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

Sorry that was the dreadful spellchecker we seem to have. It was written as monophasic - following a curve with a single up and down rather than wiggles.

 

Interestingly I was just doing a rotation in a virology lab where several people were studying mechanisms by which many common viruses actually downregulate parts of the host interferon response in the first phase of illness. I’ve only seen this in the context of type I interferons, though. Can’t say for sure if it’s relevant but an interesting tidbit nonetheless.

 

Sorry, I must have missed this before stepping away. I don’t think this is an issue—it’s well known that interferon responses can start in parenchymal cells and elevate to sepsis mediated by type I interferon signaling in DCs. [Edit: not solely DCs, of course, but blocking IFNAR seems to be protective against septic shock for this reason]

All you need is a passing dendritic cell to encounter a high enough level of interferon in the muscle and it’ll start screaming its head off with its own interferon response into the circulation.

Also as far as I know, the only person who has measured type I interferon in ME/CFS with any hope of plasma detection was Phair, and I’m pretty sure it was at baseline, not during active PEM.

Thanks again for the discussion!

 

There are thankfully some programs that will automate it for you. Though they’ll only save you time in the 10% of instances where the programs work as expected and you don’t have to waste that same amount of time on troubleshooting instead.

 

I used to think citations were essential but recently I discovered that they are generally slower.
If I have a citation I can go to PubMed and find an abstract but then mostly it says there is a paywall. So I have to go to my UCL site (it may offer to do that automatically) and click on the journal and find the volume and so on.

If I just Google the author names and if necessary a couple of title words I get the paper comes up anyway - often with a free version and also with an AI summary of other relevant material. As far as I can see precise citations and Doi's are redundant and take much longer!

 

I can see the possibility of something like that but what I can't see is why the interferon response to activity should be abnormal without there being some learnt adaptive drive that appears after an infection and can then go up and down over months and years. An acquired tendency to overreact in interferon terms could be a purely maturational shift. Ankylosing spondylitis may be an example of that - you may be programmed to start overproducing cytokines in spine aged 15 as the tissue shifts from growing point to mature enthuses. But ME/CFS does not have that sort of stereotyped age of onset.

 

Qeios have uploaded some wording changes I made earlier.

 

Unless the disease state is characterized by abnormal regulation of constitutive parenchymal interferon production such that it increases in viral infection and doesn’t get suppressed as much as it should afterwards, creating a self propagating loop as interferon is wont to do.

Which, during activity, would cause an exaggerated response to a sterile pathogenic trigger already known to cause an immune response during exercise even in healthy people.

A trigger which may potentially be more likely to occur at much lower thresholds of activity if that higher constitutive interferon production affects metabolic state, or if there’s some genetic abnormality shifting the balance of extracellular calcium (which wouldn’t have caused problems until now).

[Edit: and, since suppression of interferon transcription involves multiple transcription factors and epigenetic changes acting cumulatively, there is lots of room for influence of hormonal changes, host-virus interactions in future infections, and various stochastic processes leading to those fluctuations]

 

Just been watching the PrecisionLife presentation on the AfME webinar. They have what seems a very black-box type of way of analysing genetic data. They ended up with a dozen or so different subgroups of PwME by means I didn't understand, and were talking about precision medicine - different drugs according to your particular genetic profile (the subgroup to which you belong, I think?).

How does that sort of approach fit in with the theory you're proposing here, @Jonathan Edwards? I don't have the impression that your approach predicts a lot of subgroups, each with a tailored drug.

Is any precision medicine going on in RA? Or is everyone just on rituximab?

 

With this is mind, if I’m understanding correctly, this is why it would allow for both people feeling better and people feeling worse, depending upon the stimulus.

So I can feel better after one immune trigger, whereby the T cell population involved and which expands are ‘not ME/CFS type’. So as a percentage of overall activity my ‘ME/CFS type’ is lower, there is less of that activity, relatively speaking. The feedback loop could be said to be being curtailed a little.

While for another immune trigger, the T cell population involved which expands is ‘ME/CFS type’. So as a percentage of overall activity this is greater, there feedback loop is supported.

So one question is, while there may be some commonality between us for this to all work, there may be individual variation (I’m thinking genetic as well as antigen here)?

 

I have a Google Scholar alert for ME/CFS and it was included in my results 2 days ago, but linking to a PDF listed on the MEA website instead of Qeios. I think I did see the odd Qeios result over the years so they are included, but don't seem to rank especially high.

 

BTW, if I type 'Jonathan Edwards' into the Qeios search box, I don't get this latest paper - just some older ones. Is that just me?

 

This is Rob Phair's argument and it seems to me to be doing a lot of heavy lifting, which, without a specific mechanism borders on Deus ex Machina. As you know, I have been observing diseases as a clinician and pondering their system dynamics for about fifty years and at the moment I don't see a precedent - without an adaptive immune response - for this in other diseases. ME/CFS may well have a unique dynamic structure but it would be nice to have at least some idea what isn't suppressing 'as much as it should'.

 

I get the strong impression that they have picked up on genes that may alter susceptibility of end-organ targets to ME/CFS. As an analogy, if there was a gene for rather narrow heart valves then people with that gene would probably get more trouble with mitral valve stenosis after rheumatic fever. But the gene would not tell us anything about rheumatic fever (other than what we know, that it affects heart valves) or the best way to treat it (eradicate streptococcal infection before the fever gets a chance to start). Another gene might make you more likely to get arthritis with it, but again not tell us the mechanism.

I wonder if this is relatively easy to do if you are looking at variations in presentation of an illness rather than trying to find genes for the illness itself. I didn't feel that any of the genes they mentioned gave an obvious clue to what ME/CFS is, as a process.

My model allows for all sorts of variations based on individual susceptibilities - very much so - but the focus is on what process takes advantage of those susceptibilities.

Also, if these genes for subgroups pick out end-organ susceptibilities then they are likely just to act symptomatically. I suspect that most people with ME/CFS are hoping for a drug that gets at the common process that affects all people with it. Rituximab makes almost everyone with RA better. Penicillin (or a newer antibiotic) gets rid of all rheumatic fever, because it hits the cause.

Precision Life have 'precision medicine' as their raison d'être. That means finding individual drugs for individual people. I am not persuaded that one should major on that. Some fine tuning is sensible but mostly we are looking for treatments that deal with the core problem for everyone with a disease. In RA the choice of treatment rests largely on convenience, history of adverse reactions, and so on. RA as a diagnosis includes a few people without evidence of B cell disease and rituximab doesn't help them. Their disease probably shouldn't be called RA. But that is because they have a fundamentally different problem going on.

 

I don't believe he is right about itaconate in parenchymal cells, but yes we both are relying on parenchymal interferon production. We actually do have some precedent, and quite a few candidate mechanisms that I'm currently trying to run down. This particular system of transcriptional regulation has been well studied since the early 2000s.

Details forthcoming

 

Sasha,

same for me. I do a search for "Jonathan Edwards" and then tap on the "since 2025" option and the paper pops up. hope this helps.

 

Looking forward to it! Also feel free to give us a topic list to brush up on beforehand. We can all find some more videos and books to read

 

It comes up for me if I just google Qeios Jonathan Edwards
The machinery at Qeios seems a bit slow this time but they have done some corrections for me very quickly.

 

Yes, that disappointed me a bit. It sounds a really good way to split your focus in too many directions, or be satisfied with managing to improve some symptoms in one subset because it's enough to meet your commercial targets.