‘There’s no one long Covid’: Experts struggle to make sense of the continuing mystery

Robert Gallo apologized for still coughing. The day before President Biden tested positive for Covid-19, the famed HIV researcher said he was still recovering from a Covid infection that had left him unable to walk, put him in the hospital, and made him delusional, he said Wednesday during a roundtable discussion about long Covid.

Presented by the Global Virus Network, a coalition of leading virologists, the two-day virtual conference convened experts across disciplines and around the world to ask and answer questions about what causes long Covid, how to predict who gets it, how to treat it, and just possibly how to prevent it.

No one has the answers, but Gallo, who co-founded the group and is also director of the Institute of Human Virology at the University of Maryland School of Medicine, puts his money on the amount of virus present right from the start. “We have definitive data that vaccine reduces virus, so if we can take that as a conclusion that the amount of virus is critical to predicting the future, you have a great biomarker,” he said. “I don’t think you can wait. I agree with those clinical people who want to go forward right away.”

To a person, scientists expressed eagerness for better studies, better funding, better participation, with urgency bubbling up from specialists in cardiology, neuroscience, epidemiology, pulmonology, and immunology. Yet these are still early days for research into long Covid. Recognized since 2020, its definition is still sometimes debated, although most definitions include symptoms that persist weeks or months after acute infection and include fatigue, headache, shortness of breath, memory problems, GI issues, and joint and muscle pain.

We still need a taxonomy, Yale cardiologist Harlan Krumholz said, to sort people and their myriad symptoms into groups so that scientists with different expertise can speak the same language as they try to better understand what is going wrong.

But when something seems to work, try it, was the consensus. “I don’t think we should wait just to thoroughly understand a mechanism before trying some reasonable interventions, especially if the interventions are low risk,” Krumholz said.

Epidemiologist Sairam Parthasarathy of the University of Arizona College of Medicine – Tucson painted the picture of prevalence, setting it at 43% of all Covid cases based on pooled evidence of 50 studies. He called out the risk of long Covid as greater than the risk of developing diabetes and asthma, citing a study from Italy that estimated it at 25%. And in the U.S., disadvantaged populations, including Native American and Hispanic people, are disproportionately more likely to be hospitalized for Covid. “It’s a few that may actually carry the burden of the many, and we need to address this,” Parthasarathy said.

There are lessons from another familiar disease: cancer. Michelle Monje, a neuroscientist and neuro-oncologist from Stanford, has previously connected long Covid effects in the brain to the cognitive impairment called “chemo brain” that follows treatment with methotrexate. Now she says long Covid also resembles what happens in the cytokine storm that follows the cancer immunotherapy CAR-T. In all three cases, inflammation disrupts immune cells in the brain called microglia, which ordinarily maintain healthy neural circuit function but when inflammation strikes, become neurotoxic. In mice, she found that depleting microglia with a small molecule that targets a necessary growth factor receptor allows microglia to come back to normal and rescue the cognitive deficits after cancer therapies. “This is something that we have not yet tested, but are in the process of testing in the context of long Covid,” she said.

Monje has also zoomed in on a particular protein circulating in the brain at elevated levels sparked by inflammation in response to viral infection. In blood samples taken from patients who had relatively mild Covid in spring 2020 and then cognitive impairment later, higher levels of the chemokine CCL11 levels persisted. She also found a lot of the variability in CCL11 levels that could be explained by previous autoimmune disease.

“Inflammation causes neuroinflammation, which causes dysregulation of multiple cellular lineages. And, you know, we think that this is a relatively common mechanism contributing to cognitive impairment after Covid,” Monje said. “It really begs the question of how various immune challenges that might elicit different cytokine profiles might increase the risk for overlapping yet distinct constellations of neurological and psychiatric symptoms.”

There may be other targets to explore, and other biomarkers to guide the way. Here’s where some of the other scientists are devoting their attention:

James Harker of Imperial College London studies the impact of long Covid on the lungs, using CT scans and proteomic tools to see if there is lasting damage to the lungs after acute infection, such as the scarring known as fibrosis. As in other long Covid research, there wasn’t a strong link between severity of the disease and ongoing changes in the lungs. And the story goes beyond inflammatory responses to infection. “The proteins we see in the post-Covid lung are largely associated with things like cell death and wounding and altered oxygenation state and reactive oxygen status,” he said. “So they suggest that the lungs of those individuals might have ongoing cell death and tissue repair processes, that it is an altered metabolic state.”

James Heath of the Institute for Systems Biology in Seattle focuses his research on identifying factors that would put patients at risk of long Covid, applying multi-omic tools to query blood samples, electronic health records, and questionnaires. His work has turned up clues to what higher levels of autoantibodies — an antibody that the immune system aims at the body’s own proteins — mean when they set off a cascade of immune responses. In the cohort he studied, the presence of autoantibodies wasn’t strongly associated with disease severity, but they were linked to gastrointestinal problems and changes in exercise capacity. Some specific autoantibodies activated a particular immune pathway, one that the monoclonal antibody narsoplimab targets. Heath’s team has designed and proposed to the NIH a clinical trial to test the drug to treat long Covid patients, especially those who can be identified through the specific autoantibodies he has identified.

Speaking of NIH, Janko Nikolich-Žugich of the University of Arizona College of Medicine – Tucson called for more funding of its Recover study, a national observational study of long Covid. The $1.2 billion that was allocated to it isn’t enough, he said. Recruitment is approaching 40% of its targets, which should be met by the January deadline, but “it’s actually falling quite short of everything that needs to be done,” he said. “Funding really needs to be tripled to extract the value of this study and to mobilize broadly the scientific community to participate in it in the best possible manner in a both scientific and medical sense.”

Brett Giroir, the four-star admiral formerly on the White House Covid task force, summed up research as it stands now. “We have mechanisms that are anywhere from persistence of SARS-CoV-2 to the activation of other viruses, to autoantibodies to distant inflammation in the brain, to profibrotic signals in the lung. And who the hell knows what’s going on in the cardiac system, as Harlan [Krumholz] said, because we really don’t have a clue,” he said. “We have a multiyear NIH study ongoing that could prove to be definitive. But what can we do in the short term? We can’t wait five to 10 years for the patients that Harlan has described who are in such suffering.”

Eric Rubin, an immunologist at Harvard Medical School and editor of the New England Journal of Medicine, pointed to the power of big data. “We have lots of patients and we have data,” he said. “I think we’re still looking for a collective sort of wisdom to bring to this question.” In an interview before the conference, Rubin told STAT “there are lots of different symptoms and we haven’t done a very good job of figuring out the vast majority of those,” he said. “What hasn’t worked so far, at least not in ways that I’ve seen it, is describing a syndrome or coming up with a list of criteria for a syndrome, which is how we ordinarily approach new diseases. This entity represents actually a lot of distinct entities. There’s no one long Covid.”

Paul Utz of Stanford University called for more and larger studies to explore autoimmunity and long Covid to understand who has or develops autoantibodies and how they might be contributing. He listed what we don’t yet know: the true prevalence of autoantibodies in Covid or in long Covid; whether it’s transient or permanent; and whether patients go on to develop autoimmunity. “Recover won’t answer this,” he said. Asked about the impact of vaccination on new autoimmunity, he said, “we don’t know if vaccination prevents it, but we speculate that it does.”

Meanwhile, Krumholz is urging people to pool whatever they know to keep patients from wandering in the wilderness of at-best partial solutions for their symptoms.

“It’s about this abyss of ignorance that’s pervading the entire field,” he said. “Most of our tests are insensitive to detect abnormalities, and yet we have people sitting in front of us who are not just lightly affected.”