Preprint Registered interventional trials fail to cover patient-prioritised treatments in Post-COVID condition (PCC), 2026, Gabriel et al

[forestglip]

Registered interventional trials fail to cover patient-prioritised treatments in Post-COVID condition (PCC)

Spoiler: Authors

Gabriel, Marc; Koster, Steffen; Puta, Christian; Katzer, Katrin; Reuken, Philipp A.; Stallmach, Andreas

Purpose
Despite the substantial health and socioeconomic burden of Post-COVID condition (PCC), no disease-modifying therapy has been established, largely because of clinical heterogeneity and incompletely understood pathophysiology. Major national and international funding initiatives have been launched in response. Hundreds of interventional trials have investigated PCC, yet patients and their treating physicians largely rely on non-evidence-based treatment options in clinical practice.

Methods
Here we quantify the overlap between the registered trial landscape and patient-led practice. We deduplicated 854 registry records (ClinicalTrials.gov, DRKS, EU-CTR; snapshot 16 July 2026) into 714 unique interventional trials. Interventions were categorised as pharmacological or supportive/non-pharmacological, and endpoints as symptom-oriented or mechanism-oriented according to their therapeutic objective.

Results
Only 51 of 714 trials (7.1%) tested any of the 29 patient-guide treatments; 20 were active. Post-exertional malaise, the cardinal symptom of the ME/CFS-like phenotype, was captured in 57 trials but was a (co-)primary endpoint in only eight. Ten guide treatments - the first-line repertoire for autonomic dysfunction, mast-cell stabilisers, and several repurposed neuro-modulators - had no registered trial in any registry, and the German registry DRKS contributed no guide-relevant trial.

Registered evidence concentrated instead on SARS-CoV-2 antivirals, immunoglobulins and hyperbaric oxygen. Only three of the eight trials with PEM as a (co-)primary endpoint tested an intervention directed at a proposed driver of PEM; none addressed microvascular or bioenergetic mechanisms.

Conclusion
The registered landscape emphasises symptomatic treatment. There is a substantial unmet need for well-powered, mechanism-based, disease-modifying trials with biomarker-driven designs and objective, exertion-based endpoints.

Web | DOI | PDF | Research Square | Preprint

 

[SNT Gatchaman]

The most pronounced discrepancy between research activity and patient priorities, however, concerns neither pharmacological interventions nor a specific treatment modality, but a highly relevant patient-centred outcome: PEM.

 

[Jonathan Edwards]

I am not at all sure how you could make PEM an objective outcome. How do you decide whether or not symptoms were exacerbated by exertion? There may be some people who regularly note a temporal relation between doing things and feeling bad but my understanding is that this is not the core issue in terms of 'PEM' for ME/CFS and likely to be hard to distinguish from post-exertional fatigue in other conditions.

We went through this exercise for ME/CFS in the UK working group and nobody could think of any drugs to repurpose based on patient experience other than LDN.

 

[Jonathan Edwards]

I have read through rather quickly. I think the authors are missing out on practical realities and have an unrealistic folklore-informed conception of what should be the best targets. They make use of the Patient Led Research Collaborative work which we thought was pretty unhelpful.

I wonder if any of these authors are members here? It would be good to have a dialogue. I worry that despite interest and money being focused on ME/CFS in German-speaking countries recently, there is a reliance on folklore-informed biology that is likely to lead to wasting a lot of that money.

 

[Creekside]

There may be some people who regularly note a temporal relation between doing things and feeling bad but my understanding is that this is not the core issue in terms of 'PEM' for ME/CFS and likely to be hard to distinguish from post-exertional fatigue in other conditions.

Then what is the core issue? To me, PEM was a worsening of existing symptoms (some possibly negligible outside of PEM) a consistent delay after a consistent exertion event. I can't recall getting post-exertional fatigue that matched my ME PEM before developing ME. My ME PEM matched flu symptoms quite well, but that's easy enough to separate from PEM.

How do you decide whether or not symptoms were exacerbated by exertion?

Statistics? If climbing a 10' ladder twice consistently triggered PEM 24 hrs later, resulting in a relatively consistent severity, and this was consistent for 3 or whatever repetitions is statistically relevant, and the symptoms don't flare up without that or similar exertion events, then it should be reasonable evidence of a correlation. Some PWME have multiple triggers (foods, emotions, hypersensitivities, etc) for symptom flares, so they'd be statistically difficult for proving the trigger, but I assume that other PWME have fewer and easier to identify triggers, and could form a cohort for PEM studies.

 

[Jonathan Edwards]

To me, PEM was a worsening of existing symptoms (some possibly negligible outside of PEM) a consistent delay after a consistent exertion event.

But others describe it as very unpredictable. Moreover, I don't see this as the main burden of ME/CFS, merely as its temporal pattern. The main burden is surely the level of symptoms and disability over the long term. Trials should try to reflect that.

There is often an implication that 'PEM' is something somehow different from the symptoms of ME/CFS over and above its temporal relation to exertion. We do not have any basis for saying that as far as I know. If there is a difference it was time that it was enunciated.

Statistics? If climbing a 10' ladder twice consistently triggered PEM 24 hrs later, resulting in a relatively consistent severity, and this was consistent for 3 or whatever repetitions is statistically relevant, and the symptoms don't flare up without that or similar exertion events, then it should be reasonable evidence of a correlation.

From what I hear from members this is not realistic, for a variety of reasons.

 

[rvallee]

It doesn't really make sense to compare to patient preferences, because until there is a theoretical reason to conduct a trial it simply doesn't make sense to even fund one. Before HIV was discovered, it's doubtful that preferences in the AIDS community would align with antivirals, it simply wasn't known, and for sure they would not align with anything that ended up finding effective treatments.

The sad truth is that at least 95% of those trials simply had no plausible reason to be conducted, twice over, because the vast majority of them were explicitly promoting treatment models that have been firmly proven to be failures. A clinical trial without a plausible reason almost never makes sense.

This is not asking the right question, or not asking it the right way. It's like political polls asking people about whether the country is going in the right direction, with groups of people responding the same way for mutually exclusive reasons. These groups don't agree with each other at all, and yet the polls make it seem like the combined option is a major concern.

Patients want research that will figure out the mechanisms, then to conduct clinical trials based on plausible mechanisms. Everything else is pointless. Which points to major issues with the entire system of health care, with how they prioritize, how they deal with failure, how they are incapable of escalating efforts, rather choosing to commit even further to failure.

It remains that the vast majority of research conducted in LC has been so thoroughly wasted that it was easy to predict from the start. The fact that it still happened makes it clear that the problem isn't technical, it's not even with the patients, it's with failed institutions that are stagnant, lack any resemblance of leadership or strategic thinking, and are obviously very comfortable with perpetual failure, no matter the consequences for millions.

It's so bad that the focus actually has to move away from the patients and the illness, and work at figuring out how the hell are things so dysfunctional. Whether efforts align with patient preferences is a distant concerned compared with the fact that they don't even align with the patients at all. They are all based on fictitious models that don't even make sense, bear no connection to reality, and operate in accountability-free systems with no corrective mechanisms and not a lick of introspection or self-awareness.

It's just not normal for experts to fail this awfully. The very concept of expertise makes this impossible. And sure enough, there are no experts on this condition, none that meet the definition of what an expert is. This failure will be studied for centuries, it's the most extreme example of expert systems failure in our entire history. And sure enough, the industry is completely oblivious to it, most are completely indifferent, if not aggressively opposed to doing anything.

Business school often features famous failures, businesses that collapsed despite having a long, successful history and context. Medicine appears incapable of doing this, because the entire system is associated with everything else. The failures are not private, they are not self-contained within a group of people doing things on their own, everyone in the industry is closely associated with them here. It makes it impossible to fix this, self-regulation is simply not a thing, will never be a thing.