BACKGROUND: Prostate cancer (PCa) is one of the major causes of cancer-induced death among males. Here, we applied integrated bioinformatics analysis to identify key prognostic factors for PCa patients.

MATERIAL AND METHODS: The gene expression data were obtained from the UCSC Xena website. We calculated the differentially expressed genes between PCa tissues and normal controls. Pathway enrichment analyses found cell cycle-related pathways might play crucial roles during PCa tumorigenesis. The genes were assigned into 22 modules established via weighted gene co-expression network analysis (WGCNA).

RESULTS: The results indicated that the purple and red modules were obviously linked to the Gleason score, pathological N, pathological T, recurrence, and recurrence-free survival (RFS). In addition, Kaplan-Meier curve analysis found 8 modules were markedly correlated with RFS, serving as prognostic markers for PCa patients. Then, the hub genes in the most 2 critical modules (purple and red) were visualized by Cytoscape software. Pathway enrichment analyses confirmed the above findings that cell cycle-related pathways might play vital roles during PCa initiation and progression. Lastly, we randomly chose the PILRβ (also termed PILRB) in the red module for clinical validation. The immunohistochemistry (IHC) results showed that PILRβ was significantly increased in the high-risk PCa population compared with low-/middle-risk patients.

CONCLUSIONS: We used integrated bioinformatics approaches to identify hub genes that can serve as prognosis markers and potential treatment targets for PCa patients.

Keywords: Immunochemistry, Prostatic Neoplasms, Biomarkers, Tumor, Computational Biology, Databases, Genetic, Gene Regulatory Networks, Neoplasm Grading, Software