Reliability of Negative Prostate MRI for Biopsy Decision-Making in the Male Han Chinese Population.

BACKGROUND

Multiparametric magnetic resonance imaging (mpMRI) has been widely utilized in clinical practice for identifying clinically significant prostate cancer (csPCa). Although mpMRI demonstrates a pooled negative predictive value (NPV) of 90%, additional clinical parameters require evaluation to enhance this metric specifically for the Chinese population-given the rising incidence of PCa in China, as well as ethnic differences in average prostate volume (PV) and chronic prostatitis prevalence that may impact mpMRI's diagnostic performance.

METHODS

A retrospective analysis was conducted on 543 patients who underwent transrectal ultrasound-guided prostate biopsy at Beijing Tsinghua Changgung Hospital between November 2014 and March 2025. After applying exclusion criteria, 412 patients were enrolled, all of whom had completed prebiopsy mpMRI within 1 month prior to biopsy. Patients were stratified into four groups based on the results of MRI examination and the pathological outcomes of biopsy: MRI (-) PCa (-), MRI (+) PCa (-), MRI (-) PCa (+), and MRI (+) PCa (+) groups. Multivariate logistic regression analyses were used to assess the odd ratios (ORs) of potential predictors for csPCa, comparing the MRI (-) PCa (+) and MRI (-) PCa (-) groups. Receiver operating characteristic curves were generated to analyze the predictive values of total PSA (tPSA), free PSA (fPSA), free-to-total (f/t) PSA, PV, PSA density (PSAD), and adjusted PSAD (PSADadj, defined as PSAD × weight) for csPCa in patients with negative MRI.

RESULTS

The patient distribution was as follows: MRI (-) PCa (-) group: 27.9% (115/412), MRI (+) PCa (-) group: 36.9% (152/412), MRI (-) PCa (+) group: 2.4% (10/412), and MRI (+) PCa (+) group: 32.8% (135/412). The NPV of MRI for csPCa was 92%. Multivariate analyses indicated that PV was negatively associated with the presence of csPCa (OR = 0.940, 95% CI: 0.896-0.986, p = 0.012), while PSAD and PSADadj were positively associated with csPCa occurrence (OR = 10.288, 95% CI: 1.569-67.46, p = 0.015; OR = 1.027, 95% CI: 1.001-1.053, p = 0.043, respectively). For MRI-negative patients, PV > 55.25 mL (sensitivity = 100%, specificity = 63.2%), PSAD < 0.100 ng/mL2 (sensitivity = 100%, specificity = 25.4%), or PSADadj < 7.24 ng/mL (sensitivity = 100%, specificity = 28.1%) enhanced MRI's NPV to 100%, while PSAD < 0.205 ng/mL2 (sensitivity = 77.8%, specificity = 71.9%) and PSADadj < 24.97 ng/mL (sensitivity = 55.6%, specificity = 90.4%) improved NPV to 97.6% and 92.6%, respectively.

CONCLUSION

In Chinese men with negative prostate MRIs, PV > 55.25 mL, PSAD < 0.100 ng/mL2, or PSADadj < 7.24 ng/mL may elevate mpMRI's NPV from 92% to 100%, enabling safe avoidance of unnecessary biopsies. Prospective multicenter validation is required to confirm these findings.