27 April 2018 - Authored by:Steven Baldwin
If the CJEU adopts AG Wathelet’s opinion, it will likely be more difficult for patentees to satisfy Art 3(a) SPC Regulation. AG Wathelet expressly rejects both the core inventive advance test suggested by the referring English Court and the scope of protection test proposed by Gilead. Instead, the AG sees the wording of the claims and the interpretation of that wording as the sole basis for determining whether the active ingredient is protected by the basic patent. According to the AG the decisive test should be whether at the priority date of the patent, “it would have been obvious to a person skilled in the art that the active ingredient in question was specifically and precisely identifiable in the wording of the patent claims.”
The AG offers little guidance on what is meant by “specifically and precisely identifiable” and, as mentioned above, has linked his test to the understanding of the patent claims by the skilled person at the priority date. Such an approach may significantly complicate the SPC granting assessment carried out by the EU’s national patent offices. Further, his opinion leaves substantial doubt as to the level of specificity required when relying on generic or functionally-defined claims in the basic patent. There is still therefore a need for further guidance on at least this issue and, if the CJEU adopts the AG’s opinion, the pending Sandoz and Hexal and Sitagliptin cases on Art 3(a) will still need to be heard by the CJEU. What is however clear from this decision is that the AG favours an approach which will result in a restrictive interpretation of Art 3(a) and will therefore make it more difficult for patentees to obtain SPCs for active ingredients which have not been explicitly identified in the claims. (The full content of this article is also available here)
Brief re-cap of the facts
Teva, Accord, Lupin and Generics (t/a Mylan) are attempting to invalidate Gilead’s SPC (SPC/GB05/041) in the English High Court. The SPC is for a composition containing tenofovir disoproxil (TD) with emtricitabine. Gilead’s combination product is TRUVADA® – an anti-retroviral used in treating HIV. Gilead relies on claim 27 of its basic patent (EP 0 915 894), which claims “a pharmaceutical composition comprising a compound according to any one of claims 1-25 together with a pharmaceutically acceptable carrier and optionally other therapeutic ingredients”. It is not in dispute that claim 25 discloses TD. However, emtricitabine is not referred to anywhere in the patent. Teva et al’s position is therefore that Gilead’s SPC does not satisfy Art 3(a). (Click here for the referring court’s decision)
Analysis of the AG’s opinion
This opinion unfortunately poses as many questions as it answers. AG Wathelet rejects the core inventive advance and scope of protection tests and appears instead to adopt a disclosure test. However, he provides little guidance on when his proposed Art 3(a) test will in fact be satisfied. The AG takes as his starting point the Medeva C-322/10 case and acknowledges that “the real question which arises in the present case is with what degree of specificity or abstraction a product is “specified” in the claims of the basic patent”. His suggested answer to this question is as follows:
“A product is protected by a patent within the meaning of Article 3(a) of [the SPC Regulation] if, on the priority date of the patent, it would have been obvious to a person skilled in the art that the active ingredient in question was specifically and precisely identifiable in the wording of the claims of the basic patent. In the case of a combination of active ingredients, each active ingredient in that combination must be specifically, precisely and individually identifiable in the wording of the claims of the basic patent.”
We note the following in relation to this decision:
- “Specifically and precisely identifiable”: considering the AG’s appreciation that the core question for this preliminary reference is the degree of specificity required to satisfy Art 3(a), it is surprising that he then proceeds to offer no clear test as to what is meant by “specifically and precisely identifiable”. Indeed, whilst he argues that the CJEU has previously sought to “limit interpretation of the wording of claims to a certain degree of specificity or abstraction”, he also accepts that the name of the active ingredient or its chemical composition does not need to be referred to expressly in the claims (see para 82). Similarly, he accepts that Art 3(a) permits functionally-defined claims, but then rejects the idea that functional claims to, say, diuretics can satisfy the Art 3(a) test (see footnote 47). Given this, the most that can be said is that the AG appears to be promoting a restrictive interpretation of Art 3(a). However from a practical perspective it is unclear when an active ingredient (or combination) will be “specifically and precisely identifiable” for the purposes of Art 3(a).
- Inventions or findings after the priority date may be excluded: the AG refers explicitly to the knowledge of the skilled person at the priority date of the basic patent. If the skilled person is assumed not to have knowledge of the marketed post-priority product when conducting the Art 3(a) analysis, it is questionable whether a patentee will be able to obtain an SPC for a product which is only identified after the priority date of the basic patent. One example of this would be an SPC application with respect to a product which is developed by a third party after the priority date of the basic patent (this may in fact have influenced the AG in his opinion, see footnote 51). Moreover, in cases where the active ingredient was developed by the patentee at a later stage, it will have to be decided carefully whether the application for a SPC should be based on a later patent claiming specifically this compound rather than a patent with broad generic claims.
- Variants: the AG also appears to have intended that his test could be applied where the SPC is based on a generic (e.g. Markush) claim. He accepts that multiple variants of a chemical substance may be claimed by a patentee but held that, in order for a variant to be protected by the basic patent, it must be obvious at the priority date that the variant was “specifically and precisely identifiable” in the claims. This test appears to be an attempt to mirror one of the two possible tests proposed by the English Court of Appeal to deal with Markush claims in the Sandoz and Hexal  EWCA Civ 49 case i.e. that the specific substituents necessary to form the active ingredient falling within the Markush claim must be amongst those which the skilled person could derive, based on their common general knowledge at the priority date, from a reading of the patent claim. However, the AG has misunderstood how the Court of Appeal was using the test in that case. The Court of Appeal held that the abovementioned “common general knowledge” test was to be applied after the scope of the Markush claim had already been determined and with the skilled person having assumed knowledge of the marketed post-priority product (see para 98 of the CoA judgment). The skilled person would then consider whether the marketed compound’s substituents fell within the scope of that formulae as interpreted at the priority date. This is separate from the question of whether it would have been obvious to the skilled person that the (potentially unknown) active ingredient would have been identifiable in the wording of the claims at the priority date.
- Claim amendments: the AG is clear (see paras 79 and footnotes 41 and 46) that a patentee should not be permitted to amend its patent claims post-grant to specifically claim a marketed combination and therefore satisfy the test for Art 3(a), as was attempted by the SPC holder in the Actavis v Boehringer C‑577/13 case. However, it is difficult to reconcile this viewpoint with the rest of the AG’s opinion. He relies on a disclosure-based test, with the focus being on the wording of the claims (see para 72). This being so, it is unclear why an expressly claimed combination should not satisfy Art 3(a). It is a long-standing principle of patent law that claim amendments can be made provided that they are allowable (and satisfy the requirements as to added matter, clarity, support etc.). Further, such amendments take effect ab initio. It is therefore unclear why the SPC in the Boehringer case should not have been granted if one adopts the “specifically and precisely identifiable” test suggested by the AG in this case.
- The UK decision in Eli Lilly v HGS: the argument run by Gilead in this case that Art 3(a) will be satisfied where the active ingredient falls within the scope of protection of the claims is essentially the approach that was adopted by Warren J in Eli Lilly v HGS  EWHC 2404 (Pat) when the case was referred back to the English High Court from the CJEU. The AG has clearly stated that, in his view, this is not the correct test i.e. falling within the scope of protection of the claims is a necessary, but not sufficient, condition for the purposes of satisfying Art 3(a). If the CJEU adopts the AG’s opinion then the clear steer from the European courts will be that the English decision in Eli Lilly v HGS was decided using the wrong legal test for Art 3(a) (and was perhaps therefore wrongly decided). Similarly, adoption of the AG’s test would suggest that the CJEU will likely favour the German Court’s more restrictive approach to generic claiming in relation to Art 3(a) set forward in the Sitagliptin case, as opposed to the broader approach taken by the English Court of Appeal in Sandoz and Hexal.
We will provide a further update as soon as the CJEU judgment is published.