Search
Close
Search
Advanced Search
Search Menu
AI Discovery Assistant

Abstract

Background

In people with hepatitis B virus (HBV) infection, persistence of hepatitis B e antigen (HBeAg) is associated with clinical progression and need for treatment. HBeAg loss represents partial immune control and is a critical event in the natural history of chronic HBV.

Methods

We conducted a systematic review and meta-analysis of cohort studies that report HBeAg loss among people with untreated chronic HBV. We evaluated HBeAg loss using a random-effects model and conducted subanalysis on region.

Results

We screened 10 560 publications, performed 196 full-text analyses, and included 26 studies for meta-analysis. The pooled rate of HBeAg loss was 6.46/100 person-years (PYs) (95% confidence interval, 5.17–8.08). Meta-regression showed that older age of participants and studies in Europe were associated with higher rate of HBeAg loss. Rates per 100 PYs were 7.43 (95% confidence interval, 6.30–8.75; 1 study) in Africa, 3.24 (2.61­–4.02; 1 study) in the Eastern Mediterranean, 13.67 (11.21–16.66; 4 studies) in Europe, 7.34 (4.61­–11.70; 5 studies) in North America, and 5.53 (4.05­–7.55; 15 studies) in the Western Pacific.

Conclusions

Spontaneous HBeAg loss occurs at a rate of 6.46/100 PYs. Variations by region and age group may reflect epidemiological, immunological, or HBV genotype-related differences.

chronic hepatitis B infection, hepatitis B e antigen, hepatitis B e antigen clearance, natural history

More than 240 million people are chronically infected with hepatitis B virus (HBV) worldwide [1]. HBV is the world's leading cause of chronic liver disease, liver cancer, and liver-related deaths [1]. In 2016, the World Health Organization (WHO) announced the ambitious goal of “eliminating viral hepatitis as a major public health threat” by 2030, targeting a 95% reduction in HBV incidence and 65% reduction in HBV-related mortality rate [1]. Efforts at HBV elimination include development of novel therapeutic agents and scaling up screening and treatment programs in low- and middle-income countries [1, 2]. While not all people with chronic HBV infection require treatment with antiviral therapy, WHO elimination efforts have motivated expanded screening efforts in HBV-endemic countries to evaluate people for treatment eligibility [3].

The natural history of chronic HBV infection consists of 5 characteristic phases: hepatitis B e antigen (HBeAg)–positive chronic infection (ie, immune tolerant), HBeAg-positive chronic hepatitis (immune active or immune clearance), HBeAg-negative chronic infection (inactive carrier), HBeAg-negative chronic hepatitis (immune escape), and hepatitis B surface antigen (HBsAg)–negative infection (occult HBV infection) [4]. Transitions between these phases depend on the presence or absence of HBeAg, replication of HBV DNA, and hepatic inflammation. The 2 most important end points of chronic HBV management are the suppression of HBV DNA level to undetectable levels and the loss of HBsAg, with the latter considered a functional cure of the disease [5]. People who achieve these end points frequently have normalization of hepatic biochemical test results (eg, aminotransferase levels) and are less likely to develop cirrhosis or hepatocellular carcinoma (HCC) [5].

Before functional cure, a critical event in HBV natural history is the loss of HBeAg, which is also associated with favorable clinical outcomes. People who experience loss of HBeAg are more likely to achieve reduction of HBV DNA without antiviral treatment [6]. However, because the precore and basal core promotor regions of the HBV genome encode for the protein HBeAg, mutations in this region could abrogate HBeAg production. Some people harboring HBV with these mutations may then appear to experience loss of HBeAg but continue to have modestly elevated HBV DNA levels. These mutations are associated with older age, more active liver inflammation, lower rates of HBsAg loss, and higher rates of cirrhosis and HCC [7, 8]. Nevertheless, persistence of HBeAg is an independent risk factor for cirrhosis, HCC, and death among patients with chronic HBV infection [9–12]. Furthermore, the presence of HBeAg in pregnant women is associated with an increased likelihood of mother-to-child transmission [13, 14].

The presence of HBeAg is not sufficient indication for initiation of antiviral therapy [4], and the benefits of antiviral treatment in HBeAg-positive patients is a subject of ongoing investigation [15]. Clinicians may wish to defer initiation of drug therapy at least temporarily in patients who are likely to undergo spontaneous loss of HBeAg. Therefore, an improved understanding of the probability of spontaneous HBeAg loss may assist clinicians in weighing the risks and benefits of initiating long-term antiviral treatment in patients during the HBeAg-positive phases of chronic HBV infection. However, to our knowledge, no current studies have been published that assess HBeAg loss during the natural history of HBeAg-positive chronic HBV infection in a variety of populations. The objectives of our systematic review and meta-analysis are to determine the rate of spontaneous HBeAg loss among untreated individuals with chronic HBV infection and to investigate regional and age-related variations across populations.

METHODS

Search Strategy

We searched PubMed, Embase, and Cochrane databases for published English language studies of chronic, untreated HBV infection. We used search terms related to chronic HBV and outcomes, such as e antigen loss, e antigen seroconversion, cirrhosis, hepatocellular carcinoma, and mortality. We recognized that many cohort studies of people with chronic HBV infection may have not included HBeAg loss as a primary outcome but still measured and reported this event. Therefore, we used an expansive search strategy, which included terms associated with outcomes of cirrhosis, HCC, and associated mortality rates (see Supplementary Materials). This study was registered under PROSPERO (registration no. CRD42019122510).

Selection Criteria

We restricted our population to individuals with chronic HBV infection (presence of HBsAg for ≥6 months). We included prospective, observational cohort studies with community-based and healthcare facility (hospital or clinic)–based sampling. We excluded retrospective studies, randomized controlled trials, case-control studies, and cross-sectional studies. We also excluded studies in which participants had received antiviral or interferon therapy and studies in special subpopulations, such as patients with acute HBV infection or HIV or HCV coinfection, those receiving hemodialysis, those being treated for HCC, and liver transplant recipients.

Data Extraction

We screened titles and abstracts for inclusion using a screening tool (Supplementary Figures 1 and 2). Two authors (A. M. M. and A. F. L.) independently performed full-text review and data extraction of the following variables: authors, journal, year of publication, setting (country or countries), duration of follow-up, and method of outcome ascertainment. We recorded the study’s stated objectives and whether ascertaining the incidence rate of HBeAg loss was a prespecified study objective. Cohort data included participants’ age, sex, race, HBV genotype, and incidence of cirrhosis and HCC. The primary outcome was the annual rate of HBeAg loss. We included the proportion of HBeAg-negative patients who subsequently had HBeAg seroreversion, if reported. We extracted alanine aminotransferase (ALT) and HBV DNA values before and after HBeAg loss from studies that measured and reported them.

Statistical Analysis

We used a random-effects model to calculate a pooled annual rate of HBeAg loss among participants who were HBeAg positive at the start of the study. For studies that directly reported this metric, we used their reported value in analysis. For studies that did not specifically report this metric, we estimated the annual rate of HBeAg loss with the rate defined as events divided by total follow-up time (person-years [PYs] of observation), where events was defined as the number of participants who had HBeAg loss during that time. This approach has been used in a previously published meta-analysis [16]. Where the total follow-up time was not reported for HBeAg-positive patients, we estimated this parameter by multiplying the mean follow-up time by the number of HBeAg-positive patients at study start. Given the potential risk for bias, we estimated the annual rate of HBeAg loss for (1) all studies (n = 26), (2) studies that reported HBeAg loss as a prespecified end point (n = 17), and (3) studies that directly reported the incidence rate of HBeAg loss (n = 7). We used log transformation to estimate pooled HBeAg loss rates.

We conducted a meta-regression using the following variables: mean age in cohort, percentage of female participants in cohort, and geographic region (Africa, Eastern Mediterranean, Europe, North America, and Western Pacific). No studies from South America met the inclusion criteria. We performed separate subgroup analyses based on geographic region (Africa, Eastern Mediterranean, Europe, North America, or Western Pacific).

We assessed the risk of bias with a uniform instrument adapted from the Newcastle-Ottawa scale for assessing quality of studies in meta-analyses (Supplementary Table 1) [17]. We evaluated the statistical heterogeneity of HBeAg loss rates, using the I2 test and the Q coefficient, a weighted sum of squared deviations between individual effect estimates and the pooled effect across studies [18, 19]. We examined publication bias using a funnel plot and the Egger test. For calculations, we used R software, version 4.0.2 (2020), using the META and METAFOR statistical packages.

RESULTS

Our search yielded 10 560 unique publications (Figure 1). After screening of titles and abstracts, 196 publications remained. After full-text review, we included a total of 26 unique patient cohorts in the meta-analysis, with 7550 participants having HBeAg-positive chronic HBV infection (Table 1) [20–45]. Healthcare facility–based recruitment accounted for 5468 participants (72.4%; 18 studies), community-based recruitment for 896 (11.9%; 4 studies), mixed recruitment for 1186 (15.7%; 4 studies). Seventeen studies representing 4689 participants (62.1%) specified HBeAg loss as an outcome in the objectives of their analysis. Studies were located in 5 WHO regions: 5561 participants (73.7%) in 15 studies in the Western Pacific, 1504 (19.9%) in 5 studies in North America, 173 (2.3%) in 4 studies in Europe, 139 (1.8%) in 1 study in the Eastern Mediterranean, and 173 (2.3%) in 1 study in Africa. Genotype testing was conducted in 2661 participants (36.2%) among 10 studies; genotypes B and C accounted for 72% of participants in studies that reported genotype testing by HBeAg status.

Flow chart of article selection. Abbreviations: HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus.
Figure 1.

Flow chart of article selection. Abbreviations: HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus.

Open in new tabDownload slide
Table 1.
Open in new tab

Characteristics of Included Prospective Observational Studies Evaluating Spontaneous Hepatitis B e Antigen Loss

Study Authors (Year)RegionNo. in CohortaAge at Enrollment, Mean, yFemale Sex, %GenotypesYear of Initial EnrollmentFollow-up, PYsHBeAg Loss Rate per 100 PYs
Tsuchie et al (1984) [42]Western Pacific1393636NR19708343.6
Dragosics (1987) [29]Europe153315NR19705311.4
Kim et al (1987) [34]North America573720NR19795388.0
Liaw et al (1988) [23]Western Pacific5093113NR1979145112.9
Ruiz-Moreno (1989) [37]Europe77639NR197923113.9
Chan et al (1994) [45]Western Pacific1110NRNR19818883.5
Chang et al (1995) [22]Western Pacific365439NR198825925.4
Bozkaya et al (1996) [20]Western Pacific332744NRNR13510.4
Evans et al (1997) [30]North America4542139NR198514827.3
Fujisawa et al (2000) [32]Western Pacific36747NR19764286.1
Yuen et al (2003) [43]Western Pacific12153836NR1976293616.8
Zacharakis et al (2005) [44]Europe143439NR19926910.0
Ni et al (2004) [35]Western Pacific380641B, C, B + C199156183.9
Livingston et al (2007) [25]North America5072442A, B, C, D, F1983105464.3
Chan et al (2010) [21]Western Pacific492837C19974046.2
Tai et al (2010) [39]Western Pacific13862953NR1980104315.9
Tseng et al (2011) [41]Western Pacific185646B, C198436393.3
Yang et al (2012) [28]Western Pacific4394031B, C199131615.9
Roushan et al (2012) [36]Eastern Mediterranean139155NR199025303.2
Chiu et al (2014) [26]Western Pacific331841B, C, B + C197768193.8
Ieluzzi et al (2014) [31]Europe673024NR197236914.4
Lim et al (2015) [33]Western Pacific2011134NR198430516.0
Tong et al (2015) [40]North America953526A, B, C19899185.0
Shimakawa et al (2016) [38]Africa1731150A, E197419127.4
Chen et al (2019) [24]Western Pacific1821142B, C, B + C198636042.4
Brahmania et al (2019) [27]North America3913552A, B, C, D201124017.9
Study Authors (Year)RegionNo. in CohortaAge at Enrollment, Mean, yFemale Sex, %GenotypesYear of Initial EnrollmentFollow-up, PYsHBeAg Loss Rate per 100 PYs
Tsuchie et al (1984) [42]Western Pacific1393636NR19708343.6
Dragosics (1987) [29]Europe153315NR19705311.4
Kim et al (1987) [34]North America573720NR19795388.0
Liaw et al (1988) [23]Western Pacific5093113NR1979145112.9
Ruiz-Moreno (1989) [37]Europe77639NR197923113.9
Chan et al (1994) [45]Western Pacific1110NRNR19818883.5
Chang et al (1995) [22]Western Pacific365439NR198825925.4
Bozkaya et al (1996) [20]Western Pacific332744NRNR13510.4
Evans et al (1997) [30]North America4542139NR198514827.3
Fujisawa et al (2000) [32]Western Pacific36747NR19764286.1
Yuen et al (2003) [43]Western Pacific12153836NR1976293616.8
Zacharakis et al (2005) [44]Europe143439NR19926910.0
Ni et al (2004) [35]Western Pacific380641B, C, B + C199156183.9
Livingston et al (2007) [25]North America5072442A, B, C, D, F1983105464.3
Chan et al (2010) [21]Western Pacific492837C19974046.2
Tai et al (2010) [39]Western Pacific13862953NR1980104315.9
Tseng et al (2011) [41]Western Pacific185646B, C198436393.3
Yang et al (2012) [28]Western Pacific4394031B, C199131615.9
Roushan et al (2012) [36]Eastern Mediterranean139155NR199025303.2
Chiu et al (2014) [26]Western Pacific331841B, C, B + C197768193.8
Ieluzzi et al (2014) [31]Europe673024NR197236914.4
Lim et al (2015) [33]Western Pacific2011134NR198430516.0
Tong et al (2015) [40]North America953526A, B, C19899185.0
Shimakawa et al (2016) [38]Africa1731150A, E197419127.4
Chen et al (2019) [24]Western Pacific1821142B, C, B + C198636042.4
Brahmania et al (2019) [27]North America3913552A, B, C, D201124017.9

Abbreviations: HBeAg, hepatitis B virus e antigen; NR, not reported; PYs, person-years.

a

Number of HBeAg-positive participants at cohort start.

Table 1.
Open in new tab

Characteristics of Included Prospective Observational Studies Evaluating Spontaneous Hepatitis B e Antigen Loss

Study Authors (Year)RegionNo. in CohortaAge at Enrollment, Mean, yFemale Sex, %GenotypesYear of Initial EnrollmentFollow-up, PYsHBeAg Loss Rate per 100 PYs
Tsuchie et al (1984) [42]Western Pacific1393636NR19708343.6
Dragosics (1987) [29]Europe153315NR19705311.4
Kim et al (1987) [34]North America573720NR19795388.0
Liaw et al (1988) [23]Western Pacific5093113NR1979145112.9
Ruiz-Moreno (1989) [37]Europe77639NR197923113.9
Chan et al (1994) [45]Western Pacific1110NRNR19818883.5
Chang et al (1995) [22]Western Pacific365439NR198825925.4
Bozkaya et al (1996) [20]Western Pacific332744NRNR13510.4
Evans et al (1997) [30]North America4542139NR198514827.3
Fujisawa et al (2000) [32]Western Pacific36747NR19764286.1
Yuen et al (2003) [43]Western Pacific12153836NR1976293616.8
Zacharakis et al (2005) [44]Europe143439NR19926910.0
Ni et al (2004) [35]Western Pacific380641B, C, B + C199156183.9
Livingston et al (2007) [25]North America5072442A, B, C, D, F1983105464.3
Chan et al (2010) [21]Western Pacific492837C19974046.2
Tai et al (2010) [39]Western Pacific13862953NR1980104315.9
Tseng et al (2011) [41]Western Pacific185646B, C198436393.3
Yang et al (2012) [28]Western Pacific4394031B, C199131615.9
Roushan et al (2012) [36]Eastern Mediterranean139155NR199025303.2
Chiu et al (2014) [26]Western Pacific331841B, C, B + C197768193.8
Ieluzzi et al (2014) [31]Europe673024NR197236914.4
Lim et al (2015) [33]Western Pacific2011134NR198430516.0
Tong et al (2015) [40]North America953526A, B, C19899185.0
Shimakawa et al (2016) [38]Africa1731150A, E197419127.4
Chen et al (2019) [24]Western Pacific1821142B, C, B + C198636042.4
Brahmania et al (2019) [27]North America3913552A, B, C, D201124017.9
Study Authors (Year)RegionNo. in CohortaAge at Enrollment, Mean, yFemale Sex, %GenotypesYear of Initial EnrollmentFollow-up, PYsHBeAg Loss Rate per 100 PYs
Tsuchie et al (1984) [42]Western Pacific1393636NR19708343.6
Dragosics (1987) [29]Europe153315NR19705311.4
Kim et al (1987) [34]North America573720NR19795388.0
Liaw et al (1988) [23]Western Pacific5093113NR1979145112.9
Ruiz-Moreno (1989) [37]Europe77639NR197923113.9
Chan et al (1994) [45]Western Pacific1110NRNR19818883.5
Chang et al (1995) [22]Western Pacific365439NR198825925.4
Bozkaya et al (1996) [20]Western Pacific332744NRNR13510.4
Evans et al (1997) [30]North America4542139NR198514827.3
Fujisawa et al (2000) [32]Western Pacific36747NR19764286.1
Yuen et al (2003) [43]Western Pacific12153836NR1976293616.8
Zacharakis et al (2005) [44]Europe143439NR19926910.0
Ni et al (2004) [35]Western Pacific380641B, C, B + C199156183.9
Livingston et al (2007) [25]North America5072442A, B, C, D, F1983105464.3
Chan et al (2010) [21]Western Pacific492837C19974046.2
Tai et al (2010) [39]Western Pacific13862953NR1980104315.9
Tseng et al (2011) [41]Western Pacific185646B, C198436393.3
Yang et al (2012) [28]Western Pacific4394031B, C199131615.9
Roushan et al (2012) [36]Eastern Mediterranean139155NR199025303.2
Chiu et al (2014) [26]Western Pacific331841B, C, B + C197768193.8
Ieluzzi et al (2014) [31]Europe673024NR197236914.4
Lim et al (2015) [33]Western Pacific2011134NR198430516.0
Tong et al (2015) [40]North America953526A, B, C19899185.0
Shimakawa et al (2016) [38]Africa1731150A, E197419127.4
Chen et al (2019) [24]Western Pacific1821142B, C, B + C198636042.4
Brahmania et al (2019) [27]North America3913552A, B, C, D201124017.9

Abbreviations: HBeAg, hepatitis B virus e antigen; NR, not reported; PYs, person-years.

a

Number of HBeAg-positive participants at cohort start.

During a total of 64 877 PYs of follow-up, 3633 persons experienced loss of HBeAg. The pooled rate of HBeAg loss in all studies was 6.46/100 PYs (95% confidence interval [CI], 5.17–8.08; Figure 2). There was a high level of heterogeneity in this estimate (I2 = 98%). In 17 studies where HBeAg loss was prespecified as a study outcome, pooled HBeAg loss rate was 5.45/100 PYs (95% CI, 4.04–7.34; I2 = 98%) (Supplementary Figure 3). In 7 studies that directly reported the rate of HBeAg loss for cohort participants, the pooled HBeAg loss rate was 5.75/100 PYs (95% CI, 4.80–6.87; I2 = 87%).

Pooled incidence rate of hepatitis B e antigen (HBeAg) loss (per 100 person-years) in untreated persons with chronic hepatitis B virus [20–45]. Annual rates were combined using a random-effects model with log transformation. The vertical dashed line represents the pooled rate. Heterogeneity parameters are shown with χ2 degrees of freedom and I2 statistic. Abbreviation: CI, confidence interval.
Figure 2.

Pooled incidence rate of hepatitis B e antigen (HBeAg) loss (per 100 person-years) in untreated persons with chronic hepatitis B virus [20–45]. Annual rates were combined using a random-effects model with log transformation. The vertical dashed line represents the pooled rate. Heterogeneity parameters are shown with χ2 degrees of freedom and I2 statistic. Abbreviation: CI, confidence interval.

Open in new tabDownload slide

After excluding 1 study that did not report participants’ sex [45], we conducted a meta-regression using mean age, percentage of female participants, and geographic region (Table 2). Older age was associated with higher rates of HBeAg loss; for each additional year in the mean age of the cohort, the rate of HBeAg loss increased by 0.019/100 PYs (95% CI, .002–.037). Compared with those in the Western Pacific, participants in Europe had a higher rate of HBeAg loss (coefficient, 0.726/100 PYs [95% CI, .141–1.131]). In a univariable model, female sex was associated with a lower HBeAg loss rate (coefficient, −1.74/100 PYs [95% CI, −3.748 to .259]), but this difference was not statistically significant, and the effect was not evident in the multivariable model.

Table 2.
Open in new tab

Meta-Regression Analysis

Single CovariateMultiple Covariates
VariableCoefficient (95% CI)P ValueCoefficient (95% CI)P Value
Age (per year)0.022 (.008–.035).0020.019 (.002–.037).03
Sex (% female)a−1.744 (−3.748 to .259).090.019 (−2.112 to 2.151).99
Region
 Western PacificReference…Reference…
 Africa0.295 (−.901 to 1.491).530.438 (−.498 to 1.374).36
 Eastern Mediterranean−0.534 (−1.738 to .671).38−0.206 (−1.199 to .788).68
 Europe0.823 (.119–1.528).020.726 (.141–1.311).02
 North America0.286 (−.320 to 0.891).360.055 (−.439 to .549).83
Single CovariateMultiple Covariates
VariableCoefficient (95% CI)P ValueCoefficient (95% CI)P Value
Age (per year)0.022 (.008–.035).0020.019 (.002–.037).03
Sex (% female)a−1.744 (−3.748 to .259).090.019 (−2.112 to 2.151).99
Region
 Western PacificReference…Reference…
 Africa0.295 (−.901 to 1.491).530.438 (−.498 to 1.374).36
 Eastern Mediterranean−0.534 (−1.738 to .671).38−0.206 (−1.199 to .788).68
 Europe0.823 (.119–1.528).020.726 (.141–1.311).02
 North America0.286 (−.320 to 0.891).360.055 (−.439 to .549).83

Abbreviation: CI, confidence interval.

a

One study (Chan et al [45]) was excluded from the analysis of sex and from the combined model because it did not report the sex of its cohort participants.

Table 2.
Open in new tab

Meta-Regression Analysis

Single CovariateMultiple Covariates
VariableCoefficient (95% CI)P ValueCoefficient (95% CI)P Value
Age (per year)0.022 (.008–.035).0020.019 (.002–.037).03
Sex (% female)a−1.744 (−3.748 to .259).090.019 (−2.112 to 2.151).99
Region
 Western PacificReference…Reference…
 Africa0.295 (−.901 to 1.491).530.438 (−.498 to 1.374).36
 Eastern Mediterranean−0.534 (−1.738 to .671).38−0.206 (−1.199 to .788).68
 Europe0.823 (.119–1.528).020.726 (.141–1.311).02
 North America0.286 (−.320 to 0.891).360.055 (−.439 to .549).83
Single CovariateMultiple Covariates
VariableCoefficient (95% CI)P ValueCoefficient (95% CI)P Value
Age (per year)0.022 (.008–.035).0020.019 (.002–.037).03
Sex (% female)a−1.744 (−3.748 to .259).090.019 (−2.112 to 2.151).99
Region
 Western PacificReference…Reference…
 Africa0.295 (−.901 to 1.491).530.438 (−.498 to 1.374).36
 Eastern Mediterranean−0.534 (−1.738 to .671).38−0.206 (−1.199 to .788).68
 Europe0.823 (.119–1.528).020.726 (.141–1.311).02
 North America0.286 (−.320 to 0.891).360.055 (−.439 to .549).83

Abbreviation: CI, confidence interval.

a

One study (Chan et al [45]) was excluded from the analysis of sex and from the combined model because it did not report the sex of its cohort participants.

Region-specific rates of HBeAg loss were 5.53/100 PYs (95% CI, 4.05–7.55; I2 = 98%) in the Western Pacific; 7.34/100 PYs (4.61–11.70; I2 = 96%) in North America, and 13.67/100 PYs (11.21–16.66; I2 = 0%) in Europe (Figure 3). The sole study in the Eastern Mediterranean region reported an HBeAg loss rate of 3.24/100 PYs (95% CI, 2.61–4.02) [36], and the sole study in Africa reported a loss rate of 7.43/100 PYs (6.30–8.75) [38]. Among 10 studies reporting HBV genotypes, 4 showed that genotype C was associated with a lower rate of HBeAg loss than other genotypes [24, 25, 35, 41]. In these different cohorts, participants with HBV genotype B or B + C were 2–3 times more likely to have HBeAg loss than those with genotype C [24, 25, 35, 41]. One study with a variety of HBV genotypes also noted that participants with genotype C or F had higher rates of HBeAg reversion after HBeAg loss than genotypes A, B, or D [25].

Pooled rate of hepatitis B virus e antigen (HBeAg) loss (per 100 person-years) in untreated persons with chronic hepatitis B virus, stratified by region [20–45]. Heterogeneity parameters for each subgroup are shown with χ2 degrees of freedom and I2 statistic. Test for difference between subgroups showed Q = 95.22 and P < .001. Abbreviation: CI, confidence interval.
Figure 3.

Pooled rate of hepatitis B virus e antigen (HBeAg) loss (per 100 person-years) in untreated persons with chronic hepatitis B virus, stratified by region [20–45]. Heterogeneity parameters for each subgroup are shown with χ2 degrees of freedom and I2 statistic. Test for difference between subgroups showed Q = 95.22 and P < .001. Abbreviation: CI, confidence interval.

Open in new tabDownload slide

Only 1 study enrolled participants at the time of perinatal diagnosis of HBV [45]. In that study of 111 infants with HBV, 31 lost HBeAg within the first 8 years of life, yielding a rate of HbeAg loss of 3.49/100 PYs (95% CI, 2.46–4.96). Three studies evaluated age-specific rates of HbeAg loss within their cohorts [30, 33, 36]. One study found no association between age and HBeAg loss after adjustment for sex and ethnicity [33]. The other 2 found that participants in the second and third decades of life had 1.5–2.8 times higher rates of HBeAg loss compared with earlier in life [30, 36].

Several studies evaluated factors associated with HBeAg loss. Three studies that enrolled children found that participants whose mothers were HBeAg positive had lower rates of HBeAg loss [26, 36, 41]. Seven studies demonstrated that elevated aminotransferase levels were associated with spontaneous HBeAg loss [20, 22, 23, 27, 28, 45]. For example, 1 study showed that participants with ALT levels of 15–45 or >45 IU/L had rates of HBeAg loss that were 1.59 (95% CI, 1.09–2.32) or 2.96 (1.81–4.83) times higher, respectively, than those in participants with ALT levels <15 IU/L [28]. Several studies demonstrated a decline in HBV DNA associated with loss of HBeAg [20, 22, 43, 45]. We were unable to conduct a pooled analysis on the ALT or HBV DNA changes associated with HbeAg loss, owing to inconsistent reporting of these measures across studies.

The I2 statistic for the pooled HBeAg loss rate was 98%, indicating a high level of heterogeneity. This remained high (98%) even when we restricted the analysis to studies that prespecified HBeAg loss as a study objective. Results from meta-regression analysis of age, sex, and geographic region showed a high level of residual heterogeneity (I2 = 96%). The quality of studies varied depending on participant selection, study objective, and frequency of HBeAg testing during follow-up (Supplementary Figure 4). Only 7 studies directly reported HBeAg loss rate for their cohort (Q8 column in Supplementary Figure 4). We did not find evidence of publication bias using Egger test for funnel plot asymmetry (P = .22; Supplementary Figure 5).

DISCUSSION

In this systematic review and meta-analysis of prospective observational studies of untreated chronic HBV infection, we found that the overall pooled rate of spontaneous HBeAg loss was 6.46 per 100 PYs (95% CI, 5.17–8.08). Loss of HBeAg was associated with a decline in HBV DNA and improved long-term outcomes [6, 9]. In contrast, people with persistently positive HBeAg have higher incidences of cirrhosis and HCC and higher mortality rates [9–12]. Furthermore, pregnant women with HBeAg are more likely to transmit HBV infection to their infants [13, 14]. Our findings enhance the understanding of the natural history of chronic HBV infection and demonstrate that, while the loss of HBeAg occurred at a low rate, it occurred more frequently than other important milestones of chronic HBV infection, such as loss of HBsAg [16, 46].

Therapeutic goals in the management of chronic HBV infection are evolving and new definitions of “cure” are being considered [5, 47]. In most patients, a “sterilizing” cure with the loss of covalently closed circular DNA remains elusive and difficult to measure [47]. A functional cure, with loss of HBsAg, is more feasible to obtain and measure; however, this occurs rarely (at a rate of approximately 1% per year) and is typically preceded by HBeAg loss [5, 16, 46]. Therefore, for patients in the HBeAg-positive phases of chronic HBV, the most immediate goal is loss of HBeAg, suppression of HBV DNA, and normalization of aminotransferase levels. An earlier meta-analysis suggests that HBeAg-positive patients treated with antiviral therapy may not have higher rates of HBeAg loss than untreated patients [48].

Treatment decisions for people in the HBeAg-positive phase of disease are complex and must balance the risks of long-term antiviral therapy with the risks of continued viral replication and hepatitis in untreated chronic HBV infection. Our study provides region-specific estimates of the probability of HBeAg loss without antiviral therapy. More data are needed regarding how long to monitor HBeAg-positive disease before initiating treatment, particularly because treatment decisions must also incorporate other risk factors for progressive liver disease, adherence to antiviral therapy, and patient preferences. HBV biomarkers may offer additional data for clinicians to predict HBeAg loss and assess HBV disease activity before starting therapy. For example, preliminary studies suggest that HBV core-related antigen and HBV RNA may be predictive of HBeAg loss. Still, larger cohorts are needed to verify these findings, and the current availability of these assays is limited [47, 49].

Our analysis included studies across 5 WHO regions. The pooled rate of HBeAg loss is lower in studies from the Western Pacific region (5.53/100 PYs [95% CI, 4.05–7.55) than in those from Europe (13.67/100 PYs [11.21–16.66]), and the difference between these regions was significant in a meta-regression. Geographic differences may reflect differences in HBV genotypes and/or route of transmission (ie, mother-to-child versus horizontal transmission); several studies noted that genotype C, which is prevalent in the Western Pacific region, is associated with delayed HBeAg loss compared with other genotypes [24, 25, 35, 41]. One study analyzing 507 participants of Alaskan native origin with a distribution of genotypes A, B, C, D, and F found that participants with genotype C had a lower rate of HBeAg loss and, along with genotype F, a higher rate of HBeAg reversion compared with other genotypes [25]. Collectively, these data demonstrate that persons with genotype C are more likely to retain HBeAg later into adulthood and, therefore, may have a higher risk of liver-related complications and mother-to-child transmission. While the benefit of HBV antiviral therapy in HBeAg-positive patients remains the subject of ongoing investigation [6, 15], our study suggests that patients with genotype C may need to be prioritized for treatment as their rate of spontaneous HBeAg loss may be lower than in other populations.

We found only 1 study analyzing spontaneous HBeAg loss in sub-Saharan Africa, which was based on community serosurveys in the Gambia, where genotype E was the predominant circulating strain [38]. That study followed up 173 participants for a total of 1912 PYs and found an HBeAg loss rate of 7.43/100 PYs (95% CI, 6.30–8.75). The incidence of HCC and the HCC-associated mortality rate remain high in sub-Saharan Africa [38, 50]. An improved understanding of the natural history of HBV and treatment strategies to reduce the incidence of HCC in sub-Saharan Africa is urgently needed because the clinical burden of chronic HBV infection is increasing across the continent, where it also intersects with the HIV epidemic [50]. The rising incidence of HBV-related complications in Africa can be addressed in a number of ways, including strategies of birth dose immunization, early screening, and expanded treatment [2, 3].

Before this review, individual cohort studies led to mixed conclusions regarding age and HBeAg loss. Two studies noted a higher rate of HBeAg loss in the second and third decades of life compared with early childhood [30, 36]. A third study did not find that age was associated with varying rates of HBeAg loss after adjusting for sex and ethnicity, which highlights the variability of HBV natural history across populations with epidemiological or genotype-related differences [33]. Our study clarifies the relationship between age and HBeAg loss: meta-regression across all studies showed that older age was associated with higher HBeAg loss rates (for each additional year in the mean age of cohort, HBeAg loss rate increased by 0.019/100 PYs [95% CI, 0.2–3.7]).

Several studies demonstrated that elevations in aminotransferases are associated with, and typically precede, HBeAg loss [20, 22, 25, 27, 28, 45]. This finding is consistent with the understanding that immune-mediated clearance of HBeAg frequently results in hepatic inflammation during the “immune active” or “immune clearance” phase [5, 27]. People who experience this immune response without sustained HBeAg loss are at high risk for cirrhosis and HCC [9, 22]. Several studies in our review demonstrated that a decrease in HBV DNA levels is associated with sustained HBeAg loss, which may be indicative of decreasing viral activity in hepatocytes. This likely explains the association between loss of HBeAg and decreased incidence of cirrhosis and HCC in long-term follow-up.

The findings of our study are subject to several limitations. To obtain a thorough representation of studies without introducing selection bias, we used an expansive search strategy and narrow inclusion criteria that included only prospective observational studies. However, our sample lacked adequate representation of many HBV-endemic countries, particularly in sub-Saharan Africa, Central Asia, South-East Asia, and South America. Our estimated incidence rate of HBeAg loss in the sub-Saharan Africa and Eastern Mediterranean regions should be interpreted with caution. More studies are needed regarding HBV natural history and response to treatment in these regions. Our estimated HBeAg loss rate is met with a high level of heterogeneity, even after accounting for age, sex, and geographic region. Residual heterogeneity in our findings may be explained by unmeasured differences in populations, such as environmental exposures, immunological differences, and genetic mutations in HBV. In addition, only 7 studies directly reported the rate of HBeAg loss in their cohorts; for the remainder of studies, we estimated the rate of HBeAg loss using a previously reported method [16]. Although this method could introduce bias if the rates of loss-to-follow-up and events were not constant over time, the pooled HBeAg loss rate in these 7 studies (5.75/100 PYs) was similar to the pooled rate in studies that reported HBeAg loss as a prespecified study outcome (5.45/100 PYs) and all studies (6.46/100 PYs).

Inconsistent reporting of ALT, HBV DNA, and genetic mutations across studies prevented us from conducting pooled analyses on these important variables. Specifically, limited reporting on HBV genotypes meant that we could not quantify its effects on HBeAg loss. Finally, only 5 studies tested for HBeAg reversion [20, 25, 30, 38, 43], which may limit the precision and accuracy of our estimated pooled HBeAg loss rate and the observations related to ALT and HBV DNA changes around the time of HBeAg loss. Reporting HBeAg reversion and genotype-specific outcomes in observational cohorts and clinical trials of novel therapeutic agents could reduce bias in this estimate.

In conclusion, we found that 3633 participants (48.1%) experienced spontaneous loss of HBeAg during a total of 64 877 PYs of follow-up in this systematic review and meta-analysis, corresponding to an estimated rate of HbeAg loss of 6.46/100 PYs. Differences in the rate of HBeAg loss across age groups and geographic regions may be due to immunological, epidemiological, or HBV genotype–related differences and may have implications for future HBV treatment guidelines.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org/). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.

Notes

Author contributions. All authors had access to the study data and reviewed and approved the final manuscript.

Disclaimer. The funders had no role in study design, data collection, analysis, interpretation, or writing of the report. The article contents are solely the responsibility of the authors and do not necessarily represent the official views of the funders.

Financial support. This work was supported by the National Institutes of Health (grants T32AI007433 to A. M. M., R37AI058736 to K. A. F., and K01HL123349 to E. P. H.) and the Charles A. King Trust (award to A. M. M.). E. P. H. is also a Jerome and Celia Reich Endowed Scholar in HIV/AIDS Research at Massachusetts General Hospital.

Data availability

All data used in this study are available on reasonable request from the corresponding author.

References

1

World Health Organization
.
Global health sector strategy on viral hepatitis, 2016–2021. 2016.
 https://apps.who.int/iris/bitstream/handle/10665/246177/WHO-HIV-2016.06-eng.pdf. Accessed 20 August 2020.

2

Thomas
 
DL
.
Global elimination of chronic hepatitis
.
N Engl J Med
 
2019
;
380
:
2041
–
50
.

Google Scholar

Crossref
Search ADS
PubMed

 
3

Cooke
 
GS
,
Andrieux-Meyer
 
I
,
Applegate
 
TL
, et al.  
Accelerating the elimination of viral hepatitis: a Lancet Gastroenterology & Hepatology Commission
.
Lancet Gastroenterol Hepatol
 
2019
;
4
:
135
–
84
.

Google Scholar

Crossref
Search ADS
PubMed

 
4

Terrault
 
NA
,
Lok
 
ASF
,
McMahon
 
BJ
, et al.  
Update on prevention, diagnosis, and treatment of chronic hepatitis B: AASLD 2018 hepatitis B guidance
.
Hepatology
 
2018
;
67
:
1560
–
99
.

Google Scholar

Crossref
Search ADS
PubMed

 
5

Likhitsup
 
A
,
Lok
 
AS
.
Understanding the natural history of hepatitis B virus infection and the new definitions of cure and the endpoints of clinical trials
.
Clin Liver Dis
 
2019
;
23
:
401
–
16
.

Google Scholar

Crossref
Search ADS
PubMed

 
6

Liaw
 
YF
.
HBeAg seroconversion as an important end point in the treatment of chronic hepatitis B
.
Hepatol Int
 
2009
;
3
:
425
–
33
.

Google Scholar

Crossref
Search ADS
PubMed

 
7

Yuen
 
MF
,
Tanaka
 
Y
,
Ng
 
IO
, et al.  
Hepatic necroinflammation and fibrosis in patients with genotypes Ba and C, core-promoter and precore mutations
.
J Viral Hepat
 
2005
;
12
:
513
.

Google Scholar

Crossref
Search ADS
PubMed

 
8

Chou
 
YC
,
Yu
 
MW
,
Wu
 
CF
, et al.  
Temporal relationship between hepatitis B virus enhancer II/basal core promoter sequence variation and risk of hepatocellular carcinoma
.
Gut
 
2008
;
57
:
91
.

Google Scholar

Crossref
Search ADS
PubMed

 
9

Yang
 
HI
,
Lu
 
SN
,
Liaw
 
YF
, et al.  
Hepatitis B e-antigen and the risk of hepatocellular carcinoma
.
N Engl J Med
 
2002
;
347
:
168
–
74
.

Google Scholar

Crossref
Search ADS
PubMed

 
10

Chen
 
CJ
.
Risk of hepatocellular carcinoma across a biological gradient of serum hepatitis B virus DNA Level
.
JAMA
 
2006
;
295
:
65
.

Google Scholar

Crossref
Search ADS
PubMed

 
11

Iloeje
 
UH
,
Yang
 
H
,
Su
 
J
, et al.  
Predicting cirrhosis risk based on the level of circulating hepatitis B viral load
.
Gastroenterology
 
2006
;
130
:
678
–
86
.

Google Scholar

Crossref
Search ADS
PubMed

 
12

Iloeje
 
UH
,
Yang
 
H
,
Jen
 
C
, et al.  
Risk and predictors of mortality associated with chronic hepatitis B infection
.
Clin Gastroenterol Hepatol
 
2007
;
5
:
921
–
31
.

Google Scholar

Crossref
Search ADS
PubMed

 
13

Keane
 
E
,
Funk
 
AL
,
Shimakawa
 
Y
.
Systematic review with meta-analysis: the risk of mother-to-child transmission of hepatitis B virus infection in sub-Saharan Africa
.
Aliment Pharmacol Ther
 
2016
;
44
:
1005
–
17
.

Google Scholar

Crossref
Search ADS
PubMed

 
14

Boucheron
 
P
,
Lu
 
Y
,
Yoshida
 
K
, et al.  
Accuracy of HBeAg to identify pregnant women at risk of transmitting hepatitis B virus to their neonates: a systematic review and meta-analysis
.
Lancet Infect Dis
 
2021
;
21
:
85
–
96
.

Google Scholar

Crossref
Search ADS
PubMed

 
15

Kennedy
 
P
,
Litwin
 
S
,
Dolman
 
G
, et al.  
Immune tolerant chronic hepatitis B: the unrecognized risks
.
Viruses
 
2017
;
9
:
96
.

Google Scholar

Crossref
Search ADS
PubMed

 
16

Yeo
 
YH
,
Ho
 
HJ
,
Yang
 
HI
, et al.  
Factors associated with rates of HBsAg seroclearance in adults with chronic HBV infection: a systematic review and meta-analysis
.
Gastroenterology
 
2019
;
156
:
635
–
46.e9
.

Google Scholar

Crossref
Search ADS
PubMed

 
17

Wells
 
G
,
Shea
 
B
,
O’Connell
 
D
, et al.  
The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses
.
2018
. http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed 10 December 2021.

18

Cochran
 
WG
.
The combination of estimates from different experiments
.
Biometrics
 
1954
;
10
:
101
.

Google Scholar

Crossref
Search ADS

 
19

Higgins
 
JPT
,
Thompson
 
SG
.
Quantifying heterogeneity in a meta-analysis
.
Stat Med
 
2002
;
21
:
1539
–
58
.

Google Scholar

Crossref
Search ADS
PubMed

 
20

Bozkaya
 
H
,
Ayola
 
B
,
Lok
 
ASF
.
High rate of mutations in the hepatitis B core gene during the immune clearance phase of chronic hepatitis B virus infection
.
Hepatology
 
1996
;
24
:
32
–
37
.

Google Scholar

Crossref
Search ADS
PubMed

 
21

Chan
 
HLY
,
Wong
 
VWS
,
Wong
 
GLH
, et al.  
A longitudinal study on the natural history of serum hepatitis B surface antigen changes in chronic hepatitis B
.
Hepatology
 
2010
;
52
:
1232
–
41
.

Google Scholar

Crossref
Search ADS
PubMed

 
22

Chang
 
MH
,
Hsu
 
HY
,
Hsu
 
HC
, et al.  
The significance of spontaneous hepatitis B e antigen seroconversion in childhood: with special emphasis on the clearance of hepatitis B e antigen before 3 years of age
.
Hepatology
 
1995
;
22
:
1387
–
92
.

Google Scholar

PubMed
OpenURL Placeholder Text

 
23

Liaw
 
YF
,
Tai
 
DI
,
Chu
 
CM
, et al.  
The development of cirrhosis in patients with chronic type B hepatitis: a prospective study
.
Hepatology
 
1988
;
8
:
493
–
6
.

Google Scholar

Crossref
Search ADS
PubMed

 
24

Chen
 
H
,
Wu
 
J
,
Su
 
T
, et al.  
Baseline level of hepatitis B core antibody predicts spontaneous hepatitis B e antigen (HBeAg) seroconversion in HBeAg-positive children with a normal alanine aminotransferase level
.
Hepatology
 
2019
;
70
:
1903
–
12
.

Google Scholar

Crossref
Search ADS
PubMed

 
25

Livingston
 
SE
,
Simonetti
 
JP
,
Bulkow
 
LR
, et al.  
Clearance of hepatitis B e antigen in patients with chronic hepatitis B and genotypes A, B, C, D, and F
.
Gastroenterology
 
2007
;
133
:
1452
–
7
.

Google Scholar

Crossref
Search ADS
PubMed

 
26

Chiu
 
YC
,
Liao
 
SF
,
Wu
 
JF
, et al.  
Factors affecting the natural decay of hepatitis B surface antigen in children with chronic hepatitis B virus infection during long-term follow-up
.
J Pediatr
 
2014
;
165
:
767
–
72.e1
.

Google Scholar

Crossref
Search ADS
PubMed

 
27

Brahmania
 
M
,
Lombardero
 
M
,
Hansen
 
BE
, et al.  
Association between severe serum alanine aminotransferase flares and hepatitis B e antigen seroconversion and HBV DNA decrease in untreated patients with chronic HBV infection
.
Clin Gastroenterol Hepatol
 
2019
;
17
:
2541
–
51.e2
.

Google Scholar

Crossref
Search ADS
PubMed

 
28

Yang
 
H
,
Hung
 
H
,
Lee
 
M
, et al.  
Incidence and determinants of spontaneous seroclearance of hepatitis B e antigen and DNA in patients with chronic hepatitis B
.
Clin Gastroenterol Hepatol
 
2012
;
10
:
527
–
34.e2
.

Google Scholar

Crossref
Search ADS
PubMed

 
29

Dragosics
 
B
,
Ferenci
 
P
,
Hitchman
 
E
, et al.  
Long-term follow-up study of asymptomatic HBsAg-positive voluntary blood donors in Austria: a clinical and histologic evaluation of 242 cases
.
Hepatology
 
1987
;
7
:
302
–
6
.

Google Scholar

Crossref
Search ADS
PubMed

 
30

Evans
 
AA
,
Fine
 
M
,
London
 
WT
.
Spontaneous seroconversion in hepatitis B e antigen-positive chronic hepatitis B: implications for interferon therapy
.
J Infect Dis
 
1997
;
176
:
845
–
50
.

Google Scholar

Crossref
Search ADS
PubMed

 
31

Ieluzzi
 
D
,
Covolo
 
L
,
Donato
 
F
, et al.  
Progression to cirrhosis, hepatocellular carcinoma and liver-related mortality in chronic hepatitis B patients in Italy
.
Dig Liver Dis
 
2014
;
46
:
427
–
32
.

Google Scholar

Crossref
Search ADS
PubMed

 
32

Fujisawa
 
T
,
Komatsu
 
H
,
Inui
 
A
, et al.  
Long-term outcome of chronic hepatitis B in adolescents or young adults in follow-up from childhood
.
J Pediatr Gastroenterol Nutr
 
2000
;
30
:
201
–
6
.

Google Scholar

Crossref
Search ADS
PubMed

 
33

Lim
 
TH
,
Gane
 
E
,
Moyes
 
C
, et al.  
Serological and clinical outcomes of horizontally transmitted chronic hepatitis B infection in New Zealand Māori: results from a 28-year follow-up study
.
Gut
 
2015
;
64
:
966
–
72
.

Google Scholar

Crossref
Search ADS
PubMed

 
34

Kim
 
YI
,
Heathcote
 
J
,
Wanless
 
IR
.
The hepatitis B carrier state—a follow-up study of 100 consecutive cases
.
Clin Invest Med
 
1987
;
10
:
383
–
7
.

Google Scholar

PubMed
OpenURL Placeholder Text

 
35

Ni
 
YH
,
Chang
 
MH
,
Wang
 
KJ
, et al.  
Clinical relevance of hepatitis B virus genotype in children with chronic infection and hepatocellular carcinoma
.
Gastroenterology
 
2004
;
127
:
1733
–
8
.

Google Scholar

Crossref
Search ADS
PubMed

 
36

Roushan
 
MRH
,
Bijani
 
A
,
Ramzaninejad
 
S
, et al.  
HBeAg seroconversion in children infected during early childhood with hepatitis B virus
.
J Clin Virol
 
2012
;
55
:
30
–
3
.

Google Scholar

Crossref
Search ADS
PubMed

 
37

Ruiz-Moreno
 
M
,
Camps
 
T
,
Aguado
 
JG
, et al.  
Serological and histological follow up of chronic hepatitis B infection
.
Arch Dis Child
 
1989
;
64
:
1165
–
9
.

Google Scholar

Crossref
Search ADS
PubMed

 
38

Shimakawa
 
Y
,
Lemoine
 
M
,
Njai
 
HF
, et al.  
Natural history of chronic HBV infection in West Africa: a longitudinal population-based study from The Gambia
.
Gut
 
2016
;
65
:
2007
–
16
.

Google Scholar

Crossref
Search ADS
PubMed

 
39

Tai
 
DI
,
Tsay
 
PK
,
Chen
 
WT
, et al.  
Relative roles of HBsAg seroclearance and mortality in the decline of HBsAg prevalence with increasing age
.
Am J Gastroenterol
 
2010
;
105
:
1102
–
9
.

Google Scholar

Crossref
Search ADS
PubMed

 
40

Tong
 
MJ
,
Huynh
 
TT
,
Siripongsakun
 
S
, et al.  
Predicting clinical outcomes in patients with HBsAg-positive chronic hepatitis
.
Hepatol Int
 
2015
;
9
:
567
–
77
.

Google Scholar

Crossref
Search ADS
PubMed

 
41

Tseng
 
YR
,
Wu
 
JF
,
Ni
 
YH
, et al.  
Long-term effect of maternal HBeAg on delayed HBeAg seroconversion in offspring with chronic hepatitis B infection: spontaneous HBeAg seroconversion
.
Liver Int
 
2011
;
31
:
1373
–
80
.

Google Scholar

Crossref
Search ADS
PubMed

 
42

Tsuchie
 
H
,
Kurimura
 
O
,
Tamura
 
I
, et al.  
Incidences of HBeAg and anti-HBe in, and clinical course of hepatitis B virus carriers
.
Biken J
 
1984
;
27
:
169
–
76
.

Google Scholar

PubMed
OpenURL Placeholder Text

 
43

Yuen
 
MF
,
Yuan
 
HJ
,
Hui
 
CK
, et al.  
A large population study of spontaneous HBeAg seroconversion and acute exacerbation of chronic hepatitis B infection: implications for antiviral therapy
.
Gut
 
2003
;
52
:
416
–
9
.

Google Scholar

Crossref
Search ADS
PubMed

 
44

Zacharakis
 
GH
,
Koskinas
 
J
,
Kotsiou
 
S
, et al.  
Natural history of chronic HBV infection: a cohort study with up to 12 years follow-up in North Greece (part of the Interreg I-II/EC-project)
.
J Med Virol
 
2005
;
77
:
173
–
9
.

Google Scholar

Crossref
Search ADS
PubMed

 
45

Chan
 
CY
,
Lee
 
SD
,
Yu
 
MI
, et al.  
Long-term follow-up of hepatitis B virus carrier infants
.
J Med Virol
 
1994
;
44
:
336
–
9
.

Google Scholar

Crossref
Search ADS
PubMed

 
46

Zhou
 
K
,
Contag
 
C
,
Whitaker
 
E
, et al.  
Spontaneous loss of surface antigen among adults living with chronic hepatitis B virus infection: a systematic review and pooled meta-analyses
.
Lancet Gastroenterol Hepatol
 
2019
;
4
:
227
–
38
.

Google Scholar

Crossref
Search ADS
PubMed

 
47

Naggie
 
S
,
Lok
 
AS
.
New therapeutics for hepatitis B: the road to cure
.
Annu Rev Med
 
2021
;
72
:
93
–
105
.

Google Scholar

Crossref
Search ADS
PubMed

 
48

Xing
 
T
,
Xu
 
H
,
Cao
 
L
, et al.  
HBeAg seroconversion in HBeAg-positive chronic hepatitis B patients receiving long-term nucleos(t)ide analog treatment: a systematic review and network meta-analysis
.
PLoS One
 
2017
;
12
:
e0169444
.

Google Scholar

Crossref
Search ADS
PubMed

 
49

Coffin
 
CS
,
Zhou
 
K
,
Terrault
 
NA
.
New and old biomarkers for diagnosis and management of chronic hepatitis B virus infection
.
Gastroenterology
 
2019
;
156
:
355
–
68.e3
.

Google Scholar

Crossref
Search ADS
PubMed

 
50

Lemoine
 
M
,
Thursz
 
MR
.
Battlefield against hepatitis B infection and HCC in Africa
.
J Hepatol
 
2017
;
66
:
645
–
54
.

Google Scholar

Crossref
Search ADS
PubMed

 

Author notes

Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest.

© The Author(s) 2022. Published by Oxford University Press on behalf of Infectious Diseases Society of America. All rights reserved. For permissions, please e-mail: [email protected]
This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Topic:
Issue Section:
Major Article > Viruses
Download all slides
  • Supplementary data

    • Supplementary data

    jiac168_Supplementary_Data - zip file