ABSTRACT
PURPOSE: To validate the D'Amico risk group classification in French consecutive series and to analyse recurrence risk factor after radical prostatectomy (RP) for prostate cancer. MATERIAL: We retrospectively analyzed data collected from 10/2000 to 05/2009 for 730 consecutive patients who underwent RP for clinically localized or locally advanced prostate cancer (cT1-cT3) in our institution. Biochemical recurrence (BCR) was defined by prostate-specific antigen (PSA) of greater than 0.1 ng/ml, with rising PSA at two consecutive dosages. Postoperative survival was estimated using the Kaplan-Meier method after D'Amico's group stratification. The accuracy of the model was evaluated using the Harrell's concordance index. The impact on outcomes of preoperative PSA and pathological features was evaluated using a monovariate and multivariate Cox analysis. RESULTS: Median follow-up was 30 months [interquartile range: 13-51]. The overall 3-year and 5-year probability of freedom from BCR was 85% (95% Confidence Interval (CI), 81-88%) and 78% (95% CI, 74-83%), respectively. For low, intermediate, and high-risk group, the 5-year freedom from BCR was 92% (95% CI, 88-97%), 73% (95% CI, 65-81%) and 44% (95% CI, 28-59%), respectively (p<0.001). Harrell's concordance index was 0.71. Surgical margins were positive in 31% of all cases. In a multivariate analysis, preoperative PSA, pathological tumor stage, Gleason score and surgical margins status predicted BCR after RP. CONCLUSIONS: We externally validated the ability of the D'Amico's risk group stratification to predict disease progression following RP in European patients. Preoperative PSA, pathological stage, Gleason score and surgical margins status predicted BCR after RP in our series through a multivariate analysis.
