To examine the use of prostate-specific antigen (PSA) density (PSAD) and free to total PSA ratio (f/tPSA) in enhancing the specificity of PSA in the diagnosis of prostate cancer in patients with a total PSA (tPSA) of 4-10 ng/mL and with a normal digital rectal examination (DRE). The study comprised 77 consecutive men in whom the fPSA and tPSA levels were obtained before DRE and transrectal ultrasonography-guided sextant prostate biopsy. Prostate cancer was found in 39 patients and the histology was benign in 38. Receiver operator characteristic curves, obtained from all 77 patients, were used to determine the optimal thresholds for PSAD and f/tPSA in detecting cancer. A subset of 28 patients, including seven with prostate cancer, was identified who had a normal DRE and a tPSA of 4-10 ng/mL; PSAD and f/tPSA values were applied at the optimal thresholds to assess their use in identifying those patients with cancer. When applied to the selected group of 28 patients, the PSAD (threshold 0.15) failed to identify any with prostate cancer. The f/tPSA (threshold 0.12) yielded a sensitivity of 65% and a specificity of 38%, identifying only three of seven patients with cancer. By increasing the threshold to 0.25, six patients were correctly identified, giving a sensitivity of 86%, with a lower specificity of 14%. These findings suggest that the neither PSAD nor f/tPSA either significantly reduce the negative biopsy rate or identify patients at greater risk of prostate cancer, particularly when the tPSA is equivocal at 4-10 ng/mL.