Protective effect of an improved immunization practice of mother-to-infant transmission of hepatitis B virus and risk factors associated with immunoprophylaxis failure

Supplemental Digital Content is available in the text


Introduction
Hepatitis B virus (HBV) infection is a serious global health problem with approximately 240 million people persistently infected with HBV and 780,000 HBV-related deaths annually worldwide. [1] The risk of becoming an HBV carrier is inversely related to the age of the infected patient. Chronic HBV infections persist in 1% to 5% of infected adults, 20% to 30% of young children, and up to 90% to 95% of perinatally infected infants. HBV infection early in life is likely to cause chronic disease and subsequent complications. [2] A national survey conducted in China in 1992 showed that the prevalence of hepatitis B surface antigen (HBsAg) carriers was 9.67% among children <5 years old. [3] Approximately 20 years ago, China implemented an HBV immunoprophylaxis strategy, which led to a 90% reduction (to 0.96%) in HBsAg carriers <5 years old by 2006. [4,5] Approximately 90% of infants born to both HBsAg-positive and hepatitis B e-antigen (HBeAg)-positive mothers will become HBsAg carriers if no immunoprophylaxis is given. [6] The most effective way to prevent mother-to-child transmission (MTCT) of HBV infection is by immunizing all susceptible individuals with adequate hepatitis B immune globulin (HBIG) and hepatitis B vaccines at birth, especially newborn infants born to HBV-positive mothers. [2,[7][8][9] Although administration of HBIG and the hepatitis B vaccine at birth has significantly reduced HBV infection rates, 0.75% to 9.66% of vertical HBV transmissions have not been eliminated by this combined intervention strategy. [10][11][12][13] It has been suggested that providing antiviral therapy during the third trimester of pregnancy to mothers with high HBV viral loads would reduce the risk of perinatal transmission, but data regarding when and for how long this therapy should be administered are lacking. [14] Although many studies on perinatal HBV prevention have been carried out using various prophylaxis protocols, the hepatitis B vaccine doses used in these studies have been ambiguous or inconsistent and ranged anywhere from 5 to 20 mg. Although high maternal HBV viral loads and HBeAg-positivity have been shown to be the predominant risk factors for immunoprophylaxis failure, only a few studies have evaluated the use of larger hepatitis B vaccine doses for infants born to HBeAg-positive mothers. Also, the initial injection time for HBIG and hepatitis B vaccine administration has not yet been strictly defined. [15][16][17][18] In the present study, we conducted a prospective cohort of HBsAg-positive mothers and their infants to assess the protective efficacy of a modified immunoprophylaxis protocol. This modified protocol uses 2 different vaccine doses based on the maternal HBeAg status in combination with HBIG, and the injections are administered within 2 hours after birth. The HBV infection rate in infants and the potential risk factors associated with immunoprophylaxis failure and low protective antibody titers were evaluated.

Study design
This is a prospective, observational cohort study of HBsAgpositive mothers and their corresponding infants. On the basis of the status of HBeAg of mothers, different hepatitis B vaccine doses (20 or 10 mg) were given to the infants. Therefore, we limited our study to a nonrandomized, double-blind control design (with no appropriate control in the same HB vaccination group). For this reason, this study was not registered in the clinical trial registry. The study was approved by the Medical Ethics Committee of First Hospital of Jilin University (approval number 2012-098). Participants were informed of the purpose of the study and of their rights to keep information confidential. Written informed consent was obtained from each subject before participation. Patient records/information was anonymized and de-identified before analysis.

Study populations
For this study, all pregnant women were screened for HBV infection to improve the prevention and control of vertical HBV transmission. Pregnant women who tested positive for HBsAg at antenatal clinics in all of the districts and counties of Changchun were encouraged to go to the First Hospital of Jilin University for further HBV transmission prevention. Each pregnant woman was then given an immunization schedule that obstetricians and midwives at 25 cooperative birth hospitals would follow.
Pregnant women would not be included if they met any of the following exclusion criteria: having received antiviral medications or immune-modifying therapies during pregnancy; having coviral infections; and having any immunologically compromised conditions. Infant exclusion criteria were as follows: birth weight lower than 2500 g; and born before 37 weeks of gestational age. The study was carried out from July 2012 to April 2015, and infants were followed-up at 7 and 12 months of age. In total, 918 HBsAg-positive mothers registered and participated in the study, and 28 mothers and 27 infants were excluded due to the various reasons mentioned above. As a result, 871 infants were included in the final analysis. The flow chart depicting the study participants is summarized in Fig. 1.

Data collection
Two doctors were trained to complete the unified questionnaires when the mothers visit at the outpatient clinic. The questionnaires were administered face-to-face; the following sections were covered: socio-demographic data, gestational weeks, smoking and drinking status, family history of hepatitis B, antiviral therapy, and whether they were aware of HBV infection. Mothers were followed-up after delivery through telephone call, including delivery mode, feeding pattern, infant gender, infant weight, gestational age, and HBV vaccination dose with the time of injection. In addition, infants were continued 4 times of follow-up after birth (24 hours, 14 days, 1, and 6 months of age) for HBV vaccination information, and HBV serologic markers were measured at 7 and 12 months of age.

Immunization schedule
A combined passive and active immunization was administered to neonates born to HBsAg-positive mothers. Infants born to HBsAg-positive, HBeAg-negative mothers were injected with 100 IU of HBIG (Hualan Biological Engineering Inc., Xinxiang, China) and 10 mg of the hepatitis B vaccine (Hansenula polymorpha yeast-derived recombinant Hepatitis B vaccine; Dalian Hissen Bio-pharm Co., Dalian, China) within 2 hours after birth, followed by administration of 10 mg of the hepatitis B vaccine at 1 month and 6 months of age. Infants born to HBsAgpositive, HBeAg-positive mothers were injected with 100 IU of HBIG and 20 mg of the hepatitis B vaccine within 2 hours after birth, followed by administration of 20 mg of the hepatitis B vaccine at 1 month and 6 months of age. Administration of HBV vaccine and HBIG in accordance with the described protocol is defined as a timely and adequate injection.

Laboratory examinations
Venous blood samples were collected from pregnant women at approximately 1 month before delivery. Blood samples from infants were collected at the 7-month follow-up visit. All laboratory tests were performed in the central laboratory of the First Hospital of Jilin University. Serum samples from the mother and the infant were examined for 5 markers of HBV, including HBsAg, anti-HBs, HBeAg, anti-HBe, and anti-HBc, by chemiluminescent microparticle immunoassay (CMIA) with Abbott ARCHITECT i2000SR (Abbott Laboratories, North Chicago, IL). HBV DNA was assessed by the Roche Taqman HBV test (Roche Diagnostics, Grenzach, Germany) with a detection range from 8.9 to 109 IU/mL. Liver function tests were performed with a Synchron LXH20 autoanalyser (Beckman Coulter, Brea, CA).

Definition of immunoprophylaxis outcome
Seropositivity for HBsAg was used as the marker of HBV infection in infants. If an infant tested HBsAg-positive, the HBV DNA test was performed later to confirm immunoprophylaxis failure and HBV infection. The nonresponders, low responders, medium responders, and high responders were defined as HBsAgnegative infants with anti-HBs <10 mIU/mL, anti-HBs from 10 to 100 mIU/mL, anti-HBs from 100 to 1000 mIU/mL, and anti-HBs ≥1000 mIU/mL, respectively. We usually take anti-HBs ≥10 mIU/ mL as a seroprotective response to vaccine. In this study, enough anti-HBs titer is still needed to prevent the babies from close contact infection with mothers, so anti-HBs titer ≥100 mIU/mL was considered a protective response, whereas a titer 100 mIU/ mL was considered a low protective response.

Statistical analysis
A database of the mother-to-infant HBV transmission prevention cohort was established by the CapitalBio Life Science Data Management System (CB-LDMS-Sample). When the descriptive values were of normal distribution, t tests or nonparametric tests were used and data were expressed as mean ± standard deviation (SD). The Chi-square or Fisher exact test was used for categorical variables expressed in terms of n (%). Risk factors for immunoprophylaxis failure and low protective vaccine responses were analyzed by unconditional logistic regression with adjustment for possible confounders. The odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to estimate risk. All statistical analyses were performed with the Statistical Package for Social Science (SPSS) for Windows version 18.0 (SPSS Inc., Chicago, IL). All P values were 2-tailed and a P value < 0.05 was considered statistically significant.

Basic characteristics of mothers and infants
A total of 863 mothers and their corresponding 871 infants (8 pairs of twins) were included in the final analysis. Of these,

Factors associated with immunoprophylaxis failure
As HBV transmission to infants born to HBeAg-negative mothers was all successfully prevented, we further analyzed factors related to immunoprophylaxis failure in infants born to HBeAg-positive mothers. The results of the univariate and multivariate logistic regression analyses are summarized in Table 3. After adjustment for other risk factors, multivariate regression analysis showed that mothers with higher HBV DNA levels (≥10 8 IU/mL; Adjusted OR = 4.53, 95% CI: 1.19-17.34) and inadequate initial injections (Adjusted OR = 7.69, 95% CI: 1.71-34.59) were independently associated with an increased risk for immunoprophylaxis failure. Besides, nontimely initial injections (over than 2 hours) reached a borderline significant (Adjusted OR = 4.14, 95% CI: 1.00-17.18, P = 0.050). Other factors, including delivery mode, feeding pattern, infant gender, and birth weight, were not significantly associated with maternal HBV transmission.

Factors correlated with the infant immune response to vaccination
We divided 855 HBV-negative infants into 2 groups. One group, the responder group, included 825 (96.5%) infants with anti-HBs titers ≥100 mIU/mL after completion of a series of 3 hepatitis B vaccine injections. The second group consisted of infants whose anti-HBs titers did not reach the protective level. Possible factors related to low immune responses to vaccination are summarized in

Adverse events and safety
All of the infants who received the recombinant hepatitis B vaccine and HBIG were followed-up. No serious adverse events (SAEs) were reported, and only 36 (4.1%) of the newborns had temporary redness and fever following the first dose of the vaccine, but the symptoms resolved on their own without special treatment. No difference in the number of adverse reactions was observed between the 10 and 20 mg hepatitis B vaccine dose groups (4.4% vs 3.6%, P = 0.557).

Discussion
In this cohort study, infants born to HBsAg-positive mothers with different HBeAg statuses were given different immunization Table 1 Basic characteristics of mothers and infants in the HBeAg-negative and HBeAg-positive groups. schedules, and the results were encouraging. None of the infants born to HBeAg-negative mothers were infected with HBV. All the immunoprophylaxis failure infants were born to mothers positive for both HBsAg and HBeAg, with a immunoprophylaxis failure rate of 5.2% observed. Maternal high HBV DNA levels and HB vaccine dose insufficiency along with delayed first injection were significantly associated with MTCT. Furthermore, our results showed that the majority of infants (96.5%) achieved protective level of antibodies at 7 months of age. Relatively low birth weight <3000 g in full-term babies was associated with weak response to vaccination. Different from previous studies, we used 2 different doses of hepatitis B vaccine (20 and 10 mg) to the newborns based on their mothers' HBeAg status instead of administered with 1 same dose of vaccine. Although an appropriate postnatal hepatitis B vaccine combined with HBIG was developed to prevent MTCT in 1992 in China, however, before the year of 2014, the HBV vaccination policy in Changchun city recommends that all newborn infants receive only 5 mg hepatitis B vaccine within 12 hours of birth and again at 1 and 6 months of age, and no additional prevention measures were implemented to mothers who were HBsAg carriers. This old strategy did not take the HBeAg status and HBV DNA level of the pregnant women into consideration. Given this background, it is very important to provide an improved immunoprophylaxis protocol to prevent MTCT of HBV. This benefit engineering was unique for that it provided an individual schedule to protect these in high-risk infants of HBV infection.

Characteristics
This study showed that immunoprophylaxis was successful for the majority of infants with only an overall MTCT rate of 1.84% Table 2 HBV marker test results for infants at 7 months of age.

HBV markers
No. (%) of infants born to HBeAg-negative mothers who were positive for this marker (N = 565) No. (%) of infants born to HBeAg-positive mothers who were positive for this marker (N = 306) P * HBsAg 0 (0%) 16   (16/871), which was lower than previous reports with failure rates that ranged from 2.46% to 5.44%. [18][19][20] Larger hepatitis B vaccine doses and timely vaccination as well as high cesarean rate might contribute to the low infection rate in our study. However, a higher immunoprophylaxis failure rate (5.2%) was observed in infants born to HBeAg-positive mothers. In a study by Kubo et al, [9] among 624 infants born to 481 HBeAg-positive mothers, the overall infant HBV infection rate was 3.37%, which is lower than our results. Use of antiviral medications for parts of higherrisk populations in their study explains their low infection rates. In our study, to evaluate immunoprophylaxis efficacy only, we excluded all the mothers who had taken antiviral drugs during pregnancy. The analysis of factors related to immunoprophylaxis failure in infants born to HBeAg-positive mothers revealed that high HBV DNA levels, insufficient hepatitis B vaccine doses, and delayed first injections were significantly associated with immunoprophylaxis failure. In our cohort, all the 16 blocking failure infants were born to HBeAg-positive mothers with HBV DNA >4 Â 10 7 IU/mL. After adjusting for other factors, HBV DNA was shown to be an independent risk factor for MTCT. This result is in accordance with previous data, which suggested that maternal HBV DNA >7-8 log 10 copies/mL was significantly correlated with perinatal HBV transmission. [17,21] Maternal HBV DNA >2 Â 10 7 IU/mL level had been recommended by AASLD (2009) and APASL (2012) for antiviral treatment during pregnancy. [22,23] Our results showed that combined HBIG and hepatitis B vaccine immunoprophylaxis strategy is effective for HBeAgnegative mothers and for HBeAg-positive mothers with HBV DNA < 4 Â 10 7 IU/mL. Additional prophylaxis measures, such as late-pregnancy antiviral treatments, may be needed for HBeAgpositive mothers with high HBV viral load.
It was also found that vaccine doses and injection time are another 2 risk factors that are associated with maternal HBV transmission among infants born to HBeAg-positive mothers. The Chinese Society of Hepatology and Chinese Society of Infectious Diseases and Chinese Medical Association recommended that newborns of HBsAg positive mothers administering 10 mg recombinant yeast derived HB vaccine combined with HBIG 100 IU within 24 hours after birth (preferably within 12 hours), followed by 2 additional HB vaccine at 1 and 6 months, respectively. [24] In most studies, a range of 5 to 20 mg of hepatitis B vaccine was given to newborns. [17,25] In our study, the hepatitis B vaccine dose recommended for infants born to HBeAg-positive mothers was 20 mg, which was larger than the previous. Our results showed that this immunoprophylaxis was successful with a low infection rate and HB vaccination was well tolerated. However, immunoprophylaxis failure still occurred despite the use of combined HB vaccine and HBIG. Our data demonstrated that insufficient hepatitis B vaccine doses (<20 mg) were significantly associated with immunoprophylaxis failure (OR = 7.69, 95% CI: 1.71-34.59). Similarly, the Centers for Disease Control and Prevention (CDC) and the Advisory Committee on Immunization Practices (ACIP) also recommend that the HBV Table 4 Possible factors related to low immune responses (anti-HBs 100 mIU/mL) to vaccination (N = 855).  vaccine and HBIG be administered to at-risk infants within 12 hours of delivery. [26] If their mothers are positive for both HBsAg and HBeAg, HBIG should be administered to newborns as soon as possible. [27,28] However, there have been a variety of schedules for the initial passive-active immunization of the newborn, which vary anywhere from 3 to 24 hours after birth. [15,17,[29][30][31][32] Previous data from liver biopsies indicated that more than 2 hours is required for HBV to infect healthy liver cells [33] ; therefore, the first injection of HBIG and the hepatitis B vaccine must be administered as early as possible after birth. [11] In our study, the first dose of hepatitis vaccine and HBIG was required to be administered simultaneously within 2 hours after birth. The result showed that delayed initial injection of vaccine was correlated with immunization failure, which reach a borderline significant. [34] HBeAg, anti-HBe, and anti-HBc pass through human placenta from mother to fetus and gradually disappear over time. [35] In our study, 4 infants were HBeAg-positive at 7 months of age, but by 12 months of age, HBeAg was not detectable in these infants. The presence of anti-HBe before 12 months of age and the presence of anti-HBc before 24 months of age in infants may represent transplacental maternal antibodies rather than indicators of HBV infectious status. [35] Therefore, HBV markers should not be evaluated in newborns immediately after birth, rather the HBV infection status of infants should be evaluated at 7 months of age and should continue thereafter to determine whether a chronic infection persists.
Because adequate primary vaccination provides long-term protection, [36,37] it is necessary and important for infants to obtain adequate hepatitis B surface antibody titers at primary immunization to protect them from HBV infections and HBVrelated diseases caused by potential future exposures. In this study, all the infants received the corresponding planned immunoprophylaxis protocol with the same brand of vaccine, the majority of the infants achieved a protective level, while still 3.5% of the infants became low immune responders. In this study, some factors that may be associated with low responsiveness to the HBV vaccine were analyzed, such as maternal HBV DNA, delivery mode, feeding pattern, infant gender, gestational age, birth weight, and vaccine dose. Besides, genetic factors such as the HLA group were also reported to be associated with non/ low response to HB immunization. [38,39] We found that only birth weights <3000 g was an independent predictive factor for weak response. Because the hepatitis B vaccine doses used in this study were greater than the doses used in most of the previous studies, and infants born prematurely weighing <2500 g may lead to incomplete immune fuction development, [40] the vaccinerelated safety may not be guaranteed; for this reason, only fullterm infants with birth weights ≥2500 g were included in our study. The above results may be more significant if pre-term infants weighing <2500 g were also included. Therefore, it is essential to pay close attention to those low birth weight infants for the anti-HBs level, and hepatitis B booster vaccinations should also be considered when necessary.
This study is more convincing by the large sample size and accurate data collection. However, some limitations should be acknowledged. First, our observational cohort study was limited by the evaluation of the efficacy of immunization strategies; therefore, rigorous designed randomized controlled trials are needed in the future. Second, we can define the MTCT of our objects, but we have no evidence to confirm when the transmission actually occured (intrauterine infection, natal transmission, or postnatal transmission).
In conclusion, our study showed that this improved immunization is effective for preventing perinatal transmission of HBV, particularly for mothers who are HBeAg-negative. Maternal HBV DNA level and inadequate and delayed initial injections were the independent risk factors that predicted immunoprophylaxis failure. The majority of the infants in this study reached adequate protective antibody titers after immunoprophylaxis. A relatively low birth weight of <3000 g for full-term infants was associated with a weak response to vaccination. It is noteworthy that we are keeping on following up these children until 3 years old, and their long-term immunological outcome will be evaluated in the further research.