Online breath analysis with SUPER SESI - HRMS for metabolic signatures in children with allergic asthma
Introduction:
Improving the diagnosis and management of pediatric asthma is necessary. Breath analysis can help by assessing altered metabolism and disease-associated processes in a non-invasive manner. This study aimed to identify exhaled metabolic signatures using SUPER SESI - HRMS that can distinguish children with allergic asthma from healthy controls.
Results:
The study included 48 people with allergic asthma and 56 healthy individuals. A total of 375 significant mass-to-charge features were identified, with 134 being putatively identified as metabolites. Many of these metabolites were found to be part of common pathways or chemical families. The study found that several pathways were well-represented by significant metabolites, such as elevated lysine degradation and two downregulated arginine pathways in the asthmatic group. The study also used supervised machine learning to assess the ability of breath profiles to classify samples as asthmatic or healthy in a repeatable manner, with the results showing an area under the receiver operating characteristic curve of 0.83.
Figure 1:
Statistical analysis of m/z features in breath profiles. (A) Volcano plot representing all detected 2,315 m/z features. Dashed line: Benjamini-Hochberg adjusted p-value of 0.05. (B) First two principal components (PCs) score plot of the 134 putatively identified m/z features. Blue dots represent healthy probands and red dots asthmatic patients. 95% data ellipses were added per group for visual depiction.
Discussion:
This study showcases the first online breath analysis conducted using SUPER SESI - HRMS on a pediatric population with allergic asthma, revealing distinct breath patterns that correlate with the condition. The analysis identified a significant number of discriminative features (m/z values) that could be associated with specific metabolic pathways or chemical families. Key findings include elevated pathways related to the metabolism of lysine, tyrosine, and various fatty acids, all of which have been previously linked to pediatric asthma in different contexts. Lysine metabolism, particularly, was highlighted for its significant elevation in asthma patients, with metabolites such as succinate and glutarate being directly identified and associated with the condition. Tyrosine metabolism also showed significant upregulation, with both human and microbiotic origins suggested for the metabolites identified. The study further found an increased presence of certain fatty acids, which supports the hypothesis of upregulated ω-oxidation in allergic asthma.
Contrastingly, the study identified downregulation in the metabolism of arginine, proline, and linoleic acid, with certain metabolites like palmitoylethanolamide (PEA) showing decreased levels, aligning with their known anti-inflammatory roles. Interestingly, despite the detection of oxidative stress markers like aldehydes being inconsistent across studies, this study did not find an elevated presence in the asthmatic group, suggesting potential methodological or environmental factors influencing these outcomes.
Machine learning analysis of the metabolic profiles yielded results for diagnostic applications, with an AUC of 0.83 indicating a strong potential for using these breath signatures in predictive models.
Conclusion:
For the first time, a significant number of metabolites present in the breath of children with allergic asthma were identified using online breath analysis. These metabolites were found to be different from those present in the breath of healthy children. Most of these metabolites are known to be associated with metabolic pathways and chemical families that play a role in the pathophysiological mechanisms of asthma. Furthermore, a small group of these volatile organic compounds exhibited high potential for future clinical diagnostic applications.
