Purpose: We sought to evaluate the pathologic results and postoperative outcomes for men undergoing robot-assisted laparoscopic radical prostatectomy (RLRP) for biopsy Gleason score (GS) 8 to 10 disease. Stratification of these patients according to preoperative variables was also performed in an attempt to predict organ-confined cancer.
Patients and methods: A prospective RLRP database identified all patients with preoperative biopsy GS 8 to 10. Variables, including prostate-specific antigen (PSA), percent positive biopsy cores (%PBC), maximal percentage of cancer in biopsy core (%MCB), clinical stage, pathologic stage, pathologic GS, surgical margins status, lymph node status, time to biochemical recurrence, and recurrence rate, were evaluated. Preoperative variables were treated as continuous and categorical using PSA, %PBC and %MCB cutoffs of 10 ng/mL, 50%, and 30%, respectively.
Results: Between February 2003 and September 2007, a total of 1225 RLRPs were performed at the University of Chicago Medical Center. Seventy-two (5.9%) patients had preoperative biopsy GS 8 to 10. Two patients received neoadjuvant hormonal therapy and were excluded. Among 70 patients evaluated, 33 (47%) had organconfined (pT(2)N0) disease. Forty (60.6%) patients had pathologic downgrading to GS <or=7. Overall positive surgical margin (PSM) rate was found to be 24.2%. pT(2)- and pT(3)-PSM rate was 6% and 42.3%, respectively. In multivariate logistic regression analysis, PSA <or=10 ng/mL (P = 0.04) and %MCB <or=30% (P = 0.001) were found to be statistically significant predictors of pT(2)N0 disease.
Conclusion: Preoperative biopsy GS 8 to 10 predicts a significant likelihood of finding non-organ-confined prostate cancer on the final pathology report. Preoperative PSA <or=10 ng/mL and %MCB <or=30% may be used to predict favorable pathologic outcome for these patients during surgical counseling.