Evaluation of patients with diseases of the prostate using prostate-specific antigen density.

OBJECTIVE

To compare the efficacy of two tests, prostatic-specific antigen (PSA) and the PSA/prostate volume ratio (PSAD), as diagnostic and staging markers to discriminate patients with benign prostatic hyperplasia (BPH) from patients with cancer of the prostate (CaP).

PATIENTS AND METHODS

Prostate gland volumes were estimated in 60 patients with BPH and 88 patients with clinically organ-confined CaP by performing transrectal ultrasonography (TRUS) and using the prolate ellipse formula. Serum PSA concentration was determined using an enzyme immunoassay. In patients with BPH, the prostates were removed either by transurethral resection or retropubic prostatectomy. Patients with CaP underwent laparoscopic pelvic lymphadenectomy followed by radical perineal prostatectomy. PSAD was calculated by relating the serum PSA level to the TRUS-estimated prostate volume.

RESULTS

The median PSA level was 4.4 ng/mL in patients with BPH, 9.3 ng/mL in patients with CaP-NO disease and 24 ng/mL in those with CaP-N+ disease. However, imposing a PSA limit of 4 ng/mL for the diagnosis of CaP gave a positive predictive value of only 64.8%, whereas a limit of 10 ng/mL gave a positive predictive value of 71.4%. In contrast, the median PSAD was 0.086 ng/mL/cm3 in patients with BPH, but was 0.295 ng/mL/cm3 in patients with NO-disease and 0.775 ng/mL/cm3 in those with N(+)-disease. With a limit of 0.15 ng/mL/cm3 the positive predictive value of PSAD was 81%. Furthermore, a limit of 0.6 ng/mL/cm3 revealed a positive predictive value of 81% for the diagnosis of metastatic lymph node involvement.

CONCLUSIONS

There was a considerable overlap of PSA concentrations in patients with BPH and CaP, and PSA was not sufficiently accurate to distinguish between them. In contrast, PSAD enhanced the discrimination between BPH and CaP and may provide additional information about the status of the lymph nodes in patients with CaP.