alpha-Fetoprotein (AFP) is frequently used as a diagnostic marker for hepatocellular carcinoma (HCC). Most available data concerning AFP come from studies of patients with chronic hepatitis B or other chronic liver diseases of mixed etiologies. Limited data concerning the diagnostic value of AFP for hepatitis C virus (HCV)-related HCC have to date come only from Asian and European studies, and results are conflicting. There may be significant differences in AFP levels depending on racial backgrounds and etiologies of primary liver disease. We conducted a multicenter, retrospective, case-control study of 163 HCC patients with HCV infection and 149 control patients with HCV-related cirrhosis. The positive likelihood ratios for AFP at 0 to 20, 21 to 50, 51 to 100, and 101 to 200 ng/mL were 0.46, 1.31, 1.15, and 6.90, respectively. No controls had AFP greater than 200 ng/mL. The sensitivity of AFP for the diagnosis of HCC in African Americans with HCV infection was lower than that of patients of all other ethnic groups combined (57.1% vs. 81.6% for AFP > 10 ng/mL, P =.02, and 42.9% vs. 66.0% for AFP > 20 ng/mL, P =.05). The area under the receiver operating characteristics curve was 0.81 for non-African Americans but only 0.56 for African Americans. In conclusion, AFP greater than 200 ng/mL can be used to confirm HCC in patients with HCV-related cirrhosis and a hepatic mass. However, AFP is insensitive for the diagnosis of HCC in African Americans.