On 25 March 2021, the ECJ ended the Lundbeck pay-for-delay saga by dismissing the appeals from Lundbeck and five generic manufacturers against a European Commission ‘pay-for-delay’ decision. Following its recent Paroxetine judgment, the ECJ found that Lundbeck’s process patents did not preclude generic companies being viewed as potential competitors, particularly since the patents did not represent an insurmountable barrier to entry. In addition, the patent settlement agreements constituted infringements "by object".
A determining factor in this context was whether the net gain of the transfer value from the originator company to the generic manufacturer is sufficient to disincentivise the generic company from entering the market.
These rulings provide pharma companies with useful guidance on the antitrust assessment of their settlement agreements. However, ongoing antitrust investigations and soon-to-be-tightened merger control show that the overall pharma saga is far from over yet.
On 8 September 2016, the General Court (“GC”) dismissed in their entirety the appeals brought by Lundbeck and the generic companies Alpharma, Merck KGa, Generics UK, Arrow and Ranbaxy against the European Commission's ‘pay-for-delay’ decision (see our October 2016 newsletter). The European Court of Justice (“ECJ”) confirmed the findings of the GC and handed down separate judgments on each of the six appeals. This article focuses on the judgment of Lundbeck’s appeal, as the ECJ repeated most of the same reasoning in the appeals concerning the generic companies. These judgments largely follow the ECJ’s approach in its Paroxetine judgment (see our February 2020 newsletter)
First, the ECJ confirmed that the GC was correct to find that Lundbeck and the generic companies were potential competitors when signing the patent settlement agreements. The GC was entitled to conclude that the existence of a process patent did not preclude generic products being part of the same relevant market as the originator product Citalopram (an antidepressant medicine). Indeed, the generic companies had “real and concrete possibilities” (absent the agreement) to enter the market at the time that the agreements were concluded. The preparatory steps taken by the generic companies, such as obtaining or applying for market authorisation, demonstrated this possibility. The ECJ confirmed that there is no need for the Commission to demonstrate with certainty that the generics manufacturers would have (successfully) entered the market, or to review the strength of the patent at issue.
The ECJ also agreed with the GC that the settlement agreements constitute a restriction of competition ‘by object’. The concept of restriction ‘by object’ must be interpreted strictly. The test is whether the transfer of value is shown to be sufficiently beneficial to encourage the generic manufacturer to refrain from entering the market concerned and from competing on the merits with the originator. In this case, the value transfers broadly corresponded to the profits that the generic companies could have made when entering the market or to the damages they would have obtained if they had successfully challenged Lundbeck's patents. As a result, the transfers were of a high enough value to remove the generic manufacturers’ incentive to enter the market and thus eliminated the competitive pressure.
Given that Lundbeck also had not mentioned any pro-competitive effect associated with the agreements in its appeal to rebut this finding, the ECJ agreed with the GC that, in this case, the value transfer constituted a restriction of competition by object.
Moreover, the ECJ rejected Lundbeck’s claims that the Commission should have applied the "scope of the patent" test and taken into account that the contractual restrictions did not exceed the scope of Lundbeck's process patents. The ECJ found that even if the restrictions imposed through the settlement agreements potentially fell within the scope of Lundbeck's patents, these restrictions went beyond the protection of its IP rights.
Finally, the ECJ confirmed that the GC was right to reject arguments relating to the alleged legitimate objectives and efficiencies brought by the settlements and errors in the calculation of the fines.
The classification by the ECJ of the settlement agreements as restriction of competition ‘by object’ comes as no surprise after its February 2020 Paroxetine judgment. The ECJ judgment provides the following key takeaways:
- settlement agreements offering significant value transfers from the originator to the generic manufacturer, with the intention that the latter will refrain from entering the market, are likely to be considered as ‘by object’ restrictions;
- pro-competitive effects associated with the settlement agreements may rebut their characterisation as a ‘by object’ restriction, provided these effects are demonstrated, relevant, sufficiently significant and specifically related to the agreements;
- and in order to assess whether a company is a potential competitor, the test is whether there are "real and concrete possibilities" of it joining the market and competing with the companies present in it.
Even though the Lundbeck pay-for-delay saga has now been put to bed, the pharma quest seems far from over. The appeal by drug manufacturer and originator company Servier against the GC’s ruling on its patent settlement agreements is pending (see our January 2019 newsletter) and various antitrust investigations into pharma companies are ongoing (see for instance the recent fine and ongoing investigation into Teva). In addition, the European Commission has recently joined a Multilateral Working Group to analyse the effects of global pharma mergers, and has published Guidance on the application of ‘Article 22’ referrals to catch ‘killer acquisitions’ in sectors such as the pharmaceutical industry (see our October 2020 newsletter). ‘Forewarned is forearmed’ may thus be a fitting expression for companies active in the pharma sector.
This article was published in the Competition Newsletter of April 2021. Other articles in this newsletter: