Long COVID Atlas
Text Theme
Note

This is education, not medical advice. Tests and treatments belong in a conversation with your own clinician, who knows your full picture.

Mechanism · Immune dysregulation

Immune dysregulation & autoimmunity (pathogenic IgG)

After most infections the immune system finishes the job and stands down. In long COVID it often does not. Months later, signs of an immune system still running hot keep turning up, and in some people the antibodies meant to fight the virus appear to turn on the body's own tissue instead. That second idea, autoimmunity, is the one with the most direct evidence behind it: antibodies taken from patients can carry symptoms into a healthy animal. It is a leading mechanism, well supported for a subgroup, and not yet proven to drive every case.

Short version: the immune system stays switched on long after the virus is gone, and in some people the body's own antibodies start causing harm. Antibodies from patients can transfer symptoms to mice, which is strong evidence. No immune treatment is approved yet.

Start here: an alarm that never resets

Autoantibodies bind receptors, but also occur in healthcell membraneGPCRfunctional autoantibody (agonist)also found in some healthy people
Antibodies normally lock onto the virus. Some long COVID antibodies instead lock onto the body's own cell-surface receptors.

Your immune system runs like a building alarm. An intruder trips it, the response kicks in, the threat is cleared, and the alarm resets. In long COVID the reset step seems to fail. The threat is gone, or nearly gone, yet the alarm keeps ringing, and the cost of leaving it ringing shows up as fatigue, brain fog, racing heart, and pain across many systems at once.

There are two related problems here. One is plain over-activation: immune signaling that stays high for months when it should have settled. The other is autoimmunity, where the antibodies built to fight the virus start recognizing the body's own parts. The two feed each other, and telling them apart in any one person is still hard.

Antibodies that target your own receptors

The cells that control heart rate, blood vessel tone, and gut movement carry docking points on their surface called G-protein coupled receptors. In a share of people with long COVID, antibodies appear that bind these receptors and either block them or jam them switched on.1

Larger studies have found patterns of these receptor-targeting and blood-pressure-pathway antibodies that line up with how sick people are and with which symptoms they have.2 The honest caveat is that some of the same antibodies show up in healthy people too, so their presence alone does not prove they are causing the illness. The link is real but not settled.

moderatealso seen in some healthy controlsassociation, causation unproven

The transfer experiment: moving from link to cause

Patient antibodies make a mouse illlong COVID patientIgGmousefatigue, pain,imbalance, nerve damagetransferring the antibody transfers the illness (in a subgroup)
Purified antibodies from patients, injected into healthy mice, can reproduce long COVID-like signs. This is what moves autoimmunity from correlation toward cause.

The most convincing evidence that these antibodies do real harm comes from transfer experiments. Researchers purified the antibody fraction, the IgG, from people with long COVID and injected it into healthy mice. The mice then developed measurable problems that mirror patient symptoms, including heightened pain sensitivity and changes in movement and heart-rate control.3

This design is powerful because it reverses the usual direction of guesswork. Instead of finding antibodies in sick people and wondering whether they matter, it takes the antibodies out, puts them into a healthy body, and watches the illness follow. The effect appears to hold for a subgroup rather than everyone, which fits the picture of long COVID as several overlapping diseases rather than one.

IgG transfer reproduces symptomswell-founded for a subsetsubgroup-dependent

A nerve-injury subgroup

In some patients the antibody attack appears aimed at the nervous system. Studies have tied specific autoantibodies to neurological symptoms, and a subgroup shows blood markers of nerve and support-cell injury, the kind released when brain tissue is stressed.4

This matters because it offers a concrete route from an immune problem to a brain symptom like fog or numbness, rather than leaving the connection vague. It also suggests that the same disease label may need to be split by mechanism, since an antibody driving someone's nerve pain is not necessarily the one driving someone else's racing heart.

low to moderateneuro-injury IgG subgroup

What keeps the system switched on

Even without autoimmunity, the immune signal in long COVID stays elevated for an unusually long time. Detailed tracking has shown disturbed immune patterns persisting eight months after the first infection, well past the point where a normal response should have quieted.5

Two things plausibly keep the alarm ringing. Leftover viral material sitting in tissue gives the immune system something to keep reacting to, a slow drip of provocation. And a leaky gut lining lets bacterial fragments cross into the blood, adding a second source of low-grade inflammation. Both supply the constant irritation that an immune system stuck in the on position needs.

moderatesustained activation documented to 8 months

What it drives downstream

If the immune problem is upstream, its fingerprints should show up in the organs that misbehave in long COVID, and they do. The receptor-targeting antibodies plausibly feed the autonomic problems behind POTS, the nerve-injury subgroup tracks with brain fog, and the same inflammation appears to damage the thin lining of blood vessels, the endothelium.

Sustained immune activation can also wake up old viruses that the body normally keeps dormant, such as Epstein-Barr virus. Whether that reactivation is a cause of ongoing symptoms or just a downstream sign of a dysregulated immune system is still debated.

moderateEBV role debated

How it is measured, and what is missing

There is no single blood test that confirms immune-driven long COVID. Research panels can measure interferon signaling, T-cell patterns, and specific autoantibodies, but these stay mostly in studies because no panel is validated for diagnosis or tied to a treatment decision. A normal routine immune workup does not rule the mechanism out.

That gap is the practical problem. The transfer experiments make a strong case that antibodies cause harm in some people, which is exactly the situation where removing or blocking antibodies might help. Trials of immune-targeting treatments are the logical next step, and several are underway, but none has produced an approved therapy yet.

What we don't know

Honest about the edges of the evidence. These are open questions, not settled answers.

  • We do not know what fraction of long COVID is driven by autoimmunity versus persistent inflammation alone.
  • We cannot yet identify, in advance, which patients carry the harmful antibodies that transfer disease in animals.
  • We do not know whether the receptor-targeting antibodies start the illness or appear after other damage is done.
  • We do not know why the same autoantibodies can be present in healthy people without causing symptoms.
  • We do not know whether removing or blocking these antibodies reverses symptoms in people, only that they cause signs in mice.
  • We do not know whether EBV reactivation is a driver of ongoing symptoms or a marker of immune dysregulation.
  • We lack a validated blood test that ties a specific immune pattern to a specific treatment.

What this means for you

If your long COVID looks immune-driven, with widespread symptoms that flare and never fully settle, the useful thing to know is that this is one of the better-evidenced mechanisms, not a fringe idea. Antibodies taken from patients can carry symptoms into a healthy animal, which is about as direct as mechanism evidence gets short of a human trial. That does not mean a test can confirm it in you today, because the research panels are not yet diagnostic tools.

The practical implication is patience with the science and caution with the marketplace. Because no immune treatment is approved, any clinic selling antibody-clearing procedures or strong immune drugs for long COVID is selling ahead of the evidence. The reasonable path is to treat the downstream problems that have real management, the autonomic and sleep and pacing pieces, and to watch the immune-targeting trials, ideally by asking whether you can join one.

References

Each reference links to the source on PubMed, PMC, or the publisher.

  1. Functional autoantibodies against G-protein coupled receptors in patients with persistent post-COVID symptoms. Journal of Translational Autoimmunity, 2021.
  2. Autoantibodies targeting GPCRs and renin-angiotensin system molecules associate with symptom severity after COVID-19. Journal of Translational Medicine, 2022.
  3. Transfer of IgG from people with long COVID reproduces long COVID-like signs in mice. 2025.
  4. A causal link between autoantibodies and neurological symptoms in long COVID. 2025.
  5. Immunological dysfunction persists for eight months following initial mild SARS-CoV-2 infection. Nature Immunology, 2022.

Associated topics

MechanismViral persistence / SARS-CoV-2 reservoirsupplies the chronic provocationMechanismDysautonomia / autonomic dysfunctiondownstream of receptor antibodiesMechanismEndothelial dysfunctionvascular lining injured by inflammationSymptomBrain fog / cognitive impairmentneuro-injury IgG subgroupSymptomOrthostatic tachycardiaautonomic antibodies implicated