Interferon Treatment in Children with Chronic Hepatitis B: an Israeli experience

Transcription

1 Interferon Treatment in Children with Chronic Hepatitis B: an Israeli experience Corina Hartman MD 1, Hanoch Hager MD 4, Drora Berkowitz MD 1, Nurit Rimon PhD 2, Lucyna Bassan MD 3, Phillipe Trogouboff MD 5, Amnon Teitler MD 6 and Raanan Shamir MD 1 1 Division of Pediatric Gastroenterology and Nutrition, Department of Pediatrics, 2 Laboratory of Virology and 3 Department of Pathology, Rambam Medical Center, Haifa, Israel 4 Pediatric Gastroenterology and Nutrition Unit, Department of Pediatrics, 5 Department of Pathology and 6 Laboratory of Immunology, HaEmek Medical Center, Afula, Israel Affiliated to Technion Faculty of Medicine, Haifa, Israel Key words: chronic hepatitis B, children, interferon, Abstract Objective: To summarize the experience of two pediatric gastroenterology centers in northern Israel using interferon alpha to treat children with chronic hepatitis B. Patients and Methods: We retrospectively reviewed the medical records of 59 children with chronic HBV. Forty-nine children were treated with subcutaneous IFNa at dosages of 10 MU/m (n = 12) or 5 MU/m (n = 37) three times a week for 6 months. Children treated with 5 MU/m IFNa were a heterogeneous group with regard to their alanine aminotransferases levels: 11/37 (group 1) had ALT twice above the upper limit of normal and 26/37 (group 2) had normal or less than twice ULN elevations of ALT. Ten children were observed without treatment. Results: Sustained biochemical and virologic response occurred in 9/12 children (75%) treated with 10 MU/m IFNa, 3/11 (27%) in group 1 given 5 MU/m (P = ) and 4% (1/26) in group 2 treated with 5 MU/m. After 4 years follow-up, similar seroconversion rates were present in children who had baseline ALT higher than twice ULN, whether they were treated with 10 or 5 MU/m IFNa (83 vs. 73%). Children with normal or slightly abnormal baseline ALT also had a similar anti-hepatitis B e antigen seroconversion rate at the end of follow-up, whether they had been treated with IFNa or not (2/26, 8%, in group 2 and 4/10, 40%, in the untreated children, not significantly different). Conclusions: The results of IFNa treatment in this group of Israeli children are similar to these reported in the literature and reinforce the current consensus, which recommends INFa only for children with chronic HBV and significantly elevated liver enzymes. IMAJ 2002;4(Suppl):914±918 Hepatitis B virus infection remains a major health problem worldwide. The risk of developing chronic HBV infection after an acute infection is highest in infancy and early childhood, and chronic HBV infection is associated with significant morbidity and mortality in adult life [1]. In individuals who acquired HBV in infancy or early childhood, chronic sequelae of HBV infection including cirrhosis and hepatocellular carcinoma may develop later in life [2]. Given the above, effective therapy for chronic HBV infection in children is desirable. Currently, interferon alpha and the newer nucleoside analogue, lamivudine, are the only therapeutic agents approved for the treatment of chronic HBV in children. While little experience has been acquired with the use of lamivudine in children, IFNa treatment has been the mainstay of therapy for children with chronic HBV for several years. A 3±6 month course of IFNa in doses of 3±10 MU 3 times a week resulted in suppression of HBV replication and amelioration of liver disease in 30±40% of European adults as well as children, compared to 10±15% of control subjects [3,4]. However, the reported response rate in an Asian population was 15±20% [5]. This lower response rate is characteristic of areas with a high HBV infection prevalence (15±20% hepatitis B surface antigen carriers in the general population) and has been related to perinatal transmission and young age at infection acquisition, associated with a greater immune tolerance. Israel is a country with an intermediate prevalence of HBV infection and a HBsAg carrier frequency of 0.5±5.8%, depending on the ethnic background [6]. A 31.5% response rate to IFNa was reported in Israeli children with chronic hepatitis B, in a study in which 78% of patients were immigrant Jewish children from the former Soviet Union [7]. Having used IFNa to treat chronic HBV in a group of Israeli Arab children for several years, we summarize here the efficacy, safety, and long-term follow-up of these children treated at two pediatric gastroenterology centers in northern Israel. Patients and Methods Patients We retrospectively reviewed the medical records of 59 children diagnosed with chronic HBV between November 1987 and January 1999 at two referral pediatric gastroenterology units in northern Israel. The diagnosis was based on evidence of active viral replication (positive serum HBsAg), hepatitis B e antigen and HBV DNA) for at least 6 months and evidence of inflammation on liver biopsy. The study population was heterogeneous with regard to serum alanine aminotransferase levels. No patient had clinical or biochemical evidence of decompensated liver disease. Hepatitis D HBV = hepatitis B virus IFNa = interferon alpha ALT = alanine aminotransferase ULN = upper limit of normal HBsAg = hepatitis B surface antigen HDV = hepatitis D virus 914 C. Hartman et al. IMAJ

2 virus (anti-hdv) antibodies were negative in all patients. One child had hepatitis C virus supra-infection (anti-hcv and HCV RNA positive). Informed consent from the children's parents or legal guardians was requested before the performance of liver biopsies and the start of INF treatment. Children with serum ALT levels more than twice the upper limit of normal were treated with subcutaneous recombinant 10 MU/m INF (Intron-A, Schering-Plough, USA) 3 times a week (n = 12) in one center, and 5 MU/m 3 times a week (n = 11) in the other center, for 6 months. Children with normal or slightly elevated (less than twice ULN) serum ALT levels were treated with recombinant 5 MU/m INF (Intron-A) 3 times a week, subcutaneously, for 6 months (n = 26) or followed without treatment (n = 10). All children had their IFNa administered at bedtime together with paracetamol in order to avoid the flu-like side effects. In children treated with IFNa, clinical and laboratory evaluation including history, physical examination, complete blood count and liver function tests were performed at treatment entry, every 2 weeks in the first month and every month thereafter during treatment. Serologic and virologic tests, including HBsAg, HBeAg and HBV DNA were obtained at baseline, every 3 months during IFNa treatment, and every 6 months for an additional 18 months or until seroconversion. Yearly physical examination, liver function tests, alpha-fetoprotein and serologic tests were performed in children followed without treatment and in children treated with IFNa, thereafter. Methods HBsAg, HBeAg, anti-hbe and anti-hdv were tested by commercial radioimmunoassays (Axsym system, Abbott Laboratories, Chicago, IL, USA). Anti-HCV antibodies were tested by third-generation enzyme-linked immunassay (Axsym system). Serum HBV DNA was assayed by liquid hybridization technique, with 1.5 pg/ml being the lower limit of detection. HBV DNA levels were analyzed semiquantitatively and considered negative when less than 1.5 pg/ml, low positive between 1.5 and 100 pg/ml and high positive when more than 100 pg/ml. HCV RNA was detected in serum by polymerase chain reaction, using nested primers from the 5' noncoding region of HCV [8]. Liver biopsies were assessed for histologic activity index according to the method described by Knodell et al. [9]. Statistical analysis The differences between the two treatment groups were evaluated using a two-tailed Student's t-test. For single items the Fisher's exact test was used. The variables analyzed were age, gender, ethnic distribution, source of infection, HAI and seroconversion rates. Significance was defined as P < Results Table 1 shows the main demographic data and disease characteristics of children with chronic HBV and high serum ALT levels. As HCV = hepatitis C virus HBeAg = hepatitis B e antigen HAI = histologic activity index Table 1. Demographic and clinical characteristics and the results of IFNa treatment in children with chronic HBV and highly elevated baseline ALT Age (yr) 10 MU/m3/wk (n=12) 5 MU/m3/wk. (n=11) Median (range) 7.5 (4.5±17) 8.5 (5.3±14.5) Gender (%) Male 11 (92) 9 (82) Female 1 (8) 2 (18) Ethnic origin (%) Israeli Arabs 11 (92) 7 (64) Russian immigrant Jews 1 (8) 4 (36) Route of HBV infection (%) Intrafamilial 8 (67) 5 (45) Parenteral 1 (8) 0 Unknown 3 (25) 6 (55) ALT* Median (range) 3.5 (2.3±8.7) 3.1 (2.2±6.3) HAI score Median (range) 8 (3±14) 8 (2±12) Overall seroconversion rate (%) 10 (83) 8 (73) INFa response rate** 9 (75) 3 (27) Delayed seroconversion 1 (8) 5 (45) HbsAg seroconversion 2 (16) 0 Side effects (%) 11(92) 8(73) Flu-like symptoms 11(92) 8(73) Anorexia with weight loss** 12(100) 4(36) ALT flare-up** 10(83) 4(36) Other 3(25) 0 Follow-up (yr) Median (range) 5 (1±6) 3.7 (1±5.5) * ALT = alanine aminotransferases, expressed as times the upper limit of normal ** The difference between the two groups was statistically significant (P < 0.05) shown, the majority of children were Israeli Arab males. The route of infection was defined as intrafamilial in 13 of the 18 Arab children. In these children, HBV infection was detected in other family members, parents and/or siblings, but the mode of transmission (vertical or horizontal) could not be established. In another nine children (39%), the source of HBV infection was not identified. All biopsies showed evidence of chronic hepatitis with activity scores in the minimal to moderate range. Cirrhosis was present in one biopsy. None of the examined baseline clinical or laboratory parameters was statistically different between the children given 5 or 10 MU/ m IFNa. Sustained biochemical and virologic response to treatment ± defined as ALT normalization, sustained clearance of HBV DNA and HBeAg from serum, and seroconversion to anti-hbe, during IFNa treatment and/or within 12 months after the end of therapy ± was significantly higher in children treated with 10 MU/m IFNa- (9/12) than in children given 5 MU/m (3/11) (75% vs. 27%, P = 0.03). During the follow-up, however, delayed seroconversion (HBV DNA, HBeAg clearance and development of anti-hbe after 12 months from the end of IFNa therapy) occurred in another 6 patients, including 5 children from the 5 MU/m IFNa group, and 1 child from the 10 MU/ m group. Therefore, at the end of 4 years follow-up, the overall IMAJ Interferon for Chronic Hepatitis B in Children 915

3 seroconversion rate was similar in the two treatment groups (83 vs. 73%). Follow-up biopsies were not performed. All responders remained HBV DNA, HBeAg negative and anti- HBe positive, with normal transaminases throughout the follow-up period. Clearance of HBsAg with the appearance of anti-hbs antibodies (3 months and 3 years respectively from the end of IFNa treatment) were observed in two children, both from the group treated with 10 MU/m. The child with HCV superinfection was treated with 10 MU/m, and had both HBV (including HBsAg to anti- HBs seroconversion) and HCV clearance (negative HCV RNA). Table 2 shows the demographic and clinical characteristics of children with chronic HBV infection and normal or slightly abnormal ALT, grouped according to whether or not they were treated with IFNa. The two groups of children were comparable with respect to age, gender, ethnic origin and route of infection. Analysis of baseline disease characteristics in patients treated with IFNa as compared to patients in the observation group showed no significant differences with regard to HBV DNA and serum ALT levels. The children in the observation group had a significantly longer follow-up (P < ). Sustained biochemical and virologic response to IFNa therapy occurred in one child from group I, 4 months after the start of treatment (4%). Another child underwent delayed HBV DNA clearance and HBeAg to anti-hbe seroconversion associated with ALT normalization 18 months after the end of IFNa treatment (8%). In the observation group, 4 (40%) children underwent spontaneous HBeAg to anti-hbe seroconversion during the follow-up, from 12 to 75 months after the diagnosis of HBV infection (not statistically Table 2. Demographic and clinical characteristics of children with chronic hepatitis B and normal or minimally elevated liver aminotransferases Age (yr) Group 1 (n=11) IFNa 5 MU/m3/wk Group 2 (n=10) Observation group Median (range) 12.9 (7.9±18.5) 14.6 (13±19) Gender (%) Male 20 (77) 8 (80) Female 6 (23) 2 (20) Ethnic origin (%) Israeli Arabs 24 (92) 9 (90) Ethiopian Jews 2 (8) 1 (10) Route of infection (%) Intrafamilial 20 (77) 9 (90) Unknown 6 (23) 1 (10) ALT* Median (range) 2.2 ( ) 1.4 (1-1.8) HAI Median (range) 6 (3-8) Seroconversion rate (%) 2 (8) 4 (40) ALT normalization (%) 6 (23) 8 (80) Side effects (%) 18 (70) 0 (0) Flu-like symptoms 18(70) Other 7(27) Follow-up (mo) Median (range) 42 (25±57) 85 (12±156)** * ALT expressed as times the upper limit of normal ** The difference was statistically significant different from the seroconversion rate in the treated group). ALT normalization without HBV DNA or HBeAg clearance occurred in another 4 children from group 1 (16%) and an additional 4 children from the control group (40%) during the follow-up. In all patients HBeAg to anti-hbe seroconversion was sustained and associated with normalization of ALT levels. HBsAg clearance did not occur in any patient in this study. All children treated with IFNa completed the 6 months of therapy. The early side effects (fever, chills, malaise, myalgia and headache) seen after the initial doses were present in almost all children and decreased in intensity with further doses. Poor appetite was reported by all children from the high dose group during treatment, but only by a few in the 5 MU/m group (P = 0.04). ALT flare-up, 3±18 times above twice ULN, occurred more often in the high dose group (P = 0.04). Only in two children did ALT values rise above 500 IU/L (both from the high dose group), but none manifested signs of hepatic decompensation. Neutropenia, thrombocytopenia and urticarial rash were each present in two children, but only in the high dose group. All the adverse effects were transient: neutropenia responded promptly to transient IFNa dose reduction, and urticaria responded to oral antihistamines. None of the children in the observation group had clinical complaints related to HBV infection during the follow-up, and the difference in the adverse events between treated and observed patients was statistically significant (P = 0.001). Discussion This study relates the experience of two pediatric gastroenterology and nutrition centers in northern Israel in the treatment of children with chronic HBV infection, and confirms the results reported in the literature. We found that in children with highly abnormal ALT, the 10 vs. 5 MU/m IFNa dose regimen resulted in a significantly higher sustained response rate (75 vs. 27%) at one year after the end of treatment, but similar seroconversion rates (83 and 73% respectively) after 4 years follow-up. In addition, a 6 month course of 5 MU/m IFNa in children with chronic hepatitis B and normal or minimally elevated ALT was not more beneficial than no treatment with respect to HbeAg/anti-HBe seroconversion and ALT normalization, but was accompanied by the inherent, although transient side effects associated with IFNa therapy. Until recently, IFNa was the only option available for the treatment of children with chronic hepatitis B. In 1999, a European panel of pediatricians and hepatologists published their experience, opinions and suggestions regarding ``Interferon-alpha treatment of chronic hepatitis B in childhood: a consensus advice based on experience in European children,'' with the aim of providing a guideline for IFNa treatment in children with this disease [10]. Based on the results of 11 controlled trials in European children ± 351 treated with IFNa and 267 controls ± the consensus stated that candidates for treatment are children who are HBeAg and HBV DNA positive, with lowintermediate HBV DNA levels and abnormal ALT values, aged 2 years or more. The standard regimen proposed was 5 MU/m 2 of IFNa intramuscularly or subcutaneously three times weekly for 6 months. There was, however, much debate on every topic of the final consensus, including patient selection and IFNa schedule. The 916 C. Hartman et al. IMAJ

4 optimal dosage and duration of IFNa treatment have never ceased to be investigated in adults and children alike. Besides the commonly standard IFNa regimen, other schedules with different doses, prolonged or repeated treatments were tried in the hope of improving the response rate. Higher doses were deemed to be associated with response rates no better than rates achieved with 5 MU/m, but a much higher rate of side effects in some studies, although not in all [11±17]. Two studies compared 10 and 5 MU/m dose regimens [15,17]. Ruiz-Moreno et al. [15] compared 10 and 5 MU/m for 6 months with no treatment and found a significant difference in response rate only between the 10 MU/m group and controls (58 vs. 17%). The response rate was slightly higher with 10 compared to 5 MU/m, but the difference was not statistically significant. Gurakan et al. [17], in a randomized study, examined the efficacy of 10 and 5 MU/m IFNa for 6 months in a group of children with chronic hepatitis B. At 12 months after the end of treatment, 60% achieved complete response in the group given the higher dose, compared with a 33% response rate in the 5 MU/m group, but the difference was not statistically significant. Our study did find a statistically significant difference in the response rate between the 10 and 5 MU/m regimens (75 and 27% respectively), and although not significant, HBsAg clearance was observed only in the 10 MU/m regimen. However, more severe transient side effects were present in the 10 MU/m regimen, and after 4 years follow-up, similar high seroconversion rates were recorded for both treatment doses (83 and 73% respectively). It has been argued that the prevention of liver cancer necesitates treatment initiation before the viral DNA has been integrated into hepatocytes' host genome. In addition, ongoing HBV replication is thought to be responsible for the perpetuation of liver injury leading in the end to cirrhosis and/or hepatocellular carcinoma [18]. Therefore, the significantly faster response should favor the use of the 10 MU/m regimen. However, the importance of the difference in short-term response between the two treatment regimens remains questionable until future studies examine the occurrence of cirrhosis and hepatocellular carcinoma in children treated with different IFNa regimens or not treated at all. With regard to patient selection, even though most studies treat only children with highly elevated liver aminotransferases and low HBV DNA, other investigators will consider for treatment those children with normal ALT and all range HBV DNA levels, too. This approach was justified by these investigators based on the experience that ALT levels do not always reflect the degree of inflammatory activity present on liver biopsies, which would be a better prognostic factor for successful IFNa therapy than ALT levels [10]. The need for patient selection derived from two factors: the poor response rate to IFNa treatment in patients with chronic hepatitis B and normal ALT on the one hand, and the good longterm prognosis of these patients on the other [19]. Many studies analyzed the factors influencing the response to IFNa in adults and children with chronic hepatitis B. The most consistent variables found to be associated with high HBeAg seroconversion rate were pre-treatment ALT and HBV DNA levels [14,20]. In addition to, and independent of ALT or HBV DNA levels, other factors were found to be predictors of higher response rate to IFNa therapy, such as postnatal infection, history of acute hepatitis, younger age at the time of IFNa treatment, female gender, higher HAI on liver biopsies, IFNa dose and/or treatment duration [21,22]. The largest experience with IFNa treatment in adults and children with chronic hepatitis B and normal liver enzymes came from the studies done in the early 1990s in Asian patients. Prospective randomized trials in Chinese children with normal liver enzymes showed poor response rates to IFNa, similar to the results in Chinese adults [11,23]. Similar studies were not performed in European children. Few European studies, which stratified the study population according to pre-treatment ALT values, also showed low response rates to IFNa in patients with normal liver enzymes [16,24,25]. Based on this and on the experience in the Asian patients it is accepted that normal or minimally elevated ALT values are associated with poor response to IFNa, but how poor this response is remains unclear. In this study none of the children with normal ALT underwent seroconversion after IFNa. In children with slightly abnormal ALT, the seroconversion rate was similar whether they were or were not treated with IFNa. Therefore, poor IFNa response rates were reported in both European and Asian patients with chronic hepatitis B and normal aminotransferases, implying that this is not a particularity of Asian patients. It is accepted that the poor antiviral response in these patients is due to immune tolerance to HBV, induced by exposure to the virus in early life, which is the usual mode of infection acquisition in regions of high and intermediate endemicity. Hence, the Israeli children with chronic hepatitis B and normal ALT from this study behave similarly to the Asian patients; and when treated with IFNa, ALT was a poor prognostic sign independent of the genetic background. There are several limitations to this study. The study was retrospective, the sample size for each study group was small, there was no randomization of treatment or control groups, and the follow-up time was dissimilar. However, the study population was homogenous with respect to the demographic and clinical characteristics. Even though the results might seem unique to this particular ethnic group of native Israeli Arab children, they are in agreement with the general experience and are so straightforward that they are hard to disregard. In conclusion, in this cohort of northern Israeli children with chronic HBV infection and abnormal ALT, 10 MU/m IFNa treatment for 6 months achieved a higher complete response rate as compared to the 5 MU/m dose. However, after 4 years of followup there were no differences in seroconversion rates between the two regimens. The probability of more severe complications with the 10 MU/m regimen should be weighed against possible longterm benefits regarding prevention of cirrhosis and heptocellular carcinoma. In children with chronic hepatitis B and normal or minimal elevated liver enzymes, IFNa treatment had no short or long-term benefit compared to no treatment but was associated with significant although transient side effects. Our results reinforce the current approach, which does not recommend IFNa for children with chronic hepatitis B and minimal activity as reflected by normal or minimally elevated liver enzymes. IMAJ Interferon for Chronic Hepatitis B in Children 917

5 References 1. Di Marco V, Lo Iacono O, Camma C, et al. The long-term course of chronic hepatitis B. Hepatology 1999;30:257± Fujisawa T, Komatsu H, Inui A, et al. Long-term outcome of chronic hepatitis B in adolescents or young adults in follow-up from childhood. J Pediatr Gastroenterol Nutr 2000;30:201±6. 3. Wong DK, Chung AM, O'Rourke K, Naylor CD, Detsky AS, Hearthcote J. Effect of alpha-interferon treatment in patients with hepatitis B e antigen-positive chronic hepatitis B. A meta-analysis. Ann Intern Med 1993;119:312± Vajro P, Migliaro F, Fontanella A, Orso G. Interferon: a meta-analysis of published studies in pediatric chronic hepatitis B. Acta Gastroenterol Belg 1998;61:219± Lok AS. Alpha-interferon therapy for chronic hepatitis B virus infection in children and Oriental patients. J Gastroenterol Hepatol 1991;6(Suppl 1):15± Eliakim M. Epidemiology of viral hepatitis in Israel, Isr J Med Sci 1979;15:248± Neudorf-Grauss R, Bujanover Y, Dinari G, et al. Chronic hepatitis B virus in children in Israel: clinical and epidemiological characteristics and response to interferon therapy. IMAJ 2000;2:164±8. 8. Yoshioka K, Kakumu, S, Wakita T, et al. Detection of hepatitis C virus by polymerase chain reaction and response to interferon therapy: relationship to genotypes of hepatitis C virus. Hepatology 1992;16:293±9. 9. Desmet VJ, Gerber M, Hoofnagle JH, Manns M, Scheuer PJ. Classification of chronic hepatitis: diagnosis, grading and staging. Hepatology 1994;19:1513± Jara P, Bortolotti F. Interferon-alpha treatment of chronic hepatitis B infection in childhood: a consensus advice based on experience in European children. J Pediatr Gastroenterol Nutr 1999;29:163± Lai CL, Lok AS, Lin HJ, Wu PC, Yeoh EK, Yeung CY. Placebo-controlled trial of recombinant alpha 2-interferon in Chinese HBsAg-carrier children. Lancet 1987;ii:877± La Banda F, Ruiz Moreno M, Carreno V, et al. Recombinant alpha 2- interferon treatment in children with chronic hepatitis B. Lancet 1988;i: Ruiz Moreno M, Jimenez J, Porres JC, Bartolome J, Moreno A, Carreno V. A controlled trial of recombinant interferon-alpha in Caucasian children with chronic hepatitis B. Digestion 1990;45:26± Sokal EM, Wirth S, Goyens P, Depreterre A, Cornu C. Interferon alfa-2b therapy in children with chronic hepatitis B. Gut 1993;34(2 Suppl):S87± Ruiz-Moreno M, Rua MJ, Molina J, et al. Prospective, randomized controlled trial of interferon-alpha in children with chronic hepatitis B. Hepatology 1991;13:1035± Vajro P, Tedesco M, Fontanella A, et al. Prolonged and high dose recombinant interferon alpha-2b alone or after prednisone priming accelerates termination of active viral replication in children with chronic hepatitis B infection. Pediatr Infect Dis J 1996;15:223± Gurakan F, Kocak N, Ozen H, Yuce A. Comparison of standard and high dosage recombinant interferon alpha 2b for treatment of children with chronic hepatitis B infection. Pediatr Infect Dis J 2000;19:52± Chang MH, Chen PJ, Chen JY, et al. Hepatitis B virus integration in hepatitis B virus-related hepatocellular carcinoma in childhood. Hepatology 1991;13:316± de Franchis R, Meucci G, Vecchi M, et al. The natural history of asymptomatic hepatitis B surface antigen carriers. Ann Intern Med 1993;118:191± Ruiz-Moreno M, Camps T, Jimenez J, et al. Factors predictive of response to interferon therapy in children with chronic hepatitis B. J Hepatol 1995;22:540± Sokal EM, Conjeevaram HS, Roberts EA, et al. Interferon alpha therapy for chronic hepatitis B in children: a multinational randomized controlled trial. Gastroenterology 1998;114:988± Bruguera M, Amat L, Garcia O, et al. Treatment of chronic hepatitis B in children with recombinant alpha interferon. Different response according to age at infection. J Clin Gastroenterol 1993;17:296± Lai CL, Lin HJ, Lau JN, et al. Effect of recombinant alpha 2 interferon with or without prednisone in Chinese HBsAg carrier children. QJMed 1991;78:155± Di Bisceglie AM, Fong TL, Fried MW, et al. A randomized, controlled trial of recombinant alpha-interferon therapy for chronic hepatitis B. Am J Gastroenterol 1993;88:1887± Hoofnagle JH, Peters M, Mullen KD, et al. Randomized, controlled trial of recombinant human alpha-interferon in patients with chronic hepatitis B. Gastroenterology 1988;95:1318±25. Correspondence: Dr. R. Shamir, Division of Pediatric Gastroenterology and Nutrition, Dept. of Pediatrics, Rambam Medical Center, P.O. Box 9602, Haifa 31096, Israel. Phone: (972-4) Fax: (972-4) shamirr@netvision.net.il