Feed

Translating high-resolution multiomics data into clinically actionable biomarkers is critical for overcoming therapeutic resistance and tumor heterogeneity in prostate adenocarcinoma (PRAD). To decode the complex immunosuppressive tumor microenvironment (TME) and identify robust prognostic targets, we developed a systematic biomarker discovery pipeline integrating single-cell RNA sequencing (scRNA-seq) mapping and high-dimensional network analysis. By deconvoluting scRNA-seq profiles from over 35,000 PRAD cells, nonnegative matrix factorization (NMF) of the malignant epithelial compartment revealed nine distinct transcriptional metaprograms (MPs). High-dimensional weighted gene coexpression network analysis (hdWGCNA) pinpointed PRAD-MP7 as the core proliferative engine and nominated the malignant-specific gene YBX1 as the master prognostic hub. To establish clinical utility evidence, we validated YBX1 across six independent global PRAD cohorts, where its overexpression robustly predicted poor overall survival (OS) and relapse-free survival (RFS). In vitro functional validation via siRNA-mediated knockdown in DU-145 and PC-3 cells significantly attenuated proliferative and invasive capacities, impairing cell viability and downregulating key progression markers (Ki-67, MMP2, and MMP9). Crucially, immunogenomic profiling mapped YBX1 expression to an "immune-excluded" TME, characterized by depleted CD8+ T cell and dendritic cell infiltration alongside elevated immune checkpoint networks. Serving as a bridge to clinical translation, YBX1 effectively predicted clinical responses in three immunotherapy cohorts and demonstrated broad resistance to 12 chemotherapeutic and targeted agents. Our multiomics integration pipeline highlights YBX1 as a dual-functional oncogene that couples malignant proliferation with immune evasion, establishing it as a highly translational biomarker and an actionable target for precision PRAD management.