On January 23, 2018, the European Court of Justice (CoJ) handed down an important judgment in the Hoffman-La Roche / Novartis case (C-179/16). For the first time, the CoJ takes a stance on an emerging hot topic in EU antitrust law: disparagement – or, in more trendy terms, fake news. And the CoJ’s message is clear: the EU will show no mercy for businesses engaging in such practices.


The dispute relates to the distribution and sale in Italy of two medicinal products manufactured by Genentech, a Roche subsidiary:

  • Avastin, distributed by Roche and its subsidiary Roche Italia. Avastin received a marketing authorization (MA, mandatory under EU law) for the treatment of certain tumorous diseases in 2005.
  • Lucentis, distributed by Novartis and its subsidiary Novartis Italia. Lucentis received a MA in 2007 for the treatment of eye diseases.

Prior to the placement on the market of Lucentis, and in the absence of a therapeutic alternative, certain doctors would also prescribe Avastin for certain eye diseases. This is a practice called ‘off-label use’, authorized under Italian law. This practice continued after Lucentis received its MA two years later, due to the lower unit price of Avastin.

In a decision adopted on February 27, 2014, the Italian competition authority (Autorità Garante Della Concorrenza, AGCM) took the view that Roche and Novartis had concluded an anti-competitive agreement designed to put an end to the off-label use of Avastin for eye diseases and to boost the sales of the higher priced Lucentis product.

As part of this agreement and to further their strategy, the AGCM found that Roche and Novartis had ‘conspired to communicate to the pharmaceutical regulatory authorities, to medical professionals and to the general public statements to the effect that the cancer treatment when used off label was less safe than the ophthalmological medicine’. To that effect, they: (1) produced and disseminated opinions which could raise concerns regarding the safety of Avastin for the treatment of eye disease, while downplaying opinions to the contrary, and (2) engaged proceedings before the European Medicines Agency (EMA) to register certain side effects associated with the use of Avastin and be authorized to send formal communication on the topic to healthcare professionals (NB: the EMA eventually did not grant this authorization, but amended the summary of Avastin’s characteristics in order to mention certain side effects associated with its off-label use).

In light of these elements, the AGCM imposed a 92 million euros fine to Roche, Novartis and their respective Italian subsidiaries.

After an unsuccessful action before the Tribunale amministrativo regionale per il Lazio, Roche, Novartis and their Italian subsidiaries brought actions before the Consiglio di Stato, which referred a number of questions to the CoJ. Among other questions, the Consiglio di Stato asked the CoJ whether the practice under examination – namely, the dissemination of information suggesting that a medicinal product is not innocuous or may be less effective, when this information is not supported by scientific evidence, but cannot be indisputably excluded either – amounts to a restriction of competition by object.

The CoJ’s Take: Dissemination of Misleading Information as a Restriction by Object

The concept of restriction of competition ‘by object’ must be interpreted strictly, and can only be applied to those types of coordination that reveal a degree of harm to competition that is sufficient for it to be held that there is no need to examine their effects.

On the basis of this rule, Roche and Novartis argued that their practice did not qualify as a restriction by object. According to them, the dissemination of information regarding Avastin and Lucentis spurred from genuine concerns about the safety of Avastin in ophthalmology and aimed at protection public health, as well as the reputation of Roche as manufacturer and distributor of Avastin.

The CoJ dismissed this argument, on the basis of the following elements:

1. First, as per the previous case-law on restrictions by object, regard must be had to the context in which the practice takes place. In this case, the relevant context includes the EU rules on pharmaceutical products, which set up a pharmacovigilance system. Under these rules, the holder of a MA must, inter alia: (1) supply any new information on the product to the EMA, the Commission and the Member States, and (2) ensure that information to the public is presented objectively and in a non-misleading way.

2. Second, in assessing the anticompetitive objective of the practice, the following elements must be taken into account:

  • Pharmacovigilance obligations rest solely on the holder of the MA (in this case, Roche), and not with another undertaking marketing a competing medicinal product (g. Novartis); and
  • More importantly, information may be qualified as misleading when it aims at confusing, in a context of scientific uncertainty, the EMA, the Commission and the general public as to their perception of the risks associated with the off-label use of Avastin. The CoJ takes the view that, given the characteristics of the medicinal product market (g. the fact that doctors are typically risk adverse), the dissemination of such information will encourage doctors to refrain from using Avastin.

Interestingly, the CoJ’s analysis does not factor in the correctness of the information. This suggests that the dissemination of information which is correct but presented in a partial and selective way does amount to misleading information (to be noted that this view is explicitly spelt out in the non-binding opinion of the Advocate General). In other words, the CoJ appears to set a fairly low standard in assessing whether information is misleading or not.

In Conclusion: Beware of Disparagement!

It may have been largely unnoticed, but disparagement is slowly emerging as an antitrust trend in several European countries.

In addition to this Italian case, the French Competition Authority (FCA) has also been very active in enforcing disparagement practices, adopting four decisions between 2013 and 2017. Sanofi-Aventis, Schering-Plough and recently Janssen-Cilag were fined respectively €40.6 million, €15.3 million and €25 million for having implemented disparagement strategies in order to delay the development of competing generic drugs (decisions 13-D-11, 13-D-21 and 17-D-25). In 2014, the FCA also fined a yogurt manufacturer for disseminating misleading statements regarding the sanitary quality of the dairy products of one of its competitors.

Against this background, the CoJ’s judgment in Hoffman – La Roche / Novartis further clarifies the legal regime applicable to disparagement. In our view, the judgment conveys three key messages:

  • Disparagement can come in many forms and shapes. The Hoffmann – La Roche / Novartis case arguably features a rather unconventional and somewhat counterintuitive disparagement scenario, that is, one where the disparaged party is also part of the collusion. As such, this case illustrates the huge variety of scenarios that can fit under the header ‘disparagement.’ However, that is not to say that any form of criticism towards your competitors will get you in trouble. In practice, all cases to date deal with well-organized disparagement campaigns. Thus, a few negative comments in passing are unlikely to give rise to an investigation.
  • Disparagement goes beyond fake news, at least in certain sectors. As noted above, the concept of disparagement extends beyond the dissemination of incorrect information. The dissemination of correct information but in a partial way may also raise issues, especially when it impacts an economic sector that is highly risk-adverse. In this regard, we note that most precedents on disparagement – if not all – featured misleading health claims. It remains to be seen whether a similar theory of harm could be argued in relation to less sensitive sectors or in relation to practices unrelated to human health.
  • If prosecuted under Article 101 TFEU, effects do not need to be proven since it can be a “by object” infringement. As a result, failed disparagement strategies may also trigger antitrust enforcement.

Based on the above, it appears that the CoJ judgment sets a fairly low standard to prove antitrust infringements based on disparagement practices. This, in turn, may boost enforcement, both at EU and at national level.

Therefore, businesses operating in sectors where demand is highly risk-adverse should include disparagement issues in their compliance efforts. This is particularly relevant in sectors where human health is at stake, such as pharma and food. Conceivably, this could also be a risk in a wide array of other sectors. For instance, consistently casting doubts on the environmental safety of a given substance or pushing a bank to withdraw its support to a competitor by claiming that it is in financial difficulties may well, under certain circumstances, constitute disparagement in the eye of an antitrust authority.