ANATOMIC SITE-SPECIFIC POSITIVE MARGINS IN ORGAN-CONFINED PROSTATE-CANCER AND ITS IMPACT ON OUTCOME AFTER RADICAL PROSTATECTOMY

Objectives. The impact of a positive surgical margin in otherwise conf ined prostate cancer after radical prostatectomy remains unclear. We a nalyzed the outcome of a large number of patients with organ-confined prostate cancer according to the presence and anatomic site of margin positivity. Methods. We evaluated 2712 prostatectomy patients with Sta ge pT2NO cancer (ie, no evidence of extraprostatic disease, seminal ve sicle or regional node involvement) and no prior therapy who were trea ted by radical prostatectomy between 1987 and 1995 at Mayo Clinic. A t otal of 697 patients (26%) had positive margins. To assess the effect of margin status in the absence of treatment, 378 patients with postop erative adjuvant therapy were not considered for the study group; the final group consisted of 2334 patients. Results. Overall, 253 (58%) tu mors were positive at the apex and/or urethra, 85 (19%) at the prostat e base, 11 (2.5%) at the anterior prostate, and 174 (40%) at the poste rior prostate; 89 (20%) had at least two margins involved and 21 (8.3% ) had more than two involved. The apex/urethra was the only positive a natomic site in 183 (42%). Five-year survival free of clinical recurre nce or prostate-specific antigen (PSA) biochemical failure (postoperat ive serum PSA of 0.2 ng/mL or more) for patients with a single positiv e margin was 79% for apex or urethra, 78% for anterior/posterior, and 56% for prostate base. Five-year survival free of clinical recurrence or PSA (biochemical) failure was slightly higher for those with one ve rsus two margin-positive regions (77% versus 68%, respectively). Multi variate analysis revealed that positive surgical margins were a signif icant predictor of clinical recurrence and PSA (biochemical) failure ( relative risk [95% confidence interval]: 1.65 [1.24, 2.18]) after cont rolling for Gleason grade, preoperative PSA, and deoxyribonucleic acid (DNA) ploidy. The effect of margin positivity on recurrence at a spec ific anatomic site (versus negative margins or positive at a different anatomic site) revealed the prostate base to be the only significant anatomic site when adjusted for grade, PSA, and ploidy. Five-year surv ival free of the combined clinical or PSA failure end point for those with versus those without positive margins at the prostate base was 56 % versus 85%, respectively (P <0.0001). Conclusions. Positive surgical margins are a significant predictor of recurrence in Stage pT2NO canc er, which is independent of grade, PSA, and DNA ploidy. The impact of positive margin status on recurrence-free survival appears to be anato mic and site-specific, with prostate base positivity significantly ass ociated with poor outcome. The benefit of adjuvant therapy based on an atomic site-specific margin positivity remains to be tested in order t o optimize recurrence-free survival. (C) 1997, Elsevier Science Inc. A ll rights reserved.