To determine the role of hepatitis C virus (HCV) and hepatitis B virus (HBV) in the etiology of hepatocellular carcinoma (HCC) in Greece, blood samples and questionnaire data were obtained from 65 incident cases of HCC as well as from two groups of controls matched to the cases on gender, age, and hospital during a 16-month period in 1991-1992. HBsAg was detected in 61.5% of HCC cases, 12.3% of control patients with metastatic liver cancer and 6.2% of the other controls; anti-HCV seropositivity was 15.4%, 4.6%, and 0 in the three groups, respectively. The odds ratios in logistic regression modelling comparing the HCC cases to the combined control series were 19.2 for the presence of HBsAg and 10.1 for anti-HCV reactivity. To determine the possible role of HCV in 10 liver cancer deaths in Miyazaki, Japan, a nested case control study was performed. Five controls for each HCC case were matched by gender, age, date of screening, and human T-cell lymphotropic virus-I (HTLV-I) status. Eighty percent of 10 cases and 38% of 50 controls had any serologic marker of HCV infection; 70% and 16%, respectively, had anti-HCV. Of 9 liver cancer cases tested, 44% were positive for HCV-RNA compared to 32% of all controls. Only 1 case and 1 control were HBsAg-positive. The odds ratio (OR) estimate for the association between any marker of HCV infection and liver cancer death was 6.2; the relationship with anti-HCV was stronger (OR = infinite, p<0.001) than with HCV-RNA (OR = 2.1). HTLV-I seroprevalence was correlated with HCV infection (OR = 1.6); however, the association of anti-HCV with liver cancer death did not vary by HTLV-I status. Thus, antibody to HCV was strongly associated with death from liver cancer in this population. The concomitant presence of HTLV-I infection did not appear to alter the relationship significantly, although the study sample was small.
