We know that outbreaks like coronavirus will become more common in the future and tackling them is the Apollo programme of our time, according to Professor Marion Koopmans, head of the viroscience department at Erasmus University Medical Centre in Rotterdam, the Netherlands. 

Habitat loss such as forest clearing or mining operations can increase the chance of a disease jumping from animals to humans, says Prof Koopmans. Photo: MPG Koopmans, Erasmus MC 2020

She is a member of the European Commission’s recently established advisory panel on Covid-19 and is coordinator of the VEO project, which is developing techniques to spot new infectious diseases as they emerge and to track them when they do. Much of what they have learned already is being used in the global fight against the new coronavirus pandemic.

What is an emerging disease?

These tend to be diseases that are circulating in human or animal populations to a degree, but when there is a change of some sort, it leads to an outbreak. In the case of the COVID-19, for example, it jumped the species barrier from animals to become a new virus in humans.

Why do new diseases like this emerge?

When we reconstruct what has happened in emerging disease outbreaks, what we see is that something has changed. More and more humans share the world, and as we try to feed and accommodate them, it leads to habitat loss for wild animals.

This disturbance can lead to a change in animal behaviour that brings them into closer contact with humans. So we might see a forest being cleared or mining operations driving animals out of their usual habitat. Those interactions are an important driver as it increases the chance of a disease crossing the species barrier into humans.

Climate change also has an impact on (existing) diseases by allowing them to move into new areas. Socio-political unrest is also important as it can bring inequality or the breakdown of health systems, which can be a risk.

Which types of diseases risk becoming outbreaks in the future?

In VEO, we have grouped diseases into different scenarios that cover a lot of possible ways we can get outbreaks. The first looks at vector-borne diseases – viruses, bacteria or parasites that can infect humans but are spread by animals, typically insects like mosquitoes or ticks.

Malaria and Lyme disease are good examples of these. We might see certain species of malaria-carrying mosquitoes appearing in new areas as the climate changes. 

Then we have zoonotic diseases, which are carried by wild birds or animals and then jump the species barrier to infect humans. There are many examples of these, such as influenza and Ebola. 

We are also looking at hidden pathogens that might be released in the future, such as diseases that are currently trapped inside permafrost but could emerge as it melts.

Finally, we have rare infections that might become more of a problem in fast growing, high density urban populations.

What has COVID-19 revealed about our ability to spot new diseases?

Disease detection mostly focuses on those that we already know about – we have surveillance networks that look out for specific diseases such as flu, norovirus or measles.

The way it is organised is pathogen by pathogen. This means we still have a reactive approach to dealing with these diseases once they become an outbreak.

What is VEO trying to do?

We want to be able to see these diseases coming better by taking a different approach that looks at everything so we can spot something before it becomes a major problem.

How do you do that?

For each of the emerging disease scenarios, we are looking at what the drivers might be. So, if I am looking at mosquito-borne diseases, we would look at what influences how many mosquitoes we have, which species, what climate and habitats are important for specific species of mosquitoes to flourish.

Are there conditions that might bring in tropical diseases, and are there people in these areas, and how are they behaving? We start layering all these things together from different types of data sets until we see a convergence of risk factors.

What is the advantage of this?

Just look at what has happened with coronavirus. We have had to set everything up from scratch and it means the number of cases is outstripping our capacity to diagnose them. We don’t expect to have clear, scalable antibody tests for another couple of months.

If we can rethink our models of disease detection, we can get ahead of that by ensuring we already have tests available and make sure there won’t be a shortage of critical reagents we need for those diagnostic tests.

We can make a start on looking for treatments and developing vaccines. But you can only do that if you are able to see these events coming.

What happens when you identify a risk?

At first a flag goes up and we would intensify surveillance in that area. But we don’t know what we are looking for so we have to use techniques that allow us to spot anything out of the ordinary that might be there. One of the most powerful techniques we have for doing that at the moment is metagenomics.

‘We want to be able to see these diseases coming ... by taking a different approach.’ Prof Marion Koopmans, Erasmus University Medical Centre, the Netherlands

What is metagenomics and why is it so useful?

It is exactly what the Chinese used when they started seeing patients with unknown pneumonia. None of the usual diagnostics were either negative or not clear, so they carried out metagenomic analysis on samples from those patients.

This is where you look at all the genetic material that is there. You might find five to ten million pieces of genetic material from all sorts of bacteria and viruses, and then you compare this with samples from healthy people. You are looking for something that stands out as unusual. This is how they first discovered the new coronavirus.

How is VEO now contributing to the global effort against the coronavirus?

We are doing a lot of work translating the genetic data we have about the virus into tools we can use against it. One part of that is using it to develop rapid diagnostic tests and particularly the phylogeny (how the virus mutates as it spread) which is a critical tool in trying to understand how widespread the virus is and if there might be some level of immunity in the population already.

We are trying to use it to also understand why children are not getting as sick as older people. We are also tracking the diversity of the virus as it spreads around the world.

A third element is also trying to work with social media – we have digital epidemiologists who are trying to track Twitter feeds to understand how the disease might be spreading from new reports, for example. This has been overwhelming though because of the sheer volume of information now being shared.

We are using neural networks to analyse what is driving the information and where the reliable information is coming from. One of the first reports about the coronavirus outbreak was about an unknown disease in a Chinese newspaper. This is the sort of thing we are now trying to pick up. And of course, we are also trying to predict what the next emerging disease will be.

How close are you to being able to do that?

The project is only in its first few months, but we know that outbreak events like these are likely to become more common in the future because of the growing human population, changing climate and land use change happening around the world.

Diseases like Nipah virus (which is spread by bats and can pass to humans from infected pigs) and other bat-borne respiratory and neurotropic infections are of particular concern. 

Tackling these diseases is an Apollo programme for our time in terms of the effort, technology and scale of what is required. If we can bring the detection process forward so we can spot outbreaks coming, we can move very fast against them, track them and develop vaccines to keep them under control.