Abstract FLT3 mutations (mFLT3) in acute myeloid leukemia (AML) carry a poor prognosis.The mandatory co-occurrence of other driver mutations with mFLT3 leads to suboptimal clinical responses with single agent FLT3 inhibitors (FLT3i). Pre-existing or emergent RAS mutations are one mechanism of primary and secondary resistance to single agent FLT3i. Therefore development of combinatorial strategies to overcome resistance conferring co-mutations is critical. Herein we report a novel in-vitro combination of the FLT3/SYK inhibitor (FSI) TAK-659 with Proteasome inhibitor (PI) ixazomib, to co-operatively inhibit proliferation of mFLT3 and RAS or WT1 co-mutated AML blasts, which are otherwise less sensitive to Flt3i and FSI. Peripheral blood or bone marrow derived cells from patients with AML were isolated and treated in vitro with TAK-659 and Ixazomib alone or in combination. Cytosolic/nuclear protein lysates were obtained at 12 hours of drug/control treatment and western blotting performed for proteins of interest. Tritiated thymidine proliferation assays were also performed with titrated doses of each agent alone, and combination of fixed dose of one agent with titration of the second. Patient samples were annotated for mutations by Next Generation Sequencing. Similar experiments were carried out on cell lines MV411 (Flt3ITD/MLL), THP-1 and OCI-AML3 (both NRAS). Proliferation assays on a series of primary AML samples revealed an inherent dichotomy in response patterns. mFLT-3 samples with co-mutations in DNMT3A, NPM1 or tMLL were very sensitive to TAK-659 (IC50 <75 nM) but much less sensitive to Ixazomib alone (IC50>100nM). The co-occurrence of RAS or WT1 mutations predicted for resistance to TAK-659 but interestingly sensitized cells to PI alone and in combination. In fact, presence of RAS mutations irrespective of FLT3 status predicted for sensitivity to PI (IC50<50nM). We also observed sensitivity to Ixazomib in THP-1 and OCI-AML3, while MV4-11 was sensitive to TAK-659 but relatively resistant to Ixazomib alone. Based on genome-wide CRISPR screen, overactivity of RAC-PAK signaling mediated by PREX1 is essential to RAS mutated AML (T Wang etal. Cell 168:890, 2017). Similarly, WT1 mutation in AML is known to ablate expression of a major nuclear chaperone for beta-catenin, TBL1X, and RAC has been found to be a substitute (Y Wang et al. Mol Cell. 57:662, 2015). We therefore hypothesized that PI decreased nuclear availability of RAC1 along with its chaperone partner beta-catenin - a determinant of leukemic stem cell activity, also regulated by RAC/PAK and the proteasomal system. Indeed, in primary RAS or WT1 mutated AML samples, we observed PI treatment led to significant reduction in nuclear RAC1 levels and decreased nuclear beta-catenin levels, specifically PAK phosphorylated p675 active beta-catenin. These findings suggest PI effects RAC1, PAK and beta-catenin simultaneously. Similar results were also noted in RAS mutant cell lines. siRNA knockdown experiments also linked p62 SQSTM1 in a pathway to RAS/RAC/PAK and beta-catenin. In summary, we delineate a RAS/RAC/PAK/Beta-catenin pathway to be linked to critical events in RAS or WT1 mutated AML progression and resistance. FLT3 mutated AML with these co-mutations are insensitive to single-agent activity of FLT3i or FSI alone and the addition of PI is critical to overcome them. Citation Format: Sravanti Rangaraju, Santosh Kumar Pasupuleti, Silvia Marino, Daniela Nicoleta Petrusca, Larry D. Cripe, Bhaskar Ramdas, Lakshmi Reddy Palam, Katie J. Sargent, Hamid Sayar, Heiko Konig, Reuben Kapur, David G. Roodman, Scott H. Boswell. Exploiting the proteasome to overcome RAS and WT1 mutation mediated resistance to FLT-3 inhibition in acute myeloid leukemia [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 800.
Read full abstract