Patents Court considers validity of SPCs based on a Markush formula

The UK Patents Court has considered the validity of a supplemental protection certificate (SPC) based on a Markush formula, including the interpretation of Article 3(a) of the SPC Regulation. This article analyses the decision and considers how it sits in the context of recent cases.

Summary

Markush formulae are commonly used in patent claims as they are a means of presenting a structural formula that allows for variable constituents at one or more locations, thereby covering many permutations of constituents and a large (potentially very very large) number of conceivably associated chemical compounds. In the SPC field, they are often raised as an interesting basis on which to test the scope of the rules governing validity. In particular, considering the test under Article 3(a) of Regulation 469/2009/EC (the SPC Regulation), the formula in question would need to “protect” a particular “product” (that may be a single chemical compound), which is one of many permutations within its scope.

Continuing the recent flurry of SPC cases, Markush formulae recently got their day in court; before the UK Patents Court to be precise, with Mr Justice Arnold handing down the decision last month (Sandoz Ltd & Anor v GD Searle LLC & Anor [2017] EWHC 987 (Pat)). In this judgment, Arnold J held that the Markush formula did protect the product, despite maintaining his view that the test under Article 3(a) is not clear. In doing so, Arnold J reiterated his support for the test he proposed in Teva UK Ltd & Ors v Gilead Sciences Inc [2017] EWHC 13 (Pat), which is subject to a referral before the CJEU. If the decision in Sandoz is appealed and subject to timing, the CJEU’s response to the Teva referral may nonetheless impact on the case.

Background

G.D. Searle LLC (Searle) is the registered proprietor of SPC/GB07/038 (the SPC) which protected “Darunavir or the pharmaceutically acceptable salt, ester or prodrug thereof”. Darunavir, an anti-retroviral medication used in the treatment of human immunodeficiency virus (HIV), is marketed in Europe by companies related to Janssen Sciences Ireland UC (Janssen), the exclusive licensee, under the trade name Prezista.

The basic patent on which the SPC is based, EP (UK) No. 0 810 209 (the Patent) expired on 23 August 2013. Sandoz Limited and Hexal AG (the Claimants) brought an action against Searle and Janssen (the Defendants) for revocation of the SPC, which expires in February 2019, in order to clear the way for the launch of their generic darunavir product.

The Patent

The claims of the Patent are framed by reference to two Markush formulae, Formula I and Formula II. Formula I contains five variable substituent groups, and Formula II differs only in that it omits the stereochemistry of the left-hand two carbon atoms of the compound backbone. Formula I is as follows [in which P1, P2, R2, R3 and R4 each represent multiple variables]:


According to expert evidence, the Patent specifically disclosed approximately 100 compounds and darunavir was not one of them. Hence, darunavir was represented only by the Markush formulae in the claims, which covered a huge number of compounds, said to be between 7x10135 to 1x10377 compounds.

Article 3(a)

Article 3(a) of the SPC Regulation requires that: “the product is protected by a basic patent in force."

In considering the correct interpretation of Article 3(a), Arnold J referred to comments he made in the recent case of Teva v Gilead. In that case, he considered that the correct test to be applied in relation to Article 3(a) is still unclear, and made a referral to the CJEU for clarification. To assist the CJEU in this regard, he also proposed his own interpretation of Article 3(a), the key points of which he re-stated in this decision as follows:

• It is not sufficient for a product to be “protected” by a basic patent if dealings in the product would infringe a claim of the patent (applying what were referred to as “the Infringing Act Rules��).

• It is necessary that the product falls within at least one claim of the patent (applying what were referred to as “the Extent of Protection Rules”).

• It is not clear whether it is sufficient that the product falls within at least one claim of the patent. It appears from case law of the CJEU that it is not sufficient and that something more is required, but the tests established in Medeva (Case C-322/10, Medeva BV v Comptroller General of Patents, Designs and Trade Marks) and Eli Lilly (Case C-493/12, Eli Lilly and Company Ltd v Human Genome Sciences Inc) are unclear and difficult to apply.

• In Arnold J’s view, a product will be “protected” by the basic patent if (i) it falls within the scope of the claim, and (ii) it contains an active ingredient, or combination of active ingredients, which embodies the inventive advance (or technical contribution) of the patent.

The Decision

The key issue, in this case, was whether the Markush formula satisfied Article 3(a) when it was the only means by which the compound in question was disclosed in the Patent.

Sandoz accepted that their product would infringe the SPC if its validity was upheld (and also agreed to give 30 days’ notice before launch, notwithstanding which Searle nonetheless counter-claimed for injunctive relief).

Sandoz, therefore, relied on the second part of the test, ie that infringement was not enough. In this respect, Sandoz relied on the fact that darunavir is not specifically identified in the Patent, nor is there any teaching in the Patent directed to darunavir. In applying Article 3(a) to Markush claims, Sandoz contended that if the skilled person is not directed towards the product through the teaching of the patent (combined with their common general knowledge at the priority date), then the product is not “protected”. However, accepting that the test for interpreting Article 3(a) was not acte clair, Sandoz suggested that a referral to the CJEU was required (to follow the Teva referral).

In contrast, Searle contended that the lack of clarity on the correct interpretation of Article 3(a) was irrelevant to the present case because under the broadest tenable interpretation Daruvanir would nonetheless be protected by the Patent.

Arnold J agreed with Searle. While urging the need to produce a clear test (via the Teva referral), he held that in this case it could be seen that the Patent satisfied the test set by the CJEU in Eli Lilly, namely that it is sufficient for the active ingredient to be covered by a functional description provided that the claims "relate, implicitly but necessarily and specifically, to the active ingredient”.

He held that it is not an objection that the relevant claim covers a large number of compounds in addition to the active ingredient in question.

Arnold J also dismissed Sandoz’s objections based on the breadth of the Markush formula as being essentially an attack on excessive claim breadth, which goes to the validity of the Patent and not the SPC. As the Claimants had not put the validity of the Patent in issue in the proceedings, the presumption of validity prevailed. The judge also noted that it is not the job of patent offices to consider the justification for the breadth of claims in the underlying patents when assessing SPC applications.

Comment

In respect of validity, the approach, in this case, might be said to be in contrast to another recent decision, also by Mr Justice Arnold, in Teva v MSD (Teva UK Ltd v Merck Sharp & Dohme Corp [2017] EWHC 539 (Pat)). In that case, a detailed CGK analysis was conducted in order to assess whether a triple combination product constituted a “distinct invention” for the purposes of Article 3(c). Despite Arnold J’s reservations about adducing expert evidence, given the need for “a simple and transparent system for the grant of SPCs”, he nevertheless considered a validity-style assessment of whether the claimed combination was “obvious to try”. These decisions raise the question of whether the Court’s willingness to consider such issues will depend on the particular facts of the case, or the type of validity in issue (ie excessive claim breadth vs. obviousness).

Another interesting aspect of this case was the quantity and type of evidence submitted. Sandoz served an expert report as evidence in chief, to which Searle replied with its own expert report. Searle also served four witness statements containing evidence of foreign law, which covered the treatment of patent extensions for Markush claims in the EU, US, Japan and South Korea. However, Searle’s primary position was that no expert evidence was required in order decide this case, with which Arnold J ultimately agreed.

Regarding the foreign law evidence, and notwithstanding the fact that there was no dispute on foreign law, Arnold J commented that he found the evidence of “no real assistance” on the basis that the dispute focussed on the correct interpretation of Article 3(a) having regard to guidance provided by the CJEU.

Considering expert evidence, Arnold J considered it but held it was also unnecessary. In this respect, it is interesting to compare this to recent cases. In Teva v MSD, two expert reports were submitted despite Arnold J’s opinion that, having found the SPC invalid on the basis of Article 3(a), adducing expert evidence on the question of Article 3(c) was not necessary. In Teva v Gilead, Gilead adduced expert evidence without the permission of the court in the form of a witness statement from an expert acting as an independent fact witness. Teva did not object to its admission (with the approval of Arnold J, who commented on the report’s brief and uncontroversial nature), and also served hearsay notices in respect of a number of scientific papers. In Merck (BL O/117/16 dated 12 January 2016), the UK IPO considered expert evidence submitted by the applicant (MSD) in order to assess Article 3(c) (further analysis of that case and the associated CJEU referral is available here). These cases demonstrate the diversity of approaches used in adducing expert evidence in SPC cases.

In conclusion, while in some respects it is refreshing not to add to the number of outstanding SPC referrals to the CJEU, this case lends further support to the proposition that the position under Article 3(a) and the correct test to apply remains unclear. The parties to this case and many other stakeholders will currently be submitting their comments on this issue via the legal study being conducted by the Max Planck Institute. In the meantime, and perhaps, in any event, litigation in this area testing the requirements of the SPC Regulation looks set to continue.