By Signe Kjeldgaard Jensen
Chair person: Professor Kim Gjerum Nielsen
Opponent: Professor Mika Mäkelä
Opponent: Doctor Marie Standl
ABSTRACT
In this thesis, different endotypes of asthma were investigated. Asthma is the most common chronic disease in childhood – it represents a great burden to the affected families and has huge expenses for the health care system. However, we have limited knowledge of specific endotypes of childhood asthma – a very heterogeneous disease – and asthma is still treated as one disease with limited possibilities for targeted and personalized treatment. Therefore, enlightening different endotypes of childhood asthma can help us unravel different mechanisms and identify useful predictive biomarkers of disease, which are in urgent need – especially during early childhood. Genetic markers and polygenic risk scores (PRS) could constitute a change in paradigm trying to reach this goal. This thesis focused on 1) the role of blood eosinophils – a well-established biomarker in later life atopic disease – throughout early childhood atopic disease, 2) genetic risk of asthma as a predictor of course of disease and disease subtypes, and 3) genetic risk of high body mass index (BMI) and early-life mechanisms of childhood asthma. We delved into this by investigating different cohorts. First, we leveraged the deep data layers within the two COpenhagen Prospective Studies on Athma in Childhood (COPSAC) motherchild cohorts. The COPSAC2000 cohort is a high-risk cohort consisting of 411 children born by mothers with asthma, and the COPSAC2010 cohort is an unselected, population-based cohort consisting of 700 children. The children were followed intensively from birth and throughout childhood with extensive examinations. All procedures and diagnoses adhered to strictly followed standard operational procedures, therefore securing homogeneity in the data. Secondly, we investigated the extensive register-based cohort established by The Integrative Psychiatric Research (iPSYCH) consortium combined with registries on prescriptions and hospitalizations.
In Paper I, we investigated the predictive role and early life associations of blood eosinophil count and atopic disease in preschool children in the COPSAC cohorts. Blood eosinophil count was investigated at age 18 months and 6 years in association to current atopic disease, later development of disease, and remission of existing disease. In contrast to age 6 years, blood eosinophil count was not associated with atopic disease at age 18 months. Furthermore, blood eosinophil count at age 18 months was not associated with later atopic disease, nor was it associated with persistence of disease among children with already existing disease. Genetic risk seems to contribute differently to childhood and adult-onset asthma and asthma heritability has been shown to be highest for early-onset asthma.
Therefore, in Paper II we performed a candidate-gene-study and investigated the strongest known individual childhood asthma-loci in relation to asthma subtypes and disease characteristics. Single nucleotide polymorphisms (SNPs) were investigated individually or combined in scores based on putative similar mechanisms in relation to redeemed prescriptions for asthma medication throughout childhood in the iPSYCH cohort. Replication and association to asthma-related traits were sought in the two COPSAC cohorts. We found GSDMB and CDHR3 to be the strongest risk loci in early childhood, already evident in the first year of life for CDHR3. They showed significant interaction and seemed to represent a non-atopic endotype. Contrary, T2 related risk loci showed a slightly later onset of disease and increased risk of allergy. However, T2 related risk loci were not associated with sensitization or atopic asthma, but instead with early life lower respiratory tract infections and exacerbations. Risk loci generally displayed associations across classic, temporally defined preschool wheezing phenotypes.
In Paper III, the known association between asthma and obesity was explored. This was done by investigating the association between genetic susceptibility to high BMI using a BMI PRS and asthma related traits throughout childhood. We found the BMI PRS to be associated with increased risk of asthma exacerbations and lower respiratory tract infections in early childhood. Interestingly, this association was independent of the current BMI of the children. The findings were replicated in the large-scale iPSYCH cohort, confirming the association between the BMI PRS and increased risk of hospitalizations for asthma and lower respiratory tract infections. To elaborate on these findings, we performed a transcriptome-wide association study (TWAS) to look for shared genes with effect on BMI and gene expression in the nasal airways in the COPSAC2010 cohort. Here, we were able to identify 29 genes with pleiotropic effects.
In conclusion, Paper I implied a limited role of blood eosinophil levels in preschool atopic disease and questioned the value of early life blood eosinophil count as a predictive biomarker in a clinical setting for atopic disease in later childhood. Paper II, on the other hand, showed better performances of genetic risk loci. The susceptibility genes for early childhood asthma appeared to engage distinct disease mechanisms, evident in their age-specific effects and associations with intermediate traits. This underscored the potential of genomic approaches in identifying endotypes of early childhood asthma for research and clinical purposes. In Paper III, we were able to increase the understanding of the obesity-related asthma phenotype by showing that children with a higher BMI PRS had increased risk of respiratory infections leading to asthma exacerbations. In contrast, these associations were independent of the child’s current BMI. The studies in this thesis have contributed to our understanding of childhood asthma in preschool years and helped enlighten different subtypes of this disease entity. A traditional biomarker such as blood eosinophils might not be applicable in young children, whereas genetic risk loci could constitute a more promising predictive biomarker in future diagnostics, treatment interventions, and strategies to unravel disease mechanisms in the pursuit towards more personalized medicine of early life childhood asthma.
