Topological Data Analysis Reveals Antimicrobial Resistotypes Associated to the Microbiome in Bronchiectasis: An International Multi-centre Study

Introduction: While prior work has reported on microorganisms and resistance independently, no study to date has integrated these data for clinical value in bronchiectasis. Methods: Shotgun metagenomics to evaluate microbiome and resistance profiles was performed on the Cohort of Asian and Matched European Bronchiectasis 2 (CAMEB2) (n=251) recruited from four countries(Singapore, Malaysia, Italy, and the United Kingdom) and ‘matched’ by age, sex, exacerbation frequency and lung function. Spectral clustering was implemented to assess for resistotypes (RTs)and topological data analysis (TDA) with PCoA (as lens) applied to integrate microbiomes, RTs, and clinical outcomes. Results: Unsupervised analysis reveals two RTs: RT1, characterized bymacrolide and tetracycline resistance and RT2, abundant for multidrug efflux genes. Integrative analysis by TDA reveals several associations between specific bacteria, RTs, and clinical outcomes including (1) Haemophilus influenzae, RT1, moderate disease severity and non-exacerbator status and (2) Pseudomonas aeruginosa,RT2, and post-TB bronchiectasis aetiology.
Conclusion: Spectral clustering and integrative analysis using a TDA framework identifies non-linear complex associations between the microbiome, resistance profile and clinical outcomes within bronchiectasis with significant potential for improved patient risk stratification using next-generation sequencing. Funding: Singapore Ministry of Health's National Medical Research Council under its Clinician Scientist Award (MOH-000710) (S.H.C).