The proposal of an early diagnosis of prostate cancer through mass screening with digital rectal examination (DRE), transrectal ultrasound (TRUS) and serum tumor markers remains controversial: there is no high risk population. No study has proven that mass screening reduces the mortality from prostate cancer. However, when clinical and biological data give arguments for the presence of a cancer, every patient requires a prostate biopsy. We have studied the Positive Predictive Value (PPV) of each test in a selected population: 200 men over 50 years of age in which rectal examination or PSA assay was suspicious were investigated. Without any reference to the prostate volume, we considered that the PSA level was "suspicious" when it reached 3 times the upper reference value, or 12 ng/ml. DRE was suspicious in 73%, comprising 50% with prostate carcinoma. PSA assay was suspicious in 65%, comprising 61% with prostate carcinoma. 88% of cancers had a suspicious DRE or PSA assay. TRUS was suspicious in 89%, comprising 45% with prostate carcinoma. Ultrasound guided core biopsies were performed in each case, and allowed a positive diagnosis in 42% of cases, whereas bilateral fine-needle cyto-aspirates were positive in 87% of histologically proven carcinomas. Cytology alone was positive in 3 patients with negative biopsies. Both results show that the PPV of a suspicious DRE associated with an elevated PSA level is 84%. An increased PSA level is correlated with a cancer in 61%.(ABSTRACT TRUNCATED AT 250 WORDS)