The acute leukemias represent a biologically and clinically diverse group of disorders of the bone marrow.1 Because of the pioneering attempts to classify these disorders on the basis of cytogenetic and molecular characteristics, it is possible now to attribute distinct clinical behaviors and prognostic characteristics to the leukemias. A more fundamental understanding of the molecular pathogenesis of leukemias also has availed a rich array of potential therapeutic targets, and the FMS-related tyrosine kinase receptor 3 (FLT3) has emerged as an attractive prospect.2 Although several small-molecule inhibitors of FLT3 are being used in clinical trials, there remains a significant need to develop more effective and enduring therapies against this target. An antibody-based approach could have significant advantages over the current approaches with small molecules. In this report, we discuss recent observations on the relevance of FLT3 to leukemia pathogenesis, lessons learned from the clinical trials with small molecules in leukemias, and ongoing challenges that may be addressed better with the use of therapeutic antibodies.
FMS-related tyrosine kinase receptor 3 (FLT3) is a class III receptor tyrosine kinase that holds considerable promise as a therapeutic target in hematologic malignancies. Current efforts directed toward the development of small-molecule tyrosine kinase inhibitors of FLT3 may be limited by off-target toxicities and the development of drug resistance. Target-specific antibodies could overcome these hurdles and provide additional mechanisms to enhance the antitumor efficacy of FLT3 inhibitors. IMC-EB10 is a novel antibody directed against FLT3. The binding of IMC-EB10 to FLT3 results in antiproliferative effects in vitro and in mouse models engrafted with human leukemia cells that harbor wild-type or constitutively activated FLT3. Future clinical trials will test these notions formally and will identify the most appropriate opportunities for this member of a new generation of antileukemic therapies. Cancer 2010;116(4 suppl):1013–7. © 2010 American Cancer Society.
FLT3 in Hematopoiesis
FLT3 is a member of the class III receptor tyrosine kinase (RTK) family that is homologous to other receptors involved in the control of hematopoiesis and cancer pathogenesis, including FMS, KIT, and platelet-derived growth factor receptor.3 These RTKs are characterized by 5 immunoglobulin (Ig)-like extracellular domains, a transmembrane domain, a juxtamembrane domain, and an interrupted tyrosine kinase (TK) domain. The plasma membrane-bound form of FLT3 is a 160-kilodalton (kD) glycoprotein that forms dimers upon binding the FLT3 ligand. Because of this dimerization, the internal TK domain is activated by autophosphorylation. In turn, this activation transduces growth signals through several well characterized pathways, including those mediated by the mitogen-activated protein kinase (MAPK), phosphatidylinositol 3 kinase (PI3K), and the signal transducer and activator of transcription 5 (STAT5).2
The role of FLT3 in normal and abnormal hematopoiesis increasingly is being elucidated. FLT3 is expressed on myeloid-lymphoid progenitor cells but not on pluripotent stem cells.4 In mouse models, inactivating mutations in the mouse homologue of FLT3 are associated with little if any organismal abnormalities except for subtle defects in hematopoiesis involving B-cell populations.5 Extrapolating from such observations, the specific targeted abrogation of FLT3 function in patients should spare the compartment of normal hematopoietic stem cells and result in little or no toxic effects on either hematopoiesis or nonhematopoietic functions.
FLT3 Mutations in Leukemia
The overexpression of wild-type FLT3 has been noted in >90% of acute myeloid leukemias (AML) and nearly all B-cell acute lymphoblastic leukemias (ALL). Less frequently, it has been noted in T-cell ALL and in chronic myelogenous leukemia during blast crisis.2
A major reason for regarding FLT3 as a validated target in leukemia relates to the observation that functionally relevant mutations in this gene are associated with the pathogenesis of leukemia. Two mechanisms account for the majority of such mutations. Internal tandem duplications in the juxtamembrane domain occur in up to 30% of cells derived from patients with AML and in 5% of cells derived from patients with myelodysplastic syndrome,6 whereas point mutations in the TK domain are observed in approximately 7% of patients with AML.7 Both mutations represent gain-of-function alterations that lead to the constitutive activation of FLT3 and the robust proliferation of leukemic cells.7, 8
FLT3 mutations also have prognostic import and can be important predictors of disease recurrence.9-11 In mouse models, high levels of FLT3 in the presence of other genetic aberrations are sufficient to induce aggressive AML. The genetic background and the interplay of mutations in critical leukemia-associated genes can be powerful determinants of the tempo of the disease. For example, patients whose leukemic cells express the mutant form of nucleophosmin 1 but not the FLT3 mutation appear to have a relatively favorable prognosis.12
The organ-specific expression of FLT3 makes it an attractive molecular target for antileukemia therapy.2 To date, several small-molecule tyrosine kinase inhibitors have entered clinical development (Table 1). Generally, the inhibitory activity of these agents is not limited to FLT3: Other intracellular tyrosine kinases also are inhibited in vitro with drug concentrations that may be relevant in patients. Indeed, some small molecules directed against FLT3 were not developed a priori as FLT3 inhibitors. Nevertheless, their activity in vitro against FLT3 stimulated further development in leukemias. For example, CEP-701 was identified originally as a tyrosine kinase A inhibitor.13 Similarly, PKC412 initially was conceived as an inhibitor of protein kinase C.14 However, the ability of these molecules to inhibit FLT3 phosphorylation established the rationale for further development in FLT3-associated leukemia.
|Compound||Company||FLT3 IC50, nM||Additional Targets||Clinical Status|
|Sunitinib||Pfizer Inc., New York, NY||50||KIT, PDGFR, VEGFR||Phase 1/2 (AML)|
|CEP-701||Cephalon, Frazer, Pa||2||TrkA||Phase 3 (AML)|
|PKC412||Novartis International AG, Basel, Switzerland||10||KIT||Phase 3 (AML)|
|MLN0518||Millennium Pharmaceuticals, Inc., Cambridge, Mass||30||PDGFR, KIT||Phase 2 (AML)|
|Sorafenib||Bayer AG, Leverkusen, Germany||1.2||VEGFR, PDGFR, KIT||Phase 1 (AML)|
Although as single agents the use of these drugs has produced hematologic responses, most responses have been associated with the clearance of leukemic blasts from the systemic circulation rather than from the bone marrow. Not surprisingly, these responses appear to be transient and of short duration. Proof-of-concept studies of CEP-701 as monotherapy in patients with refractory or recurrent AML produced reductions in peripheral blast counts in 5 of 14 patients. One patient also had a decrease in the bone marrow blast count. In clinical trials, PKC412 also demonstrated clearance of peripheral blasts, but this only rarely translated into a significant reduction in the bone marrow blast count.15 It is interesting to note that some responses were associated with the inhibition of FLT3 phosphorylation. The availability of biomarkers that correlate with clinical endpoints could significantly advance the rational development of these agents.
Rationale for Antibody Development
A potential limitation of small-molecule inhibitors has been the inability of some patients to tolerate the higher doses of drugs that may be required to achieve adequate plasma concentrations and more complete inhibition of FLT3.16 Consequently, the maximum tolerated dose of some drugs—the typical paradigm in anticancer drug development—could not be reached.15-17 It is possible that toxicities associated with the inhibition of 1 or more targets beyond FLT3 could have been responsible. Another potential limitation is the development of drug resistance. Potential mechanisms include the induction of de novo mutations or selection for preexisting mutations in FLT3 upon drug exposure that abrogate the interaction of the drug with the molecular target.18, 19 Such binding-domain mutations reduce or eliminate the sensitivity of the leukemic cells to these inhibitors. The precise clinical relevance of these mechanisms remains uncertain, but it is a cause for concern with the anticipated long-term use of such agents. Finally, other mechanisms for resistance may relate to normal cellular detoxification vehicles, such as P-glycoprotein or the cytochrome P450 system, and a variety of other normal metabolic or bioavailability mechanisms that interfere with the effective intracellular concentrations of drugs.
An antibody approach may successfully overcome some or many of these shortcomings.20 Monoclonal antibodies may exert their full antitumor effect through additional mechanisms that are not applicable to small-molecule inhibitors, such as ligand blockade or the modulation of antitumor immune mechanisms through antibody-dependent cell-mediated cytotoxicity (ADCC). Because intracellular resistance mechanisms may place severe limits on the long-term use of small molecules, antibodies may be advantageous because they are not subject to such mechanisms. Most important, the target specificity of antibodies circumvents potential toxicities that result from the inhibition of multiple targets, which inevitably are a feature of small-molecule kinase inhibitors.
However, antibodies also may have several unique shortcomings, particularly the potential for hypersensitivity reactions associated with protein products. It is possible that the expression of the molecular target could be selected against over time, and the expression of FLT3 may be reduced or lost completely after a period of exposure to an anti-FLT3 antibody. Post-translational processing of cell surface receptors also is possible; this could liberate soluble forms of the extracellular domain into the circulation, which then may act as a therapeutic sink. Although neither of these mechanisms is known to apply to FLT3, 1 or more such hypothetical mechanisms could attenuate the clinical activity of anti-FLT3 antibodies.
Anti-FLT3 Monoclonal Antibodies
Several monoclonal antibodies directed against FLT3 have been described. IMC-EB10 is a fully human IgG1 monoclonal antibody that initially was isolated by phage display technology. IMC-NC7 is also a human anti-FLT3 IgG1 antibody. Both antibodies bind to FLT3 with high affinity: IMC-EB10 with a Kd of 158 pM and IMC-NC7 with a Kd of 450 pM.20-22 Much of the discussion below is focused on IMC-EB10, because this is the antibody that will be advanced to clinical development. IMC-EB10 recognizes both the wild-type and mutant forms of FLT3 and inhibits ligand-induced and ligand-independent phosphorylation, respectively. It also abrogates downstream signaling through inhibition of the MAPK, PI3K, and STAT5 pathways in a variety of AML cell culture models as well as in primary blasts from patients with AML.21, 22
With regard to other antitumor effects, IMC-EB10 appears to be more effective than IMC-NC7 in mediating ADCC in vitro. In this context, it is worth noting a somewhat paradoxical effect of IMC-EB10 in ALL that emphasizes the overriding contribution of ADCC toward its antileukemia effect. Although some ALL cell lines exposed to IMC-EB10 produce an induction rather than a suppression of FLT3 phosphorylation,22, 23 nevertheless, there is a strong antileukemic effect compared with that produced by IMC-NC7 in vivo because of the ADCC activity of IMC-EB10.23
Consistent with the in vitro data, IMC-EB10 also exhibits profound antileukemic effects in mouse models that harbor AML cells with different FLT3 genotypes. In human xenografts, exposure to single-agent IMC-EB10 significantly prolonged the survival of mice engrafted with human AML cells that expressed either wild-type or mutant FLT3.21 IMC-EB10 also reduced the engraftment of primary AML and ALL blasts in immunodeficient mouse models.20-23
Prolonged exposure of cells to small-molecule FLT3 inhibitors may lead to the selection of drug-resistant phenotypes. By contrast, to our knowledge, to date, in vivo exposure to IMC-EB10 has not lead to resistance. In typical experiments, leukemic cells that survived after exposure to IMC-EB10 remained sensitive to IMC-EB10 upon retransplantation. Moreover, leukemia cell clones that were generated specifically to become resistant to CEP-701 also were sensitive to the cytotoxicity of IMC-EB10.24 The precise mechanism of CEP-701-induced resistance in these cells is unknown; nevertheless, this observation could have profound implications for the future development of monoclonal antibodies targeting a receptor that no longer is inhibited by a previously active therapeutic agent.
In summary, several distinct mechanisms could account for the full antileukemic effects of IMC-EB10. In addition to direct blockade of the ligand and inhibition of signal transduction, the ability of IMC-EB10 to induce rapid and efficient internalization of surface receptors as well as ADCC could result in significant antileukemic effects in vivo. Finally, as a prelude to human trials, to our knowledge, IMC-EB10 has not exhibited any undesirable effects on nonleukemic cells or any overt toxicity in mice, and long-term treatment of mice with IMC-EB10 has produced no significant compromise of normal hematopoiesis.20-22
Clinical Development of IMC-EB10
Phase 1 and phase 2 clinical trials will test the validity of the antibody approach to FLT3 inhibition in AML. The primary objective of the phase 1 study will be to define the maximum tolerated dose and the pharmacokinetic profile of IMC-EB10 administered weekly to patients with AML. Patients who enter this trial may have failed to achieve complete remission to a previous standard induction regimen, may have developed recurrent disease after a response to previous antileukemia therapy, or may have been deemed ineligible for potentially curative or approved salvage options. Given preclinical data on the apparent activity of IMC-EB10 after the failure of small-molecule inhibitors, there will be no exclusion with regard to prior exposure to FLT3 inhibitors. An important exploratory objective will be to relate any observed clinical activity to the FLT3 genotype (ie, wild type or mutant). Depending on the outcome of the phase 1 trial, 1 or more phase 2 trials as monotherapy or in combination with conventional antileukemic agents may be launched.
If there is insufficient activity with a naked antibody approach, then another consideration might be the development of an immunoconjugate through the attachment of a cytotoxic payload to IMC-EB10. The ability of FLT3 to internalize after binding to an antibody makes this approach feasible. Several anti-FLT3 immunoconjugates currently are in development.25 The successful development of immunoconjugates may further enhance the therapeutic window of anti-FLT3 antibodies.
In conclusion, IMC-EB10 has demonstrated considerable promise in vitro and in animal models as an effective form of antileukemia therapy. The high therapeutic index associated with target-specific antibodies virtually devoid of off-target activities would be highly desirable in this setting. The additional gain of efficacy from ADCC could further enhance the value of the antibody strategy. The ability of IMC-EB10 to inhibit not only various genotypic isoforms of FLT3 but also AML cells that have become resistant to small-molecule inhibitors extends the range of application of this antibody in leukemia. Future clinical trials will examine these notions systematically. The eventual translation of these ideas to patients could have a significant impact on the future therapy of leukemia.
CONFLICT OF INTEREST DISCLOSURES
The articles in this supplement represent proceedings of the “12th Conference on Cancer Therapy with Antibodies and Immunoconjugates,” held in Parsippany, New Jersey, October 16-18, 2008. Unrestricted grant support for the conference was provided by Actinium Pharmaceuticals, Inc; Bayer Schering Pharma; Center for Molecular Medicine and Immunology; ImClone Systems Corporation; MDS Nordion; National Cancer Institute; National Institutes of Health; New Jersey Commission on Cancer Research; and PerkinElmer Life & Analytical Sciences. The supplement was supported by an unrestricted educational grant from ImClone Systems Corporation, a wholly-owned subsidiary of Eli Lilly and Company, and by page charges to the authors. The authors are full-time employees of ImClone Systems, a wholly owned subsidiary of Eli Lilly and Company.