Affaires RéglementairesAutorités sanitairesCNRSEnquêteÉthiqueÉtudesRechercheRecherche cliniqueTraduction

INQUIRY. Publication of studies: how the IHU and Didier Raoult ignore the law and scientific ethics

by Victor Garcia

L’EXPRESS 6 October 2021

Translation by Citizen4Science’s editorial team courtesy of L’Express – Link to the original article

The IHU teams have published hundreds of studies that violate scientific ethics, including at least four of them that do not comply with the law governing experiments on human beings.

In July, L’Express revealed that the teams of the IHU Méditerranée Infection had published dozens of studies – some signed by Didier Raoult – that came close to, or even crossed, the limits of the law governing experiments on human beings. In addition to the work on Hydroxychloroquine, L’Express highlighted the absence of legal authorizations or the disregard of scientific ethics in studies conducted on minors, homeless people or students of the University of Aix-Marseille who underwent vaginal swabs. Offences that could be punishable by penalties of several thousand euros, or even imprisonment. Although the French National Agency for the Safety of Medicines (Agence Nationale de sécurité du médicament et des produits de santé, ANSM) and the University Hospitals of Marseille (AP-HM) have indicated that they have launched internal investigations, both organizations have not yet rendered their verdict.

Since then, we have identified some 250 new problematic studies. Most of them, of questionable quality, have above all the interest of generating Sigaps (System of Interrogation, Management and Analysis of Scientific Publications points), which allow to obtain important funding. This is a legal, but unethical practice – and one that the IHU has made an art of – that L’Express had already denounced a year ago. On a more serious note, four studies seem to have been conducted without legal authorization. Once again questioned by L’Express, ANSM indicated that it had « requested additional information » from the IHU and said it was ready to take sanitary or legal measures in the event of a breach. The Marseille-based institute headed by Prof. Raoult did not respond to our questions.

Particularly sensitive studies-no trace of authorization

French law is particularly vigilant concerning research involving the human person (RIPH for Recherche Impliquant la Personne Humaine ). They are classified in three categories and must all obtain the authorization of a Committee for the Protection of Persons (CPP), which ensure the relevance of the work and the absence of risk for the participants. This legal provision aims to protect French citizens from the abuses observed in the past. The most sensitive clinical trials, called RIPH1, must also be authorized by the ANSM. If the ANSM provides consent, it automatically registers the research in the European Clinical Trials Registry (EudraCT), an online searchable database that aims to inform patients and healthcare professionals about clinical trials. In summary, RIPH1s must have an authorization number from a CPP, the ANSM and have an EudraCT number. Yet, L’Express found that there is no record of such submissions for at least three IHU clinical trials involving fecal transplants. « Clinical trials that plan to change the fecal flora of patients are always RIPH1, because this operation is considered as the administration of a drug, says the ANSM [in this online document, editor’s note], » recalls Prof. Mathieu Molimard, a specialist in clinical pharmacology at the Bordeaux University Hospital.

The first of these trials was conducted from March to November 2013. It was the subject of a study published in May 2015 in the European Journal of Clinical Microbiology & Infectious Diseases. This work, signed in particular by Prof. Didier Raoult, presents « the preliminary before-and-after results of a prospective study comparing the mortality of 61 patients with an average age of 84 years, 42 of whom were treated with antibiotics and 19 with [fecal] transplantation, » the authors write. At the time, the law did not yet refer to « RIPH1 », but to biomedical research, which had to follow the same rules. « This study is clearly interventional research classified as ‘biomedical research’ according to the public health law then in force. It therefore required a favorable opinion from ANSM, a CPP and an EudraCT number », says Bettina Couderc, member of the National Conference of Personal Protection Committees (CNCPP). However, the document does not mention any of these authorizations. And the trial does not appear anywhere in the European register. The authors simply state that they « complied with the regulations on research involving humans ». They state that they have sought the advice of the ANSM, whose recommendations on clinical trials and fecal transplants they cite. These recommendations indicate that in the case of such a study, authorizations are necessary.

The second trial was conducted between May 2016 and May 2017 as part of a thesis supervised by Prof. Jean-Christophe Lagier. The student’s work aimed to compare the results of 199 patients who received a fecal transplant or antibiotics. « This is an RIPH in the sense of the current regulations and given the acts performed, it necessarily falls under the RIPH1 regulation, » says Philippe Amiel, a lawyer member of the Inserm Research Ethics Committee and president of the institution’s deontology college. However, the thesis does not mention any authorization from a CPP or the ANSM, only a « validation by the IHU ethics committee under the number 2007-009 ». This committee has no authority in the matter and cannot replace a CPP or the ANSM. Here again, the trial is conspicuous by its absence from the European clinical trials register.

The third was launched in the summer of 2019 by Prof. Lagier, head of the infectious and tropical diseases department at the IHU. The purpose of this work is to analyze the effectiveness of fecal transplants on 3088 patients with schizophrenia in treating their depression. « There is no doubt that the research aiming to demonstrate a benefit of fecal transplantation in major depression in schizophrenic patients is a RIPH1, a trial that is all the more sensitive as it concerns patients with a psychiatric pathology, a population to which the CPP and the ANSM are particularly attentive, » analyses Philippe Amiel. « Moreover, the scientific interest of this trial can be questioned since it does not plan to include patients who do not receive treatment, which would allow to compare their results with those who benefit from a fecal transplant and thus to evaluate a potential benefit », adds Mathieu Molimard. And once again, there is no trace of this work in the European registry of clinical trials. In fact, while the IHU indicates in an internal document available online that it has been conducting trials on fecal transplants since 2013, the European registry does not mention any.

In addition to the field of fecal transplants, L’Express also discovered that the IHU conducted a study in February 2020 to evaluate a method for detecting Covid-19. This work, published on June 16, 2021 in the Journal of Clinical Medicine and signed in particular by Prof. Didier Raoult and Jean-Christophe Lagier, is not particularly sensitive since it involved self-sampling of saliva on 212 people. According to the specialists we contacted, it is nevertheless a RIPH3. « This is a prospective, interventional study that involves data collection and examinations that are not part of standard care. It therefore requires a favorable opinion from a CPP and the approval of the authorities before it can be started », notes Fabienne Blum, Doctor of Pharmacy and President of the non-profit organization Citizen4Science. However, in the document the authorization is presented as « protocol 2021-014 » delivered by the Ethics Committee of the IHU Méditerranée Infection on… April 30, 2021. Not only does the opinion of the Ethics Committee of the IHU have no authority in this matter, but it was issued more than a year after the beginning of the research. Now, the fact of falsely claiming an authorization from a CPP and/or the ANSM constitutes not only a serious ethical breach – the people who participate are deceived – but carrying out such studies without authorization is punishable by up to one year in prison and a fine of 15,000 euros, in case the offence has been characterized .

How could these studies have been published in scientific journals? Questioned by L’Express, Professor Emmanuel Andrès, from the CHRU and the Faculty of Medicine of Strasbourg, co-editor of the Journal of Clinical Medicine, explains that « a journal has no means of ‘inquisition’ or investigation and – in our journal – the fact that the authors have declared that their study fulfils the administrative requirements is a mandatory element, but sufficient for the manuscript to be reviewed and possibly published. Moreover, before the Covid controversies, no journal had the slightest doubt about the ‘integrity’ of the data and the content of the manuscripts submitted by the IHU of Marseille and/or the paper signatories. »

246 studies but a single authorization number

The Express also discovered that the IHU published 247 studies between 2011 and 2021 that all indicate the same authorization number « IFR48 09-022. » Except that the general rule is: one study, one authorization from a CPP – and from the ANSM for a RIPH1 -, one number, » reminds Bettina Couderc. It can happen that a protocol obtains an extension, but this only works if the trial is amended at the margin ». « In any case, it is unthinkable that a single authorization is given for 247 studies, we have never seen that », adds Philippe Amiel. Did the IHU use this number to suggest that it had obtained the necessary authorizations? The competent authorities, including ANSM, will have to verify this. In the meantime, an analysis of these 247 studies reveals impediments to scientific ethics and questions the quality of the work and its interest. The vast majority of these studies were signed by Prof. Raoult (238 signatures), Prof. Fournier (174) and Prof. Lagier (104), but their protocols and research subjects are very different. 184 of them concern analyses of human stools, 24 concern vaginal smears and 13 concern urine samples. The last 26 analyzed various samples: breast milk, blood, etc. « It is in any case unimaginable that a single authorization be given for 247 studies, I have never seen that and there is no research that could acquire a single authorization over 10 years, » notes Philippe Amiel.

Results of the analysis of the 247 « IFR48 09-022 » studies according to the type of sample taken from patients.

And if most of them concern French patients, some studies concern African populations, dozens of which do not indicate the authorization number of the foreign country in question, which is mandatory in the framework of multi-country research. 80 include children, some of whom are only a few months old. The Express also compared the texts of these studies, two by two. It appears that 71 of them are more than 50% similar, and some are even more than 75% identical. This suggests that the IHU could use a pre-established model, reused over and over again. Again, a « bad practice », according to the experts interviewed by L’Express.

Result of the analysis calculating the percentage of similarities between the texts of different studies.

In addition, 56 of these publications have only one or two pages, 25 have three pages, and 55 have between 5 and 10 pages. Only 8 studies have more than 20 pages. Most of them concern a very small number of patients, less than 10 in most cases. Two hypotheses are then possible: either Prof. Didier Raoult and his teams conducted a dozen large studies and cut them up into a multitude of small publications – a known method, called « the salami technique » – or they published a study each time they performed a sample on a patient. In both cases, this is not acceptable and it is not part of good scientific practice, » says Hervé Maisonneuve, a doctor specializing in scientific integrity. All researchers have the right to discover a bacterium and to publish their work, but the vast majority normally wait until they have solid elements before submitting it to a journal.

List of scientific journals that published the 247 studies and average review time before publication.

Worse still, 31 studies were published in the scientific journal only one day after receipt, 126 in less than eight days. Only about 30 were accepted after three months. However, when a journal receives a study, it must re-read it and verify its results before publishing it. This is the guarantee, for the readers – health authorities, administrations, scientists, journalists and citizens – of the quality of the information presented. And it is perfectly impossible to reread and verify in less than a day, no more than in a week, even for a three-page study, unless you have relatives on the editorial board of the journal who accept without checking, » says Hervé Maisonneuve. This process is not only suspicious, it is unacceptable. Not surprisingly, more than half (135) of these studies were published – often only a few days after receipt – in New Microbes and New Infections (NMNI), a journal whose editorial board now includes five researchers from Marseilles, three of whom are from the IHU, and which was directed by a member of the IHU until 2020.

« Sigaps is a pretty good deal »

he question remains as to why IHU researchers publish so many studies of questionable quality. Two potential reasons emerge. The first is that the more a researcher publishes, the higher his or her H-Index climbs. This bibliometric tool, invented in 2005, is criticized because it only measures the productivity of scientists, whereas it is frequently used to evaluate their quality. Didier Raoult has never failed to point out that his H-Index is one of the highest in the world and twice as high as Albert Einstein’s. The second reason is that the more studies an institution publishes, the more Sigaps points it gets, for System of interrogation, management and analysis of scientific publications. And the more points it has, the more funding it receives. « For each Sigaps point, the Ministry of Health grants about 650 euros per year for four years to the researcher’s supervisory institution. An article earns between 1 and 32 points, i.e. between €650 and €20,080; if an author obtains 20 points during the year, the institution receives about €13,000 per year for 4 years », Francis Berenbaum, head of the rheumatology department at the Hôpital Saint-Antoine (AP-HP), told us in a previous survey on the IHU’s « Sigaps System ».

As Yanis Roussel, who manages Prof. Raoult’s communication and press relations, told us, the IHU allows the AP-HM, his supervisory body, to collect about « 10 million euros per year », i.e. 15% of the annual Sigaps funding received by the AP-HM (66 million euros) via Sigaps. This process is perfectly legal and the sums do not fall directly into the pockets of the IHU, but replenish those of its supervisory body, the AP-HM. Nevertheless, local agreements between institutes are common and at the discretion of the health establishments. And when such agreements exist, the budget is generally redistributed to the various teams in proportion to their contributions to the Sigaps points, » Francis Berenbaum told us. Didier Raoult does not hide the fact that he has used the Sigaps points as a means of pressure against the AP-HM, since he indicated, in the book IHU, « Le défi de la recherche et de la médecine intégrées« , that he had carried out ‘a Sigaps signature strike’ in order to find ‘common ground [with the AP-HM, editor’s note] for the launch of the IHU’. In the annual meeting of the IHU on February 13, 2017, he summarized the matter thus: « Sigaps (…) it’s pretty good business » (at 45’39 in this video). Furthermore, it is worth remembering that « the SIGAPS pie » is not scalable. The 1.6 billion euro envelope has been relatively stable for years. The institutes that take a large share of it therefore do so at the expense of those that publish less.

Why, then, have the authorities and supervisors, such as the ANSM and the AP-HM, not reacted for years? Why has the IHU been able to conduct clinical trials that require authorizations from a CPP and the ANSM without having them? Did the ANSM forget to register these studies in the European registry? « It is extremely unlikely, » says Professor Mathieu Molimard. If this is the case, how is it possible that trials were conducted without its authorization? Were these organizations unaware of the IHU’s abuses or were they aware of them and decided to turn a blind eye? When questioned, the AP-HM did not answer our questions. The ANSM, which has had the benefit of several weeks of reflection, merely stated that « investigations and verifications on possible breaches of clinical trial regulations by the Marseille IHU are still in progress » and that « additional information has been requested » concerning the issues raised by L’Express. As in July, the French agency states that « if these investigations reveal breaches of regulations, the necessary measures will be taken, including health measures to guarantee the safety of participants and/or a new referral to the courts ». An investigation that may at least appear to be late insofar as some of the work was conducted more than ten years ago and that these actions were known for a long time by the experts in the field.

« CNRS and Inserm were very shocked by these practices and stopped their collaboration with the IHU in January2018, » recalls Philippe Amiel. And it is indeed the ANSM that is supposed to police such cases. But the sad reality is that there is no structure in France that is able to deal with this kind of situation, with strategies to circumvent the rules on such a scale. » At the end of September, the CNRS ethics committee issued a very critical opinion on the professor from Marseille, a sign that the institute is also concerned about the methods of the IHU, and probably about the lack of reaction from the authorities. Just like the French Office for Scientific Integrity, which also published a press release directly targeting Prof. Didier Raoult. Fortunately, the vast majority of French researchers do not engage in these practices. But the IHU is not alone, and a small number of scientists, well known to specialists, show themselves to be disrespectful of the laws and of scientific ethics. And if these dysfunctions are not denounced, or even reprimanded by the authorities, this could send a very bad signal to all researchers who respect ethics and the law, as well as those who work within ethics committees. After all, why follow such restrictive rules if there are no sanctions for not following them? This lack of reaction could also reinforce the distrust of citizens towards administrative and scientific institutions when their confidence is already shaken by the Covid crisis and its controversies. And at a time when some people willingly imagine ubiquitous plots by « Big Pharma », it would undoubtedly be useful to recall that abuses do not always come from the designated « evil camp » and to show that, in all cases, the sanctions provided by the laws on scientific ethics are applicable.


Citizen4Science is a non-profit organization, promoting science with a professional editorial staff. We fiercely defend our independence. To support us, make a one-time or monthly donation by cliking the link below:

Propulsé par HelloAsso

Laisser un commentaire

Votre adresse e-mail ne sera pas publiée. Les champs obligatoires sont indiqués avec *

Résoudre : *
18 ⁄ 6 =