Effect of conventional interferon‐α in patients with HBeAg‐positive chronic hepatitis B: a systematic review and meta‐analysis
Journal Articles
Overview
Research
Identity
Additional Document Info
View All
Overview
abstract
AbstractBackground Although a few studies have tested the effect of interferon‐α on chronic hepatitis B, its treatment effect remains uncertain, and the association of treatment effect with intervention characteristics has not been thoroughly explored. This study examined the effect of IFN‐α in patients with HBeAg‐positive chronic hepatitis B, and investigated the characteristics associated with treatment effect.Methods We searched MEDLINE, Scientific Citation Index, Current Content Connect, Cochrane Controlled Trial Register, and Chinese Biomedical Database, all up to 15 September 2009. We included randomized trials comparing IFN‐α to placebo, no treatment, or standard care (SC) in patients with HBeAg‐positive chronic hepatitis B. Two reviewers assessed the risk of bias and extracted data, independently and in duplicate. We conducted meta‐analyses of the included studies, and subgroup analyses to examine the association of pre‐specified characteristics (eg, dose, treatment duration) with treatment effect.Results A total of 31 randomized controlled trials, involving 2164 patients, were included. The risk of bias varied across studies. Compared with placebo, no treatment, or SC, IFN‐α improved loss of HBeAg (OR 2.36, 95% CI 1.83 to 3.04), HBV DNA undetectability (OR 2.04, 95% CI 1.28 to 3.32), HBeAg seroconversion (OR 1.82, 95% CI 1.26 to 2.62), ALT normalization (OR 1.24, 95% CI 1.01 to 1.56), and loss of HBsAg (OR 2.45, 95% CI 1.22 to 4.91). Treatment effects differed in high versus low dose, and long versus short duration of IFN‐α. The effect of high dose IFN‐α (OR 3.28, 95% CI 2.31 to 4.66) is statistically larger than that of low dose IFN‐α (OR 1.58, 95% CI 1.10 to 2.28) on loss of HBeAg (interaction P = 0.017), and longer IFN‐α treatment durations produce greater effects (OR 3.28, 95% CI 2.16 to 5.00) than do shorter durations (OR 1.94, 95% CI 0.42 to 2.66, interaction P = 0.038). High dose IFN‐α had a significant effect on HBV DNA undetectability (OR 2.80, 95% CI 2.03 to 3.86), while low dose IFN‐α did not (OR 0.93, 95% CI 0.61 to 1.41, interaction P = 0.01); longer treatments significantly improved HBV DNA undetectability (OR 2.58, 95% CI 1.62 to 4.12), but shorter durations did not (OR 1.28, 95% CI 0.83 to 1.97, interaction P = 0.024).Conclusions IFN‐α can improve serological, biomedical, and virological response. Higher doses and prolonged treatments appear to have larger treatment benefits than lower doses and shorter treatments. However, the increased adverse reactions and costs associated with higher doses and prolonged treatment warrant caution in applying these results.
