Coronavirus found in Paris sewage points to early warning system

By sampling sewage across greater Paris for more than one month, researchers have detected a rise and fall in novel coronavirus concentrations that correspond to the shape of the COVID-19 outbreak in the region, where a lockdown is now suppressing spread of the disease. Although several research groups have reported detecting coronavirus in wastewater, the researchers say the new study is the first to show that the technique can pick up a sharp rise in viral concentrations in sewage before cases explode in the clinic. That points to its potential as a cheap, noninvasive tool to warn against outbreaks, they say.

“This visibility is also going to help us predict a second wave of outbreaks,” says Sébastien Wurtzer, a virologist at Eau de Paris, the city’s public water utility. Wurtzer and his colleagues posted the study, which has not been peer-reviewed, on the preprint repository medRxiv on 17 April.

Sewers offer near–real-time outbreak data, because they constantly collect feces and urine that can contain coronavirus shed by infected humans. (Once excreted from the body, the virus degrades quickly, although scientists have found limited instances of infectious virus in fecal matter.) Polymerase chain reaction testing identifies fragments of RNA from SARS-CoV-2, the virus that causes COVID-19. Higher concentrations of virus in the wastewater corresponds to higher numbers of infected people who contribute to the sewer system.

For the Paris study, Wurtzer and his colleagues sampled wastewater from up to five Paris-area plants twice a week between 5 March and 7 April. They noted “high concentrations” of viral RNA several days before 10 March, the first day that Paris recorded multiple deaths from COVID-19. Concentrations continued to rise a few days ahead of an acceleration in clinical cases and deaths in Paris. “We have a very clear curve that precedes the curve in numbers of clinical cases, and now with confinement, we see a flattening of that curve,” says Laurent Moulin, a study co-author and a microbiologist also at Eau de Paris. He estimates it took between a half a day and 3 days for the sewage to move from toilets to the treatment plants.

Sewer monitoring can illustrate the timing and scale of outbreaks that are currently difficult to visualise because of a general lack of human testing, says Zhugen Yang, a biomedical engineer at Cranfield University’s Water Science Institute, a U.K. centre that is developing $2 tests detecting SARS-CoV-2 in sewage. “In most countries, individual tests are in short supply, and outbreak figures are based on computer modelling,” he says. “But sewer sampling gives a fairly inexpensive, evidence-based image of the actual viral load in a community.” Using computer models that incorporate data on how many viral particles individuals shed, and how they become diluted in sewage, it is even possible to translate detected viral concentrations into estimates of absolute numbers of infections in a sewage system’s catchment area, he says.

Another advantage of wastewater sampling is that it picks up virus associated with the vast number of people who are infected with SARS-CoV-2 but do not present symptoms for the disease, says Paul Bertsch, science director of land and water at the Commonwealth Scientific and Industrial Research Organisation in Australia. Although viral shedding varies among individuals and over the course of their infection, he says, a sewage system blends these variations into an average that represents the wider community. And depending on the sewage system, the warnings can come quickly. He points out that wastewater monitoring in Israel, for example, picked up a polio outbreak before any clinical cases appeared at all, according to a 2018 study.

Building on similar studies in the Netherlands and the United States, Bertsch’s group last week reported the first detection of coronavirus in Australian sewage. He and his colleagues sampled wastewater in Brisbane representing 600,000 people, in March and April. In contrast to the study in Paris, they found a peak of viral shedding that corresponded to the peak detected through direct human testing. The difference might be explained by more prevalent human testing in Australia, he says.

Bertsch says he hopes to “tap into” Australia’s existing systems for monitoring wastewater for illegal drugs to develop a national COVID-19 monitoring system that could be in place within 1 month. Later, it might even be feasible to “go up-pipe” with specialised sampling portals allowing finer-scale community sampling. “We could test by postal code, for example,” he says.

Meanwhile, as suppression measures drag on across the globe, exacting huge economic costs, some nations are easing up on restrictions and some people are protesting lockdowns. Sewage data might make them think twice, Yang says. “They could see the numbers and understand the seriousness of this seemingly invisible outbreak,” he says. “Seeing is believing.”