In 1990, a case-control study was conducted in Italy to investigate the possible association between HCV infection and hepatocellular carcinoma (HCC). Serum samples from 65 subjects with newly diagnosed hepatocellular carcinoma and 99 hospital control subjects were tested for the presence of anti-HCV by second-generation ELISA test; positive sera were assayed by RIBA anti-HCV second-generation test. In addition, samples were tested for hepatitis B surface antigen (HBsAg), antibodies to the hepatitis B core antigen (anti-HBc), and antibodies to HBsAg (anti-HBs). The presence of HCV and/or HBsAg serologic markers was significantly associated with hepatocellular carcinoma risk: the relative risk (RR) of HCC was 21.3 (95% CI = 8.8-51.5) for anti-HCV positivity in the absence of HBsAg; the relative risk of HCC was 13.3 (95% CI = 5.5-32.2) for the presence of HBsAg in the absence of anti-HCV. A higher risk (77.0) was observed when both markers were present. These findings indicate that HCV and HBsAg are independent risk factors for HCC. The results of multivariate analysis showed that the adjusted RR linking anti-HCV and HCC was 26.9 (95% CI = 9.9-72.5), the adjusted RR linking HBsAg and HCC was 11.4 (95% CI = 3.1-41.4), whereas no association (RR 1.5; 95% CI = 0.6-3.6) was found to link HCC with anti-HBc and/or anti-HBs positivity. Through the computation of population attributable risk we estimate that 25% of HCC cases occurring in Italy could be attributed to anti-HCV positivity alone and 20% to HBsAg carrier state alone.(ABSTRACT TRUNCATED AT 250 WORDS)