CFI Spinal Fluid study from Lipkin and Hornig is out.
- Thread starter aimossy
- Start date
@Valentijn
Could be. I'm around yr 20 now, so I'm definitely in the chronic phase. My sed rates were never high though. I have some of the very old records. I think I got up to a 4 once maybe. It could be that yours has a somewhat different pathogenesis than most, converging on a common pathway that gives symptoms, or just that your genetics alter your response. Hard to know. Most people with RA have high sed rates - but a few don't. They still have RA.
I'd love to be able to start correlating symptoms and test results with genetics via GWAS. That shouldn't be too hard in this day and age with 23andme costing around $100 and the computer power that we now have available to us. I know Kerr has really tried to subtype with the hope of getting more statistically significant results. I admire his work a lot. Unfortunately when he looked for SNP's, he only looked in the genes that showed altered expression in the previous study (I think - could be off on that). Interestingly, 23andme did ask me a question or 2 about whether I had CFS - so they are likely looking for associations. With enough people, you shouldn't need to subtype.
I do think the initial stages of ME are more pro-inflammatory. For example, I ran fevers for a few years every day (not too high, around 99-100.5) I had a frontal headache, and more obvious neurological symptoms. My morning temp now is often 95, and I rarely go much over 97-98. Early on, there is probably an excess of interferon-gamma, which is both Th1 polarizing and a Th1 effector cytokine.
Maybe you'll never go into the long stage. I think the long stage is marked by immune anergy in which the immune system can no longer mount an effective Th1 response, or at least it needs excessive stimulation to do so. That's speculative though.
Could be. I'm around yr 20 now, so I'm definitely in the chronic phase. My sed rates were never high though. I have some of the very old records. I think I got up to a 4 once maybe. It could be that yours has a somewhat different pathogenesis than most, converging on a common pathway that gives symptoms, or just that your genetics alter your response. Hard to know. Most people with RA have high sed rates - but a few don't. They still have RA.
I'd love to be able to start correlating symptoms and test results with genetics via GWAS. That shouldn't be too hard in this day and age with 23andme costing around $100 and the computer power that we now have available to us. I know Kerr has really tried to subtype with the hope of getting more statistically significant results. I admire his work a lot. Unfortunately when he looked for SNP's, he only looked in the genes that showed altered expression in the previous study (I think - could be off on that). Interestingly, 23andme did ask me a question or 2 about whether I had CFS - so they are likely looking for associations. With enough people, you shouldn't need to subtype.
I do think the initial stages of ME are more pro-inflammatory. For example, I ran fevers for a few years every day (not too high, around 99-100.5) I had a frontal headache, and more obvious neurological symptoms. My morning temp now is often 95, and I rarely go much over 97-98. Early on, there is probably an excess of interferon-gamma, which is both Th1 polarizing and a Th1 effector cytokine.
Maybe you'll never go into the long stage. I think the long stage is marked by immune anergy in which the immune system can no longer mount an effective Th1 response, or at least it needs excessive stimulation to do so. That's speculative though.
- Messages
- 5,256
- Likes
- 32,032
Some interesting points Eeyeore. I hoped to make it clear, as I have said before, that not all inflammation gives a high CRP (nettle sting mediated by histamine does not) but I think I said cytokine mediated inflammation, and in my experience when cytokines get switched on IL-6 and CRP tend to get brought in. CRP is driven by IL-6. I am afraid I do not buy this stuff about TH1, TH2 and TH17, since most of the cytokine is macrophage driven in most cases. The Ying, Yang and then Yong and Yung of TH subsets has been shifting around for twenty years and as far as I can see never had much to do with any real human disease. In thirty odd years of looking after seronegative spondarthropathy I am not sure that I saw much inflammation without a raised CRP, but that takes into account the different calibration of CRP in different individuals and I would take a level of 3 as potentially as abnormal as one of 15 in an individual with a low response sensitivity.
But I do agree with the basic point that not all inflammation is cytokine driven. I think the point I was trying to make was that I am not convinced that IL-8 levels 100 times normal in the blood are likely to be coming from some special IL-8 type inflammation in the brain. I think you would agree. It seems that in CSF they did not find IL-8 up much either.
I would be cautious about the trendy talk of microglia not coming from bone marrow. That is mostly based on mice and I learnt in 1979 when studying macrophage ontogeny using bone marrow transplants in mice that you can get some misleading results with these kinetic studies. My reading of the literature on human brain mononuclear phagocytes is that it is all a bit more complicated and bone marrow derived cells do come in during significant inflammation. BUt it is all circumstantial evidence because you cannot do the relevant kinetic studies on people.
But I do agree with the basic point that not all inflammation is cytokine driven. I think the point I was trying to make was that I am not convinced that IL-8 levels 100 times normal in the blood are likely to be coming from some special IL-8 type inflammation in the brain. I think you would agree. It seems that in CSF they did not find IL-8 up much either.
I would be cautious about the trendy talk of microglia not coming from bone marrow. That is mostly based on mice and I learnt in 1979 when studying macrophage ontogeny using bone marrow transplants in mice that you can get some misleading results with these kinetic studies. My reading of the literature on human brain mononuclear phagocytes is that it is all a bit more complicated and bone marrow derived cells do come in during significant inflammation. BUt it is all circumstantial evidence because you cannot do the relevant kinetic studies on people.
I've crossed the 20 year mark. In the first 10 years or so I was much more functional than now but I was more "obviously" ill as I had mild fevers daily in the early afternoons with sore throat, headache, confusion etc. Flu-like illnesses occurred every 3 weeks or so. Makes me think some kind of pathogen lifecycle. Eventually the whole thing morphed into this low body temperature anti-inflammatory state. CRP zero etc. The only inflammation I get these days is of the histamine kind.
I don't think the 3 year mark is carved in stone. It would be interesting to see a survival analysis type of analysis for IFN gamma in this dataset. I see patients on here and elsewhere who have viral symptoms 30 years into this thing while others progress to immune crapout a mere few months after onset. The immune system caves in at different rates in every patient depending on all kinds of factors probably including use of immune modulators. These days I'm using various IFN gamma inducers and I feel much more like I did in the first 10 years of my illness. The more I shift towards "Th1", the more functionality / muscle use I regain, by the way, even though I feel rubbish in the way cytokines make you feel.
I don't think the 3 year mark is carved in stone. It would be interesting to see a survival analysis type of analysis for IFN gamma in this dataset. I see patients on here and elsewhere who have viral symptoms 30 years into this thing while others progress to immune crapout a mere few months after onset. The immune system caves in at different rates in every patient depending on all kinds of factors probably including use of immune modulators. These days I'm using various IFN gamma inducers and I feel much more like I did in the first 10 years of my illness. The more I shift towards "Th1", the more functionality / muscle use I regain, by the way, even though I feel rubbish in the way cytokines make you feel.
@Jonathan Edwards -
We definitely agree on the uselessness of these panels returning absurdly high numbers for IL-8 and other cytokines. I also agree that the inter-lab variability is so high that controls and patients would have to be run (blinded) by the same lab to test if the numbers mean anything at all. I do not believe there is a magic site in the brain producing IL-8 in mass quantities (and I'm not even sure how readily IL8 can cross the BBB - if at all - I just don't know).
Yes, cytokines are partly macrophage driven, although they are also driven by CD4 cells and many other immune and non-immune cells. Certain cytokines are more strongly linked to macrophages or T-cells (or many other cells, immune and otherwise). I think the Th subset theory is pretty well accepted now. It's in Janeway and Kuby and most of the modern immunology textbooks as established fact (although some newer Th subsets are less established, e.g. Th3 and Th9), and studies have been published and well replicated in the big journals. A lot of recent work has distinguished various subsets of macrophages, initially described as M1 and M2, and the literature is very persuasive and the studies are replicated. M2 is now being subdivided, including both allergy-type polarizations and "deactivated" phenotypes. IFN-gamma + LPS clearly derives macrophages to an M1 phenotype, which is part of a Th1 type imflammatory response, producing il-6 and TNF-alpha - and as expected, elevating CRP and producing signs of inflammation. IL10 produces a deactivated M2 type phenotype, MHCII molecules are downregulated or not expressed, and TGF-beta is upregulated, probably playing a role in matrix remodelling and tissue repair. The macrophage polarization work is not as well developed as Th subsets, but it's coming along - it's not just a few small studies by any means. It does make sense that this should be the case - as you point out, macrophages are key in orchestrating the cytokine milieu and the type of immune response. I would expect similar properties of dendritic cells, although I have not read much about that yet.
There has been a lot of work on autoimmunity and Th17 cells. It seems to play a key role in seronegative autoinflammatory conditions, such as psoriasis. There are a number of IL-17 mabs in development right now - I believe one just got approved (sekukinumab I think). They have generally proven to be far more effective in psoriasis than any other drugs developed to date, including stelara, and far more effective than tnf-alpha mabs. IL-17 is an effector cytokine of an activated Th17 CD4 cell - IL23 is a polarizing cytokine, which, in the presence of TGF-beta, will polarize naive T-cells towards a Th17 phenotype. Il-23 mabs exist already for treatment (ustekinumab). It's well known that polymorphisms in the IL-23 receptor predispose to autoimmune disease, especially psoriasis, AS, PsA, USpA, and IBD, but seem to have a role in a wide array of autoimmune diseases.
Leprosy is even a good example for Th1 vs Th2 - very different illnesses (lepromatous vs tuberculoid). Similar mechanisms are probably involved with TB (unsurprisingly, since they are both mycobacteria). Systemic sclerosis is initially an inflammamtory Th1 disease, and later transitions to a Th2 disease during which there is extensive matrix deposition and fibrosis.
The Th subset paradigm (I would argue as broader markers of immune polarization) explains a wide array of illnesses that are clinically observed, as well as variation in their presentation. I believe that alterations in host immunity, possibly genetic in origin, are far more likely to play a role in ME/CFS than any particular undiscovered pathogen. If it were the latter, I believe we'd have found it long ago. I don't know many rheumatologists measuring cytokines - I think it's still mostly at a research level - and outside of furthering understanding of ME, I do not think it's likely that cytokine testing is of much use right now clinically.
I agree that the evidence strongly suggests that during periods of inflammation, macrophages/monocytes do cross the BBB and play a role in CNS immunity. I've also read a number of studies that show this. I have yet to read anything conclusive on whether or not those cells can then become microglia (doubtful) or eventually just die off. Normal maintenance in the CNS is performed by microglia, which are a self-replicating (based on my reading of the literature to date) population of cells that remain in the CNS.
While cytokine mediated inflammation is usually accompanied by elevations in IL-6 and CRP, I am not sure we can conclusively state that that is always the case. I happen to not believe that ME is an inflammatory condition. However, I think one must allow for the possibility that altered immune signalling could create aberrant cytokine fingerprints. Lipkin and Hornig showed that there was significant decoupling of expression of many cytokines. Klimas has also shown this (saw it at the conference, not sure if it's published yet). So, as you point out, IL-6 generally gets pulled in - but it's possible it might not if key cytokine expression patterns have become decoupled. Again, though, I do not believe ME is an inflammatory disease. I just think we have to be careful making assumptions from experience in other diseases.
We definitely agree on the uselessness of these panels returning absurdly high numbers for IL-8 and other cytokines. I also agree that the inter-lab variability is so high that controls and patients would have to be run (blinded) by the same lab to test if the numbers mean anything at all. I do not believe there is a magic site in the brain producing IL-8 in mass quantities (and I'm not even sure how readily IL8 can cross the BBB - if at all - I just don't know).
Yes, cytokines are partly macrophage driven, although they are also driven by CD4 cells and many other immune and non-immune cells. Certain cytokines are more strongly linked to macrophages or T-cells (or many other cells, immune and otherwise). I think the Th subset theory is pretty well accepted now. It's in Janeway and Kuby and most of the modern immunology textbooks as established fact (although some newer Th subsets are less established, e.g. Th3 and Th9), and studies have been published and well replicated in the big journals. A lot of recent work has distinguished various subsets of macrophages, initially described as M1 and M2, and the literature is very persuasive and the studies are replicated. M2 is now being subdivided, including both allergy-type polarizations and "deactivated" phenotypes. IFN-gamma + LPS clearly derives macrophages to an M1 phenotype, which is part of a Th1 type imflammatory response, producing il-6 and TNF-alpha - and as expected, elevating CRP and producing signs of inflammation. IL10 produces a deactivated M2 type phenotype, MHCII molecules are downregulated or not expressed, and TGF-beta is upregulated, probably playing a role in matrix remodelling and tissue repair. The macrophage polarization work is not as well developed as Th subsets, but it's coming along - it's not just a few small studies by any means. It does make sense that this should be the case - as you point out, macrophages are key in orchestrating the cytokine milieu and the type of immune response. I would expect similar properties of dendritic cells, although I have not read much about that yet.
There has been a lot of work on autoimmunity and Th17 cells. It seems to play a key role in seronegative autoinflammatory conditions, such as psoriasis. There are a number of IL-17 mabs in development right now - I believe one just got approved (sekukinumab I think). They have generally proven to be far more effective in psoriasis than any other drugs developed to date, including stelara, and far more effective than tnf-alpha mabs. IL-17 is an effector cytokine of an activated Th17 CD4 cell - IL23 is a polarizing cytokine, which, in the presence of TGF-beta, will polarize naive T-cells towards a Th17 phenotype. Il-23 mabs exist already for treatment (ustekinumab). It's well known that polymorphisms in the IL-23 receptor predispose to autoimmune disease, especially psoriasis, AS, PsA, USpA, and IBD, but seem to have a role in a wide array of autoimmune diseases.
Leprosy is even a good example for Th1 vs Th2 - very different illnesses (lepromatous vs tuberculoid). Similar mechanisms are probably involved with TB (unsurprisingly, since they are both mycobacteria). Systemic sclerosis is initially an inflammamtory Th1 disease, and later transitions to a Th2 disease during which there is extensive matrix deposition and fibrosis.
The Th subset paradigm (I would argue as broader markers of immune polarization) explains a wide array of illnesses that are clinically observed, as well as variation in their presentation. I believe that alterations in host immunity, possibly genetic in origin, are far more likely to play a role in ME/CFS than any particular undiscovered pathogen. If it were the latter, I believe we'd have found it long ago. I don't know many rheumatologists measuring cytokines - I think it's still mostly at a research level - and outside of furthering understanding of ME, I do not think it's likely that cytokine testing is of much use right now clinically.
I agree that the evidence strongly suggests that during periods of inflammation, macrophages/monocytes do cross the BBB and play a role in CNS immunity. I've also read a number of studies that show this. I have yet to read anything conclusive on whether or not those cells can then become microglia (doubtful) or eventually just die off. Normal maintenance in the CNS is performed by microglia, which are a self-replicating (based on my reading of the literature to date) population of cells that remain in the CNS.
While cytokine mediated inflammation is usually accompanied by elevations in IL-6 and CRP, I am not sure we can conclusively state that that is always the case. I happen to not believe that ME is an inflammatory condition. However, I think one must allow for the possibility that altered immune signalling could create aberrant cytokine fingerprints. Lipkin and Hornig showed that there was significant decoupling of expression of many cytokines. Klimas has also shown this (saw it at the conference, not sure if it's published yet). So, as you point out, IL-6 generally gets pulled in - but it's possible it might not if key cytokine expression patterns have become decoupled. Again, though, I do not believe ME is an inflammatory disease. I just think we have to be careful making assumptions from experience in other diseases.
It's not everybosy who have had a low ESR within the ME/SEID family. Mine has ranged from 20-38 over the last years. My CRP was around 5-6 but back to normal now. What does it all mean, whether this is significant, who knows.
WWIW I do think we have an inflammatory problem.
WWIW I do think we have an inflammatory problem.
Last edited:
The last time I got an actual flu (not a flu-like exacerbation of ME) was in the winter of 2006-2007. Now when I get a virus I just get worsened orthostatic intolerance, energy and muscle pain but no fever or mucus. It's very difficult to judge if I'm just having an ME crash or a new infection.
Last June I got something resembling a real cold for the first time since. It happened when I was doing well on the starch. I got mild temperature and sore throat + phlegm which lasted about three weeks. I thought it was a good sign but unfortunately I haven't gotten any normal illnesses since even though others in the house have had the flu and strep throat several times.
Last June I got something resembling a real cold for the first time since. It happened when I was doing well on the starch. I got mild temperature and sore throat + phlegm which lasted about three weeks. I thought it was a good sign but unfortunately I haven't gotten any normal illnesses since even though others in the house have had the flu and strep throat several times.
- Messages
- 5,256
- Likes
- 32,032
Dear Eeyore,
I may not be a grandmother, but I have seen a few eggs in my time. I am not sure what your background is. You have clearly read quite a lot of immunology literature but I would warn against confusing fact with fashion.
Leprosy is certainly a good Ying Yang paradigm. I was put on to a project to try to seperate 'lepromatous' and 'tuberculoid' RA in 1978, even before anybody thought to call it TH2 and TH1. The project went nowhere. When TH1/TH2 came along its was already clear that this was a pretty dubious distinction based on mice. When I set up my own lab programme it became clear to me that you do not bother with these slogans that are designed for the benefit of those too lazy to look at detail - you look at individual molecules and pathways. All the molecules are relevant in their different ways but you will never understand anything if you stuff them into Procrustean functional categories.
The functional distinction of TH2 being more related to antibody production has a grain of truth but only a grain. TNF alpha and gamma interferon are both essential for normal antibody production too. Nobody has learnt anything from TH1/TH2/TH17/TH101. What we HAVE learnt is that individual molecules are relevant to certain diseases. Spondarthritis centres not around CD4 cells but around MHC class I based interactions (?CD8,?NK receptors) and certain shared cytokine receptor thresholds. Forget TH17, just stick to IL-17.
I am very familiar with the IL-17 bandwagon in autoimmunity. The main therapeutic monoclonal was taken forward in RA by my good friend Pierre Miossec. It failed. But as you say it has proven useful in autoinflammatory conditions, which need to be clearly distinguished from autoimmunity. So all the hype about IL-17 in autoimmunity proved wrong. And RA did not turn out to be TH1 or TH2 either - nobody has ever found anything wrong with T cells in RA. I am pretty sure it is purely a problem of B cell feedback control - as I suspect is also true of most of the other autoantibody associated diseases.
The Ying Yang idea has been pushed by people like Marc Feldman and countless others for years but there has never been a scrap of evidence for 'polarisation'. Tuberculosis is not a TH1 response it is a unique aberrant reaction orchestrated by the bacterium, not the host. Specific mechanisms activate specific pathways. I don't know where the idea of scleroderma changing from TH1 to TH2 comes from but it is a fairly tale as far as I am concerned. People love to talk this talk because they cannot understand anything more complicated than Bill and Ben or Noddy and Big Ears (I actually had a slide of that once but it upset too many important people).
And I would not take too much notice of what gets into textbooks. Our department under Ivan Roitt used to write these and all sorts of guff went in. There are CD4 cells with different cytokine repertoires, indeed, but the rest seems to me to be nonsense. And as for the M1, M2 macrophage story: I cut my research teeth on macrophage subsets. There are at least six subsets in synovium and they have precious little to do with M1/M2 in mice. Maybe the most important distinction is bright CD16 expression - which explains the pattern of RA and is very local to certain tissues. The work has been well developed. It is just that the trendy mouse people do not bother to read the human literature. And defining subsets is not the point - the point is to follow the individual molecules and pathways.
There will no doubt be forms of inflammation involving cytokines that do not show up in terms of raised CRP. However, the specialised T cell cytokine patterns are, I suspect, almost entirely operative at short range so they are not very likely to show up in the circulation. Most of the gamma interferon probably comes from NK or similar cells. I realised after the last post that gout is probably a good indicator of the non-specificity of CRP - CRP will go up nicely in acute gout without any adaptive immune mechanisms being involved.
What we need in ME more than anything is a level playing field of scientific scepticism. We need to target psychobabble, but equally we should target immunobabble. Jo Cambridge and I still have a good laugh once a week about all the bilgewater in the immunology journals.
I may not be a grandmother, but I have seen a few eggs in my time. I am not sure what your background is. You have clearly read quite a lot of immunology literature but I would warn against confusing fact with fashion.
Leprosy is certainly a good Ying Yang paradigm. I was put on to a project to try to seperate 'lepromatous' and 'tuberculoid' RA in 1978, even before anybody thought to call it TH2 and TH1. The project went nowhere. When TH1/TH2 came along its was already clear that this was a pretty dubious distinction based on mice. When I set up my own lab programme it became clear to me that you do not bother with these slogans that are designed for the benefit of those too lazy to look at detail - you look at individual molecules and pathways. All the molecules are relevant in their different ways but you will never understand anything if you stuff them into Procrustean functional categories.
The functional distinction of TH2 being more related to antibody production has a grain of truth but only a grain. TNF alpha and gamma interferon are both essential for normal antibody production too. Nobody has learnt anything from TH1/TH2/TH17/TH101. What we HAVE learnt is that individual molecules are relevant to certain diseases. Spondarthritis centres not around CD4 cells but around MHC class I based interactions (?CD8,?NK receptors) and certain shared cytokine receptor thresholds. Forget TH17, just stick to IL-17.
I am very familiar with the IL-17 bandwagon in autoimmunity. The main therapeutic monoclonal was taken forward in RA by my good friend Pierre Miossec. It failed. But as you say it has proven useful in autoinflammatory conditions, which need to be clearly distinguished from autoimmunity. So all the hype about IL-17 in autoimmunity proved wrong. And RA did not turn out to be TH1 or TH2 either - nobody has ever found anything wrong with T cells in RA. I am pretty sure it is purely a problem of B cell feedback control - as I suspect is also true of most of the other autoantibody associated diseases.
The Ying Yang idea has been pushed by people like Marc Feldman and countless others for years but there has never been a scrap of evidence for 'polarisation'. Tuberculosis is not a TH1 response it is a unique aberrant reaction orchestrated by the bacterium, not the host. Specific mechanisms activate specific pathways. I don't know where the idea of scleroderma changing from TH1 to TH2 comes from but it is a fairly tale as far as I am concerned. People love to talk this talk because they cannot understand anything more complicated than Bill and Ben or Noddy and Big Ears (I actually had a slide of that once but it upset too many important people).
And I would not take too much notice of what gets into textbooks. Our department under Ivan Roitt used to write these and all sorts of guff went in. There are CD4 cells with different cytokine repertoires, indeed, but the rest seems to me to be nonsense. And as for the M1, M2 macrophage story: I cut my research teeth on macrophage subsets. There are at least six subsets in synovium and they have precious little to do with M1/M2 in mice. Maybe the most important distinction is bright CD16 expression - which explains the pattern of RA and is very local to certain tissues. The work has been well developed. It is just that the trendy mouse people do not bother to read the human literature. And defining subsets is not the point - the point is to follow the individual molecules and pathways.
There will no doubt be forms of inflammation involving cytokines that do not show up in terms of raised CRP. However, the specialised T cell cytokine patterns are, I suspect, almost entirely operative at short range so they are not very likely to show up in the circulation. Most of the gamma interferon probably comes from NK or similar cells. I realised after the last post that gout is probably a good indicator of the non-specificity of CRP - CRP will go up nicely in acute gout without any adaptive immune mechanisms being involved.
What we need in ME more than anything is a level playing field of scientific scepticism. We need to target psychobabble, but equally we should target immunobabble. Jo Cambridge and I still have a good laugh once a week about all the bilgewater in the immunology journals.
@Jonathan Edwards
First, I should say that I appreciate having doctors and researchers present on these forums. I also agree with you that we need a fair dose of scientific skepticism. I'm skeptical of my own theories as well as yours and just about everything else out there. What is obvious is that we do not really understand ME - so skepticism about any proposed theory is mandatory. I am using the tools I have at my disposal to try to understand. That doesn't include a well funded lab - so unfortunately I cannot test at each step, as I'd like to do. The reason we have so many different ideas going in so many different directions in the field of ME research is that we do not have the funding to do high-powered, well designed studies to test hypotheses, and discard those that fail. XMRV is an exception - a testable hypothesis that failed, AND was sufficiently well funded so that now all but a few have abandoned it. I do think there can be a tendency to build a house of cards if you make a reasonable, plausible jump, but then lack the funding to prove or disprove it, and then build on that, over and over. These are problems that can only be addressed by funding and the research community taking ME more seriously. I'm not in a position to make that happen right now - you may be to some degree, although it is a Herculean task and there are very few if any individuals who can drastically alter perception of ME, but you can probably do a lot more than I to help.
Immunology is complicated - and we don't understand it all yet. One of my immunology professors used to like to say, "Most of the most brilliant minds in medicine that I have known went into immunology, where they were never heard from again." It's not a discipline that lends itself to very simple models.
I think we're going to have to disagree on the T-helper subsets for now. I find the evidence in peer reviewed journals and immunology texts very persuasive. It should, however, be thought of as a model. e.g. Not every patient with multiple myeloma has the same molecular pathology (is it a translocation, point mutation, hyperdiploidy, etc.), but it does converge on a common pathway. Still, we lump them together, and it allows us to treat them. Most hematologists use the same induction therapy on all patients who are pursuing an aggressive course of treatment, which is probably a proteosome inhibitor, an imid, maybe an aklylating agent, and dex - because they generally work, regardless of the exact molecular pathology. Right now, we classify cancers largely by tissue of origin - and it is a helpful model - but eventually, we will analyze the specific mutations, and that will be a better model. A better model that creates more subtypes can tailor therapy more effectively, but it's impractical to subdivide the model to the point where every human being is his/her own category. We must strike a balance. Medicine is all about categorization - the model is used as long as it predicts results accurately. When it doesn't you go deeper, until you're working at a molecular level. Since none of us can really claim knowledge of every amino acid residue of every protein in every human being, how they all interact, etc., we cannot go that deep all the time. The complexity is too great. We must always develop models to describe phenomena and predict results. The Th17 model predicts the efficacy of IL-17 and IL-23 mabs - which seems to work. That doesn't mean that it completely describes everything. We do know that in humans there are master transcriptional regulators of different CD4 immune profiles, and we can demonstrate their binding to DNA and promoting transcription of the cytokines we consider to be part of different subsets of cells. e.g. T-bet for Th1, GATA-3 for Th2, RORC for Th17 and FoxP3 for Treg. I wouldn't claim this explains everything, only that it is a useful model for understanding what is happening. There is also a great deal of plasticity in the model - cells may switch from one profile to another (Treg's are subclassified as induced or natural Tregs). However, there is evidence for a certain degree of discreteness - stimulation of naive CD4 Th cells with both Th1 and Th2 polarizing cytokines, over a wide range of quantities of each, yields a Th2 cell that produces Th2 cytokines. The switch appears to be binary, and GATA-3 appears to directly bind to DNA and prevent transcription of T-bet.
M. leprae modifies the host's immune response for survival (not uncommon in general among pathogens). Being intracellular, a humoral response won't be very effective. However, why do some people develop one form and others another? We also know that genetics plays a major role - so it's generally the same pathogen reacting to different immune systems that yields different results. The pathogen has the same bag of tricks in each person - the clinical phenotype of the individual is not just determined by variation / strain in the pathogen. We know, for example, that polymorphisms in the TNF-alpha promoter (-308) can have very large effects on the likelihood of development of lepromatous or tuberculoid leprosy. If you prefer, you can think of this at the cytokine level. Either model will work in many cases.
In reference to seronegative spondyloarthritides, by MHC molecules I presume you are referring to HLA-B27 and related molecules - which clearly play a major role. Let's take B27 since it's the prototypical example, say 2705 - a high susceptibility variant common in Europeans. We know that it is present in the vast majority of AS cases (90% or so, I think). However, the converse is not true - the vast majority of B27 carriers do not develop AS or related diseases. We have found other genes that are involved, although we cannot explain all of the genetic risk. One of those polymorphisms is a receptor for IL23 - and we know that IL-23 does make CD4 cells make IL-17 - which, if you block, improves symptoms. Also, if you block the IL23, it improves symptoms. The effects in psoriasis - also autoinflammatory - have been dramatic with sekukinumab - far surpassing TNF blockers. In this case, Th17 models do predict responses. Furthermore, they predict that if one responds to one (of 17/23), there is a good chance one will respond to the other. Interactions between cytokines need to be understood, and to understand that we have to understand how each cytokine changes the cytokine production of other cells, and which cytokines tend to be produced together by a given cell.
Back to RA for a bit - you say nothing was ever found to be wrong with T-cells. I'm not exactly sure what that means - what was looked at, etc. You suggest that B-cell feedback is relevant, which would be consistent with the success of rituximab in treatment of RA. However, abatacept prevents T-cell activation, and is also effective in RA. TNf-alpha blockers work - but TNF-alpha is expressed by many immune cells. The way I think of it is that each individual is an array of polymorphisms that create a different balance in each person. One person may be resistant to a type of disease while another is susceptible. Environment plays a role, as does pure, quantum mechanical randomness. By targeting any number of cytokines, we can alter that balance - and right now, the results are often still unpredictable. I do not believe that any disease is purely a disease of CD4 T-cells (rheum disease - I suppose HIV is mostly a CD4 disease, although glial cells are affected too). I do believe that in rheumatic disease, T-cells play a role. You can target that role to modify the function of the whole system, or you can target other cells (say, B-cells, in RA). It doesn't necessarily matter. If I have four guys holding up a heavy piece of furniture at the 4 corners, if I knock out the legs of any of the 4, the furniture will fall. We all have Il-17 and Th17 cells that make it, but we don't all have rheumatic disease. Some people, especially with diseases like AS, have only the mildest forms and it is never diagnosed - others face a rapid progression to extreme disability despite aggressive therapy. While a given cell's polarization may be, for a time, binary, the mixture of cells and cytokines is constantly changing. It may be that Il-17 levels are normal in some people with AS - but reducing them can still help.
There are multiple ways to categorize macrophages. I can classify my books by color, by number of pages, by author's last name, or by height. They are all valid ways to classify books. I would say there are a number of ways to categorize macrophages. With flow cytometry, we now love to count cells based on clusters of differentiation. You can now run panels for lymphocyte subsets and such - they run them in ME sometimes, although I'm not sure of their clinical usefulness. So looking at CD16 expression is interesting, and probably tells us about the transcriptional regulators that are active in that cell, and thus the cytokines and other proteins it is producing and secreting. It's one interesting, and probably useful, way of categorizing macrophages - but I don't think it's the only one. Has that model worked in ME? If not, then either a) there is nothing to find in ME re: macrophages or b) there is something to find, but another model is necessary. I don't know much about your personal research in RA, but it may be specifically relevant to RA.
Your comments on gout are interesting. It supports the idea that cytokines are made by a wide array of cells acting cooperatively. Most if not all cytokines are produced by more than one cell type - although relative importance may vary.
Are you implying a lack of respect for immunology as a discipline? As all of us, you come to the table with a preconceived set of ideas, largely shaped by your training and experience. True of me as well of course. I think we all need to challenge ourselves to go beyond that. When one scientist claims something, the other is always justified in requesting proof (or evidence at least). Skepticism to a degree is healthy, but too much skepticism can be a problem. As a rheumatologist, you think in terms of certain disease paradigms. You've developed a model - one that works - for treating your patients. You probably also have realized that your current understanding of disease has not worked to date in effectively treating ME patients. That you've come here shows a willingness to try to understand the disease better by listening to patients and trying to formulate new hypotheses. I would argue, however, that we may need to break out of the normal mold a bit to explain this disease. If it were a minor variation on a common theme of well understood illnesses, I think we'd have figured it out. Some degree of out of the box thinking is beneficial, tempered of course by scientific skepticism - and ultimately driven by a desire to find the truth. Over the years I have entertained a number of theories about ME. Some were discarded due to contradictory evidence. Others remain plausible but unproven. The simple fact is that we do need to think outside the box, because thinking in the box has not yielded any good answers - and by now, it should have. All that advice applies equally to me - I need to remain skeptical, but also open minded, and listen to new points of view, and I need to be willing to abandon preconceived notions when evidence does not support them.
First, I should say that I appreciate having doctors and researchers present on these forums. I also agree with you that we need a fair dose of scientific skepticism. I'm skeptical of my own theories as well as yours and just about everything else out there. What is obvious is that we do not really understand ME - so skepticism about any proposed theory is mandatory. I am using the tools I have at my disposal to try to understand. That doesn't include a well funded lab - so unfortunately I cannot test at each step, as I'd like to do. The reason we have so many different ideas going in so many different directions in the field of ME research is that we do not have the funding to do high-powered, well designed studies to test hypotheses, and discard those that fail. XMRV is an exception - a testable hypothesis that failed, AND was sufficiently well funded so that now all but a few have abandoned it. I do think there can be a tendency to build a house of cards if you make a reasonable, plausible jump, but then lack the funding to prove or disprove it, and then build on that, over and over. These are problems that can only be addressed by funding and the research community taking ME more seriously. I'm not in a position to make that happen right now - you may be to some degree, although it is a Herculean task and there are very few if any individuals who can drastically alter perception of ME, but you can probably do a lot more than I to help.
Immunology is complicated - and we don't understand it all yet. One of my immunology professors used to like to say, "Most of the most brilliant minds in medicine that I have known went into immunology, where they were never heard from again." It's not a discipline that lends itself to very simple models.
I think we're going to have to disagree on the T-helper subsets for now. I find the evidence in peer reviewed journals and immunology texts very persuasive. It should, however, be thought of as a model. e.g. Not every patient with multiple myeloma has the same molecular pathology (is it a translocation, point mutation, hyperdiploidy, etc.), but it does converge on a common pathway. Still, we lump them together, and it allows us to treat them. Most hematologists use the same induction therapy on all patients who are pursuing an aggressive course of treatment, which is probably a proteosome inhibitor, an imid, maybe an aklylating agent, and dex - because they generally work, regardless of the exact molecular pathology. Right now, we classify cancers largely by tissue of origin - and it is a helpful model - but eventually, we will analyze the specific mutations, and that will be a better model. A better model that creates more subtypes can tailor therapy more effectively, but it's impractical to subdivide the model to the point where every human being is his/her own category. We must strike a balance. Medicine is all about categorization - the model is used as long as it predicts results accurately. When it doesn't you go deeper, until you're working at a molecular level. Since none of us can really claim knowledge of every amino acid residue of every protein in every human being, how they all interact, etc., we cannot go that deep all the time. The complexity is too great. We must always develop models to describe phenomena and predict results. The Th17 model predicts the efficacy of IL-17 and IL-23 mabs - which seems to work. That doesn't mean that it completely describes everything. We do know that in humans there are master transcriptional regulators of different CD4 immune profiles, and we can demonstrate their binding to DNA and promoting transcription of the cytokines we consider to be part of different subsets of cells. e.g. T-bet for Th1, GATA-3 for Th2, RORC for Th17 and FoxP3 for Treg. I wouldn't claim this explains everything, only that it is a useful model for understanding what is happening. There is also a great deal of plasticity in the model - cells may switch from one profile to another (Treg's are subclassified as induced or natural Tregs). However, there is evidence for a certain degree of discreteness - stimulation of naive CD4 Th cells with both Th1 and Th2 polarizing cytokines, over a wide range of quantities of each, yields a Th2 cell that produces Th2 cytokines. The switch appears to be binary, and GATA-3 appears to directly bind to DNA and prevent transcription of T-bet.
M. leprae modifies the host's immune response for survival (not uncommon in general among pathogens). Being intracellular, a humoral response won't be very effective. However, why do some people develop one form and others another? We also know that genetics plays a major role - so it's generally the same pathogen reacting to different immune systems that yields different results. The pathogen has the same bag of tricks in each person - the clinical phenotype of the individual is not just determined by variation / strain in the pathogen. We know, for example, that polymorphisms in the TNF-alpha promoter (-308) can have very large effects on the likelihood of development of lepromatous or tuberculoid leprosy. If you prefer, you can think of this at the cytokine level. Either model will work in many cases.
In reference to seronegative spondyloarthritides, by MHC molecules I presume you are referring to HLA-B27 and related molecules - which clearly play a major role. Let's take B27 since it's the prototypical example, say 2705 - a high susceptibility variant common in Europeans. We know that it is present in the vast majority of AS cases (90% or so, I think). However, the converse is not true - the vast majority of B27 carriers do not develop AS or related diseases. We have found other genes that are involved, although we cannot explain all of the genetic risk. One of those polymorphisms is a receptor for IL23 - and we know that IL-23 does make CD4 cells make IL-17 - which, if you block, improves symptoms. Also, if you block the IL23, it improves symptoms. The effects in psoriasis - also autoinflammatory - have been dramatic with sekukinumab - far surpassing TNF blockers. In this case, Th17 models do predict responses. Furthermore, they predict that if one responds to one (of 17/23), there is a good chance one will respond to the other. Interactions between cytokines need to be understood, and to understand that we have to understand how each cytokine changes the cytokine production of other cells, and which cytokines tend to be produced together by a given cell.
Back to RA for a bit - you say nothing was ever found to be wrong with T-cells. I'm not exactly sure what that means - what was looked at, etc. You suggest that B-cell feedback is relevant, which would be consistent with the success of rituximab in treatment of RA. However, abatacept prevents T-cell activation, and is also effective in RA. TNf-alpha blockers work - but TNF-alpha is expressed by many immune cells. The way I think of it is that each individual is an array of polymorphisms that create a different balance in each person. One person may be resistant to a type of disease while another is susceptible. Environment plays a role, as does pure, quantum mechanical randomness. By targeting any number of cytokines, we can alter that balance - and right now, the results are often still unpredictable. I do not believe that any disease is purely a disease of CD4 T-cells (rheum disease - I suppose HIV is mostly a CD4 disease, although glial cells are affected too). I do believe that in rheumatic disease, T-cells play a role. You can target that role to modify the function of the whole system, or you can target other cells (say, B-cells, in RA). It doesn't necessarily matter. If I have four guys holding up a heavy piece of furniture at the 4 corners, if I knock out the legs of any of the 4, the furniture will fall. We all have Il-17 and Th17 cells that make it, but we don't all have rheumatic disease. Some people, especially with diseases like AS, have only the mildest forms and it is never diagnosed - others face a rapid progression to extreme disability despite aggressive therapy. While a given cell's polarization may be, for a time, binary, the mixture of cells and cytokines is constantly changing. It may be that Il-17 levels are normal in some people with AS - but reducing them can still help.
There are multiple ways to categorize macrophages. I can classify my books by color, by number of pages, by author's last name, or by height. They are all valid ways to classify books. I would say there are a number of ways to categorize macrophages. With flow cytometry, we now love to count cells based on clusters of differentiation. You can now run panels for lymphocyte subsets and such - they run them in ME sometimes, although I'm not sure of their clinical usefulness. So looking at CD16 expression is interesting, and probably tells us about the transcriptional regulators that are active in that cell, and thus the cytokines and other proteins it is producing and secreting. It's one interesting, and probably useful, way of categorizing macrophages - but I don't think it's the only one. Has that model worked in ME? If not, then either a) there is nothing to find in ME re: macrophages or b) there is something to find, but another model is necessary. I don't know much about your personal research in RA, but it may be specifically relevant to RA.
Your comments on gout are interesting. It supports the idea that cytokines are made by a wide array of cells acting cooperatively. Most if not all cytokines are produced by more than one cell type - although relative importance may vary.
Are you implying a lack of respect for immunology as a discipline? As all of us, you come to the table with a preconceived set of ideas, largely shaped by your training and experience. True of me as well of course. I think we all need to challenge ourselves to go beyond that. When one scientist claims something, the other is always justified in requesting proof (or evidence at least). Skepticism to a degree is healthy, but too much skepticism can be a problem. As a rheumatologist, you think in terms of certain disease paradigms. You've developed a model - one that works - for treating your patients. You probably also have realized that your current understanding of disease has not worked to date in effectively treating ME patients. That you've come here shows a willingness to try to understand the disease better by listening to patients and trying to formulate new hypotheses. I would argue, however, that we may need to break out of the normal mold a bit to explain this disease. If it were a minor variation on a common theme of well understood illnesses, I think we'd have figured it out. Some degree of out of the box thinking is beneficial, tempered of course by scientific skepticism - and ultimately driven by a desire to find the truth. Over the years I have entertained a number of theories about ME. Some were discarded due to contradictory evidence. Others remain plausible but unproven. The simple fact is that we do need to think outside the box, because thinking in the box has not yielded any good answers - and by now, it should have. All that advice applies equally to me - I need to remain skeptical, but also open minded, and listen to new points of view, and I need to be willing to abandon preconceived notions when evidence does not support them.
Last edited:
- Messages
- 5,256
- Likes
- 32,032
Dear @Eeyore,
I have had debates with several people on PR over the last year or so about these issues but perhaps not with you. I have made a number of these points before but it is always worth developing things in new directions.
I never had a well funded lab but managed to prove that rituximab was an effective therapy for autoimmune disease and get it licensed for RA. One of the points I have made before is that it strikes me that the problem for ME research is not funding so much as getting a sharper focus on the plausible possibilities and getting consensus on data. Money is no use without clear ideas. And when you have a clear idea you often do not need much money. When I thought (out of the box) that rituximab might work for RA, I bought the drug with my own money and put up the IV lines myself and got the result needed to bring in the $10M of commercial support needed to get to licence.
So I think the real issue is the second one you mention - getting the research community to take ME seriously. I think we are making progress with that, at least in the UK it seems so and there are good signs from US, Oz and Scandinavia too. I am retired but I am trying to push people to take an interest.
I agree with the factual points you make about TH subsets. But what I think needs questioning is this odd idea about 'balance' of 'TH1 response' or 'TH2 response'. You may be able to give an example but I can't think of any in which this concept has been of any use. It is not a good model. And you do not have to go into minute detail to find an alternative - you look at specific dysregulatory steps in specific pathways - as we did for RA (and other autoantibody-associated problems) in our review in Immunology in 1999. There is nothing impractical about this. The difference is that the Ying Yang balance theory makes use of no specific testable system dynamic abnormalities - it is just handwaving, whereas our model for self-perpetuating B cells was based on precise pathways and precise dysregulatory mechanisms. Immunologists love diagrams with arrows going everywhere. These are useless. A testable theory has to say exactly which arrow is the critical one at each point in the sequence.
The efficacy of TH17 was not predicted by any TH17 model. It was maybe predicted by finding a genetic linkage to the IL-23 receptor. The 'TH17 model' predicted anti-IL-17 would work in RA and it was wrong because it had no basis in a specifics. Anti-IL-17 was then thrown at everything else and I have no idea whether it was found to work in psoriasis before or after the IL-23 receptor linkage story came out. But my memory is that the linkage is to ank spond more than to psoriasis. So the 'TH17 model' was useless. Why not just stick to the IL-17 pathway and its relation to IL-23.
Leprosy may well be a good example of a response that can go two ways depending on TNF thresholds, but that does not mean that this concept has any bearing on autoimmunity!
What was looked for in terms of T cells in RA? Between 1981 and not so long ago just about everything imaginable was looked for. Everybody working on RA had to work on T cells. I decided to work on synovial tissue susceptibility instead - so got no funding but made more headway than all the T cell boys and girls. People looked at T cell numbers, cytokine production, clonality, receptor usage, responses to every peptide you can think of. Specifically they looked at peptides from IgG Fc (the antigen of rheumatoid factor) and citrullinated peptides (the antigen for ACPA) and found nothing. At the end of the day the conclusion was 'Actually the T cells seem to be peculiarly unactivated.' Even Marc Feldman, who made his career out of proposing that T cells were crucial in RA and the source of TNF, ended up showing that what little T cell activation there was was not through antigen stimulation but through a bystander pathway.
The genetic risk factors for RA are Class II (on B cells) and PTPN22. The latter is involved in T cell signalling but is also crucial for B cell receptor editing and weeding out rogue antibodies in marrow. Moreover, and this is the nitty-gritty, at least a dozen anti-T cell biologics consistently failed to have a useful therapeutic effect despite in some cases rendering patients seriously T lymphopenic for decades. Whereas B cell depletion consistently reduces CRP to normal in a good proportion of patients and inhibits erosive change. It is not really a one or other of four legs situation. Killing T cells did zilch. Killing B cells worked well. Blocking costimulation is of course entirely consistent with removing the bystander help from normal T cells that autoreactive B cells need. TNF blockade works because the effector mechanism for RA is small immune complex ligation of CD16 generating TNF (about 100 times as much as Marc got in his T cell cultures).
We can write off CD16 being relevant to ME simply on the basis of tissue distribution (it is expressed in precisely those tissues affected by RA - so 'CD16 ligation disease' effectively already has a name - RA). I have discussed previously on PR with Marco about the possibility that if ME is mediated by autoantibody the effector mechanism might be through CD64 (CD32 would not fit). Microglia also have CD64.
I have great respect for good immunologists. Most immunologists working on normal mechanisms are good. Sadly in the area of pathogenesis the good ones tend, as your professor noted, to hide in obscurity, because the subject has been taken over by the Ying Yang politicos. My ideas in immunology were not really shaped by training because I had to teach myself immunology after having worked on synovial biology for ten years (having been very unimpressed by T cell babble and leprosy projects). Maybe that is why I ended up thinking out of the box and coming to the conclusion that the pathogenesis of RA was staring us in the face if people would only stop obsessing over T cells. So breaking out of molds has never been an issue for me (my first degree is in Art and Architecture). The mold I see as most constricting is the T cell subset obsession mold, to be honest. Coming to ME afresh in retirement it looks to me as if the thinking in the box has been the same old genetics/environment duo that misses the systems dynamics of dysregulatory disease. So granny says ta for the 'advice' but has some suggestions for sucking the egg from the other end!!!!
I have had debates with several people on PR over the last year or so about these issues but perhaps not with you. I have made a number of these points before but it is always worth developing things in new directions.
I never had a well funded lab but managed to prove that rituximab was an effective therapy for autoimmune disease and get it licensed for RA. One of the points I have made before is that it strikes me that the problem for ME research is not funding so much as getting a sharper focus on the plausible possibilities and getting consensus on data. Money is no use without clear ideas. And when you have a clear idea you often do not need much money. When I thought (out of the box) that rituximab might work for RA, I bought the drug with my own money and put up the IV lines myself and got the result needed to bring in the $10M of commercial support needed to get to licence.
So I think the real issue is the second one you mention - getting the research community to take ME seriously. I think we are making progress with that, at least in the UK it seems so and there are good signs from US, Oz and Scandinavia too. I am retired but I am trying to push people to take an interest.
I agree with the factual points you make about TH subsets. But what I think needs questioning is this odd idea about 'balance' of 'TH1 response' or 'TH2 response'. You may be able to give an example but I can't think of any in which this concept has been of any use. It is not a good model. And you do not have to go into minute detail to find an alternative - you look at specific dysregulatory steps in specific pathways - as we did for RA (and other autoantibody-associated problems) in our review in Immunology in 1999. There is nothing impractical about this. The difference is that the Ying Yang balance theory makes use of no specific testable system dynamic abnormalities - it is just handwaving, whereas our model for self-perpetuating B cells was based on precise pathways and precise dysregulatory mechanisms. Immunologists love diagrams with arrows going everywhere. These are useless. A testable theory has to say exactly which arrow is the critical one at each point in the sequence.
The efficacy of TH17 was not predicted by any TH17 model. It was maybe predicted by finding a genetic linkage to the IL-23 receptor. The 'TH17 model' predicted anti-IL-17 would work in RA and it was wrong because it had no basis in a specifics. Anti-IL-17 was then thrown at everything else and I have no idea whether it was found to work in psoriasis before or after the IL-23 receptor linkage story came out. But my memory is that the linkage is to ank spond more than to psoriasis. So the 'TH17 model' was useless. Why not just stick to the IL-17 pathway and its relation to IL-23.
Leprosy may well be a good example of a response that can go two ways depending on TNF thresholds, but that does not mean that this concept has any bearing on autoimmunity!
What was looked for in terms of T cells in RA? Between 1981 and not so long ago just about everything imaginable was looked for. Everybody working on RA had to work on T cells. I decided to work on synovial tissue susceptibility instead - so got no funding but made more headway than all the T cell boys and girls. People looked at T cell numbers, cytokine production, clonality, receptor usage, responses to every peptide you can think of. Specifically they looked at peptides from IgG Fc (the antigen of rheumatoid factor) and citrullinated peptides (the antigen for ACPA) and found nothing. At the end of the day the conclusion was 'Actually the T cells seem to be peculiarly unactivated.' Even Marc Feldman, who made his career out of proposing that T cells were crucial in RA and the source of TNF, ended up showing that what little T cell activation there was was not through antigen stimulation but through a bystander pathway.
The genetic risk factors for RA are Class II (on B cells) and PTPN22. The latter is involved in T cell signalling but is also crucial for B cell receptor editing and weeding out rogue antibodies in marrow. Moreover, and this is the nitty-gritty, at least a dozen anti-T cell biologics consistently failed to have a useful therapeutic effect despite in some cases rendering patients seriously T lymphopenic for decades. Whereas B cell depletion consistently reduces CRP to normal in a good proportion of patients and inhibits erosive change. It is not really a one or other of four legs situation. Killing T cells did zilch. Killing B cells worked well. Blocking costimulation is of course entirely consistent with removing the bystander help from normal T cells that autoreactive B cells need. TNF blockade works because the effector mechanism for RA is small immune complex ligation of CD16 generating TNF (about 100 times as much as Marc got in his T cell cultures).
We can write off CD16 being relevant to ME simply on the basis of tissue distribution (it is expressed in precisely those tissues affected by RA - so 'CD16 ligation disease' effectively already has a name - RA). I have discussed previously on PR with Marco about the possibility that if ME is mediated by autoantibody the effector mechanism might be through CD64 (CD32 would not fit). Microglia also have CD64.
I have great respect for good immunologists. Most immunologists working on normal mechanisms are good. Sadly in the area of pathogenesis the good ones tend, as your professor noted, to hide in obscurity, because the subject has been taken over by the Ying Yang politicos. My ideas in immunology were not really shaped by training because I had to teach myself immunology after having worked on synovial biology for ten years (having been very unimpressed by T cell babble and leprosy projects). Maybe that is why I ended up thinking out of the box and coming to the conclusion that the pathogenesis of RA was staring us in the face if people would only stop obsessing over T cells. So breaking out of molds has never been an issue for me (my first degree is in Art and Architecture). The mold I see as most constricting is the T cell subset obsession mold, to be honest. Coming to ME afresh in retirement it looks to me as if the thinking in the box has been the same old genetics/environment duo that misses the systems dynamics of dysregulatory disease. So granny says ta for the 'advice' but has some suggestions for sucking the egg from the other end!!!!
@Eeyore @Jonathan Edwards
Thanks for the fascinating discussion. I won't pretend I understand it all, but do find it an eye opener. I did a couple of online Immunology courses last year - which probably doesn't even get me into the 'Noddy and Big Ears' club, but it certainly made me a good deal llless ignorant (not hard), and whetted my appetite too.
Oops. Guilty of using the Yin Yang metaphor writing about my course here, and didn't even realise I was being unoriginal (though I did admit I was probably oversimplying badly).
Anyway, will keep up as best I can.
Thanks for the fascinating discussion. I won't pretend I understand it all, but do find it an eye opener. I did a couple of online Immunology courses last year - which probably doesn't even get me into the 'Noddy and Big Ears' club, but it certainly made me a good deal llless ignorant (not hard), and whetted my appetite too.
Oops. Guilty of using the Yin Yang metaphor writing about my course here, and didn't even realise I was being unoriginal (though I did admit I was probably oversimplying badly).
Anyway, will keep up as best I can.
Last edited:
@Jonathan Edwards
I haven’t debated anything with you in the past - I joined the forums yesterday. As such, I do not know much about your views on ME and its causation. I’d be very curious to hear them. Do you think it is autoimmune or autoinflammatory? Infectious? Genetic? Other? Or is it too early to even make guesses on general direction? Where do you think research should be looking right now?
I partly agree with your views on funding. We’ve looked at a lot of aspects of ME in small studies, and so many are contradictory. We need larger studies to confirm/refute these contradictions. An excellent example is the Hornig/Lipkin study. Prior studies on cytokine levels were all over the map. None of them noted a biphasic disease process. The large number of patients allowed them to see the differences, and to posit the biphasic response. I think that one study means we can throw all the others out. Yes, we should replicate it, but I believe it’s likely to be replicable.
I do agree there are problems with focus and direction. For one, we do way too much ‘fishing’ - a trial and error approach - and not nearly enough serious thinking about how the disease process could occur. Mostly we take theories of existing diseases and modify them a bit and think we have a new theory that is worth testing - not very creative. We should be formulating deeper, testable hypotheses. The experimental arm of science can do very little w/o the theoretical arm charting a course - which it has failed to do with ME.
We certainly agree the research community needs to take it more seriously, and that the stigma needs to disappear. A young, brilliant tenure-track professor won’t research ME. If he does, he risks losing tenure (I know some examples of this). These professors also happen to be the most driven and they produce the most research (on average) because of the strong motivation to publish before their tenure is decided. When I did bench research (developmental genetics in c. elegans), it was the assistant profs who were still in their labs at 3 in the morning. The tenured, senior professors were out at 5 pm. I am encouraged to hear from an MD that the attitude is improving over in the UK. It may be improving here in the US, but most doctors still really don’t believe it and just can’t wait to get the patient out the door. I think researchers are also frustrated because they don’t know where to start. With RA, you start at the thing you can see - a swollen joint. You work back from there until eventually you reach a molecular understanding. We can’t find much in the way of consistent, easily observed and tested, replicable abnormalities in ME - so where do you start? Everyone agrees that a badly swollen joint is a problem - no such obvious problem is seen in ME.
I think your interpretation of the modern view of T-helper subsets is dated, and the field seems to heading away from your viewpoint. Maybe that will change, only time will tell. As you point out, science has headed in wrong directions in the past. A lot of this is still at the research level, and not used by clinicians.
The idea that the cause of all RA is completely understood is oversimplified. Simplicity is great when it works, but it doesn’t always explain everything. In many cases, we find susceptibility to autoimmune/autoinflammatory diseases to be modulated by many genes. The IL23R is just one - but in that case, we mostly see connections to auto inflammatory diseases (AS, PsA, USpA, IBD, Psoriasis). The fact that we do not see a strong linkage to RA implies there is fundamentally something different about these disease processes, which any clinician who deals with rheumatic disease can tell you - they don’t look or act the same. You can’t always point to a particular enzyme in a particular pathway - because it’s not the same for all patients. There may ultimately be a common pathway at the root of it if you go far enough - but we don’t always have those answers, and the final common pathway isn’t always the ideal target. If there were one single, simple problem that was the cause of all RA, for example, we could target a drug at that point and put the disease into remission in every patient. There is more heterogeneity than that in the population. Rituximab works, but not always - and it’s used rarely here in the states compared to tnfa/tnfar mabs. The number of drugs we have for RA is large, and the process of finding which to use remains mostly trial and error. Even if you can say that the abnormality is PTPN22 or HLA-DR4/DR1/whatever, it doesn’t tell you that therapy X or Y will work.
If we look at the vast array of cytokines, and the fact that changes in any of them, or their receptors, on a genetic level alters immune function in a measurable way (e.g. even heterozygous partial loss of function mutations in IL-28B decrease response to Hep C drugs - although that has gotten much better recently). Subtle changes in promoters or enhancers can have real physiologic consequences. All of these cytokines are creating a vast network of communication. It’s not like diagramming the citric acid cycle or beta-oxidation of fatty acids. You need the huge diagrams with lots and lots of arrows to explain it all. There are a lot of players, both cell types and messenger proteins. If you drop part of it out and simplify, you may not be able to explain things anymore. Immunology is a bit unique in the number of conflicting or reinforcing signals that are sent between a wide variety of cells - it’s just more complex. All those cytokines are there for a reason, and a model that omits too much will fail.
Th17 (or Th-anything) doesn’t explain everything in immunology. It doesn’t explain RA - which appears to be an area of particular expertise for you. T cells may not be of preeminent importance in RA. Back to the furniture analogy, you can target different legs - but not everything has many legs, and definitely not the same ones. AS responds to blockers of IL23, IL17, and tnf-alpha. Sometimes it responds to anakinra (IL-1). Methotrexate, which works in RA, seems to do relatively little in AS. As predicted by the lack of antibodies, rituximab doesn’t do much in AS. The point is that you can target a number of different pathways and they all work, so these are the various guys holding up the heavy furniture. This is true in most diseases - in HIV you use combination therapy, protease inhibitors, nucleotide analogs, reverse transcriptase inhibitors, etc.
It is interesting that T-cells seemed less active in RA. It may be that t-regs were not functioning normally or adequately - but that is speculation. I’m not sure how to measure where TNF-alpha comes from in RA (or anything else) - the system is too complex to replicate in vitro with confidence, so getting a relative idea of contributions of different cell types to the total TNF-alpha pool seems difficult at best. The fact that you could make some patients T-lymphopenic without success doesn’t prove that t-cells aren’t important. The issue isn’t always number, it can also be function. If there are different subpopulations of t-cells, indiscriminate depletion may not have any beneficial effects. There are hundreds if not thousands of articles demonstrating aberrant th17/treg cells in rheumatoid arthritis. It may be a secondary phenomenon - the b cells may be primary - but it does seem relevant. Many studies also show that the ratios can be corrected by treatment and that it correlates with symptomatic improvement. You must have seen these studies - and are discounting them for a reason. Curious what that is?
Genetic risk factors for RA are extensive. HLA-DR is probably the main one, and as you point out. Interesting that in AS, the main risk factor is an HLA class 1 not specific to APC’s, and in RA it’s a class 2 specific to APC’s. There are quite a few genetic polymorphisms linked to RA outside the MHC on chromosome 6. Some are large risk factors, some small. These aren’t really clinically useful either - or tested in general in treating RA - but it doesn’t mean they aren’t real or important in understanding it. I don’t know any rheums who test right off for all the genes that confer risk, as this doesn’t provide (afaik) any clinical benefit.
If T cells are irrelevant in RA, why does abatacept work? I think you are oversimplifying. I agree that B-cells are likely at the root of the problem, and clearly play a role (or rituximab wouldn’t work), but that’s not the whole story.
I think you did break out of the mold in think about RA - and it served you well. ME isn’t RA though. There is no consistent elevation of inflammatory markers - and in fact most markers of inflammation are low in ME. CRP is low-normal, ESR is generally very low (with exceptions - but Medscape even suggests that a diagnosis of CFS should be questioned if the ESR is not very low - mine have almost always been 0), and ferritin is often found to be low. It looks more like an anergic immune system. This would be consistent with the broadly lower cytokine levels found by Lipkin and Hornig (very low p value) in long term patients. I don’t believe ME is autoimmune. People have posited that it’s immune exhaustion, but I bet as a rheumatologist you have issues with that theory, considering you can see an RA or SLE patient for many decades and their immune systems don’t become “exhausted” and stop attacking their bodies. Additionally, we do not see tissue damage and organ failure in ME - or nothing I’m aware of that is replicable. In short, we need a very different approach to understanding it.
Fibro gets lumped in with rheumatology a lot - even though it’s not known to be autoimmune or autoinflammatory. Curiously, genes that relate to fatty acid metabolism are linked to fibro and maybe ME. Genes that affect autoimmunity are not.
I haven’t debated anything with you in the past - I joined the forums yesterday. As such, I do not know much about your views on ME and its causation. I’d be very curious to hear them. Do you think it is autoimmune or autoinflammatory? Infectious? Genetic? Other? Or is it too early to even make guesses on general direction? Where do you think research should be looking right now?
I partly agree with your views on funding. We’ve looked at a lot of aspects of ME in small studies, and so many are contradictory. We need larger studies to confirm/refute these contradictions. An excellent example is the Hornig/Lipkin study. Prior studies on cytokine levels were all over the map. None of them noted a biphasic disease process. The large number of patients allowed them to see the differences, and to posit the biphasic response. I think that one study means we can throw all the others out. Yes, we should replicate it, but I believe it’s likely to be replicable.
I do agree there are problems with focus and direction. For one, we do way too much ‘fishing’ - a trial and error approach - and not nearly enough serious thinking about how the disease process could occur. Mostly we take theories of existing diseases and modify them a bit and think we have a new theory that is worth testing - not very creative. We should be formulating deeper, testable hypotheses. The experimental arm of science can do very little w/o the theoretical arm charting a course - which it has failed to do with ME.
We certainly agree the research community needs to take it more seriously, and that the stigma needs to disappear. A young, brilliant tenure-track professor won’t research ME. If he does, he risks losing tenure (I know some examples of this). These professors also happen to be the most driven and they produce the most research (on average) because of the strong motivation to publish before their tenure is decided. When I did bench research (developmental genetics in c. elegans), it was the assistant profs who were still in their labs at 3 in the morning. The tenured, senior professors were out at 5 pm. I am encouraged to hear from an MD that the attitude is improving over in the UK. It may be improving here in the US, but most doctors still really don’t believe it and just can’t wait to get the patient out the door. I think researchers are also frustrated because they don’t know where to start. With RA, you start at the thing you can see - a swollen joint. You work back from there until eventually you reach a molecular understanding. We can’t find much in the way of consistent, easily observed and tested, replicable abnormalities in ME - so where do you start? Everyone agrees that a badly swollen joint is a problem - no such obvious problem is seen in ME.
I think your interpretation of the modern view of T-helper subsets is dated, and the field seems to heading away from your viewpoint. Maybe that will change, only time will tell. As you point out, science has headed in wrong directions in the past. A lot of this is still at the research level, and not used by clinicians.
The idea that the cause of all RA is completely understood is oversimplified. Simplicity is great when it works, but it doesn’t always explain everything. In many cases, we find susceptibility to autoimmune/autoinflammatory diseases to be modulated by many genes. The IL23R is just one - but in that case, we mostly see connections to auto inflammatory diseases (AS, PsA, USpA, IBD, Psoriasis). The fact that we do not see a strong linkage to RA implies there is fundamentally something different about these disease processes, which any clinician who deals with rheumatic disease can tell you - they don’t look or act the same. You can’t always point to a particular enzyme in a particular pathway - because it’s not the same for all patients. There may ultimately be a common pathway at the root of it if you go far enough - but we don’t always have those answers, and the final common pathway isn’t always the ideal target. If there were one single, simple problem that was the cause of all RA, for example, we could target a drug at that point and put the disease into remission in every patient. There is more heterogeneity than that in the population. Rituximab works, but not always - and it’s used rarely here in the states compared to tnfa/tnfar mabs. The number of drugs we have for RA is large, and the process of finding which to use remains mostly trial and error. Even if you can say that the abnormality is PTPN22 or HLA-DR4/DR1/whatever, it doesn’t tell you that therapy X or Y will work.
If we look at the vast array of cytokines, and the fact that changes in any of them, or their receptors, on a genetic level alters immune function in a measurable way (e.g. even heterozygous partial loss of function mutations in IL-28B decrease response to Hep C drugs - although that has gotten much better recently). Subtle changes in promoters or enhancers can have real physiologic consequences. All of these cytokines are creating a vast network of communication. It’s not like diagramming the citric acid cycle or beta-oxidation of fatty acids. You need the huge diagrams with lots and lots of arrows to explain it all. There are a lot of players, both cell types and messenger proteins. If you drop part of it out and simplify, you may not be able to explain things anymore. Immunology is a bit unique in the number of conflicting or reinforcing signals that are sent between a wide variety of cells - it’s just more complex. All those cytokines are there for a reason, and a model that omits too much will fail.
Th17 (or Th-anything) doesn’t explain everything in immunology. It doesn’t explain RA - which appears to be an area of particular expertise for you. T cells may not be of preeminent importance in RA. Back to the furniture analogy, you can target different legs - but not everything has many legs, and definitely not the same ones. AS responds to blockers of IL23, IL17, and tnf-alpha. Sometimes it responds to anakinra (IL-1). Methotrexate, which works in RA, seems to do relatively little in AS. As predicted by the lack of antibodies, rituximab doesn’t do much in AS. The point is that you can target a number of different pathways and they all work, so these are the various guys holding up the heavy furniture. This is true in most diseases - in HIV you use combination therapy, protease inhibitors, nucleotide analogs, reverse transcriptase inhibitors, etc.
It is interesting that T-cells seemed less active in RA. It may be that t-regs were not functioning normally or adequately - but that is speculation. I’m not sure how to measure where TNF-alpha comes from in RA (or anything else) - the system is too complex to replicate in vitro with confidence, so getting a relative idea of contributions of different cell types to the total TNF-alpha pool seems difficult at best. The fact that you could make some patients T-lymphopenic without success doesn’t prove that t-cells aren’t important. The issue isn’t always number, it can also be function. If there are different subpopulations of t-cells, indiscriminate depletion may not have any beneficial effects. There are hundreds if not thousands of articles demonstrating aberrant th17/treg cells in rheumatoid arthritis. It may be a secondary phenomenon - the b cells may be primary - but it does seem relevant. Many studies also show that the ratios can be corrected by treatment and that it correlates with symptomatic improvement. You must have seen these studies - and are discounting them for a reason. Curious what that is?
Genetic risk factors for RA are extensive. HLA-DR is probably the main one, and as you point out. Interesting that in AS, the main risk factor is an HLA class 1 not specific to APC’s, and in RA it’s a class 2 specific to APC’s. There are quite a few genetic polymorphisms linked to RA outside the MHC on chromosome 6. Some are large risk factors, some small. These aren’t really clinically useful either - or tested in general in treating RA - but it doesn’t mean they aren’t real or important in understanding it. I don’t know any rheums who test right off for all the genes that confer risk, as this doesn’t provide (afaik) any clinical benefit.
If T cells are irrelevant in RA, why does abatacept work? I think you are oversimplifying. I agree that B-cells are likely at the root of the problem, and clearly play a role (or rituximab wouldn’t work), but that’s not the whole story.
I think you did break out of the mold in think about RA - and it served you well. ME isn’t RA though. There is no consistent elevation of inflammatory markers - and in fact most markers of inflammation are low in ME. CRP is low-normal, ESR is generally very low (with exceptions - but Medscape even suggests that a diagnosis of CFS should be questioned if the ESR is not very low - mine have almost always been 0), and ferritin is often found to be low. It looks more like an anergic immune system. This would be consistent with the broadly lower cytokine levels found by Lipkin and Hornig (very low p value) in long term patients. I don’t believe ME is autoimmune. People have posited that it’s immune exhaustion, but I bet as a rheumatologist you have issues with that theory, considering you can see an RA or SLE patient for many decades and their immune systems don’t become “exhausted” and stop attacking their bodies. Additionally, we do not see tissue damage and organ failure in ME - or nothing I’m aware of that is replicable. In short, we need a very different approach to understanding it.
Fibro gets lumped in with rheumatology a lot - even though it’s not known to be autoimmune or autoinflammatory. Curiously, genes that relate to fatty acid metabolism are linked to fibro and maybe ME. Genes that affect autoimmunity are not.
@Jonathan Edwards
I should add that autoimmunity may play some auxiliary role in ME. For example, we frequently see elevated antiphospholipid antibodies. Autoantibodies to the alpha-3 subunit of the acetylcholine receptor and several adrenergic receptors are found in POTS patients. It's not clear what role, if any, these play in the disease, but it could be important. At least one small study found antibodies against some of the heat shock proteins.
Personally, I've had a pretty extensive autoantibody screen and found nothing at all - including antiphospholipid antibodies.
I don't think this is the root cause of the disease, although it may contribute.
I should add that autoimmunity may play some auxiliary role in ME. For example, we frequently see elevated antiphospholipid antibodies. Autoantibodies to the alpha-3 subunit of the acetylcholine receptor and several adrenergic receptors are found in POTS patients. It's not clear what role, if any, these play in the disease, but it could be important. At least one small study found antibodies against some of the heat shock proteins.
Personally, I've had a pretty extensive autoantibody screen and found nothing at all - including antiphospholipid antibodies.
I don't think this is the root cause of the disease, although it may contribute.
Sorry
Sorry, cannot agree with that. Error bars are commonly used in good scientific papers after performing analysis of variance (ANOVA) to show whether means are significantly different
Sorry, cannot agree with that. Error bars are commonly used in good scientific papers after performing analysis of variance (ANOVA) to show whether means are significantly different
- Messages
- 5,256
- Likes
- 32,032
What is it that you disagree with exactly? I agree that error bars are commonly used and often in papers that are otherwise good. But my understanding is that the standard error is, as you say, a measure of variance, which is a technical measure that allows you to determine a p value for the likelihood that two populations are the same. Since the p value is normally given as well, the standard error seems to me to be just part of the mathematical working from data to p value, and when you are told the p value tells you nothing further of interest. And sticking it on to a histogram rectangle makes it look like some measure of variation, like a range or standard deviation, and it is not that.
So I entirely agree that a standard error is a respectable part of a statistical computation but I cannot see what purpose an error bar stuck on a rectangle serves If a standard deviation was given we would know the standard error anyway from the sample size, which will be in the methods section and we would also have some useful information about variation. OK, you can derive standard deviation from standard error the other way around, but why show a half way point in a calculation rather than the raw information - and of course a scatter plot would be even more informative.
So I entirely agree that a standard error is a respectable part of a statistical computation but I cannot see what purpose an error bar stuck on a rectangle serves If a standard deviation was given we would know the standard error anyway from the sample size, which will be in the methods section and we would also have some useful information about variation. OK, you can derive standard deviation from standard error the other way around, but why show a half way point in a calculation rather than the raw information - and of course a scatter plot would be even more informative.
- Messages
- 5,256
- Likes
- 32,032
Well, @Eeyore, if you have just joined you are most welcome. There are a number of people here with a lot of scientific knowledge and expertise. Most if not all new research gets pulled to pieces fairly thoroughly. My own thoughts about ME some time back when I had recently joined are on a thread entitled 'Do MEs cause CFS?'. I suggested that their might be six or more aetiological routes, with some degree of overlap, based on my own clinical experience. I might summarise things differently now, but that is where I was then.
My suggestion was that some cases may be genuinely autoimmune, in the sense of being driven by an acquired autoantibody loop, some may be autoinflammatory, as in Reiter's syndrome, which seems to leave some people with an ME-like syndrome (and of course is a sequel to intracellular infection), and some might involve a mitochondrial problem (although I am less attracted to that idea and have difficulty with the system dynamics). In addition I think there is likely to be a CNS loop involved either 'bolted on' to a primary immune loop or on its own without any immune disturbance - it might involve persistent microglial changes for instance. These are just ball park ideas but I think it very likely that pathogenesis is heterogeneous but also that the PEM pattern may indicate a fairly consistent presence of a common type of CNS regulatory loop.
One of the things that seemed most important to me in RA was Stastny's point, when he discovered the DR4 link, that the rest of the causality was mostly stochastic. Most immunologists do not seem to twig to the fact that B cell biology (and in fact T cell) is innately stochastic. All antibody genesis is random. (So it is not quantum stochasticity, it is activation induced deaminase stochasticity.) Stochasticity obviously comes into cancer and I would bet that it will be a major factor in ME since it is also a largely sporadic acquired persistent pathology. Otherwise people get stuck looking for 'environmental factors' as if they were the only alternative to genetics and spend lifetimes looking for agents that may not be there.
I think things are changing. With the MRC initiative in the UK we have several bright people at the assistant professor stage working in ME.
I agree that 'the field' is heading away from my viewpoint on T cells, if 'the field' is the herd of sheep that turns up at most immunology meetings. But so much of it is self fulfilling stuff based on manipulations in mice. The cytokine balance approach simply does not address the dynamics of an acquired sporadic illness in the way that Stastny understood it had to be. Genetic thresholds make sense in ank spond since it really a late onset genetic disease that appears when the skeleton matures. But why does the Reiter's patient suddenly change from normality to an inflammatory state that may last years?
I do not claim that the cause of RA is completely understood but it is understood well enough to know what to do about it. Unfortunately, many colleagues feel the need to get grants and write papers and since they do not understand system dynamics, so cannot understand a purely B cell problem, all convince each other that it must be to do with Tregs or some such. I think probably this is not the place to answer all of your other points on RA in detail, but they are covered in my various publications on PubMed. You do need a few T cells, but normal T cells. We know the TNF comes from the intimal macrophages - for all sorts of reasons that go back to the 1960s. For instance HEV formation occurs at a stage when there is only intimal macrophage activation and no lymphocytes in the tissue.
What we need are some safe selective plasma cell ablating agents. The oncologists will make them soon and then we will be able to induce long term remission in RA just as we can in ITP now. (I have my tongue just a little in my cheek here but I would actually put several pints of beer on this.)
I know of no studies indicating that TH17 or T regs are aberrant in RA and there are people in my department who work on this so I have followed the literature. T cells will get excited if the place is awash with TNF but as far as I know nobody has shown there is anything wrong with them. That stuff about correction with anti-TNF (from my department) is to my mind completely uninterpretable.
I agree that immune exhaustion is implausible - I don't run with that idea. And yes we need to think of a new sort of effector mechanism. CD64 is interesting in that it may be able to get into a loop with gamma interferon without necessarily showering out TNF. (And joints are TNF sensitive just as skin uses IL-1 and ME involves other things.) CD64 may be able to mediate antigen recognition using much lower affinity antibodies that may not show up on standard tests. I would not put money on this one specifically but maybe it is the sort of sideways thinking that we both think is needed. There are also antibodies with Vh usage that makes them half way between adaptive and innate in function that might be relevant. And I am pretty sure that some ME subsets will not be antibody related (maybe more a Reiter type response).
It is early days but all this stuff is being discussed amongst researchers in the UK and some of it is being investigated in the lab. There are going to be several more false dawns on the way but I think things are beginning to roll.
My suggestion was that some cases may be genuinely autoimmune, in the sense of being driven by an acquired autoantibody loop, some may be autoinflammatory, as in Reiter's syndrome, which seems to leave some people with an ME-like syndrome (and of course is a sequel to intracellular infection), and some might involve a mitochondrial problem (although I am less attracted to that idea and have difficulty with the system dynamics). In addition I think there is likely to be a CNS loop involved either 'bolted on' to a primary immune loop or on its own without any immune disturbance - it might involve persistent microglial changes for instance. These are just ball park ideas but I think it very likely that pathogenesis is heterogeneous but also that the PEM pattern may indicate a fairly consistent presence of a common type of CNS regulatory loop.
One of the things that seemed most important to me in RA was Stastny's point, when he discovered the DR4 link, that the rest of the causality was mostly stochastic. Most immunologists do not seem to twig to the fact that B cell biology (and in fact T cell) is innately stochastic. All antibody genesis is random. (So it is not quantum stochasticity, it is activation induced deaminase stochasticity.) Stochasticity obviously comes into cancer and I would bet that it will be a major factor in ME since it is also a largely sporadic acquired persistent pathology. Otherwise people get stuck looking for 'environmental factors' as if they were the only alternative to genetics and spend lifetimes looking for agents that may not be there.
I think things are changing. With the MRC initiative in the UK we have several bright people at the assistant professor stage working in ME.
I agree that 'the field' is heading away from my viewpoint on T cells, if 'the field' is the herd of sheep that turns up at most immunology meetings. But so much of it is self fulfilling stuff based on manipulations in mice. The cytokine balance approach simply does not address the dynamics of an acquired sporadic illness in the way that Stastny understood it had to be. Genetic thresholds make sense in ank spond since it really a late onset genetic disease that appears when the skeleton matures. But why does the Reiter's patient suddenly change from normality to an inflammatory state that may last years?
I do not claim that the cause of RA is completely understood but it is understood well enough to know what to do about it. Unfortunately, many colleagues feel the need to get grants and write papers and since they do not understand system dynamics, so cannot understand a purely B cell problem, all convince each other that it must be to do with Tregs or some such. I think probably this is not the place to answer all of your other points on RA in detail, but they are covered in my various publications on PubMed. You do need a few T cells, but normal T cells. We know the TNF comes from the intimal macrophages - for all sorts of reasons that go back to the 1960s. For instance HEV formation occurs at a stage when there is only intimal macrophage activation and no lymphocytes in the tissue.
What we need are some safe selective plasma cell ablating agents. The oncologists will make them soon and then we will be able to induce long term remission in RA just as we can in ITP now. (I have my tongue just a little in my cheek here but I would actually put several pints of beer on this.)
I know of no studies indicating that TH17 or T regs are aberrant in RA and there are people in my department who work on this so I have followed the literature. T cells will get excited if the place is awash with TNF but as far as I know nobody has shown there is anything wrong with them. That stuff about correction with anti-TNF (from my department) is to my mind completely uninterpretable.
I agree that immune exhaustion is implausible - I don't run with that idea. And yes we need to think of a new sort of effector mechanism. CD64 is interesting in that it may be able to get into a loop with gamma interferon without necessarily showering out TNF. (And joints are TNF sensitive just as skin uses IL-1 and ME involves other things.) CD64 may be able to mediate antigen recognition using much lower affinity antibodies that may not show up on standard tests. I would not put money on this one specifically but maybe it is the sort of sideways thinking that we both think is needed. There are also antibodies with Vh usage that makes them half way between adaptive and innate in function that might be relevant. And I am pretty sure that some ME subsets will not be antibody related (maybe more a Reiter type response).
It is early days but all this stuff is being discussed amongst researchers in the UK and some of it is being investigated in the lab. There are going to be several more false dawns on the way but I think things are beginning to roll.
just another anecdotal here, but my ESR is always raised - only sits at around 20 so nothing to get excited about. I also have raised cytokines from REDLABS and definitely do have some kind of inflammatory nature to my illness - 3 weeks of steroids last year for severe asthma exacerbation improved a number of symptoms - lymph nodes went down on back of neck for first time in 6 years, mucous in lungs improved, anal pain disappeared entirely for first time in 2 years, joint pain went away, stiffness in joints improved, ankle swelling went down for first time in 6 years. Unfortunately a whole lot of others things got worse so I wouldn't want to take steroids again and rheumatologist cant find any cause for my illness so 'M.E' it is.