Rituximab and placebo response

I invented the dosing regimens for autoimmunity and I am pretty sure dosage doesn't matter very much at all. We probably give twice as much as necessary. There are all sorts of variations on the way the dosing is described but we have no real evidence it matters. If B cells are depleted, which they are with all these variations, then the objective has been achieved. I spent some time trying to work out if different doses were important in RA but concluded that unless you go very much lower the results are similar.

 

I think it shows that people generally don't consider CBT/GET a promising treatment.

Rituximab seemed to promise a cure or at least a good treatment. One infusion and you would after a few months begin to feel better and be able to do much more for a long time, maybe even several years and then be able to get another infusion.

 

Thanks @Jonathan Edwards, v helpful information, especially as I have seen a few online place doubt on the rituxamab trial results because of this change.

 

So it is not really about either/or (for objective vs subjective outcomes, but both) and importantly, questions about activity substitution related to what you mention!

 

There’s something I could use some help in understanding.

The first phase II trial was double blinded and placebo controlled. 10/15 in the group taking rituximab were classed as improvers; 2/15 in the placebo group were classed as such.

In the phase III trial, 26% of the rituximab group improved versus 35% in the placebo group.

Why did a significantly smaller proportion of the placebo group improve in the phase II trial versus phase III? To my understanding, both placebo groups possessed the same amount of information as both trials were close labelled. Is the apparent difference likely just statistical noise from the phase II trial having too small a sample size?

 

From a statistics point of view, yes, 15 is a very small sample size and I'd be very wary of concluding anything from a sample of that size.

 

I saw rituximab described as the only approved anti EBV therapy
I wonder if assessed on a case by case basis if it might be useful to me

 

A series of posts have been moved from
Aripiprazole - Abilify

Not specifically, just a trial that when looked at carefully in common sense terms provides evidence with a high probability of being meaningful. My proof of concept trial for rituximab in systemic autoimmune disease was uncontrolled and unrandomised, but because I had detailed pharmacodynamic profiles and objective endpoints it was sufficient to turn around the drug company from no interest to investing $2M in a bigger study. Everyone could see the results must mean something.

I spent much of my career involved in trials and what matters in the end is the likelihood, in common sense terms, that the results will stand up.

I agree that if you are a physician treating ME then a personal account of improvement while using e.g. aripiprazole should be a wake up call to see if that might mean something and maybe explore, discuss with the drug company, ask colleagues about experiences etc. My comment was in relation to us as forum users hearing of someone getting better while on a drug. The problem is that with a million or so people with ME with internet access someone with ME is bound to feel better while using aripiprazole, for whatever reason. And if others hear of that and ask to try some they may well get a placebo effect and so we end up with a rash of stories of improvement on the drug. I am the same physician who decided to investigate further in the clinic context, and one of my roles here is to pick up research ideas and try to persuade funders to take them seriously, but the stories of improvement on aripiprazole are no use to me in that regard - for the likelihood reason I have given. If OMF physicians think that the context merits investigation then hopefully they will do a useful trial, but their perspective will be different from ours.

No. I think that is an unfair analysis that misses the point. I have not stated that patients are not getting better. A placebo response is getting better, so by definition I think they are. What I am unconvinced by is the assumption that they have got better because of that particular drug. We might well ask whether it matters if it its a placebo effect but the answer to that is that it does not matter much to that person but it does matter to everyone else because of the risks of psychotropic drugs (which include increased suicide risk, stroke from hypertensive crisis, permanent disfiguring involuntary movements, fetal abnormalities, etc, etc.).

I agree that endpoints matter, and deciding in advance what end-point you are going to take as evidence of true effect is crucial. The phase 2 open rituximab study is instructive because it showed that the placebo response followed what physicians and patients thought the time profile should be. The problem with invoking tachyphylaxis in ME is that it looks pretty much like what one would expect from fall off of placebo response. The irony of the phase 2 ritual trial was that the falloff of the placebo response was part of the placebo response because it was expected and another dose produced a new placebo response as expected. In other words the trial shows just how much suggestibility can come in to this. That is not gaslighting. It is documenting the way symptoms can go up and down.

I think the cyclo results tells nothing until we have a way to do a trial that can be relied on, and I think the toxicity of cyclo makes that unethical anyway. There were just as 'spectacular treatment effects' with the OPEN rituximab study as with cyclo and we know these were spurious. So there is every likelihood that the cyclo results were too. Cyclo really doesn't do much to B cell activity on its own at the doses given so it is highly unlikely that it had an effect through B cells (particularly as even rituximab did nothing). Cyclo might conceivably have had a useful effect through some other means but my estimate of the likelihood is close to zero.

All I am really trying to say here is that we have learnt that for at least some people with a diagnosis of ME there can be a huge placebo effect. That is not disbelieving or gaslighting,it is a fact of life. And my impression from watching my wife receive psychotropic drugs is that they are probably even more risky than we are told.I suspect most of them made her worse and bad enough to be tube fed. I would urge people not to play around with them.

 

This might be getting off topic but were the placebo results from rituximab trial 10.. 20.. 30% improvement or were they full recovery? I wonder what happened to them in the years since. If they relapsed..

 

The trial didn’t reveal that much to us, many unanswered questions regarding possible explanation of placebo effect, things like their choice of primary and secondary endpoints possibly being very flawed (using these 0-6 simple self-reported scales) etc.

 

As Andy says,I am not saying that at all, simply pointing to facts that have to be explained somehow and which show that we cannot dismiss major placebo responses in people diagnosed as having ME/CFS. These issues are much morecomplex than just 'what it sounds like'.

It is worth looking at the graphs appended to my NICE testimony. A good proportion patients in the open label continuation phase 2 study on rituximab showed a complete normalisation of scores. The lines go up to the top of the scale. Presumably that means they thought they were completely well.It is pretty remarkable.

And they repeatedly relapsed at around six months and got better again with a further shot - exactly as the theory said they should. In fact the irony is that I may have been responsible for these results because I pointed out I a comment in PlosOne following the 2011 paper that best results at six months would be expected, rather than the three month endpoint in the original blinded trial phase. I spoke to Fluge and Mella and explained the time course in RA. The ME patients then showed the same time course, despite the fact that we now know that the drug was not doing anything, at least not in any consistent way.

 

I am afraid that I think there is only one plausible explanation of the placebo effect - suggestibility - for the reasons given above. The time course fitted what was expected very neatly. Before I had pointed out the expected time course it had not done so.

I cannot see any unanswered questions here to be honest. The drug did not work but there was a major placebo effect due to expectation.

 

May I ask then, outside of a trial, how do you explain what is happening when a patient has a very strong, significant, and fairly fast response to a treatment when there was zero expectation? When every treatment they ever did before for years never did anything and they were expecting nothing? When they were having no disease fluctuations and their severity was terrible?

This is what I mean by unanswered questions, maybe certain patients with ME are more susceptible to placebo effect and others are really not at all unless something truly works to alleviate symptoms.

 

You are already presuming a 'response to treatment' but all that is observed is an improvement.
That would probably have to be coincidence if there was zero expectation but do people really try drugs with zero expectation? It seems a contradiction in terms.

The real problem is that other than the person in question nobody else has any guarantee about the real context. The reason why we have strict methods in science is that if we do not know the context well we have to assume all possibilities apply.

I know what it seems like when this happens to oneself, but to be of value to others - and not potentially misleading - it needs to be considered without the first person angle.

Who knows what has happened in an individualise? However, it is of interest that psychotherapists and physiotherapists are confident enough about their ability to tell an improvement is due to a treatment to be able to 'individualise' on the basis of what they know is likely to work in that case. But, as I indicated at the NICE committee, if you look at the level of improvement with therapist-delivered treatments and also at the controls it is obvious that in routine care therapists cannot possibly know if improvement was due to treatment.

Maybe put it this way. If the aims to get health care professionals to be rigorous about testing treatments so that we actually know what works it is a good idea to take onboard the same degree of rigour. Human beings are remarkably suggestible and as Richard Feynman said the easiest person to fool is yourself.

 

@Jonathan Edwards YES we do! We do try treatments with zero expectation or virtually zero expectation!

When you suffer from this illness and you have your entire life robbed from you, you lose your job and ability to work, you lose your health insurance because you don’t have the money, you have a financial crisis because you have no more income and are near homeless and lose all that you’ve ever built in your life, and it’s almost impossible to get disability because the government denies virtually everyone with ME and it takes years and lawyers, then YES you do try most everything with zero or almost zero expectation because of utter desperation.

If that desperation causes more expectation based placebo effect then more treatments would cause improvements across the board, but almost nothing ever works.

 

No one here has been arguing that we don’t need rigorous science and RCTs in this spirited discussion. What has irked me and I’m sure others is what you stated here:

That you feel there isn’t any point on this forum for members to give their anecdotal accounts of improvements on specific treatments, and that you seem to know, with such certainty, that any major, rapid, and fairly long lasting improvement anyone’s experienced is due to placebo effect or something else not related at all to treatment efficacy. I know you have no idea or basis for making such a statement.

 

@Trish that is just splitting hairs, virtually zero or low expectation is enough to make the point that is being made. There’s no effective difference.