Value of free prostate-specific antigen (Hybritech Tandem-R) in symptomatic patients consulting the urologist

Introduction: Prostate-specific antigen (PSA) is a widely used tumor marker in the detection and follow-up of adenocarcinoma of the prostate. Selectio n of candidates for prostate biopsies is hampered by the lack of specificit y resulting in a large number of unnecessary biopsies. The intention of our study was to compare the percent free PSA (f-PSA; Hybritech Tandem-R) with total PSA and age-specific PSA reference values to evaluate the clinical b enefit in detecting patients with prostate cancer (PC) in a selected group of patients consulting the urologist. The question was whether cutoff point s are influenced by this selection of patients. Methods: A total of 188 pat ients, 114 with benign prostate hyperplasia (BPH) and 74 with PC were selec ted. It is a selected group of patients consulting the urologist. Diagnosis was confirmed in the BPH and PC groups by either ultrasound-guided biopsy or transurethral resection of the prostate or suprapubic adenomectomy or cy stoprostatectomy. Total PSA (t-PSA) and f-PSA of all patients were measured before any manipulation by Tandem-R assay for f-PSA and Tandem-E assay for t-PSA (Hybritech). Mean values of age, prostate volume, t-PSA, f-PSA, perc ent F-PSA were compared in patients with BPH and PC by Mann-Whitney U test. The sensitivity and specificity of t-PSA and age-specific PSA were compare d to the sensitivities and specificities of different cutoff points of perc ent F-PSA. Results: The mean value of t-PSA, f-PSA and percent F-PSA in pat ients with BPH (n = 114) and PC (n = 74) were statistically significantly d ifferent. At PSA levels between 4 and in ng/ml 19% of negative biopsies cou ld be avoided by the use of percent F-PSA (cutoff point 25%). There was no additional benefit of age-specific PSA. At a PSA of < 4 ng/ml 6 of 7 PCs co uld be diagnosed by percent F-PSA (cutoff point 25%), whereas only 1 patien t would be diagnosed by age-specific PSA. Conclusion: Percent f-PSA seems t o decrease the biopsy rate at PSA levels from 4 to 10 ng/ml without missing a relevant number of cancers and to increase the detection rate at PSA <4 ng/ml. Our data indicate that it might be necessary to choose high cutoff p oints (25%; Tandem-E and R assay, Hybritech) in a selected study population consulting the urologist with large glands and a high prevalence of diseas e. However, this situation is not comparable to testing of screening popula tions. No benefit of age-specific PSA could be observed in this study. Copy right (C) 2000 S. Karger AG, Basel.