#0639

EXTERNAL VALIDATION OF THE PROSTATE BIOPSY COLLABORATIVE GROUP AND EUROPEAN RANDOMIZED STUDY OF SCREENING FOR PROSTATE CANCER RISK CALCULATORS IN MULTI-ETHNIC MALAYSIA

Y. Seo¹, Y. Ooi², C. Kalogirou³

¹Hospital Kuala Lumpur, Urology, Kuala Lumpur, Malaysia
²Hospital Pulau Pinang, Urology, Pulau Pinang, Malaysia
³University Hospital Wurzburg, Urology, Wurzburg, Germany

Introduction:

Prostate cancer detection rates for PSA ≤10.0 ng/ml in Malaysia (14.3%) are lower than in Western populations (33.7%). This raises concerns about the applicability of Western-developed risk calculators (RCs) in Asian cohorts. We assessed the performance of the Prostate Biopsy Collaborative Group Risk Calculator (PBCG-RC) and the European Randomized Study of Screening for Prostate Cancer (ERSPC-RC3) in a multi-ethnic Malaysian cohort.

Material and methods:

A retrospective analysis of men who underwent prostate biopsy in Penang Hospital, Malaysia between January 2019 to August 2024 was performed. Requirement criteria are age (50-80 years), prostate-specific antigen (PSA) (3.0 - 50 ng/ml), abnormal digital rectal examination (DRE) or suspicious lesions on prostate imaging. Predicted probabilities for clinically significant prostate cancer (csPCa, Gleason ≥7) were compared with histopathology results. Statistical analyses were performed using IBM SPSS Statistics version 22 (IBM, USA). Evaluation was performed by calibration, discrimination, and clinical utility using calibration plots, area under the receiver operating characteristic curves (AUCs), and decision curve analyses (DCAs), respectively.

Results:

995 prostate biopsies were eligible for final analysis. 250 men (25.2%) had csPCa. PBCG-RC (AUC: 0.78, 95% CI 0.74-0.81) and ERSPC-RC3 (AUC: 0.82, 95% CI 0.78-0.85) shows strong discrimination and predictive accuracy. Median risk of csPCa was 49% (interquartile Range [IQR] 33-65%) for PBCG-RC and 13% (IQR 4-36%) for ERSPC-RC3. Both PBCG-RC (Calibration slope: 0.91 [95% CI 0.79 - 1.0]) and ERSPC-RC3 (Calibration slope: 0.88 [95% CI 0.79 - 0.97]) underestimated the risk of csPCa. Decision curve analysis (DCA) showed clinical net benefit for risk thresholds ≥5%, with PBCG-RC performing better across a wider range (5–30%) than ERSPC-RC3 (5–25%). Re-calibration improved the clinical benefits of both models.