Hunting for Covid-19 in the nation’s wastewater
Geneticist Neil Gemmell is calling for a system to monitor wastewater to find possible Covid-19 clusters and carriers of the disease.
He’s part of a national group led by the Institute of Environmental Science and Research which has been given a $1.65 million grant to begin research.
Professor Gemmell chatted with Colin Peacock about how the monitoring could work.
Photo: Supplied / University of Otago
He says results from overseas projects and the current ESR research have been exciting, and shown wastewater could give an early warning of cases.
“One of the really important things about this sewage detection system is that most of the international studies are saying that Covid in sewage predicts our detection of disease in the population using standard approaches, by about two to three days.
“So if we take [the latest outbreak in] Auckland for example,… that was detected early because the family recognised that they probably had Covid-like symptoms, they went and got tested, and from there on the cluster slowly grew. And I think we were very fortunate there, and we owe a great deal of gratitude to that family for getting themselves tested early.
“But if we’d had a Covid sewage system in place perhaps we would have detected that cluster even earlier. Maybe a day or two earlier, I’m not saying we would have, I’m saying it’s a possibility. And one of the things I think is most important in this Covid epidemic is information early, so that we can take action early.”
“So you’re getting indications that people are carrying the disease in a population, perhaps asymptomatic, well in advance of them actually feeling sick and then going for a Covid test.”
So far, early studies of the wastewater testing have been done retrospectively in places where an outbreak has happened, including Italy, the Netherlands and Australia. Scientists used previously collected wastewater from those areas and compared tests from the wastewater with case data from that same period.
The New Zealand team were able to find the virus in wastewater from the Tahuna Wastewater Treatment plant, from late March and April.
“But perhaps one of the more compelling cases for the use of wastewater came from a study from Arizona State University about a month ago,” Gemmell says.
“What they were doing is as the college year was starting in North America they implemented a sewage detection system around their residential colleges.
“So they had a college there with 300 individuals in it, and they got a positive test on the sewage. And then they went back in and used the classical nasal tests… to explore that population of 300 individuals, and they found two asymtomatic cases.
“So their argument was that using this surveillance approach they avoided a full blown Covid cluster from emerging in one of their residential colleges.”
Gemmell says detecting Covid-19 is no easy task – either at traditional medical testing stations, or from wastewater.
“There’s actually a heck of a lot of work involved in processing sewage samples, so it’s slow and painstaking work. We haven’t got a process to automate that at the moment, and I don’t think anyone does internationally. But you are collecting a lot of information from a single sample across many individuals and that’s where the power lies.
“First you’ve got to get a sample – and of course the question is how representative is that sample of the population that you’re testing? You’ve got to concentrate that sample, and then you’ve got to extract the RNA, because it’s an RNA virus that we’re detecting.”
Chemical reactions are then used to amplify the virus by making copies of it – so there’s enough to detect.
“There’s lots of things that can go wrong there: you might not get a sample that is representative of the population, you can get different flow rates, stormwater can come into the system, there’ll be contaminants and other things there, and other things that are going to inhibit your test reaction. So it’s pretty complicated stuff, but it’s doable.”
Which places in the wastewater system the samples could be taken from and how many samples you’d need to cover a group are still important questions.
“The closer you can get to a source the fewer people you would then have to go back through and [test]. It’s sort of like our contact tracing approach, but at a sewer level,” Gemmell says.
“It’s not a creepy big brother thing, nobody’s identified through this, it’s just a sample of individuals.
“So, lets say… you’re doing it in an airport for a given time period and you get a detection there, it’d be just like if we had a detection at a site that we’d visited with our Covid app. Perhaps that detection at the sewage would then give an alert saying there has been a detection present in that locality where you were, and then the question is should you go and get tested if you’re showing symptoms?”
The Dunedin tests have given the team a chance to begin to explore the matter of sample points, he says.
“Here in Dunedin we’ve got one major treatment centre which has effectively been covering 100,000 people. But there are different sumps, collection points, and in Dunedin we’ve got a few of those and we’re working on trying to figure out how many.
“So those sort of collection points… might be areas that you’d get a representation of some thousands of individuals – so it isn’t a fine scale. But in terms of mass screening of a population – lets say you’re doing a thousand, ten thousand people, and you’re getting nothing on a regular basis – that gives us some confidence that we can go about our daily lives safe from Covid-19.
“If we get a detection then that might set off an alert system, then we get into the Ministry of Health standard screening procedures where we’re asking people who’ve been in that area to go and get a test. It’s a slightly more targeted approach.”
He’s eager to see the research move on to the next phase.
“At this stage it’s very much a try it and seen how it works scenario – if we were to set up a massive system and then discover it doesn’t really work the way we expect it to, that would be a mistake. So where we’re at is trying to find some exemplar situations where we can test to see if this will work.
“And of course there’s relatively little Covid in New Zealand, so the best places to test are in managed isolation units. Then if that’s working then we can look to spread that out to a much larger network.
“But I think a sample every couple of days wouldn’t be a bad place to start – from key locations. In New Zealand I think some of those key locations are; our airports, most of the cases we’re getting are imported cases; our port systems; our managed isolation units. They would be places of potential focus, and then perhaps a slightly broader network out from around those.”