The coronavirus can warp the body’s defenses in many ways — disarming the body’s early warning systems, for example, or causing immune cells to misfire. But a spate of new studies suggests another insidious consequence: The infection can trigger the production of antibodies that mistakenly attack the patient’s own tissues instead of the virus.
The latest report, published online this week, suggests that so-called autoantibodies can persist months after the infection has resolved, perhaps causing irreparable harm. If other studies confirm the finding, it may explain some of the lingering symptoms in people who have recovered from COVID-19. The syndrome, sometimes referred to as long COVID, can include dementia, “brain fog” and joint pain.
Autoantibodies are not new to science: They are the misguided soldiers of the immune system, tied to debilitating diseases such as lupus and rheumatoid arthritis, which arise when the body attacks its own tissues.
The newest study is small, with just nine patients, five of whom had autoantibodies for at least seven months. It has not yet undergone peer review for publication, and the authors urged caution in interpreting the results.
“It’s a signal; it is not definitive,” said Dr. Nahid Bhadelia, medical director of the special viruses unit at Boston Medical Center, who led the study. “We don’t know how prevalent it is, and whether or not it can be linked to long COVID.”
The question of autoimmunity following coronavirus infection is urgent and important, Bhadelia added. As many as 1 in 3 survivors of COVID-19 say they still experience symptoms.
“This is a real phenomenon,” she said. “We’re looking at a second pandemic of people with ongoing potential disability who may not be able to return to work, and that’s a huge impact on the health systems.”
A growing body of evidence suggests that autoimmunity contributes to the severity of COVID-19 in some people. A study published online in October found that among 52 patients with severe COVID-19, more than 70% carried antibodies against their own DNA and against proteins that help with blood clotting.
In another study, also published online in October, researchers discovered autoantibodies to carbohydrates made by the body in COVID-19 patients, which could explain neurological symptoms. And a study in the journal Science Translational Medicine in November found that half of patients hospitalized for COVID-19 had at least transient autoantibodies that cause clots and blockages in blood vessels.
The gathering research raises the worrying possibility that lingering autoantibodies might lead to autoimmune disease in some people infected with the coronavirus.
“Once these autoantibodies are induced, there is no going back,” said Akiko Iwasaki, an immunologist at Yale University. “They will be a permanent part of the person’s immune system.”
She added: “What does it do to vaccine response? What does it do to newly acquired infections? These are all questions that will have to be addressed.”
Iwasaki’s team showed in December that severely ill patients had dramatic increases in a wide array of autoantibodies that target parts of the immune system, brain cells, connective tissue and clotting factors.
“We really see broadly reactive autoantibody responses in these patients,” Iwasaki said. She had suspected that autoimmunity might play some role, but “even I didn’t expect to see this much auto-reactivity.”
Iwasaki and her colleagues collected blood from 172 patients with a range of symptoms, 22 health care workers who had been infected, and 30 uninfected health care workers.
One in five infected patients had autoantibodies to five proteins in their own bodies, and up to 80% to at least one protein, the researchers found. Patients with severe COVID-19 had many more of these antibodies, which hindered their immune responses and exacerbated illness. Of 15 patients who died during the study, 14 had autoantibodies to at least one constituent of the immune system.
The study convincingly shows that autoantibodies “alter the course of disease,” said Marion Pepper, an immunologist at the University of Washington in Seattle who was not involved in the research.
Autoimmunity after an illness is not unique to the coronavirus. Other intensely inflammatory infections, including malaria, leprosy and respiratory viruses, are also known to trigger autoantibodies. But autoimmunity and COVID-19 may be a particularly hazardous mix, experts said.
One analysis of nearly 170,000 people with rare autoimmune rheumatic diseases like lupus and scleroderma indicated that they face increased odds of death from COVID-19. And a study of more than 130,000 people found that autoimmune conditions like Type 1 diabetes, psoriasis and rheumatoid arthritis increase the risk of respiratory complications and death from COVID-19.
Some of the antibodies seemed to be the result of inborn defects in the immune system. For example, a study in the journal Science in October found that about 10% of severely ill COVID-19 patients had existing autoantibodies that attacked key components of the immune system that were supposed to kick in after exposure to the virus. Without that rapid response, the body’s defense is hopelessly delayed, fighting a losing battle against the multiplying virus.
Yet the mere presence of autoantibodies does not indicate harm. They are in the general population and don’t always lead to illness, some experts noted.
“Anywhere from 10 to 15% of the population has some level of this auto-reactivity,” said Dr. Iñaki Sanz, an immunologist at Emory University. “The issue is that you need many other events downstream of the autoantibodies to induce disease.”
At least in some patients, autoantibodies clearly emerged as a result of the illness, Iwasaki’s study showed. Extreme inflammation caused by viral infections can cause cells to burst open, spewing their contents and befuddling the immune system’s ability to distinguish “self” from “other.”
But autoantibodies induced in this manner may level off after a few months, said Dr. Shiv Pillai, an immunologist at Harvard University: “Probably in the vast majority of COVID-19 patients, autoantibodies emerge in the acute phase, then decline.”
“That being said — yes, it would be interesting if long COVID might be explained by specific autoantibodies,” he added.
Several researchers, including Bhadelia and Iwasaki, are following patients over time to see how long autoantibodies persist and whether they wreak permanent damage. Although scientists have known that acute infections can trigger their presence, the phenomenon has never been studied in such detail.
“That’s maybe the one silver lining here,” Pepper said. “We’re going to learn some fundamental principles about acute viral infections in people which haven’t been easy to study in this way before.”