Longitudinal Sensitization Patterns in Childhood and Adolescence

This is the authors' abstract. We don't add key points for this paper.

Allergen-specific immunoglobulin E (sIgE) production typically starts in early childhood and develops over time in a complex manner that is not fully understood [1]. Some allergens contribute to early life sensitization, for example, egg and milk [2], while others increase through adolescence, for example, aeroallergens [2]. Concurrent sensitization to multiple allergens and changing sensitization patterns over time are common. Understanding the temporal development of sIgE can lead to early identification of individuals at risk of developing allergic diseases, allowing for timely intervention to prevent the progression of allergies and related conditions such as asthma and atopic dermatitis (AD). Several studies have performed detailed longitudinal sIgE measurements in childhood [3] or performed patient clustering of atopic traits and diseases [4, 5]. However, there is a lack of studies comprehensively describing the temporal sensitization patterns with unified statistical models. We previously identified and replicated longitudinal sensitization patterns in the Copenhagen Prospective Studies on Asthma in Childhood 2000 birth cohort (COPSAC2000) of 411 children born to mothers with asthma [6]. An unsupervised cluster analysis revealed seven age- and allergen-specific patterns of sensitization from age 0–6 years, differentially associated with asthma, rhinitis, and AD. All seven patterns were verified in the independent BAMSE cohort with high reproducibility (R2 > 0.89). We have now gathered sensitization and atopic disease outcome data in the COPSAC2000 cohort at 13 and 18 years of age. Here, we extend these early childhood patterns to explore their development into adolescence and assess their associations with asthma, rhinitis, and AD at age 18 years. In COPSAC2000, blood samples were collected at ages ½, 1½, 4, 6, 13, and 18 years for sIgE measurements against eight aeroallergens and five food allergens. Sensitization was defined as sIgE ≥ 0.35 kUA/L. For the definitions of rhinitis, asthma, and AD, please see the Online Repository. Of the 411 children enrolled, 405 (98.5%) had at least one valid sIgE measurement. Of these, 240 (59.3%) were sensitized to at least one allergen; 132 (32.6%) to food allergens, 209 (51.6%) to aeroallergens, and 101 (24.9%) to both. Individual sensitization rates at each time point are shown in Figure S1. Sensitization rates generally increased through childhood and adolescence for aeroallergens, while food allergens peaked earlier. Most new cases of aeroallergen sensitization appeared at 13 years, where food sensitization was in remission at both 13 and 18 years (Figure S2). Using the Non-Negative Sparse Parallel Factor (NNS-PARAFAC) model, we analyzed the COPSAC2000 sensitization data (n = 405) as a three-way data array (children, age, allergens), which explained 60% of the total variation. Figure 1 shows the prevalence and development at age 0–18 years of the seven patterns: (1) dog/cat/horse; (2) timothy grass/birch; (3) molds; (4) house dust mite; (5) peanut/wheat flour/mugwort; (6) peanut/soybean; and (7) egg/milk/wheat flour. All four aeroallergen-driven patterns (1–4) and the mixed aero−/food allergen pattern (5) increased up to 13 years, with minor changes thereafter. Food sensitization to peanut/soybean (pattern 6) was stable and independent of age, peaking at 13 years, while egg/milk/wheat flour (pattern 7) increased early then decreased by 18 years. We observed a positive direction of association between all sensitization patterns and asthma, rhinitis, and AD at age 18 years, except for pattern 7 (egg/milk/wheat flour) and AD (Figure 2), although many were not significant. Adjusting for potential confounders did not alter the results noteworthily. The strongest associations were for pattern 1 (dog/cat/horse) with all three outcomes (ORs 3.75–4.80, p-values < 0.004). For asthma, patterns 1, 2, 3, and 5 were significant, with pattern 4 (house dust mite) borderline significant. For rhinitis, all four aeroallergen patterns and the mixed pattern were significant, while food allergen patterns 6–7 were not. For AD, only pattern 1 was positively significant, and pattern 7 showed an inverse association. These findings highlight the dynamic nature of childhood sensitization and its impact on atopic diseases over time. Further discussion is available in the Online Repository. We expanded seven replicated sensitization patterns from age 0–6 years to age 18 in a prospective birth cohort. These patterns develop differently through adolescence, with stronger associations with atopic disease when including data from later time points. By understanding these distinct sensitization patterns and their associations with long-term health outcomes, clinicians can make more informed decisions about the management and prevention of allergic diseases in children through lifestyle interventions, allergen avoidance, or medical management such as targeted allergen immunotherapy. A.-M.M.S. has written the first draft of the manuscript. M.A.R. and J.T. performed the statistical analyses. All co-authors have provided important intellectual input and contributed considerably to the analyses and interpretation of the data. The corresponding author had full access to the data and had final responsibility for the decision to submit for publication. J.T. reports a speaking fee from AstraZeneca. A.-M.