Long-lived plasma cell (LLPC) theory - Similarities between CFS and Lupus?

Jonathan Edwards said:
What would be nice to know is why DQ matters a lot, when very few T helper cells seem to be DQ restricted in their responses. I haven't heard a good answer to that.
Click to expand...
Ever since I learned that EBV deliberately binds to HLA-DR and DQ as a coreceptor for entry, I’ve had an inkling for a cross-presentation mix up story:

BCR binds weakly to some random antigen it encounters in the CNS, but that latently infected B cell happened to have some viral envelope fragments floating around with high affinity for MHCII that end up presented instead. Even if only a small number of T cells can get into the brain in non-pathological conditions, there’s a decent chance that some of them are EBV-specific T cell clones floating around from the initial infection. T cell signaling from the MHC handshake plus some unique features of the brain niche encourages proliferation of that B cell clone, resulting in lots of antibody for the random antigen. Immune complex interactions with microglia take it from there. Maybe enhanced B cell survival from the latent EBV and/or some other features of the brain niche makes this unfortunate series of events more likely to happen as many times as there are oligoclonal bands.

Just something I was tossing around since it would fit with some preliminary findings of higher EBV-specific T cell clones in the CSF of MS patients. I was encouraging some of my virology classmates to test whether the HLA alleles with the highest affinity for EBV are the same ones that confer MS risk but I don’t think anyone took me up on it. Though the idea is probably infeasible due to some adaptive immunity-related detail I’m unaware of since it’s not my field
 
forestglip said:
Is the idea that symptom improvement would require antibodies to be stuck to CD38, or that just one of these events is enough to interrupt a positive feedback loop for improvement even after antibodies are gone? Is the former (antibodies stay stuck) possible with the length of time (> year) that people continued to see improvement in the open label study?
Click to expand...

The former is not possible most of these molecules turn over quite quickly.

forestglip said:
Do we have some other example of where interrupting a positive feedback loop but otherwise not adding or removing anything from the body helps long term with a disease? I feel like the body's homeostasis mechanisms are too robust for this, but would love to see counterevidence.
Click to expand...

It seems to be the case for immune thrombocytopenia. One shot of rituximab can abort an otherwise chronic disease. We do not know the mechanism of the 'loop' there.

Another possible example is TB. The TB bacillus encourages macrophages to engulf it and then tells the macrophages to build a pile of cheesy junk it can hide away in. Kill the bacterium and the body stops providing the loop.

Another is chorioncarcinoma where the tumour takes advantage of the body growing a vascuar surface for it to grow on, just as happens with an embryo. The body thinks the tumour is n embryo. One shot of methotrexate and the whole thing vanishes.

In all these cases there is probably some rogue DNA keeping the loop going. For cancers and infections that DNA is in the tumour or bacterium. For autoimmunity the rogue DNA may be rearranged immunoglobulin genes in B clones or it may be rearranged T cell receptor genes in clones.

So maybe there is a need to kill some clones of cells.

But there are intriguing and more subtle loops that we can break. One is the pilonidal sinus. This is an abscess formed around hairs that have grown back in to a cavity in the skin. The cavity becomes inflamed and the healing response just closes it off more. Another version is an 'inclusion dermoid' where a bit of skin epithelium gets hidden inside the dermis and forms an internal surface. If you make a good enough hole into these cavities and open them up the problem vanishes. Another version is a corn. The pressure on the lump of keratinous skin pushes the base further and further in so that the lump gets thicker and thicker. If you pare away all the keratin and make sure shoes fit properly the corn vanishes.

These examples involve loops that depend purely on tissue arrangement. It is conceivable that an immune cell loop could form purely on the basis of tissue arrangement, although I think some rogue clones would help. If you have RA a knee joint may suddenly become involved and swell up badly but if you inject a high local dose of steroid that knee may go back to normal. So at least locally you can break loops.

There are lots of possibilities!
 
jnmaciuch said:
Ever since I learned that EBV deliberately binds to HLA-DR and DQ as a coreceptor for entry, I’ve had an inkling for a cross-presentation mix up story:
Click to expand...

Yes I think a cross presentation story would be plausible. I don't like mimicry stories. I am not quite sure where the DQ risk allele comes in here but maybe it is better at presenting some class of EBV fragment, perhaps for some nonspecific reason like DR4 probably being good at presenting citrullinated fragments in RA.
 
Jonathan Edwards said:
I am not sure one cn make very useful comparisons. One problem is what you take as response threshold. If you set the threshold right all trials wil have 55% 'responders'.
Click to expand...
Fair enough. Although if there is no response in some, and definitively a response in others, you might expect a somewhat clear separation if the baseline is somewhat similar.

And I don’t really understand how it’s possible to have a sustained partial response to a treatment if the treatment actually targets a lynchpin in the feedback loops.

I can understand relapses if the treatment temporarily targets a downstream affect of the real culprit, it’s the loops I struggle with.
 
Utsikt said:
I can understand relapses if the treatment temporarily targets a downstream affect of the real culprit, it’s the loops I struggle with.
Click to expand...

The trouble is that the immune system is complex enough to have loads of loops working in parallel and in series and with cross-talk. Our original 1999 hypothesis for autoimmunity suggested that basic antibody production involves at least two intertwined loops using different epitopes.

We have always said that where RA is a cycle lupus is a Swiss Watch with all sorts of epicycles. It is the one disease where you get a whole range of autoantibodies - but always to the same clutch of things.
 
Jaybee00 said:
Yes I think that Fluge doesn’t think many Dara injections are needed. That’s why they amended the pilot study to include a week 0 ,week 10 (only if there is no response), week 24 and week 48.

I think the multiple injections protocol came from multiple myeloma where I assume patients are churning out loads of bad plasma cells.
Click to expand...
I see. Whereas in ME under the theory you would have a tiny batch of them that are faulty?

It’s interesting that they believe the original dosage of 1800mg every 2W x4 is too much. I guess they saw the NK cell counts plummet after first dose and concluded that.
 
Last edited:
From the Fluge and Mella paper:

The patients with no clinical benefit from daratumumab intervention had lower numbers of NK cells in peripheral blood [...] at [...] follow-up until 40 weeks
Click to expand...

But it was still above (or close to) 125 x 10^6/L after 15 months for three of them (edit: higher than their baseline and the minimum required to take part in the phase 2 trial). The last dose was given at 30 weeks.

I wonder what would have happened if they had received another dose once their NK cells went up.
 
Last edited:
Assuming there is an actual response to Dara, could it be that the high NK cells in responders is just a downstream effect of whatever is actually mediating the effect of dara?

Are there any obvious things that would result in consistently high NK count that would also be affected by Dara?
 
We can see that after first dose at 2 weeks a few of the patients had NK cells that plummeted to 0, implying if NK cell ADCC of LLPC was the key, the second dose onwards had very little effect because there were no more NK cells to do any more work.

I also wonder WTF blue was on because her NK cells was the only one to be higher than baseline post Dara.
 

Attachments

  • Screenshot 2025-09-12 020645.png
    Screenshot 2025-09-12 020645.png
    139.6 KB · Views: 11
OrganicChilli said:
Not really. This is something that happened to 3 out of 4 non-responders if you look at the supplementary table. For some reason patient 9 did not have her NK cell count recorded after 15 months.
Click to expand...
Which implies in turn that NK cells are less sticky than we thought, and do not seem to correlate with illness severity (we have seen this from the scatterplot of SF36 baseline against NK cell count).
 
Last edited:
Jaybee00 said:
Yes I think that Fluge doesn’t think many Dara injections are needed. That’s why they amended the pilot study to include a week 0 ,week 10 (only if there is no response), week 24 and week 48.

I think the multiple injections protocol came from multiple myeloma where I assume patients are churning out loads of bad plasma cells.
Click to expand...
Hang on, where can I read about the planned dosing schedule in the P2?

It says here week 0, 2, 4, 24, 26
 

Attachments

  • Screenshot 2025-09-12 225703.png
    Screenshot 2025-09-12 225703.png
    233.6 KB · Views: 4
Utsikt said:
According to the participant info document for Dara P2 (in Norwegian), there will be five doses.
Click to expand...
Correct. The item I posted about was the amendment to the pilot study NOT the P2.

I assume they went with the more frequent dosing in phase 2 as more of a CYA strategy.
 
@Jonathan Edwards
@jnmaciuch

Just so I can understand again, what are the main critiques/evidences why you think it is not antibody related? I understand there are a few points:

1. Little detection of auto-antibodies.
2. No known mechanism for auto-antibodies to be created post infection
3. No evidence of auto-antibodies effect on the tissues (could this be autoantibodies that disrupt signalling pathways)?

Would you be able to explain abit more on these if they are the main points? And others.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom