PREDICTORS OF PATHOLOGICAL STAGE BEFORE NEOADJUVANT ANDROGEN WITHDRAWAL THERAPY AND RADICAL PROSTATECTOMY

Purpose: This prospective randomized trial was used to compare predict ive factors for organ confined margin negative status after radical pr ostatectomy with and without a 3-month course of neoadjuvant androgen withdrawal therapy. Materials and Methods: A total of 213 patients wit h localized adenocarcinoma of the prostate were randomized to radical prostatectomy with or without a 3-month course of 300 mg. neoadjuvant cyproterone acetate daily. Multivariate logistic regression analysis w as used to determine significant predictors of organ confined margin n egative status after radical prostatectomy in both groups. Parameters evaluated included baseline prostate specific antigen (PSA 4 or less, 4.1 to 10, greater than 10 ng./ml.), clinical stage (T2c versus T2b or less), biopsy Gleason score and percentage of surface area of biopsie s involved with cancer. The multivariate analysis was repeated with PS A density and the natural logarithm of PSA to optimize the model. Resu lts: In the radical prostatectomy alone arm a model incorporating only PSA density was the best predictor of organ confined margin negative status. In the neoadjuvant androgen withdrawal therapy arm a model inc orporating biopsy Gleason score, PSA density and clinical stage was th e best predictor. Conclusions: The conventional predictors of patholog y at radical prostatectomy, biopsy Gleason score, PSA density and clin ical stage retain significance as predictors in patients treated with a 3-month course of neoadjuvant androgen withdrawal therapy before rad ical prostatectomy.