Background
Worldwide, hepatitis B virus (HBV) is a common cause of liver disease. An estimated 350 million persons are chronically infected with HBV, and these individuals have a 15-25% risk of dying from HBV-related disease, including liver cirrhosis, hepatic decompensation, and hepatocellular carcinoma [
1,
2]. HBV is 100 times more infectious than human immunodeficiency virus and is transmitted by percutaneous or mucosal exposure to infected blood or other body fluids. Perinatal transmission, household contact, sexual contact, blood transfusion, and unsterilized injection are known as common routes of HBV transmission. The risk of mother-to-child transmission is 5-20% if the mother is positive for hepatitis B surface antigen (HBsAg) alone, but 90% if the mother is positive for hepatitis B e antigen (HBeAg) [
3]. To prevent mother-to-child transmission at or around birth, hepatitis B immunoglobulin (HBIG) is administrated for newborns born to HBsAg-positive mothers within 12 hr after birth combined with a three-dose series hepatitis B vaccine in many countries, including Japan [
4,
5]. HBIG has high levels of antibodies to HBsAg (anti-HBs), which are neutralizing antibodies against HBV. HBIG is immediately effective and protective for a few months after birth. However, the levels of anti-HBs decrease over time. Therefore, active vaccination is required to sustain sufficient levels of anti-HBs to protect young infants from HBV infection. This combination strategy can show a protective efficacy of almost 90% and results in fewer than 5% of infants becoming HBV carriers [
6‐
8].
Little is known about immunity from HBV infection in children after successful immunoprophylactic treatment, resulting in several questions about immunity post-vaccination. For example, it remains controversial whether the appearance of anti-HBs in children born to HBsAg-positive mothers implies complete protection from HBV infection after birth. Previous studies showed that serum HBV DNA was detected by polymerase chain reaction (PCR) in children born to HBsAg-positive mothers even after anti-HBs were induced by hepatitis B vaccine [
9,
10]. These findings suggested that children born to HBsAg-positive mothers have a risk of becoming HBV carriers even if immunoprophylactic treatment was successfully administered. Although the levels of serum HBV DNA are low in these anti-HBs-positive children after immunoprophylactic treatment, it is nevertheless a concern that reactivation of HBV replication could occur if these children receive immunosuppressive therapy.
In addition, the responses of HBV-specific cytotoxic T lymphocytes (CTLs) have never been evaluated in children after prophylactic treatment. HBV-specific CTLs play a major role in the control of HBV infection [
11]. Because hepatitis B vaccine is derived from surface protein, theoretically Th2 cytokines associated with helper T lymphocytes are produced in response to vaccination [
12]. To stimulate major histocompatibility complex (MHC) Class I restricted CD8+ CTLs, endogenous peptides processing and presentation is required. Although HBs peptide-specific CTLs can be induced by hepatitis B vaccine, whether CTL responses to other peptides derived from core and polymerase regions are primed remains unclear. There is, therefore, a possibility that HBV-core protein or -polymerase protein-specific CTLs would be detectable in children born to HBsAg-positive mothers if these children were exposed to HBV transiently or persistently at or after birth.
The aim of this study was to clarify the association between HBV-specific CTL responses and HBV viremia in children born to HBsAg-positive mothers after successful immunoprophylactic treatment. HBV-specific CTL responses were evaluated using enzyme-linked immunosorbent spot (ELISPOT) and MHC Class I peptide pentamer assays. Because human leukocyte antigen (HLA)-A24 is the most common HLA type in Japan, HBV-specific HLA-A24-restricted CTL epitopes were used for assays. Serum HBV DNA was measured by real-time PCR, which is a powerful tool for the detection of HBV DNA. Our data suggest that low-level viremia may occur and that this may be associated with priming of CTL responses even after apparently successful immunoprophylaxis.
Discussion
Newborns born to HBsAg-positive mothers are usually exposed to HBV around delivery, except those who contract intra-uterine infection. The administration of HBIG containing neutralizing antibodies can prevent newborns from HBV infection for a few months after birth. Active hepatitis B vaccine following HBIG administration induces active immunity and stimulates the production of sustained neutralizing antibodies against HBV in newborns. However, the overall immune response induced by both HBIG and hepatitis B vaccine in the prevention of mother-to-child transmission remains unclear. Although a small proportion of newborns treated with HBIG combined with hepatitis B vaccine become positive for HBV DNA from cord blood or peripheral blood transiently at or after birth, they eventually show no viremia [
9,
10,
19]. In contrast, a proportion of newborns receiving HBIG combined with hepatitis vaccine become positive for anti-HBs, but HBV viremia persists after birth [
10]. These observations suggest that newborns can be suffering from HBV viremia transiently or for several years after birth, despite having received immunoprophylactic treatment. Our goal was to assess the association between T-cell responses and viremia in children after successful prophylactic treatment.
HBV viremia stimulates host cellular immunity, and CTLs can be involved in viral control [
11,
17]. The induction of HBV-core or polymerase-specific CTL indicates that HBV infects host cells and that HBV proteins presented on the surface of the host cells are recognized as targets by the host. In this study, 2 (15%) of 13 children who were negative for HBsAg showed significant responses in IFN-γ ELISPOT against HBV peptides. Because HBV-specific peptides were derived for core and polymerase protein in this study, these findings suggested that HBV-specific T-cells were primed due not to hepatitis B vaccine but to the infection with HBV. In addition, one of the 2 children in whom HBV-specific T-cell responses were detectable was positive for serum HBV DNA. Although the levels of serum HBV DNA were low, viremia persisted despite HBV-specific T-cells and anti-HBs remaining detectable. These findings were similar to those in patients who suffered from acute HBV infection [
20,
21]. In acute HBV infection, clinical recovery does not imply the complete clearance of serum or intrahepatic HBV DNA. The authors of previous studies reported that HBV viremia persisted and anti-viral T-cell responses were detectable for several years after resolution of acute hepatitis B [
20‐
22]. The levels of serum HBV DNA are usually low after acute hepatitis B.
In animal models of acute hepatitis B, functional T-cell responses were generated after infection using high doses of inoculated virus [
23]. The results of one previous study suggested that exposure to a low viral load allows persistent viremia without priming functional T-cells [
24]. Similarly, detection of functional T-cells might be closely related to the load of HBV to which children are exposed before, during, and after prophylactic treatment. However, the mothers of the 2 children in whom HBV-specific T-cells were detected were negative for HBeAg. Because HBeAg-negative mothers usually have a low viral load (Table
1), viral exposure alone might not explain the induction of HBV-specific T-cells in children. Some studies suggested that HBeAg might induce T cell tolerance in vertical transmission [
25]. Therefore, viral genome mutations such as precore mutants, which are unable to secrete HBeAg, could be associated with the outcome of mother-to-child transmission [
26]. In C4, ELISPOT showed a positive result for polymerase protein but not core protein. This finding was consistent with that of a previous study, in which some patients showed positive results for both peptides and others showed positive results for only one peptide in acute HBV infection [
17]. Although CTL responses are restricted with the HLA type, CTL responses to viral epitopes were different in each individual.
Various studies have been performed using PCR to evaluate the efficacy of immunization against HBV in infants born to HBV carrier mothers [
9,
10]. In a previous study, 2 of 29 (7%) children who received HBIG and hepatitis B vaccine and became negative for HBsAg and positive for anti-HBs were positive for serum HBV DNA [
9]. Of the 2 children, one was free of HBV viremia 2 years later. The other child became positive for HBsAg and negative for anti-HBs later, because the mutation in the "a" determinant of the S region occured. In another study, serum HBV DNA was detected at 4-6 years after birth by PCR in 14 of 94 (14%) children who received prophylactic treatment [
10]. Of the 14 children, 7 were negative for HBsAg and positive for anti-HBs. Three of the 7 children showed a high viral load (> 10
5 copies/mL), and the remaining 4 children showed a low viral load (< 200 copies/mL). In the present study, real-time PCR was adopted to detect serum HBV DNA, as real-time PCR can be more sensitive than conventional PCR. In addition, serum HBV DNA was quantified more accurately. Serum HBV DNA was detected in 4 of 13 (31%) children in our study, but the levels of HBV DNA were between 100 and 80 copies/mL. The prevalence rate of HBV viremia in this study was higher than that in previous studies [
9,
10]. The difference in the prevalence was presumably caused by the difference in the detection limit of each PCR method. Anti-HBc were detected in one of the 4 children with viremia. However, the child was a 6-month-old boy. Probably, maternal anti-HBc were transferred to the child through the placenta at birth [
27]. Of the 4 children with viremia, 3 were born to HBeAg-negative mothers. In general, babies born to HBeAg-negative mothers are unlikely to become HBV carriers, even if immunoprophylactic treatment is not given. We cannot answer the question why 3 of 4 children with viremia were born to HBeAg-negative mothers. In a future study, the prevalence of viremia in a large number of children who received successful prophylactic treatment should be assessed.
Although escape mutants have been associated with persistence of serum HBV DNA after acute hepatitis B, the appearance of an escape mutant is uncommon in acute hepatitis B. In a previous study, variants in the S region were detected in 2 of 14 children who received prophylactic treatment [
10]. However, the 2 children were positive for HBsAg, negative for anti-HBs antibodies, and infected with a high viral load (>10
8 copies/mL). The previous study results suggested that the majority of patients with viremia were infected with wild-type HBV after prophylactic treatment. Although we did not sequence the S region, the mechanism of persistent viremia in children after the appearance of anti-HBs antibody might be different from that in children infected with escape variants. The level of serum anti-HBs antibodies considered protective is >10 mIU/mL [
28]. In the present study, the level of serum anti-HBsAb was < 10 mIU/mL in 4 of the 13 children who were negative for HBsAg after prophylactic treatment. Although it is controversial whether booster vaccination is required to maintain adequate levels of anti-HBs antibodies [
28‐
30], all of the 4 children were negative for serum HBV DNA. All of these children had levels of anti-HBs antibodies >10 mIU/mL at 1 year after birth. However, the levels of anti-HBs antibodies decreased to < 10 mIU/mL over time. None of the 4 children showed a significant HBV-specific T-cell response. Further studies are needed to evaluate the role of HBV-specific T-cell responses in children who lose anti-HBs after prophylactic treatment.
The results of this study raise questions about the long-term clinical impact of a low-dose HBV infection in children. For instance, children with a low-dose infection could suffer from hepatocelluar carcinoma in the future. In addition, the new emerging biological therapies for immunomodulation, including tumor necrosis factor inhibitors and anti-B cell agents such as Rituximab, will induce reactivation of HBV with a potentially fatal outcome in adult patients with chronic HBV infection or occult HBV infection [
31,
32]. Because this study was cross-sectional, we could not answer these questions. Longitudinal studies in children receiving prophylactic treatment are necessary to properly assess these issues and answer these questions.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
HK and PK contributed to the design of this study and drafted this manuscript. AI, TS, EH, AT, and TF participated in data collection and critically revised the manuscript. All the authors concurred with the submission and will take the responsibility for the manuscript.