But researchers are pressing ahead with new clinical trials for more targeted treatments built on advances in our understanding of long covid. There’s now a large body of research on the condition. “It is absolutely not mysterious,” said Hannah Davis, co-founder of the Patient-Led Research Collaborative.

“This is a campaign of human service and scientific discovery,” said Wes Ely, professor of medicine and co-director of the Center for Critical Illness, Brain Dysfunction, and Survivorship at Vanderbilt University Medical Center.

Immune dysfunction in long covid

Long covid is an umbrella term, and treating it as one disease is, in a way, equivalent to thinking cancer is one entity as opposed to the hundreds of types, each with different diagnostics, prognosis and treatments, said Ziyad Al-Aly, a clinical epidemiologist at Washington University in St. Louis.

Long covid potentially has many subtypes with different biological pathways and responses to different treatments, researchers said.

“It is a constellation of many things, and dissecting out those things is going to be the biggest challenge,” said Janna Moen, a neuroscientist at the Howard Hughes Medical Institute, who recently wrote a review about how the immune system affects the brain and spinal cord in people with long covid, which is awaiting peer review.

“It’s going to mean that everybody is going to need a different treatment. Personalized medicine is going to become very necessary,” she added.

One pathway to long covid may be through immune system dysfunction, where the “immune system goes haywire,” Topol said.

People have a 20 to 40 percent increased risk of developing different autoimmune issues following a coronavirus infection, four large datasets of millions of patients show.

Some people with long covid also have persistently elevated autoantibodies, where the immune system mistakenly produces antibodies that attack healthy tissue — particularly in the nervous system — which may play a role in the condition, said Akiko Iwasaki, professor of immunobiology at Yale University School of Medicine.

In addition, coronavirus infection can reactivate latent viruses, which may contribute to autoimmunity. The Epstein-Barr virus — known for its role in infectious mononucleosis and present in over 95 percent of adults — gets reactivated in a subset of people, which may be linked to developing long covid, Iwasaki said.

Viral persistence

Another pathway to long covid may be viral persistence, where bits of the coronavirus linger long after the acute infection.

Viral persistence could chronically activate the immune system into producing a smoldering inflammation that can weaken the blood-brain barrier, which typically shields the brain. Inflammatory molecules could cross into the brain and trigger neuroinflammation.

Many researchers were surprised to learn that viral particles can linger in the body even after the acute infection is over.

“We just didn’t think of coronaviruses having the capability of lasting more than a couple of weeks,” said Michael Peluso, an assistant professor and infectious-disease clinician at the University of California at San Francisco.

But fragments of the virus can stay in our bodies for far longer — 14 months after infection in some patients, and almost two years in some others.

Coronavirus spike protein may be particularly disruptive and inflammation-inducing, lingering in the skull, meninges — the protective tissue around the brain — and the brain itself. It is also associated with inflammation and damage in the nervous system, according to a 2024 study.

Yet another route to long covid may be through the inflammation of the endothelial cells that line blood vessels throughout the body, including the brain.

The coronavirus spike protein also binds to clotting proteins in the blood, creating microclots that are themselves pro-inflammatory and alter immune cell activity, a 2024 study reported.

All these different likely biological pathways of long covid are “all kind of related unfortunately in a way that makes it very hard to parse out exactly what the problem is,” Moen said.

Gaps in diagnosis and treatment

Despite progress in basic research, there are no validated diagnostic tests for long covid.

“It is a patient-reported syndrome,” Peluso said. Though many other medical conditions are diagnosed wholly based on how people feel, such as pain disorders and depression, “a lot of people use that to discount it or express skepticism or disparage the condition,” he said.

As a result, long covid patients face challenges in getting adequate access to care. Though long covid is now more recognized in the medical fields, the onus is still on the patient to find doctors who will listen.

“We tend to stay the same or get worse. And we’re very needy. We need a lot of medications. We need a lot of paperwork filled out,” said Moen, who as a long covid patient has seen dozens of doctors. “A lot of them just don’t like taking us as patients.”

The lack of a biomarker has also stymied clinical trials looking for evidence of improvements, researchers said.

“There really needs to be a concerted, organized and well-funded biomarker discovery effort for this problem so that we can find a diagnostic test,” Peluso said.

There are also no approved treatments for long covid.

“What’s so sad here is people are so desperate, and we don’t have anything validated,” Topol said.

Vaccinations remain the best way to prevent coronavirus infection and lower the risk of subsequent long covid by reducing the severity of the initial infection and immune response. However, coronavirus vaccination rates this winter were “dismal,” Al-Aly said.

During an infection, the “preponderance of evidence” shows that taking antivirals, such as Paxlovid, does decrease the risk of developing long covid later — though it’s not unanimous, Al-Aly said. A 2025 clinical trial of over 20,000 participants found that people treated with another antiviral, molnupiravir, felt better, had fewer and milder symptoms, and took less time off work six months after infection.

However, once long covid has developed, antivirals have not been effective, with two recent large randomized clinical trials finding no differences in outcomes compared with placebos.

Instead, the strategy is to manage the symptoms — variable as they are — in each patient.

There is no “cookie-cutter protocol” for treating long covid, Ely said. “The best clinicians evaluate them head to toe and see what their story is and how they’re suffering. And they tailor their therapy accordingly,” he said.

Davis, who has long covid, has tried blood thinners, antivirals and other repurposed drugs for her symptoms to varying degrees of success.

Other patients have tried treatments such as hyperbaric oxygen chambers and plasmapheresis, which filters out and then replaces their blood, but studies on these interventions tend to be small and without a placebo control, Topol said.

Searching for treatments

Like the virus they study, researchers and clinicians have evolved in how they tackle long covid.

Earlier clinical trials lumped together people with all kinds of long covid symptoms; now, researchers are increasingly stratifying patients by symptoms in the hopes of targeting specific biological pathways with their treatments.

This approach would also make it more likely to spot therapeutic benefits that could otherwise be missed if given to a broader patient population.

The treatment strategy has also shifted.

“All of the initial trials were really focused on targeting the virus, and now we’re starting to see trials that are focusing on targeting the immune system,” Peluso said.

Peluso, Ely and their collaborators are running a large double-blind, randomized clinical trial on a treatment seeking to turn off the immune “light switch that is creating this disease state, creating so much human suffering,” Ely said.

The trial, REVERSE-LC, will give patients baricitinib, which is an FDA-approved immunotherapeutic drug for treating other immune disorders such as rheumatoid arthritis and alopecia areata.

Another ongoing double-blind, randomized clinical trial, called ADDRESS-LC, will test a different, more specific immunomodulator called bezisterim for neurocognitive long covid.

Bezisterim is a novel anti-inflammatory agent that has been shown to selectively inhibit specific inflammatory pathways while not suppressing the immune system, said Penelope Markham, senior vice president at BioVie, a pharmaceutical company sponsoring the study.

Importantly, it can cross the blood-brain barrier, meaning it may potentially treat neuroinflammation at its source, and is being developed to treat other inflammatory neurological conditions such as Parkinson’s disease.

Still, researchers say what is coming down the pipeline is not enough compared with all the treatments that have been identified as promising.

“There’s a lot of potential shots on goal. And you have to run through this, but you can’t know until you’ve done the rigorous test,” Topol said.

The goal of these clinical trials is to “provide mercy, and mercy out of love … And my definition of mercy is my willingness to dive into their chaos and provide lifting and healing,” Ely said.

This story is the second of a three-part series on five years of long covid. The first story focused on the neurocognitive effects of long covid, and the third story will explore how postinfection illnesses aren’t new — and how we can better prepare for them in the future.