FLT3 Inhibitor Treatment Increases FLT3 Expression That Exposes FLT3-ITD+ AML Blasts to Elimination By FLT3 CAR-T Cells

Background: FMS-like tyrosine kinase 3 (FLT3) is a transmembrane protein uniformly expressed on leukemic blasts in acute myeloid leukemia (AML), and driver of leukemia-genesis in FLT3-ITD+ (Internal tandem duplication) AML. There is an increasing body of pre-clinical and clinical data suggesting that FLT3-ITD+ AML blasts respond to FLT3 inhibitor treatment by augmenting FLT3 expression in order to sustain the survival signal provided by this mutation. Here, we analyzed FLT3 expression on FLT3 wild type and FLT3-ITD+ AML cells after treatment with the FLT3 inhibitors midostaurin, quizartinib and crenolanib, and determined the antileukemia efficacy of combination treatment with FLT3 inhibitors and FLT3 CAR T cells in vitro and in vivo.

Methods: MOLM-13 and MV4;11 AML cells (both FLT3-ITD+) were cultured in the presence of IC50 doses of midostaurin, quizartinib and crenolanib, respectively to induce resistance (MOLM-13R/MV4;11R). A FLT3-CAR comprised of BV10 scFv binding domain, CD28-CD3ζ signal module and EGFRt marker was encoded in a lentiviral vector and expressed in CD8+ and CD4+ T cells of healthy donors and patients (n=6). T cell mediated cytolytic activity was evaluated in luminescence-based assay, cytokine production analyzed by ELISA and proliferation assessed by CFSE dye dilution. NSG mice (n= 4-6 per group) were engrafted with MOLM-13/ffLuc AML cells and treated with 5x10⁶ CAR T cells alone or in combination with FLT3 inhibitors.

Results: We detected a significant increase in FLT3 expression on both MV4;11 and MOLM-13 AML cells after treatment with each of the inhibitors as assessed by mean fluorescence intensity (quizartinib > crenolanib > midostaurin). The increase in FLT3 expression occurred specifically on these FLT3-ITD+ AML cell lines and was not observed on FLT3 wt AML (THP-1), acute lymphoblastic leukemia (NALM-16), mixed lineage leukemia (KOPN-8 and SEM) cell lines and normal hematopoietic stem cells. We applied single molecule sensitive super-resolution microscopy to demonstrate that the average number of FLT3 molecules (per micrometer sq.) on MV4;11 AML cells had increased from 0.80 (untreated) to 10.7 (quizartinib), 4.7 (crenolanib), and 3.3 (midostaurin) (p<.05). Of interest, midostaurin induced clustering of FLT3, while FLT3 was still present as monomers after quizartinib and crenolanib treatment. Intriguingly, the higher FLT3 density after FLT3 inhibitor treatment translated into superior antileukemia reactivity of FLT3 CAR T cells against AML cell lines and primary AML cells in vitro and in vivo. We observed the strongest increase in cytolytic activity, cytokine production and proliferation by CD8+ and CD4+ FLT3 CAR T cells after treatment with crenolanib and quizartinib, followed by midostaurin (p<.05).

We confirmed that upregulation of FLT3 occurred on MOLM-13 cells during FLT3 inhibitor therapy in NSG mice in vivo, and observed synergistic antileukemia efficacy of FLT3 CAR T cells in combination with each of the compounds. The mean frequency of FLT3 CAR T cells in mice that received FLT3 CAR T cells and an FLT3 inhibitor was 2-4 fold higher compared to mice had received FLT3 CAR T cells alone (p<.05) and was the highest in the cohort of mice that had received FLT3 CAR T cells in combination with crenolanib. FLT3 CAR T cells alone and each of the combination treatments of FLT3 CAR T & FLT3 inhibitor achieved 100% response rate which compares favorably to untreated or FLT3 inhibitor alone (0%). However, the mean fold reduction in leukemia burden (b/w day 7 and 10) was greater in all three combination treatment compare to only CAR treatment (p<.05). The most potent combination was FLT3 CAR T cells & crenolanib that accomplished the strongest reduction in leukemia burden as assessed by bioluminescence imaging and flow cytometry.

Conclusion: Collectively, the data show that FLT3 inhibitors augment cell surface expression of FLT3 in FLT3-ITD+ AML cells which leads to enhanced recognition and elimination by FLT3 CAR T cells. This is, to our knowledge, the first demonstration that small molecule inhibitors and CAR T cell immunotherapy can be used synergistically to treat a hematologic malignancy. We confirmed this principle with each of the FLT3 inhibitors in our panel, and observed the strongest antileukemia activity of FLT3 CAR T cells in combination with crenolanib. Our data encourage the clinical evaluation of this combination treatment in high risk patients with FLT3-ITD+ AML.