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

[ryanc97]

Hi everyone,

So as you know I am a big fan of Fluge and Mella and the Daratumumab/Cyclophosphamide hype train. Why is because to me it looks like the remissions induced by Dara indicate that LLPC dysfunction is the cause.

My working theory is that post infection, you have this vulnerable window where any exertion triggers a faulty evolution of a batch of B cells into faulty LLPCs that produce harmful AABs for the rest of your life giving you CFS.

But then, if LLPCs are the culprit, does this not bear similarity to Lupus, which is an auto-antibody driven disease. And Daratumumab and Teclistamab have been trialed to some success in Lupus patients.

So to me, if LLPC theory is true, then it would seem Lupus and CFS are the same kind of disease, but of course, CFS QoL is much worse than Lupus QoL, so the AABs produced are probably more damaging.

An example is that you can go to the Lupus reddit and search the term 'exercise' and you get results suggesting Lupus patients can exercise no problem.

> Two other immunologic markers were recognized in the 1950s as being associated with lupus: the biologic false-positive test for syphilis and the immunofluorescent test for antinuclear antibodies. Moore, working in Baltimore, demonstrated that systemic lupus developed in 7 percent of 148 individuals with chronic false-positive tests for syphilis and that a further 30 percent had symptoms consistent with collagen disease.

I would like to ask @Jonathan Edwards on the history of Lupus, from my understanding, some scientists in the 1950s discovered that majority of Lupus patients had this ANA antibody? How does this compare to the current state of researchers in CFS looking at autoantibodies? Like Scheibenbogen?

And how does this reconcile with LLPCs or at least one of Plasmablast/SLPC/LLPC being the problem?

TLDR - Daratumumab and Teclistamab, two treatments looked at by Fluge/Mella and Habets (ok I know he is not a proper researcher) have been used in studies to treat serious Lupus patients with success.

I also find the difference in papers administering MABs to Lupus patients quite interesting. The researchers focus on the biomarker, ANA, and do not focus on the patient response, like SF36, step count etc. They use Scores on the Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K, a 24-item weighted score of lupus activity that ranges from 0 to 105, with higher scores indicating greater disease activity).

So since the treatment (targeting lymphocytes) is the same, perhaps they are similar and we can learn something about CFS by looking at the history of Lupus, it's discovery, and it's current treatments? Are they part of the same picture or related?

When someone comes in to say no AABs have been detected in CFS, is it possible the technology just isn't there to detect the AABs produced by CFS? I know we've all heard about B2ADR AABs, GPCR AABs, is there maybe hidden ones that the current tests just dont screen for?

 

[Trish]

I had to look up LLPC. I assume you mean long lived plasma cell?

 

[ryanc97]

I had to look up LLPC. I assume you mean long lived plasma cell?

Yes, these are (one of) the final end products of the B cell lineage.

 

[Jonathan Edwards]

My working theory is that post infection, you have this vulnerable window where any exertion triggers a faulty evolution of a batch of B cells into faulty LLPCs that produce harmful AABs for the rest of your life giving you CFS.

I will have a go at commenting on various bits of your idea.

As a rule infection does not trigger autoimmunity. Almost everyone thinks it does but the evidence is strongly against, with the possible atypical exception of Gullain Barre syndrome. So a parallel with lupus does not fit here - one of the cornerstones of our Qeios piece arguments.

B cells with the wrong antibody gene rearrangements occur purely by chance but it would be possible for a variety of circumstances, including possibly infection, to facilitate the expansion of bad clones with plasma cell formation. It is hard to see how exertion would trigger this though. I think it is more likely that once the immune error has developed, then exertion triggers some short term signalling pathway that is sensitised in some way by the immunology.

But then, if LLPCs are the culprit, does this not bear similarity to Lupus, which is an auto-antibody driven disease. And Daratumumab and Teclistamab have been trialed to some success in Lupus patients.

Lupus is one of many autoantibody driven diseases. If ME/CFS involves autoantibodies then it fits in with all of them maybe. I initiated the use of B cell depleting treatment for lupus in 2000 with Maria Leandro and published the first study shortly after. Rituximab works very well but trials have been badly designed until recently. So we know that some of the autoantibodies come from short lived plasma cells in lupus but some hang around and come from long lived cells. That was the basis of Maria's PhD thesis.

 

[Jonathan Edwards]

CFS QoL is much worse than Lupus QoL, so the AABs produced are probably more damaging

The real difference is that the antibodies in lupus kill you. Before steroids were available, lupus was regarded as almost always fatal. We now recognise milder forms but even with treatment lupus still has a significant mortality due directly to the effects of antibodies.

Exercise isn't much of a problem simply because the damage in lupus is relevant to exercise - kideny failure, lung failure, cardiac failure, bleeding and infection from cytopenias etc.

some scientists in the 1950s discovered that majority of Lupus patients had this ANA antibody? How does this compare to the current state of researchers in CFS looking at autoantibodies? Like Scheibenbogen?

Basically all lupus patients have antinuclear antibodies and many have anti-DNA antibodies. The lupus clinical picture does not occur without these. Very few healthy people have anti-DNA antibodies.In contrast people with ME/CFS have no major difference from healthy controls in terms of antibodies. Slight statistical differences have been reported for GPCR, that is about all.

They use Scores on the Systemic Lupus Erythematosus Disease Activity Index 2000

These scores include evidence of renal damage, blood cell counts, rash, joint swelling etc.. These provide better markers of improvement in life threatening aspects. SF36 does not really come into it if the problem is dying from a stroke or kidney failure.

When someone comes in to say no AABs have been detected in CFS, is it possible the technology just isn't there to detect the AABs produced by CFS? I know we've all heard about B2ADR AABs, GPCR AABs, is there maybe hidden ones that the current tests just dont screen for?

It is always possible that the antibodies have been missed but people have looked hard at all sorts of tissues and nothing that could explain causatioj of symptoms has come up. Which is why in our Qeios piece we suggested that if antibodies were involved it was not as autoantibody but through some more indirect mechanism.

 

[Creekside]

My working theory is that post infection, you have this vulnerable window where any exertion triggers a faulty evolution of a batch of B cells into faulty LLPCs that produce harmful AABs for the rest of your life giving you CFS.

Does that fit with abrupt switching of state (ME 100%, then 0%, then 100% again)?

 

[ryanc97]

Well this theory is just the most likely cause of action behind the success of the Dara trial. Since Dara targets plasmablast/SLPC/LLPC, basically the antibody factories, then it must be antibody related. If there are no differences, it means the test is missing it.

At least thats how I see it.

For switching, I'm quite sure that is a result of temporary symptom relief caused by some intervention. If AABs are culprit then just treating the downstream symptoms could give improvements which patients then say is remission.

 

[EndME]

If someone believes Cyclo is effective and supposed this in shown in the upcoming trial (which is a mssive "if"), then one will have to construct a theory on how it’s possible that the Rituximab trial didn’t show such results. With your belief, that would then probably be related in finding aabs that don’t go down or as long after Rituximab or finding differences in trial design that can explain different outcomes.

What do you imagine the aabs could be binding too and causing issues if you think the aabs are “more damaging”? Because in many diseases caused by antibodies the damage to the organ is very visible, but in ME/CFS there are no signs of this?

There’s probably other theories that could work just as well. One could have something as proposed in the Qeios piece by Jonathan and his colleagues or invoke some other relevance for B-cells (I once read that something like BDNF, which is supposedly important for neurological functioning, has a bidirectional feedback loop with B-cells so something of that sort might work as well).

 

[Kitty]

For switching, I'm quite sure that is a result of temporary symptom relief caused by some intervention.

As someone who's experienced it twice, I can assure you it isn't the result of any known intervention. Nothing happened in either case, not even a cold. First time round I went to bed after a normal day in an office (which I'd been doing all the years I'd been ill), and woke up to find everything had changed.

Technically remissions of 4 – 5 years are temporary, but it's better than you get from a lot of drugs for chronic illnesses. It doesn't feel like a very apt description for having your life back for nearly a decade.

 

[ryanc97]

If someone believes Cyclo is effective and supposed this in shown in the upcoming trial (which is a mssive "if"), then one will have to construct a theory on how it’s possible that the Rituximab trial didn’t show such results. With your belief, that would then probably be related in finding aabs that don’t go down or as long after Rituximab or finding differences in trial design that can explain different outcomes.

What do you imagine the aabs could be binding too and causing issues if you think the aabs are “more damaging”? Because in many diseases caused by antibodies the damage to the organ is very visible, but in ME/CFS there are no signs of this?

There’s probably other theories that could work just as well. One could have something as proposed in the Qeios piece by Jonathan and his colleagues or invoke some other relevance for B-cells (I once read that something like BDNF, which is supposedly important for neurological functioning, has a bidirectional feedback loop with B-cells so something of that sort might work as well).

It's quite simple, this is just biology, it is explained in the video. Jump to the slide at 10:24 here from Mella's video. He explains this.

if you look at the CD20 vs CD38 expression on B cells to the plasma cell, you find the early B cells have CD20 where the late stage plasmablast/SLPC/LLPC have CD38. 20% of LLPCs express CD20, quoted from Mella. So whether you get better on Ritux depends whether your faulty plasma cells have CD20 or not.

So Rituximab targets the precursors. Cyclo targets everything. Dara targets the late stage plasma cells.

Basically Olav Mella explains everything in his Charite presentation, from there, distilling a theory is straightforward. It is effectively his theory, I'm just saying it in another way.

 

[EndME]

It's quite simple, this is just biology, it is explained in the video. Jump to the slide at 10:24 here from Mella's video. He explains this.

if you look at the CD20 vs CD38 expression on B cells to the plasma cell, you find the early B cells have CD20 where the late stage plasmablast/SLPC/LLPC have CD38. 20% of LLPCs express CD20, quoted from Mella. So whether you get better on Ritux depends whether your faulty plasma cells have CD20 or not.

So Rituximab targets the precursors. Cyclo targets everything. Dara targets the late stage plasma cells.

Basically Olav Mella explains everything in his Charite presentation, from there, distilling a theory is straightforward. It is effectively his theory, I'm just saying it in another way.

I don't think this addresses any of my questions or I don't see how because the whole point is that you will still have to develop an antibody theory that is specific to these dynamics (so for example if certain GPCRs go down for months after Rituximab then those won't fit the theory or if I'm understanding correctly the theory can't involve antibodies by short lived plasma cells being the culprit because Rituximab stops the production process of these). I have understood that precursor and lineage argument (RTX targeting CD20 vs Dara targeting CD38), but I think that is exactly why the questions above remain open. That is there is no point to invoke an antibody theory that can't work, when you already think that at some step the argument has to depend on some LLPC dynamics.

 

[ryanc97]

I don't think this addresses any of my questions or I don't see how because the whole point is that you will still have to develop an antibody theory that is specific to these dynamics (so for example if certain GPCRs go down for months after Rituximab then those won't fit the theory or if I'm understanding correctly the theory can't involve antibodies by short lived plasma cells being the culprit because Rituximab stops the production process of these). I have understood that precursor and lineage argument (RTX targeting CD20 vs Dara targeting CD38), but I think that is exactly why the questions above remain open. That is there is no point to invoke an antibody theory that can't work, when you already think that at some step the argument has to depend on some LLPC dynamics.

This CD20 vs CD38 literally explains why Rituximab did not work while Dara works. I'll explain it simpler: Rituximab does not target the ALL sources of the poison while Daratumumab does.

In 20% of cases, where CD20 is expressed on toxic LLPCs, Rituximab works. This explains why Rituximab worked for some but didn't for others. Whereas for Dara, it didn't work for people with low NK cells - related to the mechanism of the drug.

Why can't the theory not involve AAB from SLPC? You realize there are multiple sources of AAB, plasmablast/SLPC/LLPC. Ritux can deplete some SLPC, sure good, you have some symptom improvement, but the main culprit of LLPC is untouched.\

For AABs, it is quite clear there is no permanent damage in CFS. So it is causing havoc by some signalling blocking or something.

 

[Jonathan Edwards]

Did cyclophosphamide treatment reduce Ig levels? Generally you have to give a lot of cylophosphamide to get significant Ig drops.

 

[EndME]

This CD20 vs CD38 literally explains why Rituximab did not work while Dara works. I'll explain it simpler: Rituximab does not target the ALL sources of the poison while Daratumumab does.

Yes, this all understood.

In 20% of cases, where CD20 is expressed on toxic LLPCs, Rituximab works. This explains why Rituximab worked for some but didn't for others.

This argument doesn't seem sensible to me because you'd have to develop an understanding of why leaving 80% of the problematic factories untouched can result in success which probably requires a deeper understanding of the dynamics. I can understand that there might be loops that respond differently to the same mediciation across individuals but assuming from that that Rituximab must have worked for some seems to be quite a jump (the population level evidence comes first). Apart from that I also think this whole argument is in contradiction to your suggestion, because as far as I can tell it suggests that the situation is different to Lupus, because as I understand there you generally need a continued course of Rituximab, because something retains a memory to make things reappear, however it seems that the improvements in the ME/CFS Rituximab trials were permanent, which is rather the opposing observation.

Why can't the theory not involve AAB from SLPC? You realize there are multiple sources of AAB, plasmablast/SLPC/LLPC. Ritux can deplete some SLPC, sure good, you have some symptom improvement, but the main culprit of LLPC is untouched.\

The point I was trying to make is that the theory you want to develop will have to involve the main culprit. So there is no point in looking at things that can't be the main culprit to begin with. For example, those that think certain GPCR aabs are relevant or anti-DNA, and they all effectivitely go down nicely after RTX then there's no point to invoke them in a straightforward way and it's more sensible to focus on those that don't, for example like those involved in Sjorgens (according to Google). You mentioned Scheibenbogen and her antibodies above, from what I've understood all antibodies she has claimed to be of relevance fall after RTX so they automatically have to irrelevant to such a theory.

 

[Jonathan Edwards]

You seem to be not grasping the fact that Rituximab works in some and not in others.

We don't have reliable evidence for rituximab working in any ME/CFS cases. Some people became well and remained well after rituxmab. Others relapsed after a period of months. But we don't know whether in either case it had anything to do with the rituximab.

Mella is speculating on what might explain things, but the absence of any difference in the phase 3 trial makes it much most likely that it simply does not work.

 

[ryanc97]

This argument doesn't seem sensible to me because you'd have to develop an understanding of why leaving 80% of the problematic factories untouched can result in success which probably requires a deeper understanding of the dynamics.

I never said this. I said this results in failure. Do read the above

Apart from that I also think this whole argument is in contradiction to your suggestion, because as far as I can tell it suggests that the situation is different to Lupus, because as I understand there you generally need a continued course of Rituximab, because something retains a memory to make things reappear, however it seems that the improvements in the ME/CFS Rituximab trials were permanent, which is rather the opposing observation.

"In 20% of cases, where CD20 is expressed on toxic LLPCs, Rituximab works PERMANENTLY. This explains why Rituximab worked for some but didn't for others. Whereas for Dara, it didn't work for people with low NK cells - related to the mechanism of the drug."

 

[ryanc97]

We don't have reliable evidence for rituximab working in any ME/CFS cases. Some people became well and remained well after rituxmab. Others relapsed after a period of months. But we don't know whether in either case it had anything to do with the rituximab.

Mella is speculating on what might explain things, but the absence of any difference in the phase 3 trial makes it much most likely that it simply does not work.

I don't think a failed P3 negates the fact it worked for a subset of people in P2 and P1. Hence any theory must account for this. The 20% of LLPC expressing CD20 accounts for this. It could be for a majority it doesn't work because for some unknown reason, their faulty LLPCs don't have CD20!

I am just fitting a theory to ALL the data (Ritux P1,P2,P3, cyclo P1, Dara P1). This is called maximum likelihood estimation in statistics. Albeit there is no parametric model, I am just using logic.

An interesting point that does involve biology is Mella said up to 20% of LLPCs express CD20. Is this due to age? Is CD20 expression correlated to age of the plasma cell?

 

[Jonathan Edwards]

I don't think a failed P3 negates the fact it worked for a subset of people in P2 and P1.

It very much does. People got better in phase 1 and 2 trials but it can never be assumed that any specific mechanism of the treatment caused this. That is the whole point of our trial policy.

I had a detailed look at the data profiles over time from the phase 2 study. there was no consistent pattern in the blinded phase in terms of response timing. That to me suggests that there was no causal effect of the drug.

 

[ryanc97]

Did cyclophosphamide treatment reduce Ig levels? Generally you have to give a lot of cylophosphamide to get significant Ig drops.

For some reason, the paper 2015 cyclo trial does not show IGG levels at all.

 

[ryanc97]

It very much does. People got better in phase 1 and 2 trials but it can never be assumed that any specific mechanism of the treatment caused this. That is the whole point of our trial policy.

I had a detailed look at the data profiles over time from the phase 2 study. there was no consistent pattern in the blinded phase in terms of response timing. That to me suggests that there was no causal effect of the drug.

Where did you find individual time series data of the patients in P2? I can't find it in the paper.

Also I disagree, I think that some patients had remission on Ritux means something is going on.

Also, if Ritux is said to not work, that suits the theory fine because you can say Rituximab doesn't touch LLPCs. Or at least the 20% of LLPCs with CD20 are not faulty.

And also, if P3 is definitive to say whether something works or not, how many drugs have a P3 for CFS?