On September 9, 2020, the French Competition Authority (“FCA”) fined Novartis, Roche and its subsidiary Genentech €444 million for abusing their collective dominance on the market for AMD treatment. The FCA found that the parties disparaged the off-label use of Roche’s Avastin drug and spread an alarmist discourse before the public authorities in order to preserve the dominant position and high price of Novartis’ Lucentis drug.


In the late 1990ies, US pharmaceutical company Genentech developed a revolutionary active ingredient called bevacizumab[1]  (later sold under the brand name Avastin) to treat certain types of cancer. Bevacizumab was licensed to its parent company Roche for distribution outside of the United States, and was granted a market authorization by the European Commission in 2005. In parallel, having found that Avastin could also help treat age-related macular degeneration (“AMD”), Genentech developed and started selling ranibizumab, a specific molecule (sold under the name of Lucentis) that it considered more appropriate for treating AMD. Lucentis was licensed to Novartis for distribution outside of the United States, and was granted an EU market authorization in 2007. Novartis, already at that time, was one of Roche’s shareholders with a 33,33% share of the voting rights. Roche, in turn, held a majority stake in Genentech, and acquired all of its outstanding shares in 2009.

As Avastin was about 30 times cheaper that Lucentis, some doctors administered Avastin “off-label” for AMD treatment, on their own responsibility, where, on the basis of an assessment of individual patients, they concluded that it was necessary to do so to meet a patient’s specific needs. Roche, however, never applied for an authorization to market Avastin for AMD treatment. In light of the objective differences between the two molecules, a scientific debate arose on the respective safety and efficacy of Avastin and Lucentis for AMD treatment. Between 2010 and 2013, a number of international and French scientific studies were conducted on the topic.

In parallel, the French regulatory framework (which, in particular following the Mediator healthcare scandal, heavily restricted the use and possibility to reimburse drugs used “off-label”) evolved and in December 2014, the French Government adopted a decree making it easier to use a drug “off-label” even when an authorized alternative was available. [2] As a result, the Agence nationale de sécurité du médicament et des produits de santé (“ANSM”) issued a “Temporary Recommendation for Use” for the use of Avastin in the treatment for AMD.

The FCA decision

In 2014, the Italian Competition Authority fined Novartis and Roche under Article 101(1) TFEU for anti-competitively agreeing to spread misleading information on the use of Avastin in ophthalmology in order to encourage the use of Lucentis for AMD treatment. Shortly thereafter, the FCA initiated ex officio proceedings. Unlike its Italian counterpart, the FCA did not rely on Article 101(1) TFEU and instead alleged that Novartis, Roche and Genentech engaged into abuses of collective dominance infringing Article 102 TFUE and Article L. 420-2 of the French Commercial Code. On September 9, 2020, further to a five year-investigation, the FCA fined the three companies a total amount of €444 million for abusing their collective dominance between March/April 2008 and November 2013.

The alleged collective dominant position of Novartis, Roche and Genentech

The FCA found that the three companies, together, held a collective dominant position on the market for the treatment of AMD. The FCA considered that Novartis, Roche and Genentech constituted a single entity because of Novartis’ shareholding in Roche, Roche’s shareholding in Genentech, and the contractual links between Genentech and Roche, and Genentech and Novartis respectively, through the licencing agreements. The FCA considered that this single entity held a dominant position on the market for the treatment of AMD where the parties’ combined market shares exceeded 90%, until Bayer’s entry in 2013.

The alleged practices

The FCA found that (i) Novartis had allegedly disparaged Avastin before patient associations, health professionals, and the general public; and (ii) the three companies had allegedly implemented blocking tactics to delay the public authorities’ initiatives to regulate the use of Avastin in the treatment of AMD.

The first infringement. The FCA held that pharmaceutical companies have the duty to communicate in an objective, comprehensive and reliable way to doctors, public authorities, and the public in general.[3] The FCA considered, in this case, that Novartis had disseminated selective and biased data comparing Avastin and Lucentis, unduly insisting on the risks related to the use of Avastin for AMD treatment.

The FCA found that these practices reduced the “off-label” use of Avastin in ophthalmology, which in turn unduly preserved Novartis’ quasi-monopolistic position on the market and helped sustain Lucentis’ high price. As a side effect, according to the FCA, the practices unduly increased the pricing of Bayer’s drug Eylea, since Eylea’s price was set taking into account Lucentis’ (but not Avastin’s) price.

Novartis was fined €253.9 million for the first infringement.

The second infringement. The FCA found that Roche and Novartis, aided by Genentech, had intervened in an abusive way before the French public authorities. The practice allegedly aimed at delaying the public authorities’ initiatives to promote the use of Avastin for AMD treatment.

The FCA relied on (i) Roche’s initial temporary refusal to supply samples of Avastin to the Agence française de sécurité sanitaire et des produits de santé (“AFSSAPS”)[4] for the purpose of a stability assessment and (ii) Roche’s communications to the AFSSAPS, at the latter’s request. It found that Roche had unduly emphasized the side effects of Avastin’s use in ophthalmology. As for Novartis, the FCA found that it had spread a worrying discourse among public authorities based on incomplete studies presented out of context. According to the FCA, these practices had a negative impact on the evolution of health care spending, and also contributed to preventing Avastin from being used in comparative tests that could have enabled the Comité économique des produits de santé (“CEPS”) to substantially lower the price of Lucentis.[5]

Although Genentech did not directly intervene in Novartis’ and Roche’s interactions with French authorities, the FCA considered that it had helped the two other companies coordinate their message so that it remained consistent and based on the same scientific data. The FCA notably supported its conclusion by citing a number of e-mails exchanged between Genentech and the two other laboratories.

Novartis was fined €131.2 million for the second objection, while Roche and Genentech were jointly ordered to pay a €59.7 million fine.


The FCA decision is one of the rare cases, to date, where the FCA found a collective dominant position. It raises a number of critical questions about the definition of a collective dominant position in the context of a contractual licensing relationship, with one of the players involved holding, on its own, a quasi-monopolistic market position (Novartis). The nature of the abuse in such circumstances may also be discussed. The decision, moreover, sets a particularly low standard of proof to establish the abuse, as it almost entirely relies on one of the parties’ behaviour (Novartis) to demonstrate a collective scheme of practices. Finally, the decision also raises serious questions regarding the FCA’s jurisdiction to interpret scientific studies and assess the legitimacy of laboratories’ concerns over the use of medicines. Novartis has publicly announced its intention to appeal the decision.

[1]              The so-called “anti-VEGF” principle, which aims at blocking the “VEGF” (“vascular endothelial growth factor A”) protein, which contributes to the development of blood vessels and may contribute to the vascularization of cancerous tumors.

[2]              Decree No. 2014-1703 of 30 December 2014.

[3]              See Decision 20-D-11 regarding practices implemented in the treatment of Age-related macular degeneration (AMD) sector, paras. 764-767.

[4]              AFSSAPS is the former name of ANSM.

[5]              The CEPS set the price of medicines reimbursed by the French social security system by looking at the price of other medicines used for the same treatment (also called “comparators”). To qualify as a comparator, a medicine must benefit from certain authorizations or recommendations of use, or be frequently prescribed “off-label”. Therefore, in this case, as Avastin was only granted a Temporary Recommendation for Use in 2014, the CEPS did not take its price into account when setting the price for Lucentis.