Speculations about the genetics of ME/CFS and DecodeME

[poetinsf]

I felt like writing about this.

In my opinion ME/CFS is probably a label for several different poorly understood illnesses that happen to share some characteristics and which do not yet have their own name. We might call them subtypes of ME/CFS. Then, in a ME/CFS cohort there is probably also a significant percentage of patients whose primary illness was erroneously diagnosed as ME/CFS. These could be unusual presentations of common illnesses with some similarity to ME/CFS like multiple sclerosis, or adult-onset forms of rare diseases like primary mitochondrial disorders or muscle diseases that are often difficult to diagnose.

I'm of a different opinion. If you take individual symptoms, like fatigue or brain fog, it could intersect with many other diseases. In totality, however, it's unlikely that all symptoms are shared with others. And that is especially the case if you take PEM into account; there is no other disease that exhibits such a dramatic worsening after a minor exertion, and the set of just three symptoms {fatigue, brain fog, PEM} already separates ME/CFS from all other known diseases. Can there be multiple *unknown* diseases that produce the same set of symptoms? Possible, but again unlikely. What are the chances that you encounter multiple unknown diseases that produce same set of symptoms? I'm not aware of any in the annals of medicine.

That said, I think it's possible that there are multiple direct causes for ME/CFS. If ME/CFS is a hypersensitivity to exertion, for example, something could be pinching the neuroimmune system to produce that hypersensitivity. For others, the system just became hypersensitive, like allergy. The treatment would be different for each: removing the pinching will do for the former, while you'd have to figure out how to temper down the hypersensitivity in the latter. If you define the disease as {cause, symptoms, treatment} triplet, they could be classified as different diseases. But the underlying mechanism would be the same and identifying that mechanism should be a key in finding the treatment in both cases. And following the symptoms, rather than speculating on unknown possibilities apart from the symptoms, is always the best way to identify the mechanism in any debugging process.

 

[wigglethemouse]

I'm excited by the collaborations that DecodeME will facilitate when the data is made available. Subtyping by symptoms or metabolic pathways and looking for relevant genes could tease out more variants and prehaps treatment ideas.

I know of two who are doing this/have done this using the existing UK Biobank genetic data on ME/CFS. Some projects include :

1. Precision Life have published findings already on analysing the UK Biobank data.
https://www.s4me.info/threads/genet...analysis-2023-taylor-et-al.34243/#post-507747

2. OMF - Melbourne group
https://www.omf.ngo/biological-outlier-and-subtyping-software-for-myalgic-encephalomyelitis/

The analytical tool will be developed using pre-existing UK Biobank gene and metabolism data from over 1000 self- reported ME/CFS patients. Findings from this data will then be validated on gene and metabolism data we produce ourselves from blood received from 300 ME/CFS patients recruited by the Australian ME/CFS Biobank.

@MelbME - Are you able to comment if you will be extending the tools being developed to DecodeME?

3. OMF Stanford are screening a cohort for common genetic variants in the BH4 pathway using an inhouse screen which they will then compare against actual pathway measurements.
https://www.omf.ngo/genetic-and-metabolic-markers-of-bh4-deficiency-in-long-covid/

Tetrahydrobiopterin (BH4) is a vitamin-like cofactor that is made in the body by a series of enzymatic steps. However, there are common mutations in the biosynthesis pathway that are associated with cofactor deficiency. BH4 is needed for metabolism of a small number of amino acids that are precursors to neurotransmitters, and deficiency leads to cognitive impairment and a wide variety of other symptoms associated with neurotransmitter imbalance. BH4 is also needed for the enzyme nitric oxide synthase, which is a major regulator of blood flow dynamics and immune cell function.

4. Liz Worthey and colleagues at UAB are developing tools to analyse Whole Geneome Sequencing data. Jarred Younger stated the first project is out for peer review and hopes for publication by end of year. 7min 25s in
Transcript :

last study is our whole genome sequencing for mecfs this is being run by Liz worthy and camil Burch and it's funded by solve me and uh this one is using this really cool genome sequencing approach that for mecfs it basically will tell you what genetic variants you have that are most likely to be driving your chronic fatigue or whatever your primary symptom is and we finished the first project that is about to be sent out for peer review uh so I'm not going to talk about any of the results right now but I'm assuming by the end of this year it'll be published and I can tell you all about that as well ultimately I hope this is something where everyone who has mysterious pain or fatigue or cognitive issues can have this sequencing done and identify what's probably causing it.

I have high hopes that these and other groups will collaborate with DecodeME when the data is made available.

 

[Andy]

For info on applying for access to DecodeME data see https://www.decodeme.org.uk/researcher-access/.

 

[V.R.T.]

On the other hand making a drug to order now is so easy that I would be optimistic that once we have an idea what it is we want to put right a drug could be made available very quickly. I strongly suspect that there is some sort of signalling problem and signals are particularly easy targets. But we don't know what they are.

I'd like to ask you a little bit more about this if you dont mind: Could a drug to order really be made available so quickly? I was under the impression a new drug would take at least a decade to progresss through trials and reach patients. Do you have a reasonable best case scenario for how a good result from Decode ME might play out?

 

[Jonathan Edwards]

I'd like to ask you a little bit more about this if you dont mind: Could a drug to order really be made available so quickly? I was under the impression a new drug would take at least a decade to progresss through trials and reach patients. Do you have a reasonable best case scenario for how a good result from Decode ME might play out?

The very best thing would be that DecodeME would find a link to some very obscure but crucial control protein - let us say TWEAK (it exists). Suddenly we would realise that ME/CFS might be mediated by short range interaction between TWEAK and gamma-interferon sensitising nerve endings. The task is then to produce a specific TWEAK antagonist that competed for the interferon binding site. Setting up assays for competitive inhibition of binding might take 3-6 months I guess - assuming that either a drug company or a TWEAK biology lab had the wherewithal to get going as soon as the paper is read. I don't know how long it would take to synthesise candidate small molecular mass inhibitors (simple drugs) but I think these days much of that is streamlined. The task would be similar to the generation of protease inhibitors for AIDS, which was quite quick once the target was known. If the choice was to make a monoclonal antibody to do the job you just clone out antibodies with high binding, make some cDNA and go into production. I may be oversimplifying but the biotech people do all sorts of things very quickly these days.

To make the scenario even better we could make it that it was realised that this TWEAK pathway was a key factor in 'malaise' associated with all sorts of illnesses and post-surgery and whatever and looked a really good target for a drug with wide applicability. Lots of research groups would get on to it and just like the Covid vaccines you would have a product made by several in no time at all (under 2 years). Roll-out to clinical use, if the drug was really effective, could be six months, with two phase 3 trials in parallel and accelerated approval.

That is optimistic, but the main point is that once the companies have seen something as profitable they can get it through to the clinic pretty quick. There might even be repurposable monoclonals - similar to rituixmab - but which actually work.

 

[V.R.T.]

The very best thing would be that DecodeME would find a link to some very obscure but crucial control protein - let us say TWEAK (it exists). Suddenly we would realise that ME/CFS might be mediated by short range interaction between TWEAK and gamma-interferon sensitising nerve endings. The task is then to produce a specific TWEAK antagonist that competed for the interferon binding site. Setting up assays for competitive inhibition of binding might take 3-6 months I guess - assuming that either a drug company or a TWEAK biology lab had the wherewithal to get going as soon as the paper is read. I don't know how long it would take to synthesise candidate small molecular mass inhibitors (simple drugs) but I think these days much of that is streamlined. The task would be similar to the generation of protease inhibitors for AIDS, which was quite quick once the target was known. If the choice was to make a monoclonal antibody to do the job you just clone out antibodies with high binding, make some cDNA and go into production. I may be oversimplifying but the biotech people do all sorts of things very quickly these days.

To make the scenario even better we could make it that it was realised that this TWEAK pathway was a key factor in 'malaise' associated with all sorts of illnesses and post-surgery and whatever and looked a really good target for a drug with wide applicability. Lots of research groups would get on to it and just like the Covid vaccines you would have a product made by several in no time at all (under 2 years). Roll-out to clinical use, if the drug was really effective, could be six months, with two phase 3 trials in parallel and accelerated approval.

That is optimistic, but the main point is that once the companies have seen something as profitable they can get it through to the clinic pretty quick. There might even be repurposable monoclonals - similar to rituixmab - but which actually work.

Thanks, that's very clear. Fingers and toes crossed then I suppose!

 

[MelbME]

I'm excited by the collaborations that DecodeME will facilitate when the data is made available. Subtyping by symptoms or metabolic pathways and looking for relevant genes could tease out more variants and prehaps treatment ideas.

I know of two who are doing this/have done this using the existing UK Biobank genetic data on ME/CFS. Some projects include :

1. Precision Life have published findings already on analysing the UK Biobank data.
https://www.s4me.info/threads/genet...analysis-2023-taylor-et-al.34243/#post-507747

2. OMF - Melbourne group
https://www.omf.ngo/biological-outlier-and-subtyping-software-for-myalgic-encephalomyelitis/

@MelbME - Are you able to comment if you will be extending the tools being developed to DecodeME?

3. OMF Stanford are screening a cohort for common genetic variants in the BH4 pathway using an inhouse screen which they will then compare against actual pathway measurements.
https://www.omf.ngo/genetic-and-metabolic-markers-of-bh4-deficiency-in-long-covid/


4. Liz Worthey and colleagues at UAB are developing tools to analyse Whole Geneome Sequencing data. Jarred Younger stated the first project is out for peer review and hopes for publication by end of year. 7min 25s in
Transcript :

I have high hopes that these and other groups will collaborate with DecodeME when the data is made available.

Definitely interested in looking at the Decode data. Have a few things we're interested in.

Sneaky value of this type of resource is providing data for short projects to get masters students interested in this research area to perhaps do a PhD.

 

[Kitty]

I think you need to know when patients are close to it. Knowing that accurately would be a key test to timing different intervention strategies. It's critical.

This isn't really the right place for this, but as I always forget when it is the right moment, here goes!

It's worth bearing in mind that food can have an impact on the likelihood and severity of PEM in some of us. And if research participants need to travel or stay away from home to take part, or will need lunch because it's a full day in clinic, they're likely to vary their eating routines.

It might not even be relevant in the type of work you hope to do, but maybe it's worth adding to the list of questions to ask at the planning stage: do we need to encourage participants to eat their usual food at the usual time, and try to make sure it's feasible for them?

 

[Sean]

I think you need to know when patients are close to it. Knowing that accurately would be a key test to timing different intervention strategies. It's critical.

Definitely important. But...

My experience is that PEM is a permanent feature. It is always there, all that changes is how hard it is cranked up.

It is also a non-linear response, I think, which is part of why it is difficult to learn how to predict and manage.

Even the best management of it is only going to reduce the impact, which is good as far as it goes. An actual cure will require fixing the underlying problem driving PEM.

 

[Kitty]

It is also a non-linear response, I think, which is part of why it is difficult to learn how to predict and manage.

This! I was trying to say something about it when I was waffling about food yesterday, but couldn't muster a string of words that made sense.

 

[MelbME]

Definitely important. But...

My experience is that PEM is a permanent feature. It is always there, all that changes is how hard it is cranked up.

It is also a non-linear response, I think, which is part of why it is difficult to learn how to predict and manage.

Even the best management of it is only going to reduce the impact, which is good as far as it goes. An actual cure will require fixing the underlying problem driving PEM.

There is very little research on it at this stage. What you've described means it's not simple and likely multi-factorial, which I appreciate and is likely true.

Patients currently control exertion of skeletal muscles, some thinking and sensory processes (dark room, reduce noise, etc). But there are internal processes they do have control in over exerting. These are sympathetic drive, immune activation, adrenaline dumping processes (MCAS, POTS), etc. Then there is the fact that energy recovery processes can vary between people.

Even if PEM was a linear process of occurring over a certain level of exertion (not saying it is). It would still appear non-linear because a patient can only control a few factors of exertion.

Agree on the cure or best treatment opportunity being a fixing of PEM.

 

[MelbME]

This isn't really the right place for this, but as I always forget when it is the right moment, here goes!

It's worth bearing in mind that food can have an impact on the likelihood and severity of PEM in some of us. And if research participants need to travel or stay away from home to take part, or will need lunch because it's a full day in clinic, they're likely to vary their eating routines.

It might not even be relevant in the type of work you hope to do, but maybe it's worth adding to the list of questions to ask at the planning stage: do we need to encourage participants to eat their usual food at the usual time, and try to make sure it's feasible for them?

Great point and definitely something to consider. We're really trying to transition as much research to the home as possible because of these types of reasons. Of course CPET or imaging based studies still can't unfortunately.

Very interested in trying to get telehealth and home visits as being more common place for MECFS patients in Australia too. I think this is an important part of improving the care pathway that can be worked on now.

 

[Yann04]

There is very little research on it at this stage. What you've described means it's not simple and likely multi-factorial, which I appreciate and is likely true.

Patients currently control exertion of skeletal muscles, some thinking and sensory processes (dark room, reduce noise, etc). But there are internal processes they do have control in over exerting. These are sympathetic drive, immune activation, adrenaline dumping processes (MCAS, POTS), etc. Then there is the fact that energy recovery processes can vary between people.

Even if PEM was a linear process of occurring over a certain level of exertion (not saying it is). It would still appear non-linear because a patient can only control a few factors of exertion.

Agree on the cure or best treatment opportunity being a fixing of PEM.

Very happy to see you mentioning the “adrenaline dump” or “adrenaline rush” phenomenon a lot of us experience. It’s criminally underreported in the lieterature, I only know of one ME study in english that mentions it, and mostly in passing.

 

[MelbME]

Very happy to see you mentioning the “adrenaline dump” or “adrenaline rush” phenomenon a lot of us experience. It’s criminally underreported in the lieterature, I only know of one ME study in english that mentions it, and mostly in passing.

Well the care model for patients we have built up in our head is about energy preservation and we see pacing and medications to help curtail uncontrollable energy dumps as critical to that.

If you reframe the perspective around energy conservation like we do then it becomes clearer.

Energy drags: controllable exertion (movement, cognitive tasks, sensory input), uncontrollable exertion (infections, adrenaline dumping co-morbidities, overactive immune, sympathetic drive elevated)

Reduced energy provision: reduced blood flow, poor digestion and dietary substrate provision, mitochondrial inefficiency.

From this perspective PEM could simply be dropping below an energy threshold where certain vital organs drastically reduce function which is reversible but has consequence to restarting energy production systemically. Kidneys is a prime target since the gradient of substrate reabsorption needs to be maintained by a certain energy threshold. But liver could certainly also be impacted.

Anyway, this is our broad working idea for some time.

 

[Sean]

But there are internal processes they do have control in over exerting.

I think there is a 'not' missing from that sentence.

Otherwise, yes, there is a lot going on we don't have control over. It doesn't matter how carefully I manage my activity patterns, I still end up being knocked for six at random by unpredictable flare ups.

 

[Kitty]

Very happy to see you mentioning the “adrenaline dump” or “adrenaline rush” phenomenon a lot of us experience.

Energy drags: controllable exertion (movement, cognitive tasks, sensory input), uncontrollable exertion (infections, adrenaline dumping co-morbidities, overactive immune, sympathetic drive elevated)

I've no idea what's actually happening in the body, but for me the adrenaline pump feels like the backup battery kicking in.

I don't get it unless I've overexerted, and one of the problems is the inability to switch it off when the 'emergency' (i.e., trying to complete a mundane task rather than leave it half finished) is over. Nine hours later I can't sleep because the wretched thing is still going.

 

[Trish]

Nine hours later I can't sleep because the wretched thing is still going.

Yes, the 'tired but wired' sleepless night(s) I experience as an early stage of a crash.

 

[MelbME]

I've no idea what's actually happening in the body, but for me the adrenaline pump feels like the backup battery kicking in.

I don't get it unless I've overexerted, and one of the problems is the inability to switch it off when the 'emergency' (i.e., trying to complete a mundane task rather than leave it half finished) is over. Nine hours later I can't sleep because the wretched thing is still going.

Well adrenaline doesn't create energy, it pushes you to use more though it makes substrates of energy (like glucose and fatty acids) more readily available.

 

[Kitty]

Well adrenaline doesn't create energy, it pushes you to use more

That's interesting—though I guess not surprising. It might have a significant role in contributing to PEM, especially as once it starts, it's difficult to switch off.

I don't know to what extent the latter's abnormal, though. I seem to remember it being difficult to settle for a long time after an adrenaline kick before I got ill.

 

[EndME]

Is the term "adrenaline" not simply being used because patients think that is what it may feel like, not because adrenaline actually bares any relevance?

I think at those times it better to avoid medical terminology, even though most of us might know what is being referred to and maybe we should come up with a better and more accurate terminology. I think we should be careful referring to different things as "adrenaline dumps", "energy", "ATP" etc. In particular I wonder if we can maybe uncover more of what might be going on if we drop these kinds of terminologies and go back to the drawing board to describe what one is experiencing in more layman terms.