How to avoid prostate biopsy in men with Prostate Image-Reporting and Data System 3 lesion? Development and external validation of new biopsy indication using prostate health index density

Abstract

Background

To develop a customized prostate biopsy indication using prostate health index density (PHID) combined with multiparametric magnetic resonance imaging (mpMRI) and assess the reliability of the PHID cutoff value in external populations.

Methods

A total of 521 cognitive MRI/ultrasonography fusion prostate biopsies and biomarker tests for prostate-specific antigen (PSA), free PSA, and PHI were performed after mpMRI. The predictive value for clinically significant prostate cancer (csPCa; Gleason score≥7) of PSA derivatives was examined using the ROC curve. We developed a new biopsy indication utilizing a PHID cutoff based on the Prostate Image-Reporting and Data System (PI-RADS) score, which was externally validated.

Results

The combination of PHID and mpMRI (AUC = 0.884) demonstrated the highest predictive ability for csPCa, although PHID (AUC = 0.843) and PI-RADS (AUC = 0.806) individually also showed a high diagnostic value. When a PHID cutoff of 0.75 was used in men with PI-RADS 3 lesions, the negative predictive value of csPCa was 100%, and approximately half of the biopsies could be safely avoided.

Conclusion

Compared to PHID or PI-RADS scores alone, the combination of PHID and PI-RADS scores increased the accuracy of csPCa detection and the number of cases in which biopsy could be avoided. In men with PI-RADS 3 lesions, the optimal PHID cutoff ≥0.75 can prevent half of the unnecessary biopsies without missing csPCa. In men with PI-RADS 4-5 lesions, biopsies are warranted regardless of PHID values because csPCa could be accompanied by low PHID.