Review: Alpha-interferon terminates viral replications and eradicates the carrier state in chronic hepatitis B virus infection
ACP J Club. 1994 Jan-Feb;120:12. doi:10.7326/ACPJC-1994-120-1-012
Wong DK, Cheung AM, O'Rourke K, et al. Effect of alpha-interferon treatment in patients with hepatitis B e antigen-positive chronic hepatitis B: a meta-analysis. Ann Intern Med. 1993 Aug 15;119:312-23.
To determine whether alpha-interferon is effective in terminating viral replication and eradicating the carrier state in patients with chronic hepatitis B virus (HBV) infection.
Randomized controlled trials were identified using MEDLINE (1966 through June 1992) with the key words interferon type 1 and hepatitis B; bibliography review of relevant papers and review articles; and author contact.
Studies were selected if the patients had chronic HBV infection with active viral replication (hepatitis B surface antigen [HBsAg] and hepatitis B e antigen [HBeAg] in serum for > 6 mo), patients were ≥ 18 years of age, treatment was ≥ 3 months, follow-up was ≥ 6 months, and alpha-interferon was used as monotherapy. Duplicate and preliminary reports, abstracts, and letters were excluded.
Data were extracted on loss of HBsAg, HBeAg, and HBV DNA; development of antibodies to HBsAg (anti-HBs) and HBeAg (anti-HBe); normalization of serum alanine and aspartate aminotransferase (ALT and AST) levels; patient and disease characteristics; doses; improvement in histologic indices; and study quality. Intention-to-treat analysis was used, and follow-up losses were considered treatment failures.
40 trials were identified, and 16 were included in the meta-analysis. 837 patients (82% men) were studied. Alpha-interferon was administered subcutaneously or intramuscularly from daily to twice weekly with dosage from 7 million units (MU) to 30 MU/m2per week. Rates of HBeAg clearance ranged from 9.5% to 69.7% in the treatment groups and from 0% to 38.7% in the control groups; all studies showed a trend favoring treatment. Pooled analysis showed improvement in all disease markers in patients who received alpha-interferon compared with control for loss of HbsAg (P = 0.001); loss of HBeAg (P < 0.001); and loss of HBV DNA (P < 0.001) (Table). Treated patients had an increase in anti-HBs (NNT = 16) and in anti-HBe (NNT = 5), and had normalization of ALT levels (NNT = 4). High-dose treatment (> 15 MU/m2 per wk) was more effective than low-dose treatment (P < 0.001). The effect of interferon was not related to duration of treatment (3 vs 6 mo and 6 vs 12 mo), race (Asian [Chinese] vs white), ALT status, or sex. During treatment, 3 of 498 patients in the treatment group died, and 1 of 339 patients in the control group died.
Alpha-interferon is effective in terminating viral replication and carrier status in patients with chronic hepatitis B virus infection. Dose is more important than length of treatment.
Source of funding: None.
For article reprint: Dr. J. Heathcote, 399 Bathurst Street, ECW 7-006, Toronto, Ontario M5T 2S8, Canada. FAX 416-603-9195.
Table. Alpha-interferon vs control for chronic hepatitis B virus infection at up to 12 months of follow-up*
|Outcomes||Weighted event rates||RBI (95% CI)||NNT (CI)|
|Hepatitis B surface antigen||7.8%||1.8%||333% (111 to 500)||18 (12 to 46)|
|Hepatitis B e antigen||33%||12%||175% (125 to 217)||5 (4 to 7)|
|Hepatitis B virus DNA||37%||17%||118% (82 to 153)||5 (4 to 8)|
*Abbreviations defined in Glossary; RBI and CI calculated from data in article.
The meta-analysis by Wong and colleagues contributes to our understanding of the role of interferon in chronic HBV infection in several ways. First, it supports the consensus about the efficacy of alpha-interferon. Numerous small trials have shown a consistent treatment effect in the termination of the viral replicative phase of HBV infection, but the results were not statistically significant. The authors show that nonsignificant treatment effects in individual trials were a function of small numbers and insufficient power. Second, the analysis provides a more precise quantitative estimate of the size of the overall treatment effect. This is important for individual decision making and for predicting the long-term effects of treatment. Finally, the study may help resolve the current controversy about optimal dose and duration of therapy.
Despite the short-term efficacy of interferon, it is not a panacea. Only 1 in 5 treated patients will benefit, and only a few of the estimated 1 million HBV carriers in North America are candidates: those who are eAg or HBV DNA positive, are HIV negative, and have hepatitis but not decompensated liver disease. Also, limited data are available on the long-term effects of therapy on clinically important outcomes, such as liver failure and hepatocellular carcinoma (1). The magnitude of the observed absolute risk reduction may diminish with time because HBeAg to anti-HBe seroconversion occurs spontaneously without therapy. Interferon is expensive, although it appears to be economically attractive (2). Also, nucleic acid analogues are another promising therapeutic option (3, 4). Further economic and long-term clinical studies are needed before the relative roles of interferon and nucleic analogues in the treatment of hepatitis B viral infection are clarified.
Murray Krahn, MD
The Toronto HospitalToronto, Ontario, Canada