Let’s talk of this strain of COVID-19 “more contagious” (and “pre-prints”)…

Faut qu'on se parle de cette souche de COVID-19 «plus contagieuse» (et des «preprints»)...

Faut qu'on se parle de cette souche de COVID-19 «plus contagieuse» (et des «preprints»)...

Share

May 6, 2020 14h13

Share

Let’s talk of this strain of COVID-19 “more contagious” (and “pre-prints”)…

Faut qu'on se parle de cette souche de COVID-19 «plus contagieuse» (et des «preprints»)...

Faut qu'on se parle de cette souche de COVID-19 «plus contagieuse» (et des «preprints»)...

Jean-François Cliche

The Sun

BLOG / When the dust will have begun to fall on this crisis of the COVID-19, and even before if you can, it will be necessary to ask serious questions about the articles in the “pre-publication”, both in the media and in the research world. Because I’m not sure that they have done more good than harm in the current crisis.

The latest example is this story, repeated in many media of the planet, around a new strain of “now dominant” of the COVID-19 which would be “more contagious” than the original . As you might expect, this “new” has caused a lot of excitement and anxiety on social networks. And all this for frankly not much else.

The L. A. Times was the first to report this “news”, which goes back to an article placed recently on the server prepublication bioRxiv. The document deals with a mutation observed in the virus of the COVID-19 and that seems to be widespread thereafter. The authors of the study, who are serious people also interpret this as the sign that this mutation gives an advantage to the virus mutants by making them more contagious. And the tone of the prépapier actually has something of an alarmist, because he speaks of an “emergency” (urgent concern, in the summary) about the mutation in question, the poetically named “D614G”.

However, here, as in all the other studies that find themselves on bioRxiv, it is a “pre-publication”, that is to say, a sort of “rough draft” that has not yet gone through the process of peer review which leads (eventually) to a publication in scientific journals. We can’t rely on it as much as on a study review published in good and due form. Already that when a study is published, it remains “just a study” whose results need to be corroborated by other work, so imagine the caution with which we must deal with the preprints.

But neither the Timesnor the JdeM, nor most of the others do not have this nuance. Many have taken these results to cash, have taken headlines of apocalyptic and often did not even bother to seek the input of other experts.

It would have been necessary. Personally, I’ve stopped counting the number of virologists and epidemiologists that I’ve seen fart checkmarks on my network at sujetde this “new” in the last few days. Essentially, they say, it is possible that this mutation D614G to have an influence on the infectiousness of the COVID-19. It could be… maybe… But the article to bioRxiv makes no demonstration.

First, note the epidemiologist from Harvard, Bill Hanage in an illuminating series of tweets, one of the main arguments of this preprint to defend the thesis of natural selection is that the mutant variant has eventually become more widespread than the original that broke at the start in China (and a few other places in the world). But in and of itself, says Hanage, it doesn’t mean anything. It can be a sign of a greater transmissibility of the mutant, but this can also be explained simply by what geneticists call the “founder effect” : the first virus to have infected Europe belonged to the mutant strain, and it is shown that a time that China and Asia are managed to control their outbreaks, it is Europe that has become the main source of contagion. It would therefore be dealing with a strain of virus which are not more contagious than the other, but who have just been “lucky” to find himself in the right place, at the right time.

It is not yet certain of anything, but a sign that to believe Mr. Hanage that there was no doubt the case that a strain “lucky” is a chart of the prépapier (figure 3 on page 19 (33), which shows the way in which the epidemic took place in the State of Washington. This place was one of the first (and most severely) affected in North America because of its links with Asia. In the State of Washington, the epidemic started with the original strain (in orange on the graph), then the mutant strain (in blue) is gradually arriving from the east coast of the u.s., where the coronavirus was mainly happened in Europe.

Faut qu'on se parle de cette souche de COVID-19 «plus contagieuse» (et des «preprints»)...

Korber et al, 2020

But as we can see, the two strains were then declined at a pace quite comparable, which suggests that they do not transmit better than the other. Add to that the fact that pre has found exactly the same pattern in Canada, where the COVID-19’s first arrival in British Columbia (primarily in Asia), before Quebec does not know the strong epidemic that we know (from mainly Europe and the eastern United States). And it does not seem that the mutant strain has decreased less quickly in Canada either.

Faut qu'on se parle de cette souche de COVID-19 «plus contagieuse» (et des «preprints»)...

Korber et al, 2020

In addition, as pointed out by the virologist american Angela Rasmussen, the mere fact that a mutation affects a protein that the virus uses to latch on to the cells does not mean that this changes anything about its contagiousness. It is possible that this is the case, but it is necessary to make tests in vitro and in vivo before affirming, something that the authors of the prépapier who set fire to the powder did not.

In the past, we have already found mutations similar on the Ebola virus, for example, and one of these mutations, named A82V, was tested in vitro in animal cells, showing a potential infectious four times stronger than other strains of Ebola. But when we had tested the same mutation on real animals, the mutation was found to be insignificant, and this may finally be no difference, remember accurately, Ms. Rasmussen.

However, the prépapier on the COVID-19 published last week did no test : neither in vitro, nor in animals. So for the moment, nothing leads us to believe that the famous mutation D614G has any effect on the infectiousness. Maybe that is the case, maybe it is not.

And that’s not counting a separate study, it has been peer-reviewed and published, concluded last week that there is no sign at present that the COVID-19) is evolving in several sub-types with clinical characteristics or epidemiological different. Which, a priori, completely contradicts the idea of a new sub-type to be more contagious.

These are all pieces of information that were easy to find, as long as you do some research and/or interviews to validate the information. And as long as we take the preprints for what they are : preliminary results, drafts to be taken with beautiful large tweezers. Never for a thing that can be put in the newspaper without validation or warning on its value.

* * * * *

The worst is without a doubt that this was not the first time, in the course of this pandemic, a preprint was used as a starting point to a false history of fear. This winter, indian researchers had pre-published a study concluding that pieces of HIV have been found in the genome of the COVID-19. It turned out that it was completely false and the prépapier has been withdrawn by the authors themselves, but the damage was done : the idea without the foundation of a virus made in the laboratory just win an appearance of scientific validity that would not delete easily. I have to say, I still get fairly regular e-mails from readers who are questioning this study. The damages to public information are real and long-term.

Then I ask the question : what is it with these servers pre-publication ? There is no question of the close, there is a consensus on that, because these tools are necessary and very useful in research, which allow researchers to discuss their work before peer review. But they can be used as a tool of public dissemination, and this is where the problems begin.

The media, as we say, are not the only ones to commit this sin : I’ve seen many scientists cite preprints in recent weeks. I have myself done (putting always the warnings). But this is not true that the bioRxiv and the other servers of the same type are used to speed up the search, contrary to what the JdeM claims. The scholarly journals have ways to accelerate the peer review and publication when the circumstances require it — as evidenced by the numerous studies on the COVID-19 of which are published for weeks. At the risk of repeating myself : the servers to pre-serve as forum of discussion for researchers, not instruments of dissemination.

So what is it that we can do so that they continue to fulfil this role, while limiting the collateral damage ? You protect it with a password to restrict access to scholars (these web sites are all currently in free access) ? Personally, I don’t see a better solution, but it may be just a sign of my lack of imagination.

In any case, I believe that it is necessary to find a solution because if it is useful for research, public information, this system has caused serious damage already…

* * * * *

To participate in a discussion about this post, please visit my page Facebook professional.

Le Soleil

Share Button

Add a Comment

Your email address will not be published. Required fields are marked *