Patents are crucial for pharmaceutical companies. The idea behind a patent is to give an innovative drug manufacturer the opportunity to recoup its investments. When a legitimately obtained patent of an innovative producer expires, so-called ‘generic producers’ can market a drug with the same active ingredient. This will increase competition and (significantly) reduce the price of the drug.
When a patent is challenged, that is frequently grounds for innovative producers to enter into settlement agreements with generic producers. Settlement agreements can prevent pharmaceutical companies from becoming embroiled in years-long disputes about a patent’s validity. But settlement agreements may be in breach of the cartel prohibition and may give rise to high fines for all the companies involved.
Lundbeck: pay-for-delay
Danish pharmaceutical company Lundbeck also entered into settlement agreements with four generic producers. It developed the active ingredient citalopram, which formed the basis for the Cipramil antidepressant. Lundbeck obtained the exclusive right to exploit citalopram and also held a so-called ‘process patent’. A process patent gives a pharmaceutical the exclusive right to produce an active ingredient in a specific manner. From 2002, generic manufacturers were allowed to market drugs containing the active ingredient citalopram, as long as they did not violate Lundbeck’s process patent in doing so.
When Lundbeck’s patent on the active ingredient citalopram expired, the competition presented itself. Lundbeck threatened to take legal action against generic producers for infringement of its (still valid) process patent. Lundbeck then entered into settlement agreements with Merck, Arrow, Alpharma and Ranbaxy (generic producers). The generic producers undertook not to market their version of Cipramil. In return, Lundbeck would pay an amount to the generic producers. This is referred to as a pay-for-delay agreement (see here and here for earlier blogs).
The Danish competition authority informed the European Commission (the Commission), which subsequently launched an investigation and imposed hefty fines on Lundbeck and the four generic producers. Lundbeck and the generic producers challenged the cartel fines before the General Court of the European Union (the General Court). In 2016, the General Court dismissed the appeal and upheld the fines. Lundbeck and the generic producers appealed to the Court of Justice of the European Union (the Court of Justice). The Court of Justice mainly considered two questions that were relevant to the application of the cartel prohibition: (i) are Lundbeck and the generic producers competitors (or potential competitors) of each other; and (ii) did the settlement agreements have the restriction of competition as their object (and not merely as their effect)?
Potential competitor or not?
To determine whether the generic producers are potential competitors of Lundbeck, it must be assessed whether there are ‘realistic and concrete possibilities’ for the generic producers to enter the relevant market. The Court of Justice referred to its earlier case law and found that a generic producer is a potential competitor of Lundbeck if it has the ‘firm intention and inherent ability to enter the market’ and does not meet ‘barriers to entry that are insurmountable’. The Court of Justice went on to say that the Commission need not first demonstrate with certainty (i) that the generic producer will in fact enter the market concerned; and (ii) that it will be capable of retaining its place there. What matters is that there are real and concrete possibilities of the generic producer entering the market. That was the case here, according to the Court of Justice, which found that the four generic producers had indeed taken preparatory steps, such as negotiating with suppliers and obtaining marketing authorisation for citalopram.
The Court of Justice also referred to its earlier Generics ruling, which held that the existence of process patents cannot as such be regarded as an insurmountable barrier to entry for a generic producer. The generic producer can also try to produce citalopram by means of a process other than Lundbeck’s patented process. A generic producer can furthermore challenge the validity of a patent. In short, according to the Court of Justice, the General Court correctly found that Lundbeck and the generic producers were at least potential competitors of each other.
Pay-for-delay agreement: restriction of competition by object?
The Court of Justice then turned to the question whether the settlement agreements had an anticompetitive object. Restrictions by object have the restriction of competition as their object. These agreements are deemed to have such a negative effect on competition (the price, quantity and quality of goods or services) that it is unnecessary first to demonstrate that the agreements had concrete negative effects before the cartel prohibition is deemed to have been violated. Lundbeck disputed that the settlement agreements had an anti-competitive object. It believed that the restriction of the agreements fell within the scope of protection of its patents. Moreover, a settlement agreement is a legitimate and customary way of avoiding patent disputes: legitimate in the sense that Lundbeck must be able to defend itself against wrongful infringement of its patents.
The Court of Justice found that the concept of ‘restriction of competition by object’ must be interpreted restrictively. It added that if a certain ‘transfer of value’ from the innovative producer to the generic producer can have no other explanation than the commercial interest of both parties not to engage in competition on the merits, that constitutes a restriction of competition by object. A transfer of value need not relate to money, but may also relate to licences, for instance. In the present case, the Court of Justice found that it was clear that the settlement agreements prevented the generic manufacturers from entering the market, not Lundbeck’s process patent.
Pay-for-delay agreements: tension between competition law and IP law
Back in 2020 already, the Court of Justice ruled on the question whether the settlement agreements that pharmaceutical company Generics entered into with generic producers regarding the production of paroxetine violated the cartel prohibition. In its Lundbeck judgment, the Court of Justice confirmed the approach taken in Generics (2020).
Enforcement action against pay-for-delay agreements presents an area of tension between competition law and intellectual property law (see here and here). Competition law aims to protect competition. Patents offer an innovative producer a legitimate monopoly during a certain period of time. At the same time, competition law also continues to apply to the holder(s) of a valid patent. If a settlement agreement does not go beyond the protection offered by a patent, unlawful restriction of competition is unlikely to be assumed, since the protection offered by a patent is legitimate in and of itself.
According to Lundbeck, its process patent prevented the generic producers from entering the market, not the amount that Lundbeck paid them under the settlement agreement. The Court of Justice disagreed. Potential competition as described by the Court of Justice will usually arise in the pharmaceutical sector, particularly in the case of the ‘blockbusters’, long before the expiry of a patent. Particularly in the case of a blockbuster, generic producers have the incentive to ensure that they can also enter the market as soon as possible. The rationale behind a pay-for-delay agreement is crucial in assessing whether the agreement restricts competition, because the fact that generic producers had already taken preparatory steps confirmed that the process patent apparently did not prevent them from wanting to enter the market.
Patents may be strong, weak or anything in between. The stronger the patent is, the less likely it is that a generic manufacturer will be able to successfully challenge the patent. In this case, evidence showed that Lundbeck’s process patent was not particularly strong. The weakness of the process patent confirmed that the generic producers did not refuse to enter the market because of Lundbeck’s process patent. According to the Court of Justice, process patents can never constitute ‘insurmountable barriers to entry’, because process patents can always be challenged. Also, generic producers can develop alternative processes to produce the drug in a different manner. The Court of Justice thereby provided a clear framework for how competition authorities handle the assessment of pay-for-delay agreements in the case of process patents.
So has the matter now been clarified? No, several interesting pay-for-delay cases are still pending before the General Court and the Court of Justice. One such case concerns the pharmaceutical companies Teva and Cephalon. The Commission fined them in 2020 for entering into a pay-for-delay agreement (see also our blog here). They appealed that fine. In the pay-for-delay agreement entered into by Teva and Cephalon, the transfer of money played a smaller role than in the Lundbeck case, because not only the transfer of money but primarily the entry into a commercial agreement formed part of the pay-for-delay agreement. The Court of Justice must therefore address the question of whether the cartel prohibition has been violated also in those circumstances. The Court of Justice’s earlier judgments in Generics and Lundbeck appear to allow such a ruling. The General Court’s judgment will be a test as to whether a (consistent) line can be discerned (and, if so, which) on the part of the General Court and the ECJ in answering the question what form the value transfer may take to constitute a violation of the cartel prohibition.
Servier case
The Commission furthermore fined Servier and five generic producers in 2014. These fines are currently being challenged before the Court of Justice. The Commission believes that Servier not only violated the cartel prohibition, but also abused its dominant position (Article 102 TFEU) – the latter for pursuing a strategy to exclude generic producers such as Krka from market entry. The General Court did not fully subscribe to the Commission’s decision and ruled that the Commission had not demonstrated that the agreement entered into between Servier and Krka restricted competition. The General Court also found that abuse of a dominant position could not be established. On 14 July 2022, Advocate General Kokott found – unlike the General Court – that the settlement agreement between Servier and Krka did have an anti-competitive object. The Court of Justice is currently considering the Servier case.
It remains to be seen whether the Court of Justice will find that Servier abused a dominant position. It seems only logical that a pharmaceutical company can abuse its dominant position only if the agreements entered into with generic producers also violate competition law (in this case the cartel prohibition). In our view, the entry into a legitimate settlement agreement should not lead to a finding of a prohibited exclusionary strategy that constitutes abuse of a dominant position (Article 102 TFEU). That could result in drug manufacturers being deprived of a legitimate means to settle proceedings. And that would be difficult to reconcile with the fact that even the Commission found, after extensive investigation, that many settlement agreements (around 89%) proved to be unproblematic.
In short, the Lundbeck judgment answers the question when settlement agreements are or are not compatible with the cartel prohibition. The Court of Justice’s judgment in Servier/Commission will have to be awaited for a ruling on abuse of a dominant position in that context, which will most likely be issued in the near future.
Diederik Schrijvershof and Leah Peeters wrote the article ‘Court of Justice allows pay-for-delay agreements’ for the Jurisprudentie Geneesmiddelenrecht journal. That article addresses the Court of Justice’s judgment in the Lundbeck case. This blog is an abridged version of the annotation.
Follow Maverick Advocaten on Twitter and LinkedIn
