Diagnostic accuracy of percent free prostate-specific antigen in prostatic pathology and its usefulness in monitoring prostatic cancer patients

Abstract

Introduction: The aim of our study was to evaluate the clinical usefulness of percent free prostate-specific antigen (PSA) [ratio of free PSA (fPSA) to total PSA (tPSA); f/tPSA] in prostatic pathology and its usefulness in monitoring prostatic cancer patients.

Patients and methods: Our prospective study was carried out on 470 consecutive male patients referred to our outpatient urological clinic for observation. We looked for relationships between tPSA, fPSA and percent free PSA and the patient's age, prostatic volume and histologic diagnosis as assessed by prostatic biopsies or surgical specimens (benign prostatic hypertrophy, carcinoma, hypertrophy with inflammation). In all cases, we calculated the specificity, sensitivity and diagnostic accuracy of percent free PSA in the diagnosis of prostatic diseases, using cutoff values ranging from 14 to 20%. In prostatic cancer patients, we considered the relationships between the various PSA molecular forms and staging, grading and follow-up values. We also evaluated the effects of hormonosuppressive therapy on the serum markers and noted for which tPSA value percent free PSA possessed the greatest diagnostic accuracy.

Results: While tPSA and fPSA values appeared to be correlated with patient age and prostatic volume, percent free PSA did not show a relationship with these parameters. The specificity, sensitivity and overall diagnostic accuracy were better assuming a 16% cutoff value for percent free PSA than with other cutoff values. Prostatic inflammation associated with benign hypertrophy can cause false positives in both tPSA and f/tPSA measurements, since 60% of these patients have an f/tPSA ratio below 16%. In diagnosing carcinoma, the diagnostic accuracy of percent free PSA is 100% when tPSA is between 2.5 and 4.0 ng/ml. Percent free PSA is not linked with staging in prostatic cancer, but it does appear to be related to the Gleason score. In patients receiving hormonosuppressive treatment, f/tPSA decreased significantly, and more so in patients with a higher Gleason score. In patients with disease in rapid progression, percent free PSA was lower than in patients in a stable condition.

Conclusions: Based on our experience, 16% as the f/tPSA cutoff value for discriminating between benign and malignant pathologies is the best possible choice, as it provides the highest overall values of sensitivity, specificity and diagnostic accuracy (80, 61.5 and 84.5%, respectively) in the diagnosis of prostatic cancer. We believe that f/tPSA is not a definitive test for diagnosing prostatic cancer. Our observations on the behavior of percent free PSA in relation to prostatic carcinoma grading and staging and in the follow-up of carcinoma patients are interesting; however, further studies are needed to define the appropriate role of f/tPSA in patients with an established diagnosis of prostatic carcinoma and in the follow-up of patients with prostatic cancer.