Pathologic stage and biochemical recurrence after radical prostatectomy and neoadjuvant hormone therapy: Effect of treatment until PSA is undetectable

All patients receiving a minimum of 3 months of neoadjuvant hormonal therap y (NHT) before radical prostatectomy at Memorial Sloan-Fettering Cancer Cen ter were reviewed to determine if pathologic results and PSA relapse-free s urvival differed in those treated until an undetectable PSA concentration w as attained preoperatively compared with those in whom undetectable levels were never attained. A total of 286 patients undergoing radical prostatecto my between August 1990 and January 1998 received a minimum of 3 months of N HT. Follow-up ranged from 0 to 77 months (median 24 months). At the complet ion of NHT, PSA was undetectable in 109 patients (38%) (Group 1) and detect able in 177 patients (Group 2). The tumor was pathologically organ confined in 95 of 109 patients (87%) in Group 1 compared with 114 of 177 patients ( 64%) in Group 2 (P < 0.001; chi(2) test). No tumor was identified in the ra dical prostatectomy specimen in 4.6% of the patients in Group 1 compared wi th 3.4% of the patients in Group 2. Positive margin rates were similar in t he two groups: 17% in Group 1 and 22% in Group 2 (P = 0.35; chi(2) test). O n multivariate logistic regression analysis, significant predictors of path ologic organ-confined status included treatment until undetectable v detect able PSA (odds ratio 3.8; 95% confidence interval [CI] 1.9-7.8), pretreatme nt serum PSA less than or equal to 10 ng/mL v >10 ng/mL (odds ratio 3.4; 95 % CI 1.8-6.4), clinical stage T-1 v T-2/3 (odds ratio 3.3; 95% CI 1.4-7.4), and biopsy Gleason score less than or equal to 6 v greater than or equal t o 7 (odds ratio 2.5; 95% CI 1.3-4.6). Only serum PSA, biopsy Gleason score, and duration of NHT were significant predictors of PSA relapse on Cox prop ortional hazards analysis, with odds ratios of 2.4 (95% CI 1.3-4.5) for PSA >10 ng/mL v less than or equal to 10 ng/mL, 1.8 (95% CI 1.1-3.2) for Gleas on score greater than or equal to 7 v less than or equal to 6, and 2.8 (95% CI 1.1-6.6) for >8 months of NHT v 3 to 5 months of NHT. Although there wa s no significant difference in the PSA relapse-free survival rate between G roups 1 and 2 (P = 0.20; log-rank test), there was a trend toward better PS A relapse-free survival in Group 1 (p = 0.077; log-rank test) in the subset with pretreatment PSA >10 nglmL. Despite the better pathologic results wit h NHT administered until PSA is undetectable, longer follow-up is necessary to determine if PSA relapse-free survival is improved, especially in patie nts with pretreatment PSA >10 nglmL. The optimal patient and duration and t ype of NHT remain to be determined.