| |
Treatment with Peg-Interferon a-2b for HBeAg-Positive Chronic Hepatitis B: HBsAg Loss Is Associated with HBV Genotype
|
| |
| |
The American Journal of Gastroenterology
Volume 101 Page 297 - February 2006
Hajo J. Flink, M.Sc.1, Monika van Zonneveld, M.D., Ph.D.1, Bettina E. Hansen, M.Sc.2, Robert A. de Man, M.D., Ph.D.1, Solko W. Schalm, M.D., Ph.D.1, Harry L.A. Janssen, M.D., Ph.D.1, for the HBV 99-01 Study Group*
1Department of Gastroenterology and Hepatology, and 2Department of Epidemiology and Biostatistics, Erasmus MC, University Medical Center Rotterdam, The Netherlands
".....The ultimate endpoint of antiviral therapy for chronic HBV infection is loss of HBsAg, which is accompanied not only by disease remission in terms of ALT normalization, but also by a significantly decreased risk of liver failure and hepatocellular carcinoma
Zonneveld showed that over a median period of 8.8 yr, the HBsAg seroconversion rate increased up to 52% in chronic hepatitis B patients who responded to interferon therapy (26). Overall, these results support the hypothesis that Peg-interferon a, more than nucleoside analogs, can induce a complete and vigorous immune response which leads not only to loss of HBeAg, but also to loss of HBsAg....
..... Detailed analysis of our data showed that both HBeAg- and HBsAg seroconversion were strongly associated with the HBV genotype. Among responders harboring genotype A, there was an HBsAg seroclearance rate of 31%. Genotype A patients exhibited a high rate of HBeAg and HBsAg loss, not only as a result of Peg-interferon a therapy but also spontaneously...."
ABSTRACT
BACKGROUND AND AIMS: Hepatitis B surface antigen (HBsAg) loss is the hallmark of a complete response to antiviral therapy for chronic hepatitis B. In this study, we investigated the frequency of HBsAg loss after treatment with pegylated (Peg)-interferon a-2b.
METHODS: In a multicenter randomized controlled trial, 266 HBeAg-positive patients were treated for 52 wks with Peg-interferon a-2b (100 μg/wk) in combination with either lamivudine (100 mg/day) or placebo. Posttreatment follow-up was 26 wks.
RESULTS: At the end of follow-up, 95 (36%) of the 266 patients exhibited HBeAg loss, 18 (7%) HBsAg loss, and 16 (6%) HBsAg seroconversion. Addition of lamivudine did not enhance HBeAg loss, HBsAg loss, or development of anti-HBs. All 18 patients who showed HBsAg loss had normal ALT; 11 (61%) of these patients were also hepatitis B virus (HBV) DNA negative (<400 copies/mL) at the end of follow-up. Loss of HBsAg differed according to HBV genotype: 14% for genotype A, 9% for genotype B, 3% for genotype C, and 2% for genotype D (A vs D: p= 0.006).
CONCLUSIONS: One year of Peg-interferon a-2b for HBeAg-positive patients led to HBsAg loss in 7%. Our study indicates that treatment with Peg-interferon a-2b is the best therapy to achieve HBsAg clearance in patients with genotype A.
INTRODUCTION
Worldwide, over 360 million people are chronically infected with hepatitis B virus (HBV) (1). Chronic HBV infection, defined as hepatitis B surface antigen (HBsAg) positivity for more than 6 months, is associated with increased risk of developing cirrhosis, hepatic decompensation, and hepatocellular carcinoma (2, 3). Loss of HBsAg is closely related to improved long-term histological and virological outcome as well as decreased risk of hepatic decompensation and hepatocellular carcinoma (4-7). Therefore, the ultimate purpose of therapy for chronic hepatitis B is serum HBsAg negativity.
Treatment with nucleoside or nucleotide analogues is associated with reactivation of disease activity after termination of treatment or, in case of continued therapy, development of HBV variants which are resistant to treatment. Several studies with lamivudine have shown absence or less than 2% loss of HBsAg (8-11). One-year treatment with adefovir dipivoxil resulted in 1.6% loss of HBsAg, a percentage in the same range as spontaneous HBsAg seroclearance (12, 13). Treatment with standard interferon a may lead to increased loss of HBsAg (14, 15).
To date, little is known about loss of HBsAg due to treatment with pegylated (Peg)-interferon a. Therefore, the aim of the study was to investigate the incidence of and the factors predicting HBsAg loss due to treatment with Peg-interferon a-2b alone or in combination with lamivudine.
PATIENTS AND METHODS
This study was conducted as an ancillary study of a multicenter trial comparing combination therapy of Peg-interferon a-2b and lamivudine versus Peg-interferon a-2b alone (16). For this study, 266 patients were analyzed after central evaluation of eligibility. Males and females were included if they were 16 yr or older, had been positive for HBsAg and HBeAg for at least 6 months prior to randomization, and had elevated serum ALT levels of at least twice the upper limits of normal (ULN) 2 months before randomization. Exclusion criteria included antiviral or immune modulatory treatment within 6 months; antibodies against hepatitis C, hepatitis D, or human immunodeficiency virus; presence of decompensated liver disease (prothrombin time prolonged by ≥3 s, serum albumin <35 g/L, ascites, encephalopathy, history of variceal bleeding); pregnancy or recent drug or alcohol abuse. Other exclusion criteria were: inadequate hematological levels, i.e., leucopenia (≦3,000/mm3), thrombocytopenia (≦100,000/mm3), or granulocytopenia (≦1,800/mm3); serum a-fetoprotein level of more than 50 ng/mL; hypo- or hyperthyroidism; radiological evidence of hepatocellular carcinoma, or any contraindication specified for interferon a. All patients gave written informed consent and the ethics committee at the participating centers approved the protocol.
Study Design
The patients received 100 μg/wk Peg-interferon a-2b for the first 32 wks, followed by 50 μg/wk Peg-interferon a-2b from week 32 to week 52. This was combined with either 100 mg lamivudine daily or a placebo. Post-treatment follow-up lasted 26 wks.
During therapy and follow-up, patients were seen monthly for routine examination as well as biochemical and HBV DNA assessment. Transaminases were determined locally and are therefore expressed as times the ULN. HBV DNA was assessed using in-house Taqman PCR (detection limit 400 copies/mL) based on the Eurohep standard (17). HBeAg (AxSYM, Abbott, Abbot Park, IL, USA) and HBsAg (AxSYM, Abbott) status were measured at weeks 0, 32, 52 (during treatment) and week 78 (end of follow-up). HBV genotype (Inno-Lipa Assay, Innogenetics) was assessed at baseline. HBeAg and HBsAg responses were defined as loss of serum HBeAg and HBsAg at the end of follow-up, respectively. HBs- and HBe seroconversion were defined as loss of HBs or HBe antigen, respectively, with development of the corresponding antibodies. Liver histology was assessed at baseline and an optional biopsy was performed at the end of treatment. Paired biopsies were available for 110 patients. Histological scoring was performed by one experienced pathologist according to the histological activity index, modified by Ishak et al. (18). The pathologist was blind for chronological order of biopsies and treatment schedule. Improvement of histology was defined as a reduction of at least two points in the necroinflammatory score (range 0-18) and one point in the fibrosis score (range 0-6).
Statistical Analysis
All data were analyzed by means of SPSS version 10.1 (SPSS Inc., Chicago, IL). Chi-square or Fisher's exact test and the Mann-Whitney U test were used where appropriate (all 2-tailed). A p value <0.05 was considered significant.
RESULTS
Loss of HBeAg and HBsAg
At the end of follow-up, 95 (36%) of the 266 patients exhibited HBeAg loss, 18 (7%) loss of HBsAg, and 16 (6%) HBsAg seroconversion. Among patients with HBeAg loss, 19% were also HBsAg negative at the end of follow-up. Addition of lamivudine did not enhance HBeAg loss, HBsAg loss, or development of anti-HBs. At the end of follow-up, HBeAg clearance was found for 36% in the monotherapy and 35% of the combination group. HBsAg clearance at the end of follow-up was 7% for both treatment groups. Results of HBeAg and HBsAg testing were available for weeks 0, 32, 52, and 78. Percentages of HBeAg- and HBsAg clearance and HBV DNA <400 copies/mL at these time points are shown in Figure 1. Twelve patients became HBsAg negative before week 52 (end of treatment) and 6 became HBsAg negative between week 52 and week 78 (end of follow-up). Baseline characteristics for nonresponders and for those who showed loss of serum HBeAg and/or HBsAg are given in Table 1. Compared to nonresponders, baseline HBV DNA values were lower among those exhibiting HBeAg loss (p= 0.043), but not those with HBsAg loss. Compared to nonresponders and HBeAg responders, patients with HBsAg loss were significantly older (p= 0.001).
Loss of HBeAg and HBsAg in Relation to HBV DNA, ALT, and Histology
Among HBeAg responders, 59 (65%) patients had normal serum ALT, 62 (69%) HBV DNA <200,000 copies/mL, and 9 (22%) HBV DNA <400 copies/mL by PCR at the end of follow-up (Fig. 2). No differences were found between the different treatment regimens and HBV DNA response or ALT normalization. Among HBsAg responders, all patients had normal serum ALT and HBV DNA <200,000 copies/mL, and 11 (61%) had HBV DNA <400 copies/mL by PCR at the end of follow-up. In the nonresponder group, we found 28 (18%) patients with normal serum ALT, 12 (9%) with HBV DNA <200,000 copies/mL, and none with HBV DNA <400 copies/mL by PCR at the end of follow-up.
As far as histology is concerned, improved (reduction of >2 points HAI) necroinflammation scores were found for 36 (38%) of the HBeAg responders, 7 (39%) of the HBsAg responders, and 39 (23%) of the nonresponders (HBeAg and HBsAg responders versus nonresponders, p= 0.016). Among HBsAg-negative patients, the fibrosis score had improved in 2 cases and worsened in 3 cases, while 3 patients showed no change (median 0, range -2-1). Fibrosis scores were not significantly different among nonresponders, HBeAg responders, and HBsAg responders.
Loss of HBeAg and HBsAg in Relation to Genotype
The most common HBV genotypes were genotype A (n = 90), B (n = 23), C (n = 39), and D (n = 103) (Table 2). HBV genotypes A (97%) and D (95%) were found predominantly among Caucasians; among Asians, genotypes B (78%) and C (80%) were the most common (Table 2). Patients with genotype A lived in Northwest and Eastern Europe, patients with genotype B or C in all continents, and those with genotype D primarily in the Mediterranean countries and Northwest Europe.
HBV genotype was not only associated with HBeAg response, but also with HBsAg response (Fig. 3). Loss of HBsAg was found predominantly in patients harboring genotype A. Thirteen of the 90 genotype A patients (14%), 2 of the 23 genotype B (9%), 1 of the 39 genotype C (3%), and 2 of the 103 genotype D (2%) patients had lost HBsAg at the end of follow-up (genotype A versus genotype D, p= 0.006). Anti-HBs was seen in 12 patients with genotype A (13%), 2 with genotype B (9%), 0 with genotype C, and 2 with genotype D (2%). Among HBeAg responders, 31% of genotype A, 20% of genotype B, 9% of genotype C, and 8% of genotype D patients lost serum HBsAg as well; anti-HBs was seen in 29%, 20%, 0%, and 8% at the end of follow-up for genotypes A, B, C, and D, respectively.
DISCUSSION
The ultimate endpoint of antiviral therapy for chronic HBV infection is loss of HBsAg, which is accompanied not only by disease remission in terms of ALT normalization, but also by a significantly decreased risk of liver failure and hepatocellular carcinoma (4-6). Spontaneous HBsAg loss is uncommon and varies from 1% to 2% annually (19). In the current study, after 52 wks of Peg-interferon a-2b therapy in combination with lamivudine or placebo and 26 wks of follow-up, we found serum HBsAg loss in 18 patients (7%) and development of anti-HBs in 16 patients (6%). Adding lamivudine did not enhance HBeAg- or HBsAg seroclearance. Previous studies have shown that treatment with nucleoside or nucleotide analogues, such as lamivudine and adefovir, probably do not lead to an enhanced rate of HBsAg seroclearance (8-11, 20, 21). Nevertheless, in several single cases, prolonged adefovir therapy induced a sharp reduction of cccDNA and HBsAg seroconversion (22). Therefore, the hypothesis that a decrease in intrahepatic viral load itself may be associated with restoration of a noncytolytic Th1 response and HBsAg seroconversion deserves further investigation (23). A recent pilot study using tenofovir did show HBsAg seroconversion in 5 of 35 (14%) patients (24). This needs to be confirmed in a large randomized study which was initiated recently.
We previously demonstrated that HBeAg seroconversion after treatment with standard interferon a persisted longer than after lamivudine therapy (25). Furthermore, HBsAg seroconversion occurred more frequently in patients treated with interferon a than those receiving lamivudine. In keeping with the sustained response after interferon a therapy, a prolonged follow-up study by van Zonneveld showed that over a median period of 8.8 yr, the HBsAg seroconversion rate increased up to 52% in chronic hepatitis B patients who responded to interferon therapy (26). Overall, these results support the hypothesis that Peg-interferon a, more than nucleoside analogs, can induce a complete and vigorous immune response which leads not only to loss of HBeAg, but also to loss of HBsAg.
Detailed analysis of our data showed that both HBeAg- and HBsAg seroconversion were strongly associated with the HBV genotype. Among responders harboring genotype A, there was an HBsAg seroclearance rate of 31%. Genotype A patients exhibited a high rate of HBeAg and HBsAg loss, not only as a result of Peg-interferon a therapy but also spontaneously (27). This probably relates to the presence of genotype-dependent mutations located within the basic core promotor region of the genome (28). Among Asian patients, we found a higher HBeAg- and HBsAg seroconversion rate for those with genotype B than for those with genotype C. This confirms the results of previous studies that showed better results of interferon a for patients with genotype B, also due to a response predisposition caused by core promotor mutations (29-32). Although our results revealed a strong association between HBV genotype and HBsAg loss, the absolute number with HBsAg clearance is relatively low. Therefore, multivariate analysis for the prediction of HBsAg loss could not be performed. Nevertheless, the clear association between HBV genotype and loss of HBsAg should be taken into account in future studies.
Two global chronic hepatitis B studies on Peg-interferon a-2a, in a similar therapeutic regimen as our study, have recently been finalized. In a difficult-to-treat HBeAg-negative population, Marcellin et al. found 4% HBsAg seroconversion (33). This low HBsAg response can probably be explained by the fact that HBeAg-negative patients often harbor the non-A genotype and carry pre-core mutations which may hamper effective immune stimulation by interferon a. In an HBeAg-positive population treated for 1 yr with Peg-interferon a-2a, Lau and colleagues found 32% HBeAg seroconversion, 3% HBsAg loss, and 3% HBsAg seroconversion (21). When this rate of HBsAg loss is compared with our results, it is important to note that 87% of their population were Asian, for whom low HBsAg seroclearance rates after interferon a have been described previously (21). The HBsAg loss among our Asian patients was 5%.
An interesting finding of our study is the relatively large number of patients (39%) for whom serum HBV DNA could still be detected by Taqman technology after HBsAg seroclearance. Yuen et al. recently described a Chinese chronic HBV population with HBsAg seroclearance, spontaneously, or after interferon a therapy. Upon long-term follow-up, they found serum HBV DNA in only 2%, but intrahepatic HBV DNA (mainly cccDNA) in 37% of the population (4). This study and earlier studies showing a delay between post-interferon a HBsAg seroclearance and HBV DNA negativity by PCR (4, 34) suggest that with continued follow-up, HBV DNA might further disappear from serum in most patients who become HBsAg negative. This is further supported by analysis of our data from a previous study, which showed that 90% of the HBsAg-negative patients also had undetectable HBV DNA levels by PCR after prolonged follow-up (26).
In conclusion, 7% of patients treated with Peg-interferon a-2b showed HBsAg loss, a response higher than that reported for lamivudine and/or adefovir. HBV genotypes were associated with HbeAg and HBsAg seroclearance. In particular, patients with genotype A seemed to benefit from Peg-interferon a-2b therapy, with an HBsAg clearance rate of 14% overall and 31% among responders who had lost HBeAg.
|
|
| |
| |
|
|
|
