Deciphering pathogenesis and therapeutic vulnerabilities in germline RUNX1 mutated AML - PROJECT SUMMARY RUNX1 mutations have been identified in 10-20% of acute myeloid leukemia (AML) patients with up to a third being reported as germline, characterized by chemoresistance and poor prognosis. Natural history and genomic studies suggest that germline RUNX1 mutated patients typically present with three disease stages: 1) an initially indolent stage of familial platelet disorders (FPD) harboring monoallelic germline RUNX1 mutation, 2) an insidious pre-leukemic stage carrying biallelic RUNX1 mutations and 3) an end stage of fulminant, leukemic transformation (FPD-AML) with acquisition of additional somatic mutations often involving signaling effectors (most commonly FLT3). Dissecting how germline RUNX1 mutations cooperate with these somatic events to drive leukemic transformation (which remains unknown) will shed new light on AML pathogenesis and therapies. Faithful preclinical models including genetically engineered mouse models and patient derived xenografts, critical for studying leukemogenesis and therapeutic development, are largely lacking for FPD-AML. We have recently developed a model of FPD to AML progression by temporally introducing somatic events in mice carrying Runx1R188Q/Flox;Flt3ITD-frt/+;RosaFlpoER;Mx-Cre alleles in which sequential inactivation of Runx1 followed by Flt3ITD activation recapitulated the genetic events and disease progression in FPD patients. Furthermore, biallelic but not monoallelic Runx1 mutations are critical for transformation in this model. We observed that multipotent progenitors (MPPs) in this context upregulate self-renewal transcriptional programs and serially propagate leukemia upon transplant. We further employed CRISPR dropout screens and identified the histone H3 lysine 4 (H3K4) methyltransferases Mll4 and Mll5 as candidate dependencies in Runx1-mutant cells. Lastly, Mll5 mRNA expression and H3K4me3 levels are increased in biallelic Runx1 mutated cells. Thus, we hypothesize that biallelic Runx1 mutations enable an H3K4me3 driven epigenetic and transcriptional cell state in MPPs through the activity of Mll4 and/or Mll5, which further cooperates with subsequent oncogenic events (e.g. Flt3ITD) to activate self-renewal programs resulting in leukemic transformation. We will determine the mechanisms by which mutant Runx1 alone and in cooperation with Flt3-ITD drives leukemic transformation in FPD and interrogate therapeutic efficacy of inhibiting Mll4/Mll5 and its effector H3K4me3 in FPD-AML. Completion of these studies will: 1) provide greater insights into the epigenetic and transcriptional mechanism of leukemic transformation in FPD; 2) delineate the role of RUNX1 mutations in FPD-AML; 3) inform whether strategies targeting mutant RUNX1 and epigenetic regulators (MLL4/MLL5/H3K4me3)) may have therapeutic relevance in preventing or reversing myeloid transformation, thus paving the road to developing novel clinical-grade therapies. The gained knowledge may be extrapolated to other types of AML and RUNX1-mutated malignancies.
