PSA LEVELS AND THE RATE OF POSITIVE SURGICAL MARGINS IN RADICAL PROSTATECTOMY SPECIMENS PRECEDED BY ANDROGEN BLOCKADE IN CLINICAL B-2 (T2BNXMO) PROSTATE-CANCER

Objectives. Neoadjuvant hormonal therapy reduces positive margins in p atients undergoing radical prostatectomy. All patients experience a de crease in serum prostate specific antigen (PSA), but not always to a l evel that is nondetectable. The results of several prospective, random ized trials indicate that the incidence of positive margins decreases with the use of androgen deprivation prior to radical prostatectomy. I t has been suggested that a greater decline in PSA levels would result in fewer positive margins. In a recent US trial of patients with T2bN xMO prostate cancer, we reported that 18% of patients randomized to re ceive 3 months of leuprolide acetate and flutamide had positive margin s, compared to 48% of those who had radical prostatectomy alone (P < 0 .001). We correlated the PSA levels prior to and following androgen de privation and the presence of a positive margin following radical pros tatectomy (RP). Methods. One hundred and thirty-seven of 149 patients randomized to receive presurgery androgen deprivation (AD) underwent r adical prostatectomy. Of these, 135 had a PSA level obtained both prio r to androgen deprivation and prior to surgery. We analyzed the percen t positive margins in patients whose PSA values became undetectable an d in those whose values remained above 0.1 ng/mL despite androgen depr ivation. Results. Eight of 45 patients (19%) with a nadir PSA less tha n or equal to 0.1 ng/mL had a positive surgical margin and 16/92 (17%) with a nadir PSA > 0.1 ng/mL had tumor at the margin. There were no s tatistical differences in these two groups (P = 1.0 by Fisher's Exact Test [two-tailed], and the Pearson correlation was -0.015). Conclusion s. There was no correlation between an undetectable PSA and a PSA > 0. 1 ng/mL and the presence of tumor at the margin when 3 months of AD wa s given prior to RP. It is possible that longer periods of AD prior to RP will reduce PSA to an undetectable level in a higher percent of pa tients. However, these data suggest that an undetectable level will no t result in less positive margins. (C) 1997 by Elsevier Science Inc.