Prevalence of vaccine-derived hepatitis B surface antibodies in children and adolescents in Germany: results from a population-based survey, 2014–2017

Viral hepatitis B is a serious public health challenge, and a leading cause of acute and chronic liver disease globally [1, 2]. Chronic hepatitis B can lead to severe long-term sequelae such as liver cirrhosis and hepatocellular carcinoma [1]. Hepatitis B virus (HBV) infections acquired through mother-to-child transmission are often asymptomatic, and more than 90% of those infected during infancy and early childhood develop a chronic infection compared to only 5% of people who are infected as an adult [1, 3].

World Health Organization (WHO) estimates that 14 million people are living with chronic HBV infection (2019) [4], and long-term sequelae caused by chronic HBV infections are responsible for about 56,000 deaths per year in the WHO European Region [5]. In Germany, an HBV prevalence (antibody to hepatitis B core antigen (anti-HBc) and hepatitis B surface antigen (HBsAg)) of 0.3% has been found in the latest adult population-based survey (German Health Interview and Examination Survey for Adults (DEGS1, 2008–2011)) [6]. Among children aged three to 17 years in Germany, the prevalence of the surface antigen of HBV (HBsAg, current infection) was 0.2% in 2003–2006 [6, 7].

In 2016, WHO published the first global health sector strategy on viral hepatitis, offering a plan that included specific impact targets and goals to eliminate viral hepatitis as a public health problem by 2030. Childhood vaccinations are a central intervention for hepatitis B elimination [8]. The action plan for ending viral hepatitis in the WHO European Region included the target of reaching an overall coverage of 95% of the population having three doses of the HBV vaccine by 2020 to reach HBV elimination by 2030 [9, 10]. Evidence from Germany indicates that the childhood vaccination coverage is below this threshold [11].

While treatment of chronic hepatitis B infection may lead to seroconversion, there is no cure [1]. A life-long treatment is usually required to maintain a low viral load and prevent the development of long-term sequelae [1]. This, as well as the high risk of chronic infection if infected early in life, emphasizes the importance of primary prevention of hepatitis B infection during childhood.

An effective and safe vaccine has existed since 1982, and currently available vaccines offer a high protection against infection with hepatitis B [12‐14]. In Germany, vaccination against HBV has been recommended to all infants, children and adolescents since 1995 in the national immunisation schedule, as outlined by the German Standing Committee on Vaccination (STIKO) [15].

Ensuring effective protection following hepatitis B vaccination has important implications for both individuals and public health. While vaccination offers good protection and has proved to be an important intervention to reduce transmission, the exact duration of hepatitis B protection after childhood immunisation remains unclear. Some studies suggest several years or even lifelong protection [12, 16‐20], while others indicate 10–15 years [18, 21]. A study from Germany demonstrated lower anti-HBs levels after HBV childhood immunisation and reduced immunogenicity when given the polyvalent vaccine Hexavac [22], which was withdrawn from the market in 2012 [23].

Successful primary vaccination has been defined as when the individuals have anti-HBs levels ≥ 100 IU/L 4–6 weeks after vaccination by STIKO [24], whereas WHO defines an initial anti-HBs level of ≥ 10 IU/L as indication of successful vaccination [25]. However, the anti-HBs level decreases over time and furthermore does not necessarily reflect immunity [20, 26, 27].

Studies also suggest that there is a difference in anti-HBs levels after primary vaccination according to the number of doses given (three or four) and the time interval between the last two vaccine doses [28, 19]. However, important open questions about the duration of hepatitis B protection following childhood immunisation remain. One of these include whether the recommended six months between the two last doses impacts the stability of the anti-HBs level over time [29].

Until recently, either four doses of polyvalent vaccine or three doses of monovalent vaccine were recommended for infant vaccination in Germany. In June 2021, the recommendations were changed to three doses which should be provided at two, four and 11 months of age with a minimum of six months between the last two doses, regardless of vaccine type. Only children born to a mother with unknown HBV status or who are infected with HBV are offered a birth-dose simultaneously with immunoglobulin [30].

A study by Gillesberg Raiser et al. [31] found that 91.4% (95%CI: 89.7%–92.8%) of children and adolescents in Germany who did not receive a birth-dose, received three or more doses of hepatitis B vaccination. However, only 79.1% (95%CI: 76.8%–81.2%) received the last dose with at least six months between the two last doses as recommended.

We aimed to estimate the proportion of hepatitis B vaccinated children and adolescents (three to 17 years) in Germany with an anti-HBs level < 10 IU/L, 10–99 IU/L and ≥ 100 IU/L. Moreover, we wanted to assess if three versus four doses and compliance with the recommended six months interval between the last two doses are associated with an anti-HBs level ≥ 10 IU/L when taking type of vaccine and time since the last dose into account.

We used data from the second wave of the German Health Interview and Examination Survey for Children and Adolescents (KiGGS Wave 2), a nationwide population-based cross-sectional study conducted from 2014–2017 [32].

KiGGS Wave 2 is comprised of two components each with its own sampling procedure: interviews only and interviews combined with physical health examinations. A detailed description of the methodology of KiGGS and KiGGS Wave 2 is published elsewhere [32, 33]. In brief, an age-stratified population sample was drawn of three to 17-year-olds who were invited to take part in the physical health examination component. A total of 3,567 children and adolescents participated in the examination component, which corresponds to a response rate of 41.5%.

Data on sociodemographic factors and general health status through a large number of health indicators were collected using questionnaires and interviews [32]. The socio-economic status is measured through the collection of education and professional qualification, employment status of parents and net household income which are converted into a seven point index scale. This original index scale was then categorised into five groups [34]. Information on date, number and type of vaccinations was collected from vaccination cards.

Blood samples were collected from participants as part of the physical health examinations. The blood samples were analysed for antibodies against hepatitis B (anti-HBs and anti-HBc) and HBV surface antigen (HBsAg) using the Abbott Laboratories (Illinois, USA) Architect system. The anti-HBs assay used had an overall specificity of 99.67% (95% CI 99.2%–99.9%) and a sensitivity of 97.5% (95% CI 95.9%–98.6%) with a limit of blank of 0,43 IU/L, a limit of detection of 0,90 IU/L, and a range of quantitation of 2,50–1000,00 IU/L. Serum specimens with a level of anti-HBs exceeding 1000 IU/L were automatically diluted and re-analysed. Diluted samples still exceeding the upper range of quantitation (now 25,000 IU/L) were not further processed and excluded from Fig. 1, but included in the other quantitative data analyses (n = 17) [35].

We excluded participants with unknown vaccination date, or with unreadable or incomplete information on hepatitis B vaccination. Furthermore, we excluded participants with missing anti-HBs measurements, or with signs of previous or current HBV infection (anti-HBc or HBsAg positive participants).

We defined a recommended schedule as one with three or more vaccine doses with at least six months between the last two doses.

Three thousand five hundred sixty-seven children and adolescents participated in the examination part of KiGGS Wave 2. Of them, 402 participants were excluded due to not having their vaccination card, or it being illegible, and another 206 due to not having any HBV vaccination records. Six participants had signs of current or previous HBV infection and 464 had no anti-HBs titre measurements (Additional file 1). Finally, 2,489 were included in our analyses.

Of the 2,489 included participants, 50.7% were female. The median age was 11 years (IQR: 7–14). Nearly two thirds of the participants lived in West Germany (64.1%). 21.5% had a migration background (at least one parent). 76.4% of participants had received a polyvalent HBV vaccination series (Table 1).

KiGGS Wave 2 participants included in our analysis differed slightly from those excluded with regard to age and place of residence. The median age of the included participants (N = 2,489) was 11 years, compared to 9 years for the 1,078 excluded participants (p < 0.001). The proportion of participants living in East Germany (including Berlin) was slightly higher for included participants (35.9%) than excluded participants (31.2%) (p < 0.001) (Additional file 2).

In our study, we did not find that participants who had been vaccinated on the recommended schedule, or who had four doses of HBV vaccine, had higher levels of anti-HBs when taking time since last dose and type of vaccine into account, which is a result that has previously been reported in the literature [19, 22, 28]. There are several possible explanations for this. First, it is important to underline that having been vaccinated on the recommended schedule or having received four doses may have an impact on immune persistency and immunological memory, and individuals with low and decreasing anti-HBs levels may still be protected [22]. It is also likely that the impact of time since last dose (especially since antibody levels are measured at study participation) and type of vaccine is so strong that any associations with other variables are difficult to detect. Furthermore, when looking at the categories of three versus four doses and recommended versus not recommended schedule, there are only a few participants in the non-recommended schedule group and the group with three doses, especially in the anti-HBs ≥ 100 IU/L category. We found time since last dose and type of vaccine to be independently associated with anti-HBs concentration which is consistent with what has been found in other studies [22, 19, 27‐29, 36].

Our results confirmed the importance of time since last dose, and the longer the time since the last dose, the lower the anti-HBs concentration. While anti-HBs level does not equal protection, looking at time in relation to anti-HBs concentration is important as the purpose of the childhood HBV vaccine programme is to ensure protection against HBV infection throughout childhood and adolescence. Infection early in life is associated with a much larger risk of chronic infection and sequalae, and therefore reducing risk of infection in early life is important for individuals and public health. For this study, we did not have a longer follow-up time to assess whether protection may last until adulthood, which would be needed to further answer the question of the duration of HBV protection and how it depends on the level of anti-HBs.

Biological sex has been reported in the literature to have an influence on anti-HBs titre, with females in general having a better post-vaccination response [37‐40]. The differences in immune response is caused by both genes and hormones, and the immune response changes in males and females over time [41]. In our study we did not include biological sex as an individual variable in our models. Among the KiGGS participants, a higher proportion of males have received a recommended schedule in the older age groups. We tested the effect of biological sex in the multivariable models and sex did not impact the effect and association between our variables of interest and anti-HBs concentration. More than half of the participants however were also younger than 11 years of age, and therefore the effect of hormones and females reaching puberty may not be detectable among the participants in our study [41].

Our results confirmed previous findings that Hexavac has a weaker immunogenicity than Infanrix Hexa, illustrated in our data as being negatively associated with anti-HBs level ≥ 10 IU/L [22]. While in some individuals the anti-HBs concentration decreased to a lower or undetectable level, immunity may remain due to HBsAg-specific immunological memory [22]. This biological response may be linked to factors which we were unable to assess with the available data.

Thus far, the WHO and STIKO in Germany do not recommend booster vaccinations after completion of the three-dose vaccination programme [12, 30]. However, considering that more than one third of the children and adolescents in our study had an anti-HBs level under 10 IU/L, further information on duration and actual protection would help answer whether a booster vaccination would be called for to maintain immunity. One systematic review found that the protection varies, and that based on the anti-HBs titre (e.g. less than 10 IU/L), a booster dose should be administered [29]. However, the study also stresses the lack of data in particular from large sero-prevalence studies among adults that have been vaccinated against HBV as part of the childhood vaccination programme. Another larger study from Taiwan, which consisted of a series of cross-sectional serological surveys of HBV markers in four age groups between 2004 and 2012 found that booster vaccination did not add protection against HBsAg [42]. A study which measured residual immunity 10–16 years after vaccination in Canadian children [43] found that despite absence of HBV antibody concentrations ≥ 12 IU/L, most participants had an anamnestic response to a challenge dose which indicates immune memory and likely protection. There is a need for more studies in this area, and also booster vaccinations may be more applicable in high prevalence settings and or among groups at higher risk of infection, for example household contacts of people living with hepatitis B.

Our results are subject to limitations. Importantly, for the observational KiGGS Wave 2 study anti-HBs antibodies were measured at the time of study participation, therefore the time since vaccination varied considerably. Furthermore, we did not measure an acute response to vaccination, but rather a mixture of response and waning.

Moreover, only 2,489 of the 3,567 participants (69.8%) who took part in the examination arm of the KiGGS Wave 2 study were included in our analyses and we found some differences in age and geographical place of living between those included versus those excluded, however the weighted analyses accounted for this difference in the probability of selection in the sample. One explanation for the lower mean age among those excluded may be that parents tended to be less willing to agree to a venous puncture when their children were of younger age. While there may be differences that we were unable to account for, implications of study participation have likely had limited impact on our results, as sociodemographic differences may affect access to (and uptake of) vaccines, but are unlikely to influence the biological response to vaccination. Fitting and comparing statistical models is normally done using the likelihood-ratio or AIC test, however this is challenging using survey data [44]. We assessed model fit using the Wald test for each added variable [44].

While there are important limitations to our study, there are also several strengths. The KiGGS Wave 2 data are based on a probability-based nation-wide sample, with a rigorous recruitment strategy and a high participation rate. Survey weights were included to account for the unequal probabilities of selection into the sample according to the national level regarding age, sex, federal state, migration background, and the education level of the parents.

In conclusion, we found that almost two thirds of the study participants had an anti-HBs level ≥ 10 IU/L at the time of the examination. We found that having a recommended schedule or three versus four doses did not have an influence on anti-HBs level. Future studies are needed to improve the understanding of the association between anti-HBs level and adequate and sustaining protection against HBV.