Recent genomewide association studies have identified several prostate cancer susceptibility variants. However, the association between these variants and biochemical failure in prostate cancer patients receiving radical prostatectomy has not been determined. We systematically evaluated 20 prostate cancer-associated single-nucleotide polymorphisms in a cohort of 320 localized prostate cancer patients receiving radical prostatectomy. Each single-nucleotide polymorphism found to be associated with the recurrence of prostate-specific antigen was further analyzed by Kaplan-Meier analysis and Cox regression model. Three prostate cancer susceptibility single-nucleotide polymorphisms (rs1447295 at 8q24, rs7920517 and rs10993994 at 10q11) were associated with prostate-specific antigen recurrence (P < 0.02). Of these, rs7920517 and rs10993994, which were in strong linkage disequilibrium (r(2) = 0.91), also showed significant associations with poor prostate-specific antigen-free survival following radical prostatectomy (log-rank test; P < 0.01). The associations remained significant in our multivariate Cox proportional hazards analysis after adjusting for other clinicopathologic risk covariates (P < 0.01). In conclusion, loci associated with risk for prostate cancer, such as rs7920517 and rs10993994, might also be used to predict the recurrence of prostate-specific antigen in prostate cancer patients receiving radical prostatectomy.