Crenolanib is a potent inhibitor of FLT3 with activity against resistance-conferring point mutants

Abstract

Mutations of the type III receptor tyrosine kinase FLT3 occur in approximately 30% of acute myeloid leukemia patients and lead to constitutive activation. This has made FLT3-activating mutations an attractive drug target because they are probable driver mutations of this disease. As more potent FLT3 inhibitors are developed, a predictable development of resistance-conferring point mutations, commonly at residue D835, has been observed. Crenolanib is a highly selective and potent FLT3 tyrosine kinase inhibitor (TKI) with activity against the internal tandem duplication (FLT3/ITD) mutants and the FLT3/D835 point mutants. We tested crenolanib against a panel of D835 mutant cell lines and primary patient blasts and observed superior cytotoxic effects when compared with other available FLT3 TKIs such as quizartinib and sorafenib. Another potential advantage of crenolanib is its reduced inhibition of c-Kit compared with quizartinib. In progenitor cell assays, crenolanib was less disruptive of erythroid colony growth, which may result in relatively less myelosuppression than quizartinib. Finally, correlative data from an ongoing clinical trial demonstrate that acute myeloid leukemia patients can achieve sufficient levels of crenolanib to inhibit both FLT3/ITD and resistance-conferring FLT3/D835 mutants in vivo. Crenolanib is thus an important next-generation FLT3 TKI.

Trial registration: ClinicalTrials.gov NCT01657682.

Figures

Figure 1

Crenolanib. (A) The structure of crenolanib, a benzamidine-quinolone derivative. (B) KinomeScan results for crenolanib. Shown are the results of all kinases in the panel inhibited 50% or more (compared with baseline activity) by 100 nM crenolanib.

Figure 2

Activity of crenolanib against WT and ITD-mutated FLT3 in vitro. (A) FLT3/ITD cell lines (Molm14 and MV411) and FLT3 WT cell lines (SEMK2) were treated with crenolanib for 1 hour and then cells were lysed, immunoprecipitated for FLT3, and analyzed by immunoblotting for phospho- and total FLT3. (B) An aliquot of the lysate of Molm14s treated with crenolanib from (A) was reserved for analysis of downstream signaling molecules including pAKT, pMAPK, pSTAT5, and the corresponding total protein content. (C) Cytotoxicity of these doses of crenolanib in FLT3/ITD cell lines was analyzed by MTT assay. HL-60 cells, which express extremely limited WT FLT3, were used as a control. (D) Apoptosis induced by crenolanib and sorafenib 48 hours after treatment of Molm14 cells with drug was analyzed by Annexin V flow cytometry. (E) Patient blasts containing a FLT3/ITD mutation were treated with crenolanib, lysed, and immunoblotted for phospho- and total FLT3.

Figure 3

Inhibitory activity of crenolanib against FLT3 TKD mutations in primary AML samples. (A) Blasts from patient 2 (de novo D835Y), patient 3 (de novo D835Y), and patient 4 (relapsed D835V) were incubated for 1 hour with crenolanib and then lysed, immunoprecipitated for FLT3, and analyzed for phospho- and total FLT3 by immunoblotting. (B) Blasts from the same patients in panel 2 were treated with crenolanib in 96-well plates for 72 hours, and cytotoxicity was analyzed by MTT. (C) Blasts from patient 4 were also treated with quizartinib and sorafenib in this analysis. In a separate immunoblot experiment (inset), blasts from patient 4 were treated with 20 nM of quizartinib, crenolanib, and sorafenib for 1 hour and then lysed, immunoprecipitated for FLT3, and analyzed by immunoblot for phospho- and total FLT3. (D) Patient 5 had a FLT3/ITD mutation and was treated with sorafenib, responded to treatment, and then relapsed. Blasts collected after relapse harbored a D835F mutation, along with the original FLT3/ITD mutation. These blasts were treated with crenolanib, quizartinib, or sorafenib in 96-well plates for 72 hours, and cytotoxicity was analyzed by MTT.

Figure 4

Inhibition of c-KIT and erythropoiesis. (A) TF-1 cells, which express WT c-Kit, were treated with crenolanib or quizartinib for 1 hour. In the last 5 minutes of drug treatment, 20 ng of stem cell factor was added. The cells were lysed, immunoprecipitated for c-Kit, and analyzed by immunoblot for phospho- and total c-Kit. (B) Normal human donor bone marrow (n = 3) was collected and diluted to a concentration of 100 000 cells per mL in MethoCult. Various concentrations of crenolanib or quizartinib were added, and cells were plated in quadruplicate in 35-mm dishes. Each dish was viewed under a light microscope, and total numbers of granulocyte-macrophage colony-forming unit and erythrocyte burst-forming unit colonies were recorded.

Figure 5

Crenolanib inhibits FLT3 in vivo. (A) Plasma samples from patients 6 and 7 were collected as part of the phase 2 clinical trial of crenolanib. Patients received a single dose of drug on day 1, and samples were collected for PIA and pK analysis. Molm14 cells were incubated with each plasma sample for 1 hour and then lysed, immunoprecipitated for FLT3, and analyzed by immunoblot for phospho- and total FLT3. (B) The same time course samples from these patients were analyzed by mass spectroscopy for serum drug concentrations of crenolanib. (C) Blasts from patient 4 (relapsed D835V) were used for a PIA assay with steady-state plasma samples from 5 different patients enrolled in the crenolanib trial. Blasts were incubated in plasma for 1 hour and then lysed, immunoprecipitated for FLT3, and analyzed by immunoblot for phospho- and total FLT3. Densitometry was performed on this blot and serum concentrations of crenolanib from the corresponding time points were determined by pK analysis. In the figure, the densitometry result (expressed as a percentage of control) and the crenolanib concentration for a given time point are listed directly below each sample on the blot.