This week, the EU General Court partially overturned the EU Commission’s decisions in Perindopril (Servier v Commission and Krka v Commission).
The judgment was handed down pretty much on the tenth anniversary of the original dawn raids in November 2008. The raids came as a follow-up from the European Commission’s pharma sector enquiry and led to a number of infringement decisions that have also found their way up to the General Court and to the Court of Justice of the EU. The key theory which the EU Commission advanced in the cohort of these so-called ‘pay-for-delay’ cases is, very broadly, that EU competition law can intervene in patent settlement cases in certain circumstances (both under the rules on abuse of dominance and restrictive agreements). These circumstances are (again, very broadly) where (i) the settlement proposal restricts entry by an actual or potential generic competitor—the delay element, and (ii) where the originator company makes a value transfer to the potential generic entrant – the payment element. This could be by way of a lump-sum payment or through some other way (e.g., through a beneficial distribution agreement). On abuse of dominance, the theory is that unilateral conduct aimed at “shutting out a competing technology and buying out a number of competitors” constitutes an abuse. These theories are now being tested in the European Courts.
In parallel, the UK’s Competition and Markets Authority (CMA)—or OFT, as it then was—investigated similar issues in Paroxetine, a case which the European Commission pushed to the CMA because of an EU limitation issue and which the CMA pursued as the UK does not have a limitation period for competition law infringements. That case is currently under appeal before the Competition Appeal Tribunal which in turn has referred a number of questions to the European Court of Justice.
In this week’s Perindopril judgment, the General Court confirms that a patent settlement agreement can be a restriction by object (i.e., without the Commission having to prove actual anti-competitive effects) where it (i) contains an inducement in the form of a benefit for the generic company (again, the pay element), and (ii) a corresponding limitation of the generic company’s efforts to compete with the originator company (the delay element). However, the General Court annulled parts of the Commission’s decision of giving rise, in particular, to a number of significant points that are of wider relevance beyond the facts of the case:
First, the Court held that where there is (i) a genuine dispute involving litigation between the parties; and (ii) a license agreement that is directly connected with the settlement of that dispute, then the linking of the settlement and the licensing arrangement in a settlement context is not a strong indication of a reverse payment or value transfer. It could be, but that is for the Commission to prove and, in particular, it will need to be shown that any value given in the license exceeds the ‘normal’ value of the asset traded.
Second, the General Court rejected the proposition that a licensing arrangement for some, but not all, EU member states in respect of which the dispute is settled, constitutes a value transfer or some form of market sharing. For there to be an EU competition law issue, the EU Commission needs to demonstrate on the facts that the agreement is not at arm’s length (i.e., below value) as an inducement to recognize the patent in other (non-licensed) territories.
Third, and possibly most importantly, the Commission lost on abuse of dominance, where the theory does not depend on the existence of a (settlement) agreement but it does require the Commission to prove dominance on a relevant market. Here, the General Court rejected the Commission’s approach to define the economic market as comprising only the relevant molecule, rather than by therapeutic substitutability. This point is also important beyond dominance cases as over the past 10 years or so the Commission has gradually moved more and more to a molecule-based assessment, particularly in genericised markets. The General Court goes into some detail on the point that, although there may be a significant price difference once a market becomes genericised, non-price factors remain important, especially in pharma where the purchasing decision is taken by a clinician and not the end user. In the UK Paroxetine case, Mr Justice Roth sitting in the Competition Appeal Tribunal, explicitly followed the view of the Commission in Perindopril (recognising that it was novel) and included a question on this point in his referral request to the European Court.
The General Court’s ruling in Perindopril is unlikely to be the end of the road. Like the previous decision in Lundbeck, which is currently awaiting judgment from the CJEU on appeal from the General Court, it seems likely that some parties in Perindopril will appeal further. If so, there will also be an interesting interaction with the preliminary ruling request in Paroxetine which is likely to be considered before any further appeal is heard in Perindopril.
Watch this space.