Combined Transrectal Ultrasound Five-Grade Scoring System and Multiparametric MRI PI-RADS Score for Risk Prediction Modeling of Prostate Cancer and Clinically Significant Prostate Cancer in the PSA Gray Zone

Abstract

Purpose: This study aims to develop and validate predictive nomograms for prostate cancer (PCa) and clinically significant PCa (csPCa) in prostate-specific antigen (PSA) gray zone of 4−10 ng/mL, utilizing transrectal ultrasound five-grade scoring system (TRUS-5), prostate imaging reporting and data system (PI-RADS) score based on multiparametric magnetic resonance imaging (mpMRI) and clinical parameters.

Method: A retrospective analysis was conducted on a cohort of 383 patients with elevated PSA levels in the gray zone who underwent initial TRUS-guided 12-core prostate biopsy between January 2018 and December 2020. Multivariate logistic regression analyses were utilized to identify independent predictors for PCa and csPCa and to establish predictive nomograms. The diagnostic performance, calibration ability, and clinical usefulness of each model were assessed through receiver operating characteristic (ROC) curve analysis, calibration curve analysis, and decision curve analysis (DCA), respectively.

Results: Two diagnostic nomograms were developed based on five risk factors (age, free/total PSA [(f/t) PSA] ratio, prostate volume [PV], TRUS-5, and PI-RADS score) to predict PCa and five risk factors (PSA, (f/t) PSA, PV, TRUS-5, and PI-RADS score) for csPCa of PSA gray zone patients. The TRUS-5/PI-RADS combined model demonstrated higher area under the curves (AUCs) (0.865 for PCa and 0.910 for csPCa) compared to the models based on PI-RADS score or TRUS-5 (p  < 0.05), as well as the greater net benefits across a wide range of threshold probabilities for PSA gray zone patients.

Conclusion: The current study presented valuable TRUS-5/PI-RADS combined nomograms for predicting PCa or csPCa in patients within the PSA gray zone, which may serve as useful tools for the clinical management of PCa and csPCa.