NATURAL HISTORY OF HEPATITIS B INFECTION

Although chronic hepatitis B virus (HBV) infection can cause serious liver disease and is the major risk factor for hepatocellular carcinoma (HCC) worldwide, risk of adverse outcomes varies substantially between different groups of HBV carriers. Some HBV carriers remain essentially healthy even after many decades of infection. The reasons for these differences in the natural history of infection are not clearly understood, but may relate to differences in both endogenous and environmental factors that interact with the chronic infection. The focus of this laboratory, which is concerned with the natural history of chronic infection, draws on three prospective cohorts where HBV is endemic and the risks of HCC vary considerably. Current and future projects will follow individual carriers over long periods of time in an attempt to explain the observed differences in the natural history of HBV infection. Exogenous and endogenous factors influencing the clearance of active viral replication and alterations in levels of hepatitis B surface antigen (HBsAg) in adult carriers are being pursued through molecular epidemiology. In addition to shedding some light on the natural history of infection and the risk of development of serious liver disease, we have the potential to develop more effective means for identifying HBV carriers at risk for development of HCC.

A poorly understood aspect of the natural history of chronic HBV infection is the transition from the high- to low-titer viremic state that is often coincident with clearance of serum hepatitis B e antigen (HBeAg) and development of antibodies to HBeAg (anti-HBe). In populations where infection is acquired perinatally or in early childhood, this event occurs in most carriers before adulthood (Evans et al., J. Infect. Dis. 176:845, 1997) and, therefore, decades before the occurrence of HCC. Carriers who remain viremic through adulthood may have higher levels of liver damage. Attempts to find an association between a prolonged viremic state and the development of HCC, however, have been inconclusive.

In this study, we examined serologic risk factors for HCC among HBV carriers followed prospectively for up to 5 years in Haimen City, China. Haimen City is an area of extremely high mortality rates for HCC (194 per 100,000 per year in males age 30 to 64) and endemic for chronic HBV infection. In 1992-93, 60,984 men age 30 to 64 were enrolled in a prospective cohort study of HCC in Haimen City, an agricultural area of Jiangsu Province, China. The study is a collaboration with Shanghai Medical University and the Haimen City Anti-Epidemic Station (HCAS). All subjects were residents of Haimen City at the time of study entry and attended mass screenings conducted in their villages by teams from the HCAS. Each subject completed a brief questionnaire about lifestyle, health, and family history, and contributed samples of blood. All blood samples were processed within 24 hours of collection at the HCAS headquarters, and each subject was tested for HBV surface antigen (HBsAg), alanine aminotransferase, and alphafeto-protein.

In serum samples taken at study entry in this prospective cohort, we analyzed HBV markers in 388 male cases (occurring 5 years after study entry) and 1282 matched controls from Haimen City, China. All subjects were positive for HBsAg. HCC risk was strongly inversely associated with serum titer of HBsAg. Carriers in the highest quartile (60 mg/ml) compared to the lowest (4 mg/ml) were protected with a relative risk (RR) of 0.14 (95% CI 0.09-0.23).

Twenty four percent of both cases and controls were HBV DNA(+) by dot blot analysis. When compared to HBV DNA(-)s, however, those with the lowest viral titer (87 pg/ml) had increased risk (RR 2.4, 1.6-3.4) and those with the highest viral titer (525 pg/ml) were protected (RR 0.39, 0.20-0.80). Among HBV DNA(+)s, viral and HBsAg titers are strongly correlated. In multivariate models, once HBsAg titer is controlled for, HBV DNA positivity at any titer had a RR of 2.9 (2.0-4.0) and the RR for the highest quartile of HBsAg titer compared to the lowest was 0.09 (95% CI 0.04-0.16; Figure 1). Marker-associated risks were unrelated to time on study; subjects who developed HCC shortly after study entry did not differ in their HBsAg or HBV DNA-associated risks from those who developed disease 4 or more years after study entry. These data suggest the intriguing possibility that these marker titers are stable features of the high-risk carrier state and could be used as a means of identifying carriers at high risk for developing HCC.

Low HBsAg titer has not been previously reported as a determinant of HCC or other morbidity in HBV carriers and was not associated with liver enzyme elevations in our study. Differences in viral subtype, host immune response, and/or hepatotoxic environmental exposures are potential explanations for these observations.

PUBLICATIONS

1.   EVANS, A.A., O'CONNELL, A.P., PUGH, J.C., MASON, W.S., SHEN, F-M., CHEN, G-C., LIN, W-Y., DIA, A., M'BOUP, S., DRAMÉ, B. AND LONDON, W.T. Geographic variation in viral load among hepatitis B carriers with differing risks of hepatocellular carcinoma. Cancer Epidemiol. Biomark. Prev. 7:559-565, 1998.

EVANS, A.A., FINE, M.K., LONDON, W.T. Are Asian HBV carriers different from non-Asian carriers? Letter to the Editor. Lancet 351:1285-1286, 1998.

EVANS, A.A., LONDON, W.T. The epidemiology of hepatitis B virus. In Viral Hepatitis, edited by A.J. Zuckerman, H.C. Thomas. Harcourt Brace & Co. Ltd., London, pp. 107-114, 1998.

MASON, W.S., EVANS, A.A., LONDON, W.T. Hepatitis B virus replication, liver disease and hepatocellular carcinoma. In Human Tumor Viruses, edited by D. McCance. ASM Publications, Washington, DC, pp. 253-281, 1998.

^§   Fox Chase researcher

^a   S. M'Boup: LeDantec Hospital, University of Dakar, Dakar, Senegal

^b   F-M. Shen: Shanghai Medical University, Shanghai, PRC

Illustrations or unpublished data in these reports should not be used without permission of the author.

Fox Chase Cancer Center

Scientific Report 1998