Preprint Comparison of T-cell Receptor Diversity of people with Myalgic Encephalomyelitis versus controls, 2023, Dibble, Ponting et al

Pre-print

Free fulltext:
https://www.researchsquare.com/article/rs-3164397/v1

Comparison of T-cell Receptor Diversity of people with Myalgic Encephalomyelitis versus controls

Joshua J Dibble1

Ben Ferneyhough2

Matthew Roddis2

Sam Millington2

Michael D Fischer2

Nick J Parkinson2

Chris P Ponting1

Email

1 University of Edinburgh,

2 Systems Biology Laboratory UK


Objective:

Myalgic Encephalomyelitis (ME; sometimes referred to as Chronic Fatigue Syndrome or CFS) is a chronic disease without laboratory test, detailed aetiological understanding or effective therapy. Its symptoms are diverse, but it is distinguished from other fatiguing illnesses by the experience of post-exertional malaise, the worsening of symptoms even after minor physical or mental exertion. Its frequent onset after infection might indicate that it is an autoimmune disease or that it arises from abnormal T-cell activation.

Results:

To test this hypothesis, we sequenced the genomic loci of a/d, b and g T-cell receptors (TCR) from 40 human blood samples from each of four groups: severely affected people with ME/CFS; mildly or moderately affected people with ME/CFS; people diagnosed with Multiple Sclerosis, as disease controls; and, healthy controls. Seeking to automatically classify these individuals’ samples by their TCR repertoires, we applied P-SVM, a machine learning method. However, despite working well on a simulated data set, this approach did not partition samples into the four subgroups, beyond what was expected by chance alone. Our findings do not support the hypothesis that blood samples from people with ME/CFS frequently contain altered T-cell receptor diversity.

 

They are using Renyi entropy which is sort of a generalised measure of diversity which includes many different measures of diversity within it. You get a curve like one of these (taken from this website):

Where the y axis Ha(X) is the entropy a.k.a a measure of diversity, and the x axis is a parameter alpha. Different values of alpha represent different diversity metrics. At alpha=1 you are looking at shannon diversity for example.

From the methods:

So I think what they're doing is for a given cell type (CD4 or 8) and chain, every patient gets a curve like this and they compare every possible pair of patients and calculate how 'different' their curves are by calculating euclidean distances between them, generating a single number for each pair. Then they use machine learning to sort them into the groups and see how accurately they can do it. They get their p values by randomly shuffling the data labels around and trying to sort into groups based on these random labels many times, and basically seeing how their real data compares against this randomised null distribution.

Here is their logic of why they do it this way, rather than just picking one diversity measure:

 

The logic kind of seems fair enough, although I wonder if comparing the whole curve might in someway dilute the signal from whichever source of diversity is actually most relevant - but I suppose in the absence of knowing exactly what kind of diversity you are looking for this makes sense. They test their method out on a simulated dataset to validate their approach, but in doing so you are making all sorts of assumptions about what you think a dataset with clonal expansions will look like.

I think it might have been nice to validate their method on real datasets where there is a known difference in clonal expansion. I'm wondering also if such other datasets actually exist and whether changes in diversity have been seen from this kind of analysis in different diseases. I guess there would be T cell lymphomas where you would see a huge expansion of a singular clone? What about other disease? Clearly they don't see a signal in MS in this dataset for example, possibly because any relevant cells are not in the blood but in the CSF or other tissues as @Hutan alluded to. I seem to recall from various sources that measuring things like this in the blood can be tricky since the action is usually happening elsewhere, and where it is relevant it might be because you are capturing cells at the right timepoint in the process of trafficking from one place to another. I have to defer to people who know much more than I do about this sort of thing.

 

Wow, never expected Claude Shannon to come up here.

 

I realise you have written further on this. But my initial thought was 'is this so'?
I don't think we have good evidence for MS involving a specific T cell response. B cells yes, but the data on T cells has always looked iffy to me and for the autoimmune diseases I studied there is really no evidence for specific T cell responses (the clear exception is coeliac disease for sa special reason).

The problem with all this is that T cells in blood may well not show up restricted clonality because the relevant T cells are busy in tissue. But restricted clonality in target tissue isn't any good because there is no control situation - normal people do not have many T cells in these tissues. Lymphoid tissue might be a good guide but getting lymphoid samples has always been very tricky in practice. In any inflamed tissue there will be restricted T cell clonality because the T cells are recruited selectively and may expand there. In ME there aren't any tissues with T cells in of note. We all have a few everywhere but not enough to get any data from I suspect.

My gut feeling is that this approach is like trying to roll a boulder up a mountain. It is good that they tried to repeat some preliminary data from the USA but I wouldn't want to pour money into going further.

 

Lymph nodes will be packed with T cells, as they are in normal people. But biopsying lymph nodes has proved impractical and raises huge sampling problems.

T cells infiltrate normal muscle after exercise and again sampling and access are really tricky. If ME muscles were significantly more inflamed than normal people's it would have shown upon magnetic resonance spectroscopy or imaging.

In barn door pathological states like RA and lupus we tried to get useful information from lymphocyte populations in the 1980s and people have again since, with new techniques, but I don't think it has told us anything fundamental.

 

Well, PEM is said to let for months in some cases and also to be an exacerbation of symptoms already present. I find it hard to see why people who are persistently severely ill with ME would not have pathology if there is some?

 

I love seeing a study that admits it found nothing. We need more of these! It will prevent researchers from turning over the same stone repeatedly.

 

this is getting a bit off topic but I'm curious if you think there is a sensible tissue to look at/biopsy in general for ME. Or with our current level of knowledge would it be too much of a blind guess?

 

It would be difficult to get a severely ill patient in PEM to collaborate with anything that requires leaving the house, or even the stress of strangers coming into their home to take blood and do an examination.

 

I think there was enough inflammation to significantly alterT cell populations in muscle it ought to have been picked upon some cases by now. In polymyalgia we thought there was nothing much to find in muscles, just a raised ESR, but when MRI became available it showed up like a light bulb perimuscular tissues in the shoulders.

And if the pathology only occurs in people who cannot tolerate investigations we are up against it yet again!

 

No, I do not think biopsy is going to help us.