Abstract
Basement membrane plays an important role in tumor invasion and metastasis, which is closely related to prognosis. However, the prognostic value and biology of basement membrane genes (BMGs) in prostate cancer (PCa) remain unknown. In the TCGA training set, we used differentially expressed gene analysis, protein-protein interaction networks, univariate and multivariate Cox regression, and least absolute shrinkage and selection operator regression to construct a basement membrane-related risk model (BMRM) and validated its effectiveness in the MSKCC validation set. Furthermore, the accurate nomogram was constructed to improve clinical applicability. Patients with PCa were divided into high-risk and low-risk groups according to the optimal cut-off value of the basement membrane-related risk score (BMRS). It was found that BMRS was significantly associated with RFS, T-stage, Gleason score, and tumor microenvironmental characteristics in PCa patients. Further analysis showed that the model grouping was closely related to tumor immune microenvironment characteristics, immune checkpoint inhibitors, and chemotherapeutic drug sensitivity. In this study, we developed a new BMGs-based prognostic model to determine the prognostic value of BMGs in PCa. Finally, we confirmed that THBS2, a key gene in BMRM, may be an important link in the occurrence and progression of PCa. This study provides a novel perspective to assess the prognosis of PCa patients and provides clues for the selection of future personalized treatment regimens.