Background. Severe asthma (SA) is a heterogeneous pathology with diverse clinical phenotypes resulting from various endotypes (i.e. pathophysiological mechanisms) that still needed to be fully described locally to propose the most suitable therapeutic strategies. Thanks to a cohort of children with SA (n=20) and age-matched disease-control children (n=10), our group has recently evidenced a local immune signature in bronchoalveolar lavages (BALs), which allows discriminating SA from controls and, within SA children, those with frequent exacerbations from those without (Adel-Patient et al. 2021a & 2021b). The objectives of the present study was to pursue SA endotypes characterisation by analysing BAL through non-targeted metabolomics approach. Methods. BALs were first freeze-dried and normalised based on protein concentration. Then metabolomics analysis was preformed, using liquid chromatography (C18 column) coupled with high-resolution mass spectrometry (HRMS), and following an established workflow for sample preparation (Boudah et al., 2014), data acquisition and treatment (https://workflow4metabolomics.org; MetaboHub). Annotation of metabolite features was performed using an internal spectral database (mlomic R script) and confirmed by MS/MS (highest level of confidence). Non-supervised analysed (Principal Component Analysis, PCA ; Hierarchical Ascendant Clusterisation, HAC) were performed thanks to R package (ropls, complexHeatmap). Metabolite Set Enrichment Analysis (MSEA) was performed using MetaboAnalyst 5.0 and SMPDB as a metabolite set library, and statistical significance assessed using the Mann-Whitney test. Results. 94 metabolites were annotated in BAL, involved in various biological pathways. Unsupervised analyses (PCA and HAC) did not differentiate SA children from controls. However, MSEA showed that biotin, carnitine synthesis, lysine degradation, methionine metabolism and spermidine and spermine biosynthesis pathways were the most significantly enriched metabolism pathways in SA. In line with observations from others in mice, we further observed that levels of spermine (p = 0.0111) and spermidine (p= 0.0276) were significantly increased in BALs from children with SA compared to NA, thus confirming that this pathway may be relevant in the pathophysiology of asthma. Conclusion. Our study is the first one describing the BAL metabolome in children with SA. It provides new exploratory avenues, as metabolomics pathways involved/perturbed by SA. This work is funded by ANR SevAsthma (ANR-18-CE14-0011-03) References. Adel-Patient, K. et al., 2021. 1. A Comprehensive Analysis of Immune Constituents in Blood and Bronchoalveolar Lavage Allows Identification of an Immune Signature of Severe Asthma in Children. Front. Immunol., 12, 700521, doi:10.3389/fimmu.2021.700521. Adel‐Patient, K. et al, 2021. Immune Signatures Distinguish Frequent from Non‐frequent Exacerbators among Children with Severe Asthma. Allergy, 76, 2261–2264, doi:10.1111/all.14759. Boudah, S. et al., 2014. Annotation of the Human Serum Metabolome by Coupling Three Liquid Chromatography Methods to High-Resolution Mass Spectrometry. J. Chromatogr. B, 966, 34–47, doi:10.1016/j.jchromb.2014.04.025.