First published on Informa Connect Law’s CompLawBlog, April 2020

‘Pay-for-delay’ agreements

Following the European Commission’s Pharmaceutical Sector Inquiry, ‘pay-for-delay’ agreements (when a generic company delays independent entry onto the market in return for a transfer of value from the originator company) have continued to be heavily scrutinised by the competition authorities.

Although the CJEU is yet to deliver a decision in either Lundbeck or Servier (both appeals from Commission decisions and subsequently from General Court judgments), on 30 January 2020 the CJEU did hand down its judgment in Case C-307/18, Generics (UK) Limited and Others v Competition and Markets Authority (the ‘Judgment’).  The Judgment, which responded to a preliminary reference from the UK’s Competition Appeal Tribunal, ‘CAT’, was particularly significant as it was the first time that Europe’s highest court has provided guidance on the competition law assessment of  ‘pay-for-delay’ agreements.

Given the breadth of issues considered by the CJEU, this article focusses on the CJEU’s findings in relation to Article 101 TFEU and seeks to highlight some of the most interesting aspects of the Judgment on this issue.

A brief introduction to the paroxetine proceedings

The case related to GSK’s paroxetine product (marketed under the brand name Seroxat®) and  concerned a series of agreements between GlaxoSmithKline (‘GSK’) and a number of generic companies (namely, IVAX; Generics (UK) (‘GUK’); and Alpharma).  Before the expiry of the basic compound patent in January 1999 and of data exclusivity in December 2000, GSK had obtained a number of other patents relating to paroxetine including the Anhydrate patent (which claimed four paroxetine polymorphs and the process for making them).  By mid-2000, GSK became aware that a number of generic companies were contemplating entry onto the UK market.

GSK entered into agreements with IVAX, GUK and Alpharma.  For the purpose of this article it is sufficient to note that these agreements sought to settle disputes relating to, inter alia, the Anhydrate patent and, broadly speaking, provided that:

• the generic manufacturers would be supplied with a limited volume of paroxetine manufactured by GSK but were otherwise prevented from making, importing or supplying generic paroxetine; and
• GSK would make a number of ‘value transfers’ to the generic companies including, for example, contributions towards their legal costs, provision of a ‘promotional allowance’ or ‘marketing allowance’ and/or a payment in respect of the ‘production and preparation cost of launch in the UK market’ (see paragraphs 10 to 14 of the Judgment for more detail).

Notwithstanding the agreements entered into between GSK, IVAX, GUK and Alpharma, the Anhydrate patent was also the subject of litigation between GSK and other generic companies.  For example, in July 2002, the Anhydrate patent was partially invalidated in proceedings brought by BASF (leaving only two process claims in force).  In December 2003, another generic company (Apotex) successfully defended an infringement claim brought by GSK and entered the market.  IVAX, GUK and Alpharma then terminated their respective agreements with GSK and entered the market independently and without restraint.

The Competition and Markets Authority (‘CMA’) found that GSK, GUK and Alpharma had infringed the prohibition in Chapter I of the Competition Act 1998 by entering into the GSK/GUK and GSK/Alpharma agreements and that, by entering into a series of agreements, GSK had also abused its dominant position in the market for paroxetine, infringing Chapter II of the Competition Act 1998.  The CMA imposed penalties of more than £44 million.  This decision was appealed to the CAT which made a preliminary reference to the CJEU asking ten questions about the correct interpretation of both Articles 101 and 102 TFEU.  This article focuses on the Article 101 TFEU aspects of the judgment.

The CJEU’s Judgment – key points on Article 101 TFEU

The first two questions referred to the CJEU went to the issue of potential competition and, in effect, asked whether an originator and a generic can be regarded as potential competitors in circumstances where there is a genuine patent dispute between them.

The CJEU responded that a generic company will be a ‘potential competitor’ of an originator where there are ‘real and concrete possibilities’ for the generic to enter the market and compete with the originator.  According to the CJEU (see Ruling 1), when making this assessment, it is necessary to consider:

1. whether the generic has a ‘firm intention’ and an ‘inherent ability’ to enter the market and has taken ‘sufficient preparatory steps’ (including obtaining an MA, having an adequate stock of the generic product, taking legal steps to challenge the process patents held by the originator and adopting a range of marketing initiatives); and
2. whether there are any ‘insurmountable barriers’ to market entry

The CJEU gave very little weight to the existence of patents.  For example, while the CJEU noted that patents may form part of the economic and legal context of an agreement,  it did not consider that the presumption of patent validity, the uncertain outcome of disputes concerning validity or the existence of an interim injunction would prevent the originator and a generic from being ‘potential competitors’.  Moreover, the CJEU expressly concluded that the existence of a process patent for an active ingredient that is already in the public domain is not an ‘insurmountable barrier’ to market entry.  In this regard, it is worth noting that the Judgment addresses the issue of potential competition with express reference to the existence of process patents.  It is not yet clear whether the same analysis could be applied to a patent of a different type.

 The third, fourth and fifth questions referred by the CAT asked the CJEU to clarify the circumstances in which settlement agreements would constitute a restriction of competition ‘by object’.  By way of reminder, to be a restriction of competition ‘by object’ an agreement must reveal ‘in itself’ a ‘sufficient degree of harm’ to competition.  In this case, the CJEU held that a settlement agreement is a restriction of competition ‘by object’ where (see Ruling 2):

1. a generic manufacturer undertakes not to enter the market and not to continue an invalidity challenge in exchange for a transfer (or transfers) of value from an originator company; and
2. it is clear that the net gain from the transfers of value can be explained only by the commercial interests of the parties to the agreement to avoid competition on the merits; unless
3. the settlement agreement is accompanied by ‘sufficiently significant’ pro-competitive effects that are ‘specifically related to the agreements in issue’ and are capable of giving rise to a reasonable doubt that the agreement causes a sufficient degree of harm to competition – ‘minimal’ and ‘uncertain’ benefits (such as a slight reduction in price) will be insufficient to escape a ‘by object’ categorisation.

Given that the Judgment responds to a preliminary reference, the CJEU does not expressly consider each value transfer in issue in the underlying paroxetine proceedings. However, the CJEU has clarified that all transfers of value (pecuniary and non-pecuniary; direct and indirect) should be taken into account.  The CJEU has also made clear that a transfer of value may be justified, and not lead to the conclusion that there is a ‘by object’ restriction of competition, if it is appropriate and strictly necessary in the light of the legitimate objectives of the agreement.  The size of the value transfer is relevant and the net gain to the generic manufacturer must be ‘sufficiently large’ or ‘sufficiently beneficial’ to induce the generic manufacturer to delay entry.  It will be interesting to see how the CAT applies this guidance to the agreements in issue.

The sixth question referred to the CJEU asks specifically about the relevance of two factors (namely the likelihood of success in patent proceedings and the likelihood of the parties having entered into a less restrictive agreement) to the assessment of a ‘by effect’ infringement (which will be relevant in circumstances where the pro-competitive aspects of the agreement are sufficient to raise a reasonable degree of doubt that the agreement will cause a sufficient degree of harm to competition to be regarded as an infringement ‘by object’).   In answering this question, the CJEU has confirmed that it is not necessary for the referring court to establish that (in the absence of the agreement) the generic manufacturer would probably have been successful in patent litigation or that the parties would probably have concluded a less restrictive settlement agreement (see Ruling 3).

What next?

While the Judgment provides some helpful insight about the approach the CJEU is likely to adopt when analysing so-called ‘pay-for-delay’ agreements, it is inevitably focussed on the questions referred by the CAT.  Accordingly, the Judgment does not address all of the competition law issues that arise in relation to ‘pay-for-delay’ agreements.  For example, although the Judgment is fairly expansive in relation to ‘by object’ restrictions, comparatively little is said on the subject of ‘by effect’ restrictions (arguably due to the narrow nature of the question referred by the CAT).

In addition, given that the Judgment has been given in the context of a preliminary reference, it will be for the CAT to apply the guidance given by the CJEU to the facts of the case.  The conclusions reached by the CJEU in the context of the Lundbeck and Servier appeals from the General Court will be particularly interesting to the extent that they explore outstanding questions in more detail.