Prospective evaluation of prostate specific antigen and prostate specific antigen density in the detection of nonpalpable and stage T1C carcinoma of the prostate.

PURPOSE: We evaluated prospectively prostate specific antigen (PSA) and prostate specific antigen density in the detection of prostate cancer in patients with normal findings on digital rectal examination with and without normal transrectal ultrasound.
MATERIALS AND METHODS: Consecutive patients (184) with an elevated serum PSA and normal digital rectal examination underwent transrectal ultrasound with lesion directed and systematic biopsies (6 if prostatic volume was 50 cc or less and 12 if volume was more than 50 cc). Receiver operating characteristic curves, predictive values and likelihood ratios were calculated for PSA and PSA density.
RESULTS: Of the 184 patients 50 (27%) with a normal digital rectal examination had cancer compared to 30 of 112 (27%) with a normal digital rectal examination and transrectal ultrasound. Median PSA or PSA density did not differ between the positive and negative biopsy groups among patients with a normal digital rectal examination (8.4 versus 7.1 and 0.22 versus 0.14 ng./ml., respectively) or a normal digital rectal examination and transrectal ultrasound (8.2 versus 7.5 and 0.21 versus 0.14 ng./ml., respectively). PSA density was superior to PSA by receiver operating characteristic analysis for cancer detection when all PSA values or those between 4 and 20 ng./ml. were considered. However, the significance was lost for a PSA of 4 to 10 ng./ml. Likelihood ratios demonstrated insignificant changes in the post-test probability if PSA density was used to determine the need for biopsy and many cancers would have been missed.
CONCLUSIONS: PSA density should not be used to determine the need for biopsy in patients with a normal digital rectal examination and/or transrectal ultrasound.