From genotype to phenotype in early childhood asthma

Asthma and recurrent wheeze in the first years of life represent a heterogeneous and poorly understood syndrome with a need to understand to what extent phenotypes reflect distinct underlying mechanisms.

We sought to investigate whether specific genetic asthma mechanisms, represented by genetic risk loci, were associated with specific disease courses and asthma phenotypes.

Known childhood asthma risk loci (GSDMB, CDHR3, FUT2, ABO, HLA-DQA1, IL33, IL1RL1, IL13, and TSLP) were analyzed in relation to redeemed prescriptions for asthma and allergy medication from birth to age 15 years in more than 23,000 children from the iPSYCH (Integrative Psychiatric Research) study. Gene variants were studied separately and as combined scores on the basis of putative similar mechanisms. Association with atopic and nonatopic asthma phenotypes was examined in more than 6000 children from the COPSAC (Copenhagen Prospective Study on Asthma in Childhood), BAMSE (Children, Allergy, Milieu, Stockholm, Epidemiology), and CHILD (Canadian Healthy Infant Longitudinal Development) birth cohorts.

GSDMB and CDHR3 were the strongest risk loci for asthma prescriptions in the first years of life, with effects continuing into school age, although with attenuating effect size. CDHR3 was characterized by associations present already in the first year of life and a strong interaction with GSDMB genotype. Suspected T_H2-related loci were characterized by a slightly later onset around age 2 to 3 years, with increasing or stable effect size till age 15 years and increased risk of allergic rhinitis. GSDMB and CDHR3 were associated with early transient disease, whereas most other loci were associated with both persistent and late-onset disease and with both atopic and nonatopic asthma.

Risk loci of early childhood asthma seem to involve different disease mechanisms as illustrated by their specific age-related effects. However, risk loci generally showed associations across classical age- and atopy-related phenotypes, suggesting that specific asthma mechanisms are not well captured by classical phenotyping.