Host source tracking for pathogen spillover from food and animals

A novel prediction tool that uses highly informative portions of pathogen genomes to identify the original host source and track invasive infections caused by spillover from food and animals. 

Lance B. Price, PhD; Director, Antibiotic Resistance Action Center; Professor, Environmental and Occupational Health at The George Washington University, Washington DC.

We aim to solve the challenge in identifying the source of antimicrobial-resistant Escherichia coli infections in low- and middle-income communities. E. coli is the leading cause of death from antimicrobial-resistant infections globally, causing 800,000 deaths annually.

E. coli, an opportunistic pathogen, effectively colonizes the gastrointestinal tract of humans and animals. In many LMICs, misuse of readily available antimicrobials in food animals drives multidrug-resistant E. coli, which contaminates food, and colonizes and infects humans through food-to-human and animal-to-human transmission. Although traditionally associated with intestinal diseases, zoonotic E. coli can cause severe extra-intestinal infections, a fact often overlooked due to limited analytic tools.

Existing tools, effective for outbreak investigations, struggle with sporadic events and E. coli diversity in food animals. This limits detection of such infections and impact measurement, hindering development of interventions to reduce transmission.

Distinguishing human- and animal-source E. coli infections is crucial for mitigating antimicrobial-resistant infections in LMICs. This may be driven by unregulated antimicrobial use in food animals and inadequate WASH infrastructure. In Sub-Saharan Africa, 150,000 deaths occur from antimicrobial-resistant E. coli infections annually, amidst high prophylactic antimicrobial use in food animals (90% in some areas) and 32% of the population relying on unimproved water sources.

Our solution will be directly useful to three audiences:

1. The Nigerian research community: We are supporting the research community by providing a validated solution that can improve detection of sporadic zoonotic E. coli infections and enhance one health AMR research.
2. Nigerian National AMR Technical Working groups and The Nigeria Center for Disease Control and Prevention (which is responsible for One Health AMR surveillance network): We are supporting the Nigeria CDC by providing a validated solution that can be used to benchmark rates of sporadic zoonotic E. coli infections, as well as to support subsequent surveillance efforts that can be used to determine the effectiveness of any interventions, including antimicrobial stewardship, vaccines, and Water, Sanitation, and Hygiene (WASH) interventions. 
3. Communities of food animal handlers (abattoir workers), local agriculture commissioners, and waste management authorities: WASH interventions are largely absent from abattoir settings across Nigeria. Working with local commissioners and authorities to prioritize and implement readily available interventions will directly impact protection of abattoir workers and have immediate downstream impacts on reducing contamination on purchased meat.

Our solution provides a direct benefit to the well-being of the general public by generating new knowledge on the burden of zoonotic E. coli infections, producing open-access peer-reviewed publications, new and open-access data, and an open-access model.

Eventually, as an open-source, locally-validated tool, we also hope that our solution will provide additional public good that can be used to benchmark the rate of sporadic zoonotic E. coli infections, as well as for the planning and assessment of interventions for reducing antimicrobial-resistant E. coli infections in LMICs.

Lastly, we envision that our solution will provide a public good to the research and public health communities in LMICs by creating an open-access, locally validated tool that would enable researchers and public health practitioners to produce accurate host source assignments for any E. coli isolate from any source, without requiring them to collect hundreds of E. coli reference genomes, as would be needed for traditional molecular epidemiological tools.

Our solution is expected to create tangible impact for the general population in Nigeria after our scale-up, and in countries where we will deploy our solution in the future. E. coli is the leading cause of antimicrobial-resistant bacterial infection globally, and the source of these pathogens may be from human-to-human or animal/food-to-human transmissions. Our solution will delineate the source of antimicrobial-resistant E. coli infections in Nigeria, and this information can directly inform strategies in the country’s One Health Programme.

We expect underserved and vulnerable populations to be disproportionately affected by sporadic zoonotic infections because of their likely proximity to food animals and limited access to clean water and sanitation infrastructure. In addition to increased frequency of zoonotic infections, E. coli infections from food animals may have higher rates of resistance to important clinical antimicrobials due to frequent use in food-animal production. Therefore, we expect our solution will have a positive impact on vulnerable populations in rural and semi-rural communities in Nigeria by identifying zoonotic infections and enabling targeted interventions to reduce its burden. Further, our deployment leveraging the geographic, climate, and ethnic diversity across Nigeria, the most populous country in Africa, will provide findings readily applicable to other LMIC settings.

Over the next year, we anticipate a multi-stage approach to scale and maximize our impact:

• Initiate and strengthen sample and data collection and validating/piloting our solution using Nigeria E. coli genome data.
• Leveraging existing collaborations with established networks like the National Centre for Disease Control (NCDC) and the Nigeria National One Health Technical Committee to present initial/ preliminary results relevant to the National One Health and National Antimicrobial Stewardship Agenda.

Over the next 3 years:

• Initiate scale-up in collaboration with NCDC, NTWG, and Nigerian national reference laboratories, including a phased rollout across all six Nigerian geopolitical zones following successful implementation of the project in the northern and southern regions.
• Regional Replication: Utilize the established partnership in Rwanda as a springboard to replicate the project model and its successful interventions in other African countries, fostering regional collaboration and knowledge sharing.
• Generation of peer-reviewed publications to provide open-access results to a broad audience.
• Begin validation/pilot on two other continents: Central/South America (Ecuador) and Asia (Taiwan). 

As noted, we have generated publications and have presented findings to multiple US and international agencies.

To measure success against impact goals, we propose:

1. Technical milestones
   1. Finalize panel of host-associated genetic elements for predicting meat/food animal and human sources in Nigeria.
   2. Completion of a model training dataset.
   3. Achieve accurate host source assignment on animal E. coli. Specifically, >90% of E. coli collected from food animals should be inferred as originating from an animal.
2. Dissemination milestones
   1. Publish findings in peer-reviewed journals.
   2. Present findings at quarterly technical quarterly meeting of the NTWG on AMR.
   3. Assist with public awareness campaigns, and measure knowledge and behavioral changes among target audiences regarding responsible antibiotic use, WASH, and zoonotic E. coli risks, including among abattoir workers.
3. Scale-up milestones
   1. Engage in planning with the NCDC and Nigeria One Health Coordinating Committee regarding post-pilot scale-up.
   2. Inform national and regional policies related to AMR control, WASH interventions, and responsible antibiotic use in humans and animals.
   3. Discuss future iterations of validation and piloting with regional partners. Track adoption of project-derived recommendations into existing or newly developed programs by stakeholders, including the NCDC and the Nigeria National One Health Coordinating Committee.

1. Financial resources are a key barrier. Through cycles of development for our solution, we can now accurately predict host source for E. coli from the U.S. However, we need a validation/piloting step prior to deployment in different countries/regions, which is costly. As we learn more about extending in-country validation to LMICs, this barrier may be mitigated by a validation approach that is less resource-intensive. We may also create assays that don’t need whole genome data, potentially lowering scale-up costs.
2. We believe the proposed pilot project can be rapidly performed. However, in the next phase, we may face infrastructure barrier for scaling up our solution to the national level in Nigeria. Our current plan is to be able to tap into the existing national AMR Surveillance infrastructure in Nigeria by leveraging our existing collaborations.

One of the major barriers we have faced is that the typical funding mechanisms are very focused on innovation and research, but are less interested in refinement, validation, and deployment of solutions. In addition, even though AMR should be tackled with a great sense of urgency, the typical grant funding requires a months long and often even years-long process.

Our solution is an incredible fit for the Challenge Area, and our earlier studies suggest that even in the U.S., up to 18% human E. coli extra-intestinal infections – one of the most common and frequently drug-resistant infections -- may be caused by zoonotic strains from meat and/or food animals. Our preliminary studies suggest that zoonotic E. coli may cause significantly greater extra-intestinal disease in LMICs, and this type of analytics can directly impact how LMICs choose to direct their limited resources. We believe it is important for us to rapidly validate and pilot our solution in LMIC setting, and this is the reason why we are applying to the Trinity Challenge.

We look forward to continuing our collaborations with the University of Michigan, Oxford University, the Wellcome Sanger Institute, and Statens Serum Institut (Denmark) on this work. We also look to strengthen existing partnerships with Stakeholders in the One Health Initiative in Nigeria, including the Nigeria Centre for Disease Control, Nigeria and the Nigeria Veterinary Medical Association.

We would be interested in developing partnerships with:

International Centre for Antimicrobial Resistance Solutions (ICARS)

• Our team and ICARS share the same mission of working with LMICs partners closely to customize solutions for their needs. We are dedicated to developing effective, open-access analytic resources to reduce the burden of zoonotic antimicrobial-resistant bacterial infections in LMICs.

Institute For Health Metrics and Evaluation (IHME)

• If we are successful in deploying our solutions in more LMIC settings, our findings may be used for improving global estimates of zoonotic antimicrobial-resistant infection burden.

West Africa Health Organization

• We also look to integrate findings from this research into ongoing regional efforts to strengthen the One Health and AMR Surveillance Programme by the West Africa Health Organization.