Validation of FLT3-ITD As a Therapeutic Target in Human Acute Myeloid Leukemia

Abstract

Abstract 937 Background. Activating in-tandem duplication (ITD) mutations within the FLT3 juxtamembrane region are detected in ∼25% of adult acute myeloid leukemia (AML) cases and portend a poor prognosis. AC220 (quizartinib) is a potent selective investigational FLT3/KIT inhibitor with encouraging preliminary clinical activity, as evidenced by a composite complete remission rate of 43% in 53 chemotherapy refractory/relapsed AML patients evaluable for an interim analysis of the ongoing phase II study in FLT3-ITD+ relapsed/refractory AML (Cortes et al, EHA 2011 abstract 1019). Many patients who initially respond ultimately suffer disease progression. We sought to utilize the clinical efficacy of AC220 to determine if FLT3-ITD is a valid therapeutic target in human AML. A saturation mutagenesis strategy identified amino acid substitutions at three amino acid residues (F691, D835, Y842) in FLT3-ITD that confer a high degree of resistance to AC220 in vitro. Of these, substitutions at F691 and D835 conferred the greatest degree of relative resistance. Results. To assess the validity of FLT3-ITD as a therapeutic target in AML, we analyzed paired pretreatment and relapse samples obtained from a cohort of 9 FLT3-ITD-positive AML patients enrolled in the exploratory part of the ongoing phase II study of AC220 in relapsed/refractory AML (clinicaltrials.gov NCT00989261) who relapsed after initially achieving morphologic clearance of bone marrow blasts to <5% on treatment. Through RT-PCR subcloning and sequencing of the FLT3 kinase domain, we observed evolution of one or more acquired kinase domain mutations on the FLT3-ITD containing allele in all 9 cases at relapse. Clinically detected mutations were restricted to the “gatekeeper” residue F691 and/or the activation loop residue D835 (F691L, n=3, D835Y, n=4; D835V, n=2, D835F, n=1). Each of these mutations successfully transformed Ba/F3 cells to IL-3 independence and conferred substantial resistance to AC220 in cell growth and cell-based biochemical assays. All of the AC220-resistant mutations conferred substantial in vitro cross-resistance to sorafenib, which has been reported to harbor clinical activity in a small number of FLT3-ITD-associated AML cases.In an expanded analysis of genomic DNA samples from 30 patients enrolled in the exploratory part of the Phase II study who discontinued study drug for any reason, we observed the occurrence of acquired mutations in the kinase domain (D835 and F691) in a total of 10 of 30 (33%) patients at the off study timepoint. One patient had a D835Y mutation prior to going on AC220.Molecular docking studies were undertaken to provide mechanistic insights into the structural basis of resistance conferred by AC220-resistant mutations. These studies revealed that AC220 likely binds strongly to the DFG-out inactive FLT3 kinase domain. AC220 directly interacts with the gatekeeper residue F691, explaining the drug-resistance associated with the F691L mutation. Mutations at D835 and Y842 may potentially de-stabilize the inactive conformation, and result in a more active, AC220 binding-deficient FLT3 kinase. Indeed, substitutions at these residues are known to activate FLT3 in the absence of an ITD mutation.To more precisely assess the frequency and identity of resistance-conferring mutations at relapse, we analyzed a subset of samples using single molecule real-time (SMRT™; Pacific Biosciences, Menlo Park, CA) sequencing, which can generate reads of sufficient length to enable focused interrogation of the kinase domain of FLT3-ITD alleles. With this assay, more than 350 reads of >1000 nucleotides were reliably obtained. Analyses of pretreatment and relapse samples from four patients confirmed the presence of resistance-conferring FLT3-ITD kinase domain mutations at F691 or D835 at relapse in 36–60% of FLT3-ITD sequence reads. Additionally, this method detected polyclonal resistance in two of the four samples assessed. Conclusions. Our studies validate FLT3-ITD as a therapeutic target in a proportion of AML cases, and demonstrate that the clinical activity of AC220 is mediated by FLT3-ITD inhibition. AC220-resistant FLT3-ITD gatekeeper and activation loop mutations identified in clinical samples from relapsing patients represent high-value therapeutic targets for next-generation FLT3 inhibitors. Disclosures:Off Label Use: AC220 is an investigational agent and has no approved drug indication in AML. Chin:Pacific Biosciences: Employment. Hunt:Ambit Biosciences: Employment. Levis:Ambit Biosciences, Inc: Consultancy. Travers:Pacific Biosciences: Employment. Wang:Pacific Biosciences: Employment. Kasarskis:Pacific Biosciences: Employment, Equity Ownership. Schadt:Pacific Biosciences: Employment, Equity Ownership.