Antibodies that turn against elements of our own immune defences are a key driver of severe illness and death following SARS-CoV-2 infection in some people, according to a large international study. These rogue antibodies, known as autoantibodies, are also present in a small proportion of healthy, uninfected individuals — and their prevalence increases with age, which may help to explain why elderly people are at higher risk of severe COVID-19.
The findings, published on 19 August in Science Immunology1, provide robust evidence to support an observation made by the same research team last October. Led by immunologist Jean-Laurent Casanova at the Rockefeller University in New York City, the researchers found that around 10% of people with severe COVID-19 had autoantibodies that attack and block type 1 interferons, protein molecules in the blood that have a critical role in fighting off viral infections2.
“The initial report from last year was probably one of the most important papers in the pandemic,” says Aaron Ring, an immunologist at the Yale School of Medicine in New Haven, Connecticut, who was not involved in this work. “What they’ve done in this new study is really dig down to see just how common these antibodies are across the general population — and it turns out they’re astonishingly prevalent.”
The international research team focused on detecting autoantibodies that could neutralize lower, more physiologically relevant concentrations of interferons. They studied 3,595 patients from 38 countries with critical COVID-19, meaning that the individuals were ill enough to be admitted to an intensive-care unit. Overall, 13.6% of these patients possessed autoantibodies, with the proportion ranging from 9.6% of those below the age of 40, up to 21% of those over 80. Autoantibodies were also present in 18% of people who had died of the disease.
Casanova and his colleagues suspected that these devious antibodies were a cause, rather than a consequence, of critical COVID-19. There were hints that this might be the case — the group had previously found that autoantibodies were present in around 4 in 1,000 healthy people whose samples had been collected before the pandemic2. The team also found that individuals with genetic mutations that disrupt the activity of type 1 interferons are at higher risk of life-threatening disease3,4.
To examine this link further, the researchers hunted for autoantibodies in a massive collection of blood samples taken from almost 35,000 healthy people before the pandemic. They found that 0.18% of those between 18 and 69 had existing autoantibodies against type 1 interferon, and that this proportion increased with age: autoantibodies were present in around 1.1% of 70- to 79-year-olds, and 3.4% of those over the age of 80.
“There is a massive increase in prevalence” with age, Casanova says. “This largely explains the high risk of severe COVID in people in the elderly population.” He adds that these findings have clear clinical implications, and suggests that hospitals should be checking patients for these autoantibodies, as well as mutations implicated in blocking type 1 interferons. This could identify people who are more likely to become critically ill from COVID-19, helping physicians to tailor their treatment appropriately.
A sample of more than 30,000 people is “too big to ignore”, according to Ring. “It just shows that this is something that we need to think about.” He adds that researchers should now consider whether autoantibodies play a part in driving other infectious diseases. Ring’s team has already found evidence5 of autoantibodies against various immune-system components in people with COVID-19, and he and his colleagues are now investigating further. “I suspect that we’ve just started scratching the surface,” Ring says.
Nature 597, 162 (2021)
Bastard, P. et al. Science Immunol. 6, eabl4340 (2021).
Bastard, P. et al. Science 370, eabd4585 (2020).
Asano, T. et al. Science Immunol. 6, eabl4348 (2021).
Zhang, Q. et al. Science 370, eabd4570 (2020).
Wang, E. Y. et al. Nature 595, 283–288 (2021).