Major Finding: Annual incidence rates of hepatocellular carcinoma and liver-related death were 0.06% and 0.04%, respectively, among 1,932 carriers of inactive hepatitis B virus (HBV). The annual incidence rates were 0.02% for hepatocellular carcinoma and 0.02% for liver-related death among 18,137 controls.
Data Source: A prospective cohort study involving 20,069 subjects.
Disclosures: This study received grant support from Bristol-Myers Squibb Co.; Department of Health, Executive Yuan, National Health Research Institutes, Taiwan; and Academia Sinica, Taiwan. The investigators reported that there were no financial or other disclosures related to the study.
Carriers of inactive hepatitis B virus are at increased risk of hepatocellular carcinoma and liver-related death, compared with noncarriers, according to findings from a large, prospective cohort study.
The annual incidence rates of hepatocellular carcinoma and liver-related death in 20,069 participants in the study, which had a mean follow-up of more than 13 years, were 0.06% and 0.04%, respectively, among 1,932 carriers of inactive hepatitis B virus (HBV). The annual incidence rates were 0.02% for hepatocellular carcinoma and 0.02% for liver-related death among 18,137 controls, wrote Dr. Jin-De Chen of National Taiwan University Hospital, Taipei, and colleagues.
Multivariate-adjusted hazard ratios for carriers, compared with controls, were 4.6 and 2.1 for hepatocellular carcinoma and liver-related death, respectively, the investigators reported (Gastroenterology [doi: 10.1053/j.gastro.2010.01.042]).
The authors used data from the Risk Evaluation of Viral Load Elevation and Associated Liver Disease/Cancer-Hepatitis B Virus (REVEAL–HBV) Study for their analysis. Inactive HBV carriers were those who had seronegative hepatitis B e antigen status, anti–hepatitis C virus (HCV)-seronegative status, serum levels of HBV DNA less than 10,000 copies/mL without cirrhosis, hepatocellular carcinoma, or increased serum alanine transaminase (ALT) levels. Controls were participants who were seronegative for HB surface antigen and antibodies against HCV, but who had similar clinical liver features (normal serum ALT, without cirrhosis or hepatocellular carcinoma at study entry). Together, the groups contributed 262,122 person-years of follow-up.
Significant predictors of hepatocellular carcinoma in the entire cohort, compared with controls, were older age (hazard ratio of 2.7/10-year increment), high-normal baseline ALT level (HR 2.2), and alcohol drinking habit (HR 2.4). Significant predictors in inactive HBV carriers, compared with controls, were older age (HR 2.6/10-year increment) and drinking habit (HR 3.7).
The risk of hepatocellular carcinoma was higher in those with baseline serum HBV DNA levels of 300-10,000 copies/mL, compared with those with undetectable serum HBV DNA at baseline, but the difference did not reach statistical significance (HR 1.6). In inactive carriers with undetectable serum HBV DNA at baseline, alcohol drinking habit was a significant predictor of hepatocellular carcinoma (HR 6.9).
Significant predictors of liver-related death in the entire cohort, compared with controls, were similar to those for hepatocellular carcinoma, and included older age (HR 2.3/10-year increment), high-normal baseline serum ALT level (HR 1.9), and alcohol drinking habit (HR 2.1). Among inactive HBV carriers, only older age (HR 2.6/10-year increment) and alcohol drinking habit (HR 5.8) were significant predictors.
The inclusion of hepatocellular carcinoma as a time-dependent event in the analysis revealed that it was the most striking risk predictor for liver-related death (HR 611 for the entire cohort, and 451 for inactive HBV carriers), as would be expected, the investigators noted.
In those without newly developed hepatocellular carcinoma, older age (HR 2.0/10-year increment) and alcohol drinking habit (HR 2.3) were significant risk factors for liver-related death; in inactive carriers without hepatocellular carcinoma, only older age was a significant predictor (HR 2.4).
No significant differences were seen in the development of hepatocellular carcinoma or liver-related death between inactive carriers with undetectable and detectable serum HBV DNA, the investigators noted.
The study is limited by a lack of histologic data, since liver biopsies were not applicable in the large community-based study. Also, serial HBV DNA measurements were not taken to assess for continued inactive carriage, and other risk factors, such as HBV genotypes, were not assessed in those with HBV DNA levels greater than 100,000 copies/mL to determine an association with hepatocellular carcinoma in inactive carriers.
Future studies that include serial measurement of hepatitis B surface antigen serostatus and serum HBV DNA levels would help clarify the natural history of inactive HBV carriage, they concluded.