Oxford -recovery, the good news is a decoy to hide inconsistencies and serious faults

As we were publishing a paper on the shortcomings, omissions and manipulations of Oxford University's Recovery trial, a press release was issued on the 16^th June stating that a "discovery", the corticosteroid Dexamethasone improved the likelihood of survival in Covid-19 patients who received ventilatory assistance. This information has, of course, been the subject of numerous media reports at that time.

This in itself is good news, but it is not really a surprise. Nothing to shout home about.

A doctor told us: Corticosteroids have long been used in respiratory distress. It would therefore seem abnormal that a patient in an advanced state of distress should not receive it.

Once again, misinformation is circulating thanks to the Recovery trial.

What is Dexamethasone?

It is a corticosteroid whose properties are well-known and documented in several publications (including Vidal) and studies.

This is a very common corticosteroid, anti-inflammatory drug, that is already used for the treatment of COVID-19 in advanced stages of the disease. A less potent corticosteroid, methylprednisolone, is actually also used in the MATH treatment protocol developed by the Eastern Virginia Medical Group.

Under respiratory assistance, patients have a decrease in their breathing capacity, and it is therefore not surprising that a corticosteroid works at this level.

What is most surprising, is why is this new finding its way in the open only now?

In the press release dated June 16, 2020, the investigators write that the observations were made as of June 8, 2020. 

It would have taken them 8 days to alert the public to this pseudo good news.

Hydroxychloroquine, on the other hand, was the subject of an immediate press release concluding that it was withdrawn on 5^th June 2020. Professor Landray told us that the analyses had been done the day before (so on June 4 at the request of the MHRA).  At that time, the control group contained 3132 patients. It took investigators less than 24 hours to issue a press release and inform the withdrawal of hydroxychloroquine testing on the grounds that it had no benefit, but was not toxic. As a reminder, the observed lethality rate was 25% not different from placebo.

24 hours to remove hydroxychloroquine.

A quick look back at the calendar

• On March 23, a request was made to the MHRA by the recovery investigators, Prof. Horby and Prof. Landray. On March 25^th, that is two days later, the acceptance of the inclusion of hydroxychloroquine in the trial was given.
• A presentation document for hydroxychloroquine is issued on April 1, 2020.  This document is no longer available on the website.  It was replaced by a  version 3 dated April 18^th, 2020 although the name of this document has a different date and includes April 20

• On 26 May the regulator gave permission to use the remdesivir for Covid 19.
• May 27 Recovery includes remdesivir in the trial without changing the protocol.
• On June 3, Recovery made a change to its protocol going from 6 arms to 7 arms with several modifications.
• On 4 June, the regulator asked Recovery to check the effects of hydroxychloroquine.
• On 5 June Recovery, in a laconic press release, stated that the non-toxic hydroxychloroquine had no effect on Covid 19 and removed it from the test with immediate effect.
• June 5 FranceSoir interview Martin Landray.
• On June 6, Martin Landray's interview  was published and FranceSoir published a first paper questioning Recovery on the dosage used and other elements.
• June 8 Recovery conducts the analysis on dexamethasone. The information is not released until the 16^th via the press release. No one knew about it.
• On June 15, FranceSoir published a paper questioning the lies and manipulations of Recovery.
• On June 16 Recovery announces with great noise that dexamethasone reduces  mortality.  The day after the harsh accusations, revealed in our paper. 

We will go back on the diversion made by a Liberation (a french website) Liberation, where the investigators of Recovery indicated that FranceSoir had reported a false or misunderstood.  The same person of Liberation had announced on twitter, about the flawed Lancet study (a study that proved that hydroxychloroquine was not working), before having to carry out a safety effort to cover up the misinformation. It is a pity for a journalist, who claims to be a scientist, to fail on two occasions : failing to uncover the enormous fraud of the Lancet and in addition, missing the "alarming" elements of the Recovery study.

Did we disturb so much that Professors Horby and Prof. Landray did not ask for a right of response to our publication!

They obviously used the "golden opportunity" offered by Libération to manipulate the public opinion. Before, reporting that what the words that we reported in our interview of Pr Landray were in fact correct. In the end this is not critical and should not divert our attention.

We had doubts about Recovery, we wrote  about it.  Some of the elements would jump to the face of even the less attentive reader. Why wait 8 days to announce a good news whist for information on hydroxychloroquine investigators rushed to social networks?

We start to find some answers, first of all, by taking a closer look at the protocol. This protocol is in its version 6  dated May 14th, 2020. This excessive number of versions is a sign of a blatant lack of scientific mastery of a protocol that is modified, depending on the political needs of the moment. An internet tool makes it possible to make a comparison with the previous version.

In red are the elements that were removed and in green the added elements. Additionally, the trial adds a 7th arm.  The addition is for a "convalescent plasma" treatment and there is a modification of a "in parallel" test to a "factorial" test with different randomizations.

The 7th arm is the plasma arm that was not included in the previous version.  This means that patients will be able to receive plasma from patients who are cured.

In addition, we see that the need for renal replacement (dialysis) is shifted from secondary to other measures. It also includes a composite measure of mortality or the need for ventilation. We will return to the importance of this point in an upcoming paper.

It is normal for an evolutionary trial to see its protocol evolve over time.

But here, these are evolutions that can be akin to amateurism and there is no desire for transparency, which would be the signature of good faith through press releases on these elements.

By taking a closer look at the available data, several points caught our attention.

We are in the midst of an epidemic and hospitalizations are regressing, but they are much higher in Britain than in France at the same stage of the epidemic.

On June 5th, the day after the data analysis for HCQ, the base of the control cell "SOC: standard of care" had 3132  patients.  On June 8, the SOC analysis had  4321;  a difference of 1,189 patients.  On the same dates, the number of patients was 11,000 patients on June 5 and 11,500 patients on June 8, as reported by Recovery in the respective press releases.

Between June 5th, 515 hospital admissions, 6th June 438, 7th June 435 and 8th June 458 or 1846 new patients in hospitals in the United Kingdom.  The previous inclusion rate in the trial estimated by Recovery was 13%; this means that they would have included 1846-13% - 239 patients. Now, the press release says that the delta of inclusion and about 500.  Recovery therefore had to push for the inclusion of patients in the trial. Here the most questionable point is how can one have an increase the inclusion of 500 patients (difference between 11,500 on June 8 and 11,000 on June 5) and at the same time an increase in the placebo group of 1189  (4321 on June 8 to 3132 on June 5).

This doesn't add up, it does not make sense. So there's bound to be an error or manipulation somewhere. This means that the figures provided are questionable on simple things. And here we're not talking about people who sadly passed away.

Moving back to the presentation sheet to the participating doctors on hydroxychloroquine

The current version on the site is version 3 of April 20, 2020 or based on the date indicated by 'properties of the document of April 18, 2020. Another unimportant inconsistency but that adds to the picture. What matters is that this document is the justification for administering the hydroxychloroquine dosage, the first 2400mg 24h, followed by 800mg the following days for 9 days. This paper is 3 pages long and explains that they were based on pharmacokinetic analysis and a 1995 WHO paper on the use of chloroquine to define dosage.  However, the pharmacokinetic analysis is not available, the one mentioned by Professor Landray in his interview  and that forms the basis of the justification of the very high dosage of hydroxychloroquine.

Impossible to find version 2 on the Recovery website. Our requests to Recovery have been in vain. However, and not easily, we were managed to obtain version 2 of this document as well as the WHO report (70 pages), which we presented to various microbiologist and physician experts.  Without going into the details of the analysis the first elements that emerge are:

• This document is 25 pages long and includes the famous pharmacokinetic analysis that was  removed from version 3,  which for the record has been reduced to 3 pages. Why?
• The pharmacokinetic analysis of these documents extrapolates several things: chloroquine and hydroxychloroquine, extrapolation of the load dose supposedly based on WHO documents.
• The question arises as to whether these experts have not confused the units or the metrics.

Our experts reviewing the information are even more surprised by the quantities of hydroxychloroquine evoked and prescribed. Hydroxychloroquine has a fairly long half-life on already infected patients and the lethal dose is considered to be between 3000mg and 5000mg with the overdose level is at 2000mg (as defined in Vidal) with recommendation of hospitalization in the emergency room.  The WHO document reports potentially lethal doses starting at 50 mg/Kg. Patients received 2400mg on the first day, 800mg on the second 800mg on the 3rd and so on until the 9th^ème  day. The pharmacokinetic data presented in the HCQ dosing document in Recovery are the result of theoretical extrapolations and simulations that have no place in such a document where only experimental evidence must prevail.

Our clinical trial expert tells us:

This paper denotes a certain amateurism, a total ignorance of medical ethics, the trampling of good practice of clinical trials and ultimately the unconsciousness of the fact that a load dose of 2.4 g is potentially lethal depending on the age, body mass and degree of fragility of the patient.

This serious fault could explain the lack of benefit of the HCQ which would have killed as many patients as it saved.

Should the principal investigators of the trial had wanted to harm the HCQ molecule they could not have done better.

Data from the product monograph for hydroxychloroquine from Mylan Pharmaceuticals

 

The chief investigators of Recovery, Professors Horby and Landray will surely have to answer to the courts for what will remain in history as the #CovidPapers and the #RecoveryFraud.  It is at least a medical error, incompetence or potentially a crime.