Chronic infection with hepatitis B virus (HBV) is still a

Transcription

1 CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2012;10: Incidence and Determinants of Spontaneous Hepatitis B e Antigen and DNA in Patients With Chronic Hepatitis B HWAI I YANG,*,, HSIU LIAN HUNG, MEI HSUAN LEE, JESSICA LIU,, CHIN LAN JEN, JUN SU, LI YU WANG, # SHENG NAN LU,** SAN LIN YOU, UCHENNA H. ILOEJE, and CHIEN JEN CHEN,, for the Risk Evaluation of Viral Load Elevation and Associated Liver Disease/Cancer HBV (REVEAL-HBV) Study Group *Molecular and Genomic Epidemiology Center, China Medical University Hospital, Taichung, Taiwan; Graduate Institute of Clinical Medical Science, China Medical University, Taichung, Taiwan; Genomics Research Center, Academia Sinica, Taipei, Taiwan; Graduate Institute of Epidemiology and Preventive Medicine, National Taiwan University, Taipei, Taiwan; Formerly affiliated with Bristol-Myers Squibb, Co, Wallingford, Connecticut; # MacKay Medical College, Taipei, Taiwan; **Kaohsiung Chang Gung Memorial Hospital, Taipei, Taiwan; Global Health Outcomes and Outcomes Research, Bristol-Myers Squibb, Co, Wallingford, Connecticut BACKGROUND & AIMS: The spontaneous seroclearance of hepatitis B e antigen (HBeAg) and hepatitis B virus (HBV) DNA are important markers of progression of chronic HBV infection. We performed a long-term cohort study to elucidate the incidence and determinants of HBeAg and HBV DNA seroclearance in patients with chronic hepatitis B. METHODS: A total of 1289 participants with a serum HBV DNA level of 10,000 copies/ml or more and without cirrhosis when the study began ( ) were followed up until June A subset of patients that tested positive for HBeAg at baseline (n 439) was included in the analysis of HBeAg seroclearance. Cox proportional hazards models were used to estimate seroclearance rate ratios for various determinants associated with the outcomes. RESULTS: After person-years of follow-up evaluation, HBeAg seroclearance occurred in 187 participants (incidence rate, 5.9 per 100 personyears). The cumulative lifetime incidence of HBeAg seroclearance among patients who were 30 to 40, or 50, 60, 70, or 74 years old was 38.8%, 69.4%, 81.9%, 89.1%, and 95.5%, respectively. Major predictors of HBeAg seroclearance included female sex, genotype B, the precore 1896 mutant, increased serum levels of alanine aminotransferase, and low baseline serum levels of HBV DNA. The median (interquartile range) serum level of HBV DNA at the time of HBeAg seroclearance was 177,801 copies/ml (4941 3,247,560 copies/ml). HBV DNA seroclearance occurred in 199 participants (15.4%) during the mean follow-up period of 7.8 years (incidence rate, 1.97 per 100 person-years). The cumulative lifetime incidence of HBV DNA seroclearance at 40, 50, 60, 70, and 77 years old was 10.0%, 25.0%, 38.8%, 54.2%, and 82.8%, respectively. Lower levels of HBV DNA at study entry and among those with the precore 1896 wild-type variant were associated with an increased rate of HBV DNA seroclearance. Among individuals who were HBeAg-seropositive at study entry and cleared serum HBV DNA during the follow-up period, 89% had cleared HBeAg by the time they had an undetectable serum level of HBV DNA. CONCLUSIONS: Serum level of HBV DNA is the most important predictor of seroclearance of HBeAg and HBV DNA. This finding supports current clinical guidelines for antiviral treatments of chronic hepatitis B. Keywords: Long-term Follow-up Study; Virology; Epidemiology; Natural Course. Chronic infection with hepatitis B virus (HBV) is still a serious health problem because of its worldwide distribution and unfavorable sequelae. 1,2 Despite the availability of effective vaccines since 1982, more than 350 million people in the world are chronically infected. 1,2 The majority of them reside in the Asia-Pacific region, where the infection usually is acquired perinatally or in early childhood. 3,4 They are at an extraordinarily high risk of developing cirrhosis, hepatic decompensation, and hepatocellular carcinoma (HCC). 5 The natural course of chronic HBV infection acquired in early life is divided conventionally into 3 phases. 6 8 The first (immune tolerance) phase features positivity for hepatitis B e antigen (HBeAg), an extremely high amount of circulating HBV DNA, and normal levels of alanine aminotransferase (ALT) in the serum. This phase may last for years with an individual variation and is thought to be associated with no evidence of liver injury. The second (immune seroclearance) phase is characterized by episodic acute flares of serum ALT levels resulting from specific T-lymphocyte mediated cellular responses to viral antigens and apoptosis of hepatocytes. These events may result in the decrease of viral replication and HBeAg seroclearance/ seroconversion, and subsequent seroclearance of HBV DNA. However, a prolonged phase or recurrent episodes of acute liver inflammation may lead to repeated cycles of injury and regeneration, leading to necroinflammation/fibrosis and an increased risk of progression to cirrhosis and HCC. Finally, a proportion of persons are able to enter the residual inactive phase, which is characterized by low serum levels of HBV DNA, absence of HBeAg and presence of antibodies against HBeAg, and normal serum ALT levels, with the continued presence of hepatitis B surface antigen (HBsAg) in serum. Spontaneous HBsAg seroclearance may occur years after staying in this phase. Recent advances in the understanding of the natural history of chronic HBV infection revealed that replication of HBV is the fundamental element that drives the disease progression and clinical sequelae in patients with chronic hepatitis B. 6,9 12 Serum HBV DNA level is a direct measurement of HBV load in circulating blood, and HBeAg is considered a surrogate serologic marker of it. The seroclearance/seroconversion of HBeAg, decline of serum HBV DNA levels to undetectable levels (HBV DNA seroclearance), and HBsAg seroclearance/seroconversion Abbreviations used in this paper: ALT, alanine aminotransferase; CI, confidence interval; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma by the AGA Institute /$36.00 doi: /j.cgh

2 528 YANG ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No. 5 Figure 1. Flow of participants included in this study. HCV, hepatitis C virus. are important milestones in the natural history of chronic HBV infection, 12 and have become primary goals of treatment. 5,13,14 Transitions of these milestones in the natural history are important parameters to be contrasted with those under antiviral treatment. However, very few studies have investigated the incidence and determinants of the spontaneous seroclearance of these markers using a long-term follow-up community-based study. The aim of this analysis was to investigate the incidence and determinants of spontaneous seroclearance of HBeAg and HBV DNA in a long-term follow-up cohort study on the natural history of chronic hepatitis B patients with high viral loads ( 10 4 copies/ml). Patients and Methods Study Populations The study participants came from a subset of the Risk Evaluation of Viral Load Elevation and Associated Liver Disease/Cancer HBV cohort, 12,15 which was a community-based cohort enrolled from 7 townships in Taiwan from 1991 to 1992 and followed up until June At enrollment, 4155 participants were seropositive for HBsAg (Figure 1). They were invited to return for follow-up visits every 6 to 12 months. Exclusion criteria of this analysis included the lack of adequate serum samples for HBV DNA testing (n 304), the seropositivity or absence of data on antibodies against hepatitis C virus (n 198), a diagnosis of cirrhosis within 6 months of entry (n 69), baseline HBV DNA level less than 10,000 copies/ml (n 2020), and the absence of follow-up serum levels of HBV DNA (n 275). A total of 1289 participants were finally included in the analysis on seroclearance of HBV DNA in this study. The subset of HBeAg-seropositive patients at study entry (n 439) was used for the analysis on HBeAg seroclearance. The analysis of seroclearance of HBV DNA posterior to HBeAg seroclearance was based on 187 participants who serocleared HBeAg during the follow-up period. This study was approved by the Institutional Review Board of the College of Public Health, National Taiwan University, Taipei, Taiwan. Data Collection and Biomarker Tests A structured questionnaire was used by public health nurses to interview all participants in person at enrollment to obtain information on sociodemographic characteristics, dietary habits, history of cigarette smoking and alcohol intake, personal medical and surgical history, as well as family history of cancers and other major diseases. A 10-mL blood sample was drawn from each participant at each visit. The sample was fractionated on the same day of collection and frozen at 70 C. Laboratory tests are described in the Supplementary Materials and Methods section. Outcome Measure The spontaneous clearance of HBeAg in serum (HBeAg seroclearance) and decrease of serum HBV DNA levels to un-

3 May 2012 SPONTANEOUS HBeAG AND HBV DNA SEROCLEARANCE 529 detectable levels ( 100 copies/ml, HBV DNA seroclearance) were 2 major outcomes of interest. The first occasion at which a participant tested seronegative for HBeAg or had a serum HBV DNA level less than 100 copies/ml was defined to be the date of seroclearance of HBeAg or HBV DNA, respectively. Statistical Analysis The person-years of follow-up evaluation for each participant were computed from the date of enrollment to the date at the occurrence of specific study outcomes, the date of death, or the date at last follow-up visit until June 30, 2004, whichever came first. Participants who were seropositive for HBeAg or with detectable serum HBV DNA levels at their death or at the last follow-up dates were defined as censored for the specific study outcomes. The Kaplan Meier method was used to estimate the cumulative probability of seroclearance of HBeAg and HBV DNA by follow-up year, as well as cumulative lifetime probability of seroclearance of HBeAg and HBV DNA by age with the use of late entry method. Cox proportional hazards models were used to estimate univariate- and multivariate-adjusted seroclearance rate ratios with 95% confidence intervals (CIs) for associated determinants of HBeAg and HBV DNA seroclearance, respectively. Statistical significance was determined by 2-tailed tests with significance set at a P value of less than.05. Statistical software (SAS version 9.2; SAS Institute, Cary, NC) was used for statistical analyses. Results Basic Characteristics of Participants As shown in Table 1, a total of 1289 participants were included in this study. The majority of them were men (69%), aged 30 to 39 years old (36%), without habits of cigarette smoking (65%) or alcohol consumption (87%), and with serum ALT levels less than 45 IU/mL (91%) and HBeAg-seronegative status (66%). All participants had serum HBV DNA levels of 10,000 or more copies/ml at study entry, while the HBV in most of these patients was genotype B (65%) with precore G1896A mutation (58%) and basal core promoter wild-type variant (56%). Incidence Rate and Cumulative Incidence of Hepatitis B e Antigen Seroclearance At cohort entry, a total of 439 participants were HBeAgseropositive with high serum HBV DNA levels ( 10 4 copies/ ml). After person-years of follow-up evaluation, HBeAg seroclearance occurred in 187 participants, giving an incidence rate of 5.9 per 100 person-years (5.9% per year). The cumulative incidence of HBeAg seroclearance by 5 and 10 years after recruitment was 18.8% and 42.4%, respectively. After seroclearance, only 3 (1.6%) participants had HBeAg re-emergence in their serum. As shown in Figure 2A, the cumulative lifetime incidence of HBeAg seroclearance from 30 to 40, 50, 60, 70, and 74 years old was 38.8%, 69.4%, 81.9%, 89.1%, and 95.5%, respectively. Determinants of Hepatitis B e Antigen Seroclearance As shown in Table 2, sex, HBV genotype, HBV precore 1896 mutant, baseline serum ALT level, and baseline serum Table 1. Demographic and Clinical Characteristics of 1289 Study Participants Variable Number of subjects, % Sex Female 402 (31.2) Male 887 (68.8) Age, y (36.5) (30.1) (25.8) (7.7) Cigarette smoking No 841 (65.3) Yes 446 (34.7) Alcohol consumption No 1124 (87.4) Yes 162 (12.6) ALT level, U/L at entry (50.0) (40.8) (9.2) HBV DNA level, copies/ml at entry 1 million 501 (38.9) 100, , (21.3) 10,000 99, (39.9) HBeAg Positive 439 (34.1) Negative 850 (65.9) HBV genotype a BorB C 778 (64.8) C 423 (35.2) Precore mutation b Wild type 432 (42.3) G1896A 484 (47.4) Wild type G1896A 105 (10.3) BCP mutation c Wild type 568 (56.1) A1762T/G1764A 286 (28.3) Mixed type 158 (15.6) a Data were unavailable for 88 subjects. b Data were unavailable for 268 subjects. c Data were unavailable for 277 subjects. HBV DNA level were associated statistically significantly with HBeAg seroclearance in multivariate analysis. The multivariateadjusted rate ratio was 1.89 (95% CI, ) for women compared with men; 3.19 (95% CI, ) for HBV genotype B or B and C co-infection compared with genotype C; 1.53 (95% CI, ) for precore mutants compared with the wild type; 1.59 (95% CI, ) and 2.96 (95% CI, ), respectively, for baseline serum ALT levels of 15 to 44 and 45 or more compared with less than 15 U/L; and 1.98 (95% CI, ), 1.46 (95% CI, ), 2.61 (95% CI, ), and 4.36 (95% CI, ), respectively, for baseline serum HBV DNA levels of 10 7 to less than 10 8,10 6 to less than 10 7,10 5 to less than 10 6, and 10 4 to less than 10 5, compared with 10 8 or more copies/ml. Serial Serum Hepatitis B Virus DNA and Alanine Aminotransferase Levels Before Hepatitis B e Antigen Seroclearance Among 187 participants who had serocleared HBeAg during the follow-up period, 177 had available data on serum

4 530 YANG ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No U/L (interquartile range, U/L), respectively. Serum HBV DNA levels started to decline gradually from 6 years before HBeAg seroclearance (Figure 2B), and an ALT flare also was observed 6 years before HBeAg seroclearance (Figure 2C). Incidence Rate and Cumulative Incidence of Hepatitis B Virus DNA Seroclearance Among 1289 participants who had high serum HBV DNA levels ( 10 4 copies/ml), 199 (15.4%) serocleared HBV DNA during a mean follow-up period of 7.8 years. The incidence rate of spontaneous HBV DNA seroclearance was 1.97 per 100 person-years (around 2.0% per year). The cumulative incidence of HBV DNA seroclearance was 5.2% and 14.5%, respectively, at 5 and 10 years of follow-up evaluation (Figure 3A). The cumulative lifetime incidence of HBV DNA seroclearance from 30 to 40, 50, 60, 70, and 77 years old was 10.0%, 25.0%, 38.8%, 54.2%, and 82.8%, respectively (Figure 3B). In the subset of 850 participants who were HBeAg-seronegative at study entry, 163 (19.2%) serocleared HBV DNA during follow-up evaluation, giving an incidence rate of 2.45 per 100 person-years (2.5% per year). The cumulative lifetime incidence of HBV DNA seroclearance from 30 to 40, 50, 60, 70, and 77 years old in this group was 11.0%, 27.8%, 43.2%, 59.8%, and 85.0%, respectively, which was slightly higher than the rates for all participants. Determinants of Spontaneous Seroclearance of Hepatitis B Virus DNA Table 3 shows the multivariate Cox regression analyses of determinants of HBV DNA seroclearance. In both all participants and HBeAg-seronegative participants, lower HBV DNA levels at study entry, and precore 1896 wild-type variant were associated with an increased rate of HBV DNA seroclearance. The seroclearance rate ratio in all participants was 1.72 (95% CI, Table 2. Multivariate-Adjusted HBeAg Seroclearance Rate Ratio With 95% Confidence Interval for Each Predictor Included in the Cox Regression Analysis Variable Rate ratio (95% CI) P value Figure 2. (A) Cumulative lifetime probability of seroclearance of HBeAg (30 74 years old); long-term tracking of (B) serum HBV DNA levels and (C) serum ALT levels in participants with HBeAg seroclearance during follow-up evaluation. HBV DNA levels at the time of HBeAg seroclearance. Among them, only 20 (11%) had undetectable levels of HBV DNA at the time of HBeAg seroclearance. The median of serum HBV DNA and ALT levels at the time of HBeAg seroclearance were 177,801 copies/ml (interquartile range, ,247,560 copies/ml) and Sex Male Female 1.89 ( ).002 Age, y ( ).44 HBV genotype C B/B C 3.19 ( ).001 Precore 1896 Wild-type (G) Mutant (A or mixed) 1.53 ( ).043 Baseline ALT level, U/L ( ) ( ).001 HBV DNA level, copies/ml at study entry 100,000,000 10,000,000 99,999, ( ).005 1,000,000 9,999, ( ) , , ( ) ,000 99, ( ).001

5 May 2012 SPONTANEOUS HBeAG AND HBV DNA SEROCLEARANCE 531 Figure 3. (A) Cumulative probability of declining HBV DNA concentrations to specified levels by follow-up years. (B) Cumulative lifetime probability of seroclearance of HBV DNA in serum. (C) Cumulative probability of reaching an undetectable HBV DNA level after HBeAg seroclearance during follow-up evaluation. Long-term tracking of (D) serum HBV DNA levels and (E) serum ALT levels in participants who cleared circulating HBV DNA during follow-up evaluation ) and 3.45 (95% CI, ), respectively, for HBV DNA levels of 100,000 to 999,999 and 10,000 to 99,999 copies/ml compared with the levels of 1,000,000 copies/ml or more as the referent; and was 1.38 (95% CI, ) and 2.56 (95% CI, ), respectively, for participants seronegative for HBeAg at study entry. The seroclearance rate ratio for precore 1896 mutant was 0.55 (95% CI, ) in all participants and HBeAg-seronegative subset. Cumulative Incidence of Hepatitis B Virus DNA Seroclearance After Hepatitis B e Antigen Seroclearance During the Follow-up Period Among 177 participants who were HBeAg-seropositive at study entry and subsequently serocleared HBeAg during the follow-up evaluation, 20 participants serocleared HBV DNA

6 532 YANG ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No. 5 Table 3. Multivariate-Adjusted HBV DNA Seroclearance Rate Ratio With 95% Confidence Interval for Each Predictor Included in the Cox Regression Analysis All participants (n 1289) HBeAg-seronegative patients at entry (n 850) Variable Rate ratio (95% CI) P value Rate ratio (95% CI) P value Sex Male Female 1.02 ( ) ( ).77 Age, per year 0.99 ( ) ( ).63 HBV DNA level, copies/ml at study entry 1,000, , , ( ) ( ).40 10,000 99, ( ) ( ).010 HBeAg Positive (Not included) Negative 1.87 ( ).09 Baseline ALT level, U/L ( ) ( ) ( ) ( ).08 Precore 1896 Wild-type (G) Mutant (A or mixed) 0.55 ( ) ( ).008 before or on the same occasion with HBeAg seroclearance and 40 participants did not have serum HBV DNA levels for samples collected after HBeAg seroclearance. As a result, 15 of 117 participants (12.8%) eventually serocleared HBV DNA after HBeAg seroclearance during the follow-up evaluation. As shown in Figure 3C, the cumulative incidence of HBV DNA seroclearance at 5 and 10 years after HBeAg seroclearance was 11.5% and 45.5%, respectively. Hepatitis B e Antigen Serostatus at the Time of Hepatitis B Virus DNA Seroclearance and Serial Serum Hepatitis B Virus DNA and Alanine Aminotransferase Levels Before Hepatitis B Virus DNA Seroclearance Among 36 participants who were HBeAg-seropositive at study entry and serocleared HBV DNA during the follow-up evaluation, 32 (89%) had serocleared HBeAg at the time of HBV DNA seroclearance. There was a gradually decreasing trend in serum HBV DNA levels more than 10 years before HBV DNA seroclearance (Figure 3D), whereas an increased ALT level was observed 5 years before HBV DNA seroclearance (Figure 3E). Discussion It is very challenging to directly observe serial changes in important HBV seromarkers and delineate their temporality using a community-based chronic hepatitis B natural history cohort, especially when transition rates of these markers were low. It requires long-term tracking of cohort members, careful testing of blood samples collected at each participant s visit, and appropriate follow-up intervals to achieve this goal. In the present study, only 11% of participants had undetectable serum level of HBV DNA (HBV DNA seroclearance) at the time of HBeAg seroclearance, with a median serum HBV DNA level of around 10 5 copies/ml. In participants who had baseline serum HBV DNA levels of 10,000 or more copies/ml and then serocleared HBV DNA during the follow-up evaluation, most (163 of 199; 82%) were HBeAg seronegative at study entry. In the remaining 36 participants who were HBeAg-seropositive at study entry and serocleared HBV DNA during follow-up evaluation, 32 had cleared HBeAg at the time of HBV DNA seroclearance. Therefore, in our observation, almost all (195 of 199) participants who lost HBV DNA in serum cleared HBeAg before the seroclearance of HBV DNA. Our previous study also reported that at the time of HBsAg seroclearance, 95.8% of participants had undetectable serum HBV DNA levels. 16 As shown in Supplementary Tables 1 and 2, patients with baseline HBeAg seropositivity had a very low probability of seroclearance of HBV DNA and HBsAg before the HBeAg seroclearance. Likewise, patients who did not experience seroclearance of HBV DNA had an extremely low chance to clear HBsAg, regardless of HBV DNA level at baseline. Incorporating all these data, 16 our studies confirmed that seroclearance of HBeAg, HBV DNA, and HBsAg appeared to happen chronologically, with HBeAg seroclearance being the first milestone followed by HBV DNA seroclearance, and then finally the seroclearance of HBsAg. The chronologic order of seroclearance of HBeAg and HBV DNA are very distinct among patients without and under antiviral treatment. With potent antiviral therapy, a patient s serum HBV DNA level could be lowered to an undetectable level rapidly, followed by a much tardier rate of HBeAg seroclearance/seroconversion (12% 32% at the end of the first year of treatment, which is still much higher than the spontaneous HBeAg seroclearance rate in this natural history study). 8 In contrast, our natural history data showed that HBeAg seroclearance occurred first, then the seroclearance of HBV DNA. The HBeAg reappearance rate after HBeAg seroclearance in antiviral treatment was quite different from that in the natural history as observed in this study. The reappearance rate was 1.6% in the current study compared with the cumulative reappearance rates of 11%, 13%, and 16%, respectively, at 6 months, 1 year, and 2

7 May 2012 SPONTANEOUS HBeAG AND HBV DNA SEROCLEARANCE 533 years after the cessation of lamivudine therapy. 17 The rates of reaching undetectable serum HBV DNA levels were quite different in these 2 circumstances. Compared with antiviral treatments, which suppressed the virus to undetectable levels at the first year with a proportion of patients ranging from 21% to 93%, 8 the spontaneous seroclearance of HBV DNA showed a much lower rate in our study. The plausible mechanisms that underlie these differences between antiviral therapy and natural history may occur because only host immunity drives the progression of marker changes in the natural history. The host immune system brings a pressure to the HBV, lowering their amount and forcing their evolution into the immune escape precore mutant strain. This precore stop-codon mutation abolishes the synthesis of HBeAg while continuing production of virions. On the contrary, most current antiviral nucleos(t)ide analogues effectively suppress HBV replication by inhibiting viral reverse transcriptases; they cannot affect the covalently closed circular DNA, the replicative template of HBV within the infected hepatocytes. Thus, the production of HBeAg persists even after years of treatment. These mechanisms also explain why HBeAg reappears quickly after the cessation of antiviral therapy. The annual incidence of spontaneous HBeAg seroclearance/ seroconversion has been reported to be 2% to 15% in previous studies, depending on age, HBV genotype, and ALT level In the current study, the annual incidence rate of spontaneous HBeAg seroclearance was estimated to be 5.9%. The cumulative lifetime probability of HBeAg seroclearance after 30 years increased rapidly, to around 70% at 50 years and around 90% at 70 years (Figure 2A). The seroclearance of circulating HBV DNA without antiviral therapy has rarely been explored. We estimated that the incidence rate of spontaneous seroclearance of HBV DNA was around 2.0% per year for participants who had serum HBV DNA levels of 10,000 or more copies/ml at study entry. The incidence rate was 2.5% per year for participants who were HBeAg-seronegative at study entry. In our previous study, 16 the annual HBsAg seroclearance rate was 2.26% for the entire cohort, and approximately 5% among participants with serum HBV DNA levels of 10,000 or more copies/ml and decreased to undetectable levels before HBsAg seroclearance. 16 During the transitions of all these seromarkers, serum HBV DNA levels seemed to play a central role. It was consistently the most important predictor of not only the disease progression of chronic hepatitis B but also seroclearance of HBeAg, HBV DNA, and HBsAg. This fact in the natural history has been well responded to by recommendations of current guidelines, 5,13,14 in which the primary goal is persistent suppression of HBV replication at the lowest levels possible without development of resistance, hoping to accelerate seroclearance/seroconversion of HBeAg and HBsAg, and, ultimately, reducing the morbidity and mortality associated with both cirrhosis and HCC. It is interesting to note that HBV precore 1896 mutant seemed to play the opposite role in the seroclearance of HBeAg and HBV DNA. Participants infected with precore G1896A mutant HBV had a higher probability of HBeAg seroclearance than participants infected with wild-type HBV (Table 2), whereas they had a lower probability of HBV DNA seroclearance (Table 3). A study observed that the precore stop mutant to total viremia ratios tended to correlate with seroconversion of antibodies against HBeAg, but high precore mutant ratios were associated with persistently high viremia after seroconversion of antibodies against HBeAg. 23 The G-to-A mutation at nucleotide 1896 generates a stop codon (TAG) in the precore sequence, resulting in the translational defect in this region, which abolishes the synthesis and secretion of HBeAg yet continues viral replication. 24,25 However, it might provide a selection advantage in persistently infected hosts, by immune elimination of hepatocytes harboring nondefective HBV with the HBeAg expression. 26 There are some other factors that deserve further research on their role in the seroclearance of HBeAg, HBV DNA, and HBsAg. The quantification of HBsAg in serum is an area of current research interest. Recent evidence has shown that both HBsAg and HBV DNA levels decrease along the natural course of chronic hepatitis B. 27 The HBsAg kinetics could be an important candidate factor associated with seroclearance outcomes. Host genetic variants have been found to be associated with the risk of persistent infection with HBV as well as the development of HCC. 28,29 This approach potentially may be applied to studies that examine host genetic features that affect transitions of important milestones in the natural history of chronic hepatitis B. In conclusion, this study confirmed the major temporal order of seroclearance of HBeAg and HBV DNA in the natural history of chronic HBV infection. During the transitions of these milestones, serum HBV DNA level was consistently the most important predictor. Implementation with antiviral treatments might accelerate seroclearance of HBeAg, HBV DNA, and HBsAg; and hopefully reduce the morbidity and mortality resulting from subsequent adverse progression. Supplementary Material Note: To access the supplementary material accompanying this article, visit the online version of Clinical Gastroenterology and Hepatology at and at doi: / j.cgh References 1. Lavanchy D. Hepatitis B virus epidemiology, disease burden, treatment, and current and emerging prevention and control measures. J Viral Hepat 2004;11: World Health Organization. Fact sheet: hepatitis B, Available at: en/index.html. 3. Chen CJ, Wang LY, Yu MW. Epidemiology of hepatitis B virus infection in the Asia-Pacific region. J Gastroenterol Hepatol 2000; 15(Suppl):E3 E6. 4. Custer B, Sullivan SD, Hazlet TK, et al. Global epidemiology of hepatitis B virus. J Clin Gastroenterol 2004;38:S158 S Lok AS, McMahon BJ. Chronic hepatitis B: update Hepatology 2009;50: Chen CJ, Yang HI, Su J, et al. Risk of hepatocellular carcinoma across a biological gradient of serum hepatitis B virus DNA level. JAMA 2006;295: Lee WM. Hepatitis B virus infection. N Engl J Med 1997;337: Liaw YF, Chu CM. Hepatitis B virus infection. Lancet 2009;373: Yang HI, Lu SN, Liaw YF, et al. Hepatitis B e antigen and the risk of hepatocellular carcinoma. N Engl J Med 2002;347: Iloeje UH, Yang HI, Su J, et al. Predicting cirrhosis risk based on the level of circulating hepatitis B viral load. Gastroenterology 2006;130:

8 534 YANG ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No Iloeje UH, Yang HI, Jen CL, et al. Risk and predictors of mortality associated with chronic hepatitis B infection. Clin Gastroenterol Hepatol 2007;5: Chen CJ, Yang HI. Natural history of chronic hepatitis B RE- VEALed. J Gastroenterol Hepatol 2011;26: Liaw YF, Leung N, Kao JH, et al. Asian-Pacific consensus statement on the management of chronic hepatitis B: a 2008 update. Hepatol Int 2008;2: European Association for the Study of the Liver. EASL clinical practice guidelines: management of chronic hepatitis B. J Hepatol 2009;50: Chen CJ, Iloeje UH, Yang HI. Long-term outcomes in hepatitis B: the REVEAL-HBV study. Clin Liver Dis 2007;11: Liu J, Yang HI, Lee MH, et al. Incidence and determinants of spontaneous hepatitis B surface antigen seroclearance: a community-based follow-up study. Gastroenterology 2010;139: Ryu SH, Chung YH, Choi MH, et al. Long-term additional lamivudine therapy enhances durability of lamivudine-induced HBeAg loss: a prospective study. J Hepatol 2003;39: Kao JH, Chen PJ, Lai MY, et al. Hepatitis B virus genotypes and spontaneous hepatitis B e antigen seroconversion in Taiwanese hepatitis B carriers. J Med Virol 2004;72: Chu CJ, Hussain M, Lok AS. Hepatitis B virus genotype B is associated with earlier HBeAg seroconversion compared with hepatitis B virus genotype C. Gastroenterology 2002;122: Liaw YF. Hepatitis flares and hepatitis B e antigen seroconversion: implication in anti-hepatitis B virus therapy. J Gastroenterol Hepatol 2003;18: Chu CM, Liaw YF. Genotype C hepatitis B virus infection is associated with a higher risk of reactivation of hepatitis B and progression to cirrhosis than genotype B: a longitudinal study of hepatitis B e antigen-positive patients with normal aminotransferase levels at baseline. J Hepatol 2005;43: Liaw YF. Natural history of chronic hepatitis B virus infection and long-term outcome under treatment. Liver Int 2009;29(Suppl 1): Chu CM, Yeh CT, Lee CS, et al. Precore stop mutant in HBeAgpositive patients with chronic hepatitis B: clinical characteristics and correlation with the course of HBeAg-to-anti-HBe seroconversion. J Clin Microbiol 2002;40: Tong SP, Li JS, Vitvitski L, et al. Active hepatitis B virus replication in the presence of anti-hbe is associated with viral variants containing an inactive pre-c region. Virology 1990; 176: Santantonio T, Jung MC, Miska S, et al. High prevalence and heterogeneity of HBV prec mutants in anti-hbe-positive carriers with chronic liver disease in southern Italy. J Hepatol 1991; 13(Suppl 4):S78 S Okamoto H, Yotsumoto S, Akahane Y, et al. Hepatitis B viruses with precore region defects prevail in persistently infected hosts along with seroconversion to the antibody against e antigen. J Virol 1990;64: Liaw YF. Clinical utility of hepatitis B surface antigen quantitation in patients with chronic hepatitis B: a review. Hepatology 2011; 53: Kamatani Y, Wattanapokayakit S, Ochi H, et al. A genome-wide association study identifies variants in the HLA-DP locus associated with chronic hepatitis B in Asians. Nat Genet 2009;41: Zhang H, Zhai Y, Hu Z, et al. Genome-wide association study identifies 1p36.22 as a new susceptibility locus for hepatocellular carcinoma in chronic hepatitis B virus carriers. Nat Genet 2010;42: Reprint requests: Address requests for reprints to: Hwai-I Yang, PhD, Graduate Institute of Clinical Medical Science, China Medical University, Taichung 404, Taiwan. hwaii.yang@gmail.com; fax: (886) Acknowledgments Other members of the Risk Evaluation of Viral Load Elevation and Associated Liver Disease/Cancer in HBV Study Group are as follows: Y. F. Liaw from Chang-Gung Memorial Hospital and Chang-Gung University; T. H. H. Chen from the College of Public Health, National Taiwan University; S. H. Yeh from the Department of Microbiology, National Taiwan University; C. A. Sun from the Department of Public Health, National Defense Medical Center; S. Y. Chen from the Department of Public Health, Tzu Chi University; S. C. Ho and T. G. Lu from the Huhsi Health Center, Penghu County; C. C. Chen from the Kaohsu Health Center, Pingtung County; D. S. Chen, P. J. Chen, C. Y. Hsieh, H. S. Lee, P. M. Yang, C. H. Chen, J. D. Chen, S. P. Huang, and C. F. Jan from the National Taiwan University Hospital; W. C. How from the Paihsa Health Center, Penghu County; W. P. Wu and T. Y. Ou from the Provincial Penghu Hospital; K. C. Shih from the Provincial Chutung Hospital; W. S. Chung and C. Li from the Provincial Potzu Hospital; C. G. Lin from the Sanchi Health Center, Taipei County; K. E. Chu from the Shin Kong Wu Ho-Su Memorial Hospital; and M. H. Wu from the Taipei City Psychiatric Center. Conflicts of interest These authors disclose the following: Uchenna Iloeje is an employee of and holds stock in Bristol-Myers Squibb, Co; Jun Su was an employee of and held stock in Bristol-Myers Squibb, Co, previously; and Chien-Jen Chen was supported by research grants from Bristol-Myers Squibb to conduct the laboratory tests of hepatitis B virus DNA and hepatitis B e antigen for follow-up samples in this study. The remaining authors disclose no conflicts. Funding Supported by research grants from the Department of Health, Executive Yuan (Taipei, Taiwan); Academia Sinica (Taipei, Taiwan); National Health Research Institute (Taipei, Taiwan) and Bristol-Myers Squibb (Wallingford, CT) to conduct the laboratory tests for this study. The funding sources of this study had no role in the design or conduct of the study, nor did they have any role in the collection, management, analysis, and interpretation of the data as well as the preparation, review, or approval of the manuscript. All data handling and statistical analyses were performed by staff at the China Medical University and Hospital, National Taiwan University, and Academia Sinica. At no time did the funding sources have access to the data.

9 534.e1 YANG ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No. 5 Supplementary Materials and Methods Laboratory Tests Laboratory tests were performed using commercial kits: HBsAg and HBeAg by radioimmunoassay (Abbott Laboratories, North Chicago, IL), antibodies against hepatitis C virus by enzyme immunoassay with second-generation test kits (Abbott Laboratory), serum HBV DNA level by polymerase chain reaction (COBAS Amplicor; Roche Diagnostics, Indianapolis, IN), and ALT level by serum chemistry autoanalyzer (model 736; Hitachi Co, Tokyo, Japan) using commercial reagents (Biomerieux, March L Etoile, France). To trace the changes in serostatus of HBeAg and HBV DNA, these biomarkers were tested in follow-up samples collected at every visit in participants who were HBeAg-seropositive and/or had serum HBV DNA levels of 10 4 or more copies/ml using a microparticle enzyme immunoassay (AxSYM HBe 2.0; Abbott Laboratories) and real-time polymerase chain reaction (COBAS TaqMan; Roche Diagnostics), respectively. Serum ALT levels also were tested in each follow-up sample.

10 May 2012 SPONTANEOUS HBeAG AND HBV DNA SEROCLEARANCE 534.e2 Supplementary Table 1. Associations Between HBeAg Serostatus/HBeAg Seroclearance and HBV DNA at Various Baseline Serum HBV DNA Levels Baseline HBeAg-seronegative or HBeAg seroclearance during follow-up evaluation Baseline HBV DNA 10,000 99,999 copies/ml Subjects, n(%) HBV DNA, n (%) Baseline HBV DNA 100, ,999 copies/ml Subjects, n(%) HBV DNA, n (%) Bseline HBV DNA 1 million copies/ml Subjects, n(%) HBV DNA, n (%) No 4 0 (0.0) 15 0 (0.0) (0.4) Yes (23.7) (15.1) (14.2) Supplementary Table 2. Associations With HBsAg Seroclearance for HBeAg Status/HBeAg Seroclearance and HBV DNA Seroclearance at Various Baseline Serum HBV DNA Levels Baseline HBV DNA 10,000 99,999 copies/ml Baseline HBV DNA 100, ,999 copies/ml Baseline HBV DNA 1 million copies/ml Subjects, n HBsAg, n (%) Subjects, n HBsAg, n (%) Subjects, n HBsAg, n (%) Baseline HBeAg-seronegative or HBeAg seroclearance during follow-up evaluation No 4 0 (0.0) 15 0 (0.0) (0.0) Yes (9.9) (7.4) (4.9) HBV DNA during follow-up evaluation No (0.8) (0.0) (0.2) Yes (38.8) (48.7) (30.8)