Long COVID linked to exhausted, hyperactive immune cells targeting multiple viruses

Long COVID remains one of medicine's most stubborn puzzles. Patients report extreme fatigue, racing heartbeats, shortness of breath, brain fog, aching joints—symptoms that shift unpredictably and resist easy diagnosis. Some people have had their entire lives upended by it, forced to leave their jobs and abandon the routines they knew before the pandemic. The World Health Organization estimates that roughly one in ten COVID-19 infections result in long COVID, with symptoms persisting for months or years. Yet doctors have no standard treatment to offer. Existing antivirals have largely failed. Anti-inflammatory drugs might help some patients, but there is no proven therapy that works reliably for anyone.

The reason is biological complexity. Long COVID does not appear to be a single disease with a single cause. Instead, it seems to emerge from multiple mechanisms that vary considerably from one patient to the next. Understanding what drives it requires looking closely at the immune system itself—and that is where researchers at Gladstone Institutes have begun making headway. In a study published in Cell Reports Medicine in late July 2026, scientists led by Nadia Roan, a senior investigator at Gladstone and professor at UC San Francisco, identified distinctive changes in immune cells that may help explain why some people's bodies cannot seem to recover from COVID-19.

The focus was on CD8 T cells, specialized immune soldiers designed to detect and destroy cells infected with specific viruses. Each CD8 T cell is exquisitely precise—a cell trained to recognize the flu virus will target flu-infected cells but leave alone cells infected with a cold virus. Some CD8 T cells are trained to recognize SARS-CoV-2. But most prior research on long COVID had treated these cells as a general category, missing the nuance. Roan's team wanted to study virus-specific CD8 T cells in detail, and they had the tools to do it. A collaborator, Evan Newell of the Fred Hutchinson Cancer Center, had developed combinatorial tetramer technology that detects virus-specific CD8 T cells in blood without disturbing them, keeping them as close as possible to their natural state. The researchers combined this with CyTOF, an advanced technique that measures 29 different proteins on each cell—essentially creating a 29-dimensional portrait of cellular function.

The team also had access to an unusually clean dataset: blood samples from long COVID patients and matched controls who had been infected only once, before vaccines were available, limiting the variables that could complicate the data. They focused on patients with particularly severe long COVID—those experiencing many symptoms—drawn from the LIINC study, a long-term investigation of COVID-19's lasting effects led by UCSF. Notably, the researchers did not stop at SARS-CoV-2-specific cells. They also examined CD8 T cells trained against two herpesviruses: Epstein-Barr virus and cytomegalovirus. There is growing evidence that these viruses, which most people carry dormant in their bodies, may play a role in long COVID. People with long COVID have significantly elevated antibodies against Epstein-Barr virus, suggesting the virus may be reactivating or being targeted more aggressively.

The findings were striking. The three types of virus-specific CD8 T cells—those targeting SARS-CoV-2, Epstein-Barr virus, and cytomegalovirus—shared certain abnormal features in people with long COVID that were absent in those who had recovered fully. These cells showed signs of exhaustion, as if they had been worn down by repeatedly fighting the viruses they were designed to combat. But more intriguingly, all three types also appeared hyperactive in a specific way: they had heightened cytolytic capacity, meaning they could destroy infected cells with unusual intensity. During active COVID-19, patients do have many cytolytic CD8 T cells, but these levels typically decline after recovery. In long COVID patients, the levels remained elevated. Whether this persistent hyperactivity drives long COVID or results from it remains unclear—a chicken-and-egg question that future research will need to untangle.

One finding stood out with particular force: the enhanced cytolytic features were especially pronounced in women with long COVID compared to men. Long COVID itself is far more common in women, and this cellular evidence suggests the disease may operate differently across sexes. This underscores how critical it is to include women in disease research and to consider sex-specific mechanisms when designing treatments. The most profound changes, surprisingly, appeared not in SARS-CoV-2-specific cells but in cytomegalovirus-specific cells, adding weight to the hypothesis that herpesviruses may be key players in long COVID's persistence.

Roan and her colleagues are now moving forward on multiple fronts. They have developed a mouse model of long COVID that may help answer causal questions and allow rapid testing of potential therapies. They are expanding beyond blood samples to study tissue samples, where immune activation actually occurs. Some of this work will focus on the female reproductive tract, supported by a new grant to investigate sex-specific differences in long COVID, including hormonal and genetic factors. The immune changes they identified are not necessarily permanent—many people with long COVID do improve over time—and the question now is whether therapy can accelerate that recovery. These discoveries will not immediately yield a single cure, but they are beginning to map the terrain of a disease that has confounded medicine for years.