Akebia and Otsuka (Akebia) have a HIF-PH inhibitor called vadadustat, which is in Phase III clinical trials for anaemia. In an attempt to clear the way for UK launch, they sought revocation of FibroGen’s EP(UK) HIF-PH inhibitor patents.

First instance

In assessing whether the claims were excessively broad (meaning the invention cannot be performed across the breadth of the claim without undue burden), Arnold LJ reviewed relevant case law, and concluded that a two stage test needed to be carried out:

1. Does the disclosure of the patent make it plausible that the invention will work across the scope of the claims?
2. If the first answer is yes, could the invention be performed across the scope of the claims without undue burden?

Arnold LJ also found that case law precluded the court from taking functional limitations into account, when determining the scope of the claims; and the skilled person would need to identify substantially all compounds covered by the claim without undue burden, to answer yes in the second stage. For those reasons, he found that the patents implicitly promised substantially all compounds having the structural limitations (~10183 compounds!) have the claimed therapeutic efficacy. However, the patents only demonstrated five compounds having the claimed therapeutic effect, and no structure-activity relationships (SAR) had been disclosed.

Therefore, Arnold LJ found that the claims were insufficient (and “AgrEvo obvious” [see note 1 at the end of this article] for the same reasons), because it was implausible that all compounds having the structural features would have the functional features or therapeutic efficacy.

In case he was wrong, Arnold LJ also considered whether the skilled person could identify substantially all compounds covered by claim without undue burden. However, he asserted that a substantial research project would be required to identify more than a tiny fraction of additional compounds that work, with no guarantee of success. Therefore, the claims were found to be invalid again.

Appeal

FibroGen appealed against this decision of insufficiency.

In relation to plausibility, it argued that functional features are limiting; and it was plausible that compounds with the structural and functional features have the claimed therapeutic effect. Also, the patents do not imply that all compounds with the structural features work, because the technical contribution was not the known compounds in themselves. Instead, it was that certain classes are HIF-PH inhibitors, and may be used in treating anaemia by increasing endogenous EPO.

In relation to undue burden, FibroGen argued that identifying substantially all compounds is the wrong test, and inevitably causes an undue burden. FibroGen also argued that SAR analysis and lead optimisation are merely routine and iterative.

Plausibility

Regarding plausibility, the Court of Appeal acknowledged that defining an invention using general terms is permissible, if it is plausible the invention will work across the scope of the claims.

Also, a patent can be granted for a class of compounds, when a generally applicable effect has only been demonstrated for a few individual compounds; unlike when no general application has been demonstrated (following the House of Lords’ 1997 decision in Biogen v Medeva).

On that basis, the Court of Appeal reasoned that it must be possible to reasonably predict the invention will work with substantially everything in the scope of the claim. Therefore, the Court of Appeal decided to apply a three step test:

1. What falls within the scope of the claimed class?
2. What does “working the invention” actually mean?
3. On that basis, can the skilled person make a reasonable prediction that the invention will work with substantially everything falling within the scope of the claim?

In applying step 1, the Court of Appeal decided that the claims did cover compounds having the structural and functional features specified; and the functional requirements could be determined by tests that were well described in the patent.

In applying step 2, it was clear that “working the invention” meant treating the claimed form of anaemia.

Therefore, the appropriate question to ask in step 3 was whether the skilled person could reasonably predict that compounds satisfying the structural limitations, and the functional limitations (determined by tests described in the patent), are useful in treating the claimed forms of anaemia. On the evidence, the Court of Appeal found that a reasonable prediction could be made. Therefore, the first instance conclusion of implausibility was incorrect.

Undue burden

In relation to undue burden, the Court of Appeal acknowledged that claims may be defined by functional limitations, if the invention cannot otherwise be defined more precisely without unduly restricting the scope of the invention. However, Arnold LJ’s conclusion that the skilled person has to identify substantially all compounds covered by the claim was questionable.

Having reviewed case law, the Court of Appeal recognised that it is admissible to recite a group of substances in generalised form (for example, functionally), even if some substances are unsuitable for the purpose, provided the skilled person can easily determine their suitability by experiments. Also, it does not matter if some compounds within the generalised class have not been discovered yet, provided using them makes use of the invention. As functional language inherently covers undiscovered compounds, the skilled person cannot be required to identify substantially all compounds covered by the claim.

Ultimately, the Court of Appeal came to the conclusion that if compounds are defined by functional features, the reader must be able to (without undue burden):

1. Identify further compounds that satisfy the relevant test; and
2. Work substantially anywhere within the claim.

Applying these to the FibroGen patents, the Court of Appeal concluded that it was possible to identify other compounds that work. Although it would require a lot of effort, SAR analysis and lead optimisation are routine and iterative, and performing the tests in the patents is not unduly difficult. Therefore, the skilled person could identify further effective compounds without undue burden.

The Court of Appeal also decided that the skilled person would be able to work substantially anywhere in the scope of the claims, because there was no evidence that particular regions of the claim scope cannot be tested for the functional features (consistent with the Supreme Court decision of 2020 regarding Regeneron Pharmaceuticals v Kymab decision).

Therefore, the Court of Appeal concluded that the claims were not insufficient for excessive claim breadth.

Comment

In summary, this decision confirms that functional features can be limiting on the scope of a claim. However, the skilled person must be able to reasonably predict that the invention will work with substantially everything falling within the scope of the claim; and the reader must be able to identify further compounds that satisfy the relevant test, and work substantially anywhere within the claim, without undue burden.

This is a clear reminder when drafting that we should include sufficient information about tests for functional features, as well as information about general applicability of beneficial effects. However, we should be careful to avoid overly broad claims covering regions that cannot be tested.

Note

1. AgrEvo obviousness arises when claims lack a technical contribution (based on T 939/92). If an alleged technical effect is not credibly achieved for substantially everything falling within the scope of a claim, the effect cannot be used to formulate the technical problem for the assessment of obviousness. However, it is not inventive to merely provide new compounds with no beneficial properties.