Court of Appeal upholds validity of pharmaceutical patent claims with functional limitations: FibroGen Inc v Akebia Therapeutics Inc [2021] EWCA Civ 1279

The UK Court of Appeal has reviewed the law on insufficiency and its application to claims with both structural and functional limitations. In doing so the court has explained how the concepts of plausibility and undue burden should be applied when assessing the sufficiency of claims of this type, in particular to the requirement that it must be possible to perform the invention across the breadth of the claim.

The appeal concerned six patents belonging to FibroGen, which had been divided into two “families”. The court had to consider issues of obviousness and infringement, as well as insufficiency. The appeal on insufficiency was successful, with the end result that one of the “Family A” patents was held to be valid and infringed by Akebia’s vadadustat product, when used to treat CKD (chronic kidney disease) anaemia. On the first two issues the Court of Appeal upheld the first instance judgment of Lord Justice Arnold.

The two key claims started with structural limitations – the products had to be heterocyclic carboxamide compounds selected from a group of a number of different types of carboxamides – and a functional limitation, which was that the compound “inhibits hypoxia inducible factor (IHF) prolyl hydroxylase enzyme activity”. The claims then continued to specify that the compound was to be used in the manufacture of a medicament for, or for use in, “increasing endogenous erythropoietin … in the … treatment of [CKD anaemia]”. One of the claims also required that the compound had to be a compound covered by a generalised “Markush” chemical formula. Thus there were additional limitations, both functional and structural.

The question that the Court of Appeal had to resolve when considering insufficiency, and the requirement that the skilled person should be able to perform the invention across the breadth of the claim, was whether the group of compounds covered by the claim extended to all those compounds that satisfied the structural limitations, or whether the group was limited to those compounds that satisfied both the structural and the functional limitations.

The first instance judgment
In his first instance judgment, Arnold LJ considered that the Court was required to apply a two-stage enquiry: first to determine whether the disclosure of the patent in suit, read in the light of the common general knowledge, made it plausible that the invention would work across the scope of the claim; second, if it was plausible, whether the invention could be performed across the scope of the claim without undue burden. Arnold LJ considered that the Patent Court was bound by an earlier decision of the Court of Appeal in Idenix v Gilead and that this prevented the court from taking the functional limitation of the claim into account when determining which compounds were covered by the claim. Arnold LJ accordingly held that the patent implicitly promised that all compounds that satisfied the structural limitations would have the therapeutic efficacy. However, it was not plausible that this would be true. Therefore the first stage of the enquiry was not satisfied.

Neither was the second stage satisfied: it would require a very substantial research project to identify compounds that worked, other than those specifically identified in the patent, and success would not be guaranteed. It was probable that the skilled person, given sufficient resources, would be able to identify some compounds which were effective but this would only be a tiny fraction of all the compounds claimed.

The Court of Appeal’s approach
The Court of Appeal disagreed with this approach. Its judgment includes an extensive review of UK and EPO case law. The court held that the legal requirement was that it must be possible to make a reasonable prediction that the invention will work with substantially everything falling within the scope of the claim. This required, first, the identification of what fell within the scope of the claimed class; second, a determination of what was meant by saying that the invention “works”; third, in the light of the first two steps, an assessment of whether it was possible to make a reasonable prediction that the invention will work with substantially everything falling within the scope of the claim.

Importantly, when considering the first step – the identification of what fell within the scope of the claimed class – it was correct to take some types of functional features into account. The Court of Appeal drew a distinction between functional features that delineate the class of compounds claimed (which it referred to as “step one functional features”) and those that relate to the ultimate purpose for which the compounds are to be used (“step two functional features”). It was not the law that the skilled person had to be able to identify substantially all compounds that satisfied the step one structural and functional features. For example, functional language can mean that the claim covers compounds which are yet to be invented. What was necessary was that the skilled person could, without undue burden, identify compounds within the claimed class (other than those specifically identified in the patent itself). Furthermore, it had to be possible for the skilled person to work substantially anywhere within the whole claim: it must be possible for the skilled person, given any sensible compound within the structural class (or substantially any), to apply the tests without undue burden and work out if it is a claimed compound.

In the present case, the step one functional requirements features of the claim could be determined with an appropriate biochemical in vitro assay (for HIF-PH inhibition) and with a suitable animal model (to determine that there was an increase in endogenous EPO). How easily these tests could be done was relevant to the question of undue burden but not to the determination of the class of compounds covered by the claim. As for the second step – did the compounds work – this meant treating the CKD form of anaemia (i.e. this was the step two functional feature of the claim). The third step of the approach set out by the Court of Appeal then required the court to decide whether it was possible to make a reasonable prediction that the compounds that satisfied both the structural and step one functional features of the claims would be useful to treat CKD anaemia. On the evidence this was indeed plausible.

As for the requirement that the invention can be performed without undue burden, with claims of this type, it had to be possible to identify some other compounds, beyond those individually identified in the patent, which fell within the claimed class of compounds and therefore were likely to have therapeutic efficacy. In addition, it must also be possible for the skilled person to work substantially anywhere in the whole claim and, therefore, given any sensible compound within the class of compounds covered by the structural features, that the skilled person would be able to apply the functional tests without undue burden, in order to work out if the compound fell within the claimed class of compounds.

On the evidence and the findings of the first instance court, the Court of Appeal held that the skilled person (or team) would be able to find some compounds which were effective. Although it would take a great deal of work, it would be routine for the medicinal chemist and iterative in nature. That was not, in the view of the Court of Appeal, an undue burden.
The end result was that the patent was found to have met the requirements for sufficiency and the relevant claims were valid.

Comment
The clarification by the Court of Appeal that functional limitations can be taken into account when assessing sufficiency is to be welcomed. It is not always possible, in particular where an invention relates to a general principle rather than a specific product, to identify exhaustively all products that will work, yet it is often a relatively simple matter to test the activity and efficacy of any one product. In such circumstances, inclusion of a functional limitation can provide the patentee with fair protection commensurate with the contribution that they have made to the art. Were the patentee required to limit the patent’s claims only to those products which had been demonstrated to work, it could be easy for competitors to design around the patent despite taking the benefit of the invention. On the other hand, if the practical application of the functional test does impose an undue burden on the skilled person, the patent will be invalid, which will continue to prevent patentees from obtaining an unjustified monopoly on “anything that works”.

This decision also needs to be balanced against the judgment last year of the Supreme Court in Regeneron Pharmaceuticals Inc v Kymab Ltd, which by a majority held that a claim to a range of products (in that case, genetically engineered mice) was insufficient if the disclosure of the patent only enabled some, rather than all, of those products to be made without undue burden, notwithstanding that all of the claimed products would, if made, share the beneficial properties disclosed by the patent. This judgment of the Court of Appeal appears to be consistent, as the court has held that it must be possible for the skilled person to work substantially anywhere in the whole claim. It therefore appears that claims that cover a number of discrete classes of compounds, including classes that in fact cannot be put into practice by the skilled person, will still be vulnerable to a finding of insufficiency.