Multivariable Models Incorporating Multiparametric Magnetic Resonance Imaging Efficiently Predict Results of Prostate Biopsy and Reduce Unnecessary Biopsy

Abstract

Purpose We sought to develop diagnostic models incorporating mpMRI examination to identify PCa (Gleason score≥3+3) and CSPCa (Gleason score≥3+4) to reduce overdiagnosis and overtreatment. Methods We retrospectively identified 784 patients according to inclusion criteria between 2016 and 2020. The cohort was split into a training cohort of 548 (70%) patients and a validation cohort of 236 (30%) patients. Age, PSA derivatives, prostate volume, and mpMRI parameters were assessed as predictors for PCa and CSPCa. The multivariable models based on clinical parameters were evaluated using area under the curve (AUC), calibration plots, and decision curve analysis (DCA). Results Univariate analysis showed that age, tPSA, PSAD, prostate volume, MRI-PCa, MRI-seminal vesicle invasion, and MRI-lymph node invasion were significant predictors for both PCa and CSPCa (each p ≤0.001). PSAD has the highest diagnostic accuracy in predicting PCa (AUC=0.79) and CSPCa (AUC=0.79). The multivariable models for PCa (AUC=0.92, 95% CI: 0.88–0.96) and CSPCa (AUC=0.95, 95% CI: 0.92–0.97) were significantly higher than the combination of derivatives for PSA ( p =0.041 and 0.009 for PCa and CSPCa, respectively) or mpMRI (each p <0.001) in diagnostic accuracy. And the multivariable models for PCa and CSPCa illustrated better calibration and substantial improvement in DCA at threshold above 10%, compared with PSA or mpMRI derivatives. The PCa model with a 30% cutoff or CSPCa model with a 20% cutoff could spare the number of biopsies by 53%, and avoid the number of benign biopsies over 80%, while keeping a 95% sensitivity for detecting CSPCa. Conclusion Our multivariable models could reduce unnecessary biopsy without comprising the ability to diagnose CSPCa. Further prospective validation is required.

Developing Strategy to Predict the Results of Prostate Multiparametric Magnetic Resonance Imaging and Reduce Unnecessary Multiparametric Magnetic Resonance Imaging Scan

“…Our prior study, consistent with other studies, found that clinical parameters such as age, PSA derivates [total PSA (tPSA), free/total PSA (f/tPSA), and PSA density (PSAD)], and prostate volume (PV) were significant predictors for PCa and CSPCa ( 8 , 15 ). Therefore, we question whether the negative and equivocal mpMRI scans could be limited using a model based on these clinical parameters among men with an elevated PSA level.…”

“…To date, multivariate models or machine learning models for the detection of CSPCa have been developed in a growing body of literatures (8)(9)(10)(11)(12)25). Studies demonstrated that a risk-based triage strategy could reduce more unnecessary biopsy and the overdiagnosis in comparison with single parameters (8,11,25).…”

“…To date, multivariate models or machine learning models for the detection of CSPCa have been developed in a growing body of literatures (8)(9)(10)(11)(12)25). Studies demonstrated that a risk-based triage strategy could reduce more unnecessary biopsy and the overdiagnosis in comparison with single parameters (8,11,25). However, the study about developing a multivariate model to predict the results of prostate mpMRI and selecting patients who could benefit from mpMRI is limited (13,14).…”

“…Performing mpMRI and/or prostate biopsy among men with a high risk of CSPCa could be an acceptable option (7). A dozen of risk calculators incorporating clinical variables and/or novel biomarkers have been developed to predict the results of prostate biopsy and to reduce unnecessary biopsy by 36%-66% (8)(9)(10)(11)(12). However, the knowledge about developing a strategy to predict the results of mpMRI and select patients who could benefit from mpMRI is limited (13,14).…”

Construction and Comparison of Different Models in Detecting Prostate Cancer and Clinically Significant Prostate Cancer

“…In a prospective study, researchers found that the AUC of the combination of PHI and PI-RADS was higher than that of PHI (0.873 vs. 0.735, p = 0.002) or PI-RADS (0.873 vs. 0.830, p = 0.035) alone, and this study showed the superiority of combined diagnosis ( 22 ). There were also some studies that compare different cutoff values in the model of combined diagnosis and found the cutoff value that can save the maximum degree of biopsies ( 23 ). However, in their series, these studies only compared the single predictor with the multivariable model and did not establish the base model with some primary indicators (such as age, BMI, TPSA, and the medical history of DM and hypertension).…”