Predicting Zoonotic Respiratory Virus Transmission to Humans

Identifying predictors of pathogen emergence and transmission based on One Health-focused zoonotic virus surveillance and multilayer network modeling.

Gregory C. Gray, Professor and Infectious Disease Epidemiologist, School of Medicine, Global Health Institute, and School of the Environment, Duke University, North Carolina, USA

During the last 25 years the world has experienced the emergence and spread of a number of novel respiratory viruses that have caused considerable morbidity and mortality. The majority of novel infectious organisms emerge from wildlife or domesticated animals. These emerging viruses have had major impacts on international travel, trade, and economies. The World Bank estimates that the SARS epidemic cost the affected nations US$30-50 billion, and the H1N1 pandemic US$45-55 billion. Infectious disease experts have long aimed to predict the next pandemic, but they have yet to succeed, leaving public health communities scrambling to identify and contain infectious disease outbreaks once they occur. During lengthy outbreak investigations, the virus continues to transmit in humans, further adapting to the host and becoming more established.

We take a mathematical, interdisciplinary approach rooted in One Health: we focus on the interface of humans and non-human animals, using social data and rigorous network modeling to identify transmission pathways and the conditions that facilitate disease emergence. Through this research, we will identify better ways to predict, detect, and mitigate future emerging virus threats at the animal-human interface, before the viruses fully adapt to humans, enter transportation hubs, and become global pandemics.

See supplemental figures

This research aims to transform understanding of novel zoonotic disease ecology and transmission. This effort will result in a large collection of specimens and high-quality transmission trees at the animal-human population interface, which will be made available for others to analyze. Findings and perspectives from the research will be communicated through a public lecture series. 

In terms of broader impacts, we will provide crucial training to improve viral detection capacity in Vietnam and China, which are hotspots of disease emergence. This project involves conducting four prospective, controlled, One Health studies of animal workers, their animals, the animal environments, and their non-animal working associates at four sites in Asia. At each site, we will perform a prospective study of the above subjects for evidence of novel emerging zoonotic viruses.

We have identified coinvestigators in China and Vietnam who will perform the proposed studies at each site. This international collaboration will identify novel respiratory viruses in four zoonotic disease-relevant settings in Asia, and thus inform public health actions.

Each year, planned exchanges of graduate students and visiting investigators will additionally have broader impacts beyond the research findings in strengthening the scientific partnerships.

The research team will markedly improve viral detection capacity at the four collaborating institutional research sites in Vietnam and China via training and direct involvement in the research, thus contributing to public health goals. The international trainees will learn a suite of skills ranging from molecular diagnostic techniques, to survey methods and how to implement the One Health and ecological modeling approaches used in this research. Each year, planned exchanges of graduate students and investigators will contribute to these training goals while also cementing the international partnerships that are crucial to the proposed research. Funding will also support One Health training workshops in China, Vietnam and at Duke University, which will be aimed at advancing One Health research and increasing ecological, evolutionary, and mathematical training within One Health. Findings and perspectives from the research will be communicated to the public via scientific publications and through a lecture series.

This work is not an esoteric, academic big data exercise.  We see it as generating new knowledge which will lead to practical interventions in animal husbandry and public health, not only for Vietnam and China, but potentially throughout the world.  We expect the work will lead US, UK, and EU teams to focus novel pathogen surveillance efforts more directly on the human-animal interface rather than on ambitious searches for all the viruses that the world’s estimated 6 million animals of the world may harbor. We have strongly argued for this shift in focus in recent manuscripts, especially in:

Gray GC et al. Mitigating Future Respiratory Virus Pandemics: New Threats and Approaches to Consider. Viruses 2021 invited manuscript

We have also demonstrated that focusing pathogen discovery at the human-animal interface has helped us identify a number of cross-species spillover pathogens in both humans and animals. Most interesting has been our recent discovery of what is likely the eighth recognized coronavirus causing disease in humans:

Vlasova AN et al. Novel Canine Coronavirus among Hospitalized Pneumonia Patients in Malaysia. Clin Infect Dis 2021 in review 

We are confident that our multilayer network modeling approach will improve future infectious disease transmission modeling.

Successful demonstrations of One Health-oriented surveillance in Vietnam and China are likely to lead to expansions of such surveillance in other Emerging Infectious Disease hotspots throughout the world.

Successful demonstrations of this multilayer networks modeling may improve our infectious disease transmission modeling which has repeatedly been denigrated during the COVID-19 pandemic as inaccurate.

We recognize that our proposal is both data intense and ambitious, however, it is rooted in data collected in the field in lieu of artificial data generated by supposition or contingency tables. 

We posit that such field data will be held as more valuable than the many infectious disease transmission models that have been previously employed to summarize past epidemics and project future infectious disease transmission.

Our solution will measure success through the various data we collect and the outcomes we generate. 

We expect the four study sites to engage a total of 300 persons (240 animal workers and 60 non-animal exposed associates) working at multiple swine or poultry farms and multiple live animal markets in China and Vietnam.  The serial sampling will generate a total of 20,700 samples to study: 7,500 (300 participants x 25) nasal washes samples, 6,000 (240 x 25) animal oral secretion samples, 6,000 (240 x 25) animal fecal samples, and 1200 paired human sera samples (300 participants x 4 draws).

We expect to identify and characterize at least 50 novel animal viruses over the three-year study period, and to produce approximately 75 transmission trees (at least one tree per virus identified).

The novel viruses will help us anticipate future pandemic threats, their ecology, and their interventions.

The transmission trees will be used to investigate the social, ecological, and occupational drivers of infection with animal-derived pathogens and the potential for superspreading of infectious organisms among human populations.  

The research will also generate a set of methods to rigorously infer transmission trees by integrating multiple types of serially-collected data into multilayer networks.   

Our chief barriers are financial. We have pitched related projects to multiple US government agencies and the US government entities which do most such surveillance work have quite different strategies. We have reviewed how their animal pathogen centric or human pathogen centric approaches failed to predict both the 2009 and 2019 pandemics in writing and in multiple lectures but as yet found no major sponsors. See:

Gray GC, Abdelgadir A. While We Endure this Pandemic What New Respiratory Virus Threats Are We Missing? Open Forum Infect Dis 2021, ofab078. View

Gray GC, Robie ER, Studstill CJ, and Nunn CL. Mitigating Future Respiratory Virus Pandemics: New Threats and Approaches to Consider. Viruses invited

One potential pitfall is the loss of study participants through participation fatigue, government edict, or a farm owner’s disenchantment. Participants who are lost to follow-up (2 consecutive weeks) will be replaced with an individual of similar exposures and age-group. 

Other potential pitfall could involve a loss of research personnel or research supplies.  We expect these losses to be of low risk as the study duration for each site is brief (24 months) and we are prepositioning most all the study supplies needed for field work.

Duke University

University of North Carolina

Duke-NUS Medical School, Singapore

Duke Kunshan University, China

We are confident that our plan to conduct surveillance for and model respiratory virus transmission in our proposed way is a breakthrough in helping to identify and preempt prepandemic threats. We applied with the hope that the Trinity Challenge would help us in demonstrating that our plan works as proposed.

Diagnostic companies, philanthropic organizations, non-governmental organizations, etc.