Abstract
OBJECTIVE: To evaluate patterns and imaging-based predictors of positive surgical margins (PSMs) in patients with high-risk prostate cancer (PCa) undergoing nerve-sparing (NS) robot-assisted radical prostatectomy (RARP). MATERIALS AND METHODS: We retrospectively analyzed 1,235 consecutive patients with high-risk PCa treated with RARP between 2022 and 2024 at a high-volume tertiary referral center. Among them, 533 patients underwent preoperative multiparametric MRI (mpMRI) reviewed by two expert uro-radiologists and MRI-ultrasound fusion biopsy. A per-side analysis was performed to identify predictors of ipsilateral PSMs in cases where neurovascular bundle (NVB) preservation was attempted. Biochemical recurrence (BCR) was assessed, and multivariable logistic regression was used to determine independent predictors of PSMs. RESULTS: Overall, 36.1% of patients underwent non-nerve-sparing surgery, 49.5% unilateral NS, and 14.4% bilateral NS. Nerve sparing was performed on 418 surgical sides, with ipsilateral PSMs detected in 90 (21.5%). Sides with and without PSMs showed comparable nerve-sparing techniques (intra- vs interfascial) and similar 3-year BCR-free survival rates (68% vs. 69%) at a median follow-up of 36 months, although earlier biochemical failure at lower PSA thresholds was more frequent in PSM-positive sides. On multivariable analysis, larger prostate volume, apical tumor location, peripheral zone involvement, greater lesion diameter, and extracapsular extension on mpMRI were independently associated with an increased risk of ipsilateral PSMs. CONCLUSIONS: Nerve-sparing RARP may be feasible in carefully selected high-risk PCa patients. The integration of mpMRI-based predictors can enhance patient selection, optimizing the balance between oncologic safety and functional preservation. Prospective studies are warranted to validate these findings and minimize selection bias.