Performance of PI-RADS in predicting transperineal prostate biopsy result in South Asian immigrant men compared to the general Australian population.

Abstract

Introduction & Objectives: Preliminary data from the literature suggests the performance of the PIRADS criteria in predicting prostate cancer on biopsy may vary across differing ethnic groups. In particular, some studies suggest that PI-RADS 3 specifically may be associated with a far lower risk of prostate cancer in Asian individuals as compared to Caucasians. However, these studies abroad have focused on East Asians, and there is as yet no data comparing the performance of PI-RADS in South Asian immigrants to the general population in Western countries. This study aims to assess the performance of PI-RADS criteria in predicting risk of prostate cancer in South Asian immigrant men compared to the general Australian population. Method(s): Retrospective audit with ethics approval was conducted of 175 men (84 South Asian, 91 general population) undergoing transperineal prostate biopsy for elevated PSA in a private practice in Melbourne, Australia. South Asian origin was defined as birth in a SAARC country. Datapoints included demographics, PSA level and density, mpMRI index lesion PIRADS score and biopsy results. Regarding baseline statistics, South Asian men compared to the general population had higher PSA (mean 7.61 vs 5.75 ng/mL, P = 0.0443), higher PSA density (mean 0.19 vs 0.13, P = 0.0059) and younger age (mean 61.5 vs 64.3, P = 0.0442). Result(s): In men with PI-RADS 3 index lesion on MRI, South Asian origin was associated with lower risk of prostate cancer on transperineal biopsy compared to the general Australian population (OR 0.28, 95% CI 0.08-0.92, P = 0.04), after multivariate adjustment for risk factors of PSA level (OR 0.93, 95% CI 0.75-1.05, P = 0.4), PSA density >0.15 (OR 1.29, 95% CI 0.22-8.68, P = 0.8), and age (OR 1.04, 95% CI 0.97-1.12, P = 0.3). However, in those with PIRADS 4 or 5 lesions, those of South Asian origin did not have differing risk of prostate cancer on biopsy compared the general population (OR 0.94, 95% CI 0.3-2.92, P > 0.9), after multivariate adjustment for PSA level (OR 0.9, 95% CI 0.81-1.01, P = 0.059), PSA density >0.15 (OR 6.41, 95% CI 1.72-30.2, P = 0.009), and age (OR 1.1, 95% CI 1.03-1.19, P = 0.008). Conclusion(s): In our cohort of South Asian immigrant men in Australia, PI-RADS 3 was associated with a significantly lower risk of prostate cancer on biopsy than in the general Australian population. This finding mirrors studies in the literature which show PI-RADS 3 is associated with lower risk of prostate cancer in East Asian-origin Americans compared to the general American population. Our finding may be attributable to the higher described rate of prostatitis in South Asians, possibly resulting in changes akin to PI-RADS 3 on MRI. Of note, in those men with PI-RADS 4 or 5 lesions, South Asian origin was not associated with lower prostate cancer risk compared to the general population.