Multi-omics profiling links enhancer-associated MSH6 downregulation to platinum resistance, prognosis, and immune features in ovarian cancer
Platinum resistance is a major cause of poor outcome in ovarian cancer (OC), but biomarkers that reflect the proteomic and epigenetic basis of resistance remain limited. We aimed to develop a multi-omics signature associated with prognosis and to explore molecular features linked to chemoresistance in OC. Proteomic, transcriptomic, single-cell, spatial transcriptomic, and epigenomic data were integratively analyzed. A risk model was built using LASSO and Random Forest methods and validated across seven GEO datasets combined into a metavalidation cohort (n = 1049). MSH6 was further examined by ChIP-qPCR and immunohistochemistry in 71 OC samples. Immune-related associations were evaluated using IOBR and the IMvigor210 cohort. A five-protein signature (ARAF, ATM, MSH6, ASNS, SETD2) was associated with platinum resistance and survival in OC. The model consistently stratified patients into prognostically distinct risk groups across multiple cohorts. Single-cell and spatial analyses indicated that the high-risk group was enriched for stem-like epithelial features and showed reduced immune infiltration. Additional analyses supported an association between lower MSH6 expression and reduced H3K27ac enrichment at a putative enhancer locus. In the clinical cohort, low MSH6 protein expression was associated with platinum resistance and poor prognosis. Low MSH6 expression was also linked to immune-cold features and less favorable immunotherapy-related associations in external datasets. This study identifies a proteomics-derived multi-omics signature associated with platinum resistance and prognosis in OC and highlights MSH6 as a candidate marker linked to chemoresistant and immune-related features. Further mechanistic and clinical validation is needed.