Clinical phenotypes of exacerbation are associated with differences in microbial composition and diversity in COPD

<b>Background:</b> Exacerbation phenotyping is needed to improve personalised treatment. We tested the relationship of clinical exacerbation phenotypes to the sputum microbiome. <b>Methods:</b> Induced sputum was collected from COPD patients when stable and at exacerbation. Exacerbations were classified as bacterial, eosinophilic or viral/paucigranulocytic. Microbiome analysis was performed by 16s rRNA sequencing. <b>Results:</b> 46 patients were included. Mean age 70.8 years (SD±7.0), mean FEV1% predicted 63.3 (SD±22). Exacerbations were classified (bacterial=33, eosinophilic=19, viral=19). There was no obvious clustering of stable and exacerbation samples by PCoA (figure 1A, p=0.08, PERMANOVA). Differentially identified taxa by LEfSe are shown in figure 1B. Classifying exacerbations by phenotype resulted in significant differences by PERMANOVA, p=0.003 (figure 1C). Differences in alpha-diversity were observed across the 4 groups (ANOVA, p=0.0005), figure 1D. Higher CAT scores indicate more severe symptoms during exacerbations (p=0.0067). <b>Conclusion:</b> Exacerbations can be defined by clinical phenotypes which demonstrate differences in microbiome composition and diversity.