Abstract

The participating institutions of the American Cancer Society National Prostate Cancer Detection Project did 520 biopsies on 2425 men over a 3.5-year period. A total of 88 cancers were confirmed pathologically, 93% of which clinically were organ confined. In 324 men (62.3%), a recommendation for biopsy was made based solely on the results of transrectal ultrasonography (TRUS); in 69 patients (13.3%), solely on the digital rectal examination (DRE); in 116 patients (22.3%), on abnormal DRE and TRUS examinations; and in 11 patients (2.1%), in whom DRE and TRUS were normal, on elevated prostate-specific antigen (PSA) levels. The TRUS was abnormal in 80.6% of men found to have cancer, and the PSA level and DRE were abnormal for 67% and 50% of cancers, respectively. The influence of PSA level on cancer detection increased as the serum level increased above 4 ng/ml. The positive predictive values of both the DRE and TRUS were influenced significantly by the presence of an elevated PSA level (P = 0.044 and P less than 0.001, respectively). The results of this ongoing multicenter study support the following statements: (1) the prostate cancer detection rate is influenced by this diagnostic triad and (2) the detection rate of organ-confined disease can be improved substantially by early detection programs.