Teclistamab for ME/CFS

Data is Not entirly useless.
For all intents and purposes it is, as JE explained above. And the minuscule value it might produce doesn’t outweigh the risk of harm by any reasonable standards.
You do not know the individual cost functions of people, I think its unreasonable to assume that Out of all patients, there are Not some for whom It makes sense.
I severely doubt the patients are given sufficient information to give proper consent.

Regardless of that, it simply isn’t ethical to do medical experiments on people just because they are willing to say yes to it, if the experiments doesn’t produce any meaningful information. This has nothing to do with what any one individual patient wants or not, it’s about the ethical, moral and legal duty of HCPs to do no harm.
Yes, not here, but Its still making quite a wave right ?
I can’t answer for anyone else, other than giving you my impression of the sentiment in the Dara threads. Besides, if uninformed people are gazing over something has no bearing on the merit of doing that or similar things, so I don’t understand its relevance to this discussion.
Scheibenbogen hypothized this,
(…) Its insane to introduce so much uncertanty in the results just because you do not get funding for the same Dosis as in the Phase 1 Trial.
It doesn’t look like this holds up under scrutiny (from this thread):
I heard people suggest that the dosage might be different (and that the Norwegian trial couldn't provide the dosage because of lack of funding). Not sure if this is true. Here's what I could find online:

Fluge et al. 2011 (pilot RCT, n=30)
Rituximab 500 mg/m2 given twice two weeks apart, with follow-up for 12 months

Fluge et al. 2015 (phase-2, open-label, n= 29)
2 infusions (500 mg/m2) 2 weeks apart, followed by maintenance rituximab infusions after 3, 6, 10 and 15 months

Fluge et al. 2019 (phase-3, RCT, n = 152)
2 infusions of rituximab (500 mg/m2 of body surface area), 2 weeks apart
followed by 4 maintenance infusions with a fixed dose of 500 mg at 3, 6, 9, and 12 months

Japanese trial (RCT, crossover, n = 30)
four times at weekly intervals during the first three weeks of the primary and secondary evaluation periods.
(375 mg/m^2/week iv. x 4 times)

Not sure if this is a significant difference.
 
For all intents and purposes it is, as JE explained above. And the minuscule value it might produce doesn’t outweigh the risk of harm by any reasonable standards.
You would be SOOO unhappy about the Treatment protocol that I am pushing my doctor to do ;)

To get Back to topic,
I do not understand the point about the Data Not being valuable?
Afaik Habets will publish the "results" of the first 10 cases soon. And I think It is safe to assume that he will Not conciously doctor those.
Ofc a rct would be better, but why should this data be meaningless ?
 
To get Back to topic,
I do not understand the point about the Data Not being valuable?
JE has already explained that here:

It’s not about data being doctored, I have no reason to assume bad faith, but that we have no good measurements for «improvements» in ME/CFS so it’s difficult to track. And we know that saline can be followed by what appears to be substantial improvements, even in quite rigorous trials by Fluge and Mella. Without doing blinded RCT with placebo, you have no way of knowing if you’re actually achieved anything with the drug. But you get all the risk of using the drug.
 
I do not understand the point about the Data Not being valuable?

As Richard Feynman said (to quote Google AI):

"The first principle is that you must not fool yourself—and you are the easiest person to fool," emphasizes the importance of self-honesty and critical self-reflection, especially when pursuing knowledge or attempting to understand the world. It highlights the tendency for individuals to unconsciously distort information or hold biased beliefs that align with their desires, making them their own worst enemy in the pursuit of truth.

Science is 90% that. A few make it through to the 10%.
And it is also pretty easy to fool other people, especially now we have social media.
 
It’s not about data being doctored, I have no reason to assume bad faith, but that we have no good measurements for «improvements» in ME/CFS so it’s difficult to track. And we know that saline can be followed by what appears to be substantial improvements, even in quite rigorous trials by Fluge and Mella. Without doing blinded RCT with placebo, you have no way of knowing if you’re actually achieved anything with the drug. But you get all the risk of using the drug.

These seem to me like reasons why the Data needs a certain Prior to be understood in context, but Not at all a reason to completly disregard it.
I do think that this will be an indication If further Investigation ist warented.

Maybe this is the first step to having the results of a tecil Phase 3 rct in approx 7 yrs.
 
These seem to me like reasons why the Data needs a certain Prior to be understood in context, but Not at all a reason to completly disregard it.
If the data is exposed to substantial sources of bias, it can in most cases be completely disregarded. That’s why we need to do trials properly.

And due to the risk of the interventions, we need a good rationale for even doing a trial, and alternative methods of testing the hypotheses should always be considered.

I think we’re going in circles here.
 
It seems as though Dr. Habets has successfully treated an ME patient with low doses teclistamab. Much more cost effective than daratumumab apparently (< 1000€ per treatment course). This is the German maverick oncologist that's also treated ME patients successfully with daratumumab in the past.

https://twitter.com/user/status/1925124231940989201


Teclistamab is a BCMA directed T-cell engager (i.e. it "tells" cytotoxic t cells to kill BCMA expressing cells). BCMA seems to be present on later stage B cells and long lived plasma cells.

Extremely early and just a clinical anecdote but interesting nonetheless.

Has anybody heard of this in the context of ME before?


Okay I see this has gained traction here. My thoughts:

Habets is a good data source because he is willing to try whatever and all kinds of crazy things on patients. In general this is what we need, crazy doctors. Because patients themselves are willing to risk anything.

Now, if he treats more patients with Tec and has good success rate, it lends credence to Fluge and Mella's LLPC theory, because apparently the BCMA thing is on the LLPC too. So two diff treatments working on the same theory = more support for that theory.

I've prodded him on X to ask more but his reply pattern is quite random, he leaves alot of people on read.
 
JE has already explained that here:

It’s not about data being doctored, I have no reason to assume bad faith, but that we have no good measurements for «improvements» in ME/CFS so it’s difficult to track. And we know that saline can be followed by what appears to be substantial improvements, even in quite rigorous trials by Fluge and Mella. Without doing blinded RCT with placebo, you have no way of knowing if you’re actually achieved anything with the drug. But you get all the risk of using the drug.
I think you are being too picky on data here, as ME is a very data sparse environment, you need every data point you can get, even if the quality is not good.

What he is doing is good but of course the way he tweets is not very coherent which kind of sucks.

Also, saline boosted mean step count in placebo (n=74) from 3232 to 3904. Wouldn't say that's substantial improvement. In fact I would argue that 20% improvement in step count is precisely in line with a placebo effect
 
Last edited:
I think you are being too picky on data here, as ME is a very data sparse environment, you need every data point you can get, even if the quality is not good.
That’s really not how things work. If the data isn’t good enough it isn’t good enough. It doesn’t matter if we don’t have any better. This is a common misconception, I suspect because people don’t like facing how little we actually know.
Also, saline boosted mean step count in placebo (n=74) from 3232 to 3904. Wouldn't say that's substantial improvement.
Fair enough.
In fact I would argue that 20% improvement in step count is precisely in line with a placebo effect
I was mostly referring to individuals, but it’s my fault for not making that clear. But I don’t have the data at hand, and it isn’t really central to the argument about the data for dara or tecli.
 
If you ever do find the individual data for the P3 DB study, please share it, I would like to have a look too.
 
Btw, there is some evidence that Dara+Tecil Work well together. No Cross toxicity and improved efficacy in cancer patients.
I think It is Not unreasonable to think that whatever mechanism is responsible for dara and Tecil to Work in CFS could also be improved by a Cockatail of both.
Interesting! do you have studies on this? Any links? This would be an LLPC blasting cocktail
 
Habets is a good data source because he is willing to try whatever and all kinds of crazy things on patients. In general this is what we need, crazy doctors. Because patients themselves are willing to risk anything.
I think you’ll find this forum isn’t really pro-cowboy doctors. It’s called “Science for ME” for a reason.. to have a place where there are standards and rigor applied to the discussion of medical care and research into ME/CFS. There are likely few members on here who think that desperate patients should be experimented on.
I think you are being too picky on data here, as ME is a very data sparse environment, you need every data point you can get, even if the quality is not good.
I think you’re fairly new here, members like @Utsikt and @Jonathan Edwards have backgrounds in science and research and are good to listen to and learn from.
 
I think you’ll find this forum isn’t really pro-cowboy doctors. It’s called “Science for ME” for a reason.. to have a place where there are standards and rigor applied to the discussion of medical care and research into ME/CFS. There are likely few members on here who think that desperate patients should be experimented on.

I think you’re fairly new here, members like @Utsikt and @Jonathan Edwards have backgrounds in science and research and are good to listen to and learn from.

Well, while we wait till 2027 or 2030 for Fluge and Mella to get the funding to pay for 44 x 20k = 880k euros courses of Dara for the RCT, I like that this doctor is trying things.

I also have a stats/data science background and work in the analytics field with Python so I am used to looking at data. Usually hundreds of thousands of observations, but in the ME case, we don't have much.

RCTs are the gold standard for intervention but we can count on one hand how many we have in ME.
 
Last edited:
I also have a stats/data science background and work in the analytics field with Python so I am used to looking at data. Usually hundreds of thousands of observations, but in the ME case, we don't have much.

RCTs are the gold standard for intervention but we can count on one hand how many we have in ME.
Surely you understand the importance of data quality then?

And something someone from the harder sciences might be less aware of is just how unreliable data in medicine can be, especially when any kind of human judgement is involved.
 
Surely you understand the importance of data quality then?

And something someone from the harder sciences might be less aware of is just how unreliable data in medicine can be, especially when any kind of human judgement is involved.
I do, but I'm happy Habets is adding more data points even though they are anecdotes while we wait till 2030.

If he has some success with Tecs, it would add support to the LLPC theory.

I know, that's why I don't trust SF36 as much as step counts. But even then step counts can be faulty if they just shake their fitbit.

In stats the problem of the dependent variable being bad is noisy labels. I am sure it happens a tonne in medicine
 
I do, but I'm happy Habets is adding more data points even though they are anecdotes while we wait till 2030.

If he has some success with Tecs, it would add support to the LLPC theory.

I know, that's why I don't trust SF36 as much as step counts. But even then step counts can be faulty if they just shake their fitbit.
But he keeps adding data with too low quality so it won’t make a difference. If he actually wanted to contribute and not just enrich himself off of desperate patients, he would set up a proper trial.
 
Back
Top Bottom