Preprint Safety, tolerability and clinical effects of BC007 on fatigue and quality of life in patients with post-COVID syndrome (reCOVer)..,2024,Bettina et al

Safety, tolerability and clinical effects of BC007 on fatigue and quality of life in patients with post-COVID syndrome (reCOVer): a prospective, exploratory, randomised , placebo-controlled, double-blind, crossover phase IIa clinical trial

Bettina Hohberger, Marion Ganslmayer, Thomas Harrer, Friedrich Kruse, Stefanie Maas, Tobias Borst, Ralph Heimke-Brinck, Andreas Stog, Thomas Knauer, Eva Ruehl, Victoria Zeisberg, Adam Skornia, Alexander Bartsch, Armin Stroebel, Monika Wytopil, Carolin Merkel, Sophia Hofmann, Katja G. Schmidt, Petra Lakatos, Julia Schottenhamml, Martin Herrmann, Christian Mardin, Juergen Rech

Abstract

Background: As recent data suggest an involvement of GPCR-fAAb in PCS pathogenesis, neutralisation of such GPCR-fAAbs by BC007 could improve PCS symptoms. The aim of the reCOVer trial was to investigate safety, tolerability and clinical effects of BC007 on fatigue, its severity and quality of life in PCS patients.

Methods: reCOVer is a prospective, exploratory, randomized, placebo-controlled, double-blind, crossover phase IIa clinical trial with 1350 mg BC007 at the University of Erlangen-Nuernberg, Germany. Eligible participants were 18-80 years with GPCR-fAAb, whose PCS symptoms persisted ≥3 months after PCR-confirmed COVID-19, with fatigue as the major symptom (Bell score ≤60) and at least three of eight defined PCS symptoms. Participants were randomly assigned (1:1) according to a crossover design to either receive BC007 (sequence A) or placebo (sequence B) at day 0 and day 48 with a follow-up of 28 days, respectively. A crossover design was chosen to increase patient adherence. Occurrence of treatment-emergent adverse events (TEAEs) in comparison between sequence A and B from d0 to d28 and d0 to d70 were the primary and co-primary endpoint, respectively.

Findings: Between 31.10.2023 and 12.06.2024, 30 PCS patients were randomised and analysed. The trial has been concluded. Summarizing all AE rates, no statistically significant differences between sequence A und sequence B were observed within day 28 and day 70. One report of a serious adverse event, not related to treatment, was recorded. As a secondary endpoint, BC007 showed a significant improvement on self-reported fatigue and its severity, as well as quality of life.

Interpretation: As BC007 was well tolerated and showed a significant improvement of fatigue and quality of life, it might offer a therapeutic option for an autoimmune subgroup of PCS patients.

Trial registration: EudraCT, number 2022-001781-35.

Funding: German Federal Ministry of Education and Research, German Research Foundation.

Link | PDF

 

Judging by the graphs, there's no real difference between groups A and B. If the drug worked, you would see group A improve, then group B, but I don't notice anything except that group B seems a little healthier.

 

The primary outcome of treatment-emergent adverse events (TEAEs) was as follows:

So although the difference was not statistically significant (due to low sample size and rate of events), there were more than twice as many adverse events in the intervention than in the control group.

If I understand correctly, from V2 to V7, sequence A was getting the treatment and sequence B the placebo, while from V8 to V13 it was the opposite. Looking at FACIT and the Bell scale from V2 to V7, it seems that the control group started off with significantly higher values but that the intervention group caught up with them, indicating they improved more. Because the authors corrected baseline variables in their regression, I think that explains why apparently small differences were significant. Would be easier to see if they shared the raw data.

Supplementary table 12 seems like an error because the mean and sd from sequence A and B differ enormously.

 

The text is rather confusing. They write: 'Sequence A received 1350 mg BC007 followed by placebo, sequence B received placebo, followed by 1350 mg BC007.'

But then they start comparing the sequences ("no statistically significant differences between sequence A und sequence B were observed") which does not make al lot of sense: the main outcome should be to compare treatment versus placebo.

And In the flow diagram they speak of intervention A and intervention B and sequence AB, sequence BA. So what does 'A' refer to? The sequence or the treatment? They seem to mix up the two throughout the paper (or at least wrote it down in a confusing matter).

 

Assuming that the text is correct and sequence A got placebo at visit 8, then it is strange that it outperformed sequence B (which received treatment) after the crossover on the Bell scale.

Here's a visualisation of that. After crossover the placebo group seem to have performed better?


This does not correspond to the results in table 2 where they report a significant effect of BC007 of 3.43 points.


So perhaps 'sequence A' in the supplementary material does not refer to the sequence but to treatment, namely those participants that were receiving BC007?

 

I did similar analysis to ME/CFS Skeptic because the charts in the article are very difficult to read. In short, I think the treatment had a very noticeable effect in Cohort A and basically no effect in Cohort B, averaging out to a statistically significant treatment effect.

Here is FACIT-FS over time:



You can visibly see that for cohort A, the score improves immediately after infusion of the treatment (day 0) and stays improved. The starting score is 18.07 and it gradually and consistently rises to a peak of 26.23. It does dip a bit after the second (placebo) infusion but only to 24.

For cohort B, the score improves (but less than cohort A) after the first (placebo) infusion. But there is a much bigger jump after the treatment infusion (roughly double the size of the placebo effect). Unfortunately, the improvement in this cohort fades a little bit.

Cohort A improved by 6.68 points and Cohort B improved by 4.44 points 49 days after the treatment infusion. The FACIT is a 13 question survey where you answer 0 (very tired) or 4 (not tired at all) to each question. So this is equivalent to Cohort A answering each question 0.5 marks higher.

If we compare cohort A to a battery, they started with a 34.75% charge and ended up with a 55% charge after 91 days. It's a clinically significant improvement, but hardly a cure.

Here's a chart showing the relative improvement of Cohort A:



That is so clear. There's a huge treatment effect from the first infusion and basically no placebo effect from the second infusion. You can also see the improvement slows over the first 40 days and stalls/goes backwards over the following 50 days. That's consistent with the case reports from the first accidental healing at Erlangen.

(That standalone red dot is the final measurement taken on day 91, 49 days after the second infusion. It should be connected by a line to the red dot on day 28, but I couldn't figure out how to make Excel work).

But here's what the relative improvement of Cohort B looks like:



Eyeballing that, it's difficult to say there's any treatment effect. To steelman the argument, you could say that the treatment effect is stronger over the first 15 days (excluding day 2) but the gains stop. By contrast, the natural healing/placebo effect from the first infusion is reasonably consistent. The result is that 28 days after each infusion, there is less improvement from the treatment infusion than the placebo infusion in Cohort B.

Why was the treatment effect so noticeable in Cohort A but not B? Perhaps because they got the treatment first, they knew the second infusion was a placebo and there was no placebo effect to deflate the result. Perhaps BC-007 only works in some sub-groups of ME/CFS and by co-incidence they were in Cohort A but not B. Or, of course, perhaps it is because BC-007 doesn't work.

This trial reminds me of an IVIG trial that Andrew Lloyd did. The trial was a marvellous success. Long term patients went from wheelchairs to walking, not working to working. Unfortunately, this happened in both the treatment and placebo arms. I think the shock of a placebo miraculously curing so many severe patients turned Lloyd into a BPS proponent to the ever lasting harm of Aussie ME/CFS patients.

Note that I looked at the FACIT scores whereas ME/CFS Sceptic looked at the Bell scores. But I think the story is the same. Here's the charts for comparison:

 

Were sequence A and B randomly selected? If so is the difference in response low enough to be due to noise?

 

Thanks for providing this overview.

Not sure if you took this into account but I suspect that sequence A in the supplementary material means 'Treatment' at any timepoint rather than the group that first got treatment and then placebo. Otherwise it would not match with the results they report in the table.

 

I'm not sure I understand how this could be. The paper itself quite clearly describes sequence A as those who received the treatment first, placebo second. And the supplementary materials describes it as "sequence A". If they changed the definition without explaining it, it would be the worst kind of error/misconduct.

I also cannot figure out how to explain Table 2. I suspect they compared the performance of A (treatment) and B (placebo) during the first 28 days. By my calculations (simple arithmetic comparing the improvement in mean FACIT scores over the first 28 days), this has a treatment effect of 2.78 (similar to their figure of 2.91). Then for the cross-over they compared the performance of cohort B (treatment) v A (placebo). By my calculations this is a treatment effect of 1.16 (a positive treatment effect but not as strong as the 2.73 effect reported by the trial).

Maybe they are using a regression to calculate the treatment effect and that may explain the discrepancy between my numbers and theirs.

I had assumed they were randomised, but reading the paper, I cannot see an explicit confirmation of this. The paper says "The allocation was generated by the Center for Clinical Studies CCS (AStr), participant enrolment was done by the investigator (AS, TK, BH, VZ, ASk). Assignment to sequence was done by the pharmacy"