The Medeva case
To understand how this issue has developed it first helps to examine the aspects of the CJEU’s ruling in the Medeva case1 that concern Article 3(a). Medevaconcerned applications in the UK Intellectual Property Office for five SPCs. These sought protection for different permutations of combined vaccines against diphtheria, tetanus, whooping cough, poliomyelitis and/or meningitis.
In support of these applications, Medeva cited a number of marketing authorisations. Common to the marketing authorisations was the combination of two things: pertactin and filamentous haemagglutinin antigens, although each authorisation contained other antigens in varying combinations. The European patent on which Medeva based its SPC applications was for a method of preparation of an acellular vaccine against Bordetella pertussis (whooping cough agent). This vaccine, as claimed in the patent, also consisted of a combination of the two antigens pertactin and filamentous haemagglutinin. But, unlike the combinations for which the marketing authorisations were granted, these were not combined in the patent with any other antigens.
In four of the five SPC applications made by Medeva, SPC protection was sought for combinations of more active components and ingredients than were covered by the claims of the patent. So, the question was: is the patent capable of being a ‘basic patent in force’, for the purpose of article 3(a) SPC Regulation in these cases?
Prior to Medeva, different approaches were taken to this question by the national courts. Most notably, German and Dutch courts adopted an ‘infringement’ test, which asked whether the product in issue would in theory infringe the basic patent2 . If so, article 3(a) would be satisfied. The courts of France and the UK3 , on the other hand, adopted narrower ‘subject matter’ tests, in which it is asked whether the product is part of the new and inventive subject matter of the patent.
In Medeva, the CJEU ruled in favour of the ‘subject matter’ approach. Importantly, the court says that the component active ingredients are protected by a basic patent in force if they are ‘specified’ in the wording of the claims. On the facts of this case, this meant the combinations for which SPCs were sought were not protected and the SPCs were not permitted.
Shortly after the Medeva ruling, University of Queensland4 , which concerned SPCs on products obtained from a patented process, followed the reasoning in Medeva. However, in Queensland, the word used by the CJEU to describe how the subject matter of the basic patent must be disclosed was ‘identified’, rather than ʽspecifiedʼ, although it is thought that these terms share the same meaning5.
These decisions have now led to a series of references to the CJEU trying to pin down when a product for which protection is sought is identified or specified in the basic patent in force. Must the active ingredients of a combination vaccine be identified explicitly and individually, or is it sufficient for them to be referred to generically or functionally?
Before addressing that question, it is necessary to summarise the CJEU rulings since Medeva, because these provide the backdrop to the latest cases. Not all of these concern Article 3(a) directly, but they are relevant and have done little to clarify the area:
- Georgetown University v Octrooicentrum Nederland6 addresses the question of how many SPCs are allowed per patent. In its decision, the CJEU rules that it is possible, in principle, to obtain several SPCs on the basis of the same basic patent, provided that each of the products for which the SPCs are sought are protected ‘as such’ by that patent;
- In Actavis Group v Sanofi7 the CJEU rules that when the patentee has already obtained an SPC on the basis of a marketing authorisation for a single product, the patent holder is precluded from obtaining a second SPC on the basis of the same patent for a combination product when the other active ingredient is, again, not protected ‘as such’ by the patent.
- In the very similar case of Actavis v Boehringer Ingelheim8 , however, the CJEU rules that when an SPC for an active ingredient that ‘constitutes the sole subject-matter of the inventionʼ of a basic patent has already been obtained, the holder may not obtain a second SPC for a combination of that claimed ingredient with any another active ingredient.
The Eli Lilly v Human Genome Sciences ruling
Eli Lilly v Human Genome Sciences (HGS)9 is actually concerned with a patent for therapeutic antibodies to neutrokine-α, rather than a combination10 or second SPC, but it is an important decision on the issue of interpreting Article 3(a) and has added to the uncertainty requiring the latest referrals.
Both the patent owner HGS, and Eli Lilly, had developed anti-neutrokine-α antibody products. HGS’s product, belimumab, was approved to treat systemic lupus erythematosus (SLE). Eli Lilly’s product, tabalumab, was in clinical trials but had not yet been approved. When approved, HGS would have the opportunity to apply for an SPC for Eli Lilly’s tabalumab, based on its own patent but relying on Eli Lilly’s marketing authorisation. To avoid this possibility, Eli Lilly sought a declaration that any such SPC would be invalid because tabalumab was not expressly named in the relevant HGS patent. Instead, Eli Lilly argued, the patent is drafted only in functional terms to claim any antibody to neutrokine-α, and this does not identify or specify tabalumab for the purpose of Article 3(a) and the Medeva ruling.
In its preliminary ruling, the CJEU explained that ‘protected by a basic patent in force’ under Article 3(a) of the SPC Regulation does not require a product to be identified by a structural formula (in this context, the amino acid sequence of the antibody) but that a functional definition may be acceptable if, once construed -
the claims relate, implicitly but necessarily and specifically, to the active ingredient in question.
The CJEU then placed the onus back on the national court to construe the claims and answer this test. The case was, accordingly, referred back to Warren J in the English Patents Court 11.
Warren J’s interpretation was, in this case, that functional definitions could, in principle, bring an active ingredient within the protection of a basic patent. He states that determining whether the claims ‘relate implicitly but necessarily and specifically’ to the active ingredient means that if the active ingredient is covered by the claims, according to conventional methods of claim interpretation, it is protected for the purposes of Article 3(a) of the SPC Regulation. On the facts of the case, it was conceded during the course of proceedings that tabalumab fell within the scope of claim 13 of the patent, even though it was not individually identified. Tabalumab was therefore protected by the patent within the meaning of Article 3(a).
But not every court agrees with this understanding of the Eli Lilly decision.
Differing national interpretations of Eli Lilly
The differing national approaches to Article 3(a) and the CJEU ruling in Eli Lilly have come to a head in further references to the CJEU: one from the English Patents Court, concerning Truvada and another concerning a Markush formula; as well as one from the German Federal Patent Court, concerning sitagliptin.
First, the Truvada cases. The Truvada cases concern an SPC held by the defendant, Gilead Sciences Inc. The SPC covers Gilead’s HIV treatment, Truvada, consisting of a combination of tenofovir disoproxil fumarate and emtricitabine in a fixed dose tablet.
In the German Truvada case in the Regional Court of Munich12 the issue of the validity of the SPC arose in a request for a preliminary injunction against a number of generics manufacturers. Specifically, the defendants referred to a qualified notice of the German Federal Patent Court in pending nullity proceedings13, in which the court had expressed its preliminary view that the SPC in question is not valid. This was on the basis that, applying the Medeva and in particular Eli Lilly case law, emtricitabine was not identified in a sufficiently specific way in claim 27 of the basic patent in force.
The Regional Court has agreed with the Federal Patent Court that in the case of combination SPCs, the requirement of Article 3(a) of the SPC Regulation can only be considered fulfilled, if two requirements are met:
- the combination of active ingredients must be the subject of the invention of the basic patent; and
- the active ingredients must either be identified in the claims of the basic patent by means of a structural formula, or they must be given a functional definition in the claims of the basic patent in a way that the claims can be considered to relate implicitly but necessarily and specifically to the active ingredient in question.
The Regional Court also agreed that the second requirement was not met, on the basis that a structural formula means the ‘explicit’ specification of the compound in the claims; while a functional formula only ‘implicitly’ specifies the compound. The Regional Court also holds that the first requirement – that the combination of tenofovir disoproxil fumarate and emtricitabine is the subject of the invention of the basic patent – is not met. This is because the basic patent only discloses tenofovir disoproxil as a mono-compound. This is a narrower approach than that taken by the English Patents Court in Eli Lilly.
National decisions on tenofovir
Decisions in parallel cases concerning tenofovir have also come from the courts of France14, Denmark15 and Switzerland16. At first instance in France, again, in the context of a preliminary injunction, the court held that the SPC is likely to be invalid, because the words ʽtherapeutic ingredientʼ did not give a functional definition of a compound and there was nothing in the patent specification to support emtricitabine as this ingredient. In addition, emtricitabine was not part of the invention defined by the subject matter of the basic patent. In Denmark, the court also refused a preliminary injunction against Accord on the basis that the SPC’s combination of tenofovir disoproxil and emtricitabine are not protected by the basic patent, because the words ʽother therapeutic ingredientsʼ in the relevant claim did not specify emtricitabine.
The Swiss court does something different. It has so far followed an infringement test for determining whether a product is protected by a basic patent in force17, of the kind ruled out by Medeva. In the Gilead / Truvada case, the Swiss Federal Patent Court saw no reason to depart from this test, because even though Swiss SPC legislation was drafted to harmonize with the SPC Regulation, as a court of a non-EU state it is not bound to follow CJEU authority. Using the infringement test, the Swiss court holds that the combination is protected by the basic patent in force and the SPC is therefore valid.
In a parting shot, the Swiss court was also critical of the inconsistent state of CJEU authority on this issue, calling it a ʽterminologisches Durcheinanderʼ – a terminological mess. A comment echoing others made by Arnold J in the UK.
The UK tenofovir referral
In the absence of a workable set of criteria and diverging national decisions on the application of the SPC Regulation in this area, the English Patents Court in its Truvada case has requested a preliminary opinion from the CJEU on the following simple question:
What are the criteria for deciding whether “the product is protected by a basic patent in force” in Article 3(a) of the SPC Regulation?
Pending a preliminary ruling by the CJEU, Arnold J suggests that the answer is to consider the ‘inventive advanceʼ or technical contribution of the basic patent. According to this approach, a product will satisfy Article 3(a) if ‘it contains an active ingredient, or a combination of active ingredients, which embodies the inventive advance (or technical contribution) of the basic patent’. When applied to the facts of the case, the combination of agents comprising Truvada does not, the judge suggests, embody the ‘inventive advance’ of the basic patent. Arnold J considers this approach to be consistent with the objectives of the SPC Regulation to encourage and reward innovation in medicinal products and the wording in the CJEU’s decisions in Actavis v Boehringer Ingelheim and Actavis Group v Sanofi, discussed above.
The German Sitagliptin case
It was mentioned above that the German Federal Patent Court has also referred questions to the CJEU for a preliminary ruling. These concern functional definition claims in a basic patent relating to the treatment of diabetes mellitus by administration of dipeptidyl peptidase IV (DP IV) inhibitors18. While it discloses a number of specific DP IV inhibitors, the patent also refers to the fact that other, unspecified, DP IV inhibitors may be used. One such inhibitor not specified by the patent, but which is marketed by a licensee for the treatment of diabetes mellitus, is sitagliptin.
On the basis that it is not individually disclosed by the patent, an SPC application for sitagliptin was rejected by the German Patent and Trade Mark Office. On appeal, the Federal Patent Court has stated that the CJEU rulings in Medeva and Eli Lilly mean that sitagliptin is not disclosed specifically enough to form part of the subject matter of the claims of the patent. However, aware of the conflicting approach taken by the English Patents Court in Eli Lilly19 and elsewhere, the court has referred three questions to the CJEU.
The questions referred ask, in essence, whether a product is only protected by a basic patent in force for the purpose of Article 3(a) if it is provided in the claims as a specific embodiment. If so, does this mean that for a claim using a general functional definition of an active substance class, it is not enough for the product to be individualized as a specific embodiment in the teaching of the basic patent? It is also asked whether, depending on the answer to these questions, a product is protected if it is developed using independent inventive activity after the filing date of the basic patent, but nonetheless falls within the functional definition of the claims.
Markush formulae – Prezista
Then there is the case concerning a Markush formula, heard most recently by the English Court of Appeal in Sandoz v GD Searle20. While it does not deal with combinations, the court discusses the Eli Lilly ruling in order to attempt to shed light on Medeva and Article 3(a). It decides, however, that it is necessary to make another referral to the CJEU.
In this case, the claimants challenged the validity of an SPC for ʽDarunavir or the pharmaceutically acceptable salt, ester or prodrug thereofʼ. Marketed as ʽPrezistaʼ, the product is a protease inhibitor used as an anti-retroviral medication to treat HIV infection. The basic patent in force, on which the SPC relied, claimed a Markush formula generically covering darunavir. So, darunavir was not specified or identified in any of the claims of the patent, by specific name or structure, or anywhere in the specification. There was also no teaching in the patent that pointed to it.
Arnold J, at first instance, Eli Lilly made it clear that it is not necessary for the active ingredient to be identified in the claim by means of a structural formula: it is sufficient for the active ingredient to be covered by a functional description provided that the claims ʽrelate, implicitly but necessarily and specifically, to the active ingredientʼ. In other words, it was not necessary for the claim individually to name or depict the active ingredient and a Markush formula is sufficient. Nonetheless, the judge also repeats his view, given in Truvada, that a better test than those so far set out by the CJEU, would be to ask whether the product falls within the claim and that it embodies the inventive advance (or technical contribution) of the claim.
The Court of Appeal, however, while regarding it as plain that a Markush claim can in some circumstances amount to a sufficiently precise claim for the purposes of Article 3(a) SPC Regulation, does not think that Eli Lilly provides a basis for saying that an ingredient can be adequately identified by a Markush formula however broadly it is framed and however obscure the relevant substituent is. This is for two reasons, in the court’s judgement:
- Firstly, if the objective is to ensure that the patent proprietor has come close to an actual realisation of the product, then the fact that the relevant substituents cannot be arrived at from a reading of the specification and the common general knowledge may be highly relevant;
- Secondly, it is likely from the sitagliptin decision that a German court would take the view that a Markush formula may, at least in a case like the present, fail to provide protection within the meaning of Article 3(a). A decision in the UK that a Markush formula would always be adequate would therefore lead to conflicting decisions at least in these two member states.
The Court of Appeal agrees with Arnold J that a better test would be to ask whether the product that is the subject of the SPC embodies the core inventive advance of the basic patent. In a case of a product with a single active ingredient and a patent with a claim which identifies a number of compounds by means of a Markush formula, all of which compounds embody the core inventive advance of the patent, the test should be whether the skilled person, considering the claims of the patent on the one hand and the structure of the product in question on the other, would immediately recognise that the active ingredient in question is one of those specified by the formula. The Court of Appeal thinks that this test would be satisfied in the present case, but because it is not clear that it is the correct approach under EU law, the court has referred it in those terms to the CJEU for a preliminary ruling.
How can these cases on Article 3(a) be summarised? It is clear that the national judges are having difficulty in interpreting the CJEU case law, which is of course very unsatisfactory for commercial decision makers. However, it could be said, very broadly, at the moment, that the English court is taking a broader approach to Eli Lilly than Germany and some of the other EU courts. It is therefore to be hoped that one or other, or preferably both, of the two new referrals to the CJEU bring some clarity on this issue. However, as will be appreciated, much of the problem is that these rulings themselves are inconsistent.
Arnold J has expressed the need for ʽbright-line rulesʼ in future cases. Hopefully, he will get these in the Abraxis, Truvada and sitagliptin cases, but past experience suggest this cannot be relied upon.