The activated receptor tyrosine kinase FLT3 appears to play a central role in the pathogenesis of AML and is an exciting therapeutic target. Several FLT3-inhibitory agents are now progressing through clinical trials
Department of Haematology
Despite gradual refinements in chemotherapy combinations and major improvements in supportive care strategies over the last 30 years, long-term survival is still achieved in less than half of young adults with acute myeloid leukaemia (AML). This figure falls to less than 15% in older patients for whom the delivery of intensive therapy is frequently not achievable due to frailty or co-morbidity. Increasing knowledge of the biology of AML has led to the development of several novel classes of targeted agents. These include epigenetic-acting drugs (histone deacetylase inhibitors, azacytidine), farnesyl transferase inhibitors, the anti-CD33 immunoconjugate gemtuzumab ozogamicin (Mylotarg) and inhibitors of the tyrosine kinase receptor FLT3. This article will review the evolution of FLT3 inhibitors and discuss issues pertinent to their future development and clinical/implementation.
The membrane-spanning tyrosine kinase receptor FLT3 (FMS-like tyrosine kinase 3) represents a particularly attractive therapeutic target in AML. FLT3 is overexpressed in over 90% of AML patients and appears to play a central role in disease pathogenesis. Phosphorylation of FLT3 triggers activation of downstream signalling targets including proteins in the STAT, MAP kinase and AKT pathways which are involved in regulation of cell proliferation, transcription and apoptosis. FLT3-activating mutations, including internal tandem duplication (ITD) mutations within the juxtamembrane region and single-base point mutations within the tyrosine kinase domain of the receptor are among the commonest molecular lesions so far described in AML, with a combined prevalence of 30-35%.[5,6] Both types of FLT3 mutation cause autophosphorylation of the receptor, switching it to its active configuration. It is now well-established that FLT3 ITD mutations are associated with poor disease prognosis, in terms of both increased relapse risk and reduced overall survival. In patients that lack a FLT3 mutation, the overexpressed wild type FLT3 receptor may still contribute to leukaemogenesis.
The overwhelming success of the BCR/ABL kinase inhibitor imatinib mesylate in chronic myeloid leukaemia (CML) has served as proof of principle that tyrosine kinase inhibitors can elicit significant clinical responses in haematological malignancy. Imatinib monotherapy induces sustained haematological and cytogenetic remissions in the vast majority of patients with chronic phase CML. A note of caution should be sounded in AML, however, where unlike BCR/ABL in CML, FLT3 is only one among several genetic ‘hits’ that combine to cause the disease; similar to the situation in CML blast crisis or Philadelphia positive acute lymphoblastic leukaemia where BCR/ABL inhibition with imatinib produces favourable clinical responses but not outright cures.
Large scale drug discovery programmes in the pharmaceutical industry have identified more than 20 tyrosine kinase inhibitors with varying degrees of specificity towards FLT3. Most of these are heterocyclic compounds that compete with ATP for binding to the ATP binding pocket of the FLT3 receptor. Candidate compounds tend to move through a welltrodden series of pre-clinical laboratory evaluations during which FLT3 inhibitory activity is established by cell-based phosphorylayion assays, the ability to induce cytotoxicity is assessed by incubating with FLT3-expressing cell lines or primary patient material, and mouse models of FLT3-dependent leukaemia are used to demonstrate prolongation of survival in animals treated with the drug.
At least 10 compounds with FLT3 inhibitory activity have entered clinical trials at the time of writing(published results are summarised in Table 1). Some of these are highly selective for FLT3 while others additionally affect a range of other kinases. Most of the early phase clinical trials have involved administration of FLT3 inhibitors as single agents to relapsed or treatment-refractory AML patients, including at least some patients with FLT3 mutations. Although these studies have shown that inhibition of FLT3 is, in general, safe and well-tolerated, clinical responses have been broadly similar across the studies and limited in both depth and duration, including clearances of leukaemic blasts from the peripheral blood, temporary reductions in bone marrow blast numbers or periods of blood transfusion independence lasting from weeks to months. Complete remissions have only been seen in occasional patients. Importantly, the correlative laboratory results from these studies have provided good evidence linking sustained inhibition of FLT3 with clinical response. In view of the relatively modest clinical responses seen with FLT3 inhibitor monotherapy to date, even allowing for the fact that the above phase 1/2 trials included notoriously poor prognosis, heavily pre-treated or elderly patients, it appears that, to achieve their maximum clinical benefit, FLT3 inhibitors will need to be combined with drugs from other classes. There is a strong pre-clinical rationale to suggest a synergistic relationship between FLT3 inhibitors and conventional cytotoxic agents and, until targeted therapies become available for other genetic ‘hits’ that contribute to the pathogenesis of AML, the best therapeutic approach is likely to lie in combining FLT3 inhibitors with chemotherapy. Cell cycle dynamics need to be carefully considered to optimise drug sequencing as many chemotherapeutic agents require cells to be actively proliferating. ‘Pre-treatment’ with FLT3 inhibitors antagonises the effects of chemotherapy by decreasing the percentage of actively cycling cells. In contrast, synergistic in vitro cytoxicity has been demonstrated when FLT3 inhibitors are administered after chemotherapy; this finding has directly influenced the design of current drug combination studies.
The two FLT3 inhibitory compounds that have progressed furthest through clinical development to date are the indolocarbazole alkaloids lestaurtinib (Cephalon) and midostaurin (Novartis). Both of these agents are currently under assessment in combination with chemotherapy within ongoing international phase 3 trials.
Lestaurtinib (formerly CEP701) was initially identified as an inhibitor of TrkA, a member of the nerve growth receptor subfamily, but subsequently also found to potently inhibit FLT3 with an IC50 of 3nM., Lestaurtinib has good oral bioavailability and was well tolerated in two early phase clinical trials with the commonest reported toxicities being nausea and diarrhoea. In a US phase 1/2 trial, 17 heavily-pretreated patients with FLT3-mutated relapsed/refractory AML were treated with lestaurtinib at a dose of 40-80 mg twice daily. Five patients showed transient reductions in circulating blasts, with one of these patients also showing a decrease in bone marrow blasts to <5%. In a UK-based phase 2 study 29 newly-diagnosed elderly AML patients were treated with lestaurtinib at a twice-daily dose of 60 mg for 8 weeks with dose escalation to 80 mg if well tolerated. Transient clinical responses were seen in three out of five evaluable patients with mutated FLT3 and five of 22 wild type FLT3 patients. Detailed correlative laboratory analysis running alongside each of these trials confirmed that clinical responses could be predicted if a patient’s blasts were inherently sensitive to the drug (as determined by in vitro cytotoxicity assay) and the patient maintained plasma drug levels sufficient for sustained inhibition of FLT3 to <15% of baseline activity.
The combination of lestaurtinib with chemotherapy has recently been under evaluation in two international phase 3 studies. The US-based Cephalon 204 Study recently closed to recruitment after randomising 220 AML patients with FLT3-activating mutations in first relapse to receive salvage chemotherapy either alone or in combination with oral lestaurtinib, which was given after each course of chemotherapy. Preliminary reports showed encouraging rates of complete remission that correlated well with laboratory assays of FLT3 inhibition. Further results are expected in late 2009.
In late 2006, the UK-based NCRI AML15 study for newly-diagnosed adult patients below the age of 60 years opened a randomisation that allocated half of the patients identified to harbour a FLT3 mutation on diagnostic screening to receive lestaurtinib following each cycle of induction and consolidation chemotherapy. The AML15 trial recently completed recruitment, but lestaurtinib treatment will continue to be evaluated in its successor, the AML17 trial. Lestaurtinib also has inhibitory activity at nanomolar concentrations against JAK2 and early phase clinical studies are in progress in myelofibrosis and polycythaemia vera.
Midostaurin (formerly PKC412) was originally developed as an inhibitor of protein kinase C but, like lestaurtinib, was subsequently found to more-potently inhibit FLT3 (IC50 approximately 10nM). Midostaurin also has good oral bioavailability and proved to be well-tolerated and displayed clinical activity in a phase 2 study. 14 out of 20 patients with mutated FLT3 and either relapsed/refractory AML or high risk myelodysplastic syndrome showed >50% reductions in peripheral blood blast count, with bone marrow blasts falling to <5% in two cases (without blood count recovery). Again, response durations were short. The recently-opened international phase 3 RATIFY (Randomised AML Trial In FLT3 in <60 year olds) study will assess the effects of adding midostaurin (or placebo) to standard induction and consolidation chemotherapy in newly-diagnosed adults with mutant FLT3.
While the results of the above chemotherapy combination studies are keenly awaited, several questions fundamental to the future development of FLT3 inhibitors remain to be fully addressed. Many of the ‘firstgeneration’ of FLT3-inhibitory drugs, including both lestaurtinib and midostaurin actually have relatively broad kinase inhibition profiles in comparison to ‘cleaner’ more FLT3-selective ‘second generation’ agents such as AC220. As AML is a complex multigenic disease, the simultaneous inhibition of other important tyrosine kinases, such as those that stimulate angiogenesis, by broader spectrum drugs may be therapeutically advantageous but could come at the price of increased toxicity.
Most FLT3-inhibitory compounds are highly hydrophobic and bind strongly to plasma proteins resulting in greatly decreased free levels of the drug being available to bind to target, and much of the heterogeneity in clinical response to FLT3 inhibitors may be attributable to variations in patient metabolism with alpha-1 acid glycoprotein levels appearing to vary both between patients and in individual patients at different timepoints in the course of their disease. Dosing schedules may need to be individually tailored to allow for these factors.
The clinical experiences with imatinib in CML show us that acquired drug resistance is a major challenge facing the implementation of tyrosine kinase inhibitors. Resistance may potentially arise, as in CML, through the acquisition of secondary tyrosine kinase domain point mutations that interfere with drug binding.Other potential mechanisms of resistance include upregulation of expression of FLT3 or its ligand and the activation of alternative FLT3-independent cell survival and proliferation pathways.,
Finally, FLT3 mutations have been identified within the leukaemic stem cell population in a majority of patients with FLT3-ITD AML. These stem cell populations are increasingly felt to be central to the aetiology of disease relapse and it is important that future clinical studies address the role of post-remission maintenance FLT3 inhibitor therapy in their eradication.
Aberrant FLT3 signalling plays a major role in leukaemogenesis in a significant proportion of AML patients making FLT3 a tantalising target for novel therapeutic intervention. Although clinical responses to FLT3 inhibitor monotherapy have so far been relatively modest in depth and duration, early phase clinical studies have confirmed the clinical anti-leukaemic efficacy of a strategy of sustained FLT3 inhibition in FLT3-dependant AML. Given that the pathogenesis of AML involves multiple genetic ‘hits’, inhibition of FLT3 is much more likely to impact on cure rates if combined with other existing or novel treatment modalities. The results of major international phase 3 trials of lestaurtinib and midostaurin used in combination with combination chemotherapy regimens are eagerly awaited. Ultimately it is hoped that, as our knowledge of the complex molecular pathogenetic mechanisms underlying AML increases, FLT3 inhibitors will be successfully combined with a range of other molecularly targeted agents, avoiding some of the toxicities associated with traditional chemotherapy and opening up the prospect of ‘curative therapy’ to a greater proportion of patients.
1. Lowenberg B, et al. N Engl J Med 1999;341:1051-1062.
2. Estey E. Semin Oncol 2008;35:439-448.
3. Rosnet O, et al. Leukemia 1996;10:238-248.
4. Stirewalt DL, et al. Nat Rev Cancer 2003;3:650-665.
5. Nakao M, et al.Leukemia 1996;10:1911-1918.
6. Yamamoto Y, et al.Blood 2001;97:2434-2439.
7. Levis M, et al.Leukemia 2003;17:1738-1752.
8. Kantarjian H, et al. N Engl J Med 2002;346:645-652.
9. Sawyers CL. Cancer Cell 2002;1:413-415.
10. Knapper S. Br J Haematol 2007;138:687-699.
11. Levis M, et al. Blood 2004;104:1145-1150.
12. Levis M, et al. Blood 2002;99:3885-3891.
13. Knapper S, et al. Blood 2006;108:3494-3503.
14. Smith BD, et al. Blood 2004;103:3669-3676.
15. Knapper S, et al. Blood 2006;108:3262-3270.
16. Levis M, et al. Blood 2005;106:403a.
17. Weisberg E, et al.Cancer Cell 2002;1:433-443.
18. Stone RM, et al.Blood 2005;105:54-60.
19. Zarrinikar, et al. Blood 112, 859a. 2008. Ref Type: Generic
20. Chu SH, et al. Drug Resist Updat 2009;12:8-16
21. Cools J, et al. Cancer Res 2004;64:6385-6389.
22. Piloto O, et al. Blood 2007;109:1643-1652.
23. Levis M, et al. Blood 2005
24. Fiedler W, et al. Blood 2003;102:2763-2767.
25. Giles FJ, et al. Cancer 2003;97:1920-1928.
26. Fiedler W, et al. Blood 2005;105:986-993.
27. DeAngelo DJ, et al. Blood 2006;108:3674-3681.
28. Zhang W, et al. J Natl Cancer Inst 2008;100:184-198.
29. Pratz KW, et al. Blood 2009;113:3938-3946.