Serum prostate specific antigen levels in men with benign prostatic hyperplasia and cancer of prostate

Abstract

Objective: To determine the sensitivity, specificity and positive predictive value of the PSA test at the conventional cut-off value of 4 ng/ml. Design: Retrospective study. Setting: Nairobi Hospital Laboratory, Nairobi. Data Source: Results of serum Prostate specific Antigen (PSA), estimation and prostate histology specimens at the-Nairobi Hospital Laboratory between January 1997 and December 1999. Results: A total of 3309 PSA tests and 524 prostate biopsies were, processed during the study period. One hundred cases had both PSA test and histology, where the PSA test was done before the surgical procedure. Fourty nine of them had histological diagnosis of cancer of prostate (CAP) while 71 had benign prostate hyperplasia (BPH). There was considerable overlap in the PSA concentrations between the two groups, the ranges being 0.34-36 ng/ml and 1.78-4339 ng/ml for BPH and CAP respectively. 63.4% of BPH subjects had PSA concentrations above the 4 ng/ml threshold value; 29.6% being in the diagnostic gray zone of 4-10 ng/ml. In contrast only 6% of CAP subjects had PSA values in the gray zone. Using the 4 ng/ml cut-off, the sensitivity, specificity and positive predictive value of the PSA test were 89.8, 37 and 49% respectively. Conclusion: Although PSA is a sensitive test, it is not sufficiently specific to discriminate between BPH and CAP at intermediate values. Additional approaches are required to enhance its specificity for CAP screening without reducing the sensitivity.