Clinical implications of free-to-total immunoreactive prostate-specific antigen ratios

A study was performed to evaluate the free-to-total prostate-specific antigen (PSA) ratio for discriminating benign prostatic hyperplasia (BPH) or prostate cancer in the intermediate PSA range (2.0-10.0 microg/l) in patients referred for prostate evaluation. In addition, the relationship of free-to-total PSA ratio and tumor grade in prostatic cancer cases, implying a higher concentration of complex PSA in poorly differentiated cancer, was assessed for its predictive value of tumor aggressiveness at the time of diagnosis.

Seven hundred and sixteen patients referred to the out-patient clinics of two urological departments were included in this prospective study. Blood samples were taken for total immunoreactive and free PSA (IMMULITE) determinations prior to any manipulation. The patients were grouped according to their PSA levels: 2.0-4.0 microg/l, 4.0-10.0 microg/l, 10.0-20.0 microg/l and > or = 20.0 microg/l. All patients were categorized, after histological confirmation, as having BPH (n = 423) or prostate cancer (n = 293). In patients with cancer the tumor grade was also assessed.

In patients with serum immunoreactive PSA levels in the 2.0-4.0 microg/l range, a free-to-total PSA ratio lower than 22% predicted the presence of prostate cancer with a sensitivity of 67% and a specificity of 63%. The positive- and negative-predictive values were 29% and 90% respectively. Receiver-operating characteristic curve analysis indicated a free-to-total PSA ratio of 22% to be the optimum discriminatory level in this low PSA range. For patients with a serum PSA level between 4.0 and 10.0 microg/l, the threshold ratio of 18% gave a sensitivity of 70%, a specificity of 70%, a positive-predictive value of 46% and a negative-predictive value of 87%. Men with a well differentiated grade of prostate cancer had higher free-to-total PSA ratios than those with less differentiated tumors (p = 0.01).

Our data indicate that the free-to-total PSA ratio, in patients with prostatic disease and with PSA levels in the 2.0-10.0 microg/l range, gives a significant improvement in prediction of cancer over the total immunoreactive PSA value alone. Because of the correlation between a higher tumor grade and a lower free-to-total PSA ratio, this ratio may be helpful in assessing the risk of a poorly differentiated cancer.