Va lues for the free to total prostate-specific antigen ratio as a function of age: Necessity of reference range validation

Objective: To measure the levels of free prostate-specific antigen (PSA), t otal PSA, and free to total PSA ratio in a population of men with no known prostate pathology aged from 20 to 70 years. Patients and Methods: Serum to tal PSA and free PSA values were determined in 1,502 patients due for a sys tematic health examination. The digital rectal examination was only propose d for those over 50 years of age. The assays were determined on the AsXYM a pparatus, from Abbott laboratories, by MEIA technology with monoclonal anti bodies. Results: 1,274 men were available for study. The mean age was 43.6 +/- 11 years (range 20-69 years). The total PSA level was stable up to 40 y ears. Beyond that, it increased with age. There was a linear regression bet ween the age and the logarithm of the total PSA rate (r = 0.26, p < 0.0001) from 40 to 70 years. The upper limit of the normal value (95th percentile) increased from 1.07 for the 20- to 30-year age range to 2.82 for the 60- t o 70-year range. The free PSA level was stable up to 50 yea rs of age. It t hen significantly increased. The upper limit of the normal value was measur ed as 0.42 in the range of 20-30 years and as 0.53 in the range of 60-70 ye ars with an annual average increase rate of roughly 0.5%. Overall there was a linear regression between age and the free PSA rate (r = 0.12, p < 0.000 1). The upper limit of the free to total PSA ratio, measured as being 0.68 in the range of 20-29 years, dropped towards 60-69 years with an upper limi t of the normal of 0.48. The average annual reduction rate was around 0.70% . There was a linear regression between the age and the free to total PSA r atio (r = 0.17, p < 0.0001). Conclusion: These total PSA levels are lower t han the ones measured in other studies with other assay methods. These vari ations stress the importance of validating reference values of total PSA an d free PSA as a function of the assay method and the population to which th ey are applied before using them as an aid in the diagnosis of prostate can cer.