Risk of prostate cancer death following a low PSA level in the PLCO Trial

Longer-than-annual screening intervals have been suggested to improve the balance of benefits and harms in prostate cancer screening. Many researchers, societies and guideline committees have suggested that screening intervals could depend on the prostate specific antigen (PSA) result. We analyzed data from men (N=33,897) aged 55-74yrs with a baseline PSA test in the intervention arm of the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening trial (USA, 1993-2001). We estimated 5- and 10-year risks of aggressive cancer (Gleason ≥ 8 and/or Stage III/IV) and 15-year risks of prostate cancer mortality for men with baseline PSA ≤ 0.5ng/ml (N=4,862), ≤ 1ng/ml (N=15,110) and 1.01-2.5ng/ml (N=12,422). 217 men died from prostate cancer through 15yrs, though no men with PSA ≤ 1ng/ml died from prostate cancer within 5yrs (95% CI: 0.00%-0.03%). The 5-year incidence of aggressive disease was low (0.08%, 95% CI: 0.03%-0.12%) for men with PSA ≤ 1ng/ml, and higher for men with baseline PSA 1.01-2.5ng/ml (0.51%, 95% CI: 0.38-0.74). No men aged ≥ 65yrs with PSA ≤ 0.5ng/ml died from prostate cancer within 15yrs (95% CI: 0.00%-0.32%), and their 10-year incidence of aggressive disease was low (0.25%, 95% CI: 0.00%-0.53%). Compared to white men, black men with PSA ≤ 1ng/ml had higher 10-year rates of aggressive disease (1.6% vs. 0.4%, p