Advertisement

Loss-of-function variants in the filaggrin gene are a significant risk factor for peanut allergy

      Background

      IgE-mediated peanut allergy is a complex trait with strong heritability, but its genetic basis is currently unknown. Loss-of-function mutations within the filaggrin gene are associated with atopic dermatitis and other atopic diseases; therefore, filaggrin is a candidate gene in the etiology of peanut allergy.

      Objective

      To investigate the association between filaggrin loss-of-function mutations and peanut allergy.

      Methods

      Case-control study of 71 English, Dutch, and Irish oral food challenge–positive patients with peanut allergy and 1000 non peanut-sensitized English population controls. Replication was tested in 390 white Canadian patients with peanut allergy (defined by food challenge, or clinical history and skin prick test wheal to peanut ≥8 mm and/or peanut-specific IgE ≥15 kUL−1) and 891 white Canadian population controls. The most prevalent filaggrin loss-of-function mutations were assayed in each population: R501X and 2282del4 in the Europeans, and R501X, 2282del4, R2447X, and S3247X in the Canadians. The Fisher exact test and logistic regression were used to test for association; covariate analysis controlled for coexistent atopic dermatitis.

      Results

      Filaggrin loss-of-function mutations showed a strong and significant association with peanut allergy in the food challenge–positive patients (P = 3.0 × 10−6; odds ratio, 5.3; 95% CI, 2.8-10.2), and this association was replicated in the Canadian study (P = 5.4 × 10−5; odds ratio, 1.9; 95% CI, 1.4-2.6). The association of filaggrin mutations with peanut allergy remains significant (P = .0008) after controlling for coexistent atopic dermatitis.

      Conclusion

      Filaggrin mutations represent a significant risk factor for IgE-mediated peanut allergy, indicating a role for epithelial barrier dysfunction in the pathogenesis of this disease.

      Key words

      Abbreviations used:

      AD (Atopic dermatitis), ALSPAC (Avon Longitudinal Study of Parents and Children), FLG (Filaggrin), OR (Odds ratio), SPT (Skin prick test), UK (United Kingdom)
      An adverse immune response to peanut ingestion may be severe and is potentially life-threatening.
      • Bock S.A.
      • Munoz-Furlong A.
      • Sampson H.A.
      Further fatalities caused by anaphylactic reactions to food, 2001-2006.
      The prevalence of IgE-mediated peanut allergy in the United Kingdom (UK) and the United States has increased significantly over the past decades
      • Burks A.W.
      Peanut allergy.
      • Sicherer S.H.
      • Munoz-Furlong A.
      • Godbold J.H.
      • Sampson H.A.
      US prevalence of self-reported peanut, tree nut, and sesame allergy: 11-year follow-up.
      but may now have stabilized in the UK
      • Venter C.
      • Hasan Arshad S.
      • Grundy J.
      • Pereira B.
      • Bernie Clayton C.
      • Voigt K.
      • et al.
      Time trends in the prevalence of peanut allergy: three cohorts of children from the same geographical location in the UK.
      and Canada.
      • Ben-Shoshan M.
      • Kagan R.S.
      • Alizadehfar R.
      • Joseph L.
      • Turnbull E.
      • St Pierre Y.
      • et al.
      Is the prevalence of peanut allergy increasing? a 5-year follow-up study in children in Montreal.
      The prevalence of peanut allergy in preschool and school-age children is approximately 1.2% to 1.6%,
      • Sicherer S.H.
      • Munoz-Furlong A.
      • Godbold J.H.
      • Sampson H.A.
      US prevalence of self-reported peanut, tree nut, and sesame allergy: 11-year follow-up.
      • Venter C.
      • Hasan Arshad S.
      • Grundy J.
      • Pereira B.
      • Bernie Clayton C.
      • Voigt K.
      • et al.
      Time trends in the prevalence of peanut allergy: three cohorts of children from the same geographical location in the UK.
      • Ben-Shoshan M.
      • Kagan R.S.
      • Alizadehfar R.
      • Joseph L.
      • Turnbull E.
      • St Pierre Y.
      • et al.
      Is the prevalence of peanut allergy increasing? a 5-year follow-up study in children in Montreal.
      whereas the prevalence in US adults is estimated to be 0.6%.
      • Sicherer S.H.
      • Munoz-Furlong A.
      • Godbold J.H.
      • Sampson H.A.
      US prevalence of self-reported peanut, tree nut, and sesame allergy: 11-year follow-up.
      Peanut allergy is strongly heritable, with a monozygotic twin concordance of 64% compared with 7% in both dizygotic twins
      • Sicherer S.H.
      • Furlong T.J.
      • Maes H.H.
      • Desnick R.J.
      • Sampson H.A.
      • Gelb B.D.
      Genetics of peanut allergy: a twin study.
      and other siblings.
      • Hourihane J.O.
      • Dean T.P.
      • Warner J.O.
      Peanut allergy in relation to heredity, maternal diet, and other atopic diseases: results of a questionnaire survey, skin prick testing, and food challenges.
      A previously reported association with HLA class II genes
      • Howell W.M.
      • Turner S.J.
      • Hourihane J.O.
      • Dean T.P.
      • Warner J.O.
      HLA class II DRB1, DQB1 and DPB1 genotypic associations with peanut allergy: evidence from a family-based and case-control study.
      has not been replicated,
      • Shreffler W.G.
      • Charlop-Powers Z.
      • Sicherer S.H.
      Lack of association of HLA class II alleles with peanut allergy.
      and the genetic basis of this disorder remains poorly understood.
      Loss-of-function mutations in the filaggrin gene (FLG) are a strong and significant risk factor for atopic dermatitis (AD)
      • Rodríguez E.
      • Baurecht H.
      • Herberich E.
      • Wagenpfeil S.
      • Brown S.
      • Cordell H.
      • et al.
      Meta-analysis of filaggrin polymorphisms in eczema and asthma: robust risk factors in atopic disease.
      as well as asthma in association with AD,
      • Rodríguez E.
      • Baurecht H.
      • Herberich E.
      • Wagenpfeil S.
      • Brown S.
      • Cordell H.
      • et al.
      Meta-analysis of filaggrin polymorphisms in eczema and asthma: robust risk factors in atopic disease.
      allergic rhinitis,
      • van den Oord R.
      • Sheikh A.
      Filaggrin gene defects and risk of developing allergic sensitisation and allergic disorders: systematic review and meta-analysis.
      and elevated IgE, indicating sensitization to certain foods.
      • van den Oord R.
      • Sheikh A.
      Filaggrin gene defects and risk of developing allergic sensitisation and allergic disorders: systematic review and meta-analysis.
      The FLG gene encodes profilaggrin, an insoluble polyprotein that is expressed in the granular layer of the epidermis and is broken down to release filaggrin monomers in the stratum corneum.
      • Sandilands A.
      • Sutherland C.
      • Irvine A.
      • McLean W.
      Filaggrin in the frontline: role in skin barrier function and disease.
      Filaggrin plays a key role in epithelial barrier function, but this protein is expressed in neither the bronchial airways nor the upper gastrointestinal tract beyond the oral mucosa
      • De Benedetto A.
      • Qualia C.M.
      • Baroody F.M.
      • Beck L.A.
      Filaggrin expression in oral, nasal, and esophageal mucosa.
      or possibly the esophagus.
      • Blanchard C.
      • Stucke E.M.
      • Burwinkel K.
      • Caldwell J.M.
      • Collins M.H.
      • Ahrens A.
      • et al.
      Coordinate interaction between IL-13 and epithelial differentiation cluster genes in eosinophilic esophagitis.
      It has been hypothesized that allergic sensitization in the atopic state occurs via either transcutaneous or transmucosal passage of allergens, a process that may be facilitated by filaggrin deficiency.
      • Hudson T.J.
      Skin barrier function and allergic risk.
      • Leung D.Y.
      Our evolving understanding of the functional role of filaggrin in atopic dermatitis.
      Experimental evidence supporting a prominent role of epithelial barrier deficiency as a facilitating early event in allergic priming comes from the recent analysis of a filaggrin-deficient mouse mutant.
      • Fallon P.G.
      • Sasaki T.
      • Sandilands A.
      • Campbell L.E.
      • Saunders S.P.
      • Mangan N.E.
      • et al.
      A homozygous frameshift mutation in the mouse Flg gene facilitates enhanced percutaneous allergen priming.
      In view of the strong association of FLG null mutations with atopic disease and impaired skin barrier function, we aimed to investigate their role as a risk factor for IgE-mediated peanut allergy.

      Methods

       Subjects

      Seventy-one patients with peanut allergy confirmed by oral food challenge, 1000 nonsensitized controls, and the respective population controls were collected from the white European populations of England, The Netherlands, and Ireland, as follows:

       English patients with peanut allergy

      Records of thirty-five children with peanut allergy were obtained from the Avon Longitudinal Study of Parents and Children (ALSPAC).
      • Golding J.
      • Pembrey M.
      • Jones R.
      ALSPAC–the Avon Longitudinal Study of Parents and Children, I: study methodology.
      In this longitudinal birth cohort study, peanut allergy was defined by a suggestive clinical history plus a positive double-blind placebo-controlled food challenge. Subjects reported by parents as having flexural dermatitis at 2 or more time points (up to a maximum of 4) between the ages of 6 and 42 months were designated as having AD.

       Dutch patients with peanut allergy

      Twenty patients with peanut allergy were recruited from pediatric allergy clinics. Peanut allergy was defined by a positive double-blind placebo-controlled oral food challenge test to peanut. AD was defined on the basis of a dermatologist’s examination.

       Irish patients with peanut allergy

      Sixteen patients with peanut allergy were recruited from pediatric allergy clinics on the basis of a suggestive clinical history and a positive open oral food challenge test to peanut. AD was defined on the basis of a pediatrician’s examination or parental report.

       Non–peanut-sensitized control group

      ALSPAC is a longitudinal, population-based birth cohort study that originally included over 14,000 English children, as previously described.
      • Golding J.
      • Pembrey M.
      • Jones R.
      ALSPAC–the Avon Longitudinal Study of Parents and Children, I: study methodology.
      One thousand consecutive non–peanut-sensitized individuals (ie, having had a negative skin prick test [SPT] result for peanut) for whom AD status was recorded were drawn from the ALSPAC cohort as normal controls for the primary analysis. In this control group, AD was defined in the same way as for the ALSPAC patients—that is, as having flexural dermatitis at 2 or more time points between the ages of 6 and 42 months.
      Demographic data for the food challenge–positive patients with peanut allergy and non–peanut-sensitized controls are shown in Table I.
      Table IDemographic data and clinical characteristics of 71 food challenge–positive patients with peanut allergy and a nonsensitized control group
      Patients and controlsNo. of individualsAge (y),

      range (mean)
      Male sex,

      n (%)
      Coexistent AD,

      n (%)
      Patients with peanut allergy from an English population birth cohort35Cohort recruited at birth and followed up for ≥7 y19 (54.3)16 out of 22 for whom AD data are available

      (72.7)
      Dutch patients with peanut allergy203-14 (7.5)14 (70.0)17 (85.0)
      Irish patients with peanut allergy161-18 (10.5)7 (43.8)10 (62.5)
      Total patients with peanut allergy71Not applicable40 (56.3)43 (74.1)
      Non–peanut-sensitized controls from English population birth cohort1000Cohort recruited at birth and followed up for ≥7 y544 (54.4)270 (27.0)
      The English population birth cohort is ALSPAC; food challenges in the English and Dutch case collections were double-blind placebo-controlled; in the Irish patients, peanut allergy was confirmed by an open oral food challenge; % coexistent AD refers to the proportion of patients for whom AD data were available.

       Population control groups

      Within the ALSPAC population birth cohort, relevant phenotype data were available for 6895 individuals successfully typed for the 2 most common FLG null mutations (R501X and 2282del4). The 6851 individuals without peanut allergy from this collection were used as an English population control group. One hundred Dutch population control samples were obtained from healthy adult blood donors attending the University Medical Centre, Groningen. Irish population controls were 100 healthy adults from the population-based Trinity Biobank control samples.

       Replication study

      The replication of findings from the food challenge–proven patients was tested in a larger Canadian peanut allergy case collection, as follows:

       Canadian patients with peanut allergy

      From an established Canadian peanut allergy cohort,
      • Ben-Shoshan M.
      • Harrington D.W.
      • Soller L.
      • Fragapane J.
      • Joseph L.
      • St Pierre Y.
      • et al.
      A population-based study on peanut, tree nut, fish, shellfish, and sesame allergy prevalence in Canada.
      • Ben-Shoshan M.
      • Kagan R.
      • Primeau M.N.
      • Alizadehfar R.
      • Verreault N.
      • Yu J.W.
      • et al.
      Availability of the epinephrine autoinjector at school in children with peanut allergy.
      • Yu J.W.
      • Kagan R.
      • Verreault N.
      • Nicolas N.
      • Joseph L.
      • St Pierre Y.
      • et al.
      Accidental ingestions in children with peanut allergy.
      • Kagan R.
      • Hayami D.
      • Joseph L.
      • St Pierre Y.
      • Clarke A.E.
      The predictive value of a positive prick skin test to peanut in atopic, peanut-naive children.
      390 white patients with peanut allergy were recruited between July 2008 and April 2009. They were defined as having peanut allergy on the basis of a positive oral food challenge (n = 25) or, in the non–food-challenged subjects, a peanut-specific IgE ≥15 kUL−1 (n = 65) or a SPT wheal to peanut ≥8 mm (n = 214) or both sIgE ≥15 kUL−1 and SPT wheal ≥8 mm (n = 86).
      • Roberts G.
      • Lack G.
      Team tALSoPaCS. ALSPAC cohort.
      • Sporik R.
      • Hill D.J.
      • Hosking C.S.
      Specificity of allergen skin testing in predicting positive open food challenges to milk, egg and peanut in children.
      • Wainstein B.K.
      • Yee A.
      • Jelley D.
      • Ziegler M.
      • Ziegler J.B.
      Combining skin prick, immediate skin application and specific-IgE testing in the diagnosis of peanut allergy in children.
      A total of 249 (68%) of the patients defined by immunologic parameters had a clinical history of anaphylaxis to peanut on the basis of the consensus definition from the National Institute of Allergy and Infectious Disease/Food Allergy and Anaphylaxis Network,
      • Sampson H.A.
      • Munoz-Furlong A.
      • Campbell R.L.
      • Adkinson Jr., N.F.
      • Bock S.A.
      • Branum A.
      • et al.
      Second symposium on the definition and management of anaphylaxis: summary report–second National Institute of Allergy and Infectious Disease/Food Allergy and Anaphylaxis Network symposium.
      whereas 116 (32%) had a history strongly suggestive of type I hypersensitivity to peanut.
      • Hourihane J.O.
      • Kilburn S.A.
      • Dean P.
      • Warner J.O.
      Clinical characteristics of peanut allergy.
      These thresholds have gained general acceptance, and in the Canadian context, it would be ethically difficult to justify the use of a food challenge in children with values exceeding these thresholds.
      The clinical and immunologic data relating to the Canadian patients with peanut allergy are summarized in Table II.
      Table IIClinical and immunologic parameters in Canadian patients with peanut allergy
      Case definitionn
      Positive oral food challenge test to peanut25
      History of anaphylaxis to peanut
      • Sampson H.A.
      • Munoz-Furlong A.
      • Campbell R.L.
      • Adkinson Jr., N.F.
      • Bock S.A.
      • Branum A.
      • et al.
      Second symposium on the definition and management of anaphylaxis: summary report–second National Institute of Allergy and Infectious Disease/Food Allergy and Anaphylaxis Network symposium.
      (n = 249)
      Peanut-specific IgE ≥15 kU/L

      SPT to peanut ≥8 mm

      Peanut-specific IgE ≥15 kU/L and SPT to peanut ≥8 mm
      48

      141

      60
      History suggestive of a type I hypersensitivity reaction to peanut
      • Hourihane J.O.
      • Kilburn S.A.
      • Dean P.
      • Warner J.O.
      Clinical characteristics of peanut allergy.
      (n = 116)
      Peanut-specific IgE ≥15 kU/L

      SPT to peanut ≥8 mm

      Peanut-specific IgE ≥15 kU/L and SPT to peanut ≥8 mm
      17

      73

      26
      Total390
      A total of 390 patients with peanut allergy were drawn from an established Canadian case collection, defined as having peanut allergy on the basis of a positive oral food challenge (n = 25) or, in the non–food-challenged subjects, a peanut-specific IgE ≥15 kUL−1 (n = 65) or an SPT wheal to peanut of ≥8 mm (n = 214) or both sIgE ≥15 kUL−1 and SPT wheal ≥8 mm (n = 86).

       Canadian population controls

      A total of 891 Canadian controls of white ethnicity were derived from a population collection of adult volunteers from the Toronto, Ontario, area, for whom peanut allergy status and the presence or absence of coexistent AD were unknown.
      Demographic data and clinical characteristics for the replication collection are shown in Table III.
      Table IIIDemographic data and clinical characteristics of the Canadian replication study
      Patients and controlsNo. of individualsAge (y),

      range (mean)
      Male sex,

      n (%)
      Coexistent AD,

      n (%)
      White Canadian patients with peanut allergy3900-21 (9.5)239 (61.3)286/383 (69.5)
      White Canadian population controls89123-77 (57.5)281 (31.5)Unknown
      AD is defined according to parental report. Data were available in 383 Canadian patients; 266 reported AD, of whom 257 reported that a physician had made or confirmed the diagnosis of AD; % coexistent AD refers to the proportion of patients for whom AD data were available.

       Ethical considerations

      The peanut allergy case collections were approved by the relevant local research ethics committees, and all subjects or the subjects’ guardians gave written informed consent. DNA was collected and analyzed from individuals within each of the control groups with ethical approval and written informed consent.

       Genotyping for FLG loss-of-function mutations

      DNA was extracted from blood or saliva samples by using standard protocols. The FLG loss-of-function mutations that are most prevalent in the European population were genotyped by using a previously published methodology.
      • Sandilands A.
      • Terron-Kwiatkowski A.
      • Hull P.
      • O’Regan G.
      • Clayton T.
      • Watson R.
      • et al.
      Comprehensive analysis of the gene encoding filaggrin uncovers prevalent and rare mutations in ichthyosis vulgaris and atopic eczema.
      R501X and 2282del4 had previously been typed in the ALSPAC cohort; R501X, 2282del4, R2447X, and S3247X were typed in the Dutch, Canadian, and Irish case collections in this study.

       Statistical analysis

       Case-control analysis to test the association between filaggrin genotype and peanut allergy

      The Fisher exact test and logistic regression were used to test for association between peanut allergy and the FLG loss-of-function mutations as a combined null genotype. The rationale for creating a combined null genotype for FLG is based on biochemical and immunohistochemical studies demonstrating that each FLG null mutation has an equivalent biological effect.
      • Sandilands A.
      • Sutherland C.
      • Irvine A.
      • McLean W.
      Filaggrin in the frontline: role in skin barrier function and disease.
      • Sandilands A.
      • Terron-Kwiatkowski A.
      • Hull P.
      • O’Regan G.
      • Clayton T.
      • Watson R.
      • et al.
      Comprehensive analysis of the gene encoding filaggrin uncovers prevalent and rare mutations in ichthyosis vulgaris and atopic eczema.
      The combined null genotype in the pooled English/ Dutch/Irish case collection used data relating to the 2 mutations (R501X and 2282del4) because these were available for all patients and the controls; the combined null genotype in the Canadian patients and controls relates to the 4 mutations (R501X, 2282del4, R2447X, and S3247X), and these 4 mutations were also analyzed in the subgroups of Dutch and Irish patients with their relevant control populations.
      Analysis was performed with the statistical analysis package Stata (version 9; StataCorp LP, College Station, Tex). Logistic regression models the log odds of disease as a linear function of variables encoding allele effects at the relevant loci. A genotype variable coded as 0, 1, or 2 according to the number of mutant alleles carried at the FLG locus was included in the regression equation.

       Covariate analysis to control for coexistent AD

      FLG null mutations are strongly and significantly associated with AD,
      • Rodríguez E.
      • Baurecht H.
      • Herberich E.
      • Wagenpfeil S.
      • Brown S.
      • Cordell H.
      • et al.
      Meta-analysis of filaggrin polymorphisms in eczema and asthma: robust risk factors in atopic disease.
      and AD is more prevalent among patients with peanut allergy than in the general population.
      • Lack G.
      • Fox D.
      • Northstone K.
      • Golding J.
      Factors associated with the development of peanut allergy in childhood.
      • Boyce J.A.
      • Assa’ad A.
      • Burks A.W.
      • Jones S.M.
      • Sampson H.A.
      • Wood R.A.
      • et al.
      Guidelines for the diagnosis and management of food allergy in the United States: summary of the NIAID-sponsored Expert Panel Report.
      We therefore investigated the role of coexistent AD as a risk factor and potential confounder in the peanut-FLG association by using covariate analysis in the food challenge–positive dataset. This analysis was not possible in the Canadian dataset because AD status in the control population is unknown. Covariate analysis was performed by including AD as a variable in the logistic regression equation using Stata.

      Results

       FLG null mutations are strongly and significantly associated with peanut allergy demonstrated by positive food challenge

      Analysis of the 71 patients with peanut allergy defined by positive oral challenge, when compared with 1000 non–peanut-sensitized controls, showed a strong and significant association of FLG loss-of-function mutations with peanut allergy (odds ratio [OR], 5.3; 95% CI, 2.8-10.2; Fisher exact test, P = 3.0 × 10−6; Table IV).
      Table IVGenotyping results and statistical analysis of filaggrin loss-of-function mutations in 71 food challenge–proven patients with peanut allergy and 1000 non–peanut-sensitized controls
      Genotypes and statistical testsPatients (n)Controls (n)
      No FLG mutations detected (homozygous wild-type)59963
      One wild-type and 1 FLG mutant allele (heterozygous)1037
      Two FLG loss-of-function mutations (homozygous or compound heterozygous)20
      Total711000
      Proportion of individuals carrying FLG loss-of-function mutations (%)16.93.7
      Fisher exact testP = 3.0 × 10−6
      OR (95% CI)5.3 (2.8-10.2)
      This combined population dataset relates to the 2 most prevalent FLG loss-of-function mutations (R501X and 2282del4) in English (n = 35), Dutch (n = 20), and Irish (n = 16) patients with peanut allergy demonstrated by oral food challenge. The designations homozygous wild-type, heterozygous, and 2 null mutations represent the combined null genotype results for the 2 available mutations; individuals with 2 null mutations may be homozygous for an FLG null allele or compound heterozygous, having 2 different FLG null alleles.
      Comparison of the English (n = 35) and Dutch (n = 20) patients with peanut allergy with their respective population controls in separate case-control analyses also showed a statistically significant association between FLG null mutations and peanut allergy in each population subgroup; analysis of the Irish patients with peanut allergy (n = 16) in a separate case-control analysis did not give a statistically significant result because of the small number of patients (Table V).
      Table VGenotyping results and statistical analysis of filaggrin loss-of-function mutations in patients with peanut allergy and matched controls from English, Dutch, and Irish populations
      Genotypes and statistical testsEnglish (ALSPAC)DutchIrish
      Patients (n)Controls (n)Patients (n)Controls (n)Patients (n)Controls (n)
      No FLG mutations detected (homozygous wild-type)28636817951393
      One wild-type and 1 FLG mutant allele (heterozygous)74801527
      Two FLG loss-of-function mutations (homozygous or compound heterozygous)032010
      Total3568512010016100
      Proportion of individuals carrying FLG loss-of-function mutations (%)20.07.115.05.018.87.0
      Fisher exact testP = .0251P = .0335P = .0640
      OR (95% CI)3.2 (1.4-7.2)3.5 (1.1-11.4)3.3 (1.0-11.7)
      The ALSPAC is a longitudinal, population-based birth cohort study.
      • Golding J.
      • Pembrey M.
      • Jones R.
      ALSPAC–the Avon Longitudinal Study of Parents and Children, I: study methodology.
      This English population was screened for the 2 most common FLG null mutations (R501X and 2282del4). In the Dutch and Irish patients and controls, the 4 most prevalent FLG loss-of-function mutations were assayed (R501X, 2282del4, R2447X, and S3247X). The designations homozygous wild-type, heterozygous, and 2 null mutations represent the combined null genotype results for the available mutations in each population.

       FLG null mutations are associated with peanut allergy in a Canadian case-control study

      Table VI summarizes the genotype data and association analysis for the Canadian patients and controls. The highly significant association is replicated (P = 5.4 × 10−5) with an estimated OR of 1.9 (95% CI, 1.2-2.6).
      Table VIGenotyping results and statistical analysis of filaggrin loss-of-function mutations in a case-control study in the Canadian population
      Genotypes and statistical testsPatients (n)Controls (n)
      No FLG mutations detected (homozygous wild-type)315793
      One wild-type and 1 FLG mutant allele (heterozygous)6694
      Two FLG loss-of-function mutations (homozygous or compound heterozygous)94
      Total390891
      Proportion of individuals carrying FLG loss-of-function mutations (%)19.211.0
      Fisher exact testP = 5.4 × 10−5
      OR (95% CI)1.9 (1.4-2.6)
      Patients and controls were screened for the 4 most prevalent FLG loss-of-function mutations (R501X, 2282del4, R2447X, and S3247X). The designations homozygous wild-type, heterozygous, and 2 null mutations represent the combined null genotype results for the 4 available mutations.

       Association between FLG null mutations and peanut allergy is not solely attributable to coexistent AD

      Covariate analysis using both AD and FLG as predictors in the logistic regression model demonstrated the strong association of atopic eczema with peanut allergy in the 71 food challenge–positive patients, with an OR of 7.4 (95% CI, 4.1-13.7). However, in these same patients, having controlled for AD, the residual OR for FLG in association with peanut allergy was 3.8 (95% CI, 1.7-8.3; P = .0008).

      Discussion

      The major known risk factors for IgE-mediated peanut allergy are family history of the disease and coexistent atopy.
      • Lack G.
      • Fox D.
      • Northstone K.
      • Golding J.
      Factors associated with the development of peanut allergy in childhood.
      However, despite the clinical importance of peanut allergy and its strong heritability, no significant genetic risk factors have previously been confirmed by replication studies. Taken together, our experimental data from 4 populations of European origin demonstrate a strong and significant association of loss-of-function mutations within the FLG gene with clinically significant peanut allergy. Furthermore, we have confirmed, by using covariate analysis in the food challenge–positive patients, that this association does not simply represent a coassociation with AD. The association between FLG and peanut allergy is not fully accounted for by confounding with coexistent AD, because the statistically significant association of the combined FLG null genotype with peanut allergy persists even after controlling for AD (P = .0008).
      The differing case collections chosen for this genetic epidemiologic study offer complementary insights into the possible link between FLG genotype and peanut allergy. The white European dataset had a positive result to food challenge testing that was double-blind and placebo-controlled in the English (n = 35) and Dutch (n = 20) patients and an open challenge in the Irish (n = 16) patients. Double-blind placebo-controlled oral food challenge is the gold standard in a research setting, but open or single-blind challenges are more standard clinical practice,
      • Burks A.W.
      Peanut allergy.
      • Boyce J.A.
      • Assa’ad A.
      • Burks A.W.
      • Jones S.M.
      • Sampson H.A.
      • Wood R.A.
      • et al.
      Guidelines for the diagnosis and management of food allergy in the United States: summary of the NIAID-sponsored Expert Panel Report.
      particularly in children. The ALSPAC cohort is a large, unselected population birth cohort in which data have been collected on a prospective basis, providing the opportunity for a case-control study in which the controls are appropriately matched to patients and in which data on coexistent AD have been collected. The Canadian collection of patients with peanut allergy, however, provides a much greater statistical power because of the larger number of individuals with peanut allergy available for analysis. There is a striking age difference between the Canadian patients and the control population. However, because AD does not affect survival pattern, this study design would not be predicted to affect the findings or interpretation of data.
      The higher OR of disease in the combined English, Dutch, and Irish dataset compared with the Canadian study may be explained by the use of a hypernormal (nonsensitized) English control population, whereas the Canadian population controls would include individuals with current or previous peanut sensitization or allergy, resulting in a lower OR. However, it is interesting to note that the estimated OR is 3.2 to 3.5 (95% CI, 1.0-11.7) in each of the population subgroups when they are compared with their respective unselected population controls (Table V). Alternatively, or in addition, there may be unidentified FLG null mutations within the white Canadian population that have not been included in the statistical analysis, therefore possibly leading to an underestimation of the FLG association. We limited our analysis in these large studies to the 2 or 4 most common FLG mutations and are therefore likely to have underestimated the total contribution of FLG null alleles to this disease association. We limited our analyses to these selected mutations because the FLG gene is large and its DNA sequence is highly repetitive, making genotyping technically difficult.
      • Sandilands A.
      • Sutherland C.
      • Irvine A.
      • McLean W.
      Filaggrin in the frontline: role in skin barrier function and disease.
      It would be interesting and informative to dissect further the association of FLG status with AD, food sensitization, and allergy. However, the separation of AD, food allergy, and food sensitization data is complex and challenging because these phenotypes show considerable overlap. Furthermore, the definition of peanut allergy in the Canadian patients necessarily includes peanut sensitization data, and our analysis is limited because data on AD and food allergy phenotypes are not available for the Canadian control population. The analysis of FLG genotype and peanut allergy with AD as a confounding factor was therefore limited to the food challenge–positive patients and nonsensitized controls.
      The estimated OR for FLG as a risk factor for peanut allergy is 5.3 (95% CI, 2.8-10.2) in the pooled analysis of English/Dutch/Irish food challenge–positive patients for whom 2 FLG loss-of-function mutations were assessed; the equivalent OR is 1.9 (95% CI, 1.4-2.6) in the Canadian collection (Table VI), 3.5 (95% CI, 1.1-11.4) in the Dutch collection, and 3.3 (95% CI, 1.0-11.7) in the Irish collection (Table V), for whom 4 FLG null mutations were assessed. A total of over 20 FLG null mutations have now been identified within populations of European ancestry. In the Irish population, the most intensively studied white European population with respect to FLG, R501X and 2282del4 together represent about 66% of the FLG mutations; adding R2447X and S3247X covers about 95% of the known FLG mutations.
      The prevalence of food challenge–positive peanut allergy within this subset of the 1989 to 1990 ALSPAC birth cohort is 0.6%, consistent with other published data.
      • Burks A.W.
      Peanut allergy.
      The proportion of these patients carrying 1 or more FLG null mutations (associated with an estimated OR of 3.2) is 20.0% (Table V). Similarly, in the Canadian study, 19.2% of the patients with peanut allergy carry 1 or more FLG null mutations, with an estimated OR of 1.9 (Table VI). These data demonstrate the striking contribution of a single gene in the etiology of a complex trait resulting from multiple genetic and environmental effects.
      The mechanisms by which filaggrin deficiency in the skin may lead to AD and other atopic disorders have been debated extensively.
      • Hudson T.J.
      Skin barrier function and allergic risk.
      • Elias P.M.
      • Schmuth M.
      Abnormal skin barrier in the etiopathogenesis of atopic dermatitis.
      • Hanifin J.M.
      Evolving concepts of pathogenesis in atopic dermatitis and other eczemas.
      It is proposed that impairment of the epidermal barrier allows the penetration of allergens and irritants, resulting in local inflammation as well as a systemic atopic response. There is experimental evidence in support of this hypothesis from mouse models
      • Fallon P.G.
      • Sasaki T.
      • Sandilands A.
      • Campbell L.E.
      • Saunders S.P.
      • Mangan N.E.
      • et al.
      A homozygous frameshift mutation in the mouse Flg gene facilitates enhanced percutaneous allergen priming.
      • Strid J.
      • Hourihane J.
      • Kimber I.
      • Callard R.
      • Strobel S.
      Disruption of the stratum corneum allows potent epicutaneous immunization with protein antigens resulting in a dominant systemic Th2 response.
      • Man M.Q.
      • Hatano Y.
      • Lee S.H.
      • Man M.
      • Chang S.
      • Feingold K.R.
      • et al.
      Characterization of a hapten-induced, murine model with multiple features of atopic dermatitis: structural, immunologic and biochemical changes following single versus multiple oxazolone challenges.
      • Oyoshi M.K.
      • Murphy G.F.
      • Geha R.S.
      Filaggrin-deficient mice exhibit TH17-dominated skin inflammation and permissiveness to epicutaneous sensitization with protein antigen.
      as well as clinical epidemiologic studies showing that FLG null mutations are a risk factor for asthma only in the subtype of asthma that is associated with AD.
      • Rodríguez E.
      • Baurecht H.
      • Herberich E.
      • Wagenpfeil S.
      • Brown S.
      • Cordell H.
      • et al.
      Meta-analysis of filaggrin polymorphisms in eczema and asthma: robust risk factors in atopic disease.
      Similarly, it has been suggested that peanut sensitization may result from transcutaneous exposure to peanut allergens in the environment
      • Fox A.T.
      • Sasieni P.
      • du Toit G.
      • Syed H.
      • Lack G.
      Household peanut consumption as a risk factor for the development of peanut allergy.
      or in topical preparations,
      • Lack G.
      • Fox D.
      • Northstone K.
      • Golding J.
      Factors associated with the development of peanut allergy in childhood.
      particularly because early-onset AD is a risk factor for peanut allergy.
      • Lack G.
      • Fox D.
      • Northstone K.
      • Golding J.
      Factors associated with the development of peanut allergy in childhood.
      This contrasts with the oral route of exposure, which is thought to be tolerogenic.
      • Sicherer S.H.
      • Sampson H.A.
      Peanut allergy: emerging concepts and approaches for an apparent epidemic.
      However, intestinal permeability is known to be increased amongst some patients with AD,
      • Pike M.G.
      • Heddle R.J.
      • Boulton P.
      • Turner M.W.
      • Atherton D.J.
      Increased intestinal permeability in atopic eczema.
      and it has been suggested that intestinal permeability may be an additional route of allergen penetration in peanut allergy and other food allergies.
      • Sicherer S.H.
      • Sampson H.A.
      Food allergy.
      The extent and functional significance of filaggrin expression in the gastrointestinal tract below the oral mucosa is currently unknown but warrants further investigation. This study focused on the association between FLG loss-of-function mutations and peanut allergy; the possibility of association with other food allergies has not been addressed and also warrants further investigation. One previous study of FLG genotype and sensitization to foods was conducted as a subgroup analysis within a German cohort study on asthma risk.
      • Marenholz I.
      • Kerscher T.
      • Bauerfeind A.
      • Esparza-Gordillo J.
      • Nickel R.
      • Keil T.
      • et al.
      An interaction between filaggrin mutations and early food sensitization improves the prediction of childhood asthma.
      This analysis showed a strong synergistic interaction between the FLG-null alleles and early food sensitization in the disease transition from eczema to asthma (relative excess risk, 2.64) but no significant association of FLG status with food sensitization per se in 185 children with AD. Although the subgroup sample size was rather small, these contrasting data may reflect differences in the pathogenesis of food sensitization versus food allergy as well as possible differences in the pathogenesis of allergy to different foods.
      The well recognized strong association of AD with peanut allergy is demonstrated in the results of covariate analysis, in which eczema shows an estimated OR of 7.4 compared with the FLG estimated OR (after accounting for AD) of 3.8. Hence, AD is a stronger independent risk factor for peanut allergy than FLG null mutations. However, it is interesting and noteworthy that the FLG genotype effect is not solely accounted for by a coassociation with AD, a finding that contributes to the debate outlined regarding the routes and mechanisms of allergic sensitization. It is possible that the association of FLG loss-of-function mutations persists because of incorrect assignment of patients with AD to the non-AD subgroup (ie, incomplete case ascertainment of AD on the basis of these definitions). The case definition used for AD, flexural dermatitis reported by parents on at least 2 occasions (in the ALSPAC cohort), could possibly have failed to include children with mild, localized, or transient disease in the AD case collection. However, the high prevalence of coexistent AD in the ALSPAC cohort, at 27.1% in subjects without peanut allergy (Table III) and 27.6% in the cohort as a whole, in comparison with published UK studies,
      • Shamssain M.
      Trends in the prevalence and severity of asthma, rhinitis and atopic eczema in 6- to 7- and 13- to 14-yr-old children from the north-east of England.
      suggests that few patients have been missed by this case definition. Absence of established AD according to the case definitions used in these studies does not preclude a significant epidermal barrier defect in filaggrin-deficient individuals. Murine models of filaggrin deficiency have shown enhanced percutaneous allergen sensitization in 3 studies.
      • Fallon P.G.
      • Sasaki T.
      • Sandilands A.
      • Campbell L.E.
      • Saunders S.P.
      • Mangan N.E.
      • et al.
      A homozygous frameshift mutation in the mouse Flg gene facilitates enhanced percutaneous allergen priming.
      • Man M.Q.
      • Hatano Y.
      • Lee S.H.
      • Man M.
      • Chang S.
      • Feingold K.R.
      • et al.
      Characterization of a hapten-induced, murine model with multiple features of atopic dermatitis: structural, immunologic and biochemical changes following single versus multiple oxazolone challenges.
      • Oyoshi M.K.
      • Murphy G.F.
      • Geha R.S.
      Filaggrin-deficient mice exhibit TH17-dominated skin inflammation and permissiveness to epicutaneous sensitization with protein antigen.
      These mice appear to have an inherent epithelial barrier defect that facilitates allergen priming in the absence of clinical evidence of cutaneous inflammation.
      Here we have shown that FLG mutations confer a risk for peanut allergy in the absence of clinical evidence of AD. In contrast, although FLG null alleles confer a significant risk of asthma, with a reproducible OR of 1.8, this effect is confined to those with pre-existent or coexistent AD.
      • Rodríguez E.
      • Baurecht H.
      • Herberich E.
      • Wagenpfeil S.
      • Brown S.
      • Cordell H.
      • et al.
      Meta-analysis of filaggrin polymorphisms in eczema and asthma: robust risk factors in atopic disease.
      The mechanistic pathways through which FLG null alleles contribute risk to these related but distinct atopic phenotypes are as yet unclear.
      In conclusion, FLG null mutations not only represent a highly significant genetic risk factor for AD but also are the single most significant genetic risk for peanut allergy that has been identified to date. The robustness of this association is demonstrated by replication within 3 populations— English, Dutch, and Canadian—plus an association close to significance in an Irish collection. The association of FLG mutations with peanut allergy adds further insight into the pathogenesis of the atopic state and the critical role of epithelial barrier function in health and disease.
      Key messages
      • Loss-of-function variants in the filaggrin gene are strongly and significantly associated with peanut allergy.
      • The association of filaggrin mutations with peanut allergy cannot fully be accounted for by the confounding effect of coexistent AD.
      • Filaggrin mutations are a novel risk factor for IgE-mediated peanut allergy, indicating a role for epithelial barrier dysfunction in the pathogenesis of this disease.
      We are grateful for all the mothers who took part in the ALSPAC study, the midwives’ cooperation and help in recruitment, and the whole of the ALSPAC Study Team. We acknowledge the help of Mary Fujiwara, McGill University.

      References

        • Bock S.A.
        • Munoz-Furlong A.
        • Sampson H.A.
        Further fatalities caused by anaphylactic reactions to food, 2001-2006.
        J Allergy Clin Immunol. 2007; 119: 1016-1018
        • Burks A.W.
        Peanut allergy.
        Lancet. 2008; 371: 1538-1546
        • Sicherer S.H.
        • Munoz-Furlong A.
        • Godbold J.H.
        • Sampson H.A.
        US prevalence of self-reported peanut, tree nut, and sesame allergy: 11-year follow-up.
        J Allergy Clin Immunol. 2010; 125: 1322-1326
        • Venter C.
        • Hasan Arshad S.
        • Grundy J.
        • Pereira B.
        • Bernie Clayton C.
        • Voigt K.
        • et al.
        Time trends in the prevalence of peanut allergy: three cohorts of children from the same geographical location in the UK.
        Allergy. 2010; 65: 103-108
        • Ben-Shoshan M.
        • Kagan R.S.
        • Alizadehfar R.
        • Joseph L.
        • Turnbull E.
        • St Pierre Y.
        • et al.
        Is the prevalence of peanut allergy increasing? a 5-year follow-up study in children in Montreal.
        J Allergy Clin Immunol. 2009; 123: 783-788
        • Sicherer S.H.
        • Furlong T.J.
        • Maes H.H.
        • Desnick R.J.
        • Sampson H.A.
        • Gelb B.D.
        Genetics of peanut allergy: a twin study.
        J Allergy Clin Immunol. 2000; 106: 53-56
        • Hourihane J.O.
        • Dean T.P.
        • Warner J.O.
        Peanut allergy in relation to heredity, maternal diet, and other atopic diseases: results of a questionnaire survey, skin prick testing, and food challenges.
        BMJ. 1996; 313: 518-521
        • Howell W.M.
        • Turner S.J.
        • Hourihane J.O.
        • Dean T.P.
        • Warner J.O.
        HLA class II DRB1, DQB1 and DPB1 genotypic associations with peanut allergy: evidence from a family-based and case-control study.
        Clin Exp Allergy. 1998; 28: 156-162
        • Shreffler W.G.
        • Charlop-Powers Z.
        • Sicherer S.H.
        Lack of association of HLA class II alleles with peanut allergy.
        Ann Allergy Asthma Immunol. 2006; 96: 865-869
        • Rodríguez E.
        • Baurecht H.
        • Herberich E.
        • Wagenpfeil S.
        • Brown S.
        • Cordell H.
        • et al.
        Meta-analysis of filaggrin polymorphisms in eczema and asthma: robust risk factors in atopic disease.
        J Allergy Clin Immunol. 2009; 123 (e7): 1361-1370
        • van den Oord R.
        • Sheikh A.
        Filaggrin gene defects and risk of developing allergic sensitisation and allergic disorders: systematic review and meta-analysis.
        BMJ. 2009; 339: b2433
        • Sandilands A.
        • Sutherland C.
        • Irvine A.
        • McLean W.
        Filaggrin in the frontline: role in skin barrier function and disease.
        J Cell Sci. 2009; 122: 1285-1294
        • De Benedetto A.
        • Qualia C.M.
        • Baroody F.M.
        • Beck L.A.
        Filaggrin expression in oral, nasal, and esophageal mucosa.
        J Invest Dermatol. 2008; 128: 1594-1597
        • Blanchard C.
        • Stucke E.M.
        • Burwinkel K.
        • Caldwell J.M.
        • Collins M.H.
        • Ahrens A.
        • et al.
        Coordinate interaction between IL-13 and epithelial differentiation cluster genes in eosinophilic esophagitis.
        J Immunol. 2010; 184: 4033-4041
        • Hudson T.J.
        Skin barrier function and allergic risk.
        Nat Genet. 2006; 38: 399-400
        • Leung D.Y.
        Our evolving understanding of the functional role of filaggrin in atopic dermatitis.
        J Allergy Clin Immunol. 2009; 124: 494-495
        • Fallon P.G.
        • Sasaki T.
        • Sandilands A.
        • Campbell L.E.
        • Saunders S.P.
        • Mangan N.E.
        • et al.
        A homozygous frameshift mutation in the mouse Flg gene facilitates enhanced percutaneous allergen priming.
        Nat Genet. 2009; 41: 602-608
        • Golding J.
        • Pembrey M.
        • Jones R.
        ALSPAC–the Avon Longitudinal Study of Parents and Children, I: study methodology.
        Paediatr Perinat Epidemiol. 2001; 15: 74-87
        • Ben-Shoshan M.
        • Harrington D.W.
        • Soller L.
        • Fragapane J.
        • Joseph L.
        • St Pierre Y.
        • et al.
        A population-based study on peanut, tree nut, fish, shellfish, and sesame allergy prevalence in Canada.
        J Allergy Clin Immunol. 2010; 125: 1327-1335
        • Ben-Shoshan M.
        • Kagan R.
        • Primeau M.N.
        • Alizadehfar R.
        • Verreault N.
        • Yu J.W.
        • et al.
        Availability of the epinephrine autoinjector at school in children with peanut allergy.
        Ann Allergy Asthma Immunol. 2008; 100: 570-575
        • Yu J.W.
        • Kagan R.
        • Verreault N.
        • Nicolas N.
        • Joseph L.
        • St Pierre Y.
        • et al.
        Accidental ingestions in children with peanut allergy.
        J Allergy Clin Immunol. 2006; 118: 466-472
        • Kagan R.
        • Hayami D.
        • Joseph L.
        • St Pierre Y.
        • Clarke A.E.
        The predictive value of a positive prick skin test to peanut in atopic, peanut-naive children.
        Ann Allergy Asthma Immunol. 2003; 90: 640-645
        • Roberts G.
        • Lack G.
        Team tALSoPaCS. ALSPAC cohort.
        J Allergy Clin Immunol. 2005; 115: 1291-1296
        • Sporik R.
        • Hill D.J.
        • Hosking C.S.
        Specificity of allergen skin testing in predicting positive open food challenges to milk, egg and peanut in children.
        Clin Exp Allergy. 2000; 30: 1540-1546
        • Wainstein B.K.
        • Yee A.
        • Jelley D.
        • Ziegler M.
        • Ziegler J.B.
        Combining skin prick, immediate skin application and specific-IgE testing in the diagnosis of peanut allergy in children.
        Pediatr Allergy Immunol. 2007; 18: 231-239
        • Sampson H.A.
        • Munoz-Furlong A.
        • Campbell R.L.
        • Adkinson Jr., N.F.
        • Bock S.A.
        • Branum A.
        • et al.
        Second symposium on the definition and management of anaphylaxis: summary report–second National Institute of Allergy and Infectious Disease/Food Allergy and Anaphylaxis Network symposium.
        Ann Emerg Med. 2006; 47: 373-380
        • Hourihane J.O.
        • Kilburn S.A.
        • Dean P.
        • Warner J.O.
        Clinical characteristics of peanut allergy.
        Clin Exp Allergy. 1997; 27: 634-639
        • Sandilands A.
        • Terron-Kwiatkowski A.
        • Hull P.
        • O’Regan G.
        • Clayton T.
        • Watson R.
        • et al.
        Comprehensive analysis of the gene encoding filaggrin uncovers prevalent and rare mutations in ichthyosis vulgaris and atopic eczema.
        Nat Genet. 2007; 39: 650-654
        • Lack G.
        • Fox D.
        • Northstone K.
        • Golding J.
        Factors associated with the development of peanut allergy in childhood.
        N Engl J Med. 2003; 348: 977-985
        • Boyce J.A.
        • Assa’ad A.
        • Burks A.W.
        • Jones S.M.
        • Sampson H.A.
        • Wood R.A.
        • et al.
        Guidelines for the diagnosis and management of food allergy in the United States: summary of the NIAID-sponsored Expert Panel Report.
        J Allergy Clin Immunol. 2010; 126: 1105-1118
        • Elias P.M.
        • Schmuth M.
        Abnormal skin barrier in the etiopathogenesis of atopic dermatitis.
        Curr Allergy Asthma Rep. 2009; 9: 265-272
        • Hanifin J.M.
        Evolving concepts of pathogenesis in atopic dermatitis and other eczemas.
        J Invest Dermatol. 2009; 129: 320-322
        • Strid J.
        • Hourihane J.
        • Kimber I.
        • Callard R.
        • Strobel S.
        Disruption of the stratum corneum allows potent epicutaneous immunization with protein antigens resulting in a dominant systemic Th2 response.
        Eur J Immunol. 2004; 34: 2100-2109
        • Man M.Q.
        • Hatano Y.
        • Lee S.H.
        • Man M.
        • Chang S.
        • Feingold K.R.
        • et al.
        Characterization of a hapten-induced, murine model with multiple features of atopic dermatitis: structural, immunologic and biochemical changes following single versus multiple oxazolone challenges.
        J Invest Dermatol. 2008; 128: 79-86
        • Oyoshi M.K.
        • Murphy G.F.
        • Geha R.S.
        Filaggrin-deficient mice exhibit TH17-dominated skin inflammation and permissiveness to epicutaneous sensitization with protein antigen.
        J Allergy Clin Immunol. 2009; 124 (93 e1): 485-493
        • Fox A.T.
        • Sasieni P.
        • du Toit G.
        • Syed H.
        • Lack G.
        Household peanut consumption as a risk factor for the development of peanut allergy.
        J Allergy Clin Immunol. 2009; 123: 417-423
        • Sicherer S.H.
        • Sampson H.A.
        Peanut allergy: emerging concepts and approaches for an apparent epidemic.
        J Allergy Clin Immunol. 2007; 120 (quiz 4-5): 491-503
        • Pike M.G.
        • Heddle R.J.
        • Boulton P.
        • Turner M.W.
        • Atherton D.J.
        Increased intestinal permeability in atopic eczema.
        J Invest Dermatol. 1986; 86: 101-104
        • Sicherer S.H.
        • Sampson H.A.
        Food allergy.
        J Allergy Clin Immunol. 2010; 125: S116-S125
        • Marenholz I.
        • Kerscher T.
        • Bauerfeind A.
        • Esparza-Gordillo J.
        • Nickel R.
        • Keil T.
        • et al.
        An interaction between filaggrin mutations and early food sensitization improves the prediction of childhood asthma.
        J Allergy Clin Immunol. 2009; 123: 911-916
        • Shamssain M.
        Trends in the prevalence and severity of asthma, rhinitis and atopic eczema in 6- to 7- and 13- to 14-yr-old children from the north-east of England.
        Pediatr Allergy Immunol. 2007; 18: 149-153