De Meirleir is using Rituximab?

[Jonathan Edwards]

My problem is that there seems to be evidence that supports both sides interpretations. If it was just a question of being ´in a position to assess it,´ then surely all the immunologists would agree with you?

What evidence runs contrary to what I am suggesting? I totally agree that B cells produce cytokines but they do so in environments where 'inflammation' does not apply. Most immunologists are not in a position to understand this, it seems. Whether it is because they are not very good at putting together complex dynamic ideas or whether they have just had no training in tissue microanatomy I have never really understood but I suspect it is the former. Immunodynamics are so complicated that most people simply cannot get their heads around it. Honest people like my wife, who is highly intelligent and a medical expert in another field, just say they cannot handle it. The problem is that people who have become immunologists who are not able to get their heads around the complexity latch on to simple catchy ideas in review articles without worrying about the fct that they do not add up.

Perhaps a practical point would help. Cytokines are signals from one cell to another. In general they operate in microenvironments - so the concentration generated in a cell cluster of volume 1 nanolitre may be millions or even billions of times higher than the concentration resulting when the cytokine diffuses out into blood and ECF. So cytokines in blood in general do not cause inflammation.

What is inflammation? It is an increase in permeability to fluid of blood vessels and an exit of white cells into the tissue. What would that be in spleen, where B cell produce cytokines? The problem is that splenic vessels are totally permeable anyway and the white pulp tissue is totally made up of white cells anyway. So you cannot really have inflammation of the spleen. The spleen is a bag of inflammation to start with in a sense.

The other key point is that cytokine production by B cells is mostly just a step in the co-operative process that leads to antibody production. So blocking B cell cytokines is just blocking antibody production and if you take away all the B cells it does not matter which step you think you are blocking because all the steps have gone. It would be different if B cells stimulated T cells to go off and cause inflammation in autoimmunity on their own. The problem with this is that nobody has ever found any self-reactive t cells that might do this in autoimmunity. Virtually all immunologists believe in autoreactive T cells despite none ever having been found, just as people believe in gods and free will despite them never having been found. But to take autoreactive T cells seriously we need some empirical evidence for their existence.

At root I guess the problem may be that not that many people in academic disciplines are actually any good at assessing what theory is truly supported by evidence. That may seem surprising but the 2008 crash showed it to the case for 95% of economists and there does not seem to be a single psychiatrist in the UK capable of seeing that the evidence for using CBT in ME is zero. And then there is Donal Trump running for president. Human discernment does seem to be a bit unreliable in a range of fields! By and large people go on believing what they want to believe.

 

[msf]

Well, the thing is I don´t know whether the evidence for one view is better than the other. If they need to prove autoreactive T cells exist, don´t you also need to prove that there are autoantibodies in RA?

Here is some (possibly circumstantial) evidence for B cells causing inflammation in MS: http://www.ncbi.nlm.nih.gov/pubmed/...tiple+sclerosis+and+B+cell+depletion+therapy,

Anyway, I´m glad that I don´t need to decide whose model is correct, as I do not see myself getting Rituximab anytime in the near to medium-term future.

 

[Chrisb]

As a result I am only too happy to act as an expert witness for the defence in cases of negligence in relation to inappropriate use.

@Jonathan Edwards

Is that quite what you mean? Plaintiff, perhaps?

 

[Jonathan Edwards]

Well, the thing is I don´t know whether the evidence for one view is better than the other. If they need to prove autoreactive T cells exist, don´t you also need to prove that there are autoantibodies in RA?

The existence of auto-antibodies in RA has been documented since 1948, with thousands of papers describing the exact binding sites, the affinities, the subclasses, the physical chemistry of complex formation etc etc. Nobody has the slightest doubt about their existence. There are standard diagnostic tests based on their measurement used in thousands of hospitals across the world. They are called rheumatoid factors. My point is that although immunologists believe that the T cell equivalent underlies the presence of the antibody it has never been found, despite forty years of looking and getting negative results. It is a bit like the difference between Starbucks paper cups and the Holy Grail. One we have evidence for, the other we do not! It is quite extraordinary how blind people seem to be to the difference, but one learns that that is the way the human mind tends to work.

 

[Jonathan Edwards]

@Jonathan Edwards

Is that quite what you mean? Plaintiff, perhaps?

Indeed, I got it the wrong way around.

 

[Jonathan Edwards]

Here is some (possibly circumstantial) evidence for B cells causing inflammation in MS: http://www.ncbi.nlm.nih.gov/pubmed/?term=R. Li et al., “Pro-inflammatory GM-CSF–producing B cells in multiple sclerosis and B cell depletion therapy,

MS is interesting in that it may not be so much a disease of autoantibody production as a disease of B cell survival in brain. What we do not know anything much about is why B cells should be able to survive in brain in MS patients but since it occurs in apparently random episodes in adult life it seems likely that the survival is a function of a specific surface antibody. That would explain why the same antibody species are found in the CSF in MS patients repeatedly over many years - the oligoclonal bands. It does not seem to be a matter of a cytokine functional subset - which ought to lead to a polyclonal picture.

In this context the Canadian paper is interesting and I am waiting to hear more on this. I am not holding my breath becuase I have never heard of any of these people and there are no numerical data in their abstract. Moreover, they start out with this same mantra about suddenly discovering B cells do other things (which we knew forty years ago). They also talk of IL-10 producing regulatory B cells which I am pretty sure do not really exist. But somewhere in this story there may be something new and relevant.

It seems very plausible that if B cells are allowed to survive in brain they will cause inflammation there. One mechanism was described in the 1990s - local antibody production, regardless of specificity, can cause demyelination through microglial activation. So maybe B cells in brain would be pro-inflammatory. But then the crucial question is how B cells come to survive in MS brains when presumably they are normally forbidden from surviving in brain - or maybe even entering brain in the normal situation, regardless of their capacity to produce cytokines.

A crucial piece of evidence about MS with rituximab, which we predicted when first suggesting its use in 2002, is that new lesions stop immediately after rituximab is given - in contrast to all other autoimmune diseases where improvement takes weeks. That means that MS lesions are not due to circulating antibody but to the effects of B cells that have accessed brain locally. Whether the local action is through antibody or cytokine we do not know and certainly in this case I would be happy with either possibility as we have nothing to distinguish at present.

 

[msf]

The existence of auto-antibodies in RA has been documented since 1948, with thousands of papers describing the exact binding sites, the affinities, the subclasses, the physical chemistry of complex formation etc etc. Nobody has the slightest doubt about their existence. There are standard diagnostic tests based on their measurement used in thousands of hospitals across the world. They are called rheumatoid factors. My point is that although immunologists believe that the T cell equivalent underlies the presence of the antibody it has never been found, despite forty years of looking and getting negative results. It is a bit like the difference between Starbucks paper cups and the Holy Grail. One we have evidence for, the other we do not! It is quite extraordinary how blind people seem to be to the difference, but one learns that that is the way the human mind tends to work.

Sorry, I was a bit vague - some of the things I have read stated that there is not much evidence for RF having a causative role in RA. But then I just found a summary that says that there is considerable evidence for this - so back to square one!

 

[Jonathan Edwards]

Sorry, I was a bit vague - some of the things I have read stated that there is not much evidence for RF having a causative role in RA. But then I just found a summary that says that there is considerable evidence for this - so back to square one!

Fair enough. Up until 1996 I was unimpressed by the idea that rheumatoid factors had a causative role. I was unimpressed by lymphocytes and immunology in general - partly because immunologists seemed to talk hot air most of the time. I was proud of not being an immunologist and worked on stromal fibroblasts and macrophages and their role in RA. Then I gave a presentation at Grand Rounds on the specialised stromal features of joint tissue and walking back to my office I realised that I had just given a very powerful set of reasons for thinking that the inflammation in RA must be due to immune complexes. My next thought was that it would be interesting to know if anyone had done a study of ocal expression of immunoglobulin receptors in joint tissue. So I scanned PubMed and to my surprise came up with a nice paper on this on which I was an author! I had completely forgotten that in 1988 a visiting German PhD student had looked at this in collaboration with myself and her immunology supervisor. So I discovered that I had to rather rapidly become an immunologist after all and specifically a B cell immunologist. That wasn't too hard because I had been working alongside Jo Cambridge for ten years and in the department where thyroid autoantibodies were first reported so I had been fairly immersed in the subject even if I had tried to keep it at arms length.

It is still not clear whether rheumatoid factors or anti-citrillinated protein antibodies are the more pathogenic in RA but we are pretty sure that autoantibodies have to play two distinct complementary roles in most autoimmune diseases and these roles are not necessarily closely tied to specificity. It looks likely that subsets of both types of antibody can play both roles but it is very hard to prove since interventions like rituximab tend to affect all antibody types to a degree. What we really need are physical chemistry techniques that actually show the binding kinetics in relevant contexts. There was a guy called Kunkel who made huge progress in this line in the 1960s and 1970s but the techniques are hugely labour intensive and by 1983 people just dropped them and switched to quick and easy ELISAs. With about ten million dollars one could synthesise the right probes and tune up the right detection apparatus but nobody was interested ten years ago so I got out of science. The time is probably still not right, because science politics has become almost as trivial as the US Republican primaries, but maybe in a decade someone will get down to showing exactly what does what.

 

[msf]

Thanks for taking the time to explain your position, Prof. Edwards. A slight quibble though - I thought you had previously said that anti-citrullinated protein antibodies weren´t proper autoantibodies? Or is that just semantics?

 

[Jonathan Edwards]

Thanks for taking the time to explain your position, Prof. Edwards. A slight quibble though - I thought you had previously said that anti-citrullinated protein antibodies weren´t proper autoantibodies? Or is that just semantics?

Indeed they are not proper autoantibodies but they can fulfil the roles ascribed to autoantibodies. One could even claim that anti-DNA antibodies are not proper autoantibodies I guess. All these words are shorthand in one way or another.

 

[perrier]

MS is interesting in that it may not be so much a disease of autoantibody production as a disease of B cell survival in brain. What we do not know anything much about is why B cells should be able to survive in brain in MS patients but since it occurs in apparently random episodes in adult life it seems likely that the survival is a function of a specific surface antibody. That would explain why the same antibody species are found in the CSF in MS patients repeatedly over many years - the oligoclonal bands. It does not seem to be a matter of a cytokine functional subset - which ought to lead to a polyclonal picture.

In this context the Canadian paper is interesting and I am waiting to hear more on this. I am not holding my breath becuase I have never heard of any of these people and there are no numerical data in their abstract. Moreover, they start out with this same mantra about suddenly discovering B cells do other things (which we knew forty years ago). They also talk of IL-10 producing regulatory B cells which I am pretty sure do not really exist. But somewhere in this story there may be something new and relevant.

It seems very plausible that if B cells are allowed to survive in brain they will cause inflammation there. One mechanism was described in the 1990s - local antibody production, regardless of specificity, can cause demyelination through microglial activation. So maybe B cells in brain would be pro-inflammatory. But then the crucial question is how B cells come to survive in MS brains when presumably they are normally forbidden from surviving in brain - or maybe even entering brain in the normal situation, regardless of their capacity to produce cytokines.

A crucial piece of evidence about MS with rituximab, which we predicted when first suggesting its use in 2002, is that new lesions stop immediately after rituximab is given - in contrast to all other autoimmune diseases where improvement takes weeks. That means that MS lesions are not due to circulating antibody but to the effects of B cells that have accessed brain locally. Whether the local action is through antibody or cytokine we do not know and certainly in this case I would be happy with either possibility as we have nothing to distinguish at present.

Dr Wahls seems to have had some luck improving her MS with a very rigid diet. Do you think this is possible? Or is she having some inexplicable remission.

 

[Jonathan Edwards]

Dr Wahls seems to have had some luck improving her MS with a very rigid diet. Do you think this is possible? Or is she having some inexplicable remission.

I would have thought her remission was entirely explicable on the basis that you can never predict whether someone with MS will have another attack or when!

 

[msf]

So, I looked at this thread again, and read over the points Prof. Edwards made. It seems like he is very sure of the specific local role of cytokine production by B cells, and who am I to dispute that? I had another thought though - rather than by affecting the general cytokine balance, could Ritux be having its effect through destroying antibodies against a pathogen, rather than by destroying auto-antibodies? By destroying these antibodies, a possible non-optimal response to a pathogen would be altered, wouldn´t it? Is there any way of determining which of these is happening, except for saying that if it was the latter, the patient would be expected to get worse? As the articles I posted above show, that did not seem to be the case with some infections and Ritux. Also, the idea of a pathogen could be misleading here - what if the pathogen in question is actually a commensal in others, and it is the non-optimal response that makes it a pathogen? Sorry if I am asking the same sort of questions again, but it is still not clear to me that dampening the body´s inflammatory response to a pathogen is necessarily a bad thing for every patient in every case. For instance, if the body´s response to sepsis was diminished, wouldn´t patients be expected to survive longer?

 

[Jonathan Edwards]

So, I looked at this thread again, and read over the points Prof. Edwards made. It seems like he is very sure of the specific local role of cytokine production by B cells, and who am I to dispute that? I had another thought though - rather than by affecting the general cytokine balance, could Ritux be having its effect through destroying antibodies against a pathogen, rather than by destroying auto-antibodies? By destroying these antibodies, a possible non-optimal response to a pathogen would be altered, wouldn´t it? Is there any way of determining which of these is happening, except for saying that if it was the latter, the patient would be expected to get worse? As the articles I posted above show, that did not seem to be the case with some infections and Ritux. Also, the idea of a pathogen could be misleading here - what if the pathogen in question is actually a commensal in others, and it is the non-optimal response that makes it a pathogen? Sorry if I am asking the same sort of questions again, but it is still not clear to me that dampening the body´s inflammatory response to a pathogen is necessarily a bad thing for every patient in every case. For instance, if the body´s response to sepsis was diminished, wouldn´t patients be expected to survive longer?

I don't think I said anything about affecting cytokine balance - which is a concept I do not believe in.

Rituximab does not destroy antibodies, or even plasma cells. It destroys B cells going through maturation. If a pathogen is present there is normally a mature antibody response with plasma cell differentiation within 3 weeks. So ay time after that one would not expect rituximab to have much effect on antibodies to microbes. And we have a large body of evidence indicating it has virtually no effect on antibodies to microbes. Rituximab would only interfere with a response to a microbe if the microbe arrived after the rituximab was given, so not a microbe responsible for the illness being treated.

If you dampen the response to sepsis, as was the plan for TNF inhibitors to treat sepsis, the patient dies very quickly indeed. Anti-TNF for sepsis was a complete disaster. There are a few very specific situations where response to infection is counterproductive - TB is the best example - but in general not.

 

[msf]

Hehe, you caught me using short-hand again. I am aware of Rituximab´s target.

So how come an ongoing autoimmune disease ´needs´ an ongoing ´supply´ of antibodies, but an ongoing response to an ongoing infection (which is what I was suggesting) doesn´t?

Thanks for the sepsis example. I guess some people thought it was worth a try though, which leads me to the next question: if dampening the response can help (as in your TB example), how can we be sure it won´t in any particular disease? It may be unlikely, but it still seems like a possibility worth keeping in mind.

Oh, and the cytokine balance stuff was my suggestion - I wasn´t trying to assign it to anyone else.

 

[Marky90]

So how come an ongoing autoimmune disease ´needs´ an ongoing ´supply´ of antibodies, but an ongoing response to an ongoing infection (which is what I was suggesting) doesn´t?

Just chiming in to guess that it`s simply because the antibodies to self-X always have X present, whilst that`s not the case for infection. Where the antibodies don`t become active before immunological memory is alerted of the non self presence.

 

[msf]

Just chiming in to guess that it`s simply because the antibodies to self-X always have X present, whilst that`s not the case for infection. Where the antibodies don`t become active before immunological memory is alerted of the non self presence.

It is the case in the case I mentioned above, that of an ongoing response to an ongoing infection.

 

[Daffodil]

wow. interesting that KDM has changed his view a little. he told me earlier on that rituxan would result in a bunch of deaths. wonder why he never brought it up as a possibility for me. maybe he uses it on the sickest, bedridden patients.?

maybe it can help with symptoms even if it depletes some B cells? maybe he gives a little and then does it again every few months?

 

[Gingergrrl]

wonder why he never brought it up as a possibility for me.

From my own experience, it would only be used in patients with rock solid evidence of autoimmune antibodies. It seems that KDM tests for Lyme and pathogens and his focus is very different. If I did not have proof of auto-antibodies from three different labs spanning the course of a year, combined with significant improvement from an autoimmune dose of IVIG, we would not be pursuing Rituximab and I would agree that it is too risky. But in certain cases, I feel it is worth the risk, especially if every precaution known to man is taken along with it

 

[Jonathan Edwards]

wow. interesting that KDM has changed his view a little. he told me earlier on that rituxan would result in a bunch of deaths. wonder why he never brought it up as a possibility for me. maybe he uses it on the sickest, bedridden patients.?

maybe it can help with symptoms even if it depletes some B cells? maybe he gives a little and then does it again every few months?

Dr De Meirleir has made it clear he has no idea about immunology. Presumably he says whatever he thinks will impress the individual patient. Giving a little is in my view tantamount to medical negligence and I doubt he has ever tried that.