Objective: To determine associations of lower urinary tract symptoms (LUTS) with prostate-specific antigen (PSA) levels and screen-detected localized and advanced prostate cancer.
Subjects and methods: A case-control study nested within the UK population-based ProtecT (Prostate testing for cancer and Treatment) study. Men aged 50-69 years were invited for PSA testing and those with a PSA level of >or=3.0 ng/mL were invited for biopsy. We determined whether LUTS were associated with a PSA level of >or=3.0 ng/mL and prostate cancer using logistic regression models adjusted for age, family history of prostate cancer and PSA level as appropriate. Areas under receiver operating characteristic curves (AUC) were compared between models with and without symptoms.
Results: In all, 65 871 men had a PSA test: 7251 had a PSA level of >or=3.0 ng/mL including 2467 subsequently diagnosed with prostate cancer (2119 localized, 348 advanced). LUTS were positively associated with a PSA level of >or=3.0 ng/mL: odds ratios (ORs) were 1.18 (95% confidence interval, CI 1.01-1.38), 1.69 (95% CI 1.32-2.16), and 1.60 (95% CI 1.33-1.93) for daytime urination frequency (hourly vs less frequent), urgency and hesitancy (most/all the time vs never), respectively. LUTS among men with a PSA level of >or=3 ng/mL were negatively associated with prostate cancer: ORs were 0.44 (95% CI 0.22-0.83), 0.74 (95% CI 0.63-0.87), and 0.83 (95% CI 0.73-0.94) for nocturia (4+ vs 0), leakage and hesitancy (occasionally/sometimes vs never), respectively. LUTS improved the prediction of a PSA level of >or=3.0 ng/mL (AUC 0.635 vs 0.606, P < 0.001) and prostate cancer (AUC 0.661 vs 0.638; P < 0.001).
Conclusions: A history of LUTS before PSA testing marginally improves the prediction of an individual's risk for prostate cancer; men with a PSA level of >or=3 ng/mL and LUTS were more likely to be diagnosed with benign disease than prostate cancer.