Trial Report Plasma exchange therapy for the post COVID-19 condition: a phase II, double-blind, placebo-controlled, randomized trial, 2025, España-Cueto+

Plasma exchange therapy for the post COVID-19 condition: a phase II, double-blind, placebo-controlled, randomized trial
España-Cueto, Sergio; Loste, Cora; Lladós, Gemma; López, Cristina; Santos, José Ramón; Dulsat, Gemma; García, Anna; Carmezim, João; Carabia, Julia; Ancochea, Águeda; Fernández-Prendres, Carla; Morales-Indiano, Cristian; Quirant, Bibiana; Martínez-Cáceres, Eva; Sanchez, Anna; Parraga, Ivonne Graciela; Chamorro, Anna; San José, Alba; Abad, Elena; Muñoz-Moreno, Jose A.; Prats, Anna; Fumaz, Carmina R.; Coll-Fernández, Roser; Estany, Carla; Torrano, Pamela; Puig, Jordi; Clotet, Bonaventura; Tebé, Cristian; Massanella, Marta; Paredes, Roger; Mateu, Lourdes

The post-COVID-19 condition (PCC) is a highly debilitating and persistent postinfectious syndrome that affects millions of people worldwide and has no effective treatment. Therapeutic plasma exchange (TPE) has the potential to improve the PCC by clearing the peripheral soluble pro-inflammatory immune milieu derived from acute or persistent SARS-CoV-2 infection.

In a phase II, double-blind, placebo-controlled, randomized trial, fifty subjects with PCC were randomly assigned (1:1) to receive six sessions of either TPE or a sham plasma exchange and were followed for 90 days (ClinicalTrials.gov registration: NCT05445674). The primary endpoint was safety; secondary endpoints included functional status, symptomology, quality of life, neurocognitive symptoms, and peripheral biochemistry, hematology, coagulation and inflammation parameters.

Both study arms had a similarly favorable safety profile. There were no diferences between groups in any of the efficacy parameters evaluated. Whereas TPE is safe, it did not lead to any discernible improvement of the PCC in this clinical trial.

Link | PDF (Nature Communications) [Open Access]

 

Is it weird to get excited by an honest reporting of the findings?

 

This study has limitations. The most obvious one is its small sample size, which does not allow to fully rule out an effect of TPE on PCC. However, our results are enough to suggest that, if such effect exists, it size its likely to be small and, unless we are ble to identify better treatment candidates, the efficiency of pursuing a larger study of this invasive procedure is low.

 

They tried to:
Participants in the TPE arm performed six sessions of TPE using human serum albumin on days 1, 3, 8, 10, 15, and 17. The placebo arm got six sessions of sham plasma exchange, which involved infusing 200 to 250 cc of sterile saline solution 0.9% on the same days and time that would have been allotted for TPE, using an intravenous line as done with the rest of the participants. Throughout the treatment period, as well as after the final TPE/placebo sessions and during follow-up visits, all participants underwent clinical and safety assessments, and blood samples were collected. Subsequent follow-up visits were conducted on days 22, 45, and 90.

(…)

These interviews were conducted to evaluate the existence of persistent symptoms as well as the progression of their symptoms on days 0, 8, 15, 22, 45, and 90, with the goal of determining whether 2 or 4 cycles may provide a similar effect as 6. Extensive lab analysis profiles were also performed at the same days 0, 8, 15, 22, 45, and 90

 

They haven't looked specifically at patients with increased autoantibodies from what I can tell

 

I don't think it is plausible that autoantibodies cause symptoms in post Covid illness.

If 'Long Covid' was due to an autoantibody it would be against one antigen. But inasmuch as there is evidence for any autoantibodies they seem to describe 10 or twenty antigens as being targeted. But each of those should produce a quite distinct illness. There should be a Long Covid with nephritis, another sort with arthritis, another with hypertension, another with recurrent infection - and so on.

The acid test of causes of disease is always in the detail. It is not good just vaguely suggesting, as we see researchers do, that a flush of autoantibodies might be making people unwell. If they were they would be making people unwell in very specific ways, just as antithyroid or anti-DNA antibodies do. I would be prepared to bet £1000 that autoantibodies are not relevant. And this study provides an honest negative answer in support of that.

 

I don't think so. An autoantibody causing an illness will have clear hallmark clinical expression that does not overlap anyhow with other features. If there are ten autoantibodies causing illness at least three of them should produce something objective like a rash or neuropathy. The picture looks absolutely nothing like my experience of the spectrum of autoimmune disease in rheumatology.

Well, plasma exchange is pretty unreliable at producing benefit in autoimmune disease generally, so yes, very much so. Antibodies can easily come back in three weeks even if they go down. And just reducing circulating antibody doesn't necessarily produce clinical benefit in that time frame.

 

Appears that there is also no short term positive effects with saline infusions, even though some doctors like to prescibe those for things like POTS?

 

Thanks for your insights - would you mind commenting on the following?

It doesn't seem to be as easy as "auto-antibodies = Long COVID", that's for sure. Also, the autoantibodies that are most commonly discussed are the ones against vasoactive receptors such as adrenergic and muscarinic receptors, right?
According to a presentation from the University of Erlangen (https://www.fau.tv/clip/id/53443) these autoantibodies alone aren't an issue. When they get in contact with inflamed or hypoxic tissues though they become functionally active. It must, according to their hypothesis, thus be a combination of endothelial dysfunction and the presence of vasoactive autoantibodies. (Hence, the IA-Surv study by Dr. Mathes of Charité/Havelhöhe only allows participants without persistent spike protein that could cause endothelial inflammation.)
I was under the impression that the Wirth Scheibenbogen hypothesis also starts off with hypoperfusion (due to vasoactive autoantibodies)?
Would you not agree that a causal relationship between e.g. POTS (as a symptom of Long COVID) and vasoactive autoantibodies is not far fetched?
And isn't the latest hypothesis that Long COVID is not a homogenous group but rather consists of several sub groups? Of which one might well be an autoimmune group?
Many questions

Also, in line with @Hutan according to supplementary data, the TPE patients only saw about 70% reduction in IgG titer by the end of their treatment. I was under the impression that when using IgG depletion for autoimmune diseases, we're looking at reductions in the ballpark of 90%?

 

If the saline was infused in such a way as to maintain constant circulatory volume then it might not tell us anything, though.

 

A causal relationship between POT, whether in Long Covid or not, and vasoactive autoantibodies is very far fetched. The antibodies in question have for decades been known to be present in normal people at much the same level. On several occasions they have been claimed to be associated with a variety of conditions. Scheibenbogen's data shows clearly that levels in ME/CFS are not materially different from normals, even if there is a slight shift in average level. It is inconceivable to me that they are pathogenic with these data. Pathogenic antibodies show a very clear cut separation from controls, as a directly causal factor must do.

The presence of other contributing factors does not change the analysis. These autoantibodies cannot plausibly be causal.

Long Covid very obviously consists of a variety of different problems, including healing lung damage, post-ICU syndrome, post-viral fatigue type pictures and other things like anosmia. But there is no reason to suggest that this would fit with several autoantibodies. We are not expecting the lung and post-ICU and anosmia problems to be autoantibody driven. They have explanations. The rest show variation in symptoms but it is nothing like the variation between for instance rheumatoid, lupus and sceroderma, each of which has completely different pathologies that can be related to different antibodies.

I have been studying data on autoantibodies for forty years. There are sore thumb antibodies and noise antibodies. Rheumatoid patients have both and dozens of noise antibodies distracted people over the years. TheME/CFS and Long Covid antibody findings so far are noise. Expected noise. The problem is that today's immunologists are so obsessed with gathering vast quantities of data without thinking what it means that they have not learnt the lessons of the past.

 

Is this noise just natural variance?

And what could explain consistent slight, but clinically insignificant shifts in autoantibodies?

 

Is there such a consistent shift? My understanding of Scheibenbogen and others work is that nobody can demonstrate a consistent shift and in fact have demonstrated that it is anything but consistent:

Hohberger et al find elevated levels of elevated levels of ß2-AAB and M2-AAB (https://www.sciencedirect.com/science/article/pii/S2589909021000204?via=ihub), then Scheibenbogen and Bergquist find no correlation to disease severity but raised levels of M3-AAB and M4-AAB (https://pubmed.ncbi.nlm.nih.gov/34589868/), whilst a Long-Covid study (https://www.frontiersin.org/articles/10.3389/fimmu.2022.981532/full#f2) is able to find a small correlation of disease severity with levels of ß2-AAB, however these levels are now lower, not higher whilst in the next study ß2-AAB are supposed to be slightly elevated in ME/CFS patients (https://www.sciencedirect.com/science/article/pii/S0889159115300209?via=ihub), rather than lower.

I'm sure there's more results contradicting each other. Is anyone able to make this into a consistent picture?

I think the most consistent result may be that of elevated ANA levels, but that also seems rather inconsistent, but at least they never appear to be lower in ME/CFS?

 

Thanks for your comment.

No doubt about that, that's why I mentioned the talk from University of Erlangen in which they state a predisposing factor is needed for the antibodies to gain function. In healthy patients/normal people without predisposing factors, they would not cause problems.
They cite the following paper there, specifically figure 1: https://journals.lww.com/transplant...emia_and_transplantation_predispose_to.2.aspx

"These data demonstrate that AT1R-Abs can induce renal vascular contraction under predisposing conditions such as in ischemic or transplanted kidneys."

How about "necessary but not sufficient" instead of "causal"?

Could you comment on why you believe it is far-fetched to assign certain cases of POTS to functionally active autoantibodies (e.g. due to predisposing conditions such as endothelialitis)? As to me it does not sound implausible - and many therapeutics used to treat POTS (e.g. beta blockers and midodrine) act on the same receptors and in opposing fashion to the observed autoantibodies.

Do you assign the (slight and transient) efficacy of repeated immunoadsorption seen here to placebo? Are you expecting the current double-blind trial to lack significance?

Would you mind sharing your opinion on their hypothesis on Na/K-ATPase and ME?

Thanks again for your insights.

 

I might have misunderstood some previous info, or just have read inaccurate interpretations. I agree that it doesn’t seem like anything is consistent.

 

I don't know either. Certainly some people seem to be seeing or perhaps more so wanting to be seeing a consistent shift. My inability to see one means nothing, perhaps others can explain what this consistent shift is supposed to be?

 

This is an open-label, uncontrolled trial with a subjective primary outcome measure.

If you compare the results with the placebo effects observed in the blinded placebo-controlled trials on ME/CFS-patients by Fluge and Mella, you’ll find that the results of the trial you linked are well within the observed effects of both known ineffective treatments and actual placebos.

https://www.s4me.info/threads/six-y...me-cfs-2024-rekeland-fluge-mella-et-al.39478/