Objective  To explore the value of clinical and biparametric magnetic resonance imaging (bpMRI) in diagnosing extraprostatic extension (EPE) of prostate cancer (PCa).

Methods This retrospective study assigned 107 patients into EPE (n=42) and organ-limited (n=65) groups based on their postoperative pathology after radical prostatectomy from August 2018 to May 2024 at Wuhu Second People's Hospital. The differences in the following clinical risk indicators were compared between the groups: age, total prostate specific antigen (tPSA), prostate volume, prostate specific antigen density (PSAD) and International Society of Urological Pathology (ISUP) score for prostate puncture. The differences in MRI indicators, prostate imaging reporting and data system (PI-RADS) score and bpMRI were also identified. Binary Logistic regression analysis was used to construct clinical and joint models for diagnosing EPE, and screening independent influencing factors. The ROC curve analyze the independent influencing factors and diagnostic performance of the models. The DeLong test was used to compare the differences between the AUC models. A nomogram was draw, and performance evaluated.

Results The differences in tPSA, PSAD, ISUP score for prostate puncture, PI-RADS score, and bpMRI were statistically significant between the two groups (P<0.05). The clinical model AUC was 0.821; while the AUCs of the combined model and independent influencing factors PSAD (OR=25.992), ISUP score for prostate puncture (OR=1.676), and bpMRI (OR=10.729) were 0.899, 0.813, 0.770, and 0.793 respectively (P<0.001). The combined model was superior to the clinical model (Z=2.502 and P=0.012). The average AUC for 5-fold cross-validation was 0.887, with high model calibration and a threshold range of 5%-85%, indicating clinical benefits.

Conclusions The combined model nomogram derived from clinical and bpMRI indicators is highiy valuable for diagnosing PCa EPE.