The preliminary results of a feasibility study of a randomized trial for the early detection of prostate carcinoma are reported. 4,229 healthy men aged 60-75 years were invited to undergo digital rectal examination (DRE) and transrectal ultrasonography (TRUS); 1,284 of them responded and were thus examined. Subjects with suspicious findings at DRE and/or TRUS underwent transperineal US-guided biopsy and prostate-specific antigen (PSA) determination. Subjects with equivocal findings were controlled after 6 months with DRE, TRUS and PSA. The screening program was rather simple (examination time < 10') and cost was limited (cost x subject = US $25). So far, 27 biopsies (2.1%) have been performed. Eighteen cancers have been detected (1.41%), the prevalence/expected incidence ratio being 10.6:1. Stage at diagnosis was B1 in 12, B2 in 3, C in 2 and D in one case, respectively. Cancer had been suspected at DRE in 12, at TRUS in 17 and at PSA (cutoff = 4 micrograms/ml) in 15 of 18 cases, respectively. This study provides evidence that screening for prostate cancer is feasible at a moderate cost and diagnostic anticipation is relevant. A prospective randomized trial is needed to assess whether early detection has any impact at all on mortality. When designing such a study, the effect of compliance, which was low in our experience, on the statistical power should be carefully evaluated. The possibility of pre-screening with PSA, which is much more accepted and might improve attendance rates, should also be evaluated by specific studies.