Combinatorial inhibition of LSD1 and Menin induces therapeutic differentiation in AML

Acute myeloid leukemia (AML) is characterized by differentiation arrest and uncontrolled proliferation. Differentiation therapy aims to treat AML by de-repressing latent myeloid maturation programs to induce cell cycle arrest and subsequent cell death. This approach is curative in the promyelocytic AML subtype, but has met with limited success in other subtypes. Genes such as LSD1 have emerged as intriguing non-APL AML differentiation therapy targets, but results as monoagents in clinical trials have been mixed. Here, we performed differentiation-specific CRISPR screens to identify targets whose inhibition synergizes with LSD1 inhibition to induce terminal differentiation of non-APL AML cells. Intriguingly, the MLL co-factor Menin scored as the top hit. Using cell lines, primary patient samples, and mouse AML models, we find that dual inhibition of LSD1 and Menin is a highly promising approach for differentiation therapy. Mechanistically, we determine that inhibition of Menin downregulates drivers of proliferation and stemness such as MEIS1, and inhibition of LSD1 induces inflammatory and interferon-related pro-myeloid differentiation expression programs. Surprisingly, we find that this combination is effective in selected AML models without mutations in MLL or NPM1, thus nominating dual inhibition of LSD1 and Menin as an attractive therapeutic approach for a mutationally diverse set of non-APL AMLs.