University of Groningen
The multifactorial nature of food allergy
van Ginkel, Cornelia Doriene
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2018
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
van Ginkel, C. D. (2018). The multifactorial nature of food allergy. Rijksuniversiteit Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
266 CHAPTER 12
268
PART IV Chapter 12 - Summary and general discussion
267
SUMMARY AND GENERAL DISCUSSION
Chapter 1 provides a brief introduction to food allergy including its extensive impact, both
socially and financially.1,2 In addition, we describe the literature available regarding
environmental factors associated with sensitization to foods and food allergy. Chapter 2 gives
an overview of the literature regarding the genetics of allergic disorders, including food allergy. From both chapters, we can conclude that although food allergy is an important clinical problem, compared to other atopic diseases, it has been studied to a lesser extent. Defining food allergy in large populations is challenging which complicates the identification of both environmental and genetic risk factors.
In Part I of this thesis we discuss the prevalence of food allergy among Dutch adults and
adolescents and identify environmental, behavioral and familial variables associated with food allergy. In the adults from the Lifelines cohort and adolescents from Dutch high schools we classified subjects as “likely to have food allergy” based on questionnaires we developed and interpreted based on an extensive literature search. In addition, we studied the association between food allergy and both breastfeeding and family history. These latter associations were studied in children who had a double-blind placebo-controlled food challenges (DBPCFC) as part of regular tertiary paediatric allergy care because of a history consistent with an IgE mediated reaction after ingestion of a food.
In Chapter 3 we studied the prevalence of both likely and questionable, self-perceived food
allergy in adults of the Lifelines cohort. In addition, we identified risk factors associated with food allergy and studied the association with psychopathology and health-related quality of life (H-RQOL). Of this cohort of 78 890 Dutch adults, 4.0% reported food allergy with a culprit food, symptom and characteristics which we classified as consistent with food allergy. Additionally, 8.1% was classified as having questionable, self-perceived food allergy without these features. Reported asthma, eczema and nasal allergy increased the risk of food allergy. Female sex and an younger age were associated with an increased risk of food allergy as well. In addition, we reported an association between the living environment during childhood and the risk of food allergy in adult life. Adults who lived in a more urban environment during childhood had a higher risk of being food allergic compared to those who lived on a farm during childhood. This confirms a previous study on childhood food allergy in the US3and
indicates that the protective effect of living on a farm continues into adulthood. Potentially, an increased microbial diversity, higher vitamin D levels and less exposure to ambient pollutants could decrease the risk of food allergy.3 In previous studies, exposure to farm milk
was reported to be inversely associated with sensitization to foods in 7 606 children4 and with
higher numbers of regulatory T cells in 298 children.5 This might be due to the consumption
of bovine miRNAs in cow’s milk which are altered by high-heat treatment as applied to
268 commercial milk.6 In addition, early farm exposure was associated with hypomethylation of
the IL-13 and CD14 gene,7,8 indicating epigenetic effects of the early life environment.
We had a special interest in the group with questionable, self-perceived food allergy. These subjects also reported a poorer H-RQOL compared to controls, just like subjects with likely food allergy. Interestingly, they also reported significantly more burn-out and depression compared to controls without food allergy. This was not seen for subjects with likely food allergy and it remains unclear whether this is a cause or consequence of questionable, self-perceived food allergy. Interestingly, previous research indicated that children with suspected food allergy benefit from a positive DBPCFC confirming the food allergy.9 This suggests that subjects suffer from uncertainty regarding their food allergic status
and this may partly explain these findings. But more importantly, although the diagnosis may be questionable from a purely medical perspective, this is a group with a high morbidity which makes them a priority for future research.
In Chapter 4 we studied the prevalence of food allergy in adolescents from Dutch high schools.
In addition, we examined the frequency of epinephrine auto-injector (EAI) prescription among high-risk individuals. Based on a short questionnaire we concluded that 17.3% of the 2 284 adolescents reported so called “problems with food”. We contacted these subjects for a more detailed telephone questionnaire and concluded that 70/120 adolescents considered themselves as being food allergic. However, only 48/120 (40%) were eventually classified as likely to be food allergic based on the reported foods and symptoms. To summarize, the calculated questionnaire-based prevalence of likely and self-perceived but questionable food allergy was 6.2% and 4.0%, respectively. A meta-analysis of this study and 17 other european studies up to 2012 concluded that the overall prevalence of self-reported food allergy was 5.9% with a higher prevalence in children and northern european countries.10 Future studies
with more strict definitions of food allergy including the DBPCFC are indicated to limit bias introduced by false-positive diagnoses of food allergy.
In addition, we show an alarming under-prescription of EAIs since less than 1 in 30 high-risk adolescents had actually been prescribed this potentially life-saving device. This study was performed in 2009 and seven years later, in 2016, this same research was conducted again to study whether the prevalence of food allergy and EAI prescription in adolescents increased. Fortunately, the prevalence of food allergy did not increase but EAI ownership only marginally improved.11 In addition, a recent study with mystery guest visits
concluded that food-allergic patients are often not, or incorrectly instructed by pharmacists on how to use an EAIs.12 These studies indicatethat awareness about food allergy, anaphylaxis
and especially its management still need to be increased.
In Chapter 5 we describe the association between a longer duration of breastfeeding and a
lower risk of food allergy as diagnosed by DBPCFC. No confounders were identified in this association, including parental atopic conditions and early atopic manifestations. This indicates that this protective effect is relevant for children both with and without a positive
Chap
ter 12
269 Summary and general discussion - Chapter 12 PART IV
267
SUMMARY AND GENERAL DISCUSSION
Chapter 1 provides a brief introduction to food allergy including its extensive impact, both
socially and financially.1,2 In addition, we describe the literature available regarding
environmental factors associated with sensitization to foods and food allergy. Chapter 2 gives
an overview of the literature regarding the genetics of allergic disorders, including food allergy. From both chapters, we can conclude that although food allergy is an important clinical problem, compared to other atopic diseases, it has been studied to a lesser extent. Defining food allergy in large populations is challenging which complicates the identification of both environmental and genetic risk factors.
In Part I of this thesis we discuss the prevalence of food allergy among Dutch adults and
adolescents and identify environmental, behavioral and familial variables associated with food allergy. In the adults from the Lifelines cohort and adolescents from Dutch high schools we classified subjects as “likely to have food allergy” based on questionnaires we developed and interpreted based on an extensive literature search. In addition, we studied the association between food allergy and both breastfeeding and family history. These latter associations were studied in children who had a double-blind placebo-controlled food challenges (DBPCFC) as part of regular tertiary paediatric allergy care because of a history consistent with an IgE mediated reaction after ingestion of a food.
In Chapter 3 we studied the prevalence of both likely and questionable, self-perceived food
allergy in adults of the Lifelines cohort. In addition, we identified risk factors associated with food allergy and studied the association with psychopathology and health-related quality of life (H-RQOL). Of this cohort of 78 890 Dutch adults, 4.0% reported food allergy with a culprit food, symptom and characteristics which we classified as consistent with food allergy. Additionally, 8.1% was classified as having questionable, self-perceived food allergy without these features. Reported asthma, eczema and nasal allergy increased the risk of food allergy. Female sex and an younger age were associated with an increased risk of food allergy as well. In addition, we reported an association between the living environment during childhood and the risk of food allergy in adult life. Adults who lived in a more urban environment during childhood had a higher risk of being food allergic compared to those who lived on a farm during childhood. This confirms a previous study on childhood food allergy in the US3and
indicates that the protective effect of living on a farm continues into adulthood. Potentially, an increased microbial diversity, higher vitamin D levels and less exposure to ambient pollutants could decrease the risk of food allergy.3 In previous studies, exposure to farm milk
was reported to be inversely associated with sensitization to foods in 7 606 children4 and with
higher numbers of regulatory T cells in 298 children.5 This might be due to the consumption
of bovine miRNAs in cow’s milk which are altered by high-heat treatment as applied to
268 commercial milk.6 In addition, early farm exposure was associated with hypomethylation of
the IL-13 and CD14 gene,7,8 indicating epigenetic effects of the early life environment.
We had a special interest in the group with questionable, self-perceived food allergy. These subjects also reported a poorer H-RQOL compared to controls, just like subjects with likely food allergy. Interestingly, they also reported significantly more burn-out and depression compared to controls without food allergy. This was not seen for subjects with likely food allergy and it remains unclear whether this is a cause or consequence of questionable, self-perceived food allergy. Interestingly, previous research indicated that children with suspected food allergy benefit from a positive DBPCFC confirming the food allergy.9 This suggests that subjects suffer from uncertainty regarding their food allergic status
and this may partly explain these findings. But more importantly, although the diagnosis may be questionable from a purely medical perspective, this is a group with a high morbidity which makes them a priority for future research.
In Chapter 4 we studied the prevalence of food allergy in adolescents from Dutch high schools.
In addition, we examined the frequency of epinephrine auto-injector (EAI) prescription among high-risk individuals. Based on a short questionnaire we concluded that 17.3% of the 2 284 adolescents reported so called “problems with food”. We contacted these subjects for a more detailed telephone questionnaire and concluded that 70/120 adolescents considered themselves as being food allergic. However, only 48/120 (40%) were eventually classified as likely to be food allergic based on the reported foods and symptoms. To summarize, the calculated questionnaire-based prevalence of likely and self-perceived but questionable food allergy was 6.2% and 4.0%, respectively. A meta-analysis of this study and 17 other european studies up to 2012 concluded that the overall prevalence of self-reported food allergy was 5.9% with a higher prevalence in children and northern european countries.10 Future studies
with more strict definitions of food allergy including the DBPCFC are indicated to limit bias introduced by false-positive diagnoses of food allergy.
In addition, we show an alarming under-prescription of EAIs since less than 1 in 30 high-risk adolescents had actually been prescribed this potentially life-saving device. This study was performed in 2009 and seven years later, in 2016, this same research was conducted again to study whether the prevalence of food allergy and EAI prescription in adolescents increased. Fortunately, the prevalence of food allergy did not increase but EAI ownership only marginally improved.11 In addition, a recent study with mystery guest visits
concluded that food-allergic patients are often not, or incorrectly instructed by pharmacists on how to use an EAIs.12 These studies indicatethat awareness about food allergy, anaphylaxis
and especially its management still need to be increased.
In Chapter 5 we describe the association between a longer duration of breastfeeding and a
lower risk of food allergy as diagnosed by DBPCFC. No confounders were identified in this association, including parental atopic conditions and early atopic manifestations. This indicates that this protective effect is relevant for children both with and without a positive
270
PART IV Chapter 12 - Summary and general discussion
269 family history of atopy and is independent of reverse causation. However, we did show that atopic parents breastfeed their children more often. But even after adjusting for this confounder, there was no significant association of food allergy with breastfeeding for any duration versus bottle-feeding. Previous studies were contradictory or inconclusive and rarely studied food allergy as diagnosed by DBPCFCs. Interestingly, a previous study reported an association between breastfeeding and sensitization to foods which was influenced significantly by specific gene variants.13 Future studies are indicated to replicate our findings
and to study whether this previously reported interaction with genes is also relevant for food allergy. Although this study is likely to represent the best knowledge yet available regarding the association of breastfeeding and food allergy as diagnosed by DBPCFC, it had some limitations. Unfortunately, there was no data available regarding the exclusivity of breastfeeding and the introduction of solid foods. As published in 2015, early oral exposure to peanut was associated with development of tolerance14 which is likely to confound the
association between breastfeeding and food allergy. Therefore, future studies are indicated to identify the optimal duration of breastfeeding and timing of introduction of potential allergenic foods to decrease the risk of food allergy.
The influence of family history on the risk of food allergy is studied in Chapter 6 since in clinical
practice, concern about food allergy in family members of patients is common. We concluded that in tertiary care, having a positive family history of food allergy, including having a sibling with a positive DBPCFC, has little to no influence on the risk of becoming sensitized or clinically allergic to foods. Only a positive family history of rhinoconjunctivitis paradoxally decreased the risk of food allergy and as previously published, a family history of eczema increased the risk of cow’s milk allergy.15 The low risk of sibling-associated food allergy confirms a previous
study which showed that only 13.6% of siblings were both sensitized and reported symptoms to the same food. The authors concluded that sensitization without reactivity was more common among siblings. This implies that in tertiary care, children with a sibling with confirmed food allergy or a positive family history of food allergy should undergo DBPCFC based on clinical suspicion alone, as the diagnosis cannot be anticipated or inferred based on the family history.
The association between a positive family history of eczema and food allergy is interesting. Eczema is a common atopic skin disorder characterized by dry, pruritic skin and an impaired skin barrier.16 The latter may permit intact proteins to pass the barrier and elicit
an immune response. The dual-allergen-exposure hypothesis proposes that low-dose cutaneous exposure to food allergens triggers Th2 responses and IgE production by B cells while early oral exposure induces tolerance by stimulating regulatory T cells and Th1 cells.17
In addition, being fist born born was associated with an increased risk of sensitization to foods but not with an increased risk of food allergy as diagnosed by the DBPCFC. When having a sibling is considered as a proxy of early life infection, these results indicate that early life infections only influences the risk of sensitization. This is supported by eleven studies on
270 the use of probiotics to prevent food allergy which had disappointing results, as reviewed in 2013.18 Two of them reported only a lower risk of sensitization.19,20
These first four studies focus on environmental, behavioral and familial factors associated with food allergy. Part II of this thesis focuses on genetic risk factors associated with food
allergy. Just like in Part I, we studied adults from the Lifelines cohort and children diagnosed with food allergy based on the DBPCFC, the gold standard. With this last definition of food allergy we were able to validate whether previous identified variants were relevant in sensitization to foods, food allergy or both. We studied these variants in a candidate-gene design using both family-based association testing and case-control studies. In addition, we identified new genetic variants associated with food allergy in two hypothesis-free genome-wide association studies.
The association between food allergy and loss of function (LOF) variants within the filaggrin (FLG) gene is discussed in Chapter 7. This gene is located on chromosome 1q21 within the
epidermal differentiation complex which regulates skin barrier function. The filaggrin protein helps aggregate the epidermal cytoskeleton to form a protein-lipid barrier and fits within the dual-allergen-exposure hypothesis for the pathogenesis of food allergy, as discussed above. Further evidence for this hypothesis is provided by a study in mice showing that epicutaneous exposure to peanut protein on barrier-disrupted skin causes Th2-type immunity with high levels of peanut-specific IgE and prevents the development of oral tolerance to peanut.21
Other studies reported that in children with eczema or carrying FLG mutations, increased early life environmental exposure to peanut in household dust was associated with an increased risk of sensitization to peanut and peanut allergy as confirmed by food challenges.22,23 In a
review and meta-analysis of previous studies, FLG LOF variants were identified as strong risk factors for atopic dermatitis24, allergic rhinitis and sensitization to peanut.25 In children with
atopic dermatitis, they were also associated with asthma.25 Cases and controls in our study
had any positive or only negative DBPCFC(s), respectively. Therefore, the results of our study indicate that FLG is not only important for the pathway leading to sensitization, but also important for the clinically more relevant pathway leading to food allergy. The association was not confounded by eczema, rhinoconjunctivitis or asthma by history or sIgE to the food tested by DBPCFC. A more recent study replicated the association between FLG LOF variants and both sensitization to foods and food allergy in children of 2 years of age.26 In addition, they reported
an association between food allergy and a neonatal barrier defect, measured by increased trans epidermal water loss in the first few days of life, even when the children did not have eczema.26 Skin barrier improving interventions are promising, especially since they are shown
to decrease the risk of eczema by 50%.27 As reviewed recently, three small studies did not find
a significant association between regular emollient application and a decreased risk of food allergy although trends were reported.27 Therefore, larger trials are underway and in 5 years,
we hope to know whether improvement of the skin barrier with emollients leads to an lower risk of food allergy.
Chap
ter 12
271 Summary and general discussion - Chapter 12 PART IV
269 family history of atopy and is independent of reverse causation. However, we did show that atopic parents breastfeed their children more often. But even after adjusting for this confounder, there was no significant association of food allergy with breastfeeding for any duration versus bottle-feeding. Previous studies were contradictory or inconclusive and rarely studied food allergy as diagnosed by DBPCFCs. Interestingly, a previous study reported an association between breastfeeding and sensitization to foods which was influenced significantly by specific gene variants.13 Future studies are indicated to replicate our findings
and to study whether this previously reported interaction with genes is also relevant for food allergy. Although this study is likely to represent the best knowledge yet available regarding the association of breastfeeding and food allergy as diagnosed by DBPCFC, it had some limitations. Unfortunately, there was no data available regarding the exclusivity of breastfeeding and the introduction of solid foods. As published in 2015, early oral exposure to peanut was associated with development of tolerance14 which is likely to confound the
association between breastfeeding and food allergy. Therefore, future studies are indicated to identify the optimal duration of breastfeeding and timing of introduction of potential allergenic foods to decrease the risk of food allergy.
The influence of family history on the risk of food allergy is studied in Chapter 6 since in clinical
practice, concern about food allergy in family members of patients is common. We concluded that in tertiary care, having a positive family history of food allergy, including having a sibling with a positive DBPCFC, has little to no influence on the risk of becoming sensitized or clinically allergic to foods. Only a positive family history of rhinoconjunctivitis paradoxally decreased the risk of food allergy and as previously published, a family history of eczema increased the risk of cow’s milk allergy.15 The low risk of sibling-associated food allergy confirms a previous
study which showed that only 13.6% of siblings were both sensitized and reported symptoms to the same food. The authors concluded that sensitization without reactivity was more common among siblings. This implies that in tertiary care, children with a sibling with confirmed food allergy or a positive family history of food allergy should undergo DBPCFC based on clinical suspicion alone, as the diagnosis cannot be anticipated or inferred based on the family history.
The association between a positive family history of eczema and food allergy is interesting. Eczema is a common atopic skin disorder characterized by dry, pruritic skin and an impaired skin barrier.16 The latter may permit intact proteins to pass the barrier and elicit
an immune response. The dual-allergen-exposure hypothesis proposes that low-dose cutaneous exposure to food allergens triggers Th2 responses and IgE production by B cells while early oral exposure induces tolerance by stimulating regulatory T cells and Th1 cells.17
In addition, being fist born born was associated with an increased risk of sensitization to foods but not with an increased risk of food allergy as diagnosed by the DBPCFC. When having a sibling is considered as a proxy of early life infection, these results indicate that early life infections only influences the risk of sensitization. This is supported by eleven studies on
270 the use of probiotics to prevent food allergy which had disappointing results, as reviewed in 2013.18 Two of them reported only a lower risk of sensitization.19,20
These first four studies focus on environmental, behavioral and familial factors associated with food allergy. Part II of this thesis focuses on genetic risk factors associated with food
allergy. Just like in Part I, we studied adults from the Lifelines cohort and children diagnosed with food allergy based on the DBPCFC, the gold standard. With this last definition of food allergy we were able to validate whether previous identified variants were relevant in sensitization to foods, food allergy or both. We studied these variants in a candidate-gene design using both family-based association testing and case-control studies. In addition, we identified new genetic variants associated with food allergy in two hypothesis-free genome-wide association studies.
The association between food allergy and loss of function (LOF) variants within the filaggrin (FLG) gene is discussed in Chapter 7. This gene is located on chromosome 1q21 within the
epidermal differentiation complex which regulates skin barrier function. The filaggrin protein helps aggregate the epidermal cytoskeleton to form a protein-lipid barrier and fits within the dual-allergen-exposure hypothesis for the pathogenesis of food allergy, as discussed above. Further evidence for this hypothesis is provided by a study in mice showing that epicutaneous exposure to peanut protein on barrier-disrupted skin causes Th2-type immunity with high levels of peanut-specific IgE and prevents the development of oral tolerance to peanut.21
Other studies reported that in children with eczema or carrying FLG mutations, increased early life environmental exposure to peanut in household dust was associated with an increased risk of sensitization to peanut and peanut allergy as confirmed by food challenges.22,23 In a
review and meta-analysis of previous studies, FLG LOF variants were identified as strong risk factors for atopic dermatitis24, allergic rhinitis and sensitization to peanut.25 In children with
atopic dermatitis, they were also associated with asthma.25 Cases and controls in our study
had any positive or only negative DBPCFC(s), respectively. Therefore, the results of our study indicate that FLG is not only important for the pathway leading to sensitization, but also important for the clinically more relevant pathway leading to food allergy. The association was not confounded by eczema, rhinoconjunctivitis or asthma by history or sIgE to the food tested by DBPCFC. A more recent study replicated the association between FLG LOF variants and both sensitization to foods and food allergy in children of 2 years of age.26 In addition, they reported
an association between food allergy and a neonatal barrier defect, measured by increased trans epidermal water loss in the first few days of life, even when the children did not have eczema.26 Skin barrier improving interventions are promising, especially since they are shown
to decrease the risk of eczema by 50%.27 As reviewed recently, three small studies did not find
a significant association between regular emollient application and a decreased risk of food allergy although trends were reported.27 Therefore, larger trials are underway and in 5 years,
we hope to know whether improvement of the skin barrier with emollients leads to an lower risk of food allergy.
272
PART IV Chapter 12 - Summary and general discussion
271 In a second candidate gene study, described in Chapter 8, we confirmed the association of
signal transducer and activator of transcription factor 6 (STAT6) gene with both sensitization
and food allergy. By inducing expression of GATA-3, STAT6 enhances expression of cytokines which stimulate the differentiation of naïve CD4+ T cells to the Th2 subset and suppresses Th1 and Th17 cell responses28. Th2 cells secrete cytokines including IL-4, IL-5 and IL-13 that work
in combination with mast cells, macrophages and eosinophils to promote allergic responses to (food)allergens29. In activated B cells, STAT6 promotes immunoglobulin class switching to
both IgE and IgG1 and the expression of antigen presenting cell surface molecules, as reviewed by Goenka and Kaplan30. Therefore, these mechanisms could explain our observed association
with the presence of food allergy as well as sIgE levels. Recently, a study reported that 90 % of mice having both Filaggrin mutations and T-cell specific expression of constitutively active
Stat6 developed atopic dermatitis, compared to 40% of mice carrying only Filaggrin
mutations31. It was reported that expression of Filaggrin in keratinocytes is diminished by
increased STAT6 and Th2 related cytokines IL-4 and IL-1332, suggesting a role of STAT6 in the
dual-allergen-exposure hypothesis17. Finally, it was demonstrated that formation of
allergen-specific regulatory T cells was decreased due to enhanced IL-4R-STAT6 signalling in mice33,
suggesting a role for STAT6 in tolerance to foods.
In addition, environmental exposures as identified in Part I of this thesis may have epigenetic effects. Epigenetics refers to the regulation of DNA expression by methylation, chromatin modifications, or regulatory RNA molecules. These factors do not affect the DNA sequence itself, yet are at least partially heritable. Recently, DNA methylation differences in several genes including STAT6 were reported in farmers’ compared to non-farmers’ children8.
This indicates that STAT6 might be a mediator in the previously discussed association between a more urban living environment during childhood and a lower risk of food allergy in adulthood.
To summarize, both Filaggrin and STAT6 seems to play a pivotal role in food allergy through various cell types, both dependent and independent of sIgE. Inhibiting STAT6 and restoring the epithelial barrier seems an appealing therapeutic strategy to treat food allergy. However, the complex interplay of STAT6 with cytokines and transcription factors such as GATA3 suggests that variants in their corresponding genes might have an interactive effect on
STAT6, complicating the pathophysiologic pathway involved.
Within adults and adolescents of the Lifelines and Prevention and Incidence of Asthma and Mite Allergy (PIAMA) study, respectively, we performed a Genome Wide Association Study on questionnaire defined food and peanut allergy, as described in Chapter 9. The PIAMA birth
cohort includes 3 963 Dutch children from the general population born in 1996-1997. These had follow-up visits and questionnaires from birth onwards and completed the food allergy questionnaire at the age of 17 years. In 392 cases with food allergy (including 47 cases with peanut allergy) and 9470 controls, we identified 6 and 25 independent top hits associated with food and peanut allergy, respectively. Unfortunately, we did not replicate these variants
272 in children with food allergy diagnosed by food challenges recruited at our own center and as part of a large German study. Potentially, this could be caused by an age effect or it may indicate that the identified variants are only relevant in food allergy in adults with different culprit foods compared to food allergy in children. Both discovery cohorts included adults and adolescents from the general population and the most frequently reported culprit foods among cases were apple, hazelnut, walnut, shellfish and almond. Children with food allergy within both replication cohorts most frequently had positive food challenges for hen’s egg, peanut, cow’s milk and/or cashew. This confirms previous research since the majority of childhood allergies to milk, egg and wheat resolve, whereas allergies to peanut, tree nuts and shellfish tend to persist more frequently34. These latter allergies are therefore more common
in adults, together with allergies to foods cross-reacting with tree-pollen35.
Interestingly, we replicated 10 SNPs previously associated with peanut allergy in genome wide studies in children, including well-known HLA-DQB1 variants, indicating that this effect continues into adulthood. The Human Leukocyte Antigen (HLA) class II genes encode molecules that present antigens, including food allergens, to naïve CD4+T lymphocytes. These stimulate B-cell antibody production, mediating the allergic response.
The SHARE consortium recently published a list of 136 SNPs to be associated with any allergic disease (including asthma, rhinitis, and eczema), yet food allergy was not addressed in that study 36. Of the 121 variants available in our discovery analyses, we showed for the first
time that 10 are likely to be associated with food allergy and peanut allergy as well. However, the effect sizes decreased for up to 35% after correction for other atopic morbidities. This indicates that these variants have a more general effect on allergy, including food allergy. We studied whether variants identified in a Canadian genome wide association study on peanut allergy were also relevant in Dutch children with food allergy diagnosed by DBPCFC, as discussed in Chapter 10. The discovery analysis was performed in 850 Canadian cases with
peanut allergy and 926 hyper-controls. The latter were subjects with no history of asthma, airway hyper responsiveness, atopy, eczema, allergic rhinitis, or food allergy and negative skin prick tests. Cases were defined by a variety of parameters but the majority was sensitized to foods and had a convincing history of peanut allergy. The discovery analysis identified SNPs in
HLA (discussed in chapter 11) and a variant on chromosome 2, close to integrin a6(ITGA6) as
genome-wide significantly associated with food allergy. The latter was not replicated in both cohorts of Dutch children. However, in the meta-analysis using 5 studies (in Canadian, 2 American, 2 Dutch populations) this less common variant (MAF= 0.02) reached genome wide significance for association with both food and peanut allergy. Another variant, close to
angiopoietin 4 (ANGPT4) was replicated in our Dutch cohorts. Previously, this gene was
associated with permeability of blood vessels and promotes lymphatic dilatation37. In a
meta-analysis of four studies, chromosome 11 open reading frame 30 (C11orf30) was identified as a susceptibility locus for both food and peanut allergy. As discussed in chapter 2 of this thesis, this gene was previously associated with asthma38, atopic dermatitis39–41 and
Chap
ter 12
273 Summary and general discussion - Chapter 12 PART IV
271 In a second candidate gene study, described in Chapter 8, we confirmed the association of
signal transducer and activator of transcription factor 6 (STAT6) gene with both sensitization
and food allergy. By inducing expression of GATA-3, STAT6 enhances expression of cytokines which stimulate the differentiation of naïve CD4+ T cells to the Th2 subset and suppresses Th1 and Th17 cell responses28. Th2 cells secrete cytokines including IL-4, IL-5 and IL-13 that work
in combination with mast cells, macrophages and eosinophils to promote allergic responses to (food)allergens29. In activated B cells, STAT6 promotes immunoglobulin class switching to
both IgE and IgG1 and the expression of antigen presenting cell surface molecules, as reviewed by Goenka and Kaplan30. Therefore, these mechanisms could explain our observed association
with the presence of food allergy as well as sIgE levels. Recently, a study reported that 90 % of mice having both Filaggrin mutations and T-cell specific expression of constitutively active
Stat6 developed atopic dermatitis, compared to 40% of mice carrying only Filaggrin
mutations31. It was reported that expression of Filaggrin in keratinocytes is diminished by
increased STAT6 and Th2 related cytokines IL-4 and IL-1332, suggesting a role of STAT6 in the
dual-allergen-exposure hypothesis17. Finally, it was demonstrated that formation of
allergen-specific regulatory T cells was decreased due to enhanced IL-4R-STAT6 signalling in mice33,
suggesting a role for STAT6 in tolerance to foods.
In addition, environmental exposures as identified in Part I of this thesis may have epigenetic effects. Epigenetics refers to the regulation of DNA expression by methylation, chromatin modifications, or regulatory RNA molecules. These factors do not affect the DNA sequence itself, yet are at least partially heritable. Recently, DNA methylation differences in several genes including STAT6 were reported in farmers’ compared to non-farmers’ children8.
This indicates that STAT6 might be a mediator in the previously discussed association between a more urban living environment during childhood and a lower risk of food allergy in adulthood.
To summarize, both Filaggrin and STAT6 seems to play a pivotal role in food allergy through various cell types, both dependent and independent of sIgE. Inhibiting STAT6 and restoring the epithelial barrier seems an appealing therapeutic strategy to treat food allergy. However, the complex interplay of STAT6 with cytokines and transcription factors such as GATA3 suggests that variants in their corresponding genes might have an interactive effect on
STAT6, complicating the pathophysiologic pathway involved.
Within adults and adolescents of the Lifelines and Prevention and Incidence of Asthma and Mite Allergy (PIAMA) study, respectively, we performed a Genome Wide Association Study on questionnaire defined food and peanut allergy, as described in Chapter 9. The PIAMA birth
cohort includes 3 963 Dutch children from the general population born in 1996-1997. These had follow-up visits and questionnaires from birth onwards and completed the food allergy questionnaire at the age of 17 years. In 392 cases with food allergy (including 47 cases with peanut allergy) and 9470 controls, we identified 6 and 25 independent top hits associated with food and peanut allergy, respectively. Unfortunately, we did not replicate these variants
272 in children with food allergy diagnosed by food challenges recruited at our own center and as part of a large German study. Potentially, this could be caused by an age effect or it may indicate that the identified variants are only relevant in food allergy in adults with different culprit foods compared to food allergy in children. Both discovery cohorts included adults and adolescents from the general population and the most frequently reported culprit foods among cases were apple, hazelnut, walnut, shellfish and almond. Children with food allergy within both replication cohorts most frequently had positive food challenges for hen’s egg, peanut, cow’s milk and/or cashew. This confirms previous research since the majority of childhood allergies to milk, egg and wheat resolve, whereas allergies to peanut, tree nuts and shellfish tend to persist more frequently34. These latter allergies are therefore more common
in adults, together with allergies to foods cross-reacting with tree-pollen35.
Interestingly, we replicated 10 SNPs previously associated with peanut allergy in genome wide studies in children, including well-known HLA-DQB1 variants, indicating that this effect continues into adulthood. The Human Leukocyte Antigen (HLA) class II genes encode molecules that present antigens, including food allergens, to naïve CD4+T lymphocytes. These stimulate B-cell antibody production, mediating the allergic response.
The SHARE consortium recently published a list of 136 SNPs to be associated with any allergic disease (including asthma, rhinitis, and eczema), yet food allergy was not addressed in that study 36. Of the 121 variants available in our discovery analyses, we showed for the first
time that 10 are likely to be associated with food allergy and peanut allergy as well. However, the effect sizes decreased for up to 35% after correction for other atopic morbidities. This indicates that these variants have a more general effect on allergy, including food allergy. We studied whether variants identified in a Canadian genome wide association study on peanut allergy were also relevant in Dutch children with food allergy diagnosed by DBPCFC, as discussed in Chapter 10. The discovery analysis was performed in 850 Canadian cases with
peanut allergy and 926 hyper-controls. The latter were subjects with no history of asthma, airway hyper responsiveness, atopy, eczema, allergic rhinitis, or food allergy and negative skin prick tests. Cases were defined by a variety of parameters but the majority was sensitized to foods and had a convincing history of peanut allergy. The discovery analysis identified SNPs in
HLA (discussed in chapter 11) and a variant on chromosome 2, close to integrin a6(ITGA6) as
genome-wide significantly associated with food allergy. The latter was not replicated in both cohorts of Dutch children. However, in the meta-analysis using 5 studies (in Canadian, 2 American, 2 Dutch populations) this less common variant (MAF= 0.02) reached genome wide significance for association with both food and peanut allergy. Another variant, close to
angiopoietin 4 (ANGPT4) was replicated in our Dutch cohorts. Previously, this gene was
associated with permeability of blood vessels and promotes lymphatic dilatation37. In a
meta-analysis of four studies, chromosome 11 open reading frame 30 (C11orf30) was identified as a susceptibility locus for both food and peanut allergy. As discussed in chapter 2 of this thesis, this gene was previously associated with asthma38, atopic dermatitis39–41 and
274
PART IV Chapter 12 - Summary and general discussion
which is supported by a previous study which identified a variant in this gene as associated with a higher risk of poly-sensitization (specific IgE or skin prick test to at least 4 allergens)43.
Unfortunately, C11orf30 was not genotyped in both Dutch cohorts with food allergic children and the identified risk variants were not associated with food allergy and peanut allergy in Dutch adults and adolescents, as described in chapter 9.
As discussed above, peanut allergy was repeatedly and specifically associated with variants in the HLA class II genes. In chapter 11, the results of the Canadian association study on peanut
allergy regarding chromosome 6 were presented separately. Gene variants close to
HLA-DQB1, HLA-DQA2 and HLA-DRA were associated with peanut allergy, both in the Canadian
discovery analysis and the Dutch cohorts with children with food allergy as confirmed by DBPCFCs. Interestingly, the association with food allergy in general was not significant in both Dutch cohorts, just as reported for other HLA variants in chapter 9 which indicates that the associations with HLA class II genes are truly specific for peanut allergy. In addition, the odds ratios increased with increasing severity of food allergy. In a stratified analysis by asthma status, the majority of odds ratios were not significantly different. In the replication analysis described in chapter 9 we previously showed that odds ratios of variants within HLA class II genes increased after adjusting for atopic comorbidities. This indicates that these variants have an effect independent of other atopic morbidities.
CONCLUDING REMARKS AND FUTURE PERSPECTIVES
As described in this thesis, the prevalence of self-reported food allergy among Dutch adults and adolescents is considerable and it significantly impacts health-related quality of life. Among adults with self-reported food allergy, only one-third reports characteristics consistent with food allergy. Among the remainder with questionable, self-perceived food allergy, burn-out and depression were more often reported compared to controls withburn-out food allergy. The high prevalence and considerable psychological burden of questionable, self-perceived food allergy makes this phenomenon a priority for future research. A recent study concluded that people who are more open to experience, one of the Big Five personality traits, had the greatest difficulty managing their food allergy in daily life.44 It is likely that their personality
conflicts with the behavioral consequences of having food allergy.44 Future studies which take
personality into account are indicated to develop strategies to improve H-RQOL in patients with food allergy.
We report a persistent alarming under-prescription of epinephrine auto-injectors since less than 1 in 30 high-risk adolescents had actually been prescribed this potentially life-saving device. Fortunately, an allergy management support system is developed which provides diagnostic and management recommendations for general practictioners45. This system
requires further testing and validation but hopefully it will result in less over-diagnosis of food allergy and improved prescription of potentially lifesaving EAIs to patients at high risk of anaphylaxis.
It seems that there are multiple pathways leading to both sensitization to foods and food allergy, which involve both environmental and genetic factors. As indicated by the
274 association between eczema, food allergy and the effect of loss-of-function variants within the FLG gene, a decreased barrier function is likely to be associated with both sensitization to foods and food allergy. It is hypothesized that sensitization to food allergens and food allergy occurs when the antigen is encountered by other sites such as airways or skin in contrast to the gut, where oral tolerance is likely to be the default response46,47. In addition, detergents
and surfactants are recently associated with an impaired skin barrier48 but their role in food
allergy has not been established yet. Our data on FLG strongly supports the role of a reduced skin barrier in the pathway leading to clinical food allergy. Therefore, we recommend that intervention trials aimed at improving the skin barrier also address food allergy confirmed by DBPCFCs as an outcome.
The association between early farm exposure and adult food allergy confirms previous studies reporting an association between farm exposure and a decreased risk of asthma49,50.
DNA methylation alterations in STAT6 were reported in farmers’ children8 indicating a
gene-environment interaction. Furthermore, the consumption of farm milk was previously associated with higher numbers of regulatory T cells in children5 and levels of specific
regulatory T cells were lower in young children with food allergy51. In addition, higher levels
of specific regulatory T cells were associated with development of oral tolerance52 and were
seen in subjects undergoing oral immunotherapy53.
Future studies are indicated to identify genes robustly associated with the difference between sensitization to foods and food allergy since most identified genetic risk factors seem to be associated with both conditions. We therefore recommend to phenotype cases using double-blind placebo-controlled food challenges with challenge-negative subjects as controls. Ultimately, meta-analyses are indicated to generate more power which gives the opportunity to reveal food allergy SNPs with lower effect and study specific food allergies separately. For studies regarding food allergy in general, we would recommend to study pediatric and adult food allergy separately since they might reflect different phenotypes.
The studies in this thesis are all observational in design in which we tried to carefully adjust for confounding factors. The identified risk factors including duration of breastfeeding, comorbidities, family history, birth order and living environment during childhood are unsuitable to study in a more prospective study design, such as a randomized controlled trial. Therefore, this may represent the best data available regarding their role in food allergy.
As described for other multifactorial diseases, it might be true that in the pathogenesis of food allergy “genes load the gun but environment pulls the trigger”. This thesis focused both on environmental and genetic risk factors associated with food allergy but did not address epigenetic mechanisms mediating both effects. Epigenetics add another layer of complexity to the pathogenesis and future studies are indicated to unravel how the identified environmental and genetic risk factors interact and lead to food allergy. Another interesting theme is whether the identified factors affect treatment response, tolerance development and whether they are relevant for personalized treatment of food allergic patients.
Chap
ter 12
275 Summary and general discussion - Chapter 12 PART IV
which is supported by a previous study which identified a variant in this gene as associated with a higher risk of poly-sensitization (specific IgE or skin prick test to at least 4 allergens)43.
Unfortunately, C11orf30 was not genotyped in both Dutch cohorts with food allergic children and the identified risk variants were not associated with food allergy and peanut allergy in Dutch adults and adolescents, as described in chapter 9.
As discussed above, peanut allergy was repeatedly and specifically associated with variants in the HLA class II genes. In chapter 11, the results of the Canadian association study on peanut
allergy regarding chromosome 6 were presented separately. Gene variants close to
HLA-DQB1, HLA-DQA2 and HLA-DRA were associated with peanut allergy, both in the Canadian
discovery analysis and the Dutch cohorts with children with food allergy as confirmed by DBPCFCs. Interestingly, the association with food allergy in general was not significant in both Dutch cohorts, just as reported for other HLA variants in chapter 9 which indicates that the associations with HLA class II genes are truly specific for peanut allergy. In addition, the odds ratios increased with increasing severity of food allergy. In a stratified analysis by asthma status, the majority of odds ratios were not significantly different. In the replication analysis described in chapter 9 we previously showed that odds ratios of variants within HLA class II genes increased after adjusting for atopic comorbidities. This indicates that these variants have an effect independent of other atopic morbidities.
CONCLUDING REMARKS AND FUTURE PERSPECTIVES
As described in this thesis, the prevalence of self-reported food allergy among Dutch adults and adolescents is considerable and it significantly impacts health-related quality of life. Among adults with self-reported food allergy, only one-third reports characteristics consistent with food allergy. Among the remainder with questionable, self-perceived food allergy, burn-out and depression were more often reported compared to controls withburn-out food allergy. The high prevalence and considerable psychological burden of questionable, self-perceived food allergy makes this phenomenon a priority for future research. A recent study concluded that people who are more open to experience, one of the Big Five personality traits, had the greatest difficulty managing their food allergy in daily life.44 It is likely that their personality
conflicts with the behavioral consequences of having food allergy.44 Future studies which take
personality into account are indicated to develop strategies to improve H-RQOL in patients with food allergy.
We report a persistent alarming under-prescription of epinephrine auto-injectors since less than 1 in 30 high-risk adolescents had actually been prescribed this potentially life-saving device. Fortunately, an allergy management support system is developed which provides diagnostic and management recommendations for general practictioners45. This system
requires further testing and validation but hopefully it will result in less over-diagnosis of food allergy and improved prescription of potentially lifesaving EAIs to patients at high risk of anaphylaxis.
It seems that there are multiple pathways leading to both sensitization to foods and food allergy, which involve both environmental and genetic factors. As indicated by the
274 association between eczema, food allergy and the effect of loss-of-function variants within the FLG gene, a decreased barrier function is likely to be associated with both sensitization to foods and food allergy. It is hypothesized that sensitization to food allergens and food allergy occurs when the antigen is encountered by other sites such as airways or skin in contrast to the gut, where oral tolerance is likely to be the default response46,47. In addition, detergents
and surfactants are recently associated with an impaired skin barrier48 but their role in food
allergy has not been established yet. Our data on FLG strongly supports the role of a reduced skin barrier in the pathway leading to clinical food allergy. Therefore, we recommend that intervention trials aimed at improving the skin barrier also address food allergy confirmed by DBPCFCs as an outcome.
The association between early farm exposure and adult food allergy confirms previous studies reporting an association between farm exposure and a decreased risk of asthma49,50.
DNA methylation alterations in STAT6 were reported in farmers’ children8 indicating a
gene-environment interaction. Furthermore, the consumption of farm milk was previously associated with higher numbers of regulatory T cells in children5 and levels of specific
regulatory T cells were lower in young children with food allergy51. In addition, higher levels
of specific regulatory T cells were associated with development of oral tolerance52 and were
seen in subjects undergoing oral immunotherapy53.
Future studies are indicated to identify genes robustly associated with the difference between sensitization to foods and food allergy since most identified genetic risk factors seem to be associated with both conditions. We therefore recommend to phenotype cases using double-blind placebo-controlled food challenges with challenge-negative subjects as controls. Ultimately, meta-analyses are indicated to generate more power which gives the opportunity to reveal food allergy SNPs with lower effect and study specific food allergies separately. For studies regarding food allergy in general, we would recommend to study pediatric and adult food allergy separately since they might reflect different phenotypes.
The studies in this thesis are all observational in design in which we tried to carefully adjust for confounding factors. The identified risk factors including duration of breastfeeding, comorbidities, family history, birth order and living environment during childhood are unsuitable to study in a more prospective study design, such as a randomized controlled trial. Therefore, this may represent the best data available regarding their role in food allergy.
As described for other multifactorial diseases, it might be true that in the pathogenesis of food allergy “genes load the gun but environment pulls the trigger”. This thesis focused both on environmental and genetic risk factors associated with food allergy but did not address epigenetic mechanisms mediating both effects. Epigenetics add another layer of complexity to the pathogenesis and future studies are indicated to unravel how the identified environmental and genetic risk factors interact and lead to food allergy. Another interesting theme is whether the identified factors affect treatment response, tolerance development and whether they are relevant for personalized treatment of food allergic patients.
276
PART IV Chapter 12 - Summary and general discussion
275
REFERENCES
1. Fox M, Mugford M, Voordouw J, Cornelisse-Vermaat J, Antonides G, De La Hoz Caballer B, et al. Health sector costs of self-reported food allergy in Europe: A patient-based cost of illness study. Eur J Public Health. 2013;23(5):757–62.
2. Flokstra-de Blok BMJ, Dubois AEJ, Vlieg-Boerstra BJ, Oude Elberink JNG, Raat H, DunnGalvin A, et al. Health-related quality of life of food allergic patients: comparison with the general population and other diseases. Allergy. 2010 Feb;65(2):238–44. 3. Gupta RS, Springston EE, Smith B, Warrier
MR, Pongracic J, Holl JL. Geographic variability of childhood food allergy in the United States. Clin Pediatr (Phila). 2012;51(9):856–61.
4. Loss G, Apprich S, Waser M, Kneifel W, Genuneit J, Büchele G, et al. The protective effect of farm milk consumption on childhood asthma and atopy: The GABRIELA study. J Allergy Clin Immunol. 2011;128(4). 5. Lluis A, Depner M, Gaugler B, Saas P,
Casaca VI, Raedler D, et al. Increased regulatory T-cell numbers are associated with farm milk exposure and lower atopic sensitization and asthma in childhood. J Allergy Clin Immunol. 2014;133(2). 6. Kirchner B, Pfaffl MW, Dumpler J, Von
Mutius E, Ege MJ. MicroRNA in native and processed cow’s milk and its implication for the farm milk effect on asthma. J Allergy Clin Immunol. 2016;137(6):1893–1895e13. 7. Slaats GGG, Reinius LE, Alm J, Kere J,
Scheynius A, Joerink M. DNA methylation levels within the CD14 promoter region are lower in placentas of mothers living on a farm. Allergy. 2012 Jul;67(7):895–903. 8. Michel S, Busato F, Genuneit J, Pekkanen J,
Dalphin J-C, Riedler J, et al. Farm exposure and time trends in early childhood may influence DNA methylation in genes related to asthma and allergy. Allergy. 2013 Mar;68(3):355–64.
9. van der Velde JL, Flokstra-de Blok BMJ, de Groot H, Oude-Elberink JNG, Kerkhof M, Duiverman EJ, et al. Food allergy-related quality of life after double-blind, placebo-controlled food challenges in adults, adolescents, and children. J Allergy Clin
Immunol. 2012 Nov;130(5):1136–1143.e2. 10. Nwaru BI, Hickstein L, Panesar SS, Muraro
A, Werfel T, Cardona V, et al. The epidemiology of food allergy in Europe: A systematic review and meta-analysis. Allergy Eur J Allergy Clin Immunol. 2014;69(1):62–75.
11. Saleh-Langenberg J, Bootsma GM, van Ginkel CD, Kollen BJ, Flokstra-de Blok BMJ, Dubois AEJ. The prevalence of food allergy and epinephrine auto-injectors in Dutch food-allergic adolescents. Pediatr Allergy Immunol. 2016;1–4.
12. Saleh-Langenberg J, de Vries S, Bak E, Kollen BJ, Flokstra-de Blok BMJ, Dubois AEJ. Incomplete and incorrect epinephrine auto-injector training to food-allergic patients by pharmacists in the Netherlands. Pediatr Allergy Immunol. 2017;28(3):238–44. 13. Hong X, Wang G, Liu X, Kumar R, Tsai H-J,
Arguelles L, et al. Gene polymorphisms, breast-feeding, and development of food sensitization in early childhood. J Allergy Clin Immunol. 2011 Aug;128(2):374–81.e2. 14. Perkin MR, Logan K, Tseng A, Raji B, Ayis S,
Peacock J, et al. Randomized Trial of Introduction of Allergenic Foods in Breast-Fed Infants. N Engl J Med.
2016;374(18):1733–43.
15. van den Berg ME, Flokstra-de Blok BMJ, Vlieg-Boerstra BJ, Kerkhof M, van der Heide S, Koppelman GH, et al. Parental eczema increases the risk of double-blind, placebo-controlled reactions to milk but not to egg, peanut or hazelnut. Int Arch Allergy Immunol. 2012 Jan;158(1):77–83. 16. Eichenfield LF, Tom WL, Chamlin SL,
Feldman SR, Hanifin JM, Simpson EL, et al. Guidelines of care for the management of atopic dermatitis: section 1. Diagnosis and assessment of atopic dermatitis. J Am Acad Dermatol. 2014 Feb;70(2):338–51. 17. Lack G. Epidemiologic risks for food allergy.
J Allergy Clin Immunol. 2008 Jun;121(6):1331–6.
18. Marrs T, Bruce KD, Logan K, Rivett DW, Perkin MR, Lack G, et al. Is there an association between microbial exposure and food allergy? A systematic review. Pediatr Allergy Immunol. 2013;24(4):311–
276
20.
19. Morisset M, Aubert-Jacquin C, Soulaines P, Moneret-Vautrin DA, Dupont C. A non-hydrolyzed, fermented milk formula reduces digestive and respiratory events in infants at high risk of allergy. Eur J Clin Nutr. 2011;65(2):175–83.
20. Kuitunen M, Kukkonen K, Juntunen-Backman K, Korpela R, Poussa T, Tuure T, et al. Probiotics prevent IgE-associated allergy until age 5 years in cesarean-delivered children but not in the total cohort. J Allergy Clin Immunol. 2009;123(2):335–41. 21. Strid J, Hourihane J, Kimber I, Callard R,
Strobel S. Epicutaneous exposure to peanut protein prevents oral tolerance and enhances allergic sensitization. Clin Exp Allergy. 2005 Jun;35(6):757–66. 22. Brough H a, Simpson A, Makinson K,
Hankinson J, Brown S, Douiri A, et al. Peanut allergy: Effect of environmental peanut exposure in children with filaggrin loss-of-function mutations. J Allergy Clin Immunol. 2014 Oct;134(4):867–875.e1. 23. Brough HA, Liu AH, Sicherer S, Makinson K,
Douiri A, Brown SJ, et al. Atopic dermatitis increases the effect of exposure to peanut antigen in dust on peanut sensitization and likely peanut allergy. J Allergy Clin Immunol. 2015;135(1):164–70.
24. O’Regan GM, Sandilands A, McLean WHI, Irvine AD. Filaggrin in atopic dermatitis. J Allergy Clin Immunol. 2009 Sep;124(3 Suppl 2):R2-6.
25. van den Oord RAHM, Sheikh A. Filaggrin gene defects and risk of developing allergic sensitisation and allergic disorders: systematic review and meta-analysis. BMJ. 2009 Jul 9;339:b2433–b2433.
26. Kelleher MM, Dunn-Galvin A, Gray C, Murray DM, Kiely M, Kenny L, et al. Skin barrier impairment at birth predicts food allergy at 2 years of age. J Allergy Clin Immunol. 2016;137(4):1111–1116.e8. 27. Lowe AJ, Leung DYM, Tang MLK, Su JC,
Allen KJ. The skin as a target for prevention of the atopic march. Ann Allergy, Asthma Immunol. 2018;120(2):145–51.
28. Paul WE, Zhu J. How are TH2-type immune responses initiated and amplified? Nat Rev Immunol. 2010;10(4):225–35.
29. Abbas AK, Lichtman AH, Pillai S. Cellular and Molecular Immunology. 8th ed.
Philadelphia: Elsevier, Saunders; 2015. 30. Goenka S, Kaplan MH. Transcriptional
regulation by STAT6. J Immunol Res. 2011;50(1):87–96.
31. Sehra S, Krishnamurthy P, Koh B, Zhou HM, Seymour L, Akhtar N, et al. Increased Th2 activity and diminished skin barrier function cooperate in allergic skin inflammation. Eur J Immunol. 2016;46(11):2609–13. 32. Howell MD, Kim BE, Gao P, Grant A V.,
Boguniewicz M, DeBenedetto A, et al. Cytokine modulation of atopic dermatitis filaggrin skin expression. J Allergy Clin Immunol. 2007;120(1):150–5.
33. Rivas MN, Burton OT, Wise P, Charbonnier L, Georgiev P, Oettgen HC, et al. Regulatory T cell reprogramming towards a Th2 cell-like lineage impairs oral tolerance and promotes food allergy. Immunity. 2016;42(3):512–23.
34. Sicherer SH, Sampson H a. Food allergy: Epidemiology, pathogenesis, diagnosis, and treatment. J Allergy Clin Immunol. 2014 Feb;133(2):291–307; quiz 308.
35. Werfel T, Asero R, Ballmer-Weber BK, Beyer K, Enrique E, Knulst AC, et al. Position paper of the EAACI: Food allergy due to
immunological cross-reactions with common inhalant allergens. Allergy Eur J Allergy Clin Immunol. 2015;70(9):1079–90. 36. Ferreira MA, Vonk JM, Baurecht H,
Marenholz I, Tian C, Hoffman JD, et al. Shared genetic origin of asthma , hay fever and eczema elucidates allergic disease biology. Nat Genet. 2017;49:1752–7. 37. Kesler CT, Pereira ER, Cui CH, Nelson GM,
Masuck DJ, Baish JW, et al. Angiopoietin-4 increases permeability of blood vessels and promotes lymphatic dilation. FASEB J. 2015;29(9):3668–77.
38. Moffatt MF, Phil D, Gut IG, Demenais F, Strachan DP, Bouzigon E, et al. A Large-Scale, Consortium-Based Genomewide Association Study of Asthma. N Engl J Med. 2010;363(13):1211–21.
39. Esparza-Gordillo J, Weidinger S, Fölster-Holst R, Bauerfeind A, Ruschendorf F, Patone G, et al. A common variant on chromosome 11q13 is associated with atopic dermatitis. Nat Genet. 2009 May;41(5):596–601.
40. Ellinghaus D, Baurecht H, Esparza-Gordillo J, Rodríguez E, Matanovic A, Marenholz I, et
