The Mediator case is back in court. The appeal trial, which started on Monday, January 9, is scheduled to last six months. At first instance, the Servier laboratories and their former number 2, Jean-Philippe Seta, had been found guilty of aggravated deception, and of manslaughter and involuntary injury. Servier was sentenced to a fine of 2.7 million euros (in addition to damages to the victims of 183 million euros), and Jean-Philippe Seta to a four-year suspended prison sentence and a fine of 90,600 euros. On the other hand, the laboratory had been acquitted of the offenses of improperly obtaining marketing authorization and fraud to the detriment of Social Security, hence the appeal of the Paris prosecutor’s office and the civil parties (Servier to him also appealed).
On the occasion of the opening of this new trial, L’Express questioned Séverine Carré-Pétraud, editorial director of Prescribe, an independent medical journal specializing in the analysis of the benefit-risk balance of drugs. His observation is as clear as it is maddening: thirteen years after the Mediator scandal broke, “too many drugs continue to expose patients to disproportionate risks”. Decryption.
L’Express: Do all lessons have been learned of the Mediator case?
Séverine Carré-Petraud: This amounts to asking whether or not such a scandal could happen again. Hard to say. But one thing is certain: even today, both prescribers and patients must be cautious, because doubts about drugs continue to benefit companies more than patients. A marketing authorization does not necessarily mean that the risk-benefit balance is really favorable. Similarly, reimbursement is not always synonymous with therapeutic progress. In the reports published each year by Prescribe, we show that about half of the 100 to 130 new drugs put on the market each year or new indications show no clinical progress. A little less than a third provide minimal to significant progress, and around 15% of drugs or new indications expose them to more risks than existing drugs.
“On the fact that a drug can be authorized without proof of effectiveness or progress compared to existing treatments, nothing has changed.”
The writing of Prescribe had been the first to alert on the Mediator…
Its active principle being a derivative of amphetamines, we said very quickly that it could present undesirable effects close to those already known with these molecules: pulmonary arterial hypertension, valvulopathy… In 1997, two anorectics from the same family, also developed by Servier, have been withdrawn. But not the Pick, then marketed as an antidiabetic and hypolipidemic. That same year, Prescrire had made it clear that there was no reason to use this drug, as we had no evidence of efficacy in the indications for which it was sold, and there were signs of serious adverse effects. In the years that followed, we regularly recalled the risks presented by this product, which was finally withdrawn in 2009.
Since then, what changes have taken place?
On the fact that a drug can be authorized without proof of effectiveness or progress compared to existing treatments, nothing has changed. A large majority of treatments arrive on the market without the laboratories having demonstrated any therapeutic progress. New products are often compared to a placebo, but not to other drugs. So we don’t know if they do better than other effective treatments that are already available.
Moreover, a good number of drugs are authorized on the basis of data which evaluate biological criteria and not criteria of the least clinical complication – fewer hospitalizations, fewer deaths… For example, in the case of diabetes, patients would like find out if a particular drug will reduce the risk of stroke or kidney failure. But concretely, the only available data most often concern the effect of the product on blood sugar levels. The treatments are not evaluated over a sufficient period of time to be able to conclude otherwise. However, in the long term, they can lead to serious adverse effects that go beyond the interest of lowering blood sugar. We have seen this with gliptins, a large family of antidiabetics approved in the early 2000s: they certainly lower blood sugar levels, but serious adverse effects have been observed which make the benefit-risk balance unfavorable.
It’s the same for many treatments, including those for cancer. They too are often evaluated on biological or imaging criteria. But at the time when they benefit from a marketing authorization, one generally does not know their effect on the lifespan of the patients. Even though they can expose patients to serious and even fatal adverse effects.
Are the risks presented by drugs not better monitored since the Mediator affair?
In the case of Mediator, we were able to see that the foreseeable risk of serious adverse effects linked to pharmacological proximity to amphetamines had not been taken into account. It is difficult to know if this type of possible side effect is now better anticipated.
“Reporting of adverse events seems to have increased, whether by patients or caregivers.”
At the time, no one had either taken into account the first reports of adverse effects linked to benfluorex. This opacity on the results of pharmacovigilance has decreased, in particular due to European regulations that France has been obliged to adopt. But it still takes too long for these alerts to translate into drug withdrawals. The example of pholcodine, a cough suppressant authorized for a long time, is a good illustration of this: we realized in the 2010s that there were serious signals of cross-allergy with anesthetic drugs, but that is only the end of 2022 that the European Medicines Agency has decided to withdraw it. Ditto for fenspiride (Pneumorel for its trade name), another medicine from the Servier group marketed since the 1970s against coughs: it was only withdrawn in 2019 despite various signals, particularly cardiac.
But have there been any positive developments since the Mediator scandal?
Sure. Reporting of adverse events seems to have increased, whether by patients or caregivers. The Mediator case also stimulated the use of data from the SNDS, the national health data system, to assess health products.
Finally, the Transparency Commission of the High Authority for Health, which issues opinions on the reimbursement of drugs, is much better organized, with fewer ties of interest, more requirements and transparency. It represents a useful filter, because when it decides against the reimbursement of a drug, companies can choose not to market it.
Despite everything, you regularly point out medications that should be withdrawn…
Our latest report published at the beginning of December 2022 shows that 88 drugs marketed in France still expose patients to disproportionate risks, even though they do not present a clinical benefit, or the pathology in which they are used is not serious.
How do you achieve this result?
Our analyzes are based on our own documentation, which is very extensive. We analyze the opinions of health agencies in Europe, the United States, Australia, etc. We also ask companies for additional information, even if they rarely play the game. We attach as much importance to adverse effects as to efficacy data, we take into account information on the pharmacological groups to which the molecules belong. It is by crossing all these data that we establish our conclusions. And moreover, of course, we have no links of interest with the laboratories.
Can you give us some examples of these drugs that you think should be withdrawn from the market?
Among the best known to the general public are, for example, nasal decongestants based on pseudoephedrine. They may have transient symptomatic effects on runny nose during colds, but they expose people to cardiovascular effects that can go as far as stroke or infarction. The same goes for Toplexil, a cough suppressant that can cause neurological disorders (sedation, hallucination, convulsions, etc.). Or even Maxilase, against sore throats, devoid of any demonstrated effect beyond that of a placebo, but which can cause sometimes serious skin or allergic disorders.
We should also mention oral diclofenac (Voltarene), a non-steroidal anti-inflammatory drug that presents more cardiovascular risks than other drugs in the same category. Or medicinal clays like Smecta, naturally polluted with lead. Or certain products for reflux and nausea, such as domperidone (Motilium) and metopimazine (Vogalene and Vogalib). They are used in gastroenterology, whereas they are neuroleptics. They therefore expose you to the same adverse effects as other neuroleptics, with in particular the risk of heart rhythm disorders, stroke, and sudden death. We have been warning about all these drugs for a long time, and many others. But it is clear that not much is happening. Even today, patients can continue to take them, and we can only regret it.