Preprint: Treatment of Long COVID symptoms with triple anticoagulant therapy 2023 Laubshder, Pretorius et al

[rvallee]

This process you are describing has yielded exactly nothing in decades. So I don't think much of it. Hasn't even produced a single bit of useful data in basically dozens of similar conditions and barely delivers anything in general. Medical breakthroughs are made in labs, not because of a perfect study that pleases bureaucrats.

No need to be involved in a failed process to see it has failed, just as you don't need to be a programmer to notice when software doesn't work. The failure is visible for all to see.

The idea that research should begin at a later phase makes as much sense as public testing of software that isn't even in an alpha phase, doesn't even have a user interface. This could go on for a millennia and it wouldn't even amount to anything.

It's good to be critical but seeing this makes it pretty clear how many of you would have dismissed work like Barry Marshall's. Or in fact any work that isn't mostly bureaucratic checkboxing. Maybe rightfully so, but breakthrough research never looks like previous research. That's the breaking through part. It breaks textbooks and conventions.

I understand this slow process is the norm. And it has failed us miserably and hundreds of millions more. We need to move fast and break things, and it's going to have to happen because none of this damn system is capable of actually delivering squat. It's likely going to happen anyway, precisely because of how inept the whole system is revealing itself to be.

 

[Hutan]

This wasn't really research though, I'm not sure why you are defending it @rvallee. They had 90 patients who were already there, really wanting the treatment to work, they were treated and then they were asked how they felt. It's very much along the lines of a BPS study.

They could have at least blinded the evaluation of the blood slides (initial and after treatment). It could have been done quite easily, making up some slides with the blood of healthy controls, and ensuring that the people evaluating the slides did not know which slide was which. We still don't know if the microclots are an artefact of blood processing or really part of the pathology of some diseases.

And why not go straight to a small RCT? The researchers were already feeling convinced about the safety and the efficacy of the treatment.
20 patients given the treatment + 20 patients given a placebo - blood samples before and after treatment; symptom survey
20 patients with a recent Covid infection and no persistent symptoms - blood samples
Evaluate the blood samples in a blinded fashion.
That's not a hard trial to do, and it would tell us if a bigger trial was worthwhile.

 

[SNT Gatchaman]

It's good to be critical but seeing this makes it pretty clear how many of you would have dismissed work like Barry Marshall's.

I'm pretty sure that hypercoagulability is going to be shown to be an important factor in LC, and likely ME. There are just too many signals and indeed we had a decent paper that had good evidence of this in LC (Analysis of thrombogenicity under flow reveals new insights into the prothrombotic state of patients with post-COVID syndrome, 2022).

I suspect that this work on microclotting will be developed and ultimately could be shown to be happening in vivo — but we're a long way from it being merely an artefact of hypercoagulability as Jo has said from the outset. The team's hypothesis includes the idea that fibrin degradation products aren't high (#alltestsarenormal) because fibrinolysis is impaired, so fibrin is not being degraded; and microclots persist when conventional thinking says they should degrade or simply be filtered. We do have potentially supportive secondary evidence in vivo, eg the hyperpolarised Xenon MRI studies.

If so, then yes that would be a paradigm shift in understanding thrombosis and the prothrombotic state and this could be a Marshall moment: "why would anyone treat a stress ulcer with antibiotics?" -> "why would anyone treat fatigue or brain fog with anticoagulants?". Similarly, as the team points out in their paper: using anticoagulants not to make people hypocoagulable, with inherent risks of major bleeding events, but instead to return people from hypercoagulable to normal. If that's able to be monitored and confirmed then treated patients would have similar risks of bleeding as a normal healthy person.

Apart from the major problems in study design as outlined up-thread (it's a case series not a trial), I think one of the other main problems is pushing too hard for this to be The Explanation. This preprint shows a subgroup of patients who didn't improve and who still had hyperactivated platelets and microclots. I suspect at best this would be a symptomatic treatment, while some people naturally recover, but that eg upstream immunometabolic dysregulation still needs to be addressed in the non-spontaneous recoverers.

There's handwaving, eg "widespread endothelialitis inhibit the transport of oxygen at a capillary/cellular level. This provides a ready explanation for the symptoms of Long COVID." It's going to be much more complex than that. Perhaps it might contribute and from my own experience of "vascular/blood weirdness" I'm suspicious that it does — but it's a long way from the whole picture. I recall Resia Pretorius did recently tweet that they thought microclotting was downstream.

I would like to see demonstration of microclots in vivo, eg Martin Kräter's microfluidics studies or maybe via flow cytometry which I believe this team is also looking at. Studying responses with such objective markers to correlate subjective symptoms against (controlled) triple therapy might then be able show something useful — whether effective or ineffective.

This was not a trial but we are advocating for trials. It was simply a snapshot of a clinician-initiated treatment regimen from 1 dr’s practice. Various groups are working on planning of trials.

-the numbers are relatively small
-these results need to be replicated in randomised, double-blind, placebo-controlled trials with some urgency.
These studies need to have clear objective physiological & laboratory outcomes

 

[rvallee]

I think maybe we're just not seeing this from the same perspective. I don't see this as an evidence-based type trial for this particular treatment. That process is generally worthless and frankly I wrote if off entirely as likely to produce anything useful.

Simply asking about symptoms is also useless in itself, it's the correlation that matters. It's not the treatment that is being tested, it's the hypothesis.

That's probably a product of the fact that we're trying to figure out the pathophysiology, whereas normally it is known and the EBM process is about trying to target it. But the pathophysiology here can only be figured out from large numbers, whereas distinct diseases can often do much of this phase with a few validated patients. We don't even know how to really validate patients.

This study tells us very little about this particular treatment, and even if it had been a 1,000-strong quantum-randomized double-blinded controlled trial, no MD worth their license would prescribe this treatment.

Research is a process. Maybe we're too used to having one shot only and that's it, but there are supposed to be subsequent steps. It's not supposed to be THE one trial, it's the first one. No effective treatment will ever be found from a single trial, nor would a single one tell us much about whether it is really effective.

The only reason the trial is worth anything is because of the observed effect on the microclots detection and correlation with symptoms. If valid, it tells us that it is a viable research path. I don't know if it's valid, and there is no textbook for this because it's not previously known. Medicine is still finding entire new organs, though small ones, so I think people should chill a bit on this.

I'm not comparing the two studies or saying this is it for us but it's a lot like how bacteria could not possibly survive in the stomach lining. Science is a process and anything new will, by definition, be controversial as not being textbook. That's normal.

It's like Ron Davis said about his difficulty getting funded from the NIH: they don't do exploratory research. This is normally done at universities from tenured budgets, which is what then leads to grants for formal studies.

We have to do the academic stuff using the grant model because no one can get tenured working on any of this. It's terrible but this study is in the observe-hypothesize-test model of scientific research, not an evidence-based paradigm.

I don't know if that's really how the researchers intend it, but that's where I see its value. We have to work with the wrong tools because we are denied the standard ones. It's bound to be messy.

But it's really clear that the current model only works as well as the overall level of technological progress. So any breakthrough will likely involve a new piece or method of technology.

It's not so much about method as getting a clear signal, and whatever the method used, once the technology needed to understand this is figured out, it will be pretty obvious.

 

[Hutan]

I think maybe we're just not seeing this from the same perspective. I don't see this as an evidence-based type trial for this particular treatment. That process is generally worthless and frankly I wrote if off entirely as likely to produce anything useful.

I'm still not understanding. I still believe that trials with a sound set-up (RCT) can help us know what is going on. The real problem is that we have seen so very few good ones in ME/CFS.

I haven't written off the idea of microclots. But I need to see some more solid evidence before I believe in them - blinded assessments resulting in samples from people with LC or ME/CFS being separated from samples from healthy controls.

I agree with Resia Pretorius that it was not a trial. It's great that the people involved are calling for trials. So, that's good, hopefully the evidence is coming.

Research is a process. Maybe we're too used to having one shot only and that's it, but there are supposed to be subsequent steps. It's not supposed to be THE one trial, it's the first one. No effective treatment will ever be found from a single trial, nor would a single one tell us much about whether it is really effective.

Fair enough. But it's also fair enough for people to say 'this isn't evidence yet'.

 

[Hubris]

That process is generally worthless and frankly I wrote if off entirely as likely to produce anything useful.

Maybe I'm not understanding, but this is completely wrong. RCT is often the only way to know if a treatment works and if the trial is large enough and the results significant you can trust that it does work. You can try asking anyone who works at a hedge fund and analyses pharma stocks for a living and they will tell you this.

Why did i pick this example? Because it's a job where you have to perform well if you want to earn your salary - pick the wrong companies and you are fired. So you have to develop a reality based approach. You have to know what actually works and what doesn't. It's not like medical researchers where you can spend your entire life publishing poor studies, contribute absolutely nothing to the science and still get paid.

 

[SNT Gatchaman]

That's probably a product of the fact that we're trying to figure out the pathophysiology, whereas normally it is known and the EBM process is about trying to target it. But the pathophysiology here can only be figured out from large numbers, whereas distinct diseases can often do much of this phase with a few validated patients. We don't even know how to really validate patients.

A possible counter-example might be the findings in the metformin vs fluvoxamine vs ivermectin study There they did a decent attempt at an RCT (admittedly as an add-on to the original acute study) with good attempts at blinding and look to be showing a significant preventative effect against Long Covid with a short 14 day course) of metformin instituted within 3 days of infection. That might give us a treatment (or at least a partial preventative) first and then secondarily possibly tell us something about the pathophysiology of LC.

The trick in that case though is we don't know enough about how metformin works to answer immediately what the pathophysiological mechanism might be. Metformin's effects are pleiotropic, just some of them being effects on metabolism/mitochondria/respiratory chain, as well as the microcirculation, NO signalling and vascular permeability. As a partial structural analog, it also reduces serum ADMA, which is one of the (not) excreted metabolites that was significantly different in Maureen Hanson's recent urinary study.

 

[Andy]

Despite what we might be accused of, I, and others on this thread, are perfectly capable of holding two thoughts. One, that the idea of microclots is worth investigating, even if just to rule it out, and two, that in order to do so it will actually need more than just a series of case reports. And as SNTG highlights, the authors of this paper acknowledge this as well.

Declaring that research is broken and and demanding that therefore things should be done in a different way doesn't achieve anything in the real world. We have collectively worked for a long time to highlight how BPS research into ME/CFS, and other areas, does not follow existing best research practices and therefore is fatally flawed; to then argue that biomedical research should not need to follow the same existing best research practices because the "process is generally worthless" undermines all of that effort.

 

[SNT Gatchaman]

And why not go straight to a small RCT? The researchers were already feeling convinced about the safety and the efficacy of the treatment.

They may not be able to gain ethics at this stage. They do have ethics to do an observational study, where a clinician has already independently decided to offer treatment with informed consent, but it might be viewed differently when the committee would be giving permission for this treatment (or placebo) - which everyone would naturally be wary of, because it sounds inherently dangerous. The calculation should weigh the dire state of the patients as well as the mechanistic principles being proposed.

It does seem from the tweets that other international centres are going ahead. Hopefully the level of evidence will rise over the year - RCTs will take time to enrol and report - but maybe we'll see something relating to newer/different fundamental science investigations before then.

 

[Hubris]

They may not be able to gain ethics at this stage.

Have the ethical boards considered that maybe, just maybe, spewing out one unblinded study after another for decades, luring desperate patients into trying dozens of harmful drugs that don't work, might in fact be more unethical than doing proper RCT studies? If the researchers have limited funds (and also time but nevermind that, nobody cares if patients have to wait 40 years for a treatment) why not let them do a RCT right away?

As absurd as this sounds, i can believe it. I've had personal experience with researchers struggling to get ethical approval for a stool microbiome study (which they had already done many times in other illnesses), which literally poses no risk to the patient (you just poop and mail it to them). What reason could you possibly have for denying approval? Oh, i know - someone on the ethics board thinks ME is a fake illness so any biomedical research is inherently immoral and a waste of money.

My question:
Is there anyone who is supposed to oversee and correct these blatant issues with ethics boards or is it another case of "don't worry, we police ourselves because we can never be wrong" kind of deal? Well, we are talking about medicine after all so i guess there is no accountability as usual.

Sometimes i understand why nobody wants to study ME. You find so many roadblocks ahead of you, get treated like shit by every colleague, at a certain point you just choose something else.

 

[rvallee]

I'm still not understanding. I still believe that trials with a sound set-up (RCT) can help us know what is going on. The real problem is that we have seen so very few good ones in ME/CFS.

I haven't written off the idea of microclots. But I need to see some more solid evidence before I believe in them - blinded assessments resulting in samples from people with LC or ME/CFS being separated from samples from healthy controls.

I agree with Resia Pretorius that it was not a trial. It's great that the people involved are calling for trials. So, that's good, hopefully the evidence is coming.


Fair enough. But it's also fair enough for people to say 'this isn't evidence yet'.

I think that the last few years have made it clear that pragmatic trials without a theoretical foundation are completely useless, and probably harmful because of the huge wasted opportunity cost (decades wasted on CBT-GET means having given up on real research all this time).

RCTs are useless here because they need to be large and extremely rigorous to be useful. And that's extremely expensive so trials are usually a 3-step process. Only a full process is valuable, same as the process to validate vaccines. It's massively expensive to do this, it's not a working model for trial-and-error.

This model only works when the pathophysiology is understood and treatments are being compared for their effect on the pathophysiology. And even then, unless the effect is massive, it's probably worthless.

Nothing matters other than understanding the biology. The rest is just noise. Finding any treatment that has an effect is only significant if it leads to understanding the pathophysiology. I mean it when I say this paradigm is useless, said it many times before. It's not just for when I like the outcome. I'd likely be more suspicious of any treatment that only has preliminary EBM "evidence", frankly. This paradigm is basically like a reverse-seer: it's almost perfectly wrong all the time.

And here as far as I'm concerned is not an outcome, merely a step. This is indeed evidence for nothing, it's a process of figuring out the cause and that's many complicated steps.

If there's anything to any part of the evidence found so far, whether to do with something in the blood, or something with the vascular/endothelial system, it all has to be figured out before it can help patients. The process of trying random treatments in typical BPS/EMB fashion is the worst idea ever, and that's not at all how I see it here.

When it comes to asking for controls, there is no evidence that the clots disappear naturally. So no need for controls here. I said trial a few times but this really is a basic science study that simply has to rely on awkward tools out of necessity.

 

[rvallee]

Maybe I'm not understanding, but this is completely wrong. RCT is often the only way to know if a treatment works and if the trial is large enough and the results significant you can trust that it does work. You can try asking anyone who works at a hedge fund and analyses pharma stocks for a living and they will tell you this.

It's a massively expensive process that only works in some contexts. I don't think it works outside of drug development at all.

We can even look at examples where it technically went though the process, but the evidence is very unreliable, in the form of SSRIs. The biology isn't known, either for the problem or the solution, so the evidence is all over the place and generally poor or conflicted.

Medicine has made everything to be about pathophysiology or bust, they demand it. It's not the system we deserve but it's the one we have to work with. Or ideally around, soon enough.

 

[Trish]

Oddly enough, I think one of the most important reasons to do double blind clinical trials in cases like this is precisely because the underlying biology hasn't been elucidated yet.

We need trials so people don't end up wasting resources, time and potential harm from doctors going on prescribing potentially harmful and/or useless things like LDN, abilify, this triple antiocoagulant protocol, unless there is real evidence that they actually work.

That's why the rituximab trial was important - to find out whether it works or not once and for all so we could all move on and people not be put through such treatment any more.

So I say, yes, they do need to get on and do at least one really properly run blinded clinical trial of this treatment before even more people get drawn in to a potentially dangerous treatment unless it really does work.

Of course at the same time they should also go on doing properly run studies of the microclot phenomenon to establish whether it really is something fundamental to Long Covid or an artefact of the processing of the blood or whatever or a downstream effect of some other factor of the covid virus that comes back once anticoagulation treatment stops.

While we are in a situation where people are experimenting with all sorts of apparently random treatments out of desperation to find a cure, we really need the doctors prescribing these treatments to step up and do proper pilot studies that can provide useful information, whether it's crossover trials, controlled trials, blinded trials, open label with objective endpoints, or whatever.

 

[Hutan]

Is there anyone who is supposed to oversee and correct these blatant issues with ethics boards or is it another case of "don't worry, we police ourselves because we can never be wrong" kind of deal? Well, we are talking about medicine after all so i guess there is no accountability as usual.

Obviously things will vary from country to country and possibly from ethics board to ethics board. The system needs to be improved so that there is more consistency. In my country, there is a system for overseeing ethics boards, there is accountability. It's imperfect, but there is a substantial desire for improvement and some evidence of that change already happening.

Reviewing an application for approval can often be a really time-consuming job, and usually the people on the boards are squeezing in this work between their normal day jobs. Payment, on an hourly basis, is typically minimal. The people reviewing a particular application aren't necessarily the reviewers with the most relevant knowledge. So, I wouldn't be criticising the people trying to evaluate applications across the board; mostly they aren't the core problem. It's the system where things need to get better, and I think it can. Perhaps AI can help improve ethics approvals.

Have the ethical boards considered that maybe, just maybe, spewing out one unblinded study after another for decades, luring desperate patients into trying dozens of harmful drugs that don't work, might in fact be more unethical than doing proper RCT studies?

I totally agree, poorly designed studies not only don't move knowledge forward, they are unethical.

 

[Hutan]

I think that the last few years have made it clear that pragmatic trials without a theoretical foundation are completely useless, and probably harmful because of the huge wasted opportunity cost (decades wasted on CBT-GET means having given up on real research all this time).

RCTs are useless here because they need to be large and extremely rigorous to be useful. And that's extremely expensive so trials are usually a 3-step process. Only a full process is valuable, same as the process to validate vaccines. It's massively expensive to do this, it's not a working model for trial-and-error.

I don't think an RCT has to be large to be useful. A small one can identify things that are worth more study.

And actually, maybe fewer, better quality trials when it comes to treatments would be a good thing, be less wasteful and lead to less misinformation. Maybe researchers can make better use of databases, like the US Veterans Affairs database that looked retrospectively at the question of whether paxlovid reduces post-Covid symptoms. They identified that there might be something there, and now RCT's trialling treatments have a good basis, even without the pathophysiology being understood. A useful treatment can provide a clue to the pathophysiology.

Those of us who have spent any time looking at the studies posted in this forum know that the majority of them are rubbish. Cutting out the confusion produced by reports of people feeling better as a result of an open-label treatment might actually get us to useful treatments quicker.

Medicine has made everything to be about pathophysiology or bust, they demand it. It's not the system we deserve but it's the one we have to work with. Or ideally around, soon enough.

Nothing matters other than understanding the biology. The rest is just noise. Finding any treatment that has an effect is only significant if it leads to understanding the pathophysiology.

I assume you advocating for biology to be understood? For sure, researchers try stuff in the lab and get ideas that they want to test more rigorously. That's science, and it's fine. But, how do we know that the scientist has found something useful?

When it comes to asking for controls, there is no evidence that the clots disappear naturally. So no need for controls here. I said trial a few times but this really is a basic science study that simply has to rely on awkward tools out of necessity.

But, we still don't really know if the in vitro microclots indicate a pathology, if they really are more common in the blood of people with Long covid than in healthy people. That's why healthy controls are needed. I'm not sure how whether the clots disappear naturally or not is relevant.