S27 The relationship between neutrophilic inflammation and the airway microbiome using novel full length 16s rRNA sequencing in bronchiectasis

<h3>Introduction</h3> The relationship between neutrophilic inflammation and microbiome dysbiosis, two central features of bronchiectasis pathophysiology, is not fully understood. We used novel full length 16s rRNA sequencing, which provides detailed characterisation of bacterial communities at species level, to investigate the relationship between microbiome composition and markers of neutrophilic inflammation in bronchiectasis airways. <h3>Methods</h3> The neutrophil serine protease Proteinase-3 (PR3) and the neutrophil granule protein Olfactomedin-4 (OLFM4), were measured by ELISA in sputum supernatant from patients enrolled in the pan-European, multicentre EMBARC-BRIDGE study. Results were linked to sputum levels of neutrophil elastase (NE), a known bronchiectasis severity marker, and to relative abundance of taxa at species level in sputum measured by full length 16s rRNA sequencing. <h3>Results</h3> 143 patients were included. Mean age was 65 yrs (± 16) and 48% were female. Sputum PR3 and OLFM4 were strongly correlated (r_s=0.73, p&lt;0.001). Both were moderately correlated with NE (r_s=0.54 and r_s=0.4 respectively, p&lt;0001). Median PR3 and OLFM4 were significantly higher in patients with <i>Pseudomonas aeruginosa</i> infection (p=0.002 and p&lt;0.001), mucus plugging on chest computed tomography (p=0.006 and p=0.045) and hospitalisation for severe exacerbation (p=0.07 and p=0.05). Patients were grouped into high and low biomarker levels based around the median concentration. Table 1 shows bacterial species with highest relative abundance in samples grouped by high and low PR3, OLFM4 and NE. Linear discriminant analysis Effect Size (LEfSe) showed that the proteobacteria <i>Haemophilus influenzae</i> and <i>P. aeruginosa</i> were the most differentially abundant species defining patients with high PR3, OLFM4 and NE. In patients with low sputum PR3 and OLFM4, <i>Rothia mucilaginosa</i> was the most differentially abundant species; whereas unclassified <i>Streptococcus</i> differentiated those with low NE. Alpha diversity, measured using the Shannon index, was lower among patients with high PR3 (p&lt;0.001), OLFM4 (p&lt;0.001) and NE (p=0.034). Beta diversity analysed by PERMANOVA revealed distinct clusters defined by high and low PR3, OLFM4 and NE levels. High sputum PR3, OLFM4 and NE levels are associated with microbiome dysbiosis, characterised by predominance of proteobacteria and reduced alpha diversity. PR3 and OLFM4 were more strongly associated with dysbiosis than NE and warrant further investigation as biomarkers in bronchiectasis.