Generics (UK) Limited & Ors v AstraZeneca AB

Plausibility is not a freestanding validity requirement. It is the standard used to assess whether a patent makes a claimed technical effect sufficiently credible and whether the claimed invention makes a technical contribution. For a single chemical compound, the application read with common general knowledge must make it plausible that the compound has the asserted utility. A bare assertion or mere possibility is insufficient. Post-published or secret information cannot supply missing disclosure, although later evidence may test or confirm an effect already made plausible. A selection from a prior-art genus is inventive only where the selected compound makes a real technical advance through a useful property. Showing the same property, or merely making it plausible, does not overcome arbitrary selection.

AstraZeneca appealed against the decision of Dr Michael Tappin KC, sitting as a Deputy High Court Judge, which held the patent invalid on grounds of lack of inventive step and insufficiency: [2025] EWHC 1012 (Pat). The patent concerned dapagliflozin, an SGLT2 inhibitor, and related supplementary protection certificates.

The appeal concerned the interpretation of the patent’s assay disclosure; whether a product claim could attract a less demanding plausibility standard than a medical-use claim; the effect of G 2/21; and whether dapagliflozin was an arbitrary selection from the genus disclosed in WO 01/27128 A1. The central questions were whether the application made the asserted utility plausible and whether it disclosed a technical contribution over the prior art.

1. Disposition. The appeal was dismissed unanimously. The patent, and consequently the supplementary protection certificates, remained invalid.
2. Plausibility. Articles 56 and 83 of the European Patent Convention, given effect in the United Kingdom by the Patents Act 1977, do not themselves mention plausibility. The concept is a case-law standard. Applying the majority decision in Generics (UK) Ltd v Warner-Lambert Co LLC [2018] UKSC 56 and this Court’s decision in Sandoz Ltd v Bristol-Myers Squibb Holdings Ireland Unlimited Company [2023] EWCA Civ 472, the application had to make it plausible, when read with common general knowledge, that dapagliflozin had the claimed utility. The same standard applied to the product claim and the medical-use claim.
3. A bare assertion, an unspecified assay result, or a mere possibility was insufficient. The patent described a method using a generic reference to an inhibitor, without identifying dapagliflozin, reporting a result, or disclosing its potency. Repeated statements that dapagliflozin was an SGLT2 inhibitor useful for treating diabetes could not bootstrap the required disclosure.
4. The Court held that the skilled team would follow the patent’s cross-reference to WO 128 and then the cited Tanabe Seiyaku papers. Those materials disclosed strict structural requirements for activity and gave the skilled team a legitimate reason to doubt that dapagliflozin was an SGLT2 inhibitor, and therefore that it would treat diabetes. Later success, clinical evidence, or secret pre-filing data could not cure the deficient application.
5. G 2/21 and inventive step. The Enlarged Board’s formulation in G 2/21 [2023] OJ EPO A85 required the technical effect to be encompassed by the technical teaching and embodied by the same originally disclosed invention. The Court found that the Boards of Appeal and other national courts had not yet reached a settled interpretation capable of justifying departure from Sandoz v BMS.
6. Arbitrary selection. A selection from a prior-art genus is inventive only if the selected compound makes a real technical advance through a useful, different or improved property. Mere plausibility of the same asserted property is not itself a technical contribution. The patent’s assay disclosure was materially identical to WO 128, and AstraZeneca did not contend that dapagliflozin was superior to the compounds disclosed there, including Example 12. The selection was therefore arbitrary.

• Court of Appeal (Civil Division): Appeal dismissed. The Court upheld the finding that the patent was invalid for lack of inventive step and insufficiency. [2025] EWCA Civ 903.
• High Court, Patents Court: Dr Michael Tappin KC, sitting as a Deputy High Court Judge, held the patent invalid on both pleaded grounds. [2025] EWHC 1012 (Pat).