| Summary: | Antimicrobial resistance (AMR) represents a significant challenge to global health comprising substantial complexity driving its emergence and spread. Thus, surveillance efforts are essential to monitor trends, identify emergence and develop interventions, and of particular importance in curtailing rapid global dissemination facilitated by plasmids and other mobile genetic elements. Accordingly, the development and deployment of novel surveillance approaches is prioritised in the World Health Organisation’s global AMR action plan. Wastewater-based epidemiology (WBE) is of increasing interest as a convenient surveillance approach, leveraging the pooling of human excreta to generate information on human populations at scale. Whilst WBE has found success in the surveillance of polioviruses and SARS-CoV-2, its application to monitoring population-level AMR prevalence remains limited to date. Part of the challenge lies in the paucity of clearly reported studies and validation of methodology which hinders effective iterative development.
In this thesis, I employed metagenomic sequencing methods to investigate several fundamental questions of methodology and study design regarding the application of WBE to population-level AMR surveillance. By initially conducting a comprehensive synthesis of shortcomings and knowledge gaps in the existing literature, I tailored my experimental chapters to specifically address gaps in knowledge whilst employing best practices gleaned from the review. I benchmarked the performance of multiple sequencing and bioinformatic approaches in the context of WBE AMR surveillance. I also investigated the impact of initial wastewater sampling methods on profiling taxonomic and resistome composition, with parallel characterisation of short-term temporal fluctuation and associated drivers. Lastly, I conducted a multi-site longitudinal survey to explore factors associated with wastewater composition such as antibiotic prescribing and residue concentrations.
A common recurring theme throughout this thesis is the importance of validating individual methods to systematically identify sources of variation when developing workflows for WBE. The work I have conducted highlights the complexity involved in the analysis and deconvolution of wastewater for AMR surveillance, and demonstrates the need for clearer reporting of high-quality metadata to facilitate Iterative development of optimal practice. Additionally, interdisciplinary collaboration is vital for this field where insight from colleagues in areas of chemical and viral surveillance as well as the often- overlooked expertise of individuals working outside of research but within the wastewater industry is invaluable. Future work underscored by collaboration and the sharing of granular metadata will significantly accelerate the development of wastewater-based epidemiology into a robust surveillance tool for the mitigation of antimicrobial resistance.
|
|---|