The ethics of experimentation with treatments based on unevidenced hypotheses - discussion thread

Moved posts

That's why I'm in favour of testing cytokine responses. ME's symptoms are similar to viral infections too, and those symptoms are due to the immune signals, not the virus itself. I rate ME studies involving immune signals more likely to reveal something useful than ones involving mitochondrial function.

 

Does viral infections cause PEM on a regular basis?

 

Seems to contribute to switching ME on for many of us. There could be an underlying start to ME that smoulders for years and then gets fully triggered by a flu etc.

 

Many appears to have a viral trigger for their onset, but I don’t think PEM is a normal reaction to an infection. So I’m not sure that the infection is driving PEM, rather it’s causing something that then causes PEM.

Which is why I’m sceptical to testing cytokines on healthy people because 1) it will probably not produce PEM, and 2) if it does, it might no be reversible.

And infections have been associated with a worsening of ME/CFS, so it’s probably not safe to test on someone that already has PEM.

 

I think there is an underlying problem first and an infection switches things up and introduces PEM? I don't know, just throwing thoughts and experiences out there.

It could be a number of 'hits' that further things.

 

Sure, but when we don’t know we might risk causing ME/CFS in healthy people that would otherwise not have gotten it, or making patients worse. I don’t see how that risk can be justifiable.

 

Sorry, I think we are talking about two different things. I'm not talking about treatment or testing here. I don't understand any of that side of the conversation.

 

Apologies, I got confused!

 

Moved posts

I don't know, but since the symptoms are similar, they may use at least part of the same mechanism.

I certainly wasn't suggesting that. I was suggesting it for PWME, to see if one of those signals reliably triggered or affected symptoms.

Would there be a possibility of long-term worsening for the subject? Yes. Is there a possibility of long-term worsening if that subject doesn't do any experiments? Yes. Is there a possibility that experimentation could result in knowledge that could lead to treating ME? Yes. The question is what those relative probabilities and outcomes are.

I should point out that my personal experience with ME experimentation is that I never had any long-term worsening from it. I'd be willing to try individual cytokines. Someone who tried some "known to be safe" treatment and suffered long-term worsening would have a different perspective.

 

We’ve discussed this elsewhere, but those are not the only considerations. You also have to account for the ethics of providing said interventions to someone. That has to be done by someone with a legal and ethical obligation to not cause harm.

And any worsening when not doing experiments is much more removed from the action than worsening caused directly by the experiments. So you can’t necessarily equate those outcomes even if they have the same probability.

There’s also the question of the likelihood of gaining any useful information from the experiments. Which is probably very, very low because the symptoms are not similar enough if infections don’t regularly lead to PEM. And we don’t even know where to look.

 

I'm not specifically interested in PEM; I'm interested in the symptoms shared by ME (in or not in PEM) and viral infections. My symptoms from food responses, viral infections, and PEM were similar. My symptoms from physically and cognitively triggered PEM were similar but maybe not exactly the same. To me, immune response seems a likely connection between all those.

I don't have much interest in philosophizing about ethics. I don't think there's a universal absolute concept of ethics to judge situation by. I feel that it's not ethical to force someone to undergo risky experiments, but I also feel that it's unethical to prevent someone who has little to lose and feels that the risk of an experiment is low to be able to experiment on themselves. There are cases where an individual can be judged to be unable to judge risk vs benefits well enough (by the judge's personal judgement), but other cases where it's not easy to judge such ability.

The US has a program that allows patients with severe diseases and no approved treatments to try certain drugs. I think it would be unethical to prescribe those to people with less severe diseases without fully disclosing risks and benefits according to the best available data, but that doesn't cover all situations.

Ethics is messy, which is probably why I don't think deeply about it.

 

Which is why we should think deeply about it in order to avoid doing things we actually were not okay with..

 

Ah, but if it takes lots of deep thinking to be able to judge how okay or not okay we are with something, is it really that important? The really important situations are likely fairly clear.

Regarding "avoiding doing harm", is even that simple? Is treatment known to be harmful to a patient with ME? No, since we don't understand ME. Even a known-to-be-lethal-dose-in-healthy-people isn't 100.00% sure to be harmful to a PWME. Likewise, anything isn't 100.00% guaranteed to not be beneficial to a PWME. It's possible that something could be an effective treatment, and withholding that might be condemning a bedbound patient to unnecessary prolonged suffering. So, for any given dose of something, there's no absolute judgement whether giving the dose will be more harmful than not giving the dose. How do you fairly judge fuzzy definitions based on fuzzy and incomplete data?

If someone developed a treatment that was theoretically effective for ME, and maybe even worked on mice (How can we tell?), would you insist that severe, at the point of asking for assisted suicide, patients wait 20 years for a proper long-term study of safety?

 

That’s an absurd assertion.

Clarity can’t be used to determine importance - only importance can. And lack of ethical clarity certainly does not automatically mean that something isn’t important.

I can think of many topics that are important, but not ethically clear. Like assisted suicide. But let’s not discuss those here.

We don’t need certainty about harm. We only need to be reasonably sure that it can be harmful.

I don’t think we’re going to get past this point. The likelihood is incredibly small that it will be of benefit, because most substances are of no benefit for most conditions. That’s a statistical fact that you can’t just ignore. We know more than just «we can’t know for certain that it won’t help».

There are no drugs that would require 20 years of studies to assess the safety within reason. There is no reason to discuss this example.

And the ethics really are not clear about giving someone something in what is essentially a trade for their suicide. Because if you accept that - where do you draw the line? What can you expect or mandate someone to give to someone if they threaten to commit suicide if they don’t get it?

It would open a whole can of worms, that coincidentally, has no ethical clarity.

You would also have to consider the alternative: if we know that a drug could work because we know the mechanism(s) of ME/CFS, according to e.g. Jonathan Edwards, it would be fairly quick to set up a trial. And if it works - it would really work so it would be very obvious. Which means that the person in question might get access to the drug in a year or two.

If they knew that piece of information, maybe they would be able to hold on for a while longer? I think it’s fair to assume it would at least lower the risk of suicide.

 

It isn't that difficult. Dogs are roughly two feet long but vary hugely. Whales are roughly 50 feet long but vary hugely. But we are pretty sure whales are bigger than dogs. The risks of most drugs are hugely underappreciated. You get to realise that when, like me and all other doctors, people have died as a result of your prescriptions. The chances of benefit in any given disease from a given drug, even if it sort of targets the right pathway, is small, usually vanishingly small.

 

Is that going to be a problem for us, if research identifies a dodgy pathway in ME/CFS?

 

I would also be interested to know the answer to this, especially in the context of this comment.

So is it the case that if we have a first approximation of ME/CFS mechanisms a drug would have a much better chance of succeeding, or is it still a small chance?

 

No I think that takes the remark out of context. If we know what we want to achieve we can be highly selective about choosing a drug or make a new one from scratch. When it came to choosing rituximab for RA I knew there were maybe ten out of 20,000 drugs that targeted the right area but I could see that only one of them could possibly be worth trying - largely because we already knew the others didn't work.

 

My personal experience is that I've encountered several surprising treatments for my ME. Cumin blocking and even curing PEM? I would have considered the chance of that effectively zero ... but it worked amazingly well for me. I developed intolerance to something in plant embryos; what could possibly treat that? Squash would not have been on my list, yet it did fix that problem. Of course, maybe that was due to a microbe in the squash, rather than a chemical, but whatever it was worked.

My personal experience is also that I haven't suffered any lasting harm from any experiments. I don't know what the statistics say about risk from a specific treatment or the likelihood of benefits, but if both are very low probabilities, it would be hard to make a clear decision.

My personal experience is also mostly treatments that worked after a few, or even a single dose, so the risk of harm should be quite low. I'm not suggesting putting people on multi-year drug experiments for the tiny chance of a benefit. I'm thinking more about single dose experiments.