AN ANALYSIS OF THE TIME-COURSE OF POSTOPERATIVE PROSTATE-SPECIFIC ANTIGEN FAILURE IN PATIENTS WITH POSITIVE SURGICAL MARGINS - IMPLICATIONSON THE USE OF ADJUVANT THERAPY

Objectives. The role of adjuvant therapy in the postprostatectomy sett ing for positive margin patients is an unresolved issue. The purpose o f this study is to provide the rationale for patient selection in Phas e III trials that test the impact of adjuvant therapy on survival in p ositive margin prostate cancer patients. Methods. Early (12 months or less) and delayed (more than 12 months) postoperative prostate-specifi c antigen (PSA) failure have been correlated with distant and local fa ilure, respectively, as the site of first failure. In this study, a Co x regression multivariate analysis was used to determine the significa nt independent clinical and pathologic predictors of-early and delayed postoperative PSA failure in 143 margin-positive prostate cancer pati ents. Results. Margin-positive patients with positive pelvic lymph nod es, seminal vesicle invasion, or prostatectomy Gleason sum 8 or higher were excluded. For the remaining patients, a prostatectomy Gleason su m of 7, preoperative PSA more than 20 ng/mL, and an endorectal coil ma gnetic resonance imaging (erMRI) scan showing extensive disease were s ignificant independent predictors of early postoperative PSA failure. Conversely, a prostatectomy Gleason sum of 6 or less, preoperative PSA 20 ng/mL or less, and an erMRI showing limited disease predicted dela yed PSA failure. Conclusions. Preliminary data suggest that the patter n of first failure can be predicted by the time course of rise in the postoperative PSA. The preliminary results of this study suggest that patient selection for clinical trials examining the efficacy of postop erative adjuvant therapy in the positive margin patient may be determi ned on the basis of the clinical and pathologic characteristics that p redict early versus delayed postoperative PSA failure.