Identification of genes that predict the biochemical recurrence of prostate cancer.

Prostate cancer (PCa) is one of the most prevalent cancer types in men. Biochemical recurrence continues to occur in a large proportion of patients after radical prostatectomy. Thus, prognostic biomarkers are required to determine which treatment is suitable. In the present study, RNA-sequencing gene expression data from The Cancer Genome Atlas was used in order to develop a risk-score staging system based on the expression of eight genes. Cox multivariate regression was used to predict the outcome of patients with PCa. The biomedical recurrence-free survival of patients with low-risk scores was significantly longer compared with patients with high-risk scores (P=5x10(-7)). This result was further validated using another dataset, GSE70769, from the National Center for Biotechnology Information. The prognostic values of other clinical information and risk scores were evaluated for 5-year biochemical recurrence. The prognostic value of the risk score was determined using an area under curve value of 0.819, predicting the 5-year biochemical recurrence of patients with PCa. The risk score was identified to be significantly associated with primary tumor stage (P<0.01), Gleason score (P<0.01), and lymph node invasion (P<0.05), but was independent of age. Cox multivariate regression revealed that the risk score was an indicator for prediction of biochemical recurrence. Thus, the risk score is a valuable and robust indicator for predicting the biochemical recurrence of patients with PCa.