Background: We investigated whether age influences the predictiveness of genetic risk score (GRS) for prostate cancer (PCa) in a Chinese hospital-based biopsy cohort.
Methods: We included consecutive patients who underwent prostate biopsies in two tertiary centers between 2012 and 2014. GRS was calculated using 24 PCa-associated genetic variants and its predictiveness was assessed by area under curve (AUC).
Results: Of 1120 men tested, 724 with prostate-specific antigen (PSA) < 20 ng/ml were selected for further analysis. Patients were divided into 3 groups by age cutoffs at 60 and 70 years. GRS significantly predicted PCa for all patients (AUC: 0.561; 95% CI: 0.514-0.609) and was an independent predictor in multivariate analysis for the 60-70 year-olds (AUC: 0.612, 95% CI: 0.541-0.684), but not for patients aged < 60 years or ≥ 70 years. For PCa with Gleason score ≥ 7, GRS discriminative ability was 0.582 (95% CI=0.527-0.637) for all patients, and 0.647 (95% CI: 0.541-0.684) for the 60-70 year-old group.
Conclusion: GRS significantly increased clinical prediction of PCa and high-grade disease in Chinese men aged 60-70 years, which implies that men in this age group would benefit most from genetic testing.
Keywords: age; genetic score; multivariate model; predictive performance; prostate cancer.