Early Childhood Atopic Phenotypes and the Development of
Allergic Respiratory Disease
To the Editor,
Pediatric atopic dermatitis (AD) is a chronic, pruritic, inflammatory
skin disorder that affects up to 20% of children
worldwide1. Often the earliest sign of atopy, AD has
been recognized as the start of the “atopic march”, described as the
progression of AD to allergic respiratory diseases (ARD) including
asthma and allergic rhinitis2. Although these atopic
conditions often share a common T2 enriched pathway influenced by both
genetic and environmental factors, not all children with AD have
subsequent risk of ARD. Additionally, recent investigations dispute the
theory that the temporal progression of the atopic march occurs in a
sequential pattern3. Early AD can facilitate allergen
sensitization due to a dysfunctional skin barrier4.
Both aeroallergen and food sensitization has been associated with risk
of ARD5, but it is less clear whether AD may partially
mediate that risk. Additionally, food allergy has been recognized to be
part of the atopic march, however its role in the march to ARD is less
well identified. It is also unclear how these risks may appear in
ethnically diverse populations. Distinct atopic phenotypes may better
predict risk of ARD. Our objective was to identify whether associations
between early food sensitization, aeroallergen sensitization, or food
allergy (FA) and the subsequent risk of ARD by age 10 was modified by
the development of early AD by age 2 years.
We analyzed data from our racially and socioeconomically diverse birth
cohort, Wayne County Health, Environment, Allergy and Asthma
Longitudinal Study (WHEALS) that enrolled pregnant women 21–45 years of
age and their offspring. Recruitment period was from September
2003–December 2007. Eligibility and recruitment are described in
previous publications 6 and all study protocols were
approved by the Henry Ford Health System Institutional Review Board.
Offspring sensitization to aeroallergens (Alternaria, cat,
cockroach, dog, Dermatophagoides farina, ragweed, timothy grass),
milk, egg, or peanut was determined at 2 years of age by sIgE≥0.35 IU/mL
and skin prick testing (SPT; wheal size ≥3 mm larger than the saline
control defined a positive test). As sensitization does not always
translate to clinical allergy, we also formed an algorithm to determine
those most likely to have true IgE-mediated food
allergy7. Two allergists reviewed subjects with at
least one of the following criteria: (1) at least one food (milk, egg or
peanut) with sIgE ≥0.35 IU/mL; (2) a positive SPT; or (3) parental
report of infant symptoms potentially related to food allergy plus at
least one sIgE >0.10 IU/mL. To standardize classifying
infants to the presence of IgE-mediated food allergy (IgE-FA),
physicians were asked to combine professional experience with
investigator-developed protocols based on the Guidelines for the
Diagnosis and Management of Food Allergy in the United
States8. A third allergist independently reviewed and
ruled on discordant decisions. Data on asthma and AR by age 10 diagnosed
by the study physician was collected using clinical history, physical
exam, spirometry, and methacholine test.
Adjusted relative risk (aRR) was calculated using Poisson regression
with robust error variance and following adjustment for sex, child’s
race, parental history of asthma, parental history of AR, BMI z-score at
age 2, delivery mode, 1-month breastfeeding status, prenatal indoor dog
exposure, prenatal indoor cat exposure, and 1-month daycare status.
Of the 1258 mother-child pairs enrolled in WHEALS, 347 had sufficient
data for analyses (Supplemental Figure 1). Demographics are shown in
Table 1. The overall rate of early AD by age 2 years was 25.4% (88 out
of 347 subjects). Supplemental Table 1 shows the overall rates of asthma
and allergic rhinitis by age 10 by 2-year AD status. AD by age 2 years
significantly modified the association between FA at 3-5 years and the
risk of ever having asthma by age 10 (p=0.027) (Figure 1). In the
absence of AD, FA to milk, egg, or peanut was associated with an
increased risk of ever having asthma (aRR 3.36(1.71, 6.58),
p<0.001), while no difference was observed in the presence of
AD (aRR 1.24(0.57, 2.68), p=0.99). Food sensitization in the absence of
AD was associated with increased risk of ever asthma (aRR 2.04(1.03,
4.05), p=0.038), but was not associated with ever AR (aRR 1.10(0.81,
1.48), p=0.97). Food sensitization in the presence of AD was not
associated with ever asthma (aRR 0.89(0.45,1.78), p=0.99) or ever
allergic rhinitis (aRR 1.45(0.97-2.16), p=0.078).
In terms of aeroallergen sensitization, AD by age 2 did not
significantly modify the association between aeroallergen sensitization
at age 2 and the risk of ARD by age 10. This was true for aeroallergen
sensitization overall and when sub-analyzed by seasonal versus perennial
(Table 2, Figure 1). However, among those without AD by age 2, perennial
aeroallergen sensitization was associated with an increased risk of ever
having asthma by age 10 (aRR 2.15 (1.06, 4.36), p=0.031). This
association was not significant for those with AD by age 2 (aRR 1.68
(0.78, 3.63), p=0.26).
Our findings among a racially and socioeconomically diverse birth cohort
suggest that early AD modifies the relationship between FA and the risk
of ever having asthma by age 10. However, the association between FA and
increased risk of ever having asthma was only seen among those without
AD by age 2, which does not support the previously reported atopic
progression of disease as described by the atopic march. This held true
after correcting for several environmental and parental factors that may
increase risk of ARD in our cohort. Our findings may represent a
distinct atopic phenotype more characteristic among non-White subgroups,
as our cohort is 67% self-identified Black. Previous reports have
highlighted the differences in AD phenotypes among ethnically diverse
subgroups9. Additionally, recent reports highlight
atopic trajectories differ among White and Black children, with Black
children more likely to have asthma without FA, AR, or allergen
sensitization10. Due to sample size, we were unable to
assess the differing trajectories in whites versus blacks. However,
because our cohort is composed of 64.8% black children, we believe that
black race may be contributing to the outcomes of our study as previous
studies have reported atopic trajectories that are different in Black
children10. Future studies investigating these
endotypes that differ by ethnicity would be beneficial to identify
potential immunological markers that would guide therapies for
ethnically diverse populations and allow appropriate anticipatory
guidance.
Keywords : Atopic march, atopic dermatitis, food allergy, food
sensitization, aeroallergen sensitization, asthma, allergic rhinitis
Key message: Identifying early atopic phenotypes may help
identify later ARD risk. This study reiterates that the “march” is not
always a chronological process, but rather a complex relationship
between heterogenous allergic phenotypes.