It is estimated that at least 350 million people are

Transcription

1 Sustained Disease Remission after Spontaneous HBeAg Seroconversion Is Associated with Reduction in Fibrosis Progression in Chronic Hepatitis B Chinese Patients Chee-Kin Hui, 1,2 Nancy Leung, 4 Tony W.H. Shek, 5 Hung Yao, 6 Wai-Ki Lee, 7 Jak-Yiu Lai, 8 Sik-To Lai, 8 Wai-Man Wong, 9 Lawrence SW. Lai, 10 Ronnie T.P. Poon, 3 Chung-Mau Lo, 3 Sheung-Tat Fan, 3 and George K.K. Lau 1,2 ; for the Hong Kong Liver Fibrosis Study Group Recently, controversies have arisen about whether hepatitis B e antigen (HBeAg) seroconversion can result in regression of fibrosis, thus improving the clinical outcome of Chinese patients with chronic hepatitis B. In this study, we determined if spontaneous HBeAg seroconversion is associated with regression of fibrosis in Chinese chronic hepatitis B patients. We evaluated the histology of liver samples from 128 HBeAg-positive treatment-naive Chinese patients who had undergone 2 liver biopsies over the years. Regression of fibrosis was defined as a decrease in fibrosis stage of at least 1 point. Sustained disease remission was defined as HBeAg seroconversion and hepatitis B virus (HBV) DNA < 10 4 copies/ml at follow-up liver biopsy. The mean duration ( standard error of the mean) between the initial and follow-up liver biopsies was months. Regression of fibrosis was higher in patients with sustained disease remission (5 of 13 [38.5%] versus 22 of 115 [19.1%], P < ), patients who were younger (20-29 years old) at initial liver biopsy (17of 54 [31.5%] versus 10 of 74 [13.5%], P ), and patients with genotype B (17of 43 [39.5%] versus 10 of 85 [11.8%], P 0.004). On multivariate analysis, sustained disease remission (relative risk [RR] 3.00, 95% confidence interval [95% CI] , P 0.01) and being years old at initial liver biopsy (RR 2.94, 95% CI , P 0.04) were independently associated with regression of fibrosis. The rate of fibrosis progression was lower in patients with sustained disease remission than in those who remained HBeAg positive (median 0 fibrosis units/year, range 2.00 to 0.70 fibrosis units/year, versus median 0.51 fibrosis units/year, range 0 to 2.03 fibrosis units/year, P 0.02). Conclusion: Spontaneous sustained remission of disease is associated not only with little progression of fibrosis but also with regression of fibrosis. (HEPATOLOGY 2007;46: ) Abbreviations: ALT, alanine aminotransferase; anti-hbe, hepatitis B e antibody; HAI, histologic activity index; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma. From the 1 Department of Medicine, 2 Research Centre of Infection and Immunity, and 3 Department of Surgery,University of Hong Kong, Queen Mary Hospital, Hong Kong Special Administrative Region (SAR), China; 4 Department of Medicine, Alice Ho Miu Ling Nethersole Hospital, Hong Kong SAR, China; 5 Department of Pathology, Queen Mary Hospital, Hong Kong SAR, China; 6 Department of Pathology, Princess Margaret Hospital, Hong Kong SAR, China; 7 Department of Pathology, Tuen Mun Hospital, Hong Kong SAR, China; 8 Department of Medicine, Princess Margaret Hospital, Hong Kong SAR, China; 9 St. Paul s Hospital, Hong Kong SAR, China; and 10 Department of Medicine, Caritas Medical Centre, Hong Kong SAR, China. Received December 20, 2006; accepted April 3, Supported by the Hong Kong Liver Foundation (grant to G.K.K.L.). Address reprint requests to: Dr. Chee-Kin Hui, Department of Medicine, University of Hong Kong, Queen Mary Hospital, 102 Pokfulam Road, Hong Kong SAR, China. ckhui23@gmail.com; fax: (852) Copyright 2007 by the American Association for the Study of Liver Diseases. Published online in Wiley InterScience ( DOI /hep Potential conflict of interest: Nothing to report. It is estimated that at least 350 million people are chronically infected with the hepatitis B virus (HBV) worldwide. 1 Even in the United States, which has a lower prevalence of HBV, an estimated 1.25 million individuals are infected with the virus. 2 Despite the availability of effective hepatitis B vaccines, new infections with HBV still occur. In the United States alone, approximately 100,000 people are acutely infected each year, some of whom progress to chronic hepatitis B. 3 Individuals with chronic hepatitis B are at increased risk of developing cirrhosis, hepatic decompensation, and hepatocellular carcinoma (HCC). An estimated 5000 people in the United States die annually from complications of HBV, such as cirrhosis and HCC. 4 One of the key events in the evolution of chronic hepatitis B is hepatitis B e antigen (HBeAg) seroconversion, which is a prerequisite for clearance of hepatitis B surface antigen (HBsAg). This may occur either spontaneously, at a rate of 5%-10% per year, or following therapy with 690

2 HEPATOLOGY, Vol. 46, No. 3, 2007 HUI ET AL. 691 interferon alpha or nucleoside/nucleotide analogs. 5,6 Accumulating evidence in the past 25 years suggests that the risk of developing cirrhosis and HCC is higher for HBV carriers who undergo HBeAg seroconversion later in life, have persistently high replication of HBV, and have a longer duration of active hepatitis. The annual incidence of cirrhosis has been estimated to be 2%-6% for HBeAg-positive and 8%-10% for HBeAgnegative patients. 7,8 The higher rate of cirrhosis among HBeAg-negative patients is related to older age and more advanced liver disease at presentation. 8 Among HBeAgpositive patients, the rate of cirrhosis development is higher in those who remained HBeAg-positive during follow-up. 7 Several studies have also shown that patients who had HBeAg reversion had an increased risk of cirrhosis when compared with those with sustained HBeAg seroconversion. 7,9,10 In contrast, a recent study showed that disease activity continues to progress in a proportion of patients after HBeAg seroconversion. This study from Taiwan found that although HBeAg seroconversion conferred a favorable outcome to some patients, active hepatitis did occur after HBeAg seroconversion, leading to cirrhosis and HCC. 10 In a study by McMahon et al., neither seroconversion from HBeAg to hepatitis B e antibody (anti-hbe) nor even the loss of HBsAg was found to protect patients from the development of HCC. 9 Long-term follow-up studies have found that interferon alpha therapy has minimal benefit in reducing the risk of cirrhosis, HCC, and liver-related mortality, except for the few patients with sustained HBeAg seroconversion. 11,12 However, these studies mainly addressed the impact of HBeAg seroconversion on clinical outcome rather than liver histological response. 8,13-15 It remains uncertain whether HBeAg seroconversion together with reduced viral replication results in regression of fibrosis in patients with chronic hepatitis B. To better define the natural history and prognostic factors of chronic hepatitis B, we evaluated liver histology from all patients who underwent 2 or more liver biopsies over a 10-year period. Hepatic fibrosis was used as a primary marker for disease regression or progression. Patients and Methods A database of all HBsAg- and HBeAg-positive adult patients who had undergone 2 or more liver biopsies at Queen Mary Hospital, Queen Elizabeth Hospital, Princess Margaret Hospital, Alice Ho Miu Ling Nethersole Hospital, and Tuen Mun Hospital in Hong Kong, China, between 1994 and 2005 was obtained, and clinical charts of these patients were reviewed. Consecutive patients were included into this study if they fulfilled the following criteria: (1) HBsAg-positive for at least 6 months, (2) HBeAg-positive for at least 6 months, (3) treatment-naive before the initial liver biopsy, (4) had not received antiviral treatment with either nucleoside/nucleotide analogs or immunomodulators during the period between the initial and follow-up liver biopsies, (5) had serum alanine aminotransferase (ALT) above the upper limit of normal (7-53 U/l for men and 7-31 U/l for women), (6) had an alcohol intensity of less than 10 g/day as defined, 16,17 and (7) liver biopsy specimen showed at least 9 portal tracts to ensure accurate interpretation of histologic analysis. Patients were excluded from the study if they had: (1) previous therapy with nucleoside/nucleotide analog or immunomodulators; (2) coinfection with hepatitis C virus (HCV), human immunodeficiency virus, or hepatitis D virus; or (3) HCC at initial liver biopsy. In the database, a total of 406 treatment-naive chronic HBV patients were identified to have had 1 liver biopsy. Two hundred and seventy-eight (68.5%) were excluded (241 were treated with anti-hbv therapy after the liver biopsy, whereas 37 patients only had 1 liver biopsy). One hundred and twenty-eight patients (31.5%) fulfilled the inclusion criteria for the study. A follow-up liver biopsy was performed in 51 of the 128 patients (39.8%) before subsequent recruitment into the following clinical trials: an unpublished conventional interferon alpha study conducted in 1994 (n 23), a phase 3 pegylated interferon alpha-2a study (n 11), 18,19 a phase 3 lamivudine study (n 7), 20 a phase 2 combination therapy (n 4), 21 a phase 3 telbivudine study (n 3), 22 and a phase 2 clevudine study (n 3). 23 The remaining 77 of these 128 patients (60.2%) underwent follow-up liver biopsy for reassessment of chronic HBV disease, as these patients were assessed before commercial testing for serum HBV DNA was available. Assessment of Fibrosis. All liver biopsy specimens were traced and assessed by 1 hepatic pathologist (T.S.), blinded to the chronological sequence of the liver biopsies and the biochemistry and virological studies, using the modified histologic activity index (HAI) score for grading inflammation and necrosis and the Ishak fibrosis score for staging fibrosis. 24,25 The necroinflammatory components of the modified HAI score include periportal inflammation or periseptal interface hepatitis (0-4), confluent necrosis (0-6), focal lytic necrosis, apoptosis and focal inflammation (0-4), and portal inflammation (0-4). The Ishak score was used for fibrosis staging, which stages fibrosis from 0 to 6. Regression of fibrosis was defined as at least a 1-point decrease in Ishak fibrosis stage and a worsening of fibrosis as at least a 1-point increase in Ishak fibrosis stage. The rate of fibrosis progression was calculated as the difference

3 692 HUI ET AL. HEPATOLOGY, September 2007 in fibrosis score between the first and follow-up (or second) liver biopsies divided by the duration of the interval between the biopsies and expressed as fibrosis units/ year. 16,17,26 Virological Studies. Patients were followed up every 3-6 months in the outpatient clinic after the first liver biopsy. Serum was tested for HBsAg, HBeAg, and anti- HBe with the enzyme-linked immunosorbent assay (ELISA; Abbott Laboratories, Chicago, IL) on every follow-up visit. Anti-HCV, anti-hepatitis D, and anti-hiv were retrospectively tested by commercially available ELISA (Abbott GmbH Diagnostika, Wiesbaden-Delkenheim, Germany). HBV DNA at the first and follow-up liver biopsies was retrospectively quantified by real-time polymerase chain reaction assay as previously described with a linear range of 10 2 to 10 8 copies/ml on stored serum samples HBV genotype was performed as described. 30 HBeAg seroconversion was defined as loss of HBeAg accompanied by the development of anti-hbe for 2 consecutive readings 3 months apart. Sustained disease remission or inactive HBsAg carrier was defined as HBeAg negative with positive anti-hbe, normal serum ALT levels, and HBV DNA less than 10 4 copies/ml at the follow-up liver biopsy as previously defined. 31 Written informed consent was obtained from all patients, and the research protocol for this study was approved by the local institutional review board. Statistical Analysis. All statistical analysis was performed with the Statistical Package for Social Sciences Program version 12 (SPSS, Chicago, IL). The primary end point of this study was regression of fibrosis at follow-up liver biopsy. The Kaplan-Meier method for univariate analysis was used to identify predictors for regression of fibrosis at follow-up liver biopsy, and the Mantel-Haenszel log-rank test was applied for comparisons. Variables with a P value 0.10 in the univariate analysis were included in the final Cox proportional analysis model with a forward stepwise procedure to determine the most significant factors associated with regression of fibrosis. The validity of each Cox proportional hazards model was checked by examining the corresponding martingale and standardized score residuals. The chi-square and Student t tests were performed as appropriate. Continuous variables are expressed as means standard errors of the mean (SEM) except for the modified HAI score, fibrosis stage, and rate of fibrosis progression, which are presented as medians with ranges. All P values were 2-tailed, and P values less than 0.05 were considered significant throughout this study. Fig. 1. Graph showing correlation between age at initial liver biopsy and fibrosis stage on initial liver biopsy. Results Liver Histology of Initial Liver Biopsy. The mean SEM of portal tracts in the initial biopsy was The median modified HAI score of the initial liver biopsy was 9 (range 2-18), and median fibrosis stage was 2 (range 0-5). Of the 128 patients, 7 patients (5.5%) had no fibrosis, 72 patients (56.3%) had portal fibrotic expansion (fibrosis stage 1 or 2), 32 patients (25.0%) had bridging fibrosis (stage 3 or 4), and 17 patients (13.3%) had cirrhosis (stage 5 or 6). Correlation Between Age and Initial Histology, There was a positive correlation between age at initial liver biopsy and fibrosis stage on initial liver biopsy (r 0.69, P ; Fig. 1). Those years old at the initial liver biopsy had a significantly lower fibrosis stage compared with those years old at the initial liver biopsy (median 1, range 0-5, versus median 2, range 0-5, respectively, P 0.03). Similarly, there was also a positive correlation between age and the modified HAI score at initial liver biopsy (r 0.20, P 0.03). There was a trend that patients years old at the initial liver biopsy had a lower modified HAI score on initial liver biopsy (median 8, range 2-15) compared with those years old at the initial liver biopsy (median 10, range 2-18, P 0.07). Clinical Outcome. HBeAg seroconversion occurred in 28 of the 128 patients (21.9%) at the follow-up liver biopsy. Thirteen of the 28 patients (46.4%) with HBeAg seroconversion sustained disease remission at the follow-up liver biopsy. Mean serum ALT level ( SEM) at the follow-up liver biopsy in those with sustained disease remission was lower than in those who developed HBeAg seroconversion but had HBV DNA 10 4 copies/ml (23 9 versus U/l, P 0.02).

4 HEPATOLOGY, Vol. 46, No. 3, 2007 HUI ET AL. 693 Table 1. Clinical and Histologic Characteristics of Patients with and without Regression of Fibrosis on Follow-Up Liver Biopsy Characteristics Regression of fibrosis (n 27) No regression of fibrosis (n 101) P value Mean age at initial biopsy (years) Age years years * ALT at initial liver biopsy (U/L) Normal serum ALT at follow-up liver biopsy, U/L Yes 9 16 No HBV DNA at initial liver biopsy (log 10 copies/ml) * Sex (M:F) 22:5 81: * HBeAg seroconversion at follow-up liver biopsy Yes 5 23 No * Sustained disease remission Yes 5 8 No * Initial modified HAI score * Initial fibrosis stage: F0-F F3-F * Modified HAI score on repeat biopsy Genotype B C * *By log rank. ALT, alanine aminotransferase; HAI, histologic activity index; HBeAg, hepatitis B e antigen. All patients with loss of HBeAg developed anti-hbe at the follow-up liver biopsy. None of the 28 patients with HBeAg seroconversion developed HBeAg seroreversion at the follow-up liver biopsy. No patient lost HBsAg. Sixteen of the 43 patients (37.2%) with genotype B compared with 12 of the 85 patients (14.1%) with genotype C had HBeAg seroconversion at the follow-up liver biopsy (P 0.003). Liver Histology at Follow-Up Liver Biopsy. Mean duration ( SEM) between initial and follow-up liver biopsies was months. Mean number of portal tracts ( SEM) in the follow-up biopsy was At follow-up liver biopsy, 7 patients (5.5%) had no fibrosis, 73 patients (57.0%) had portal fibrotic expansion, 16 patients (12.5%) had bridging fibrosis, and 32 patients (25.0%) had cirrhosis. Among the 128 patients, 27 patients (21.1%) had regression of fibrosis. There was no progression of fibrosis (median 2, range 0-6) at follow-up liver biopsy compared to the initial liver biopsy (median 2, range 0-5, P 0.15). Patients with Regression of Fibrosis at Follow-Up Liver Biopsy. Table 1 compares patients with and without regression of fibrosis at follow-up liver biopsy. Patients with regression of fibrosis at follow-up liver biopsy had a lower modified HAI score at follow-up liver biopsy when compared with those without regression of fibrosis (P 0.01). The factors that correlated with regression of fibrosis at follow-up liver biopsy were sustained disease remission (P on log-rank; Fig. 2A), age at initial liver biopsy (P on log-rank; Fig. 2B), genotype B(P on log-rank; Fig. 2C), and normal serum ALT at the follow-up liver biopsy (P 0.04 on log-rank) (Table 1). Table 2 shows the results of multiple regression analysis with the use of a multivariate Cox proportional hazards model. Sustained disease remission (adjusted relative risk [RR] 3.00, 95% confidence interval [CI] , P 0.01) and age at initial liver biopsy (adjusted RR 2.94, 95% CI , P 0.04) were independent factors associated with regression of fibrosis at follow-up liver biopsy. There was a trend that genotype B (P 0.06) was independently associated with regression of fibrosis at follow-up liver biopsy.

5 694 HUI ET AL. HEPATOLOGY, September 2007 those who had HBeAg seroconversion but serum HBV DNA 10 4 copies/ml (median 9, range 5-16, P 0.04) and those who remained HBeAg positive (median 8, range 1-18, P 0.03) at the follow-up liver biopsy. Sustained Disease Remission Was Associated with Slower Rate of Fibrosis Progression. Seventeen of the 128 patients (13.3%) with cirrhosis on initial liver biopsy were excluded from this calculation, as patients with cirrhosis would not be expected to have fibrosis progression. The median rate of fibrosis progression in the remaining 111 patients was 0 fibrosis units/year (range 2.00 to 2.03 fibrosis units/year). Patients with sustained disease remission (n 12) had the lowest rate of fibrosis progression (median 0 fibrosis units/year, range 2.00 to 0.70 fibrosis units/year) compared with those with HBeAg seroconversion but serum HBV DNA 10 4 copies/ml (n 12; median 0.21 fibrosis units/year, range 1.01 to 1.00 fibrosis units/ year, P 0.03) and those who remained HbeAg-positive (n 87; median 0.51 fibrosis units/year, range 0 to 2.03 fibrosis units/year, P 0.02). Patients years old at the initial liver biopsy had a lower rate of fibrosis progression compared with those years old at initial liver biopsy (median 0 fibrosis units/year, range 4.00 to 1.00, versus median 0.21 fibrosis units/year, range 2.00 to 2.03 fibrosis units/ year, respectively, P 0.04). The rate of fibrosis progression was also slower in patients with genotype B (median 0 fibrosis units/year, range 4.00 to 2.03 fibrosis units/year) compared with those with genotype C (median 0.02 fibrosis units/year, range 2.32 to 1.33 fibrosis units/year, P 0.05). Age Was Not Associated with Sustained Disease Remission. There was no correlation between age of pa- Table 2. Adjusted Relative Risk of Regression of Fibrosis on Follow-Up Liver Biopsy according to Various Factors Adjusted relative risk 95% Confidence interval P value Fig. 2. Graph showing cumulative regression of fibrosis according to (A) sustained disease remission, (B) age, and (C) genotype. Sustained Disease Remission Was Associated with Lower Modified HAI Score at Follow-Up Liver Biopsy. Patients with sustained disease remission had a lower modified HAI score (median 7, range 5-11) than Sustained disease remission Yes No 1 Age years years 1 Genotype B C 1 Normal serum ALT at follow-up liver biopsy Yes No

6 HEPATOLOGY, Vol. 46, No. 3, 2007 HUI ET AL. 695 Fig. 3. Graph showing correlation between (A) regression of fibrosis and change in serum HBV DNA level between the initial and the follow-up liver biopsies, and (B) degree of regression of fibrosis and change in serum HBV DNA level between the initial and the follow-up liver biopsies. tients at initial liver biopsy and sustained disease remission (r 0.16, P 0.15). Sustained disease remission was experienced by 7 of the 54 patients (13.0%) who were years old at the initial liver biopsy compared with 6 of the 74 patients (8.1%) who were years old at the initial liver biopsy (P 0.37). Correlation Between Change in HBV DNA at Initial and Follow-Up Liver Biopsies and Regression of Fibrosis. There was a positive correlation between regression of fibrosis and change in serum HBV DNA between the initial and follow-up liver biopsies (r 0.55, P 0.001; Fig. 3A). The degree of regression of fibrosis also correlated positively with change in serum HBV DNA level between the initial and follow-up liver biopsies (r 0.18, P 0.04; Fig. 3B). Discussion The natural history of chronic hepatitis B acquired perinatally has been categorized as having 3 phases. 32 The first phase is characterized by active replication of HBV, positivity for HBeAg, and normal to low levels of aminotransferase. The second phase, characterized by immune clearance, usually occurs from the age of 15 to 35 years, during which hepatic flares may occur as the result of specific T-lymphocyte-mediated cellular responses to viral antigens and apoptosis of hepatocytes. This phase subsides as viral replication declines. The third and residual phase (inactive HBsAg carrier state) is characterized by the absence of HBeAg and the presence of anti-hbe, persistently normal aminotransferase levels, and low or undetectable serum HBV DNA. Liver biopsy performed during this phase usually shows mild hepatitis and minimal fibrosis. On the other hand, significant fibrosis or inactive cirrhosis may be observed, and it was postulated that these patients accrued severe liver injury during the preceding immune clearance phase. Although spontaneous HBeAg seroconversion is associated with a reduced risk of cirrhosis and HCC, the effect of HBeAg seroconversion on regression of fibrosis or the rate of fibrosis progression is uncertain. In this study, the rate of fibrosis progression was estimated from the fibrosis stage of 2 consecutive liver biopsies of treatment-naive patients performed a mean of 43.9 months apart. Our results show that HBeAg seroconversion alone does not necessarily result in a regression of fibrosis. Using a more stringent definition of sustained disease remission, that is, HBeAg seroconversion together with serum HBV DNA below 10 4 copies/ml, multivariate analysis showed it to be independently associated with regression of fibrosis when compared with those who had HBeAg seroconversion but HBV DNA 10 4 copies/ml or patients who remained HbeAg-positive. This is consistent with recent findings that patients with serum HBV DNA 10 4 copies/ml are at higher risk of developing HCC or cirrhosis. 33,34 This study helps to explain why progression to cirrhosis is higher in those with active hepatitis after spontaneous HBeAg seroconversion. 10 Patients with spontaneous HBeAg seroconversion but HBV viremia 10 4 copies/ml at follow-up liver biopsy had a higher rate of fibrosis progression and thus an increased risk of progression to cirrhosis and HCC. 10 The higher number of patients with sustained disease remission in the Taiwan study (66.8%) is probably a result of the higher threshold of HBV DNA ( copies/ml) used to define sustained disease remission after HBeAg seroconversion. 10 In this study, we adopted the definition from the recently revised

7 696 HUI ET AL. HEPATOLOGY, September 2007 National Institutes of Health Consensus guideline on chronic hepatitis B, where sustained disease remission or inactive HBsAg carrier was defined as HBeAg seroconversion, normal serum ALT level, plus HBV DNA less than 10 4 copies/ml. 31 Aside from improvement in fibrosis stage, patients with sustained disease remission also had a lower modified HAI score. This gives further credence to our hypothesis that achieving HBeAg seroconversion alone is inadequate for disease remission. Active disease or active inflammation in the liver is still possible when HBV DNA is more than 10 4 copies/ml after HBeAg seroconversion. Therefore, in addition to HBeAg seroconversion, sustained disease remission is required in order to achieve regression of fibrosis and hepatic inflammation. Similarly, in chronic HCV patients, hepatic inflammation is associated with fibrosis progression even in those with persistently normal ALT levels. For example, chronic HCV patients with persistently normal ALT levels who have minimal or no fibrosis and necroinflammation are less likely to develop fibrosis progression on longitudinal liver biopsy than those with moderate fibrosis or necroinflammation. 35,36 Sustained disease remission, especially if achieved early, may represent a truly inactive state and confer a favorable prognosis. This is supported by a long-term epidemiological study that found no difference in the survival between HBsAg-positive healthy blood donors and HBsAg-negative controls over a 30-year period. 14 HBeAg seroconversion with sustained suppression of HBV DNA to less than 10 4 copies/ml can result in regression of fibrosis or reduction in the rate of fibrosis progression per year, thus conferring a better outcome in this subset of patients, as reported. 14,37 Furthermore, a greater decrease in serum HBV DNA level between initial and follow-up liver biopsies also resulted in a higher degree of regression of fibrosis. Age was also found to be an important factor associated with regression of fibrosis. We postulate that if sustained disease remission was achieved at a younger age, fibrosis might not occur, or regression of fibrosis might follow. The study also demonstrates that in chronic hepatitis B, fibrosis progression tends to be slower during the initial 2 decades, and the rate of fibrosis progression increases after the second decade. This finding also explains previous observations that those with chronic hepatitis B with spontaneous loss of HBsAg can have a reduced risk of cirrhosis or HCC, provided that the loss of HBsAg occurred at a younger age, in the absence of HCV coinfection, and preceded the development of cirrhosis. 38,39 Patients who lose HBsAg at an older age are more likely to already have significant fibrosis or even cirrhosis prior to HBsAg loss. Thus, regression of fibrosis and decrease in the rate of fibrosis progression is less likely. Our study also confirms findings from Taiwan that genotype B is associated with a higher rate of spontaneous HBeAg seroconversion and a lower rate of fibrosis progression. 40,41 A few important clinical recommendations can be derived from this study. First, HBeAg-positive patients with chronic hepatitis B with raised serum ALT levels should be treated early, because younger patients have milder fibrosis stage and modified HAI score and can develop regression of fibrosis with sustained disease remission. However, we noted that younger patients did not seem to have achieved sustained disease remission at a higher rate. Second, antiviral therapy is recommended for those who are HbeAg-negative but with serum HBV DNA 10 4 copies/ml, because this group of patients has higher fibrosis progression rate than those with sustained disease remission. Because of the retrospective nature of this study, it has a few limitations. A bias may have resulted from the follow-up liver biopsies not being performed per protocol or at a discrete time point after the initial liver biopsy. Progression of fibrosis is not likely to be linear, because disease activity waxes and wanes, especially during flares of disease. Likewise, the calculated rate of fibrosis progression, consisting of score fractions, assumes that fibrosis progression is linear. 16,17,26 This invented fraction has limited accuracy because of the absence of histology during the in-between points and the points being defined at the start of the study. Thus, the invented fractions also lack meaning. 42 Although still considered a gold standard, histological scoring systems can be unreliable when measuring fibrosis progression because of potential sampling error New imaging systems, such as the Fibroscan, a noninvasive and reproducible method for measuring liver stiffness, may eventually be found to be more diagnostically superior. 46,47 However, it must be noted that the Fibroscan has yet to be sufficiently validated for routine clinical use. Finally, patients in this study who were followed up without therapeutic intervention underwent serial liver biopsies for a number of reasons. This study started at a time when conventional interferon was the only approved treatment for chronic hepatitis B, from 1994 to Few patients accepted interferon therapy because of poor response rate, disproportionately high cost, and associated side effects. In conclusion, our study shows that sustained disease remission defined as spontaneous HBeAg seroconversion with HBV DNA less than 10 4 copies/ml is associated with regression of fibrosis or reduction in the rate of fibrosis progression per year. Antiviral therapy should be consid-

8 HEPATOLOGY, Vol. 46, No. 3, 2007 HUI ET AL. 697 ered in those patients who develop HBV DNA 10 4 copies/ml after HBeAg seroconversion because they are still at risk of their fibrosis progressing. Acknowledgment: AS Lok for her sound advice, YY Lau for her help and A Wong. References 1. Lee W. Hepatitis B virus infection. N Engl J Med 1997;337: McQuillan GM, Coleman PJ, Kruszon-Moran D, Moyer LA, Lambert SB, Margolis HS. Prevalence of hepatitis B virus infection in the United States: the National Health and Nutrition Examination Surveys, Am J Public Health 1999;89: Hepatitis B Foundation. What is hepatitis B? Statistics. Available at: Accessed April 20, Poterucha JJ, Wiesner RH. Liver transplantation and hepatitis B. Ann Intern Med 1997;126: van Nunen AB, Hansen BE, Suh DJ, Lohr HF, Chemello L, Fontaine H, et al. Durability of HBeAg seroconversion following antiviral therapy for chronic hepatitis B: relation to type of therapy and pretreatment serum hepatitis B virus DNA and alanine aminotransferase. Gut 2003;52: Liaw YF, Chu CM, Su IJ, Huang MY, Lin DY, Chang-Chien CS. Clinical and histological events preceding hepatitis B e antigen seroconversion in chronic type B hepatitis. Gastroenterology 1983;84: Lok AS, Lai CL, Wu PC, Leung EK, Lam TS. Spontaneous hepatitis B e antigen to antibody seroconversion and reversion in Chinese patients with chronic hepatitis B virus infection. Gastroenterology 1987;92: Fattovich G. Natural history and prognosis of hepatitis B. Semin Liver Dis 2003;23: McMahon BJ, Holck P, Bulkow L, Snowball M. Serologic and clinical outcomes of 1536 Alaska Natives chronically infected with hepatitis B virus. Ann Intern Med 2001;135: Hsu YS, Chien RN, Yeh CT, Sheen IS, Chiou HY, Chu CM, Liaw YF. Long-term outcome after spontaneous HBeAg seroconversion in patients with chronic hepatitis B. HEPATOLOGY 2002;35: Craxi A, Di Bona D, Camma C. Interferon-alpha for HBeAg positive chronic hepatitis B. J Hepatol 2003;39(suppl 1):S99-S Lau DT, Everhart J, Kleiner DE, Park Y, Vergalla J, Schmid P, et al. Long-term follow-up of patients with chronic hepatitis B treated with interferon alfa. Gastroenterology 1997;113: Yang HI, Lu SN, Liaw YF, You SL, Sun CA, Wang LY, et al. Hepatitis B e antigen and the risk of hepatocellular carcinoma. N Engl J Med 2002; 347: Manno M, Camma C, Schepis F, Bassi F, Gelmini F, Miselli F, et al. Natural history of chronic HBV carriers in Northern Italy: morbidity and mortality after 30 years. Gastroenterology 2004;127: Niederau C, Heintges T, Lange S, Goldmann G, Niederau CM, Mohr L, et al. Long-term follow-up of HBeAg positive patients treated with interferon alfa for chronic hepatitis B. N Engl J Med 1996;334: Hui CK, Lau E, Monto A, Kim M, Luk JM, Poon RT, et al. Natural history of patients with recurrent chronic hepatitis C virus and occult hepatitis B co-infection after liver transplantation. Am J Transplant 2006; 6: Hui CK, Lau E, Wu H, Monto A, Kim M, Luk JM, et al. Fibrosis progression in chronic hepatitis C patients with occult hepatitis B co-infection. J Clin Virol 2006;35: Lau GK, Piratvisuth T, Luo KX, Marcellin P, Thongsawat S, Cooksley G, et al. Peginterferon alfa-2a (40kD) (PEGASYS) monotherapy and in combination with lamivudine is more effective than lamivudine monotherapy in HBeAg positive chronic hepatitis B: results from a large, multinational study. N Engl J Med 2005;352: Marcellin P, Lau G, Bonino F, Farci P, Hadziyannis S, Jin R, et al. Peginterferon alfa-2a alone, lamivudine alone and the two in combination in patients with HBeAg-negative chronic hepatitis B. N Engl J Med 2004; 351: Lai CL, Chien RN, Leung NW, Chang TT, Guan R, Tai DI, et al. A one-year trial of lamivudine for chronic hepatitis B. N Engl J Med 1998; 339: Lau GK, Cooksley H, Ribeiro RM, Powers KA, Bowden S, Mommeja- Marin H, et al. Randomized, double-blind study comparing adefovir dipivoxil (ADV) plus emtricitabine (FTC) combination therapy versus ADV alone in HBeAg ( ) chronic hepatitis B: efficacy and mechanisms of treatment response [Abstract]. HEPATOLOGY 2004;40(Suppl 1):666A. 22. Lai CL, Leung N, Teo EK, Tong M, Wong F, Hann HW, et al. A 1-year trial of telbivudine, lamivudine, and the combination in patients with hepatitis B e antigen-positive chronic hepatitis B. Gastroenterology 2005; 129: Marcellin P, Mommeja-Marin H, Sacks SL, Lau GK, Sereni D, Bronowicki JP, et al. A phase II dose-escalating trial of clevudine in patients with chronic hepatitis B. HEPATOLOGY 2004;40: Bisceglie AM, Conjeevaram HS, Fried MW, Sallie R, Park Y, Yurdaydin C, et al. Ribavirin as therapy for chronic hepatitis C. A randomized, double-blind, placebo-controlled trial. Ann Intern Med 1995;123: Ishak KG, Baptista A, Bianchi L, Callea F, De Groote J, Gudak F, et al. Histological grading and staging of chronic hepatitis. J Hepatol 1995;22: Poynard T, Mathurin P, Lai CL, Guyader D, Poupon R, Tainturier MH, et al. A comparison of fibrosis progression in chronic liver diseases. J Hepatol 2003;38: Hui CK, Sun J, Au WY, Lie AK, Yueng YH, Zhang HY, et al. Occult hepatitis B virus infection in hematopoietic stem cell donors in a hepatitis B virus endemic area. J Hepatol 2005;42: Hui CK, Cheung WW, Au WY, Lie AK, Zhang HY, Yueng YH, et al. Hepatitis B reactivation after withdrawal of preemptive lamivudine in patients with hematological malignancy upon completion of cytotoxic chemotherapy. Gut 2005;154: Hui CK, Cheung WW, Zhang HY, Au WY, Yueng YH, Leung AY, et al. Kinetics and risk of de novo hepatitis B infection in HBsAg negative patients undergoing cytotoxic chemotherapy. Gastroenterology 2006;131: Lau GK, Leung YH, Fong DY, Au WY, Kwong YL, Lie A, et al. High hepatitis B virus (HBV) DNA viral load as the most important risk factor for HBV reactivation in patients positive for HBV surface antigen undergoing autologous hematopoietic cell transplantation. Blood 2002;99: Lok AS, McMahon BJ. Chronic hepatitis B. HEPATOLOGY 2007;45: Hui CK, Lau GK. The natural history of chronic hepatitis B infection. In: Buti M, Esteban R. First edition. Barcelona: Grup 3, Viral Hepatitis 2006: Chen CJ, Yang HI, Su J, Jen CL, You SL, Lu SN, et al. Risk of hepatocellular carcinoma across a biological gradient of serum hepatitis B virus DNA level. JAMA 2006;295: Iloeje UH, Yang HI, Su J, Jen CL, Chen CJ. Risk Evaluation of Viral Load Elevation and Associated Liver Disease/Cancer-In HBV (the RE- VEAL-HBV) Study Group. Predicting cirrhosis risk based on the level of circulating hepatitis B viral load. Gastroenterology 2006;130: Hui CK, Belaye T, Montegrande K, Wright TL. A comparison in the progression of liver fibrosis in chronic hepatitis C between persistently normal and elevated transaminase. J Hepatol 2003;38: Hui CK, Monto A, Belaye T, Lau E, Wright TL. Outcomes of interferon- and ribavirin treatment for chronic hepatitis C patients with normal serum aminotransaminases. Gut 2003;52: Martinot-Peignoux M, Boyer N, Colombat M, Akremi R, Pham BN, Ollivier S, et al. Serum hepatitis B virus HBV DNA levels and liver histology in inactive HBsAg carriers. J Hepatol 2002;36: Chen YC, Sheen IS, Chu CM, Liaw YF. Prognosis following spontaneous HBsAg clearance in chronic hepatitis B patients with or without concurrent infection. Gastroenterology 2002;123:

9 698 HUI ET AL. HEPATOLOGY, September Hui CK, Lie A, Au WY, Leung YH, Ma SY, Cheung WW, et al. A long-term follow-up study on hepatitis B surface antigen positive patients undergoing allogeneic hematopoietic stem cell transplantation. Blood 2005;106: Kao JH, Chen PJ, Lai MY, Chen DS. Hepatitis B genotypes correlate with clinical outcomes in patients with chronic hepatitis B. Gastroenterology 2000;1818: Chan HL, Hui AY, Wong ML, Tse AM, Hung LC, Wong VW, et al. Genotype C hepatitis B virus infection is associated with an increased risk of hepatocellular carcinoma. Gut 2004;53: Standish RA, Cholongitas E, Dhillon A, Burroughs AK, Dhillon AP. An appraisal of the histopathological assessment of liver fibrosis. Gut 2006;55: Braunstein H. Needle biopsy of the liver in cirrhosis. Arch Pathol 1956; 62: Waldstein SS, Szanto PB. Accuracy of sampling by needle biopsy in diffuse liver disease. Arch Pathol 1950;50: Hubscher SG. Histological grading and staging in chronic hepatitis: clinical applications and problems. J Hepatol 1998; Castera L, Vergniol J, Foucher J, Le Bail B, Chanteloup E, Haaser M, et al. Prospective comparison of transient elastography, Fibrotest, APRI, and liver biopsy for the assessment of fibrosis in chronic hepatitis C. Gastroenterology 2005;128: Foucher J, Chanteloup E, Vergniol J, Castera L, Le Bail B, Adhoute X, et al. Diagnosis of cirrhosis by transient elastography (FibroScan): a prospective study. Gut 2006;55: