Since the tumor protein p53 (TP53), a transcription factor, plays a crucial role in prostate cancer development and progression, we hypothesized that sequence variants in TP53 binding sites might affect clinical outcomes in patients with prostate cancer. We systematically evaluated 41 single nucleotide polymorphisms (SNPs) within genome-wide predicted TP53 binding sites in a cohort of 1,024 prostate cancer patients. The associations of these SNPs with prostate cancer characteristics and clinical outcomes after radical prostatectomy for localized disease and after androgen-deprivation therapy (ADT) for advanced disease were assessed by Kaplan-Meier analysis and Cox regression model. ARAP2 rs1444377 and TRPS1 rs722740 were associated with advanced stage prostate cancer. FRK rs171866 remained as a significant predictor for disease progression; DAB2 rs268091 and EXOC4 rs1149558 remained as significant predictors for prostate cancer-specific mortality (PCSM); and EXOC4 rs1149558 remained as a significant predictor for all-cause mortality after ADT in multivariate models that included clinicopathologic predictors. In addition, the numbers of protective genotypes at DAB2 rs268091 and EXOC4 rs1149558 showed a cumulative effect on PCSM (P for trend = 0.002). Our results suggested that SNPs within TP53 binding sites might be valuable biomarkers for prostate cancer outcome prediction.