EU General Court provides guidance to assess whether patent settlements between originator and generics infringe antitrust law (Servier)

On 12 December 2018 the General Court of the EU (“GC”) has passed a set of parallel judgments (we will referrer to Case T-691/14,Servier and Others v. Commission and Case T-684/144,Krka v. Commission) on the incompatibility with Article 101 TFEU of a number of settlement agreements in patent disputes between Servier, which is the originator of Perindopril (a drug primarily intended for the treatment of hypertension and heart failure), and five manufacturers of that drug’s generic versions.

The controversy was raised by a decision of the Commission, which characterized those patent settlements as “pay-for delay” agreements, i.e. as agreements by which the originator pays generic producers to delay market entry with the main purpose of protecting the original drug from generic price competition beyond the expiry of the patent protection. As a result, the Commission fined Servier and the generic competitors a total of € 427.7 million; in turn Servier and the generic producers appealed the decision before the GC.

Notably, the GC already passed a series of judgements on the same issue in 2016 when it upheld the decision of the Commission in the Lundbeck case and ruled for the first time that pharmaceutical “pay-for-delay” agreements are anti-competitive “by object”, which means they can be presumed contrary to Article 101 TFEU without conducting an empirical analysis of the actual and potential effects of such agreements on competition. This judgment (which is under appeal before the ECJ) has been strongly criticized by the industry and by commentators on the grounds that it clashes with IP and competition law principles, particularly with the one according to which for an agreement to fall within the prohibition of Article 101 it must restrict competition that would otherwise have existed (so called “counterfactual analysis” of the agreement).

Indeed, defendants and commentators have argued that the generic manufacturer could not (as a matter of law or of fact) be considered an actual or potential competitor on the same market of the patented drug since market entry would be precluded by patent rights; and therefore the settlement agreement could not be presumed harmful for competition since there would be no competition between the parties absent the contested settlement. Since the matter entails an evaluation of the merits and validity of the patent rights which are the object of the dispute settled with the contested agreement, the Commission’s and GC’s pronouncements that the patent settlement is anti-competitive by object seems to interfere with the fundamental protection of IP rights, as neither the Commission nor the Courts of the EU are competent to rule on the validity of patents.

Nonetheless, the GC in Servier and Krka has now once again confirmed its view: generic manufacturers are potential competitors of originators and therefore outright pay-for-delay agreements are anti-competitive “by object”. However, it made important statements which aim at better circumscribing the boundaries for the assessment of patent settlements agreements under Article 101, though still leaving room for discretion in the appraisal. It also reduced the fine imposed on Servier by more than € 100 million and completely annulled a € 10 million fine imposed on generic producer Krka.

First, coherently with the case law of the ECJ, the GC highlights that the guarantees provided for in Article 17(1) of the Charter of Fundamental Rights also apply to intellectual property rights along with presumption of validity of patent rights. Hence, it is “a priori legitimate” and, in principle, “pro-competitive” for the parties to a patent dispute to conclude a settlement agreement rather than pursuing litigation before a court. However, for the purpose of reconciling patent law and competition law in the context of such settlements “a balance must be struck between, on the one hand, the need to allow undertakings to make settlements, the increased use of which is beneficial for society and, on the other hand, the need to prevent the risk of misuse of settlement agreements, contrary to competition law, leading to entirely invalid patents being maintained and, especially in the medicinal products sector, an unjustified financial burden for public budgets” (para. 219-252 in Servier as summarized in para. 131 in Krka).

Second, in the light of the above, the GC recognizes that non-marketing (or non-competition) and non-challenge clauses are strictly necessary for, or inherent to, the settlement of some “genuine disputes” relating to patents in which the interested party “has already had the opportunity to challenge the validity of the patent concerned and ultimately acknowledges that validity”. Thus, the mere presence in settlements agreements of such clauses, whose scope is limited to that of the patent in question, does not justify a finding of a restriction of competition “by object” where the agreements are based on the recognition by the parties of the validity of the patent and, consequently, of the infringing nature of the generic products concerned (para. 138-141 in Krka).

However, and third, the GC added that where a generic manufacturer is “induced” to accept such a patent settlement (and the inherent non-marketing and non-challenge clauses) by an “abnormal” benefit conceded by the originator, typically a “reverse payment” (i.e. a payment from the originator to the generic), which is not justified by any other plausible explanation, the picture changes and the agreements in question “must be regarded as market exclusion agreements, in which the stayers are to compensate the goers. Such agreements actually constitute a buying-off of competition and must therefore be classified as restrictions of competition by object” (para 143-150 in Krka). And this is so even if there is a genuine dispute and the patent could legitimately be regarded as valid by the parties to the agreement (para. 152 in Krka).

Further and interestingly, the GC clarified that to find that there is a “reverse payment” (or another abnormal inducive benefit), the Commission can resort by analogy to the “arm’s length” concept (which is similar to “normal competitive conditions”) developed in the field of State aid to determine whether a State has acted like a private investor (para. 173 in Krka).

Notably, on a side but relevant note, the GC pointed out that, “although neither the Commission nor the Courts of the EU are competent to rule on the validity of patents, […] those institutions may, in the context of their respective powers and without ruling on the intrinsic validity of the patent, find that is has been used abnormally, in a manner which has no relation to its specific subject matter” (para. 145 n Krka); and “inducing a generic manufacturer to accept non-marketing and non-challenge clauses [through a reverse payment] constitutes an abnormal use of the patent” (para. 146 in Krka). This statement clearly puts pay-for-delay agreements in the pharma sector within the realm of “abuse or misuse of rights”.

With regard specifically to settlement agreements between Servier and Krka, the GC found that the Commission failed to show to the requisite standard that there was a “reverse payment” from the originator to the generic manufacturer to make the latter accept restrictive clauses in the settlement agreements. In particular, the GC noted that: (i) there was a genuine dispute; (ii) there was a linking of the license agreement to the settlement agreement without strong indication of an abnormal inducement; and (iii) the burden to prove the existence of a “masked” reverse payment was up to the commission (see para 248-2050 in Krka). Further, (iv) the GC found that the Commission also erred in assessing the effects of the agreements between Servier and Krka, as it was not shown how the invalidation of the patent in question and thus market entry could have occurred faster absent the contested agreements. As a result, it annulled the relevant part of the Commission’s decision and the related fine.

The Servier judgements are currently under appeal before the ECJ.

On errors in assessing dominance of Servier in the relevant markets.

In addition to fining Servier and the five generic manufacturers for concluding the contested pay-for-delay agreements in breach of Article 101, the Commission also fined Servier (EUR 41,27 million) for abusing of its dominant position in breach of Article 102 TFEU on four national markets for perindopril formulations (in France, Poland, the United Kingdom and the Netherlands) and on the perindopril API technology market. Allegedly, the abuse consisted in implementing an overall exclusionary strategy against generic manufacturers by: (i) concluding the contested pay-for-delay settlements; and (ii) acquiring an independent, non-infringing technology (the AZAD Technology) to produce perindopril API.

However, in a very rare stance, the GC annulled this part of the decision too on the grounds that the Commission erred in identifying the relevant markets and thus in finding that Servier was dominant. In particular, the GC found that the Commission defined the product market too narrowly and identified four “manifest” errors of assessment in this respect:

• the Commission wrongly weighed the importance of price competition since the demand for prescription medicines is driven by prescribing physicians (rather than patients), who choose a treatment mainly depending on therapeutic considerations (rather than the cost of treatment, which is generally reimbursed by the NHS);
• the Commission wrongly focused on certain specific characteristics of perindopril while ignoring other ACE inhibitors, which treat the same disease and may be deemed a substitute by prescribing physicians;
• the Commission did not take sufficiently into account evidence that pointed to the absence of significant differences between efficacy and side effects of perindopril and competing products;
• similarly, the Commission underestimated the evidence pointing to competitive constraints exercised on perindopril by rival products. This part of the judgment represents an important development since the ECJ as well as Italian courts have consistently limited their power of review over authorities’ definitions of relevant markets to “manifest errors” in the assessment or reasoning, being reluctant to get into the merits of complex economic matters that require the specific and superior technical competence remitted to competition authorities. By virtue of such superior technical competence, competition authorities have always been accorded a margin of discretion (or “technical discretion”) in defining relevant markets, which could not be reviewed by courts beyond the finding of manifest factual errors in the assessment or in the reasoning. Indeed, full annulments of market definitions are not unprecedented but certainly very rare (reportedly, in 5 cases only out of more than 600 annulment proceedings at EU level).

This judgement does not change the relevant criterion (finding of a “manifest” error in the authority’s factual assessment or reasoning), though it seems to mark a new trend where EU courts are less reluctant to narrow the margin of discretion accorded to the Commission or at least to require a higher standard of rigorousness and thoroughness in the assessment and reasoning over relevant markets’ definitions in Article 102 infringements. If this trend is confirmed, it is likely to similarly condition the depth of scrutiny of national courts over the competition authorities’ assessment of relevant markets in abuse cases.