AG Saugmandsgaard Øe provides guidance on the application of EU competition law in the pharmaceutical sector

**Update**: see here the comment on the final judgment of the CJEU.

On 21 September Advocate General Saugmandsgaard Øe provided his Opinion to the CJEU on some key issues regarding competition law in the pharmaceutical sector. The request for a preliminary ruling was referred by the Italian Supreme Administrative Court (“Consiglio di Stato”) in relation to a cartel case where the Italian Competition Authority (“ICA”) fined Roche and Novartis for a total amount of 180 million euros.

The facts of the case

According to the ICA, the companies colluded to artificially reduce the demand by medical practitioners for Avastin, a drug marketed by Roche for the treatment of cancer but prescribed off-label in the treatment of age-related macular degeneration (“AMD”), to the advantage of Lucentis, a drug marketed by Novartis and officially approved for the treatment of AMD but ten times more expensive than Avastin.

This goal would have allegedly been achieved through (i) the production and dissemination of opinions which called into doubt the safety and the efficacy of the off-label use of Avastin, downplaying the value of scientific evidence to the contrary; (ii) the request by Roche, against its interest, to the European Medicines Agency (“EMA”) for an amendment to Avastin’s SPC in order to warn the public (doctors and patients) on the alleged risks connected to the off-label use of the drug and get the medicine removed from the list of the reimbursed drugs; (iii) the request by Roche of an authorisation to send a “direct healthcare professional communication” (“DHPC”) to medical practitioners illustrating the amendments to Avastin’s SPC.

The ICA found that the parties had an incentive to implement this strategy since Roche would have earned royalties through its subsidiary Genentech, licensor of the patent covering Lucentis, and Novartis’ would have earned directly from the sales of Lucentis and from its 33% shareholding in Roche.

The questions referred to the CJEU and the AG’s answers

The questions referred by the Consiglio di Stato to the CJEU and addressed by AG Saugmandsgaard Øe can be recapped as follows.

1) Is the scope of marketing authorisations (“MAs”) binding on competition authorities for the purpose of defining the relevant market? In other words, do off-label and authorised medicines have to be included in the same relevant market?

AG Saugmandsgaard Øe took a substantial approach in examining the interplay between the content of MAs and the definition of the relevant market under EU competition law. More specifically, he drew attention to the fact that the relevant market always comprises those products which are regarded as objectively interchangeable by the demand side. Therefore, although sectorial regulation may play a role in influencing the degree of substitutability of the products, it has to be assessed taking into account the characteristics of the product and the actual competitive conditions.

Having said that, in the case under review AG Saugmandsgaard Øe identified the demand side of the market with prescribing medical practitioners and recognised that, in general, MAs are likely to guide doctors in choosing the most appropriate treatment on a case by case basis. However, he considered Avastin and Lucentis as part of the same relevant market since medical practitioners prescribed Avastin for a long time for ophthalmological indications in spite of the uncertainty regarding the lawfulness of this practise in presence of a drug approved for that indication.

2) Should the parties of a patent licensing agreement (i.e.: Genentech and Novartis) be regarded as competitors if the licensee operates on the relevant market solely by virtue of that agreement and, if they are not, can their collusive conduct harm competition?

AG Saugmandsgaard Øe recalled that, under the Technology Transfer Block Exemption Regulation, the parties to a licensing agreement are competing undertakings in the relevant market for the contract products if, absent the agreement, they would have been actual or potential competitors in that market. Therefore, in the present case, since Novartis had not planned to commence R&D in relation to drugs for the treatment of AMD or other ocular vascular pathologies, Novartis and Genentech had to be considered non-competing undertakings.

Nevertheless, in the view of the AG, the restrictions agreed by the two parties, although not expressly set out in the licensing agreement, would give rise to a significant restriction of competition. In this regard, the AG argued that the strategy implemented by Roche and Novartis eliminated all the competitive constraints on Lucentis by influencing the actions of the pharmaceutical regulatory authorities and medical practitioners in order to eventually alter the demand of Avastin.

3)  Should a concerted practice intended to emphasise that a drug is worse than another be regarded as a restriction of competition by object when that assumption cannot be indisputably excluded?

In this regard, the AG recalls that a restriction of competition by object occurs when an agreement or a concerted practise reveals a sufficient degree of harm to competition to render the examination of the effects superfluous. The criteria to be taken into account for the purpose of this assessment are (i) the content of the agreement, (ii) the objectives it seeks to attain and (iii) the economic and legal context in which it is concluded. Also (iv) the subjective intention of the parties, while not a necessary factor, may be a relevant factor for that purpose. The examination of these features allows to check whether there is an alternative explanation for the coordination other than the pursuit of an anticompetitive aim.

AG Saugmandsgaard Øe applied the above criteria to the conduct at issue in the proceedings and concluded that the circumstance that the parties exploited a scientific uncertainty for the purpose of discouraging the off-label use of Avastin in such a way as to alter demand in favour of Lucentis would amount to a restriction of competition by object. This was the case especially because the information disseminated by Roche and Novartis was presented selectively or incompletely and the companies omitted to provide a correct picture of the risks, as they were aware that in Italy medical practitioners are particularly sensitive to safety considerations surrounding the off-label prescribing of pharmaceutical products.


The opinion delivered to the CJEU by AG Saugmandsgaard Øe does not bind the CJEU, which will be free to address the referred questions differently. However, it is interesting to see that in regard of all the issued raised by the Italian Court, the AG adopted a substantial approach and reaffirmed the independent applicability of EU competition law in the disputes which involve also other branches of law, like IP law or sectorial legislation.

That said, if the CJEU follows the advice provided by AG Saugmandsgaard Øe, there will be three major developments in the case law related to the pharmaceutical sector: (i) MAs will not bind competition authorities in the definition of the relevant market; (ii) licensing agreements between non-competing undertakings may contain clauses which eliminate all the competitive pressure on the licensing parties; (iii) the reliability of the scientific information concerning the efficacy of a drug and the risks related to it can influence the decision-making process of the competition authorities.

Print Friendly, PDF & Email