Acute hepatitis B in the vaccination era: Predictors of functional cure and chronicity

Abstract

BACKGROUND

Despite the existence of an effective vaccine and treatment for the hepatitis B virus (HBV) more than 880000 people die each year from complications of cirrhosis and hepatocellular carcinoma associated with chronic HBV (CHB) infection. Limited information exists regarding the effect of antiviral treatment or the effect of patient background and viral factors on acute HBV outcome in a population with wide vaccine availability. Previous studies have demonstrated that widespread availability of the vaccine in developed countries has shifted the epidemiology of acute HBV infection to vulnerable subpopulations.

AIM

To evaluate the hypothesis that this epidemiologic shift in susceptibility of populations to acute HBV infection may alter the immune responses during acute infection and affect the rates of functional cure (FC) and viral persistence.

METHODS

We performed a retrospective evaluation of all patients above the age of 18 with acute HBV infection in Southern Israel between the years 2004-2022. Evaluated outcomes were FC, defined as hepatitis B surface antigen (HBsAg) clearance, or CHB infection defined as positive HBsAg at 6 months past the acute event. Clinical, biochemical, metabolic, and socioeconomic variables were analyzed.

RESULTS

Among 78 patients with acute HBV infection, 51 (65.3%) achieved FC, while 27 (34.6%) progressed to CHB infection. Patients who achieved FC had significantly higher peak alanine aminotransferase levels at presentation than those who developed CHB infection (2117 ± 1509 IU/mL vs 1103 ± 1540 IU/mL; P = 0.01). No differences in early initiation of nucleos(t)ide analogue therapy were found between patients who achieved FC and patients who developed CHB. Progression to CHB infection was significantly more common among patients with a background of low socioeconomic status (SES; 70% vs 41%; P = 0.04) and diabetes mellitus (DM; 26% vs 3.9%; P = 0.007). In univariable analysis, DM presence and a high SES were associated with a significant effect on HBsAg persistence (odds ratio = 8.57 and 0.46, respectively).

CONCLUSION

In the vaccination era, we postulate that acute HBV carries a higher-than-expected risk of chronicity, particularly among patients with metabolic disease and those from socioeconomic disadvantage backgrounds. Early risk stratification, targeted follow-up for high-risk cases, and further study of metabolic and social factors in FC are warranted.

Key Words: Hepatitis B; Diabetes mellitus; Socioeconomic status; Vaccination; Nucleos(t)ide analogues

Core Tip: The widespread availability of the universal hepatitis B virus (HBV) vaccination reduced acute HBV incidence in developed countries. Acute HBV usually resolves with functional cure in > 95% of adults. In our study we evaluated all acute HBV cases in southern Israel between 2004-2022 and found a striking 34% chronicity rate. Further analysis revealed that a background of diabetes and low socioeconomic status were found to strongly predict chronic infection. These findings highlight the need for risk stratification in acute HBV patients from vulnerable populations.

INTRODUCTION

Hepatitis B virus (HBV) is a small hepatotropic DNA virus that persistently remains a cause of chronic liver disease around the globe[1]. It is estimated that more than 2 billion persons have been exposed worldwide to HBV of whom approximately 240 million have chronic infection[2]. Although an effective vaccine and treatment exist, more than 880000 people die each year from complications of cirrhosis and hepatocellular carcinoma (HCC)[3]. Widespread availability of the vaccine in developed countries[4] has shifted the epidemiology of acute HBV in developed countries to vulnerable subpopulations[5].

Chronic HBV (CHB) is defined by the persistence of hepatitis B surface antigen (HBsAg) > 6 months following acute HBV infection[6]. The evolution of acute HBV infection to CHB likely involves both host and viral factors[7,8] although precise mechanisms are not well defined[9-13]. An intriguing novel host factor is the early introduction of nucleos(t)ide analogues (NAs) during the acute infection. American and European Liver society guidelines recommend supportive treatment during acute HBV and advocate for treatment with NAs only in patients with severe hepatitis potentially progressing towards acute liver failure[14,15]. However, there is a paucity of data regarding the effect of antiviral therapy[16] administered during acute HBV infection on the risk of progression to CHB. Furthermore, little is known regarding the impact of a low socioeconomic status (SES) and metabolic comorbidities such as obesity and diabetes mellitus (DM), which are often overrepresented among vulnerable unvaccinated populations[17-20], on the transition from acute to CHB.

In this study, we aimed to evaluate the hypothesis that the likelihood of achieving a functional cure (FC) vs progression from acute HBV infection to CHB in adult patients is influenced by patient background characteristics, pre-existing medical conditions and NA treatment.

MATERIALS AND METHODS

Study population

We performed a retrospective evaluation of all male and female patients, members of Clalit Health Services (CHS) in the Southern Israel district, above the age of 18 years who were diagnosed with an acute HBV infection between January 2004 to December 2022.

CHS provides medical services for a population of nearly 1 million people living in Southern Israel and has a comprehensive computerized database with records of medical history, laboratory and imaging tests. In Israel, where HBV genotype D[21,22] is the predominant genotype, a universal 3 dose vaccination program beginning within 12 hours from birth has been implemented since 1992[23] and has achieved widespread coverage.

Clinical definitions and data sources

A diagnosis of an acute HBV infection was established when a patient presented with newly positive tests for HBsAg and hepatitis B core (HBc) IgM antibody, in the absence of any prior evidence of HBV infection or immunity, including previous positivity for anti-hepatitis B surface, HBsAg, IgG or IgM anti-HBc, hepatitis B envelope antigen (HBeAg) or anti-hepatitis B envelope. Included patients were stratified into two groups: Those who demonstrated HBsAg loss with/without seroconversion within 6 months following acute HBV diagnosis (FC) and those who retained HBsAg 6 months or longer after acute HBV diagnosis (CHB). Patients lacking sufficient data to determine progression to CHB or achievement of FC following acute HBV infection, were excluded. Patients presenting with acute HBV infection while receiving immunosuppressive treatment in whom HBV reactivation could not be ruled out, were also excluded from the analysis. Patients with antibodies against hepatitis delta were excluded. Validation of the results was performed through manual review of HBsAg status in the CHS patient records. Patients with a PCR positive diagnosis of hepatitis C virus (HCV) or antibodies to HCV without a known negative PCR result were excluded. Patients with a diagnosis of human immunodeficiency virus were also excluded.

Clinical parameters including demographic characteristics, diagnostic information and laboratory data were collected if obtained within 1 month of the acute HBV infection. For laboratory markers with multiple measurements surrounding the acute event, peak values were selected to appropriately represent severity of disease. SES was assigned according to the Central Bureau of Statistics index, based on each patient’s residential zip code. DM was defined based on an HgbA1C score of at least 6.5% or use of anti-glycemic medications. Dyslipidemia and hypertension were identified based on documented diagnoses in the medical record or the use of corresponding pharmacologic therapies. Detection of anti-HBc and HBsAg was performed with the Alinity system from Abbot, Illinois, United States. HBV DNA was quantified via Real-Time Qualitative PCR assay, on the GeneXpert system (Cepheid, California, United States). HBV DNA testing was performed only in patients who provided a second blood sample following a new diagnosis of acute HBV. Anti-HDV antibodies were detected using the LIAISON^® XL MUREX Anti-HDV (DiaSorin, Saluggia, Italy). Data were extracted using the Clalit Research Data Sharing Platform powered by MDClone.

Study outcomes

The primary clinical outcome evaluated was HBsAg positivity following 6 months of the acute HBV infection. Secondary outcomes included comparisons of clinical and demographic characteristics between patients who achieved FC and those who progressed to CHB infection.

Ethical statement

The study was approved by the institutional review board of Soroka University Medical Center and was conducted in compliance with the Declaration of Helsinki, Good Clinical Practice guidelines, and local regulatory requirements. For this study of existing patient data, the Institutional Review Board of Soroka University Medical Center granted a full waiver of informed consent (Approval No. 0138-23-SOR) on September 26, 2023, under study protocol SCRC23018.

Statistical analysis

Patient characteristics were summarized using frequencies and percentages for categorical variables and means with standard deviations for continuous variables. Univariate comparisons, were performed using the χ² test or Fisher’s exact for categorical variables appropriate, and the Student’s t test, Welch two-sample t-test, or the Wilcoxon rank-sum test for continuous variables, depending on the data distribution and. A directed acyclic graph (DAG) was constructed based on clinical knowledge and prior literature, to formalize assumed causal relationships and to identify potential confounder. Odds ratios (ORs) with 95% confidence intervals (95%CIs) were calculated to quantify the strength and direction of associations. Based on DAG framework, age, sex, and SES were identified as confounders and included in a DAG-informed multivariable logistic regression model. After testing several iterations of the model, we ultimately retained SES as a continuous variable (1 = low, 3 = high) and excluded age from the final model. Variables not identified as confounders or considered downstream of the exposure were not adjusted for to avoid overadjustment bias. A two-sided P value ≤ 0.05 was considered statistically significant. All analyses were conducted using R (version 2023.09.1+494). Key packages, including survival, survminer, and gtsummary, were used for survival analysis, data visualization, and the generation of regression tables.

RESULTS

Study population and acute HBV outcome

During the study period from January 2004 to December 2022, the Soroka University Medical Center virology laboratory identified 181 patients who tested positive for HBsAg and IgM anti-HBc (Figure 1). Eighty patients were excluded due to prior positive HBV serology or missing data regarding HBV infection outcomes. A total of 78 patients were included in the final analysis. Of these, 51 patients (65.3%) achieved FC while 27 patients (34.6%) progressed to CHB. In both the CHB and FC groups the majority of patients were male (85% and 63% respectively, P = 0.03, Table 1). The mean age at the time of acute HBV infection was 37.7 years (range: 19.9-71.8 years) in the FC group and 42.06 years (range: 18.8-78.5 years) in the CHB group (P = 0.3). The majority of patients in both groups were Israeli born (63% in the FC group and 59% in the CHB group; P = 0.4). A significant difference in SES was observed between the groups. Patients in the CHB were significantly more likely to come from a low SES compared to those in the FC group (70% vs 41%, P = 0.04; Figure 2). DM was significantly more prevalent in the CHB group than the FC group (26% vs 3.9%, P = 0.007). No significant differences were observed between the groups with respect to ethnicity, smoking status, body mass index (BMI), diagnosis of hypertension or dyslipidemia, history of alcohol abuse or intravenous drug use.

Figure 1 Study population flow chart. HBV: Hepatitis B virus; HBsAg: Hepatitis B surface antigen; HCV: Hepatitis C virus; HIV: Human immunodeficiency virus.

Figure 2 Distribution of patients within the functional cure and chronic hepatitis B virus groups according to socioeconomic status, country of birth, gender and ethnicity. A: Socioeconomic status distribution by hepatitis B surface antigen (HBsAg) status; B: Gender distribution by HBsAg status; C: Country of birth distribution by HBsAg status; D: Ethnicity distribution by HBsAg status. HBsAg: Hepatitis B surface antigen; SES: Socioeconomic status; FC: Functional cure; CHB: Chronic hepatitis B virus.

Table 1 Patient characteristics, n (%) or mean ± SD.

	Total (n = 78)	Chronic HBV (n = 27)	Functional cure (n = 51)	P value1
Male	55 (71)	23 (85)	32 (63)	0.03
Ethnicity				0.3
Arab	20 (26)	9(33)	11 (22)
Jewish	58 (74)	18 (67)	40 (78)
Age (years)	39.2 ± 15.5	42.0 ± 17.2	37.7 ± 14.6	0.3
Region of birth				0.4
Israel	48 (62)	16 (59)	32 (63)
Africa	8 (10)	5 (19)	3(5.9)
Middle East	7 (9)	3 (11)	4 (7.8)
Russia/Former Union of Soviet Socialist Republics	8 (10)	1 (3.7)	7 (14)
Europe	2 (2.6)	1 (3.7)	1 (2)
Other	5 (6.4)	1 (3.7)	4 (7.8)
History of alcohol use	7 (9)	2 (7.4)	5 (9.8)	> 0.9
Active smoking	8 (16)	2 (15)	6 (17)	0.3
IV drug use	5 (6.4)	2 (7.4)	3 (5.9)	> 0.9
BMI (kg/m2)	25.4 ± 7.1	26.7 ± 6.1	24.7 ± 7.5	0.2
Diabetes mellitus	9 (12)	7 (26)	2 (3.9)	0.007
Hypertension	17 (22)	7 (26)	10 (20)	0.5
Dyslipidemia	5 (6.4)	3 (11)	2 (3.9)	0.3
Anti-HBs positive	30 (48)	8 (33)	22 (58)	0.059
HBeAg positive	49 (64)	18 (67)	31 (62)	0.7
HBV DNA (IU/mL)2	1.48 × 106 ± 3.5 × 106	1.25 × 106 ± 9.2 × 105	1.54 × 106 ± 3.9 × 106	0.8

¹Pearson’s χ² test; Wilcoxon rank sum test.

²Data available for only 25.9% and 58.8% of patients in the chronic hepatitis B virus and functional cure groups respectively.

HBV: Hepatitis B virus; BMI: Body mass index; HBeAg: Hepatitis B envelope antigen; Anti-HBs: Anti-hepatitis B surface.

Clinical differences between patients who achieved FC and those who progressed to CHB infection

Patients in the FC group presented with a more severe hepatitis compared with those in the CHB group (Table 2). Peak alanine aminotransferase (ALT) levels were significantly higher in the FC group than in the CHB group (2117 ± 1509 IU/mL vs 1103 ± 1540 IU/mL, P = 0.01, Figure 3). Peak aspartate aminotransferase (AST) levels were also higher in the FC group than in the CHB group, although this difference did not reach statistical significance (1378 ± 1046 IU/mL vs 838 ± 1166 IU/mL, P = 0.058). Peak total bilirubin levels and peak alkaline phosphatase levels were also significantly higher in the FC group than in the CHB group (12.0 ± 8.5 mg/dL vs 6.5 ± 8.6 mg/dL, P = 0.01 and 207 ± 105 IU/mL vs 159 ± 87 IU/mL, P = 0.04 respectively). No statistically significant differences were observed in gamma glutamyl transferase levels or markers of hepatic synthetic function (Table 2). A comparison of the mean liver enzyme levels during the acute presentation rather than the peak values, yielded similar results (Supplementary Table 1). In both the FC and CHB groups, the majority of patients were positive for HBeAg (63% vs 67%, P = 0.8). HBV DNA results were available for 25.9% of the CHB group and 58.8% of patients in the CHB group. Among patients with available PCR data, no statistically significant difference was observed in mean HBV DNA levels between the groups (P = 0.8). No significant differences were observed between the groups with respect to initiation of NA therapy. During the first month following the diagnosis of acute HBV infection, 7.8% of patients in the FC group and 11% of patients in the CHB group initiated NA treatment (P = 0.4). In both groups, entecavir was the most frequently prescribed NA. No statistically significant difference was observed in the MELD-3 scores between the CHB and the FC groups (8.19 ± 3.07 vs 9.34 ± 3.17, P = 0.2).

Figure 3 Comparison of liver enzymes and total bilirubin during acute hepatitis B virus presentation in patients who achieved functional cure and patients with chronic hepatitis B virus. HBsAg: Hepatitis B surface antigen; ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; GGT: Gamma glutamyl transferase; Alk Phos: Alkaline phosphatase; FC: Functional cure; CHB: Chronic hepatitis B virus.

Table 2 Clinical presentation of patients with chronic hepatitis B virus and patients who achieved functional cure, n (%) or mean ± SD.

	Total (n = 78)	Chronic HBV (n = 27)	Functional cure (n = 51)	P value1
ALT (IU/mL)	1770 ± 1585	1103 ± 1540	2117 ± 1509	0.01
AST (IU/mL)	1193 ± 1111	838 ± 1166	1378 ± 1046	0.058
Total bilirubin (mg/dL)	10.2 ± 8.87	6.52 ± 8.62	12.0 ± 8.5	0.01
Direct bilirubin (mg/dL)	4.31 ± 3.98	4.26 ± 5.89	6.87 ± 4.99	0.07
Alkaline phosphatase (IU/mL)	190 ± 101	159 ± 87	207 ± 105	0.04
GGT (IU/mL)	219 ± 144	175 ± 128	236 ± 148	0.12
Albumin (g/dL)	3.78 ± 0.57	3.86 ± 0.54	3.74 ± 0.59	0.4
INR	1.39 ± 0.85	1.24 ± 0.42	1.46 ± 1.0	0.2
Creatinine (mg/dL)	0.81 ± 0.31	0.89 ± 0.43	0.77 ± 0.23	0.2
Hemoglobin (g/dL)	13.78 ± 1.88	14.26 ± 1.82	13.53 ± 1.89	0.1
Platelets (× 103/micL)	230 ± 99	216 ± 121	238 ± 85	0.4
Na (mEq/L)	138.5 ± 3.0	138.7 ± 3.1	138.4 ± 2.9	0.7
MELD-3	8.97 ± 3.16	8.19 ± 3.07	9.34 ± 3.17	0.2
NA treatment				0.4
Entecavir	6 (7.6)	2 (7.4)	4 (7.8)
Lamivudine	1 (1)	1 (3.4)	0 (0)
No treatment	71 (91)	24 (89)	47 (92)

¹Welch two sample t-test.

ALT: Alanine aminotransferase; AST: Aspartate aminotransferase; GGT: Gamma glutamyl transferase; INR: International normalized ratio; NA: Nucleos(t)ide analogue; HBV: Hepatitis B virus.

Factors associated with the progression to CHB infection

A pre-existing diagnosis of DM at the time of acute HBV infection was associated with a significantly increased risk of progression to chronic infection (OR = 8.57, 95%CI: 1.88-61.01). High SES was significantly associated with a reduced risk of chronic infection (OR = 0.46; 95%CI: 0.22-0.89). Male gender was associated with an increased risk of chronic infection (OR = 3.41; 95%CI: 1.11-12.97). In contrast, age and BMI were not significantly associated with progression to chronic infection (Table 3). Initiation of NA therapy near the onset of acute HBV infection was not significantly associated with progression to chronic infection (OR = 1.47, 95%CI: 0.27-7.19, P = 0.63). However, the analysis was substantially underpowered because only a very small number of patients in either group received antiviral therapy limiting any meaningful inference regarding treatment effect. To formalize assumed causal relationships and to identify potential confounders, a DAG-informed multivariable logistic regression analysis was performed. However, due to the relatively small sample size, the multivariable model did not achieve sufficient precision or reliability (Supplementary Figure 1). In an additional sensitivity analysis, excluding patients with either DM or low SES did not materially alter the association between these factors and the risk of CHB infection (Supplementary Table 2). However, male sex was no longer an independently associated with the outcome after the exclusion of patients with low SES. A subgroup analysis of only the CHB and FC patients with HBeAg positivity (62.8% of the cohort) revealed results consistent with the full cohort analysis (Supplementary Figure 2 and Supplementary Table 3).

Table 3 Univariate analysis odds ratio analysis of factors leading to hepatitis B surface antigen persistence.

Predictor	Odds ratio (confidence interval)	P value
Age	1.02 (0.99-1.05)	0.243
Gender	3.41 (1.11-12.97)	0.0458
BMI	1.04 (0.97-1.12)	0.264
Diabetes	8.57 (1.88-61.01)	0.0109
SES (high)	0.46 (0.22-0.89)	0.0289

BMI: Body mass index; SES: Socioeconomic status.

DISCUSSION

In this study, we retrospectively evaluated a cohort of patients diagnosed with acute HBV in a setting with an established vaccination program, with the aim of identifying factors associated with the progression to CHB infection. In comparison with historical cohorts[6], our study identified a notably high proportion of adult patients (34%) who remained HBsAg positive and progressed to CHB infection according to current disease definitions. DM was identified as a significant risk factor for progression to CHB infection with an OR of 8.57. Conversely, higher SES was associated with a significantly reduced risk of progression to CHB with an OR of 0.46. Despite the small sample size and wide confidence interval, these findings suggest that in a setting with widespread vaccine availability, unvaccinated individuals from socioeconomically disadvantaged and medically vulnerable populations may be at increased risk of persistent HBsAg positivity. Additionally, patients who achieved FC exhibited more severe hepatitis, as reflected by substantially higher ALT and AST levels, compared with patients who remained HBsAg positive.

Although complete eradication of HBV is not achievable due to the persistence of covalently closed circular DNA and integrated viral DNA within hepatocytes[24], FC is typically achieved in more than 95% of adults following acute infection[14,15]. In contrast, we observed an approximately sixfold higher rate of HBsAg persistence at 6 months compared with historically reported rates. This unexpectedly high rate of HBsAg persistence, was confirmed by extensive manual review to exclude misclassifications or data inaccuracies. In addition, a subgroup analysis of only the HBeAg positive patients of our cohort produced comparable findings. Notably, A similarly high rate of HBsAg persistence was recently reported in a North American cohort of adults with acute HBV infection, prompting the authors to question the adequacy of current definitions of acute HBV resolution[17]. Together, these observations suggest that contemporary host- and population-level factors may meaningfully influence the immune response to acute HBV infection and warrant focused investigation.

DM represents a biologically plausible host-level factor contributing to impaired viral clearance following acute HBV infection. DM has been associated with dysfunction of both adaptive and innate immune responses, including impaired T-cell, B-cell, Macrophage and Natural killer cell function[25-28], as well as with worse clinical outcomes in CHB, including cirrhosis and HCC[29,30]. Consistent with these observations a recent large retrospective study from Israel demonstrated a higher prevalence of DM among patients with CHB infection[18]. However, data specifically addressing the impact of DM on HBsAg clearance following acute HBV infection remain limited. In this context, our findings suggest that diabetes-associated immune dysfunction may impair effective viral control during acute infection, thereby increasing the likelihood of HBsAg persistence. This hypothesis is further supported by the observation that patients achieving FC demonstrated higher ALT levels than those progressing to chronic infection, consistent with a more vigorous immune response.

Our findings also suggest that low SES may be an independent significant risk factor for HBsAg persistence. While CHB is known to disproportionately affect socioeconomically disadvantaged populations[19,20], the role of SES in viral clearance following acute infection has not been previously described. Potential explanations include the higher prevalence of comorbidities such as obesity[31], DM[32] and malnutrition[33] in low-SES populations, which may collectively contribute to impaired immune function, as well as emerging evidence linking gut microbiota composition to immune-mediated HBV clearance. Differences in gut microbiome composition across socioeconomic strata may theoretically partially contribute at least in part to the observed disparities and warrant further investigation[34-36]. Finally, we cannot rule out the possibility of residual confounding arising from differences in health care access health literacy, or other unmeasured socioeconomic factors that may have contributed to this association.

The principal strengths of this study include a rigorously validated cohort defined by serologic criteria, comprehensive longitudinal follow-up and access to unified electronic medical records capturing SES, comorbidities, laboratory data, and antiviral therapy. As the sole tertiary referral center in Southern Israel, our institution provides centralized care for acute HBV minimizing lost to follow up and ensuring uniform ascertainment of clinical outcomes.

Acute HBV infection remains uncommon in adults and the cohort size is comparable to contemporary studies[17]. However, key limitations of this study include its retrospective design and modest sample size, which limited statistical power particularly for evaluating the impact of NA therapy during the acute phase, and contributed to wide confidence intervals observed for key associations, including DM and SES. Additionally, although multivariable analysis was performed and demonstrated similar directional trends, the stability of these estimates was limited, further underscoring the need for larger, adequately powered cohorts to more precisely define risk relationship and mitigate residual confounding. Residual confounding cannot be excluded and the absence of data on HBV genotype (presumed to be predominantly genotype D), escape mutants, viral kinetics, transmission routes and interval between symptom onset and clinical presentation further limits mechanistic interpretation of our findings. In addition, despite the manual validation of all acute HBV cases included in our study, the possibility that some cases represented unrecognized HBV reactivation rather than true acute HBV infection cannot be completely excluded. Finally, the single center design and the region-specific characteristics of the study population may limit the generalizability of our findings.

CONCLUSION

This study demonstrates an unexpectedly high rate of progression from acute to CHB infection in adults, with DM and low SES emerging as significant risk factors for HBsAg persistence. These findings highlight the need for prospective studies examining metabolic, socioeconomic, and host-microbiome determinants of immune response during acute HBV infection and underscore the ongoing importance of comprehensive HBV vaccination strategies, to prevent chronic disease particularly in vulnerable populations.

ACKNOWLEDGEMENTS

This study was conducted as part of the requirements for MD degree from the Medical School for International Health at the Faculty of Health Sciences, Ben-Gurion University of the Negev.