PURPOSE This study evaluated the diagnostic performance of two tumor markers, alpha-fetoprotein (AFP) and des-gamma-carboxy prothrombin (DCP), a combination of these tests and other common tests for patients developing hepatocellular carcinoma (HCC) during long-term follow-up for HCV-related liver disease. METHODS We reviewed the medical records of 144 patients who developed HCC after at least one-year follow up for HCV-related liver disease in Yamaguchi University Hospital or Tenri Hospital and demographic, clinical and laboratory data were collected retrospectively. To evaluate diagnostic test performance, we defined the oldest data collected from each patient at least one year before the diagnosis of HCC as negative control data. RESULTS Sensitivity and specificity were 0.52, 0.61 for AFP with a cut-off of 20 ng/ml compared to 0.47 and 1.00 for DCP with a cut-off of 40 mAU/ml, respectively. Combination tests for AFP at 100 ng/ml and DCP at 40 mAU/ml yielded the best sensitivity at 0.84 with a specificity of 0.50. Multivariate logistic regression model using AFP, AST, LD, hemoglobin, prothrombin time and male ratio as variables showed significantly better diagnostic performance, i.e. sensitivity 0.85, specificity 0.74, than single test or combination test using AFP and DCP alone. CONCLUSIONS AFP and DCP alone were not sufficient for detection of HCC developing during long-term follow-up of HCV-related cirrhosis. Logistic regression including AFP and commonly used laboratory tests showed superior diagnostic accuracy compared to that of tumor marker tests alone.