An Update of a Theory

I've decided to prepare myself for the results showing no clear link, or being a bit too confusing to interpret straight away. It would be so good to have something to focus on, though, even if it still leaves a long way to go to trial disease modifying treatment. Researchers will struggle to get major grants as things stand, as they've so little to go on.

 

Please see the Fluge cycloME paper

https://www.s4me.info/threads/intra...-study-2020-rekeland-mella-fluge-et-al.14925/

Patients positive for HLA-DQB1*03:03 and/or HLA-C*07:04 (n = 12) had significantly higher response rate compared to patients negative for these alleles (n = 28), 83 vs. 43%, respectively.


And this paper
Human Leukocyte Antigen alleles associated with Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS): Fluge, Mella et al 2020

https://www.s4me.info/threads/human...syndrome-me-cfs-fluge-mella-et-al-2020.14329/

We report novel HLA associations pointing toward the involvement of the immune system in ME/CFS pathogenesis.

 

Well, my sensitivity abruptly ended after food poisoning flushed my digestive tract out, so it seems likely some microbe was messing up my immune response somehow. The hallmark of type IV sensitivity seemed to be a 48-72 hr delay, very consistent. My delay was +/- just a few minutes, and the immunologist I asked said she couldn't think of any other process with that precision and length of delay. It didn't vary with the type of food or the amount, or any other variable I was aware of. I'm really impressed with the precision and consistency of that biological timing circuit.

 

I am aware that Fluge has reported some associations but I don't think these have been confirmed by others. DecodeME should tell us but so far MHC linkage has not been consistent. Fluge only has small samples.

The response data are I think hard to interpret. I don't know what the correction factor should be for any statistical analysis on this and again the numbers are small. A higher response rate relating to DQ and C would be odd because these are involved in quite different aspects of lymphocyte behaviour.

I guess it is possible that Fluge has picked up subsets of ME/CFS with specific immune mechanisms but the validity of these things ultimately rests on hard statistics from big numbers I think.

 

Seems like this research is the the sort of thing which might identify underlying chromic activation of the immune system - [Max Planck Institute Director receives funding for Silent Flagellin in Chronic Inflammatory and Auto-immune Disease (SilentFlame) - https://www.bio.mpg.de/327802/ruth-ley-receives-erc-advanced-grant-2024]

As per Jonathan's comments, there may be clues from GWAS [DecodeME] - so even if "Silent Flagellin" aren't the specific stimulus, it may be possible to identify that activation of the immune system is likely - cue search for stimulus.

Also Dr Li* highlighted that there seems to be immune activation, and a (xenobiotic) stimulus coming from the gut [NIH metabolomics webinar], and that metabolite coverage was poor in ME. So basically there seems to be an opportunity to get data, re potential cause, via metabolomics. GWAS & metabolomics can be used without a prior theory (agnostic) or focused to test a theory - potentially finding the unexpected!
*https://www.jax.org/research-and-faculty/faculty/shuzhao-li

Whole genome sequences/rare variant studies [e.g. families with more than 1 member affected & at least one severe] are another way to potentially identify cause.

 

Thanks for this lengthy idea collection @Jonathan Edwards. I cannot claim to have understood much of it yet, but I wonder how one could begin to examine the above ideas.

Regarding the analogy to pregnancy and autoimmune disease: Do we have any (anecdotal) evidence on the role of pregnancy in ME/CFS? My impression from a few internet stories had been that some people had reported symptomatic improvement during pregnancy and I hadn't heard anything about worsening of symptoms which is probably what one be more likely to expect as a pregnany seems to come with a lot of "PEM-inducing" efforts, but I cannot say anything about the reliabilty about these anecdotes or whether this rate of improvement surpasses the natural rate of improvement present in mild/moderate ME/CFS patients (presumably mainly patients on the "milder spectrum" of ME/CFS would get pregnant) and presumably people on the internet are more likely to report "improvement stories" than the opposite. If data existed on this, would improvements during pregnancy be more indicative of the involvement in B-cells in women or are there abundantly alternative explanations involving NK or T-cell processes (for example has improvement during pregnancy been reported in T cell diseases) that could also explain such a phenomenon, if it even exists?

Your observations seem to imply that it is very important for studies to record whether patients had a sudden onset following infection (so possibly more likely to be T cell related) or whether they had a gradual onset (so possibly more likely to be B cell related), which unfortunately is not commonly enough reported. A similarly easy thing, which I haven’t seen asked/recorded, is asking whether there is a history of autoimmune diseases running in the family, should that be something that is recorded and could recording a history of "enthusiastic T-cell diseases" also be fruitful (if they are all linked to MHC Class I genes and DecodeME shows that ME/CFS isn't then it probably wouldn't be useful, but maybe some diseases not associated to MHC Class I genes could be picked up)?

The exception to this would probably be the link between MS and EBV, however that seems to be a very slow or gradual process taking place over years whilst the association between EBV and ME/CFS seems to be operating on a much smaller time scale.

Would it even be plausible to you that B-cell infection with EBV could somehow lead to a very fast onset of ME/CFS as autoimmune disease?

I can’t claim to understand anything about the CD16 and T-cell mechanisms you are proposing here, but would this not be something that Efgartigimod could potentially target?

I once again had a quick glimpse at the RituxME paper to see if there were perhaps marginally different response rates between sexes. I am 100% certain the Norwegian group will have looked at this and probably saw no differences, but I couldn’t find this data, does anybody have it at hand?

 

Hi @EndME I have not heard much about any changes in ME/CFS with pregnancy. It does not seem to disappear or get very much worse at least.

It is difficult to predict what pregnancy would tell us anyway. RA and lupus are fairly closely similar autoimmune diseases in many respects but pregnancy has the opposite effect on each of them - worse for lupus, better for RA.

It would be interesting to know what the effect of pregnancy on psoriasis is - I don't know that.

Onset of disease shortly after an infection points to T cell mediation as a likely option. For antibody mediated disease onset can be abrupt and it may even be precipitated by infection (can occur with lupus) but mostly abrupt onset seems to come out of the blue. Blood bank data suggest that even in these cases autoantibodies may have been present for up to ten years beforehand.

Family history of autoimmune disease tends not to be very helpful since taken as a whole these are common. I think identifiable linkage to MHC Class I or II genes is more likely to tell us something. So far those links do not seem to be coming up.

The relation between EBV and MS still seems to me a bit uncertain but it would be very plausible that EBV infection had a permissive effect on later development of MS. That of course would not be a close temporal link, as you indicate. MS isn't quite an autoimmune disease at least so far in that we do not see a well defined autoantibody. We see antibody production in the wrong tissue but it is still not clear what they are specific for.

Efgartigimod targets FcRn, which is involved in IgG homeostasis. CD16 is FcRIIIa and specifically mediates cytotoxicity by T or NK like cells.

 

The flaky ladies in my family said it resolved in pregnancy, and after each child it was less severe when it returned.

I only saw my mam with a rash on exposed skin once. By the time I was old enough to notice she only usually got it in her ears and belly button, and with her sister (who'd had a rough time with it in her teens) it only ever appeared at the hairline on her neck. They never remembered seeing their mum, who'd had seven pregnancies, with a rash at all.

There were apparently other members of my parents' and grandparents' generations who had a similar experience, but I didn't hear it from them directly. I seem to be the only one who got the arthritis, but I don't have psoriasis. (Always the contrary one!)


ETA: Another thing is that, at least in my mam and auntie, it vanished altogether somewhere round the age of 60. I don't know this to be the case, but my mam assumed it was because remnants of the hormonal cycle carry on after menopause for quite some time, albeit erratically.

 

From Psoriasis and pregnancy (J Cutan Med Surg. 2002 Nov-Dec;6(6):561-70)

 

I can’t pretend I understand your model Jo, but it’s good to know you’re thinking about it so deeply.

I think we all wish there were many more people with the necessary experience, knowledge and intelligence to be thinking so deeply about the problem. Writing this makes me wonder if it would be useful to publish a summary of what we know we know about ME/CFS – the type of knowledge that you have used to come up with your model – including what we know it isn’t, and things which are yet be determined, in order to stimulate more people like you to start thinking about how to solve this problem.

Apologies if I’m being dim, but is there a risk that these signals may be missed in DecodeME because of corrections for multiple comparisons if the predictions are not incorporated into the analysis prior to the release of the data?

Will specific predictions be tested in DecodeME? Or will it all be data crunching to see if there are any correlations that retain significance after adjusting for multiple comparisons?

@Andy @Simon M

 

Agree…would be useful for all these new “Long Covid” researchers. Also would be helpful to list potential/likely druggable targets/pathways—even if there is some potential speculation.

I think JE might argue that there has not been that much new since the 2016 paper (present new theory excluded).