Preprint A Proposed Mechanism for ME/CFS Invoking Macrophage Fc-gamma-RI and Interferon Gamma, 2025, Edwards, Cambridge and Cliff

[forestglip]

I think it would be more productive to address inconsistencies or instances of alleged misunderstanding directly, point by point.

I looked at one claim in the response, high urinary creatine as objective evidence of muscle damage in ME/CFS.

Thread 'Urinary creatine'

Jun 14, 2025

I just wanted to look at at further evidence for one claim supporting biochemical pathology from the response from Margaret Williams to Jonathan Edwards' hypothesis paper:

I mention just one more illustration: in 2001 evidence was presented by SCM Richards et al (including psychiatrist Anthony Cleare, who works with and co-authors papers on ME/CFS with Simon Wessely) at the British Rheumatologists’ Conference in Edinburgh showing that 53% of ME/CFS patients were excreting in their urine significant levels of creatine and other muscle-related metabolites including choline and...

• forestglip

• 

She cites one unpublished datapoint. Chris Armstrong's team published a paper that seems to show the opposite, but maybe @MelbME can clarify. It doesn't seem worth going through every point in detail after looking at one that looks to me to use cherrypicked data.

I'd rather she provide one claim, the most convincing one, rather than a laundry list of claims that comes close to a gish gallop.

 

[Jonathan Edwards]

On one hand, I see accounts from reputable clinicians saying "there is no observable inflammation" (Edwards), and on the other hand, reputable clinicians saying "there is observable inflammation" (Klimas).

So what conclusion is to be drawn here for the non-clinician on the sidelines?

I don't think a non-clinician should be drawing any conclusions from what clinicians say. They need to be familiar with the facts as much as the clinicians - certainly the pathobiological facts.

And in our paper I summarize those facts - the appearance of the tissue and more specifically the imaging findings show no inflammation. I mention possible caveats and why I think they are insubstantial.

 

[wigglethemouse]

I think we have let a lot slide in this paper as it is a hypothesis paper. This seems to be the beans.

We suggest that in ME/CFS, antibodies with relatively low afnity for antigens available on an ongoing
basis contribute to disease indirectly by being taken up, together with such antigens, by high-afnity FcγRI on
gamma interferon-primed macrophages and thereby facilitating presentation of antigen to T cells

1. What evidence do we have for any "junk antibodies" in ME/CFS. I can only think of the Lipkin 1999 noting antibodies behave strange in ME/CFS. Perhaps his later B cell paper as well...

2. What evidence do we have for primed macrophages in ME/CFS? The closest we have is Hanson/Grimson writing in their immune cell transcriptomics paper that we need to study macrophages given the changes that team found in monocytes.

3. I have not seen data highlighting FcγRI receptors in ME/CFS. The papers I've seen in general seem to indicate you would see more of those receptors if they were being activated.

4. There could be an argument that there are moire activated T cells in ME/CFS but where is the evidence.

To me those are the things I would like evidence for in order to move forward with this hypothesis. The real question is how are we going to convince other researchers to start testing the hypothesis when there is so little evidence for it?

 

[Yann04]

1. What evidence do we have for any "junk antibodies" in ME/CFS. I can only think of the Lipkin 1999 noting antibodies behave strange in ME/CFS. Perhaps his later B cell paper as well...

As far as I understand this would basically be impossible to test for. There wouldn’t be one unform disease causing junk antigen, it would be a “personalised mix”, and it would be related to very slight shifts in plasma B-cell populations that wouldn’t look relevant statistically.

But I may have misunderstood the theory.

 

[wigglethemouse]

As far as I understand this would basically be impossible to test for.

I should have added that Jonathon hinted there would be a paper coming regarding antibodies. He also hinted we should reread the hypothesis when DecodeME is published. Perhaps there is something else.

I'm hoping when we get some additional evidence we will see some more meat in this hypothesis. I think the Hanson/Grimson lab were going to do a deeper dive on the immune cells of interest in their previous paper so perhaps something will appear there which will be helpful in the context of this paper. However, I think Cornell had their research funding frozen..........

 

[Eddie]

Yes, @EndME has it right. This isn't a matter of differences in language or interpretation. It is a matter of brute fact.

It isn't a matter of brute fact, the conclusion one draws depends entirely on how the evidence is intrepretated. Evidence is not some objective entity that points towards one true conclusion, evidence means something because a mind interprets it as changing a point of view. If these disagreements were matters of brute fact there would be no need for evidence and no need for debate.

Nancy Klimas doesn't even know what she means by inflammation I suspect. It is more like abracadabra or even biopsychosocial. That is the norm in biomedical research too. "All just words" as a fellow doctoral student once said to me.

Perhaps she doesn't have a specific definition, that what she means by inflammation is a cultural meme that refers to all sorts of processes. But as is the case with abracadabra and biopsychosocial, these words still mean something, even if what they mean varies more person to person than other terms. To the individual using the word it might not be clear what exactly it means. But that is all the more reason why the disagreement arises, because the terms are being used in different ways to express different and perhaps uncertain meanings. I would go even further and say that this is not just a problem with biomedical research but with all communication, a word or symbol or a gesture has as many meanings as there exists points of view. I think the only real solution is to try and help others to understand what your point of view is by explaining it in terms where agreement does exist.

 

[DMissa]

I am not quite sure what you are recommending @DMissa . Over the years we have been through the details of methodological and interpretation problems here. Margaret is suggesting that I am dismissing things without having given them consideration and she has now been doing this for ten years despite my explaining the details on the public forums - and many others pointing out flaws I have considered. All she ever does is appeal to the authority of professors, while at the same time casting aspersions on the professors she doesn't agree with.

In other diseases there are solid facts about biology relevant the process. In MECFS I don't see anything of that sort. If there was it would be used clinically by physicians who aren't attached to a fringe private clinic.

Is there anything Margaret mentions that you would consider established biology for MECFS? Anything claimed by the people she quotes you would like to defend?

Well for example what I asked about was the inflammation/Klimas issue and the response which boiled down to "the other person doesn't understand this" didn't give me much to go on.

edit: I now see another more recent reply about what was in the paper:

And in our paper I summarize those facts - the appearance of the tissue and more specifically the imaging findings show no inflammation. I mention possible caveats and why I think they are insubstantial.

This would have been a good way to avoid this. I guess the burden then falls on the other party to provide the evidence to the contrary. This was a more useful reply.

Is there anything Margaret mentions that you would consider established biology for MECFS? Anything claimed by the people she quotes you would like to defend?

I was referring more to what you'd commented about Klimas, specifically.

I don't think a non-clinician should be drawing any conclusions from what clinicians say. They need to be familiar with the facts as much as the clinicians - certainly the pathobiological facts.

Discerning what is fact without depth of experience in a particular subject is a challenge that provides more than reasonable grounds to seek advice from those with that experience. Is this not one major benefit of collaboration? Being aware of one's own, current, sphere of confident interpretation is important - as is seeking discussion to expand it.

If I were to hand you some respirometry and extracellular acidification traces, plus some background reading, I could not expect you to apply anywhere close to the depth of understanding and accuracy of interpretation that I have developed by years of experience. And it would make perfect sense for you to seek my advice, particularly if there was something you did not feel confident in drawing conclusions about.

I am not seeking discussion in lieu of assessing the facts for myself - I am seeking it to provide context to help myself better discern what is or isn't fact. It is not a replacement, it is a supplement. And just like any other bit of information it would be measured and scrutinised as best can be done, and not taken on trust alone.

 

[wigglethemouse]

I think a clinician's view of ME/CFS is much broader than a researchers. Clinicians treat co-existing illnesses, some of which may have inflammation as defined by Jonathon. I think it's pretty clear by now that people throw the term "inflammation" about a lot but everyone seems to have a different interpretation of what that means.

The paper is very clear by what they term as inflammation for the purpose of this paper. I get redness and swelling as a comorbidity as defined in the paper for inflammation but I don't see this in the majority so I'm okay with what he has written vs what a clinician may say.

ME/CFS does not involve inflammation as such. No tissues show swelling, heat, or redness, even if there is pain, and magnetic resonance imaging has not shown inflammatory oedema.

A researcher is trying to find what is common and not common to a percentage of patients and present a hypothesis for that group. In this thread Jonathon and collaborators present a hypothesis that could fit the common clinical picture based on how the disease presents for the majority of ME/CFS patients.

I'd like to see more debate, reviews, and feedback on the Synthesis section of the paper to help determine if it has "legs" or not. With those contributions the hypothesis can be refined, adjusted, tested, or torn up for something new.

 

[hotblack]

I am not quite sure what you are recommending @DMissa . Over the years we have been through the details of methodological and interpretation problems here

Should we consider the audience that isn’t deep into this? I wonder what someone unversed in all the history would make of this and if they’d grasp the point about things having been found but there being no reliable evidence for causal mechanisms? I suppose it depends on the target audience and aim of the paper, is the aim to get current researchers to think differently or new researchers to do so?

 

[Jonathan Edwards]

I think we have let a lot slide in this paper as it is a hypothesis paper. This seems to be the beans.


1. What evidence do we have for any "junk antibodies" in ME/CFS. I can only think of the Lipkin 1999 noting antibodies behave strange in ME/CFS. Perhaps his later B cell paper as well...

2. What evidence do we have for primed macrophages in ME/CFS? The closest we have is Hanson/Grimson writing in their immune cell transcriptomics paper that we need to study macrophages given the changes that team found in monocytes.

You have some reasonable questions there @wigglethemouse.

1. All antibodies are in the right contex antibodies to junk. As Ivan Roitt used to say, if you have 3gm of antibody in a pint of blood you have 3gm of anti- streptococcus antibody. This is hyperbole but the point is that every antibody species will bind with some minimal affinity to every antigen. Biology sets a functional cut off of a dissociation constant maybe at 10^ -8 most of the time. In an ELISA you set it with your concentrations and washing times.

BUT thermodynamics implies that the relevant Kd will change if antibody is rebound to something like a receptor (or in an ELISA plastic). The idea is that we all have antibodies that are well chosen for specificity to pathogens in most contexts but that in the context of FcgR1 this is less precise. Similar arguments apply even more so to T cell receptors. You can get any T cell to respond to any antigen if you ramp up secondary signals.

We are suggesting that the antibody issue may be more serious for women as part of a less precise vetting of B cells - which is suggested by their higher rate of frankly autoimmunity.

To test for women having more 'junk antibody' activity you might need a special assay using FcR1 but if it were just a shift in Kd you could probably do it with standard ELISA technology.

That would be a brilliant test for the theory! Somebody should go off and do it at once!

But it would be no good looking for more junk antibody in pwMECFS because the theory does not predict that. In the theory the interaction in the context of FcR1 is purely permissive and normal. The idea is that you need an expanded population of overresponsive T cells to turn what normally an innocent noise factor into pathology.

2 & 3. Similarly, the theory does not predict you will find more primed macrophages with FcR1 in MECFS except where they are hidden away in microcompartments interacting with T cells. That could be spleen or Peyers patches maybe.

4. Is much the same.

I guess the case is built on the argument that the epidemiology indicates something like this must be going on but hidden from all current testing methods. We are deliberately building a theory that would explain why nothing is showing up yet.

There is more to come but I need some breakfast.

But the idea of testing female sera for more low affinity activity seems to me a very nice experiment if one could persuade someone to fund it. Maybe Audrey could do it - she has already shown some odd things about MECFS Igs.

 

[Sasha]

Should we consider the audience that isn’t deep into this? I wonder what someone unversed in all the history would make of this and if they’d grasp the point about things having been found but there being no reliable evidence for causal mechanisms? I suppose it depends on the target audience and aim of the paper, is the aim to get current researchers to think differently or new researchers to do so?

Actually, answering Margaret's question in some depth might be helpful as an explainer for the wider audience of PwME as to why our vast literature of 9,000 papers that's always getting cited as proof that ME/CFS is biological isn't very useful.

Someone said upthread that her big list of papers looked close to a Gish gallop but it seems entirely reasonable to me that if she thinks there's a ton of strong evidence, she would want to show its extent. I'm not suggesting responding in depth to each citation but maybe giving the general principles for assessing the literature.

 

[Trish]

Actually, answering Margaret's question in some depth might be helpful as an explainer for the wider audience of PwME as to why our vast literature of 9,000 papers that's always getting cited as proof that ME/CFS is biological isn't very useful.

Someone said upthread that her big list of papers looked close to a Gish gallop but it seems entirely reasonable to me that if she thinks there's a ton of strong evidence, she would want to show its extent. I'm not suggesting responding in depth to each citation but maybe giving the general principles for assessing the literature.

Surely that's what this whole forum is about. We take research papers one by one and look at the findings as reported, and factors that might make those findings signs of genuinely useful difference in biology for ME/CFS and which may be artefacts from, for example, too small samples, poor matching of pwME and controls, lack of correction for multiple comparisons and other statistical weaknesses, overinterpretation of differences when there is a large overlap between groups, and so on. And in many cases, lack of replication by another lab with another set of patients.

 

[Sasha]

Surely that's what this whole forum is about. We take research papers one by one and look at the findings as reported, and factors that might make those findings signs of genuinely useful difference in biology for ME/CFS and which may be artefacts from, for example, too small samples, poor matching of pwME and controls, lack of correction for multiple comparisons and other statistical weaknesses, overinterpretation of differences when there is a large overlap between groups, and so on. And in many cases, lack of replication by another lab with another set of patients.

Yes it is, but have you seen the list of studies that Margaret presents? It's far to many to critique individually, so spelling out the principles as you have cold be a good way to make a response manageable and yet informative - maybe digging into the first one or two to give an example of specific critique.

 

[Sean]

I would love to know who Margeret really is.

IIRC, it is more than one person, and the name is just a nom de plume.

 

[Peter T]

IIRC, it is more than one person, and the name is just a nom de plume.

‘Margaret Williams’ has done tremendous work collecting information in general and also specificity references for biomedical research, for which she/they deserve credit. (I had not realised she was not just one person.)

However, just as with the BPS coterie’s work, press of information alone is not sufficient to answer questions on aetiology. Quality of this information needs a critical evaluation. However, I suspect we are now getting to the point where one person alone can not produce a meaningful synthesis of all these studies. I had previously fantasised that my super lottery win, if it ever happened, could pay for a serious medical text book and related professorial chair giving an overview on the biomedical evidence relating to ME/CFS. (I even had been undertaking ‘brain retraining’ aimed at becoming supper rich).

I wonder now that, unless a breakthrough study renders all this redundant, that the best way forward is publication of such as the current ‘Proposed Mechanism’. Such models are ultimately likely to be ‘wrong’, as seen with Newtonian mechanics, however hopefully the lengthy debates they provoke that prove them ‘wrong’ give a us hook to hang the existing research on and to promote further research asking better questions.

 

[Sasha]

Apologies if I've posted this idea before - not sure if it ever made it out of my head - but I'm wondering if a natural experiment has already been done regarding some of the proposed treatments in this paper, that would help test its validity.

Fluge and Mella tested rituximab for ME/CFS because a couple (?) of ME/CFS patients who got dosed with rituximab for their cancer showed marked improvement in their ME/CFS. I'm wondering if PwME are similarly getting dosed for other diseases that they have, with some of the treatments proposed here.

If they are, do we have any means of finding out about their outcomes? For example, who gets bortezomib? Can we contact clinicians who treat those patients and ask what happened to any patients who also have ME/CFS and got bortezomib? Can we do some kind of fancy AI-assisted search of social media to tot up PwME claiming to have got better when treated with bortezomib? In both cases, can we do some sort of comparison search for drugs we think wouldn't work, as a kind of control condition (I know that would be a big mess, potentially, but am wondering if the big mess could be corrected for)?

 

[V.R.T.]

Margaret Williams’ has done tremendous work collecting information in general and also specificity references for biomedical research, for which she/they deserve credit.

The issue is that the recent piece on Simon Wessely was front loaded with so many poorly evidenced claims of irrefutable biological evidence that left the following discussion (which to be honest I only skimmed but it looked much better) of Wessely's many misdeeds vunerable to easy attack from psychobehavioral allies who can just say look at all of this nonsense they believe.

It delegitimised an imo very necessary expose on someone who has caused deep harm to this community.

So yes, hopefully we have some research evidence coming that will get us back on the same page somewhat. Because it is so frustrating to watch well meaning people invalidate their own arguments in that way.

 

[V.R.T.]

Apologies if I've posted this idea before - not sure if it ever made it out of my head - but I'm wondering if a natural experiment has already been done regarding some of the proposed treatments in this paper, that would help test its validity.

Fluge and Mella tested rituximab for ME/CFS because a couple (?) of ME/CFS patients who got dosed with rituximab for their cancer showed marked improvement in their ME/CFS. I'm wondering if PwME are similarly getting dosed for other diseases that they have, with some of the treatments proposed here.

If they are, do we have any means of finding out about their outcomes? For example, who gets bortezomib? Can we contact clinicians who treat those patients and ask what happened to any patients who also have ME/CFS and got bortezomib? Can we do some kind of fancy AI-assisted search of social media to tot up PwME claiming to have got better when treated with bortezomib? In both cases, can we do some sort of comparison search for drugs we think wouldn't work, as a kind of control condition (I know that would be a big mess, potentially, but am wondering if the big mess could be corrected for)?

I know Ron Davis' team were doing this kind of work with medical records regarding potential drugs.

 

[Utsikt]

It isn't a matter of brute fact, the conclusion one draws depends entirely on how the evidence is intrepretated. Evidence is not some objective entity that points towards one true conclusion, evidence means something because a mind interprets it as changing a point of view. If these disagreements were matters of brute fact there would be no need for evidence and no need for debate.

My (limited) understanding is that the majority of the uncertainty in science lies in how accurately the evidence is able to capture the underlying reality, and not in how the evidence is interpreted.

So the caveat is mostly «if A is correct, B must be true», and B is just given by logic.

The problem we’re discussing here is that for the studies mentioned by MW, if A is true, any one of B to Z could be true. Yet MW claims that A definitely means that G is true, and ignores the various other meanings that are equally supported or made possible by evidence A.

There is also the issue of how we might not be able to imagine all of the possible explanations for A, and therefore wrongly believe that A supports B, and only B. Essentially an issue of unknown unknowns.

As for your last point - you can’t assume that all people are perfectly rational. And you’d still need evidence to establish something as a fact, regardless, otherwise it would be an opinion.

 

[wigglethemouse]

To test for women having more 'junk antibody' activity you might need a special assay using FcR1 but if it were just a shift in Kd you could probably do it with standard ELISA technology.

But the idea of testing female sera for more low affinity activity seems to me a very nice experiment if one could persuade someone to fund it. Maybe Audrey could do it - she has already shown some odd things about MECFS Igs.

Thank you for taking the time to respond in such detail.

In 2020 I took part in a very small unpublished pilot study that found "low level" connective tissue disease antibodies (sjogrens, lupus, dermatomyositis etc) vs healthy controls using a modern ELISA like array in a research lab. I think the results changed with longitudinal sampling, and did not hold up when tested with standard clinical test for autoimmunity or gold standard wet lab for most participants. Is this the kind of result you would expect to see?

That researcher moved on and I recently heard that another researcher did another similar pilot project with the same research test lab. I think on the whole both researchers did not know what to make of the results except to confirm that it might fit the picture of non-specific B cell clonal activation. Would that description seem relevant to your hypothesis?