January 27, 2015

EU Antitrust Developments in the Pharmaceutical Sector in 2014

Arnold & Porter Advisory

This client advisory gives a brief overview of EU antitrust developments in the pharmaceutical sector in 2014. These developments center around two well-known themes.

First, to what extent can pharmaceutical companies use the regulatory process or the patent rights obtained in that process without running afoul of Art. 101 and/or Art. 102 of the Treaty on the Functioning of the European Union (TFEU). This question first arose in AstraZeneca.1 In the past two years, the European Commission (EC) has focused on patent settlements, in particular those involving value transfers between the originator and one or more generic companies.

As in the USA, these agreements are dubbed "pay-for-delay" settlements. While the EC's approach is not dissimilar to that followed by the US Federal Trade Commission, the situation in Europe differs from that in the USA because it will likely take several more years before its supreme court – the European Court of Justice (ECJ) – will weigh in.2 That said, in some Member States, appeal courts have already upheld decisions of national competition authorities (NCAs) prohibiting pharmaceutical companies from making certain uses of the regulatory process or their patent rights to delay entry of potential competitors.

The second theme is an evergreen theme in EU law: parallel trade. Given the strong market integration overtones that have governed antitrust enforcement in Europe since its birth, the EC is unlikely to abandon its long standing policy to ban territorial restraints in distribution agreements that limit cross-border intra-brand competition within the EU.

However, the EC can set priorities in its enforcement policy. This is what it did last year, when it decided not to pursue its investigation in a parallel trade case that had been on its desk for about 15 years (after the ECJ had issued a final ruling confirming that it had not assessed carefully enough the pharmaceutical company's efficiency defense in support of a pricing policy aimed at stemming parallel exports from Spain to other Member States, in particular the United Kingdom).


We will first review the EC's Monitoring Report fifth Report on the Monitoring of Patent Settlements (infra 1).3 We will then turn to the EC's prohibition decision of 9 July 2014 in Servier (infra 2).4 This decision, which banned practices aimed at delaying price competition of generic products, follows two prohibition decisions on "pay-for-delay" practices in 2013: Lundbeck5 and Johnson & Johnson/Novartis6. Thereafter, we will put the Servier decision in a broader policy perspective by pointing at some regulatory developments that occurred last year: the EC's revision of the Technology Transfer Block Exemption Regulation (TTBER) and its communication giving guidance on restrictions of competition "by object" for the purpose of defining which agreements may benefit from the de minimis Notice (infra 3).7 We will close with a brief review of a few major antitrust cases decided by NCAs – merely in order to illustrate that the EC and its counterparts in the 28 Member States do not act in splendid isolation (infra 4).

1.  Monitoring Report on Settlement Agreements

In July 2009, when it completed its pharmaceutical sector inquiry, the EC had announced that it would monitor agreements whereby originator and generic companies settle their patent-related disputes. Like its predecessors, the fifth Report is only a monitoring report. It gives an overview of the various types of settlement agreements (146 in total) without assessing their legality under Art. 101 of the TFEU. That said, the typology used in the Report reflects the EC's views as to which agreements warrant special antitrust scrutiny. It will take a few more years before the General Court or the ECJ will issue judgments that will shed light on the standard of legality governing settlement agreements and as long as there will be no non-confidential versions of the EC's decisions in Lundbeck, Johnson & Johnson/Novartis and Servier, the EC's monitoring reports remain the only source of policy guidance in this field. In its last report, as in previous reports, the EC distinguishes between three types of agreements.

First, the EC views 66 agreements as unproblematic because they do not in any way limit the generic company's entry into the market, i.e., they leave that company entirely free to market its own generic product in the relevant geographic market, under the conditions chosen by itself. In 29 cases, these agreements were concluded on a "walk away" basis: both parties agreed to simply discontinue their litigation without undertaking any further commitment/obligation vis-à-vis each other. In the other 37 cases, there was a "value transfer" in the form of compensation of litigation costs and/or damages. The compensation was either paid by the originator (e.g., because it had originally obtained an interim injunction against a generic company's product, but then feared to lose the main case) or by the generic company (e.g., because it had entered the market at its own risk but pending the litigation, the patents concerned expired).

Second, the EC counts 69 agreements that do limit the generic company's market entry in the sense that the latter agreed to postpone its entry until after the patent(s) at issue had expired (e.g., because it accepted the validity of the originator's patent or decided not to appeal an unfavorable court judgment). None of these agreements involved a value transfer operated by the originator. In some instances, it was the generic company who agreed to compensate the originator company for its litigation costs and/or damages.

Third, the remaining 11 agreements are reported to have limited the generic company's market entry and involved a value transfer operated by the originator. These are the only agreements for which the EC has indicated in the past that they warrant closer antitrust scrutiny and it repeats this warning in its 2014 report by stating that these agreements are "likely to attract the highest degree of antitrust scrutiny".8

However, in 9 instances, the agreements only contained a non-assert clause pursuant to which the originator committed not to invoke its patent against the generic company, thereby in fact allowing early generic entry. In most cases, this clause was supplemented by a license. The EC qualifies this clause as a form of value transfer offered by the originator to the generic company, but acknowledges that an agreement containing such a non-assert clause (possibly combined with a license) "is not likely to attract the highest degree of antitrust scrutiny" because it "includes no other limitative provision than determining the date of the generic entry with the originator's undertaking not to challenge such entry (a 'pure early entry')".9 The EC adds that, in any event, it would engage in an effects analysis of such agreements and verify "whether the license granted to the generic company may in fact have pro-competitive effects".10 It would also be ready to "consider arguments raised by parties pointing to any potential pro-competitive effects of the settlements".11

The 2 other agreements involving a value transfer truly limited generic entry because the generic company accepted to enter the market only at the time the patent lapsed and received a compensatory cash payment from the originator. The EC views these agreements as the ones that call for heightened scrutiny. Although, the EC hastens to add that "this is not to suggest that agreements falling in this category would always be incompatible with EU competition law" and that an assessment would need to be conducted "on the basis of the circumstances of each individual case"12, there is little doubt that it will continue to adopt a "by object' approach when assessing these agreements, as it did in Lundbeck and in Servier (infra).

The "by object" approach is reserved for agreements which restrict competition "by their very nature". According to settled case law, this is the case for agreements that "reveal a sufficient degree of harm to competition that it may be found that there is no need to examine their effects".13

2.  Servier

In this case, the EC identified two infringements: on the one hand, a unilateral practice whereby Servier had engaged in a form of output restriction, which constituted an abuse of monopoly power within the meaning of Art. 102 TFEU and, on the other hand, patent settlement agreements between Servier and a number of generic companies that violated Art. 101 TFEU. According to the EC, Servier engaged in these two practices in order to extend its monopoly power for its best-selling cardiovascular, anti-blood pressure medicine containing the active substance perindopril.

According to the EC, Servier abused its dominant position by preventing generic companies from obtaining access to non-protected technology that would have enabled them to develop patent-free products in competition with its own product. More specifically, Servier did so by acquiring the most advanced technology in 2004 (i.e., the year after the patent protection for perindopril had expired) and subsequently never using this technology.

Furthermore, between 2005 and 2007, Servier entered agreements with five generic companies in order to settle disputes over the validity of its patents. According to the EC, "this was not an ordinary transaction where two parties decide to settle a patent claim outside of court to save time and costs" because four generic companies received considerable cash payments (totaling several tens of millions of euros) from Servier and one of them received a license to sell Servier's product in seven countries in exchange for a promise not to enter other markets in Europe.

In its press release, the EC qualifies the value transfers in question (i.e., cash payments or licenses) as means whereby Servier and the generic companies shared monopoly rents and thereby caused harm to patients, national health systems and taxpayers.  In this regard, it observes that "experience shows that effective generic competition drives prices down significantly" and that "in 2007, prices of generic perindopril dropped on average by 90% compared to Servier's previous price level in the UK (…) when the only remaining legal challenger in the UK obtained the annulment of Servier's then most important patent".

The reference to "experience" is not accidental. According to settled case law, an agreement constitutes an infringement "by object" when "experience shows that [it] leads to falls in production and price increases, resulting in poor allocation of resources to the detriment, in particular, of consumers" (emphasis added).14

In its press release, the EC also quotes various internal company documents to show that the parties' objective was indeed to make money on the back of consumers. These quotes are not accidental either. Pursuant to settled case law, before qualifying an agreement as an infringement "by object", the EC must have regard to "the content of its provisions, its objectives and the economic and legal context of which it forms a part".15

3. TTBER and the Guidance communication on de minimis agreements

This is not the place to review the revised TTBER in detail.16 However, it is worth flagging a couple of points.

First, the Guidelines accompanying the TTBER contain a short section on "pay-for-restriction in settlement agreements". Referring to Lundbeck, the EC states that these agreements usually do not involve a transfer of technology rights "but are based on a value transfer from one party in return for a limitation on the entry and/or expansion on the market of the other party and may be caught by Art. 101-1 TFEU". The term "may" strikes us as an understatement, given the above mentioned "by object" approach in Lundbeck and more recently Servier. Where the settlement does contain a transfer of technology rights in the form of a license, the EC argues that two contract clauses are so pernicious that they deprive the entire agreement of its presumed legality. The EC mentions the allocation of markets or customers (without prejudice to certain specific exceptions, which we do not discuss here) and the restriction of the licensee's ability to exploit its own technology rights (without prejudice to a broadly worded exception) as "hard core" restrictions that take the entire agreement outside the scope of the TTBER.17

The accompanying Guidelines also refer to "excluded restrictions", i.e., contract clauses which are presumably unlawful but do not affect the validity of the entire agreement. No specific reference to the pharmaceutical sector is made but the following two clauses are probably particularly relevant for pharmaceutical players: clauses, pursuant to which the licensee is required to license back to the licensor on an exclusive basis any improvements made to the licensed technology and clauses, pursuant to which the licensee is precluded from challenging the validity of the licensor's technology.[[TTBER, Art. 5-1 (a) and (b).]]

The EC has offered a one-year grace period for companies to amend the terms of existing agreements as the revised TTBER will enter in force only as of 1 May 2015.18

Given the EC's "by object" approach to settlement agreements involving a transfer of value, it goes without saying that pharmaceutical companies have an interest in reviewing their contracts carefully in order to make sure that they fit the model agreement that is block exempted under the TTBER.

This is all the more advisable since the EC has indicated in its de minimis Notice that agreements containing hard core restrictions will raise concern even if the parties to these agreements hold a market share that does not exceed 15% (in a vertical context) or 10% (in a horizontal context).19 It would thus seem wise for parties to a settlement agreement that could be viewed as allocating markets or customers (because they involve some form of value transfer) to spend some time assess the legality of their contracts under Art. 101 TFEU.

4. EU Member States Pharmaceutical Cases

We cannot review in any detail a number of interesting antitrust decisions adopted by NCAs in the pharmaceutical sector last year. However, it is useful to mention just a few of them as these decisions show that the EC and the NCAs are "on the same page" when they challenge agreements or practices that restrict the market entry of potential suppliers of cheaper medicines.

Thus, in Italy, on 14 January 2014, the highest Administrative Court upheld the NCA's decision from January 2012 fining Pfizer for an abuse of a dominant position on the relevant market. Pfizer had allegedly used a divisional patent for latanoprost (the active substance in Xalatan, a medicine used for the treatment of glaucoma) in order to delay the market entry of generic products.20 It apparently did not help Pfizer that it had never used this patent for the purpose of launching itself a new product.

On 27 February 2014, the Italian NCA fined Roche and Novartis (which holds 33% in Roche) for allegedly agreeing to protect Novartis' best seller drug Lucentis, a medicine used to treat a serious deterioration of vascular eyesight conditions, against potential competition from Roche's drug Avastin, a medicine officially registered as an anti-cancer drug but also used "off label" to treat the same eye disease as Lucentis – but at a price that was more than 10 times cheaper.21 In particular, according to the NCA, the companies had agreed to raise concerns about the safety of Avastin in order to influence doctors' and health services' prescriptions of the product.

On 18 December 2014, the Paris Court of Appeal upheld the NCA's decision to fine Sanofi-Aventis for allegedly denigrating generic substitutes of its Plavix drug for heart patients.22 Sanofi had been fined for allegedly deploying a communication strategy aimed at convincing doctors to prescribe Plavix and encouraging pharmacists which were dispensing generic versions of it to switch to Plavix. The Court noted that creating doubts in the minds of prescribing doctors regarding the therapeutic efficacy of a generic product compared to that of the originator drug is an abusive practice.


On 27 May 2014, the EC rejected a longstanding complaint of Spanish wholesalers regarding GSK's pricing policy aimed at limiting parallel exports of its pharmaceutical products from Spain to other EU Member States.23 These wholesalers had insisted that the EC re-activate its investigation since the ECJ had ruled in 2009 that this pricing policy constituted an infringement "by object" of Art. 101 TFEU, even if that agreement cannot be "presumed to deprive final consumers of the advantages of effective competition in terms of supply or price" (for instance, because the parallel traders pocket all these advantages for themselves).24

However, the EC saw no point in pursuing the matter because (i) GSK had already ceased this policy in 1998 and had never re-introduced it since then, (ii) this policy, which had only been in operation for a couple of months, was manifestly not hampering the Spanish wholesalers' capacity to export the medicines today and (iii) in any event, the Spanish antitrust authority and courts were well placed to decide whether GSK's pricing policy met the conditions set forth in Art. 101-3 TFEU.25

It follows that there are two scenarios in which pharmaceutical suppliers might get away with measures aimed at impeding parallel trade under the EU antitrust provisions.

First, the antitrust authority might decide that it is not – or no longer – a priority for it to investigate a parallel trade issue that one or more complainants (typically intermediaries involved in such trade) have raised, as the EC did when it rejected the complaint of Spanish wholesalers against GSK's pricing policy.

Second, the companies can invoke Art. 101-3 TFEU in an effort to convince the EC (or one of its counterparts in a Member State) or a national court that their policy creates efficiencies from which consumers will benefit. For instance, they might argue that parallel trade generates losses that negatively impact on their research and development budgets, and thus on their capacity to develop innovative medicines. This is what GSK claimed in support of its pricing policy in Spain. Let us not forget that the General Court as well as the ECJ concluded that the EC had discarded that claim too easily. Adequate access to medicines in the export countries might also be a justification, as the ECJ noted in Lelos, when it observed that suppliers can refuse to deal with wholesalers who "order medicines in quantities which are out of all proportion to those previously sold by the same wholesalers to meet the needs of the market in that Member State".26

One last comment: It would be prudent not to read into the rejection of the Spanish wholesalers' complaint that the EC has lost interest in parallel trade cases in the pharmaceutical sector.

In its decision, the EC actually mentions twice that it opened an investigation into pricing schemes for the distribution of medicines in Spain in January 2012 of several suppliers, including GSK, and that this investigation is still pending.27

  1. EC decision of 15 June 2005 in case COMP/A. 37.507/F3, upheld by the General Court in its judgment of 1 July 2010 in case T-321/05, AstraZeneca AB and AstraZeneca plc v. European Commission, {2010} ECR II-2805 and by the European Court of Justice in its judgment of 6 December 2012 in case C-457/10 P, AstraZeneca AB and AstraZeneca plc v. European Commission, not yet published but available on the Court's website.

  2. For a comment of recent post Actavis developments, we refer to the client advisory of our DC colleagues.

  3. EC Report on the Monitoring of Patent Settlement Agreements of 5 December 2014 (full text available on DG COMP's website).

  4. EC decision of 9 July 2014 in case AT 39612 (press release IP/14/799). This decision has been challenged (see pending cases T-679/14, T-680/14, T-682/14, T-684/14 and T-691/14).

  5. EC decision of 19 June 2013 in case AT 39226 (press release IP/13/563). This decision has been challenged (see pending cases T-460/13, T-467/13, T-469/13, T-471/13 and T-472/13).

  6. EC decision of 10 December 2013 in case AT 39685 (press release IP/13/1233). The parties have not challenged this decision.

  7. EC Regulation No 316/2014 of 21 March 2014 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of technology transfer agreements, O.J. L 93/17, 28 March 2014 and its Communication with Guidelines, O.J. C 89/3 of 28 March 2014. As for the de minimis point, see the EC's staff working document of 25 June 2014 (SWD 2014-198 final).

  8. Cit. note 2, paragraph 17.

  9. Id., footnote 15.

  10. Id., paragraph 47.

  11. Id., footnote 16.

  12. Id., paragraph 17.

  13. See most recently, ECJ judgment of 11 September 2014 in case C-67/13 P, Groupement Cartes Bancaires (CB) v. European Commission, not yet published, but available on the Court's website, paragraphs 49-50.

  14. Id., paragraph 51.

  15. Id., paragraph 53.

  16. See A. Gutermuth, "Revised EU competition rules for technology licensing agreements took effect on 1 May 2014" in The In-House Lawyer, 8 May 2014.

  17. TTBER, Art. 4-1 (c) and (d).

  18. TTBER, Art. 10.

  19. See EC's staff working document, cit. note 6, p. 8.

  20. Italian Council of State, Judgment of 14 January 2014, published on 12 February 2014 (overruling the appeal court's judgment) in case n° 693 Italian Competition Authority v. Pfizer Italia.

  21. Decision of 27 February 2014 in case n°1760, Roche-Novartis / Farmaci Avastin e Lucentis.

  22. Paris Court of Appeal, Judgment of 18 December 2014 in case n°1760, Sanofi SA and Sanofi-Aventis France.

  23. EC decision of 27 May 2014 in case AT 36957 (non-confidential version available on DG COMP's website).

  24. ECJ judgment of 6 October 2009 in joined cases C?501/06 P, C-513/06 P, C-515/06 P and C?519/06 P, GSK Services Unlimited v. European Commission, paragraphs 62-63.

  25. Id., paragraphs 23, 24, 26 and 39.

  26. ECJ judgment of 16 September 2008 in joined cases C-468/06 to C-478/06, Lelos and others v. GSK AEVE, ECR I-7139, paragraph 76.

  27. Cit. note 23, paragraphs 11 (f) and 27.


Subscribe Link

Email Disclaimer