Combination of Menin and IDH Mutant Inhibition in Acute Myeloid Leukemia

The interaction between menin and mixed lineage leukemia 1 (MLL1) protein contributes to abnormal proliferation in the MLL1-(KMT2A) rearranged and nucleophosmin 1 (NPM1)-mutated acute myeloid leukemia (AML). Indeed, small molecule inhibition of the menin/MLL1 interaction was shown to inhibit leukemia in both AML subtypes. Approximately 25% of NPM1 mutated AML also carry an IDH 1 or 2 mutation. Carrying both NPM1- and IDH mutations is associated with worse overall survival and relapse free survival, compared to having only the NPM1 mutation. MI-3454, a small molecule inhibitor of MLL1/menin interaction, has been shown to be effective in AML with KMT2A rearrangements and NPM1 mutations. A close analogue of this compound, KO-539 (Ziftomenib) is now being tested in clinical trials in refractory and relapsed AML patients (NCT 04067336). The aim of this study is to determine the efficacy of the combination of MI-3454 and IDH mutant inhibitors in AML carrying concurrent NPM1 and IDH1/2 mutations. Blood or bone marrow aspirate samples from patients with AML harboring both NPM1 and IDH1 and/or IDH2 mutations were obtained and used to assess the effect of combinatorial treatment of menin and IDH mutant inhibitors. Colony formation assays with these AML patient samples were performed in methylcellulose media with cells treated with a concentration range of MI-3454, ivosidenib (IDH1 inhibitor) or enasidenib (IDH2 inhibitor) as single agents and in combinations. At the end point of the experiments, colonies were counted. Cells were collected for 1) cytohistology to assess cell morphology 2) flow cytometry to evaluate markers of differentiation and apoptosis and 3) gene expression studies to assess the effect of the combinations on the expression level of genes implicated in leukemogenesis and differentiation. In the samples with concurrent NPM1 and IDH1 mutations, the combination of MI-3454 and ivosidenib caused synergistic reduction in colony number compared to the single agents, with combination indices (CI) ranging from 0.44 to 0.94. Furthermore, cells treated with this combination demonstrated more differentiating phenotypes than single agent treated cells. In the samples with concurrent NPM1 and IDH2 mutations, the combination of MI-3454 and enasidenib caused synergistic reduction in colony formation versus either single agent MI-3454 or enasidenib, with CI values 0.12-0.64. The combination of these agents also yielded stronger differentiation than single agents alone based on both morphologic assessment and enhanced expression of CD11B differentiation marker. Mechanistically, the combination of MI-3454 and IDH1/2 mutant inhibitors also more significantly decreased the expression of HOXA9, MEIS1, and HOXB2, which are oncogenes implicated in leukemogenesis, over single agents. The expression of MNDA was also significantly enhanced by these combinations, supporting differentiation. Overall, our study demonstrates that combination of MI-3454 and IDH mutant inhibitors outperforms the effect of single agents with respect to their anti-leukemic effects in clinically relevant pre-clinical models. These results support that combination of menin and IDH mutant inhibitors could be utilized in clinical trials to improve the ability of patients with AML to achieve remission.