In early June, as controversy swirled over a high-profile study of hydroxychloroquine and COVID-19, Christian Funck-Brentano started receiving aggressive emails, texts and tweets.
Funck-Brentano, professor of medicine and clinical pharmacology at Sorbonne Université and Pitié-Salpêtrière University Hospital in Paris, and a colleague had published an editorial in The Lancet alongside the study — purportedly based on data from a company called Surgisphere — citing the results and sounding an alarm about the cardiac risks of hydroxychloroquine. Within days, that paper was retracted, along with a paper in The New England Journal of Medicine about different medications also allegedly based on Surgisphere data.
“I’m 65 and I have no real concerns,” Funck-Brentano told Retraction Watch, but his co-author was in an earlier stage of his career, and Funck-Brentano wanted to handle the situation in a way that would cause the least collateral damage.
Funck-Brentano remained convinced that the risks of hydroxychloroquine in those with well-studied reasons to take it — lupus and malaria, for example — are quite low, but that they are much higher in patients with COVID-19.
What to do?
The core of the editorial is still valid, Funck-Brentano told Retraction Watch, because hydroxychloroquine has serious side effects in some populations:
The content of the comments remains OK except for praising [Mandeep] Mehra’s study as very important and valuable.
When Funck-Brentano saw the pre-publication version of the Lancet paper, he said:
I was as surprised as anyone. When I read the paper, I was impressed that the database conformed to [International Organization For Standards] and FDA rules. Of course I trusted that.
So did The Lancet. Then questions began swirling around the findings. Funck-Brentano said:
We made a comment on an accepted article that mentioned that the database was extremely clean. In another case, I recommended rejection of an article based on a telephone survey of doctors. How can you know?
Until The Lancet paper came out, no one criticized the NEJM paper. But hydroxychloroquine is like a religion for some people.
Funck-Brentano told us that he wrote to Sepai, asking him to accept an audit, because such audits can be kept confidential.
He never wrote back to me.
He also wrote to Mehra.
He wrote back right away and said he was surprised it couldn’t be audited.
As questions swirled around the paper, Stuart Spencer, an editor at The Lancet, told Funck-Brentano to wait for the audit — which turned out to be impossible because Surgisphere would not cooperate. The journal’s ethical review committee then contacted Funck-Brentano, letting him know that the paper would be retracted and that the commentary would have to be, too.
He contacted Hervé Maisonneuve, a researcher in Paris who has been active in scientific integrity, for advice.
Retracting a comment is not the same as retracting an original paper. I wondered, how can we make this retraction a positive retraction?
Would it be possible to have a retract and republish comment almost at the same time?
The journal — which, along with the JAMA journals has been using the “retract and replace” approach — agreed to that. Funck-Brentano and his colleague, along with a new co-author, wrote a new version, and it was accepted a few weeks ago.
The retraction and republication appears today. It begins:
A recent Lancet Article by Mandeep Mehra and colleagues, which reported adverse events associated with hydroxychloroquine or chloroquine treatment in patients with COVID-19, has been retracted by three of the authors,1 along with our linked Comment that provided a commentary on the Article and its findings,2 because the veracity of the data underlying this observational study could not be assured by the study authors. Subsequently, an article that used data from the same Surgisphere database, and which was authored by some of the same individuals, was retracted from the New England Journal of Medicine. Nevertheless, the debate about hydroxychloroquine for COVID-19 continues. Here, we provide a discussion of what is currently known about its proven and potential harms.
Funck-Brentano said:
This is my first retraction story in my whole career. I am extremely concerned about making good science. I look back and see some findings that no longer hold up.
Funck-Brentano recalled a 1988 paper he wrote on encainide, calling it a promising drug. A year later, a study showed that the drug killed patients more than twice as often as placebo. That study, he said,
was a beautiful study. It made me drastically change my view on treatments and medicine. I remember many were saying it was not ethical when it was planned. Big mistake in reasoning. I use still this example with my students.
Unlike The Lancet study, however, his 1988 paper was based on solid data at the time.
Like Retraction Watch? You can make a tax-deductible contribution to support our work, follow us on Twitter, like us on Facebook, add us to your RSS reader, or subscribe to our daily digest. If you find a retraction that’s not in our database, you can let us know here. For comments or feedback, email us at team@retractionwatch.com.
“But hydroxychloroquine is like a religion for some people.”
How true, no matter which side you take. So where do these religious types sit on the zinc aspect – other than mostly agnostic it would seem?
I think more atheist than agnostic on zinc, possibly because there is a lot of mumbo jumbo on zinc as a miracle mineral. Try searching YouTube for zinc and you will see. Most doctors I have talked to seem to think supplements of any kind are basically a hoax.
Still, one would think that researchers testing hydroxychloroquine – where its primary mode of action against the virus is thought to be as a zinc ionophore – would be paying attention to zinc levels. Also that doctor in NY who looks a bit crazy but sounds very sane and logical has very eloquently made the point he is using HCQ because of its ability to get zinc into cells. It’s a mystery, but when one believes in a dogma, it is hard to change,
“hydroxychloroquine – where its primary mode of action against the virus is thought to be as a zinc ionophore”
Thought by whom, though? Clearly not by its main proponent Raoult, who never mentions zinc. Not by the HCQ advocates who claim that it blocks the virus from bonding to hemoglobin.
Passive voice is doing a lot of work there.
Dear Smut – No idea where you get your information, but all the analyses I have seen emphasise the zinc ionophore aspect and, as I noted, Zelenko has been specific on this point. Raoult is only one among many HCQ proponents – but true I haven’t seen him say anything on zinc – and this is interesting and quite possibly relevant to my atheist suggestion.
What is most interesting to me in all the brouhaha is the dwelling on HCQ as if it were a magic bullet, which it clearly is not, and nothing on its mode of action. If it works, then why? Isn’t that the question anyone would immediately ask?
Zinc is NOT the primary mode of action.
If you understood HCQ at all, you would know it is acts as a weak base (base vs acid).
It is this feature that helps to elminate the virus.
“But hydroxychloroquine is like a religion for some people.”
Anti-hydroxychloroquine is also a religion for some, apparently because Orange Man Bad. The obvious elephant in the room, not getting mentioned, is that these papers with bad data were getting pushed through and then touted for political reasons.
I could be wrong, but I get the feeling that if Orange Man had never mentioned hydroxychloroquine, we would all be taking it now as prophylaxis while getting back to the business of living and reopening our economies.
I am left confused by this account. Am I correct that the chloroquine papers and comments on them are retracted? Also that the majority opinion is that those drugs have no supportable benefit against Covid19 and also are hazardous to sufferers? That they remain of benefit for their previously approved uses, though with a lower (?) level of risk than for Covid19 patients?
Perhaps a takeaway is that any publication’s acceptance of a study ought to explicitly include permission for auditing
any aspect of the study on demand, if this is not already the practice.
It’s not that hcq is holy but that it was instantly, mendaciously and vociferously declared unholy. Time has told.
Precisely.
When the goal is to flatten the curve, you use every tool available.
That is pragmatic, not dogmatic.
https://indianexpress.com/article/india/vadodara-administration-drive-hcq-helping-in-containing-covid-19-cases-say-docs-as-analysis-begins-6486049/
How to explain the discrepancy?
Sgt Grumble, correct. And so obvious for whoever is watching and listening. Not blind and deaf or prejudging.