Finding Vredo pat.pending

Transcription

1 februari 2014 berichten industriële eigendom 41 Finding Vredo pat.pending The Dutch Supreme Court decision on escitalopram in Lundbeck v Tiefenbacher et al. Alexander Tsoutsanis* Background 1. One of the evergreens in the field of global patent litigation is the Lundbeck saga on escitalopram. The patent claiming (i.a.) this compound sparked litigation across the world, including for example in the United States, but also in the United Kingdom, Germany and The Netherlands, which consistently rank among the top tier venues for patent litigation in Europe. 2. Lundbeck is no stranger to success. It proved particularly successful in developing and marketing antidepressant drugs, including since 1989 Cipramil with the active ingredient citalopram and since 2002 Cipralex with the active ingredient escitalopram. 1 Both turned out to be Lundbeck s best-selling blockbuster drugs of all times. Cipralex alone is responsible for approximately half of Lundbeck s entire turnover, grossing 498 million in 2012, and 452 million in Around 2006, a number of generic pharmaceutical companies across Europe were seeking to invalidate Lundbeck s patent for escitalopram (EP ). The dispute hit the Netherlands in 2008, with Tiefenbacher and Centrafarm leading the charge. 3 According to Lund beck, this was merely a case of generic competitors trying to clear the way into their lucrative market, free-riding on its R&D efforts. 4 Conversely, Tiefenbacher c.s. alleged that this was a case of making new drugs from old, as escitalopram was merely a purified clone, consisting of the separated (+) enantiomer of Lundbeck s existing non-separated (±) racemic compound citalopram. * Alexander Tsoutsanis is a lawyer with DLA Piper and a Senior Researcher and Lecturer at the University of Amsterdam. He also sits on the editorial board of this law journal. 1 Also marketed as Lexapro. 2 Lundbeck Full Year Report 2012, 6 Feb. 2013, p. 8 available on 3x0x633265/970471a3-466c-476f-8aa7-e215ad59aaf6/Read_the_full_ year_2012_release_in_pdf.pdf, last accessed 24 Jan Kitchin J. holds in Ch.Div. 4 May 2007 (Generics (UK) Ltd et al. v H. Lundbeck A/S) par. 2: Escitalopram was launched in 2002 and accounts for approximately 60% of Lundbeck s turn-over. It is currently the world s top selling branded antidepressant in terms of volume. 3 Ratiopharm was also involved in first and second instance, but settled and was no longer involved when the Dutch Supreme Court rendered its decision. 4 B. van den Broek, The Lundbeck Saga, in: Liber Amicorum Willem Hoyng, DeLex 2013, p Separating (+) from (±). That was the gist of Lundbecks invention which EP 066 aimed to patent. Lundbeck patented it well, not merely by claiming the process (independent claim 6), but also the product (independent claim 1) and some pharmaceutical compositions (claim 2-5). Tiefenbacher c.s. argued that both the process and product claims of EP 066 for escitalopram were invalid for lack of (i) novelty and (ii) inventive step in light of Lundbeck s own earlier patents, both in respect of the basic substance citalopram itself (US / NL ) as well as in respect of various production processes (such as US / EP ) In the Netherlands, the dispute already lasts 6 years, resulting in decisions by the District Court in The Hague (2009), the Court of Appeal (2012), to the decision of the Dutch Supreme Court last summer (2013). 6 The outcome: the Supreme Court vacates the contested decision, and refers it back to the Court of Appeal. The Dutch Supreme Court 6. The decision of the Dutch Supreme Court touches on four main points, which can be summarized as follows: a. substance protection: a patentee can also get protection for a substance (compound) made by a (novel and) inventive process, even if the composition and the possible characteristics of such substance are already known (par. 4.3). By far, this is the most important aspect from this case. It will be discussed in more detail below. b. burden of proof: the applicant for invalidity bears, in principle, the onus of establishing evidence of lack of inventive step (par ). This even applies in situations where the required evidence is on establishing that no other known methods to obtain the novel compound existed. Although this builds on the established rule of burden of proof in Dutch procedural law, it is less evident in view of the decision of the Technical Board of Appeal in T-990/96 relied upon by the Court of Appeal, which suggests a different onus of proof in exceptional situations. c. problem-solution approach: upon determining inventive step, the Dutch courts are not under the obligation to apply the problem-solution approach (par ). This was for example already held by Advocate-General Huydecoper in Global Bio Technology v Ajinomoto. 7 While it is true that there is no such 5 Insufficiency was also invoked by Ratiopharm as a third, more alternative, ground, but this argument was never decided upon by the courts. This is a key difference with the parallel escitalopram proceedings in the United Kingdom. 6 The Hague District Court 8 April 2009 HA ZA , HA ZA , 2172; The Hague Court of Appeal 24 Jan. 2012, ECLI:NL:GHSGR:2012:BV1963; Dutch Supreme Court 7 June 2013, ECLI:NL:HR:2013:BZ4115 (H. Lundbeck A/S v Alfred E. Tiefenbacher, Centrafarm BV). 7 Advocate-General Huydecoper in Global Bio Technology v Ajinomoto (ECLI:NL:PHR:2012:BY7837) par. 39. Electronic copy available at:

2 42 berichten industriële eigendom februari 2014 obligation, the problem-solution approach is nevertheless often applied by the District Court and Court of Appeal in The Hague. 8 d. Supplementary Protection Certificate: the Supreme Court does not revisit the fact-driven assessment of the Court of Appeal that escitalopram is a different compound than citalopram and has its own therapeutic effect, warranting its own SPC. The Court was unable to factor in the more recent view points of the CJEU on SPCs in Actavis v Sanofi (C-443/12), Georgetown University (C-484/12) and Eli Lilly and Company Ltd v. Human Genome Sciences (C-493/12). 9 Absolute substance protection 7. For the Netherlands, the importance, and practical significance, of this decision lies in the Supreme Court s recognition of compound protection. This was the point of law on which the Supreme Court vacated the decision of the Court of Appeal. 8. In order to put the Supreme Court s decision in perspective, it is necessary to first understand the decision of the Court of Appeal. With regard to absolute substance protection, the decision was quite elaborate, acknowledging both the case-law of the Technical Boards of Appeal of the European Patent Office as well as foreign decisions rendered with regard to the same patent elsewhere. 9. The pivotal point in this case was not whether chemical substances as such are eligible for absolute substance (compound) protection. In the Netherlands, since 1978, it has been accepted that chemical substances can be described and patented in the form of a product claim. 10 Apart from product claims, the Dutch Patent Act, in line with the European Patent Convention, i.a. also acknowledges process claims : such claims do not protect a product as such, but the process from which the product results. 11 Both types of claims can be included in the same patent application And is also recognized as such in various opinions by Advocates- General of the Dutch Supreme Court, see e.g. the opinion of Advocate-General Langemeijer in Rockwool International v. Saint-Gobain Isover (ECLI:NL:PHR:2008:BB5066) par and also in Schneider (Europe) v. Cordis (ECLI:NL:PHR:2009:BG7411) par For example, the Court s succinct reference (par ) that the (only) objective of Regulation 469/2009 is the extended protection of medicinal products for which a patent has been granted is already outdated in view of the more balanced approach taken by the CJEU in the more recent ruling of e.g. Actavis v Sanofi. In said case it was held that in line with recital 10 of the Regulation extended protection is not a goal itself but needs to be balanced against other interests at stake, including those of public health. The CJEU also held that (par. 40) the objective of the Regulation is not to compensate the holder fully for the delay to the marketing of his invention or to compensate for such delay in connection with the marketing of that invention in all its possible forms, including in the form of combinations based on that active ingredient. 10 See e.g. Pres. District Court The Hague (Van Peursem) 15 July 2005 IEPT (Teva v Abbott) par Article 53-1(b) Dutch Patent Act; 64-2 EPC. 12 In addition, there are also use claims, claiming a certain use / application of a (known) product. 10. The difference between both types of claims is the following: a. A product claim enables the patentee to prohibit any use of the substance, irrespective of its method by which this was made, and irrespective of its application. This is therefore called absolute protection. b. A process claim enables the patentee to not only prohibit any application of a process, but also any use of any substances directly obtained by such process. 11. The above is settled case law. It was expressly confirmed by the Enlarged Board of Appeal in the Mobil oil case: There are basically two different types of claim, namely a claim to a physical entity (e.g. product, apparatus) and a claim to a physical activity (e.g. method, process, use). These two basic types of claim are sometimes referred to as the two possible categories of claim. In this decision, however, the word category is used to refer generally to the various different possible classifications of claim. Within the above two basic types of claim various sub-classes are possible (e.g. a compound, a composition, a machine; or a manufacturing method, a process of producing a compound, a method of testing, etc.). Furthermore, claims including both features relating to physical activities and features relating to physical entities are also possible. There are no rigid lines of demarcation between the various possible forms of claim. (..) 5. (..) An initial question to be considered in such a case is the protection which is conferred by a claim to a physical entity such as a compound, per se. It is generally accepted as a principle underlying the EPC that a patent which claims a physical entity per se, confers absolute protection upon such physical entity; that is, wherever it exists and whatever its context (and therefore for all uses of such physical entity, whether known or unknown). 12. The pivotal point before the Court of Appeal was not the scope of protection of product claims. The concept of absolute protection was acknowledged by the Court of Appeal. 14 Rather, the key issue was how to assess the validity of such product claims, in connection with the process claims mentioned in the same patent. Precisely because of the absolute protection mentioned above in Mobil oil, the Court of Appeal held that the validity of both claims needs to be assessed separately, and therefore each need to comply with the 13 Enlarged Board of Appeal 11 December 1989 G 2/88 (Mobil oil/friction reducing additive), par. 2.2 and A previous lower instance decision can be found in e.g. Pres. District Court The Hague (Van Peursem) 15 July 2005 IEPT (Teva v Abbott) par , which did state that there is little precedent on such absolute substance protection both in case-law as well as literature. Electronic copy available at:

3 februari 2014 berichten industriële eigendom 43 requirement of, amongst others, (a) novelty, and (b) inventive step: 15 It follows from this (in so far as relevant here) that the substance must be per se both novel and inventive (and not just novel, as Lundbeck asserts in its pleading notes on appeal, under 6) in order to qualify for such far-reaching protection. The question of whether the method for preparing the substance is novel and inventive is a separate matter that has nothing to do with the question of whether the substance itself is novel and inventive, since even a novel and inventive substance can most certainly be prepared by applying a method that is in itself wholly and completely known. It is consistent with this that the same applies in the case of a productby-process claim; here too the product must be per se novel and inventive, that is independent of the novelty and inventiveness of the process, if absolute protection is to be conferred on the substance (cf. the Guidelines (2010) Part C Chapter III- 1 1 under 4.12). 13. Lundbeck took a different position. It argued that, if the process claim is novel and inventive, the product claim for the substance must be novel per se, but could derive its inventive step from the notion that it is the result of a novel and inventive process: Lundbeck argued (pleading notes on appeal, under 6): The party who enriches the state of the art with a new product is (logically) given absolute substance protection. After all, he has enabled all further possible applications of that substance. 14. The Court of Appeal dismissed this argument: (...) This assertion is incorrect. Absolute substance protection is granted to the party who first enriched the state of the art with a product that is both novel ánd inventive. If, as is the case here, the inventive step lies only in the method by which the (novel, but not inventive) product is first obtained, the state of the art is enriched with this method and with the product obtained using this method, for which method and product protection is then granted. As long as third parties are unable to obtain the product via an essentially different method of preparation, as is the case here according to Lundbeck, in fact substance protection applies. To the extent, however, that essentially different methods for preparing the product are added to the state of the art, the products obtained with these methods, if used by third parties, do not fall under the method claim of the patent. 15. The Court continued with a reference to decision T-296/87, acknowledging that particular circumstances can render a (novel) enantiomer inventive. Such circumstances include e.g. situations consisting of more than one asymmetric carbon atom resulting in an exponentially increasing number of isomers. It may also involve cases where the racemate itself is known, but is not part of development trends in the field, rendering separating and providing its enantiomers in itself to be inventive. The Court concluded that all such circumstances fall within the normal assessment of inventive step, and (therefore) do not support Lundbeck s position that the discovery of a novel and inventive method for preparing an enantiomer would be sufficient for granting protection for the enantiomer per se, in the event that the enantiomer is novel, but not inventive. 16. The Court of Appeal also carefully examined decision T-595/90 and concluded that: Analogous to this decision, one could find that a novel, but not inventive enantiomer could nonetheless be regarded as inventive if the method by which it was prepared was novel and inventive and if, at the priority date, there were no known methods by which the enantiomer could be obtained. The burden of proof with regard to this last point would rest on the holder of the patent directed at the enantiomer, as could be derived from the decision T 990/96 from the Technical Board of Appeal, ( ) Relying on this decision, the Court considered Lundbeck to bear the burden of proof, and subsequently held that it had failed to offer any such evidence. 17. If the lengthy reasoning of the Court of Appeal above shows anything, it must be that it clearly sought to follow the legal guidance set out by the Technical Boards of Appeal of the European Patent Office, acknowledging both the concept of absolute protection (G 2/88) as well as the possibility that in particular circumstances it is possible to consider an enantiomer to be inventive, because of the novel and inventive method it was obtained by (T-296/87). The Court of Appeal merely felt that Lundbeck failed to live up to its obligation to adduce evidence warranting such particular circumstances to be present in the case at hand (T-990/96). 18. So why did the Supreme Court vacate the decision of the appellate court? Pursuant to the decision of the Supreme Court, Lundbeck s main argument was that the Court of Appeal failed to appreciate that solving in an inventive way, the problem that an average person skilled in the art (and hence the state of the art) cannot dispose of a substance with expected favourable properties means that said substance that is provided in said manner and with which the state of the art is consequently enriched, is (novel and) inventive The Hague Court of Appeal 24 Jan. 2012, ECLI:NL:GHSGR:2012:BV196, par The Hague Court of Appeal 24 Jan. 2012, ECLI:NL:GHSGR:2012:BV196, par Dutch Supreme Court 7 June 2013, ECLI:NL:HR:2013:BZ4115 (H. Lundbeck A/S v Alfred E. Tiefenbacher, Centrafarm BV) par. 4.2.

4 44 berichten industriële eigendom februari The key difference between Lundbeck s position and the approach taken by the appellate court is that in Lundbeck s theory, considering such substance to be inherently inventive, is not confined to particular circumstances. In practice, this could imply that assessing inventive step of the substance is not required, as long as there is no method in the art to make it, and the claimed method to make available such previously unavailable substance is the first to do so, in an inventive manner. 20. The Dutch Supreme Court follows Lundbeck s argument, albeit in less strict words than advanced in Lundbeck s submission summarized above: 18 A substance of which the composition and possible characteristics are in themselves known and that for that reason follows from the state of the art in an obvious manner as intended by article 56 of the European Patent Convention (EPC) and article 6 of the Dutch Patents Act 1995, can nonetheless be non-obvious in the sense of these provisions and is therefore patentable if on the priority date the state of the art does not include a method for obtaining this substance and with the claimed method this substance can therefore be obtained for the first time in an inventive manner. After all, in that case also the substance, although in itself known in terms of composition and possible features, does not follow from the state of the art in an obvious manner. Therefore, in that case also a substance patent can be obtained. This is standard jurisprudence of the Technical Boards of Appeal of the European Patent Office (reference is made to the decisions that are referred to in section 2.7 of the opinion of the Advocate General). The highest courts in Germany and the United Kingdom also ruled with regard to the present patent in this sense (BGH 10 September 2009, Xa ZR 130/07 and although inventive step as such was no longer at issue in this ruling House of Lords, UK, 25 February 2009, [2009] UKHL 12). 21. The reasoning is in line with an earlier decision by the District Court of The Hague which also signaled doubts as to the correctness of the ruling of the Court of Appeal. 19 However, there is one concern. Finding Vredo 22. One can wonder whether the reasoning of the Supreme Court is in sync with the basic tenets the same Supreme Court handed down twenty years earlier in the well-known decision Vredo v Veenhuis. 20 Pursuant to that case, involving patents, each judicial decision shall at least be motivated in such extent that it gives sufficient insight in the underlying reasoning to render 18 Dutch Supreme Court 7 June 2013, ECLI:NL:HR:2013:BZ4115 (H. Lundbeck A/S v Alfred E. Tiefenbacher, Centrafarm BV) par District Court The Hague (Brinkman) 14 Aug ECLI:NL:RBSGR:2012:BX7359 (H. Lundbeck S/A v Sandoz BV) par Dutch Supreme Court 4 June 1993 NJ 1993/659 (Vredo v. Veenhuis) par the decision both for the litigants and for third parties in case of the availability of appeals: including the higher court verifiable and acceptable. It also adds that this duty to state reasons, is enshrined in article 121 of the Constitution, 20 Wet Ro, 59 and 429k Rv; its extent, depends on the circumstances of each case. 23. The circumstances in this case certainly warrant a better explanation than the reasoning above, for a number of reasons. 24. First, the (rather succinct) reasoning of the Supreme Court deviates from the (rather elaborate) opinion of the Advocate-General Hammerstein. In view of those contrasting views, it would be helpful if the Supreme Court explained as to why it took a different view, especially as it involves a point of law which was not explicitly recognized by the Dutch Supreme Court before Second, the mere reference to foreign precedents elsewhere does not release the Court of its duty to state reasons, based on its own examination of the law. This is also what those foreign courts did, and do, themselves. 26. Third, a quick glance on the proceedings in the United Kingdom, Germany and the Netherlands reveals that those cases are less comparable to the Dutch case than the Court s references above tend to suggest. Even though the patent was the same, various grounds, submissions and decisions were different in each of those jurisdictions: a. With regard to the United Kingdom, the House of Lords decision of 2009 predominantly centered on insufficiency (Article 83 EPC), while the Dutch Supreme Court and Court of Appeal decision do not. It is true that the arguments on insufficiency before the House of Lords were similar to the arguments on inventive step. 22 However, the lower instance decisions of the Patents Court (2007) and the Court of Appeal (2008), (also) touched on novelty and inventive step, and could therefore be considered more, or at least equally, relevant. Both decisions were not acknowledged by the Dutch Supreme Court. 23 The same applies to the decision of the Patents Court in 2013, which reveals that a new invalidity attack on EP 066 was staged in the United Kingdom, based on expert evidence submitted in the Dutch proceedings. It is telling that, despite the previous decisions, the Patents Court considered this new attack to have sufficient flesh to the bone to warrant 21 This also does not follow from e.g. Dutch Supreme Court 15 April 1966 NJ 1966/439 (Hoechst v. Nogepha). This case was also rendered before Dutch patent law, in 1978, lifted the ban on substance patents. The Dutch Supreme Court in Lundbeck v. Tiefenbacher also does not refer to Hoechst v. Nogepha. 22 House of Lords 25 Feb [2009] UKHL 12 (Generics (UK) Limited and others v H. Lundbeck A/S) par Patents Court 4 May 2007, [2007] EWHC 1040 (Generics (UK) Limited and Ors v H. Lundbeck A/S); Court of Appeal 10 April 2008 [2008] EWCA Civ 311 (H. Lundbeck A/S v. Generics (UK) Limited and Ors).

5 februari 2014 berichten industriële eigendom 45 a real prospect of successfully contending that the patent lacked inventive step. 24 b. The reference to the Bundesgerichtshof also constitutes less authoritative precedent, than the Dutch Supreme Court suggests: the decision discloses that this is one the rare cases in which the BGH fasttracked the proceedings and proceeded with judgment based on oral expert evidence, instead of the usual written expert opinion. The legality of the decision of the BGH to which the Supreme Court refers was even formally contested because of this, citing violations of due process. 25 In view of the above, the reference to the Bundesgerichtshof is less persuasive than the Dutch Supreme Court assumes. It also begs the question why the Court did not mention the contrasting views of the Bundespatentgericht which held EP 066 to be invalid. 26 In Dutch literature, scholars such as Brinkhof, have long called upon the Dutch Supreme Court to more proactively acknowledge foreign decisions of other patent courts. 27 In that perspective, the Supreme Court s approach to explicitly recognize and refer to authoritative precedent abroad is to be applauded. However it should not bar the Court from itself verifying whether such precedent is also persuasive and (still) relevant to the case at hand. 28 By doing so, the Court would not unnecessarily expose itself to the kind of criticism, the present ruling is already facing today Similar concerns also pertain to the fact that the Supreme Court did not explain in which way the case law of the Technical Boards of Appeal supports its position. 30 It merely referred to the decisions mentioned 24 Patents Court 12 April 2013 [2013] EWHC 739 (Resolution Chemicals Ltd v H. Lundbeck A/S) par , 140. Leave for full trial was confirmed after the Supreme Court s decision in appeal: Court of Appeal 29 July 2013 [2013] EWCA 924. The case later settled. 25 BGH 12 Nov Xa ZR 130/ BPatG 27 Aug Ni 9/05, 3 Ni 22/06, 3 Ni 54/06, 3 Ni 13/ J.J. Brinkhof, Wensen omtrent de wijze van motiveren door het Unified Patent Court, BIE 2013 p. 159, recently criticizing the quality of patent decisions of the Dutch Supreme Court and also previously in J.J. Brinkhof, Over The Desire for Harmonisation en The Avenue to Disunity. Zit de Nederlandse octrooirechtspraak wel op het juiste spoor?, Spoorbundel 2007, p with further references. 28 See also R.E. Ebbink, Beschermingsomvang in Nederland anno 2012, B who criticized the Supreme Court for disregarding foreign precedent and held that acknowledging decisions of the BGH and the Supreme Court on scope of protection is a must: Not to slavishly follow those two, who also themselves are not entirely in agreement, but to engage itself with those big brothers in a auseinanderzusetzen, as the Germans say (inofficial translation. In Dutch: Niet om die twee die onderling ook niet geheel overeenstemmen klakkeloos te volgen, maar om zich met de grote broers auseinanderzusetzen, zoals de Duitsers dat zeggen. ). Although Ebbink made its comments on scope of protection, the message is the same: acknowledging foreign precedent is good, slavishly following foreign precedent is bad, acknowledging and explaining the relevance of foreign precedent is best. 29 See e.g. Th.C.J.A. Van Engelen, Kun je wat bekend is uitvinden?, Ars Aequi 2014 p , casting significant doubts on whether the foreign precedent the Dutch Supreme Court refers to, is on point. 30 This triggers Van Engelen, Ars Aequi 2014 p. 56 to argue that e.g. T-595/90 is off-point as the absolute substance protection suggested in that decision merely involved a known desideratum. by the Supreme Court s Advocate-General. By not explaining why, and where, the Court of Appeal erred, the Supreme Court leaves right holders, and righteous citizens, guessing. The only decision the Advocate-General expressly cites from is case T-595/90. This precedent was however already acknowledged by the Court of Appeal. If the Supreme Court felt the Court of Appeal misapplied the legal standard contained therein it should state so, and why. Doubts as to the validity of the caselaw of the Technical Boards of Appeal, also comes from the Boards of Appeal itself. This follows from the article of Moufang, himself a member of the Boards of Appeal, in GRUR, signaling that, in relation to escitalopram, things are less clear-cut than the so-called standard case-law of the same Boards of Appeal suggest. 31 Practical significance 28. For legal practice in the Netherlands, the upshot is that the Supreme Court expressly adopts the approach set out in T-595/90 and subsequent decisions of the Technical Boards of Appeal. Even a product or substance is obvious in the sense that its composition and possible characteristics can be envisaged as such, it may nevertheless be non-obvious and claimable as such if there is no known way or applicable (analogy) method to make it and the claimed methods for its preparation are therefore the first to achieve this in an inventive manner. Inventors developing a (novel and) inventive process for obtaining a (novel) product which can be fully envisaged but not yet made, can therefore obtain not only a process claim for inventing the process, but can also monopolize the substance under a product claim. The result: even if a competitor would use a different method to obtain the same substance, this would still infringe the product claim, and can therefore be prohibited. Amsterdam, February See R. Moufang, Stoffschutz im Patentrecht, GRUR 2010, p