Benefits and pitfalls of functional patent claims (and why the UK is out of step with the EPO on claim construction): Astellas v Teva [2023] EWHC 2571 (Pat)

http://ipkitten.blogspot.com/2023/11/benefits-and-pitfalls-of-functional.html

In the recent UK decision Astellas v Teva [2023] EWHC 2571 (Pat) Mr Justice Mellor in the High Court considered sufficiency, inventive step and infringement of Astellas’ formulation patent for mirabegron.  The construction of a functional limitation in the claim was central to the case, highlighting the benefits and pitfalls of functional claim language. The decision also deviates from EPO practice on sufficiency and the inventive step of selection inventions. Particularly, in construing the claim for sufficiency and inventive step, the judge imported the purported technical effect of the invention as a limitation into the claim. The claim was then found inventive based solely on the technical effect as the distinguishing feature. The patentee also found themselves in an inventive step and infringement squeeze. Particularly, it was not proved that the alleged infringement had the purported technical effect of the invention relied on for inventive step, despite all of the other features of the invention being present. To this Kat, the court’s approach was thus oddly circular and based on a failure to distinguish between the claimed features of an invention and its technical effect.  

Case background

Astellas alleged that sale of Teva’s and Sandoz’s generic mirabegron products would infringe Astella’s formulation patent EP (UK) 2345410. The patent related to a slow release formulation of mirabegron. Teva and Sandoz counterclaimed that the patent was invalid for insufficiency and lack of inventive step. 

The absorption of an oral drug may be affected by how much food a patient has in their stomach. The problem purportedly solved by the claimed invention was to reduce this “food effect” on the absorption of mirabegron. The defendants argued that it was not possible from the data provided in the application as filed to make a reasonable prediction (i.e. it was not “plausible”) that the claimed formulations did in fact reduce the food effect on mirabegron. The defendants also argued that the claim was obvious given that the claimed formulation would have been produced in routine formulation development. Finally, it was submitted that the claim was not in any case infringed, as the generic products did not have all of the claimed features. 

Construing the claim to include the purported technical effect

UK case law currently diverges from the EPO on the criteria for considering “plausibility” under the heading of either sufficiency or inventive step (IPKat). At the EPO, the standard of evidence for a technical feature is considered in sufficiency if the technical feature is claimed, e.g. as in a second medical use claim relating to the technical effect of treating disease. The standard of evidence for a technical effect is considered under inventive step if the technical effect is not a claimed feature (IPKat), e.g. in a product claim defined by structural features. Under recent Boards of Appeal EPO case law, the technical effect does not always even have to be mentioned in the application as filed in order to be relied on for inventive step (IPKat). By contrast, the UK courts currently consider the standard of evidence for both claimed and unclaimed technical effects of the invention relied on for inventive step under the umbrella test of plausibility (IPKat). 

Regardless of the construction point, the judge considered the relevant legal test to be that of “plausibility” under the heading of sufficiency. The judge was very clear that, in contrast to the EPO, and in line with UK case law, plausibility is a legal test under sufficiency in which it was necessary to establish the evidence for which an invention works (i.e. solves the problem it purports to solve) (para. 233). The judge concluded that it did not matter for the sufficiency point whether or not the functional limitation of “reduced food effect” was construed as an express limitation of the claim. The judge thus considered under sufficiency whether it was “plausible” that substantially all compounds following under the claim would have the technical effect of reducing the effects of food.

The UK ab initio plausibility test 

On “plausibility”, the judge considered the Supreme Court decision in Warner Lambert as requiring there to be “some reasons for supposing” that a claimed invention has its purported technical effect (para. 226). Importantly these reasons for believing also have to be “disclosed in the specification” (para. 231). The UK approach to the standard of evidence for a technical effect can be characterised as an “ab initio plausibility” standard, whereby it must be plausible in view of the application as filed that an invention has the purported technical effect. This case law puts the UK out of step with the EPO Enlarged Board of Appeal decision G 2/21 (IPKat) and its interpretation by the Boards of Appeal (IPKat).

The question of plausibility in the present case required consideration of a large quantity of data, helpfully summarised by the judge in a table in paragraph 276. In view of the application as filed and extensive expert testimony, the judge was convinced that there were sufficient reasons for supposing that the claimed formulations would lead to the desired technical effect of a reduced food effect. In view of these data, Mr Justice Mellor consequently rejected the insufficiency attack (para. 278). 

Inventive step for a selection invention

The judge next went on to consider the question of inventive step. The judge’s construction of the claim to include the technical effect of “reduced food effect” was critical to whether the claims could be considered non-obvious. 

The judge found that if the claim was not construed to include the feature that the formulation had a “reduced food effect”, it would have been obvious for a skilled team to produce formulations having all the other features of the claim. Particularly, the skilled person would likely make such formulations as part of routine formulation development (para. 413). It would be a matter of chance whether any of these formulations had a “reduced food effect” (para. 430). The judge considered that a claim lacking the “reduced food effect” limitation would be obvious in view of this possibility (para. 422). 

However, the judge did construe the claim as including the “reduced food effect” limitation. As established for plausibility, the judge was also convinced that the range of selected parameters specified in the claim would lead to a “reduced food effect”. The judge was further convinced that the skilled team would not know to select or test these parameters in order to produce a formulation with a reduced food effect (para 413). The judge concluded that the claim construed to include the technical effect of a “reduced food effect” was inventive. 

To this Kat, the judge has here failed to consider the appropriate criteria for the inventive step of a selection invention, such as the one defined by the claim in question. The invention was characterised by the judge themselves as the selection of the range of parameters (“dissolution rate”) for a formulation (para. 419). The selection of these parameters, as established by the judge under sufficiency, was to provide the unexpected technical effect of a reduced food effect (para. 277). Under normal inventive step analysis for a selection invention of this type, the selection of the parameter range should therefore be considered inventive regardless of whether the unexpected technical effect per se was construed as a feature of the claim. Particularly, it is established case law of the EPO Board of Appeal that the selection of range is inventive if it has an unexpected technical effect, regardless of whether it would have been obvious for the skilled person to randomly produce something falling within the range (EPO Guidelines for Examination, G-VII, 12). In the present case, the judge found it necessary for the claim to also explicitly specify the technical effect of the selected range as a limitation. This seems to be rather a circular approach to inventive step, and implies that the claim lacked essential features of the invention.

Inventive step / infringement squeeze

The judge thus found the patent sufficiently disclosed and inventive in view of the prior art. However, unfortunately for Astellas, Mr Justice Mellor went on to find the patent not infringed by the generics. The key question for infringement was whether Astellas had established that Sandoz’s formulation had the feature of “reduced food effect”. Construction was therefore also critical to the question of infringement. 

The judge had already concluded for sufficiency and inventive step that the claim should be construed as including the limitation that the formulation had a “reduced food effect”. For infringement, the judge noted that it needed to be proved that the alleged infringement had a reduced food effect (para 447).  Particularly, the judge noted that “plausibility is a different standard to that required to prove infringement and I have held that, to prove infringement, at least some reduction in food effect must be demonstrated” (para 447). Whilst the alleged infringing formulation had all of the other features of the claim, the judge was not convinced that a “reduced food effect” had thereby been proved (para. 447). The claim was therefore found not infringed. 

Again, the judge’s conclusion appears to imply that essential features of the invention were missing from the claim. The curious importation of the unexpected technical effect into the claim also had the effect of creating an obviousness/infringement squeeze for Astellas. For the judge, if the claim was alternatively construed as not including the feature of “reduced food effect”, it would be infringed but also obvious, given the conclusion that the skilled team would have produced such formulations by chance. As such, if the claim was infringed it would then become invalid for obviousness (para. 450). 

Final thoughts

To this Kat, UK patent law appears confused on the fundamental question of whether it is necessary for the inventive effect of an invention to be construed as a feature of the claim itself. The UK tendency is to consider the technical effect of an invention as a claimed feature of the invention. This tendency leads to the consideration of the evidence standard for both an invention and its purported technical effect under the same heading of “plausibility” in a sufficiency analysis. By contrast, the EPO considers the evidence standard for an unclaimed technical effect under inventive step (IPKat). 

The judge in the present case, by incorporating the technical effect of the invention as a feature of the claim, also found themselves in an oddly circular position on inventive step and infringement. The claim became inventive solely for the reason that it possessed the technical effect as a claimed feature. However, given that it had not been proved that the technical features of the claim would necessarily produce the technical effect, an alleged infringement having all of these features was not considered infringing. This analysis begs the question of whether the alleged infringing embodiment would be novelty destroying. To this Kat, a critical issue here was also a failure to recognise the invention as a selection invention based on an unexpected technical effect. Under EPO case law relating to selection inventions, it is not necessary for the technical effect of a selected range to be specified in the claim. 

Leaving aside the peculiarities of UK patent law, however, the present decision is interesting for another reason. The decision also highlights the potential benefits and pitfalls of functional claim language. In contrast to the US, broad functional claim language is generally acceptable in Europe. A functional claim can provide broader coverage than a claim defined only by structural features and may also be harder for potential infringers to design around. However, as demonstrated in the present case, a potentially huge pitfall of functional claim language is that, to prove infringement of such a claim, it is necessary to prove that the alleged infringement also has the claimed function. It is therefore wise to restrict functional claim language to clear and easy to measure functional properties. If the decision in Astellas v Teva is any guide, it would also be wise to be extremely careful with descriptions of purported technical effects of the invention in the specification, in case these are construed into the claim by an English judge. 

Further reading

Plausibility

• Plausibility demystified – a review of EPO case law before G 2/21 (Feb 2023)
• UK divergence from the EPO on plausibility (Sandoz v BMS), Part 1: Is the “plausibility” test the same for both sufficiency and inventive step? (July 2023)
• UK divergence from the EPO on plausibility (Sandoz v BMS), Part 2: Interpretation of G 2/21 (July 2023)
• Chocolate teapots and nuclear bombs: Whole range sufficiency of mechanical inventions (T 0149/21) (July 2023)
• Interpretation of G 2/21: Inventive step may be supported solely by post-published data (T 0116/18) (Sep 2023)
• Reliance on a silent technical effect: Application of G 2/21 to semiconductors (T 2465/19) (Oct 2023)

Image credit: DALLE-3

Content reproduced from The IPKat as permitted under the Creative Commons Licence (UK).