Targeting menin in T-lineage acute lymphoblastic leukemia

T-lineage acute lymphoblastic leukemia (T-ALL) lacks effective targeted therapies, with poor outcomes in relapsed/refractory disease. HOXA-high T-ALL is biologically aggressive and often resistant to standard therapy. Menin inhibitors, recently approved for KMT2A-rearranged leukemias, may be effective in T-ALL, but biomarkers of response remain undefined. This study aims to evaluate the efficacy of menin inhibition in T-ALL and identify molecular predictors of sensitivity. We tested menin inhibitors (ziftomenib, revumenib, VTP-50469) in 14 primary T-ALL samples and 8 cell lines, representing HOXA-high and HOXA-low genotypes. In vitro sensitivity assays, xenograft mouse models, transcriptomics, proteomics, and phosphoproteomics were used to characterize drug response. MEF2C modulation experiments and combination studies with CDK1/2 and ERK1/2 inhibitors were performed in vitro and in vivo. Menin inhibitors suppressed leukemic growth in a subset of HOXA-high and HOXA-low primary human T-ALL samples. Similarly, ziftomenib was effective in reducing tumor burden in xenografts without major toxicity. Upon treatment, we observed down-regulation of canonical menin targets (HOXA, MEIS1, MEF2C) and upregulation of T-cell differentiation programs. Phosphoproteomic studies identified MEF2C S222 phosphorylation-mediated by CDK1/2 and ERK1/2-as a predictor of ziftomenib sensitivity in T-ALL. MEF2C overexpression promoted proliferation and ziftomenib resistance, while knockdown impaired growth. Ziftomenib synergized with CDK1/2 and ERK1/2 inhibitors in vitro and improved survival in xenografted mice. In conclusion, a subset of T-ALL, defined by high p-MEF2C S222, is sensitive to menin inhibition. Combining ziftomenib with CDK or ERK inhibition offers synergistic efficacy, supporting biomarker-driven clinical trials of this strategy in relapsed/refractory T-ALL.