Multi-omics driven immune classification of colorectal cancer: Implications for immunotherapy efficacy prediction and enhancement with WNT signaling inhibition.

The heterogeneity of tumor immune microenvironment accounts for differential prognosis and immunotherapy responsiveness among colorectal cancer (CRC) patients. Here, we developed a novel immune classification for colorectal cancer through integrative multi-omics analysis. Immune-related gene expression, mutation, and methylation profiles were collected from two large-scale multi-omics CRC cohort (TCGA, n = 627; COCC, n = 977) to perform multi-omics factor analysis (MOFA) and establish the Multi-Omics Tumor Immune Features-based Clusters of CRC (MotifCC). The molecular differences and tumor immune variations among different clusters were analyzed and experimentally validated. Potential molecular targets for converting immune-cold tumors into hot tumors were validated with cell lines, patient-derived organoids (PDOs), and animal models. Integrative multi-omics analysis revealed three novel clusters with distinct characteristics. Cluster1, with high WNT pathway activation, had the worst prognosis and a tumor-promoting microenvironment. This immune-cold subtype may benefit from a combination of immunotherapy and WNT-targeted treatment. Cluster2 was characterized by low immune infiltration and high glycolysis intensity. These patients may benefit from treatment with metabolism inhibitors. Cluster3 was distinguished by high gene methylation, high tumor mutation burden, microsatellite instability, and better response towards immunotherapy. We successfully established the MotifCC, a novel multi-omics immune classification system for CRC. This system effectively captures the multi-omics heterogeneity of CRC, providing valuable insights for tailoring immunotherapy strategies and potentially improving patient outcomes.