Relation between treatment efficacy and cumulative dose of alpha interferon in chronic hepatitis B. European Concerted Action on Viral Hepatitis (Eurohep).

BACKGROUND/AIMS: Alpha interferon (IFN) is an established treatment of chronic hepatitis B. The effect has been shown to be dose related, recommended dose regimens being associated with a doubling of the spontaneous, baseline HBeAg to anti-HBe seroconversion rate. However, the efficacy of IFN treatment in relation to the dose of IFN actually received remains to be established. The aim of this study was to estimate the relative efficacy of IFN as a function of the cumulative IFN dose. In addition we determined if and when a patient returns to his baseline chance of seroconversion after stopping IFN therapy.

MATERIALS AND METHODS

Individual patient data from 10 clinical controlled trials were available for the present analysis, in all, 746 patients, of whom 491 received IFN and 255 were untreated controls. The data were analyzed performing a time-dependent Cox regression analysis of the relative efficacy of IFN using the cumulative IFN dose administered up to any given time during the observation period and the time after termination of therapy as explanatory variables.

RESULTS

In the proposed model, the chance of HBeAg disappearance for a treated patient relative to no therapy was estimated to 2.1 at a cumulative dose of 100 MU and leveled out at about 2.8 at a cumulative dose of 500 MU. The effect of IFN was shown to decay rapidly after discontinuation and after 3 months a patient could be considered to be back to his baseline chance of HBeAg disappearance. These findings show that IFN administered at a dose of 15-30 MU/week should be considered effective (relative efficacy approximately 2) already after 1-2 months of treatment.

CONCLUSIONS

The present findings do not lend any support to the concept that IFN treatment becomes less effective when a certain total dose of IFN has been administered or that the treatment effect reaches beyond 3 months after stopping IFN.