By Ann-Marie Malby Schoos
Chariman: Anders Juul
Opponent: John Warner
Opponent: Kirsten Ohm Kyvik
Diagnosing IgE-mediated inhalant- and food allergy is challenging. Many physicians rely on a thorough interview on clinical symptoms, but allergy symptoms can arise from many different organ systems and vary in severity, and symptoms can be related to many other diseases (e.g. enlarged adenoids, common cold, chronic obstipation, diarrhea etc.). In the youngest children, the diagnosis is further complicated by the reliance on observation of symptoms by the parents. Therefore, objective measures of allergy including skin prick testing (SPT), measurement of serum specific IgE (sIgE) level, and the most recent component resolved diagnosis (CRD) are important parts of the allergy diagnosis, particularly in young children. Furthermore, asthma and eczema are linked to sensitization in childhood by referral to these diseases as “atopic diseases”. The atopic asthma and the atopic eczema are considered unique endotypes different from their non-atopic counterparts. The evidence in support of the atopic endotype is uncertain in particular since “atopic” is used irrespective of class of sensitization (food and/or inhalant); test-system (though SPT and sIgE have low agreement); specific or total IgE.
The aims of this PhD thesis were to investigate the different ways to measure allergy in children, and to challenge the term “atopic” disease with suggestions of new ways to interpret sensitization.
In study I we described the agreement between SPT and sIgE in children aged 0 to 6 years from the Copenhagen Prospective Studies on Asthma in Childhood (COPSAC2000) cohort. Furthermore, we examined how the test results predicted reported symptoms of allergic rhinitis and food allergy. We measured sensitization to 8 inhalant and 6 food allergens at ½, 1½, 4, and 6 years of age and found a poor to moderate agreement between SPT and sIgE, which was particularly poor for food sensitization and worsened with age mainly due to children with elevated sIgE levels and negative SPT results. Both tests had a low predictive value for clinical food allergy and symptoms of allergic rhinitis. These findings are important for both clinicians and researchers as choice of assessment method, age of the child, and type of investigated allergens all have an impact on test results. The lack of agreement between tests and a poor correlation to clinical disease underscore that allergy testing should only be done in children with meaningful symptoms and not as a screening tool, but it also stresses the need for new ways to analyze sensitization.
In study II we performed a longitudinal analysis of the possible associations between allergic sensitization, asthma, and eczema during childhood (0 to 13 years) in the COPSAC2000 cohort. We found that asthma and allergic sensitization were not associated traits in preschool children (0 to 6 years), but found a clear and consistent relationship between eczema and sensitization. Contrary, at 13 years of age we found a significant association between asthma and sensitization, but none between eczema and sensitization. This suggests a very limited role of allergy in the etiology of preschool asthma and questions the relevance of the term “atopic asthma” in this age group, since it is unlikely to represent a true endotype.
In study III we wanted to explore latent patterns of sensitization during the first 6 years of life in the COPSAC2000 cohort using an unsupervised multidimensional clustering technique (PARAFAC). Subsequently, these patterns were investigated in relation to asthma, rhinitis, and eczema. We found 7 latent age- and allergen-specific patterns of sensitization whereof only one was associated to asthma (pattern 1: dog/cat/horse). All seven patterns were replicated in the independent, unselected BAMSE (Children Allergy Milieu Stockholm Epidemiology [in Swedish]) birth cohort (R2>0.89). Our study suggests the presence of specific sensitization patterns in early childhood differentially associated to development of clinical outcomes supporting that they represent biological meaningful sensitization phenotypes.
In study IV we studied allergic sensitization to different protein-groups diagnosed from component resolved diagnostics in high-risk children from the Mount Sinai Pediatric Allergy Clinic in New York, USA. We aimed to examine how sensitization to these protein-groups was associated with asthma and allergic rhinitis. Fifteen protein-groups were extracted from the results, and we found that sensitization to components in the lipocalin protein-group was the only group associated with asthma. Sensitization to the PR-10, pectate lyase and grass pollen protein-groups was associated to allergic rhinitis. This study suggests a novel way of analyzing and interpreting allergen sensitization based on protein-groups.
In conclusion, we have found that diagnosing allergy and the so-called “atopic diseases” based on a single test such as SPT or sIgE at a single time-point gives very little meaning. Sensitization is something that develops during childhood and different sensitization phenotypes exist that are defined by type and number of involved allergens, age at debut, and temporal development. These phenotypes are differentially associated to clinical disease and can be explained by sensitization to common protein-groups that are present across different allergens.