#0882

Redefining Prostate Cancer Screening: Unlocking the Potential of the Free-to-Total PSA Ratio in Men with Low PSA

S. Sii¹, J. Tempo¹, L. Qu¹, N. Papa², M. Perera¹, I. Thompson³, J. Ischia¹, N. Fleshner⁴, D. Bolton¹, D. Woon¹

¹Austin Health, Urology, Melbourne, Australia
²Austin Health, Melbourne, Australia
³University of Texas, Urology, San Antonio, United States
⁴University of Toronto, Urology, Toronto, Canada

Introduction:

Prostate cancer (PCa) screening currently relies on serum prostate-specific antigen (PSA) testing, yet its utility in certain PSA ranges remains limited. Within the PSA range of 4–10 ng/ml, a low free-to-total PSA ratio (FTR) typically predicts malignancy. However, the role of FTR in identifying clinically significant prostate cancer (csPCa) in men with PSA levels ≤4 ng/ml has been less explored. This study evaluates whether the FTR enhances detection of csPCa in this low PSA cohort, offering a complementary tool for refining prostate cancer screening.

Material and methods:

Data were analyzed from the Prostate Cancer Prevention Trial, including patients with PSA ≤4 ng/ml who underwent biopsy within one year of their PSA measurement. Associations between FTR and the presence of csPCa (ISUP grade ≥2) were assessed using logistic regression models adjusted for age and PSA. The predictive accuracy of FTR was evaluated with a rescaled Brier score, and decision curve analysis assessed its clinical utility.

Results:

In a cohort of 406 men, csPCa was diagnosed in 34% (139 patients), and 50% (204 patients) had any grade PCa. Among those with an FTR ≤0.15, 46% were found to have csPCa, compared to 22% with an FTR ≥0.20. For a 60-year-old man with a PSA of 3 ng/ml, the probability of csPCa was 61% with an FTR of 0.05, declining to 18% at an FTR of 0.30. A clear negative relationship between increasing FTR and probability of csPCa was observed. A model containing FTR additional to PSA and age provides greater net benefit as per decision curve analysis and likely superior discrimination and calibration measured by a higher index of predictive accuracy.