Examines KMT2 mutations defining gene expression states in colorectal cancer, suggesting implications for targeted therapy.
3671 Background: The KMT2A–D (MLL1–4) family of H3K4 methyltransferases are critical epigenetic regulators of enhancer landscapes and cell identity. At least one of the genes is mutated in 20-25% of colorectal cancer. Their role in shaping the tumor microenvironment (TME) remains distinct from their established roles in leukemia, where they are better described. We tested whether KMT2 mutations are associated with specific immune/stromal phenotypes that could inform therapeutic stratification. Methods: We analyzed a retrospective institutional CRC cohort using targeted NGS (n = 2,267; 565 KMT2 -mutant [MUT], 1,702 wild-type [WT]) and a transcriptomic subset with matched RNA-seq (n = 353; 73 MUT, 280 WT) from primary and metastatic specimens. Differential expression (DESeq2) and pathway enrichment (Hallmark GSEA) compared MUT vs. WT, adjusting for batch and tissue site. VST-based module z-scores (cancer associated fibroblast [CAF]/stroma, epithelial, IFN response, cell-cycle) were computed. xCell, EPIC, CIBERSORTx, and ESTIMATE scores were analyzed by linear regression on z-scored outcomes (MUT vs WT) adjusted for batch and tissue site; Benjamini–Hochberg false discovery rate (FDR). β denotes the adjusted mean difference in standardized score (MUT–WT) in SD units. Overall survival was assessed by Cox proportional hazards models. Results: KMT2 -MUT tumors demonstrate a unique tumor biology, with higher epithelial (β = 0.30, FDR = 0.011), IFN response (β = 0.24, FDR = 0.021), and cell-cycle (β = 0.24, FDR = 0.021) modules with lower CAF/stroma (β = −0.29, FDR = 0.011). Consistent with this, GSEA showed enrichment of IFN-α/γ and proliferation programs (MYC/E2F/G2M) and depletion of epithelial-mesenchymal transition (EMT), myogenesis, and apical junction pathways (FDR < 0.05). TME composition tools demonstrated lower stromal content [lower CAFs (EPIC, β = −0.288, FDR = 0.102), lower fibroblasts/stroma scores (xCell, β≈−0.36/−0.35, FDR = 0.079), and trend toward lower StromalScore (ESTIMATE, β = −0.24, FDR = 0.20)], and trends to higher Th2 (xCell, β = 0.40, FDR = 0.059) and NK cells (xCell, β = 0.38, FDR = 0.079). After adjusting for age, sex, ECOG performance status, primary tumor site, stage, RAS and BRAF mutation status, and MSI status, overall survival did not differ significantly between tumors with and without KMT2 family mutations, although there was a trend toward improved outcomes (HR 0.85, 95% CI 0.70–1.00; p = 0.078). Conclusions: KMT2 family mutations in CRC characterize a tumor subset with retained epithelial identity, active proliferation, and interferon signaling, but markedly reduced stromal infiltration and EMT programs. This distinct biology suggests KMT2 loss may impede the acquisition of aggressive mesenchymal features, highlighting a potential biomarker for stromal-modulating or immune-based therapies.
No takes yet. Share an insight, caveat, or question.
Payapwattanawong et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: