Wf. Thon et al., PROSTATE-SPECIFIC ANTIGEN DENSITY - A RELIABLE PARAMETER FOR THE DETECTION OF PROSTATE-CANCER, World journal of urology, 14(1), 1996, pp. 53-58
We compared the prostate-specific antigen density (PSAD) in clinically
and surgically staged patients with specimen-confined prostate cancer
(n = 57) and in patients with benign hyperplasia (n = 69), who underw
ent transvesical adenomectomy. The PSAD was calculated from the preope
rative PSA level and the specimen volume. The prostate volume was dete
rmined by dividing the prostate weight by the specific gravity of the
tissue. The mean tissue values used for PSAD calculation were 51.9 g i
n men with prostate cancer (PCA) and 62.9 g in men with benign prostat
ic hyperplasia (BPH). The PSAD values showed significant differences (
BPH 0.19 versus PCA 0.37, P = 0.029), Receiver operator characteristic
(ROC) curves demonstrated the best cutoff value to be 0.15, with the
sensitivity being 58%; the specificity, 51% and the positive predictiv
e value of PCA, 49%. At a serum PSA level below 10 ng/ml, the best cut
off value was 0.1 and the positive predictive value was 51%. The PSAD
results we calculated from an accurate prostate volume (surgical estim
ate) show that PSAD is not a significant predictor of prostate cancer.