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

[Hutan]

We have moved some posts about daratumumab and the drug development process to the Daratumumab trial thread.

 

[Robert 1973]

The first reviewer, Brett A. Lidbury, writes:

My only comments are to encourage clarification on the comment, “Furthermore, the symptoms of ME/CFS overlap with those of viral infection. As a result, there is not the stereotyped history of post-infective illness seen, for instance, in Reiter’s syndrome, in which new symptoms appear several days after the acquisition of infection”, under the “Relation to Intracellular Infection” sub-heading. I am not sure that I agree with this position and would quote recent Long COVID examples where 12 weeks are needed to see the appearance of idiopathic, long-term fatigue, PEM, and associated symptoms (not including the immediate post-infection period where the impacts of acute inflammation may be observed via cardiac, lung, or other organ sequelae).

Jonathan replies:

I am not sure exactly what the concern is about the evolution of symptoms. We were making the point that many features of ME/CFS, including fatigue, malaise, orthostatic intolerance, nausea and sensitivity to light and sound, are also features of acute viral infections such as EBV and Covid-19. This makes it more difficult to clearly separate the acute viral illness from the subsequent long term illness. The requirement of several weeks observation before a diagnosis of ME/CFS is made reflects this overlap. The PEM pattern of exacerbation after exertion may be hard to tease out early on when malaise and fatigue can vary, perhaps with periods of viraemia, and can also appear to follow exertion, although that is hard to pin down. Unlike Reiter's syndrome there is no very clear distinction between the initial viral and later ME/CFS symptoms themselves.

My experience was that some of my ME/CFS symptoms came on quite suddenly and were very different to the flu-like symptoms I had from the infection, or any symptoms I had had before. Specifically, I thought I had developed an allergy to something(s).

As I’ve written before, for a long time I struggled to accept my ME/CFS diagnosis. One of the things that persuaded me that the diagnosis was correct was reading an account by a physician (can’t remember who) who wrote that ME/CFS patients often present by saying that they think they have developed an allergy to something that is making them feel unwell.

I would be interested to know if this type of complaint might be a way of predicting whether people are likely to have unresolving ME/CFS rather than resolving PVFS. I’m not suggesting that everyone with ME/CFS has this type of symptom, but I wonder if having it may mean you are less likely to recover. I also wonder if the same may apply to alcohol intolerance and other distinctive symptoms.


Writing the above returns me to the discussion we had earlier in the thread which I wasn’t able to continue due to my limited capacity:

This is the fallacy of assuming that a diagnosis is a diachronic concept despite being made at time X.
I realise that this is very counterintuitive and inconvenient but those are the sorts of things I tend to turn my attention to!! If we talk of 'different illness' we have to be clear whether we are talking about cause or effect and whether it is now or how things turn out.

Diagnostic criteria are a mirage because they take none of this into account.

I think I understand what you’re saying here but there is a difference between a diagnosis and a definition of a disease for the purposes of a hypothesis.

Your hypothesis paper and your Concept of ME/CFS paper both reference the CCC and the 2021 NICE guideline. If you’re defining ME/CFS in accordance with CCC or NICE, then the statement that “Adult-onset illness mostly never fully resolves” appears to be misleading – at least to me and to others on here who probably know considerably more about the characteristics of ME/CFS and the epidemiological data that many of scientists who we hope will be reading your paper. If, on the other hand, for the purpose of your hypothesis you are defining ME/CFS as an illness that mostly never fully resolves in cases of adult onset (which would be a valid definition), then I think you need to make that clear.

You said before that you may re-word the paragraph in question, and I agree that that would be worthwhile, if you’ve not already done so.

I’m sorry to bang on about this but I think it’s important. I also apologise if I’m repeating myself or anything anyone else has said – I’ve not managed to keep up with this thread and I’m running on very limited capacity.

 

[Jonathan Edwards]

If you’re defining ME/CFS in accordance with CCC or NICE, then the statement that “Adult-onset illness mostly never fully resolves” appears to be misleading

I don't really follow that. Most people who study the illness appear to agree that most people who fit CCC have an illness that never fully resolves. It isn't part of CCC but then it shouldn't be, because CCC is designed to identify people whose illness has a particular pattern now. NICE doesn't really have a definition. It gives features that indicate that you should consider ME/CFS. (Which in this context isn't very helpful.)

This is basically the diachronic/synchronic point.

 

[Hutan]

Your hypothesis paper and your Concept of ME/CFS paper both reference the CCC and the 2021 NICE guideline. If you’re defining ME/CFS in accordance with CCC or NICE, then the statement that “Adult-onset illness mostly never fully resolves” appears to be misleading – at least to me and to others on here who probably know considerably more about the characteristics of ME/CFS and the epidemiological data that many of scientists who we hope will be reading your paper. If, on the other hand, for the purpose of your hypothesis you are defining ME/CFS as an illness that mostly never fully resolves in cases of adult onset (which would be a valid definition), then I think you need to make that clear.

You said before that you may re-word the paragraph in question, and I agree that that would be worthwhile, if you’ve not already done so.

I agree. I think precision around this issue is important in ME/CFS papers. It is useful if authors make a clear definition of the disease they believe they are talking about.

Most people who study the illness appear to agree that most people who fit CCC have an illness that never fully resolves. It isn't part of CCC but then it shouldn't be, because CCC is designed to identify people whose illness has a particular pattern now.

I don't think we have solid evidence that most people who fit the CCC at six months have an illness that never fully resolves. From my reading of papers and accounts and the experience of my own family, I don't actually believe that that is true.

In fact, given the large numbers of people who probably meet modern definitions of ME/CFS and who do recover in the first 12 months, I actually think it is quite likely that most people who fit such definitions at any one time do recover.

 

[Robert 1973]

I agree with @Hutan.

[Edit: From earlier in this thread:]

I think the idea that ME/CFS is a disease that does not fade away in a [monophasic] fashion over a period of months after an infection is widely accepted. A number of research studies require subjects to have had the illness for at least two years in recognition of the fact that monophonic resolving PVF can take nearly that long.

But CCC only requires symptoms for 6 months for adults, and as far as I’m aware it makes no reference to monophasic improvement, which I suspect may be highly relevant in predicting long-term outcomes – although I’m not aware of any data on that.

 

[Utsikt]

I don't think we have solid evidence that most people who fit the CCC at six months have an illness that never fully resolves. From my reading of papers and accounts and the experience of my own family, I don't actually believe that that is true.

In fact, given the large numbers of people who probably meet modern definitions of ME/CFS and who do recover in the first 12 months, I actually think it is quite likely that most people who fit such definitions at any one time do recover.

I’ve had different experiences.

My only perspective on this is a couple of years in LC support groups. In those, most people had no idea about what PEM was. I did neither. Most took it to mean PESE, PEF rapid fatiguability, and many talked of any immediate symptoms as PEM. I’ve seen people say that they get «PEM in their arms» from holding a book.

Most of the people that recovered did not retrospectively describe delayed reactions to exertion, only prolonged and disproportionate. It was «I feel terrible after doing this», not «something hit me like a train the day after doing this».

In short, most of the recoveries I’ve observed by pwLC were people that experienced similar symptoms as me, similar disability as me, but dissimilar patterns of symptoms in relation to exertion. They did not get the «living as if on a credit card» analogy, only «things have gotten more expensive». But they still said they had PEM.

There are obviously a ton of sources of bias with my own observations and memory here.

 

[horton6]

Most of the people that recovered did not retrospectively describe delayed reactions to exertion, only prolonged and disproportionate. It was «I feel terrible after doing this», not «something hit me like a train the day after doing this».

It can take a long time to recognise the patterns relating exertion to later symptoms, though; our reality is so different to the common understanding of how exertion impacts bodies. For the first ten years of being sick, I thought it was a problem that just mysteriously disappeared and reappeared nonsensically. So I wouldn't put total faith in the way that newly sick people are interpreting their patterns of illness and which events they're connecting symptoms to.

 

[Jonathan Edwards]

I don't think we have solid evidence that most people who fit the CCC at six months have an illness that never fully resolves.

You may be right but we have traditionally had a figure of around 80% not resolving. That may be too high but it is what is quoted. Do we have any better data?

All these arguments in this section of the paper are general and not intended to be precise. They make a point about the overall dynamics. I am quite surprised that patients are concerned about saying that adult disease mostly does not resolve because they are usually keen to point out that it is
n't just a phase that will go away.

But CCC only requires symptoms for 6 months for adults, and as far as I’m aware it makes no reference to monophasic improvement,

But it wouldn't. That is the point. It applies at a point in time. What it does do is indicate relapsing after exertion. I personally think that ME/CFS is better characterised by fluctuations over a longer period than just one episode of PEM but that isn't practical for a set of criteria.

Sets of criteria mostly belong to Emerson's 'foolish consistency' in my view. In all my years working on RA I never knew what the diagnostic criteria were.

 

[Hutan]

You may be right but we have traditionally had a figure of around 80% not resolving. That may be too high but it is what is quoted. Do we have any better data?

What is the source of the 80%? The data is of course a mess, with poor characterisation of ME/CFS, especially in those early months.

I am quite surprised that patients are concerned about saying that adult disease mostly does not resolve because they are usually keen to point out that it is
n't just a phase that will go away.

We mentioned before the problem with saying that hardly anyone recovers from ME/CFS is that creates a situation where anyone who reasonably believes that they had ME/CFS and recovers feels special, and may promote whatever they were trying at the time as the cure. If lots of people are recovering in that first year or two, and they do seem to, then that is a lot of people feeling as though they have something important to share about brain retraining or LP or gratitude journalling.

It also creates a disconnect between the clinicians who see people recover from what sounds to them to be ME/CFS, and the patient charities seemingly promoting the doom and gloom of 'only 5% of adults with ME/CFS ever recover', or whatever. It creates problems in treatment studies that don't properly account for natural recovery.

 

[Robert 1973]

I am quite surprised that patients are concerned about saying that adult disease mostly does not resolve because they are usually keen to point out that it is
n't just a phase that will go away.

I think it’s more accurate to say that it is not unusual for patients be be diagnosed with ME/CFS and recover within 2 years but that recovery is unusual thereafter – as I think has been said earlier in this thread.

There are many reports of people recovering from ME/CFS following interventions. In my experience patients on here are usually keen to point out that natural recovery is the most likely explanation, particularly in the first 2 years.

I don’t know if there is useful data on recovery rates following diagnosis using CCC. @Dolphin may know.

Edit: crossed with @hutan’s post.

 

[Utsikt]

Kielland and Lie showed that of the people that were diagnosed with G93.3 (including PVFS) in 2016 in Norway (CCC is often used, but Fukuda occurs as well), pretty much nobody recovered fully, and very few partially.

It must be noted that G93.3 can only be used by specialists, and not GPs, so the data might be skewed towards the more ill patients.

The data might also be skewed by survivorship bias, because most people started using more sick leave and had a negative wage development many years before their diagnosis, so the data doesn’t capture the early years of the people that did recover.

But I think it’s still fair to say that if you get a diagnosis after a few years of being sick, chances are that you’ll stay sick.

I wish we had better data for the early years.

The analysis in this article suggests that the wage income for persons diagnosed with G93.3 starts to deviate from their control groups' wages years before the diagnosis. The sharpest decline starts around 3 years before and lasts until a year after diagnosis, then stabilizes at a very low level. The consumption of sick-leave benefits starts increasing even before that, and also taxable transfers begin to deviate from the control groups’ as early as 5–6 years before diagnosis, gradually replacing wage income and sick-leave benefits.

Only a small share of the G93.3 cases who made no or very low wages around the time of the diagnosis experienced a positive wage development in the following years. In the 2016 cohort, 57% of the 1524 cases showed no improvement, 38% deteriorated, and only 5% improved somehow over the 3 years studied. Improvements were small. Only three of the 733 persons who made less than NOK(2018) 100,000 earned wages corresponding to a median wage 3 years later. Although less certain, the longer-term timelines based on the full sample of cases from 2009 to 2018 suggest similar trajectories.

 

[Utsikt]

This is basically the diachronic/synchronic point.

Could you explain this point in layman’s terms?

 

[Jonathan Edwards]

Could you explain this point in layman’s terms?

Very simply: A synchronic concept can be applied as a class across examples occurring at the same time - that are synchronous or simultaneous. A diachronic concept is applied across events or entities occurring over extended time. But as soon as you start thinking about it you realize things get much more complicated and so the terms are used in various ways in various contexts.

People with MECFS today would be a synchronic grouping. All Norwegians who ever had MECFS has a diachronic aspect. Maybe a good illustration is the distinction between prevalence and lifetime prevalence - which may cause confusion.

Diagnostic criteria are usually designed to identify people with features now. But we may be more interested in what features they have at other times. Ironically the main point of a diagnosis is to predict future health.

So if we want a diachronic concept of a disease diagnostic criteria may not be the best way to delineate it.

The best example in biology is a species. This is a concept based on ideas of inheritance but it can only be defined for individuals now that might belong to closed mating groups. Since individuals constantly change with mutation and natural selection there is no historic group in the past that is the same. There is no dividing line over time between H. erectus, H. sapiens and Neanderthal. But at a point in history they would be separable.

 

[EndME]

I’ve now managed to get around reading this piece. Whilst I’m not qualified to say anything about the details of the theory I thought it might be useful to write down some of my thoughts on some of the things (much of which has also been discussed) which I believe to be central to the piece but not to the specific theory:

• The development of the theory is entirely orthogonal to what we have otherwise seen in ME/CFS research. Previous ME/CFS theories (mitochondrial theories, ATP theories, viral persistence ideas, neuroinflammation, GPCR aabs, microclots…) have been based on assuming one already knows where the problem lies, often using a meaningless study as evidence. Here the opposite is done: The theory is based on all negative findings (no inflammation, Rituximab showing no efficacy, no structural abnormalities, no organ damage, no pathogenic antibodies and all other negative study results). It then combines this with what is known about onset age and sex ratio. This seems fairly obvious but I think it has to be reiterated that this is very different to what we’ve typically seen.
• This is a useful exercise in the light of positive results by DecodeME as well: A theory based on DecodeME will have to account for the clinical picture and all the negative findings sensibly.
• I think one can possibly argue that when trying to come up with a theory that combines the negative results and the sex ratio, the sex ratio could be a weak point of developing a single theory. With the current diagnostic uncertainty, it’s hard to understand what exactly the sex ratio might tells us, especially if for example the condition may be made up of different conditions both with a very different sex ratio (say a T-cell mechanism slightly male dominated and a B-cell mechanism with a 9:1 ratio).
• An open question to me @Jonathan Edwards : Age distribution is discussed in the text and the hypothesis tries to handle the discrepancy of the permissive role of diverse antibodies climbing with age and age of onset of ME/CFS not climbing until old age by introducing mechanisms that decline after early adulthood, making the full pathological loop less likely to initiate, but as far as we know once acquired, ME/CFS resolvence seems almost independent of age and as mentioned in the text, perhaps may even be reversed to age of onset with higher chances of recovery at lower age? If the B-cell repertoire broadens with age and there is an accumulation of low-affinity IgG antibodies which increases the likelihood of FcγRI activation, but ME/CFS onset doesn’t increase with age because there are other mechanisms that counterbalance this (macrophages or T-cells being less “eager” less IFNg) if the second things are part of the positive feedback loops how do you not get higher chances of resolution of symptoms in older people? Just because there is generally more junk around or why would the feedback loops be age independent despite having the symptoms being to resolve on its own (which might be quite different to some of the illnesses mentioned in the text)? If I think of young people as having a new engine that is once powerful and sensitive and as such with higher chances of overheating when things have gone wrong and older people carrying a slower but more stable engine even if some of the antibody rust on it is quite questionable, what exactly is the fire extinguisher they seem to carry at least equally?
• One comment: I noticed the perhaps slightly more obscure reference to Margaret Mar and was wandering whether a reference to GWI in the same section might be useful as a nerve agent has been implicated there and there are results from smaller GWAS on GWI.

 

[Jonathan Edwards]

Thanks for the comments.

The development of the theory is entirely orthogonal to what we have otherwise seen in ME/CFS research.

Jo and I have seen orthogonality as the only interesting approach to science since we met in 1983ish. We were a bit worried Jackie was less that way inclined, but she didn't turn a hair.

With the current diagnostic uncertainty, it’s hard to understand what exactly the sex ratio might tells us, especially if for example the condition may be made up of different conditions both with a very different sex ratio (say a T-cell mechanism slightly male dominated and a B-cell mechanism with a 9:1 ratio).

I agree that the sex ratio may be misleading and on its own would be a weak point to build a theory on. If anything one would build an autoantibody theory. But the combination of all the considerations drove me away from that simple solution.

If the much higher diagnosis rate in women is not something to do with either antibody spectrum or perhaps some broader factor that drives antibody spectrum (which as we say could be interferon gamma regulation) I find it hard to see an obvious alternative beyond the popular assumption that women worry more and so are more likely to think they have a weird illness. Which I just don't believe is the answer.

for example the condition may be made up of different conditions both with a very different sex ratio (say a T-cell mechanism slightly male dominated and a B-cell mechanism with a 9:1 ratio).

This is certainy a possibiity. But for me the combined analysis of age and sex distribution makes this a less parsimonious approach with little attraction. ME/CFS in women might parallel lupus dynamics and in men Reiter's dynamics but the shapes of the age profiles look to me more like it is the same problem.

If the B-cell repertoire broadens with age and there is an accumulation of low-affinity IgG antibodies which increases the likelihood of FcγRI activation, but ME/CFS onset doesn’t increase with age because there are other mechanisms that counterbalance this (macrophages or T-cells being less “eager” less IFNg) if the second things are part of the positive feedback loops how do you not get higher chances of resolution of symptoms in older people?

I think because unwinding requires some further readjustment of the adaptive immune cell repertoire - if you like the plasticity of the T cell repertoire (because B cell dynamics do not change much with age). Early on it may be easier for rogue T cell populations to arise or expand but equally easier for balancing out populations to come along and say 'hey, you shouldn't really be doing that'. As time goes by the T cell repertoire is believed to become more static, with thymic involution. It doesn't become immutable, and thymic activity does not cease, but it makes sense that the T cell recognition system is learning how to produce a balanced defence system mostly in early life. Unlike B cells, T cell populations include both responders and suppressors/regulators and either may expand.

T cells are not by any means irrelevant to autoantibody driven disease. In RA it looks as if they provide inappropriate help on a fairly broad basis. But for lupus and ITP T cell input may require more specific populations that may get thrown up in young adulthood but are less likely to arise de novo in old age.

But all in all I think your summary is very in line with what we would want people to read in the paper.

 

[Utsikt]

Would someone be able to help me by explaining what «orthogonality» means in this context?

 

[OrganicChilli]

Would someone be able to help me by explaining what «orthogonality» means in this context?

As someone who had the same question and had to Google: independent or not influenced by existing ideas or assumptions

 

[Jonathan Edwards]

Literally, it means in a direction that has nothing in common with what has been indicated before - at right angles. It generates a cotangent of ZERO.

Or maybe more relevantly a cosine of ZERO.

 

[Utsikt]

As someone who had the same question and had to Google: independent or not influenced by existing ideas or assumptions

Literally, it means in a direction that has nothing in common with what has been indicated before - at right angles. It generates a cotangent of ZERO.

Thank you!

So it’s orthagonal in relation to previous ideas, not the data. That’s what was throwing me off.

 

[Jonathan Edwards]

So it’s orthagonal in relation to previous ideas, not the data. That’s what was throwing me off.

I think what EndME meant was that it was orthogonal in approach. Everyone else was taking the metro here or there and we were suggesting going up the Eiffel Tower.