SPECIAL FEATURE Vol.6 No.2 (April 2001)

Antiviral Strategies for Chronic Hepatitis B Infection

Prof. Lai Ching-Lung
Department of Medicine, Queen Mary Hosptial, The University of Hong Kong

The usual endpoint for the treatment of chronic hepatitis B is loss of hepatitis B e antigen (HBeAg) with or without the development of antibody against HBeAg (anti-HBe).

In the Caucasian patients who mostly acquire the disease during adolescence and who progress directly to the immune clearance phase, loss of HBeAg usually signifies stable disease with little subsequent complications. This has been proven with long-term follow-up of untreated and interferon-treated Caucasian patients. In 317 asymptomatic patients (94% negative for HBeAg) followed up for 16 years in Montreal, the risk of cirrhosis complications and hepatocellular carcinoma was small (Villeneuve et al., 1994). In a long-term follow-up study of interferon treatment, the overall survival and survival without complications were significantly longer in patients negative for HBeAg, irrespective of whether the loss of HBeAg was secondary to interferon treatment or was spontaneous (Niederau et al., 1996).

In contrast in the Asian patients who mostly acquire the hepatitis B infection in the early years of life, the disease tends to progress after HBeAg seroconversion. In a Taiwan study of 684 Chinese patients with follow-up biopsies, the annual incidence for the histologic development of cirrhosis was 2.4% for HBeAg-positive patients. However, patients who were positive for anti-HBe still continued to develop cirrhosis at an annual rate of 1.3% (Liaw et al, 1988). Chang et al. (1995) found that the biopsies in 30 Chinese children who had HBeAg seroconversion at the median age of 5.5 years were comparable to those who remained HBeAg positive. Two of the anti-HBe-positive children had cirrhosis with one developing hepatocellular carcinoma at the age of 11 years. In an ongoing follow-up of 3214 Chinese patients at Queen Mary Hospital, the median age of HBeAg seroconversion was 34.6 years. For the 379 patients with complications of cirrhosis or hepatocellular carcinoma, the median age of onset of complications (including ascites, spontaneous bacterial peritonitis, variceal bleeding, encephalopathy and hepatocellular carcinoma) was 56-59 years. The majority (66.7-79.4%) of these patients were positive for anti-HBe. Thus in Asian patients, HBeAg seroconversion is not a good endpoint for efficacy of treatment. Histologic improvement, preferably with HBV DNA levels not detectable by PCR assay, are probably more appropriate short-term endpoints for efficacy of treatment.

There are two possible modes of attack for the treatment of chronic hepatitis B: (1) enhancement of immune clearance of infected hepatocytes, e.g. interferon a, therapeutic vaccines, and (2) suppression of viral replication, e.g. lamivudine, newer nucleoside analogues.

For drugs acting through immunomodulation, the immune system is enhanced to effect the clearance of infected hepatocytes. However, there must be sufficient hepatocyte regeneration capacity in order for the destroyed cells to be replaced. This is why interferon, and most likely therapeutic vaccines, can cause hepatic decompensation in patients who have already developed cirrhosis of the liver. The prognostic factors for poor response to interferon include: males, long duration of infection, Asia population, low alanine transaminase levels, high HBV DNA levels, low to mild histologic activity, precore mutants, homosexuals and other patients with impaired immunity. The first six of these criteria all apply to the Hong Kong Chinese patients.

The nucleoside analogue that is registered for use in hepatitis B patients in Hong Kong, Mainland China and the United States of America is lamivudine. Unfortunately the use of lamivudine as monotherapy is associated with the possible development of the YMDD variant hepatitis virus. The prognostic factors favouring the development of YMDD variants include: large body weight and body mass index, high HBV DNA levels, increasing age, males and, surprisingly, non-Asians.

Lamivudine, and most nucleoside analogues, acts by suppression of viral replication through competitive inhibition of reverse transcriptase and DNA polymerase. It has little effect on the non-replicative component of the hepatitis B virus inside the nucleus of infected hepatocytes, i.e., the covently closed circular DNA (cccDNA). Hence it has to be taken on a long-term basis.

The theoretical duration of therapy of nucleoside analogues has been calculated using mathematical models, based on the calculation of the regeneration rate and death rate of uninfected cells, the virus production rate and death rate of infected cells and the clearance rate of free virus in the circulation. It has been calculated that the half life and life span of the free virus are 24 hours and 35 days respectively, and the half-life and daily turnover rate of infected cells are 10-100 days and 1-7% respectively. Assuming an 100% efficacy of viral suppression, one year of treatment by a nucleoside analogue will decrease the proportion of infected cells to 8 x 10-2-10-11 of the initial value (Nowak et al., 1996). Assuming a 99.3% efficacy (for adefovir dipivoxil) 517 days of treatment should reduce the viral load down to only one copy of HBV DNA. However in practice a longer, probably much longer, period of treatment is required. Part of the reason may be that these calculations do not take into account the presence of the cccDNA in the infected cells, the cccDNA only declining very slowly, if at all, through natural cell death and cell division.

The future of hepatitis B therapy lies in combination therapy, either combining an immunomodulator with a nucleoside analogue or combining two to three nucleoside analogues. Two trials of combination of interferon with lamivudine (8 weeks of lamivudine followed by 16 weeks of interferon and lamivudine) did not show significant improvement over lamivudine or interferon alone with respect to HBeAg seroconversion and histology (Schalm et al., 2000). Newer trials with different timing in the combination of these two drugs are currently being carried out.

Steroid priming prior to giving lamivudine to patients with normal transaminases is reported to be useful in inducing HBeAg seroconversion through the immunomodulatory effects of steroid (Liaw et al., 2000). However steroid withdrawal has been reported to give rise to dangerous, even fatal, exacerbation in patients with chronic hepatitis B. The use of steroid priming is not generally advocated.

Combining two to three nucleoside analogues is analogous to the "combinatorial ledge" in the control of human immunodeficiency virus (HIV). In controlling HIV, combining three drugs would make the development of resistance highly unlikely whereas combining two drugs only will almost certainly lead to resistance. There thus seems to be a "ledge" or threshold beyond which resistance is unlikely to develop. Since hepatitis B virus mutates at a much slower rate compared to HIV, a combination of two drugs may be sufficient to make resistance unlikely for hepatitis B infection. But whether this is true can only be confirmed or disproved by future clinical studies.