It has been proven that chronic hepatitis B virus (HBV) infection may have a progressive course ending in liver cirrhosis or hepatocellular carcinoma. 1,2 Interferon alfa (IFN-␣) has been used as therapy for this disease, leading to a sustained loss of HBV DNA and hepatitis B e antigen (HBeAg) (and seroconversion to anti-HBe) and biochemical improvement (normalization of alanine transaminase [ALT] values) in around 25% to 40% of the treated patients. 2-10 However, the remaining patients do not achieve a response, including those patients who relapse after achieving an initial response and patients who do not respond. These patients may potentially benefit from another cycle of IFN-␣ treatment.Attempts in pilot studies of retreatment with IFN-␣ for previous nonresponders provided no definite conclusions on its possible efficacy in chronic HBV. 11 No controlled trial evaluating the benefits of retreatment of chronic HBV with IFN-␣ has been published. The objective of our study has been to determine whether a second course of IFN-␣, given during 6 months, is safe and acceptable, and whether it results in a significantly higher treatment response (loss of HBeAg and HBV DNA) than no treatment in previously nonresponder patients with chronic HBV.
PATIENTS AND METHODSStudy Design. This study was designed as a multicenter (see Appendix) randomized controlled trial in patients who failed to respond to a previous cycle of IFN-␣. Assuming that patients with no treatment show a 5% response, and patients with retreatment 25% (a 20% difference), 7 a total number of 145 patients was needed to detect a significant difference at the 5% level with a power of 80%, taking into account a possible 25% drop-out rate during the study. The inclusion period started in April 1992 and ended in December 1996. At that time, 62 patients who were HBeAg-and HBV DNA-positive in serum with abnormal ALT values (documented at least 3 times within the 6 months screening period) and a chronic HBV 12 documented in a liver biopsy obtained within 1 year before inclusion had been enrolled. The patients should have undergone a previous treatment of only one course of any type of IFN-␣ treatment (recombinant or lymphoblastoid) for at least 12 weeks with a minimum of 13.5 million units (MU) per week 8,13 ; at the end of the first IFN-␣ cycle ( at least 6 months before enrollment in this study) patients should have been HBeAg-positive.The following patient exclusion criteria were used: (1) evidence of any other type of liver disease, i.e., anti-hepatitis C virus or anti-hepatitis D virus positive, alcohol, ascites, bleeding varices, or Abbreviations: HBV, hepatitis B virus; HBeAg, hepatitis B virus e antigen; ALT, alanine transaminase; HBsAg, hepatitis B virus surface antigen; IFN-␣, interferon alfa.From the