Characterisation of the gut-lung axis microbiome in clinically stable patients with chronic obstructive pulmonary disease.

Background Airway and gut dysbiosis have been reported in Chronic Obstructive Pulmonary Disease (COPD); however, their relationship and association with clinical features remain poorly understood. We aimed to characterise the lung and gut microbiome in patients with stable COPD and controls. Methods Prospective, multicentre, longitudinal and controlled study of n = 60 stable patients with COPD and n = 30 controls. In them, we analysed 16S rRNA-seq in oropharyngeal (OP) swabs, sputum, bronchoalveolar lavage fluid (BALF) and stool. Weighted gene co-expression network analysis (WGCNA) was employed in each sample type to identify modules of co-abundant bacteria associated with clinical traits. Findings We found that the microbiome in airway and stool samples was highly dissimilar both in patients and controls, with 0.37% of this diversity associated to COPD. The microbiome taxa associated with COPD in OP swabs and sputum were highly similar, but different from BALF, suggesting that OP swabs can be a surrogate sample of sputum. Finally, using WGCNA, we identified: (a) 5 modules in OP swabs and 3 in sputum associated with FEV1, but some of them were also associated with exacerbations, dyspnoea and inhaled steroid (ICS) use; (b) In BALF 4 modules associated with FEV1 and dyspnoea, and 2 modules with ICS; and, finally, (c) in stool, 1 module related to FEV1, 1 to exacerbations and 3 with ICS. Interpretation The gut and lung microbiomes in patients with COPD are distinct, but both clinically relevant as both present bacterial associations with airflow limitation, exacerbation history, and ICS use.