Leukemia (2013) 27, 260–268 & 2013 Macmillan Publishers Limited All rights reserved 0887-6924/13 www.nature.com/leu REVIEW FLT3 inhibition: a moving and evolving target in acute myeloid leukaemia AYH Leung, C-H Man and Y-L Kwong Internal tandem duplication (ITD) of the fms-like tyrosine kinase 3 (FLT3) gene is a gain-of-function mutation common in acute myeloid leukaemia (AML). It is associated with inferior prognosis and response to chemotherapy. Single base mutations at the FLT3 tyrosine kinase domain (TKD) also leads to a gain of function, although its prognostic significance is less well defined because of its rarity. The clinical benefits of FLT3 inhibition are generally limited to AML with FLT3-ITD. However, responses are transient and leukaemia progression invariably occurs. There is compelling evidence that leukaemia clones carrying both ITD and TKD mutations appear when resistance to FLT3 inhibitors occurs. Interestingly, the emergence of double ITD and TKD mutants can be recapitulated in vitro when FLT3-ITD þ leukaemia cell lines are treated with mutagens and FLT3 inhibitors. Furthermore, murine xenotransplantation models also suggest that, in some cases, the FTL3-ITD and TKD double mutants actually exist in minute amounts before treatment with FLT3 inhibitors, expand under the selection pressure of FLT3 inhibition and become the predominant resistant clone(s) during the drug-refractory phase. On the basis of this model of clonal evolution, a multipronged strategy using more potent FLT3 inhibitors, and a combinatorial approach targeting both FLT3-dependent and FLT3-independent pathways, will be needed to improve outcome. Leukemia (2013) 27, 260–268; doi:10.1038/leu.2012.195 Keywords: FLT3; internal tandem duplication; tyrosine kinase domain mutation; inhibition INTRODUCTION Clinically, FLT3-ITD occurs most frequently in AML with normal 2,3 Acute myeloid leukaemia (AML) is a group of heterogeneous karyotype, trisomy 8, t(6;9) and t(15;17), and is generally 17 diseases sharing in common an abnormal increase in myeloblasts associated with leukocytosis on presentation. Although the in the circulation and bone marrow. Intensive chemotherapy and complete remission (CR) rate does not appear to be affected, 18 allogeneic haematopoietic stem cell (HSC) transplantation are the AMLs with FLT3-ITD have higher relapse rates and therefore 19 principal treatment modalities. However, these approaches have inferior disease-free and overall survivals, particularly in AML 20 21 22 reached an impasse, with a cure rate of only 30–40%.1 Internal with larger ITD size, higher allelic burden and multiple ITDs. tandem duplication (ITD) of the fms-like tyrosine kinase 3 (FLT3) Therefore, FLT3 inhibition has become a legitimate therapeutic gene is a gain-of-function mutation, and represents one of the option, and clinical trials of FLT3 inhibitors in AML have been 23–25 most common genetic alterations in AML, occurring in nearly 30% ongoing for a decade. To date, more than 20 small molecule of cases.2,3 FLT3 is a class III receptor tyrosine kinase consisting of inhibitors against FLT3 have been reported, of which eight of 26 five extracellular immunoglobulin-like domains, a transmembrane them have been evaluated in clinical trials. Structurally, most of domain, a juxtamembrane domain and two tyrosine kinase these inhibitors are heterocyclic compounds, inhibiting FLT3 domains (TKD1 and TKD2).4 It is highly expressed in haemato- activity by competing with ATP for binding to the ATP-binding poietic stem and progenitor cells.5 Upon binding to FLT3 ligand, it pocket of the TKD. Functionally, they are generally multikinase dimerizes and autophosphorylates, resulting in activation of inhibitors. Their clinical activities appear to be mediated by FLT3 tyrosine kinase activity, which in turn activates the phosphoino- inhibition, so that their efficacies are limited to AML carrying FLT3- sitide-3-kinase/AKT and RAS/mitogen-activated protein kinase ITD, and correlated with inhibition of FLT3 phosphorylation and 27–31 pathways.6,7 FLT3-ITD involves an in-frame duplication of 3–400 hence its downstream signalling effectors. Although basepairs at the juxtamembrane or TKD1 domains. It results inhibition of FLT3 may be achievable, clinical efficacy is less than in alteration of cellular signalling, including the constitutive convincing, being limited by the invariable leukaemia progression 24 activation of FLT3 independent of ligand binding;8 activation of despite continuous treatment. Comprehensive reviews on 24,32–39 STAT5 (signal transducer and activator of transcription 5) via SRC the mechanisms of drug resistance have been published. kinase;9 phosphorylation of a transcription factor FOXO3A;10 The proposed mechanisms include persistent activation of FLT3 downregulation of the equilibrative nucleoside transporter 1 for signalling due to overexpression of FLT3 ligand and protein, cytarabine;11 and induction of reactive oxygen species production.12 activation of antiapoptotic signals and protection of leukaemia- These molecular consequences result in damage13 and defective initiating cells (LIC) by the bone marrow niche (Table 1). Although repair of DNA, and increased cellular proliferation and resistance to therapeutic strategies have been formulated on these proposi- apoptosis.14–16 Comprehensive description of normal and aberrant tions, they are based primarily on in vitro data and laboratory FLT3 signalling has been reviewed previously.4,14–16 observations. Division of Haematology and Bone Marrow Transplantation, Department of Medicine, Queen Mary Hospital, Hong Kong, China. Correspondence: Professor Y-L Kwong, Division of Haematology and Bone Marrow Transplantation, Department of Medicine, Professorial Block, Queen Mary Hospital, Pokfulam Road, Hong Kong, China. E-mail: [email protected] Received 11 May 2012; revised 26 June 2012; accepted 6 July 2012; accepted article preview online 16 July 2012; advance online publication, 3 August 2012 FLT3 inhibition AYH Leung et al 261 Table 1. Proposed mechanisms of resistance to FLT3 inhibitors Proposed mechanisms Proposed means of targeting References Overexpression of FLT3 ligands Targeting FLT3 ligand as a therapeutic strategy Sato et al.34 Overexpression of FLT3 protein Degrading FLT3 proteins by inhibition of chaperone heat-shock protein 90 Weisberg et al.37, Nishioka et al.81 and Al Shaer et al.82 Protection by bone marrow niche Reducing CXCL12 signalling in stromal cells by the HDM2 antagonist Nutlin-3a Kojima et al.32 via stabilization of p53 Disrupting interaction between LSCs and niche (for example, CXCR4 antagonists) Parmar et al.33 Over-riding microenvironment-mediated resistance to FLT3 inhibitors by combination Weisberg et al.83 treatment with dasatinib and JAK inhibitors Overexpression of antiapoptotic genes Targeting STAT pathway or survivin with inhibitors or shRNA Zhou et al.38 Targeting MCL-1 to restore sensitivity to FLT3 inhibitors Breitenbuecher et al.39 and Sto¨lzel et al.35 Emergence of resistant clones See text Abbreviations: FLT3, fms-like tyrosine kinase 3; JAK, janus kinase; LSC, leukaemia stem cells; shRNA, short hairpin RNA; STAT, signal transducer and activator of transcription. Table 2. Clinical trials of FLT3 inhibitors in FLT3-mutated AML patients Inhibitor Dosing schedulea Other drugs Number of patients Outcome References ITD TKD mutation Lestaurtinib 40–60 mg b.i.d. None 16 1 BR: 30%; HI: 16% Smith et al.30 Lestaurtinib 60 mg b.i.d., days 1–28 None 2 3 BR: 60%; HI: 60% Knapper et al.84 Lestaurtinib 80 mg b.i.d., days 7–112 Mitoxantrone, etoposide 103b 11b No improvement in survival Levis et al.49 cytarabine, days 1–5 Midostaurin 75 mg t.i.d. None 18 2 CR: 5%; BR: 35%; HI: 30% Stone et al.85 Midostaurin 50 mg b.i.d. None 12 6 BR: 67%; HI: 50% Fischer et al.28 Sunitinib 50–75 mg daily, days 1–28 None 2 2 CRi: 25%; PR: 75% Fiedler et al.51 80 mg b.i.d., days 29–58 Tandutinib 150–700 mg b.i.d. None 8 1 BR: 25%; HI: 25% DeAngelo et al.50 KW-2449 25–500 mg b.i.d., days 1–14 None 11 0 BR: 45% Pratz et al.52 Sorafenib 600 mg daily–400 mg b.i.d. None 7c 3b CRi: 11%; BR: 56% Zhang et al.53 Sorafenib 200–800 mg daily None 6 0 CR: 33%; BR: 66% Metzelder et al.86 Sorafenib 400–600 mg b.i.d., 14–28 days None 2 0 BR: 50% Pratz et al.87 Sorafenib 400 mg b.i.d., induction: 7 days; Idarubicin, cytarabine 13 2 CR þ CRp: 100% Ravandi et al.55 maintenance: 28 days AC220 200 mg daily None 62 0 CR þ CRp þ CRi: 45% Cortes et al.54 Sorafenib 400 mg b.i.d. None 13 0 CRi: 46%, nCRi: 46% Man et al.61 Sorafenib 400 mg b.i.d. None 65 0 CR þ CRi þ CMR: 38% Metzelder et al.88 HI þ BR þ PR: 61% Midostaurin 50 mg b.i.d. Daunorubicin Cytarabine 9 4 CR: 92% Stone et al. 56 Abbreviations: AML, acute myeloid leukaemia; b.i.d., twice daily; t.i.d., three times daily; BR, blast response; CMR, complete molecular response; CR, complete remission; CRp, CR without platelet recovery; CRi, CR with incomplete haematological recovery; FLT3, fms-like tyrosine kinase 3; HI, haematological improvement; ITD, internal tandem duplication; nCRi, near CRi but with focal blast prominence in trephine biopsy that could not be enumerated; PR, partial remission; TKD, tyrosine kinase domain. aUnless stated, continued until maximal response or unacceptable toxicity. All medications were orally administered. bTwo patients had both ITD and TKD mutation. cOne patient had both ITD and TKD mutation. There is now compelling clinical evidence that the resistance to full-blown AML when additional genetic alterations, such as FLT3 inhibitors may arise from the emergence of FLT3-ITD þ NUP98-HoxD13, are present.47 These data can be considered as clones carrying additional TKD mutations.