Some comments on this paper, and general elements to discuss :
1) Predatory Journal
The publication journal (Fortune Journals) is widely identified as predatory, lacking transparent peer review and accepting articles within days. This undermines the scientific integrity of the process.
2) Fake affiliation
Affiliations are not clearly academic: Authors represent BonSens.org, ReinfoCovid, or self-declared research institutes, without institutional oversight or verification.
Xavier Azalbert’s affiliation was discussed in a press article and proved to be fake : he studied economics in Toulouse in 1988 but the Toulouse school of economics was founded in 2011. Christian Perronne’s affiliation is also fake. https://www.lepoint.fr/sante/exclusif-le-juteux-business-bresilien-de-complotistes-antivax-francais-27-10-2024-2573797_40.php
3) Introduction
The whole introduction does not contain any reference and states the author’s opinion without any factual basis to rely on. Some elements are misleading, like the word “knowledge shared by the Chinese” which actually refers to a communication by Gao et al. without any data, mentioning chloroquine as a promising treatment against covid19 in ongoing clinical trials. Another conceptual mistake by the authors is about the role of a placebo group in a clinical trial. The authors present this placebo group as a way to test if a drug is well tolerated and to assess its toxicity. Actually, a placebo group is mainly useful to identify drug efficiency compared to not taking the drug, to gather evidence on the efficiency of a treatment. Depicting a placebo group as a way to assess a drug’s potential toxicity is a methodological incomprehension.
The authors also mention the retracted Gautret et al., which is an established scientific fraud (especially because control and treatment groups were not tested with the same sensibility) as established by pr. Valérie Giordanengo’s comment on the retraction notice (https://www.sciencedirect.com/science/article/pii/S0924857924003327). Sufficient concerns had been raised in 2023 already, long before Lacout et al. was published (https://linkinghub.elsevier.com/retrieve/pii/S0040-5957(23)00081-1). The authors describe it as sufficient evidence to treat patients without any further clinical research.
4) First studies on hydroxychloroquine
The authors present first studies on hydroxychloroquine from China : the aforementioned Gao et al. which had no data, and another one which was not even published at that time, from Chen Z et al. This second study was a double blinded clinical trial, the main author could not have had access to the data during the trial, but he actually communicated throughout the study in the chinese press (https://web.archive.org/web/20200612175709/http://www.bjnews.com.cn/feature/2020/02/18/691194.html). The study started on February 2nd and lasted until February 28th with 300 patients planned initially. On February 17th, the investigators stated that 19 out of 20 patients had significant improvements. On February 28th, the paper was finally published with 19 out of 31 patients having had significant improvements. This means that after February 17th, out of 11 patients enrolled in the hydroxychloroquine group, not a single one experienced any form of improvement. This publication can’t be relied on to advance that hydroxychloroquine would be efficient : the initial protocol was not respected, the double-blind was not respected and the data communicated is inconsistent.
They also qualify as “stupid” the fact that the paper has been retracted, which is an insult and is not scientific language.
Finally, molecular mechanisms are discussed. Although some mechanisms are real, they are interpreted in a single way. For instance, the authors emphasize that hydroxychloroquine’s main action is to reduce endocytosis : this is useful to suppress viral entry in a cell, but has also other consequences on defense mechanisms, like reducing the ability to present antigens by Antigen Presenting Cells. The immunomodulation effect is only described as effective against a “cytokine storm” but never addressed as a mechanism that could lower immune defenses against a virus during an infection.
5) Allegation of biased studies on hydroxychloroquine
The authors claim that the Recovery study overdosed hydroxychloroquine, thus leading to a higher death rate in the hydroxychloroquine group. They did not mention the rationale published in the Recovery protocol, which was to use a well tolerated loading dose equivalent to the higher end of the concentrations encountered in steady state patients treated for rheumatoid arthritis (https://www.recoverytrial.net/files/protocol-archive/recovery-protocol-v2-0-2020-03-23.pdf). There is also a misleading concept that is discussed by the authors. The initial dose is 4 tablets, with 4 tablets 6 hours later and then 2 tablets every 12 hours. The authors counted day 2 tablets in the initial dose and compared it to a single uptake of 12 tablets. This is clearly not the same.
Besides, the authors criticize the Recovery protocol (https://www.nejm.org/doi/10.1056/NEJMoa2022926?url_ver=Z39.88-2003) by stating that it is a late hydroxychloroquine uptake compared to Raoult’s protocol (https://www.imrpress.com/journal/RCM/22/3/10.31083/j.rcm2203116). When comparing both data, we observe that in Raoult’s case, hospitalization starts at day 4 after first symptoms on average, while it starts at day 6 in the recovery protocol. Raoult’s treatment does not start during the first days like the authors state in their paper. Since the half-life of hydroxychloroquine is very long, it can be considered that there is not much decrease in concentration in the human body, and a cumulative dose during a few days of treatment can be used to address the high dose allegation by the authors. In fact, when accumulating the daily doses since beginning of the treatment, Raoult’s administered dose exceeds that of Recovery until day 15 after first symptoms. See following table for more details (values in mg) :
6) Cardiac toxicity
The authors remind us that hydroxychloroquine’s cardiac toxicity is a very rare side-effect, and mention various studies showing 9 deaths out of 100 000 patients when combining azithromycin and hydroxychloroquine. The authors consider that the balance is positive based on flawed data : with no positive effect of hydroxychloroquine administration, this treatment caused at least 9 unnecessary deaths due to cardiac toxicity out of 100 000 treated patients. But the authors omit to mention that hydroxychloroquine could raise the deaths caused by covid19 infection through other mechanisms, like those affecting the immune system. They also call the recovery study a “sabotage”, with no evidence to back their conspiratorial claim.
7) On retracted papers
Mehra et al. is presented as a “Lancetgate” while the retraction process was actually transparent and fast compared to what one could call the “Raoultgate” with the retraction of Gautret et al. But the authors then describe the retraction of another study, Pradelle et al., omitting to mention the harassment campaign they orchestrated against the last author, professor Lega (https://sfpt-fr.org/vie-de-la-sfpt/communiques-sfpt/2055-communiqu%C3%A9-de-presse-%C3%A0-propos-du-retrait-d-une-publication-de-la-revue-biomedicine-pharmacotherapy-02-2024). Associating him and his university to nazis (https://www.francesoir.fr/opinions-tribunes/covid-l-affaire-lega-est-elle-la-reminiscence-de-l-eugenisme-et-de-l-euthanasie-0), Xavier Azalbert triggered an intense online harassment campaign with the help of some of his co-authors (https://www.lepoint.fr/sante/comment-medecins-et-scientifiques-sont-cibles-par-un-groupuscule-pro-raoult-25-02-2024-2553390_40.php). While the author state that Lega did not answer and make raw data available, he actually did answer and his letters have been made public (https://retractionwatch.com/2024/11/01/how-an-article-estimating-deaths-from-hydroxychloroquine-use-came-to-be-retracted/).
Finally, the authors criticize a published meta-analysis, Fiolet et al., for having extracted data before registering to the prospero platform. This was also part of a harassment campaign still visible online, to which the authors had responded (https://www.clinicalmicrobiologyandinfection.com/article/S1198-743X(20)30613-3/fulltext#appsec1).
In a last development, they quote retrospective, inconclusive studies to claim that hydroxychloroquine would be efficient. Of note, Davido et al. was first posted on a document repository online, with a very flawed methodology worse than immortal time bias since patients who died before day 3 of treatment and assigned to the hydroxychloroquine group were simply swapped into the “no-hydroxychloroquine” group. Interestingly, the absolute number of patients described in the initial protocol did not add up correctly (24+1+1 = 28 according to the authors data), showing strong shortcomings in the follow-up of this study.
8) Online harassment supported by Science Guardians
One of the authors, Xavier Azalbert, is publicly known not only for his advocacy around controversial COVID-19 treatments, but also for his targeted harassment campaign against fact-checker Antoine Daoust, founder of Fact & Furious. His methods have been extensively documented : Xavier Azalbert is an entrepreneur who bought France Soir media, but lost its “CPPAP” classification, granting tax advantages for donations in France. Since Fact & Furious had this agreement, he blackmailed Antoine Daout to force him into selling him his media. The recording of this blackmailing has been made public (https://lofficine.eu/temoignage/la-methode-azalbert-partie-1/). His harassing campaign included over 150 tweets, publication of private information, and prolonged defamation efforts. According to multiple sources, this led Daoust to a suicide attempt in December 2023, and ultimately to a court ruling against Azalbert for defamation and moral prejudice (https://www.franc-tireur.fr/le-calvaire-dun-fact-checker and https://www.conspiracywatch.info/affaire-fact-furious-xavier-azalbert-face-a-la-justice.html).
In 2021 already, Xavier Azalbert started his harassment campaign against Lonni Besançon among other targets. See this post : https://archive.is/VD0Vp. He also spread rumors about disciplinary actions against Lonni Besançon, with no factual basis.
Among the harassers, Olivierfente was one of the most active online accounts. He was very active in looking up every instance of Lonni using the word “mère” (mother in french) and removing any form of context (hiding the sea lioning and harassment campaigns initiated against him online, to which any human being would have lost temper at times). See November 2024 post https://archive.is/ciPDp.
Science Guardians is dedicated to amplifying “voices of authors who carry the heaviest burden in research yet are most vulnerable”, standing “against harassment and intimidation, ensuring sage and supportive environment for all”, according to this communication : .
But on social networks, the harassing group around Xavier Azalbert is promoted and supported by the official Science Guardians account. First there are interactions with Olivierfente here : https://archive.is/Lo30r. Then there are interactions with zebulon75018 who has repeatedly depicted me as a pig or a cockroach https://archive.is/OYfUv. Xavier Azalbert is being literally reposted by the official account :
Science guardians also used the elements provided by this harassing group to attack Lonni Besançon on his online communication, never addressing the factual elements he has been showing in scientific journals or post-reviewing platforms.
This is not a personal attack, but rather a call to reflect on how scientific communication, ethical responsibility, and harassment often overlap in high-stakes health debates. When authors who engage in public intimidation also push controversial scientific claims—especially via predatory journals—it raises questions not just about credibility, but about the ecosystem of influence and pressure that surrounds post-truth science. Science Guardians should not associate with authors like those on this publication to harass science integrity specialists.
Points for peer discussion:
Should critiques of flawed studies be held to higher standards, especially when they become public flashpoints? How can they be supported when confronted with harassment campaigns ? How should platforms treat scientific contributions from individuals involved in public disinformation or harassment campaigns?
What safeguards can post-peer-review platforms implement to avoid amplifying ideologically driven science masquerading as evidence-based critique?
In conclusion, this case is not just about the scientific merit of one article—it is a test of integrity for the broader scientific conversation. I look forward to engaging with others here on both the methodological and ethical dimensions.