Status: Funded - Closed
The early life metabolome in the origins of childhood asthma and allergy
Bo Chawes, M.D., Ph.D.
BACKGROUND: Childhood asthma and allergy presumably arise from gene-diet-microbe interactions during pregnancy and early infancy causing immune deregulation, chronic inflammation and subsequently symptomatic disease. The underlying biological pathways are however poorly understood.
GAP: We propose to investigate whether early life metabolic dysregulation is an intermediate step leading from inflammation to clinical disease penetrance.
HYPOTHESIS: Programming of childhood asthma and allergy disease is likely reflected in early metabolic changes leading to disease penetrance why metabolomics may provide early biomarkers of disease development and severity.
METHODS: Study subjects comprise children of (1) COPSAC2000 (N=411) and COPSAC2010 (N=700), two ongoing clinical birth cohorts followed closely at visits to our research unit for development of asthma and allergy; and (2) COPSAC registry (N=1500), a case-control register based cohort of children hospitalized for asthma and matched controls. Metabolic profiling will be performed in (1) COPSAC2000: dried blood spots age 1week, plasma age 6yrs, urine age 1month (2) COPSAC2010: dried blood spots age 1week, urine age 1month, exhaled breath from age 1week and onwards, and (3) COPSAC registry: dried blood spots age 1week.
RESULTS: We found that high blood levels of LDL-C were associated with concurrent asthma by age 7yrs (p=0.03) and poorer lung function (p=0.02). High levels of HDL-C levels were associated with improved lung function p=0.02), decreased bronchial responsiveness (p=0.05) and a decreased aeroallergen sensitization (p=0.006). High levels of triglycerides were associated with aeroallergen sensitization (p=0.02). The preliminary results from the urine metabolomics profiles obtained at age 1 month have a significant capacity for predicting asthma status at age 7 years with an AUC of approx. 0.65- 0.70, with 20 metabolites responsible for the separation. The breathomics profiles have been analyzed by PLS-lda and SVM algorithms using a setup train model on 70% and test on the remainder with repeated holdout scheme. The preliminary results show strong batch/breath collection effects, accounting for a predictive capability of clinical outcomes such as exacerbations of wheeze and asthma of approx. 60-70% correct.
IMPACT: The primary project finding is that profiles in exhaled breath and urine collected early in life long time before symptoms arise are predictive of asthma development. When the key
metabolites are identified, they may be utilized in the future for screening high-risk children.
Furthermore, we discovered that an unhealthy blood lipid profile characterizes children with
asthma and allergies, which suggests communality and shared risk factors with other
inflammatory complex disorders.