Filaggrin gene variants and atopic diseases in early childhood assessed longitudinally from birth

Pediatr Allergy Immunol 2010: 21: 954–961 Filaggrin gene variants and atopic diseases inearly childhood assessed longitudinally frombirth Bønnelykke K, Pipper CB, Tavendale R, Palmer CNA, Bisgaard H.
Filaggrin gene variants and atopic diseases in early childhood assessed Pipper1, Roger Tavendale2, Colin N. A.
Pediatr Allergy Immunol 2010: 21: 954–961.
1Copenhagen Studies on Asthma in Childhood; Copenhagen University Hospital, Gentofte, Denmark,2Population Pharmacogenetics Group, Biomedical Copenhagen Prospective Study on Asthma in Childhood (COPSAC) Research Centre, University of Dundee, NinewellsHospital and Medical School, Dundee, UK was one of the discovery cohorts of the association between eczemaand variants in the filaggrin coding gene (FLG). Here, we study theFLG-associated risk of asthma symptoms in early life and describe thetemporal relationship in the development of the different FLG-asso-ciated atopic outcomes: asthma, sensitization and eczema, assessedlongitudinally from birth. A high-risk cohort of 411 children wasassessed in a prospective clinical study from birth to school-age.
Asthma, acute severe asthma exacerbations, sensitization and eczemawere diagnosed prospectively by the investigators. FLG variantsR501X and Del4 were determined in 382 Caucasians. Filaggrin vari-ants increased risk of developing recurrent wheeze, asthma and asthmaexacerbations (hazard ratio 1.82 [1.06–3.12], p = 0.03), which wasexpressed within the first 1.5 yr of life. Children with filaggrin variants Key words: filaggrin; gene; asthma; sensitization; had a marked and persistent increase in acute severe asthma exacer- bations from 1 yr of age (incidence ratio 2.40 [1.19–4.81], p = 0.01) Klaus Bønnelykke, Copenhagen Studies on Asthma in and increased risk of asthma by age 5 (odds ratio 2.62 [1.12–6.11], Childhood, University of Copenhagen, Gentofte p = 0.03). FLG variants increased the risk of eczema, manifesting Hospital, Ledreborg Alle 34, DK-2820 Gentofte, fully in the first year of life (point prevalence ratio for age 0–5 was 1.75 [1.29–2.37]; p-value = 0.0003) contrasting the increased risk of specific sensitization by age 4 (odds ratio 3.52 [1.72–7.25], p = 0.0007) but not age 1.5. This study describes a FLG-associated pattern of atopic diseases characterized by the early onset of asthma symptomsand eczema and later development of sensitization. The association of Data from this manuscript have been presented as an filaggrin variants with asthma suggests skin barrier dysfunction as a abstract at the ATS congress 2009 but not elsewherein abstract or any other form.
novel, and potentially modifiable, mechanism driving early childhoodasthma.
Copenhagen Prospective Study on Asthma in Such association between skin barrier dysfunc- Childhood (COPSAC) was one of the discovery tion and airway disease is intriguing as it cohorts of loss-of-function variants in the gene indicates a novel and potentially modifiable encoding filaggrin (FLG) as major determinants mechanism in asthma pathogenesis. To under- of eczema (1) and this has been consistently stand this mechanism, there is a need for replicated (2–10). An association with asthma longitudinal studies in early life describing the was also suggested in the discovery cohorts (1) temporal relationship in the development of and confirmed in subsequent reports on asthma occurring in association with eczema (2–4, 6, 8– Also, asthma in early childhood differs from 10) though one study reported no association separately. The majority of children with asthma Filaggrin, early asthma and sensitization symptoms in early life outgrow their symptoms birth. Wheeze was translated to the parents as before school-age, but the symptom burden in wheeze or whistling sounds, breathlessness or this age-group is high with severe exacerbations recurrent troublesome cough severely affecting and hospitalization being more common than the well-being of the infant, and was recorded as composite dichotomized scores (yes/no) as We therefore analyzed FLG variants against previously described in details (13). The doctor at the longitudinal clinical diagnoses of recurrent the clinical research unit reviewed symptom wheeze, acute severe asthma exacerbations, definition and the diary entries with the parents asthma and eczema together with assessments at the 6-monthly clinical sessions as well as at of sensitization and describe the temporal rela- acute severe asthma exacerbations. Recurrent tionship in the development of the different wheeze was predefined as five episodes within FLG-associated atopic outcomes through the 6 months, each episode lasting at least three first 5 yr of life in the prospective clinical study of a high-risk birth-cohort ÔCOPSACÕ (12, 13).
four consecutive weeks documented from thediaries.
Asthma was diagnosed according to the inter- national guidelines as previously detailed (13)based on recurrent wheeze as defined previously; COPSAC is a longitudinal clinical study of a research unit doctor to be typical of asthma birth-cohort of 411 infants born to mothers with (e.g., exercise-induced symptoms, prolonged noc- a history of asthma; the recruitment of which was turnal cough, recurrent cough outside common previously described in detail (12, 13). Data cold, symptoms causing wakening at night); in validity and quality control procedures follow need of intermittent rescue use of inhaled b2- ÔGood Clinical PracticeÕ guidelines. History is agonist; and responding to a 3-month course of collected online during visits to the COPSAC inhaled corticosteroids and relapsing when stop- clinical research unit. Objective measurements ping treatment. Asthma was also diagnosed are double-checked against source data and the without a previous history of wheeze in case of database subsequently locked. An audit trail is acute severe asthma exacerbation as defined The COPSAC clinical research unit provided regular as well as acute clinical assessments for defined from need of oral prednisolone or all participating children, who attended this clinic high-dose inhaled corticosteroid for wheezy instead of family practitioners or other health symptoms prescribed at the discretion of the care providers for diagnosis and treatment of any doctor at the clinical research unit or acute respiratory or skin-related symptoms. Regular hospitalization for treatment for such symptoms visits were scheduled at the COPSAC clinical research unit at 6-monthly intervals, and addi- tional visits were arranged immediately upon the defined as onset of any of the previously onset of symptoms. At every visit, the child was mentioned diagnoses and this measure was used seen by both doctor and nurses of the research clinic and a full physical examination and historywas obtained using structured questions and Eczema. Skin lesions were described at both closed response categories focusing on the childÕs scheduled and acute visits according to prede- airway and skin symptoms, as well as recent fined morphology and localization; eczema was history of medication, health care utilization, defined based on the Hanifin-Rajka criteria as The study was approved by the Ethics Com- Specific IgE at age 1½ and 4 was determined mittee for Copenhagen (KF 01-289/96) and The Danish Data Protection Agency (2008-41-1754).
Sweden) against the most common food (egg, and inhalant allergens (cat, dog, horse, birch,timothy grass, Dermatophagoides pteronyssinus,mugwort, molds). Values ‡0.35 kU/l were con- sidered indicative of sensitization and was ana- Recurrent wheeze. Respiratory symptoms were recorded daily by the parents in diaries from The proportion of the cross-sectional diagnosis (asthma by age 5 and sensitization by age 4) attributed to FLG mutations was calculated as 2282del4 was performed as previously described 100*(population prevalence-prevalence in wild- All analyses were made in sas version 9.1.
The association between FLG variants and end-points were analyzed by a two-level dominant genetic model combining the two SNPs R501X The clinical follow-up rate of the COPSAC cohort was 95% by age 1; 90% by age 2; 85% by age 3; 79% by age 4; and 76% by age 5.
Of 411 infants from the COPSAC cohort, 382 asthma-related events was explored graphically Caucasians were genotyped for the loss-of-func- by Kaplan–Meier plot and judged non-paramet- tion variants R501x, 2282del4 (1). The mutated rically by a log-rank test. Quantification was alleles R501x and 2282del4 were present in 18 performed in terms of hazard ratios in a Cox regression. P-values correspond to Wald testsand asymptotic 95% Wald confidence intervalswere calculated for the log(hazard ratio) and Age at onset of recurrent wheeze, asthma and Incidences of the event of one or more acute acute severe asthma exacerbations all exhibited severe asthma exacerbations per child were calcu- similar profiles (data on file), and diagnoses were lated in the age-spans (0–1, 1–2, 2–3, 3–4, 4–5) for closely correlated. We therefore analyzed the each FLG level. Age-adjusted incidence ratios for composite end-point asthma-related phenotype acute severe asthma exacerbations were analyzed in the survival analysis. Ninety-five of 382 by a log-linear model taking into account within children developed an asthma-related phenotype child correlation by an independence working (Fig. 1). The overall hazard ratio owing to FLG GEE approach. P-values correspond to robust variants was 1.82 [1.06–3.12], p = 0.03. The Wald tests and asymptotic 95% robust Wald Kaplan–Meier curve shows that differentiation confidence intervals were calculated for the in the development of an asthma-related pheno- log(incidence ratios) and back-transformed.
type was clearly present in the first 18 months of Point prevalence of eczema was calculated for life 0- to 1.5-yr hazard ratio 2.44 [1.17–4.71], ages (½, 1½, 2½, 3½ and 4½) for each FLG p = 0.02, where after it could no longer be level. Age-adjusted prevalence ratio for eczema recognized statistically; 1.5- to 5-yr hazard ratio was analyzed by a log-linear model taking into account within child correlation by an indepen-dence working GEE approach. P-values corre-spond to robust Wald tests and asymptotic 95%robust Wald confidence intervals were calculatedfor the log(prevalence ratio) and back-trans-formed.
The associations between FLG variants and the events current asthma at age 5, sensitizationat age 4, and sensitization at age 1.5 werequantified in terms of odds ratios by logisticregression. P-values correspond to likelihoodratio tests and asymptotic 95% Wald confidenceintervals were calculated for the log(odds ratios)and back-transformed.
For each outcome, effect modification through eczema was examined by including an interactionin the model between FLG status and eczemastatus prior to registration of that outcome. In Fig. 1. Asthma-related phenotype. Estimated cumulativerisk of developing asthma-related phenotype with and addition, analyses stratified on eczema status without filaggrin coding gene variants, log-rank test p-value = 0.03. Numbers at risk are given below the graph.
Filaggrin, early asthma and sensitization Fig. 3. Asthma; point-prevalence. The odds ratio of asthma Fig. 2. Acute severe asthma exacerbations. Yearly incidence by age 5 for mutated versus non-mutated was 2.62 of one or more acute severe asthma exacerbations requiring high-dose steroid intervention or hospitalization stratifiedon age and filaggrin coding gene (FLG) variants. Theoverall age-adjusted incidence ratio owing to FLG variantswas estimated to 2.40 [1.19–4.81], p-value = 0.01.
Yearly incidences of acute severe asthma exacer-bation are shown in Fig. 2. Incidences wereclearly elevated from infancy because of FLGvariants and this elevation persisted throughoutall 5 yr. The overall age-adjusted incidence ratiobecause of FLG variants was estimated to 2.40[1.19–4.81], p-value = 0.01.
Yearly point-prevalence of asthma is shown inFig. 3. Incidences were elevated because of FLG Fig. 4. Sensitization; point prevalence. By age 4, the odds variants, and this elevation increased throughout ratio of sensitization for mutated versus non-mutated was all 5 yr. The odds ratio of asthma by age 5 for 3.53 [1.72–7.25], p-value = 0.0007.
(1.12;6.11), p-value = 0.03. The proportion of The profile of the yearly point prevalence of current asthma attributed to FLG was 12.3%.
eczema suggested a similar profile after 6 monthsof age in children with and without FLGmutations (Fig. 5). The overall age-adjusted point-prevalence ratio owing to FLG variants Point-prevalence of specific IgE is shown in was estimated to 1.75 [1.29–2.37], p-value = Fig. 4 for age ½, 1½ and 4. The effect from FLG mutations occurs later than the asthma and There was no evidence of effect modification eczema phenotypes with no effect for the first through atopic dermatitis in any of the end- 2 yr, but by age 4, the odds ratio of sensitization points. Direct effects adjusted for eczema status for mutated versus non-mutated was 3.53 [1.72– were largely comparable in size to the total 7.25], p-value = 0.0007. The proportion of cur- effects. Stratified analyses based on eczema status rent specific allergy attributed to FLG was supported this conclusion except for sensitization where the effect size seemed larger in childrenwith eczema (Table 1).
FLG variants were strongly associated with thedevelopment of eczema manifesting fully in the FLG variants conferred clinically significant risk first year of life as previously reported (1, 16).
of investigator-diagnosed early asthma pheno- diseases. Also, cross-sectional studies relying on questionnaire-diagnosed eczema carry a risk of recall bias with parents of asthmatic childrenbeing more likely to recall early skin symptoms.
Accurate phenotyping is the major strength of our study and improves the power of the genetic association analyses. The meticulous prospective clinical monitoring, diagnoses and treatment of lung and skin symptoms through the first 5 yr of life was carried out solely by the investigators.
The cohort was seen regularly at 6-month inter- vals as well as for acute lung and skin symptoms controlled diagnosis and treatment according topredefined algorithms. The diagnostic accuracy Fig. 5. Eczema; yearly point prevalence. The overall age- from prospective clinical monitoring is the key adjusted prevalence ratio owing to filaggrin coding gene difference between this clinical study and tradi- variants was estimated to 1.75 [1.29–2.37], p-value = tional epidemiologic cohorts and is of particular importance in the clinical diagnosis of wheezewhere evaluation and perceptions of the terms types (recurrent wheeze, acute severe asthma are variable among practitioners and caregivers exacerbations and asthma) and sensitization in and eczema where inter-observer variation is a addition to eczema. The temporal pattern of FLG-associated atopic diseases was character- The statistical power of the study is limited by ized by early onset of asthma symptoms and the relatively low number of cases with FLG eczema, a persistent risk of acute severe asthma variants. However, we were able to demonstrate exacerbations, and later development of sensiti- statistically significant and biologically plausible zation. This suggests skin barrier dysfunction as effects. The external validity of the study is a novel and potentially modifiable mechanism limited by the high-risk nature of the cohort. A driving asthma symptoms in early childhood.
recent study reported no parental-of-origin effect This is the first longitudinal study describing of FLG, suggesting that the selection of high-risk the effect of FLG variants on the temporal mothers and not fathers only did not result in a clinical expressions of asthma, eczema and sen- specific maternal FLG effect (5). The carrier sitization assessed objectively by clinical examin- ations from birth. The onset and programming 11%, which is slightly higher than the reported of atopic disease occurs in early life and only longitudinal studies from birth can provide between 8.8% and 9.6% (1, 17, 18). Diagnosing insight into the temporal relationship and poten- asthma in early life is difficult with differences in tial causal interrelation of atopic manifestations diagnostic tradition between doctors and coun- of genetic variants. The novel contribution of our tries, so even though the risk of misclassification study is the prospective day-by-day monitoring was minimized by the standardized diagnostic of disease onset. This provides detailed insight into the age-dependent disease presentations.
Earlier studies relied on cross-sectional assess- have been reported (19). However, these new ments of disease rather than exact age of disease SNPs are substantially less prevalent and qual- debut and therefore could not appropriately itatively different with some residual function as demonstrated by protein immunoreactivity and a Table 1. Association of filaggrin gene variants with asthma outcomes and sensitization stratified by eczema status (eczema ever from 0 to 5 yr of age) Filaggrin, early asthma and sensitization significantly lower penetrance of eczema, and size but also because of the specific function and they were therefore not included in this study.
location of FLG in the skin pinpointing skin FLG loss-of-function variants were associated barrier dysfunction as an important pathogenetic with the increased risk of acute severe asthma step for at least a proportion of atopic disease.
exacerbation in the first years of life. This has not previously been reported but is in line with a keratohyalin granules within the stratum corne- previous report of FLG-associated increased risk um that provides a physical barrier, which of exacerbations in older children and adults reduces water loss and protects the body from potentially harmful environmental exposures, FLG variants increased the risk of recurrent such as allergens, toxic chemicals and infectious wheeze in infancy and investigator-diagnosed organisms (22). Two independent variants in the current asthma by age 5, contributing 10.8% of FLG gene (R510X and 2282del4), carried by the cases. Other studies have shown association approximately 9% of people of European origin, with asthma symptoms in pre-school-age (3, 5, result in complete loss of processed functional 21) and others showed association with asthma FLG in the epidermis (1). Homo- or heterozyg- later in life (2–4, 6, 8), while one study reported otes for these FLG variants alleles have varying no association with asthma (11). Recent meta- degrees of impaired skin barrier (23). This leads analyses conclude that FLG is a risk factor for to increased risk of eczema, which is inherited as asthma although not as strong as for eczema (9, a semi-dominant trait with high penetrance in FLG null homozygotes or compound heterozyg- FLG variants also increased the risk of sensi- otes and reduced penetrance in heterozygotes (1).
tization contributing 12.3% of the cases in line The association with eczema has been consis- with previous reports (3, 4, 8, 10). Interestingly, tently replicated (2–10), making FLG the stron- the risk of sensitization increased after onset of gest known genetic risk factor for eczema and asthma symptoms and eczema. This temporal having changed the paradigm of eczema patho- relationship has not previously been reported.
genesis from a focus on immunological deviation We did not find evidence of eczema mediating the FLG-associated risk of asthma or sensitiza- tion. Because of the longitudinal diagnosis of dysfunction to allergic sensitization may be eczema, we could study the effect of eczema increased skin permeability to allergens. Recent diagnosis prior to other atopic traits that has not animal studies support this mechanism by show- been performed previously. However, our study ing increased uptake of intact allergen through was not powered for such analyses, which can the skin in FLG-deficient mice and resulting explain why we find no statistical significant increased IgE sensitization and skin inflamma- interaction. Previous studies have only found tion (24). It is not clear to which extent increased risk of asthma in association with FLG-mediated sensitization requires skin inflam- mation as association with allergic rhinitis has We previously demonstrated that FLG-associ- also been shown independent of eczema (8).
ated risk of eczema is specific for the develop- ment of symptoms during the first year of life disease is not yet understood. FLG is expressed (16) and here extend the temporal description to in the skin (1) and in the outer layers of the oral asthma symptoms before 2 yr of age and later and nasal mucosa (25) but not in the respiratory sensitization. This temporal pattern suggests that epithelium of the nose (8, 25) or the lower gene-environmental research targeting FLG-defi- airways (25, 26). This suggests that FLG-associ- cient individuals should focus on infancy. In ated asthma is mediated through a systemic, agreement with this, we recently demonstrated possibly immunological, mechanism stimulated interaction between FLG variants and cat at through the impaired skin barrier. The late effect birth on development of eczema (16). Relevant on sensitization compared to the early increased outcomes other than eczema will be early asthma risk of asthma symptoms in this study suggests symptoms and later sensitization. The strong that sensitization does not mediate the pathway effect of FLG mutation resulting in a large from FLG deficiency to airway disease. This is proportion of children with risk variants getting supported by a report of interaction between disease makes it a useful predictor for future FLG variants and early sensitization on asthma mechanistic as well as interventional studies.
risk suggesting that these represent 2 distinct The association between FLG gene variants mechanisms in the pathogenesis of asthma (27).
and atopic diseases is intriguing, both because of One interpretation of our findings is that skin the consistent replication and considerable effect barrier defect may have independent effects on the different atopic traits (eczema, asthma and sensitization). This is in line with studies showing The authors thank the children and parents participating in that the eczema-asthma association is a relation- the COPSAC cohort as well as the COPSAC study team ship of co-existence, with asthma symptoms including Kirsten Mathiesen, Lena Vind Lotte Klansø. Theguarantor of the study is HB who has been responsible for often debuting together with or before eczema, the integrity of the work as a whole, from conception and rather than a progressive Ôatopic MarchÕ leading design to acquisition of data, analysis and interpretation of data and writing of the manuscript. KB drafted the manu- The observation of FLG-associated risk of script, analyzed and interpreted the data, contributed to early severe asthma exacerbation is novel and acquisition of data and provided important intellectual important for this difficult to control disease input. RT, CP and CBP contributed to the analyses andinterpretation of the data, revising the manuscript critically entity. Asthma exacerbations in young children and final approval of the version to be published. All represent a severe disease burden with major authors have seen and approved the final version of the impact on quality of life for patients and manuscript and none of the authors have any conflicts of socioeconomic costs for the health care system.
Asthma heterogeneity is pronounced in this age-group as evidenced by observations of different temporal patterns and association with interme- COPSAC is funded by private and public research funds.
diate phenotypes. Atopic diseases are likely to Grants above 100.000 Euro was donated by: The LundbeckFoundation; the Pharmacy Foundation of 1991; Augustinus represent several specific endotypes, subtypes of Foundation; the Danish Medical Research Council; The disease associated with distinct clinical features and defined functionally by the underlying The funders did not have any role in study design, data molecular mechanisms or by treatment response collection and analysis, decision to publish, or preparation (29). This heterogeneity is probably a main cause for inadequate management and there is anurgent need for a better understanding of disease mechanisms to improve treatment and preven- None of the authors report any conflict of interest relevant tion of disease. Genetic risk factors are causally related to disease and identification of these may therefore identify endotypes with different etiol-ogy and response to treatment. These FLG 1. Palmer CN, Irvine AD, Terron-Kwiatkowski A, findings suggest that a subgroup of children with et al. Common loss-of-function variants of the epider- pre-school asthma has disease initiated by skin mal barrier protein filaggrin are a major predisposingfactor for atopic dermatitis. Nat Genet 2006: 38: 441–6.
barrier dysfunction that maybe can be prevented 2. Brown SJ, Relton CL, Liao H, et al. Filaggrin null by targeting the skin defect. Chromosome 17q21 mutations and childhood atopic eczema: a population- (ORMDL3) gene variants were recently discov- based case-control study. J Allergy Clin Immunol 2008: ered as a risk factor for early asthma and we demonstrated that this genetic variation is asso- 3. Henderson J, Northstone K, Lee SP, et al. The bur- ciated with a totally different phenotype charac- den of disease associated with filaggrin mutations: apopulation-based, longitudinal birth cohort study.
terized also by early asthma exacerbations but J Allergy Clin Immunol 2008: 121: 872–7.
not with eczema or sensitization (30). It is likely 4. Marenholz I, Nickel R, Ruschendorf F, et al.
that such genetically defined phenotypes repre- Filaggrin loss-of-function mutations predispose to sent distinct endotypes with different pathoge- phenotypes involved in the atopic March. J Allergy Clin netic mechanisms, environmental triggers and treatment responses, and further understanding 5. Muller S, Marenholz I, Lee YA, et al. Association of Filaggrin loss-of-function-mutations with atopic der- of the underlying mechanisms may be a break- matitis and asthma in the Early Treatment of the Atopic through for the management of disease.
Child (ETAC) population. Pediatr Allergy Immunol In conclusion, this study describes a FLG- associated pattern of atopic diseases in early 6. Palmer CN, Ismail T, Lee SP, et al. Filaggrin null childhood characterized by early onset of ecze- mutations are associated with increased asthma severityin children and young adults. J Allergy Clin Immunol ma, early onset of asthma with severe exacerba- tions and later development of sensitization. Our 7. Weidinger S, Illig T, Baurecht H, et al. Loss-of- findings suggest a specific endotype of pre-school function variations within the filaggrin gene predispose asthma initiated by skin barrier dysfunction but for atopic dermatitis with allergic sensitizations. J not driven by sensitization. FLG provides a Allergy Clin Immunol 2006: 118: 214–9.
predictor useful for future targeted research into 8. Weidinger S, OÕSullivan M, Illig T, et al. Filaggrin mutations, atopic eczema, hay fever, and asthma in prevention of atopic diseases that should focus children. J Allergy Clin Immunol 2008: 121: 1203–9.
Filaggrin, early asthma and sensitization 9. Rodriguez E, Baurecht H, Herberich E, et al. Meta- thyosis vulgaris and atopic eczema. Nat Genet 2007: 39: analysis of filaggrin polymorphisms in eczema and asthma: robust risk factors in atopic disease. J Allergy 20. Basu K, Palmer CN, Lipworth BJ, et al. Filaggrin null mutations are associated with increased asthma exac- 10. van den Oord RA, Sheikh A. Filaggrin gene defects erbations in children and young adults. Allergy 2008: and risk of developing allergic sensitisation and allergic disorders: systematic review and meta-analysis. BMJ 21. Schuttelaar ML, Kerkhof M, Jonkman MF, et al.
Filaggrin mutations in the onset of eczema, sensitiza- 11. Rogers AJ, Celedon JC, Lasky-Su JA, Weiss ST, tion, asthma, hay fever and the interaction with cat Raby BA. Filaggrin mutations confer susceptibility to exposure. Allergy 2009: 64: 1758–65.
atopic dermatitis but not to asthma. J Allergy Clin 22. Candi E, Schmidt R, Melino G. The cornified enve- lope: a model of cell death in the skin. Nat Rev Mol Cell 12. Bisgaard H. The Copenhagen Prospective Study on Asthma in Childhood (COPSAC): design, rationale, 23. Smith FJ, Irvine AD, Terron-Kwiatkowski A, et al.
and baseline data from a longitudinal birth cohort Loss-of-function mutations in the gene encoding filag- study. Ann Allergy Asthma Immunol 2004: 93: 381–9.
grin cause ichthyosis vulgaris. Nat Genet 2006: 38: 337– 13. Bisgaard H, Hermansen MN, Buchvald F, et al.
Childhood asthma after bacterial colonization of the 24. Fallon PG, Sasaki T, Sandilands A, et al. A homo- airway in neonates. N Engl J Med 2007: 357: 1487–95.
zygous frameshift mutation in the mouse Flg gene 14. Halkjaer LB, Loland L, Buchvald FF, et al. Devel- facilitates enhanced percutaneous allergen priming. Nat opment of atopic dermatitis during the first 3 years of life: the Copenhagen prospective study on asthma in 25. Presland RB, Dale BA. Epithelial structural proteins childhood cohort study in high-risk children. Arch of the skin and oral cavity: function in health and dis- ease. Crit Rev Oral Biol Med 2000: 11: 383–408.
15. Paganelli R, Ansotegui IJ, Sastre J, et al. Specific 26. Ying S, Meng Q, Corrigan CJ, Lee TH. Lack of IgE antibodies in the diagnosis of atopic disease. Clin- filaggrin expression in the human bronchial mucosa.
ical evaluation of a new in vitro test system, UniCAP, in J Allergy Clin Immunol 2006: 118: 1386–8.
six European allergy clinics. Allergy 1998: 53: 763–8.
27. Marenholz I, Kerscher T, Bauerfeind A, et al. An 16. Bisgaard H, Simpson A, Palmer CN, et al. Gene- interaction between filaggrin mutations and early food environment interaction in the onset of eczema in in- sensitization improves the prediction of childhood fancy: filaggrin loss-of-function mutations enhanced by asthma. J Allergy Clin Immunol 2009: 123: 911–6.
neonatal cat exposure. PLoS Med 2008: 5: e131.
28. Illi S, von ME, Lau S, et al. The natural course of 17. Barker JN, Palmer CN, Zhao Y, et al. Null mutations atopic dermatitis from birth to age 7 years and the in the filaggrin gene (FLG) determine major suscepti- association with asthma. J Allergy Clin Immunol 2004: bility to early-onset atopic dermatitis that persists into adulthood. J Invest Dermatol 2007: 127: 564–7.
29. Anderson GP. Endotyping asthma: new insights into 18. Stemmler S, Parwez Q, Petrasch-Parwez E, Epplen key pathogenic mechanisms in a complex, heteroge- JT, Hoffjan S. Two common loss-of-function muta- neous disease. Lancet 2008: 372: 1107–19.
tions within the filaggrin gene predispose for early onset 30. Bisgaard H, Bonnelykke K, Sleiman PM, et al.
of atopic dermatitis. J Invest Dermatol 2007: 127: 722–4.
Chromosome 17q21 gene variants are associated with 19. Sandilands A, Terron-Kwiatkowski A, Hull PR, asthma and exacerbations but not atopy in early et al. Comprehensive analysis of the gene encoding childhood. Am J Respir Crit Care Med 2009: 179: 179– filaggrin uncovers prevalent and rare mutations in ich-

Source: http://www.dbac.dk/userfiles//sites/copsac.com/files/filaggrin.pdf