High prevalence of IgE sensitization to inactivated influenza vaccines, yet robust IgG4 responses, in a healthy pediatric population

Abstract Background Anaphylaxis following influenza vaccination is a rare but serious problem. The underlying immune responses are not well understood. This study elucidated the IgE and IgG antibody responses in healthy children and adolescents following inactivated influenza vaccines (IIVs). Methods The efficacy and safety of quadrivalent IIV (QIV) and trivalent IIV (TIV) were compared in healthy subjects aged 0–18 years. Serum IIV‐specific IgE, IgG, and IgG4 levels (sIgE, sIgG, and sIgG4) were measured with ImmunoCAP. Hemagglutination inhibition (HI) assay was performed for each influenza virus subtype. Sera from earlier patients who developed anaphylaxis to different IIVs were similarly tested. Results A total of 393 subjects were enrolled: 96 were 6 months−2 years old, 100 were 3–5 years old, 100 were 6–12 years old, and 97 were 13–18 years old. No anaphylaxis was observed. Generally, QIV and TIV induced similar antibody responses. IIV‐sIgE levels rose significantly after vaccination in the 6 months–2 years old and 3–5 years old groups, did not change in the 6–12 years old group, and decreased in the 13–18 years old group. In contrast, the IIV‐sIgG4/sIgE ratio increased significantly after vaccination in all age groups. Sensitized subjects had significantly higher HI titers and IIV‐sIgG levels in the youngest age group and higher IIV‐sIgG4 levels in all age groups compared with the non‐sensitized. The IIV‐sIgG4/sIgE ratio in five patients with anaphylaxis was significantly lower than in age‐matched healthy subjects. Conclusion IIVs induce IgE sensitization in healthy children but also robust IgG4 responses that may protect them from anaphylaxis.


| INTRODUCTION
Immunization programs are vital public health interventions to reduce the disease burden caused by infectious pathogens. Nonetheless, growing hesitation on immunization because of adverse reactions threatens to hamper future immunization programs. Inactivated influenza vaccines (IIVs) used to control and manage seasonal influenza can cause minor and severe adverse reactions, the latter serving as a contraindication to future immunization. One of the severe adverse events (AEs) that can occur following influenza vaccination is anaphylaxis, a rare and life-threatening immediate hypersensitivity reaction. 1 Influenza vaccines are administered to a large number of people annually, and the cumulative incidence of anaphylaxis has been the highest among all vaccines. 2 Previously, egg allergy was thought to represent a high risk for anaphylaxis because vaccine strains are propagated in the allantoic cavity of embryonated chicken eggs and the final products contain a trace amount of egg proteins that might cause an egg-allergic patient to develop anaphylaxis; however, cumulative evidence now clearly demonstrates the safety of IIV for individuals with egg allergy of any severity. 3 Instead, we found that hemagglutinin (HA) protein, the active ingredient in IIV, was a causal allergen of anaphylaxis. 4 During a surge in IIV-associated anaphylaxis (IVA) in young children aged 3-8 years old in Japan in the 2011-2012 influenza season, we observed high IgE responses to HA in IVA patients but not in healthy or egg-allergic patients without any AEs. 4 This suggests that production of IgE antibodies to IIV may be a central issue.
To date, studies on the mechanism of IVA and the presence of IgE antibodies following influenza vaccination have been few and are mostly case-series or small-size studies. [5][6][7][8] In addition, immune responses to vaccination that may be related to anaphylaxis are not well understood. Therefore, there is a need for further, large-scale studies to evaluate IgE sensitization and immune responses to influenza vaccines that might shed light on the mechanism of IVA and lead to the production of safer vaccines.
The present study was a multicenter clinical trial that evaluated the IgE and IgG antibody responses to IIVs in healthy Japanese children and adolescents.

| Vaccines
The vaccines used in this study were trivalent IIV (TIV) and quadrivalent IIV (QIV). The IIVs were produced by inoculating the virus into the allantoic cavity of embryonated hen eggs and subsequently inactivated with formaldehyde. They were produced at four locations: the Chemo-Sero-Therapeutic Research Institute (currently, KM Biologics Co., Ltd.; Kumamoto, Japan), Daiichi Sankyo Co., Ltd. (Tokyo, Japan), the Research Foundation for Microbial Diseases of Osaka University (Osaka, Japan), and Denka Co., Ltd.

| Clinical study design
This study was designed (1) to compare the immunogenicity and safety of TIV and QIV because of their difference in protein content (more protein in QIV) and (2) to investigate the IgE and IgG responses related to allergic sensitization to the vaccines (UMIN000015324). The study was conducted at 11 centers located throughout Japan during the 2014-2015 influenza season. The study was approved by the ethics committee of Mie National Hospital and conducted in compliance with the study protocol (approval numbers: 26-13 and 27-11). Guardians of all subjects provided informed consent. Healthy volunteers aged between 0 and 18 years were enrolled in the study and categorized into four age groups: 6 months-2 years, 3-5 years, 6-12 years, and 13-18 years. The subjects in each age group were randomized to receive a subcutaneous injection of either TIV or QIV at a 1:2 allocation ratio. Subjects younger than 13 years old received two doses of either TIV or QIV at a 4-week interval, whereas those in the 13-18 years old group received a single dose of either vaccine.

| Sample collection and antibody measurement
Blood samples were collected before the first dose and 4 weeks after each dose. Influenza vaccine-specific IgE (sIgE), IgG (sIgG), and IgG4 (sIgG4) were measured using the ImmunoCAP ® assay system (Phadia AB, Uppsala, Sweden). QIV, as an influenza vaccine antigen, was covalently coupled to the activated solid phase through the amino groups of the proteins according to standard ImmunoCAP methodology. The data were expressed as kilounits per liter (kU A /L) for specific IgE and milligrams per liter (mgA/L) for specific IgG and IgG4. IgE positivity was defined as an IgE titer >0.1 kUA/L, which is the lower limit of detection and is considered to have the potential to cause hypersensitivity reactions, such as anaphylaxis. The hemagglutination inhibition (HI) assay 9 was performed to test for the presence of putative protective antibodies against the influenza virus strains contained in the vaccines.

| Safety
Local and systemic AEs were recorded by the subjects' parents in a diary within 7 days after each vaccine dose.

| Statistical analysis
Demographic data were compared among the age groups using the Kruskal-Wallis test (continuous data) or chi-square test (categorical data). The Mann-Whitney test was performed to compare antibody levels between two independent groups. Wilcoxon's signed-rank test or Friedman's test with Dunn's post-test were performed to compare changes in antibody levels before and after vaccination.

| Role of the funding sources
The study was funded by unconditional grants from the Chemo-Sero-Therapeutic Research Institute (currently, KM Biologics Co., Ltd.), Daiichi Sankyo Co., Ltd., the Research Foundation for Microbial Diseases of Osaka University, and Denka Co., Ltd. The sponsors had no role in data monitoring, statistical analysis, or data interpretation. The authors were responsible for the study design, data collection, analysis, interpretation of the data, and writing of the report. The authors had complete independence over the conduct, integrity, and publication of the study.

| Characteristics of the subjects
The study enrolled a total of 393 volunteers, about 100 in each age group (Table 1). The gender ratio showed slight male predominance in all age groups. The proportion of the subjects with a history of previous influenza infection was 19.8% in the 6 months-2 years group and increased in the older age groups, up to 69.1% in the 13-18 years group. Subjects who had no previous influenza vaccination were 40% in the 6 months-2 years group and 11%, 7%, and 2% in the three older age groups, respectively.
The number of subjects with physician-diagnosed allergic diseases was similar in each of the four age groups, with asthma at about 30%, allergic rhinitis at 50%, and egg allergy at less than 10% (Table 1).

| Safety
Tables S1 and S2 summarize the AEs that occurred within 7 days after vaccination. Generally, no severe AEs, including anaphylaxis, were observed. There were also no AEs that were immediate allergic reactions, such as urticaria. Despite the concern that QIV would cause more AEs than TIV because of its higher protein content, the incidence of AEs did not differ between TIV and QIV (Table S1). The distribution of AEs differed among the four age groups, as is often experienced in general practice (Table S2). The results were comparable to those in a recent metaanalysis. 10

| Similar immunogenicity of TIV and QIV
Next, we compared TIV and QIV in terms of the HI geometric mean titer (Table S3), geometric mean ratio (Table S4), seroconversion rate (Table S5), and seroprotection rate (Table S6) 11 for each constituent influenza virus strain. The immunogenicities of the vaccines were similar, excluding the response to B/Victoria, which was not contained in TIV, as reported previously. 12 Likewise, the TIV and QIV groups showed no significant differences in either the pre-or postvaccination levels of IIV-sIgE ( Figure S1). The IIV-sIgG and IIV-sIgG4 levels were also similar between TIV and QIV (data not shown). Therefore, in subsequent analyses, we integrated the TIV and QIV data.

| Influenza vaccine-specific IgE antibody
We evaluated the IIV-sIgE level in each age group. The percentage of sensitized children (sIgE > 0.1 kU A /L) increased from 30% to 43% in the 6 months-2 years group following vaccination ( Figure 1A, bottom).
Although the percentage of sensitization did not change after vaccination in the older age groups, more than half of the subjects were found to be sensitized, above 70% in the 3-5 years group, and about 60% and 50% in the 6-12 years and 13-18 years groups, respectively ( Figure 1B-D).
We also observed a significant increase in the IIV-sIgE level after the first and second vaccine doses in the 6 months-2 years group ( Figure 1A). In the 3-5 years group, a significant increase was observed after the second dose ( Figure 1B). No post-vaccination increase was seen in the 6-12 years group ( Figure 1C). Interestingly, the IIV-sIgE level decreased significantly in the 13-18 years group ( Figure 1D).

| Factors associated with IgE sensitization to the influenza vaccines
Logistic regression analysis was used to explore the relative impor-  (Table S7).

| Influenza vaccine-specific IgG4 and the sIgG4/sIgE ratio
We then compared the levels of sIgG4, which could function as a blocking antibody to allergic reactions, between the sensitized and non-sensitized subjects. The IIV-sIgG4 level was significantly higher in the sensitized subjects than in the non-sensitized subjects, not only in the 6 months-2 years group ( Figure 3A), but also in all age groups ( Figure 3B-D). Accordingly, the sIgG4/sIgE ratio increased significantly following vaccination in all age groups ( Figure 4). Together, the influenza vaccines induced robust sIgG4 responses, especially in sensitized subjects, which resulted in an increased sIgG4/IgE ratio in F I G U R E 2 Comparison of HI titer (A-D) and influenza vaccine-specific IgG (E-H) after vaccination between sensitized and non-sensitized subjects in age groups of 6 months to 2 years (A, E), 3 to 5 years (B, F), 6 to 12 years (C, G), and 13 to 18 years (D, H). Gridlines indicate the putative protection level. ****p < 0.0001; ns, not significant; Mann-Whitney test healthy children and adolescents who had no allergic AEs following the vaccination.

| Patients with anaphylaxis had a low IgG4/IgE ratio
Finally, we investigated the sIgG4/sIgE ratio in five patients who developed anaphylaxis following different influenza vaccinations (Table 2) in comparison with age-matched healthy subjects in the present study. The IIV-sIgG4/sIgE ratio in the patient group was significantly lower than in the healthy counterparts ( Figure 5A). The receiver operating characteristic (ROC) analysis showed the area under the curve to be 0.9766 with high specificity and sensitivity at a cut-off of 1.295 ( Figure 5B).

| DISCUSSION
In the present study, we demonstrated that sensitization to IIV was common in a healthy pediatric population and that the IIV-sIgE level Most importantly, the IIV-sIgG4 level was significantly higher in sensitized subjects than in non-sensitized subjects in all age groups, leading to a significant increase in the sIgG4/sIgE ratio following vaccination.
In addition, the sIgG4/sIgE ratio was significantly lower in patients with IVA than in age-matched healthy subjects. These results suggest that IgE sensitization to IIV may be only one aspect of a robust immune response in healthy children, and induction of IgG4 antibodies to IIV may serve as a mechanism to protect them from anaphylaxis.
The presence of IgE antibodies to IIV was first identified in a case-series study. 7 Later, we used ELISA to measure the IgE antibody titers in 117 healthy children following 2011/12 TIV. 5  In our current study, factors associated with IgE sensitization were age, 3-5 years old, and allergic rhinitis. Proportion of IgE sensitization was highest in the age group and decreased with increasing age. Natural desensitization is observed in some allergens such as food 14 and mosquito saliva 15 with unknown mechanisms. Our observation may be analogous to the phenomenon. We assume that allergic rhinitis was detected as a risk factor because it represents atopic predisposition that are susceptible to allergen sensitization. High prevalence of rhinitis in the subject population also contributed to high statistical power.
Because influenza infection often causes asthma exacerbation, the Advisory Committee on Immunization Practices, United States, recommends that all adults and children aged ≥6 months with asthma receive an influenza vaccination annually, 16 even during the COVID-19 pandemic. 17 An important concern had been whether influenza vaccines would provide sufficient immunity in asthma patients, but it F I G U R E 5 (A) Comparison of influenza vaccine-specific IgG4/IgE ratio between earlier patients who developed anaphylaxis after influenza vaccination on a separate occasion and age-matched subjects in the present cohort. Bars indicate the median. ****p < 0.0001; Mann-Whitney test. (B) Receiver operating characteristic (ROC) curve of influenza vaccine-specific IgG4/IgE ratio for diagnosis of anaphylaxis was reported that asthma patients' antibody production was sufficient. 18 However, there have been no reports of comparison of the protective antibody responses between asthma/allergic subjects and their healthy counterparts. In this study, contrary to grim expectations, 13 we found that HI titers following vaccination were significantly higher in vaccine-sensitized young children than in non-sensitized subjects.
These results suggest that production of IgE antibodies to influenza vaccine may be a normal, robust, immune response.
Yet binding of IgE to a vaccine component can cause mast cell/ basophil activation and pose a risk of developing anaphylaxis after vaccination. 4,8 Why did none of our present subjects develop any allergic reactions, despite a high rate of sensitization to the vaccine?
Perhaps it was because the sensitized subjects had higher IIV-sIgG4 levels than the non-sensitized subjects. IgG4 behaves as a monovalent antibody because of dynamic Fab arm exchange 19 ; it thus lacks the ability to cross-link target antigens, has high antigen-binding affinity. 20 IgG4's low affinity for Fc receptors 21 limits its ability to activate inflammatory cells, including mast cells. Collectively, IgG4 antibodies function as "blocking" antibodies against allergic reactions. 22,23 The concurrent elevation of sIgG4 and sIgE seen in this study was expected because immunoglobulin class-switching to IgG4 and IgE is dependent on Th2 cytokines such as IL-4 and IL-13. 24-27 Yet our current results showed that production of sIgG4 exceeded sIgE in our healthy pediatric population, which corresponded to a significant increase in the sIgG4/IgE ratio after vaccination. On the contrary, the patients who developed anaphylaxis following the influenza vaccination had a significantly lower sIgG4/sIgE ratio compared with the healthy subjects. These facts may be attributed to IL-10 and IL-10 producing regulatory T cells 28,29 and regulatory B cells 30 that promote IgG4 production and inhibit IgE production. In fact, we examined cytokine production in whole blood cultures for some of the subjects in this study and observed significant post-vaccination production of IL-10 and IL-4. 31 Importantly, IL-10 production was observed not only in naïve subjects but also in primed subjects (possessing neutralizing antibodies prior to vaccination), whereas IL-4 production was seen only in naïve subjects, 31

CONFLICTS OF INTEREST
The authors declare the following financial interests/personal relationships that may be considered as potential competing interests. Shigeru Suga reports that financial support was provided by Chemo-Sero-

PEER REVIEW
The peer review history for this article is available at https://publons. com/publon/10.1111/irv.13053.

DATA AVAILABILITY STATEMENT
The data supporting the findings of this study are available upon request from the corresponding author. ORCID