Abstract PR031: Baseline prostate-specific antigen levels in men aged 65 to 80 and fatal prostate cancer: Implications for risk-stratified screening among older men

Abstract

Abstract Introduction: Though most deaths from prostate cancer occur in older men, screening guidelines recommend against screening prostate-specific antigen (PSA) for this group. We aimed to determine if total or free PSA level in men age 65-80 years predicts long-term risk of fatal prostate cancer over 39 years of follow-up Methods: We leveraged a nested case-control study within the Physicians’ Health Study, a randomized primary prevention trial among US male physicians initiated in 1982. Among men with a blood sample at enrollment, there were 37 fatal prostate cancer cases that occurred during follow-up, age-matched to 148 controls (within 2 years). Total and free PSA levels were measured in pre-diagnostic blood samples. Conditional logistic regression was used to estimate odds ratios and 95% confidence intervals (CI) for the association between baseline total PSA and fatal prostate cancer. We estimated Area Under the Curve (AUC) to evaluate predictive ability of total and free PSA. We also estimated the cumulative incidence of fatal prostate cancer by baseline PSA category. Results: Median (interquartile range) PSA at blood draw was 1.7 ng/mL (0.9-3.2) among controls, and median follow-up from blood draw to end of study was 19 years. The cumulative risk of fatal prostate cancer at 30 years was 0.9% for PSA <1.5 ng/mL, 7.9% for PSA ≥1.5 ng/mL, 8.9% for PSA ≥3 ng/mL, and 14.8% for PSA ≥5 ng/mL. Relative to those with total PSA <1.5 ng/mL, the odds ratios for fatal prostate cancer were 9.7 (95% CI 2.8, 33.0) for those with total PSA ≥1.5 ng/mL, 12.1 (95% CI 3.3, 44.2) for PSA ≥3 ng/mL, and 26.0 (95% CI 6.4, 105.1) for PSA ≥5 ng/mL. Adding the free/total PSA ratio to total PSA in men with total PSA ≥2 ng/mL improved prediction for fatal prostate cancer, with the AUC rising from 0.71 (95% CI 0.59, 0.83) to 0.82 (95% CI 0.73, 0.91). Restricting to those who survived at least 10 years after blood draw, the cumulative risk of fatal prostate cancer at 30 years remained higher for those with elevated baseline PSA: 1.1% for PSA <1.5 ng/mL, 7.1% for PSA ≥1.5 ng/mL, 7.0% for PSA ≥3 ng/mL, and 12.0% for PSA ≥5 ng/mL. Conclusions: A baseline PSA in older men age 65 to 80 predicts long-term risk of fatal prostate cancer across decades-long follow-up. The ratio of free to total PSA improved prediction in men with PSA ≥2 ng/mL and should be considered for reflex testing. These findings suggest healthy older men with PSA ≥3 ng/mL and an estimated life expectancy of at least 10 more years remain at increased risk for fatal prostate cancer and should consider shared decision-making about ongoing PSA monitoring. Citation Format: Hannah E. Guard, Kendrik Yim, Kathryn M. Wilson, Sigrid V. Carlsson, Lorelei A. Mucci, Travis Gerke, Mark A. Preston. Baseline prostate-specific antigen levels in men aged 65 to 80 and fatal prostate cancer: Implications for risk-stratified screening among older men [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Innovations in Prostate Cancer Research and Treatment; 2026 Jan 20-22; Philadelphia PA. Philadelphia (PA): AACR; Cancer Res 2026;86(2_Suppl):Abstract nr PR031.