Mutations in the FMS-like tyrosine kinase 3 (FLT3) receptor tyrosine kinase occur in ~30% of acute myeloid leukemias (AMLs) and are associated with a poor prognosis. Activating mutations in FLT3 predominantly occur as internal tandem duplications in the juxtamembrane region (FLT3-ITD), resulting in constitutive activation of canonical signaling pathways including STAT5, RAS/MAPK, and PI3K/mTORC1. Selective, potent 2nd and 3rd generation FLT3 inhibitors have shown significant clinical activity in patients with relapsed FLT3-ITD+ AML; however, clearance of FLT3 mutant clones does not consistently occur and most patients will eventually progress after an initial response. This challenges the model of FLT3 mutant AML being oncogene addicted and suggests that redundant signaling pathways may regulate AML cell survival after FLT3 inhibition. We hypothesized that these cells are resistant to FLT3 inhibition due to cytokines within the bone marrow microenvironment, resulting in low levels of primary resistant cells that will ultimately result in relapse. We have shown that primary FLT3-ITD+ AML cells escape FLT3 inhibition in vitro in the presence of cytokines produced normally in the bone marrow and that this effect is abrogated in patient cells by co-treatment with a JAK kinase inhibitor. To understand the mechanism of this rescue, we have studied a FLT3-ITD+ AML cell line (MV4;11). GM-CSF and IL-3 in particular are able to rescue cells from apoptosis following FLT3 inhibition. Interestingly, despite reactivating the canonical signaling pathways, GM-CSF and IL-3 maintain cell survival without rescuing proliferation, consistent with the clinical scenario. Both GM-CSF and IL-3 signal through JAK kinases. Co-treatment with the JAK1/2 inhibitor, ruxolitinib, or STAT5 knockdown re-sensitized AML cells to FLT3 inhibition when grown in the presence of cytokines. Inhibition of MEK or mTORC1 was unable to fully abrogate cytokine-mediated rescue of cell survival. These data demonstrate that JAK-dependent cytokine signaling mediate rescue from FLT3 inhibition through STAT5, but is not dependent on reactivation of MAPK or mTORC1 pathways. PIM kinases are a non-canonical target downstream of FLT3-ITD and downstream of STATs in cytokine signaling. PIM kinases are transcriptionally regulated. All three PIM isoforms were downregulated upon FLT3 inhibition. Both GM-CSF and IL-3 rescued PIM1 and PIM2 expression versus IL-6, a cytokine that is unable to protect from FLT3 inhibition. Co-treatment with a novel pan-PIM kinase inhibitor (INCB053914), was sufficient to block IL-3 rescue of cell survival. Altogether, these data demonstrate that FLT3-ITD+ AML cells utilize redundant signaling pathways activated by bone marrow cytokines through a JAK-STAT5-PIM axis to escape cell death from FLT3 inhibition. Pre-clinical testing with patient-derived xenografts is planned to validate these findings in an in vivo model and assess the feasibility of these combination approaches to improve clinical responses to FLT3 inhibitor therapy.


Sung: Radius Health Inc.: Equity Ownership. Koblish: Incyte Corporation: Employment, Equity Ownership. Perl: Seattle Genetics: Other: Advisory board; Astellas: Consultancy; Actinium Pharmaceuticals: Other: Scientific Advisory Board; Daiichi Sankyo: Consultancy; Novartis: Other: Advisory Board; Arog Pharmaceuticals: Consultancy; Pfizer: Other: Advisory Board; Asana Biosciences: Other: Scientific advisory board. Carroll: Incyte Pharmaceuticals: Research Funding; Astellas Pharmaceuticals: Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.