Predicting prostate specific antigen outcome preoperatively in the prostate specific antigen era

Purpose: We evaluated the ability of previously defined risk groups to predict prostate specific antigen (PSA) outcome 10 years after radical prostatectomy in patients diagnosed with clinically localized prostate cancer during the PSA era. Materials and Methods: Between 1989 and 2000, 2,127 men with clinically localized prostate cancer underwent radical prostatectomy, including 1,027 at Hospital of the University of Pennsylvania (study cohort) and 1,100 at Brigham and Women's Hospital (validation cohort). Cox regression analysis was done to calculate the relative risk of PSA failure with the 95% confidence interval (CI) in patients at intermediate and high versus low risk. The Kaplan-Meier actuarial method was used to estimate PSA outcome 10 years after radical prostatectomy. Results: Compared with low risk patients (stages Tlc to 2a disease, PSA 10 ng./ml. or less and Gleason score 6 or less) the relative risk of PSA failure in those at intermediate (stage T2b disease or PSA greater than 10 to 20 ng./ml. or less, or Gleason score 7) and high (stage T2c disease, or PSA greater than 20 ng./ml. or Gleason score 8 or greater) risk was 3.8 (95% CI 2.6 to 5.7) and 9.6 (95% CI 6.6 to 13.9) in the study cohort, and 3.3 (95% CI 2.3 to 4.8) and 6.3 (95% CI 4.3 to 9.4) in the validation cohort. The 10-year PSA failure-free survival rate in the 1,020 patients in the low, 693 in the intermediate and 414 in the high risk groups was 83%, 46% and 29%, respectively (p <0.0001). Conclusions: Based on 10-year actuarial estimates of PSA outcome after radical prostatectomy 3 groups of patients were identified using preoperative PSA, biopsy Gleason score and 1992 clinical T category.