Prostate cancer risk assessment in men with an initial P.S.A. below 3 ng/mL: results from the Göteborg randomized population-based prostate cancer screening trial.

OBJECTIVE: To evaluate the long-term outcome of men with an initial prostate-specific antigen (PSA) level below 3 ng/mL and whether the free-to-total (F/T PSA) ratio is a useful prognostic marker in this range.

MATERIALS AND METHODS: This study is based on 5,174 men aged 50-66 years, who in 1995-1996 participated in the first round of the Göteborg randomized screening trial (initial T-PSA level <3 ng/mL). These men were subsequently invited biennially for PSA and F/T PSA screening until they reached the upper age limit (on average 69 years). Biopsy was recommended if PSA ≥ 3 ng/mL.

RESULTS: After a median follow-up of 18.9 years, 754 men (14.6%) were diagnosed with prostate cancer (PC). The overall cumulative PC incidence was 17.2%. It increased from 7.9% among men with T-PSA of ≤0.99 ng/mL to 26.0% in men with T-PSA levels of 1-1.99 ng/mL and 40.3% in men between 2-2.99 ng/mL (p < 0.001). The initial PSA was also related to the incidence of Gleason ≥7 PC (3.7% vs 9.7% vs 10.9%) and PC death (0.3% vs 1.1% vs 1.5%). Adding F/T PSA did not improve PC prediction in terms of Harrell concordance index (base model 0.76 vs 0.76) nor improvement of the likelihood of the model (p = 0.371).

CONCLUSIONS: Some men with initial PSA < 3 ng/mL will be diagnosed too late, despite participating in an organized screening program, indicating that prompt diagnosis is justified in these men. PC incidence and risk of PC death was associated with PSA., but F/T PSA had no predictive value.