Introduction

Hepatitis E virus (HEV) infection is a major cause of acute viral hepatitis worldwide, leading to 20 million HEV infections and 3.3 million symptomatic cases each year1. Four HEV genotypes are predominantly detected globally. Genotypes 1 and 2 are exclusive to humans and are primarily transmitted via contaminated water. Genotypes 3 and 4 are zoonotic, infecting various animals such as swine, wild boars and deer, and occasionally causing sporadic hepatitis E in humans across almost all countries2,3.

Certain populations are particularly vulnerable to severe complications, fulminant disease, or death following HEV infection. Among patients with underlying chronic liver disease, superinfection of HEV represents a potential trigger for acute-on-chronic liver failure4. In pregnant women, HEV infection can lead to acute liver failure and obstetric complications. A high maternal case fatality rate of 3.2–70% has been reported5, coupled with an increased risk of adverse pregnancy outcomes such as stillbirths, intrauterine deaths, and maternal-fetal transmission, resulting in neonatal infection6. Immunosuppressed patients, including transplant recipients and human immunodeficiency virus (HIV) carriers, are also at increased risk of developing chronic hepatitis E with rapid progression to cirrhosis3,7,8,9. Furthermore, HEV poses a significant burden in fragile, conflict-affected and vulnerable settings, where large-scale outbreaks have occurred. From 2010 to 2020, twelve outbreaks in forcibly displaced populations in sub-Saharan Africa resulted in over 30,000 cases and more than 610 deaths10.

A recombinant HEV vaccine, HEV239 (Hecolin, Xiamen Innovax, China), has been available for prophylactic use since its approval in China in 2011 and Pakistan in 2020. Its efficacy and safety were established in a large-scale randomized controlled phase 3 trial11. Recent data further demonstrated that the HEV239 vaccine provides high efficacy that persists for at least 10 years12. In 2022, despite facing several challenges, the South Sudan Ministry of Health and Médecins Sans Frontières (MSF) conducted the first Hepatitis E vaccination campaign to control an outbreak13. World Health Organization (WHO) is collaborating with experts and global partners to develop a generic protocol for using the hepatitis E vaccine in outbreak scenarios1. However, there is a critical need for gathering additional usage data on various HEV vaccination regimens. This endeavor is essential for formulating an optimal vaccination strategy in emergency scenarios, with the aim of maximizing accessibility while minimizing costs.

Thus far, the results of the 10-year follow-up observation have suggested that two doses of HEV239 (0 months and 1 month) provide long-lasting protection12, and a phase 2 trial study in rural Bangladesh also showed that two doses of HEV239 elicited a strong antibody response that remain positive for at least two years in almost all recipients14. However, data regarding the longer-term antibody responses to the two-dose regimen remain limited. Moreover, few studies have provided insights into the utilization of a single dose of HEV239 globally, including the strength and persistence of its antibody response. In this report, we present the findings from an extension of the phase 3 efficacy study of HEV239, evaluating immune responses to different vaccination regimens based on serum samples collected at various intervals post-vaccination.

Results

Baseline characteristics of the participants

As shown in Fig. 1, 14,094 participants from Anfeng and Qindong were randomly assigned to either the HEV239 or placebo group, with each participant receiving at least one dose of the vaccine or placebo. Of these, 14,069 participants were tested for anti-HEV antibodies prior to the first administration, including 7037 in the HEV239 group and 7032 in the placebo group. The majority of participants were from Anfeng (81.21%, 11,426/14,069), while the remaining 2643 participants were from Qindong. Ultimately, 12,082 participants underwent anti-HEV antibody testing at month 7 and were included in the analysis set, comprising 6028 in the HEV239 group and 6054 in the placebo group.

Fig. 1: Flowchart of study participants.
figure 1

14,094 participants from Anfeng and Qindong were randomly assigned to either the HEV239 or placebo group, each receiving at least one dose of the vaccine or placebo. In the HEV239 group, 11 missed anti-HEV result at 0 m and 1009 missed anti-HEV result at 7 m. In the placebo group, 14 missed anti-HEV result at 0 m and 978 missed anti-HEV result at 7 m. 12082 participants were included in the analysis set, with 6028 in the HEV239 group and 6054 in the placebo group.

Full size image

In the HEV239 group of the analysis set, vaccine doses were distributed as follows: 88 participants received one dose, 373 participants received two doses (184 at months 0 and 1; 189 at months 0 and 6), and 5567 received three doses. Baseline characteristics such as age, township, and anti-HEV IgG seroprevalence were similar across these groups with different doses, but sex distribution differed between some groups (Table 1; Supplementary Table 1).

Table 1 Baseline characteristics of the participants in different vaccination regimen group
Full size table

Long-term anti-HEV IgG positivity rates

There was a distinct difference in the baseline seroprevalence of anti-HEV IgG between the two townships, with Qindong showing a higher baseline seroprevalence (59.64%) compared to Anfeng (44.11%) (p < 0.0001). Additionally, the unsynchronised sampling timepoints further precluded a combined analysis of immunogenicity persistence data from both townships. Therefore, we conducted separate analyses for each township.

As shown in Table 2, among baseline seronegative participants in Anfeng, 53.3% (8/15) of those who received a single dose remained seropositive at months 91. Meanwhile, seropositive rates were 61.2% (30/49) for two doses at months 0 and 1, 59.6% (34/57) for two doses at months 0 and 6, and 73.0% (1270/1740) for three doses, all significantly higher than the placebo group (p < 0.0001). Concerning baseline seropositive participants in Anfeng, all 14 individuals in the single-dose group maintained their serostatus at month 91 (Table 3). The seropositivity rate was maintained by 94.1% (32/34) at month 91 for two doses at months 0 and 1 and was 95.6% (43/45) for two doses at months 0 and 6. With the three-dose regimen, 97.9% (1409/1439) of recipients remained seropositive at month 91. In the placebo group with baseline seropositivity, 81.9% (1258/1536) remained seropositive at month 91, which was significantly lower than those observed in the two-dose group (the 0–6-month schedule) (p = 0.018) and the three-dose group (p < 0.0001).

Table 2 Anti-HEV IgG seropositivity among participants who were seronegative at baseline
Full size table
Table 3 Anti-HEV IgG seropositivity among participants who were seropositive at baseline
Full size table

In Qindong, due to the limited sample size, the seropositivity rates among baseline seronegative participants showed fluctuations across different dosage groups during the follow-up period, but the seropositivity rate in the three-dose group remained high (87.3%, 254/291) at month 103, which was significantly higher than that of the placebo group (p < 0.0001). Among participants seropositive for anti-HEV IgG at baseline, nearly all of them retained antibody positivity at month 103, regardless of the number of doses administered. In contrast, only 89.4% (470/526) of placebo recipients who were seropositive at baseline had detectable antibody at the 103-month mark.

Differences in antibody kinetics between different doses or regimens

The trends in anti-HEV IgG geometric mean concentration (GMC) dynamics exhibited similar patterns across the four vaccine dosage groups, showing a decline after reaching a peak at the month 7 and eventually stabilizing around the month 55 or 67. Although peak levels varied among the groups, the plateau antibody levels were relatively similar (Fig. 2).

Fig. 2: Differences in antibody kinetics between different doses or regimens.
figure 2

Data are presented as GMC and 95% CI in the figure. Time points of the missing data were not listed in the figure (month 91 in Anfeng, and month 79 in Qindong). Comparisons of antibody dynamics between different dosage groups were conducted using the generalized estimating equation (GEE) model. # denotes no differences in GMC were observed between any of the dosage groups.

Full size image

The comparative analyses of antibody dynamics across different dosage groups revealed noteworthy findings. Among the participants in Anfeng who were seronegative at baseline, an increase in the number of vaccine doses corresponded with a marked enhancement in antibody response, demonstrating a clear dose-response relationship. The GMC levels in the three-dose group were significantly higher than those observed in both two-dose groups (the 0–1- and 0–6-month schedules) and the single-dose group (all p-values < 0.0001). Similarly, the GMC levels in each two-dose group were significantly higher than those in the single-dose group, with p-values of 0.0237 and <0.0001 for the 0–1-month and 0–6-month schedules, respectively. Interestingly, between the two-dose regimens, the 0–6-month schedule induced significantly higher GMC levels than the 0–1-month schedule (p = 0.0086). A similar dose-response relationship was observed among baseline seronegative participants in Qingdong, but the sample sizes in the single-dose and two-dose groups limited the robustness of these observations (Supplementary Table 2).

Among participants who were seropositive for anti-HEV IgG at baseline, the GMC levels were comparable across different dosage groups, particularly after 43 months. Generally, for baseline seropositive participants, all dosages regimens induced similarly durable and high antibody responses, significantly exceeding their baseline levels. By month 67, the GMC levels of Anfeng participants with baseline seropositivity who received a single dose, two doses at months 0 and 1, two doses at months 0 and 6, and three doses were 1.17 WU/mL (95% CI 0.65–2.11), 1.60 WU/mL (95% CI 1.10–2.32), 1.72 WU/mL (95% CI 1.28–2.30) and 1.85 (95% CI 1.77, 1.94), respectively, indicating a 2.4- to 3.6-fold increase over baseline levels. Additionally, in Anfeng, the GMC in the three-dose group was significantly higher than in the one-dose group (p = 0.0446), with no statistical differences observed among the other groups. In contrast, no differences in GMC were observed between any of the dosage groups in Qindong (Supplementary Table 3).

Discussion

This study is the first to present data on the durability of the anti-HEV IgG response elicited by one or two doses of the HEV239 vaccine in phase 3 trial participants with different baseline serostatus over 103 months. Furthermore, the long-term antibody dynamics were systematically analyzed across four vaccination regimens: a single dose, two doses (0–1 month), two doses (0–6 month), and three doses, with a placebo group as a control.

In hyperendemic regions, the hepatitis E vaccine is of particular public health importance for outbreak management15. However, a three-dose regimen, requiring over six months to complete, necessitates minimal migration and high recipient compliance, and confronts issues such as prolonged vaccine storage and transportations, posing extreme challenges especially when controlling outbreaks among refugees and internally displaced populations. Therefore, a reduced and shorter vaccination schedule (single or two doses) could simplify, economize, and enhance the feasibility of vaccination campaigns, enabling a rapid response to a hepatitis E outbreak and promoting broader vaccine adoption.

A few studies have provided valuable insights on the protection conferred by the two-dose HEV239 vaccine regimen. One such study was a phase 2 trial in Bangladesh involving 100 healthy including women of child-bearing age, which found that the two-dose regimen (0–1 month) demonstrated good safety and immunogenicity over two years, as well as providing protection against HEV infection during this period14. Another study was a 10-year extension follow study based on a large-scale phase 3 cohort involving 112,604 participants, which reported a 100% of efficacy against hepatitis E within 30 months of receiving two doses12. Based on these findings, the Strategic Advisory Group of Experts on Immunization (SAGE), stated in the meeting in May of this year, “in fragile and conflict-affected settings, the 2-dose schedule (at 0 and 1 month) is an acceptable alternative to the 3-dose schedule.”16 However, additional data are needed to support the use of the two-dose regimen of hepatitis E vaccination, as the long-term persistence of antibodies and the optimal interval between doses remain unclear. More importantly, there is a significant lack of data on the single dose.

In our study, the trends in anti-HEV IgG GMC dynamics indicated a decline from month 7 onwards, consistent with the immunological patterns observed in previous vaccines of the same technical type17. The antibody levels of plateau were relatively similar among groups. Both the single-dose and two-dose regimens of HEV239 demonstrated notable immunogenicity and persistence. Even among baseline seronegative individuals, more than half retained detectable specific antibodies at month 91 (within about 7.6 years) post-vaccination with two doses, and the GMC levels at month 67 remained above the detection limit. Compared to baseline seronegative participants, seropositive individuals exhibited more prolonged and higher anti-HEV IgG response in this study, indicating that pre-existing immunity contribute to the more persistent antibody response in seropositive participants. Notably, even after a single dose of the HEV239 vaccine, individuals with pre-existing immunity achieved high and sustained antibody levels for over 103 months, comparable to those observed with the two-dose regimen.

In a prior study based on this cohort, we have analyzed the relationship between antibody concentrations and the risk of HEV subclinical infection, demonstrating that even marginal antibody levels (0.077–0.25 WU/mL) significantly reduced the risk of HEV subclinical infection, providing protection for 74% [Relative risk (RR) = 0.26, 95%CI 0.11 to 0.65] of individuals exposed to the HEV18. And antibody levels within the ranges of 0.077–1.0 WU/mL and ≥1.0 WU/mL corresponded to RRs of subclinical infection of 0.58 (95%CI 0.27 to 0.65) and 0.09 (95%CI 0.04 to 0.17), respectively. The theoretical estimate of antibody threshold sufficient to lower the risk of hepatitis E disease might be even lower. Data from the Anfeng subset, which encompassed a larger sample size, revealed that antibody GMC levels in seronegative participants vaccinated with either one or two doses remained above the detection threshold at 67 months (0.11–0.12 WU/mL), while seropositive individuals exhibited much higher levels (1.17–1.72 WU/mL). These findings indicated that a one- or two-dose vaccination regimen may offer substantial protective benefits. However, it remains challenging to interpret individual antibody levels until the exact nature of protective anti-HEV immune responses—such as protective antibody thresholds and the role of cellular immunity—is clarified. Tracking antibody levels before the onset of breakthrough cases could be a method for understanding the protective antibody thresholds against hepatitis E disease. Unfortunately, based on the hepatitis surveillance results, only one breakthrough case was identified in the HEV239 vaccine recipients within our study population (from the three-dose group in Anfeng).

In addition to the lack of large-scale studies on pre-breakthrough antibody levels in vaccinated individuals, the complexity of viral immune-escape mechanisms also complicates the determination of the specific IgG antibody threshold that confers protection against HEV. Although the relationship between protection and antibody levels remains ambiguous, this study provides foundational data on the immune persistence with one or two doses, supporting ongoing efforts to determine the protective antibody threshold.

Our study possesses several limitations that warrant consideration. Firstly, the sample sizes for the single-dose and two-dose groups were limited, and there were missing data points at certain follow-up times (month 91 in Anfeng, and month 79 in Qindong), which may introduce some uncertainty in assessment of long-term antibody kinetics. Secondly, due to the budget and feasibility of the clinical trial, we staggered the sampling timepoints in the two townships and the collection frequency was reduced to once every two years since month 55. Thirdly, our analysis did not account for the potential boosting effect of natural infections on HEV vaccine antibody responses, which could lead to an overestimation of antibody levels.

In conclusion, one or two doses of the hepatitis E vaccine elicited durable antibody response and may offer substantial protective benefits.

Methods

Subjects and intervention

This study was an extension to a double-blind, randomised, placebo-controlled, phase 3 clinical trial of HEV239 conducted in 11 rural townships of Dongtai County, Jiangsu Province, China from August 2007 to October 2017 (NCT01014845). As previously described11,12,19, a total of 112 604 healthy adults aged 16–65 years were enrolled and randomly assigned in a 1:1 ratio to receive three doses of HEV239 or placebo (hepatitis B vaccine) at months 0, 1 and 6. Among them, 14 094 participants from the two townships, Anfeng and Qindong were designated as the immunogenicity subset. Serum samples were collected at months 0, 7, 19, 31, 43, 67, and 91 among participants from Anfeng, and were collected at months 0, 7, 13, 19, 31, 43, 55, 79, and 103 among participants from Qindong. Participants who completed serum collection at both 0 and 7 months were included in the analysis set of this study.

The study was done in accordance with the Declaration of Helsinki. This report is an extension analysis of a 10-year follow study based on a large-scale phase 3 trial of HEV239 vaccine, approved by the ethics committee of the Jiangsu Provincial Centre for Disease Control and Prevention and Xiamen University, Xiamen, China. Written informed consent was obtained from each trial participant, and the protocol and other relevant materials have been provided previously12.

Laboratory measurements

The immunogenicity or antibody persistence was evaluated by measuring anti-HEV IgG levels at different time-points. The anti-HEV IgG levels were quantified using commercial ELISA kits (Beijing Wantai, China), in accordance with the instructions. The concentrations of anti-HEV IgG were concurrently assessed using the WHO reference reagent (NIBSC code 95/584) and reported in WHO units (WU). The cut-off value of assay was 0.077 WU/mL. For negative samples, the anti-HEV IgG concentrations were assigned a value of 0.0385 WU/mL (half of the cut-off value).

Statistical analysis

In immunogenicity analysis, we present the seropositivity and the GMC with 95% confidence intervals (CIs) at each serum collection timepoint. Comparisons of antibody dynamics (GMC levels and their trends over time) between different groups were conducted using the generalized estimating equation (GEE) model20. Sex and age were included as covariates in the GEE model to correct for potential biases. For analyses involving three or more groups, p values were further corrected for multiple comparisons using Bonferroni method.

The analysis was performed using SAS software (version 9.1; SAS Institute, Cary, NC, USA). All reported p values are two-sided with α = 0.05.