Coordinated surveillance of foodborne pathogens and antimicrobial resistance: insights from the PATH-SAFE pilot

The Pathogen Surveillance in Agriculture, Food and Environment (PATH-SAFE) programme is a UK-based, cross-government program of work, with the overall aim of piloting a better national surveillance system for monitoring and tracking foodborne pathogens (FBP) and antimicrobial resistance (AMR) in the agri-food system. Such a broad aim has required the collaboration of not just the core, government partners who initiated the program, but also a wide range of partners from across academia, industry and further government agencies. The program itself has taken the form of multiple, aligned, thematically linked projects, conceptualized and delivered by the participant organizations, and coordinated by a central Program Team based within the Food Standards Agency (FSA). Collaboration among the partners has been crucial in order to deliver an ambitious program of work, that addresses the policy aims and requirements of multiple partners. The cross-government approach was further embedded by the Treasury’s Shared Outcome Fund requirement for departments to work collaboratively to test innovative ways of working. The program will draw to an end in March 2025 and this paper summarizes the program's achievements to date and considers how they can be taken forward beyond the lifetime of the program.

1. New tools for biosurveillance

Many of the projects within the program were aimed at developing a suite of tools to help end users across government better understand food safety and AMR risks. These range from molecular detection tools to analytical or modelling tools for characterizing or assessing risk. The molecular diagnostics include development of LAMP assays for foodborne viruses and bacteria (UK Health Security Agency (UKHSA) and 20/30 Labs), a duplex RT-qPCR for quantifying norovirus [1] and a long amplicon approach for simultaneously typing two genes in mixed populations of norovirus genogroups I and II [2], all of which can be applied to pathogen surveillance. As well as developing diagnostics to achieve or enable PATH-SAFE specific surveillance approaches, the program also undertook a broader consideration of the application of onsite diagnostics (OSD) for the rapid detection of FBP in a range of different sample types, outside of a conventional laboratory setting. This work, delivered by Fera Science Ltd, undertook a landscape review of OSD, applied a newly-adapted technology readiness level assessment structure, then piloted two OSD options with public- and private-sector end users. This work is ongoing and aims to develop a framework to map out the steps that would be required to deploy OSD for official food safety testing.

A key element of effective biosurveillance is the ability to share and analyze data in new ways, while building on existing protocols, pipelines and ways of working. In the PATH-SAFE context, the Environmental Surveillance System (developed by Deloitte, UKHSA, Veterinary Medicines Directorate (VMD) and the Environment Agency (EA) permits the sharing and visualization of a range of AMR related data from environmental sample types such as river water samples, including minimum inhibitory concentration and genetic data from bacterial isolates, and antibiotic residue data. In parallel, work undertaken by Food Standards Scotland (FSS) developed machine learning source attribution models for predicting the source or host of origin for Escherichia coli isolates. In addition to these tools, one of the central outputs from the program has been the creation of the PATH-SAFE Genomic Data Platform. This is a user-friendly platform for the analysis and visualization of bacterial genomes and associated metadata. Initially focused on Salmonella enterica subspecies enterica, platform capabilities are now being expanded to accommodate E. coli and Listeria monocytogenes, thus covering three of the FBP most frequently implicated in UK foodborne outbreaks. The platform is designed to enable cross-agency collaboration for investigations, integrate with the existing workflows of the participating government departments, and open up the interrogation of bacterial genomic datasets to non-bioinformaticians; therefore, complementing, but not replacing, existing bioinformatic expertise and accredited pipelines. Delivered by Digital Epidemiology Services (www.digitalepi.group), the platform has been developed using a consortium approach, bringing together longstanding, world class expertise and infrastructure in pathogen genomics; Pathogenwatch (https://pathogen.watch) developed by the Centre for Genomic Pathogen Surveillance (CGPS), Enterobase (https://enterobase.warwick.ac.uk/) [3], PubMLST (https://pubmlst.org/) [4] and Climb Big Data (https://www.climb.ac.uk/) [5]. Development has also been guided by End User Groups including representatives from participating departments, and by four Community Input Advisory Groups around; Technical (analytics); AMR; Foodborne Disease Data Standards; and International alignment. This type of collaborative working has been at the heart of the entire program and offers a robust paradigm for future surveillance efforts to employ.

2. Expanding our understanding of FBP & AMR in the agri-environment

The collaborative and sharing approach that underpins PATH-SAFE and has strengthened throughout the duration of the program is further exemplified by the collection of projects that have jointly generated and analyzed FBP genomic data. These projects have collectively produced baseline data on the genomic diversity of a range of FBPs across the four nations of the UK, describing what “normal” looks like. They have concurrently provided pilot data to populate the Genomic Data Platform, and addressing knowledge gaps around AMR in specific scenarios.

Some of this baseline data was generated through the sequencing of archived isolates, which can provide important historical context against which future outbreak sequences can be compared [6], in a cost-effective way [7]. This allowed background genomic data to be generated from areas where there has been a historic dearth of data. For example, in Northern Ireland, collaborations between a Belfast NHS Trust public health microbiology laboratory, the Centre for Environment, Fisheries and Aquaculture Science (Cefas) and UKHSA enabled isolates of Listeria monocytogenes from food to be sequenced and compared with sequences from Great Britain. Meanwhile the Agri-Food and Biosciences Institute (AFBI) and FSA worked together to select and sequence isolates of Salmonella and L. monocytogenes from food and animal sources, with advice on Listeria analysis provided by the Quadram Institute. Other work at the University of Oxford on Campylobacter has involved sequencing thousands of isolates from a diverse range of collaborators, in order to help elucidate transmission dynamics through the agri-food chain, and the rise in AMR.

Furthermore, a suite of projects to address knowledge gaps around AMR were developed by the Animal and Plant Health Agency (APHA) and VMD, and then delivered in collaboration with other partners including the Agricultural Industries Confederation, the Department for the Environment, Food and Rural Affairs (Defra), FSA, National Milk Records and Scotland’s Rural College. These projects focused on furthering our understanding of AMR presence, diversity, and transmission using genomics both in a project already monitoring AMR through phenotypic methods [8] and in understudied compartments of the agri-food chain, particularly ruminants (via bulk milk and cattle and sheep at slaughter) and animal feed, by collecting novel samples, isolating relevant bacteria, identifying phenotypic AMR and performing Whole genome sequencing (WGS). The WGS data from these projects is now undergoing further analysis and will be available for future comparisons, meaning that beyond filling individual knowledge gaps the program has also created an opportunity to generate insights into transmission across agri-food chains. The results of this work will be published in the near future, making them available to wider stakeholders and helping guide future AMR surveillance and intervention efforts.

3. The face of surveillance in the future?

Many of the projects above have generated genomic data from archives or new surveys using long established methodologies, albeit in new sectors. As we think about what a future national surveillance system might look like, the Program has also undertaken projects to trial new ways of surveying for FBP and AMR. The EA developed and commissioned several scoping reviews or pilot studies to inform the surveillance options for AMR in the environment [9], including river water and bathing waters [10], wild flora and fauna [11], shellfish [12] and bioaerosols [13,14]. Work on the use of bivalve shellfish for monitoring AMR and FBP will be taken forward in the final year of the Program, building on methods already developed by Cefas, EA and FSS within PATH-SAFE. Air monitoring for AMR and FBP was also successfully trialled by Queen’s University Belfast (QUB), in their project piloting air and near-source wastewater monitoring in a care home setting in Northern Ireland.

Of all the matrices trialled within PATH-SAFE, wastewater has been the most widely investigated, building on the success of the SARS-CoV-2 wastewater monitoring program [15,16]. Cefas undertook a pilot study in the Taw-Torridge river catchments in Devon, to compare wastewater, river water and shellfish monitoring for of a range of FBP, and examined findings against a complementary desk study by the Ribble Rivers Trust, assessing likely sources of FBP in the River Ribble catchment in Lancashire. Cefas also commissioned a desk study from Virosafety Limited on the potential risk of norovirus transmission to fresh produce via irrigation water; worked with the Public Analyst laboratories in Scotland to isolate and sequence Salmonella from wastewater; and added value to RNA extracts generated by the UKHSA SARS-CoV-2 wastewater monitoring program, to assess the utility of wastewater for monitoring norovirus at a national scale [17]. A Welsh pilot undertaken by Bangor University is examining the flow, dynamics and risk posed by AMR and FBP in wastewater discharges, particularly those originating from healthcare settings. In the final year of the program, a collaborative piece of work is being undertaken to generate FBP prevalence and WGS data from business as usual UKHSA polio wastewater monitoring, which can subsequently be compared with data from clinical samples collected through the third study of Infectious Intestinal Disease in the UK; an exciting opportunity to align two large surveillance efforts to generate real insights into the effectiveness of wastewater monitoring for FBP. As well as this landscape or national scale surveillance, near source wastewater monitoring is also being investigated, both in the continuing QUB work, and through a project being led by APHA to sample abattoir wastewater for AMR, building on a pilot conducted earlier in PATH-SAFE. All the work described above is relevant both in terms of estimating contamination of the environment with AMR and FBP and monitoring a human or animal population (wastewater-based epidemiology).

4. Conclusion

PATH-SAFE is a pilot program, with a vision that the tools, techniques and partnerships developed within the different projects will become key components of a future biosurveillance solution. A significant barrier to effective biosurveillance is the complexity of sharing data among departments with different access policies. The collaborative approach of the PATH-SAFE Programme has been instrumental in resolving some of those challenges and has now expanded beyond the Program, such that PATH-SAFE finds itself at the heart of a related cluster of biosurveillance efforts. The Biological Security Strategy and National Biosurveillance Network both emerged after PATH-SAFE and worked with the Program during their development phases. PATH-SAFE has also been instrumental in informing cross-government strategic initiatives like the National Action Plan for Confronting Antimicrobial Resistance 2024 to 2029, multipartner research consortia such as the EPSRC-funded Digital Health Hub for AMR, and the UKRI Transdisciplinary AMR Networks. As PATH-SAFE draws to an end in 2025, the knowledge gained will be published and the pilot solutions that have been developed will be passed on to successor programs and/or incorporated into business-as-usual in government departments. Overall, we therefore strongly recommend similar sharing of expertise and knowledge among complementary initiatives to help underpin the UK's future response to biosecurity and AMR threats.

Funding Statement

The project discussed in this manuscript was funded by HM Treasury through the Shared Outcomes Fund. In the final year of the program, the program was match-funded by the partners.

Financial disclosure

The project discussed in this manuscript was funded by HM Treasury through the Shared Outcomes Fund. In the final year of the program, the program was match-funded by the partners. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Competing interests disclosure

Edward Haynes was employed by Fera Science Ltd during the course of this work. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Writing disclosure

No writing assistance was utilized in the production of this manuscript.