Plasma Metabolomes Identify Distinct Asthma Metabotypes: Findings From the COREA Cohort

Phenotype-based asthma classification has limitations that have motivated endotype-based approaches grounded in pathophysiology. However, existing biomarkers offer limited specificity and clinical accessibility. Plasma metabolomics, which reflects disease-specific metabolic states, offers a promising strategy for asthma classification. This study aimed to identify metabolic subgroups (metabotypes) of adult asthma using plasma metabolomics, with potential implications for personalized treatment.

Plasma samples from 407 patients with asthma in the Cohort for Reality and Evolution of Adult Asthma in Korea (COREA) were analyzed using Biocrates AbsoluteIDQ p400 HR kit with liquid chromatography-mass spectrometry. After preprocessing, 281 metabolites were natural-log-transformed, standardized, and partitioned by k-means clustering. The number of clusters was determined by a multi-criteria assessment combining cluster validity indices, consensus clustering, stability analysis, and cross-algorithm agreement.

Four metabotypes with distinct lipid-class signatures were identified. Group 1 (n = 123) was characterized by elevated ether-linked phosphatidylcholines, comprising younger patients with the earliest symptom onset and female predominance. Group 2 (n = 53) showed elevated lyso-phosphatidylcholines and altered amino acid metabolism (elevated glutamate, reduced glutamine), representing a metabolically intermediate, non-T2-high subgroup. Group 3 (n = 158) exhibited globally reduced sphingomyelins and broadly lower phosphatidylcholines in middle-aged, non-obese patients. Group 4 (n = 73) showed markedly elevated triacylglycerols (TG) and diacylglycerols (DG) with the highest body mass index (BMI), consistent with a non-T2, obesity-related metabotype. The TG/DG signature of Group 4 remained robust after adjustment for BMI, age, and sex.

Plasma metabolomics identifies clinically meaningful asthma metabotypes, supporting integration of metabolomic profiling into personalized asthma management.