Nomograms to predict 3-year biochemical recurrence after robotic-assisted laparoscopic radical prostatectomy based on preoperative and perioperative variables of 774 patients.

Abstract

115 Background: Predictors of biochemical recurrence after robotic-assisted laparoscopic radical prostatectomy (RALP) are not well reported in the literature. We wanted to investigate preoperative predictors as well as the influence of nerve sparing and positive surgical margin status on 3-year biochemical recurrence. Methods: 774 patients with at least 3 year follow up had undergone RALP by a single surgeon at our institution. Biochemcial recurrence was defined as a postoperative PSA >0.2 ng/ml. Multivariable logistic regression models were used to develop the biochemical recurrence predictive nomograms: nomogram 1- age, BMI, PSA density, clinical stage, biopsy Gleason, percent positive cores, perineural invasion; nomogram 2- age, BMI, PSA density, clinical stage, biopsy Gleason, percent positive cores, perineural invasion, nerve sparing, positive surgical margins (none, unifocal, or multifocal). The predictive accuracy of the models was assessed in terms of discrimination and calibration. Results: Both nomograms discriminated well between patients that recurred and those that did not (bootstrap corrected c-indices of 0.766 and 0.806 for nomograms 1 and 2 respectively). Nomogram 1 was well calibrated, but nomogram 2 over- predicted the probability of biochemical recurrence in patients at >30% risk. Conclusions: Our nomogram based on age, BMI, PSA density, clinical stage, biopsy Gleason, percent positive cores, and perineural invasion on preoperative biopsy has a good predictive ability to differentiate between RALP-treated patients that biochemically recur by 3 years from those that do not. Adding nerve sparing and surgical margin status further improved discriminatory ability but at the expense of over-prediction for patients at high risk. These nomograms may be used to guide the use of nerve sparing and the management of positive margins in men undergoing RALP for clinically localized prostate cancer. No significant financial relationships to disclose.