BACKGROUND:The extracellular matrix (ECM) plays a critical role in the tumor microenvironment (TME). However, the prognostic relevance of matrisome-related genes (MRGs) in bladder cancer (BLCA) remains poorly understood. This study aimed to establish a matrisome-related gene signature for prognostic stratification in bladder cancer and to further characterize its associations with tumor microenvironmental features and candidate compounds.
METHODS:We analyzed transcriptomic and clinical data from TCGA and GEO datasets to identify prognostic MRGs. Consensus clustering identified distinct MRG clusters. Subsequently, a nine-gene prognostic signature (SERPINE2, RBP7, SPINK4, CD3D, CLDN5, TSPAN8, FADS1, CLIC3, and KRT4) was developed using univariate Cox, LASSO, and multivariate Cox regression analyses. Immune infiltration, RNA stemness score (RNAss), and transcriptome-based predicted drug sensitivity were assessed in high- and low-risk groups. Molecular docking was performed to explore structurally plausible drug-target interactions.
RESULTS:The signature demonstrated robust predictive accuracy for overall survival in internal and external cohorts. High-risk patients exhibited higher tumor stage and grade, a stroma- and myeloid-enriched TME, and distinct ECM remodeling patterns. Differentially expressed genes were enriched in ECM-receptor interaction and immune-related pathways. Drug susceptibility analysis identified 16 compounds, including NU7441 and Luminespib, with lower predicted IC50 values in the high-risk group. Molecular docking further supported the structural plausibility of interactions between these compounds and selected risk genes, although these findings should be interpreted as exploratory.
CONCLUSIONS:This study highlights the prognostic relevance of MRGs in BLCA. The nine-gene signature may serve as a useful framework for risk stratification in BLCA, while the identified risk genes and candidate compounds provide a basis for further biological and experimental investigation rather than direct therapeutic inference.