THE COMBINATION OF PREOPERATIVE PROSTATE-SPECIFIC ANTIGEN AND POSTOPERATIVE PATHOLOGICAL FINDINGS TO PREDICT PROSTATE-SPECIFIC ANTIGEN OUTCOME IN CLINICALLY LOCALIZED PROSTATE-CANCER

Purpose: The independent clinical and pathological predictors of time to postoperative prostate specific antigen (PSA) failure were used to identify prostate cancer patients at high risk for this end point. Mat erials and Methods: A Cox regression multivariate analysis was used to determine the prognostic significance of preoperative PSA, pathologic al stage, prostatectomy Gleason score and margin status in predicting the time to postoperative PSA failure in 862 men with palpable (T2) or PSA detected (Tlc) prostate cancer. The 2-year PSA failure rates with 95% confidence intervals were calculated using the results of Cox reg ression analysis and a bootstrap procedure with 2,000 replications, re spectively, and are presented in nomogram format stratified by preoper ative PSA, pathological stage, prostatectomy Gleason score and margin status. Results: Preoperative PSA (p = 0.0001), pathological stage (p less than or equal to 0.002), margin status (p = 0.0001) and prostatec tomy Gleason score (p = 0.034) were independent predictors of time to postoperative PSA failure, Conclusions: Patients at high risk for earl y PSA failure could be identified postoperatively on the basis of preo perative PSA and prostatectomy pathology. Adjuvant therapy trials in t hese select patients may be justified.