Cohen, E.A. Fox, D. Neuberg, J. Clark, D.G. Gilliland, and J.D. Griffin, Patients with acute myeloid leukemia and an activating mutation in FLT3 respond to a small-molecule FLT3 tyrosine kinase inhibitor, PKC412. Blood. 2005; 105:54–60. 62. D.J. DeAngelo, R.M. Stone, M.L. Heaney, S.D. Nimer, R.L. Paquette, R.B. Klisovic, M.A. Caligiuri, M.R. Cooper, J.M. Lecerf, M.D. Karol, S. Sheng, N. Holford, P.T. Curtin, B.J. Druker, and M.C. Heinrich, Phase 1 clinical results with tandutinib (MLN518), a novel FLT3 antagonist, in patients with acute myelogenous leukemia or high-risk myelodysplastic syndrome: safety, pharmacokinetics, and pharmacodynamics. Blood. 2006; 108:3674–81.
202
R.W. Stam and R. Pieters
63. M. Levis, R. Pham, B.D. Smith, and D. Small, In vitro studies of a FLT3 inhibitor combined with chemotherapy: sequence of administration is important to achieve synergistic cytotoxic effects. Blood. 2004; 104:1145–50. 64. L. Mollgard, S. Deneberg, H. Nahi, S. Bengtzen, K. Jonsson-Videsater, T. Fioretos, A. Andersson, C. Paul, and S. Lehmann, The FLT3 inhibitor PKC412 in combination with cytostatic drugs in vitro in acute myeloid leukemia. Cancer Chemother Pharmacol. 2008; 62:439–48. 65. S. Kasper, F. Breitenbuecher, Y. Hoehn, F. Heidel, D.B. Lipka, B. Markova, C. Huber, T. Kindler, and T. Fischer, The kinase inhibitor LS104 induces apoptosis, enhances cytotoxic effects of chemotherapeutic drugs and is targeting the receptor tyrosine kinase FLT3 in acute myeloid leukemia. Leuk Res. 2008; 32:1698–708. 66. M. Levis, P. Brown, B.D. Smith, A. Stine, R. Pham, R. Stone, D. Deangelo, I. Galinsky, F. Giles, E. Estey, H. Kantarjian, P. Cohen, Y. Wang, J. Roesel, J.E. Karp, and D. Small, Plasma inhibitory activity (PIA): a pharmacodynamic assay reveals insights into the basis for cytotoxic response to FLT3 inhibitors. Blood. 2006; 108:3477–83. 67. G.W. Krystal, Mechanisms of resistance to imatinib (STI571) and prospects for combination with conventional chemotherapeutic agents. Drug Resist Updat . 2001; 4:16–21. 68. M.W. Drummond, and T.L. Holyoake, Tyrosine kinase inhibitors in the treatment of chronic myeloid leukemia: so far so good? Blood Rev. 2001; 15:85–95. 69. S.H. Chu, and D. Small, Mechanisms of resistance to FLT3 inhibitors. Drug Resist Updat. 2009; 12:8–16. 70. F. Heidel, F.K. Solem, F. Breitenbuecher, D.B. Lipka, S. Kasper, M.H. Thiede, C. Brandts, H. Serve, J. Roesel, F. Giles, E. Feldman, G. Ehninger, G.J. Schiller, S. Nimer, R.M. Stone, Y. Wang, T. Kindler, P.S. Cohen, C. Huber, and T. Fischer, Clinical resistance to the kinase inhibitor PKC412 in acute myeloid leukemia by mutation of Asn-676 in the FLT3 tyrosine kinase domain. Blood. 2006; 107:293–300. 71. O. Piloto, M. Wright, P. Brown, K.T. Kim, M. Levis, and D. Small, Prolonged exposure to FLT3 inhibitors leads to resistance via activation of parallel signaling pathways. Blood. 2007; 109:1643–52. 72. T.M. Kohl, C. Hellinger, F. Ahmed, C. Buske, W. Hiddemann, S.K. Bohlander, and K. Spiekermann, BH3 mimetic ABT-737 neutralizes resistance to FLT3 inhibitor treatment mediated by FLT3-independent expression of BCL2 in primary AML blasts. Leukemia. 2007; 21:1763–72. 73. F. Breitenbuecher, B. Markova, S. Kasper, B. Carius, T. Stauder, F.D. Bohmer, K. Masson, L. Ronnstrand, C. Huber, T. Kindler, and T. Fischer, A novel molecular mechanism of primary resistance to FLT3-kinase inhibitors in acute myeloid leukemia. Blood. 2009; 113:4063-73. 74. U. Mony, M. Jawad, C. Seedhouse, N. Russell, and M. Pallis, Resistance to FLT3 inhibition in an in vitro model of primary AML cells with a stem cell phenotype in a defined microenvironment. Leukemia. 2008; 22:1395–401. 75. N.P. Shah, C. Tran, F.Y. Lee, P. Chen, D. Norris, and C.L. Sawyers, Overriding imatinib resistance with a novel ABL kinase inhibitor. Science. 2004; 305:399–401. 76. F. Heidel, D.B. Lipka, F.K. Mirea, S. Mahboobi, R. Grundler, R.K. Kancha, J. Duyster, M. Naumann, C. Huber, F.D. Bohmer, and T. Fischer, Bis(1H-indol-2-yl)methanones are effective inhibitors of FLT3-ITD tyrosine kinase and partially overcome resistance to PKC412A in vitro. Br J Haematol. 2009; 144:865–74.
Chapter 11
The Role of Tyrosine Kinase Inhibitors in the Treatment of ALL S. Wilson
11.1 Tyrosine Kinases Proteins kinases have been a fertile field for investigation in oncology as these proteins are intimately involved in the transduction of growth signals from receptor to nucleus. These enzymes catalyse the phosphorylation of proteins by transferring a terminal gamma phosphate from adenine triphosphate (ATP) to the free hydroxyl group of an amino-acid residue, via covalent bonding. The residue to which the phosphate is transferred determines the family of kinase, viz. serine/threonine or tyrosine kinases (TK). Phosphorylation is the most common post-translational modification event occuring to a protein within a cell and is utilised in all aspects of cellular processes as a switch, modulating specific functions of the effected protein. The human genome project has identified 518 protein kinases (kinome), of which 90 are classified as TK [1]. This family of proteins is vital for normal embryological development and in the homeostatic processes of adult cells. TKs play key roles in proliferation, migration, adhesion, metabolism, glucose utilisation, differentiation, cell survival and inter- and intracellular signalling [2]. Of the numerous kinases implicated in contributing to malignancies, over 45 are TK [3]. The underlying aetiology that leads to increased kinase activity may be due to chimeric fusion oncoproteins (BCR/ABL [4], HIP-1-PDGFRbeta [5] in CML), point mutations allowing disruption of kinase autoregulation (FLT3 D835 [6]), increased or aberrant expression of TK (IGF1R in osteosarcoma [7]) or decreased negative regulation of kinase activity, as evident in deletions or mutations of the tumour suppressor phosphatase PTEN seen in glioblastoma multiforme [8]. Some examples are given in Table 11.1. TK can be divided into two families, namely, receptor TK (RTK) and nonreceptor TK (Non-RTK), defined by their relationship with the cell membrane. The former consists of single-pass proteins that span the membrane, composed of an S. Wilson (*) School of Cancer Sciences, University of Birmingham, Birmingham, West Midlands, UK e-mail:
[email protected] V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3_11, © Springer Science+Business Media, LLC 2011
203
204
S. Wilson
Table 11.1 Examples of Receptor Tyrosine Kinases (TRKs) implicated in malignancies Tyrosine Kinase Malignancy References EGFR NSCLC, brain Dacic et al. [9], Shinojima et al. [10] ErbB2 – 4 Breast cancer, gastric, colorectal Lee et al. [11] c-KIT AML, melanoma LeFevre et al. [12], Tohda et al. [13] FGFR 1 AML (8p11 myeloproliferative Feng et al. [14], Macdonald et al. [15] syndrome), prostate FGFR3 MM Intini et al. [16] FLT3 AML, ALL Kottaridis et al. [17], Braoudki et al. [18] IGF1R Sarcomas, breast, prostate Sachdev [19], Werner [20] Eph A and B R Breast, hepatocellular carcinoma Zantek et al. [21] RET AML, thyroid carcinoma Santoro et al. [22], Takahashi et al. [23] PDGFR Brain tumours, GIST, leukemia Heinrich et al. [24], Fleming et al. [25], Smith et al. [26], Golub et al. [27] Axl AML Holland et al. [28], Rochlitz et al. [29]
extracellular domain, transmembranous region, regulatory juxtamembranous component, a relatively conserved catalytic kinase domain and c-terminus. In general, RTKs possess the facility to directly communicate extracellular growth factor signals to the cytoplasm, as they contain both the ligand-binding ectodomain and the cytoplasmic kinase domain (Fig. 11.1). Twenty classes of RTK have been described, dependent on the composition of the N-terminal external domain. Non-RTKs are cytoplasmic proteins that do not possess an extracellular or membrane-spanning domain and are predominantly involved in intracellular signalling. They rely on receptors from other families to transduce signals, e.g. JAK proteins are phosphorylated by cytokine receptors [30]. Phosphorylation may be induced by other cytoplasmic kinases Most TKs, besides the a2b2 tetrameric insulin receptor family, are monomeric proteins in their inactive state. Activation occurs on appropriate ligand binding to the receptor, which in turn leads to dimerisation and phosphorylation of specific tyrosine residues on the cytoplasmic component of the protein [31]. Each tyrosine kinase possesses an intrinsic kinase domain, which acts as the site for catalytic transfer of phosphate to tyrosine residues. The domain consists of sites for binding ATP and peptide substrate, and this function is tightly regulated. An important regulatory mechanism is the activation loop that acts as a gatekeeper, regulating access of the substrate to the peptide binding site. Juxtaposition of the kinase domains, which occurs after ligand-induced dimerisation, leads to trans (auto) phosphorylation of the activation loop, which adopts an “active” conformation allowing for ATP and substrate binding and subsequent phosphorylation of the kinase domain to occur. Activation of the intrinsic kinase domain increases its stability and catalytic efficiency, allowing for further phosphorylation of tyrosine residues on the C-terminus of the protein. These function as docking sites for signalling/adaptor proteins expressing src homology 2 (SH2) and phosphotyrosine binding (PTB) domains. The phosphorylation of these tyrosine sites provide a high-affinity binding site for intracellular kinases, leading to activation of intracellular
11 The Role of Tyrosine Kinase Inhibitors in the Treatment of ALL
205
Fig. 11.1 Mechanism of activation and downstream effectors of receptor tyrosine kinases: Proliferation, survival, migration, metabolism, differentiation
signal pathways, e.g. Akt/PI3K, MAPK, PKC, STAT. These pathways modulate cellular responses including pro-survival and apoptotic signalling, mechanisms for proliferation, metastasis and chemoresistance.
11.2 Tyrosine Kinases in Leukemia There is a wealth of information defining the role of aberrant kinase activity in the aetiology of chronic myeloid leukemia (CML) and solid tumours, as well as the impact pharmacological inhibition has on the management and outcome of these diseases. Dysregulation of the tyrosine kinase family has been proven to play an aetiological role in Philadelphia-positive disease and in paediatric ALL; other TKs are also emerging as key players, including FLT3 [18] and JAK2 [32]. The Philadelphia (9;22) translocation protein is demonstrated in 20–30% of adult ALL [33, 34] and in 4–5% of childhood ALL [35]. This chimeric oncoprotein is formed
206
S. Wilson
by the reciprocal translocation of the “break cluster region” gene on chromosome 22 and the non-receptor tyrosine kinase Abelson gene on chromosome 9. The resultant fusion gene codes for an oncogenic constitutively activated protein that increases the signalling of proliferative and pro-survival pathways including JAK/ STAT, Ras/Raf/MEK/ERK and PI3K/Akt [36–38]. The Src family of non-receptor TK has been implicated in the aetiology of Philadelphia-positive ALL [39]. These proteins are activated by kinase-independent direct interaction with the BCR/ABL protein and subsequently lead to phosphorylation of pro-survival downstream effectors, e.g. STAT5 and PI3K.
11.3 Tyrosine Kinase Inhibitors Tyrosine kinase inhibitors (TKIs) are a disparate group of compounds with a variety of chemical structures, united in their ability to inhibit protein kinases. The first kinase inhibitor staurosporine was isolated from Streptomyces staurosporus [40], which demonstrated a high affinity for binding and inhibiting protein kinases; however, due to its poor specificity, it is currently only utilised as a research tool. Other agents with the ability to inhibit TK were developed in following decade, including genistein and herbimycin. While able to target TK, these compounds did not display specificity within the tyrosine kinome. The first of the TKIs in clinical use, imatinib, was modelled on staurosporine’s ability to competitively inhibit ATP binding to sites in the intrinsic catalytic domain. Evidence suggests the mechanism of inhibition induced by this 2-phenylaminopyrimidine compound is allosteric in nature [41]. The drug binds to the “inactive form” of the BCR/ABL oncoprotein near the ATP-binding site and alters the conformation of the conserved DFG (Asp-Phe-Glyc) motif of the activation loop. This “DFG out” conformational change prevents ATP binding and prevents the transfer of the phosphate group to the substrate. This is believed to be the underlying mechanism for imatinib’s ability to inhibit c-KIT and PDGFR. Development of the second generation of TKIs was spurred on by the observation of a high degree of resistance to imatinib described in CML patients who had converted to the accelerated and blast phase of the disease [42]. The mechanisms of imatinib resistance is thought to primarily result from point mutations of residues found in the phosphate-binding loop (P loop), imatinib-binding site or activation loop [43–45]. Over 60 mutants that impair imatinib sensitivity have been described [46, 47]. The most notorious mutation is T315I, a substitution of isoleucine for threonine at amino acid 315, which completely precludes imatinib binding to the BCR/ABL protein [48, 49]. Other mechanisms of resistance include over-expression of the BCR/ABL protein [48] and increased expression of drug efflux pumps (p-glycoprotein) [50, 51]. In an attempt to overcome resistance of BCR/ABL mutants to kinase inhibition, Novartis AG developed the ABL, BCR/ABL, PDGFR and KIT inhibitor nilotinib
11 The Role of Tyrosine Kinase Inhibitors in the Treatment of ALL
207
(AMN-107). The original aminopyrimidine core of imatinib was maintained; however, moieties were altered in an attempt to increase the binding potency to the inactive ABL protein. The drug exhibits kinetic features of an ATP competitive inhibitor and demonstrates significantly increased potency, greater than 20 times imatinib, against Ba/F3 murine pro-B cell lines expressing non-mutated BCR/ABL as well as activity against PDGFR and c-KIT [52, 53]. Nilotinib has been demonstrated to have activity against a wide variety of imatinib-resistant clones, except against the T315I mutation [47]. Another approach to abrogating resistance in Philadelphia-positive disease was undertaken by Bristol-Myers Squibb (BMS) with the development of the dual Abl/ Src inhibitor dasatinib (BMS-354825). In Philadelphia-positive ALL, co-opting of BCR/ABL-independent pro-survival pathways, such as the Src family kinases (SFK), may explain the limited duration of remission and the early relapses seen in patients treated with imatinib monotherapy [54]. The basic structure of the dasatinib differs from imatinib and nilotinib as it is a thiazole carboxamide compound. This structure allows for combined BCR/ABL and Src family kinase inhibition (Src, Fyn, Yes and Lck), along with PDGFR a and b, c-KIT and ephrin A2 [55, 56]. Unlike the Novartis agents, dasatinib can bind to the active and inactive form of the BCR/ABL oncoprotein [57, 58]. In in vitro experiments, dasatinib demonstrates over 300 times the potency of imatinib against the unmutated forms of BCR/ ABL [45]. Dasatinib overcomes some causes of imatinib resistance by its ability to bind to clinically relevant mutations and its inherent SFK inhibition [55]. The combination of these two factors makes dasatinib an attractive agent in the treatment of Philadelphia-positive ALL. However, it does not exhibit activity against the T315I or F317L mutations [59]. A second dual BCR/ABL-Src inhibitor, bosutinib (SKI-606), has been developed by Wyeth Pharamceuticals, now Pfizer Inc. This novel 4-anilino-3-quinolinecarbonitrile compound was initially investigated as a Src kinase inhibitor; however, it exhibited concomitant Abl inhibition. The pattern of inhibition is consistent with ATP competitive inhibition [60]. In preclinical studies, bosutinib exhibited activity against cell lines that were resistant to imatinib due to mutations and BCR/ABL over-expression and is over 100 times as potent when compared to imatinib against wild-type BCR/ ABL [61]. Bosutinib does not demonstrate significant inhibition of class III RTKs, namely PDGFR or c-KIT, which may afford it a more attractive toxicity profile [61]. Unfortunately, like other Abl and Src-Abl inhibitors, the T315I mutation appears to be resistant to the drug [61, 62]. Bafetinib (INNO-406) is new dual Abl-Src family kinase inhibitor under development by Innovive Pharmaceuticals Inc., which is structurally related to imatinib and nilotinib. This compound demonstrates a unique pattern of action, inhibiting the Lyn kinase alone and no other members of the Src Family. In preclinical studies, it demonstrates 55 times more potency than imatinib in unmutated BCR/ABLexpressing cell systems [63, 64]. Cells expressing BCR/ABL mutants have been demonstrated to be sensitive to INNO-406, including the F317L and F317V mutations; however, the T351I mutation demonstrates INNO-406 resistance [65].
208
S. Wilson
Table 11.2 List of receptor tyrosine kinase inhibitors approved for use by Food and Drug Administration (FDA) – www.fda.org Drug Targets FDA-approved indications Chronic myeloid leukemia (CML) Dasatinib BCR/ABL, Src, PDGFR b, ephrin A2, c-KIT Erlotinib EGFR Lung and pancreatic cancer Gefitinib EGFR Lung cancer Imatinib BCR/ABL, PDGFRb, c-KIT CML Gastrointestinal stromal tumour (GIST) Lapatinib EGFR, HER2 Breast cancer Nilotinib BCR/ABL, PDGFRb, c-KIT CML Sorafenib VEGFR1 – 3, PDGFRb, Raf, c-KIT, FLT3 Hepatocellular carcinoma (HCC) Renal cell carcinoma (RCC) Sunitinib VEGFR 1 – 3, PDGFRb, c-KIT, RET, GIST, RCC CSF – 1, FLT3 Pazopanib VEGFR2, PDGFRb, c-KIT RCC
At present, all of the licenced TKIs are oral preparations; however, the absolute bioavailability data of these compounds is published only for imatinib, gefitinib and erlotinib. In general, the maximum plasma levels of TKIs are rapidly achieved, with the drug being quickly absorbed and Tmax ranging from 2 to 6 h [66–68]. Most of the compounds should be taken prior to eating, as absorption is enhanced in an acid environment; however, in practice, this does appear to make much practicable difference except in the case of nilotinib. The bioavailability of nilotinib is dramatically changed when taken with fatty food, an increase by 82% in bioavailability when taken 30 min after a fatty meal [69]. All TKIs are highly protein-bound, between 91 and 98% and most commonly to albumin and alpha-1-glycoprotein [70, 71]. The inhibitors in current clinical practice have long half-lives with the exception of dasatinib (3–5 h) [70]. They also demonstrate significant penetration of extracellular compartment, apart from the central nervous system. Distribution across the blood–brain barrier has not been described in all TKIs; however, in those that have been investigated, the degree of diffusion into the cerebrospinal fluid (CSF) appears limited with significantly lower levels in comparison with plasma levels [72]. One of the possible reasons for this limited diffusion is the finding that imatinib and nilotinib are targets of p-glycoprotein efflux pumps, which may actively extrude metabolites from the CSF [73]. TKIs demonstrate hepatic metabolism [74], via oxidation and hydroxylation, predominantly by the CYP3A4 system. Imatinib and nilotinib are capable of inhibiting the enzyme systems that metabolise them, which has the potential to alter subsequent metabolism [75]. Imatinib is metabolised into a N-demethylated piperazine derivative, which is as potent as the parent compound and requires significant dose alteration in hepatic reduction by 25%. Most of the metabolised drugs are faecally excreted, with a smaller amount renally excreted [68].
11 The Role of Tyrosine Kinase Inhibitors in the Treatment of ALL
209
11.4 Clinical Application of Tyrosine Kinase Inhibitors There are currently a significant number of early-phase trials being undertaken into the use of TKIs in leukemia (see www.clinicaltrials.gov). While the current focus of clinical research is on the impact TKIs exert on Philadelphia-positive disease, there is some evidence emerging for their use in Philadelphia negative disease. The earliest trials of imatinib in Philadelphia-positive ALL were commenced in the late 1990s. As part of a larger dose-escalation trial, patients with blast crisis CML and those with refractory or relapsed Philadelphia-positive ALL were enrolled to successive dose cohorts ranging from 300 to 1,000 mg [76]. Of 58 patients enrolled, 10 patients had CML with lymphoid blast crisis and 10 had Philadelphia-positive ALL, with no prior history of CML. In this combined cohort of 20 patients, a reduction of peripheral blasts was described in 70% of patients. A further four patients had evidence of a complete haematological response (CHR) (20%), as defined by reduction of blasts in marrow <5%, disappearance of blasts in peripheral blood and normalisation of the platelets and white cell count. Seven patients had a marrow response (35%), defined as a decrease in marrow blasts to either no more than 5% or between 5 and 15%, regardless of peripheral blast count. No difference in the response rate was exhibited between the patients with de novo ALL or those with previous chronic phase CML. These patients unfortunately relapsed by a median of 3 months. A phase 2 trial of imatinib enrolled 56 refractory or relapsed patients, 48 with Philadelphia positive ALL and 8 with CML in lymphoid blast crisis. The dosing of imatinib was variable, 400–800 mg, at the discretion of the investigator and depending on initial response [54]. Of the 48 Philadelphia-positive ALL patients, 9 (19%) achieved a CHR, 5 achieved (10%) a Complete Marrow Response, as defined by marrow blasts <5% without peripheral blood recovery, and 15 (31%) achieved a partial marrow response, with blasts <15% in blood and marrow. The median overall survival rate was 4.9 months, with a survival advantage for those who displayed evidence of some degree of response. Of the 5 patients who subsequently underwent allogeneic stem cell transplantation, 2 were still alive 10 months after commencing imatinib. These findings were confirmed by Wassman [77], with 70% of patients achieving a haematological response. Thirty per cent of patients achieved CHR and 29% achieved a marrow response. Once again, the duration of remission was shortlived with median relapse by 4 months. While these studies did show that a response was achievable in a disease as refractory as Philadelphia-positive ALL, the therapeutic use of imatinib monotherapy is limited due to the existence of primary resistance of approximately 30% and the rapid acquisition of secondary resistance during treatment. The first phase I trial investigating nilotinib was undertaken in a sample of 119 adults with imatinib-resistant CML or Philadelphia-positive ALL [66]. Individuals were assigned to a cohort of an escalating dose from 50 to 1,200 mg once daily or 400 to 600 mg twice daily. No dose-limiting toxicities were described in the cohorts under 600 mg daily and the drug was well tolerated. During Kantarjian’s study,
210
S. Wilson
nilotinib was proven to prolong the QTc interval and concerns regarding cardiac events in patients with pre-existing conditions and those receiving medication that prolongs the QTc have been raised. While the responses in patients with chronic and accelerated phases of CML were encouraging, of the 13 patients enrolled with Philadelphia-positive ALL, there was 1 complete molecular response and 1 partial haematological response. The potential role for nilotinib in ALL is uncertain, and further phase II trials need to be undertaken. As discussed previously, the underlying pathology in Philadelphia-positive ALL involves dysregulation of the Src family as well as expression of the Philadelphia oncoprotein. The first clinical dual Src/Abl kinase inhibitor to be investigated was dasatinib by Talpaz et al. [78] in a dose-escalation study of patients with Philadelphia-positive leukemia who were resistant or intolerant to imatinib. This phase 1 study enrolled 84 patients, 10 of whom had Philadelphia-positive ALL or CML in lymphoid blast crisis. Patients with chronic phase CML had varied dosing schedules between 15 and 240 mg daily, in either once or twice-daily dosing. All patients with Philadelphia-positive ALL were treated with twice-daily regimens. Major Hematologic responses were seen in 80% of these patients, with cytogenetic responses evident in 90% of these patients; however, the duration of response was short-lived. The median time to progression was 4 months. The drug was well tolerated, and although myelosuppression was the most commonly experienced significant toxicity, no patients withdrew as a result of drug-induced side effects. The Src-Abl Tyrosine kinase inhibition Activity Research Trials (START) were instituted to assess the efficacy of dasatinib in all phases of CML and Philadelphiapositive ALL in patients who were resistant or intolerant of imatinib. The START-L arm (CA180015) enrolled patients with lymphoid blast phase CML and de novo ALL. All acute leukemic patients had received extensive therapy prior to commencing 70 mg twice daily dasatinib [79]. All candidates had previously received imatinib and 15 (42%) had undergone allogeneic SCT [42]. Rapid responses were described with a major haematological response rate of 42% observed, of which 10 of 15 patients were progression free by 8 months. More impressively, significant cytogenetic responses were documented, with major cytogenetic responses observed in 58%. Mutational analysis was undertaken in 89% of the enrolled patients and 18 different BCR/ABL mutations were described in 25 patients, 24 of whom were resistant to imatinib. Besides the T315I mutation, which was present in a quarter of the patients analysed, dasatinib exhibited a good response rate against all leukemias bearing other imatinib-resistant mutations. The most common toxicity described was grade 3–4 cytopaenias; however, the aetiology of toxicities leading to interruption of dasatinib doses were nonhaematological in nature. Bosutinib is a specific dual Src-Abl inhibitor with very little PDGFR and c-KIT activity. This specificity makes it an attractive agent, and early drug trials support the purported lack of off-target effects. Phase I/II trials of bosutinib in the treatment of ALL have only been presented as abstracts thus far, as further trials are currently underway. The first part of the reported trial was a dose escalation in 18 subjects with chronic phase CML, who were resistant or intolerant to imatinib
11 The Role of Tyrosine Kinase Inhibitors in the Treatment of ALL
211
doses of 400–600 mg per day [80]. The recommended bosutinib dose for chronicphase CML was identified a dose of 500 mg once daily [80]. The second phase of the trial expanded the inclusion criteria to patients diagnosed with CML in accelerated phase and blast crisis and Philadelphia-positive ALL who had failed imatinib or other TKI therapy. Preliminary results of the expanded cohort of 57 patients, including 14 Philadelphia-positive ALL, were favourable [80]; 33% demonstrated CHR and 66% demonstrated major cytogenetic response. A recent update on 72 patients [81] demonstrated that an overall CHR was achieved in 18% of subjects and that by a median of 7.6 weeks a complete cytogenetic response of 22% was achieved. The most prevalent grade 3–4 toxicities were haematological in nature, with only 4% of patients experiencing grade 3–4 fluid retention, none of which are thought to be related to therapy. The safety and efficacy of bafetinib (INNO-406) is currently under investigation. The phase I trial initially enrolled 56 patients to receive 30–480 mg once daily. This was increased to twice-daily dosing on analysis of pharmacokinetic results [82, 83]. The drug was well tolerated until 480 mg twice-daily dosing where DLTs of liver dysfunction and thrombocytopaenia were described. No responses were described in the Philadelphia-positive ALL cohort, nor with accelerated or blast phase CML. The findings of the International Randomized Study of Interferon and imatinib (IRIS) trial confirmed the use of imatinib as the gold standard for chronic-phase CML [84] and the benchmark against which all the efficiency of all TKIs would be measured. In a disease as biologically diverse as ALL, it is highly unlikely that a single agent will be able to manage all aspects of the disease, and combining targeted therapies to inhibit overlapping pathways may be an attractive way forward. While there are a number of new potential targets, inhibition of these kinases, as yet, has not been translated into clinical practice. At present, the best defined role for TKIs in the management of ALL is as combination therapy in addition to conventional cytotoxics in an attempt to improve the remission rate in Philadelphia-positive ALL and prepare eligible patients for transplantation. The remission rate after induction for Philadelphia-positive ALL with cytotoxic agents alone is 80–90% [85–87]. With the addition of imatinib to a variety of intensive chemotherapeutic regimens, complete remission rates in excess of 90% have been described, and this translates into an increase in potential transplant candidates [88–92]. Concerns had been raised with regard to additional toxicity with the combination of imatinib and chemotherapy; however, in young subjects these fears seem to be unfounded. No additional significant toxicities were described when comparing combination therapy with historical controls [89]. A recent trial has demonstrated significant benefit derived in elderly patients treated with imatinib and steroids [93]. The 30 patients enrolled were treated with steroids, an initial 7-day pre-phase of 10 mg/m2, which increased to 40 mg/m2 in addition to 800 mg per day of imatinib for 45 days. An impressive CR rate of 100% for the 29 evaluable patients was reported, with a complete molecular response demonstrated in one patient. The median survival for all patients was 20 months, with a haematological remission time of 8 months.
212
S. Wilson
The main aim of treatment with combination therapy is to improve remission rates and, therefore, prepare patients for transplantation. In adult studies, the role of imatinib at induction and consolidation has demonstrated benefit, the use of imatinib as a maintenance agent in the post-stem-cell-transplant population is uncertain. Ribera has recently reported a small phase II trial of 30 patients in an attempt to answer this question [94]. Imatinib and multi-agent chemotherapy was utilised at induction and consolidation, with CR rates after induction 90% and complete molecular remission described in 21% of patients after induction and 56% after consolidation. Of the 27 patients to achieve CR, 21 were eligible and underwent stem cell transplantation of a variety of sources. The proposed duration of imatinib post-stem-cell transplantation was 1 year; however, this could only be administered for a median duration of 9 months. Therapy was interrupted due to relapse, graft-versus-host disease, grade 3–4 haematological and gastrointestinal toxicity and 1 non-relapse death. Median follow-up reported was 4.1 years, with median disease-free and overall survival of 30%. Evidence from a recent paediatric trial of patients with Philadelphia-positive ALL [95] has suggested the 3-year EFS of intensive chemotherapy with imatinib may be comparable to those who underwent matched related and matched unrelated transplants. There are currently only a few studies examining the efficacy of combining second-generation TKI and multi-agent chemotherapy strategies. Ravandi et al. has reported on a phase 2 trial of 22 treatment-naive patients and six previously treated Philadelphia-positive ALL patients, receiving 100 mg per day of dasatinib with alternating cycles of hyper-CVAD and high-dose cytarabine and methotrexate [96]. CR was achieved in 93% of evaluable patients after one cycle, of which 81% were a complete cytogenetic remission. With a median follow-up of 10 months, 21 patients were still alive and 18 were in CR. Of the 5 patients to relapse, 4 developed new ABL mutations, 3 of which were the multi-resistant T315I mutation. Another trial in an older population, median age 71 years, has also shown promise [97]. Twenty-two patients with de novo Philadelphia-positive ALL were treated with dasatinib 140 mg per day and vincristine and dexamethasone induction and achieved a complete marrow molecular remission rate of 28%. After consolidation with methotrexate, cytarabine and asparaginase, CHR rates of 95.2% were reported. While this is an impressive remission rate, 4 patients discontinued therapy after induction due to serious adverse events. This may be a limiting factor in the combination of second-generation TKIs and conventional chemotherapy.
11.5 Conclusions Striving for personalised medicine has become the accepted mantra of all haematooncologists treating patients in the 21st century. While we are still some distance off having an individualised treatment strategy for each patient, great strides have been made in stratifying therapy along biological lines. Historically, survival in
11 The Role of Tyrosine Kinase Inhibitors in the Treatment of ALL
213
the pre-imatinib era of Philadelphia-positive ALL was 10–15% with standard chemotherapy protocols, and this has been increased to 30–40%. Much like the early era of discovery of new conventional cytotoxics, we are still uncertain with regard to the full potential of combinations of TKIs. Interesting evidence has been demonstrated in preclinical work utilising combinations of TKIs and conventional cytotoxics. Nilotinib is known to be a multidrug resistance protein (p-glycoprotein) efflux pump substrate and has recently been demonstrated to inhibit the ATP-binding cassette (ABC) proteins, ABCB1 and ABCG2 [98]. In vitro data demonstrated that the combination of nilotinib with conventional agents (anthracyclines, vinca alkaloids and taxanes) enhanced both the intracellular concentrations and cytotoxicity of these agents. The possibilities of synergistic anti-neoplastic effects of TKI combinations have not been fully explored. Intracellular concentrations of dasatinib have been shown to increase when used in conjunction with nilotinib [99], which may raise the inhibitory concentration of dasatinib above the threshold required to inhibit Abl kinases bearing mutations resistant to current therapy. Preclinical evidence has demonstrated that the inhibition of the serine/threonine aurora kinase blocks cell-cycle progression and induces apoptosis in a variety of cell lines, including those with the first- and second-generation TKI-resistant mutations. Most aurora kinases developed so far appear to have a significant “off target” effect, including Abl, FLT3 and Janus kinase 2 (JAK2) inhibition to varying degrees. The multikinase inhibitors AT9283, danusertinib (PHA-739358) and XL-228 are currently being evaluated in early-phase drug trials and all have demonstrated objective responses against subjects with proven imatinib-resistant mutations [100–102]. New TKIs are being developed for treatment of solid tumours, myelodysplastic syndromes and acute myeloid leukemia, all of which appear to be driven by receptor and nonreceptor TKs, and research into childhood ALL has given some tantalising insights that may prove to be attractive therapeutic targets in the future. Genomewide analysis of children with Philadelphia-negative precursor-B ALL demonstrated alteration of IKZF1, a gene coding for the IKAROS transcription factor, important for B-cell differentiation [103]. Interestingly, the gene expression profile for these patients was similar to those with observed in association with BCR/ABL, and accordingly this cohort had a poor clinical outcome. Further investigation of these high-risk patients demonstrated activating mutations in the JAK family [32]. Mutations of JAK have also been described in approximately 20% of adults with T-ALL, which are also associated with poor response to treatment and a reduction in overall prognosis [104]. Small molecules that target the JAK family are under investigation in AML and myeloproliferative disease [105–107]. These compounds have shown promise with clinical improvement seen in myelofibrosis and objective response in those with activating JAK mutations [108]. Clinical trials using these agents in ALL patients with JAK mutations need to be undertaken. Another TK of interest is Fms-like tyrosine kinase 3 (FLT3), a receptor involved in normal B-cell development and is expressed on the majority of ALL, both B and T lineage [109].
214
S. Wilson
Targeting TK is clearly an important therapeutic option for many types of leukemias, and the exploration of their potential development in acute leukemias is its infancy. The rapidly increasing understanding of the relative roles of different types of TK and how they interact in driving the leukemic phenotype will be important in informing intelligent combinations of their inhibitors in clinical practice.
References 1. Manning G, Whyte D, Martinez R, et al. The protein kinase complement of the human genome. Science. 2002; 298(5600): 1912–1934. 2. Vlahovic G, Crawford J. Activation of tyrosine kinases in cancer. Oncologist. 2003; 8: 531–538. 3. Blume – Jensen P, Hunter T. Oncogenic kinase signalling. Nature. 2001; 411: 355–356. 4. Nowell P, Hungerford D. A minute chromosome in human chronic granulocytic leukemia. Science. 1960; 132: 1497. 5. Ross TS, Bernard OA, Berger R, et al. Fusion of huntingtin interacting protein 1 to plateletderived growth factor beta receptor (PDGFbetaR) in chronic myelomonocytic leukemia with t(5;7)(q33;q11.2). Blood. 1998; 91(12): 4419–4426. 6. Nakao M, Yokota S, Iwai T, et al. Internal tandem duplication of the flt3 gene found in acute myeloid leukemia. Leukemia. 1996; 10(12): 1911–1918. 7. Burow S, Andrulis I, Pollak M, et al. Expression of insulin-like growth factor receptor, IGF-1, and IGF-2 in primary and metastatic osteosarcoma. J. Surg. Oncol. 1998; 69: 21–27. 8. Rasheed BK, Stenzel TT, McLendon RE, et al. PTEN gene mutations are seen in high-grade but not in low-grade gliomas. Cancer Res. 1997; 57: 4187–4190. 9. Dacic S, Flanagan M, Cieply K, et al. Significance of EGFR protein expression and gene amplification in Non – Small Cell Lung Carcinoma. Am Journal Clinical Pathology. 2006; 1256: 860–865. 10. Shinojima N, Tada K, Shirasishi S, et al. Prognostic Value of Epidermal Growth Factor Receptor in Patients with Gliobastoma Multiforme. Cancer Res. 2003; 63: 6962–6970. 11. Lee JW, Soung YH, Seo SH, et al. Somatic Mutations of ERBB2 Kinase Domain in Gastric, Colorectal and Breast Carcinomas. Clin Cancer Res. 2006; 12: 57–61. 12. LeFevre G, Glotin AL, Calipel A, et al. Roles of stem cell factor/c-KIT and effects of Glivec/ STI571 in human uveal melanoma cell tumorigenesis. J Biol Chem. 2004; 30: 31769–31779. 13. Tohda S, Yang GS, Ashman LK, et al. Relationship between c-KIT expression and proliferation in acute myeloblastic leukemia cell lines. J Cell Physiol. 1993; 154: 410–418. 14. Feng S, Wang F, Matsubara A, et al. Fibroblast growth factor receptor 2 limits and receptor 1 accelerates tumorigenicity of prostate epithelial cells. Cancer Res. 1997; 57: 5369–5378. 15. Macdonald D, Reiter A, Cross N. The 8p11 myeloproliferative syndrome: a distinct clinical entity caused by constitutive activation of FGFR1. Acta Haematol. 2002; 107: 101–107. 16. Intini D, Baldini L, Fabris S, et al. Analysis of FGFR3 gene mutations in multiple myeloma patients with t(4;14). Br J Haematol. 2001; 114: 362–364. 17. Kottaridis PD, Gale RE, Frew ME, et al. The presence of a FLT3 internal tandem duplication in patients with acute myeloid leukemia (AML) adds important prognostic information to cytogenetic risk group and response to the first cycle of chemotherapy: analysis of 854 patients from the United Kingdom Medical Research Council AML 10 and 12 trials. Blood. 2001; 98: 1752–1759. 18. Braoudki M, Karpusas M, Katsibardi K, et al. Frequency of FLT3 mutations in childhood lymphoblastic leukemia. Medical Oncology. 2009; 26: 460–462.
11 The Role of Tyrosine Kinase Inhibitors in the Treatment of ALL
215
19. Sachdev D, Yee D. Disrupting insulin – like growth factor signalling as a potetnital cancer therapy. Mol Cancer Ther. 2007; 6: 1–12. 20. Werner H, Le Roith D. The role of the insulin – like growth factor system in human cancer Adv Cancer Res. 1996; 68: 183–223. 21. Zantek ND, Azimi M, Fedoro-Chaiken M, et al. E-cadherin regulates the function of the EphA2 receptor tyrosine kinase. Cell Growth Differ. 1999; 10: 629–638. 22. Santoro M, Rosati R, Grieco M, et al. The ret proto-oncogene is consistently expressed in human pheochromocytomas and thyroid medullary carcinomas. Oncogene. 1990; 5: 1595–1598. 23. Takahashi M, Buma Y, Taniguchi M. Identification of the ret proto-oncogene products in neuroblastoma and leukemia cells. Oncogene. 1991; 6: 297–301. 24. Heinrich MC, Corless CL, Duensing A, et al. PDGFRA activating mutations in gastrointestinal stromal tumors. Science. 2003; 299: 708–710. 25. Fleming TP, Saxena A, Clark WC, et al. Amplification and/or overexpression of plateletderived growth factor receptors and epidermal growth factor receptor in human glial tumors. Cancer Res. 1992; 52: 4550–4553. 26. Smith JS, Wang XY, Qian J, et al. Amplification of the platelet derived growth factor receptor-A (PDGFRA) gene occurs in oligodendrogliomas with grade IV anaplastic features. J. Neuropathol. Exp. Neurol. 2000; 59: 495–503. 27. Golub TR, Barker GF, Lovett M, et al. Fusion of PDGF receptor beta to a novel ets-like gene, tel, in chronic myelomonocytic leukemia with t(5;12) chromosomal translocation. Cell. 1999; 77: 307–316. 28. Holland SJ, Powell MJ, Franci C, et al. Multiple Roles for the Receptor Tyrosine Kinase AXL in Tumor Formation. Cancer Res. 2005; 65: 9294–9303. 29. Rochlitz C, Lohri A, Bacchi M, et al. Axl expression is associated with adverse prognosis and with expression of Bcl-2 and CD34 in de novo acute myeloid leukemia (AML): results from a multicenter trial of the Swiss Group for Clinical Cancer Research (SAKK). Leukemia. 1999; 13: 1352–1358. 30. Behrmann I, Smyczek T, Heinrich PC, et al. Janus kinase (Jak) subcellular localization revisited: the exclusive membrane localization of endogenous Janus kinase 1 by cytokine receptor interaction uncovers the Jak receptor complex to be equivalent to a receptor tyrosine kinase. J Biol Chem. 2004; 279: 35486–35493. 31. Ullrich A, Schlessinger J. Signal transduction by receptors with tyrosine kinase activity. Cell. 1990; 61: 203–212. 32. Mullighan CG, Zhang J, Harvey RC, et al. JAK mutations in high-risk childhood acute lymphoblastic leukemia. Proc Natl Acad Sci U S A. 2009; 106(23): 9414–9418. 33. Rieder H, Bonwetsch C, Janssen LA, et al. High rate of chromosome abnormalities detected by fluorescence in situ hybridization using BCR and ABL probes in adult acute lymphoblastic leukemia. Leukemia. 1998; 12: 1473–1481. 34. Secker-Walker LM, Craig JM, Hawkins JM, et al. Philadelphia positive acute lymphoblastic leukemia in adults: age distribution, BCR breakpoint and prognostic significance. Leukemia. 1991; 5: 196–199. 35. Pui CH, Crist WM and Look T. Biology and clinical significance of cytogenetic abnormalities in childhood acute lymphoblastic leukemia. Blood. 1990; 76: 1449–1463. 36. Neshat MS, Raitano AB, Wang HG, et al. The survival function of the Bcr-Abl onocogene is mediated by Bad-dependent and -independent pathways: roles for phosphatidylinositol 3-kinase and Raf. Mol Cell Biol. 2000; 20: 1179–1186. 37. Lin TS, Mahajan S, Frank DA. STAT signaling in the pathogenesis and treatment of leukemias. Oncogene. 2000; 19: 2496–2504. 38. Skorski T, Nieborowska-Skorska M, Szczylik C, et al. c-RAF-1 Serine/Threonine Kinase Is Required in BCR/ABL-dependent and Normal Hematopoiesis. Cancer Res. 1995; 55: 2275–2278. 39. Hu Y, Pelletier S, Buchdunger E, et al. Requirement of Src kinases Lyn, Hck and Fgr for BCR-ABL1-induced B-lymphoblastic leukemia but not chronic myeloid leukemia. Nat Genet. 2004; 36(5): 453–461.
216
S. Wilson
40. Omura S, Iwai Y, Hirano A, et al. A new alkaloid AM-2282 of Streptomyces origin. Fermentation, isolation and preliminary characterization. J. Antibiotics. 1977; 30: 275. 41. Buchdunger E, Zimmerman J, Mett H, et al. Inhibition of the Abl-protein – tyrosine kinase in vitro and in vivo by a 2 – phenylaminopyrimidine derivative. Cancer Res.1996; 56(1): 100–104. 42. Ottman O, Dombret H, Martinelli G, et al. Dasatinib induces rapid hematologic and cytogenetic responses in adult patients with Philadelphia chromosome-positive acute lymphoblastic leukemia with resistance or intolerance to imatinib: interim results of a phase 2 study. Blood. 2007; 100(7): 2309–2315. 43. Carter TA, Wodicka LM, Shah NP, et al. Inhibition of drug-resistant mutants of ABL, KIT, and EGF receptor kinases. Proc Natl Acad Sci U S A. 2005; 102(31): 11011–11016. 44. Shah NP, Nicoll JM, Nagar B, et al. Multiple BCR-ABL kinase domain mutations confer polyclonal resistance to the tyrosine kinase inhibitor imatinib (STI571) in chronic phase and blast crisis chronic myeloid leukemia Cancer Cell. 2002; 2: 117–125. 45. O’Hare T, Walters DK, Stoffregen EP et al. In vitro activity if Bcr-Abl inhibitors AMN-107 and BMS – 354825 against clinically relevant imatinib – resistant Abl kinase domain mutants. Cancer Res. 2005; 65: 4500–4505. 46. Hochhaus A, Kreil S, Corbin AS, et al. Molecular and chromosomal mechanisms of resistance to imatinib (STI571) therapy. Leukemia. 2002; 16(11): 2190–2196. 47. Weisberg E, Manley PW, Breitenstein W, et al. Characterization of AMN107, a selective inhibitor of native and mutant BCR/ABL. Cancer Cell. 2005; 7 : 129–141. 48. Gorre ME, Mohammed M, Ellwood K, et al. Clinical resistance to STI571 cancer therapy caused by BCR-ABL gene mutation or amplification. Science. 2001; 293: 876–80. 49. Ricci C, Scappini B, Divoky V, et al. Mutation in the ATP-binding Pocket of the ABL Kinase Domain in an STI571-resistant BCR/ABL-positive Cell Line. Cancer Res. 2002; 62(21): 5995–5998 50. Mahon FX, Deininger MW, Schultheis B, et al. Selection and characterization of BCR-ABL positive cell lines with differential sensitivity to the tyrosine kinase inhibitor STI571: diverse mechanisms of resistance. Blood. 2000; 96: 1070–1079. 51. Illmer T, Schaich M, Platzbecker U, et al. Pglycoprotein- mediated drug efflux is a resistance mechanism of chronic myelogenous leukemia cells totreatment with imatinib mesylate. Leukemia. 2004; 18: 401–408. 52. Manley PW, Cowan-Jacob SW, Mestan J. Advances in the structural biology, design and clinical development of Bcr-Abl Kinase inhibitors for the treatment of chronic myeloid leukemia. Biochim. Biophys. Acta. 2005; 1754: 3–13. 53. Verstovsek S, Akin C, Manshouri T, et al. Effects of AMN107, a novel aminopyrimidine tyrosine kinase inhibitor, on human mast cells bearing wild-type or mutated codon 816 c-kit. Leukemia Res. 2006; 30: 1365–1370. 54. Ottman OG, Druker BJ, Sawyers CL, et al. A phase 2 study of imatinib in patients with relapsed or refractory Philadelphia chromosome-positive acute lymphoblastic leukemia. Blood. 2002; 100: 1965–1971. 55. Lombardo LJ, Lee FY, Chen P et al. Discovery of N-(2-chloro-6-methyl-phenyl)-2-(6-(4-(2 hydroxyethyl)-piperazin-1-yl)-2-methylpyrimidin-4-ylamino) thiazole-5-carboxamide (BMS-354825), a dual Src/Abl kinase inhibitor with potent antitumor activity in preclinical assays. J Med Chem. 2004; 47: 6658–6661. 56. Huang F, Reeves K, Han X, et al. Identification of candidate molecular markers predicting sensitivity in solid tumors to dasatinib: rationale for patient selection. Cancer Res. 2007; 67: 2226–2238. 57. Tokarski JS, Newitt JA, Chang CY, et al. The structure of dasatinib (BMS-354825) bound to activated ABL kinase domain elucidates its inhibitory activity against imatinib-resistant ABL mutants. Cancer Res. 2006; 66: 5790–5797. 58. Gambacorti-Passerini C, Gasser M, Ahmed S et al. Abl inhibitor BMS354825 binding mode in Abelson kinase revealed by molecular docking studies. Leukemia. 2005; 19: 1267–1269.
11 The Role of Tyrosine Kinase Inhibitors in the Treatment of ALL
217
59. Soverini S, Colarossi S, Gnani A, et al. Resistance to dasatinib in Philadelphia-positive leukemia patients and the presence or the selection of mutations at residues 315 and 317 in the BCR-ABL kinase domain. Hematologica. 2007; 92(3): 401–404. 60. Golas JM, Arndt K, Etienne C, et al. SKI-606, a 4 – Anilino – 3 – quinolinecarbonitrile Dual Inhibitor of Src and Abl Kinases, Is a Potent Antiproliferative Agent against Chronic Myelogenous Leukemia Cells in Culture and Causes Regression of K652 Xenografts in Nude Mice. Cancer Res. 2003. 61. Puttini M, Coluccia AML, Boschelli F, et al. In vitro and in vivo activity of SKI-606, a novel Src-Abl inhibitor, against imatinib – resistant Bcr-Abl+ neoplastic cells. Cancer Res. 2006; 66(23): 11314–11322. 62. Soverini S, Tasco T, Grafone S, et al. Binding mode of the tyrosine kinase inhibitor bosutinib (SKI-606) to Abl kinase. J Clin Oncol (2007) ASCO Annual Meeting Proceedings (PostMeeting Edition). Vol 25, No 18S (June 20 Supplement), 2007: abstract 7049. 63. Kimura S, Naito H, Segawa H, et al. NS-187, a potent and selective dual Bcr-Abl/Lyn tyrosine kinase inhibitor, is a novel agent for imatinib-resistant leukemia. Blood. 2005; 106: 3948–3954. 64. Niwa T, Asaki T, Kimura S. NS-187 (INNO-406), a Bcr-Abl dual tyrosine kinase inhibitor. Anal Chem Insights. 2007; 2: 93–106. 65. Deguchi Y, Shinya K, Ashihara E, et al. Compartison of imatinib, dasatinib, nilotinib and INNO-406 in imatinib – resistant cell lines. Leuk Res. 2008; 32(6): 980–983. 66. Kantarjian H, Giles F, Wunderle L, et al. Nilotinib in Imatinib-Resistant CML and Philadelphia Chromosome – Positive ALL. NEJM. 2006; 354(24): 2542–2551. 67. Christopher LJ, Cui D, Wu C, et al. Metabolism and disposition of dasatinib after oral administration to humans. Drug Metab Dispos. 2008, 36(7): 1357–1364. 68. Gschwind HP, Pfaar U, Waldmeier F, et al. Metabolism and disposition of imatinib mesylate in healthy volunteers. Drug Metab Dispos. 2005; 33 (10): 1503–1512. 69. Tanaka C, Smith T, KantarJlan H, et al. Clinical pharmacokinetics (PK) of AMN107, a novel inhibitor of Bcr-Abl, In healthy subjects and patients with Imatinib resistant or intolerant chronic myelogenous leukemia (CML) or relapsed/refractory Ph+ acute lymphocytic leukemia (Ph+ ALL). J Clin Oncol. 2006; 24(Suppl 18): b3095. Abstract. 45. 70. Kamath AV, Wang J, Lee FY, et al. Preclinical pharmacokinetics and in vitro metabolism of dasatinib (BMS-354825): a potent oral multi-targeted kinase inhibitor against SRC and BCR-ABL. Cancer Chemother Pharmacol. 2008; 61 (3): 365–376. 71. Gambacorti-Passerini C, Zucchetti M, Russo D et al., Alpha1 acid glycoprotein binds to imatinib (STI571) and substantially alters its pharmacokinetics in chronic myeloid leukemia patients. Clin Cancer Res. 2003; 9(2): 625–632. 72. Neville K, Parise RA, Thompson P, et al. Plasma and Cerebrospinal Fluid Pharmacokinetics of Imatinib after Administration to Nonhuman Primates. Clin Cancer Res. 2004; 10: 2525–2529. 73. Dai H, Marbach P, Lemaire M, et al. Distribution of STI-571 to the brain is limited by P-glycoprotein-mediated efflux. J Pharmacol Exp Ther. 2003; 304(3): 1085–1092. 74. Kagan M, Tran P, Fischer V, et al. Safety, pharmacokinetics (PK), metabolism, and mass balance of [14C]-AMN1 07, a novel amlnopyrimidine inhibitor of Bcr-Abl tyrosine kinase, in healthy subjects. Blood. 2005; 106: 4887. 75. van Erp NP, Gelderblom H, Karlsson MO, et al. Influence of CYP3A4 inhibition on the steady-state pharmacokinetics of imatinib. Clin Cancer Res. 2007; 13(24): 7394–400. 76. Druker BJ, Sawyers CL, Kantarijan H, et al. Activity of a specific inhibitor of the BCR-ABL tyrosine kinase in the blast crisis of chronic myeloid leukemia and acute lymphoblastic leukemia with the Philadelphia chromosome. N Eng J Med. 2001; 344: 1038–1042. 77. Wassman B, Pfeifer H, Scheuring UJ, et al. Early prediction of response in patients with relapsed or refractory Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph+ ALL) treated with imatinib. Blood. 2004; 103: 1495–1498. 78. Talpaz M, Shah N, Kantarjian H, et al. Dastinib in Imatinib-Resistant Philadelphia Chromosome-Positive Leukemias. N Eng J Med. 2006; 354(24): 2351–2541.
218
S. Wilson
79. Dombret H, Ottman OG, Rosti G, et al. Dasatinib (SPRYCEL) in patients (pts) with Philadelphia chromosome-positive acute lymphoblastic leukemia who are imatinib-resistant (im-r) or –intolerant (im-i): updated results from the CA180-015 START-L study. Blood. (Suppl.) 2006; 108: 286. 80. Gambacorti-Passerini C, Kantarjian H, Bruemmendorf T, et al. Bosutinib (SKI-606) Demonstrates Clinical Activity and Is Well Tolerated among Patients with AP and BP CML and Ph+ ALL. Blood (ASH Annual Meeting Abstracts). Nov 2007; 110: 473. 81. Gambacorti-Passerini C, Kantarjian HM, Baccarani M, et al. Activity and tolerance of bosutinib in patients with AP and BP CML and Ph+ ALL. Journal of Clinical Oncology. 2008 ASCO Annual Meeting Proceedings (Post-Meeting Edition). 82. Kantarjian HM, Cortes J, le Coutre P, et al. A Phase I Study of INNO-406 in Patients with advanced Philadelphia (Ph+) Chromosome-Positive Leukemias Who Are Resistant or Intolerant to Imatinib and Second Generation Tyrosine Kinase Inhibitors. Blood (ASH Meeting Abstracts). 2007; 110: abstract 469. 83. Kantarjian H, le Coutre P, Cortes J, et al. Phase I study of INNO – 406, a dual Abl/Lyn kinase inhibitor, in Philadelphia chromosome – positive leukemias after imatinib resistance or intolerance. Cancer. 2010;116: 2665–2672. 84. O’Brien SG, Guilhot F, Larson RA, et al. Imatinib compared with interferon and low-dose cytarabine for newly diagnosed chronic-phase chronic myeloid leukemia. N Engl J Med. 2003; 348: 994–1004. 85. Secker-Walker LM, Prentice HG, Durrant J, et al. Cytogenetics adds independent prognostic information in adults with acute lymphoblastic leukemia on MRC trial UKALL XA – MRC Adult Leukemia Working Party. Br J Haematol. 1997; 96: 601–610. 86. Wetzler M, Dodge RK, Mrozek K, et al. Prospective karyotype analysis in adult acute lymphoblastic leukemia: The Cancer and Leukemia Group B experience. Blood. 1999; 93: 3983–3993. 87. Annino L, Vegna ML, Camera A, et al. Treatment of adult acute lymphoblastic leukemia (ALL): Long-term follow-up of the GIMEMA ALL 0288 randomized study. Blood. 2002; 99: 863–871. 88. Thomas DA, Kantarjian HM, Cortes JE, et al. Outcome after frontline therapy with the hyper-CVAD and imatinib mesylate regimen for adults with de novo or minimally treated Philadelphia (Ph) positive acute lymphoblastic leukemia (ALL). JCO ASCO Annual Meeting Proceedings (Post-Meeting Edition) 2008; 26, 15S:7019. 89. Yanada M, Takeuchi J, Sugiura I, et al. High Complete Remission Rate and Promising Outcome by Combination of Imatinib and Chemotherapy for Newly Diagnosed BCR-ABLPositive Acute Lymphoblastic Leukemia: A Phase II Study by the Japan Adult Leukemia Study Group. J Clin Oncol. 2006; 24(3): 460–466. 90. Wassman B, Pfeifer H, Goekbuget N, et al. Alternating versus concurrent schedules of imatinib and chemotherapy as front-line therapy for Philadelphia-positive acute lymphoblastic leukemia (Ph+ ALL). Blood. 2006; 108: 1469–1477. 91. Chalandon Y, Thomas X, Hayette S, et al. First results of the GRAAPH-2005 study in younger adult patients with de novo Philadelphia positive acute lymphoblastic leukemia [abstract]. Blood. 2008; 112: 11. 92. de Labarthe A, Rousselot P, Huguet-Rigal F, et al. Imatinib combined with induction or consolidation chemotherapy in patients with de novo Philadelphia chromosome-positive acute lymphoblastic leukemia: results of the GRAAPH-2003 study. Blood. 2007; 109(4): 1480–1413. 93. Vignetti M, Fazi P, Cimino G, et al. Imatinib plus steroids induces complete remissions and prolonged survival in elderly Philadelphia chromosome-positive patients with acute lymphoblastic leukemia without additional chemotherapy: results of the Gruppo Italiano Malattie Ematologiche dell’Adulto (GIMEMA) LAL0201-B protocol. Blood. 2007; 109(9): 3676–3678. 94. Ribera J-M, Oriol A, Gonzalez M, et al. Concurrent intensive chemotherapy and imatinib before and after stem cell transplantation in newly diagnosed Philadelphia-positive acute lymphoblastic leukemia. Final results of the CSTIBES02 trial Hematologica. 2010; 95(1): 87–95.
11 The Role of Tyrosine Kinase Inhibitors in the Treatment of ALL
219
95. Schultz KR, Bowman WP, Aledo A, et al. Improved early event-free survival with imatinib in Philadelphia chromosome–positive acute lymphoblastic leukemia: A Children’s Oncology Group study. J Clin Oncol. 2009; 27: 5175–5181. 96. Ravandi F, Thomas D, Kantarjian H, et al. Pjase II study of combination of hyperCVAD in frontline therapy of patients with Philadelphia Chromosome (Ph) Positive Acute Lymphoblastic Leukemia (ALL). Blood (ASH Annual Meeting Abstracts). 2008; 112. abstract 2921. 97. Rousselot P, Cayuela J-M, Recher C, et al. Dasatinib (Sprycel (R) and chemotherapy for First-Line Treatment in Elderly Patients with De Novo Philadelphia Positive ALL: Results of the First 22 Patietns Included in the EWALL-Ph-01 Trial (on Behalf of the European Working Group on Adult ALL (EWALL)). Blood (ASH Annual Meeting Abstracts). 2008; 112: abstract 2920. 98. Tiwari AK, Sodani K, Wang SR, et al. Nilotinib (AMN-107, Tasigna) reverses multidrug resistnce by inhibiting the activity of the ABCB1.Pgp and ABCG2/BCRP/MXR transporters. Biochemical Pharmacology. 2009; 78(2): 153–161. 99. Hiwase DK, White D, Zrim S, et al. Nilotinib – mediated inhibition of ABCB1 increases intracellular concentration of Dasatinib in CML cells: implications for combination TKI therapy. Leukemia. 2010; 24: 658–660. 100. Cortes-Franco J, Dombret H, Schafhausen P, et al. Danusertib hydrochloride (PHA-739358), a multi-kinase Aurora inhibitor, elicits clinical benefit in advanced myeloid leukemia and Philadelphia chromosome positive acute lymphoblastic leukemia. Blood (ASH Annual Meeting Abstracts). 2009; 114: abstract 864. 101. Foran JM, Ravandi F, O’Brien SM, et al, Phase I and pharmacodynamic trials of AT9283, an aurora kinase inhibitor, in refractory leukemia. J Clin Oncol ASCO Annual Meeting Proceedings (Post - Meeting Edition). 2008; 26: abstract 2518. 102. Cortes J, Paquette R, Talpaz M, et al. Preliminary clinical activity in a phase I trial of the BCR-ABL/IGF-1R/Aurora kinase inhibitor XL228 in patients with Ph++ leukemias with either failure to multiple TKI therapies or with T315I mutation. Blood (ASH Annual Meeting Abstracts). 2008; 112(11): abstract 3232. 103. Mullighan CG, Goorha S, Radtke I, et al. Genome – wide analysis of genetic alterations in acute lymphoblastic leukaemia. Nature. 2007; 446: 758–764. 104. Flex E, Petrangeli V, Stella L, et al. Somatically acquired JAK1 mutations in adult acute lymphoblastic leukemia. J Exp Med, 2008; 205(4): 751–758. 105. Verstovsek S, Kantarjian H, Pardanani A, et al. INCB018424, an oral, selective JAK2 inhibitor, shows significant clinical activity in a phase I/II study in patients with primary myelofibrosis (PMF) and post polycythemia vera/essential thrombocythemia myelofibrosis (post-PV/ ET MF) Blood. 2007; 110: abstract 558. 106. Verstovsek S, Kantarjian, HM, Pardanani AD, et al. The JAK inhibitor, INCB018424, demonstrates durable and marked clinical responses in primary myelofibrosis (PMF) and postpolycythemia/essential thrombocythemia myelofibrosis (post PV/ET-MF) Blood. 2008; 112: abstract 1762. 107. Moliterno AR, Roboz GJ, Carroll M, et al. An open-label study of CEP- 701 in patients with JAK2 V617F-positive polycythemia vera and essential thrombocytosis. Blood. 2008; 112: abstract. 99. 108. Verstovsek S, Kantarjian H, Mesa RA, et al. Long – Term Follow up and Optimized Dosing Regimen of INCB018424 in Patients with Myelofibrosis: Durable Clinical, Functional and Symptomatic Responses with Improved Hematological Safety. 51st American Society of Hematology Annual Meeting. 2009; abstract 756. 109. Drexler HG. Expression of FLT3 receptor and response to FLT3 ligand in leukemic cells Leukemia. 1996; 10(4): 588–599.
Chapter 12
Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia Arend von Stackelberg
12.1 History In 1975, Kohler et al. produced the first monoclonal antibodies (moAbs) by fusing a myeloma cell line with a specific antibody-producing B cell. This combined the unlimited growth potential of myeloma cells with the predetermined antibody specificity of normal immune spleen cells from an immunized mouse. The technique called somatic cell hybridization results in a hybridoma [1]. Humans receiving murine moAbs produce human anti-mouse antibodies (HAMA) leading to inactivation of the moAbs and allergic complications. The chimerization or humanization of moAbs via innovative recombinant DNA technology leads to a better tolerance to the compounds and has allowed for using the natural effector mechanisms of destroying with moAb-coated targets [2]. Whereas chimeric moAbs have antigenbinding parts (variable regions) of the mouse antibody and the effector parts (constant region, Fc) of a human antibody, in humanized compounds the mousederiving part of the moAb is reduced to the antigen-binding site (hypervariable region). The methodological approach to engineer the composition of moAbs is a broad and intensively developing field [3].
12.1.1 Structure of Monoclonal Antibodies (moAbs) A physiological IgG antibody consists of four polypeptide chains, two identical heavy chains and two identical light chains, which are covalently linked by interchain disulfide bonds. The chains have a variable (V) and a constant (C) region. The constant region of light chains consists of one C-domain, and the heavy chains have
A. von Stackelberg (*) Pädiatrische Onkologie/Hämatologie, Charité, OHC, Augustenburger Platz 13353, Berlin, Germany e-mail:
[email protected] V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3_12, © Springer Science+Business Media, LLC 2011
221
222
A. von Stackelberg CDRs Heavy chain
VH/L region
Light chain Disulfid bonds
CH1/L region Hinge region
Complement Interaction site
Fab domain
CH2 region CH3 region
Fc domain
Fig. 12.1 Structure of an immunoglobulin (Ig) G antibody. The light chain consists of a variable (VL) and a constant (CL) region. The heavy chain consists of a variable region (VH) and 3 constant regions (C1–3). The C2 region contains sites interacting with complement. The C2 and C3 domains of the two heavy-chain dimers form the Fc domain. The variable regions contain the antigen-specific complementarity-determining regions (CDR), called hypervariable regions. The V- and CL/C1 regions form the Fab domain, which is linked to the Fc domain via the flexible hinge region. Two pairs of a heavy and a light chin are linked together with disulphide bonds
two more C-domains, which together with two other C-domains form the dimeric heavy chain Fc domain. The four V-regions each together with one C-domain form the Fab-domain. The Fab and the Fc domains are linked by a flexible hinge region. The Fc domain interacts with Fc-receptor-positive immune effector cells inducing cellular cytotoxic activity against the target cell. Furthermore, it contains regions interacting with the complement system and initiating the complementdependent cytotoxicity (CDC). The V-region includes hypervariable regions (complementarity-determining regions – CDR), which ensure the specific recognition of the antigen (Fig. 12.1). In chimeric moAbs, the constant regions of a murine antibody are replaced with human C-regions. For humanization, additionally the V-region is replaced with human polypeptide chains, except the hypervariable region (Fig. 12.2). Fully human recombinant antibodies do not contain non-human components. The specific determination against human antigens is achieved via recombinant technology [4].
12.2 The Clinical Use of MoAbs MoAbs can be used for anti-tumour or anti-leukemic treatment as unconjugated compounds or as immunoconjugates. In both cases, the antibodies bind to the specific antigen, in case of acute lymphoblastic leukemia (ALL) to typical
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
a
CDRs
b
223
CFRs
VH/L region CH1/L region
Fab domain
CH2 region CH3 region
Fc domain
Fig. 12.2 (a) Chimaeric antibody: constant regions are human (dark grey), variable regions are murine (light grey). (b) Humanized antibody: constant regions and conserved framework regions (CFRs) of the variable region are human. Only the CDR, also called hypervariable regions, remain murine
precursor B-cell (PBC) or T-cell antigens. In unconjugated antibodies, the antibodymediated physiologic immune mechanisms are responsible for anti-leukemic activity. Immunoconjugates deliver toxic compounds directly to the antigenexpressing cell. Several toxins have been applied such as cytostatic drugs, bacterial toxins, and radionuclides [5]. For optimal activity of unconjugated antibodies, the persistence of the antigen–antibody complex at the cell surface and its exposure to the immune effector mechanisms is required. Activity of immunoconjugates is better if the complex is rapidly internalized leading to a maximal effect of the toxin directly inside of the cell. This has been shown by comparing the cytotoxic effect of anti-CD19 versus anti-CD22 immunoconjugates. The anti-CD22- immunoconjugate complex was internalized at a much higher degree and led to a higher cytotoxicity. Consequently, CD22 is a more suitable target for immunoconjugate approaches, whereas CD19 is a better target for unconjugated antibodies exploiting the physiologic mechanisms of the immune system [6, 7]. The selection of the appropriate target antigen depends furthermore on the rate and intensity of the antigen expression on the target cell surface and its expression on other non-tumour cells and tissues. Among antibody isotypes, IgG1 and IgG3 exhibit best activation of ADCC and CDC and are therefore preferred for the development of unconjugated cytotoxic moAbs. In contrast, IgG2 and IgG4 are suitable isotypes for blocking a binding site, for inducing internalization or for exhibiting agonistic activity [4].
224
A. von Stackelberg
12.2.1 Target Antigens on Lymphoblastic Leukemia Cells for MoAbs A variety of antigens are expressed on the surface and in the cytoplasm of lymphoblastic leukemia cells and add to the immunologic characterization of the leukemia [8]. Antigens suitable as targets for moAbs are listed and characterized in Table 12.1. Among the B-cell specific antigens, CD79a, CD19 and CD22 are highly expressed (>95%) in all B-cell precursor (BCP) acute lymphoblastic leukemias, whereas CD10 and CD21 are present in >95% of commonand pre-B, but not in pro-B leukemias and CD20 is present in only 10–35% of the BCP leukemias. Among T-lineage antigens, CD2, CD5 and CD7 are considered as pan-T-lineage antigens, which are expressed throughout maturation. CD3, which is part of the T-cell receptor (TCR) complex, is present at a cytoplasmic level during T-cell maturation and is expressed at the surface only in mature T-cells. In T-ALL, CD7 and cytoplasmic CD3 can be considered as reliably expressed antigens in all subtypes, whereas CD1, CD2, CD5, CD4, CD8 and CD25 are variably expressed among different subtypes and clones and, therefore, less suitable for systematic immunotherapy [9]. CD3, CD5, CD7 and CD25 are rapidly internalized upon binding to antibodies and, therefore, suitable targets of immunotoxins. CD4, CD8 and in part CD2 are mostly not internalized and thus suitable for unconjugated antibodies inducing ADCC and CDC [10]. CD52, CD45 and HLA-DR are highly expressed on nearly all ALL cells. They are also expressed on many other haematopoietic cells and thus not as selective for anti-leukemic treatment. Myeloid antigens are aberrantly expressed in some lymphoblastic leukemia clones. Therefore, CD33 targeted therapy has also been used in individual patients with refractory CD33-positive ALL (Table 12.1).
12.2.2 Mechanisms of Immune Effector Cells Immune effector cells expressing antibody Fcg-receptors and capable of inducing antibody-dependent anti-tumour effects include natural killer (NK) cells, monocytes/macrophages, neutrophils and dendritic cells (DCs). They accomplish this via an array of antibody-dependent effector functions, including direct killing of antibody-coated target cells via lysis, apoptosis and phagocytosis, or indirect effects such as cytokine/chemokine release and promotion of adaptive immune responses [11]. Whereas the Fcg-receptor subtypes FcgRIIIa and FcgRIIa mediate activating effects, the FcgRIIb expressed on neutrophil granulocytes and monocytes mediates inhibition of activity. A high ratio of FcgRIIa/b seems to be essential for optimal antibody activity [12, 13]. There is evidence from experimental models that cytokines also modulate the FcgR-mediated anti-tumour activity [14].
>95%
CD79a
T-lineage antigens CD2 –
>95%
CD22
75%a
–
–
–
11–37%
50
45
140
33–37
T cells, natural killer (NK) cells, and thymocytes
Immature and maturing B-cells (not on stem, pro-B and plasma cells) Immature and maturing B-cells (not on stem-, pro-B and plasma cells) All B-cell maturation steps (pro-B until plasma cell)
No data
Rapid
Rapid
Very slow
[9]
[115, 116]
[80, 81]
[62, 63]
(continued)
[140, 141] Co-stimulatory glycoprotein for lymphocyte activation. Interaction with antigen presenting or target cells via ligand CD58
BCR co-receptor; Interaction with CD21 after binding to antigen-activated complement C3d; enhances BCR mediated signals Transmembrane calcium channel; multimerization induced by BCR ligation Sialic-acid-binding transmembrane IG-like lectin (SIGLEC), BCR signalling inhibition CD79a and b form as transmembrane heterodimers with sIg the BCR complex. Enhances BCR signalling via ITAM
CD20
95
Intermediate
–
>95%
CD19
Zinc-dependent metalloprotease, degradation of small peptides
Rapid
[201]
Kidney, liver, intestines, placenta, choroid plexus, brain, gonads, adrenal cortex, leucocytes B-(precursor) cells, no stem-/plasma cells; dendritic-, follicular cells
20%
100
[53, 54]
B-lineage antigens CD10 95%
[42]
References
Table 12.1 Expression and characteristics of selected targets for monoclonal antibodies (moAbs) in paediatric ALL Expression in BCP lineage Expression in Molecular ALL T-lineage ALL weight (kDa) Physiological expression Internalization Biologic function Target Lineage independent antigens CD52 90% 80% 25 Lympho-, monocytes, dendritic None GPI anchored protein, unknown cells, male genital tract function HLA-DR >95% >95% All haematopoietic cells N.d. HLA-class II receptor
–
20%a
–
?
CD4
CD5
CD7
CD25
10–15%
75%
>95%
>90%a
55%a
Surface 55%;a cytoplasmic >95%a
67
55
67
55
Molecular weight (kDa)
Myeloid cells, monocytes, dendritic cells, downregulated in granulocytes (probably not on stem cells)
Activated T-lymphocytes, NK-cells, macrophages
T-cells and the B1a subset of B-cells Pan T-cell antigen; expression on myeloid subsets
Helper and regulatory T-cells, post thymic T-cell malignancies
All T-cells; cytoplasmic in expression prethymic T-cell malignancies cytoplasmatic expression
Physiological expression
Rapid
Rapid
Rapid
Rapid
Slow
Intermediate
Internalization
Sialic-acid-binding transmembrane immunoglobulin-like lectin (SIGLEC), inhibitory regulation of immune cells
CD3 polypeptide chains associated with a and b-subunits of TCR, intracellular regions represent signalling domain of the TCR complex Four immunoglobulin-like domains, a hydrophobic transmembrane domain, a long cytoplasmic tail. Co-receptor for the TCR complex Cysteine-rich scavenger receptor family glycoprotein (Leu-1) Transmembrane protein, member of the IgSF, with one N-terminal domain. Regulation of cytokine production a-subunit of the IL2 receptor (IL2Ra). Mediates IL2 immunomodulation in context with b and g subunits
Biologic function
[172, 173]
[140, 168]
[163]
[157]
[154]
[29]
References
BCP B-cell precursor; BCR B-cell receptor; GPI Glycosylphosphatidylinositol; IgSF immunoglobulin super family; IL2 interleukin 2; ITAM immunoreceptor tyrosine-based activation motif; kDa kilodalton; N.d. not done; ref. references; sIg surface immunoglobulin; TCR T-cell receptor a Stackelberg, ALL-REZ BFM study, unpublished data
Myeloid antigens CD33 20–33%
–
CD3
Table 12.1 (continued) Expression in BCP lineage Expression in ALL T-lineage ALL Target
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
227
12.2.2.1 NK-Cells NK cells are the best appreciated among anti-tumour effectors, as their activity is readily measured in commonly used ADCC assays. NK cells have the unique distinction that they typically only express the activating receptor FcgRIIIa and are not subject to regulation by the inhibitory receptor FcgRIIb. The major FcgRinduced effector activities of NK cells are cytolysis of target cells through lytic granule release or apoptosis via secretion of TNF family ligands (e.g. TNF, FasL) [15]. They are also potent producers of cytokines such as interferon-g. NK cells are regulated through a series of activating receptors, such as NKG2D, and inhibitory receptors of the killer Ig-like receptor (KIR) family. KIRs suppress killing when they interact with autologous (self) class I major histocompatibility complex (MHC) molecules on normal cells. Target cells lacking a matching MHC are, however, attacked by NK cells, via binding to activating NK receptors, establishing the so-called “missing-self” paradigm of NK cell function. Engagement of the activating receptor FcgRIIIa (CD16a) by antibody-coated (opsonized) target cells can partially override the KIR inhibitory signal, resulting in killing of even autologous cells. Interestingly, additional killing can be observed when anti-KIR or anti-MHC blocking antibodies are combined with anti-tumour antibodies [11].
12.2.2.2 Macrophages, Neutrophils and Dendritic Cells All cells of the myeloid lineage, including monocytes/macrophages, neutrophils and DCs, express FcgRIIa and at least one splice variant of the inhibitory receptor FcgRIIb. Monocytes/macrophages and DCs also express FcgRIIIa and FcgRI depending on their source and activation state [13, 16]. Neutrophils express FcgRIIIb rather than FcgRIIIa, and FcgRI when activated by G-CSF [17–19]. Macrophages and neutrophils are classic phagocytes, and can phagocyte opsonized target cells through engagement of FcgRs. They can also induce apoptosis of target cells through the release of reactive nitrogen and oxygen intermediates, or lyse them through the release of cytolytic granules. Furthermore, macrophages and DCs act as professional antigen-presenting cells (APCs). FcgR-mediated phagocytosis, besides leading to the destruction of a target cell, can facilitate a potentially more robust anti-tumour effect known as cross-priming, in which these cells process and present tumour-derived antigens on their surface class I MHC, thus acquiring the ability to activate T cells. Cross-priming can activate cytotoxic T lymphocytes (CTLs) that recognize MHC–tumour antigen complexes, ultimately leading to attack on the tumour cells. This effect, catalysed by an anti-tumour antibody, may theoretically lead to long-lasting adaptive anti-tumour immunity and long-term remission. Indeed, such effects are sometimes taken into account to explain the long-term responses observed in lymphoma patients after therapy with the anti-CD20 antibody rituximab [20, 21].
228
A. von Stackelberg
VH anti CD19
VL anti CD19
VH anti CD3
VL anti CD3
Fig. 12.3 Bispecific T-cell engaging antibody MT103 (Blinatumomab): the compound consists of a single chain antibody of 2 Fv domains with variable light (VL) and heavy chain (VH) regions specific for CD19 and CD3 which are linked through a Glycine–Serine linker
12.2.2.3 Cytotoxic T-Cells Cytotoxic T-cells are important effector cells of the immune system for tumour control. They can recognize foreign antigens presented by MHC-class I molecules and induce lysis of a suspicious target cell by producing cell membrane-damaging perforin and by expression of pro-apoptotic Fas-ligand and TNF-a. However, they do not contribute to the antibody-mediated immune responses, since they have no Fc-receptor. T-cell engaging bispecific antibodies directed against a T-cell specific antigen (i.e. CD3) and a target specific antigen (i.e. CD19) link T-cells to target cells and thus can induce T-cell-mediated cell death and possibly a T-cell-mediated tumour immunity (Fig. 12.3).
12.2.3 Activity of Unconjugated Antibodies An antibody itself usually is not responsible for killing target cells, but instead marks the cells for other components or effector cells of the body’s immune system to attack, or it can initiate signalling mechanisms in the targeted cell that leads to
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
229
the cell’s self-destruction. The former two attack mechanisms are referred to as antibody-mediated CDC, antibody-dependent cellular cytotoxicity (ADCC) and antibody-dependent cell-mediated phagocytosis (ADCP). ADCC involves the recognition of the antibody by immune cells that engage the antibody-marked cells and either through their direct action or through the recruitment of other cell types leads to death of the tagged cells. The efficiency of ADCC and ADCP depends on the immune status of the recipient and the availability of macrophages and NK-cells. CDC is a process where a cascade of different complement proteins become activated, usually when several IgGs are in close proximity to each other, either with direct cytotoxic effect carried out by the membrane attack complex (MAC) or with indirect cytotoxic effect by attracting other immune effector cells to this location. There have been successful efforts to optimize monoclonal antibody– complement interactions and complement activity to enhance the anti-tumour potency of the monoclonal antibody [22, 23]. However, the exact role of CDC in anti-tumour activity of moAbs remains to be clarified. Fc-domain engineering is a fascinating and rapidly growing field with the aim to optimize the affinity of the antibody to Fc-g-receptors of effector cells of the immune system and thus to improve anti-tumour efficacy of unconjugated moAbs [11, 24]. A better affinity can be achieved selectively for the NK-cell specific Fc-g-IIIa-receptor (CD16), leading to a directly enhanced NK-cell-mediated cytotoxicity or for the macrophage specific Fc-g-IIa (CD32) receptor, thus enhancing the activation of APCs and possibly the adaptive T-cell immunity [25, 26]. For example, a CD20 moAb with enhanced affinity to CD16 induced ADCC at lower concentrations and a greater ADCC and NK-cell activation at antibody saturation levels than the unmodified rituximab antibody [27].
12.2.4 Immunoconjugates, Radioimmunotherapy MoAbs give the opportunity to deliver toxic compounds as immunoconjugates directly to cells bearing the specific target. Clinical and preclinical investigations have delineated several variables that appear to be important determinants of the efficacy of immunoconjugate therapy, including antibody specificity, antigen density, tumour size, presence of shed antigen and the rate of antibody internalization and metabolism by target cells [28, 29]. As suitable toxins, truncated bacterial toxins such as Pseudomonas or Diphtheria toxins have been used [5]. The toxic antibiotic compound calicheamicin was found to be more active in lymphoblastic leukemia cells when compared to myeloid leukemia or normal lymphocytes [30]. In AML cells, the expression of P-glycoprotein (PGP) correlates with resistance to calicheamicin, an effect that can be blocked with the PGP antagonist cyclosporine [31]. An alternative approach of immunoconjugates is linking of enzymes to a monoclonal antibody, which activates a prodrug after binding to the target antigen to generate a cytotoxic compound that acts locally by killing the target cells without harming cells not expressing the target antigen [32]. Radioimmunotherapy has been developed to deliver radiation therapy to a target by linking the radionuclide to a
230
A. von Stackelberg
specific antibody. Different radionuclides have been applied. Whereas the b emitter 131Iodine has been most frequently used due to its comparably easy availability, it entails the disadvantages of rather low-energy b emission, additional g emission and comparably long half-life. 90Yttrium lacks g emission and has higherenergy b emission with a penetration distance of up to 11 mm, which is advantageous for the treatment of solid tumours. Radionuclides with shorter distance such as a emitters (213Bismuth, 211Astatine) or Auger electron emitters with extremely short distance (<100 nm, 111Indium, 125Iodine) are ideal for very small size tumours or leukemias. The latter ones are ideally transferred into the target cell via endocytosis, causing more radiation when linked with internalizing moAbs [33, 34]. Pretargeted radioimmunotherapy was developed to prevent the long exposure to radiolabelled antibodies of conventional radioimmunotherapy until target sites are saturated. In a two-step approach, a bifunctional or streptavidin-linked target-specific antibody is applied first as a preparative agent until target saturation. After a washout step of the unbound agent, the administration of a radiolabelled anti-antibody or biotin as second agent leads to rapid and specific binding to the preparative agent [35].
12.2.5 Bispecific Antibodies Bispecific antibodies are artificially engineered compounds with variable regions against two different specific antigens. The construction of such antibodies is based on the elements of moAbs including variable and constant parts of light and heavy chains, linked together in different manner. Particularly attractive are so-called bispecific T-cell-engaging antibodies (BiTEs®). They link cytotoxic T-cells via an anti-CD3 part with a tumour-typical antigen and induce cytotoxic activity against the target cell [36]. This effect can be demonstrated in MHC-class I-positive and negative target cells indicating a mechanism circumventing escape mechanisms of tumours cells from immune control. Immunological synapses can be demonstrated to be present during the cytolytic process, suggesting a physiologic immune reaction mediated by the antibodies [37]. A bispecific antibody against CD19 and CD3 was able to link CD3-positive T-cells to CD19-positive non-Hodgkin lymphoma (NHL) or leukemia cells and induce substantial cytotoxicity and T-cell mediated antitumour immunity (Fig. 12.3) [38, 39]. Several other approaches with bispecific antibodies have been followed as well. A bispecific antibody against two B-cell NHL/leukemia antigens (CD19/CD22) had a higher reactivity against the B-cell neoplasm than the combination of monospecific moAbs alone [40].
12.2.6 Nomenclature of MoAbs The nomenclature of monoclonal antibodies has been determined by the United States Adopted Name (USAN) Council (http://www.ama-assn.org). All names of
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
231
moAbs start with an variable prefix determining the individual drug such as ri- for rituximab or epra- for epratuzumab, followed by an affix determining the clinical target of the drug as -tu(m)- for tumour, -li(m)- for immune system, or -ci(r)- for cardiovascular, followed by an affix determining the antibody origin as -o- for mouse, -u- for human, -xi- for chimeric or -zu- for humanized, followed by the suffix -mab for moAb. A conjugate such as a toxin or a radionuclide is added as second word such as ozogamicin for the toxin calicheamicin (gemtuzumab ozogamicin – Mylotarg®) or tiuxetan as chelator for yttrium-90 (90Y-Ibritumomab tiuxetan – Zevalin®) [41].
12.3 Lineage-Independent Antigens 12.3.1 CD52 12.3.1.1 Biologic Characteristics of CD52 Antigen CD52 is a glycosylphosphatidylinositol (GPI) anchored protein of unknown function. It is expressed on mature lymphocytes, monocytes, DCs and the male genital tract, but not on stem cells. Among malignant cells, CD52 is highly expressed in low-grade B cell lymphoproliferative disorders and most BCP and T-lymphoblastic lymphomas and leukemias. In contrast, all pro-B or t(4;11)-positive ALL samples, most of the chronic T-lineage lymphomas, and about half of the pre-T-ALL leukemias are negative for CD52 [42, 43]. CD52 is expressed on the surface of positive cells at a high density compared to other antigens. Although CD52 is well expressed on both B- and T-lineage ALL, it has not been systematically used for treatment of patients with ALL. One reason is certainly the depletion of the whole lymphocytic compartment, leading to prolonged immunodeficiency with susceptibility to viral infections and immunoglobulin deficiency.
12.3.1.2 Preclinical and Clinical Use of Anti-CD52 MoAbs Alemtuzumab (Campath®) is a humanized monoclonal antibody against CD52, which has been approved by the FDA for treatment of refractory CLL (Chronic Lymphocytic Leukemia). The cytotoxic mechanisms of alemtuzumab are mediated by CDC and ADCC [44, 45]. Furthermore, the induction of caspase-independent apoptosis has been demonstrated [46]. Apart from infusion-related toxicities mostly of grade I-II, which could be largely avoided by subcutaneous application of the drug, neutropenia, thrombocytopenia and opportunistic infections, in particular of viral origin due to the depletion of lymphocytes, have been observed [47]. Alemtuzumab has been shown to be effective in relapsed/refractory CLL, but in particular in untreated CLL and in combination with chemotherapy. Furthermore, alemtuzumab demonstrated activity in patients with diverse T-cell NHL (T-cell pro-lymphocytic leukemia, peripheral T-cell lymphoma or cutaneous T-cell
232
A. von Stackelberg
lymphoma). Alemtuzumab has been used as an element of conditioning regimen prior to allogeneic and autologous stem-cell transplantations for prevention of graft-versus-host disease and graft rejection. In particular, in the context of total body irradiation, it was associated with high infectious, mostly viral, toxicity. A better tolerability was observed in the context of non-myeloablative conditioning [46]. In several groups, alemtuzumab has been applied as part of the conditioning regimen for allogeneic stem-cell transplantation (SCT) of children with ALL (Table 12.2) [48]. In comparison with ATG, alemtuzumab revealed higher rates of adenovirus infections [49]. Another trial showed a longer delay of immune reconstitution after alemtuzumab compared to ATG [50]. The rate of post-transplant lymphoproliferative disease (LPD) was lower after alemtuzumab compared to other T-cell depleting regimens due to its additional cytotoxic activity against B-lineage lymphocytes [51]. 12.3.1.3 Immunoconjugates with CD52 Alemtuzumab has been linked with 188Rhenium. So far, a favourable in vitro behaviour and biodistribution could be demonstrated as precondition for further preclinical development (Table 12.2) [52].
12.3.2 HLA-DR HLA-DR is an antigen representing the monomeric domain of an HLA-class II antigen. MHC class II molecules are non-covalently associated heterodimers of two transmembrane glycoproteins. HLA-DR, one of three MHC II molecules, is expressed on antigen-presenting cells, including B lymphocytes, monocytes and DCs, and plays a pivotal role in antigen presentation and the induction of antigenspecific immune responses. HLA-DR is also expressed on a wide variety of B-lineage lymphoma and leukemia cells, which suggests that it may be a good target for antibody-based immunotherapy (Table 12.1). The HLA-DR molecule is extremely polymorphic due to its highly variable b-chain with more than 500 different HLA-DRB alleles identified so far [53]. Thus, a variety of moAbs can recognize only a selected part of HLA-DR antigens and, therefore, fail to react against a defined proportion of HLA-DR-positive malignancies. The humanized 1D10 (Apolizumab, Remitogen®, Protein Design Labs, Inc.) antibody is directed against a polymorphic epitope of the b-chain HLA-class II antigens. It has been shown to be reactive against about half of human HLA-DR-positive malignancies [54, 55]. Since the most prominent cytotoxic effect of apolizumab was ADCC induced by granulocytes, a combination therapy with G-CSF has been evaluated in clinical trials, showing some efficacy in adult B-cell lymphoma [18]. Recently, a fully human HLA-DR antibody HD8 has been developed, showing substantial potency not only for activation of ADCC and CDC but also for induction of
CD52 HLA-DR
Rat, humanized Murine
Linked to b-emitter 188rhenium Anti-HLA-DR–streptavidin conjugate as pretarget, linked via biotin with radionuclide
HLA-class II b-chain Murine, humanized HLA-class II b-chain moAb, recognizes 50% of HLA-DRpositive malignancies. ADCC via granulocytes HLA-DR Human IgG1-type Anti-HLA-DR, binds as well HLA-DQ and DP, ADCC/P, CDC
References
Preclinical investigations Preclinical
Preclinical
[52] [57]
[56]
Phase I-III studies in adult [46–48] CLL, clinical experience in adult T-NHL, part of allo-SCT conditioning regimen Phase I/II in adult B-cell [18, 55, 56] lymphoma
Studies
Characteristics and stage of development ADCC antibody-dependent cellular cytotoxicity; CDC complement-dependent cytotoxicity; CLL chronic lymphatic leukemia; HLA human leukocyte antigen; T-NHL T-cell non-Hodgkin lymphoma
Immunoconjugates 188 Re-alemtuzumab Lym1-SA 90Y-DOTA-biotin
HD8
Apolizumab (HU1D10, Remitogen®, Protein Design Labs, Inc.)
Table 12.2 MoAbs directed against lineage-independent antigens CD52 and HLA-DR Compounds Targets Species Characteristics Unconjugated CD52 Rat, humanized ADCC, CDC, apoptosis Alemtuzumab (Campath®, Genzyme, Inc. [US]; Bayer Schering, Inc.[EU])
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia 233
234
A. von Stackelberg
apoptosis via HLA-R signalling in vitro and in mouse models [56]. HD8 showed reactivity against nearly all HLA-DR-positive cell lines and human malignant cells. Furthermore, it recognizes HLA-DPB and DQB alleles (Table 12.2). In a preclinical setting, the activity of pretargeted radioimmunotherapy targeting HLA-DR has been evaluated. The antibody–streptavidin conjugate was first infused until saturation of the antigen sites, followed by a clearing agent, and finally by radiolabelled biotin (90Y-DOTA-biotin). This approach was found to be more effective against lymphoma cell lines and less toxic in mice than the application of directly labelled antibody [57].
12.4 B-Lineage-Specific Antigens 12.4.1 CD10 CD10, common acute lymphoblastic leukemia antigen (CALLA), in context with other tissues referred to as neprilysin, is a zinc-dependent metalloprotease enzyme that degrades a number of small secreted peptides. One of these is the amyloid beta peptide, whose abnormal misfolding and aggregation in neural tissue has been implicated as a cause of Alzheimer’s disease. Synthesized as a membrane-bound protein, the neprilysin ectodomain is released extracellularly after it has been transported from the Golgi apparatus to the cell surface. Neprilysin is expressed in a variety of organs and tissues including kidney, lung, neuronal and stromal cells. Therefore, it is not suitable as target for specific anti-leukemic therapy. Some investigators have used anti-CD10 antibodies for ex vivo purging prior to autologous SCT [58].
12.4.2 CD19 12.4.2.1 Biologic Characteristics of the CD19 Antigen CD19 is a 95-kDa transmembrane glycoprotein of the immunoglobulin superfamily, containing two extracellular immunoglobulin-like domains and an extensive cytoplasmic tail. CD19 is expressed in B-cells, dendritic and follicular cells. In B-cells, CD19 is expressed by early pre-B cells from the time of heavy chain gene rearrangement until plasma cell differentiation, being lost in mature plasma cells. CD19 interacts with CD77 playing a role in germinal centre formation, B-cell homing and apoptosis. It is B-lineage-specific and functions as a positive regulator of B-cell receptor signalling in conjunction with CD21 and CD81. CD19 is a critical signal transduction molecule that regulates B lymphocyte development, activation and differentiation. The CD19/CD21/CD81/CD225 (Leu-13) signalling complex modulates the threshold for the B-cell antigen receptor (BCR). CD21 enables CD19
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
235
to be cross-linked to the BCR via complement activation after pre-immune recognition of an immunogen by the complement system, thus reducing the number of B-cell receptor molecules, which must be ligated to enable B-cell activation. Consequently, CD19 acts as a BCR co-receptor. The internalization of CD19 upon antibody binding seems to be only moderate and does not substantially impair tumour inhibition via effector cells (Table 12.1) [59–62]. 12.4.2.2 Preclinical and Clinical Use Anti-CD19 MoAbs CD19 is an attractive target for cancers of lymphoid origin due to its high expression in most (>90%) PBC acute lymphoblastic leukemias (ALL) and non-Hodgkin lymphomas (NHL). CD19 cell surface expression is lower relative to CD20, but it begins earlier and persists longer throughout B-cell maturation [63]. CD19 has been a focus of immunotherapy development for over 20 years, and several CD19specific antibodies have been evaluated for the treatment of B-lineage malignancies in vitro, in mouse models and in clinical trials (Table 12.3). 12.4.2.3 Unconjugated Anti-CD19 MoAbs In a transgenic mouse model, CD19-positive B-cells and malignant CD19-positive B-cell lymphomas/leukemias were efficiently depleted by unmodified anti-CD19 antibodies mainly via Fc-receptor-g (FcR g)-mediated macrophage activity (ADCP). In this model, the B-cell depletion was twice as durable as the one after anti-CD20 moAb, suggesting also the potentially more severe side effect of immunoglobulin deficiency after effective anti-CD19 therapy [64, 65]. Some responses have been observed after treating patients with refractory B-cell lymphoma with a murine anti-CD19 moAb, but the response is of limited duration [66]. Several attempts have been undertaken to improve ADCC and ADCP by engineering the Fc domain of CD19 targeting moAbs. By this manipulation, the cell-mediated cytotoxicity could be enhanced up to 100–1,000 fold compared to the conventional antibody [60]. The humanized and Fc-engineered antibody XmAb®5574 has been developed by the company Xencor and is currently being investigated in early clinical trials. 12.4.2.4 CD19 Antibody–Drug Conjugates CD19 has been used as target for antibody–drug conjugates. One attempt using a murine anti-CD19 moAb with the protein synthesis inhibitor ricin in patients with CD19-positive relapsed/refractory NHL, CLL or ALL showed feasibility and some clinical response in phase I studies with bolus or continuous infusion. However, in phase II, no relevant responses have been observed. The efficacy was certainly impaired by the use of the pure murine antibody, leading to the development of human anti-mouse antibody (HAMA) in a substantial proportion of the patients
CD19, CD3
Blinatumomab (Micromed, Inc.)
Murine Murine Murine Murine
CD19 CD19 CD19
CD19
Murine
Murine
Anti-CD19 + Idarubicin Anti-CD19 + liposomal daunorubicin Anti-CD19 single chain Fv + pseudomonas Exotoxin A Anti-CD19 single Fv chain + TRAIL. Induction of apoptosis
Anti-CD19 + ricin
[69] [7, 70] [71] [74]
Xenograft model
[67, 68]
[38]
[79]
[40]
[60]
References
Clinical experience in phase I/II Xenograft model Xenograft model Xenograft model
Clinical trials phase I/II
Xenograft model
Xenograft model
Xenograft model, adult phase I ongoing
Studies
Characteristics and stage of development ADCC antibody-dependent cellular cytotoxicity; CDC complement-dependent cytotoxicity; Fv variable region; TRAIL tumour necrosis factor-related apoptosisinducing ligand
Ida-anti-CD19 Anti-CD19 liposomal DNR Anti-CD19 Pseudomonas toxin A scFvCD19:sTRAIL
CD19
Cd19, CD16
Sctb ds [19 × 16 × 19]
Immunoconjugates Anti-B4-blocked-ricin
Murine
CD19, CD22
Bispecific DT2219 Murine
Fc-engineered, humanized anti-CD19, induces ADCC, CDC
Murine humanized
Single chain, bispecific anti-CD19/CD22, linked to diphtheria toxin Single chain with 3 Fv determinants, 2 distal Fv anti-CD19, 1 central anti-CD16 Single chain, bispecific anti-CD19/CD3
Characteristics
Species
Table 12.3 MoAbs directed against CD19 Compounds Targets Unconjugated XmAb5574 (Xencor Inc.) CD19
236 A. von Stackelberg
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
237
[67, 68]. In a mouse model, an anti-CD19-Idarubicin conjugate showed anti-leukemic efficacy, which was less toxic and more effective than the same dose of unconjugated idarubicin [69]. Furthermore, it could be demonstrated that liposomal daunorubicin (DNR) displayed a higher anti-leukemic activity in mice being conjugated to antiCD19 compared to anti-CD20. This was attributed to the faster internalization of CD19 and a greater intracellular release of the cytotoxic drug [7]. This effect could not be shown for liposomal vincristine, but a combination of anti-CD19 conjugated liposomal DNR with anti-CD20 conjugated liposomal VCR revealed the best anti-leukemic effect [70]. Another preclinical approach used a single-chain anti-CD19 Fv fragment fused to a derivate of pseudomonas toxin A, achieving substantial anti-B-cell neoplasm activity in vitro and in mouse models [71]. However, the internalization of a CD19-antibody complex is still low compared to CD22 and is inversely correlated to the co-expression of CD21 [72]. A construct that conjugates microtubule-destabilizing agent monomethyl auristatin E (MMAE) to the humanized anti-CD19 antibody hBU12 via a protease-sensitive valine–citrulline (vc) dipeptide linker was able to overcome the inhibitory effect of CD21 on antiCD19 activity and exhibited good anti-tumour effect in rituximab-refractory lymphoma cells [73]. With the aim to selectively induce apoptosis in CD19-positive target cells, the tumour necrosis factor-related apoptosis-inducing ligand TRAIL was fused to a CD19-specific single-chain Fv antibody fragment (scFv), resulting in the compound scFvCD19:sTRAIL. scFvCD19:sTRAIL was able to induce apoptosis in several CD19-positive tumour cell lines, whereas normal blood cells remained unaffected. The effect could be enhanced by simultaneous treatment with valproic acid (VPA) or ciclosporin. Treatment of patient-derived acute B-lymphoblastic leukemia (B-ALL) and chronic B-lymphocytic leukemia (BCLL) cells resulted in strong tumouricidal activity [74]. However, it is has been shown that the intracellular delivery of toxic compounds via CD22 is much more effective than via CD19. Thus, the development of unconjugated anti-CD19 has been prioritized, whereas CD22 has been selected as more suitable target for immunotoxins [6]. 12.4.2.5 Bispecific Antibodies Involving Anti-CD19 A bispecific anti-CD19 anti-CD22 antibody linked to diphtheria toxin (DT2219) was more effective in a mouse B-cell leukemia/lymphoma xenograft model than the single antibodies linked to the toxin [40]. A bispecific antibody MT103 (blinatumomab, Micromet, Inc.) targeting CD19 and CD3 and thus engaging T-cells showed a high and striking anti B-cell lymphoma/leukemia activity at low doses in vitro and in vivo [75]. The compound consists of a single chain antibody of 2 Fv domains linked through a Gly-Ser linker (Fig. 12.3). The application in clinical trials was a continuous infusion over several weeks, thus providing a continuous level of the active compound [38, 39]. The anti-tumour effect was better than with rituximab and was not impaired by the addition of dexamethasone [76, 77]. Furthermore, blinatumomab induced a reduction of leukemia burden below the detection limit in
238
A. von Stackelberg
adult patients with ALL and persistence of minimal residual disease (MRD) after chemotherapy [78]. By analogy, a NK-cell-engaging antibody was developed on the basis of a single-chain construct with two anti-CD19 domains and one central anti-CD16 domain. The avidity to CD19 was threefold greater, and a comparable ADCC could be achieved at 10–40 fold lower concentrations than with a bispecific anti-CD19/CD16 antibody with only one anti-CD19 domain [79].
12.4.3 CD20 12.4.3.1 Biologic Characteristics of the CD20 Antigen CD20 is a 33–37 kDa, non-glycosylated B-lymphocyte-specific integral membrane phosphoprotein and belongs to the membrane-spanning 4A gene family (MS4A). A 44 amino-acid extracellular loop contains the epitopes for antibody binding. CD20 is highly expressed in mature normal and malignant B-cells and variably expressed in normal and malignant PBCs. It is not expressed in very early (pro) BCPs and stem cells, neither in plasma cells. Aberrant CD20 expression is occasionally observed on malignant T cells. CD20 has no known natural ligand, and its function has not been clarified in detail. However, it seems to be implicated in the regulation of transmembrane calcium conductance. After B-cell receptor ligation, intracellularly stored calcium is released, CD20 forms oligomers within lipid rafts and functions as a calcium channel allowing calcium influx and refilling of the stores, leading to cellcycle progression and B-lymphocyte proliferation (Table 12.1) [80, 81]. 12.4.3.2 Preclinical and Clinical Use of Anti-CD20 MoAbs In the early 1980s, moAbs against CD20 (B1) were developed and investigated [82]. Antibody binding to CD20 induces the formation of plasma membrane rafts, leading to concentration of src kinases and apoptotic signalling molecules [83]. Anti-CD20 moAbs have been found to show two different types of activity classified as type I and II. Type I induces ADCC and potent CDC, whereas Type II antibodies lack CDC activity. Nevertheless, the type II moAb tositumomab was shown to be more effective in depleting B-cells than the type I moAb Rituximab [84]. In 1997, rituximab, a chimeric type I anti-CD20 antibody was licensed by the Food and Drug Administration as the first anti-cancer monoclonal antibody. Subsequently, the remarkable anti-tumour activity of rituximab with chemotherapy was shown in B-cell lymphoma [85]. In PBC malignancies, CD20 is expressed in less than 50% of the clones. However, it has been shown that the gene and the antigen expression of CD20 in PBC ALL is upregulated during induction therapy, most likely due to the glucocorticoid effect [86, 87]. Although the antigen is a less suitable target for a broader use in ALL and has been used as therapeutic target only anecdotally in patients not responding to conventional therapy, the broad development of
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
239
different CD20-directed compounds demonstrates exemplarily what is possible in contemporary immunotherapy [88–90]. 12.4.3.3 Unconjugated Anti-CD20 MoAbs The human/mouse chimeric CD20 moAb rituximab (Rituxan®, Biogen Idec/ Genentech, Inc./MabThera®, Roche, Inc.) is rapidly becoming a standard element of the treatment of mature B-cell malignancies. It has been developed as single agent but is particularly effective when combined with conventional chemotherapies such as cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP) [85, 91]. In paediatric patients, only anecdotal data on rituximab plus chemotherapy in mature B-cell lymphoma have been reported, but larger prospective studies are ongoing [92]. The development of rituximab in ALL is less advanced due to the limited expression of the antigen on BCP clones. Several case reports and small patient series confirmed a promising/encouraging effect of the drug in adult ALL as a single agent and in combination with chemotherapy. However, prospective trials are ongoing [93]. In paediatric patients with ALL, the few case reports that have been published using Rituximab as single agent [90] or in combination with chemotherapy in refractory or multiple-relapse patients demonstrate its anti-leukemic activity [89, 94, 95]. Although unconjugated moAbs show a very favourable toxicity profile and are, therefore, ideal for combination with chemotherapy, some patients experienced severe and even fatal cytokine release syndrome [96]. A variety of alternative unconjugated moAbs against CD20 have been developed with the aim to improve anti-tumour activity compared to the gold standard, rituximab. Among preclinically developed humanized antibodies, veltuzumab (Immunomedics, Inc.) has reached clinical development. The drug has antigen specificity nearly identical to rituximab with conserved framework regions (CFRs) derived from the humanized antibody epratuzumab (anti-CD22). Preclinical data showed significantly better anti-tumour activity compared to rituximab, most probable due to one amino-acid change in the variable heavychain region [97]. In phase I/II studies in adult patients with refractory B-cell lymphoma, who were heavily pretreated with chemotherapy and rituximab, objective responses in 44% and complete remissions in 27% could be achieved at comparably low doses [98]. Among several murine antibodies, 1K1791 best inhibited cell proliferation and induced caspase-independent apoptosis [99]. Humanized and human forms have been developed that exhibit superior CDC and ADCC as compared to rituximab in vitro [100]. The fully human anti-CD20 antibody ofatumumab has been shown to exhibit a better CDC compared to rituximab [101]. Ofatumumab showed promising response in CLL and follicular lymphoma in phase I/II trials and is further investigated in phase II trials in this indication [102]. In 2009, ofatumumab (Genmab, Inc.) has been approved for refractory CLL by the FDA and the EMEA. These novel antibodies may be interesting compounds for future clinical development promising better anti-tumour activity than the established standard rituximab (Table 12.4).
CD20 CD20
Immunoconjugates 20-2b Rituximab/alliinase
Linked to tetrameric IFNa Local enzymatic activation of prodrug alliin to alicin Linked with 131iodine
Bivalent: better ADCC but no CDC compared to par-ental antibodies; Hexavalent: specific action against malignant B-cells Bispecific for CD20 and CD3 (T-cell engaging) and trifunctional with additional Fc-domain
[97, 98]
[85, 91–95]
References
[109] [32]
[107, 108]
In vitro, xenograft model, pilot phase I post SCT
In vitro, xenograft model In vitro, xenograft model
[103, 104]
In vitro, xenograft model
Phase I/II in adult patients with CLL [102] and FL, approved by FDA/EMEA for refractory CLL in 2009
Phase I/II/III in adults single agent and combination therapy, FDA approval since 1997 Phase I/II in adult patients
Studies
131
I-tositumomab (Bexxar®, CD20 Clinical phase I-III trials, FDA [110, 112] GlaxoSmithKline, Inc.) approval 2003 for NHL 90 [112, 114] Clinical phase I-III trials, FDA CD20 Mouse Y-Ibritumomab-Tiuxetan Linked with b-radiator 90Yttrium via approval 2002/EU 2004; phase (Zevalin®, Spectrum chelator tiuxetan I in children with relapsed/ Pharmaceuticals, Inc.[US]; refractory NHL Bayer Schering, Inc. [EU]) Characteristics and stage of development ADCC antibody-dependent cellular cytotoxicity; CDC complement-dependent cytotoxicity; C-domain constant domain; CLL chronic lymphatic leukemia; EMEA European medicines agency; EU European union; Fc constant region CH2/3; FDA Food and Drug Administration; HLA human leukocyte antigen; NHL non-Hodgkin lymphoma; SCT stem-cell transplantation; TNF tumour necrosis factor; US United States
Humanized Chimeric mouse/human Mouse
Rat/mouse
CD20, CD3
Bi20 (fBTA05)
Human
CD20, CD22 Humanized
CD20
Ofatumumab (HuMax-CD20, Genmab Inc.)
Hypervariable regions similar to rituximab, conserved framework regions derived from epratuzumab Directed against smaller extracellular loop of CD20, higher binding activity compared to rituximab
Unconjugated IgG1 antibody with human C-domains and murine Fv domains
Chimeric mouse/human Humanized
Characteristics
Species
Bispecific Anti-CD20/22 bi- or hexavalent
CD20
Veltuzumab (Immunomedics, Inc.)
Table 12.4 MoAbs directed against CD20 Compounds Targets Unconjugated Rituximab (Rituxan®/ CD20 MabThera®, Roche, Inc.)
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
241
12.4.3.4 Bispecific Antibodies Recently, bispecific humanized antibodies constructed from veltuzumab and epratuzumab have been tested in preclinical setting. The anti-CD20/22 construct showed higher ADCC but not CDC as compared with the parental compounds and did not exhibit substantially higher anti-tumour activity in vitro [103]. A hexavalent antibody construct linking four FAB fragments of epratuzumab (anti-CD22) with the anti-CD20 antibody veltuzumab was shown to act more specifically against malignant B-cells compared to the parental antibodies [104]. BiTEs directed against CD3 and CD20 have been developed and investigated in vitro [105, 106]. Recently, the CD20/CD3 bispecific antibody with a Fc-domain and, therefore, trifunctional capacity, Bi20 (fBTA05), has been developed and shows remarkable activity against CD20+ lymphoma cells in preclinical studies by engaging and activating T-cells and inducing Fc-mediated cellular response [107]. In a pilot phase I study, the compound induced prompt but transient responses in patients with recurrent/refractory CLL or high-grade lymphoma post allogeneic SCT in concert with either donor lymphocyte infusion or retransplantation, by inducing graft versus leukemia activity without substantial graft-versus-host disease (Table 12.4) [108]. 12.4.3.5 Immunoconjugates with CD20 Generally, CD 20 has been shown to be a less attractive candidate for immunotoxin therapy due to lack of internalization of the antigen–antibody complex after binding [7]. In contrast, a stable expression on the cell surface is compatible with effective radioimmunotherapy. Using the modular Dock-and-Lock method, interferon alpha (IFNa) has been linked to the humanized anti-C20 antibody veltuzumab. The resulting immunoconjugate 20-2b exhibited superior ADCC against lymphoma cells in vitro and in xenografts compared to veltuzumab and IFNa but lacks CDC [109]. Arditti et al. report on an immunoconjugate of rituximab linked to the enzyme alliinase, which activates the garlic-derived prodrug alliin to the cytotoxic compound allicin, which exhibited interesting anti-lymphoma activity in vitro and in mouse models [32]. CD20 has extensively been used as target for radioimmunotherapy in B-NHL. One clinically developed compound is 131I-tositumomab (Bexxar®, GlaxoSmithKline, Inc.) with the murine monoclonal antibody directed against CD20 linked with 131Iodine. The drug has been extensively used and has proven to be effective in patients with relapsed or refractory B-cell NHL [110]. A second well-established radiolabelled anti-C20 compound is 90Y-Ibritumomab-Tiuxetan (Zevalin®, Spectrum Pharma ceuticals, Inc.[US]; Bayer Schering, Inc.[EU]). In a randomized trial in patients with refractory NHL, it was more effective than rituximab alone [111]. Furthermore, radiolabelled CD20 has been added to conditioning regimens prior to autologous or allogeneic SCT as a more targeted alternative to total body irradiation [112]. A single centre observation indicates an equal efficacy but lower toxicity of 131 I-tositumomab compared to 90Y-Ibritumomab-Tiuxetan [113]. A COG phase I
242
A. von Stackelberg
study gave first impressions on safety and tolerability of 90Y-Ibritumomab-Tiuxetan in children with relapsed/refractory NHL, which is being confirmed in a phase II study currently [114]. No other data on radioimmunotherapy in children are published, in particular not in children with ALL. However, in patients with sufficient CD20 expression, CD20 directed radioimmunotherapy is an attractive approach that should be prospectively evaluated in future trials (Table 12.4).
12.4.4 CD 22 12.4.4.1 Biologic Characteristics of the CD22 Antigen CD22 is a 140-kDa transmembrane immunoglobulin-like lectin, which specifically binds sialic acid at its N-terminus. The presence of immunoglobulin domains makes CD22 a member of the immunoglobulin superfamily. CD22 acts as an accessory co-receptor that modulates BCR signalling upon ligation. Independently from ligation, the BCR-associated kinase Lyn phosphorylates tyrosines of the cytoplasmatic CD22 tail immunoreceptor tyrosine-based inhibitory motifs (ITIMs). These inhibit BCR signalling, enhance calcium efflux and lead to endocytosis [115]. Furthermore, it is involved in the CD19/21 and CD40 signalling regulation, peripheral B-cell homeostasis and survival and the promotion of BCR-induced cell cycle progression [116]. CD22 is expressed on immature and maturing B-cells, however, not on stem-, pro-B cells and plasma cells (Table 12.1). 12.4.4.2 Preclinical and Clinical Use of Anti-CD22 MoAbs As antigen restricted to B cells and being expressed on the majority of PBC lymphoblastic leukemias, CD22 is a suitable antigen for immunotherapy. Its rapid endocytosis upon ligand binding qualifies it as ideal target for immunoconjugated toxins that can exhibit their cytotoxicity against the target cell intracellularly [117]. 12.4.4.3 Unconjugated Anti-CD22 MoAbs A variety of anti-CD22 antibodies against different epitopes have been developed and mostly used for functional analyses [118]. Epratuzumab is a humanized IgG1 anti-CD22 antibody directed against the third extracellular domain (epitope B) of CD22. Whereas Epratuzumab displays no direct apoptosis and no CDC against lymphoma cells in vitro as compared to rituximab, moderate ADCC has been noted [119]. Although Epratuzumab induces a rapid internalization of the antigen antibody complex, under certain circumstances, Fc-receptor-bearing effector cells get enough opportunity to bind to the complex and exhibit cytotoxic activity. In clinical trials, epratuzumab has shown objective responses in 15–20% of patients with
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
243
relapsed and refractory CD22-positive diffuse large B-cell or indolent NHL [120, 121]. Furthermore, a combination of epratuzumab and rituximab revealed improved responses and CR rates compared to rituximab alone [122]. In children with relapsed ALL, epratuzumab has been investigated within a 14-days single drug window (360 mg/m², 2 times per week), followed by a 4-week combination with multidrug chemotherapy (360 mg/m², once a week). Toxicity was mostly limited to infusion related symptoms. Among 15 patients, 1 had a reduction of leukemic blasts in the blood, 3 had progressive disease and others had stable disease during the 2-week single drug window. Among 12 evaluable patients, 9 achieved a CR with the combination cycle and 7 of them with a negative MRD status post induction. Furthermore, one patient with partial response, one with stable disease and one with progressive disease were registered after the combination cycle. Although substantial anti-leukemic activity of the single drug has not been observed, the response in combination with chemotherapy and the high rate of MRD negativity in the high-risk patient population lead to the conclusion that epratuzumab plus chemotherapy has promising anti-leukemic activity and should be further investigated in prospective trials [123]. 12.4.4.4 Bispecific CD22 Targeting Antibodies A bispecific anti-CD22/anti-CD3 mouse moAb linked to ricin-A has been developed with the aim to deliver cytotoxicity to CD22-positive cells and activate at the same time a T-cell mediated immune response. T-cells were not affected by the immunotoxin and the whole construct exhibited a strong cytotoxic activity against B-cell lymphoma lines in vitro and in a SCID mouse model [124]. A bispecific anti-CD22/anti-CD28 was shown to enhance the anti-tumour activity of a T-cell engaging anti-CD19/anti-CD3 construct by co-stimulating the T-cells via the co-stimulatory T-cell antigen CD28 (Table 12.5) [125]. Preclinical experiences with a bispecific immunoconjugate targeting CD19/CD22 and linked to diphtheria toxin have been described in Sect. 12.4.2.5 [40], and a bispecific anti-CD22/antiCD20 construct has been mentioned in Sect. 12.4.3.4 [103]. 12.4.4.5 Immunoconjugates with CD22 The ribosome-inactivating protein saporin demonstrated a much better cytotoxicity in vitro against B-cell lymphoma lines when linked to a CD22-directed moAb compared to antibodies against CD19 or 37, demonstrating the better internalization with CD22 [126]. The immunoconjugate was also active in single patients with refractory B-NHL [127]. BL22 (CAT-3888,Cambridge Antibody Technology Group) is a recombinant immunotoxin containing a truncated form of the bacterial toxin Pseudomonas exotoxin A attached to an Fv fragment of a recombinant human anti-CD22 monoclonal antibody (Fig. 12.4). After binding to CD22, the compound is rapidly
CD22/CD28
BsAb anti-CD22/anti-CD28
Murine, humanized
Murine
CD22
CD22
Murine, humanized
CD22
Immunotoxin calicheamicin internalized after CD22 ligation Truncated Pseudomonas toxin. Recombinant antibody/toxin construct. CAT-8015 with improved CD22 affinity Linked to 90Yttrium
References
[136, 137, 202]
[117, 129, 130] Phase I/II trials in adults with HCL/CLL. Phase I trial in paediatric ALL Phase I/II trials in adults with refractory B-cell lymphoma
[132, 133]
In vitro and in xenografts
[125]
[124]
Preclinical, adult phase I/II [119–123] in lymphoma, paediatric phase I/II in relapsed ALL
Characteristics and study settings BsAb bispecific antibody; C-domain constant domain; CLL chronic lymphatic leukemia; HCL hairy-cell leukemia
90
Y-epratuzumab
Immunoconjugates Inotuzumab ozogamicin (Wyeth, Inc.) CAT-3888 (BL22) and CAT-8015 (HA22) (AstraZeneca, Inc.)
Murine
CD22/CD3
Bispecific BsAb anti-CD22/anti-CD3
Murine
Targets third extracellular domain, rapid internalization
Murine, humanized
Studies
SCID mouse model Linked to the toxin ricin-A. Toxicity against B cells, not T-cell. Additive T-cell mediated cytotoxicity T-cell co-stimulation via CD28 In vitro
Characteristics
Species
Table 12.5 MoAbs directed against CD22 Compounds Targets Unconjugated Epratuzumab (Immunomedics, Inc.) CD22
244 A. von Stackelberg
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
245
VH/L region
Truncated Pseudomonas toxin
Fig. 12.4 Recombinant immunotoxin containing a truncated form of the bacterial toxin Pseudomonas exotoxin A attached to the Fv fragment (variable region of a light and a heavy chain) of a recombinant human monoclonal antibody. The compound BL22 (CAT-3888, Cambridge Antibody Technology Group) targets CD22, the V-regions are linked with a disulphide bond, the compound Tac(Fv)-PE38 (LMB-2) targets CD25, V-regions are linked with a peptide linker (G4S)3
internalized, the toxin is processed and leads to cell death via caspase-mediated apoptosis [128]. The compound has shown activity in hairy-cell leukemia in phase I and II trials. Dose-limiting toxicities were haemolytic uremic syndrome in single patients, hypolbuminaemia, transaminase elevations, fatigue and oedema [129, 130]. A phase I trial in children with ALL demonstrated feasibility, acceptable toxicity and transient anti-leukemic activity in relapsed/refractory disease [131].
246
A. von Stackelberg
A second-generation recombinant compound, CAT-8015, with an optimized affinity to CD22 has been developed and compared to the parental compound CAT-3888 showed improved anti-tumour activity in vitro and in animal models [117]. After initial development and production by the National Cancer Institute (NCI), these compounds have been aquired by AstraZeneca for further development and marketing. CMC-544, inotuzumab ozogamicin, is a humanized monoclonal antibody directed against CD22 and linked to the toxin calicheamicin. A strong cytotoxic activity against the ETV6-RUNX1 REH cell-line was demonstrated in vitro and in mouse models [132, 133]. A superior cytotoxic efficiency compared to rituximab could be demonstrated as well as synergistic action of both compounds, if combined [133]. The drug is developed by the company Wyeth and is currently investigated in phase I/II trials for adult NHL. The mouse anti-CD22 antibody LL2 has been linked with 131Iodine. Treatment of patients with recurrent B-NHL in a phase I/II study resulted in a 30% response rate and quite severe haematological toxicity [134]. To diminish haematological toxicity, unconjugated anti-CD22 was administered prior to radioimmunotherapy, leading not only to reduced toxicity but also to reduced efficacy of the treatment [135]. Advancing this approach, the humanized anti-CD22 antibody epratuzumab was linked with the b emitter 90Yttrium. Improved anti-lymphoma activity was observed when given at a fractionated schedule [136]. The best therapeutic efficacy in a mouse model was achieved combining 90Y-epratuzumab with the anti-CD20 antibody veltuzumab (Table 12.5) [137].
12.4.5 CD79 CD79 is a transmembrane protein building the B-cell receptor complex covalently as CD79a and b heterodimer together with the surface immunoglobulin. Both CD79 chains contain an immunoreceptor tyrosine-based activation motif (ITAM) in their intracellular tails that propagates BCR signalling. CD79a is expressed as the earliest B-cell-specific antigen and maintained throughout maturation until the mature plasma cell [9]. It is not expressed on stem cells or on other tissues or cells of the body. As part of the BCR complex CD79 is internalized upon BCR antigen binding. The capacity of rapid internalization and the exclusive expression on B cells and B-lineage malignancies make it a suitable target for immunotherapy, in particular with immunoconjugates. Humanized moAbs against CD79a and b have been investigated in vitro and in xenograft models to determine their capacity to deplete B cells and B-NHL cell lines [138]. Whereas unconjugated antibodies were not able to destroy B cells, antibodies linked to immunotoxins (maleimidomethyl-cyclohexane-carboxylate, MCC) exhibited a strong cytotoxicity against B-NHL cells, in particular, when directed against CD79b. These data show that CD79 can be a suitable target for immunotoxic therapy and that more preclinical data with B-NHL and lymphoblastic leukemia cells should be generated [139].
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
247
12.5 T-Lineage-Specific Antigens For T-ALL, the development of immunotherapy is much less advanced. Few specific compounds have been developed, and no clinical trials in childhood ALL have been performed systematically. T-lineage-specific antigen directed therapy leads to T-cell depletion and thus to complications, in particular caused by viral infections.
12.5.1 CD2 The CD2 receptor is a co-stimulatory 50-kDa transmembrane glycoprotein with two extracellular immunoglobulin-like domains for antigen-specific lymphocyte activation found on T cells, natural killer (NK) cells and thymocytes. It is expressed on the majority but not all T-lineage ALL clones. CD2 plays a role in adhesion between activated T-cells or NK-cells and target cells and interacts with antigenpresenting or target cells via its ligand CD58 [140]. Internalization of CD2 upon binding to antibodies was less prominent compared to other T-lineage antigens such as CD3, 5 or 7 (Table 12.1) [141]. MEDI-507 (siplizumab) is a humanized monoclonal antibody directed against CD2. The antibody was effective against adult T-cell leukemia in a SCID mouse model [142]. The cytotoxic effect was predominantly attributed to NK-cell mediated ADCC [143]. The drug was primarily developed to treat graft-versus-host disease (GvHD) after allogeneic SCT. In a randomized phase I trial in adults with GvHD with corticosteroid co-medication, siplizumab was not associated with a higher toxicity compared to a placebo arm, but the therapeutic effect was only moderate [144]. In a phase I trial for children with GvHD, the drug was effective in reducing GvHD but was associated with severe viral infections in some patients [145]. Siplizumab showed efficacy in a phase I trial in adult patients with T-cell malignancies. However, while there was a marked T-cell depletion in all patients, 14% of patients developed EBV-associated LPD. This complication was attributed to the additional depletion of NK cells, leading to a complete deficiency of cells capable of controlling EBV-infected B cells [146]. Consequently, a clinical trial for treatment of adult T-cell lymphoma (sponsored by the NCI) is combining siplizumab with rituximab and chemotherapy with the aim to exploit the anti-lymphoma potency of siplizumab while avoiding the risk of EBV-associated LPD by depleting the B-cell compartment with rituximab (Table 12.6) [147].
12.5.2 CD3 CD3 represents a series of intermediate molecular-weight polypeptide chains (CD3a, CD3a, CD3a and CDa ) closely associated with a and g-subunits of the TCR that recognizes antigen-peptide epitopes presented by MHC molecules in a
CD4
CD5
CD7
CD25
Zanolimumab (HuMax-CD4; Genmab, Inc.)
T101
DA7
Tac(Fv)-PE38 (LMB-2)
Recombinant
Mouse
Mouse
Human
Mouse
VL/H domain linked with truncated diphtheria toxin
Prevention of MHC-II interaction, induction of ADCC Transient T-cell depletion, internalization of CD5 moAb linked to immunotoxin ricin A.
Unconjugated, active via ADCC, CDC, apoptosis
Characteristics Unconjugated, active via ADCC
Phase I in adult T-cell lymphoma. DLT severe vascular leak, little anti-tumour activity Preclinical data
Phase I adult T-cell lymphoma
Studies Phase I trial for GvHD treatment and for adult T-cell malignancies, induction of EBV-LPD Approved for treatment of allorejection after organ transplantation and for GvHD treatment. Anecdotal reports in T-ALL Phase I/II in adult cutaneous T-cell lymphoma
[5, 170]
[166, 167]
[159]
[156]
[140, 151]
References [142, 144–146]
Characteristics and study settings ADCC antibody-dependent cellular cytotoxicity; CDC complement-dependent cytotoxicity; DLT dose-limiting toxicity; EBV Epstein–Barr virus; GvHD graft–versus–host disease; LPD lymphoproliferative disease; VL/H variable light and heavy chain region
CD3
Muromonab-CD3 (OKT3®; Janssen Pharmaceutical, Inc.)
Table 12.6 MoAbs directed against T-lineage antigens Compounds Targets Species Siplizumab (MEDI-507) CD2 Rat, humanized
248 A. von Stackelberg
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
249
class-restricted manner. The intracellular regions of the CD3-subunits represent the signalling domains of the TCR complex that mediates T-cell activation. CD3 is expressed by all maturation stages of T-cells. Cytoplasmic CD3 is expressed on all T-ALLs, whereas surface membrane expression of the antigen is rare [148]. CD3 is rapidly internalized. A rapid endocytosis and degradation of anti-CD3 has been demonstrated after binding to the CD3 antigen. Thus, though CD3 is an attractive target for immunoconjugates, so far only unconjugated antibodies have been developed (Table 12.1) [29, 149]. Muromonab-CD3 (OKT3®; Janssen Pharmaceutica, Inc.) is a murine IgG2a monoclonal antibody directed against the 20-kDa CD3a subunit. It is approved for allograft rejection after organ transplantation and has been widely used in the treatment of acute GvHD after allogeneic SCT. T-cell depletion is induced via CDC, ADCC and direct apoptosis. Furthermore, binding of muromonab-CD3 to the CD3a subunit and interaction with Fc-receptor-bearing effector cells may result in activation of T-cell signalling and thus release of cytokines, which constitutes one of the major side effects of the drug. Furthermore, immunoregulatory T-cells that contribute to promoting immunological tolerance in autoimmunological and alloimmunological settings are also eliminated [150]. The substantial T-cell depletion results in immunodeficiency with risk for opportunistic infections and EBV-associated LPD [140]. Single reports on the efficacy of muromonab-CD3 in CD3-positive ALL have been published [151]. A humanized derivate of muromonab-CD3, teplizumab (hOKT3g1[Ala-Ala]), with a modified Fc-domain and markedly reduced ADCC activity has shown to be equally effective in modulating T-cell activity in mixed lymphocyte cultures [152]. It has been developed for the T-cell directed immunosuppressive treatment of type-1 diabetes in phase I-III trials. No experiences on treatment of acute GvHD or T-cell malignancies are reported (Table 12.6) [152, 153].
12.5.3 CD4 CD4 is a 55-kDa membrane glycoprotein with four immunoglobulin-like domains, a hydrophobic transmembrane domain and a long cytoplasmic tail [154]. CD4 acts as co-receptor for the TCR complex. It is expressed on helper and regulatory T-cells, and it recognizes antigens presented by MHC Class II molecules in association with the TCR. CD4 represents an attractive target for post-thymic T-cell malignancies with CD4+ phenotype (Table 12.1) [140]. Based on a murine anti-CD4 monoclonal antibody, a chimeric construct (cM-T412) was developed, which showed activity against cutaneous T-cell lymphoma with an acceptable side-effect profile [155]. Furthermore, a fully human IgG1a anti-CD4 monoclonal antibody, zanolimumab (HuMax-CD4; Genmab, Inc.), showed activity in patients with psoriasis and in patients with refractory cutaneous T-cell lymphoma (Table 12.6) [156].
250
A. von Stackelberg
12.5.4 CD5 CD5 (Leu-1) is a 67-kDa cysteine-rich scavenger receptor family glycoprotein expressed on T-cells and the B1a subset of B cells [157]. CD5 acts as a co-receptor regulating the signalling strength of the TCR as well as regulating immune tolerance. CD5 is internalized upon binding to specific antibodies (Table 12.1) [158]. Murine moAbs against CD5 have shown limited and transient efficacy in adult T-cell lymphoma. Treatment was complicated by a high rate of neutralizing antimouse antibodies and CD5 antigens were down regulated in the course of therapy [159]. The same antibody labelled with 90Yttrium showed limited activity in patients with refractory CD5-positive malignancies but was also associated with a high rate of human anti-mouse antibodies preventing any further treatment with the drug [160]. An immunoconjugate of anti-CD5 moAb and ricin A chain (XZ-CD5) did not show efficacy in patients with rheumatologic diseases and was not effective against alloreactive T-cells in the treatment of acute GvHD (Table 12.6) [161, 162].
12.5.5 CD7 CD7 is a transmembrane protein, member of the immunoglobulin super family, with one N-terminal domain. CD7 seems to be involved in the regulation of cytokine production. It is rapidly internalized upon binding to an antibody and, therefore, a suitable target for immunotoxins [163]. CD7 is expressed on T-cells of all maturation stages and on most T-ALL clones (Table 12.1). In a xenograft model of SCID mice with human T-ALL, the anti-CD7-Saporin immunotoxin was found to have a significant anti-leukemic activity, which apart from the direct cytotoxic effect, can in part be contributed to ADCC [164]. The best anti-leukemic activity was registered when combining anti-CD7-saporin with anti-CD38-saporin [165]. An anti-CD7 mouse antibody linked with ricin A was developed, primarily as specific agent against GvHD post-allogeneic SCT [166]. However, the drug was also investigated in a phase I trial for adults with refractory T-cell lymphoma. At the maximum tolerated dose, limited by quite severe vascular leak syndromes, only minor responses have been observed (Table 12.6) [167].
12.5.6 CD25 CD25 is the 55 kDa a-subunit of the IL2 receptor (IL2Ra). It forms a high-affinity heterodimer with the b- (CD122) and g-subunit (CD132). Alone, CD25 has a low affinity to IL2. CD25 is expressed at low levels in unstimulated T-cells, NK-cells and macrophages. In activated T-cells, it is highly expressed, and so is the case in many T- and some B-cell neoplasias [140]. In childhood T-ALL, CD25 expression has been found in about 75% of the investigated clones (Table 12.1) [168].
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
251
An inhibitory humanized IgG2 antibody daclizumab (Zentapax®) has been developed for prevention of IL2 activation of T-cells in prevention of allograft rejection, in particular in organ transplantation [169]. A recombinant immunotoxin targeting CD25 has been constructed, fusing CD25 specific variable regions of a light chain and a heavy chain linked with a truncated pseudomonas toxin, Tac(Fv)-PE38 (LMB-2). So far, only preclinical data have been reported and show a good activity of the compound in CD25-positive malignancies in vitro and in xenograft models [5, 170]. An alternative way to successfully target CD25 may be mentioned here. A recombinant protein has been composed fusing truncated diphtheria toxin with a human IL-2 fragment, Denileukin diftitox (DAB389IL-2, Ontak®). The compound uses the natural ligand of the functional IL2 receptor delivering the toxin into the target cell after internalization (Table 12.6) [171].
12.6 Myeloid Antigens 12.6.1 CD33 12.6.1.1 Biologic Characteristics of the CD33 Antigen CD33 is a 67-kDa transmembrane immunoglobulin-like lectin, which, in analogy to CD22, binds sialic acid at its N-terminus and, therefore, belongs to the SIGLEC family of lectins. Containing two immunoglobulin domains, CD33 is a member of the immunoglobulin superfamily. CD33 is preferably expressed on cells of the myeloid lineage, monocytes, macrophages and DCs but as well in some lymphoid cells and malignant lymphoblastic clones [172]. Closely related to CD33 are nine other SIGLECs (SIGLEC 5–11, 14, 16), which are differentially expressed on all cells of the innate immune system, including monocytes, macrophages, DCs, neutrophils, eosinophils, basophils and mast cells. This group of antigens exhibits an inhibitory regulative function via an intracellular pathway. The extracellular sialic acid receptor can also be activated by the host–pathogen interaction [173]. Otherwise, SIGLEC(s) on macrophages, DCs and other myeloid cells are believed to function as endocytic receptors in innate immune recognition of sialylated pathogens, including both bacteria (e.g. Neisseria meningitidis) and viruses. Consequently, CD33 is an antigen with rapid internalization upon binding with antibody and, therefore, an antigen in particular suitable for immunotoxins (Table 12.1) [174].
12.6.1.2 Preclinical and Clinical Use of Anti-CD33 MoAbs The use of CD33 as therapeutic target has in particular been investigated in acute myeloid leukemias, in which the antigen is highly expressed. In childhood ALL,
252
A. von Stackelberg
CD33 is co-expressed in some clones and may, therefore, be a suitable target in these cases. 12.6.1.3 Unconjugated Anti-CD33 MoAbs M195 is a monoclonal mouse antibody directed against CD33. In a phase I trial in patients with relapsed/refractory AML, no objective tumour responses have been observed, and treatment was limited by development of anti-mouse antibodies [175]. Based on these experiences, Lintuzumab (Hum195), a humanized derivate has been developed, which showed modest but observable activity as single agent in phase I and II trials in adult patients with relapsed/refractory AML [176, 177]. In particular with higher doses and repeated applications, activity and very good tolerability could be recently demonstrated in elderly patients [178]. In patients with newly diagnosed acute promyelocytic leukemia, in situations where cytological remission is associated with persistence of a positive MRD signal (as quantified by PCR), Lintuzumab as a single agent was able to induce MRD-negative remissions [179]. In a prospective phase III trial, the randomized introduction of Lintuzumab with mitoxantrone, etoposide and cytarabine during induction in adult AML reported good tolerance of the antibody, but no difference in CR and survival rates between both arms. (Table 12.7) [180].
12.6.1.4 Bispecific Antibodies Compared to monospecific CD33 antibodies, the bispecific construct BsAb CD33/ CD64 combining the anti-CD33 with anti-CD64 specificity exhibited a longer antiproliferative effect in AML cell lines and human AML cells in vitro. The inhibition was initiated more rapidly with the bispecific antibody compared to exposure to both unlinked antibodies and also induced a stronger ADCC (Table 12.7) [181].
12.6.1.5 Immunoconjugates with CD33 Gemtuzumab ozogamicin (Go,Mylotarg®) is the humanized monoclonal antibody HuM195 directed against CD33 linked with the toxin calicheamicin. The drug has been used in phase I/II trials for treatment of adult relapsed/refractory AML and achieved CR rates of 26% as single drug, [182] or of above 50% in combination with cytarabine [183, 184]. Furthermore, Go has been combined in a reduced dose of 3 mg/m² with a cytarabine, idarubicin, fludarabine (FLAI) regimen as induction therapy in adult patients with AML and achieved a CR rate of 90% with acceptable toxicity [185]. Single patients including children with acute lymphoblastic leukemia and CD33 co-expression have successfully been treated with Go alone or in combination [30, 186]. One known and dose-limiting side effect of Go is hepatotoxicity
CD33
Immunoconjugates Gemtuzumab ozogamicin (Fa Wyeth)
Murine, humanized
Murine, humanized
Murine
Murine, humanized
Nuclear localizing sequence routes Auger electron emitter to nucleus of target cell
Immunotoxin calicheamicin internalized after CD33 ligation
Enhancing the inhibitory function of CD33 by simultaneous binding to CD64, enhanced ADCC
Phase I study adult AML
Internalization after binding to antigen, no anti-tumour activity Better avidity to CD33 than M195, active against AML
Phase I-III trials in adult AML, singlepatient experience in childhood ALL In vitro
Preclinical studies
Phase I/II/III studies adult AML
Studies
Characteristics
Characteristics and study settings ADCC antibody-dependent cellular cytotoxicity; AML acute myeloid leukemia
111
CD33
CD33, CD64
Bispecific BsAB anti-CD33/ anti-CD64
-In-NLS-Hum195
CD33
HuM195 (Lintuzumab)
Table 12.7 MoAbs directed against CD33 Compounds Targets Species Unconjugated M195 CD33 Murine
[191, 192]
[182–185]
[181]
[176, 177, 180]
[175]
References
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia 253
254
A. von Stackelberg
and veno-occlusive disease in the context of allogeneic SCT. A reason for this is the expression of CD33 in Kupffer cells, which are specialized cells of the monocyte– macrophage system along the hepatic endothelium [187]. The CD33 antibody Hum195 linked to the recombinant plant toxin gelonin showed activity against CD33-positive cell lines in vitro but was not further developed for clinical trials [188]. CD33 has been investigated and used as target for radioimmunotherapy. The antibody Hum195 has been liked with the radionuclides 131Iodine and 90Yttrium, which exhibited substantial activity against AML cells but were associated with severe myelo- and hepatotoxicity [189]. 131Iodine anti-CD33 has been successfully used within the conditioning regimen of patients with AML prior to stem-cell transplantation [190]. To avoid this “crossfire” toxicity [111] in with its shortest emission of Auger electrons within a distance of less than 1 mm has been linked to Hum195. Additionally inserted nuclear localizing sequences (NLS) were shown to successfully route the radionuclide close to the nucleus, resulting in lethal activity against the target cell without harming surrounding tissue (111-In-NLS-Hum195; Table 12.7) [191, 192].
12.7 Discussion MoAbs represent a new therapeutic principle in the treatment of malignancies. The mechanism of action of classical chemotherapy is non-specific, with a mostly antiproliferative or apoptosis-inducing effect on malignant and normal cells. Accordingly, side effects have to be taken into account. In contrast, moAbs exhibit activity against cells expressing the specific antigen. The more exclusively the antigen is expressed on the tumour cells, the more targeted is the cytotoxic effect. Side effects depend on the extent of expression of the antigen on healthy cells. For the treatment of childhood ALL, antibodies should be applicable for most patients. Potential target antigens should be expressed by ideally all leukemic clones of one immunologic lineage, allowing for the conduct of systematic prospective controlled treatment trials. The broad panel of other antigens, expressed only in subgroups of ALL clones, is suitable only for interventional targeted treatment of individual patients in refractory disease. Since the numbers of such patients are small, a valid evaluation of the role of such moAbs in the context of multidrug chemotherapy will probably not be feasible.
12.7.1 The Most Important MoAbs Against Precursor B-Cell or T-ALL The most advanced development of moAbs in haematological malignancies has been achieved in treatment of adult B-cell NHL. Many of the compounds developed
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
255
for this indication target antigens that are as well expressed in childhood BCP ALL. The other broad field of development is antibodies against T-cell antigens, mostly developed for treatment of acute GvHD or for adult peripheral T-cell lymphoma. Again, many of these antigens are as well expressed on childhood T-ALL cells. Two cytotoxic mechanisms can be induced with moAbs: the physiologic cytotoxic capacity of the immune system and the targeted administration of toxins and radionuclides via immunoconjugation. The classical cellular cytotoxicity can be induced by binding of an unconjugated monoclonal antibody to the target antigen and reaction of Fc-receptor-positive effector cells, namely, NK-cells and granulocytes against the target cells. Additionally, the physiologic CDC can be induced by unconjugated antibodies. The efficacy of this approach has been extensively proven by the use of rituximab in CD20-positive B-cell lymphoma [85].
12.7.1.1 Unconjugated Antibodies The ideal target in ALL for unconjugated moAbs would be an antigen that is reliably and selectively expressed on the majority of the leukemic clones and that is not rapidly internalized. These criteria are best fulfilled by CD19, which is highly expressed in nearly all B-lineage ALL clones and remains rather stably on the cell surface (Table 12.1). The most promising unconjugated anti-CD19 antibody is the humanized antibody XmAb®5574 (Xencor) with a modified Fc domain. Currently, it is undergoing preclinical testing before entering early-phase clinical trials [60]. An alternative and promising approach is the bispecific murine single chain antibody anti-CD19/CD3 MT103 (Blinatumomab, Micromet, Inc.). This compound has been shown to be effective in B-cell lymphoma and adult B-lineage ALL and will hopefully enter clinical trials in childhood ALL in the near future. As a single agent, the drug was able to induce durable complete remissions in advanced and stage IV B-cell lymphoma, and MRD-negative remissions in patients with ALL and MRD persistence after conventional chemotherapy [38, 78]. Blinatumomab seems to be a promising immunotherapeutic compound for childhood ALL having entered the stage of clinical development. Although CD22 is a rapidly internalizing antigen and, therefore, better suited for targeting with immunotoxins, the unconjugated humanized anti-CD22 antibody epratuzumab has been shown to exhibit cytotoxicity against CD22-positive malignant cells. Epratuzumab has been used in combination with chemotherapy in childhood relapsed ALL with remarkable efficacy in a phase II trial [123]. Thus, the compound is one of the best developed moAbs in childhood ALL. The clinical efficacy of epratuzumab is interesting enough to be further investigated within a curative multidrug strategy in randomized prospective trials. For T-lineage ALL, the availability of suitable immunotherapeutic compounds is much more difficult. CD4 and 8 have been described as antigens being stably expressed on the cell surface. CD2 is expressed in the majority of paediatric T-ALL clones and showed a moderate propensity to internalization. Siplizumab, a
256
A. von Stackelberg
h umanized anti-CD2 antibody, has shown efficacy in adult T-cell lymphoma but was associated with EBV-associated LPD due to its T- and NK-cell depleting potency. A combination with rituximab to avoid this complication has been investigated in adult T-cell lymphoma [147] and could be an attractive approach for relapsed or refractory paediatric T-ALL as well. CD4 is variably expressed in paediatric T-ALL and, therefore, not suitable as target for a systematic approach but rather for individualized treatment in single patients failing cure with conventional therapy. Zanolimumab (HuMax-CD4), a humanized anti-CD4 antibody effective in adult post-thymic (mostly cutaneous) T-cell lymphoma could be a drug with interesting anti-leukemic potency in combination with conventional chemotherapy for treatment of individual patients with CD4-positive T-ALL. CD3 is also expressed in a subset of childhood T-ALL clones only and, therefore, rather a target for individual therapy than for prospective trials. Although CD3 has been described to internalize rather rapidly, the unconjugated antibody OKT3 has successfully been used for anti-leukemic therapy in single patients with refractory leukemia. CD8 has not been considered as a target for therapeutic moAbs. Antibodies directed against lineage-independent antigens have the disadvantage of targeting immunophenotype subgroups not involved in the leukemic phenotype. Alemtuzumab (Campath®) targeting CD52 has been shown to be active in adult T-NHL as single agent and in particular in combination with chemotherapy. Alemtuzumab might be an interesting agent in paediatric patients with refractory ALL, who are to be transplanted and need to achieve a remission prior to transplantation. However, when used for T-cell depletion as part of the conditioning regimen prior to SCT, alemtuzumab could not prevent subsequent relapses in patients with high MRD burden prior to SCT [193]. The long-lasting immunosuppressive effect is thought to interfere with the graft-versus-leukemia effect, an essential mechanism of allogeneic immune control of the disease. Thus, currently other T-cell depleting regimens mostly involving anti-thymocyte globulin are preferred in context with SCT in paediatric ALL. MoAbs targeting HLA-DR have not as yet reached the evaluation stage in childhood ALL. Apolizumab targets an epitope being expressed only in subgroups of HLA-DR-positive malignancies; alternative drugs with a broader spectrum have been used only in preclinical settings so far [55, 56]. 12.7.1.2 Conjugated Antibodies The most suitable targets for immunoconjugates are antigens highly and selectively expressed on ALL cells with rapid internalization after binding with the antibody, thus delivering the toxic compound right into the target cell. The best suitable antigen in PBC ALL is CD22, which is highly expressed in nearly all clones and which has been shown to internalize rapidly after binding (Table 12.1). Two immunoconjugates directed against CD22 are promising: Inotuzumab ozogamicin (Wyeth) is a humanized monoclonal antibody linked with calicheamicin, a powerful toxin that has been shown to be in particular effective in lymphoblastic leukemia cells [132].
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
257
CAT-3888 and its optimized successor CAT-8015 are recombinant anti-CD22 antibodies linked with Pseudomonas toxin A. A phase I trial with CAT-3888 in paediatric patients with relapsed/refractory ALL showed transient activity at dose levels below any dose-limiting toxicities [117]. Both compounds are certainly most suitable for development in paediatric ALL in the near future (Table 12.5). Another suitable target highly expressed on all precursor B-ALL clones with rapid internalization upon antibody binding is CD79. However, antibodies of immunotoxins against CD79 have not been developed for clinical use so far. The best expressed T-lineage antigens are CD5 and 7, both rather rapidly internalizing antigens. Targeting CD5 and CD7 has not been well developed. Efficacy of pure murine moAbs linked to ricin A was impaired by development of alloimmune antibodies against mouse antigens. Furthermore, viral infections due to T-cell depletion have been observed [159, 166, 167]. Humanized or even fully human moAbs linked to more potent toxins such as calicheamicin have not been developed for CD5 and 7. This would be a desirable field for future development, but since the patient population with CD5- or 7-positive haematological malignancies is small, the value of this approach is limited. CD33 is aberrantly expressed in some B- and T-lineage ALL clones and is a rapidly internalizing antigen suitable for immunotoxins. Gemtuzumab ozogamicin (Mylotarg®) has shown activity in AML and in CD33-positive ALL. It is a suitable drug for individual therapy of children with refractory CD33-positive ALL but not for a broader development in ALL. Radioimmunotherapy has been developed sparsely in paediatric ALL. Singlepatient experience exists with the CD20-directed compound 90Y-IbritumomabTiuxetan in the context of conditioning regimens prior to SCT. This is indeed the most interesting indication for this category of substances. It would be a desirable goal to eliminate total body irradiation as the most toxic element of classical conditioning regimens in ALL. It could be a subject of future investigations, whether TBI could be replaced by targeted immunoradiotherapy. Whereas the leukemic cells would be hit by radiation due to the targeted transport of the radionuclide, other tissues such as the lungs, the eyes, the brain, etc. would be protected from damage. Candidate compounds for that indication in B-lineage ALL could be 90Y-epratuzumab (anti-CD22), 131I-tositumomab (Bexxar®) or 90Y-Ibritumomab-Tiuxetan (Zevalin®) in CD20-positive leukemias, and for B- and T-lineage ALL 188Re-alemtuzumab (anti-CD52) or Lym1-SA 90Y-DOTA-biotin (anti-HLA-DR).
12.7.2 Targeting or not the Leukemic Stem Cell A clear definition of the leukemic stem cell has not been achieved yet. The plasticity of precursor cells imply a differentiation but also a development backwards towards a more immature status. Since cells of different maturation stages have been shown to generate leukemia in xenograft models, the term “leukemia-propagating cells”
258
A. von Stackelberg
was introduced as an alternative to the term “leukemia stem cell [194]”. A goal of anti-leukemic therapy must be to hit the leukemia-propagating cells, which are able to maintain the generation of leukemic clones besides reducing the burden of cells that do not have that capacity. The danger of targeted therapy with monoclonal antibody is indeed, that the bulk of leukemia cells is hit and reduced, whereas leukemia-propagating cells not expressing the target antigen would persist and induce relapse. One prominent example is the detection of single CD19 negative precursor cells with the capacity to induce leukemia in xenograft model among CD19-positive ALL samples [195]. The risk to miss the tumour stem cell or other tumour cells with downregulated target can be dramatically decreased by combining two targeted strategies, which could be effective even, if one of the targets is not sufficiently expressed by the tumour cell. Several preclinical and clinical studies have demonstrated better anti-tumour effect in the case of combination immunotherapy [165].
12.7.3 Time Point and Duration of Treatment, Co-Stimulatory Strategies The optimal duration of treatment with moAbs remains controversial. Most moAbs have been used for remission induction. In adult patients with B-cell lymphomas, rituximab has been used for maintenance therapy. Randomized trials have shown a survival advantage of patients with rituximab maintenance; however, relapses were postponed not prevented [196–198]. In paediatric patients with ALL, moAbs have been used so far only interventionally in patients with refractory disease with the aim to achieve the best possible remission prior to allogeneic SCT. This picture may change considering the impressive efficacy of single-agent moAbs such as blinatumomab, which can be given as maintenance therapy in patients with relapse post-SCT. Sequential analyses of leukemic blasts in the course of treatment have shown that different antigens behave differentially: some of them were downregulated such as CD10 and CD34, and others were upregulated, such as CD19 and CD20. This finding is important for antigen-directed therapy scheduled at a later time point of treatment, since the initially well-expressed target may be downregulated later on or vice versa [199]. One other important effect that has to be considered when designing a trial with moAbs is so-called tumour sink. At diagnosis, a large amount of target antigen binds rapidly a quantity of drug that would be overdosed in the situation of MRD. Accordingly, side effects on normal cells expressing the target antigen would be more severe in this situation [200]. Accurate pharmacokinetic studies have to be performed to determine the ideal antibody dose in different clinical situations. Diverse co-stimulatory approaches have been used to enhance the cytotoxic effect of moAbs. This includes the use of growth factors such as G-CSF and GM-CSF [55], or interleukins such as IL12 or IL2 [55]. The significance of these
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
259
additional immunomodulating strategies should be addressed in preclinical models and remains to be determined in controlled trials in the clinical setting.
12.8 Summary Development of moAbs for treatment of haematological malignancies is a rapidly growing field, mainly driven by their use in adults with B-cell or peripheral T-cell lymphomas. A variety of compounds are effective in childhood ALL. Unconjugated humanized antibodies usually do not induce acute severe side effects and may be easily combined with chemotherapy. Immunoconjugates exhibit more severe side effects and may be effective enough as single agents or combined very carefully. The most promising compounds for broad and prospective controlled development in paediatric ALL are the bispecific T-cell engaging anti-CD19 antibody blinatumomab, the unconjugated anti-CD22 humanized antibody epratuzumab and the unconjugated human anti-CD19 antibody XmAb5574. Others are the immunotoxins inotuzumab ozogamicin (humanized anti-CD22+calicheamicin), and CAT-8015 (recombinant anti-CD22+truncated pseudomonas toxin) with clinical experience in adult phase I trials. For treatment of T-ALL alemtuzumab, targeting the antigen CD52, may be considered. Additionally, siplizumab (humanized anti-CD2) may have a role in CD2-positive diseases. For a variety of other antigens such as CD3, CD4, CD20 and CD33, effective compounds are available that may be used in individual patients with refractory disease expressing the respective antigen. Radioimmunotherapy may be applied in individual refractory disease; furthermore, it might be a treatment option with the potential to replace TBI prior to allogeneic SCT in future trials. Well-organized multinational phase I/II and III trials are needed to investigate the panel of promising moAbs in paediatric ALL and to determine their importance in the context of contemporary multimodal treatment protocols. Within these trials, the optimal timing, dosage and duration of treatment with different moAbs remain to be determined. MoAbs exhibit a completely different mechanism of anti-leukemic action compared to conventional chemotherapy and certainly will change substantially the design of treatment strategies for childhood ALL in the future.
References 1. Kohler G, Milstein C. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature. 1975;256:495–497. 2. Riechmann L, Clark M, Waldmann H, Winter G. Reshaping human antibodies for therapy. Nature. 1988;332:323–327. 3. Lazar GA, Desjarlais JR, Jacinto J, Karki S, Hammond PW. A molecular immunology approach to antibody humanization and functional optimization. Mol Immunol. 2007;44:1986–1998. 4. Labrijn AF, Aalberse RC, Schuurman J. When binding is enough: nonactivating antibody formats. Curr Opin Immunol. 2008;20:479–485.
260
A. von Stackelberg
5. Kreitman RJ. Recombinant immunotoxins containing truncated bacterial toxins for the treatment of hematologic malignancies. BioDrugs. 2009;23:1–13. 6. Du X, Beers R, Fitzgerald DJ, Pastan I. Differential cellular internalization of anti-CD19 and -CD22 immunotoxins results in different cytotoxic activity. Cancer Res. 2008;68:6300–6305. 7. Sapra P, Allen TM. Internalizing antibodies are necessary for improved therapeutic efficacy of antibody-targeted liposomal drugs. Cancer Res. 2002;62:7190–7194. 8. Gudowius S, Recker K, Laws HJ, Dirksen U, Troger A, Wieczorek U, Furlan S, Gobel U, Hanenberg H. Identification of candidate target antigens for antibody-based immunotherapy in childhood B-cell precursor ALL. Klin Padiatr. 2006;218:327–333. 9. Bene MC. Immunophenotyping of acute leukemias. Immunol Lett. 2005;98:9–21. 10. Preijers FW, Tax WJ, De Witte T, Janssen A, vd Heijden H, Vidal H, Wessels JM, Capel PJ. Relationship between internalization and cytotoxicity of ricin A-chain immunotoxins. Br J Haematol. 1988;70:289–294. 11. Desjarlais JR, Lazar GA, Zhukovsky EA, Chu SY. Optimizing engagement of the immune system by anti-tumor antibodies: an engineer’s perspective. Drug Discov Today. 2007;12: 898–910. 12. van Mirre E, Breunis WB, Geissler J, Hack CE, de Boer M, Roos D, Kuijpers TW. Neutrophil responsiveness to IgG, as determined by fixed ratios of mRNA levels for activating and inhibitory FcgammaRII (CD32), is stable over time and unaffected by cytokines. Blood. 2006;108:584–590. 13. Pricop L, Redecha P, Teillaud JL, Frey J, Fridman WH, Sautes-Fridman C, Salmon JE. Differential modulation of stimulatory and inhibitory Fc gamma receptors on human monocytes by Th1 and Th2 cytokines. J Immunol. 2001;166:531–537. 14. Dranoff G. Cytokines in cancer pathogenesis and cancer therapy. Nat Rev Cancer. 2004;4:11–22. 15. Kashii Y, Giorda R, Herberman RB, Whiteside TL, Vujanovic NL. Constitutive expression and role of the TNF family ligands in apoptotic killing of tumor cells by human NK cells. J Immunol. 1999;163:5358–5366. 16. Boruchov AM, Heller G, Veri MC, Bonvini E, Ravetch JV, Young JW. Activating and inhibitory IgG Fc receptors on human DCs mediate opposing functions. J Clin Invest. 2005;115:2914–2923. 17. Michon JM, Gey A, Moutel S, Tartour E, Meresse V, Fridman W, Teillaud JL. In vivo induction of functional Fc gammaRI (CD64) on neutrophils and modulation of blood cytokine mRNA levels in cancer patients treated with G-CSF (rMetHuG-CSF). Br J Haematol. 1998;100:550–556. 18. Rech J, Repp R, Rech D, Stockmeyer B, Dechant M, Niedobitek G, Gramatzki M, Valerius T. A humanized HLA-DR antibody (hu1D10, apolizumab) in combination with granulocyte colony-stimulating factor (filgrastim) for the treatment of non-Hodgkin’s lymphoma: a pilot study. Leuk Lymphoma. 2006;47:2147–2154. 19. Dechant M, Bruenke J, Valerius T. HLA class II antibodies in the treatment of hematologic malignancies. Semin Oncol. 2003;30:465–475. 20. Selenko N, Majdic O, Jager U, Sillaber C, Stockl J, Knapp W. Cross-priming of cytotoxic T cells promoted by apoptosis-inducing tumor cell reactive antibodies? J Clin Immunol. 2002;22:124–130. 21. Selenko N, Maidic O, Draxier S, Berer A, Jager U, Knapp W, Stockl J. CD20 antibody (C2B8)-induced apoptosis of lymphoma cells promotes phagocytosis by dendritic cells and cross-priming of CD8+ cytotoxic T cells. Leukemia. 2001;15:1619-1626. 22. Idusogie EE, Wong PY, Presta LG, Gazzano-Santoro H, Totpal K, Ultsch M, Mulkerrin MG. Engineered antibodies with increased activity to recruit complement. J Immunol. 2001;166:2571–2575. 23. Dall’Acqua WF, Cook KE, Damschroder MM, Woods RM, Wu H. Modulation of the effector functions of a human IgG1 through engineering of its hinge region. J Immunol. 2006;177:1129–1138. 24. Nimmerjahn F, Ravetch JV. Divergent immunoglobulin g subclass activity through selective Fc receptor binding. Science. 2005;310:1510–1512.
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
261
25. Richards JO, Karki S, Lazar GA, Chen H, Dang W, Desjarlais JR. Optimization of antibody binding to FcgammaRIIa enhances macrophage phagocytosis of tumor cells. Mol Cancer Ther. 2008;7:2517–2527. 26. Lazar GA, Dang W, Karki S, Vafa O, Peng JS, Hyun L, Chan C, Chung HS, Eivazi A, Yoder SC, Vielmetter J, Carmichael DF, Hayes RJ, Dahiyat BI. Engineered antibody Fc variants with enhanced effector function. Proc Natl Acad Sci U S A. 2006;103:4005–4010. 27. Bowles JA, Wang SY, Link BK, Allan B, Beuerlein G, Campbell MA, Marquis D, Ondek B, Wooldridge JE, Smith BJ, Breitmeyer JB, Weiner GJ. Anti-CD20 monoclonal antibody with enhanced affinity for CD16 activates NK cells at lower concentrations and more effectively than rituximab. Blood. 2006;108:2648–2654. 28. Press OW, Howell-Clark J, Anderson S, Bernstein I. Retention of B-cell-specific monoclonal antibodies by human lymphoma cells. Blood. 1994;83:1390–1397. 29. Press OW, Hansen JA, Farr A, Martin PJ. Endocytosis and degradation of murine anti-human CD3 monoclonal antibodies by normal and malignant T-lymphocytes. Cancer Res. 1988;48: 2249–2257. 30. Zwaan CM, Reinhardt D, Jurgens H, Huismans DR, Hahlen K, Smith OP, Biondi A, van Wering ER, Feingold J, Kaspers GJ. Gemtuzumab ozogamicin in pediatric CD33-positive acute lymphoblastic leukemia: first clinical experiences and relation with cellular sensitivity to single agent calicheamicin. Leukemia. 2003;17:468–470. 31. Linenberger ML, Hong T, Flowers D, Sievers EL, Gooley TA, Bennett JM, Berger MS, Leopold LH, Appelbaum FR, Bernstein ID. Multidrug-resistance phenotype and clinical responses to gemtuzumab ozogamicin. Blood. 2001;98:988–994. 32. Arditti FD, Rabinkov A, Miron T, Reisner Y, Berrebi A, Wilchek M, Mirelman D. Apoptotic killing of B-chronic lymphocytic leukemia tumor cells by allicin generated in situ using a rituximab-alliinase conjugate. Mol Cancer Ther. 2005;4:325–331. 33. Press OW, Shan D, Howell-Clark J, Eary J, Appelbaum FR, Matthews D, King DJ, Haines AM, Hamann P, Hinman L, Shochat D, Bernstein ID. Comparative metabolism and retention of iodine-125, yttrium-90, and indium-111 radioimmunoconjugates by cancer cells. Cancer Res. 1996;56:2123–2129. 34. Sharkey RM, Goldenberg DM. Targeted therapy of cancer: new prospects for antibodies and immunoconjugates. CA Cancer J Clin. 2006;56:226–243. 35. Meredith RF, Buchsbaum DJ. Pretargeted radioimmunotherapy. Int J Radiat Oncol Biol Phys. 2006;66:S57–59. 36. Kufer P, Lutterbuse R, Baeuerle PA. A revival of bispecific antibodies. Trends Biotechnol. 2004;22:238–244. 37. Offner S, Hofmeister R, Romaniuk A, Kufer P, Baeuerle PA. Induction of regular cytolytic T cell synapses by bispecific single-chain antibody constructs on MHC class I-negative tumor cells. Mol Immunol. 2006;43:763–771. 38. Bargou R, Leo E, Zugmaier G, Klinger M, Goebeler M, Knop S, Noppeney R, Viardot A, Hess G, Schuler M, Einsele H, Brandl C, Wolf A, Kirchinger P, Klappers P, Schmidt M, Riethmuller G, Reinhardt C, Baeuerle PA, Kufer P. Tumor regression in cancer patients by very low doses of a T cell-engaging antibody. Science. 2008;321:974–977. 39. Loffler A, Kufer P, Lutterbuse R, Zettl F, Daniel PT, Schwenkenbecher JM, Riethmuller G, Dorken B, Bargou RC. A recombinant bispecific single-chain antibody, CD19 X CD3, induces rapid and high lymphoma-directed cytotoxicity by unstimulated T lymphocytes. Blood. 2000;95:2098–2103. 40. Vallera DA, Todhunter DA, Kuroki DW, Shu Y, Sicheneder A, Chen H. A bispecific recombinant immunotoxin, DT2219, targeting human CD19 and CD22 receptors in a mouse xenograft model of B-cell leukemia/lymphoma. Clin Cancer Res. 2005;11:3879–3888. 41. WHO Expert Committee on Specifications for Pharmaceutical Preparations. World Health Organ Tech Rep Ser. 2008:1–138. 42. Rodig SJ, Abramson JS, Pinkus GS, Treon SP, Dorfman DM, Dong HY, Shipp MA, Kutok JL. Heterogeneous CD52 expression among hematologic neoplasms: implications for the use of alemtuzumab (CAMPATH-1H). Clin Cancer Res. 2006;12:7174–7179.
262
A. von Stackelberg
43. Golay J, Cortiana C, Manganini M, Cazzaniga G, Salvi A, Spinelli O, Bassan R, Barbui T, Biondi A, Rambaldi A, Introna M. The sensitivity of acute lymphoblastic leukemia cells carrying the t(12;21) translocation to campath-1H-mediated cell lysis. Haematologica. 2006;91:322–330. 44. Nuckel H, Frey UH, Roth A, Duhrsen U, Siffert W. Alemtuzumab induces enhanced apoptosis in vitro in B-cells from patients with chronic lymphocytic leukemia by antibody-dependent cellular cytotoxicity. Eur J Pharmacol. 2005;514:217–224. 45. Mone AP, Cheney C, Banks AL, Tridandapani S, Mehter N, Guster S, Lin T, Eisenbeis CF, Young DC, Byrd JC. Alemtuzumab induces caspase-independent cell death in human chronic lymphocytic leukemia cells through a lipid raft-dependent mechanism. Leukemia. 2006;20:272–279. 46. Gribben JG, Hallek M. Rediscovering alemtuzumab: current and emerging therapeutic roles. Br J Haematol. 2009;144:818–831. 47. Keating M, Coutre S, Rai K, Osterborg A, Faderl S, Kennedy B, Kipps T, Bodey G, Byrd JC, Rosen S, Dearden C, Dyer MJ, Hillmen P. Management guidelines for use of alemtuzumab in B-cell chronic lymphocytic leukemia. Clin Lymphoma. 2004;4:220–227. 48. Kennedy-Nasser AA, Bollard CM, Myers GD, Leung KS, Gottschalk S, Zhang Y, Liu H, Heslop HE, Brenner MK, Krance RA. Comparable outcome of alternative donor and matched sibling donor hematopoietic stem cell transplant for children with acute lymphoblastic leukemia in first or second remission using alemtuzumab in a myeloablative conditioning regimen. Biol Blood Marrow Transplant. 2008;14:1245–1252. 49. Myers GD, Krance RA, Weiss H, Kuehnle I, Demmler G, Heslop HE, Bollard CM. Adenovirus infection rates in pediatric recipients of alternate donor allogeneic bone marrow transplants receiving either antithymocyte globulin (ATG) or alemtuzumab (Campath). Bone Marrow Transplant. 2005;36:1001–1008. 50. Shah AJ, Kapoor N, Crooks GM, Weinberg KI, Azim HA, Killen R, Kuo L, Rushing T, Kohn DB, Parkman R. The effects of Campath 1H upon graft-versus-host disease, infection, relapse, and immune reconstitution in recipients of pediatric unrelated transplants. Biol Blood Marrow Transplant. 2007;13:584–593. 51. Curtis RE, Travis LB, Rowlings PA, Socie G, Kingma DW, Banks PM, Jaffe ES, Sale GE, Horowitz MM, Witherspoon RP, Shriner DA, Weisdorf DJ, Kolb HJ, Sullivan KM, Sobocinski KA, Gale RP, Hoover RN, Fraumeni JF, Jr., Deeg HJ. Risk of lymphoproliferative disorders after bone marrow transplantation: a multi-institutional study. Blood. 1999;94: 2208–2216. 52. De Decker M, Bacher K, Thierens H, Slegers G, Dierckx RA, De Vos F. In vitro and in vivo evaluation of direct rhenium-188-labeled anti-CD52 monoclonal antibody alemtuzumab for radioimmunotherapy of B-cell chronic lymphocytic leukemia. Nucl Med Biol. 2008;35: 599–604. 53. Robinson J, Waller MJ, Parham P, de Groot N, Bontrop R, Kennedy LJ, Stoehr P, Marsh SG. IMGT/HLA and IMGT/MHC: sequence databases for the study of the major histocompatibility complex. Nucleic Acids Res. 2003;31:311–314. 54. Kostelny SA, Link BK, Tso JY, Vasquez M, Jorgensen BH, Wang H, Hall WC, Weiner GJ. Humanization and characterization of the anti-HLA-DR antibody 1D10. Int J Cancer. 2001;93:556–565. 55. Stockmeyer B, Schiller M, Repp R, Lorenz HM, Kalden JR, Gramatzki M, Valerius T. Enhanced killing of B lymphoma cells by granulocyte colony-stimulating factor-primed effector cells and Hu1D10 – a humanized human leucocyte antigen DR antibody. Br J Haematol. 2002;118:959–967. 56. Tawara T, Hasegawa K, Sugiura Y, Tahara T, Ishida I, Kataoka S. Fully human antibody exhibits pan-human leukocyte antigen-DR recognition and high in vitro/vivo efficacy against human leukocyte antigen-DR-positive lymphomas. Cancer Sci. 2007;98:921–928. 57. Pagel JM, Orgun N, Hamlin DK, Wilbur DS, Gooley TA, Gopal AK, Park SI, Green DJ, Lin Y, Press OW. A comparative analysis of conventional and pretargeted radioimmunotherapy of B-cell lymphomas by targeting CD20, CD22, and HLA-DR singly and in combinations. Blood. 2009;113:4903–4913.
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
263
58. Uckun FM, Kersey JH, Haake R, Weisdorf D, Ramsay NK. Autologous bone marrow transplantation in high-risk remission B-lineage acute lymphoblastic leukemia using a cocktail of three monoclonal antibodies (BA-1/CD24, BA-2/CD9, and BA-3/CD10) plus complement and 4-hydroperoxycyclophosphamide for ex vivo bone marrow purging. Blood. 1992;79: 1094–1104. 59. Hasegawa M, Fujimoto M, Poe JC, Steeber DA, Tedder TF. CD19 can regulate B lymphocyte signal transduction independent of complement activation. J Immunol. 2001;167: 3190–3200. 60. Horton HM, Bernett MJ, Pong E, Peipp M, Karki S, Chu SY, Richards JO, Vostiar I, Joyce PF, Repp R, Desjarlais JR, Zhukovsky EA. Potent in vitro and in vivo activity of an Fc-engineered anti-CD19 monoclonal antibody against lymphoma and leukemia. Cancer Res. 2008;68: 8049–8057. 61. Anderson KC, Bates MP, Slaughenhoupt BL, Pinkus GS, Schlossman SF, Nadler LM. Expression of human B cell-associated antigens on leukemias and lymphomas: a model of human B cell differentiation. Blood. 1984;63:1424–1433. 62. Uckun FM, Jaszcz W, Ambrus JL, Fauci AS, Gajl-Peczalska K, Song CW, Wick MR, Myers DE, Waddick K, Ledbetter JA. Detailed studies on expression and function of CD19 surface determinant by using B43 monoclonal antibody and the clinical potential of anti-CD19 immunotoxins. Blood. 1988;71:13–29. 63. Tedder TF, Inaoki M, Sato S. The CD19-CD21 complex regulates signal transduction thresholds governing humoral immunity and autoimmunity. Immunity. 1997;6:107–118. 64. Yazawa N, Hamaguchi Y, Poe JC, Tedder TF. Immunotherapy using unconjugated CD19 monoclonal antibodies in animal models for B lymphocyte malignancies and autoimmune disease. Proc Natl Acad Sci U S A. 2005;102:15178–15183. 65. Vlasveld LT, Hekman A, Vyth-Dreese FA, Melief CJ, Sein JJ, Voordouw AC, Dellemijn TA, Rankin EM. Treatment of low-grade non-Hodgkin’s lymphoma with continuous infusion of low-dose recombinant interleukin-2 in combination with the B-cell-specific monoclonal antibody CLB-CD19. Cancer Immunol Immunother. 1995;40:37–47. 66. Hekman A, Honselaar A, Vuist WM, Sein JJ, Rodenhuis S, ten Bokkel Huinink WW, Somers R, Rumke P, Melief CJ. Initial experience with treatment of human B cell lymphoma with anti-CD19 monoclonal antibody. Cancer Immunol Immunother. 1991;32:364–372. 67. Grossbard ML, Lambert JM, Goldmacher VS, Spector NL, Kinsella J, Eliseo L, Coral F, Taylor JA, Blattler WA, Epstein CL, et al. Anti-B4-blocked ricin: a phase I trial of 7-day continuous infusion in patients with B-cell neoplasms. J Clin Oncol. 1993;11:726–737. 68. Multani PS, O’Day S, Nadler LM, Grossbard ML. Phase II clinical trial of bolus infusion anti-B4 blocked ricin immunoconjugate in patients with relapsed B-cell non-Hodgkin’s lymphoma. Clin Cancer Res. 1998;4:2599–2604. 69. Rowland AJ, Pietersz GA, McKenzie IF. Preclinical investigation of the antitumour effects of anti-CD19-idarubicin immunoconjugates. Cancer Immunol Immunother. 1993;37: 195–202. 70. Sapra P, Allen TM. Improved outcome when B-cell lymphoma is treated with combinations of immunoliposomal anticancer drugs targeted to both the CD19 and CD20 epitopes. Clin Cancer Res. 2004;10:2530–2537. 71. Schwemmlein M, Stieglmaier J, Kellner C, Peipp M, Saul D, Oduncu F, Emmerich B, Stockmeyer B, Lang P, Beck JD, Fey GH. A CD19-specific single-chain immunotoxin mediates potent apoptosis of B-lineage leukemic cells. Leukemia. 2007;21:1405–1412. 72. Ingle GS, Chan P, Elliott JM, Chang WS, Koeppen H, Stephan JP, Scales SJ. High CD21 expression inhibits internalization of anti-CD19 antibodies and cytotoxicity of an anti-CD19drug conjugate. Br J Haematol. 2008;140:46–58. 73. Gerber HP, Kung-Sutherland M, Stone I, Morris-Tilden C, Miyamoto J, McCormick R, Alley SC, Okeley N, Hayes B, Hernandez-Ilizaliturri FJ, McDonagh CF, Carter PJ, Benjamin D, Grewal IS. Potent antitumor activity of the anti-CD19 auristatin antibody drug conjugate hBU12-vcMMAE against rituximab-sensitive and -resistant lymphomas. Blood. 2009;113:4352–4361.
264
A. von Stackelberg
74. Stieglmaier J, Bremer E, Kellner C, Liebig TM, ten Cate B, Peipp M, Schulze-Koops H, Pfeiffer M, Buhring HJ, Greil J, Oduncu F, Emmerich B, Fey GH, Helfrich W. Selective induction of apoptosis in leukemic B-lymphoid cells by a CD19-specific TRAIL fusion protein. Cancer Immunol Immunother. 2008;57:233–246. 75. Molhoj M, Crommer S, Brischwein K, Rau D, Sriskandarajah M, Hoffmann P, Kufer P, Hofmeister R, Baeuerle PA. CD19-/CD3-bispecific antibody of the BiTE class is far superior to tandem diabody with respect to redirected tumor cell lysis. Mol Immunol. 2007;44: 1935–1943. 76. Brandl C, Haas C, d’Argouges S, Fisch T, Kufer P, Brischwein K, Prang N, Bargou R, Suzich J, Baeuerle PA, Hofmeister R. The effect of dexamethasone on polyclonal T cell activation and redirected target cell lysis as induced by a CD19/CD3-bispecific single-chain antibody construct. Cancer Immunol Immunother. 2007;56:1551–1563. 77. d’Argouges S, Wissing S, Brandl C, Prang N, Lutterbuese R, Kozhich A, Suzich J, Locher M, Kiener P, Kufer P, Hofmeister R, Baeuerle PA, Bargou RC. Combination of rituximab with blinatumomab (MT103/MEDI-538), a T cell-engaging CD19-/CD3-bispecific antibody, for highly efficient lysis of human B lymphoma cells. Leuk Res. 2008;33:465–473. 78. Topp M, Goekbuget N, Kufer P, Zugmaier G, Degenhard E, Neumann S, Horst HA, Viardot A, Schmid M, Ottmann OG, Schmidt M, Reinhardt C, Baeuerle PA, Nagorsen D, Hoelzer D, Bargou RC. Treatment with Anti-CD19 BiTE Antibody Blinatumomab (MT103 / MEDI-538) Is Able to Eliminate Minimal Residual Disease (MRD) in Patients with B-Precursor Acute Lymphoblastic Leukemia (ALL): First Results of An Ongoing Phase II Study. ASH Annual Meeting Abstracts. 2008;112:1926. 79. Kellner C, Bruenke J, Stieglmaier J, Schwemmlein M, Schwenkert M, Singer H, Mentz K, Peipp M, Lang P, Oduncu F, Stockmeyer B, Fey GH. A novel CD19-directed recombinant bispecific antibody derivative with enhanced immune effector functions for human leukemic cells. J Immunother. 2008;31:871–884. 80. Cragg MS, Walshe CA, Ivanov AO, Glennie MJ. The biology of CD20 and its potential as a target for mAb therapy. Curr Dir Autoimmun. 2005;8:140–174. 81. Walshe CA, Beers SA, French RR, Chan CH, Johnson PW, Packham GK, Glennie MJ, Cragg MS. Induction of cytosolic calcium flux by CD20 is dependent upon B Cell antigen receptor signaling. J Biol Chem. 2008;283:16971–16984. 82. Tedder TF, Forsgren A, Boyd AW, Nadler LM, Schlossman SF. Antibodies reactive with the B1 molecule inhibit cell cycle progression but not activation of human B lymphocytes. Eur J Immunol. 1986;16:881–887. 83. Janas E, Priest R, Wilde JI, White JH, Malhotra R. Rituxan (anti-CD20 antibody)-induced translocation of CD20 into lipid rafts is crucial for calcium influx and apoptosis. Clin Exp Immunol. 2005;139:439–446. 84. Beers SA, Chan CH, James S, French RR, Attfield KE, Brennan CM, Ahuja A, Shlomchik MJ, Cragg MS, Glennie MJ. Type II (tositumomab) anti-CD20 monoclonal antibody out performs type I (rituximab-like) reagents in B-cell depletion regardless of complement activation. Blood. 2008;112:4170–4177. 85. Coiffier B, Lepage E, Briere J, Herbrecht R, Tilly H, Bouabdallah R, Morel P, Van Den Neste E, Salles G, Gaulard P, Reyes F, Lederlin P, Gisselbrecht C. CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. N Engl J Med. 2002;346:235–242. 86. Rhein P, Scheid S, Ratei R, Hagemeier C, Seeger K, Kirschner-Schwabe R, Moericke A, Schrappe M, Spang R, Ludwig WD, Karawajew L. Gene expression shift towards normal B cells, decreased proliferative capacity and distinct surface receptors characterize leukemic blasts persisting during induction therapy in childhood acute lymphoblastic leukemia. Leukemia. 2007;21:897–905. 87. Dworzak MN, Schumich A, Printz D, Potschger U, Husak Z, Attarbaschi A, Basso G, Gaipa G, Ratei R, Mann G, Gadner H. CD20 up-regulation in pediatric B-cell precursor acute lymphoblastic leukemia during induction treatment: setting the stage for anti-CD20 directed immunotherapy. Blood. 2008;112:3982–3988.
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
265
88. Corbacioglu S, Eber S, Gungor T, Hummerjohann J, Niggli F. Induction of long-term remission of a relapsed childhood B-acute lymphoblastic leukemia with rituximab chimeric anti-CD20 monoclonal antibody and autologous stem cell transplantation. J Pediatr Hematol Oncol. 2003;25:327–329. 89. Claviez A, Eckert C, Seeger K, Schrauder A, Schrappe M, Henze G, von Stackelberg A. Rituximab plus chemotherapy in children with relapsed or refractory CD20-positive B-cell precursor acute lymphoblastic leukemia. Haematologica. 2006;91:272–273. 90. Morris ES, Vora A. Remission induction with single agent Rituximab in a child with multiply relapsed precursor-B ALL. Br J Haematol. 2007;139:344–345. 91. McLaughlin P, Grillo-Lopez AJ, Link BK, Levy R, Czuczman MS, Williams ME, Heyman MR, Bence-Bruckler I, White CA, Cabanillas F, Jain V, Ho AD, Lister J, Wey K, Shen D, Dallaire BK. Rituximab chimeric anti-CD20 monoclonal antibody therapy for relapsed indolent lymphoma: half of patients respond to a four-dose treatment program. J Clin Oncol. 1998;16:2825–2833. 92. Attias D, Weitzman S. The efficacy of rituximab in high-grade pediatric B-cell lymphoma/ leukemia: a review of available evidence. Curr Opin Pediatr. 2008;20:17–22. 93. Gokbuget N, Hoelzer D. Novel antibody-based therapy for acute lymphoblastic leukemia. Best Pract Res Clin Haematol. 2006;19:701–713. 94. Kiss F, Buslig J, Szegedi I, Scholtz B, Kappelmayer J, Kiss C. Early relapse after rituximab chemoimmunotherapy. Pediatr Blood Cancer. 2008;50:372–375. 95. Kraal K, Schalij-Delfos N, van Buchem M, Egeler M, Ball L. Optic nerve relapse in a child with common acute lymphoblastic leukemia, treated with systemic anti-CD-20 (rituximab). Haematologica. 2005;90 Suppl:ECR24. 96. Seifert G, Reindl T, Lobitz S, Seeger K, Henze G. Fatal course after administration of rituximab in a boy with relapsed all: a case report and review of literature. Haematologica. 2006;91:ECR23. 97. Goldenberg DM, Rossi EA, Stein R, Cardillo TM, Czuczman MS, Hernandez-Ilizaliturri FJ, Hansen HJ, Chang CH. Properties and structure-function relationships of veltuzumab (hA20), a humanized anti-CD20 monoclonal antibody. Blood. 2009;113:1062–1070. 98. Morschhauser F, Leonard JP, Fayad L, Coiffier B, Petillon MO, Coleman M, Schuster SJ, Dyer MJ, Horne H, Teoh N, Wegener WA, Goldenberg DM. Humanized anti-CD20 antibody, veltuzumab, in refractory/recurrent non-Hodgkin’s lymphoma: phase I/II results. J Clin Oncol. 2009;27:3346–3353. 99. Nishida M, Usuda S, Okabe M, Miyakoda H, Komatsu M, Hanaoka H, Teshigawara K, Niwa O. Characterization of novel murine anti-CD20 monoclonal antibodies and their comparison to 2B8 and c2B8 (rituximab). Int J Oncol. 2007;31:29–40. 100. Nishida M, Teshigawara K, Niwa O, Usuda S, Nakamura T, Ralph P, Newman R, Padlan EA. Novel humanized anti-CD20 monoclonal antibodies with unique germline VH and VL gene recruitment and potent effector functions. Int J Oncol. 2008;32:1263–1274. 101. Teeling JL, French RR, Cragg MS, van den Brakel J, Pluyter M, Huang H, Chan C, Parren PW, Hack CE, Dechant M, Valerius T, van de Winkel JG, Glennie MJ. Characterization of new human CD20 monoclonal antibodies with potent cytolytic activity against non-Hodgkin lymphomas. Blood. 2004;104:1793–1800. 102. Hagenbeek A, Gadeberg O, Johnson P, Pedersen LM, Walewski J, Hellmann A, Link BK, Robak T, Wojtukiewicz M, Pfreundschuh M, Kneba M, Engert A, Sonneveld P, Flensburg M, Petersen J, Losic N, Radford J. First clinical use of ofatumumab, a novel fully human anti-CD20 monoclonal antibody in relapsed or refractory follicular lymphoma: results of a phase 1/2 trial. Blood. 2008;111:5486–5495. 103. Qu Z, Goldenberg DM, Cardillo TM, Shi V, Hansen HJ, Chang CH. Bispecific anti-CD20/22 antibodies inhibit B-cell lymphoma proliferation by a unique mechanism of action. Blood. 2008;111:2211–2219. 104. Rossi EA, Goldenberg DM, Cardillo TM, Stein R, Chang CH. Hexavalent bispecific antibodies represent a new class of anticancer therapeutics: 1. Properties of anti-CD20/CD22 antibodies in lymphoma. Blood. 2009;113:6161–6171.
266
A. von Stackelberg
105. Gall JM, Davol PA, Grabert RC, Deaver M, Lum LG. T cells armed with anti-CD3 x antiCD20 bispecific antibody enhance killing of CD20+ malignant B cells and bypass complement-mediated rituximab resistance in vitro. Exp Hematol. 2005;33:452–459. 106. Xiong D, Xu Y, Liu H, Peng H, Shao X, Lai Z, Fan D, Yang M, Han J, Xie Y, Yang C, Zhu Z. Efficient inhibition of human B-cell lymphoma xenografts with an anti-CD20 x anti-CD3 bispecific diabody. Cancer Lett. 2002;177:29–39. 107. Stanglmaier M, Faltin M, Ruf P, Bodenhausen A, Schroder P, Lindhofer H. Bi20 (fBTA05), a novel trifunctional bispecific antibody (anti-CD20 X anti-CD3), mediates efficient killing of B-cell lymphoma cells even with very low CD20 expression levels. Int J Cancer. 2008;123:1181–1189. 108. Buhmann R, Simoes B, Stanglmaier M, Yang T, Faltin M, Bund D, Lindhofer H, Kolb HJ. Immunotherapy of recurrent B-cell malignancies after allo-SCT with Bi20 (FBTA05), a trifunctional anti-CD3 ´ anti-CD20 antibody and donor lymphocyte infusion. Bone Marrow Transplant. 2009;43:383–397. 109. Rossi EA, Goldenberg DM, Cardillo TM, Stein R, Chang CH. CD20-targeted tetrameric interferon-{alpha}, a novel and potent immunocytokine for the therapy of B-cell lymphomas. Blood. 2009;114(18):3864–3871. 110. Wahl RL. Tositumomab and (131)I therapy in non-Hodgkin’s lymphoma. J Nucl Med. 2005;46 Suppl 1:128S–140S. 111. Witzig TE, Gordon LI, Cabanillas F, Czuczman MS, Emmanouilides C, Joyce R, Pohlman BL, Bartlett NL, Wiseman GA, Padre N, Grillo-Lopez AJ, Multani P, White CA. Randomized controlled trial of yttrium-90-labeled ibritumomab tiuxetan radioimmunotherapy versus rituximab immunotherapy for patients with relapsed or refractory low-grade, follicular, or transformed B-cell non-Hodgkin’s lymphoma. J Clin Oncol. 2002;20:2453–2463. 112. Krishnan A, Nademanee A, Fung HC, Raubitschek AA, Molina A, Yamauchi D, Rodriguez R, Spielberger RT, Falk P, Palmer JM, Forman SJ. Phase II trial of a transplantation regimen of yttrium-90 ibritumomab tiuxetan and high-dose chemotherapy in patients with nonHodgkin’s lymphoma. J Clin Oncol. 2008;26:90–95. 113. Jacene HA, Filice R, Kasecamp W, Wahl RL. Comparison of 90Y-ibritumomab tiuxetan and 131I-tositumomab in clinical practice. J Nucl Med. 2007;48:1767–1776. 114. Cooney-Qualter E, Krailo M, Angiolillo A, Fawwaz RA, Wiseman G, Harrison L, Kohl V, Adamson PC, Ayello J, vande Ven C, Perkins SL, Cairo MS. A phase I study of 90yttrium-ibritumomab-tiuxetan in children and adolescents with relapsed/refractory CD20-positive non-Hodgkin’s lymphoma: a Children’s Oncology Group study. Clin Cancer Res. 2007;13: 5652s–5660s. 115. Nitschke L. CD22 and Siglec-G: B-cell inhibitory receptors with distinct functions. Immunol Rev. 2009;230:128–143. 116. Tedder TF, Poe JC, Haas KM. CD22: a multifunctional receptor that regulates B lymphocyte survival and signal transduction. Adv Immunol. 2005;88:1–50. 117. Alderson RF, Kreitman RJ, Chen T, Yeung P, Herbst R, Fox JA, Pastan I. CAT-8015: a second-generation pseudomonas exotoxin A-based immunotherapy targeting CD22-expressing hematologic malignancies. Clin Cancer Res. 2009;15:832–839. 118. Engel P, Nojima Y, Rothstein D, Zhou LJ, Wilson GL, Kehrl JH, Tedder TF. The same epitope on CD22 of B lymphocytes mediates the adhesion of erythrocytes, T and B lymphocytes, neutrophils, and monocytes. J Immunol. 1993;150:4719–4732. 119. Carnahan J, Stein R, Qu Z, Hess K, Cesano A, Hansen HJ, Goldenberg DM. Epratuzumab, a CD22-targeting recombinant humanized antibody with a different mode of action from rituximab. Mol Immunol. 2007;44:1331–1341. 120. Leonard JP, Coleman M, Ketas JC, Chadburn A, Furman R, Schuster MW, Feldman EJ, Ashe M, Schuster SJ, Wegener WA, Hansen HJ, Ziccardi H, Eschenberg M, Gayko U, Fields SZ, Cesano A, Goldenberg DM. Epratuzumab, a humanized anti-CD22 antibody, in aggressive nonHodgkin’s lymphoma: phase I/II clinical trial results. Clin Cancer Res. 2004;10:5327–5334. 121. Leonard JP, Coleman M, Ketas JC, Chadburn A, Ely S, Furman RR, Wegener WA, Hansen HJ, Ziccardi H, Eschenberg M, Gayko U, Cesano A, Goldenberg DM. Phase I/II trial of epratuzumab (humanized anti-CD22 antibody) in indolent non-Hodgkin’s lymphoma. J Clin Oncol. 2003;21:3051–3059.
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
267
122. Leonard JP, Schuster SJ, Emmanouilides C, Couture F, Teoh N, Wegener WA, Coleman M, Goldenberg DM. Durable complete responses from therapy with combined epratuzumab and rituximab: final results from an international multicenter, phase 2 study in recurrent, indolent, non-Hodgkin lymphoma. Cancer. 2008;113:2714–2723. 123. Raetz EA, Cairo MS, Borowitz MJ, Blaney SM, Krailo MD, Leil TA, Reid JM, Goldenberg DM, Wegener WA, Carroll WL, Adamson PC. Chemoimmunotherapy reinduction with epratuzumab in children with acute lymphoblastic leukemia in marrow relapse: a Children’s Oncology Group Pilot Study. J Clin Oncol. 2008;26:3756–3762. 124. Shen GL, Li JL, Vitetta ES. Bispecific anti-CD22/anti-CD3-ricin A chain immunotoxin is cytotoxic to Daudi lymphoma cells but not T cells in vitro and shows both A-chain-mediated and LAK-T-mediated killing. J Immunol. 1994;152:2368–2376. 125. Bohlen H, Manzke O, Patel B, Moldenhauer G, Dorken B, von Fliedner V, Diehl V, Tesch H. Cytolysis of leukemic B-cells by T-cells activated via two bispecific antibodies. Cancer Res. 1993;53:4310–4314. 126. Bonardi MA, French RR, Amlot P, Gromo G, Modena D, Glennie MJ. Delivery of saporin to human B-cell lymphoma using bispecific antibody: targeting via CD22 but not CD19, CD37, or immunoglobulin results in efficient killing. Cancer Res. 1993;53:3015–3021. 127. French RR, Hamblin TJ, Bell AJ, Tutt AL, Glennie MJ. Treatment of B-cell lymphomas with combination of bispecific antibodies and saporin. Lancet. 1995;346:223–224. 128. Decker T, Oelsner M, Kreitman RJ, Salvatore G, Wang QC, Pastan I, Peschel C, Licht T. Induction of caspase-dependent programmed cell death in B-cell chronic lymphocytic leukemia by anti-CD22 immunotoxins. Blood. 2004;103:2718–2726. 129. Kreitman RJ, Wilson WH, Bergeron K, Raggio M, Stetler-Stevenson M, FitzGerald DJ, Pastan I. Efficacy of the anti-CD22 recombinant immunotoxin BL22 in chemotherapyresistant hairy-cell leukemia. N Engl J Med. 2001;345:241–247. 130. Kreitman RJ, Squires DR, Stetler-Stevenson M, Noel P, FitzGerald DJ, Wilson WH, Pastan I. Phase I trial of recombinant immunotoxin RFB4(dsFv)-PE38 (BL22) in patients with B-cell malignancies. J Clin Oncol. 2005;23:6719–6729. 131. Wayne AS, Kreitman RJ, Findley HW, Lew G, Delbrook C, Steinberg SM, StetlerStevenson M, Fitzgerald DJ, Pastan I. Anti-CD22 immunotoxin RFB4(dsFv)-PE38 (BL22) for CD22-positive hematologic malignancies of childhood: preclinical studies and phase I clinical trial. Clin Cancer Res. 2010;16:1894–1903. 132. Dijoseph JF, Dougher MM, Armellino DC, Evans DY, Damle NK. Therapeutic potential of CD22-specific antibody-targeted chemotherapy using inotuzumab ozogamicin (CMC-544) for the treatment of acute lymphoblastic leukemia. Leukemia. 2007;21:2240–2245. 133. DiJoseph JF, Dougher MM, Kalyandrug LB, Armellino DC, Boghaert ER, Hamann PR, Moran JK, Damle NK. Antitumor efficacy of a combination of CMC-544 (inotuzumab ozogamicin), a CD22-targeted cytotoxic immunoconjugate of calicheamicin, and rituximab against non-Hodgkin’s B-cell lymphoma. Clin Cancer Res. 2006;12:242–249. 134. Vose JM, Colcher D, Gobar L, Bierman PJ, Augustine S, Tempero M, Leichner P, Lynch JC, Goldenberg D, Armitage JO. Phase I/II trial of multiple dose 131Iodine-MAb LL2 (CD22) in patients with recurrent non-Hodgkin’s lymphoma. Leuk Lymphoma. 2000;38:91–101. 135. Linden O, Tennvall J, Cavallin-Stahl E, Darte L, Garkavij M, Lindner KJ, Ljungberg M, Ohlsson T, Sjogreen K, Wingardh K, Strand SE. Radioimmunotherapy using 131I-labeled anti-CD22 monoclonal antibody (LL2) in patients with previously treated B-cell lymphomas. Clin Cancer Res. 1999;5:3287s–3291s. 136. Linden O, Hindorf C, Cavallin-Stahl E, Wegener WA, Goldenberg DM, Horne H, Ohlsson T, Stenberg L, Strand SE, Tennvall J. Dose-fractionated radioimmunotherapy in non-Hodgkin’s lymphoma using DOTA-conjugated, 90Y-radiolabeled, humanized anti-CD22 monoclonal antibody, epratuzumab. Clin Cancer Res. 2005;11:5215–5222. 137. Mattes MJ, Sharkey RM, Karacay H, Czuczman MS, Goldenberg DM. Therapy of advanced B-lymphoma xenografts with a combination of 90Y-anti-CD22 IgG (epratuzumab) and unlabeled anti-CD20 IgG (veltuzumab). Clin Cancer Res. 2008;14:6154–6160.
268
A. von Stackelberg
138. Li Y, Chen F, Putt M, Koo YK, Madaio M, Cambier JC, Cohen PL, Eisenberg RA. B cell depletion with anti-CD79 mAbs ameliorates autoimmune disease in MRL/lpr mice. J Immunol. 2008;181:2961–2972. 139. Polson AG, Yu SF, Elkins K, Zheng B, Clark S, Ingle GS, Slaga DS, Giere L, Du C, Tan C, Hongo JA, Gogineni A, Cole MJ, Vandlen R, Stephan JP, Young J, Chang W, Scales SJ, Ross S, Eaton D, Ebens A. Antibody-drug conjugates targeted to CD79 for the treatment of nonHodgkin lymphoma. Blood. 2007;110:616–623. 140. Morris JC, Waldmann TA, Janik JE. Receptor-directed therapy of T-cell leukemias and lymphomas. J Immunotoxicol. 2008;5:235–248. 141. Matthay KK, Abai AM, Cobb S, Hong K, Papahadjopoulos D, Straubinger RM. Role of ligand in antibody-directed endocytosis of liposomes by human T-leukemia cells. Cancer Res. 1989;49:4879–4886. 142. Zhang Z, Zhang M, Ravetch JV, Goldman C, Waldmann TA. Effective therapy for a murine model of adult T-cell leukemia with the humanized anti-CD2 monoclonal antibody, MEDI507. Blood. 2003;102:284–288. 143. Branco L, Barren P, Mao SY, Pfarr D, Kaplan R, Postema C, Langermann S, Koenig S, Johnson S. Selective deletion of antigen-specific, activated T cells by a humanized MAB to CD2 (MEDI-507) is mediated by NK cells. Transplantation. 1999;68:1588–1596. 144. Adkins D, Ratanatharathorn V, Yang H, White B. Safety profile and clinical outcomes in a phase I, placebo-controlled study of siplizumab in acute graft-versus-host disease. Transplantation. 2009;88:198–202. 145. Brochstein JA, Grupp S, Yang H, Pillemer SR, Geba GP. Phase-1 study of siplizumab in the treatment of pediatric patients with at least grade II newly diagnosed acute graft-versus-host disease. Pediatr Transplant. 2010;14(2):233–241. 146. O’Mahony D, Morris JC, Stetler-Stevenson M, Matthews H, Brown MR, Fleisher T, Pittaluga S, Raffeld M, Albert PS, Reitsma D, Kaucic K, Hammershaimb L, Waldmann TA, Janik JE. EBV-related lymphoproliferative disease complicating therapy with the anti-CD2 monoclonal antibody, siplizumab, in patients with T-cell malignancies. Clin Cancer Res. 2009;15:2514–2522. 147. Wilson WH, National Cancer Institute, US. Siplizumab, combination chemotherapy, and rituximab in treating patients with T-cell or natural killer-cell non-hodgkin lymphoma. In: ClinicalTrials.gov [Internet]. Bethesda (MD): National Library of Medicine (US). NLM identifier: NCT00832936 [cited 2009 Oct 16]; 2008. 148. Ledbetter JA, Schieven GL, Kuebelbeck VM, Uckun FM. Accessory receptors regulate coupling of the T-cell receptor complex to tyrosine kinase activation and mobilization of cytoplasmic calcium in T-lineage acute lymphoblastic leukemia. Blood. 1991;77:1271–1282. 149. Press OW, Vitetta ES, Farr AG, Hansen JA, Martin PJ. Evaluation of ricin A-chain immunotoxins directed against human T cells. Cell Immunol. 1986;102:10–20. 150. Chatenoud L, Bach JF. Resetting the functional capacity of regulatory T cells: a novel immunotherapeutic strategy to promote immune tolerance. Expert Opin Biol Ther. 2005;5 Suppl 1:S73–81. 151. Gramatzki M, Burger R, Strobel G, Trautmann U, Bartram CR, Helm G, Horneff G, Alsalameh S, Jonker M, Gebhart E, et al. Therapy with OKT3 monoclonal antibody in refractory T cell acute lymphoblastic leukemia induces interleukin-2 responsiveness. Leukemia. 1995;9:382–390. 152. Popma SH, Griswold DE, Li L. Anti-CD3 antibodies OKT3 and hOKT3gamma1(Ala-Ala) induce proliferation of T cells but impair expansion of alloreactive T cells; aspecifc T cell proliferation induced by anti-CD3 antibodies correlates with impaired expansion of alloreactive T cells. Int Immunopharmacol. 2005;5:155–162. 153. Kaufman A, Herold KC. Anti-CD3 mAbs for treatment of type 1 diabetes. Diabetes Metab Res Rev. 2009;25:302–306. 154. Leahy DJ. A structural view of CD4 and CD8. Faseb J. 1995;9:17–25. 155. Knox S, Hoppe RT, Maloney D, Gibbs I, Fowler S, Marquez C, Cornbleet PJ, Levy R. Treatment of cutaneous T-cell lymphoma with chimeric anti-CD4 monoclonal antibody. Blood. 1996;87:893–899.
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
269
156. Kim YH, Duvic M, Obitz E, Gniadecki R, Iversen L, Osterborg A, Whittaker S, Illidge TM, Schwarz T, Kaufmann R, Cooper K, Knudsen KM, Lisby S, Baadsgaard O, Knox SJ. Clinical efficacy of zanolimumab (HuMax-CD4): two phase 2 studies in refractory cutaneous T-cell lymphoma. Blood. 2007;109:4655–4662. 157. Lozano F, Simarro M, Calvo J, Vila JM, Padilla O, Bowen MA, Campbell KS. CD5 signal transduction: positive or negative modulation of antigen receptor signaling. Crit Rev Immunol. 2000;20:347–358. 158. Sieber T, Schoeler D, Ringel F, Pascu M, Schriever F. Selective internalization of monoclonal antibodies by B-cell chronic lymphocytic leukemia cells. Br J Haematol. 2003;121:458–461. 159. Dillman RO, Beauregard J, Shawler DL, Halpern SE, Markman M, Ryan KP, Baird SM, Clutter M. Continuous infusion of T101 monoclonal antibody in chronic lymphocytic leukemia and cutaneous T-cell lymphoma. J Biol Response Mod. 1986;5:394–410. 160. Foss FM, Raubitscheck A, Mulshine JL, Fleisher TA, Reynolds JC, Paik CH, Neumann RD, Boland C, Perentesis P, Brown MR, Frincke JM, Lollo CP, Larson SM, Carrasquillo JA. Phase I study of the pharmacokinetics of a radioimmunoconjugate, 90Y-T101, in patients with CD5-expressing leukemia and lymphoma. Clin Cancer Res. 1998;4:2691–2700. 161. Olsen NJ, Brooks RH, Cush JJ, Lipsky PE, St Clair EW, Matteson EL, Gold KN, Cannon GW, Jackson CG, McCune WJ, Fox DA, Nelson B, Lorenz T, Strand V. A double-blind, placebo-controlled study of anti-CD5 immunoconjugate in patients with rheumatoid arthritis. The Xoma RA Investigator Group. Arthritis Rheum. 1996;39:1102–1108. 162. Przepiorka D, LeMaistre CF, Huh YO, Luna M, Saria EA, Brown CT, Champlin RE. Evaluation of anti-CD5 ricin A chain immunoconjugate for prevention of acute graft-vs.-host disease after HLA-identical marrow transplantation. Ther Immunol. 1994;1:77–82. 163. Pauza ME, Doumbia SO, Pennell CA. Construction and characterization of human CD7specific single-chain Fv immunotoxins. J Immunol. 1997;158:3259–3269. 164. Flavell DJ, Warnes S, Noss A, Flavell SU. Host-mediated antibody-dependent cellular cytotoxicity contributes to the in vivo therapeutic efficacy of an anti-CD7-saporin immunotoxin in a severe combined immunodeficient mouse model of human T-cell acute lymphoblastic leukemia. Cancer Res. 1998;58:5787–5794. 165. Flavell DJ, Boehm DA, Noss A, Warnes SL, Flavell SU. Therapy of human T-cell acute lymphoblastic leukemia with a combination of anti-CD7 and anti-CD38-SAPORIN immunotoxins is significantly better than therapy with each individual immunotoxin. Br J Cancer. 2001;84:571–578. 166. van Oosterhout YV, van Emst L, Schattenberg AV, Tax WJ, Ruiter DJ, Spits H, Nagengast FM, Masereeuw R, Evers S, de Witte T, Preijers FW. A combination of anti-CD3 and antiCD7 ricin A-immunotoxins for the in vivo treatment of acute graft versus host disease. Blood. 2000;95:3693–3701. 167. Frankel AE, Laver JH, Willingham MC, Burns LJ, Kersey JH, Vallera DA. Therapy of patients with T-cell lymphomas and leukemias using an anti-CD7 monoclonal antibody-ricin A chain immunotoxin. Leuk Lymphoma. 1997;26:287–298. 168. Karawajew L, Ruppert V, Wuchter C, Kosser A, Schrappe M, Dorken B, Ludwig WD. Inhibition of in vitro spontaneous apoptosis by IL-7 correlates with bcl-2 up-regulation, cortical/mature immunophenotype, and better early cytoreduction of childhood T-cell acute lymphoblastic leukemia. Blood. 2000;96:297–306. 169. Nashan B, Light S, Hardie IR, Lin A, Johnson JR. Reduction of acute renal allograft rejection by daclizumab. Daclizumab Double Therapy Study Group. Transplantation. 1999;67:110–115. 170. Kreitman RJ, Bailon P, Chaudhary VK, FitzGerald DJ, Pastan I. Recombinant immunotoxins containing anti-Tac(Fv) and derivatives of Pseudomonas exotoxin produce complete regression in mice of an interleukin-2 receptor-expressing human carcinoma. Blood. 1994;83: 426–434. 171. Foss FM. DAB(389)IL-2 (ONTAK): a novel fusion toxin therapy for lymphoma. Clin Lymphoma. 2000;1:110-116; discussion 117. 172. Tanimoto M, Scheinberg DA, Cordon-Cardo C, Huie D, Clarkson BD, Old LJ. Restricted expression of an early myeloid and monocytic cell surface antigen defined by monoclonal antibody M195. Leukemia. 1989;3:339–348.
270
A. von Stackelberg
1 73. Crocker PR. Siglecs in innate immunity. Curr Opin Pharmacol. 2005;5:431–437. 174. O’Reilly MK, Paulson JC. Siglecs as targets for therapy in immune-cell-mediated disease. Trends Pharmacol Sci. 2009;30:240–248. 175. Scheinberg DA, Lovett D, Divgi CR, Graham MC, Berman E, Pentlow K, Feirt N, Finn RD, Clarkson BD, Gee TS, et al. A phase I trial of monoclonal antibody M195 in acute myelogenous leukemia: specific bone marrow targeting and internalization of radionuclide. J Clin Oncol. 1991;9:478–490. 176. Caron PC, Dumont L, Scheinberg DA. Supersaturating infusional humanized anti-CD33 monoclonal antibody HuM195 in myelogenous leukemia. Clin Cancer Res. 1998;4: 1421–1428. 177. Feldman E, Kalaycio M, Weiner G, Frankel S, Schulman P, Schwartzberg L, Jurcic J, VelezGarcia E, Seiter K, Scheinberg D, Levitt D, Wedel N. Treatment of relapsed or refractory acute myeloid leukemia with humanized anti-CD33 monoclonal antibody HuM195. Leukemia. 2003;17:314–318. 178. Raza A, Jurcic JG, Roboz GJ, Maris M, Stephenson JJ, Wood BL, Feldman EJ, Galili N, Grove LE, Drachman JG, Sievers EL. Complete remissions observed in acute myeloid leukemia following prolonged exposure to lintuzumab: a phase 1 trial. Leuk Lymphoma. 2009;50:1–9. 179. Jurcic JG, DeBlasio T, Dumont L, Yao TJ, Scheinberg DA. Molecular remission induction with retinoic acid and anti-CD33 monoclonal antibody HuM195 in acute promyelocytic leukemia. Clin Cancer Res. 2000;6:372–380. 180. Feldman EJ, Brandwein J, Stone R, Kalaycio M, Moore J, O’Connor J, Wedel N, Roboz GJ, Miller C, Chopra R, Jurcic JC, Brown R, Ehmann WC, Schulman P, Frankel SR, De Angelo D, Scheinberg D. Phase III randomized multicenter study of a humanized anti-CD33 monoclonal antibody, lintuzumab, in combination with chemotherapy, versus chemotherapy alone in patients with refractory or first-relapsed acute myeloid leukemia. J Clin Oncol. 2005;23:4110–4116. 181. Balaian L, Ball ED. Inhibition of acute myeloid leukemia cell growth by mono-specific and bi-specific anti-CD33 x anti-CD64 antibodies. Leuk Res. 2004;28:821–829. 182. Larson RA, Sievers EL, Stadtmauer EA, Lowenberg B, Estey EH, Dombret H, Theobald M, Voliotis D, Bennett JM, Richie M, Leopold LH, Berger MS, Sherman ML, Loken MR, van Dongen JJ, Bernstein ID, Appelbaum FR. Final report of the efficacy and safety of gemtuzumab ozogamicin (Mylotarg) in patients with CD33-positive acute myeloid leukemia in first recurrence. Cancer. 2005;104:1442–1452. 183. Chevallier P, Delaunay J, Turlure P, Pigneux A, Hunault M, Garand R, Guillaume T, AvetLoiseau H, Dmytruk N, Girault S, Milpied N, Ifrah N, Mohty M, Harousseau JL. Long-term disease-free survival after gemtuzumab, intermediate-dose cytarabine, and mitoxantrone in patients with CD33(+) primary resistant or relapsed acute myeloid leukemia. J Clin Oncol. 2008;26:5192–5197. 184. Kobayashi Y, Tobinai K, Takeshita A, Naito K, Asai O, Dobashi N, Furusawa S, Saito K, Mitani K, Morishima Y, Ogura M, Yoshiba F, Hotta T, Bessho M, Matsuda S, Takeuchi J, Miyawaki S, Naoe T, Usui N, Ohno R. Phase I/II study of humanized anti-CD33 antibody conjugated with calicheamicin, gemtuzumab ozogamicin, in relapsed or refractory acute myeloid leukemia: final results of Japanese multicenter cooperative study. Int J Hematol. 2009;89:460–469. 185. Candoni A, Martinelli G, Toffoletti E, Chiarvesio A, Tiribelli M, Malagola M, Piccaluga PP, Michelutti A, Simeone E, Damiani D, Russo D, Fanin R. Gemtuzumab-ozogamicin in combination with fludarabine, cytarabine, idarubicin (FLAI-GO) as induction therapy in CD33positive AML patients younger than 65 years. Leuk Res. 2008;32:1800–1808. 186. Chevallier P, Mahe B, Garand R, Talmant P, Harousseau JL, Delaunay J. Combination of chemotherapy and gemtuzumab ozogamicin in adult Philadelphia positive acute lymphoblastic leukemia patient harboring CD33 expression. Int J Hematol. 2008;88:209–211. 187. Chevallier P, Prebet T, Turlure P, Hunault M, Vigouroux S, Harousseau JL, Blaise D, Ifrah N, Milpied N, Mohty M. Prior treatment with Gemtuzumab Ozogamicin and the risk of veno-occlusive disease after allogeneic haematopoietic stem cell transplantation. Bone Marrow Transplant. 2010;45:165–170.
12 Monoclonal Antibodies in Paediatric Acute Lymphoblastic Leukemia
271
188. Pagliaro LC, Liu B, Munker R, Andreeff M, Freireich EJ, Scheinberg DA, Rosenblum MG. Humanized M195 monoclonal antibody conjugated to recombinant gelonin: an anti-CD33 immunotoxin with antileukemic activity. Clin Cancer Res. 1998;4:1971–1976. 189. Burke JM, Jurcic JG, Scheinberg DA. Radioimmunotherapy for acute leukemia. Cancer Control. 2002;9:106–113. 190. Burke JM, Caron PC, Papadopoulos EB, Divgi CR, Sgouros G, Panageas KS, Finn RD, Larson SM, O’Reilly RJ, Scheinberg DA, Jurcic JG. Cytoreduction with iodine-131-antiCD33 antibodies before bone marrow transplantation for advanced myeloid leukemias. Bone Marrow Transplant. 2003;32:549–556. 191. Chen P, Wang J, Hope K, Jin L, Dick J, Cameron R, Brandwein J, Minden M, Reilly RM. Nuclear localizing sequences promote nuclear translocation and enhance the radiotoxicity of the anti-CD33 monoclonal antibody HuM195 labeled with 111In in human myeloid leukemia cells. J Nucl Med. 2006;47:827–836. 192. Kersemans V, Cornelissen B, Minden MD, Brandwein J, Reilly RM. Drug-resistant AML cells and primary AML specimens are killed by 111In-anti-CD33 monoclonal antibodies modified with nuclear localizing peptide sequences. J Nucl Med. 2008;49:1546–1554. 193. Knechtli CJ, Goulden NJ, Hancock JP, Grandage VL, Harris EL, Garland RJ, Jones CG, Rowbottom AW, Hunt LP, Green AF, Clarke E, Lankester AW, Cornish JM, Pamphilon DH, Steward CG, Oakhill A. Minimal residual disease status before allogeneic bone marrow transplantation is an important determinant of successful outcome for children and adolescents with acute lymphoblastic leukemia. Blood. 1998;92:4072–4079. 194. le Viseur C, Hotfilder M, Bomken S, Wilson K, Rottgers S, Schrauder A, Rosemann A, Irving J, Stam RW, Shultz LD, Harbott J, Jurgens H, Schrappe M, Pieters R, Vormoor J. In childhood acute lymphoblastic leukemia, blasts at different stages of immunophenotypic maturation have stem cell properties. Cancer Cell. 2008;14:47–58. 195. Cox CV, Diamanti P, Evely RS, Kearns PR, Blair A. Expression of CD133 on leukemiainitiating cells in childhood ALL. Blood. 2009;113:3287–3296. 196. Ghielmini M. Patient benefits of maintenance immunotherapy. Leuk Res. 2006;30 Suppl 1: S22–26. 197. Vidal L, Gafter-Gvili A, Leibovici L, Dreyling M, Ghielmini M, Hsu Schmitz SF, Cohen A, Shpilberg O. Rituximab maintenance for the treatment of patients with follicular lymphoma: systematic review and meta-analysis of randomized trials. J Natl Cancer Inst. 2009;101: 248–255. 198. Moccia A, Ghielmini M. Monoclonal antibodies for the treatment of hematologic malignancies: schedule and maintenance therapy. Semin Hematol. 2008;45:75–84. 199. Gaipa G, Basso G, Maglia O, Leoni V, Faini A, Cazzaniga G, Bugarin C, Veltroni M, Michelotto B, Ratei R, Coliva T, Valsecchi MG, Biondi A, Dworzak MN. Drug-induced immunophenotypic modulation in childhood ALL: implications for minimal residual disease detection. Leukemia. 2005;19:49–56. 200. Koon HB, Severy P, Hagg DS, Butler K, Hill T, Jones AG, Waldmann TA, Junghans RP. Antileukemic effect of daclizumab in CD25 high-expressing leukemias and impact of tumor burden on antibody dosing. Leuk Res. 2006;30:190–203. 201. Consolini R, Legitimo A, Rondelli R, Guguelmi C, Barisone E, Lippi A, Cantu–Rajnoldi A, Arico M, Conter V, Cocito MG, Putti MC, Pession A, Masera G, Biondi A, Basso G. Clinical relevance of CD10 expression in childhood ALL. The Italian Association for Pediatric Hematology and Oncology (AIEOP). Haematologica. 1998;83:967–973. 202. Sharkey RM, Brenner A, Burton J, Hajjar G, Toder SP, Alavi A, Matthies A, Tsai DE, Schuster SJ, Stadtmauer EA, Czuczman MS, Lamonica D, Kraeber-Bodere F, Mahe B, Chatal JF, Rogatko A, Mardirrosian G, Goldenberg DM. Radioimmunotherapy of nonHodgkin’s lymphoma with 90Y-DOTA humanized anti-CD22 IgG (90Y-Epratuzumab): do tumor targeting and dosimetry predict therapeutic response? J Nucl Med. 2003;44: 2000–2018.
Chapter 13
Therapeutic Utility of Proteasome Inhibitors for Acute Leukemia Joya Chandra and Claudia P. Miller
13.1 Introduction Cells expressing chromosomal translocations, seen in specific acute lymphoblastic leukemia (ALL) subsets and described elsewhere in this monograph, are thought to rely more heavily on the ubiquitin-proteasome pathway, which is the primary disposal mechanism for mutated, damaged, or aberrantly expressed proteins [31]. Therefore, leukemic cells are considered to rely upon this processing pathway more than normal lymphocytes. This is a possible Achilles heel and provides an opportunity for targeted therapy [59, 63]. The terminal step in the ubiquitin-proteasome pathway is the ATP-dependent degradation of substrate proteins by the proteasome, a multisubunit catalytic complex. In the last decade, inhibiting the 20S proteasome has become a therapeutically achievable endeavor. Bortezomib (Velcade; Millenium Pharmaceuticals), the first and only proteasome inhibitor approved by the US Food and Drug Administration (FDA), is currently in use for the treatment of specific hematological malignancies such as multiple myeloma and mantle cell lymphoma [94]. Several lines of evidence suggest that acute leukemia patients might benefit from the use of proteasome inhibitors. A number of independent studies have shown that primary cells derived from leukemia patients, as well as human leukemia cell lines and mouse models for leukemia are susceptible to cell death by proteasome inhibition. In 2004, a Phase-I clinical trial of bortezomib in 15 refractory/relapsed acute leukemia patients was conducted at the University of Texas M.D. Anderson Cancer Center and identified a maximum tolerated dose and demonstrated significant proteasome inhibition at that dose and lower doses [17]. Thus, this class of drugs holds promise for improving outcomes in currently incurable acute leukemia patients. Consequently, second-generation proteasome inhibitors are emerging with unique properties as compared to bortezomib. Also, synergy observed in combining
J. Chandra (*) Department of Pediatrics Research, Children’s Cancer Hospital at M.D. Anderson, University of Texas M.D. Anderson Cancer Center, 1515 Holcombe Boulevard, Houston, TX 77030, USA e-mail:
[email protected] V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3_13, © Springer Science+Business Media, LLC 2011
273
274
J. Chandra and C.P. Miller
proteasome inhibitors with other anticancer agents portend additional avenues of applying these agents clinically. Here, we review proteasome inhibitors and their mechanism of action in leukemia cells, clinical trials with proteasome inhibitors, and discuss potential combination therapies with proteasome inhibitors that may show promise for use in ALL.
13.2 Proteasome Inhibitors 13.2.1 Rationale for Targeting the Proteasome The life of a protein in a cell can range from a few minutes to days. Proteins are constantly being recycled in the cell, broken up into short peptides and amino acids and replaced by newly generated proteins. The proteasome, which forms part of the ubiquitin-proteasome pathway (Fig. 13.1), plays a critical role in the cell by degrading up to 80% of intracellular proteins [18, 105]. Attachment of a polyubiquitin chain (which is coordinated by a series of specific enzymatic reactions) designates proteins to be disposed of by the proteasome [36]. The entire complex responsible for the degradation of ubiquitinated proteins is known as the 26S proteasome and is composed of two regulatory 19S subunits that flank a 20S catalytic core [87]. The regulatory subunits recognize and unfold ubiquitinated proteins in an ATPdependent manner, and transfer the unfolded proteins into the chamber of the 20S proteasome [102]. The barrel-shaped catalytic core is made up of alpha and beta subunits, which form four stacked rings. The two outer rings consist of only alpha subunits, whereas the middle rings are composed of beta subunits. The three proteolytic activities reside in the beta subunits. They consist of a chymotrypsin-like (cleaving after hydrophobic regions), a caspase-like (hydrolyzing acidic residues),
Fig. 13.1 Ubiquitin-proteasome pathway. A protein selected to be degraded by the proteasome is first tagged with a polyubiquitin complex with the aid of E1, E2, and E3 enzymes in an ATPdependent manner. Next, the 19S cap of the proteasome recognizes the polyubiquitin-tagged protein and removes the ubiquitin molecules and unfolds the protein, allowing entry into the 20S catalytic core. Three enzymatic activities residing in the 20S core digest the peptide, breaking the protein into small amino-acid fragments
13 Therapeutic Utility of Proteasome Inhibitors for Acute Leukemia
275
and a trypsin-like (targeting the basic regions) activities. These activities are located in specific beta subunits, in the b5, b1, and b2 subunits, respectively, of the 20S proteasome [78]. These enzymatic activities are responsible for hydrolyzing the proteins that pass through the inner core of the proteasome. Proteins destined to be degraded by the proteasome include abnormal, damaged, unfolded, or mutated proteins, as well as normal, short-lived (i.e., transcription factors) and regulatory proteins (i.e., cell-cycle proteins). Since the majority of intracellular proteins are disposed of this way, many cellular processes are affected by the proteasome. Inhibiting the proteasome impacts a wide range of signaling pathways, including those involved in cell growth and survival, tumor suppression, and apoptosis. Despite initial concerns that inhibiting the proteasome could be nonselectively detrimental to all cells, studies show that several types of cancer cells are more sensitive to proteasome inhibitors compared to normal counterparts [63]. Early studies in a human leukemia cell line, HL60, demonstrated that proliferating cells but not quiescent cells underwent rapid apoptosis in response to proteasome inhibition [22]. Similarly, Masdehors et al. showed that CLL-derived lymphocytes were sensitive to proteasome inhibitor-induced apoptosis, whereas lymphocytes isolated from healthy individuals were not [63]. One explanation for these observations is that transformed cells proliferate more rapidly than normal cells, thus having a rapid rate of protein turnover. Therefore, cancer cells are more reliant upon proteasome activity to handle this relative protein overload, whereas normal cells are less so. This theory is partially supported by studies showing increased levels of proteasomal subunits, a greater amount of ubiquitinated proteins, and higher proteasomal proteolytic activity in human transformed cells from hematological malignancies and solid tumors [51] as compared to normal counterparts.
13.2.2 Development of Proteasome Inhibitors Chemical proteasome inhibitors were originally used to advance our understanding of the role of the proteasome in cellular processes. In the last decade, these compounds have evolved from biochemical tools into candidate molecules for cancer therapy [30]. Based on their chemical structure, proteasome inhibitors are roughly divided into five classes: peptide aldehydes, peptide boronates, peptide vinyl sulfones, peptide epoxyketones, and b-lactones [52]. For all the proteasome inhibitors, a requisite chemical feature is a pharmacore that interacts with a threonine residue on the active site of the b subunits of the proteasome. The first compound described as a proteasome inhibitor is lactacystin. This is a microbial metabolite isolated from Streptomyces that despite significant potency is not suitable for clinical use [2]. Peptide aldehydes were also identified to show activity against the proteasome early on. However, peptide aldehydes were not elaborated for clinic use due to their promiscuous targeting of other proteases, namely, cathepsins and calpains. Another drawback of the peptide aldehydes was that as reversible inhibitors, they rapidly dissociated from
276
J. Chandra and C.P. Miller
the proteasome and, furthermore, could be inactivated by oxidation [2]. The first clinically relevant proteasome inhibitors were formulated using the backbone of alpha-amino boronic acids, initially synthesized by Matteson and subsequently shown to be good chymotrypsin inhibitors by Lienhard [64]. In 1999, Adams et al. replaced the aldehyde group of peptide aldehydes with a boronic acid, synthesizing several selective, potent peptide boronate proteasome inhibitors [1]. This resulted in the development of the peptide boronate PS-341, now known as bortezomib [4]. Bortezomib (marketed by Millenium Pharmaceuticals under the name, Velcade) predominantly inhibits the chymotrypsin-like activity, which is the rate-limiting activity of the proteasome. Recently, bortezomib has also shown weak inhibition of the trypsin-like and caspase-like activities in isolated proteasomes [19]. It is unclear how inhibition of specific activities of the proteasome impacts the cytotoxicity of the proteasome inhibitors. However, interestingly, bortezomib was found to possess a broad and unique cytotoxic pattern in vitro toward a panel of sixty human tumor cell lines from the National Cancer Institute (NCI) [1]. It was also active against non-small-cell lung, colon, melanoma, ovarian, breast, renal and prostate cancer cell lines. In mouse xenografts, bortezomib exhibited antitumor properties, reducing tumor burden in models for prostate cancer [103], squamous cell carcinomas [100], and mantle cell lymphoma. These observations led to the testing of bortezomib in clinical trials, which showed single-agent activity against multiple myeloma and mantle cell lymphoma and resulted in its FDA approval for these hematological malignancies in 2003 and 2006, respectively. One key difference between the naturally derived proteasome inhibitors as compared to the man-made synthetic proteasome inhibitors is in whether they bind irreversibly or reversibly to the proteasome. Lactacystin, like other naturally occurring proteasome inhibitors, is an example of an irreversible compound. Although first investigated for effects on neurite growth, lactacystin, a novel metabolite isolated from Streptomyces [25, 77], was subsequently found to specifically inhibit proteasome enzymatic activities by binding irreversibly to the chymotrypsin-like and trypsin-like site and reversibly to the caspase-like site [26]. Interestingly, inhibition of proteasome activities by lactacystin correlated with lactacystin-induced effects on neurite growth [26] providing the first example of a biologically relevant functional effect of proteasome inhibition. Unfortunately, as was the case for the peptide aldehydes, lactacystin has also been shown to target other intracellular proteases, primarily cathepsin A and tripeptidyl peptidase II. This, coupled with toxicity issues in mice, has rendered it unsuitable for clinical use [3]. Epoxomicin is another naturally derived proteasome inhibitor belonging to the epoxyketone class of compounds. Epoxomicin was isolated from Actinomycetes and originally showed antitumor effects against B16 melanoma tumors in mice [35]. Work by Meng et al. later demonstrated that the target of epoxomicin was the proteasome and that it covalently binds to the proteolytic b subunits of the proteasome [66]. It primarily inhibits the chymotrypsin-like activity and to a weaker extent the trypsin-like and caspase-like activities [66].
13 Therapeutic Utility of Proteasome Inhibitors for Acute Leukemia
277
13.2.2.1 Next-Generation Proteasome Inhibitors The success of bortezomib has been a major validation of the idea that targeting the proteasome is of therapeutic value. Consequently, next-generation proteasome inhibitors have been identified and developed (Fig. 13.2). One of the main distinguishing features of these new inhibitors compared to bortezomib is the irreversible manner in which they bind to proteasomal catalytic b subunits. Marizomib (NPI0052; salinosporamide A) is a novel, orally active, nonpeptide small-molecule proteasome inhibitor developed by Nereus Pharmaceuticals, Inc. and was originally derived from a marine bacterium [60]. Structurally, it is similar to omuralide, which is the active form of lactacystin. Marizomib was recently reported to have in vitro and in vivo efficacy in multiple myeloma [12] and acute leukemias [67]. In vitro studies also showed marizomib to induce apoptosis more potently and more rapidly than bortezomib in lymphocytes derived from chronic lymphocytic leukemia patients [96]. Marizomib is currently in clinical trials for solid tumors, leukemias, lymphomas and multiple myeloma at multiple institutions, including the University of Texas M.D. Anderson Cancer Center, Memorial Sloan–Kettering Cancer Center, and the Multiple Myeloma Research Consortium (MMRC).
Fig. 13.2 Proteasome inhibitors target the catalytic activities of the proteasome. Shown is a crosssection of a b ring of the 20S catalytic core composed of seven b subunits. Specifically, the b1, b2, and b5 subunits contain the proteolytic activities responsible for breaking up proteins. They are the caspase, trypsin, and chymotrypsin-like activities, respectively. The FDA-approved proteasome inhibitor, bortezomib, and the next-generation proteasome inhibitors, marizomib (NPI-0052) and carfilzomib (identified by chemical structure), selectively target these enzymatic activities as indicated by arrows
278
J. Chandra and C.P. Miller
Promising attributes of marizomib that portend efficacy in cancers where b ortezomib was not effective as a single agent include its unique chemical structure [60], its irreversible binding to isloated proteasome [34], and its ability to block a broader range of proteasome activities. Unlike bortezomib, which was developed to inhibit the rate-limiting chymotrypsin-like activity of the proteasome, marizomib has been shown to target the chymotrypsin-like, caspase-like, and trypsin-like activities of isolated human erythrocyte 20S proteasome (Fig. 13.2). Recent studies have shown that bortezomib is also capable of inhibiting all three activities in isolated proteasomes; however, it appears that marizomib is able to inhibit the chymotrypsin-like and trypsin-like activities more effectively both in vitro and in vivo [104]. Elegant work from Alfred Goldberg’s group has shed light upon the ability of these proteasome inhibitors to have pleiotropic effects on all three proteasome enzymatic activities [46]. Inhibition of the chymotrypsin-like activity allosterically regulates the other two activities. Therefore, it follows that a more potent inhibition of the chymotrypsin-like activity by marizomib as compared to other proteasome inhibitors can influence the degree of inhibition of the caspase-like and trypsin-like activities. Studies using peptide substrates of the proteasome indicate that inhibition of multiple sites of the proteasome is necessary to block a significant fraction of cellular protein degradation [47]. Thus, marizomib’s unique and more potent profile of inhibition has implications for proteasome processing of substrates and can impact the manner in which the compound induces apoptosis in cancer cells. Studies in multiple myeloma and leukemia model systems provide support for this theory. For example, exposure to marizomib caused induction of apoptosis in CD 138+ tumor cells from multiple myeloma patients that were refractory to bortezomib treatment in vivo [12]. Also, the combination of marizomib with bortezomib resulted in synergistic cell death in multiple myeloma and lymphoma cell lines, reinforcing the idea that each proteasome inhibitor is promoting a different subset of proapoptotic substrates [13, 15, 95]. In both myeloma and leukemia lines, marizomib induced apoptosis relies upon caspase-8 more strongly than caspase-9 [12, 67]. Together, these data suggest that differential proteasome inhibition can result in activation of different apoptotic pathways. Carfilzomib (PR-171; Onyx/Proteolix, Inc.) is a newly described, irreversible proteasome inhibitor that was isolated from Actinomycetes [50]. Carfilzomib is structurally similar to epoxomicin and thus is classified as an epoxyketone. The b5 proteasomal subunit is the primary target of carfilzomib, and it inhibits the chymotrypsin-like activity in multiple myeloma cells [20]. This second-generation proteasome inhibitor has been shown to induce apoptosis in a variety of multiple myeloma cell lines and CD38+ plasma cells isolated from patients refractory to bortezomib or standard multiple myeloma therapy [50, 82]. In vitro, carfilzomib has also shown cytotoxicity in other hematological diseases. Proteasome inhibition with carfilzomib caused cell death in patient samples from non-Hodgkin lymphoma, chronic lymphocytic leukemia, and acute myeloid leukemia. Importantly, carfilzomib showed synergistic antiproliferative effects in a multiple myeloma cell line when combined with dexamethasone [50, 82], which is used as frontline multiple myeloma therapy and is also given as part of initial anti-ALL treatment. Currently,
13 Therapeutic Utility of Proteasome Inhibitors for Acute Leukemia
279
carfilzomib is in Phase-II clinical trials in patients with refractory or relapsed multiple myeloma and advanced solid tumors and Phase-I trials in lymphoma patients. It is also being clinically evaluated in combination with dexamethasone and lenalidomide in newly diagnosed, previously untreated multiple myeloma patients.
13.3 Molecular Pharmacology of Proteasome Inhibitors Despite structural and mechanistic differences between the proteasome inhibitors, a common biochemical feature these compounds share is their ability to induce cell death. The majority of reports focus on apoptosis as the mode of cell death caused by proteasome inhibitors; however, autophagy and necrosis have also been observed in specific contexts. Essentially, two models exist with regard to the mechanism by which proteasome inhibitors cause cell death – one relies on accumulation of specific proteasomal substrates, whereas the other model invokes a nonspecific accumulation of total cellular proteins (i.e., backing up of the garbage disposal).
13.3.1 Proapoptotic Proteins Inhibiting the proteasome results in stabilization of various intracellular proteins, and among these are proteins that mediate apoptosis. One such group of proteins that accumulate as a consequence of proteasome inhibition belong to the BCL-2 family, which encompasses both proapoptotic and antiapoptotic proteins [108]. Prosurvival Bcl-2 proteins (Bcl-2, Bcl-XL, Bcl-w, Mcl-1 and A1) heterodimerize with proapoptotic proteins (Bim, Bid, Puma, Noxa, Bad, Bax, or Bak), preventing them from initiating apoptosis via a mitochondrial pathway. Of the proapoptotic family members, a subset called the BH-3-only Bcl-2 family members (Bim, Bid, Puma, Noxa, and Bad) bind to antiapoptotic Bcl-2 proteins, sequestering them, and allowing the oligomerization and localization of Bax and Bak to the mitochondrial membrane. This results in permeabilization of the mitochondrial membrane and release of proapoptotic proteins from the mitochondria, resulting in caspase activation. Thus, the ratio of proapoptotic to antiapoptotic proteins is a major determinant as to whether the fate of the cell is to undergo apoptosis or continue surviving. Theoretically, for a proteasome inhibitor to induce apoptosis via this pathway, accumulation of proapoptotic family members would be expected. Although Bax and Bak expression do not seem to be dramatically altered as a result of proteasome inhibition, other proapoptotic proteins, such as Bim, Noxa and Puma, have been shown to be stabilized with proteasome inhibitors [21, 28, 32, 54]. Interestingly, in ALL, the anti-apoptotic protein Bcl-2 has been shown to be widely expressed, and its overexpression has been linked with the lack of responsiveness to induction of chemotherapy [9]. Therefore, accumulation of proapoptotic family members by proteasome inhibitors offers a unique mode of targeting Bcl-2 expression. In fact,
280
J. Chandra and C.P. Miller
combining proteasome inhibition with therapeutic agents that specifically target Bcl-2 such as ABT-737, which relies upon Noxa, causes synergistic apoptosis in melanoma cells [69]. Another BH-3 protein, Puma, is a proteasome substrate, and accumulation of this protein is observed in cells treated with proteasome inhibitors. Interestingly, Puma overexpression experiments in which indirect activation of Bax and Bak are observed provides further support for this idea [21]. However, antiapoptotic proteins such as Mcl-1 have also been shown to accumulate as a consequence of proteasome inhibition. But further studies in multiple myeloma cells treated with bortezomib reveal that this increased Mcl-1 protein is also cleaved [32], and the cleavage fragment has a proapoptotic rather than an antiapoptotic role.
13.3.2 Activation of Caspases Caspase activation is considered a central/obligate event during apoptosis. Caspases are a 14-member family of cysteine proteases that cleave substrates at aspartate residues, act in cascades, and promote apoptosis [107]. Induction of cell death by proteasome inhibitors is accompanied by caspase activation in most cases; however, caspase-independent cell death also occurs. Generally speaking, caspase activation can be initiated by signals extrinsic or intrinsic to the cell. The most commonly studied extrinsic pathways are those that involve ligation of a death receptor such as Fas or members of the TNF-R superfamily. The intrinsic pathway refers to the initiation of a caspase cascade beginning with mitochondrial perturbations. However, increasing attention is being directed toward endoplasmic reticular (ER) stress, which is linked to the unfolded protein response (UPR) as another intrinsic pathway of caspase activation [37, 101]. Emerging reports highlight ER stress as the most likely explanation for how proteasome inhibition can be directly linked to caspase activation [23].
13.3.3 Endoplasmic Reticulum Stress and Caspase Activation The accumulation of unfolded proteins in the cell’s ER causes stress, launching a cellular defense called the UPR. The UPR is a conserved physiological response characterized by signal transduction events initiated by three ER-localized transmembrane proteins: IRE1, PERK, and ATF6 [61]. Prolonged ER stress is linked to caspase activation within ER. Two caspases, initially considered to be important in inflammatory responses, have been implicated in ER stress – caspase-12 and caspase-4 [45]. Caspase-12, although the first ER-resident caspase described, was identified in the murine system and is not universally expressed in human tissues. Also, inhibition of caspase-4, either by chemical inhibitors or by genetic means, does not consistently block cell death by proteasome inhibitors. Owing to the incon-
13 Therapeutic Utility of Proteasome Inhibitors for Acute Leukemia
281
sistencies in detecting caspase-12 or -4 activation in proteasome-inhibitor-treated cells, the more commonly studied initiator caspases remain also the most extensively studied in the context of proteasome inhibition [14]. Caspase-8 is one of the initiator caspases associated with the death receptor pathway, namely, Fas/FasL engagement. Caspase-9, another well-studied initiator caspase, is engaged by intrinsic death programs originating from mitochondrial disturbances. It is noteworthy that ER stress, via caspase-12 or caspase-4, can cause caspase-9 activation. Activation of downstream effector caspases such as caspase-3 occurs as a consequence of initiator caspase activation. These effector caspases are responsible for cleavage of the key substrates that are associated with the morphological features of apoptosis. Bortezomib induces apoptosis through activation of caspase-8, -9, -3, and the ER-resident caspase-4 [14, 24]. In considering the relative contributions of these caspases, Mitsiades et al. have shown that bortezomib, which preferentially blocks the chymotrypsin-like proteasomal activity, relies equally on activating caspase-8 or -9 to induce apoptosis [72]. On the contrary, recent in vitro studies in multiple myeloma and acute leukemia determined that the second-generation proteasome inhibitor marizomib is more dependent on caspase-8 activation to induce apoptosis, since caspase-8 inhibitors were far more effective than caspase-9 inhibitors at blocking cell death [12, 67]. In Jurkat cells (a T-cell ALL cell line) lacking expression of caspase-8, proteasome inhibition of all three proteolytic activities by marizomib was detected, indicating that caspase-8 activation is downstream of proteasome inhibition and that this unique profile of proteasome inhibition could be linked to the reliance on caspase-8 [67]. Interestingly, carfilzomib, which like bortezomib specifically inhibits the chymotrypsin-like activity, was shown to activate both intrinsic and extrinsic initiator caspases in multiple myeloma cells [50]. Together, caspase activation data obtained with bortezomib, marizomib, and carfilzomib support the notion that inhibiting multiple proteasome activities may be more selective at activating specific caspases, whereas blocking the chymotrypsinlike activity results in activation of multiple caspases. However, it is important to point out that although marizomib targets all three activities of the proteasome, in whole-leukemia cells it demonstrates differential inhibition of proteasome activities (inhibiting the chymotrypsin- and caspase-like activities more potently than the trypsin-like activity) [67]. Thus, further studies need to be conducted to conclusively link caspase activation with the degree and duration of proteasome inhibition at single or multiple proteolytic sites.
13.3.4 Oxidative Stress Proteasome inhibition has been linked to the generation of reactive oxygen species (ROS), and studies indicate that ROS contributes to the cell death observed with these agents [10]. This is likely due to metabolic differences and hyperactivated ROS-generating enzymes.
282
J. Chandra and C.P. Miller
Transformed cells are thought to possess higher levels of oxidative stress than nontransformed cells. Therefore, the generation of ROS by proteasome inhibitors represents a mechanism for selectively killing cancer cells, since a higher ROS threshold would predispose these cells to oxidative stress-dependent cell death. Evidence for this possibility has been shown in studies in various cancer systems that demonstrated increased intracellular peroxide and superoxide levels occurring concomitantly with proteasome inhibitor-induced apoptosis [27, 55, 67, 86]. Exposure to diverse antioxidants including glutathione-bolstering agents and vitamin E analogs prior to proteasome inhibitors prevent cancer cells from undergoing cell death [27, 55, 67]. While proteasome inhibitors are widely accepted to raise ROS levels, the exact source of the oxidative stress is still unclear. A role for mitochondrial sources of ROS was suggested by Ling et al. in a human lung cancer cell line treated with bortezomib [55]. Coadministration of bortezomib and rotenone, antimycin-A (inhibitors of mitochondrial electron-transport chain) or cyclosporine A (inhibitor of mitochondrial permeability transition pore) [55], blunted ROS levels, suggesting that mitochondria is playing a role in bortezomib-induced oxidative stress. In a comparison analysis between marizomib and bortezomib, we have shown that marizomib raises ROS levels to a higher degree than bortezomib in a human T-cell ALL cell line [68]. Given the unique profile of inhibition of the proteasome by marizomib as compared to bortezomib, this suggests that there might exist a correlation between degree of proteasome inhibition and ROS generation. Further studies are being conducted to address this possibility.
13.3.5 Targeting NFkB: Inhibiting Survival The success of proteasome inhibitors in multiple myeloma has largely been attributed to the ability of these compounds to interfere with the activation of the transcription factor, nuclear factor-kappa B (NFkB). Activation of NFkB via regulation of its endogenous inhibitor IkB has been demonstrated to play important roles in proliferation, inflammation, survival, angiogenesis, metastasis, and, somewhat counterintuitively, even cell death [42, 43]. These variable downstream effects of NFkB activation are dependent upon the subset of genes being driven by this transcription factor. Furthermore, activation of NFkB is proteasome dependent in at least two distinct steps [42]. One of the most commonly described proteasomedependent pathways for NFkB regulation relies on the inhibitory protein IkB (inhibitor of NFkB), which binds to and sequesters components of the multimeric NFkB transcription factor in the cytosol, thereby preventing translocation to nuclei and access to DNA. In order for IkB to release NFkB, allowing its translocation to the nucleus and transcription of its target genes, the inhibitory protein must be degraded by the proteasome. This proteasomal degradation is preceded by phosphorylation of IkB, which is implemented by a cascade of specific kinases, such as NIK (NFkB inducing kinase) and IKK (IkB kinase) family members. A second way in which NFkB is regulated by the proteasome involves the Rel family of proteins.
13 Therapeutic Utility of Proteasome Inhibitors for Acute Leukemia
283
RelA and RelB are heterodimeric components of the active NFkB transcription complex. Prior to binding with the appropriate molecular partner to form a dimer, proteasomal cleavage of the Rel protein must occur. In multiple myeloma, endogenous constitutive activation of NFkB has been described and is attributed to mutations in genes encoding regulators of the NFkB signaling pathway [38]. Gene expression profiling in multiple myeloma cell lines and in patient samples from independent laboratories have demonstrated mutations in 20% of the samples analyzed. Both gain-of-function mutations in positive regulators of NFkB signaling as well as loss-of-function mutations in negative regulators of NFkB signaling have been identified [29]. Bortezomib has been shown to dampen NFkB signaling, although there is no doubt that other signaling pathways are also affected. Furthermore, microarray profiling of multiple myeloma cells indicates a gene signature consistent with heightened sensitivity to bortezomib. Numerous structurally distinct proteasome inhibitors have been found to impact NFkB activity, providing yet another link between proteasome inhibition and diminished NFkB activation. Chauhan et al. directly compared the effects of bortezomib and marizomib on NFkB activation in HEK293 cells expressing a luciferase NFkB reporter gene [12]. Results showed that marizomib was a more potent inhibitor of this transcription factor compared to bortezomib [12], consistent with data indicating that marizomib is a more robust proteasome inhibitor than bortezomib. In ALL cells, constitutive activation of NFkB has been described, pointing toward potential efficacy of proteasome inhibitors [48]. A study examining DNAbinding activity by electrophoretic mobility shift assay (EMSA) in specimens from over 40 childhood ALL patients demonstrated that the majority had either classical NFkB activation (the active transcription factor composed of p65 and p50 heterodimers) or p50 homodimers. Furthermore, proteasome inhibition caused accumulation of phosphorylated IkB [48]. Consistent with these findings, bortezomib and other proteasome inhibitors potently induce apoptosis in ALL cells within the dose range seen to exert cytotoxic effects in multiple myeloma cells [67]. Since NFkB is usually activated as a survival response to cellular stress induced by cytotoxic agents, radiation, or DNA damage, it is notable that bortezomib sensitized tumor cells to chemotherapy or radiotherapy (activators of NFkB) [97]. In ALL cells, similar chemosensitizing effects of bortezomib have been observed when combined with other classes of agents such as alkylating agents, nucleoside analogs, epigenetically targeted agents, and kinase inhibitors [58, 106, 109]. Despite all the suggestive links between NFkB inhibition and proteasome inhibition, several studies, including some conducted in multiple myeloma (a disease where clinical efficacy of the proteasome inhibitors is well established), indicate that NFkB inhibition is not the only determinant of apoptosis induction. Utilizing a specific inhibitor of IkB, PS-1145, Hideshima et al. showed that both bortezomib and PS-1145 blocked NFkB activation [38]. However, only modest growth inhibition (20–50%) of proliferating multiple myeloma cells was observed with PS-1145, whereas bortezomib achieved 80–90% growth inhibition in these cells [38]. These results suggest that a NFkB-centric approach is not as effective as proteasome inhibition in promoting cytotoxicity.
284
J. Chandra and C.P. Miller
13.3.6 Resistance to Proteasome Inhibitors Even in multiple myeloma where major antitumor activity was observed in advanced and refractory cases, resistance to bortezomib is a challenge. Several mechanisms of resistance have been described by several groups and can be grouped into two general categories: direct alterations in the proteasome, such as b subunit amplification, or downstream effects on apoptosis pathways. For example, expression of antiapoptotic proteins such as Bcl-2, IAP proteins, and c-FLIP have been implicated in resistance to bortezomib [65]. Interestingly, increased expression of antioxidant proteins and heat-shock proteins has also been noted in lines resistant to bortezomib. Hsp27, which is a heat-shock protein possessing antioxidant properties, was among the first described molecules conferring resistance to bortezomib [11]. Employing an unbiased approach to identifying determinants of resistance, Anderson and coworkers characterized the gene expression profiles of sensitive and resistant cell lines to bortezomib [98]. From that work, they reported overexpression of heatshock protein, Hsp-27, in bortezomib-resistant lines. Antisense targeting of Hsp-27 protein restored sensitivity of these cells to proteasome inhibition. Conversely, overexpressing Hsp-27 in the sensitive lymphoma cell conferred resistance to bortezomib. Several investigators have also reported upregulation of other heat-shock proteins, Hsp-72 and -90, as a consequence of proteasome inhibition [71]. Another important regulator of sensitivity to proteasome inhibitors is the proteasome itself. Both Kraus et al. and Ostrowska et al. have reported proteasome activity profiles in samples from patients with acute leukemias [49, 83]. Chymotrypsin-like activity was elevated in plasma of newly diagnosed ALL and AML patients compared to healthy individuals. Importantly, Kraus et al. showed that the expression ratio of the b2 subunit compared to both the b1 and b5 subunits influences sensitivity to bortezomib [49]. More recently, studies by Lu et al. have implicated overexpression of the proteasomal b5 subunit containing the activity preferentially blocked by bortezomib, as a contributor toward bortezomib resistance in cells derived from a lymphoblastic leukemia cell line (JurkatB) [56]. Resistant lines were generated by repeated exposure to bortezomib and subsequent selection of established Jurkatb (Jurkat B)-resistant cells. Characterization of these cells revealed increased expression of the b5 subunit, as well as a point mutation [57]. Furthermore, introduction of this mutant b5 subunit into parental Jurkat cells resulted in resistance to apoptosis and inhibition of chymotrypsin-like activity with bortezomib.
13.4 Proteasome Inhibitors in the Clinic 13.4.1 Single-Agent Activity The discovery of bortezomib’s activity against various tumor cell lines and its reduction of tumor burden in mouse models led to its assessment in the clinic [41]. Phase-I clinical trials evaluated the effects of bortezomib on patients with refractory
13 Therapeutic Utility of Proteasome Inhibitors for Acute Leukemia
285
or relapsed therapies in solid tumors and hematological malignancies. As a single agent, in a study of 43 patients, bortezomib produced a partial response in one patient with non-small-cell lung carcinoma, and stable disease was observed in three other patients presenting with malignant melanoma, nasopharyngeal carcinoma, and renal cell carcinoma [5]. Doses administered ranged from 0.13 to 1.56 mg/m2 on days 1, 4, 8, and 11 in a 3-week cycle. At the highest dose (1.56 mg/m2), dose-limiting toxicities observed included diarrhea and sensory neuropathy. Analysis of wholecell lysates revealed a 68% inhibition reached after 1 h of administration with bortezomib (1.56 mg/m2). A second study was conducted involving 53 patients with refractory solid tumors, the majority presenting with androgen-independent prostate cancer [85]. These patients were treated with increasing doses of bortezomib (0.13–2.00 mg/m2) once weekly for 4 weeks every 5 weeks. Proteasome inhibition was monitored after 1 h of bortezomib administration, and more than 75% inhibition was detected at the 2.0 mg/m2 dose. Overall, proteasome inhibition was dosedependent and partial recovery of activity was observed by the next dose. Dosedependent gastrointestinal complications and hypotension were commonly observed but were manageable. Partial responses were seen in two prostate cancer patients; however, they displayed enlarged lymph nodes (lymphadenopathy). Two other patients enrolled in the study demonstrated a decrease in prostate-specific antigen (PSA) levels by more than 50%. Based on these solid tumor trials, the recommended continuation dose for bortezomib for Phase-II trials was 1.6 mg/m2. In advanced hematological diseases, bortezomib (escalating doses from 0.40 to 1.38 mg/m2) was used in a Phase-I study with 27 patients [79]. Of these, nine patients with multiple myeloma had a response with one patient achieving complete response, while two patients with mantle cell lymphoma and follicular lymphoma had partial responses. These effects were seen on a dosing schedule of bortezomib twice weekly for 4 weeks with a 2-week recovery. Proteasome activity was reduced in a dose-dependent manner, achieving 74% inhibition with 1.38 mg/m2 bortezomib. This reduction was detected within one hour of dosing and returned to normal levels after 72 h. Significant but manageable toxicities included thrombocytopenia, hyponatremia, fatigue, malaise, and peripheral neuropathy. These results led to the pivotal study of uncontrolled myeloma management with proteasome inhibition therapy (SUMMIT) Phase-II trials involving 202 patients with recurrent or refractory multiple myeloma who had failed at least two prior therapies [92]. Bortezomib was administered on days 1, 4, 8, and 11 followed by a 10-day recovery for 3 weeks for up to eight cycles at 1.30 mg/m2. Toxicities were observed in 25% of patients and included fatigue, nausea, diarrhea, thrombocytopenia, and sensory neuropathy. However, this treatment resulted in a 10% complete response or nearcomplete response, and stable disease was observed in 59% of the patients. Overall response rate, including complete, near-complete, and partial response, was 27% [91]. The median duration of response was 12.7 months, and median overall survival was 16 months. Furthermore, unlike previous therapies, which had only achieved 3 months for median time to progression, patients responding to bortezomib had more than doubled their median time to progression to 6.6 months. These promising results led to the approval of bortezomib in May 2003 by the FDA for the treatment of relapsed and refractory multiple myeloma. On March 2005, it
286
J. Chandra and C.P. Miller
Fig. 13.3 Time line. Shown are important events in the history of the development of proteasome inhibitors
was approved for multiple myeloma patients that had received at least one prior therapy. These events as well as other key events in the clinical development of proteasome inhibitors are depicted in the timeline presented in Fig. 13.3. Another important Phase-II study was CREST (Clinical Response and Efficacy Study of bortezomib in the Treatment of refractory myeloma), in which two doses of bortezomib (1.0 and 1.3 mg/m2) were evaluated in 54 patients with recurrent multiple myeloma using the same schedule employed for the SUMMIT trial [90]. Responses were reported for both dose groups. Patients who received 1.0 mg/m2 had a 33% response rate, while the other group had a 50% response rate. Dexamethasone (20 mg) was additionally administered to patients with suboptimal response after two cycles with bortezomib; in these patients, the overall response rate achieved with the combination of the two compounds was 44% for 1.0 mg/m2 and 62% at the higher 1.3 mg/m2 dose. Monotherapy with bortezomib or in combination with dexamethasone achieved a median duration of response of 9.5 and 13.7 months for the 1.0 and 1.3 mg/m2 groups, respectively. Most common secondary effects included but were not limited to fatigue, nausea, diarrhea, constipation, peripheral neuropathy, limb pain, and thrombocytopenia. Patients receiving the higher dose experienced more adverse effects compared to patients exposed to lower doses of bortezomib, suggesting a dose effect. Twelve patients halted treatment due to adverse events, and five patients had to discontinue because of neuropathy. However, it is important to consider that some of these patients already had preexisting neuropathy as a consequence of previous therapy. The Phase-III trial (APEX, Assessment of Proteasome Inhibition for EXtending Remission Trial) was an international study that involved refractory multiple myeloma patients (up to three prior therapies) [93]. This study compared the effects
13 Therapeutic Utility of Proteasome Inhibitors for Acute Leukemia
287
of bortezomib against dexamethasone. Patients were randomized to be administered either dexamethasone or bortezomib, with the option of switching to bortezomib if patients progressed with dexamethasone. The trial was terminated because patients receiving bortezomib did significantly better. Overall response was 38% with bortezomib and 18% with dexamethasone. In an attempt to extend the success of bortezomib in multiple myeloma to other heme malignancies, studies were conducted in non-Hodgkin lymphoma (NHL) including mantle cell lymphoma [33, 73] based on the partial responses observed in the earlier Phase I trial. Bortezomib was given two times a week for 2 weeks on a 3-week cycle at a concentration of 1.5 mg/m2. Of 24 patients, 58% had a partial or better response [76]. In a study involving patients with relapsed or refractory B-cell NHL, 12 patients out of 29 responded to bortezomib therapy. Six of these patients were complete responders and another 6 demonstrated partial response. Similar toxicities were seen as with the multiple myeloma trials – thrombocytopenia, fatigue, and neuropathy. Results from these clinical trials demonstrated that patients with relapsed or refractory multiple myeloma and mantle cell lymphoma benefited from treatment with bortezomib. These successful outcomes led to the approval of the use of bortezomib for these diseases in 2003 and 2006. Most recently, bortezomib has been approved to be used in previously untreated multiple myeloma patients. In leukemia trials, bortezomib as a single agent was evaluated by Cortes and colleagues. A Phase-I dose-escalation trial in 15 adult patients with recurrent or refractory acute leukemias was conducted using 0.75, 1.25, or 1.5 mg/m2 [17]. The maximum tolerated dose was 1.25 mg/m2 given two times a week for 4 weeks followed by a 2-week rest. Dose-limiting toxicities seen were gastrointestinal toxicities, hypotension, and fluid retention. Although biological activity (significant proteasome inhibition) was observed, not even a partial response was detected. Four patients displayed decreased blast counts but relapsed during the rest period. Based on these studies, single-agent bortezomib was not considered suitable for this population, and instead, evaluation of bortezomib in combination with other agents was recommended for acute leukemias. The Children’s Oncology Group enrolled 12 patients with refractory acute leukemia in a Phase-I trial with bortezomib to examine toxicities and tolerable doses [6]. Nine of the 12 patients in the study had refractory ALL. The proteasome inhibitor was administered twice weekly for 2 weeks every 21 days, starting with a dose of 1.3 mg/m2 (which was the tolerable dose established by Cortes et al.). Owing to complication from prior therapies or disease progression, seven patients did not complete the first course of bortezomib. Two pre-B ALL patients receiving 1.7 mg/ m2 presented dose-limiting toxicities, which included grade-3 confusion and a cranial nerve XII palsy, grade-4 hypotension and febrile neutropenia, and grade-3 creatinine elevation. Overall results did not show a definitive antitumor response with bortezomib, although isolated cases hinted at potential activity. For example, a patient with pre-B ALL exposed to 1.3 mg/m2 bortezomib had stable disease as demonstrated by decrease of bone marrow lymphoblasts after one cycle but showed progression of disease by the next cycle. Another patient also experienced stable
288
J. Chandra and C.P. Miller
disease for two cycles, showing a decrease in white blood cell counts; however, bone marrow lymphoblasts remained unchanged. An adolescent with relapsed and refractory pre-B ALL had a complete response but only after following bortezomib treatment with dexamethasone, adriamycin, and vincristine. Four additional patients also received a steroid regimen after bortezomib, showing minimal response or transient decrease in blood cell counts but not in bone marrow lymphoblasts. Previous to this, only one complete response using bortezomib has been reported in relapsed childhood ALL in a combination study with bortezomib and dexamethasone [8]. These results in refractory pre-B ALL suggest that bortezomib may be best used to sensitize this population to standard chemotherapy. Trials with the next-generation proteasome inhibitors are currently ongoing. Marizomib is being tested in Phase-I clinical trials at several institutions across USA in patients with solid tumors, lymphomas, leukemias and multiple myeloma. Initial results from these Phase-I trials demonstrate that at lower doses than bortezomib, marizomib potently blocks proteasome activity in peripheral white blood cells for a longer period of time than bortezomib. Most important, marizomib did not induce the dose limiting toxicities often associated with bortezomib. A Phase-IB trial has recently been started with marizomib in combination with vorinostat. Carfilzomib is also being evaluated in Phase-I trials in patients with lymphomas and in Phase-II trials in refractory multiple myeloma and solid tumors. Results from these trials are eagerly awaited.
13.4.2 Combination Studies Clinical studies have been designed and executed in heme malignancies to examine the ability of bortezomib to augment the function of conventional therapeutics such as dexamethasone, doxorubicin, and thalidomide. The doxorubicin studies utilized bortezomib in combination with pegylated liposomal doxorubicin in 22 patients [7]. Scheduling included bortezomib on days 1, 4, 8, and 11 and liposomal doxorubicin on day 4, every 21 days. No dose-limiting toxicities were noted; however, side effects included thrombocytopenia, neutropenia, fatigue, neuropathy, and gastrointestinal toxicities. Forty-four per cent of patients had complete response or partial response [81], and partial responses were achieved in AML and NHL patients [80]. In the last decade, the clinical development of bortezomib has been a major success for multiple myeloma patients. Another clinical success for the same disease has been the approval of the immunomodulatory agent thalidomide [74]. Given their distinct mechanisms of action, a logical question was to address the clinical efficacy of the two drugs together [99]. Bortezomib and thalidomide were evaluated in 79 patients with advanced or refractory multiple myeloma. Patients were administered bortezomib on days 1, 4, 8, and 11 and thalidomide on days 1–21 for 21 days. Grade-3 peripheral neuropathy has been reported but was not considered dose-limiting. Preliminary reporting of this trial indicates that 17% of patients achieved complete or near CR and 52% of patients achieved CR or PR.
13 Therapeutic Utility of Proteasome Inhibitors for Acute Leukemia
289
13.5 Potential Combination Studies for the Treatment of ALL As discussed above, bortezomib was clinically effective against specific hematological malignancies, namely, multiple myeloma and mantle cell lymphoma. In cancers, such as ALL, in vitro data and/or animal studies suggested potential efficacy, but clinical responses have not been demonstrated. The question is, therefore, how to best utilize the proteasome inhibitors in ALL therapeutics. Combination therapies have the potential to minimize toxicities seen with bortezomib and the newer proteasome inhibitors. Synergy between new classes of compound and proteasome inhibitors may achieve this goal as well as provide promise in diseases where monotherapy with bortezomib is limited, such as ALL. Some of these synergistic combinations, two of which are highlighted below, utilize very low doses of the respective compounds; therefore, therapy-related long term toxicites (a major concern in children and adolescents) could conceivably be minimized. Table 13.1 describes the agents being used in these trials and the sites where they are being conducted as of February 2009.
13.5.1 Proteasome Inhibitors and HDAC Inhibitors Histone deacetylase inhibitors (HDACi) target histone deacetylases (HDAC), enzymes that regulate the acetylation status of histones in conjunction with histone acetylases (HATs) [40]. HDACs remove the acetyl group that is added by HATs on lysine residues of histones. Addition and removal of acetyl groups from histones regulates chromatin conformation. Alterations in chromatin conformation enable specific transcription factor complexes to bind DNA, thus modulating gene transcription and expression. In general, HDACi promote hyperacetylation of histones, leading to an open chromatin conformation favoring gene transcription. However, if the open chromatin conformation attracts negative regulators of gene transcription, the opposite effect may take place. Furthermore, acetylation events can occur on any protein, not just histones, and so the putative extra-histone acetylation effects of HDACs and HATs broadens the impact of treating whole cells with HDACi. In vitro studies evaluating the effects of proteasome inhibitors in combination with HDACi have shown promise in several cancer models, including hematological cancers. To date, vorinostat is the sole HDACi approved by the FDA and represents the newest therapy option for patients with cutaneous T-cell lymphoma [44]. Dozens of structurally distinct HDACi are in various stages of clinical and preclinical development and have been grouped based on which HDAC family members they act upon. Several mechanisms of synergy have been reported between proteasome inhibitors, primarily bortezomib and MG-132, and these various HDACi. Most consistently reported are increased levels of oxidative stress, downregulation of antiapoptotic proteins (Mcl-1, XIAP) and upregulation of Bim, activation of c-Jun NH terminal kinase, p53 induction, and endoplasmic reticulum stress [10, 62].
Pan-HDAC inhibitor
Bcl-2, Bcl-XL Mcl-1
Solid tumors
Recruiting patients Recruiting patients
Recruiting patients Start date – 1/2009
Recruiting patients Recruiting patients
Recruiting patients
Phase I Phase II
Phase II Phase II
Phase III Phase I/II
Phase I/II
Recruiting patients
Recruiting patients
Phase I
Phase I
Status
Stage
Matthew Spear, MD; Nereus Pharmaceuticals
Merck, Multicenter Study Joseph Toscano, MD; University of CaliforniaDavis A. Keith Stewart, MD; Mayo Clinic
Barbara Pro, MD; MDACC
Merck, Multicenter Study
Charles Erlichman, MD; Mayo Clinic George Wilding, MD; University of Wisconsin Lubomir Soko, MD, PhD; Moffitt Cancer Center
PI/Institution
Combination studies involving bortezomib and marizomib marizomib that are currently under clinical evaluation Hsp heat-shock protein; HDAC histone deacetylase; PI principal investigator; MDACC M.D. Anderson Cancer Center. Source: www.clinicaltrials.gov
Marizomib Vorinostat
Obastoclax Mesylate (GX015-070)
Mantle-cell lymphoma relapsed/refractory diffuse large B-cell lymphoma Relapsed/refractory multiple myeloma Relapsed/refractory T-cell non-Hodgkin’s lymphoma Multiple myeloma Relapsed/refractory non-Hodgkin’s lymphoma Relapsed/refractory Multiple myeloma
Table 13.1 Combination trials with proteasome inhibitors Agents Targets Tumor types Bortezomib 17-AAG Hsp-90 Advanced solid tumors lymphoma Vorinostat Pan-HDAC inhibitor Solid tumors
290 J. Chandra and C.P. Miller
13 Therapeutic Utility of Proteasome Inhibitors for Acute Leukemia
291
Our lab has examined the effects of marizomib with HDACi as compared to bortezomib and HDACi and found that the former combination is more potent than the latter in leukemic lymphocytes [67, 68]. The marizomib and HDACi combination, much like marizomib alone, relies upon ROS and caspase-8 to induce apoptosis. Additionally, we also observed overlapping functions between marizomib (NPI-0052) and two structurally distinct HDACi: entinostat (MS/SNDX-275) and vorinostat (Zolinza). Despite the fact that entinostat is an inhibitor of HDAC family members 1–3, whereas vorinostat is a pan-HDAC inhibitor, we saw similar results with the two compounds. Both HDACi were able to target the proteasome by reducing mRNA levels of b catalytic subunits of the proteasome. This reduction had a functional effect, since we also detected inhibition of proteasomal proteolytic activities with HDACi in leukemia cells. Our findings were consistent with a previous report conducted in multiple myeloma where gene expression profiling in vorinostat-treated cells showed downregulation of the b5 subunit and the chymotrypsin-like activity [70]. Surprisingly, marizomib alone also affected the primary target of the HDACi by altering acetylation status and levels of total histone-H3, whereas bortezomib did not [68]. This hyperacetylation was stronger in cells treated with a combination regimen of marizomib/HDACi and was regulated by oxidative stress and caspase-8 activation, since antioxidants or lack of caspase-8 reversed the acetylation effect. These observations were extended to primary lymphocytes isolated from leukemia patients, adding to the potential mechanisms of synergy observed with HDACi and proteasome inhibitors. With Phase-1B clinical trials in various malignancies currently ongoing with marizomib in combination with vorinostat, this combination is ripe for testing in an acute leukemia setting.
13.5.2 Proteasome Inhibitors and BCL-2 Antagonists Numerous agonists of Bcl-2 have been developed spanning the spectrum from antisense strategies to small-molecule inhibitors to natural products [89]. ABT-737 is a relatively new synthetic small-molecule inhibitor produced by Abbott Laboratories using NMR-guided, structure-based drug design, representing the most deliberate attempt at designing a Bcl-2 inhibitor [76]. ABT-737 binds to antiapoptotic proteins, Bcl-2, Bcl-XL, and Bcl-w, but not Mcl-1, through their BH-3 domain, preventing binding to proapoptotic proteins. The sequestering of antiapoptotic proteins by ABT-737 allows Bax and Bak to oligomerize and induced apoptosis via a mitochondrial pathway. ABT-737 acts as a BH-3 mimetic and has shown antitumor activity against small-cell lung carcinoma, lymphomas, CLL, AML, and ALL [53]. Overexpression of Mcl-1 has been identified as a mechanism of resistance, since ABT-737 does not target this antiapoptotic protein [16]. However, proteasome inhibitors have been shown to stabilize the proapoptotic BH-3-only protein Noxa [88]. Noxa is able to bind and neutralize Mcl-1 protein, thus providing a rationale to combine these two agents [75]. Furthermore, data from our lab
292
J. Chandra and C.P. Miller
indicate that ABT-737 in a human ALL cell line increases oxidative stress by diminishing glutathione levels [39], giving additional support to combination studies with proteasome inhibitors. In preclinical studies, ABT-737 has been found to enhance bortezomib’s cytotoxicity in cell lines and primary cells from disease types such as multiple myeloma, B- and T-cell lymphomas, and CLL [84]. In vivo efficacy between ABT-737 and bortezomib was explored in a mantle cell lymphoma xenograft mouse model, which showed favorable responses to the combination compared to either single agent (two complete responses out of six mice) [84]. This study also identified that administration schedules affected outcome, since different scheduling of the same doses resulted in either no benefits or enhanced toxic side effects. Clearly, pharmacokinetic and pharmacodynamic interactions need to be taken into account in designing these combination studies. The newer proteasome inhibitors also appear to display higher, at least, additive effects when combined with ABT-737 in vitro. Using combination indices, we have observed synergistic effects with ABT-737 and marizomib (unpublished observations) in ALL cell lines. A preliminary report from Paoluzzi et al. has reported synergistic apoptosis induction with ABT-737 and carfilzomib. Overall, these promising preclinical studies with ABT-737 and various proteasome inhibitors warrant further combination studies in vivo and in the clinic.
13.6 Summary The proteasome inhibitors have dramatically impacted the treatment of some, but not all, hematological malignancies. While multiple myeloma patients and mantle cell lymphoma patients have reaped the benefits of this class of compounds, leukemia patients – adults and children alike – have not benefited as significantly. As clinical trials with newer proteasome targeting agents and combination trials using these drugs move forward, it will be important to include leukemia patients of all ages in these trials. Only through those studies will we progress in our goals of offering effective treatment options in otherwise incurable disease and raise the cure rate for children and young adults diagnosed with ALL while at the same time cause as few therapy-related long-term toxicities as possible so that survivors might enjoy an optimal quality of life.
References 1. Adams J, Palombella VJ, Sausville EA, Johnson J, Destree A, Lazarus DD, Maas J, Pien CS, Prakash S, Elliott PJ. Proteasome inhibitors: a novel class of potent and effective antitumor agents. Cancer Res. 1999; 59: 2615–2622 2. Adams J. Potential for proteasome inhibition in the treatment of cancer. Drug Discov Today. 2003; 8: 307–315
13 Therapeutic Utility of Proteasome Inhibitors for Acute Leukemia
293
3. Adams J. The development of proteasome inhibitors as anticancer drugs. Cancer Cell. 2004; 5: 417–421 4. Adams J, Kauffman M. Development of the proteasome inhibitor Velcade (Bortezomib). Cancer Invest. 2004; 22: 304–311 5. Aghajanian C, Soignet S, Dizon DS, Pien CS, Adams J, Elliott PJ, Sabbatini P, Miller V, Hensley ML, Pezzulli S, Canales C, Daud A, Spriggs DR. A phase I trial of the novel proteasome inhibitor PS341 in advanced solid tumor malignancies. Clin Cancer Res. 2002; 8: 2505–2511 6. Bachmann AS. Proteasome inhibitors in pediatric cancer treatment. Hawaii Med J. 2008; 67: 247–249 7. Blade J, Sonneveld P, San Miguel JF, Sutherland HJ, Hajek R, Nagler A, Spencer A, Robak T, Cibeira MT, Zhuang SH, Harousseau JL, Orlowski RZ. Pegylated liposomal doxorubicin plus bortezomib in relapsed or refractory multiple myeloma: efficacy and safety in patients with renal function impairment. Clin Lymphoma Myeloma. 2008; 8: 352–355 8. Brown ER, Bostrom B, Zhang PL. Morphoproteomics and bortezomib/dexamethasoneinduced response in relapsed acute lymphoblastic leukemia. Ann Clin Lab Sci. 2004; 34: 203–205 9. Campana D, Coustan-Smith E, Manabe A, Buschle M, Raimondi SC, Behm FG, Ashmun R, Arico M, Biondi A, Pui CH. Prolonged survival of B-lineage acute lymphoblastic leukemia cells is accompanied by overexpression of bcl-2 protein. Blood. 1993; 81: 1025–1031 10. Chandra J. Oxidative stress by targeted agents promotes cytotoxicity in hematological malignancies. Antioxid Redox Signal. 2008; 11(5): 1123–1137. 11. Chauhan D, Li G, Shringarpure R, Podar K, Ohtake Y, Hideshima T, Anderson KC. Blockade of Hsp27 overcomes Bortezomib/proteasome inhibitor PS-341 resistance in lymphoma cells. Cancer Res. 2003; 63: 6174–6177 12. Chauhan D, Catley L, Li G, Podar K, Hideshima T, Velankar M, Mitsiades C, Mitsiades N, Yasui H, Letai A, Ovaa H, Berkers C, Nicholson B, Chao TH, Neuteboom ST, Richardson P, Palladino MA, Anderson KC. A novel orally active proteasome inhibitor induces apoptosis in multiple myeloma cells with mechanisms distinct from Bortezomib. Cancer Cell. 2005; 8: 407–419 13. Chauhan D, Hideshima T, Anderson KC. A novel proteasome inhibitor NPI-0052 as an anticancer therapy. Br J Cancer. 2006; 95: 961–965 14. Chauhan D, Bianchi G, Anderson KC. Targeting the UPS as therapy in multiple myeloma. BMC Biochem. 2008; 9 Suppl 1: S1 15. Chauhan D, Singh A, Brahmandam M, Podar K, Hideshima T, Richardson P, Munshi N, Palladino MA, Anderson KC. Combination of proteasome inhibitors bortezomib and NPI0052 trigger in vivo synergistic cytotoxicity in multiple myeloma. Blood. 2008; 111: 1654–1664 16. Chen S, Dai Y, Harada H, Dent P, Grant S. Mcl-1 down-regulation potentiates ABT-737 lethality by cooperatively inducing Bak activation and Bax translocation. Cancer Res. 2007; 67: 782–791 17. Cortes J, Thomas D, Koller C, Giles F, Estey E, Faderl S, Garcia-Manero G, McConkey D, Ruiz SL, Guerciolini R, Wright J, Kantarjian H. Phase I study of bortezomib in refractory or relapsed acute leukemias. Clin Cancer Res. 2004; 10: 3371–3376 18. Coux O, Tanaka K, Goldberg AL. Structure and functions of the 20S and 26S proteasomes. Annu Rev Biochem. 1996; 65: 801–847 19. Crawford LJ, Walker B, Ovaa H, Chauhan D, Anderson KC, Morris TC, Irvine AE (2006) Comparative selectivity and specificity of the proteasome inhibitors BzLLLCOCHO, PS-341, and MG-132. Cancer Res. 66: 6379–6386 20. Demo SD, Kirk CJ, Aujay MA, Buchholz TJ, Dajee M, Ho MN, Jiang J, Laidig GJ, Lewis ER, Parlati F, Shenk KD, Smyth MS, Sun CM, Vallone MK, Woo TM, Molineaux CJ, Bennett MK. Antitumor activity of PR-171, a novel irreversible inhibitor of the proteasome. Cancer Res. 2007; 67: 6383–6391
294
J. Chandra and C.P. Miller
21. Ding WX, Ni HM, Chen X, Yu J, Zhang L, Yin XM. A coordinated action of Bax, PUMA, and p53 promotes MG132-induced mitochondria activation and apoptosis in colon cancer cells. Mol Cancer Ther. 2007; 6: 1062–1069 22. Drexler HC. Activation of the cell death program by inhibition of proteasome function. Proc Natl Acad Sci U S A. 1997; 94: 855–860 23. Egger L, Madden DT, Rheme C, Rao RV, Bredesen DE. Endoplasmic reticulum stressinduced cell death mediated by the proteasome. Cell Death Differ. 2007; 14: 1172–1180 24. Fennell DA, Chacko A, Mutti L. BCL-2 family regulation by the 20S proteasome inhibitor bortezomib. Oncogene. 2008; 27: 1189–1197 25. Fenteany G, Standaert RF, Reichard GA, Corey EJ, Schreiber SL. A beta-lactone related to lactacystin induces neurite outgrowth in a neuroblastoma cell line and inhibits cell cycle progression in an osteosarcoma cell line. Proc Natl Acad Sci U S A. 1994; 91: 3358–3362 26. Fenteany G, Standaert RF, Lane WS, Choi S, Corey EJ, Schreiber SL. Inhibition of proteasome activities and subunit-specific amino-terminal threonine modification by lactacystin. Science. 1995; 268: 726–731 27. Fribley A, Zeng Q, Wang CY. Proteasome inhibitor PS-341 induces apoptosis through induction of endoplasmic reticulum stress-reactive oxygen species in head and neck squamous cell carcinoma cells. Mol Cell Biol. 2004; 24: 9695–9704 28. Fribley AM, Evenchik B, Zeng Q, Park BK, Guan JY, Zhang H, Hale TJ, Soengas MS, Kaufman RJ, Wang CY. Proteasome inhibitor PS-341 induces apoptosis in cisplatin-resistant squamous cell carcinoma cells by induction of Noxa. J Biol Chem. 2006; 281: 31440–31447 29. Gilmore TD. Multiple myeloma: lusting for NF-kappaB. Cancer Cell. 2007; 12: 95–97 30. Goldberg AL, Rock K. Not just research tools – proteasome inhibitors offer therapeutic promise. Nat Med. 2002; 8: 338–340 31. Goldberg AL. Functions of the proteasome: from protein degradation and immune surveillance to cancer therapy. Biochem Soc Trans. 2007; 35: 12–17 32. Gomez-Bougie P, Wuilleme-Toumi S, Menoret E, Trichet V, Robillard N, Philippe M, Bataille R, Amiot M. Noxa up-regulation and Mcl-1 cleavage are associated to apoptosis induction by bortezomib in multiple myeloma. Cancer Res. 2007; 67: 5418–5424 33. Goy A, Younes A, McLaughlin P, Pro B, Romaguera JE, Hagemeister F, Fayad L, Dang NH, Samaniego F, Wang M, Broglio K, Samuels B, Gilles F, Sarris AH, Hart S, Trehu E, Schenkein D, Cabanillas F, Rodriguez AM. Phase II study of proteasome inhibitor bortezomib in relapsed or refractory B-cell non-Hodgkin’s lymphoma. J Clin Oncol. 2005; 23: 667–675 34. Groll M, Huber R, Potts BC. Crystal Structures of Salinosporamide A (NPI-0052) and B (NPI-0047) in Complex with the 20S Proteasome Reveal Important Consequences of betaLactone Ring Opening and a Mechanism for Irreversible Binding. J Am Chem Soc. 2006; 128: 5136–5141 35. Hanada M, Sugawara K, Kaneta K, Toda S, Nishiyama Y, Tomita K, Yamamoto H, Konishi M, Oki T. Epoxomicin, a new antitumor agent of microbial origin. J Antibiot (Tokyo). 1992; 45: 1746–1752 36. Hershko A, Heller H, Elias S, Ciechanover A. Components of ubiquitin-protein ligase system. Resolution, affinity purification, and role in protein breakdown. J Biol Chem. 1983; 258: 8206–8214 37. Hetz CA. ER stress signaling and the BCL-2 family of proteins: from adaptation to irreversible cellular damage. Antioxid Redox Signal. 2007; 9: 2345–2355 38. Hideshima T, Chauhan D, Richardson P, Mitsiades C, Mitsiades N, Hayashi T, Munshi N, Dang L, Castro A, Palombella V, Adams J, Anderson KC. NF-kappa B as a therapeutic target in multiple myeloma. J Biol Chem. 2002; 277: 16639–16647 39. Howard AN, Bridges KA, Meyn RE, Chandra J. ABT-737, a BH-3 mimetic, induces glutathione depletion and oxidative stress. Cancer Chemother Pharmacol. 2009; 65: 41–54 40. Jones P, Steinkuhler C. From natural products to small molecule ketone histone deacetylase inhibitors: development of new class specific agents. Curr Pharm Des. 2008; 14: 545–561
13 Therapeutic Utility of Proteasome Inhibitors for Acute Leukemia
295
41. Jung L, Holle L, Dalton WS. Discovery, Development, and clinical applications of bortezomib. Oncology (Williston Park). 2004; 18: 4–13 42. Karin M, Ben-Neriah Y. Phosphorylation meets ubiquitination: the control of NF-[kappa]B activity. Annu Rev Immunol. 2000; 18: 621–663 43. Karin M, Yamamoto Y, Wang QM. The IKK NF-kappa B system: a treasure trove for drug development. Nat Rev Drug Discov. 2004; 3: 17–26 44. Khan O, La Thangue NB. Drug Insight: histone deacetylase inhibitor-based therapies for cutaneous T-cell lymphomas. Nat Clin Pract Oncol. 2008; 5: 714–726 45. Kim I, Xu W, Reed JC. Cell death and endoplasmic reticulum stress: disease relevance and therapeutic opportunities. Nat Rev Drug Discov. 2008; 7: 1013–1030 46. Kisselev AF, Akopian TN, Castillo V, Goldberg AL. Proteasome active sites allosterically regulate each other, suggesting a cyclical bite-chew mechanism for protein breakdown. Mol Cell. 1999; 4: 395–402 47. Kisselev AF, Garcia-Calvo M, Overkleeft HS, Peterson E, Pennington MW, Ploegh HL, Thornberry NA, Goldberg AL The caspase-like sites of proteasomes, their substrate specificity, new inhibitors and substrates, and allosteric interactions with the trypsin-like sites. J Biol Chem. 2003; 278: 35869–35877 48. Kordes U, Krappmann D, Heissmeyer V, Ludwig WD, Scheidereit C. Transcription factor NF-kappaB is constitutively activated in acute lymphoblastic leukemia cells. Leukemia. 2000; 14: 399–402 49. Kraus M, Ruckrich T, Reich M, Gogel J, Beck A, Kammer W, Berkers CR, Burg D, Overkleeft H, Ovaa H, Driessen C. Activity patterns of proteasome subunits reflect bortezomib sensitivity of hematologic malignancies and are variable in primary human leukemia cells. Leukemia. 2007; 21: 84–92 50. Kuhn DJ, Chen Q, Voorhees PM, Strader JS, Shenk KD, Sun CM, Demo SD, Bennett MK, van Leeuwen FW, Chanan-Khan AA, Orlowski RZ. Potent activity of carfilzomib, a novel, irreversible inhibitor of the ubiquitin-proteasome pathway, against preclinical models of multiple myeloma. Blood. 2007; 110: 3281–3290 51. Kumatori A, Tanaka K, Inamura N, Sone S, Ogura T, Matsumoto T, Tachikawa T, Shin S, Ichihara A. Abnormally high expression of proteasomes in human leukemic cells. Proc Natl Acad Sci U S A. 1990; 87: 7071–7075 52. Lee DH, Goldberg AL. Proteasome inhibitors: valuable new tools for cell biologists. Trends Cell Biol. 1998; 8: 397–403 53. Letai A. BCL-2: found bound and drugged! Trends Mol Med. 2005; 11: 442–444 54. Li C, Li R, Grandis JR, Johnson DE. Bortezomib induces apoptosis via Bim and Bik upregulation and synergizes with cisplatin in the killing of head and neck squamous cell carcinoma cells. Mol Cancer Ther. 2008; 7: 1647–1655 55. Ling YH, Liebes L, Zou Y, Perez-Soler R. Reactive oxygen species generation and mitochondrial dysfunction in the apoptotic response to Bortezomib, a novel proteasome inhibitor, in human H460 non-small cell lung cancer cells. J Biol Chem. 2003; 278: 33714–33723 56. Lu S, Chen Z, Yang J, Chen L, Gong S, Zhou H, Guo L, Wang J. Overexpression of the PSMB5 gene contributes to bortezomib resistance in T-lymphoblastic lymphoma/leukemia cells derived from Jurkat line. Exp Hematol. 2008; 36: 1278–1284 57. Lu S, Yang J, Song X, Gong S, Zhou H, Guo L, Song N, Bao X, Chen P, Wang J. Point mutation of the proteasome beta5 subunit gene is an important mechanism of bortezomib resistance in bortezomib-selected variants of Jurkat T cell lymphoblastic lymphoma/leukemia line. J Pharmacol Exp Ther. 2008; 326: 423–431 58. Lu S, Chen Z, Yang J, Chen L, Zhou H, Xu X, Li J, Han F, Wang J. The effects of proteasome inhibitor bortezomib on a P-gp positive leukemia cell line K562/A02. Int J Lab Hematol. 2009; 32(1 Pt 1):e123–e131. 59. Ma W, Kantarjian H, O’Brien S, Jilani I, Zhang X, Estrov Z, Ferrajoli A, Keating M, Giles F, Albitar M. Enzymatic activity of circulating proteasomes correlates with clinical behavior in patients with chronic lymphocytic leukemia. Cancer. 2008; 112: 1306–1312
296
J. Chandra and C.P. Miller
60. Macherla VR, Mitchell SS, Manam RR, Reed KA, Chao TH, Nicholson B, Deyanat-Yazdi G, Mai B, Jensen PR, Fenical WF, Neuteboom ST, Lam KS, Palladino MA, Potts BC. Structureactivity relationship studies of salinosporamide A (NPI-0052), a novel marine derived proteasome inhibitor. J Med Chem. 2005; 48: 3684–3687 61. Malhotra JD, Kaufman RJ. The endoplasmic reticulum and the unfolded protein response. Semin Cell Dev Biol. 2007; 18: 716–731 62. Marchion D, Munster P, Development of histone deacetylase inhibitors for cancer treatment. Expert Rev Anticancer Ther, 2007; 7: 583–598 63. Masdehors P, Omura S, Merle-Beral H, Mentz F, Cosset JM, Dumont J, Magdelenat H, Delic J. Increased sensitivity of CLL-derived lymphocytes to apoptotic death activation by the proteasome-specific inhibitor lactacystin. Br J Haematol. 1999; 105: 752–757 64. Matteson DS. alpha-Amido boronic acids: a synthetic challenge and their properties as serine protease inhibitors. Med Res Rev. 2008; 28: 233–246 65. McConkey DJ, Zhu K. Mechanisms of proteasome inhibitor action and resistance in cancer. Drug Resist Update. 2008; 11: 164–179 66. Meng L, Mohan R, Kwok BH, Elofsson M, Sin N, Crews CM. Epoxomicin, a potent and selective proteasome inhibitor, exhibits in vivo antiinflammatory activity. Proc Natl Acad Sci U S A. 1999; 96: 10403–10408 67. Miller CP, Ban K, Dujka ME, McConkey DJ, Munsell M, Palladino M, Chandra J. NPI-0052, a novel proteasome inhibitor, induces caspase-8 and ROS-dependent apoptosis alone and in combination with HDAC inhibitors in leukemia cells. Blood. 2007; 110: 267–277 68. Miller CP, Rudra S, Keating MJ, Wierda WG, Palladino M, Chandra J. Caspase-8 dependent histone acetylation by a novel proteasome inhibitor, NPI-0052: a mechanism for synergy in leukemia cells. Blood. 2009; 113(18):4289–4299. 69. Miller LA, Goldstein NB, Johannes WU, Walton CH, Fujita M, Norris DA, Shellman YG. BH3 Mimetic ABT-737 and a Proteasome Inhibitor Synergistically Kill Melanomas through Noxa-Dependent Apoptosis. J Invest Dermatol. 2008; 129(4):964–971. 70. Mitsiades CS, Mitsiades NS, McMullan CJ, Poulaki V, Shringarpure R, Hideshima T, Akiyama M, Chauhan D, Munshi N, Gu X, Bailey C, Joseph M, Libermann TA, Richon VM, Marks PA, Anderson KC. Transcriptional signature of histone deacetylase inhibition in multiple myeloma: biological and clinical implications. Proc Natl Acad Sci U S A. 2004; 101: 540–545 71. Mitsiades CS, Mitsiades NS, McMullan CJ, Poulaki V, Kung AL, Davies FE, Morgan G, Akiyama M, Shringarpure R, Munshi NC, Richardson PG, Hideshima T, Chauhan D, Gu X, Bailey C, Joseph M, Libermann TA, Rosen NS, Anderson KC. Antimyeloma activity of heat shock protein-90 inhibition. Blood. 2006; 107: 1092–1100 72. Mitsiades N, Mitsiades CS, Poulaki V, Chauhan D, Fanourakis G, Gu X, Bailey C, Joseph M, Libermann TA, Treon SP, Munshi NC, Richardson PG, Hideshima T, Anderson KC. Molecular sequelae of proteasome inhibition in human multiple myeloma cells. Proc Natl Acad Sci U S A. 2002; 99: 14374–14379 73. O’Connor OA, Wright J, Moskowitz C, Muzzy J, MacGregor-Cortelli B, Stubblefield M, Straus D, Portlock C, Hamlin P, Choi E, Dumetrescu O, Esseltine D, Trehu E, Adams J, Schenkein D, Zelenetz AD. Phase II clinical experience with the novel proteasome inhibitor bortezomib in patients with indolent non-Hodgkin’s lymphoma and mantle cell lymphoma. J Clin Oncol. 2005; 23: 676–684 74. Ocio EM, Mateos MV, Maiso P, Pandiella A, San-Miguel JF. New drugs in multiple myeloma: mechanisms of action and phase I/II clinical findings. Lancet Oncol. 2008; 9: 1157–1165 75. Okumura K, Huang S, Sinicrope FA. Induction of Noxa sensitizes human colorectal cancer cells expressing Mcl-1 to the small-molecule Bcl-2/Bcl-xL inhibitor, ABT-737. Clin Cancer Res. 2008; 14: 8132–8142 76. Oltersdorf T, Elmore SW, Shoemaker AR, Armstrong RC, Augeri DJ, Belli BA, Bruncko M, Deckwerth TL, Dinges J, Hajduk PJ, Joseph MK, Kitada S, Korsmeyer SJ, Kunzer AR, Letai A, Li C, Mitten MJ, Nettesheim DG, Ng S, Nimmer PM, O’Connor JM, Oleksijew A, Petros AM, Reed JC, Shen W, Tahir SK, Thompson CB, Tomaselli KJ, Wang B, Wendt MD, Zhang H,
13 Therapeutic Utility of Proteasome Inhibitors for Acute Leukemia
297
Fesik SW, Rosenberg SH. An inhibitor of Bcl-2 family proteins induces regression of solid tumours. Nature. 2005; 435: 677–681 77. Omura S, Fujimoto T, Otoguro K, Matsuzaki K, Moriguchi R, Tanaka H, Sasaki Y. Lactacystin, a novel microbial metabolite, induces neuritogenesis of neuroblastoma cells. J Antibiot (Tokyo). 1991; 44: 113–116 78. Orlowski M, Wilk S. Catalytic activities of the 20 S proteasome, a multicatalytic proteinase complex. Arch Biochem Biophys. 2000; 383: 1–16 79. Orlowski RZ, Stinchcombe TE, Mitchell BS, Shea TC, Baldwin AS, Stahl S, Adams J, Esseltine DL, Elliott PJ, Pien CS, Guerciolini R, Anderson JK, Depcik-Smith ND, Baghat R, Lehman MJ, Novick SC, O’Connor OA, Soignet SL. Phase I trial of the proteasome inhibitor PS-341 in patients with refractory hematologic malignancies. J Clin Oncol. 2002; 20:4420–4427 80. Orlowski RZ, Voorhees PM, Garcia RA, Hall MD, Kudrik FJ, Allred T, Johri AR, Jones PE, Ivanova A, Van Deventer HW, Gabriel DA, Shea TC, Mitchell BS, Adams J, Esseltine DL, Trehu EG, Green M, Lehman MJ, Natoli S, Collins JM, Lindley CM, Dees EC. Phase 1 trial of the proteasome inhibitor bortezomib and pegylated liposomal doxorubicin in patients with advanced hematologic malignancies. Blood. 2005; 105: 3058–3065 81. Orlowski RZ, Nagler A, Sonneveld P, Blade J, Hajek R, Spencer A, San Miguel J, Robak T, Dmoszynska A, Horvath N, Spicka I, Sutherland HJ, Suvorov AN, Zhuang SH, Parekh T, Xiu L, Yuan Z, Rackoff W, Harousseau JL. Randomized phase III study of pegylated liposomal doxorubicin plus bortezomib compared with bortezomib alone in relapsed or refractory multiple myeloma: combination therapy improves time to progression. J Clin Oncol. 2007; 25: 3892–3901 82. Orlowski RZ, Kuhn DJ. Proteasome inhibitors in cancer therapy: lessons from the first decade. Clin Cancer Res. 2008; 14: 1649–1657 83. Ostrowska H, Hempel D, Holub M, Sokolowski J, Kloczko J. Assessment of circulating proteasome chymotrypsin-like activity in plasma of patients with acute and chronic leukemias. Clin Biochem. 2008; 41: 1377–1383 84. Paoluzzi L, Gonen M, Bhagat G, Furman RR, Gardner JR, Scotto L, Gueorguiev VD, Heaney ML, Manova K, O’Connor OA.The BH3-only mimetic ABT-737 synergizes the antineoplastic activity of proteasome inhibitors in lymphoid malignancies. Blood. 2008; 112: 2906–2916 85. Papandreou CN, Daliani DD, Nix D, Yang H, Madden T, Wang X, Pien CS, Millikan RE, Tu SM, Pagliaro L, Kim J, Adams J, Elliott P, Esseltine D, Petrusich A, Dieringer P, Perez C, Logothetis CJ. Phase I trial of the proteasome inhibitor bortezomib in patients with advanced solid tumors with observations in androgen-independent prostate cancer. J Clin Oncol. 2004; 22: 2108–2121 86. Perez-Galan P, Roue G, Villamor N, Montserrat E, Campo E, Colomer D. The proteasome inhibitor bortezomib induces apoptosis in mantle-cell lymphoma through generation of ROS and Noxa activation independent of p53 status. Blood. 2006; 107: 257–264 87. Peters JM, Cejka Z, Harris JR, Kleinschmidt JA, Baumeister W. Structural features of the 26 S proteasome complex. J Mol Biol. 1993; 234: 932–937 88. Qin JZ, Ziffra J, Stennett L, Bodner B, Bonish BK, Chaturvedi V, Bennett F, Pollock PM, Trent JM, Hendrix MJ, Rizzo P, Miele L, Nickoloff BJ. Proteasome inhibitors trigger NOXAmediated apoptosis in melanoma and myeloma cells. Cancer Res. 2005; 65: 6282–6293 89. Reed JC, Pellecchia M. Apoptosis-based therapies for hematologic malignancies. Blood. 2005; 106: 408–418 90. Richardson P. Clinical update: proteasome inhibitors in hematologic malignancies. Cancer Treat Rev. 2003; 29 Suppl 1: 33–39 91. Richardson PG, Barlogie B, Berenson J, Singhal S, Jagannath S, Irwin D, Rajkumar SV, Srkalovic G, Alsina M, Alexanian R, Siegel D, Orlowski RZ, Kuter D, Limentani SA, Lee S, Hideshima T, Esseltine DL, Kauffman M, Adams J, Schenkein DP, Anderson KC. A phase 2 study of bortezomib in relapsed, refractory myeloma. N Engl J Med. 2003; 348: 2609–2617 92. Richardson PG, Barlogie B, Berenson J, Singhal S, Jagannath S, Irwin DH, Rajkumar SV, Srkalovic G, Alsina M, Anderson KC. Extended follow-up of a phase II trial in relapsed,
298
J. Chandra and C.P. Miller
refractory multiple myeloma:: final time-to-event results from the SUMMIT trial. Cancer. 2006; 106: 1316–1319 93. Richardson PG, Sonneveld P, Schuster M, Irwin D, Stadtmauer E, Facon T, Harousseau JL, Ben-Yehuda D, Lonial S, Goldschmidt H, Reece D, Miguel JS, Blade J, Boccadoro M, Cavenagh J, Alsina M, Rajkumar SV, Lacy M, Jakubowiak A, Dalton W, Boral A, Esseltine DL, Schenkein D, Anderson KC. Extended follow-up of a phase 3 trial in relapsed multiple myeloma: final time-to-event results of the APEX trial. Blood. 2007; 110: 3557–3560 94. Richardson PG, Mitsiades C, Schlossman R, Ghobrial I, Hideshima T, Munshi N, Anderson KC. Bortezomib in the front-line treatment of multiple myeloma. Expert Rev Anticancer Ther. 2008; 8: 1053–1072 95. Roccaro AM, Leleu X, Sacco A, Jia X, Melhem M, Moreau AS, Ngo HT, Runnels J, Azab A, Azab F, Burwick N, Farag M, Treon SP, Palladino MA, Hideshima T, Chauhan D, Anderson KC, Ghobrial IM. Dual targeting of the proteasome regulates survival and homing in Waldenstrom macroglobulinemia. Blood. 2008; 111: 4752–4763 96. Ruiz S, Krupnik Y, Keating M, Chandra J, Palladino M, McConkey D. The proteasome inhibitor NPI-0052 is a more effective inducer of apoptosis than bortezomib in lymphocytes from patients with chronic lymphocytic leukemia. Mol Cancer Ther. 2006; 5: 1836–1843 97. Russo SM, Tepper JE, Baldwin AS, Jr., Liu R, Adams J, Elliott P, Cusack JC, Jr. Enhancement of radiosensitivity by proteasome inhibition: implications for a role of NF-kappaB. Int J Radiat Oncol Biol Phys. 2001; 50: 183–193 98. Shringarpure R, Catley L, Bhole D, Burger R, Podar K, Tai YT, Kessler B, Galardy P, Ploegh H, Tassone P, Hideshima T, Mitsiades C, Munshi NC, Chauhan D, Anderson KC. Gene expression analysis of B-lymphoma cells resistant and sensitive to bortezomib. Br J Haematol. 2006; 134: 145–156 99. Srikanth M, Davies FE, Morgan GJ. An update on drug combinations for treatment of myeloma. Expert Opin Investig Drugs. 2008; 17: 1–12 100. Sunwoo JB, Chen Z, Dong G, Yeh N, Crowl Bancroft C, Sausville E, Adams J, Elliott P, Van Waes C. Novel proteasome inhibitor PS-341 inhibits activation of nuclear factor-kappa B, cell survival, tumor growth, and angiogenesis in squamous cell carcinoma. Clin Cancer Res. 2001; 7: 1419–1428 101. Vermeulen K, Van Bockstaele DR, Berneman ZN. Apoptosis: mechanisms and relevance in cancer. Ann Hematol. 2005; 84: 627–639 102. Voges D, Zwickl P, Baumeister W. The 26S proteasome: a molecular machine designed for controlled proteolysis. Annu Rev Biochem. 1999; 68: 1015–1068 103. Williams S, Pettaway C, Song R, Papandreou C, Logothetis C, McConkey DJ. Differential effects of the proteasome inhibitor bortezomib on apoptosis and angiogenesis in human prostate tumor xenografts. Mol Cancer Ther. 2003; 2: 835–843 104. Williamson MJ, Blank JL, Bruzzese FJ, Cao Y, Daniels JS, Dick LR, Labutti J, Mazzola AM, Patil AD, Reimer CL, Solomon MS, Stirling M, Tian Y, Tsu CA, Weatherhead GS, Zhang JX, Rolfe M. Comparison of biochemical and biological effects of ML858 (salinosporamide A) and bortezomib. Mol Cancer Ther. 2006; 5: 3052–3061 105. Wolf DH, Hilt W. The proteasome: a proteolytic nanomachine of cell regulation and waste disposal. Biochim Biophys Acta. 2004; 1695: 19–31 106. Yan H, Wang YC, Li D, Wang Y, Liu W, Wu YL, Chen GQ. Arsenic trioxide and proteasome inhibitor bortezomib synergistically induce apoptosis in leukemic cells: the role of protein kinase Cdelta. Leukemia. 2007; 21: 1488–1495 107. Yi CH, Yuan J. The Jekyll and Hyde functions of caspases. Dev Cell. 2009; 16: 21–34 108. Youle RJ, Strasser A. The BCL-2 protein family: opposing activities that mediate cell death. Nat Rev Mol Cell Biol. 2008; 9: 47–59 109. Yu C, Friday BB, Lai JP, Yang L, Sarkaria J, Kay NE, Carter CA, Roberts LR, Kaufmann SH, Adjei AA. Cytotoxic synergy between the multikinase inhibitor sorafenib and the proteasome inhibitor bortezomib in vitro: induction of apoptosis through Akt and c-Jun NH2terminal kinase pathways. Mol Cancer Ther. 2006; 5: 2378–2387
Chapter 14
Targeting Epigenetic Pathways in ALL Pamela Kearns
14.1 Introduction Epigenetics refers to mitotically heritable changes in gene expression that do not involve a change in DNA sequence. Epigenetic control of gene expression is now recognised as an important mechanism in the initiation and prognosis of human malignancies including leukemias [1, 2]. Both the process of methylation of cytosines in CpG island regions situated within gene promoter regions and changes in chromatin conformation mediated by histone acetylation lead to transcriptional silencing. A growing number of epigenetically silenced genes are being recognised in cancer, and cancer type-specific patterns of hypermethylation are emerging. In addition, genes predicted as relevant to tumorigenesis have been identified as being under epigenetic control. Pharmacological targeting of these processes is of increasing interest as a therapeutic modality, but its application in the treatment of ALL remains to be proven.
14.2 DNA Methylation The methylation of the C5 position of the cytosine residues in CpG dinucleotides is a known mechanism of epigenetic silencing and is maintained by a family of DNA methyltransferases (DNMTs). There is a non-random distribution of CpG dinuleotides in the human genome. A CpG island is a region of more than 500 base pairs that are GC-rich to the level of greater than 55%. Approximately 40–60% of mammalian genes have a CpG island located within their promoter region. These are highly conserved regions and are normally not methylated but when methylated are associated with transcriptional silencing.
P. Kearns (*) School of Cancer Sciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK e-mail:
[email protected] V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3_14, © Springer Science+Business Media, LLC 2011
299
300
P. Kearns
14.3 Histone Acetylation Eukaryotic chromatin is structured around the core nucleosome, which comprises an octamer of proteins called histones. Chromatin exists in two different configurations: open and transcriptionally active or closed and transcriptionally inactive. The configuration of chromatin is influenced by post-translational modifications of histones, including acetylation and methylation. Acetylation is associated with transcriptionally active regions, as opposed to hypo-acetylated regions that are associated with inactive euchromatic or heterochromatic regions. Histone methylation occurs in both active and inactive areas of chromatin. Methylation of lysine 9 on the N-terminus of histone 3 (H3-K9) is associated with silent chromatin and can be triggered by CpG methylation. In contrast, methylation of lysine 4 on histone 3 (H3-K4) denotes activity and is found predominantly in the promoter regions of active genes. The multiple possible variations in histone modification have been referred to as the histone code, and the full implication of these variations on cellular function remains to be unravelled [2]. Histone acetylation is mediated by histone acetyl transferases (HATs) and reversed by a family of histone deacetylators (HDACs). Eighteen isoforms of human HDACs have been described, and they can be divided into four classes (HDAC 1, 2, 3 and 8), based on their structural homology with yeast HDACs. It has been observed that the function of HDACs extends beyond histones. They also deacetylate non-histone proteins [3], and this has complicated interpretation of the effects of therapeutic HDAC inhibitors.
14.4 Epigenetic Dysregulation in Cancer The epigenetic state of a cell is significantly disrupted in malignant cells. A pattern of global DNA hypomethylation with hypermethylation of CpG islands is observed in the majority of cancer types. In addition, aberrant HDAC activity is also observed in many malignancies. In the early 1980s, it was observed that a substantial proportion of CpGs were hypomethylated in cancer compared to normal tissue [4]. Subsequently, the phenomenon of global 5’methylcytosine hypomethylation in cancer has been confirmed in many cancer types, with cancer cells having between 20 and 60% less methylated CpG sites. Loss of methylation in the promoter region of the LINE (long interspersed nuclear elements) retrotransposons is reported in the many types of malignancies and an inverse relationship to microsatellite instability has been reported [5]. The further functional significance of hypomethylation in cancer is becoming recognised with the evidence of related activation of oncogenes, for example MAGE expression in melanoma [6], cyclinD2 in gastric cancer [7] and HRAS in several types of carcinomas [8–10]. More intensively investigated is the phenomenon of aberrant hypermethylation of CpG islands in and around gene promoter regions. This was first described for the calcitonin gene [11], but the RB gene was the first tumour suppressor gene reported to be repressed by site-specific methylation [12]. Since then, a growing list of
14 Targeting Epigenetic Pathways in ALL
301
other genes have been shown to be epigenetically silenced, and many are functionally associated with tumorigenesis, including cell cycle regulation, DNA repair, apoptosis and angiogenesis [13–15]. Aberrant hypermethylation has been reported in nearly all types of malignancies, including leukemias. Not all epigentically controlled genes are methylated in every cancer type, and the observation of cancer specific profiles of DNA hypermethylation is now well established. They were first described by candidate gene approaches [1, 16] and are now substantiated for several cancer types by more recent genomewide DNA methylomic techniques [17–19]. The mechanisms leading to aberrant DNA hypermethylation is not fully elucidated, although elevations in the expression of DNMTs are observed in solid tumours and leukemias, with clear differences in the pattern of changed DNMT expression between tumour types [20]. How this translates to tumour-type specific subsets of methylated gene promoter CpG islands is not known. Disease-specific oncogenic transcription factors mediating DNA methylation may be a possible mechanism of specific profiles of hypermethylation.
14.5 Aberrant Epigenetic Pathways in ALL Studies of aberrant DNA methylation in ALL have graduated from early studies that focussed on single genes, through examination of the methylation status of multiple gene sets to the more recent genomewide studies simultaneously analysing CpG islands throughout the genome. The initial single-gene analyses in acute leukemias identified promoter hypermethylation in a wide variety of genes, including E-cadherin [21], calcitonin [22], oestrogen receptor [23], hypermethylation in cancer-1 (HIC-1) [24] and p15INK4b [25, 26] and many others. Further studies of multiple gene sets identified the importance of concomitant methylation of multiple genes. Increased DNA methylation, both in terms of numbers of genes and methylation density, is consistently reported as an independent factor of poor prognosis in leukemia [27–31]. There is a clear selection bias in single gene and multiple candidate gene studies. Genomewide analyses of DNA methylation substantially reduced this bias and have identified disease-specific methylation profiles in acute leukemias. Even the earliest microarray-based DNA methylation methodology, examining 249 CpG sites relating to 57 genes, could differentiate between myeloid and B-precursor lymphoid leukemia [32]. Further genomewide DNA methylation analyses are revealing unanticipated molecular pathways and have been successfully used to identify novel methylation target genes in ALL [33, 34]. Taylor et al. [33] identified 262 aberrantly methylated CpG sites related to promoter regions of genes involved in cellular processes including cell growth, transcription, nucleotide binding, apoptosis and cell signalling. Mapping of methylated loci revealed nonrandom distribution across chromosomal sites suggesting the possibility of methylation “hotspots”, notably in chromosomes 11, 18 and 19. This interesting but as yet unexplained phenomenon has been observed in other cancers, with the regions targeted for methylation appearing to be cancer specific.
302
P. Kearns
The DNA methylome observed in adult ALL has been compared to paediatric ALL and was not thought to differ [35]; however, the study was small with limited patients numbers, and only a subset of seven genes was analysed. More comprehensive studies are awaited. Differences are likely to be identified based on leukemia biology rather than purely age. This is borne out by the identification of DNA methylation profiles specific to defined chromosomal aberrations, including studies of paediatric ALL characterised by hyperdiploidy and ETV6/RUNX1 fusion [36] and infant leukemias characterised by the mixed-lineage leukemia (MLL) translocations [37]. Although the family of DNMTs responsible for DNA methylation is known, the mechanism leading to CpG island hypermethylation in malignant cells is not fully understood. Increased DNMT activity has been reported in malignant cells, including AML [38]; however, this does not provide an explanation for the specificity of CpG islands that become hypermethylated. Interactions between genetic and epigenetic lesions in carcinogenesis are very probable, and the high frequency of chromosomal translocations in leukemias has made this a fertile area for research. Some notable fusion proteins associated with chromosomal translocations described in leukemia have been shown to play key roles in epigenetic modifications, including t(15;17) and MLL gene rearrangements. The translocation t(15;17) generates the PML-RARa transcript, the crucial initiating event in acute promyelocytic leukemia. PML-RARa has been shown to recruit DNMTs and HDACs to the CpG island of the RARb2 gene, resulting in transcriptional silencing [39, 40]. Moreover, treatment with retinoic acid can reverse the methylation associated gene silencing [41]. Chromatin immunoprecipitation (ChIP)-chip experiments identified chromatin modifications induced by PMLRARa that repress critical targets in leukemogenesis [42]. PML-RARa was shown to induce heterochromatin formation in all of 372 target genes identified, and associated HDAC1 recruitment, loss of histone 3 acetylation and loss of histone H lysine 9 trimethylation was observed. The PML-RARa genomic targets had functional diversity including cell cycle control and apoptosis. The PML-RARa mediated induction to chromatin modifications associated with repression of these key genes defines a mechanism for a disease-specific “epigenome” and supports an important genetic and epigenetic interaction in leukemogenesis. The MLL gene is a transcriptional regulator and a member of the Drosophila trithorax family, and the MLL gene product MLL1 possesses H3-K4 methyltransferase activity [43, 44]. Chromosomal translocations fusing the MLL gene with over 50 partners are associated with a specific subgroup of acute leukemias with predominance in infancy and an associated poor prognosis. Over 80% of infant ALL carry an MLL gene rearrangement and are characterised by gene expression profiles that are distinguishable from other ALL subtypes [45]. It has been recently reported that MLL-rearranged ALL is also characterised by differential genomewide DNA methylation profiles [37]. The CpG island methylation patterns were shown to differ between subtypes of MLL gene rearrangement. MLL translocations t(4:11) and t(11:19) were associated with heavily methylated profiles compared to t(9:11) and wild-type MLL. Interestingly, this difference also translated to relapse-free survival, with more heavily methylated genome carrying a worse outcome. It is interesting to note that not all hypermethylated
14 Targeting Epigenetic Pathways in ALL
303
genes identified in this study translated to repressed gene expression, highlighting the complexity of the relationship between epigenetic factors and gene expression. The density of methylation in the CpG islands and also the potential interaction with histone modifications may influence the gene silencing effect.
14.6 Epigenetic Pathways as a Therapeutic Target Unlike genetic alterations, for example gene deletions and mutations, epigenetic events do not alter the primary DNA sequence. Epigenetic processes are biochemical changes that affect DNA and the associated proteins. These processes determine the configuration of chromatin and modify gene expression [46]. Importantly, they are reversible and, therefore, can be therapeutically targeted. Currently, there are two types of clinically applicable epigenetic modulators that are able to upregulate epigenetically silenced genes: the DNA methyltransferase inhibitors (DNMTi) and the histone deacetylase inhibitors (HDACi). Clinical trials are beginning to inform us of how they can be best applied in the treatment of haematological malignancies.
14.7 Clinical Trials of DNA Methyltransferase Inhibitors There are 2 DNMTi in clinical use: 5-azacytidine and 5-aza-2’-deoxycytidine (decitabine). Both are nucleoside analogues that were originally investigated as cytotoxic agents for leukemias over 3 decades ago, when they were administered at high doses (5-azacytidine 600–1,500 mg/m2 and decitabine 1,500–2,500 mg/m2). Clinical studies in leukemia escalating to the maximum tolerated doses demonstrated promising responses but the dose-limiting toxicity was prolonged myelosuppression. Dosedependent anti-leukemia activity for decitabine was reported in phase I trials for paediatric relapsed and refractory leukemia (0.75–80 mg/kg), but the associated significant prolonged myelosuppression limited further development [47, 48]. Both 5-azacytidine and decitabine are activated by phosphorylation to their triphosphate forms. 5-azacytidine is predominantly incorporated into RNA, resulting in disassembly of polyribosomes, altered RNA methylation and a defective receptor function of transfer RNA. Decitabine triphosphate is incorporated into DNA-depleting DNMTs, resulting in DNA hypomethylation. At the higher doses, decitabine forms DNA adducts that inhibit DNA synthesis and induce cell death, whereas at lower doses, the DNA hypomethylation causes changes in gene expression profiles that induce differentiation, reduce proliferation and increase apoptosis [49]. The therapeutic potential of the now recognised hypomethylation activity of both 5-azacytidine and decitabine at low doses (5-azacytidine 50–75 mg/m2, decitabine 100–150 mg/m2) is under evaluation in several types of leukemia, notably myelodysplastic syndromes (MDS) and myeloid leukemias. The Cancer and Leukemia Group B (CALGB) performed a series of clinical trials investigating low-dose 5-azacytidine in MDS. The outcome of the phase II trials (CALGB 8421
304
P. Kearns
[50] and CALGB 8921 [50]) and the pivotal randomised phase III trial (CALGB 9221 [50, 51]) that compared low-dose 5-azacytidine to best supportive care confirmed the role for low-dose 5-azacytidine in the management of MDS. CALGB 9221 demonstrated 10% complete response rate (CR), 1% partial response rate (PR) and 36% haematological improvement (HI) in the 5-azacytidine arm compared to no CR or PR in the supportive-care arm. The crossover design of CALGB 9221 allowing supportive care arm patients to move to the 5-azacytidine arm precluded conclusions on overall survival; however, a subsequent randomised trial reported by Fenaux 2009 et al [52] confirmed that low-dose 5-azacytidine (20 mg/ m2/day × 14 days every 28 days) conferred a survival advantage in high-risk MDS compared to conventional treatment regimes (51% vs. 26% at 2 years) [52]. Decitabine, a more potent hypomethylating agent inducing a different profile of gene re-expression compared to 5-azacytidine [53], is also being explored at lowdose schedules in the treatment of MDS and AML. As with 5-azacytidine, the early-phase clinical trials of decitabine demonstrated that low-dose schedules were associated with promising response rates in MDS. This led to a Phase III trial randomising decitabine against best supportive care [54]. Decitabine was administered at 15 mg/m2 per dose, 8 hourly, ×3 days every 6 weeks and produced 9% CR, 8% PR and 13% HI compared to no responses in the supportive-care arm. Although the responses were durable (median 41 weeks), they did not translate to a statistically significant overall survival advantage (time to AML or death 12.1 months vs. 7.8 months, p = 0.16) [54]. On the basis of the results of these Phase III trials, the FDA has approved both 5-azacytidine and decitabine for the treatment of MDS. Studies investigating the therapeutic value of DNMTi in paediatric MDS are planned but will be challenging due to the low patient numbers in the younger age group. Both the dose and the duration of exposure to hypomethylating agents can influence their ability to inhibit DNA methylation, reduce clonogenicity and reactivate tumour suppressor genes [55, 56]. The successful clinical application of DNMTi in MDS and other haematological malignancies will require further optimisation of the dose and schedule both as single agents and in combination with other epigenetic modifiers and conventional cytotoxic drugs. An important challenge is the identification of biomarkers that can either predict response or monitor efficacy of hypomethylating agents. Reduction in global methylation is measurable shortly after commencement of treatment and peaks at 10–15 days with recovery at 4–6 weeks; however, attempts to identify re-expression of single genes as biomarkers have produced contradictory results. Some clinical studies have suggested a correlation between response and induced hypomethylation affecting specific genes, but this is not a consistent finding in all studies and indeed an inverse relationship has been reported in CML [57]. The optimum response biomarker for hypomethylating agents, therefore, remains unclear and is complicated by the multiplicity of the downstream effects of hypomethylation. Epigenetic regulation of genes includes not only the re-expression of methylation-silenced genes in multiple pathways, proapoptosis, proliferation, differentiation and immune regulation, but also downregulation of oncogenes via reactivation of epigentically regulated microRNAs. Beyond MDS, the clinical investigation of DNMTi has been predominantly in AML. Central review of clinical samples in the Kantarjian trial resulted in reclas-
14 Targeting Epigenetic Pathways in ALL
305
sification of 12 MDS patients to AML; 9 had been randomised to receive decitabine treatment and 5 (56%) achieved an objective response [54]. Subsequently, several low-dose, long-exposure schedules of decitabine have been explored in AML, predominantly in the elderly population where intensive cytotoxic chemotherapy is and not well tolerated. Moderate responses are reported: Lubbert and Minden [58] investigated decitabine in patients with AML (median age 72 years) using an induction regime of 135 mg/m2 continuous intravenous infusion over 72 h every 6 weeks for four courses. Patients in CR could receive maintenance therapy with 20 mg/m2 intravenously for 3 days every 8 weeks. Interestingly, the median time to response was 13 weeks, and objective responses were observed in 31% (14% CR and 17% PR) [58]. Similarly, a study of elderly AML patients (median age 69 years) treated with an alternative schedule of decitabine (20 mg/m2 intravenously for 5 days every 4 weeks) produced an objective response rate (CR and CRi) of 26% [59]. Further studies investigating low-dose schedules of both 5-azacytidine and decitabine in AML, including high-risk subgroups, in induction regimens and exploring their use as maintenance therapy are ongoing. Studies in paediatric AML are still awaited. Although preclinical evidence supports the importance of epigenetic processes in ALL, including in the paediatric age group, to date the clinical exploration of epigenetic modifiers in ALL has been limited, and clinical studies using the lowdose schedules in ALL are needed. There is evidence to suggest that epigenetic modifiers may have specific application in certain subgroups of paediatric acute leukemias due to the epigenetic–genetic crosstalk that has been described in association with the chromosomal translocations that characterise subgroups of childhood acute leukemia. The translocations fuse the DNA-binding domain of a transcriptional activator to a transcriptional repressor, leading to decreased expression of target genes that regulate myeloid differentiation, thereby causing the block in myeloid differentiation that characterises acute leukemia [60, 61]. The accumulating evidence that histone acetylation and promoter methylation contribute to the transcriptional repression activity of fusion proteins, including AML1-ETO [62], PML-RARa [63] [64] and ETV6-RUNX1 [65] would support a potential therapeutic advantage for epigenetic modifiers in these leukemia subtypes. In addition, the constitutive upregulation of HOX genes essential to leukemogenesis associated with abnormalities in the MLL gene at 11q23 has been shown to involve DNMTs and histone acetylase activity [66, 67]. Recently, infant ALL with the MLL rearrangement has been reported to exhibit global hypermethylation and interestingly this was associated with increased sensitivity to decitabine [68]. Leukemias carrying these fusion proteins may prove to be particularly sensitive to epigenetic modulators and would be an important target group for the development of these agents. A single case report of the use of decitabine 15 mg/m2 as a 3-h continuous intravenous infusion, 3 times a day, every 3 days in combination with intravenous dexamethasone as reinduction therapy in a 10-year old with multiply relapsed ALL produced a complete remission that was sustained 8 months post second allogeneic transplant [69]. While this is an interesting result, the full utility of hypomethylating agents in ALL requires formal exploration in the context of clinical trials.
306
P. Kearns
14.8 Clinical Trials of Histone Deacetylase Inhibitors (HDACi) HDACi are another class of epigenetic modifiers undergoing clinical evaluation in haematological malignancies. Acetylation of the NH2-terminal tails of histones is catalysed by histone acetylation transferases (HAT) and results in an open chromatin configuration that promotes gene transcription. Conversely, deacetylation is mediated by histone deacetylases (HDAC) and leads to gene repression. HDAC inhibition, therefore, promotes gene re-expression through chromatin relaxation [70]. Several HDACi have been evaluated clinically in early-phase clinical trials, but disappointingly, the results of their use in the treatment of leukemias do not strongly support their efficacy as single agents [71–73]. Nevertheless, there is clear evidence of synergy in vitro between DNMTi inhibitors and HDACi inhibitors, both in reactivation of gene expression [74] and in anti-leukemic activity [75]. Trials combining decitabine or 5-azacytidine with valproic acid, SAHA, depsipeptide and other newer HDACi are producing promising results. A randomised Phase II study of decitabine vs. decitabine plus valproic acid in MDS and AML that has been recently reported did not show significant difference in overall responses between the two arms. Comparison of molecular responses showed no difference in global hypomethylation (as measured by LINE-1) between the two arms and no correlation with responses; however, sustained hypomethylation and re-expression of specific genes, P15, ATM and miR124, did appear to correlate with response [76].
14.9 MicroRNAs; A Potential New Direction in Epigenetic-Targeted Therapy MicroRNAs (miRNAs) are non-coding RNA of approximately 22 nucleotides, and over 500 have been identified to date in the human genome. They are formed from precursor RNAs that are processed by RNase III endonucleases. MiRNAs bind perfectly or imperfectly to their complementary mRNAs, resulting in degradation or translational repression, respectively. They influence cellular processes including differentiation, apoptosis and proliferation, but the mechanisms by which they function are not yet fully understood [77]. There is evidence to suggest that specific patterns of miRNA expression are associated with myeloid and lymphoid leukemias, and a role in leukemogenesis is suggested [78]. Although it is understood that miRNAs are transcribed by RNase II endonucleases and their expression negatively regulated by their target mRNAs, emerging evidence suggests additional epigenetic regulation. Both DNA methylation and histone modification can influence miRNA expression [79]. Upregulation of miRNAs following exposure to HDAC inhibitors and DNMT inhibitors has been demonstrated in several cancer types [80, 81]. Roman-Gomaz et al. have recently demonstrated that aberrant miRNA methylation in ALL was associated with clinical outcome [82]. The relationship between miRNAs and epigenetic modifications warrants further investigation and represents
14 Targeting Epigenetic Pathways in ALL
307
a potential new therapeutic target for the future, both directly and in contributing to the mechanism of action of known epigenetic modifiers.
14.10 Future Role for Epigenetic Modifiers in Paediatric ALL Paediatric studies investigating the use of epigenetic modifiers including decitabine, valproic acid, MS275 and SAHA, both as single agents and in combination, are underway. Based on the evidence from the trials in adults, the effective application of these agents in the treatment of paediatric haematological malignancies will require careful study of the appropriate dosing and scheduling as well as a better understanding of their true mechanisms of action to aid the development of markers predictive of response.
References 1. Esteller M. Cancer epigenetics: DNA methylation and chromatin alterations in human cancer. Adv Exp Med Biol. 2003;532:39–49. 2. Esteller M. Epigenetics in cancer. N Engl J Med. 2008;358:1148–1159. 3. Glozak MA, Sengupta N, Zhang X, Seto E. Acetylation and deacetylation of non-histone proteins. Gene. 2005;363:15–23. 4. Feinberg AP, Vogelstein B. Hypomethylation distinguishes genes of some human cancers from their normal counterparts. Nature. 1983;301:89–92. 5. Estecio MR, Gharibyan V, Shen L, et al. LINE-1 hypomethylation in cancer is highly variable and inversely correlated with microsatellite instability. PLoS One. 2007;2:e399. 6. De Smet C, De Backer O, Faraoni I, Lurquin C, Brasseur F, Boon T. The activation of human gene MAGE-1 in tumor cells is correlated with genome-wide demethylation. Proc Natl Acad Sci U S A. 1996;93:7149–7153. 7. Oshimo Y, Nakayama H, Ito R, et al. Promoter methylation of cyclin D2 gene in gastric carcinoma. Int J Oncol. 2003;23:1663–1670. 8. Metter J, Cho C. Tissue-specific hypomethylation of the human c-K-ras gene. Nucleic Acids Res. 1989;17:7089–7099. 9. Bhave MR, Wilson MJ, Poirier LA. c-H-ras and c-K-ras gene hypomethylation in the livers and hepatomas of rats fed methyl-deficient, amino acid-defined diets. Carcinogenesis. 1988;9:343–348. 10. Feinberg AP, Vogelstein B. Hypomethylation of ras oncogenes in primary human cancers. Biochem Biophys Res Commun. 1983;111:47–54. 11. Baylin SB, Fearon ER, Vogelstein B, et al. Hypermethylation of the 5’ region of the calcitonin gene is a property of human lymphoid and acute myeloid malignancies. Blood. 1987;70:412–417. 12. Greger V, Passarge E, Hopping W, Messmer E, Horsthemke B. Epigenetic changes may contribute to the formation and spontaneous regression of retinoblastoma. Hum Genet. 1989;83:155–158. 13. Momparler RL. Cancer epigenetics. Oncogene. 2003;22:6479–6483. 14. Feinberg AP, Tycko B. The history of cancer epigenetics. Nat Rev Cancer. 2004;4:143–153. 15. Santini V, Kantarjian HM, Issa JP. Changes in DNA methylation in neoplasia: pathophysiology and therapeutic implications. Ann Intern Med. 2001;134:573–586. 16. Esteller M, Corn PG, Baylin SB, Herman JG. A gene hypermethylation profile of human cancer. Cancer Res. 2001;61:3225–3229.
308
P. Kearns
17. Estecio MR, Issa JP. Tackling the methylome: recent methodological advances in genome-wide methylation profiling. Genome Med. 2009;1:106. 18. Koga Y, Pelizzola M, Cheng E, et al. Genome-wide screen of promoter methylation identifies novel markers in melanoma. Genome Res. 2009;19:1462–1470. 19. Martinez R, Martin-Subero JI, Rohde V, et al. A microarray-based DNA methylation study of glioblastoma multiforme. Epigenetics. 2009;4:255–264. 20. Clark SJ, Melki J. DNA methylation and gene silencing in cancer: which is the guilty party? Oncogene. 2002;21:5380–5387. 21. Melki JR, Vincent PC, Brown RD, Clark SJ. Hypermethylation of E-cadherin in leukemia. Blood. 2000;95:3208–3213. 22. Leegwater PA, Lambooy LH, De Abreu RA, Bokkerink JP, van den Heuvel LP. DNA methylation patterns in the calcitonin gene region at first diagnosis and at relapse of acute lymphoblastic leukemia (ALL). Leukemia. 1997;11:971–978. 23. Yao J, Huang Q, Zhang XB, Fu WL. Promoter CpG methylation of oestrogen receptors in leukemia. Biosci Rep. 2009;29:211–216. 24. Melki JR, Vincent PC, Clark SJ. Cancer-specific region of hypermethylation identified within the HIC1 putative tumour suppressor gene in acute myeloid leukemia. Leukemia. 1999;13:877–883. 25. Quesnel B, Fenaux P. P15INK4b gene methylation and myelodysplastic syndromes. Leuk Lymphoma. 1999;35:437–443. 26. Tsellou E, Troungos C, Moschovi M, et al. Hypermethylation of CpG islands in the promoter region of the p15INK4B gene in childhood acute leukemia. Eur J Cancer. 2005;41:584–589. 27. Melki JR, Vincent PC, Clark SJ. Concurrent DNA hypermethylation of multiple genes in acute myeloid leukemia. Cancer Res. 1999;59:3730–3740. 28. Melki JR, Clark SJ. DNA methylation changes in leukemia. Semin Cancer Biol. 2002;12:347–357. 29. Hess CJ, Errami A, Berkhof J, et al. Concurrent methylation of promoters from tumor associated genes predicts outcome in acute myeloid leukemia. Leuk Lymphoma. 2008;49:1132–1141. 30. Garcia-Manero G, Bueso-Ramos C, Daniel J, Williamson J, Kantarjian HM, Issa JP. DNA methylation patterns at relapse in adult acute lymphocytic leukemia. Clin Cancer Res. 2002;8:1897–1903. 31. Roman-Gomez J, Jimenez-Velasco A, Castillejo JA, et al. Promoter hypermethylation of cancer-related genes: a strong independent prognostic factor in acute lymphoblastic leukemia. Blood. 2004;104:2492–2498. 32. Scholz C, Nimmrich I, Burger M, et al. Distinction of acute lymphoblastic leukemia from acute myeloid leukemia through microarray-based DNA methylation analysis. Ann Hematol. 2005;84:236–244. 33. Taylor KH, Pena-Hernandez KE, Davis JW, et al. Large-scale CpG methylation analysis identifies novel candidate genes and reveals methylation hotspots in acute lymphoblastic leukemia. Cancer Res. 2007;67:2617–2625. 34. Kuang SQ, Tong WG, Yang H, et al. Genome-wide identification of aberrantly methylated promoter associated CpG islands in acute lymphocytic leukemia. Leukemia. 2008;22:1529–1538. 35. Garcia-Manero G, Jeha S, Daniel J, et al. Aberrant DNA methylation in pediatric patients with acute lymphocytic leukemia. Cancer. 2003;97:695–702. 36. Davidsson J, Lilljebjorn H, Andersson A, et al. The DNA methylome of pediatric acute lymphoblastic leukemia. Hum Mol Genet. 2009;18:4054–4065. 37. Stumpel DJ, Schneider P, van Roon EH, et al. Specific promoter methylation identifies different subgroups of MLL-rearranged infant acute lymphoblastic leukemia, influences clinical outcome, and provides therapeutic options. Blood. 2009;114:5490–5498. 38. Melki JR, Warnecke P, Vincent PC, Clark SJ. Increased DNA methyltransferase expression in leukemia. Leukemia. 1998;12:311–316. 39. Rethmeier A, Aggerholm A, Olesen LH, et al. Promoter hypermethylation of the retinoic acid receptor beta2 gene is frequent in acute myeloid leukemia and associated with the presence of CBFbeta-MYH11 fusion transcripts. Br J Haematol. 2006;133:276–283.
14 Targeting Epigenetic Pathways in ALL
309
40. Glasow A, Barrett A, Petrie K, et al. DNA methylation-independent loss of RARA gene expression in acute myeloid leukemia. Blood. 2008;111:2374–2377. 41. Esteller M, Guo M, Moreno V, et al. Hypermethylation-associated Inactivation of the Cellular Retinol-Binding-Protein 1 Gene in Human Cancer. Cancer Res. 2002;62:5902–5905. 42. Hoemme C, Peerzada A, Behre G, et al. Chromatin modifications induced by PMLRARalpha repress critical targets in leukemogenesis as analyzed by ChIP-Chip. Blood. 2008;111:2887–2895. 43. Guenther MG, Jenner RG, Chevalier B, et al. Global and Hox-specific roles for the MLL1 methyltransferase. Proc Natl Acad Sci U S A. 2005;102:8603–8608. 44. Slany RK. The molecular biology of mixed lineage leukemia. Haematologica. 2009;94:984–993. 45. Armstrong SA, Staunton JE, Silverman LB, et al. MLL translocations specify a distinct gene expression profile that distinguishes a unique leukemia. Nat Genet. 2002;30:41–47. 46. Baylin SB. DNA methylation and gene silencing in cancer. Nat Clin Pract Oncol. 2005;2 Suppl 1:S4–11. 47. Rivard GE, Momparler RL, Demers J, et al. Phase I study on 5-aza-2′-deoxycytidine in children with acute leukemia. Leuk Res. 1981;5:453–462. 48. Momparler RL, Rivard GE, Gyger M. Clinical trial on 5-aza-2′-deoxycytidine in patients with acute leukemia. Pharmacol Ther. 1985;30:277–286. 49. Momparler RL. Pharmacology of 5-Aza-2′-deoxycytidine (decitabine). Semin Hematol. 2005;42:S9–16. 50. Silverman LR, McKenzie DR, Peterson BL, et al. Further analysis of trials with azacitidine in patients with myelodysplastic syndrome: studies 8421, 8921, and 9221 by the Cancer and Leukemia Group B. J Clin Oncol. 2006;24:3895–3903. 51. Silverman LR, Demakos EP, Peterson BL, et al. Randomized controlled trial of azacitidine in patients with the myelodysplastic syndrome: a study of the cancer and leukemia group B. J Clin Oncol. 2002;20:2429–2440. 52. Fenaux P, Mufti GJ, Hellstrom-Lindberg E, et al. Efficacy of azacitidine compared with that of conventional care regimens in the treatment of higher-risk myelodysplastic syndromes: a randomised, open-label, phase III study. Lancet Oncol. 2009;10:223–232. 53. Flotho C, Claus R, Batz C, et al. The DNA methyltransferase inhibitors azacitidine, decitabine and zebularine exert differential effects on cancer gene expression in acute myeloid leukemia cells. Leukemia. 2009;23:1019–1028. 54. Kantarjian H, Issa JP, Rosenfeld CS, et al. Decitabine improves patient outcomes in myelodysplastic syndromes: results of a phase III randomized study. Cancer. 2006;106:1794–1803. 55. Lemaire M, Chabot GG, Raynal NJ, et al. Importance of dose-schedule of 5-aza-2′-deoxycy tidine for epigenetic therapy of cancer. BMC Cancer. 2008;8:128. 56. Khan R, Aggerholm A, Hokland P, Hassan M, Hellstrom-Lindberg E. A pharmacodynamic study of 5-azacytidine in the P39 cell line. Exp Hematol. 2006;34:35–43. 57. Oki Y, Aoki E, Issa JP. Decitabine--bedside to bench. Crit Rev Oncol Hematol. 2007;61:140–152. 58. Lubbert M, Minden M. Decitabine in acute myeloid leukemia. Semin Hematol. 2005;42:S38–42. 59. Cashen AF, Schiller GJ, O’Donnell MR, Dipersio JF. Multicenter, Phase II Study of Decitabine for the First-Line Treatment of Older Patients With Acute Myeloid Leukemia. J Clin Oncol. 2010; 28: 556–61. 60. Friedman AD. Leukemogenesis by CBF oncoproteins. Leukemia. 1999;13:1932–1942. 61. Tenen DG, Hromas R, Licht JD, Zhang DE. Transcription factors, normal myeloid development, and leukemia. Blood. 1997;90:489–519. 62. Klisovic MI, Maghraby EA, Parthun MR, et al. Depsipeptide (FR 901228) promotes histone acetylation, gene transcription, apoptosis and its activity is enhanced by DNA methyltransferase inhibitors in AML1/ETO-positive leukemic cells. Leukemia. 2003;17:350–358. 63. Grignani F, De Matteis S, Nervi C, et al. Fusion proteins of the retinoic acid receptor-alpha recruit histone deacetylase in promyelocytic leukemia. Nature. 1998;391:815–818. 64. Lin RJ, Nagy L, Inoue S, Shao W, Miller WH, Jr., Evans RM. Role of the histone deacetylase complex in acute promyelocytic leukemia. Nature. 1998;391:811–814.
310
P. Kearns
65. Heibert SW, Lutterbach B, Durst K, et al. Mechanisms of transcriptional repression by the t(8;21)-, t(12;21)-, and inv(16)-encoded fusion proteins. Cancer Chemother Pharmacol. 2001;48 Suppl 1:S31–34. 66. Whitman SP, Liu S, Vukosavljevic T, et al. The MLL partial tandem duplication: evidence for recessive gain-of-function in acute myeloid leukemia identifies a novel patient subgroup for molecular-targeted therapy. Blood. 2005;106:345–352. 67. Dorrance AM, Liu S, Yuan W, et al. Mll partial tandem duplication induces aberrant Hox expression in vivo via specific epigenetic alterations. J Clin Invest. 2006;116:2707–2716. 68. Schafer ES, Irizarry R, Negi S, et al. Promoter Hypermethylation in MLL-r Leukemia: Biology and Therapeutic Targeting. ASH Annual Meeting Abstracts. 2009;114:3472. 69. Yanez L, Bermudez A, Richard C, Bureo E, Iriondo A. Successful induction therapy with decitabine in refractory childhood acute lymphoblastic leukemia. Leukemia. 2009;23:1342–1343. 70. Marks PA, Richon VM, Miller T, Kelly WK. Histone deacetylase inhibitors. Adv Cancer Res. 2004;91:137–168. 71. Gojo I, Jiemjit A, Trepel JB, et al. Phase 1 and pharmacologic study of MS-275, a histone deacetylase inhibitor, in adults with refractory and relapsed acute leukemias. Blood. 2007;109:2781–2790. 72. Giles F, Fischer T, Cortes J, et al. A phase I study of intravenous LBH589, a novel cinnamic hydroxamic acid analogue histone deacetylase inhibitor, in patients with refractory hematologic malignancies. Clin Cancer Res. 2006;12:4628–4635. 73. Garcia-Manero G, Assouline S, Cortes J, et al. Phase 1 study of the oral isotype specific histone deacetylase inhibitor MGCD0103 in leukemia. Blood. 2008;112:981–989. 74. Cameron EE, Bachman KE, Myohanen S, Herman JG, Baylin SB. Synergy of demethylation and histone deacetylase inhibition in the re-expression of genes silenced in cancer. Nat Genet. 1999;21:103–107. 75. Tong WG, Wei Y, Stevenson W, et al. Preclinical antileukemia activity of JNJ-26481585, a potent second-generation histone deacetylase inhibitor. Leuk Res. 2010; 34: 221–8. 76. Castoro RJ, Raynal NJ, Huang X, et al. DNA Methylation and Gene Expression Analysis in a Phase II Randomized Study of Decitabine Vs. Decitabine Plus Valproic Acid in MDS and AML. ASH Annual Meeting Abstracts. 2009;114:3808. 77. Ambros V. The functions of animal microRNAs. Nature. 2004;431:350–355. 78. Yendamuri S, Calin GA. The role of microRNA in human leukemia: a review. Leukemia. 2009;23:1257–1263. 79. Chuang JC, Jones PA. Epigenetics and microRNAs. Pediatr Res. 2007;61:24R–29R. 80. Saito Y, Friedman JM, Chihara Y, Egger G, Chuang JC, Liang G. Epigenetic therapy upregulates the tumor suppressor microRNA-126 and its host gene EGFL7 in human cancer cells. Biochem Biophys Res Commun. 2009;379:726–731. 81. Saito Y, Liang G, Egger G, et al. Specific activation of microRNA-127 with downregulation of the proto-oncogene BCL6 by chromatin-modifying drugs in human cancer cells. Cancer Cell. 2006;9:435–443. 82. Roman-Gomez J, Agirre X, Jimenez-Velasco A, et al. Epigenetic regulation of microRNAs in acute lymphoblastic leukemia. J Clin Oncol. 2009;27:1316–1322.
Chapter 15
Incorporating New Therapies into Frontline Protocols Paul S. Gaynon and Theresa M. Harned
15.1 Introduction Treatment outcomes of children and adolescents with newly diagnosed childhood acute lymphoblastic leukemia have steadily improved over the last 40 years – without the introduction of any new agents (see Table 15.1) [1–3]. However, apprehension mounts that we are reaching the limits of treatment intensity and exhausting our surprising capacity to manipulate usefully the sequences and schedules of conventional agents. Traditionally, conventional wisdom has dictated that new agents be both explored and tested definitively in relapse patients first – to the extent that any new agent has been tested at all. As the list of agents employed after relapse still looks much the same as the list of agents employed in primary therapy, new agents have had little impact thus far. Any progress to date derives from fortuitous modifications of platform therapy and better allocation of appropriate patients to hematopoietic stem cell transplant (HSCT). [4, 5]. While improvement in outcome for patients after relapse has proved elusive despite substantial progress in chemotherapy, supportive care, and HSCT [6, 7], treatment gains have been won for newly diagnosed patients with some regularity [8–13]. We hypothesize that candidate agents or multiagent treatment blocks that provide only transient response after relapse may contribute more readily to cure in newly diagnosed patients. Enhanced understanding of leukemia biology may identify potential treatment targets. Laboratory techniques to ascertain cytogenetic/molecular subsets [14], to assess minimal residual disease (MRD) [15–17] and to explore aberrant proliferative and apoptotic pathways [18–20], may help us identify patient subsets where extreme risk of relapse warrants early consideration of novel agents and facilitate patient grouping around some shared therapeutic susceptibility. Dexamethasone provides
P.S. Gaynon (*) Therapeutic Advances in Childhood Leukemia Consortium, Institute for Pediatric Clinical Research, Children’s Hospital Los Angeles, University of Southern California, Los Angeles, CA, USA e-mail:
[email protected] V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3_15, © Springer Science+Business Media, LLC 2011
311
312
P.S. Gaynon and T.M. Harned
Table 15.1 Improved outcomes in childhood acute lymphoblastic leukemia [1, 8, 10, 29] EFS Experimental Study Population Subset Intervention versus control (%) CCG-105 Intermediate ±Delayed risk intensification CCG-1882 Higher risk Slow responders Longer + stronger 75 intensification 55 CCG-1922 Standard risk Dexamethasone 77 85 CCG-1961 Higher risk Rapid responders Stronger 80 intensification 71 CCG-1991 Standard risk Slow responders Stronger 92.1 intensification 88.7 EFS event-free survival
one example where replacement of one drug by a second drug at an appropriate dose, both exploiting the vulnerability of ALL blasts to activation of the glucocorticoid receptor, has yielded improvements in event-free survival (EFS) for newly diagnosed patients [8, 21, 22]. The impact of imatinib in newly diagnosed Philadelphia chromosome-positive (Ph+) ALL [23] provides early evidence that a candidate agent that provides only occasional, transient complete remissions after relapse may contribute to cure for newly diagnosed patients with leukemias that share a vulnerability.
15.2 Why New Agents May More Readily Contribute to Cure in Primary Therapy: An Operational Definition of Relapse Einstein insisted that physics comes first and mathematics only follows [24]. We might suggest that biology comes first and computation follows. Relapse rate is a number. Relapse is a biological process. In this “operational” discussion, we shall attempt to order and name likely time-ordered steps without any claim to detailed molecular understanding. ALL is a heterogeneous disease, and the various mechanisms of treatment failure are likely diverse, multiple, and often redundant. A number of blasts persist at the end of induction (Fig. 15.1). Most newly diagnosed ALL patients achieve morphologic remissions after 4 weeks of treatment with fewer than 5% marrow blasts by conventional microscopy and with recovery of peripheral blood counts. Many or even most are MRD-negative with no detectable blasts by flow cytometry or by PCR at a sensitivity of 10−4 [15–17]. However, if treatment stops after induction, leukemia invariably returns. Some number of blasts must persist at the end of induction in all cases, whether the
15 Incorporating New Therapies into Frontline Protocols
313
2nd Relapse
1st Relapse/ Treatment
CR2
Remission Treatment
CR1
Diagnosis Fig. 15.1 Evolution of resistance in a leukemic blast population Sensitive blasts (open circle) comprise the predominate population at presentation. Resistant blasts ( filled circle), a small minority at presentation escape therapy and predominate at relapse. Refractory blasts ( filled circle), absent at diagnosis, are first present at relapse where they comprise a small minority but ultimately predominate and result in the patient’s demise
number is above or below our threshold of detection. At some point, distribution may be anatomically heterogeneous and no longer assessable by routine bone marrow sampling [25]. ALL is oligoclonal at presentation. Treatment selects the most resistant subclones. Choi et al. find that one clone, defined by a specific immunoglobulin heavychain rearrangement, may predominate at presentation and a second clone, defined by a distinct if related specific immunoglobulin heavy-chain rearrangement, predominates at end induction or relapse [26]. Not all subclones may be marked by a distinctive immunoglobulin or T-receptor rearrangement. Mullighan et al. find that the predominant cell at relapse, in most cases, may be traced back via comparative genomic hybridization to the predominant clone at presentation or to a shared ancestral clone [20]. Sensitive subclones are progressively eradicated through continued treatment. The clone responsible for relapse must persist through initial therapy, with treatmentinduced cell death eventually overbalanced by cell proliferation. Molecularly, treatment failure is linked to increased proliferation and/or decreased cell death [18]. Blast proliferation, whether balanced by treatment-induced cell kill or escaping treatment and leading to relapse, provides an opportunity for further evolution and genetic or epigenetic change to allow further decreased cell death and/or increased proliferation. A smaller resistant clone at presentation may delay relapse. Henderson et al. have linked time to relapse with the rarity of the relapse clone at presentation [27]. “Stronger” postinduction intensification has decreased the incidence of
314
P.S. Gaynon and T.M. Harned
relapse – likely by altering the balance between blast proliferation and blast death – but curiously extending conventional intensification beyond 6 months has had no impact on outcome [28, 29]. At relapse, remission reinduction rates remain high. Numerically, most blasts retain some sensitivity to conventional chemotherapy, but the leukemic population has evolved toward greater resistance, and MRD “negativity” is rare in second remission, especially when the duration of first remission is less than 36 months [30, 31]. Reaccumulation of blasts at relapse provides an opportunity for conception of a still covert, yet more resistant subclone. Survival after relapse depends on presenting characteristics such as age, WBC, and immunophenotype and site and timing of relapse [6, 7]. However, that survival appears independent of prior therapy. Freyer et al. compared survival after relapse for patients on the more and less effective arms of CCG-1961. As most trials show EFS gains exceeding survival gains, patients relapsing after a less effective therapy were expected to have better survival than the smaller number of patients relapsing after a more effective therapy. Such was not the case. Survival was identical for the 163 patients relapsing despite a less intensive therapy and the 107 patients relapsing after a more intensive therapy [31]. When blasts have lesser vulnerability to treatment than the host, they may be said to be refractory. When treatment is given and blast counts fall but recover before peripheral blood elements and/or the marrow remains replaced with blasts, the blasts may be said to be refractory. Patients die from refractory leukemia. With subsequent relapses, resistance increases further until no useful clinical response is evident, i.e., refractoriness to all known chemotherapy, should supportive care be adequate to support increasingly ill patients through increasingly aggressive treatment. The refractory subclone is usually present only at relapse, where it may be numerically rare. As modest modifications of primary chemotherapy have enhanced outcomes repeatedly, patients rarely have an occult refractory subclone at presentation. By definition, the refractory subclone would not be vulnerable to conventional therapies. The refractory clone likely arises “late” as the clone is usually small at first relapse, and most first-relapse patients and many second-relapse patients still achieve remission. One can imagine a novel agent that may alter the therapeutic index between an otherwise refractory blast and host. Such an agent, though elusive, remains a desired objective. Secondary resistance is likely different than primary resistance. Remission reinduction rates depend on duration of prior remission with no hint that any one specific treatment helps any one patient subset, with the exception of nelarabine [32, 33] in T-cell disease. However, review of a variety of putatively non-crossresistant multiagent combinations shows very similar remission induction rates in second and subsequent relapse (Table 15.2) [9]. Two corollaries follow. Generally, chemotherapeutic agents – alkylators, antibiotics, antimetabolites, and glucocorticosteroids – work through apoptosis. Secondary resistance in the clinic may arise from defects in shared cell death pathways rather
15 Incorporating New Therapies into Frontline Protocols
315
Table 15.2 Complete remission after second or subsequent relapse [9, 82, 83] Trial Complete remission rate POG 8866 30/74 (41%) Vincristine, prednisone, l-asparaginase POG 9160 30/82 (37%) Idarubicin, cytarabine High-dose cytarabine, l-asparaginase 22/52 (42%) High-dose cytarabine, l-asparaginase 8/22 (36%) Ifosfamide, etoposide 8/20 (40%) Topotecan, vinorelbine, thiotepa, dexamethasone, 6/17 (35%) and gemcitabine Vindesine prednisone, l-asparaginase 30/69 (43%) Cytarabine mitoxantrone 8/14 (57%) Totals 142/350 (41%)
than changes in drug retention, drug activation, or drug target activation. Agents that restore these pathways would seem to be of highest priority. Second, as was the case with mature B-cell leukemia [34], T-cell ALL [35], and now perhaps Ph+ ALL [23], newly diagnosed patients – lacking the refractory subclone – may be more available for cure than relapse patients. No prior improvement in outcome was noted in the treatment of relapsed mature B-cell leukemia and T-cell ALL. One may hypothesize that agents and combinations that enhance response in relapse may enhance cure when employed for newly diagnosed patients, prior to the existence of the refractory clone.
15.3 Right Drug for the Right Target for the Right Patient Choosing the right novel drug is the most pressing challenge. Most candidate agents fail, despite sincere enthusiasm and urgent clinical need. Only the most promising candidate agents move from the laboratory to the clinic, and of these, only one in ten is licensed (see Fig. 15.2a) [36]. New agents emerge with idiosyncratic arrays of preclinical assays, often with no positive control (a known clinically active drug) or negative control (known clinically inactive drug). Preclinical data are often incompletely obtained and spottily reported. Comparisons of various agents in a class – usually controlled by different pharmaceutical companies - are rare. No in vitro or xenograft assay has been validated to predict disease-specific clinical activity in humans [37, 38]. However, the preponderance of activity may have importance. The more models and dose levels showing drug effect, the more likely is clinical value. Negative trials are as important as positive trials. Work by the Pediatric Preclinical Testing Consortium is depicted in Table 15.3 [39, 40]. Various agents are studied against a battery of pediatric cell lines in a mouse xenograft model. Vincristine and cyclophosphamide are given as positive
316
P.S. Gaynon and T.M. Harned
a
Identify Candidate Compound Pre-Clinical Screen in vitro in vivo
Explore Mechanism of Action
Clinical Trials
b Identify Disease-Specific Mechanism(s) of Resistance
Identify Candidate Agent
Identify Population sharing Mechanism
Preclinical Efficacy, Specificity Toxicity Pharmacokinetics
Clinical Trials
Fig. 15.2 Alternative schemes for identification of new agents Table 15.3 Pediatric preclinical consortium [40, 47, 84–88] AZD Cell line Vcr Cpm DDP Bort 2171 Rapa ALL-2 10 0 0 0 2 ALL-3 10 2 0 1 2 ALL-4 9 10 2 0 1 0 ALL-7 10 10 2 0 5 ALL-8 8 10 2 8 1 6 ALL-16 7 10 5 6 10 ALL-17 10 8 2 9 0 2 ALL-19 10 10 0 6 0 2 Median 10 10 2 6 0 2
Dasat 0 4 8 0 2 0 2 2
ABT 263 0 2 10 10 2 8 5
Median 0 2 2 5 7 8.5 2 4 3.5
Responses 0
2 4 6 8 10 Sustained Prolonged Prolonged complete survival Partial Complete survival response >150% response Response < 150% No response T-cell: ALL-8, ALL-16 B-cell precursor: ALL-2, ALL-3, ALL-4 (PH+), ALL-7, ALL-17, ALL-19 Diagnosis: ALL-3, ALL-4, ALL-7, ALL-16, ALL-17 Relapse: ALL-2, ALL-8, ALL-19 Vcr vincristine; Cpm cyclophosphamide; DDP cisplatin; Bort bortezomib; Rpa rapamycin; Dasat dasatinib
controls and cisplatin as a negative control for pediatric ALL. Data for five drugs are depicted. Dasatinib shows most activity in a Ph+ ALL cell line. ABT 263, a bcl-2 inhibitor, shows most activity in T-cell ALL lines. Curiously, none of the new agents shows as much activity as our old standbys, cyclophosphamide and vincristine, themselves inadequate for relapse patients.
15 Incorporating New Therapies into Frontline Protocols
317
Clinical value in other cancers and populations (i.e., “paraclinical” data) assures continued drug availability and improve prospects. Agents derived from an understanding of leukemia biology rather than high throughput screening in unvalidated “models” may prove more productive (see Fig. 15.2b). An understanding of the critical role of glucocorticoids in ALL and the pharmacokinetic advantages of dexamethasone [41] led to successful trials in newly diagnosed patients. An understanding of the role of BCR-ABL in Ph+ leukemia led to imatinib [42, 43]. In successive patient cohorts, the AALL0031 trial added more days of imatinib treatment to an intensive antileukemic platform for newly diagnosed patients [23]. While the aggressive platform alone showed no advantage over past efforts, the addition of imatinib resulted in a striking improvement in outcome. An understanding of imatinib resistance and tyrosine kinase domain mutations in chronic myelogenous leukemia next led to dasatinib [44, 45]. One looks to the day when some understanding of the molecular mechanisms of treatment failure will guide selection of agents for known, validated targets, rather than indiscriminate screening of candidate agents in various unvalidated “models” with secondary exploration of likely mechanisms for the apparently active compounds. If a “model” is not validated, what does it model? Simultaneously, means are required to identify the patient population where a specific resistance mechanism operates. Observations to date point to the simultaneous emergence of resistance to a variety of putatively non-cross-resistant agents [9]. However, agents with dissimilar mechanisms of action likely share common cell death pathways, which may prove to be the crucial bottleneck. Agents that enhance apoptosis by inhibiting BCL-2 [46, 47] or survivin [48], by lowering glutathione [49–51], by raising ceramide [52, 53], or by directly affecting mitochondria [54] may deserve special interest. Experience with targeted agents to date stresses the importance of target validation [55]. A cancer may proliferate independently of an overexpressed gene. Targeted agents often have effects beyond their putative targeted action [56]. Resistance mechanisms may be diverse and several distinct mechanisms may operate in various subclones in the same patient, above or below any threshold of detection. These considerations again stress that success may be more likely for patients with newly diagnosed disease than for patients with further evolved relapsed disease. Knowledge of drug pharmacology is also critical. Dexamethasone is less protein bound than prednisolone and provides better CNS levels [41]. Penetration may also be better for other, less well-recognized compartments [57]. Specific activity has been controversial but the most recent data suggest that dexamethasone is about 6–7 times more active than prednisolone in vitro in keeping with the commonly accepted 6–7:1 clinical activity ratio [58]. Five randomized trials are depicted in Table 15.4. Two trials with prednisone and dexamethasone ratios of 7.5:1 and 10:1 show no advantage for dexamethasone [59, 60]. Three trials with ratios of 6.7, 6.1, and 6 find a statistically significant EFS advantage for dexamethasone [8, 21, 22]. BFM reports that replacement of 21 days of prednisone with dexamethasone
318 Table 15.4 Dexamethasone versus prednisone in childhood 21, 22, 59, 60] Dexamethasone Prednisone Study Population n mg/m2 CCG SR 1,060 6 40 Japan SR 231 8/6 IR 128 60/40 MRC All 1,603 6.5 40 BFM All 3,655 10 60 6 EORTC All 1,853 60 SR standard risk; IR intermediate risk
P.S. Gaynon and T.M. Harned acute lymphoblastric leukemia [8,
Ratio 6.7 7.5 6.1 6 10
EFS (%) 85 77 84 81 84 76 84 79 82 82
P value 0.002 0.625 0.01 0.008 0.94
resulted in a 1/3 reduction in relapses and an improvement in EFS, albeit with increased mortality (1.4 vs. 0.5%) in induction. Results show no advantage in survival to date except in the T-cell subset. A change in one agent in induction resulted in an improvement in 6-year EFS, a paradigm for the potential of “new” agents in newly diagnosed patients.
15.4 Right Accompanying Therapy, Right Dose, and Right Schedule in the Right Trial A useful test of an agent requires the proper agent be tested in the proper population and the proper dose and schedule with the proper accompanying therapy (Table 15.5). Proper controls are also mandatory. Phase III represents a candidate agent’s critical definitive test. A once-failed agent may not get a second chance. Drugs are used in combination in the treatment of childhood ALL. A drug may be active alone but adds nothing in combination. A drug may have no single-agent activity, e.g., leukovorin, but contribute when used in the proper sequence and combination, e.g., leukovorin fluorouracil [61]. Novel treatment elements must enhance or replace present elements. Ifosfamide has striking single activity in relapsed rhabdomyosarcoma [62] but offered no advantage when substituted for cyclophosphamide [63]. The best use of an agent requires the “right stuff” (Table 15.5). Ideally, some disease-specific activity data seem important prior to committing to a multiyear “definitive” randomized trial. The best therapy is the proper control in phase III trials. POG 9404 added methotrexate 5 g/m [2] to a DFCI platform for T-cell ALL [64]. Whole-brain irradiation was prudently delayed for 17 weeks in the experimental arm to allow for intravenous methotrexate and avoid leukoencephalopathy. However, radiation was
15 Incorporating New Therapies into Frontline Protocols
319
Table 15.5 The right stuff [9] Right drug Right target Right disease Right population Right context Right schedule Right dose
also delayed in the bowdlerized “control” regimen with no clinical imperative, and outcomes were 15% points worse than the DFCI experience. In clinical trials, the experimental arm often elicits hope and enthusiasm. The control arm may lack zealous defenders and lie vulnerable to various supposedly “harmless” untested modifications. The Children’s Cancer Group first tested a “novel” agent, dexamethasone, in CCG-1922, where we substituted dexamethasone for prednisone in induction and maintenance [8]. The control arm kept dexamethasone in delayed intensification as the best therapy despite appeals for a “cleaner” pure dexamethasone versus pure prednisone comparison. The critical study question was the value of dexamethasone in induction and maintenance in the context of the best known treatment, not pure dexamethasone versus pure prednisone as our best therapy already included dexamethasone in delayed intensification. With better outcomes, the number of patients who must be subjected to a new intervention to earn the opportunity to cure one additional patient, i.e., the number needed to treat (NNT) has increased geometrically. Many novel interventions fail but increase morbidity with no compensatory gain [65]. Identification of patient subsets at higher risk of treatment failure balances potential harms and benefits and decreases NNT. Treatment intensity now approaches the limits of feasibility and insights into the biology of leukemia have led to promising candidate agents targeted to specific leukemic subsets. In parallel with identification of agents to address the specific mechanisms of treatment failure, we must develop methods to determine the patients in which these mechanisms likely operate. More effective, not necessarily more intensive, therapy is the goal.
15.5 Clinical Trials in the Children’s Oncology Group In the Children’s Cancer Group, dexamethasone was arguably the first new agent tested in newly diagnosed patients [8]. An immunotoxin, B43-PAP [66, 67], was examined briefly in newly diagnosed higher risk B-precursor ALL with a slow initial response to therapy on CCG-1961 [12]. Drug supply problems aborted the study and pose a challenge for any novel agent.
320
P.S. Gaynon and T.M. Harned
Table 15.6 New agents in clinical trial in primary treatment of childhood ALL Preclinical Human toxicity Human “paraclinical”a Disease-specific In vitro Xenograft data activity data Agent activity data Dasatinib + + Adult + + Nelarabine + Pediatric adult + + Lestaurtinib + Adult + Clofarabine + Pediatric adult + + a Paraclinical-response data in other human cancers and populations
The Children’s Oncology Group, successor to CCG and POG, studied imatinib successfully in newly diagnosed Ph+ ALL [23]. The COG ALL Committee has followed up on the success with imatinib and pursued novel agents in newly diagnosed patients with energy (Table 15.6). Cytogenetic/molecular subsets such as Ph+ ALL and infants with MLL gene rearrangements and higher-risk (HR) T-cell and B-precursor subsets have substantial risk of relapse and potential shared vulnerability. AALL0622 substitutes dasatinib [45, 68] for imatinib on an intensive AALL0031 chemotherapy platform for Ph+ ALL. AALL0434 adds nelarabine [32, 33] to the COG “augmented BFM platform” [29]. AALL0631 adds lestaurtinib [69, 70] to an intensive chemotherapy platform [71] for infants with MLL rearrangements. AALL0622 Ph+ ALL trial employs historical controls, while other trials include randomization after completion of a feasibility phase. Future plans include a trial of clofarabine, cyclophosphamide, and etoposide [72, 73] in newly diagnosed HR B-precursor ALL patients, who are MRD-positive at the end of induction in the successor to the current trial, AALL0232.
15.5.1 Dasatinib Dasatinib is a second-generation tyrosine kinase inhibitor with known activity in CML and Ph+ ALL [74]. As with imatinib, responses after relapse are common but transient. Dasatinib has greater tyrosine kinase activity than imatinib. Only the T315I mutation is resistant to dasatinib. Ottman et al. found that the single agent dasatinib induced complete cytogenetic remission in 21/36 imatinib-resistant or -intolerant CML patients [68]. None of the 6 patients with the kinase domain mutation, T315I, had a response. AALL0622 substitutes dasatinib for imatinib on the AALL0031 platform.
15.5.2 Lestaurtinib Lestaurtinib (CEP-701) is an orally active small-molecule inhibitor of several receptor tyrosine kinases, with specificity for the tropomyosin receptor kinases TrkA, TrkB, and TrkC and Fms-like tyrosine kinase 3 (FLT3) [69, 75]. AALL0631
15 Incorporating New Therapies into Frontline Protocols
321
rests heavily on in vitro data showing major impact on infant ALL cell lines and patient samples with MLL rearrangements where FLT3 is consistently overexpressed though rarely mutated [76]. Cytotoxicity depended strongly on schedule. Lestaurtinib following chemotherapy had synergistic effect, while simultaneous administration was additive, and chemotherapy following lestaurtinib was antagonistic. Regrettably, no clinical data in adult ALL or infant ALL are available. Preliminary reports show a whiff of value for lestaurtinib in adult AML with FLT3 overexpression and internal tandem duplication [77].
15.5.3 Nelarabine Nelarabine, an ara-G prodrug, provided a 50% CR rate in first relapse T-cell ALL [32]. Curiously, little or no preclinical or clinical data are available for nelarabine in combination. A pilot trial inserted nelarabine in a BFM platform with promising preliminary results [78]. On AALL0434, “naked” nelarabine is being inserted on the augmented BFM platform [29] prior to each cyclophosphamide, cytarabine, and thiopurine pulse and in early maintenance for higher risk T-cell patients, with modest erosion of platform intensity.
15.5.4 Clofarabine The Children’s Oncology Group trials, CCG-1882, CCG-1961, and CCG-1991 highlight the importance of the first 6 months of post induction intensification [10, 11, 29]. Higher risk patients receive three 4-week blocks including cyclophosphamide, cytarabine, and thiopurine corresponding to BFM “Protocol IB and IIB” in the first 7 months of therapy [78, 79]. These blocks were first introduced in the BFM 70 study. CCG-105 randomized standard-risk patients to receive or not to receive Protocol IB and found no advantage [13]. An AIEOP trial omitted Ib and noted worsening outcomes [80]. A candidate regimen to test against Ib/IIb has long been sought. Clofarabine is a third-generation adenosine analog. In addition to effects on ribonucleotide reduction, incorporation into DNA and RNA, and interference with DNA repair, clofarabine also targets the mitochondria directly, inducing apoptosis in both resting and dividing cells [81]. The combination of clofarabine, cyclophosphamide, and etoposide has a substantial CR rate in second and subsequent relapse [72] and seems poised for a test to see whether this block may replace the BFM Ib and IIb blocks in newly diagnosed patients. Fine-tuning of this “block” from a “remission or bust” induction block to one component of multiblock therapy where subsequent therapy requires prompt predictable count recovery is critical.
322
P.S. Gaynon and T.M. Harned
15.6 Conclusion Treatment is the most important prognostic factor. As with T-cell and mature B-cell ALL, and now possibly with Ph+ ALL, better primary treatment offers a better prospect of cure for former high-risk subsets. Leukemia is oligoclonal and evolves with each recurrence. Predominant leukemia blast population may hide rare subclones that ultimately determine clinical outcome. Understanding of the disease-specific mechanism(s) of treatment failure may provide insight into their remedies. As in the past, opportunities for cure are more likely in newly diagnosed than relapse patients and will require both boldness as well as careful integration of novel agents and combinations into current treatment platforms.
References 1. Gaynon PS, Trigg ME, Heerema NA, Sensel MG, Sather HN, Hammond GD, et al. Children’s Cancer Group trials in childhood acute lymphoblastic leukemia: 1983-1995. Leukemia. 2000 Dec;14(12):2223–33. 2. Schrappe M, Reiter A, Zimmermann M, Harbott J, Ludwig WD, Henze G, et al. Long-term results of four consecutive trials in childhood ALL performed by the ALL-BFM study group from 1981 to 1995. Berlin-Frankfurt-Munster. Leukemia. 2000 Dec;14(12):2205–22. 3. Silverman LB, Declerck L, Gelber RD, Dalton VK, Asselin BL, Barr RD, et al. Results of Dana-Farber Cancer Institute Consortium protocols for children with newly diagnosed acute lymphoblastic leukemia (1981-1995). Leukemia. 2000 Dec;14(12):2247–56. 4. Einsiedel HG, von Stackelberg A, Hartmann R, Fengler R, Schrappe M, Janka-Schaub G, et al. Long-term outcome in children with relapsed ALL by risk-stratified salvage therapy: results of trial acute lymphoblastic leukemia-relapse study of the Berlin-Frankfurt-Munster Group 87. J Clin Oncol. 2005 Nov 1;23(31):7942–50. 5. Roy A, Cargill A, Love S, Moorman AV, Stoneham S, Lim A, et al. Outcome after first relapse in childhood acute lymphoblastic leukemia - lessons from the United Kingdom R2 trial. British journal of haematology. 2005 Jul;130(1):67–75. 6. Gaynon PS, Qu RP, Chappell RJ, Willoughby ML, Tubergen DG, Steinherz PG, et al. Survival after relapse in childhood acute lymphoblastic leukemia: impact of site and time to first relapse – the Children’s Cancer Group Experience. Cancer. 1998 Apr 1;82(7):1387–95. 7. Nguyen K, Devidas M, Cheng SC, La M, Raetz EA, Carroll WL, et al. Factors influencing survival after relapse from acute lymphoblastic leukemia: a Children’s Oncology Group study. Leukemia. 2008 Dec;22(12):2142–50. 8. Bostrom BC, Sensel MR, Sather HN, Gaynon PS, La MK, Johnston K, et al. Dexamethasone versus prednisone and daily oral versus weekly intravenous mercaptopurine for patients with standard-risk acute lymphoblastic leukemia: a report from the Children’s Cancer Group. Blood. 2003 May 15;101(10):3809–17. 9. Gaynon PS. Childhood acute lymphoblastic leukemia and relapse. British journal of haematology. 2005 Dec;131(5):579–87. 10. Matloub Y, Bostrom BC, Hunger SP, Angiolillo AL, Cole C, Thomson B, Devidas M, Heerema NA, Mei KL, Buckley PJ, Carroll WL, Winick N, Sather H, Nachman JB. & Gaynon PS. Escalating Dose Intravenous Methotrexate without Leukovorin Rescue during Interim Maintenance Is Superior to Oral Methotrexate for Children with Standard Risk Acute Lymphoblastic Leukemia (SR-ALL): a Children’s oncology Group Study 1991. Blood. 2008; 112, 9–10.
15 Incorporating New Therapies into Frontline Protocols
323
11. Nachman JB, Sather HN, Sensel MG, Trigg ME, Cherlow JM, Lukens JN, et al. Augmented post-induction therapy for children with high-risk acute lymphoblastic leukemia and a slow response to initial therapy. The New England journal of medicine. 1998 Jun 4;338(23):1663–71. 12. Seibel N, Krailo M., O’Neill K, Franklin J, Uckun F. & Reaman, G. Phase I Study of B43-PAP Immunotoxin in combination with Standard 4-Drug Induction for Patients wiht CD 19+ Aacute Lymphoblastic Leukemia (ALL) in Relapse, a Children’s Cancer Group Study. Pediatric Research. 1999;45, 775. 13. Tubergen DG, Gilchrist GS, O’Brien RT, Coccia PF, Sather HN, Waskerwitz MJ, et al. Improved outcome with delayed intensification for children with acute lymphoblastic leukemia and intermediate presenting features: a Childrens Cancer Group phase III trial. J Clin Oncol. 1993 Mar;11(3):527–37. 14. Schultz KR, Pullen DJ, Sather HN, Shuster JJ, Devidas M, Borowitz MJ, et al. Risk- and response-based classification of childhood B-precursor acute lymphoblastic leukemia: a combined analysis of prognostic markers from the Pediatric Oncology Group (POG) and Children’s Cancer Group (CCG). Blood. 2007 Feb 1;109(3):926–35. 15. Borowitz MJ, Devidas M, Hunger SP, Bowman WP, Carroll AJ, Carroll WL, et al. Clinical significance of minimal residual disease in childhood acute lymphoblastic leukemia and its relationship to other prognostic factors: a Children’s Oncology Group study. Blood. 2008 Jun 15;111(12):5477–85. 16. Cave H, van der Werff ten Bosch J, Suciu S, Guidal C, Waterkeyn C, Otten J, et al. Clinical significance of minimal residual disease in childhood acute lymphoblastic leukemia. European Organization for Research and Treatment of Cancer–Childhood Leukemia Cooperative Group. The New England journal of medicine. 1998 Aug 27;339(9):591–8. 17. van Dongen JJ, Seriu T, Panzer-Grumayer ER, Biondi A, Pongers-Willemse MJ, Corral L, et al. Prognostic value of minimal residual disease in acute lymphoblastic leukemia in childhood. Lancet. 1998 Nov 28;352(9142):1731–8. 18. Bhojwani D, Kang H, Moskowitz NP, Min DJ, Lee H, Potter JW, et al. Biologic pathways associated with relapse in childhood acute lymphoblastic leukemia: a Children’s Oncology Group study. Blood. 2006 Jul 15;108(2):711–7. 19. Mullighan CG, Goorha S, Radtke I, Miller CB, Coustan-Smith E, Dalton JD, et al. Genomewide analysis of genetic alterations in acute lymphoblastic leukemia. Nature. 2007 Apr 12;446(7137):758–64. 20. Mullighan CG, Phillips LA, Su X, Ma J, Miller CB, Shurtleff SA, et al. Genomic analysis of the clonal origins of relapsed acute lymphoblastic leukemia. Science (New York, NY). 2008 Nov 28;322(5906):1377–80. 21. Mitchell CD, Richards SM, Kinsey SE, Lilleyman J, Vora A, Eden TO. Benefit of dexamethasone compared with prednisolone for childhood acute lymphoblastic leukemia: results of the UK Medical Research Council ALL97 randomized trial. British journal of haematology. 2005 Jun;129(6):734–45. 22. Schrappe M, Zimmermann M, Moricke A, Mann G, Valsecchi MG, Bartram CR, Biondi A, Panzer-Grumayer R, Schrauder A, Locatelli F, Reiter A, Basso G, Niggli F, Arico M. & Conter V. Dexamethasone in Induction Can Eliminate One Third of All Relapses in Childhood Acute Lymphoblastic Leukemia (ALL): Results of an International Trial in 3655 Patients (Trial AIEOP-BFM ALL 2000). Blood. 2008;112, 9. 23. Schultz KR, Bowman WP, Slayton W, Aledo A, Devidas M, Sather H, Borowitz MJ, Davies SM, Trigg M, Pasut B, Jorstad D, Eslinger T, Burden LE, Wang C, Rutledge R, Gaynon PS, Carroll AJ, Heerema NA, Winick N, Hunger S, Carroll WL. & Camitta B. Improved Early Event Free Survival (EFS) in Children with Philapdelphia Chromosome-Positive (Ph+) Acute Lymphoblastic Leukemia (ALL) with Imatinib in Combination with High Dose Chemotherapy: Children’s Oncology Group (COG) Study AALL0031. Blood. 2007;110, 8a. 24. Isaacson W. Einstein: His Life and Universe. Simon & Schuster. 2007. 25. Hagenbeek A. & Martens ACM. Minimal Residual Disease in Acute Leukemia: Lessons Learned from Animal Models. Hematol Blood Trans. 1990;33, 31–35.
324
P.S. Gaynon and T.M. Harned
26. Choi S, Henderson MJ, Kwan E, Beesley AH, Sutton R, Bahar AY, et al. Relapse in children with acute lymphoblastic leukemia involving selection of a preexisting drug-resistant subclone. Blood. 2007 Jul 15;110(2):632–9. 27. Henderson MJ, Choi S, Beesley AH, Sutton R, Venn NC, Marshall GM, et al. Mechanism of relapse in pediatric acute lymphoblastic leukemia. Cell cycle (Georgetown, Tex.). 2008 May 15;7(10):1315–20 28. Matloub Y, Angiolillo A, Bostrom B, Hunger SP, Hachman J, Sather H, Carroll WL, Winick N, & Gaynon PS. Double Delayed Intensification (DDI) Is Equivalent to Single DI (SDI) in Children with National Cancer Institute (NCI) Standard-risk Acute Lymphoblatic Leukemia (SR-ALL) Treated on Children’s Cancer Group (CCG) Clinical Trial 1991 (CCG-1991). Blood. 2006;108, 47a. 29. Seibel NL, Steinherz PG, Sather HN, Nachman JB, Delaat C, Ettinger LJ, et al. Early postinduction intensification therapy improves survival for children and adolescents with high-risk acute lymphoblastic leukemia: a report from the Children’s Oncology Group. Blood. 2008 Mar 1;111(5):2548–55. 30. Eckert C, Biondi A, Seeger K, Cazzaniga G, Hartmann R, Beyermann B, et al. Prognostic value of minimal residual disease in relapsed childhood acute lymphoblastic leukemia. Lancet. 2001 Oct 13;358(9289):1239–41. 31. Raetz EA, Borowitz MJ, Devidas M, Linda SB, Hunger SP, Winick NJ, et al. Reinduction platform for children with first marrow relapse of acute lymphoblastic Leukemia: A Children’s Oncology Group Study [corrected]. J Clin Oncol. 2008 Aug 20;26(24):3971–8. 32. Berg SL, Blaney SM, Devidas M, Lampkin TA, Murgo A, Bernstein M, et al. Phase II study of nelarabine (compound 506U78) in children and young adults with refractory T-cell malignancies: a report from the Children’s Oncology Group. J Clin Oncol. 2005 May 20;23(15):3376–82. 33. Kisor DF. Nelarabine use in leukemias. Drugs Today (Barc). 2006 Jul;42(7):455–65. 34. Cairo MS, Gerrard M, Sposto R, Auperin A, Pinkerton CR, Michon J, et al. Results of a randomized international study of high-risk central nervous system B non-Hodgkin lymphoma and B acute lymphoblastic leukemia in children and adolescents. Blood. 2007 Apr 1;109(7):2736–43. 35. Goldberg JM, Silverman LB, Levy DE, Dalton VK, Gelber RD, Lehmann L, et al. Childhood T-cell acute lymphoblastic leukemia: the Dana-Farber Cancer Institute acute lymphoblastic leukemia consortium experience. J Clin Oncol. 2003 Oct 1;21(19):3616–22. 36. Kamb A, Wee S, Lengauer C. Why is cancer drug discovery so difficult? Nature reviews. 2007 Feb;6(2):115–20. 37. Johnson JI, Decker S, Zaharevitz D, Rubinstein LV, Venditti JM, Schepartz S, et al. Relationships between drug activity in NCI preclinical in vitro and in vivo models and early clinical trials. British journal of cancer. 2001 May 18;84(10):1424–31. 38. Suggitt M, Bibby MC. 50 years of preclinical anticancer drug screening: empirical to targetdriven approaches. Clin Cancer Res. 2005 Feb 1;11(3):971–81. 39. Houghton PJ, Adamson PC, Blaney S, Fine HA, Gorlick R, Haber M, et al. Testing of new agents in childhood cancer preclinical models: meeting summary. Clin Cancer Res. 2002 Dec;8(12):3646–57. 40. Houghton PJ, Morton CL, Tucker C, Payne D, Favours E, Cole C, et al. The pediatric preclinical testing program: description of models and early testing results. Pediatric blood & cancer. 2007 Dec;49(7):928–40. 41. Balis FM, Lester CM, Chrousos GP, Heideman RL, Poplack DG. Differences in cerebrospinal fluid penetration of corticosteroids: possible relationship to the prevention of meningeal leukemia. J Clin Oncol. 1987 Feb;5(2):202–7. 42. Druker BJ, Kantarjian H, Sawyers CL, Resta D, Reese SF, Peng B, Ford J. & Talpaz M. Activity of an ABL specific tyrosine kinase inhibitor in patients with BCR-ABL positive acute leukemias, including chronic myelogenous leukemia in blast crisis. Blood. 1999;94, 697a. 43. Druker BJ, Talpaz M, Resta D, Peng B, Buchdunger E, Ford J. & Sawyers CL. Clinical Efficacy and Safety of an ABL specific tyrosine kinase inhibitor as targeted thearpy for chronic myelogenous leukemia. Blood. 1999;94, 368a.
15 Incorporating New Therapies into Frontline Protocols
325
44. Hochhaus A. Management of Bcr-Abl-positive leukemias with dasatinib. Expert review of anticancer therapy. 2007 Nov;7(11):1529–36. 45. Sonmez M, Cobanoglu U, Ovali E, Omay SB. Use of dasatinib in the patient with Philadelphia chromosome-positive acute lymphoblastic leukemia with resistance to imatinib and renal failure. Journal of clinical pharmacy and therapeutics. 2008 Jun;33(3):329–30. 46. Kang MH, Kang YH, Szymanska B, Wilczynska-Kalak U, Sheard MA, Harned TM, et al. Activity of vincristine, L-ASP, and dexamethasone against acute lymphoblastic leukemia is enhanced by the BH3-mimetic ABT-737 in vitro and in vivo. Blood. 2007 Sep 15;110(6):2057–66. 47. Lock R, Carol H, Houghton PJ, Morton CL, Kolb EA, Gorlick R, et al. Initial testing (stage 1) of the BH3 mimetic ABT-263 by the pediatric preclinical testing program. Pediatric blood & cancer. 2008 Jun;50(6):1181–9. 48. Hansen JB, Fisker N, Westergaard M, Kjaerulff LS, Hansen HF, Thrue CA, et al. SPC3042: a proapoptotic survivin inhibitor. Molecular cancer therapeutics. 2008 Sep;7(9):2736–45. 49. Friesen C, Kiess Y, Debatin KM. A critical role of glutathione in determining apoptosis sensitivity and resistance in leukemia cells. Cell death and differentiation. 2004 Jul;11 Suppl 1:S73–85. 50. Morales MC, Perez-Yarza G, Nieto-Rementeria N, Boyano MD, Jangi M, Atencia R, et al. Intracellular glutathione levels determine cell sensitivity to apoptosis induced by the antineoplasic agent N-(4-hydroxyphenyl) retinamide. Anticancer research. 2005 MayJun;25(3B):1945–51. 51. Yoshida A, Takemura H, Inoue H, Miyashita T, Ueda T. Inhibition of glutathione synthesis overcomes Bcl-2-mediated topoisomerase inhibitor resistance and induces nonapoptotic cell death via mitochondrial-independent pathway. Cancer research. 2006 Jun 1;66(11):5772–80. 52. Baran Y, Salas A, Senkal CE, Gunduz U, Bielawski J, Obeid LM, et al. Alterations of ceramide/ sphingosine 1-phosphate rheostat involved in the regulation of resistance to imatinib-induced apoptosis in K562 human chronic myeloid leukemia cells. The Journal of biological chemistry. 2007 Apr 13;282(15):10922–34. 53. Maurer BJ, Metelitsa LS, Seeger RC, Cabot MC, Reynolds CP. Increase of ceramide and induction of mixed apoptosis/necrosis by N-(4-hydroxyphenyl)- retinamide in neuroblastoma cell lines. Journal of the National Cancer Institute. 1999 Jul 7;91(13):1138–46. 54. Genini D, Adachi S, Chao Q, Rose DW, Carrera CJ, Cottam HB, et al. Deoxyadenosine analogs induce programmed cell death in chronic lymphocytic leukemia cells by damaging the DNA and by directly affecting the mitochondria. Blood. 2000 Nov 15;96(10):3537–43. 55. Benson JD, Chen YN, Cornell-Kennon SA, Dorsch M, Kim S, Leszczyniecka M, et al. Validating cancer drug targets. Nature. 2006 May 25;441(7092):451–6. 56. Zhukov NV, Tjulandin SA. Targeted therapy in the treatment of solid tumors: practice contradicts theory. Biochemistry. 2008 May;73(5):605–18. 57. Styczynski J, Koltan A, Wysocki M. Resistance to glucocorticoids in childhood acute lymphoblastic leukemia: impact of relationship between ex vivo sensitivity and in vivo concentration on risk factor analysis. Neoplasma. 2006;53(2):168–73. 58. Haarman EG, Kaspers GJ, Pieters R, Rottier MM, Veerman AJ. Circumvention of glucocorticoid resistance in childhood leukemia. Leukemia research. 2008 Sep;32(9):1417–23. 59. Bertrand Y, Suciu S, Benoit Y, Robert A, Nelken B, Uyttebroeck L, Lutz P, Ferster A, Baruchel A, Norton L, Margueritte G, Rohrlich PS, Munzer M, Plantaz D, Millot F, Boutard P, Sirvent N, Hoyoux C, Baila L. & Otten J. Dexamethasone (DEX) (6 mg/sm/d) and Prednisolone (PRED) (60 mg/sm/d) in Induction Therapy of Childhood ALL Are Equally Effective: Results of the Second Interim Analysis of EORTC Trial 58951. Blood. 2008;112, 9. 60. Igarashi S, Manabe A, Ohara A, Kumagai M, Saito T, Okimoto Y, Kamijo T, Isoyama K, Kajiwara M, Sotomatsu M, Sugita K, Maeda M, Yabe H, Kinoshita A, Kaneko T, Hayashi Y, Ikuta K, Hanada R. & Tsuchida M. No advantage of dexamethasone over prednisolone for the outcome of standard- and intermediate-risk childhood acute lymphoblastic leukemia in the Tokyo Children’s Cancer Study Group L95-14 protocol. J Clin Oncol. 2005;23, 6489–6498. 61. Doroshow JH, Multhauf P, Leong L, Margolin K, Litchfield T, Akman S, et al. Prospective randomized comparison of fluorouracil versus fluorouracil and high-dose continuous infusion
326
P.S. Gaynon and T.M. Harned
leucovorin calcium for the treatment of advanced measurable colorectal cancer in patients previously unexposed to chemotherapy. J Clin Oncol. 1990 Mar;8(3):491–501. 62. Pappo AS, Etcubanas E, Santana VM, Rao BN, Kun LE, Fontanesi J, et al. A phase II trial of ifosfamide in previously untreated children and adolescents with unresectable rhabdomyosarcoma. Cancer. 1993 Mar 15;71(6):2119–25. 63. Crist WM, Anderson JR, Meza JL, Fryer C, Raney RB, Ruymann FB, et al. Intergroup rhabdomyosarcoma study-IV: results for patients with nonmetastatic disease. J Clin Oncol. 2001 Jun 15;19(12):3091–102. 64. Asselin B, Shuster J, Amylon M, Halperin E, Hutchison R, Lipshultz S. & Camitta B. Improved event-free survival (EFS) with high dose methotrexate (HDM) in T-cell lymphoblastic leukemia (T-ALL) and advanced lymphoblastic lymphoma (T-NHL): a Pediatric Oncology Group (POG) study. In: Proc ASCO. 2001; Vol. 20, p. 367. 65. Kumar A, Soares H, Wells R, Clarke M, Hozo I, Bleyer A, et al. Are experimental treatments for cancer in children superior to established treatments? Observational study of randomised controlled trials by the Children’s Oncology Group. BMJ (Clinical research ed.) 2005 Dec 3;331(7528):1295. 66. Ek O, Gaynon P, Zeren T, Chelstrom LM, Myers DE, Uckun FM. Treatment of human B-cell precursor leukemia in SCID mice by using a combination of the anti-CD19 immunotoxin B43-PAP with the standard chemotherapeutic drugs vincristine, methylprednisolone, and L-asparaginase. Leukemia & lymphoma. 1998 Sep;31(1-2):143–9. 67. Uckun FM, Manivel C, Arthur D, Chelstrom LM, Finnegan D, Tuel-Ahlgren L, et al. In vivo efficacy of B43 (anti-CD19)-pokeweed antiviral protein immunotoxin against human pre-B cell acute lymphoblastic leukemia in mice with severe combined immunodeficiency. Blood. 1992 May 1;79(9):2201–14. 68. Ottmann O, Dombret H, Martinelli G, Simonsson B, Guilhot F, Larson RA, et al. Dasatinib induces rapid hematologic and cytogenetic responses in adult patients with Philadelphia chromosome positive acute lymphoblastic leukemia with resistance or intolerance to imatinib: interim results of a phase 2 study. Blood. 2007 Oct 1;110(7):2309–15. 69. Smith BD, Levis M, Beran M, Giles F, Kantarjian H, Berg K, et al. Single-agent CEP-701, a novel FLT3 inhibitor, shows biologic and clinical activity in patients with relapsed or refractory acute myeloid leukemia. Blood. 2004 May 15;103(10):3669–76. 70. Tickenbrock L, Muller-Tidow C, Berdel WE, Serve H. Emerging Flt3 kinase inhibitors in the treatment of leukemia. Expert opinion on emerging drugs. 2006 Mar;11(1):153–65. 71. Drever Z, Steuber C, Bowman W, Murray J, Coppes M, Dinndorf P. & Camitta B. High risk infant ALL-improved survival with intensive chemotherapy. In: PROCEEDINGS OF THE ANNUAL MEETING-AMERICAN SOCIETY OF CLINICAL ONCOLOGY. 1998; Vol. 17, pp. 2032-2032. AMERICAN SOCIETY OF CLINICAL ONCOLOGY. 72. Hijiya N, Franklin J, Rytting M, Cooper T, Chu R, Barry E, McCarthy C, Abichandani R. & Carroll W. A phase I study of clofarabine in combination with cyclophosphamide and etoposide: A new regimen in pediatric patients with refractory or relapsed acute leukemia. In: ASCO Meeting Abstracts. 2007; Vol. 25, p. 9529. 73. Kantarjian HM, Jeha S, Gandhi V, Wess M, Faderl S. Clofarabine: past, present, and future. Leukemia & lymphoma. 2007 Oct;48(10):1922–30. 74. Brave M, Goodman V, Kaminskas E, Farrell A, Timmer W, Pope S, et al. Sprycel for chronic myeloid leukemia and Philadelphia chromosome-positive acute lymphoblastic leukemia resistant to or intolerant of imatinib mesylate. Clin Cancer Res. 2008 Jan 15;14(2):352–9. 75. Markovic A, MacKenzie KL, Lock RB. FLT-3: a new focus in the understanding of acute leukemia. The international journal of biochemistry & cell biology. 2005 Jun;37(6):1168-72. 76. Brown P, Levis M, McIntyre E, Griesemer M, Small D. Combinations of the FLT3 inhibitor CEP-701 and chemotherapy synergistically kill infant and childhood MLL-rearranged ALL cells in a sequence-dependent manner. Leukemia. 2006 Aug;20(8):1368–76. 77. Levis M, Smith B, Beran M, Baer M, Erba H, Cripe L, Coutre S, Advani A, Perl A. & Devetten M. A Randomized, Open-Label Study of Lestaurtinib (CEP-701), an Oral FLT3 Inhibitor, Administered in Sequence with Chemotherapy in Patients with Relapsed AML
15 Incorporating New Therapies into Frontline Protocols
327
Harboring FLT3 Activating Mutations: Clinical Response Correlates with Successful FLT3 Inhibition. In: Blood. 2005; Vol. 106, pp. 403-403. 2005, The American Society of Hematology. 78. Henze G, Langermann HJ, Bramswig J, Breu H, Gadner H, Schellong G, et al. [The BFM 76/79 acute lymphoblastic leukemia therapy study (author’s transl)]. Klinische Padiatrie. 1981 May;193(3):145–54. 79. Riehm H, Langermann HJ, Gadner H, Odenwald E. & Henze G. The Berlin Childhood Acute Lymphoblastic Leukemia Therapy Study, 1970-1976. Am J Pediatr Hematol Oncol. 1980;2, 299–306. 80. Pession A, Valsecchi MG, Masera G, Kamps WA, Magyarosy E, Rizzari C, et al. Long-term results of a randomized trial on extended use of high dose L-asparaginase for standard risk childhood acute lymphoblastic leukemia. J Clin Oncol. 2005 Oct 1;23(28):7161–7. 81. Harned TM, Gaynon PS. Treating refractory leukemias in childhood, role of clofarabine. Therapeutics and clinical risk management. 2008 Apr;4(2):327–36. 82. Anderson J, Krivit W, Chilcote R, Pyesmany A, Chard R, Hammond D. Comparison of the therapeutic response of patients with childhood acute lymphoblastic leukemia in relapse to vindesine versus vincristine in combination with prednisone and L-asparaginase: a phase III trial. Cancer treatment reports. 1981 Nov-Dec;65(11-12):1015–9. 83. Wells RJ, Odom LF, Gold SH, Feusner J, Krill CE, Waldron P, et al. Cytosine arabinoside and mitoxantrone treatment of relapsed or refractory childhood leukemia: initial response and relationship to multidrug resistance gene 1. Medical and pediatric oncology. 1994;22(4):244–9. 84. Houghton PJ, Morton CL, Kolb EA, Gorlick R, Lock R, Carol H, et al. Initial testing (stage 1) of the mTOR inhibitor rapamycin by the pediatric preclinical testing program. Pediatric blood & cancer. 2008 Apr;50(4):799–805. 85. Houghton PJ, Morton CL, Kolb EA, Lock R, Carol H, Reynolds CP, et al. Initial testing (stage 1) of the proteasome inhibitor bortezomib by the pediatric preclinical testing program. Pediatric blood & cancer. 2008 Jan;50(1):37–45. 86. Kolb EA, Gorlick R, Houghton PJ, Morton CL, Lock RB, Tajbakhsh M, et al. Initial testing of dasatinib by the pediatric preclinical testing program. Pediatric blood & cancer. 2008 Jun;50(6):1198–206. 87. Maris JM, Courtright J, Houghton PJ, Morton CL, Gorlick R, Kolb EA, et al. Initial testing of the VEGFR inhibitor AZD2171 by the pediatric preclinical testing program. Pediatric blood & cancer. 2008 Mar;50(3):581–7. 88. Tajbakhsh M, Houghton PJ, Morton CL, Kolb EA, Gorlick R, Maris JM, et al. Initial testing of cisplatin by the pediatric preclinical testing program. Pediatric blood & cancer. 2008 May;50(5):992–1000.
Index
A Aberrant methylation, 16 Absolute neutrophil count (ANC), 77 ABT–263, 128–130 ABT–737, 128–130, 291–292 Accelerated titration designs, 64 Acute lymphoblastic leukemia (ALL), 131–132 Acute myeloid leukemia (AML), 132–133 Acute promyelocytic leukemia (APML), 157 ADCC. See Antibody dependent cellular cytotoxicity (ADCC) ADCP. See Antibody dependent cell-mediated phagocytosis (ADCP) Adenosine triphosphate (ATP), 204, 206, 207 Adrenocorticotrophic hormone (ACTH), 2 Alkaloid vincristine, 2 Allogeneic stem cell transplantation (Allo-SCT), 25, 67 All-trans retinoic acid (ATRA), 157 Animal models leukemic stem cell, 109–110 NOD/SCID mouse, 108 preclinical testing, 110–111 recipient animals, 108–109 SCID, 106–107 in vivo xenograft model, 106 Antagonist aminopterin, 2 Antibody dependent cell-mediated phagocytosis (ADCP), 229 Antibody dependent cellular cytotoxicity (ADCC), 229, 249 Anti-CD19 vs. anti-CD22 immunoconjugates, 223 Antigen presenting cells (APCs), 227 Antisense oligonucleotides, 122 APML. See Acute promyelocytic leukemia (APML)
Apoptosis and cancer Bcl–2 family protein interactions, 119, 120 Extrinsic and intrinsic pathways, 117, 118 Apoptotic defects, 160 Area under the curve (AUC) values, 6 Asparagine synthetase (AS), 7 Ataxia telangiectasia mutated (ATM), 179 Athymic mice, 106 ATP. See Adenosine triphosphate (ATP) Avascular necrosis, 5 B B-cell development, 193 B-cell precursor (BCP), 254–257 Bcl–2 family proteins ABT–737/263, 128–130 apoptosis and cancer, 117–121 HA14–1, 125–126 haematological malignancy doxycycline, 121 oblimersen, 122–125 leukemia acute lymphoblastic, 131–132 acute myeloid, 132–133 chronic lymphocytic, 134 chronic myeloid, 133–134 lymphoma, 130–131 myeloma, 134–135 obatoclax, 126–128 paediatrics, 135 Bcl–2 homology (BH), 118 BCLL. See B-lymphocytic leukemia (BCLL) BCP. See B-cell precursor (BCP) Beige mutation, 107 Berlin-Frankfurt-Munster (BFM), 9, 321 Bispecific T-cell engaging antibodies (BiTEs®), 228, 230
V. Saha and P. Kearns (eds.), New Agents for the Treatment of Acute Lymphoblastic Leukemia, DOI 10.1007/978-1-4419-8459-3, © Springer Science+Business Media, LLC 2011
329
330 Blinatumomab, 259 B-lymphocytic leukemia (BCLL), 237 Bortezomib, 273, 276 C CALLA. See Common acute lymphoblastic leukemia antigen (CALLA) Cancer therapeutics. See Proteasome inhibitors, therapeutic utility Carfilzomib, 278–279 CAT–8015, 259 CCG. See Children’s Cancer Group (CCG) CD2, 247, 248 CD3, 247–249 CD4, 248, 249 CD5, 248, 250 CD7, 248, 250 CD10, 234 CD25, 248, 250–251 CD19 antigen antibody–drug conjugates, 235, 237 biological characteristics, 234–235 bispecific antibodies, 236–238 preclinical and clinical uses, 235–236 unconjugated anti-CD19, 235 CD20 antigen biological characteristics, 238 bispecific antibodies, 241 immunoconjugates, 241–242 preclinical and clinical uses, 238–239 unconjugated anti-CD20, 239–240 CD22 antigen biological characteristics, 242 bispecific antibodies, 243, 244 immunoconjugates, 243, 245–246 preclinical and clinical uses, 242 unconjugated anti-CD22, 242–243 CD33 antigen biological characteristics, 251 bispecific antibodies, 252, 253 immunoconjugates, 252–254 preclinical and clinical uses, 251–252 unconjugated anti-CD33, 252, 253 CD52 antigen biological characteristics, 231 immunoconjugates, 232 preclinical and clinical uses, 231–232 CD79 anitgen, 246 CDC. See Complement dependent cytotoxicity (CDC) CDDP. See Cis–diamminedichloridoplatinum (CDDP) Cellular cytotoxic activity, 222 Central nervous system (CNS) disease, 170
Index Central neurotoxicity, 176 Cerebrospinal fluid (CSF), 170, 208 Childhood acute myeloid leukemia, 69 Children’s Cancer Group (CCG), 67–68, 319, 321 Children’s Oncology Group (COG), 61, 319–321 CHOP. See Cyclophosphamide, doxorubicin, vincristin, prednisone (CHOP) Chronic lymphocytic leukemia (CLL), 134 Chronic myeloid leukemia (CML), 133–134 Cis-diamminedichloridoplatinum (CDDP), 40 Cisplatin, 40–41 Clinical trials. See also Early phase trials; Trial design and analyses clofarabine, 170–174, 321 combination studies, 288 dasatinib, 320 epigenetic pathways DNA methyltransferase inhibitors, 303–305 histone deacetylase inhibitors (HDACi), 306 forodesine, 180–181 gemcitabine, 183 lestaurtinib, 320–321 nelarabine, 175–176, 321 proteasome inhibitors, therapeutic utility combination studies, 288 single-agent activity, 284–288 single-agent activity Assesment of Proteasome Inhibition for EXtending remission-trial (APEX), 286 bortezomib, 284–288 Clinical Response and Efficacy Study of bortezomib in the Treatment of refractory myeloma (CREST), 286 dexamethasone, 287 relapsed and refractory pre-B ALL, 288 SUMMIT, 285 Clofarabine, 167–174, 321 CNS-directed therapy, 14 COG. See Children’s Oncology Group (COG) Common acute lymphoblastic leukemia antigen (CALLA), 234 Complementarity-determining regions (CDR), 222 Complement dependent cytotoxicity (CDC), 222, 229 Complete remission/response (CR), 49, 285, 288, 315 Complete response with insufficient platelet recovery (CRp), 77, 171
Index Confidence intervals (CI), 89 Consolidation and intensification MRC UKALL X protocol, 10 sensitivity/resistance patterns, 11–12 Continual reassessment method (CRM), 63–64 Continuation therapy methotrexate, 13 thioguanine, 13 TPMT, 12–13 CpG island, 299 CRLF2. See Cytokine receptor-like factor 2 (CRLF2) CRM. See Continual reassessment method (CRM) CRp. See Complete response with insufficient platelet recovery (CRp) CSF. See Cerebrospinal fluid (CSF) Cyclophosphamide, doxorubicin, vincristin, prednisone (CHOP), 76, 239 Cytokine receptor-like factor 2 (CRLF2), 29 Cytotoxicity, 172, 178, 180, 182, 184 Cytotoxic T-cells, 228 Cytotoxic T lymphocytes (CTLs), 227 D Dasatinib, 320 Daunorubicin (DNR), 9, 122 Death receptor pathway. See Extrinsic pathway Decitabine, 303–305 Denaturing high performance liquid chromatography (dHPLC), 190 Dendritic cells (DCs), 227 Deoxycytidine kinase (dCK) activity, 180 dHPLC. See Denaturing high performance liquid chromatography (dHPLC) Diffuse large B-cell lymphoma (DLBCL ), 131 DLTs. See Dose-limiting toxicities (DLTs) DNA hypermethylation, 300–301 DNA hypomethylation, 300 DNA methylation, 299, 301–302 Dose-limiting toxicities (DLTs) backbone toxicity rate, 70, 71 non-hematologic definition, 70, 71 exclusion approach, 71–72 functional approach, 72 grade 3 and 4 toxicities, 70 Drug testing concentric model, 52 pharmacokinetics, 52–53 PPTP aim, 47
331 evaluation methods, 49, 50 heat map representation, 48, 49 types, response curves, 47–48 recommendations, 45–46 therapeutic synergy, 51 transgenic mouse models, 46 treatment outcomes, 318–319 E Early phase trials phase II trials and design issues endpoint, 77–78 randomized phase II designs, 76–77 single arm phase II design, 76 phase I trials and design issues accelerated titration designs, 64 continual reassessment method, 63–64 conventional, 62–63 conventional pediatric, 63 dose limiting toxicities, 69–70 obstacles, conventional phase I trials, 8 platform approach, 65–68 rolling six design, 64–65 window approach, 68–69 preclinical evaluation, childhood leukemia, 62 Ectopic models, 106 Effective concentration for 50% effect (EC50), 127 EFS. See Event-free survival (EFS) time EGFP. See Enhanced green fluorescent protein (EGFP) Electrophoretic mobility shift assay (EMSA), 283 Endoplasmic reticulum (ER) stress, 280–281 Enhanced green fluorescent protein (EGFP), 44 Epigenetic pathways aberrant DNA methylation, 301–303 clinical trials DNA methyltransferase inhibitors, 303–305 histone deacetylase inhibitors (HDACi), 306 definition, 299 DNA methylation, 299 dysregulation in cancer, 300–301 epigenetic modifiers, paediatric ALL, 307 epigenetic-targeted therapy, 306–307 histone acetylation, 300 microRNA (miRNA), 306 therapeutic target, 303 transcriptional silencing, 299 Epigenetic processes, 303
332 Epoxomicin, 276 Epratuzumab anti-CD22 antibody, 242–243 nomenclature of MoAbs, 230–231 platform approach, 67 unconjugated antibody, 255 window approach, 68 Equipoise, 85 ER stress. See Endoplasmic reticulum (ER) stress European Medicine’s Agency (EMEA), 87, 90 Event-free survival (EFS) time, 84, 86, 87, 318 Ewing sarcoma family tumours (ESFT), 135 Extracellular signal related kinase (ERK), 127, 206 Extrinsic pathway, 117, 118 F FDA. See Food and Drug Administration (FDA) Fetal liver kinase 2 (FLK2), 189 Fixed dose rate (FDR), 182 FLK2. See Fetal liver kinase 2 (FLK2) FLT3. See Fms-like tyrosine kinase 3 (FLT3) FLT3 inhibition, 73–74 FLT3 inhibitors aberrant activation, 190–191 acivation in MLL rearranged ALL, 192–193 in AML, 193–194 clinical evaluation in AML, 194–196 drug resistance mechanisms, 197 imatinib/dasatinib, 197 mutations, 191 mutations in AML and ALL, 191–192 regulation in hematopoietic cells, 189–190 therapeutics, 196–197 FLT3 ligand (FLT3L), 189 Fms-like tyrosine kinase 3 (FLT3), 15–16 Food and Drug Administration (FDA), 208 G Gemtuzumab ozogamicin (GMTZ) CD33, 252, 257 nomenclature, MoAbs, 231 preclinical testing, 111 Gene expression profiling (GEP), 5, 27 Glucocorticoid receptors (GR), 4 Glycogen synthase kinase 3 isoforms a(GSK3a), 151 Glycogen synthase kinase 3 isoforms b(GSK3b), 151
Index GMTZ. See Gemtuzumab ozogamicin (GMTZ) Graft-versus-host-disease (GvHD), 247 Green fluorescent protein (GFP), 129 GvHD. See Graft-versus-host-disease (GvHD) H Haematological malignancies, 170, 180, 181 Haematological toxicity, 176, 183 Haematopoietic stem cell (HSC), 28, 145 Haematopoietic stem cell transplant (HSCT), 173 HATs. See Histone acetylases (HATs) HDACi. See Histone deacetylase inhibitors (HDACi) Heat shock protein, 284 Hepatic toxicity, 170, 183 Histone acetylases (HATs), 289 Histone acetylation, 300 Histone deacetylase (HDAC), 289, 306 Histone deacetylase inhibitors (HDACi), 289–291 Histone methylation, 300 HSCT. See Haematopoietic stem cell transplant (HSCT) Hsp–27 protein, 284 Human anti-mouse antibodies (HAMA), 235 Hyperdiploid ALL, 192 Hypervariable region, 222, 223 I ICN. See Intracellular NOTCH (ICN) IL–2 receptor common g chain (IL2rg), 145 Immune effector cell mechanisms, 224–228 Immunoreceptor tyrosine-based activation motif (ITAM), 246 Immunoreceptor tyrosine-based inhibitory motifs (ITIMs), 242 Induction therapy anthracyclines, 8–9 basic template, 3 L-asparaginase dexamethasone, 8 Escherichia coli and Erwinia carotovora, 7 steroids corticosteroids, 4 dexamethasone, 4 gene expression profile, 5 vincristine, 5–6 Inhibitory concentration (Ki), 128, 129, 213
Index Inhibitory concentration for 50% effect (CI50), 130 Inotuzumab ozogamicin, 259 Insulin like growth factor 1 (IGF–1), 127 Interferon alpha (IFNa), 241 Internal tandem duplication (ITD), 190 International Conference on Harmonization (ICH), 97 International normalized ratio (INR), 74 Intracellular NOTCH (ICN), 153 Intravenous (IV), 10, 171, 172, 179, 305 Intrinsic pathway, 117, 118 In vitro experimental models, 43 In vivo experimental models orthotopic ALL xenograft models, 45 transgenic mouse models, 44 In vivo xenograft model, 106 ITAM. See Immunoreceptor tyrosine-based activation motif (ITAM) J Janus family kinase (JAK), 29 Jun amino terminal kinase (JNK), 126 Juxtamembrane (JM) domain, 189, 190 K Killer Ig-like receptor (KIR), 227 Kinase insertion domain receptor (KDR), 194 L Lactacystin, 275, 276 L-asparaginase induction therapy dexamethasone, 8 E. coli and Erwinia carotovora, 7 preclinical evaluation, 40–41 preclinical therapies, 26 Lestaurtinib, 193–194, 320–321 Lestaurtinib monotherapy, 194–195 Leukemia stem cells (LSC) animal models, 109–110 cancer stem cell hypothesis, 143 characteristics, 144, 145 critical stem cell pathways aberrant cell signaling, 157–160 apoptotic defects, 160 differentiation defects, 156–157 self-renewal, 151–156 identification and cellular characteristics hierarchy, 145, 147
333 immunocompromised recipient mouse strain, 145 pathways and potential therapeutic interventions, 148 xenotransplantation, 145, 146 Leukemogenesis, 191 Lymphokine activated killer cells (LAK), 107 Lymphoma, 130–131 Lymphoproliferative diseases (LPD), 232 M Macroautophagy, 126 Macrophage colony-stimulating factor receptor (FMS), 189 Macrophages, 227 Maintained complete response (MCR), 48, 49 Major histocompatibility complex (MHC), 227, 232 Maleimidomethyl-cyclohexane-carboxylate (MCC), 246 Mammalian target of rapamycin (mTOR), 31, 46, 62, 131 MAPK kinase (MEK), 126 Marizomib, 277–278 Maximum tolerated dose (MTD), 45, 53, 62–63, 183 MCR. See Maintained complete response (MCR) MDS. See Myelodysplastic syndrome (MDS) Medical Research Council (MRC), 10, 11 Membrane attack complex (MAC), 229 6-Mercaptopurine (6-MP), 2, 12, 13, 178 Methotrexate CNS-directed therapy, 14 consolidation and intensification, 10 continuation therapy, 13 induction therapy, 3 sensitivity/resistance patterns, 11–12 Methyl thiazolyl diphenyltetrazolium bromide (MTT), 4 MHC. See Major histocompatibility complex (MHC) Midostaurin, 193–195 MIMP. See Mitochondrial inner membrane potential (MIMP) Minimal residual disease (MRD) end-reinduction, 78 platform approach, 66–67 trial design, 91, 92 Missense point-mutation, 190 Mitochondrial inner membrane potential (MIMP), 134
334 Mitochondrial outer membrane potential (MOMP), 119 Mitochondrial pathway. See Intrinsic pathway Mitogen activated protein kinase (MAPK), 126 Mixed lineage leukemia (MLL). See FLT3 inhibitors moAbs. See Monoclonal antibodies (moAbs) Monoblastic/monocytic AML, 192 Monoclonal antibodies (moAbs) B-lineage specific antigens CD10, 234 CD19, 234–238 CD20, 238–242 CD22, 242–246 CD79, 246 clinical use bispecific antibodies, 230 immune effector cell mechanism, 224, 227–228 immunoconjugates, 229–230 lymphoblastic leukemia cell antigens, 224–226 nomenclature, 230–231 radioimmunotherapy, 229–230 unconjugated antibody activity, 228–229 human anti-mouse antibodies (HAMA), 221 leukaemic stem cell, 257–258 lineage independent antigens CD52, 231–233 HLA-DR, 232–234 myeloid antigens, 251–254 against precursor B-cell/T-ALL adult B-cell NHL, 254 conjugated antibodies, 256–257 cytotoxic mechanisms, 255 unconjugated antibodies, 255–256 structure, 221–222 T-lineage specific antigens CD2, 247, 248 CD3, 247–249 CD4, 248, 249 CD5, 248, 250 CD7, 248, 250 CD25, 248, 250–251 treatment/co-stimulatory strategies, duration of, 258–259 Mouse embryonic fibroblast (MEF), 129 Mouse models, 189 6MP. See 6-Mercaptopurine (6-MP) MRD. See Minimal residual disease (MRD) MTD. See Maximum tolerated dose (MTD)
Index mTOR. See Mammalian target of rapamycin (mTOR) Multiple Myeloma Research Consortium (MMRC), 277 Myelodysplastic syndrome (MDS), 128, 195, 304 Myeloid antigens. See CD33 antigen Myeloma, 134–135 Mylotarg® CD33, 252, 257 nomenclature, moAbs, 231 preclinical testing, 111 N National Cancer Institute (NCI), 40, 41, 62, 246, 276 Natural killer (NK), 44, 224, 247 NCI. See National Cancer Institute (NCI) Nelarabine, 321 Neurotoxicity, 176 Neutropenia, 172, 173, 177, 180 Neutrophils, 227 NHL. See Non-Hodgkin Lymphoma (NHL) NK cells, 227 NOD. See Non-obese diabetic (NOD) NOD/SCID mouse, 108 Non-Hodgkin Lymphoma (NHL), 124, 131, 230, 235, 241–243, 246, 287 Non-obese diabetic (NOD), 44, 109, 145, 148–149 NOTCH 1, 16 NPI–0052. See Marizomib NSG strain, 109–111 Nuclear localizing sequences (NLS), 254 Nucleoside analogues clofarabine ALL, 167 clinical trials, 170–174 paediatric leukemia, 167 pharmacokinetics, 169–170 pharmacology, 167–169 forodesine clinical trials, 180–181 immunodeficiency, 177 pharmacology, 178–180 PNP deficiency, 177 gemcitabine clinical trials, 183 pharmacokinetics, 182–183 pharmacology, 182 nelarabine clinical trials, 175–176 multiagent chemotherapy, 177
Index multidrug therapy, 177 pharmacokinetics, 175 pharmacology, 174–175 purine nucleoside phosphorylase (PNP) deficiency, 174 relapse, 177 T-cell ALL, 177 toxicities, 176–177 Nude mutation, 107 Number needed to treat (NNT), 319 O Obatoclax, 126–128 Orthotopic models, 106 Outer mitochondrial membrane (OMM), 134, 160 P Paediatrics, 135 Parthenolide (PTL), 160 Partial remission/response (PR), 48, 123, 171, 285, 287, 288 Partial thromboplastin time (PTT), 74 PD. See Progressive disease (PD) PDGFR. See Platelet-derived growth factor receptor (PDGFR) Pediatric preclinical testing program (PPTP) aim, 47 evaluation methods, drug testing, 49, 50 heat map representation, 48, 49 types, response curves, 47–48 PEG asparaginase, 7 Peptide aldehydes, 275 Peptide boronate PS–341. See Bortezomib Peripheral neurotoxicity, 176 Per protocol (PP), 100, 101 P-glycoprotein (PGP), 6, 134, 208, 229 Pharmacokinetics clofarabine, 169–170 gemcitabine, 182–183 nelarabine, 175 preclinical evaluation, 52–53 Pharmacology clofarabine, 167–169 forodesine, 178–180 gemcitabine, 182 nelarabine, 174–175 Pharmacore, 275 3+3 phase I trial design, 62 Phase I trials and design issues accelerated titration designs, 64 continual reassessment method, 63–64
335 conventional, 62–63 conventional pediatric, 63 dose limiting toxicities, 69–70 obstacles, conventional phase I trials, 65 platform approach, 65–68 rolling six design, 64–65 window approach, 68–69 Platelet-derived growth factor receptor (PDGFR), 189, 206–208, 210 Platform approach advantage and disadvantage, 68 COG AALL01P2 triple re-induction, 66 early and late marrow relapses, 67 MRD, 66–67 three-block approach, 67 Polymerase Chain Reaction (PCR), 13, 192, 312 Postamendment block 1 therapy, 66 PPTP. See Pediatric preclinical testing program (PPTP) PR. See Partial remission/response (PR) Preclinical evaluation drug testing concentric model, 52 pharmacokinetics, 52–53 PPTP, 47–50 recommendations, 45–46 therapeutic synergy, 51 transgenic mouse models, 46 evaluation criteria, robust preclinical models, 42 in vitro experimental models, 43 in vivo and in vitro panels, 41 in vivo experimental models orthotopic ALL xenograft models, 45 transgenic mouse models, 44 L-asparaginase and cisplatin, 40–41 Pre-clinical therapies cell-death pathway triggers, 32 conceptual model, 29–30 cytogenetic analysis, 26 global profile, 29 heterogeneity, treatment response, 28 kinase inhibitors, 32 L-asparaginase, 26 monoclonal antibodies, 32 relapse occurence, extramedullary sites, 27–28 origin, 27 g-secretase, 33 steroid-sensitising adjuvants, 31–32 T-lineage ALL, 33 transcriptional profile, 29 tumour microenvironment, 28–29
336 Progressive disease (PD), 48, 49, 68, 243 Prospective meta-analysis strategy (PMAS), 99 Proteasome inhibition, 74–76 Proteasome inhibitors, therapeutic utility ALL treatment, potential combination studies BCL–2 antagonists, 291–292 HDAC inhibitors, 289–291 bortezomib, 273 clinical trials combination studies, 288 single-agent activity, 284–288 development, 275–276 haematological malignancies, 273 molecular pharmacology caspase activation, 280–281 endoplasmic reticulum stress, 280–281 NFkB-inhibiting survival, 282–283 oxidative stress, 281–282 pro-apoptotic proteins, 279–280 resistance, 284 next-generation, 277–279 20S/26S proteasome, 274 ubiquitinated proteins, 274–275 ubiquitin-proteasome pathway, 273, 274 Puma, 280 R Randomized phase II designs, 76–77 Reactive oxygen species (ROS), 191, 281 Receptor tyrosine kinase (RTK), 189 Recommended phase II dose (RPIID) phase I end point, 69 single-arm phase II design, 76 Retinoblastoma gene (RB), 155–156 Ribonucleotide reductase, 182 Rolling six design, 64–65 ROS. See Reactive oxygen species (ROS) RPIID. See Recommended phase II dose (RPIID) RT-PCR analysis, 192 S SCID. See Severe-combined immunodeficient (SCID) SCT. See Stem cell transplantation (SCT) SD. See Stable disease (SD) Second mitochondrial derived activator of caspases (Smac), 118 Self potentiation, 182, 183
Index Serine/threonine aurora kinase, 213 Severe-combined immunodeficient (SCID), 106–107 Single arm phase II design, 76 Small interfering RNA (SiRNA), 127, 157 Sorafenib/Tandutinib, 195 Src-Abl Tyrosine kinase inhibition Activity Research Trials (START), 210 Stable disease (SD), 48, 123, 243, 285, 287 Standard dose schedule (STD), 183 Steel factor receptor (KIT), 189 Stem cell pathways aberrant cell signaling BCR-ABL1 and NUP241-ABL1 signaling, 158–159 FLT3-TKD mutant signaling, 159 and JAK mutations, 159–160 apoptotic defects, 160 differentiation defects, 156–157 self-renewal bHLH proteins and their binding partners, dysregulation, 154–155 clustered and non-clustered HOX genes, 151–152 NOTCH mutations, 152–153 PTEN pathway, 153–154 RB/TP53 tumour suppressor network, 155–156 Stem cell transplantation (SCT), 67, 158, 212, 256 Stem cell tyrosine kinase 1 (STK1), 189 Subcutaneous models, 106 Sunitinib, 193–194 T T-cell lymphoma, 176 T-cell malignancies, 174, 178, 179 Thioguanine, 13 Thiopurine methyltransferase (TPMT) continuation therapy, 12–13 location, 28 Thrombocytopenia, 177 TKD. See Tyrosine kinase domain (TKD) T-lymphoblasts, 175 TNF-related apoptosis inducing ligand (TRAIL), 237 TPMT. See Thiopurine methyltransferase (TPMT) Transgenic zebrafish models, 44 Transmembrane (TM) domain, 189, 190 Treatment outcomes Children’s Oncology Group, clinical trials
Index clofarabine, 321 dasatinib, 320 lestaurtinib, 320–321 nelarabine, 321 dexamethasone, 311 drug selection clinically active/inactive, 315 dexamethasone vs. prednisone, 317 imatinib, 317 positive/negative control, 315–316 resistance mechanisms, 317 drug testing, 318–319 hematopoietic stem cell transplant (HSCT), 311 imatinib, 312 relapse, operational definition blast proliferation, 313 blast re-accumulation, 314 cell death, 313 chemotherapeutic agents, 314 oligoclonal ALL, 312–313 refractory sub-clone, 314 Trial design and analyses adaptive designs adaptive seamless design, 95 group sequential designs, 94 heterogeneity, 94 Bayesian approach criticism, 96 hypothesis testing, 96–97 interval estimates, 97 posterior distribution, 95 biological activity, 91 efficacy, 92 experimental research clinical aspects, 83 comparative trials, 85 definition, 83 equipoise, 85 translational trials, 84 treatment tolerability, 84–85 good clinical practice (GCP) principles, 97–99 international clinical trial logistics and implementation, 99–100 prospective meta-analysis strategy (PMAS), 99 requirements, 99 lymphoblastic leukemia high-risk sub-populations, 85 intermediate risk patients, 86 regulatory approval process, 87 treatment optimization, 86
337 protocol structure and content, 97–98 statistical analysis frequentist approach, 101 heterogeneity, 101–102 intention to treat (ITT), 100 interim analyses, 101 per protocol (PP), 100 as treated (TP), 100 superiority and non-inferiority add-on trial, 88 confidence intervals, 89 head-to-head trial, 88 surrogate end points failure, 93 validation, 92–93 Tropomyosin-related kinase (TRK), 194 Tumour microenvironment, 28–29 Tumour Necrosis Factor (TNF), 227, 237 Tyrosine kinase domain (TKD), 157, 189–191 Tyrosine kinase (TK) inhibitors bafetinib, 207 BCR/ABL oncoprotein, 206–207 bosutinib, 207 clinical application bafetinib, 211 bosutinib, 210–211 CML, 209–211 combination therapy, 212 dasatinib, 210, 212, 213 haematological and cytogenetic responses, 211 imatinib, 208, 209, 211, 212 dasatinib, 207, 208 hepatic metabolism, 208 imatinib, 206–207, 209 intracellular signalling pathways, 204–205 JAK mutations, 213 leukemia aetiology childhood ALL, 205 chimeric oncoprotein, 205–206 Philadelphia positive ALL, 206 Philadelphia translocation protein, 205 multi-agent chemotherapy, 212 myeloproliferative disease, 213 nilotinib, 207–209, 213 Philadelphia positive ALL, 209–212 post stem cell transplantation, 212 protein phosphorylation, 203–204 receptor and non-receptor TK families, 203 regulatory mechanism, 204 staurosporine, 206
338
Index
U Unfolded protein response (UPR), 280
W Window approach, 68–69
V Valproic acid (VPA), 237 Variable (V) region, 221 Vascular endothelial growth factor receptor (VEGFR), 194, 208 Veno-occlusive disease (VOD), 13
X Xid mutation, 107 XmAb5574, 259