Serum eosinophil-derived neurotoxin levels in machine learning-derived respiratory symptom clusters in adults

<bold>Background:</bold> Eosinophil-derived neurotoxin (EDN) has gained attention as a biomarker in asthma. Its relationship with respiratory symptoms, however, has not been explored. <bold>Aims and objectives</bold>: To describe serum EDN levels across respiratory symptom clusters derived using machine learning in a representative adult population. <bold>Methods:</bold> Within the ongoing West Sweden Asthma Study, unsupervised cluster analysis was performed in subjects with a serum EDN measurement (N=2,939) based on the presence/absence of 29 self-reported respiratory symptoms. Differences in serum EDN values between the respiratory clusters were compared using Mann–Whitney U test, stratified by sex. <bold>Results:</bold> Six respiratory symptom clusters were identified. EDN was slightly higher in clusters with allergic wheezing and coughing. It was consistently higher in men than in women across clusters (<bold>Figure 1</bold>). <bold>Conclusions:</bold> Respiratory symptom patterns correlate with EDN in adults, with men having consistenty higher values than women. Respiratory symptom patterns may reflect underlying pathophysiological pathways. <fig><object-id>erj;64/suppl_68/PA441/F1</object-id><object-id>F1</object-id><object-id>F1</object-id><graphic></graphic></fig> <bold>Figure 1. Respiratory symptom patterns and eosinophil-derived neurotoxin values in adults</bold>