Background: The ratio of gamma-seminoprotein (g-Sm) and prostate-specific antigen (PSA) has been regarded as being superior over PSA alone as a discriminator between prostate cancer and benign prostatic diseases. In previous studies, PSA and g-Sm were measured by the Eiken kit and the old-version or revised Chugai kit, respectively. We compared the power of g-Sm ratio with that of PSA alone when using Markit-M PSA assay and the revised Chugai g-Sm assay. Methods: Fifty-three patients with prostate cancer having no metastasis and 116 with benign prostatic diseases were enrolled in this study. Prostate-specific antigen was measured by Markit-M kit and g-Sm was measured by the revised Chugai kit. The discrimination power of g-Sm ratio and PSA alone was evaluated with receiver operating characteristic (ROC) curves. Comparisons between prostate cancer and benign diseases were performed with Mann-Whitney U-test and Fisher's exact test.
Results:The optimal cut-off value was set at 3.1 ng/mL for PSA and 0.935 for g-Sm ratio. Sensitivity, specificity and positive predictive value of PSA alone were 81.1, 81.0 and 66.2%, respectively, while those of g-Sm ratio were 73.6, 90.5 and 78.0%, respectively. There was no statistical significance in each value between PSA and g-Sm ratio. Areas under the ROC curves of PSA and g-Sm ratio were 0.881 and 0.866, respectively (P > 0.05). Conclusion: Contrary to the previous reports, g-Sm ratio and PSA were not different in the discrimination between prostate cancer and benign prostatic diseases, which suggested that the discrimination power of g-Sm ratio, and presumably that of the free PSA to total PSA ratio as well, could be considerably influenced by the assay kits for serum PSA and/or g-Sm (free PSA) used. Therefore, the clinical significance of g-Sm ratio should be evaluated for each PSA assay kit.