Purpose: We evaluated racial differences in the operating characteristics of prostate specific antigen (PSA) and digital rectal examination as screening tests for early detection of prostate cancer.
Materials and methods: We screened 18,527 white and 949 black men 50 years old or older using serum PSA measurement and digital rectal examination. We recommended biopsies if either test was suspicious for cancer. For PSA greater than 4.0 ng./ml. and rectal examination we calculated relative sensitivity (percentage of men with cancer who had a positive test), specificity (percentage of men without cancer who had a negative test) and positive predictive value (percentage of men with a positive test in whom cancer was detected) for the prediction of prostate cancer stratified by race.
Results: In white and black men PSA greater than 4.0 ng./ml. detected more cancers than rectal examination (p < 0.002) with a trend for a greater increase in sensitivity in black men. PSA was associated with fewer false-positives than rectal examination in white (p < 0.0001) but not in black (p > 0.05) men. Positive predictive value for prostate cancer of PSA and rectal examination was greater in black than in white men (48 versus 34 and 38 versus 22%, respectively).
Conclusions: PSA detects more cancers than rectal examination in both races, although this advantage is more pronounced in black men. In white men PSA yielded fewer false-positive results than rectal examination. However, PSA had more false-positive results than rectal examination in black men. Cancer was detected in a higher percentage of black men with PSA greater than 4.0 ng./ml. and, therefore, the risk of cancer associated with PSA greater than 4.0 ng./ml. differs by race. In a screening setting the widely accepted 25 to 30% positive predictive value for PSA greater than 4.0 ng./ml. may only apply to white men. A higher risk estimate of 36 to 60% is more accurate in black men.