The “other side of the coin” in Long COVID / ME-CFS / post-vaccinal syndromes: anti-M3 autoimmunity

[Manuel]

Much attention has focused on patients with HPA-axis dysfunction (cortisol).
But a large subgroup can look clinically identical — with a different mechanism:

Autoimmune dysautonomia driven by anti-M3 (muscarinic) autoantibodies

The M3 receptor regulates key parasympathetic functions:
• orthostatic regulation
• gland secretion (tears, saliva, sweat)
• GI motility
• ocular accommodation

When altered, patients may develop:

POTS, severe fatigue crashes, dry eyes/mouth, heat intolerance, slow digestion and visual fatigue.

Notably, visual symptoms are often not just “dry eye”: impaired ciliary muscle accommodation forces the eye to overwork.

Some patients improve with pyridostigmine (Mestinon), which enhances cholinergic signaling, and with volume expansion — treating physiology but not the immune driver.

-Immunogenetic context

Multiple chronic immune conditions share predisposition via ancestral HLA-II haplotypes:

DR2-DQ6, DR3-DQ2, DR4-DQ8

Immune reactivity against M3 presented through HLA-II has been described in small cohorts (including DR2-DQ6 carriers):
https://pmc.ncbi.nlm.nih.gov/articles/PMC7455086/

DR2-DQ6 is also linked to myasthenia gravis:
https://pmc.ncbi.nlm.nih.gov/articles/PMC3348874/

This supports a shared framework:
genetic predisposition + persistent trigger → loss of tolerance → different autoantibody targets → different phenotypes

Practical implication:
In dysautonomic Long COVID / ME-CFS, consider testing anti-M3 and myasthenia antibodies — mechanism matters.

Not one disease, but overlapping pathways.


Full explanation in the thread:

 

[Jonathan Edwards]

I don' think we have any reliable evidence supporting this proposal. Please post speculations as speculations, not as if they were medical facts.

 

[Manuel]

I don' think we have any reliable evidence supporting this proposal. Please post speculations as speculations, not as if they were medical facts.

I completely agree that we have to be careful not to present hypotheses as established medical facts.

That said, there are published studies showing GPCR autoantibodies, including muscarinic receptors such as M3, in cohorts of ME/CFS and Long COVID patients. The evidence is still emerging and not definitive, but this is not invented speculation.

There are also reports and small studies suggesting that pyridostigmine can improve dysautonomia and POTS symptoms, which is physiologically consistent with impaired cholinergic signaling.

I’m not claiming this applies to all patients, only that it may represent a biologically plausible subgroup worth investigating.

For transparency, I tested positive for anti-M3 myself, which is why I take this topic seriously, but my posts are based on published immunology, not only personal experience.

 

[Jonathan Edwards]

That said, there are published studies showing GPCR autoantibodies, including muscarinic receptors such as M3, in cohorts of ME/CFS and Long COVID patients. The evidence is still emerging and not definitive, but this is not invented speculation.

The data for ME/CFS are ten years old and I have scrutinised them in detail with the investigator. They provide reliable evidence of no difference compatible with a causal role. There is nothing to emerge. The evidence is clearly negative. These antibodies occur at the same range of levels in normal people.

 

[Jonathan Edwards]

There are also reports and small studies suggesting that pyridostigmine can improve dysautonomia and POTS symptoms, which is physiologically consistent with impaired cholinergic signaling.

A role for cholinergic signalling is plausible but pyridostigmine has been handed out off label for a long time now and if it actually worked usefully someone would have done a proper trial. The absence of proper trials for these things is a pretty good sign that nobody truly believes they do much more than provide a placebo effect.

The evidence is strongly negative.

 

[Manuel]

A role for cholinergic signalling is plausible but pyridostigmine has been handed out off label for a long time now and if it actually worked usefully someone would have done a proper trial. The absence of proper trials for these things is a pretty good sign that nobody truly believes they do much more than provide a placebo effect.

The evidence is strongly negative.

The improvement is usually modest, I agree and that actually makes physiological sense.

Pyridostigmine does not eliminate autoantibodies. If someone has anti-M3, the drug only increases available acetylcholine to partially compensate for impaired signaling. So the expected effect would be symptom reduction, not full resolution. It’s a functional workaround, not a cure.

The only way to truly remove the autoimmune driver in the future would be immune-targeted therapies (for example CAR-T–based approaches against autoreactive B cells), but we are clearly not there yet for these conditions.

And regarding trials, they are still being done. There is an ongoing study evaluating pyridostigmine in Long COVID/MECFS:
https://clinicaltrials.gov/study/NCT06366724

So I wouldn’t interpret the absence of large definitive trials yet as proof of no effect, more as a reflection of how underfunded and heterogeneous these patient populations still are.

 

[Jonathan Edwards]

The only way to truly remove the autoimmune driver in the future would be immune-targeted therapies (for example CAR-T–based approaches against autoreactive B cells), but we are clearly not there yet for these conditions.

But, as I said, we have clear evidence of no autoimmune driver. The antibody levels are all within the normal range.

 

[Jonathan Edwards]

So I wouldn’t interpret the absence of large definitive trials yet as proof of no effect, more as a reflection of how underfunded and heterogeneous these patient populations still are.

Not proof but strong evidence. The absence of trials has nothing to do with underfunding in my view. If someone really thought they had an effective drug they would get funding for a trial. It isn't that hard.

 

[Manuel]

But, as I said, we have clear evidence of no autoimmune driver. The antibody levels are all within the normal range.

I would respectfully disagree with such a categorical statement.

Saying there is “clear evidence of no autoimmune driver” assumes something that, in my view, the data do not support.

In our experience, working alongside clinicians, we repeatedly see patients with Long COVID and ME/CFS who develop autoimmune diseases. Some develop Hashimoto’s thyroiditis, others Addison’s disease, others anti-enterocyte antibodies, and we are now seeing a growing number with anti-M3 positivity. We have also seen cases with positive ANA, antiphospholipid antibodies, anti-β2 adrenergic and other GPCR autoantibodies, anti-TPO, anti-parietal cell antibodies, and in some cohorts markers consistent with small fiber neuropathy–associated autoimmunity.

Beyond individual clinical experience, published studies have reported increased prevalence of GPCR-targeting antibodies (including β-adrenergic and muscarinic receptors), ANA positivity, antiphospholipid antibodies, and other immune abnormalities in subsets of Long COVID and ME/CFS patients.

So I don’t think it is accurate to say there are no indications of autoimmune involvement in these conditions. What we likely have is heterogeneity, not absence.

 

[Jonathan Edwards]

In our experience, working alongside clinicians, we repeatedly see patients with Long COVID and ME/CFS who develop autoimmune diseases. Some develop Hashimoto’s thyroiditis, others Addison’s disease, others anti-enterocyte antibodies, and we are now seeing a growing number with anti-M3 positivity. We have also seen cases with positive ANA, antiphospholipid antibodies, anti-β2 adrenergic and other GPCR autoantibodies, anti-TPO, anti-parietal cell antibodies, and in some cohorts markers consistent with small fiber neuropathy–associated autoimmunity.

But you will see all that without Covid and the story simply does not hang togwther. A scientific story has to hang together in detail numerically and this doesn't.

 

[Manuel]

But you will see all that without Covid and the story simply does not hang togwther. A scientific story has to hang together in detail numerically and this doesn't.

I think there’s an important nuance here.
In Long COVID, more clearly than in pre-pandemic ME/CFS, we were able to observe something quite specific: previously healthy individuals, with no known autoimmune disease, developed persistent symptoms after SARS-CoV-2 infection, and measurable autoimmune phenomena appeared.

The scale and timing of the pandemic allowed this to be seen much more clearly and in a compressed timeframe than what historically happened with ME/CFS, where diagnoses were slower, more heterogeneous, and less systematically followed.

That doesn’t automatically prove causality in every case, but it does show that a defined infectious trigger can precede the emergence of chronic immune dysregulation and overt autoimmunity.

In the general population, of course autoimmune diseases also develop without SARS-CoV-2, usually after other environmental triggers in genetically susceptible individuals. That’s not controversial; it’s basic immunology.

But saying “these autoimmune findings also exist outside COVID” is not the same as saying they are irrelevant within it.

It would be like arguing that lupus is not autoimmune because multiple sclerosis also exists. Different autoimmune diseases share mechanisms, genetic predispositions, and immune pathways, but they manifest through different dominant autoantibody targets and tissue contexts.

Ultimately, many of these chronic inflammatory conditions may represent variations of autoimmune processes that we classify based on the predominant target or clinical expression. The challenge is not to deny heterogeneity, it is to identify which mechanisms apply to which subgroup.

 

[Jonathan Edwards]

A scientific story has to hang together in detail numerically and this doesn't.

As I said.

 

[Manuel]

As I said.

It seems we don’t agree on this point, and that’s okay.
In my view, there are sufficient data and clinical observations to support autoimmune involvement in at least a subset of these patients.
We may simply be interpreting the same heterogeneity differently.

 

[hotblack]

The challenge is not to deny heterogeneity, it is to identify which mechanisms apply to which subgroup.

So is your belief that there is nonsuch thing as ME/CFS but rather a collection of different conditions with different pathways requiring different treatments? If so can you say what conclusive evidence you have for this?

Surely it is important to challenge any hypothesis and it is upon those proposing to prove what they are putting forward stands up to scrutiny?

I hear heterogeneity talked about a lot but haven’t seen any proof. It often sounds like the parable of the blind men and the elephant.

 

[EndME]

In my view, there are sufficient data and clinical observations to support autoimmune involvement in at least a subset of these patients.

What subset? Even if you only consider those ME/CFS patients with elevated GPCR-aab then these elevations are not what you see in autoimmune diseases . These elevations of "subgroups" are within 2SD deviations from the mean and are seen in roughly equal proportions in healthy controls and nobody is suggesting "subgroups" of healthy controls.

Nobody is suggesting subgroups can't exist, they very possibly do. But the possibility of their existence cannot lead to the conclusion that they must be this-and-that without any meaningful data to support this.

 

[Manuel]

So is your belief that there is nonsuch thing as ME/CFS but rather a collection of different conditions with different pathways requiring different treatments? If so can you say what conclusive evidence you have for this?

Surely it is important to challenge any hypothesis and it is upon those proposing to prove what they are putting forward stands up to scrutiny?

I hear heterogeneity talked about a lot but haven’t seen any proof. It often sounds like the parable of the blind men and the elephant.

Yes, that is essentially my position.

What we currently call “ME/CFS” appears to include multiple biological subgroups that share a similar clinical phenotype (fatigue, PEM, dysautonomia, cognitive issues), but may arise from different underlying mechanisms.

And this is not theoretical, we see it reflected in patients’ laboratory data ordered by their own physicians.

For example:

• A subgroup with hypocortisolism / HPA-axis impairment
– some of them have anti-M3 positivity
– others do not

• Another subgroup with completely normal or even elevated HPA function
– but positive GPCR autoantibodies (including anti-M3 in some cases)

That alone already suggests biological heterogeneity under the same diagnostic label.

And it would not be surprising if additional subgroups exist depending on the type of immune dysregulation or autoimmunity involved.

This does not mean the syndrome “doesn’t exist.” It means the current clinical definition likely groups together overlapping but mechanistically distinct pathways. The challenge is to define them more precisely, and that requires ongoing biological characterization, not assuming uniformity from shared symptoms.

 

[EndME]

That alone already suggests biological heterogeneity under the same diagnostic label.

If two friends go on a fishing expedition and one catches a trout and the other a salmon that does not automatically mean the fishers belong to different subgroups of fishermen if on average you sometimes catch a salmon and sometimes a trout. Nobody has provided any data that things look different for GPCR-aabs in ME/CFS and everybody is just providing the data of the above fishing expedition, even if they might have problems in analysing this data.

 

[Manuel]

If two friends go on a fishing expedition and one catches a trout and the other a salmon that does not automatically mean the fishers belong to different subgroups of fishermen if on average you sometimes catch a salmon and sometimes a trout. Nobody has provided any data that things look different for GPCR-aabs in ME/CFS and everybody is just providing the data of the above fishing expedition, even if they might have problems in analysing this data.

I don’t think that analogy really fits the situation.

Catching different fish on different days is normal variation within the same underlying system.

Here we’re talking about something different: measurable physiological differences between patients.

If one group shows reduced cortisol production (whether due to autoimmune adrenal involvement, HPA-axis dysregulation, or direct tissue impairment) and another group shows completely normal or even elevated cortisol levels but has circulating GPCR autoantibodies associated with autoimmune signaling, those are not random variations of the same parameter.

They reflect different biological states:

• In one case, a hormonal axis is functionally impaired.
• In the other, the hormonal axis is intact, but there is evidence of immune-mediated receptor signaling disturbance.

That is not “trout vs salmon on different days.” That is a difference in the underlying physiology.

The question is not whether both groups share fatigue and PEM, they clearly do. The question is whether the mechanisms generating those symptoms are identical. And when we see consistent endocrine differences and immune-marker differences in real lab data, it is reasonable to consider that they are not.

That doesn’t invalidate the syndrome. It suggests it may be biologically heterogeneous, which is common in medicine.

 

[EndME]

I don’t think that analogy really fits the situation.

Catching different fish on different days is normal variation within the same underlying system.

Here we’re talking about something different: measurable physiological differences between patients.

If one group shows reduced cortisol production (whether due to autoimmune adrenal involvement, HPA-axis dysregulation, or direct tissue impairment) and another group shows completely normal or even elevated cortisol levels but has circulating GPCR autoantibodies associated with autoimmune signaling, those are not random variations of the same parameter.

I think it is very fitting because nobody has ever presented data that suggests that a subgroup in ME/CFS exists by presenting data that is inconsistent with the same phenomenon occuring in similar rates in healthy controls. There's lots of talk and no data to support it and when they present data, as in the case of GPCR-aab publications, the data presented just contradicts their point. The data precisely suggests that they are indeed "random variations of the same parameter", I'm not aware of a single study that suggests otherwise, if it would be any different ME/CFS would not be treated as it is today.

Here we’re talking about something different: measurable physiological differences between patients.

We are not. There's millions of physiological measurables differences between patients. There's subgroups of people with red hair, subgroups of people with brown eyes, subgroups of people that are tall, subgroups of people that are stretchy and subgroups of people called Tommy. That does not indicate anything that this would be pathologically relevant.

The problem is that these proportions of people exist to the same degree in healthy populations. Nothing suggests that their illness is due to marginally different levels in cortisol (which depend on an abudance of other factors) or a different factor.

I think we'll just have to disagree on this. If someone has data that suggest otherwise they may please present it.

 

[hotblack]

Nobody is suggesting subgroups can't exist, they very possibly do. But the possibility of their existence cannot lead to the conclusion that they must be this-and-that without any meaningful data to support this.

This is a good point. I’d like to see numbers on what percentage of people with ME/CFS are seen to have these other conditions.

We’ve fought for a long time to have ME/CFS recognised as a single condition and anyone proposing we dismantle that needs to be providing very conclusive evidence to do so. Especially as doing so muddies the waters a great deal and could set us back a long way.

I’m all for following the science and what the facts tell us, but I’ve yet to see anything that tells us there are conclusively many subgroups of conditions.

Catching different fish on different days is normal variation within the same underlying system.

Who is to say this isn’t the case in ME/CFS, we have no evidence to the contrary. It’s the case with many other conditions.

If you have evidence please show it so we can judge it. I’m not even seeing a cohesive narrative here just repetition of speculation. And as others here know I’m all for a bit of speculation but not for presenting it as anything but.