CellTrend test for ME/CFS (Prof. Dr. Carmen Scheibenbogen, University Hospital Charite Berlin, etc.)

[BurnA]

@deleder2k @Valentijn
I agree with what you both say.
My point is there is no statistical evidence which can be used to deduce if you will respond to RTX or not.
@Valentijn uses the word 'some' @deleder2k uses the word 'may'.

So you are both correct - SOME people with elevated autoantibodies MAY respond to RTX.

Where does that leave us ?

Lets look at the permutations :

If i take the test and have elvated autoantibodies, I may (or may not ) respnd to RTX.
If I take the test and dont have elevated autoantibodies I may ( or may not ) respond to RTX.

How has taking the test improved my decision making process with regards to RTX. ?

On the subject of POTS or OI.
Both are easy to diagnose without identifying autoantibodies.
Regarding POTS, this paper states that of the patients they had data on the ones with POTS did NOT have elevated autoantibodies.

This subjest is discussed more in another thread - here is a summary post :

Does Dysautonomia International have a forum, by the way, similar to the www.dinet.org dysautonomia forums? And have there been many people there who have been treated with rituximab?

Indeed, where does that leave us? Let's recap:

The 2014 study on 14 patients with POTS found that:
Beta 1 adrenergic receptor autoantibodies were present in all 14 POTS patients,
Beta 2 adrenergic receptor autoantibodies were present half of these patients.

So it looks like beta 1 adrenergic receptor autoantibodies are the most predicative of POTS.


Fluge and Mella's 2015 study looked for autoantibodies to all alpha and beta adrenergic receptor subtypes, but only found beta 2 adrenergic receptor autoantibodies.

So given that Fluge and Mella had four ME/CFS patients with POTS in their cohort, you would have expected that these researchers would find beta 1 adrenergic receptor autoantibodies in these patients, but they did not.

I don't know the answer to this conundrum.

 

[TiredSam]

I think we probably already know that there are too many false positives and false negatives for this test to be useful as a way to decide whether or not to use rituximab. There may be some statistical correlation with response but that is not going to give you a clinical decision - that would need a pretty black and white difference.

I have to say I am very disappointed that this is being advertised as a diagnostic test for ME/CFS at this stage. This is bad science driven by money. We have already had enough of this and we do not want any more. Patients will clutch at straws and as far as I can see this will be of no use to them or their physicians at this stage. The original paper on the antibodies was interesting but it was a statistical relation, not the basis of a diagnosis.

This was presented to me at my self-help group in Germany months ago as "There will be an announcement about a diagnostic test for ME/CFS from Prof. Dr. Scheibenbogen soon". Of course she can't be responsible for what people gossip about, but it did get all our hopes up that we may finally be able to present ourselves at doctors and social services without being told that the most appropriate treatment / condition of getting benefit payments is to go on a two week rehab for depression, where you are encouraged to exercise and have your health ruined further.

So it looks like a bit of a disappointment. I wonder if I'll be offered it when I go to the Charite in Berlin, and if so, how the results will be used.

 

[Blue]

TiredSam, I have read similar claims from (I guess the same) German charity regarding these tests. I don't think that Prof. Scheibenbogen herself said anything like that though (I didn't come across anything). She doesn't make the impression of a person who makes unreasonable claims, but quite the opposite.

 

[DanME]

I have to chime in. I know Prof Scheibenbogen personally from my advocacy work. She is a very competent and critical researcher and has been fighting for our cause for the last eight years. All against the constant pressure from the (German) psych lobby. She is a good one, trust me!

As far as I know the current blood test offered by Celltrend has never been advertised by Prof Scheibenbogen or her immunological institute at the Charité. Celltrend developed the test for her last puplished study and they have probably the rights to offer it to patients on their own. While I think it might be unfortunate to call it a blood based diagnostic of ME/CFS, it is clearly stated, that the study only found antibodies in 20-30% of the patients. Also they do not advertise the blood test as a prognostic marker for RTX treatment.

It's completely up to the patient to take it or not. In my opinion, the benifit is only personal at this point. It could be interesting to know, if oneself has these antibodies or not. Also it could be useful for getting social benefits or being taken seriously by sceptical doctors. It's not useful for making any clinical disiscions.

 

[Jonathan Edwards]

While I think it might be unfortunate to call it a blood based diagnostic of ME/CFS, it is clearly stated, that the study only found antibodies in 20-30% of the patients.

I think the problem is worse than that, because some of the controls had levels similar to the 'positives' in the patients. There may well be an interesting statistical difference that may tell us something about a subset of ME one day. However it is not just that only a subset of ME are positive - a subset of normal people are positive too so even being positive tells you nothing about yourself as someone with ME. The wording quoted gives the clear impression that this is some sort of diagnostic test when it is not. I also have a lot of respect for Dr Scheibenbogen but I would not personally want to be quoted as associated with such a commercial venture.

 

[BurnA]

I don't see the benefit of this test.

We know that some of the patients who responded had elevated autoantibodies and that these decreased during response.

We also know that some of the patients with elevated autoantibodies didn't respond and their autoantibodies didn't decrease.

We know that other patients responded who didn't have elevated autoantibodies.

I may have made an error and would like to be corrected if so.

I stated that some patients without elevated autoantibodies responded.

This was based on the fact that 30% of the study group had elevated autoantibodies yet up to 60% responded.
However the 60% response is from the 25 phase 2 Norwegian patients not the complete study population. My understanding is they mixed German patients in with Norwegian for the overall study group .

Can anyone advise ? Do we know if All responders had elevated autoantibodies ?
We know not all patients with elevated autoantibodies responded.

Thanks and apologies if I mislead.

 

[CFS_PhD]

In the recent Scheibenbogen paper, we do not know if only patients with ADRB2 or M3/4 antibodies responded to rituximab or if patients without elevations in any of these antibodies also responded.

You can look at Figure 3 and see that there are responders without elevated ADRB2 antibodies, but without having the actual data you cannot tell if these patients were the patients with elevated M3/4 antibodies or not (and vice versa).

If you look at the ADRB2/M4 correlation plot in Figure 1C, it looks like patients primarily had elevations in one of the antibodies but not both. However, it appears that they did not plot all CFS patients on these plots.

As they screened only 12 antigens, it is certainly possible that people could respond to rituxan without having elevations in these specific antibodies due to other pathogenic autoantibodies. As Jonathan Edwards and all of the clinical data for Rituxan have pointed out, it is also possible to have pathogenic autoantibodies and not respond to Rituxan due to the inability of Rituxan to affect plasma cells and possibly other mechanisms.

 

[CFS_PhD]

My 2 cents on this paper and the test being available is that it is a move in the right direction. I will make my case below for each and feel free to correct any mistakes that I make or disagree with me.

The Paper
Bottom Line: Finding an average elevation in IgG to ADRB2 and/or M3/4 that is statistically significant despite only 20-30% of patients even having any elevation at all should be followed up on as it seems unlikely to be due to chance.

Rationale: I think that the majority of people on this forum are of the opinion that there are different etiologies and subgroups of patients with ME/CFS, some of which could be autoimmune. If this were true, then you would not expect to ever find a serology in the majority of ME/CFS patients. What you see in this paper is a relatively small subset of patients with a strong enough measurable immune response to significantly drive up the average titer level in a mixed population of patients in which the majority show no measurable immune response. Additionally, it is not as if the p value in these studies was barely <0.05. The p-values were <0.001 and <0.0001, meaning that if they only did one comparison, you would expect to see this result by chance alone only 1 in 1000 to 1 in 10,000 experiments. Even when you correct for multiple comparisons, because they assessed 12 different antigens, they still would beat the threshold for statistical significance by a fair margin (0.05/12 = 0.004).

Bottom Line: The fact that a few control patients showed elevated titers on par with titer elevations in CFS patients does not detract from the data.

Rationale: Every serological test for rheumatic disease is judged based on sensitivity (If a person has a disease, how often will the test be positive) and specificity (If a person does not have the disease how often will the test be negative). In the whole CFS/ME patient population, the sensitivity of this test would not be great but this might be the result of a mixed patient population. In terms of specificity, not a single serological test can claim to be specific for any disease. There are always a small percentage of completely healthy people that test positive in any serological test. Yet without the use of serological tests, we would be in the dark ages where subjective physician judgement would be the gatekeeper to treatment for any rheumatic disease without completely obvious outward clinical manifestations. So how would the observed specificity in this study compare to the gold standard rheumatic tests? Let's compare:

This study:
108 healthy controls
healthy controls testing positive: 1-11 (depending on test and exact cutoff used)
% false positives - 0.925% - 10.18%
Specificity Range - 89.82% - 99.9%

Specificities of Commonly Used Tests:
System Lupus:
anti-dsDNA - 97.4%
anti-Sm - 96-98%
MCTD:
anti-RNP - as low as 84%
RA:
RF - 85-90%

The observed specificity is therefore in line with all other commonly used serological tests. The fact that no serological test is 100% specific is why serological tests are generally not considered diagnostic for any rheumatic disease and it is the clinical presentation (symptoms) and the serolgical test that go into making a diagnosis. Every person on this forum meets the clinical presentation for CFS/ME, therefore the only thing standing in the way of being diagnosed (and treated appropriately) is identifying a diagnostic test with good sensitivity and adequate specificity. In the small sample size of this study, the specificity does not seem to be a problem. That being said, my experience with rheumatologists is that they believe their clinical judgement way more than any diagnostic test despite the fact that rheumatic diseases are incredibly commonly misdiagnosed; putting patients in a situation where their doctors are overly skeptical of diagnostic tests and overly confident in their ability to diagnose without them. In my opinion, when you mix this attitude with low sensitivity of a test possibly due to a mixed patient population, you have a recipe for nothing ever being convincing enough to pursue or believe might be relevant.

Bottom Line: The finding that clinical response to rituxan is correlated with decreased titers is intuitive and encouraging but needs more data points to be convincing.

Rationale: The fact that titers decreased following treatment in responders but not non-responders sounds great and would make biological sense. The overall problem is the low n-value for these experiments. The significant decreases in titers in responders seems to be driven by a few select patients. This doesn't seem that problematic to me but in order for clinical response to be convincingly tied to rituxan treatment, more non-responders have to be assayed and shown to not have decreases in their titer levels.

Bottoms Line: If I were reviewing this paper for publication I would have asked them to display much more individual patient data so that we could easily see if patients were positive for one or multiple antigens simultaneously. This is because the biggest concern I have with the results is the possibility that a few CFS samples positive for multiple antigens due to some unknown experimental error is what is driving the significant elevations. This would be much less likely to be the case if most patients had only one elevated antigen.

The Test

1) Motivation for selling the tests to patients:

Based on the price of the tests (~$30 each), I cannot see how it could possibly be construed as an attempt by this company to rip off patients and make a quick buck. Let's break this down:

If they sold 3000 tests in the coming year (1000 patients decided to do all 3 tests), their YEARLY revenue on this test would be $90,000. Let's assume they have a margin of 70% (could not even be close to this high based on cost of making these tests), they would have $63,000 after cost of goods. How much does it cost to employ people with benefits to run these tests? I'm going with at least $63,000. Who's profiting from this test in the near future? No one.

Bottom line: the test is currently priced to sell as an unvalidated test with the hopes that they will eventually (with a lot more data) become validated and be used a lot more frequently.

2) Specificity of test: covered above.

3) Sensitivity of test: Unknown and requires a lot more data points to determine if it usefully defines a subset of ME/CFS patients.

I will probably inquire about having the tests run. If it comes back above their arbitrary cutoff point (>95th percentile I believe), I will definitely have a healthy control do it as well to mitigate the risk of a shitty test. Whats the harm at $100 if you have any reason to believe that you could have an autoimmune etiology? The worst that happens is it comes back negative, which it should in even the majority of patients on this forum, at which point you've possibly wasted $100. If it comes back positive, you can argue with your doctor who won't do anything about it and then you can decide for yourself if you are willing to go to more extreme measures to attempt immunomodulation. To me, information is not the enemy. The paper and test are not perfect, but I certainly wouldn't expect them to be in this disease and only time and more data points will determine if it's real (which I'm certainly not saying with confidence that it is). However, there is one guaranteed way to not find out if it's real, and that's to blow it off without proper consideration. Hopefully I've at least provided some food for thought on the topic.

 

[CFS_PhD]

One point I did not mention above is the possibility that the patients with high titers are the group with OI/NMH. I could see this as a possibility but does it really matter? If you have bad OI and CFS and you can't parse the two apart but you end up figuring out that the OI is due to autoantibodies, wouldn't the first logical thing to do be to try to knock down those antibodies and see if it resolves all of your symptoms?

I could definitely get on board with future studies looking at OI in CFS to determine if this is the case or not though.

 

[SOC]

@CFS_PhD, thank you for that detailed assessment of this paper. I don't have the knowledge of this particular topic to easily evaluate this paper and unfortunately the cognitive dysfunction that came along with ME has severely limited my ability to process technical papers even in my own field. Therefore, I always very much appreciate when one of my fellow PWME takes the time to go through a complex paper and tease out the important details for us.

 

[BurnA]

One point I did not mention above is the possibility that the patients with high titers are the group with OI/NMH. I could see this as a possibility but does it really matter? If you have bad OI and CFS and you can't parse the two apart but you end up figuring out that the OI is due to autoantibodies, wouldn't the first logical thing to do be to try to knock down those antibodies and see if it resolves all of your symptoms?

I could definitely get on board with future studies looking at OI in CFS to determine if this is the case or not though.

Thanks for your extremely in depth analysis.

I tend to agree about getting the test just out of curiosity, initially I didn't see any point (and I still don't think there is a good reason ) but it could be interesting to know if we have these autoantibodies or not.

( if people on PR do take this test it would certainly be interesting to pool the results along with noting any OI symptoms )

Regarding more data points, am I right in saying that until there are more verified rtx responders they can't provide any further data in this regard ?

One thing that would be great to establish is how many responders had no elevated autoantibodies. If this is a subset that what do you think this suggests ?

 

[CFS_PhD]

@SOC, Thanks and no problem. Having CFS/ME and having spent several years doing biomedical research, probably the greatest lesson that I have learned in life is that every person knows much less than they think they know. In the process of earning your PhD, you have to do what's called a qualifying examination, where the whole point is for you to study and learn as much as humanly possible about a few select topics and then other PhDs question you about those topics until you have to admit that no matter how much you know about something, there is always more unexplainable than there is explainable. This fact of life is why "curing" or even significantly modifying a disease is very very very hard and takes a very long time to even develop drugs that work fairly well.

Because I realize how little I know in the grand scheme of things, my analysis is just an analysis and my opinion is just an opinion. There are undoubtedly many aspects that I have not considered in this paper. That is why I appreciate and encourage additional input. But I am happy to offer my analysis.

Having first hand knowledge of how hard it is to develop and commercialize drugs that work for any disease, I cannot emphasize enough that one of my biggest disappointments in the medical community today is the choice that doctors make everyday to not even attempt to use the drugs that we do have to try and help people that are obviously suffering. Doctors tend to hide behind the idea of "evidence based medicine" and therefore only use drugs where several studies have shown that they obviously should use a drug to treat a condition. However, this can simply not be applied to real life where you actually understand very little of what drives many diseases and there are not good studies for every possible drug that makes sense to use (more often because there is not financial incentive for a company to pay the 10-100 million to run such a study for every indication). Even when you have a drug that makes sense to use but there is not definitive data making it obvious that it should be used, doctors will then argue that drugs have tox side effects. This is also true, but the vast majority of drugs on the market today have relatively minor side effects, especially when you consider that the alternative to not treating is to allow patients to suffer every second of their life with an average quality of life less than observed in MS or Schizophrenia. There are exceptions to this as always, as some drugs have serious and life threatening tox issues.

So to me, the arguements that doctors make to not trying treatments that could help ME/CFS patients are ludicrous and demonstrate the complete inability of doctors to understand how debilitating of a disease this is. Either that or they just simply don't care. There are obviously exceptions to this, MDs who actually use the available tools that they have to make a difference, but I think all CFS patients know that these docs are few and far between.

Because of the state of the medical community as I see it, when a paper like this comes out that is certainly not definitive and needs to be validated in a lot more patients and in separate studies, yet has potentially relevant and applicable findings, it is crucial that those findings are highlighted, explained and disseminated. Because there is always more unexplainable than explainable (with our current amount of knowledge), a scientific paper usually brings up more questions than answers to a scientist. For this reason, it is usually very easy for any scientist who simply doesn't want to believe that something might be important to find a reason to dismiss the finding and keep asking for more experiments at nausea before they believe that it is worth truly considering (google reviewer #3). This is a very inefficient process that at the end of the day delays getting treatments to those who need them. In my opinion, any physician worth their salary should be able to do more than just follow a treatment paradigm that has been given to them by the countries regulatory body that is a distillation of treatment options including only those that have been shown in many studies to be of indisputable use. They should be able to synthesize as much of the evidence that is out there to logically work through a problem using all of the tools that are available to help their patients.

Sorry for the digression, as you can imagine I get very mentally exhausted after working all day

 

[CFS_PhD]

@BurnA , It would be extremely difficult (basically impossible) to show that someone has no elevated autoantibodies that could be clinically significant as the number of protein coding genes in the genome is somewhere in the 20k-30k range. As both intracellular and cell surface proteins can be autoantigenic, it is therefore theoretically possible for your body to mount an autoimmune response to up to 30k different self-antigens. This is one reason why it is so hard to pin down autoimmune diseases...who knows what protein(s) are involved and where does one start to test? It is also probably why over the course of time, more and more unexplained diseases become associated with autoimmunity. This paper looked at 12 different cell surface receptors (proteins) involved in neurotransmitter function based on the justification that CFS/ME patients have cognitive dysfunction and they therefore thought it most likely that the autoimmune disease would involve inhibition of normal neurotransmitter function. Fair rationale if you ask me.

I would logically think that if a patient has a meaningful clinical response to Rituxan, it is highly likely that they have some unexplained autoimmune issue going on. But I am not an expert on B-cells/Rituxan/Rheumatolgy, like Jonathan Edwards is.

As far as adding more clinical data to further test the findings in this paper: There is utility in both just testing more patients without Rituxan treatment and treating patients both before and after Rituxan treatment. The more patients that they can test the diagnostic on, the more accurate their findings should be (with a greater n-value, the true sensitivity and specificity will be determined). But ultimately, I'm sure they would like to have a lot more data points before and after rituxan treatment to see if their findings hold.

 

[Jonathan Edwards]

Having first hand knowledge of how hard it is to develop and commercialize drugs that work for any disease, I cannot emphasize enough that one of my biggest disappointments in the medical community today is the choice that doctors make everyday to not even attempt to use the drugs that we do have to try and help people that are obviously suffering. Doctors tend to hide behind the idea of "evidence based medicine" and therefore only use drugs where several studies have shown that they obviously should use a drug to treat a condition. However, this can simply not be applied to real life where you actually understand very little of what drives many diseases and there are not good studies for every possible drug that makes sense to use (more often because there is not financial incentive for a company to pay the 10-100 million to run such a study for every indication). Even when you have a drug that makes sense to use but there is not definitive data making it obvious that it should be used, doctors will then argue that drugs have tox side effects. This is also true, but the vast majority of drugs on the market today have relatively minor side effects, especially when you consider that the alternative to not treating is to allow patients to suffer every second of their life with an average quality of life less than observed in MS or Schizophrenia. There are exceptions to this as always, as some drugs have serious and life threatening tox issues.

So to me, the arguements that doctors make to not trying treatments that could help ME/CFS patients are ludicrous and demonstrate the complete inability of doctors to understand how debilitating of a disease this is. Either that or they just simply don't care. There are obviously exceptions to this, MDs who actually use the available tools that they have to make a difference, but I think all CFS patients know that these docs are few and far between.

Because of the state of the medical community as I see it, when a paper like this comes out that is certainly not definitive and needs to be validated in a lot more patients and in separate studies, yet has potentially relevant and applicable findings, it is crucial that those findings are highlighted, explained and disseminated. Because there is always more unexplainable than explainable (with our current amount of knowledge), a scientific paper usually brings up more questions than answers to a scientist. For this reason, it is usually very easy for any scientist who simply doesn't want to believe that something might be important to find a reason to dismiss the finding and keep asking for more experiments at nausea before they believe that it is worth truly considering (google reviewer #3). This is a very inefficient process that at the end of the day delays getting treatments to those who need them. In my opinion, any physician worth their salary should be able to do more than just follow a treatment paradigm that has been given to them by the countries regulatory body that is a distillation of treatment options including only those that have been shown in many studies to be of indisputable use. They should be able to synthesize as much of the evidence that is out there to logically work through a problem using all of the tools that are available to help their patients.

Sorry for the digression, as you can imagine I get very mentally exhausted after working all day

Dear CFS_PhD,
I appreciate your enthusiasm for this study but I think you are being a bit unreasonable and even thoughtless in some of your comments here. Doctors 'hide behind' the need for good evidence because the alternative is very neatly illustrated by PACE - using treatments without good evidence. We need a level playing field of argument.

The reason why doctors do not use unlicensed drugs is that by the time they have finished internship and junior residency with some acute medicine experience they will be aware that they have already caused the death of about five people through treating them. An average orthopedic surgeon will cause the death of about one person a year through treatment. A lot of women in medicine cannot cope with the guilt and opt for low risk specialities. The risk ones are staffed by macho individuals, some of whom practice medicine a bit like playing chicken on a motorbike. I can assure you that most doctors have very good human reasons for not prescribing where there is no good evidence. I have had patients die from gastric bleeding from simple painkillers. I have seen a patient die from rituximab pneumonitis. I have seen the effects of Stevens-Johnson syndrome. Things go wrong all the time I am afraid.

Moreover, most of the unproven treatments that one sees around are fairly clearly being used by physicians whose main objective is to corner a market in a particular practice. If they genuinely wanted to test the treatments they would publish their findings. Why has so little been published on the use of unlicensed treatments in ME? We haere of physicians claiming they have used treatments on hundreds of cases - so where are the data?

So I completely disagree. An insistence on evidence can and should be applied to poorly understood disease. The alternative is staring us all in the face - PACE.

You seem to be suggesting that there are people who are unenthusiastic about this study and are dismissing it. Where did that idea come from? In the research community there is a great buzz about this study. People are wanting to replicate the findings. Efforts are being made to set up further rituximab studies but you may not realise just how difficult it is to do that without making all the same mistakes in PACE. If there is a significant risk of death or severe complications trials are not justified unless they provide evidence that can be generalised from. Simply handing out treatments on a one by one basis is a complete waste of time.

The issue raised on this thread is that the data from Dr Scheibenbogen have been used to justify a 'diagnostic' test and that does not look justified. From what I can see the chances of a PWME getting a positive test is about one in four. And if they get a positive the chances that it has anything to do with them having ME, rather than just being like a normal person is not much better than fifty fifty. That tells you pretty much nothing. Your calculations in the previous post are very impressive but unfortunately you have missed out some rather crucial factors that mean they bear no comparison to standard diagnostic tests like ANA and RF. Nice try but these things have to be done carefully.

Yes, we are all fired up with enthusiasm for the Norway-Germany collaboration. But Dr Fluge himself is very clear that he is against what you are suggesting - trying out rituximab without a clear evidence base. He would much rather that such suggestions were not made until he was sure he was actually on to something.

 

[CFS_PhD]

@Jonathan Edwards,

Thanks for your reply. I appreciate your input and hoped that you would provide it.

I don't have the time this morning to make a full reply but I will make a brief one an hopefully have time to reply further this evening.

1) Can you please provide actual reasoning behind these statements with transparent calculations? It is generally very unhelpful to be so dismissive without explaining how you have come to this conclusion and what assumptions you have made along the way. I have purposely been very transparent in my thought process and would appreciate it if you could do the same so that everyone can judge your rationale and therefore opinion accordingly. You have great experience in this area and I think everyone would benefit from understanding how you come to the opinions that you do on topics such as this:

"And if they get a positive the chances that it has anything to do with them having ME, rather than just being like a normal person is not much better than fifty fifty. That tells you pretty much nothing. Your calculations in the previous post are very impressive but unfortunately you have missed out some rather crucial factors that mean they bear no comparison to standard diagnostic tests like ANA and RF. Nice try but these things have to be done carefully."

2) I feel that I have been very clear that I fully believe that studies should be ongoing and more data is always good on all topics. I am all for having as much evidence as possible to be able to calculate the risk benefit ratio as accurately ass possible for any treatment in any disease. The difference is what do you do when you have no data either way and you have a devistating life altering (and sometimes basically ending) disease? Do you sit around waiting and refusing any sort of treatment or do you provide all the known information about risk to the patient and work with patients on an individual basis to determine what level of risk they feel comfortable taking based on how severely their disease is affecting them?

I argue for the latter and I also argue that if you cannot handle the risks that come along with treating very bad diseases (the possibility of very bad side effects, even death) then you absolutely should not be practicing in that area. Lets take MS for instance. There have now been several cases where people have died or been severely disabled by PML because of treatment. Should the opportunity to receive those treatments that can greatly improve QoL and come with this small but serious risk be restricted by MDs who know nothing of what it is like to live with the disease because they are "afraid" of killing someone? ABSOLUTELY NOT. Without having the disease you have no basis for accurately defining what the true risk benefit ratio is in these cases. Therefore, you as a doctor should transparently and clearly convey the risk and allow patients a significant voice in their care.

As far as the PACE trial - it is another treatment option. It is another tool. Would I be upset if as part of a treatment plan where all potential options were considered that CBT and GET were tried? Absolutely not. Some fraction of people might get benefit. Do I think that it will benefit all PWME? No. It is these overstated goals, to find THE treatment for the disease that get us in trouble, whether that be CBT/GET/Rituxan/Antivirals, etc. All reasonable options should be used until the day where we have enough data to support that some medicines should absolutely not be used and some are much better than others. We are far, far, far away from that point.

3) Why has so little been published on the use of unlicensed treatments in ME? We haere of physiciansclaiming they have used treatments on hundreds of cases - so where are the data?

What would you have them measure to show you benefit in the clinic outside of a formal clinical trial that takes millions of dollars to run? It is symptomatic improvement that they claim. There is no biomarker, therefore there is no rationale for taking any treatment into the clinic. Even better reason for you to be as transparent as possible about your thoughts on these and other biomarkers for CFS/ME as they are the limiting factor in this whole game and should not be so easily dismissed without thorough consideration of their possible utility.

 

[Gijs]

Science requires clinical trials and proper measurements. You don't help patiënts by shooting a rabbit with a shotgun and don't know what kills him. I have severe ME/POTS and at this stage i wouldn't use rituximab because it is no candy.

 

[Jonathan Edwards]

1) Can you please provide actual reasoning behind these statements with transparent calculations? It is generally very unhelpful to be so dismissive without explaining how you have come to this conclusion and what assumptions you have made along the way. I have purposely been very transparent in my thought process and would appreciate it if you could do the same so that everyone can judge your rationale and therefore opinion accordingly. You have great experience in this area and I think everyone would benefit from understanding how you come to the opinions that you do on topics such as this:

"And if they get a positive the chances that it has anything to do with them having ME, rather than just being like a normal person is not much better than fifty fifty. That tells you pretty much nothing. Your calculations in the previous post are very impressive but unfortunately you have missed out some rather crucial factors that mean they bear no comparison to standard diagnostic tests like ANA and RF. Nice try but these things have to be done carefully."

The problem is that things are too complicated for it to be possible to explain fully in an online post. But I will try one or two simple points. Firstly, how useful would a test with 100% specificity and sensitivity be? Answer - completely useless because it tells you nothing you have not already ascertained (in order to know it is 100%). The value of diagnostic tests is not a function of these measures. It is a function of what useful new information is provided. That depends in a complex way on what information you already have. For MCTD for instance, characteristic structural changes in the cuticles tells you something that you cannot get any other way and which has enormous predictive value. People with the characteristic changes are always heading for deep trouble. That is presumably because the changes are a direct reflection of a very unusual form of microvascular damage. Anti-RNP simply tells you that it is very unlikely that you are looking at a clinical feature due to something quite different that just happens to look similar. Second point: specificities around 85-90 are certainly common, as in this case. But these are acceptable if sensitivity is around 70% but hopeless if sensitivity is around 25%. The informational value of a test is a very complex function of either specificity and sensitivity being high (but not too high as indicated before) with the necessary height being contingent on the other measure. Almost nobody who publishes on these tests knows about any of this and will boast about how high the numbers are regardless but if you talk to someone who understands information theory you will realise that test value is not quite what it seems.

2) I feel that I have been very clear that I fully believe that studies should be ongoing and more data is always good on all topics. I am all for having as much evidence as possible to be able to calculate the risk benefit ratio as accurately ass possible for any treatment in any disease. The difference is what do you do when you have no data either way and you have a devistating life altering (and sometimes basically ending) disease? Do you sit around waiting ...

There is a very simple answer to this and your points under item 3. You do not sit on your arse. You do what I did in RA and FLuge and Mella are doing in ME. When I started treating patients my first thought was that if I could justify treating one case I could justify treating five. And with a loan from my father I could buy five lots of drug. Treating one person would have convinced nobody but treating five might. So I sent out a proposal for peer review. I did not ask for money because I knew nobody would fund it (same for Fluge and Mella, they used in house money). The experts said OK it is reasonable to try. So I treated five people, giving the IVs myself and monitoring pulse and BP for hours myself. Why? Because I thought it was important do be sure it was done right. I documented every detail of the progressof the five cases and immediately sent it off for publication. The Lancet and other big journals said it was of no interest but I got someone to print it. Then I started again with another five and a different dose and more groups until I had treated 22 with different doses and published again. By then enough friends and colleagues had enough faith to allow some funds for drug from the hospital. And at the same time, because I had documented and published everything in detail I wa able to persuade the drug company that at least if they did not believe in it they should do a trial to prove me wrong so that they did not miss out on 5billion bucks - which they are now happily raking in.

And so the story goes on and it was almost exactly the same for the Norwegians. If you really think these things are worth doing you get on and do them. It takes ten years out of your life but it is worth it. Compare that with ME physicians who hand out drugs and do not document and publish. Nobody is impressed enough to fund proper trials. The fault is not with the funding organisations. It is with the physicians who have not provided the case because they have not provided systematic data. Not being able to get a million dollar grant is never an excuse. If you do things properly you get the million dollar grant. The Norwegians have a multimillion dollar government funded trial. The others have little to show. When I started working on rheumatoid disease it was as much a mystery as ME is now. You just have to put the work in and document properly.

 

[msf]

I think it´s a bit silly to say ME physicians don´t publish - I can´t think of one who hasn´t (maybe Kauffman, but he works with Kogelnik, who has). And some of them have published trials of repurposed drugs (Valcyte is one example) in ME patients, like you did with Ritux in RA. Most of the are using drugs on label for comorbid conditions they find in ME patients, which most doctors don´t bother to look for.

 

[Jonathan Edwards]

I think it´s a bit silly to say ME physicians don´t publish - I can´t think of one who hasn´t (maybe Kauffman, but he works with Kogelnik, who has). And some of them have published trials of repurposed drugs (Valcyte is one example) in ME patients, like you did with Ritux in RA. Most of the are using drugs on label for comorbid conditions they find in ME patients, which most doctors don´t bother to look for.

Sure they publish, but not in the way you need to get something licensed, as I see it. Every single case needs documenting in detail if you want to convince the right people you are serious. YOu need to show that at every stage you are maximising opportunities for proper control conditions. You need to do formal dose response studies. There are good reasons why these techniques have become standard.

Lots of people other than Fluge and Mella seem to be using rituximab but where are the data?

 

[A.B.]

Lots of people other than Fluge and Mella seem to be using rituximab but where are the data?

Fluge and Mella are doing a fantastic job as far as I can tell from an amateur perspective.