France: Recent Court decisions reversing the burden of proof for pharmaceuticals companies who may have supplied drug which caused cancer


On two occasions, 24 September 2009[1] and 28 January 2010[2], the First Civil Chamber of the Cour de Cassation (Supreme Court) has decided in favour of the joint and several liabilities of two pharmaceutical companies without it being known which one had produced the molecule (diethylstilboestrol, “DES”) which caused the damage. This amounts to a reversing the burden of proof and imposing on the pharmaceutical company the responsibility of providing proof that its product was not the cause of the damage.

The facts

The legal experts stated categorically that the patient had indeed been exposed, nearly 40 years earlier, to DES and that this had caused her cancer. What was not known however and could not be determined was whether the patient had been exposed to DES produced by UCB or Novartis.

The decision of the courts

In the first decision to be considered here, the Cour de Cassation held that since “the DES was indeed the direct cause of the tumour, Mme Y... having been exposed in the womb to the molecule in dispute… it was therefore the responsibility of each party to prove that its product was not the cause of the damage”. Similarly, in the second decision, the Supreme Court stated: “…in the case of the victim’s  exposure to the molecule in dispute, each of the drug companies which has marketed a product containing the molecule must prove that its product is not the cause of the damage…”.

These decisions therefore establish a reversal of burden of proof and the need to provide negative proof, which by its very nature is almost impossible to do.

The scope of the decisions

In a somewhat cryptic press release[3] published by the Supreme Court regarding the decision of 24 September 2009 states that “Reversing the burden of proof in favour of the victim, the Cour de Cassation decided that it was the responsibility of each of the pharmaceutical companies to prove that its product was not the cause of the damage.

It shall then fall upon the Court of Appeal of Paris…to seek, with a view to indemnifying victims, whether one of the companies can show that its product was not administered to the claimant”.

This press release may therefore leave open the possibility that these decisions shall in fact be strictly limited to the need to compensate the victims. Legal analysis leads however to the limiting or at least circumscribing, of the scope of the decisions of the 24 September 2009 and 28 January 2010:

  • The facts of the case occurred before the law of 19 May 1998 which transposed the Directive; the risks of the same scenario recurring in the future are therefore limited.

  • In implementing the Directive, each Member State may choose to utilise any latitude set out in the Directive (this has been clearly decided, on at least four occasions[4]). Yet the freedom to dispose of the need for a claimant to establish that the damage is caused by a particular product was not part of the latitude granted by the Directive of 25 July 1985; the terms of Article 1386-1 of the Civil Code, according to which the claimant must bring proof that damage is attributable to the administration of a particular product, re-enforces this position.

  • A moral concern of the courts may have influenced their worrying decisions here: the Court of Appeal of Versailles, then the Cour de Cassation both ruled that UCB had committed a prolonged error in the management of warning signals that it was receiving in relation to DES and noted a failing “in its duty of vigilance”[5]; This may explain the practice of the ‘collective’ penalties, which are, in our view, not justified in law.

Indeed, the rare cases in which the Supreme Court has upheld the existence of collective liability have concerned cases in which collective behaviour, and, what is more, wrongdoing, was the cause of the chain of events leading to the damage, without it being possible to identify any specific perpetrator. Such liability has not previously been applied to pharmaceutical companies however.

It seems that the same result could be achieved – an order made jointly and severally against both producers - without having to go through such ‘difficult’ legal hoops, by holding that the patient had alternatively ingested the DES produced by one and then the other pharmaceutical company all within the framework of appraisal by the judges on the merits [6]  of the evidence submitted to it[7].


It seems premature to conclude that the solution found by the courts in these rulings is now inscribed in French law and in any case, such a solution would only apply for the purposes of the indemnification of the victims of older cases. However, such rulings do call into question whether the pharmaceuticals companies have received a fair trial.

[1]  Cass. civ. 1ère 24 September 2009 n° 08-16097.
[2]  Cass. civ. 1ère 28 January 2010 n° 08-18837.
[3]  Press release relating to decisions n° 878 and 880 of 24 September 2009 of the First Civil Chamber, published on the Cour de cassation website
[4]  ECJ 25 April 2002, three decisions rendered against the French (C-52/00) and Greek (C-154/00) states, and the kingdom of Spain(C-183/00) respectively; a solution reaffirmed by the ECJ 10 January 2006 case SKOV C-402/03.
[5]  CA Versailles 10 April 2004 -RG 07/02477: “despite doubts (…) about the effectiveness of diethylstilboestrol  and its harmlessness which the literature ( …) spoke of, (…) UCB took no measure  although it should have acted, even in presence of conflicting  results (…)”  
[6]  Cass. civ. 1st 22 May 2008, n° 07-17200; Cass. civ. 1st 22 May 2008, n° 05-10593; Cass. civ. 1st 22 January 2009 n° 07-16449; Cass. civ. 1st 09 July 2009 n° 08-11073; Cass. civ. 1st 24 September 2009 n° 08-16097.
[7]  This track was suggested by the Court of Appeal of Versailles of 29 November 2007, decision against which appeal was dismissed on 24 September 2009 (n° 08-10081).