Clinical bioinformatics analysis of airways inflammation in asthma

Thesis

Asthma is a prevalent and heterogeneous respiratory disease that imposes a substantial global burden. This thesis investigated three key factors that contribute to airways inflammation in asthma: type 2 (T2) inflammatory phenotypes, bacterial infections, and sex differences, utilising large-scale clinical data and multi-omics analyses in patients with asthma.

In the RASP-UK cohort, transcriptomic profiling of bronchial biopsy and brush samples from 52 severe asthmatics and 20 healthy controls demonstrated that corticosteroid-resistant T2-high asthma showed upregulation of T2-dependent genes, adaptive immune response, impaired ciliary function, and epithelial development, whereas T2-low asthma showed upregulation of Th1- and IL-17-associated genes, neuroimmune and interferon-γ signalling pathways, and enrichment of neutrophils and mast cells. T2-intermediate asthma displayed a mixed molecular profile between T2-high and T2-low endotypes with activation of pathogen-defence pathways and enrichment of mast and NK cells.

A 27-year retrospective analysis of 4,350 sputum cultures from 1,106 patients with asthma revealed that 31.4% of routine samples were culture-positive, with Haemophilus influenzae, Pseudomonas aeruginosa, and Moraxella catarrhalis being the most frequently isolated pathogens. 45.8% of isolated bacteria were antibiotic-resistant, with multidrug resistance common in Staphylococcus aureus and Escherichia coli. The temporal microbial succession was observed within patients, where H. influenzae was replaced by P. aeruginosa. Factors independently associated with positive sputum cultures included older age, higher neutrophils, and use of fluticasone propionate, but not beclomethasone or budesonide.

Sex-specific analyses of 568 adults in the U-BIOPRED cohort revealed distinct clinical and molecular patterns in asthma. In mild/moderate asthma, clinical characteristics were largely comparable, although women exhibited a higher prevalence of non-type 2 phenotypes. In severe asthma, men exhibited more pronounced airflow limitation, reflected by lower FEV1 and greater residual volume, while women reported higher symptom burden and exacerbations, alongside enhanced airway remodelling, respiratory epithelial mitochondrial dysfunction, and heightened immune activation. Microbiome profiling indicated a higher prevalence of pathogenic bacteria (Haemophilus, Moraxella, Pseudomonas) in women with severe asthma. Circulating androgen levels were inversely associated with asthma severity in both sexes, suggesting a protective role.

In conclusion, this thesis characterised the roles and mechanisms of these three factors in airways inflammation with advanced clinical data analysis approaches, providing insights for informed personalised management in asthma.

Authors: Shen, J

• DOI: 10.5287/ora-r1yw4y6r7
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: inflammation; asthma; bioinformatics analysis