A comparison of six commercial assays for total and free prostate specific antigen (PSA): the predictive value of the ratio of free to total PSA

OBJECTIVE: To evaluate the clinical utility of six commercial assays for free prostate specific antigen (fPSA) and the derived ratio of fPSA to total PSA in distinguishing between patients with prostate cancer or benign prostate hyperplasia (BPH). MATERIALS AND METHODS: Each assay was evaluated against a panel of serum samples comprising those from patients with prostatic disease, other malignancies, normal subjects and sera containing substances which might interfere with the immunoassay. RESULTS: The levels of total (tPSA), fPSA and their ratio (f/tPSA) were compared among the different samples. All assays showed similar specificities for prostate carcinoma but differed in the positive predictive values of f/tPSA. CONCLUSIONS: Although all six assays were equimolar, there were differences in calibration, particularly for fPSA. The ability of f/tPSA to discriminate between benign and malignant prostatic disease depended on the assay used. The f/tPSA is not valid when the tPSA and fPSA assays are obtained from different manufacturers.