Multivariable model integrating PHI and mpMRI for detecting csPCa in biopsy-naïve men

Area of Science:

• Urology
• Oncology
• Radiology

Background:

• Prostate cancer (PCa) diagnosis can be improved by integrating blood biomarkers and multiparametric magnetic resonance imaging (mpMRI).
• Validated models combining both tools for risk-adapted clinical decisions are scarce.

Purpose of the Study:

• To evaluate and internally validate a multivariable model for predicting clinically significant prostate cancer (csPCa) in biopsy-naïve men.
• The model integrates clinical, analytical (Prostate Health Index - PHI), and imaging (mpMRI) parameters.

Main Methods:

• Prospective observational study of 183 biopsy-naïve men (age 50-75) with PSA 4-10 ng/mL and/or abnormal DRE.
• A multivariable logistic regression model was built using PHI, PSA density, PSA free/total ratio, PIRADS score, and age.
• Internal validation via bootstrap resampling; diagnostic accuracy assessed and compared to simplified strategies.

Main Results:

• The model achieved an AUC of 0.841, with 100% sensitivity and 66.7% specificity for csPCa at a 17% risk threshold.
• It enabled avoidance of 49.4% of biopsies without missing csPCa cases.
• External validation of a similar PHI-mpMRI nomogram showed robust accuracy (AUC 0.89).

Conclusions:

• The developed model accurately predicts csPCa, outperforming individual tools like PHI or PIRADS alone.
• Its application can enhance diagnostic efficiency and reduce unnecessary prostate biopsies.
• Combining PHI and mpMRI offers a promising strategy for improved PCa diagnosis and management.