Development of a Nomogram Incorporating Multiparametric MRI and Clinical Indicators for Predicting Positive Surgical Margins After Radical Prostatectomy.

BACKGROUND: This study aimed to develop a nomogram integrating multiparametric magnetic resonance imaging (mpMRI) features and clinical indicators for the preoperative prediction of positive surgical margins (PSM) after radical prostatectomy (RP). METHODS: This retrospective study enrolled 304 patients who underwent RP for prostate cancer (PCa). The patients were divided into PSM (n = 98) and negative surgical margin (NSM, n = 206) groups based on histopathological results. Preoperative clinical variables, including prostate-specific antigen (PSA), PSA density (PSAD), biopsy Gleason Grade Group and clinical T-stage, were collected. The mpMRI parameters assessed included Prostate Imaging Reporting and Data System (PI-RADS) v2.1 score, tumour location, extracapsular extension (ECE), tumour maximum diameter, capsule contact length (CCL) and apparent diffusion coefficient (ADC) value. RESULTS: Compared with the NSM group, the PSM group had significantly higher preoperative PSA, higher PSAD, a higher proportion of biopsy Gleason Grade Group 4-5, and a higher clinical T-stage proportion. On mpMRI, the PSM group exhibited a higher proportion of PI-RADS v2.1 score 5 lesions, more frequent transition zone tumours, higher suspicion of ECE, larger tumour maximum diameter, longer CCL and lower ADC value. Multivariate analysis identified PSAD, PI-RADS v2.1 score, suspected ECE, tumour maximum diameter, CCL, ADC value, and biopsy Gleason Grade Group as independent predictors of PSM (all p < 0.05). The nomogram achieved an area under the curve of 0.907 (95% confidence interval (CI): 0.873-0.941). CONCLUSIONS: A nomogram combining mpMRI and clinical indicators may aid preoperative prediction of PSM after RP, potentially facilitating patient risk stratification.