Integration of Multiparameter MRI into Conventional Pretreatment Risk Factors to Predict Positive Surgical Margins After Radical Prostatectomy

Abstract

Introduction/BackgroundPositive surgical margins (PSMs) after radical prostatectomy (RP) can increase the risk of biochemical recurrence in prostate cancer (PCa) patients. However, the prediction of the likelihood of PSMs in patients undergoing similar surgical procedures remains a challenge. We aim to develop a predictive model for PSMs in patients undergoing non–nerve-sparing RP. Patients and MethodsIn this retrospective study, we analyzed data from PCa patients who underwent minimally invasive non–nerve-sparing RP at our hospital between June 2017 and June 2021. We identified independent risk factors associated with PSMs using clinical and MRI-based parameters in univariate and multivariate logistic regression analyzes. These factors were then used to develop a nomogram for predicting the probability of PSMs. The predictive performance was validated using calibration and receiver operating characteristic curve, area under the curve ,and decision curve analysis. ResultsMultivariate analyzes revealed prostate-specific antigen density, tumor size, tumor location at the apex, tumor contact length, extracapsular extension (ECE) level, and apparent diffusion coefficient value as independent risk factors. A nomogram was developed and validated with high accuracy (C-index = 0.78). Furthermore, we found that 44.2% of patients diagnosed with organ-confined disease had ECE after surgery, and 29.1% of patients with Gleason scores ≤7 had higher pathological scores. Interestingly, the tumor burden calculated from PCa biopsy cores was overestimated when compared to postoperative PCa specimens. ConclusionWe developed a reliable nomogram for predicting the risk of PSMs in PCa patients undergoing non–nerve-sparing RP. The study highlights the importance of incorporating these parameters in personalized surgical management.