Combination SPCs: shaken and stirred

The creation of supplementary protection certificates (SPCs) in the EU in the 1990s was a significant victory for innovators in the medicinal and agrochemical fields. By providing up to five years of extra patent protection for authorised medicinal and plant protection products, which require market authorisation (MA) before entry on the market, SPCs are some of the most valuable IP rights available. Their aim is to compensate patentees for the period of a patent term that is lost during the lengthy regulatory processes.

A key provision of the SPC regulation (EC 469/2009) is article 3(a), which stipulates that the “product” (ie, active ingredient or combination of active ingredients of the medicinal product) that is the subject of the SPC must be “protected by a basic patent in force”. However, despite its fundamental importance, it remains unclear what this definition actually means.

Since SPCs are governed by EU legislation, whenever national courts are unclear on how the law is to be interpreted, clarification may be sought by referring questions to the Court of Justice of the EU (CJEU). The CJEU has already handled several references on this issue (including on July 25, 2018, as discussed later), and another two are currently pending.

Spectrum of specificity

In a case in the English Court of Appeal that led to one of the two pending CJEU references (Sandoz v Searle), Thomas Mitcheson QC summed up the issue of the definition of article 3(a) as a “spectrum of specificity”. Table 1 summarises this ‘spectrum’ by reference to recent SPC case law, beginning at the top with generic terms covering the active ingredient(s) and increasing the specificity of the definition moving down the table.

At the broad end of the spectrum, Medeva (C-322/10) makes it clear that the active ingredients must be “specified in the wording of the claims of the basic patent” in order to be “protected”, and hence there has to be at least some identifiable definition of each active ingredient in the claims themselves. It is not sufficient for the product merely to infringe the claims of the patent.

At the narrow end of the spectrum, there would appear to be little doubt that active ingredients expressly named or defined by a full structural formula would be adequately “specified in the wording of the claims” in order to be “protected”.

The difficulty is in the middle: ie, what degree of specificity is required for active ingredients to be “protected”? For example, is it sufficient for a claim to incorporate a Markush formula covering a large number of compounds, one of which is the product for which an SPC is sought? Is it sufficient for the product to be defined in functional terms, and if so, how precisely should this function be characterised in the claims?

Until recently, only one CJEU judgment had given guidance here. In Lilly (C-493/12), the CJEU ruled that a functional formula is not precluded and that the claims must “relate implicitly but necessarily and specifically” to the active ingredient, taking into account the extent of protection rules (eg, article 69 EPC and the protocol thereto).

The referring UK court subsequently found that a functional definition of an antibody identified by its binding specificity was sufficient to meet the CJEU’s test. However, the CJEU’s phrase “relate implicitly but necessarily and specifically” is itself imprecise and has left many wondering what the intention was. It was inevitable therefore that, hot on the heels of Lilly, further references would follow.

Teva v Gilead

On July 25, 2018, the CJEU handed down its judgment in the case of Teva v Gilead (C-121/17), which provides additional clarification in relation to combination products.

Gilead obtained an SPC covering its anti-retroviral/anti-HIV combination product containing tenofovir disoproxil and emtricitabine, marketed as Truvada (a registered trademark). However, the basic patent (EP 0915894) on which the SPC was based made no reference to emtricitabine, and the only relevant claim recited: “27. A pharmaceutical composition comprising a compound according to any one of claims 1-25 (of which claim 25 recites tenofovir disoproxil) together with a pharmaceutically acceptable carrier and optionally other therapeutic ingredients.”

In the referring UK court case at the High Court (2017), Teva and others challenged the validity of Gilead’s SPC, asserting that the combination of tenofovir disoproxil and emtricitabine was not “protected by a basic patent” (article 3[a]) because emtricitabine is not specified in the wording of claim 27 and the term “other therapeutic ingredients” does not specify any active ingredient structurally, functionally or otherwise.

In contrast, Gilead argued that the product in question falls within the extent of protection of claim 27 and that the term “other therapeutic ingredients” relates “implicitly but necessarily” to emtricitabine.

The UK judge, Mr Justice Arnold, considered the definition of article 3(a) to be unclear and referred the following question to the CJEU:

“What are the criteria for deciding whether ‘the product is protected by a basic patent in force’ in article 3(a) of Regulation No 469/2009?”

Alongside this reference to the CJEU, Arnold proposed a possible article 3(a) test based on whether the active ingredient or combination of active ingredients embodied the “inventive advance” of the basic patent. In this proposed test, article 3(a) should be satisfied provided the “product” embodies that “inventive advance”.

A few months before the recent CJEU judgment, Advocate General Melchior Wathelet proposed that the active ingredients should be “specifically and precisely identifiable” at the priority date of the patent.

Two-pronged test

The judgment of the CJEU appears to adapt both of these approaches into a two-pronged test, as follows:

“Article 3(a) ... must be interpreted as meaning that a product composed of several active ingredients with a combined effect is ‘protected by a patent in force’ within the meaning of that provision where, even if the combination of active ingredients of which that product is composed is not expressly mentioned in the claims of the basic patent, those claims relate necessarily and specifically to that combination.

For that purpose, from the point of view of a person skilled in the art and on the basis of the prior art at the filing date or priority date of the basic patent:

Hence, in the context of combination products, the test for compliance with article 3(a) now includes two independent criteria: (i) the combination must “fall under the invention”; and (ii) each of the active ingredients must be specifically identifiable as of the filing/priority date.

“The court did not explicitly define what the criterion ‘fall under the invention’ means, but it appears that the combination per se must be the focus of the patent.”

The criteria are to be assessed from the point of view of the skilled person, making use not only of the claims (see paragraph 34 of the judgment) but also the description and drawings, viewed in light of the prior art and general knowledge of the person skilled in the art (see paragraphs 38, 43 and 47–49).

It is now up to the UK court to decide, on the specific facts of the case, whether the combination of tenofovir disoproxil and emtricitabine satisfies these criteria.

The combination must fall under the invention (criterion [i])
The court did not explicitly define what the criterion “fall under the invention” means, but it appears that the combination per se must be the focus of the patent. For example, if the focus of the patent was ‘A’, but in optional embodiments it was claimed in combination with ‘B’, then only ‘A’ and not ‘A + B’ would arguably “fall under the invention”.

Support for this can be drawn from paragraph 48, which states that it is necessary for the skilled person to understand whether “the product to which the claims of the basic patent relate is a specification required for the solution of the technical problem disclosed by that patent”. Thus, if the presence of ‘B’ is optional in the claims alongside ‘A’, then the specification ‘A + B’ is arguably not required for the solution of the technical problem.

This appears to echo Arnold’s “inventive advance” test, and the requirement for an active ingredient to “constitute the subject matter of the invention” from Actavis v Boehringer Ingelheim Pharma (C-577/13).

This criterion does not have clear basis in the SPC regulation, but the CJEU seems to draw support for it from articles 4 and 5, which outline the subject matter of protection of the SPC and the effects of the certificate. In particular, it seems that the intention is to ensure that the reward of the SPC is strictly aligned to the invention of the patent whose commercial exploitation has been delayed.

Thus, if the invention of the patent is solely focused on a single active ingredient ‘A’, the court appears to believe that an SPC to ‘A’ provides sufficient supplementary protection (when considering infringement of the SPC) as it can be used to oppose any marketing of any medicinal product containing ‘A’, either alone or in combination with other active ingredients (eg, ‘A + B’). Of course, it may be possible for ‘A + B’ to be considered a focus of the patent in addition to ‘A’ alone, for example, if ‘A + B’ represents a second invention. However, the court was silent on this.

By ensuring that the combination must be the focus of the invention, criterion (i) may act as a safeguard against ‘boilerplate language’ in patent specifications, which otherwise makes it easy to attempt to cover a broad range of further embodiments of the “invention” that do not themselves contribute to the solution to the technical problem.

Indeed, the judgment makes it clear that it is the court’s intention to prevent so-called “evergreening”, ie, the unwarranted extension of patent protection for products. In paragraph 42, the court discusses the unacceptable possibility of being awarded successive SPCs for products comprising a “protected” active ingredient in combination with other substances which do not constitute the subject matter of the invention.

We note that there is already jurisprudence under article 3(c) preventing the obtaining of further SPCs to evergreen a primary active ingredient by simply adding additional active ingredients, even if these are specified within the claims. For example, in Actavis v Sanofi (C443/12), the CJEU held that the basic objective of the SPC regulation was not met by granting a further SPC for a combination product (in addition to an SPC for a single active ingredient ‘A’) where the combination of active ingredients (‘A + B’) does not represent the core inventive advance. By its present judgment, the CJEU has provided a further safeguard against evergreening that is generally applicable to all combination products, ie, regardless of whether an earlier SPC has been obtained for a single active ingredient within the combination.

Although the term “falls under the invention” is used in the singular in the judgment, it is commonplace for basic patents to focus on multiple unified inventions. Previous judgments appear to make it clear that there are scenarios where more than one product may be “protected as such” by a patent under article 3(a) (see Georgetown (C-484/12)).

For example, this could be envisaged where independent claims are directed to (i) one or more active ingredients of a genus; (ii) active ingredients that represent equally alternative solutions; or (iii) both a single active ingredient and an independently inventive combination of that single active ingredient with another active ingredient. It is possible that further guidance from the CJEU may be needed to interpret the term “the invention” in such scenarios.

Each active ingredient must be specifically identifiable as of the filing/priority date (criterion [ii])
Again, the court did not explicitly define the meaning of the criterion “specifically identifiable”. It seems fairly clear from the statement itself that “specifically identifiable” does not require all of the active ingredients to be expressly written out in the claim, or even in the description and drawings (because account can be taken of the prior art when determining the skilled person’s understanding of the claim), but we are left with scant guidance on just how specific the identification needs to be.

In particular, it is not clear whether the test should be interpreted narrowly, requiring the skilled person as of the filing or priority date to have actual knowledge of the specific active ingredients to which the claims relate, because either (i) all of the active ingredients are expressly written out in the description or drawings; or (ii) the claim recites, for example, a genus (eg, a diuretic), and the description provides a non-exhaustive list of examples of that genus that the skilled person could supplement with other known examples of that genus, including the active ingredient(s) in question, from the prior art and his general knowledge.

In other words, must the skilled person be aware of the existence of each of the active ingredients and thereby be able specifically to identify each of them as an example of the terms in the claims?

At the other extreme, should the test be interpreted broadly to include the situation in which at least one of the active ingredients in the combination is not known at the priority date but is the result of a future invention (or the result of other non-routine research activity), subject to its still falling within the inventive advance or focus of the basic patent?

Paragraph 50 of the judgment appears to rule out this highly permissive interpretation by indicating that it should not be possible to take into account “results from research which took place after the filing date or priority date of the basic patent”, which runs counter to the objective of the SPC regulation. Hence, an innovator who arrives at a new embodiment by non-routine means, even very shortly after the priority date, would very probably need to make such a combination of new active ingredients the subject of its own patent to be eligible for later SPC protection.

Even if “specifically identifiable” does not permit the subsequent “discovery” of active ingredients by inventive means, would it nevertheless read on to what the skilled person would identify by routine means? The court did not explain whether the test would read on to what the skilled person would identify by routine means, nor indeed what might be considered routine. What may be key here is the choice of the word “identifiable”, given that in normal patent law the skilled person is empowered to arrive at new active ingredients by routine work, providing that this does not require the exercise of inventive activity.

In any case, it appears that patent applicants will need to make sure that they have provided clear and thorough explanations of how active ingredients can be considered “specifically identifiable” in order to stand the best chance of satisfying the requirements of criterion (ii), even in situations where the precise identity of one or more of the active ingredients has not been finalised as of the priority date.

On the other hand, writing into the specification all of the possible combinations that can be envisaged at that time may destroy one’s chances of obtaining later patents to those combinations.

Conclusions

It is not clear to what extent criterion (ii) would be applied to Markush claims, and whether every substituent of an active ingredient must be available to the skilled person at the filing or priority date in order for it to be “specifically identifiable”. We eagerly await the upcoming judgments in Royalty Pharma (C650/17) and Sandoz v Searle (C-114/18) and the possible clarity that they will bring.

Given the added burdens on national offices and SPC applicants to deal with the new two-pronged test, and the likelihood of increased litigation by generics companies, it seems that this judgment will leave the world of combination SPCs both shaken and stirred.

Joel Beevers is an assistant at Potter Clarkson. He can be contacted at:  joel.beevers@potterclarkson.com

Michael Pears is a partner at Potter Clarkson. He can be contacted at:  michael.pears@potterclarkson.com