European Court of Justice Paroxetine Ruling Provides Guidance on Patent Settlement Agreements

Jan Blockx (University of Antwerp)1


Twelve years ago, on January 15, 2008, the European Commission used its power to conduct unannounced inspections in a sector inquiry for the first time. The pharmaceutical sector inquiry it launched that day was a response to perceived delays in the entry of generics into pharmaceutical markets. The main focus of the inquiry was patent settlement agreements, i.e. agreements whereby the originator of a drug and a manufacturer of generics settle litigation over the validity of the patent protection for a pharmaceutical product and/or the infringement of said patent by the generics manufacturer. The Commission was particularly concerned over agreements in which the originator committed to a significant payment (or other value transfer) to the generics manufacturer and the latter agreed in return to delay its entry into the market (pay-for-delay).

Following the sector inquiry, the Commission has found that several such agreements constitute anticompetitive agreements (and in one case an abuse of dominance), that are contrary to EU competition law.2 The General Court of the European Union has had the opportunity to scrutinize two of these Commission decisions,3 but so far, the Court of Justice has not yet ruled on such cases. This first ruling by the supreme court of the EU came on January 29, 2020, not in an appeal against a General Court judgment (these cases are still pending), but in a preliminary ruling related to a decision by the UK Competition and Markets Authority finding that a patent settlement agreement relating to the antidepressant paroxetine infringed competition law.4

In the preliminary ruling procedure, the Court of Justice merely responds to legal questions referred to it by the national court. The Court of Justice does not adjudicate on the facts of the case and will rely on the description of the facts given by the national court to the extent that this is relevant in responding to the legal questions. After the Court of Justice’s ruling, the national court will have to decide on the merits of the case in light of the responses provided by the Court of Justice. The preliminary ruling procedure therefore allows the Court of Justice to clarify legal issues in a fairly abstract manner, thereby highlighting general principles which will also be relevant in other cases. Both in competition law cases and in other fields of European law, judgments in preliminary ruling procedures have therefore generally been more influential than other types of judgments of the Court of Justice.

The Court of Justice usually requests an opinion by one of its Advocates-General before issuing a judgment. It also did so in this case and Advocate-General Kokott issued her opinion on January 22, 2020.5 The fact that the judgment was issued a mere week after the opinion suggests that the views of the Court are very much aligned with those of the Advocate-General and the opinion can therefore provide additional elucidation of the views of the Court.

I will focus this brief comment on the Paroxetine judgment on three issues: (i) the notion of potential competition; (ii) when patent settlement agreements constitute restrictions of competition by object; and (iii) when a patent settlement strategy can constitute an abuse of dominance. The Court also ruled on a number of other points, but these concern facts which have less relevance in other cases, so I will leave these aside here.

Mere Existence of Process Patents Does Not Preclude Ootential Competition

A first set of questions of the Competition Appeal Tribunal, the referring UK court, concerns whether generic manufacturers can be viewed as (at least potential) competitors of originators if the latter are holders of a patent for the pharmaceutical drug which the generics want to produce. A specific feature in the case before the national court was that the originator did no longer hold a patent on the active ingredient of paroxetine, but merely a secondary patent for the process to produce it. The companies in this case nevertheless argued that, because of the process patent, the generics were estopped from producing the drug in question and therefore could not be viewed as potential competitors.

The Court did not accept this reasoning. It held instead that the existence of a process patent “cannot, as such, be regarded as an insurmountable barrier” to entry by the generics.6 Following the reasoning of Advocate-General Kokott, the court pointed out that a generic could enter the market notwithstanding the patent, if either (i) the patent would be invalidated; or (ii) it could enter the market without infringing the patent (by using a different production process). As regards the first point, the Court held that national competition authorities and courts were not meant to assess the validity of the patent, but that “uncertainty as to the validity of patents covering medicines is a fundamental characteristic of the pharmaceutical sector.”7 The European regulatory framework indeed provides that a marketing authorization for a generic drug can be obtained without the need to demonstrate the invalidity of the patent. The legal presumption of validity of a patent did not change this conclusion, according to the Court.

Since the process patent is not an insurmountable barrier to entry, the Court considered that the question of whether the generic is a potential competitor of the originator depends merely on whether the generic has “a firm intention and an inherent ability” to enter the market.8 That depends on whether the generic has taken preparatory steps to enter (such as building up stock and requesting marketing authorizations) but could also be confirmed by additional factors such as the “perception” of the originator, which could be attested by its willingness to pay the generics to postpone its entry into the market, in other words by the content of the patent settlement agreement itself.9


Value Transfer as Yardstick for Problematic Patent Settlements

The next questions of the referring court concerned whether and when a patent settlement agreement constitutes a restriction of competition “by object” in the sense of Article 101 TFUE, where it is not necessary to assess the effects of the practice.

The Court stated that “challenges to the validity and scope of a patent are part of normal competition in the sectors where there exist exclusive rights in relation to technology.”10 That does not mean, however, that patent settlement agreements are always anticompetitive. However, according to the Court, they can be restrictive by object if (i) they are a sham, i.e. “designed with the sole aim of disguising a market-sharing agreement or market-exclusion agreement”11 or if (ii), although there is a genuine dispute underlying the settlement, the settlement agreement provides for a transfer of value which “cannot have any explanation other than the commercial interest of both the holder of the patent and the party allegedly infringing the patent not to engage in competition on the merits.”12 To assess the transfer of value not merely monetary payments should be taken into account but also indirect transfers such as, for example, the profits to be obtained from a distribution contract. Patent settlement agreements which merely compensate for the costs of or disruption caused by the litigation, or which remunerate for the supply of goods or services, or even discharge cross-undertakings in damages which may have been given in proceedings, are not problematic. The fact that the transfer of value was inferior to the profits the generics could have obtained if it had been successful in litigation was not considered relevant for the Court, nor the fact that the patent settlement did not exceed the scope or remaining period of validity of the patent.


A Strategy to Exclude Through Patent Settlements Can Amount to an Abuse of Dominance

A final set of questions from the Competition Appeal Tribunal concerned the question of whether, in addition to an anticompetitive agreement or concerted practice in the sense of Article 101 TFEU, a dominant undertaking could also commit an abuse of dominance prohibited by Article 102 TFEU through pay-for-delay agreements. While the Court long ago confirmed that anticompetitive agreements may at the same time constitute an abuse of dominance,13 the issue at hand was whether the overall strategy of concluding pay-for-delay agreements (rather than the individual agreements by themselves) could constitute an abuse.

The Court held that this was the case because the anticompetitive effects of such a contract-oriented strategy could exceed the anticompetitive effects inherent in each of the agreements. As a consequence, the fact that one of the agreements was exempted from the prohibition on anticompetitive agreements in this case, did not prevent that agreement from being taken into account to establish an abuse in the overall strategy. The Court furthermore qualified its earlier case law, according to which the anticompetitive intent of the dominant undertaking can be a relevant factor to determine whether there is an abuse,14 by saying that such anticompetitive intent “must be taken into consideration.”15 On the other hand, according to the court, any objective justifications such as efficiency gains invoked by the dominant undertaking must be assessed objectively, and cannot be excluded because they are merely accidental.


Through this first ruling, the Court of Justice provides important guidance for the antitrust assessment of patent settlement agreements. While the judgments in the pending appeals in two other patent settlement cases provide opportunities for further clarification, a number of important points can already be noted.

Firstly, generics manufacturers that are actively preparing entry into a pharmaceutical market for which the patent of the active ingredient has expired, are potential competitors of the originator of the drug. Since patent settlements which have been found problematic in the past mostly concern process patents, this is clearly an important finding. Part of the reasoning of the Court to support this view is transposable to situations where the originator does not merely hold a process patent but also the patent for the active ingredient itself, though the Court clearly indicated that its ruling was limited to settlements regarding process patents.

Secondly, by clarifying that some patent settlements are problematic from an antitrust perspective, the ruling may make it easier for originators to rebuff frivolous claims by generics manufacturers who merely try to get a piece of the pie in a market. Since the ruling makes it more difficult to settle such claims, generics manufacturers that are not serious about entering will be discouraged from pursuing a settlement.

Finally, while some of the arguments on abuse of dominance are related to facts which are specific to this case, the Court’s insistence on the possibility that a certain strategy constitutes an abuse, has much wider implications. This view builds on observations the Court made in other rulings, including the 2017 Grand Chamber Intel judgment,16 and suggests that the practices of dominant undertakings cannot be assessed in isolation, but need to be considered in their totality.

Click here for a PDF version of the article

1 Visiting lecturer (part time) and researcher at the University of Antwerp, Belgium.

2 Commission decision of June 19, 2013 in Case 39.226 Lundbeck; Commission decision of December 10, 2013 in Case 39.685 Fentanyl; Commission decision of July 9, 2014 in Case 39.612 Perindopril (Servier). One investigation is still ongoing: Case 39.686 Cephalon.

3 See the judgments of September 8, 2016 in the Cases T-460/13 Sun Pharmaceutical Industries and Ranbaxy (UK) v. Commission ECLI:EU:T:2016:453, T-467/13 Arrow Group and Arrow Generics v. Commission ECLI:EU:T:2016:450, T-469/13 Generics (UK) v. Commission ECLI:EU:T:2016:454, T-470/13 Merck v. Commission ECLI:EU:T:2016:452, T-471/13 Xellia Pharmaceuticals and Alpharma v. Commission ECLI:EU:T:2016:460 and T-472/13 Lundbeck v. Commission ECLI:EU:T:2016:449, and the judgments of December 12, 2018 in the Cases T-677/14 Biogaran v. Commission ECLI:EU:T:2018:910, T-679/14 Teva UK and Others v. Commission ECLI:EU:T:2018:919, T-680/14 Lupin v. Commission ECLI:EU:T:2018:908, T-682/14 Mylan Laboratories and Mylan v. Commission ECLI:EU:T:2018:907, T-684/14 Krka v. Commission ECLI:EU:T:2018:918, T-691/14 Servier and Others v. Commission ECLI:EU:T:2018:922, T-701/14 Niche Generics v. Commission ECLI:EU:T:2018:921 and T-705/14 Unichem Laboratories v. Commission ECLI:EU:T:2018:915.

4 Case C-307/18 Generics (UK) and Others ECLI:EU:C:2020:52.

5 Opinion of AG Kokott in Case C-307/18 Generics (UK) and Others ECLI:EU:C:2020:28.

6 Case C-307/18 Generics (UK) and Others ECLI:EU:C:2020:52, para 46.

7 Case C-307/18 Generics (UK) and Others ECLI:EU:C:2020:52, para 51.

8 Case C-307/18 Generics (UK) and Others ECLI:EU:C:2020:52, para 54.

9 Case C-307/18 Generics (UK) and Others ECLI:EU:C:2020:52, paras 56-57.

10 Case C-307/18 Generics (UK) and Others ECLI:EU:C:2020:52, para 81.

11 Case C-307/18 Generics (UK) and Others ECLI:EU:C:2020:52, para 76.

12 Case C-307/18 Generics (UK) and Others ECLI:EU:C:2020:52, para 87.

13 See already Case C-85/76 Hoffmann-La Roche ECLI:EU:C:1979:36, para 116.

14 C-549/10 P Tomra and Others v. Commission ECLI:EU:C:2012:221, para 20. Note, however, that the Court already seems to go further in para 19 of the Tomra judgment.

15 Case C-307/18 Generics (UK) and Others ECLI:EU:C:2020:52, para 164 (emphasis added).

16 Case C-413/14 P Intel v. Commission ECLI:EU:C:2017:632.