Background: The present study was undertaken to investigate the value of gamma-seminoprotein (g-Sm) and the g-Sm/prostate specific antigen (PSA) ratio in combination with serum PSA in detecting prostate cancer. Methods: Prostate specific antigen, g-Sm and the g-Sm/PSA ratio were evaluated in 112 patients with untreated prostate cancer and 90 patients without prostate cancer who had serum PSA and gSm levels above their respective detection limits.
Results:When data for all of the patients were analyzed, serum PSA and g-Sm levels were significantly higher and the g-Sm/PSA ratio was significantly lower in patients with prostate cancer than patients without prostate cancer. The serum PSA and g-Sm levels significantly increased and the g-Sm/PSA ratio significantly decreased with advancing clinical stage in patients with prostate cancer. Among the patients with serum PSA levels ranging from 1.8 to 6 ng/mL, the g-Sm/PSA ratio was significantly lower (P < 0.05) and g-Sm levels were lower (P = 0.054) in the patients with prostate cancer than in those without prostate cancer, but serum PSA levels were not significantly different (P = 0.53). A receiver operating characteristic (ROC) analysis demonstrated that the areas under the ROC curves were 0.54 for PSA, 0.65 for g-Sm and 0.69 for the g-Sm/PSA ratio for prediction of prostate cancer in the PSA range from 1.8 to 6 ng/mL, although the ROC analysis suggested that the g-Sm/PSA ratio does not provide significant advantage over PSA in detecting prostate cancer when all of the patients were analyzed.
Conclusions:These results suggest that the g-Sm/PSA ratio and g-Sm may facilitate differentiation between patients with and without prostate cancer who have intermediate PSA levels.