Crenolanib (CP-868,596) Is a Potent and Selective Type I FLT3 Inhibitor That Retains Activity Against AC220 Resistance-Causing FLT3 Kinase Domain Mutants

The clinically active FLT3 inhibitor AC220 is vulnerable in vitro to resistance-conferring mutations at 3 residues in the FLT3-ITD kinase domain: the “gatekeeper” residue F691, and activation loop (AL) residues D835 and Y842. Mutations at 2 of these residues, F691 and D835, were identified in 8/8 FLT3-ITD+ acute myeloid leukemia (AML) patients who relapsed on AC220, including mutations at residue D835 in 6/8 patients. A molecular docking analysis suggests that mutations at D835 favor the active “DFG-in” kinase conformation and thereby impair binding of AC220, which presumably requires an inactive conformation for binding (Smith, et al, Nature 2012). Therefore, we predict that a type I inhibitor capable of binding the “DFG-in” active conformation of FLT3 will be required to inhibit AC220-resistant FLT3-ITD AL mutants. Crenolanib (CP-868,596) is a potent, selective ATP-competitive inhibitor of the FLT3-related receptor tyrosine kinases PDGFR-a and -b. Notably, crenolanib retains activity against the imatinib-resistant PDGFR-a D842V mutation, which is analogous to the AC220-resistant FLT3-ITD/D835V mutation. Low micromolar concentrations of crenolanib have been safely achieved in a phase I study of solid tumor patients with a half-life of ∼14 hours (Lewis et al, JCO 2009). We hypothesized that crenolanib may be a Type I inhibitor of FLT3 that retains activity against FLT3 mutant isoforms, including AC220-resistant FLT3 AL mutants, which are highly cross-resistant to multiple FLT3 TKIs.

In vitro binding studies demonstrate that crenolanib binds preferentially to the phosphorylated form of ABL (K_d =140nM vs K_d=440nM for non-phosphorylated ABL), confirming crenolanib is a Type I inhibitor. Additionally, crenolanib potently binds native FLT3 in vitro (K_d=0.26nM) and retains affinity for FLT3 harboring substitutions at D835 (H/V/Y; K_d= 0.24, 0.048 and 0.26nM respectively). Crenolanib demonstrates substantially more potent in vitro binding affinity for the compound FLT3-ITD/D835V mutant than AC220 (K_d=0.05nM vs K_d=210nM). In cellular assays, crenolanib induces apoptosis and inhibits the proliferation of the patient-derived FLT3-ITD+ cell lines MV4;11 and Molm14 with an inhibitory concentration 50 (IC50) of 5.2 and 9nM, respectively. FLT3 autophosphorylation and downstream signaling in MV4;11 and Molm14 cells were inhibited at similar concentrations. Treatment with crenolanib prolonged survival in a murine bone marrow transplant model of FLT3-ITD+ leukemia. Crenolanib inhibits the proliferation of BaF/3 cells transformed with FLT3-ITD (IC50 7.8 nM), and retains activity in BaF/3 cells harboring highly AC220-resistant FLT3-ITD/D835V/Y/F and FLT3-ITD/Y842C/H mutants (IC50 15–19nM). Crenolanib also potently suppresses the growth of BaF/3 cell lines containing the FLT3-activating point mutations D835V and D835Y in the absence of ITD (IC50 3.1nM), which we have recently found to be associated with AC220 resistance (Smith et al, ASH 2012, submitted). The FLT3-ITD/F691L mutation confers modest resistance to crenolanib (IC50 49.7nM). Western blot analysis reveals dose-dependent decrease in FLT3 autophosphorylation and downstream signaling. Crenolanib potently inhibits the proliferation of an AC220-resistant Molm14 subclone that harbors a D835Y mutation (IC50 15.4nM). In these cells and native Molm14 cells, crenolanib appears to retain maximal biochemical inhibition of FLT3 autophosphorylation and downstream signaling at nanomolar concentrations in human plasma, indicating relatively low plasma protein binding. Finally, treatment with crenolanib inhibited FLT3 autophosphorylation in human primary FLT3-ITD+ AML cells, including those from a patient who developed resistance to AC220 associated with a D835 mutation.

Crenolanib is a Type I inhibitor of FLT3 that retains activity in the low nanomolar range against native and AC220-resistant FLT3-ITD mutant isoforms in in vitro binding studies, cell line and murine leukemia models, as well as in primary human AML cells. Crenolanib therefore has the potential to be clinically active in AML patients with activating FLT3-ITD or AL mutations, and to recapture clinical response in patients with acquired AC220-resistant kinase domain mutations. Clinical trials of crenolanib in TKI-naïve and TKI-pretreated FLT3-mutant AML are currently being planned or have recently been initiated.

Perl:Astellas Pharmaceuticals: Consultancy. Carroll:GlaxoSmithKlein: Research Funding. Shah:Ariad: Consultancy, Research Funding.