Trial Report Plasma cell targeting with the anti-CD38 antibody daratumumab in ME/CFS -a clinical pilot study, 2025, Fluge et al

ryanc97 said:
I’m sorry but correlation is not causation is the biggest cop out people use here when you see a strong correlation, saying correlation not causation is just a way of saying I will ignore that correlation because I want to.
Click to expand...
If you’d like to give a detailed explanation of what the mechanism is and why there is proof beyond doubt please do, I’m all for hearing it.

I’m comfortable that I’m coming at all this from a place of ignorance and lack of experience, and find questions are a good way to learn.
 
Last edited:
ryanc97 said:
Yes, and I’m saying it’s an easy cop out when you want to disregard the correlation. Just say “correlation not causation” and you’re done!
Click to expand...
I would say that's the start of the discussion for any good scientist whose data shows associations, They would ask questions of whether it shows causation, and if so in which direction, examining things like reliaibilty and validity of the data, confounding factors, sample size, reproducability, how representative the samples are, whether the interpretation makes sense biologically etc.

We see, particularly in the psychobehavioural studies, all sorts of claims being made on the basis of correlations in questionnaire data that are not based on sound evidence, but on wishful thinking and prejudice.
 
hotblack said:
Separately to this I was wondering if the balance of activating and inhibitory signals could somehow be a bit off for NK cells in people with ME/CFS, so wondered if this could play into things here.
Click to expand...
A bit more background on this can be found in this write-up, some quotes below

Whether or not the NK cell kills these cells depends on a balance of signals from activating receptors and inhibitory receptors on the NK cell surface. Activating receptors recognise molecules that are expressed on the surface of cancer cells and infected
cells, and ‘switch on’ the NK cell. Inhibitory receptors act as a check on NK cell killing. Most normal healthy cells express MHC I receptors which mark these cells as ‘self’. Inhibitory receptors on the surface of the NK cell recognise cognate MHC I, and this ‘switches off’ the NK cell, preventing it from killing.
Click to expand...

The genes for both MHC I and the NK cell inhibitory receptors which recognise them vary a lot between individuals.
Click to expand...

NK cell varieties also change with age and are affected by chronic viral infections such as cytomegalovirus (CMV).
Click to expand...
 
hotblack said:
Presumably this balance matters even for a particular cell type, like in the theory here? It’s not just ‘cell with this receptor gets killed’ it’s a probabilistic thing depending upon receptor makeup, how much daratumumab binds and other factors on deciding if a cell is killed?

Separately to this I was wondering if the balance of activating and inhibitory signals could somehow be a bit off for NK cells in people with ME/CFS, so wondered if this could play into things here.
Click to expand...
Yes, efficacy of cell killing by NKs would definitely be influenced by several factors, including balance of competing signals as you mention here. And just as many things would influence simply the proliferation capacity of NK cells at baseline (affecting their number), which was my point with the “alternative” theory there—you could propose that NK cell baseline number is predictive of Dara responsiveness because:

1) there are more NK cells available for killing early in the treatment course, or
2) NK cell number at baseline is an indirect indicator of functional “killing ability” in NKs that makes them better at targetting LLPCs, or
3) NK cell number at baseline is an indirect indicator of some other signaling pathway that influences both NK proliferation and disease state, but that has nothing to do with autoantibodies from LLPCs

All three have similar plausibility on their own grounds, though 3 encompasses a much larger number of possible mechanisms to rule out. If it is true that NK cell killing happens pretty immediately after treatment initiation and 8w is about how long it takes for relevant antibodies to clear out, then my previous critique of 1 doesn’t hold. Though 1 and 2 are still hampered by the same explanatory issues as any autoantibody theory for ME/CFS, which I think has been discussed ad nauseam on other threads so I won’t go into that here again.

And, of course, all 3 theories are grounded in the assumption that these preliminary trial results are showing real improvement from the drug [edit: and the NK correlation isn’t just a fluke] which may turn out to be wrong anyways.
 
Last edited:
I wanted to see how the baseline NK correlation compares when using steps instead of SF-36.

They say that the correlation between baseline NK and change in SF-36 is r=0.77, p=0.012. Weirdly, I get the same r of 0.77, but I can't seem to get the same p, and instead get 0.0089. Close but a little bit smaller.

Maybe someone else can figure out why it's not exactly the same. I calculated the SF-36 Max Difference using what they said in figure 6 for baseline. For maximum SF-36, I assumed they used the maximum up to week 40, since that's the last week where everyone had data.
Caption for fig. 6 said:
maximum increase in SF-36 Physical Function during follow-up (maximum SF-36 PF minus baseline SF-36 PF; the baseline value for SF-36 PF is the mean of seven recordings from week −12 to week 0)
Click to expand...
Here's the data I used for the correlation, which seems to match fig. 6H.
Participant,SF-36 Max Difference from Baseline,NK-cells at baseline
1,75,251
2,51.4285714285714,136
3,0,80
4,0,107
5,53.5714285714286,169
6,17.8571428571429,85
7,39.2857142857143,294
8,58.5714285714286,383
9,10.7142857142857,115
10,50,195
Click to expand...

When looking at the Spearman correlation between baseline NK and change in steps from start to finish, I get r=0.87 and p=0.0012. Good to see that it's even higher when using two objective variables.
 
jnmaciuch said:
Are you using scipy stats? They used SPSS so my guess would be that there is some difference in defaults for how the p-value is calculated from the correlation coefficient (Fisher’s vs. permutation, probably)
Click to expand...
Yeah, that's what I tried first. I also tried this in R and got the same result:
cor.test(nk_bl, sf36_diff, method="spearman", exact=FALSE)
Click to expand...
Maybe it is some other calculation in SPSS.
 
jnmaciuch said:
my guess would be that there is some difference in defaults for how the p-value is calculated from the correlation coefficient (Fisher’s vs. permutation, probably)
Click to expand...
Good call, I think you're right. I found an R function that calculates Spearman p value with permutations (perm.cor.test). I ran it a few times with 100,000 permutations and the p value ranges from around .0115 to .0122, so right around what they reported.

Using this method for the NK vs diff steps correlation: p=0.00211
 
forestglip said:
Good call, I think you're right. I found an R function that calculates Spearman p value with permutations (perm.cor.test). I ran it a few times with 100,000 permutations and the p value ranges from around .0115 to .0122, so right around what they reported.

Using this method for the NK vs diff steps correlation: p=0.00211
Click to expand...
Good to have that explanation! And it’s good that the manuscript seems to have used permutation, since that’s the better method for small sample sizes.
 
CD38 mRNA is down in bulk PBMC RNAseq data I have from ME/CFS and Long COVID w/PEM versus healthy and post-COVID controls. not sure if this helps anyone's thinking but thought i'd throw it in (may be explicable by cell type proportion shifts)
 
Last edited:
Regarding NK cell count vs function there is this research

A Flow Cytometry-Based Whole Blood Natural Killer Cell Cytotoxicity Assay Using Overnight Cytokine Activation - PMC

Background: Measurement of natural killer (NK) cell function has important clinical utility in several diseases. Although the flow cytometry (FC)-based 4-h NK cytotoxicity assay using peripheral blood mononuclear cells (PBMCs) in the clinical ...
pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov

Which says “absolute NK cell count correlates more with the results of the WB NK cytotoxicity assay than the PBMC cytotoxicity assay.”

@jnmaciuch this might be of interest to you?

maybe this sheds light on the count vs function debate.

Also for point no3, I know someone, a ME patient, who increased his NK cells from 70 to 320 in a year. I myself increase it from 49 to 92 last month.

So if that was representative of some process nothing to do with LLPC, does that mean you manage to rewire your ME such that it’s responsive to Dara?
 
Last edited:
jnmaciuch said:
…I doubt anybody could “rewire” their disease in any meaningful way
Click to expand...
I get what you mean, but to me if it’s possibility number 3 that you mentioned then this needs to be explained too if people can boost NK cells.

This links to the idea of retesting the non responders with another dose of Dara since they had mean 150 NK with little spread at the end of the study.
 
Last edited:
voner said:
@ryanc97,

how confident are you that your NK cell tests are accurate?
Click to expand...

I mean, I did one on Aug 3rd and one on Sep 5th. Both were done under my immunologist around the same time of day (2pm), sent to the same hospital. So I don’t see any reason why it would not be accurate.

The only confounder is that on the first test I was recovering from some allergy or mild flu, had a lot of congestion and some sneezing.
 
Last edited:
ryanc97 said:
I mean, I did one on Aug 3rd and one on Sep 5th. Both were done under my immunologist around the same time of day (2pm), sent to the same hospital. So I don’t see any reason why it would not be accurate.

The only confounder is that on the first test I was recovering from some allergy or mild flu, had a lot of congestion and some sneezing.
Click to expand...

A few years back I tried to get an understanding years of my own blood tests that indicated NK cell count being extremely low and having some other anomalous immune results from blood tests. I ended up consulting with the head of the immunology department at a University research Hospital. he had spent his whole career in the innate immunity field. he said he only trusted trust any NK tests in the United States that came from one specific university hospital lab. He said he actually re-assayed for three times before he started to believe the numbers.

FWIW, I also remember that Dr. John Chia only would use ARUP for his NK cell testing.

i’m not claiming any expertise in this area. I’m just trying to give you some feedback… This discussion about daratumumab is certainly intriguing…
 
Last edited:
ryanc97 said:
I get what you mean, but to me if it’s possibility number 3 that you mentioned then this needs to be explained too if people can boost NK cells
Click to expand...
Not necessarily, if multiple things can affect NK proliferation besides the one potential link that influences disease state. Which is almost certainly the case—proliferation pathways are tied in to many many different signals.

Plus, I would expect two tests results for one person to be heavily influenced by expected fluctuation in those tests, either due to natural variation in cell type proportions day to day or month to month, or differences in test protocol such as sample handling affecting rates of cell death in the vial.
 
You must log in or register to reply here.
Back
Top Bottom