An integrated metagenomic approach to investigating disease heterogeneity in sepsis due to community-acquired pneumonia

Thesis

Sepsis is defined as life-threatening organ dysfunction caused by a dysregulated host response to infection. There is an increasing global burden of disease and it is associated with high mortality, long-term disability and shortened life expectancy. Clinical management of sepsis remains supportive rather than curative and progress in sepsis research has been severely constrained by a heterogeneous disease phenotype, limiting the interpretation of clinical trials and the development of effective therapeutic interventions. One source of heterogeneity is the pathogen but the frequent failure of clinical microbiology to identify the infecting organism in sepsis has limited efforts to understand the effect of disease heterogeneity involving the pathogen. Community-acquired pneumonia (CAP) is the most common cause of sepsis and clinical microbiology is unable to provide a diagnosis in 47% of such cases, suggesting that alternative methods such as clinical metagenomics may enable improved diagnostics.

Clinical metagenomics involves the application of next-generation sequencing technologies to characterise all the DNA and/or RNA present in a sample, enabling analysis of the entire microbiome as well as the human host genome or transcriptome from patient samples. This thesis presents the development and validation of Castanet, a method for targeted metagenomic sequencing using probe-based enrichment. Clinical metagenomic data is presented for 573 patients admitted to intensive care with sepsis due to CAP, including 447 patients for whom clinical microbiology did not identify a pathogen. In addition, digital droplet PCR data is presented for the most frequently identified causative bacteria (Streptococcus pneumoniae) and reactivated virus (Epstein-Barr virus) in the metagenomic cohort. Finally, this thesis explores how improved resolution of microbiology in the sepsis cohort can be applied to transcriptomic and genomic-based approaches to understand the host response in sepsis. This includes exploration of Epstein-Barr virus reactivation, differential gene expression analysis for different pathogens, and analysis of the association between specific HLA alleles and susceptibility to different pathogens.

This thesis demonstrates the utility of integrating metagenomic data with other omic methods to enable improved understanding of the heterogeneous host response in sepsis, with opportunities for a precision medicine approach.

Authors: Goh, C

• DOI: 10.5287/ora-vmbvvobve
• Type of award: DPhil
• Level of award: Doctoral
• Awarding institution: University of Oxford
• Language: English
• Keywords: Metagenomics; Sepsis; Community-acquired pneumonia
• Subjects: Functional genomics; Septicemia; Bacterial diseases; Metagenomics; Virus diseases; Medicine