Unbiased clustering of children with asthma or pre-school wheeze using the U-BIOPRED electronic nose platform

Rationale: Children with asthma or pre-school wheeze represent a heterogeneous group, characterized by a variety of underlying pathophysiological molecular mechanisms. Recent 9omics9 technologies provide composite molecular samples in inflammatory airway diseases [Wheelock ERJ 2013]. This includes breathomics, non-invasive metabolomics in exhaled air. Aim: To reveal phenotypes by unbiased cluster analysis based on metabolomic fingerprints from exhaled breath by electronic nose (eNose) in children with asthma or pre-school wheeze. Methods: This was a cross-sectional analysis from a subset of the paediatric U-BIOPRED cohort. Exhaled volatile organic compounds trapped on adsorption tubes were analyzed by an centralized eNose platform [Brinkman ERS 2012 A4307]. Ward clustering based on Similarity Profile Analysis [Clarke JEMBE 2008] was performed on eNose platform data, followed by ANOVA and chi-square tests. Results: 106 children were included (age (IQR)7.6 (4.0-13.0)yrs, 62% male, skin prick test (SPT) positive 54%). Five clusters were delineated that differed significantly regarding age, asthma control, asthma related quality of life, SPT. Conclusion: In this preliminary analysis, unbiased fingerprinting by eNose provides clinical clusters of children with asthma or pre-school wheeze. This suggests that metabolomics in exhaled air is suitable for phenotyping of airways disease in childhood.