Promoter hypermethylation as an independent prognostic factor for relapse in patients with prostate cancer following radical prostatectomy.

PURPOSE: To analyze the prognostic significance of six epigenetic biomarkers (APC, Cyclin D2, GSTP1, TIG1, Rassf1A, and RARbeta2 promoter hypermethylation) in a homogeneous group of prostate cancer patients, following radical prostatectomy alone. PATIENTS AND METHODS: Biomarker analyses were done retrospectively on tumors from 74 prostate cancer patients all with a Gleason score of 3 + 4 = 7 and minimum follow-up period of 7 years. Using quantitative methylation-specific PCR, we analyzed six gene promoters in primary prostate tumor tissues. Time to any progression was the primary end point, and development of metastatic disease and/or death from prostate cancer was a secondary point. The association of clinicopathologic and biomolecular risk factors to recurrence was done using the log-rank test and Cox proportional hazards model for multivariate analysis. To identify independent prognostic factors, a stepwise selection method was used. RESULTS: At a median follow-up time of 9 years, 37 patients (50%) had evidence of recurrence: biochemical/prostate-specific antigen relapse, metastases, or death from prostate cancer. In the final multivariate analysis for time to progression (TTP), the significant factors were age > 60 [hazard ratio (HR), 0.4; 95% confidence interval (95% CI), 0.2-0.8; P = 0.01], hypermethylation of GSTP1 (HR, 0.23; 95% CI; 0.09-0.64; P = 0.004), and hypermethylation of APC (HR, 3.0; 95% CI, 1.42-6.32; P = 0.004). In another multivariate analysis, a profile of hypermethylation of APC and cyclin D2 hypermethylation was significant as well: if either any one was hypermethylated (HR, 1.84; 95% CI, 0.92-3.72; P = 0.09) or if both were hypermethylated (HR, 4.3; 95% CI, 1.52-12.33; P = 0.01). CONCLUSIONS: Methylation status of selected genes in the prostate cancer specimen may predict for time to recurrence in Gleason 3 + 4 = 7 patients undergoing prostatectomy. These results should be validated in a larger and unselected cohort.