Broad Analysis of Serum and Intrathecal Antimicrobial Antibodies in Multiple Sclerosis Underscores Unique Role of EBV, 2025, Florence Pache et al

[forestglip]

The plot from the Nath paper might reflect an artefact of controls being historic though. I wonder if the cells had been frozen. B cells are not good at making antibody unless given strong encouragement with something like EBV or PMA and they tend to sulk and die in vitro most of the time. Freezing could well be a big problem. But maybe the cells had been studied fresh previously?

I think methodology was identical. They didn't retest old samples - they used old data:

Autonomic function and CSF immunophenotypic analysis obtained in this study were compared with data collected before the SARS-CoV-2 pandemic from healthy volunteers (HVs) tested under other NIH studies NCT02669212 or NCT01875588. The autonomic testing and CSF analysis pipelines were identical in methodology to allow for direct comparisons of the neuro-PASC and HV groups.

One of those listed studies was on ME/CFS, Lyme, and healthy controls, so if there is data on B cells that Hutan mentioned for HC, there should be data for ME/CFS as well to see how it compares to long COVID. I can't find a published paper with this data though.

Edit: Oh, the data is from healthy volunteers in the big deep phenotyping study. (And another group of healthy controls in an HIV study.)

 

[Jonathan Edwards]

Does this mean that the research you would want to examine this idea is already under way?

I think that study could be very useful. So far we have one or two isolated cases of dorsal root ganglion changes that might have been by chance or artefact. Any further findings in any of a sample of five cases would be important.

 

[Jonathan Edwards]

I think methodology was identical. They didn't retest old samples - they used old data:

That is encouraging but B cell survival in vitro is so temperamental that it would be easy to get an artifactual difference. You would really want to process samples blind at least as control-matched pairs and not break the blinding until you had a complete study.

 

[forestglip]

The deep phenotyping study on B cells in CSF in ME/CFS vs HV:

Two PI-ME/CFS participants had detectable antibody secreting B-cells in cerebrospinal fluid; these participants did not have oligoclonal bands or autoantibodies.

So only 2 out of 16 with ME/CFS had any antibody-secreting B cells. I looked at the data on mapmecfs. For "B cell subset - Antibody secreting cell (%)", none of the 20 HV have any, and for ME/CFS one has 4% and one has 10.5%. These are the only two ME/CFS participants that have any of these cells too:

B cell subset - Naïve (%)
B cell subset - Unswitched memory (%)
B cell subset - Switched memory (%)

And one healthy volunteer had the other types of B cells.

Edit: And just in case it's useful, there was no difference between groups for count of total B cells in CSF or B cells as a percentage of lymphocytes in CSF. This data is also in Supplementary File 15, although one healthy volunteer is on mapmecfs but not the supplementary file for some reason.


Edit 2: I don't really understand why almost all participants have B cells in the CSF but nothing is shown for the B cell subsets. It doesn't say 0, it's just empty for most participants for all the B cell subsets, including antibody secreting. Probably equipment just not sensitive enough for these specific markers.

Edit 3: Healthy people shouldn't have this many B cells in their CSF, right? Their other lymphocytes seem too high too. I posted on the deep phenotyping thread.

Edit 4: Pretty sure the data is in cells/mL, but the labels are cells/ul. So I think the numbers in the plot of B cell absolute count are per mL, not uL. So seems like normal values.

Edit 5: The two participants with antibody secreting cells also had the two highest B cell counts of the ME/CFS group. (Around 30 and 56, while everyone else was under 12.) The two healthy participants that had higher B cell counts than 30 (around 49 and 133) did not have detectable antibody secreting cells.

I think it is possible that more ME/CFS than healthy participants have ASCs, but the equipment could not detect it at levels below 4%. Or the data is just missing. Most participants don't have data for any subtypes of B cells even though almost all of them have non zero total B cells.