IgE‐reactivity profiles to allergen molecules in Russian children with and without symptoms of allergy revealed by micro‐array analysis

Abstract Background The analysis of longitudinal birth cohorts with micro‐arrayed allergen molecules has provided interesting information about the evolution of IgE sensitization in children. However, so far no cross‐sectional study has been performed comparing IgE sensitization profiles in children with and without symptoms of allergy. Furthermore, no data are available regarding molecular IgE sensitization profiles in children from Russia. Methods We recruited two groups of age‐ and gender‐matched children, one (Group 1: n = 103; 12.24 ± 2.23 years; male/female: 58/45) with symptoms and a second (Group 2: n = 97; 12.78 ± 2.23 years; male/female: 53/44), without symptoms of allergy according to international ISAAC questionnaire. Children were further studied regarding symptoms of allergy (rhinitis, asthma, atopic dermatitis) according to international guidelines, and skin prick testing with a panel of aeroallergen extracts was performed before sera were analyzed in an investigator‐blinded manner for IgE specific to more than 160 micro‐arrayed allergen molecules using ImmunoCAP ISAC technology. Results IgE sensitization = or >0.3 ISU to at least one of the micro‐arrayed allergen molecules was found in 100% of the symptomatic children and in 36% of the asymptomatic children. Symptomatic and asymptomatic children showed a comparable IgE sensitization profile; however, frequencies of IgE sensitization and IgE levels to the individual allergen molecules were higher in the symptomatic children. Aeroallergen sensitization was dominated by sensitization to major birch pollen allergen, Bet v 1, and major cat allergen, Fel d 1. Food allergen sensitization was due to cross‐sensitization to PR10 pollen and food allergens whereas genuine peanut sensitization was absent. Conclusion This is the first study analyzing molecular IgE sensitization profiles to more than 160 allergen molecules in children with and without symptoms of allergy. It detects similar molecular IgE sensitization profiles in symptomatic and asymptomatic children and identifies Bet v 1 and Fel d 1 as the predominant respiratory allergen molecules and PR10 proteins as the major food allergens and absence of genuine peanut allergy in Moscow region (Russia).


| INTRODUC TI ON
Molecular allergology utilizes allergen molecules for improving the diagnosis of allergy and serves as a basis for the generation of new molecular allergy vaccines. 1,2 It has started with the isolation of the first allergen-encoding DNA sequences [3][4][5] and the first demonstration of the potential usefulness of recombinant allergen molecules for allergy diagnosis approximately 30 years ago. 6,7 Today, molecular allergy diagnosis is considered an important part of routine allergy diagnosis. 8,9 A major step in molecular allergy diagnosis was the development of multiplex allergy tests which are based on chips containing a large and comprehensive panel of micro-arrayed allergen molecules which allow testing for IgE reactivity to multiple allergen molecules with small amounts of serum or other body fluids. 10 Due to the fact that allergen micro-arrays require only small volumes of serum, they are ideally suited for the assessment of IgE sensitization profiles in children where blood sampling and in vivo provocation testing may be challenging. 11 Several studies have demonstrated the usefulness of micro-array-based determination of IgE sensitization profiles in children of which a few may be mentioned. For example, the evolution of the IgE reactivity profiles to individual grass pollen allergen molecules has been studied in birth cohorts providing important information regarding the development of IgE sensitization profiles in early childhood. 12,13 Using recombinant food allergen molecules, it has become possible to identify risk allergen molecules which may predict severity of food allergy reactions in children. 14,15 The deciphering of molecular IgE sensitization profiles using micro-arrayed allergen molecules has been shown to be valuable for the personalized management of children suffering from multiple sensitizations 16,17 and for refining the prescription of allergen-specific immunotherapy, allergen avoidance or diet. 18,19 Furthermore, molecular diagnosis seems to be useful for the assessment of respiratory allergies such as rhinitis and asthma [20][21] as well as for certain types of food allergy, for example, oral allergy syndrome which is caused by cross-reactivity between sensitizing pollen allergens and food allergens. 22,23 Interestingly, birth cohort studies indicate that it may be possible to predict, if a yet asymptomatic child may develop symptomatic allergy later in life based on early IgE sensitization profiles to micro-arrayed allergen molecules. 24 The assessment of maternal IgG antibodies transmitted by pregnant women to their offspring and determination of the IgE sensitization in the children thereafter has provided evidence that maternal IgG may protect children against allergic sensitization after birth. 25 However, micro-arrayed allergen molecules have opened yet another important field in allergy research, the assessment of regional molecular IgE sensitization profiles in populations from different parts of the world revealing interesting peculiarities of sensitization profiles. 26 For example, house dust mite sensitization is very common in certain European countries whereas it is rare in other European countries due to climate reasons. 24,27 The molecular profiling of IgE sensitization patterns is very important for the development of allergen-specific treatments and prevention. Unfortunately, such data are not available for certain large parts of the world such as Asia with only one study indicating lack of clinically relevant pollen allergen sensitization in the tropical climate of the Philippines. 28 No assessments of molecular IgE sensitization profiles have yet been published for children from other large areas of the world such as North America, Africa and Russia.
Our study is the first detailed analysis of molecular IgE sensitization profiles in a cohort of Russian children from the Moscow region using more than 160 micro-arrayed allergen molecules. Another unique aspect of this study is that we have enrolled two groups of age-and gender-matched children, one group of children who according to the internationally accepted ISAAC questionnaire 29 exhibited symptoms of allergy and one group in which children did not suffer from any symptoms of allergy. This study design allowed us to investigate the molecular IgE sensitization profiles in asymptomatic children and to compare them with those of symptomatic Conclusion: This is the first study analyzing molecular IgE sensitization profiles to more than 160 allergen molecules in children with and without symptoms of allergy. It detects similar molecular IgE sensitization profiles in symptomatic and asymptomatic children and identifies Bet v 1 and Fel d 1 as the predominant respiratory allergen molecules and PR10 proteins as the major food allergens and absence of genuine peanut allergy in Moscow region (Russia). allergic children. Our results reveal a unique IgE sensitization profile of Russian children from the Moscow region to respiratory and food allergens, which is similar in symptomatic and asymptomatic children but differs in the two groups regarding frequencies and intensities (ie, IgE levels) of IgE recognition of the allergen molecules.

| Characterization of children with and without symptoms of allergy
In this study, 200 children attending the National Research Center-Institute of Immunology Federal Medical-Biological Agency of Russia, Moscow, Russia, with their parents were enrolled with the goal to establish two equally sized, age-and gender-matched groups of children. Permission from the local ethics committee and written informed consent from the parents were obtained. Then, a stepwise assessment was performed comprising:

| ISAAC questionnaire-based assessment
The parents were asked to fill out the "International Study of Asthma and Allergies in Childhood (ISAAC)" questionnaire, 29 which is the most frequently internationally used questionnaire to establish if a subject suffers from symptoms of IgE-associated allergy. The ISAAC questionnaire is available in multiple languages. We used an ISAAC version which had been translated into Russian language. Evidence for asthma was defined by a positive answer to the following questions, either "Has your child ever had wheezing or whistling in the chest at any time in the last 12 months?" or "In the last 12 months, has your child had a dry cough at night, apart from a cough associated with a cold or chest infection?" Allergic rhinitis was suspected after a positive answer to one of the following questions "Has your child ever had any of the following symptoms for at least one hour on most days (or on most days during the season if your symptoms are seasonal): watery, runny nose, sneezing (especially severe or even bouts of sneezing), nasal obstruction, nasal itching, conjunctivitis (red, itchy eyes), postnasal drip?" or "Have you ever had allergic rhinitis?" Questionnaire-based evidence for atopic dermatitis required a positive answer to one of the following questions "Has your child ever had atopic dermatitis?" or "Has your child ever had an itchy rash that was coming and going for at least six months?" Based on the results from the ISAAC questionnaire, two groups of children were formed, one group (n = 103) with symptoms of allergy and a second group (n = 97) without symptoms of allergy ( Figure 1).

| Clinical assessment
After the assignment of children in the symptomatic and asymptomatic group, a further clinical assessment of the children was performed, which included a detailed case history and a thorough physical examination ( Figure 1). The clinical diagnosis of allergic rhinitis was based on recommendations by the European Academy of Allergy and Clinical Immunology 30 and ARIA guidelines. 31 The diagnosis of asthma was performed according to guidelines of the Global Initiative for Asthma/Global Strategy for Asthma Management and Prevention (available from: www.ginas thma.com). 32 Atopic dermatitis was diagnosed based on international guidelines. 33 In addition to ISAAC questionnaire-based assessment, food allergy was assessed by an additional questionnaire checking symptoms of oral allergy syndrome (eg, pruritus of the lips, tongue, oral mucosa, burning sensations of the tongue, swelling of the lips or the tongue, swelling of the oral mucosa, laryngeal swelling, inflammation of the tongue or of the oral mucosa, perioral skin symptoms) urticaria, wheezing, dyspnea, nausea/vomiting, and gastrointestinal disorders that are associated with ingestion of common food allergen sources. The diagnosis of birch pollen-related oral allergy syndrome was based on a validated questionnaire approach. 34 This questionnaire included questions regarding pruritus of the lips, tongue, oral mucosa, burning sensations of the tongue, swelling of the lips or the tongue, swelling of the oral mucosa, laryngeal swelling, inflammation of the tongue or of the oral mucosa, perioral skin symptoms, wheezing, dyspnea, nausea/vomiting, and gastrointestinal disorders which were associated with ingestion of apple, peach, carrot, nuts, or other fruits and vegetables. The questionnaire-based assessment of OAS has been reported to have comparable diagnostic accuracy compared to oral provocation testing. 34

| Skin prick testing
All children were then subjected to skin prick testing according to current guidelines using a panel of allergen extracts from respiratory

| Micro-array-based determination of IgE reactivity to more than 160 allergen molecules
MeDALL allergen chips which had been developed based on the ImmunoCAP ISAC technology (Thermofisher, Phadia, Uppsala, Sweden) as described were used for measurement of specific IgE in sera from both groups of children ( Figure 1). 36  slightly superior sensitivity in direct comparison to ImmunoCAP due to low background signals and a sensitive, fluorescence-based detection system. 36 Intra-assay variation was determined to be approximately 8% in the IgE range of 1-130 ISU-IgE. 36

| Statistical analysis
Results (IgE levels) are given in medians, absolute (n), or relative numbers (%), where appropriate. The level of allergen-specific IgE for each positive allergen molecule was compared between groups using the nonparametric U test. A P value of ≤.05 was considered statistically significant. Data were analyzed by using IBM SPSS Statistics.

| Demographic and clinical characterization of children with and without symptoms of allergy
Based on ISAAC questionnaire data, we recruited 103 children (Group 1:58 males, 45 females) with a median age of 12.24 years who reported symptoms of allergy and a second group comprising 97 children (53 males, 44 females) with a median age of 12.7 years without symptoms of allergy (Table 1). Thus, the symptomatic and asymptomatic group of children were well-matched regarding age and gender distribution. According to ISAAC and further detailed clinical assessment, rhinoconjunctivitis was by far the most common symptom among the 103 symptomatic children (n = 88) followed by atopic dermatitis (n = 55), asthma (n = 40), and food allergy (n = 35). Oral allergy syndrome (OAS) (n = 34) was by far the dominating symptom of food allergy followed by urticaria (n = 6) and anaphylaxis (n = 1) ( Table 1)

| IgE sensitizations to major birch pollen allergen Bet v 1 and cat-derived Fel d 1 dominate the hierarchy of aeroallergen sensitization
Sensitization to aeroallergens was dominated by two major allergen molecules, major birch pollen allergen Bet v 1 (Group 1:63.1%; group 2:25.7%) and the major cat allergen Fel d 1 (Group 1:61.1%; group 2:15.4%) ( Figure 2A; Table S1). In the symptomatic children, the following allergen molecules which are indicative for certain allergen sources were next in the hierarchy: major TA B L E 1 Demographic and clinical characterization of children with and without symptoms of allergy from Moscow region, Russia (n = 200) were recognized by approximately 10% of the symptomatic children ( Figure 2A; Table S1).

| Children with symptoms of allergy show similar molecular IgE sensitization profiles to aeroallergens as children without symptoms of allergy: frequencies and IgE levels are different
The hierarchy of aeroallergen molecules recognized by asymptomatic children was very similar to that of symptomatic children ( Figure 2B). Again major birch pollen allergen, Bet v 1 (25.7%),  and major cat allergen Fel d 1 (15.4%) were the by far most frequently recognized allergens (Figure 2A,B; Table S1) but allergen-specific IgE levels were significantly lower in asymptomatic children as compared to the symptomatic children (Table S1). The further hierarchy of IgE recognition was also similar to that of the symptomatic children Phl p 1 (9.2%) > Art v 1 (6.2%) = Cry j 1 (6.2%) > Cup a 1 (5.1%) = Pla a 2 (5.1%), and again allergen-specific IgE levels were significantly lower than in the symptomatic children (Table S1). Thus, children without symptoms had similar IgE recognition profiles as compared to symptomatic children but specific IgE levels and sensitization rates were lower.

| Food allergen sensitization is dominated by cross-reactivity to PR10 allergens, whereas genuine peanut allergy is rare
IgE sensitization to food allergens was dominated by PR10 proteins which cross-react with the major birch pollen allergen Bet v 1. 38 The major hazelnut allergen Cor a 1 (52.4%) and the major apple allergen Mal d 1 (51.4%) were the most frequently recognized PR10 allergens in the symptomatic children followed by Ara h 8 (peanut), Pru p 1 (peach), Gly m 4 (soybean), Api g 1 (celery), and Act d 8 (kiwi) (Table S2; Figure 3A). IgE sensitization to the PR10 allergens dominated also in the children without symptoms but fewer children were sensitized and specific IgE levels were significantly lower as in the symptomatic children (Table S2; Figure 3B).

| Indoor allergen sensitization is dominated by cat followed by dog, whereas HDM is less common and has an unusual recognition profile
Besides the major birch pollen allergen, Bet v 1, the major cat allergen Fel d 1 was the most frequently recognized allergen in our   Figure 4D). After Fel d 1, Fel d 2 (19.4%), Fel d 7 (12.6%), and Fel d 4 (11.6%) followed as next frequently recognized cat allergen molecules in group 1 (Table S1; Figure 4D). IgE levels specific for Fel d 7 and Fel d 4 were higher than those for Fel d 2 ( Figure 4D; Table S1).
Usually, up to 50% of allergic subjects are sensitized to HDM allergen but in our population HDM allergy was quite rare because Der p 2 the most frequently recognized allergen showed IgE reactivity only with 9.7% of children from group 1 (Table S1). Furthermore, the frequencies of IgE recognition of the individual HDM allergen molecules among symptomatic children were unusual because Der p 7 > Der p 5 > Der p 21 were more frequently recognized than Der p 1 and Der f 1 which are usually major HDM allergens for more than 90% of HDM-allergic patients ( Figure 4E).
IgE sensitization to mold allergens was rather rare in our population and the profile was also unusual. For example, Alt a 6 was more often recognized by IgE than the major Alternaria allergen Alt a 1 ( Figure 4F).

| IgE recognition of other allergen molecules
None of our children showed IgE reactivity to any of the 5 tested   symptomatic children, and only one child from the asymptomatic group 2 showed MUXF 3-specific IgE reactivity (Table S3).

| D ISCUSS I ON
Our study is the first to perform a meticulous analysis of IgE reactivity profiles to more than 160 allergen molecules in children from Russia (Moscow region) using micro-arrayed allergen molecules. A unique feature of our study is that we have recruited two age-and gender-matched groups of children, each approximately 100, one with symptoms of allergy and one without symptoms of allergy.
For the assignment of children to the symptomatic group 1 and the asymptomatic group 2, we used the internationally standardized ISAAC questionnaire which is established all over the world in different languages. 29 Our study was designed to include a symptomatic and an asymptomatic group of children to investigate if and how many asymptomatic children show IgE sensitizations to certain allergen molecules and if so, what the differences of molecular IgE sensitizations between the two groups may be. According to the analysis of molecular IgE sensitization profiles with micro-arrayed allergen molecules in population-based birth cohort studies, it seems that 50%-60% of children exhibit IgE sensitizations against at least one of the >160 allergen molecules present on the allergen microarray. 24 In fact, we found that each of the children with symptoms of allergy (ie, 100%) showed IgE reactivity to at least one of the allergen molecules analyzed. Interestingly, 36% of the asymptomatic children also displayed IgE reactivity to at least one of the tested allergen molecules which indicates that more than 30% of persons without symptoms of allergy have a specific IgE sensitization which is in agreement with the results from population-based birth cohorts. The clinically silent IgE sensitization was also suggested by the positive reactions obtained by skin testing in the asymptomatic children (Table 1).
We know that certain "allergens" have a very low allergenic activity and hence do not induce allergic symptoms in sensitized subjects such as IgE-reactive carbohydrates. 40,41 One possible result of the comparison of the molecular IgE sensitization profiles in symptomatic and asymptomatic children could therefore have been that asymptomatic children are primarily sensitized against such low allergenic molecules. Such a result was actually obtained for a population from the Philippines where we detected a high rate of carbohydrate sensitizations in asymptomatic subjects. 28 However, it is one of the major findings of our study that both symptomatic and asymptomatic children were sensitized against the same allergen molecules. The only difference between the groups was that fewer of the asymptomatic children were sensitized and that allergen-specific IgE levels were significantly lower among the asymptomatic children. This result underlines the importance of allergen-specific IgE levels for the development of symptoms. We have not conducted a longitudinal assessment of the asymptomatic children to study if increases of allergen-specific IgE levels as they occur after frequent allergen contact may be associated with the development of symptoms.
However, the longitudinal analysis of children in birth cohorts has shown that with increasing age more children become symptomatic and that this is associated with increases of allergen-specific IgE levels. 13 It is therefore conceivable that children from the asymptomatic group may become symptomatic later in life when their allergen-specific IgE levels increase.
Another major result of our study is that it is the first to estab-  (Table 1). This result has important implications for allergen-specific forms of treatment because allergen-specific immunotherapy (AIT) with a vaccine conferring cross-protection to Bet v 1 and the cross-reactive PR10 food allergens would be needed to treat birch pollen and associated food allergy in our population. 44 Other forms of food allergy were quite rare. We found that bovine serum albumin, Bos d 6, which may be considered as cow´s milk allergen, was frequently recognized.
However, the frequent sensitization to Bos d 6 seems to be due to a In summary, our study is the first cross-sectional study to compare molecular IgE sensitization profiles to a large number of allergen molecules in symptomatic and asymptomatic children in an as yet not investigated part of the world. The study design employed by us will be useful also for other populations and areas in the world.
The obtained results have important implications for allergen-specific forms of treatment and prevention which can be implemented in regional allergy treatment and prevention programs especially in childhood.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/pai.13354.