What causes long COVID? For nearly half of cases, new research suggests an answer.

A study from researchers at Mass General Brigham may pave the way for effective treatments

Darlene Gildersleeve, who has battled long COVID, worked on a math lesson with her daughter, Abigail, in 2023.Craig F. Walker/Globe Staff

Scientists and doctors have suspected for years that one cause of the mysterious condition known as long COVID may be reservoirs of the virus that remain hidden in the bodies of its victims long after their acute infections have passed.

Earlier this month, a research team led by Boston-area scientists unveiled a study suggesting this is true for almost half of those suffering from the condition.

“It’s unlikely that persistent virus is the cause of all long COVID symptoms,” said Dr. David Walt, co-director of the Mass General Brigham Center for COVID Innovation and the study’s senior author. “What is more probable is that it’s one of the causes.”

The Centers for Disease Control and Prevention estimates almost 7 percent, or close to 18 million Americans, suffer from long COVID, a chronic condition often marked by a series of debilitating symptoms that can include extreme exhaustion, difficulty breathing, neurological problems, chronic pain, and brain fog. Almost five years after the syndrome was first identified, there are still no approved blood tests to diagnose long COVID, no clinically validated treatments, and no cure. Due to the myriad ways the disease manifests — there are more than 300 reported symptoms — many now believe long COVID may encompass several distinct conditions, with distinct causes.

Because long COVID is often characterized by an overactive immune system, those who study the condition have long suspected its most likely cause is hidden viral reservoirs that researchers have found difficult to detect and locate.

Walt’s team applied an experimental blood test 1,000 times more sensitive than the standard laboratory tests to 1,569 blood samples taken from 706 individuals in different parts of the country who enrolled in studies after their first COVID infections. Of those patients, about 80 percent had reported symptoms associated with long COVID.

The ultrasensitive tests, developed in Walt’s lab, detected fragments of the virus’s signature spike proteins in roughly 43 percent of the samples drawn from those who reported long COVID symptoms. They also found fragments of the proteins in 21 percent of those who had no symptoms. (It was not clear why the persistent virus did not make them sick, too.) The presence of the fragments, which were in such small amounts they would not be detectable using standard technologies, are a smoking gun, specialists say.

“Viruses hide, that’s what they do,” said David Putrino, an expert on long COVID, who runs a clinic at New York Mount Sinai Hospital and was not involved in the new research. “If you’re finding evidence of them in plasma, that probably tells you that there’s a lot more going on in tissue where viruses like to hide, like nerve tissue, gut tissue, joint tissue, et cetera.”

Though others have detected the presence of viral fragments in long COVID patients before, Putrino said the new study is among the most comprehensive. The fact that it was funded by the National Institutes of Health (NIH), he added, is “very important,” suggesting that federal officials may finally be willing to listen to front-line long COVID researchers and patient advocates.

The traditional medical establishment, including the NIH, Putrino said, has been slow to embrace the viral persistence theory, because it defies conventional medical thinking. Until recently, many argued that viruses that replicate using RNA, the type of genetic material found in the virus that causes COVID-19, are incapable of persisting in the body.

Others said the Harvard study offered little new information. Dr. Ziyad Al-Aly, chief of research and development at the VA St. Louis Health Care System, noted that among most long COVID researchers, the idea that viral fragments are driving many of its symptoms is widely accepted and has been the prevailing theory since the condition was first identified. Many previous studies, he said, have already found the persistence of viral fragments in about 50 percent of the patients tested.

“We keep saying the same thing,” Al-Aly said. “We’re stuck in the same loop. To move the field forward, we need to ask new questions: why some people experience viral persistence, how viral persistence is leading to tissue damage and subsequently symptoms of long COVID. And really most importantly of all, how to address it.”

Walt does not disagree.

“Just knowing that you have a persistent viral infection, which is the underlying cause of long COVID, is meaningless unless there’s a follow on treatment,” he said.

His new study, he added, is part of a larger effort to develop one. The key to the current study was a single molecule test called SIMOA that he developed in his lab more than a decade ago to detect cancer. It uses tiny beads attached to antibodies that are engineered to seek out and stick to specific proteins, which can then be labeled with fluorescent markers. The markers are then detected and counted by advanced imaging devices.

The test is now being used in a second NIH-funded study launched in July called RECOVER-VITAL, aimed at finding a cure. The 900-person, multisite clinical trial, which ends its data collection phase in January, is testing the use of the antiviral drug Paxlovid on patients who test positive for the presence of persistent viral proteins using Walt’s blood test, patients with long COVID who do not test positive for viral fragments, and a control group. The trial is also experimenting with different durations of the therapy, with some receiving a 15-day regimen, and some a 25-day regimen.

Al-Aly notes that a previous study that gave Paxlovid to long COVID patients for two weeks failed to eradicate the condition. But Walt has high-hopes for the new study, which, unlike previous studies, will identify which patients have remnants of the virus present and extend the period of time they receive Paxlovid.

“The expectation is that patients who have the presence of spike protein that receive the antiviral and who presumably clear the virus will no longer have the circulating antigens present in their bloodstream,” he said. “The hope is that we’ll see those individuals recover and return to their normal healthy state.”