Combinatorial Targeting of XPO1 and FLT3-ITD Exerts Synergistic Anti-Tumor Effects in FLT3-Mutated Acute Myeloid Leukemias

Internal tandem duplication (ITD) mutations of the fms-like tyrosine kinase-3 (FLT3) gene are common in patients with acute myeloid leukemia (AML) and associated with poor disease prognosis (Kottaridis et al., 2003). FLT3-targeted therapies using novel small-molecule inhibitors such as sorafenib are under development with some notable effects (Adnane et al., 2005). However, acquired resistance to these agents is frequent following long-term therapy. One of the main resistance mechanisms is the acquisition of point mutations in the tyrosine kinase (TK) domains (TKD) ofFLT3, in addition to ITD mutations (Smith et al., 2012; Zhang et al., 2014). Another novel target for AML therapy is exportin-1 (XPO1), one of the major transporters of proteins and mRNAs from the nucleus. High expression of XPO1 is associated with decreased overall survival in AML patients (Kojima et al., 2013). and its levels were increased in patients with FLT3-ITD mutations. Interestingly, targeting CRM1 with the small molecule inhibitor KPT-185 down-regulated FLT3 expression (Ranganathan et al., 2012). The XPO1 inhibitor selinexor (PKT-330) has been developed recently and has shown encouraging anti-tumor effects against AML-initiating cells in primary human AML-engrafted NSG mice (Etchin et al., 2015) and in ongoing clinical trials.

In this study, we investigated the anti-tumor activity of combined treatment with sorafenib and selinexor in FLT3-mutant AML cells, including those with ITD and TKD point mutations, (Smith et al., 2012; Zhang et al., 2014). Selinexor alone exerted marked cell killing in human and murineFLT3-mutant AML cells, including those harboring D835Y or ITD with Y842C or F691L mutations. Interestingly, FLT3 and MAPK or FLT3 and AKT signaling pathways were activated by the selinexor treatment for 24 h in murine FLT3-mutant AML cells. Combination treatment with selinexor and sorafenib demonstrated synergistic or additive pro-apoptotic effects in TKD point- and FLT3-ITD-mutanted AML cells, respectively. Mechanistically, the combination treatment triggered significant suppression of phospho-ERK and phospho-AKT compared to selinexor treatment alone in all examined AML cell lines. Of note, the combination treatment promoted the retention of ERK, AKT, NFκB, and FOXO3a the in nucleus, which could explain the low levels of phospho-ERK and phospho-AKT observed in the cytosol. In addition, a 5-day in vitro combination treatment with extremely low 5 to 10 nM concentrations of each agent promoted myeloid differentiation of FLT3-ITD mutant MOLM14 cells. No cell killing was observed using this treatment scenario. Furthermore, the combination treatment demonstrated profound anti-tumor efficacy in a human FLT3-mutated leukemia model in NOG mice. The median survival was doubled from 16 days in the vehicle-treated mice to 31 or 23 days with selinexor or sorafenib treatments, respectively, and this effect was increased to 51 days in the mice that received the combination treatment (p = 0.0001).

Our findings provide a pre-clinical basis for a combinatorial treatment strategy targeting XPO1 and FLT3-ITD in FLT3-mutated AML patients, especially in those patients who have acquired resistance to FLT3-targeted therapy.