The French medical watchdog the Agence nationale pour la sécurité du médicament (ANSM) could have prevented thousands of women from having sub-standard breast implants made by French company Poly Implant Prothèse (PIP) had it reacted more quickly. That is the damning verdict of an internal report at the agency – then called Afssaps – written by a member of staff and which has been seen by Mediapart.
The confidential report claims that Afssap could have ordered the sub-standard silicone gel implants to be withdrawn from use in 2007 or 2008 had it responded more promptly to the growing evidence that the PIP implants were at risk from rupturing and causing potential health problems for women. In the end the implants were not removed from the market until March 2010 when an Afssaps inspection revealed that they contained a gel that had not been authorised for medical usage. Had the implants been withdrawn when the internal report claims they should have been, between 6,000 and 10,000 French women – roughly a third of all women who had PIP implants in France – would have avoided having the sub-standard products implanted.
PIP was considered one of the world's leading suppliers of implants with more than 42,000 women in Britain receiving the implants, more than 30,000 in France, 25,000 in Brazil and 15,000 in Colombia, and the scandal of the faulty implants caused a global health scare with an estimated 300,000 women in 65 countries affected.
PIP's founder Jean-Claude Mas and four members of his senior staff went on trial in Marseille in April this year accused of aggravated fraud by cutting costs in most of the breast implants it sold. A verdict is expected later this year.
The internal report seen by Mediapart was seized by investigators during a search of ANSM offices carried out in late April 2012. This was part of an ongoing investigation being carried out by examining magistrate Annaïck Le Goff into allegations of deaths and injuries or illness linked to the implant scandal.
In a statement ANSM insisted that its official report in February 2012, handed to then health minister Xavier Bertrand, into the agency's handling of the scandal had fully examined its own actions and that it had not hidden anything. The internal report seen by Mediapart, written two months later in early April 2012, had been a “personal and retrospective” analysis by a member of staff based on exactly the same facts covered by the official report, the agency said.
However, unlike the internal report seen by Mediapart, the official report in 2012 did not say that the agency should have reacted to events in 2007 or at the latest 2008. The official report insisted that post-market surveillance – the system in which manufacturers are obliged to report adverse incidents with their products to the relevant competent authorities – did not throw up any “significant alerts” between 2002 and 2008. It was only in 2009 that the alarm was raised.
Though there have been other indications that Afssaps was slow to react to the unfolding implant scandal, notably in warnings from Marseille surgeon Christian Marinetti in late 2008 about “defective material” and the possible harmful effects of the gel, the report seen by Mediapart is the first time that criticism of its actions has come from inside the agency.
Using the same data on post-market surveillance available to Afssaps, the internal report notes that “the rise in the number of ruptures of PIP implants began from 2006”. Even allowing time for the information to show up, the author claims that “Afssaps should have reacted from 2007 and at the latest 2008” to the growing problem with the implants. This would have prevented the use of “between 12,891 and 20,933 units (given to a minimum of between 6,445 and 10,466 women), in other words a third of [French] women with PIP implants”.
The same internal document also criticises the medical watchdog for seeking to reassure women about the PIP implants even though it only possessed incomplete toxicology data relating to any potentially harmful effects of the gel used in them. “It is incomprehensible that a competent health authority, basing itself on data on genotoxicity and short-term local irritation, can deliver reassuring conclusions on gels whose manufacturing process, physical and chemical character, and danger evaluation...are unknown,” says the author. In some cases women could have been exposed to the gel for more than ten years.
The report says that Afssaps should have recommended the removal of the implants immediately after it suspended their use in March 2010. Yet in April 2011 the agency was still advising women to have a “clinical examination and a scan every six months”, and did not say the implants should be removed if there was no rupture. It was not until December 2011 that the French health ministry recommended the preventative removal of PIP implants “even without signs of deterioration of the implant”. This recommendation was not the result of a better risk evaluation but followed the death from lymphatic cancer of a patient who had had a PIP implant – though no link has been established between the cancer and the implant.
“The decision to postpone [recommending the] removal of PIP implants from women for an estimated period of one or two years led as a consequence to an additional exposure that could have been avoided, as taking into account the absence of toxicology data, the benefit of the doubt should have been given to patients,” says the internal report's author.
Impact on fertility?
The confidential document does not come to any conclusions about the risk of cancer caused by the gel – either slowly seeping into the breast or after a rupture – because there was no data on which to reach a verdict one way or the other. Once the implant scandal came to light the agency ordered a number of tests on PIP gel samples. These tests showed that the gel could be an irritant but that there was no evidence of genotoxicity, which can lead to a risk of cancer. However, the data gathered by Afssaps was not enough to establish a full toxicological profile of the gels. Moreover, the 2010 results on genotoxicity were considered unreliable and had to be confirmed by new tests in 2011.
In any case, none of these factors allow one to exclude the risk of cancer definitively. Yet on ANSM's website there is a link to “important information regarding the monitoring of women who had PIP implants”, according to which “there is no data allowing one to conclude there is an increased risk of cancer”. To support this claim the agency cites the advice given by France's national cancer institute the Institut national de cancer (Inca) dated December 22nd, 2011, which concludes by referring to “the absence of increased risk of breast cancer among women with implants in comparison with the general population”.
The problem with this view, however, is that by the end of 2011 at least 15 cases of breast cancer among women who had PIP implants had come to light. By May 2013 the number had reached 70.
The only way to be sure that there is no increased cancer risk for patients with PIP breast implants is by carrying out a rigorous epidemiological study, which has not been done. The message has nonetheless been conveyed via the media that, though an irritant, PIP gel is not carcinogenic, even though this is not based on conclusive evidence.
Independent of any risk of cancer, the PIP gels contain a compound called siloxane D4 which is classed as 'reprotoxic', meaning it can have an affect on reproduction. In approved medical gel – for example the gel Nusil – the concentration of D4 is kept to very low levels, 50 parts per million (ppm), to remove any risk. In September 2010 Afssaps had tests carried out at a laboratory in the southern city of Montpellier which showed that PIP gels contained up to 400ppm or 500 ppm of D4. The implant shell itself contained a 1% concentration. This suggests that PIP gels could have reprotoxic properties.
The issue is complicated by the fact that the composition of PIP gels varied over time and has not been systematically studied batch by batch. This means that no one knows exactly what the composition of a implant gel given to a particular patient is. But in any case, by 2010 the medical watchdog agency surely had enough elements of concern in its possession to recommend to women with PIP implants to have them removed as a precaution.
For the author of the internal report the overriding question, and one that is still unanswered, is: “Why did Afssaps decide to postpone the decision to [recommend] to women [the removal of] PIP implants even though the available toxicological data, even in the absence of potential genotoxicity, did now allow for the carrying out of a state-of-the-art risk evaluation”?
This question did not appear in the official report handed to health minister Xavier Bertrand in February 2012. This report, which sought to trace the relationship between the agency and PIP since 1995, was co-signed by the ministry of health’s own health safety unit the Direction générale de la santé (DGS). But Mediapart understands that most of this official report was written by experts at Afssaps.
The official document agreed that the tests carried out by the agency were not enough to exclude the risk that the gels were toxic, and that ultimately there was sufficient evidence for the “precautionary removal of implants and the screening of women who had had implants”. But why didn't the agency make this recommendation earlier, as soon as it was aware of the high levels of siloxane D4? In Sweden this year the medical watchdog recommended the removal of PIP implants due to the “risks of serious inflammation” because of the high level of D4. Yet to this day Afssap's successor ANSM still advises on its website that surgeons should offer to remove patients' implants “as a preventative measure and not as a matter of urgency”.
'Nothing was hidden'
On the key question of when the agency first knew there was a problem with the PIP implants, the figures cited in the official report on post-market surveillance back its later conclusion that there was no significant factor justifying raising the alarm until 2009. However, this can perhaps be explained by the fact that the statistical study that shows there was a rise in ruptures from 2006 onwards does not appear in the official report.
The author of the confidential report, on the other hand, did seek to get access to the full data on all reported problems with the implants. The agency refused but did supply the author with a table of partial data on reported ruptures in the years 2001 to 2009. It was from this table that the author was able to show that the number of implant ruptures had increased from 2006.
Why the agency did not include these figures in its own report is unclear. According to an email seen by Mediapart, and dated January 16th, 2012, the author of the confidential report tried to get their analysis included in the summary of the official report, but without success. Instead the official report only details problems with the PIP implants reported by surgeons, and has no detailed analysis of the statistics provided by the manufacturer. Yet it is well-known in medical circles that health professionals have a tendency to under-report undesirable effects of medicines or medical products.
Enlargement : Illustration 3
Instead, the official 2012 report simply notes that an “increase in the rate of ruptures coming from the manufacturer” was indeed noted between 2007 and 2009, but explains that this rise was “considered to be an artefact linked to an over-declaration by PIP”. In other words, the agency says that it did not take account of the incidents highlighted by PIP itself because it judged that the manufacturer was declaring too many.
Nor does the Afssap official report contain the full analysis showing the high levels of siloxane D4 in the gels. This was despite the fact that one expert on a working group helping with the report – not the author of the confidential study – asked for such an analysis to be included. Instead it simply refers somewhat elliptically to the fact that D4 can cause fertility problems, and in another section notes the “high levels in siloxanes D 4 to D13” without drawing any conclusions from this for the health of women with the implants.
In summary the official agency report lacks some of the telling detail included in the later unofficial confidential report, even though they were based on the same data. And this is detail which appears to have considerable bearing on deciding whether the medical watchdog could or should have acted sooner to put an end to the PIP implant scandal.
In a statement the ANSM strongly denied having hidden or removed any potentially embarrassing information from its 2012 report, which it says included a “chronology of the facts, an analysis of the post-market surveillance data, the results of inspections carried out by the health authorities, a summary of the available toxicology data”. The agency says that two months after its own report was published a member of staff “handed over to management at their own initiative a personal and retrospective analysis carried out on the basis of the same toxicological and post-market surveillance data” as the official report. It concludes that “the official report made public two months earlier was absolutely not in any way purged of elements contained in a document that was produced later”.
Contacted by Mediapart, ANSM's director Professor Dominique Maraninchi said that their report had been transparent. “We have never wanted to withdraw information. I maintain that all the source data is present.” Professor Maraninchi also said that the internal report was “not very contradictory” of the official one, which he insisted did not spare the agency either.
----------------------------------------------
English version by Michael Streeter