Understanding Autoimmunity in ME/CFS
Dr. Carmen Scheibenbogen has been a pioneer in researching autoimmunity in myalgic encephalomyelitis/ chronic fatigue syndrome (ME/CFS). Image courtesy of Dr. Carmen Scheibenbogen.
By Bronc and Eric Pyrrhus
For many decades now, people with myalgic encephalomyelitis/ chronic fatigue syndrome (ME/CFS) across the world have had to deal with medical establishments wedded to inaccurate scientific dogma regarding their illness. Alongside this, there has been a disastrous lack of funding for biomedical research into the illness.
Thankfully, there is a growing body of open-minded scientists who have challenged the dogma and are engaged in research designed to find the causes, discover a reliable biomarker, and develop effective treatments.
One such scientist is Dr. Carmen Scheibenbogen. Dr. Scheibenbogen is Professor of Clinical Immunology at the Charité university hospital in Berlin. She is also a member of the Biomarkers Working Group of EUROMENE, an EU non-profit organisation dedicated to researching ME/CFS.
We spoke with Dr. Scheibenbogen to find out more about her research and the possibility of new diagnostics and treatments.
The original findings
In 2015, Dr. Scheibenbogen and colleagues published a now-classic paper which reported that, in 29.5% of patients with ME/CFS, they found elevated levels of certain antibodies when compared to the levels found in healthy people.
These antibodies, which occur naturally at low levels, are attracted to receptors on autonomic nerve cells. Since these antibodies are attracted to regular nerve cell receptors, rather than being attracted to an infection, these antibodies are called autoantibodies.
Dr. Scheibenbogen worked with a company called CellTrend to refine a diagnostic test that could identify elevated levels of such antibodies.
In a 2021 publication, Dr. Scheibenbogen and colleagues used the CellTrend test to look for a link between levels of these antibodies and the symptoms of ME/CFS. They found that levels of some antibodies correlated with fatigue and muscle pain in patients with an infection-triggered onset, but not in patients without an infection-triggered onset.
A possible treatment?
In those patients who show elevated levels of these antibodies, what would happen if you tried to lower the levels of these antibodies? To find out, Dr. Scheibenbogen and colleagues performed a small pilot study on 10 patients who had elevated levels of antibodies that are attracted to a specific receptor on autonomic nerve cells.
In this study, they used a process called immunoadsorption to temporarily reduce the total level of all antibodies in those patients. The immunoadsorption significantly decreased the level of antibodies attracted to the specific receptor, in nine of the 10 patients.
After the immunoadsorption, the total level of antibodies in study patients was raised back to a normal level, by giving the patients a dose of new antibodies via intravenous immunoglobulin (IVIG) transfusion.
Nine of the 10 patients showed a significantly decreased level of antibodies attracted to the specific receptor, before receiving the IVIG transfusion. Seven of the nine patients reported an improvement of symptoms, after the IVIG transfusion. Three of those seven patients reported that their improvement lasted for at least a year, but the rest all relapsed within a year.
In a follow-up study, five of the patients in the original pilot study — who had reported improvement of symptoms but then relapsed — were re-treated about two years later with a slightly modified treatment schedule. Four of the five patients reported improvement of symptoms again, but one didn't. The improvement of symptoms lasted approximately 6-12 months.
Where does autoimmunity come from?
Antibodies are created by the immune system to be specifically attracted to infections. If a person has elevated levels of antibodies that are attracted to regular nerve cell receptors, rather than attracted to an infection, it is a form of autoimmunity.
But how could such autoimmunity occur in the first place? What would cause these antibody levels to rise above the levels seen in healthy people?
Immunologists have elucidated a few different ways that this type of auto-immunity could occur. Of these possibilities, two scenarios are seen as most likely: epitope spreading and molecular mimicry. Both of these scenarios are triggered by an infection.
An infection typically leads the immune system to develop antibodies that are specifically attracted to the infection, leading to inflammation surrounding the infection.
Infection of a nerve cell will attract antibodies and lead to inflammation. Created with biorender.com for Phoenix Rising.
Infection of a nerve cell will attract antibodies and lead to inflammation. Created with biorender.com for Phoenix Rising.
In the process, some infected cells may break open, releasing their contents. In addition, the inflammation may cause collateral damage to the surrounding cells, which might also break open and release their contents.
In epitope spreading, this inflammation leads to development of antibodies that are attracted to regular human molecules found near the infection, rather than attracted to the infection itself.
If there is already a low level of such antibodies, epitope spreading may cause an enhancement of these antibodies — such that these antibodies either are increased in number or are more strongly attracted to the regular human molecule.
For example, an infection near (or even in) an autonomic nerve cell would generate antibodies that are specifically attracted to the infection. But the inflammation surrounding the infection could engulf the autonomic nerve cell — possibly leading to an increased level of antibodies that are attracted to regular receptors on the nerve cell.
In molecular mimicry, an infection leads the immune system to develop antibodies that are designed to be attracted to the infection, but which also happen to be attracted to some regular human tissue.
In this case, the antibodies are designed to be attracted to a specific molecule from the infection. However, the shape of this specific molecule just happens to be similar to the shape of some other molecule found in regular human tissue.
Since the shapes of the two molecules happen to be so similar, antibodies can not distinguish between the two molecules and end up being attracted to both.
For example, a virus may have a specific molecule that just happens to have a similar shape as a regular nerve cell receptor. When a person is infected with the virus, existing antibodies that are attracted to regular nerve cell receptors will then be attracted to the virus as well. The immune system will then try to fight the virus by making more of these antibodies — leading to an increased level of antibodies that are attracted to regular receptors on nerve cells.
Dr. Scheibenbogen explains further
We spoke with Dr. Scheibenbogen over the summer to find out more about her research. The following is an edited excerpt of that discussion.
How did you get involved in the field of ME/CFS research?
Well, because I was given responsibility for these patients. I have training in hematology/oncology and I worked in that until 15 years ago. And then I was put in charge of a clinic for immunodeficiency patients, and in this clinic there was also a clinic for patients with ME/CFS.
At that time I did not know much about the disease but I rapidly recognised that there was something going completely wrong with these patients. Because, at that time, these patients were considered as psychosomatic or suffering from some sort of depression, and they were considered as weird, difficult patients. I recognised that this was not true at all.
There were many young patients who really wanted to get healthy again, and they would do anything to achieve it. They were not depressed. They just had to face a very difficult situation. Then I decided "okay, since we know so little in Germany let’s look around in the world," guessing that other people would know more.
Then I recognised something that I could not believe — that worldwide this disease was not well understood and that there was no treatment at that time to offer to these patients.
That was really a big surprise for me, coming from hematology/oncology where so much has been known already for 15 years, and where personalised medicine was already available. It was a shock for me that we had forgotten this severe and frequent disease.
Well, then I decided that I have to take responsibility — not only for learning more about the disease and how we can help the patients — but also for trying to get the pharmaceutical industry to come in and try to develop drugs.
The problem at that time was that there was not much knowledge about the mechanism of the disease. So I looked around and I recognised that in Norway there were two active oncologists, Øystein Fluge and Olav Mella, and I visited them — and actually that was my starting point for doing research on the role of auto-antibodies in ME/CFS.
I really think that now the time is changing, and that we really can achieve the development of therapies pretty soon. My great hope is that, within the next few years, we will have therapies to treat most sufferers.
In a recent study published last year you and your colleagues looked for links between antibody levels and symptoms in ME/CFS.
You found that levels of some antibodies — which are attracted to autonomic nerve cell receptors — significantly correlated with symptoms of fatigue and muscle pain in patients with infection-triggered onset, but not in patients without an infection-triggered onset.
Do you think that these antibodies in patients with infection-triggered onset could be a contributor to autonomic nerve dysfunction — perhaps a contributor to the blood vessel dysautonomia underlying both orthostatic intolerance and aerobic exercise intolerance?
First let me explain a little bit about these antibodies and what is different about them. What we know is that these are so-called natural antibodies, so actually all of us do have such antibodies, and these antibodies help us to control important functions.
One of these functions is in the autonomic nervous system, where we have two major players:
- Adrenaline, also known as epinephrine, which activates adrenergic receptors on autonomic nerve cells, and mediates the stress responses.
- Acetylcholine, which activates cholinergic receptors on autonomic nerve cells.
The antibodies, which are attracted to autonomic nerve cell receptors, seem to get altered during diseases — not only in ME/CFS, but also in autoimmune diseases and neurodegenerative diseases like Alzheimer's disease.
Actually, the first such disease, which was studied already more than 20 years ago, was a heart disease called dilated cardiomyopathy. In this disease these autoantibodies were described for the very first time.
Now, it becomes pretty soon clear, if you have to deal with ME/CFS patients, that they suffer from severe autonomic dysregulation.
So everything you do automatically (via autonomic nerves) — like your heart is beating, you are breathing, you distribute your blood into your muscles when you run, or into your brain when you think — it's not working properly in these patients.
The other thing which is quite evident, for an immunologist, is that immune dysregulation plays an important role in this disease, because the disease is triggered by an infection and we know that infections strongly stimulate the immune system.
In many patients, after an infection, it may take a while until the immune system calms down again and this is a dangerous time to develop autoimmunity.
We also have some evidence that some patients have other autoimmune diseases already, for example thyroid inflammation. They all tell you that autoimmune diseases are running in their family. Therefore, it wasn’t that difficult to put these two findings together, and think "okay, this may be an auto-immune disease."
So, if you consider that ME/CFS is an autoimmune disease — then what is the target? It’s not unlikely that the target may be the autonomic nervous system, since it is so strongly dysregulated. There is no obvious other target in this disease. We have no inflamed joints, we have no organs which are targeted.
It was not that difficult for me to study these antibodies because, as I said, some researchers had already studied such antibodies attracted to adrenaline receptors, for more than two decades.
Therefore, I also started to analyse autoantibodies attracted to these adrenaline receptors or attracted to the acetylcholine receptors, in ME/CFS patients.
That was a study that was published in 2016 in which we found that we had a subset of patients which had higher antibodies attracted to one of the adrenaline receptors, or higher antibodies attracted to one of the acetylcholine receptors. That was the starting point to focus more on these antibodies for my group.
Now, coming back to your question — what role do these antibodies play in autonomic nerve dysregulation? I think they play an important role, although it's not that easy really to prove that these antibodies result in this autonomic nerve dysregulation in people.
One way to study the role of these antibodies is with isolated cells in the laboratory, and that's what we did. So we analysed cells that had these receptors and we showed that if we mixed these cells with blood from patients, then the cell's response to adrenaline stimulation changes.
We also used genetically modified mice, who completely lack these receptors, to confirm these findings. That's another study which was published in 2020, on the function of these receptors.
Another way to study the function of these antibodies is to look to see if the levels of antibodies correlate with symptom severity in patients. That's what we did in the paper you just mentioned.
We found correlations with fatigue, with muscle pain, also with the severity of the disease as assessed by the Bell score, with cognitive impairment. We analysed the autonomic function in this study only using the COMPASS questionnaire and there we could not find a strong correlation.
But what we are very much convinced of is that one of the major problems of these antibodies is that the blood flow is impaired, since adrenaline receptors play an important role in distributing your blood throughout the body.
You cannot increase the amount of blood you have, but when you start to run, you need much more blood in your muscles. Therefore, your autonomic nerves need to redistribute blood away from the bowel or away from the brain. If this redistribution is not working properly, then your muscles do not get enough blood.
Similarly, if you do not redistribute more blood into your brain while you are talking, then you cannot concentrate properly and you may suffer brain fog. We think that the autonomic regulation of blood vessels is severely impaired in ME/CFS, and this may be one of the major problems that these autoantibodies cause.
The more research into autoimmunity that we have, the further away we get from psychosomatic discourse or the suggestion of depression, so it's very encouraging to hear that.
One more comment — these are natural antibodies. It's not just the level of antibodies which tells us that these are not working properly. It's also their function, which does not necessarily relate to the level of antibodies.
Therefore, it's more difficult to study the role of these antibodies as compared to other, simpler autoantibodies — which are just there when you have a disease, and are not there if you are healthy.
Do you think that these elevated antibodies in patients could develop as a response to an infection in the autonomic nerves themselves, perhaps via epitope spreading?
Actually, there are two mechanisms which we think may be possible. One is epitope spreading.
So what does epitope spreading mean? When you have an infection all your immune cells are trying to get rid of this virus. You have the immune system making certain inflammatory molecules, you have your T cells destroying infected cells, and you have your B cells.
These B cells produce antibodies which specifically bind to the virus and try to neutralise the virus. When you have an infection for the very first time, these antibodies do not bind very well. If you have an ongoing infection, or if you have a second infection with the same virus, then your immune system tries to make better antibodies.
So B cells are able to play around with the structure of their antibodies, and those B cells that make antibodies which bind better have a survival advantage.
When you have an infection, the immune system is not able to perfectly target only the infection, so you always have some collateral damage. In addition to B cells that make antibodies attracted to the infection, you also have B cells producing antibodies attracted to other structures, and this may include B cells producing antibodies attracted to your own body's tissue.
If you already have B cells producing antibodies attracted to adrenaline receptors, then what may happen is that you make antibodies which bind more strongly to adrenaline receptors, or which bind a little bit differently — and this may cause a complete change in the adrenaline receptor's function.
The other possibility (and again I'm sorry, immunity is complex), is called molecular mimicry. So what does this mean?
We know that there are structures in many viruses and bacteria which do have similarities to structures of our own tissues. Usually we do not have immune cells producing antibodies attracted to our own tissues, and if we do have such immune cells, they are usually calmed down.
But it may be that when you have an immune response against a virus, that you elicit such a strong antibody response against a viral structure that this antibody can then cross-react with a structure from your own tissues. As this immune response is so strong, this antibody may do some harm.
Actually, we just published a study in which we analysed the antibody response against EBV proteins. Epstein Barr Virus (EBV) is a virus which is well-known to trigger ME/CFS in a subset of patients and this is a pretty large virus with many, many viral proteins.
What we did is we cut these proteins into several thousand pieces, and then we analysed the antibody responses against all these pieces, hoping to find some pieces which were recognised by antibodies from patients with ME/CFS, but not by antibodies from healthy people.
So we were hoping to find a piece of the virus which attracts antibodies from ME/CFS patients. Actually, that is what we found. This project was started already 10 years ago. We published the first data several years ago and this data was quite frustrating because we didn't find that much difference between patients and healthy people at that time.
More recently, we have reanalysed this data. We now have modern bioinformatic technologies, and what we found now is that we could identify two pieces of these EBV proteins which attracted antibodies from the majority of ME/CFS patients.
We found that many ME/CFS patients have a stronger antibody response to one piece of the EBV virus, the so-called poly-arginine region. By doing a modern bioinformatic analysis and looking into sequence homologies, we found that there are several structures in our own body's proteins which are similar — including adrenaline receptors and mitochondrial proteins.
The other interesting thing is that this poly-arginine region is not specific to EBV, but is also found in other viruses — for example enteroviruses or human papillomaviruses (HPV), which are also known to trigger ME/CFS.
Therefore, we think this is a very important finding because it provides a first clue of molecular mimicry. In addition, it may represent an antibody which could be useful in developing a diagnostic test.
We are currently studying other cohorts to confirm this, and we are also going into more detail to see if we can confirm that there really is molecular mimicry, and if these antibodies then also alter the function of the adrenaline receptors. If we can really prove this, then this would be very convincing that we have to target these antibodies.
We were very happy and excited that we found this. Actually it was quite frustrating, there was a lot of work and we had so little success several years ago and I was actually so convinced that within the EBV signature we would find an answer — a potential target of auto-immune responses.
Now I think this is pretty convincing data. Of course, we have to do more studies but this is a very important starting point.
In 2018 and 2020 you published two papers that examined attempts to remove these antibodies by a therapy called immunoadsorption. Can you explain how this therapy works and what success you have had with it?
So, based on our concept that autoantibodies play an important role in ME/CFS and cause the disease, we were looking for therapies.
Many of you know the Rituximab story — that the Norwegians were successful in treating patients in two smaller studies and then failed to prove efficacy in a large study. The large study had several problems. One major one was that they couldn't give them the same dose because of a shortage of money.
To further develop therapies we took a chance on the technique of immunoadsorption, a type of apheresis where you wash out certain proteins directly from the blood of patients.
Actually it is similar to dialysis. You take the blood from the patient and it runs into a machine. Within the machine the cells in the blood are separated from the plasma, the patients get the cells back, and then the plasma runs through a filter. In this filter the antibodies stick. So we sort of wash all antibodies out of the plasma and then we give the plasma back to the patients as well.
This is a technique which actually is a German development and has been used for decades, for the treatment of autoantibody-mediated disease and also to get rid of antibodies in the setting of organ transplants.
Therefore we did two small trials, just to get a proof of principle that we are right by our hypothesis that when we remove the antibodies, the patients should get better. If you repeat the process several times then you get rid of most, or 90%, of all the antibodies in the patient.
What we observed was quite impressive, because several of these patients got rapidly better within a few days. The first things that got better were the flu-like feeling, arthralgias and muscle pain. Actually fatigue took longer. Cognitive impairment was much better after a week. So we had several patients with really impressive responses.
Of course, this was not a controlled study, so there's always the placebo effect. At that time, however, we couldn't convince the company to do a follow-up study as a placebo-controlled trial.
We had to wait until the coronavirus pandemic occurred. Now we are collaborating with the neurologists from our hospital, preparing a trial for immunoadsorption again — now in both ME/CFS and post-COVID ME/CFS patients.
We will also study repeated immunoadsorption, because we know that with just one week of therapy we cannot cure the disease. We think that we need to repeat it two more times.
What further research is needed into potential treatments for ME/CFS?
Our next step is to do both clinical trials and to go into more detailed research.
We urgently need therapies, and by doing clinical trials we can also learn a lot about disease mechanisms. We do these clinical trials, and also do a lot of biomarker analyses in these patients before and after therapy. By doing it this way, we can achieve both developing therapies and learning more about the disease.
There is another problem of the blood vessels called endothelial dysfunction, which has been demonstrated by several groups over recent years. Endothelial dysfunction can result in diminished blood flow and might also explain some of the disease symptoms. Auto-antibodies can cause this endothelial dysfunction.
We will also start a trial with a drug targeting the blood vessel dysfunction. This drug comes from therapy of cardiovascular disease. Hyperbaric oxygen therapy, which can help some patients with Long COVID, is something we may also study in more detail.
Of course, we will also do much more research into the function of these auto-antibodies against adrenaline receptors.
And we are happy that we now have funding from the German government, so we will start doing this research within a network of five different universities in Germany. We will study all different aspects of how they regulate autonomic functions, and how they regulate mitochondrial functions. Bhupesh Prusty will be one of our partners and we will also study how they regulate the immune system.
Here in the UK, the National Institute of Clinical Excellence has issued new guidelines for medical professionals on how to diagnose and look after ME/CFS patients. These have removed harmful psychological therapies such as graded exercise therapy.
What approach do public health authorities in Germany take towards looking after people with ME/CFS?
In Germany we do not have guidelines for the treatment of ME/CFS, we only have guidelines on fatigue, on diagnosis and management of fatigue.
Within these guidelines there is a chapter on ME/CFS — which was actually terrible because it claimed that probably ME/CFS was not a disease at all, and there's very little evidence, that it's a psychosomatic disease and all these stupid things about trials on graded exercise therapy and cognitive behavioural therapy.
These are now currently revised and they will be published again pretty soon. We were very lucky that both Uta Behrends — who is an ME/CFS clinician for children and adolescents in Germany — and myself are contributing to this chapter. Now we are at least 90% in agreement with what has been written, and all of the patient groups were asked to contribute.
This is very important because these are the guidelines which are broadly used and, together with the guidelines we have on management of Long COVID, which also discusses ME/CFS, this now reaches many physicians.
There are still some who think that ME/CFS is something like burnout, as well as those who claim now that Long COVID is all in the mind, not a disease — but these are getting fewer and fewer.
There are many physicians now who know what ME/CFS is, and know what PEM is — which hardly anybody knew five years ago. We now have a lot of training programmes and write a lot of articles on ME/CFS, and it's really been a big change.
Can you tell us about your current research into ME/CFS?
We are very happy that now we have a collaborative research group. Our lab in Berlin will now specifically focus on the findings we had in EBV.
We will study in more detail the potential molecular mimicry of EBV with the adrenaline receptors, but also with other proteins. We'll study this on a molecular and on a functional level, hoping that we can get a good diagnostic test from this. It would be great if we had a simple antibody test in which we detect an antibody that is specific for ME/CFS.
On the other main project, we are currently trying to develop a platform for therapies. Recently we got the announcement that the German government will give funding for a platform for doing clinical trials in ME/CFS and Long COVID.
This platform will be at the Charité university hospital, because we have here the best knowledge on how to diagnose ME/CFS, how to assess symptom severity, how to do clinical trials and we will also have a biomarker platform.
We will have several centres in Germany starting to do clinical trials, and we hope that we will start soon. As we try to develop therapies as fast as possible, we will focus on therapies that are already licensed for the therapy of other diseases. Because in this setting, we could develop therapies much faster.
We are also looking to learn more about which patients will profit from which therapies, because we are convinced that there will not be only one therapy helping all sufferers.
As in other diseases, we may need several drugs because we think there are probably different phases of the disease. Or probably there are also patients in which autoantibodies do not play such an important role or do not play a role at all.
By doing thorough biomarker studies we will hopefully learn to predict which patients would respond to which therapy. In the end we hope that, in five years, we will have a portfolio of different therapies and we will be able to really do personalised treatment as we do already in other therapies.
I think this may really come true within a few years and that is of course because of the huge impact that Long COVID has made. Because that really led to the change that allowed us to get funding. Without funding and without support from the pharmaceutical industry, it has been very difficult during the last ten years to do clinical studies worldwide and that is changing now.
Congratulations on the good news about funding. This gives us a lot of hope. Thank you for all the work that you are doing.
We can really have hope now because it's not just us in Germany, NIH also just made a call for funding for clinical trials in Long COVID. The most severe Long COVID patients do seem to suffer from ME/CFS, so in the end it's also about developing therapies for ME/CFS.
Note: The above article was based on a discussion conducted with Dr. Scheibenbogen on June 1, 2022. You can watch both parts of the discussion here:
Acknowledgment: Bronc would like to thank his daughter for conducting the interview and his wife for transcribing the interview.
Bronc is a former historian who is active in his local ME support group. He enjoys interviewing scientists involved in ME research to help himself and others better understand their illness.
Eric Pyrrhus is a scientist with an interest in flaviviruses, coronaviruses, and imaging technology. With undergraduate training at Columbia University and the University of Pennsylvania, and graduate training at U.C. Berkeley and UCSF Medical School, he has studied biomedical sciences, bioinformatics, biomedical imaging, biosensors, computer science, artificial intelligence, and business administration.