Clinical trials & what we need from them

[Hip]

This thread was spilt from :
https://www.s4me.info/threads/cbt-jen-brea-twitter.3013/

We re not aiming to make GPs think ME is a biological disease.

I am not proposing we make GPs believe anything without evidence to support it. What I would like to see is the UK taking some initiative, by looking at the various treatments used by US doctors, and for those that currently have the strongest evidence base, performing large-scale studies to see if they really do work or not.

If all that money which was wasted on the shoddily-performed PACE trial went into decent large scale studies on some of the treatments used in the US, we might by now have some solid evidence.

Saying that we currently have no reliable evidence for efficacy is true, but it should not stop there. The preliminary evidence is available, and that should be enough to warrant larger-scale studies to see whether than preliminary evidence pans out or not.

That's what happened with rituximab. Rituximab all began on the flimsiest of anecdotal case evidence for efficacy in ME/CFS, but ended up in a large-scale clinical trial.

The current evidence base for the efficacy of treatments like Valcyte, Valtrex, Ampligen, oxymatrine, tenofovir, low-dose naltrexone, GcMAF is higher than the original anecdotal evidence for rituximab, so it does not seem right that these treatments have not been pursued in larger-scale studies.

I appreciate that the research work on rituximab was down to the highly commendable focused efforts and persistence of Fluge and Mella, along with your own support for this enterprise. Nevertheless, I don't think we should be saying "there is no reliable evidence for the treatments used in the US". In the UK I think we should instead be saying, "there are a lot of promising treatments used in the US, and there is a need to perform large-scale studies on these to properly assess their efficacy".

 

[Hip]

The other issue in the UK is clinical freedom, which seems more limited here compared to the US.

In Dr Martin Lerner's blinded, randomized placebo-controlled trial treating 27 EBV ME/CFS patients with Valtrex, this culminated with improvements over placebo at 6 months.

But when Dr Sarah Myhill, who I understand is a private GP, not an NHS doctor, started prescribing Valtrex for ME/CFS on the basis of Lerner's study (which is a pretty safe drug), the GMC stepped in and prohibited her from using this drug as an ME/CFS treatment. See: Dr Myhill Stripped of her Right to Prescribe Valtrex off label.


So we have a situation here in the UK where: (a) nobody seems interested in performing larger-scale studies on the treatments used in the US, so that they might be deployed by the NHS if they prove efficacious; and (b) if you as a private doctor try to use one of these treatments, the GMC may step in and stop you.

 

[Jonathan Edwards]

I am not proposing we make GPs believe anything without evidence to support it. What I would like to see is the UK taking some initiative, by looking at the various treatments used by US doctors, and for those that currently have the strongest evidence base, performing large-scale studies to see if they really do work or not.

Why didn't the physicians who tried these do the trials. Why should we waste time in the UK testing treatments touted by people who cannot be bothered to do trials. The rituximab situation was quite different. The people who saw the anecdotal evidence scrabbled the money together to do a phase 2 study themselves and then had enough evidence to persuade a funding body to go to phase 3. I had to do the same with rite for RA. I gathered my own preliminary evidence and publish it in a way that was sufficiently convincing to get $3M from the manufacturers. There is no way I would waste my time on testing something with an anecdotal reputation produced by a colleague who could not get it together to produce credible evidence. The USA healthcare system spends about three times as much per head as we do. Money is sloshing around. So why not do the trials in the US?

Nevertheless, I don't think we should be saying "there is no reliable evidence for the treatments used in the US".

That is exactly the case as far as I am concerned. Because there is none. The placebo effect is very powerful - as the phase 2 rituximab study showed. I forget the Lerner study but if it had shown anything convincing it would have been repeated. The great majority of medical trials are rubbish. It is only too easy to get a result that looks positive - as PACE shows.

 

[Jonathan Edwards]

The Lerner trial abstract is not written clearly enough for it to be worth reading more. A properly performed trial has a more coherent and informative abstract. The wording is muddled in a way that makes me pretty certain that the results reported are not those of a randomised double blind controlled trial. If you do a trial properly you make sure that is clear.

 

[Indigophoton]

The placebo effect is very powerful - as the phase 2 rituximab study showed.

One thing that slightly puzzles me - the phase 2 study was double-blind and placebo-controlled, so, given the phase 3 results, presumably the conclusion is that the phase 2 study was simply too small to yield reliable results.

However, phase 2 was blinded and placebo controlled, and the results were quite different in the two arms. Why was there still apparently a much bigger placebo response in the ritux arm? Is that just a statistical fluke?

Apologies if this is an idiot question, my brain only works at 20% these days.

 

[Hip]

Why didn't the physicians who tried these do the trials. Why should we waste time in the UK testing treatments touted by people who cannot be bothered to do trials.

Perhaps I am just an idealist, but I think we should feel responsible for doing these trials in the UK because of the nature of our medical system: the NHS will not offer treatments until there is a high level of evidence of efficacy, and the GMC will not allow UK doctors to use treatments from the US or anywhere else which have lower levels of evidence.

If that's the locked down system the UK runs by, I feel the UK really needs to run trials and get the evidence. Either that, or provide UK doctors with more clinical freedom, so that GPs can employ treatments like those used in the US which have lesser evidence of efficacy, without the GMC coming after them.

But I wish more ME/CFS physicians would publish studies. I guess not every physician has scientific research proclivities. And there are so few ME/CFS physicians in the world anyway. Dr Lerner's first published study on antiviral treatment for ME/CFS was in 2001, when he was 72, and he published a few more antiviral studies afterwards, but died in 2015.

That is exactly the case as far as I am concerned. Because there is none. The placebo effect is very powerful - as the phase 2 rituximab study showed. I forget the Lerner study but if it had shown anything convincing it would have been repeated. The great majority of medical trials are rubbish. It is only too easy to get a result that looks positive - as PACE shows.

At least Lerner used objective measures of improvement: Lerner's EIPS ME/CFS scale is based on the number of hours of activity patients report engaging in each day (an objective measure), rather than the very unreliable subjective measurement scales used in the PACE study and other studies run by psychologists (asking for example how fatigued you feel subjectively). Any study using such subjective measurements of how you feel or how tired you are should be immediately discounted.

The Lerner trial abstract is not written clearly enough for it to be worth reading more. A properly performed trial has a more coherent and informative abstract. The wording is muddled in a way that makes me pretty certain that the results reported are not those of a randomised double blind controlled trial. If you do a trial properly you make sure that is clear.

I also found Lerner's writing quite unclear and disorganized, and I found it hard to read his papers.

The Lerner Valtrex study that was blinded and placebo controlled found that after 6 months, the average improvement on the EIPS scale in the placebo group was 0.41 points, and 1.12 points for the treated group (Table V in the study). The study then ran for a total of 36 months, during which the treatment group improved by an average of 3.2 points in total (but there was no placebo arm during these 36 months).

 

[Jonathan Edwards]

One thing that slightly puzzles me - the phase 2 study was double-blind and placebo-controlled, so, given the phase 3 results, presumably the conclusion is that the phase 2 study was simply too small to yield reliable results.
Why was there still apparently a much bigger placebo response in the ritux arm? Is that just a statistical fluke?

Yes it is a statistical fluke. The trial is actually quite a good lesson in how easy it is to find such flukes. The primary endpoint failed to show anything. At six months there was a difference that would have been statistically significant if it was the predefined primary endpoint. But that is just because the status of patients was wandering around wildly over time. I spent a lot of time looking at the longitudinal profiles and was always worried that they did not have a clear pattern.

 

[Indigophoton]

Yes it is a statistical fluke. The trial is actually quite a good lesson...

Thank you for the clarification.

 

[Hip]

At six months there was a difference that would have been statistically significant if it was the predefined primary endpoint. But that is just because the status of patients was wandering around wildly over time.

In terms of trying to run better clinical trials for ME/CFS, I wonder if it would help to select a cohort whose level of ME/CFS severity has been pretty stable over say the previous 3 to 5 years. This might tend to prevent these wild random fluctuations in status during the trial.

So in selecting the cohort for a trial, patients could be questioned as to whether their ME/CFS severity varies in a random way, and those who do report such random variations could be excluded from the trial.

Of course, those patients who generally have random variations in severity might comprise a different subset of the illness, in which case you would be excluding that subset. So another approach might be to incorporate those patients with random variations in a trial, but treat them as a separate group in your statistical analysis.

 

[Trish]

I think one possible way to deal with testing a fluctuating condition like ME in a clinical trial, and gauging whether an effect of a treatment is real, is to have participants wear a step and pulse monitor for several months to a year before they enter the trial, throughout the trial and for a year or more afterwards depending on the trial's length of followup.

There must be some way statistically of distinguishing a long term trend in a patient from a random fluctuation.

If the 2 day CPET weren't so likely to cause harm, I'd suggest doing this before, immediately after, and at long term follow up too. And of course taking lots of blood to be tested for changes once we know what changes are meaningful for ME.

 

[Jonathan Edwards]

Fluctuations aren't actually that much of a problem. And the best test of a good treatment is the eyeball test in my experience. When a treatment works a line that looks like a bad drawing of the himalayas slips smoothly into one that looks like the pacific ocean on a calm day - flat.

 

[Hip]

I think one possible way to deal with testing a fluctuating condition like ME in a clinical trial, and gauging whether an effect of a treatment is real, is to have participants wear a step and pulse monitor for several months to a year before they enter the trial, throughout the trial and for a year or more afterwards depending on the trial's length of followup.

I've often wondered how accurate a picture step monitors will provide of the activity level of a patient. If you are someone who prefers more sedentary activities anyway — like reading, writing, painting, working at a computer or tablet, etc — you might become very active in these activities as your health improves, but that may not necessarily be reflected in the number of steps you have walked.

So when an ME/CFS patient improves via a treatment in a trial, as their brain fog clears and their ability to walk longer distances returns, depending on their inclinations, some patients may start running around all over the town, clocking up their steps, whereas others might go the library to read, or spend all day working on their computer, enjoying and utilizing their new-found mental clarity resulting from their brain fog lifting.

 

[Hoopoe]

So as an ME/CFS patient improves via a treatment in a trial, as their brain fog clears and their ability to walk longer distances returns, depending on their inclinations, some patients may start running around all over the town, clocking up their steps, whereas others might go the library to read, or work on their computer, enjoying and utilizing their new-found mental clarity resulting from their brain fog lifting.

I find it difficult to picture an improvement in health without some improvement in activity levels, even in the most intellectually inclined person. That's based on my experience as intellectually inclined person. When in good health, a walk to the park is enjoyable and effortless and a good way to sort one's thoughts and relax.

 

[Hip]

I find it difficult to picture an improvement in health without some improvement in activity levels, even in the most intellectually inclined person.

I agree, there will likely be some increase in walking in the more cerebrally-inclined patients, even if it's only a walk to the bookshop. But it is possible that increase may be less than you get with a person who is more physically active by nature.

The sedentary vs physically active types does not just map onto cerebral vs non-cerebral: there are also people who are naturally couch potatoes, and if their ME/CFS improves, they may just return to their original couch potato ways.

 

[Trish]

I've often wondered how accurate a picture step monitors will provide of the activity level of a patient. If you are someone who prefers more sedentary activities anyway — like reading, writing, painting, working at a computer or tablet, etc — you might become very active in these activities as your health improves, but that may not necessarily be reflected in the number of steps you have walked.

I take your point, but I think, from my sample of size 2 (self and daughter), both book worms and home bodies by inclination, our level of activity has still dropped dramatically since we've been housebound ill.

I can't remember the last time one of us walked around a shopping centre, or a library, or to visit a friend, go to the cinema etc - all just everyday stuff that adds to the step count significantly.

Even if I never did any long hikes again and never went back to tap dancing classes, just being well enough to do everyday stuff healthy people take for granted would quadruple the 1000 or less 'steps' I manage daily at the moment inside my house.

And for the years I was mildly affected and still working, I had to cut back drastically on social activities, get shopping delivered, cut out anything not completely essential.

I agree it's not a perfect gauge of improvement - things like 2 day CPET, step tests, blood tests, employment records, symptom diaries etc etc should also be used to help get a full picture.

I take your point also that some people with ME are more cognitively affected than physically, and there should be some way of measuring this too over time.

 

[Hip]

What we really need is a disease mechanism, it would help us interpret test results better and suggest what kind of trials should be conducted.

We do need to home in on a disease mechanism.

I do like to the general ideas outlined by @Jonathan Edwards, that ME/CFS may be caused by some dysfunctions in immune signaling, which then leads to a chronic aberrant immune response. That immune dysfunction ultimately might become the best target for treating ME/CFS.

An aberrant immune response could also neatly explain the viral data we observe in ME/CFS. In the case of enterovirus for example, we know that there is a chronic intracellular infection in ME/CFS patients' muscles and intestinal tissues. And we know that when viral load is reduced by antiviral or immunomodulatory treatment, ME/CFS symptoms substantially improve.

However, that intracellular infection alone may not explain ME/CFS, because you also find the same intracellular infections in a percentage of healthy people. But if you add an aberrant immune response into the equation, then that intracellular infection plus an aberrant response to it might give rise to the viral subset of ME/CFS.

 

[Trish]

And we know that when viral load is reduced by antiviral or immunomodulatory treatment, ME/CFS symptoms substantially improve.

Clinical trial evidence for this?

 

[Hip]

Clinical trial evidence for this?

From one of Chia's papers:

Of 10 patients with stable, persistently high titer of neutralizing antibody against CBV-3 and CBV- 5 were treated with ribavirin for 4 months, and 5 patients were subsequently treated with interferon/ribavirin for 2 to 6 months.

Seven of 10 patients had significant improvement of fatigue and viral symptoms during ribavirin therapy, along with more than a fourfold decrease of neutralizing antibody. Most patients relapsed within 1 to 2 weeks of medication discontinuation, with subsequent rebound of neutralizing antibody to pre-treatment baseline

Chia also found that levels of enterovirus in stomach tissues (direct evidence of the virus) went down in patients that improved on oxymatrine treatment.

 

[arewenearlythereyet]

So 10 patients? Interesting but no cigar (for me)

 

[Hip]

So 10 patients? Interesting but no cigar (for me)

You don't expect actual useful things like funding for ME/CFS antiviral studies do you? Good luck with that.

Dr Chia wanted to do a study on oxymatrine treatment for enterovirus ME/CFS, but could not get funding. He ended up doing a quasi-study, the good results of which are available, but it is not published. In 5 patients he took stomach biopsies before and after oxymatrine, and showed the viral loads in the tissues went down as patients improved. But that sort of thing really needs funding to do properly.