Factual and Procedural Background
The Defendant, Company A, was proprietor of a European Patent relating to a compound called dapagliflozin, used in the treatment of diabetes, which expired in May 2023. The Patent was challenged by the Claimants, three generic pharmaceutical companies, on grounds of invalidity, specifically lack of inventive step and insufficiency. The Claimants contended that the Patent did not make it plausible that dapagliflozin is an effective SGLT2 inhibitor useful for treating diabetes and that it was an arbitrary selection from a broad class of compounds disclosed in an earlier patent application (WO 128) without demonstrating any advantage. The Patent and the earlier application disclosed assays for SGLT2 inhibitory activity but contained no assay results for dapagliflozin. After a seven-day trial, a Deputy High Court Judge held the Patent invalid on both grounds, and Company A appealed with permission on eight grounds. The appeal was expedited due to the commercial importance of dapagliflozin.
Legal Issues Presented
- Whether the Patent made it plausible that dapagliflozin is an SGLT2 inhibitor useful for treating diabetes, addressing the sufficiency and inventive step requirements.
- Whether the concept of plausibility, as developed in case law and the European Patent Convention (EPC), applies differently to claims to a single chemical compound versus claims comprising functional technical features.
- How to interpret and apply the decision of the Enlarged Board of Appeal in case G 2/21 concerning reliance on post-published evidence and the concept of plausibility in assessing inventive step.
- Whether the Patent constitutes an inventive selection or an arbitrary selection from a prior-disclosed genus of compounds, and thus whether it makes a technical contribution over the prior art.
- The appropriate standard of plausibility to apply to claims to the compound per se compared to claims to its use, including whether the standards differ for sufficiency of disclosure and inventive step.
Arguments of the Parties
Appellant's Arguments
- The Patent should be interpreted as disclosing that dapagliflozin had been tested in the assay described, and that the repeated references to dapagliflozin as an SGLT2 inhibitor imply positive assay results.
- The standard of plausibility established in Warner-Lambert is too high and should be revisited in light of G 2/21, particularly for claims to a compound per se.
- The plausibility standard for claim 2 (compound per se) should be lower (ab initio implausibility) than for claim 15 (use claim), which should follow the Warner-Lambert standard (ab initio plausibility).
- The Patent makes a technical contribution by rendering plausible the utility of dapagliflozin, distinguishing it from the prior art, even if WO 128 broadly disclosed a genus of compounds including dapagliflozin.
- The judge erred in his approach to arbitrary selection by comparing the Patent’s contribution only to the prior art’s assertions rather than the prior art’s technical contribution.
Respondents' Arguments
- The Patent does not disclose any experimental data or a priori reasoning sufficient to make plausible that dapagliflozin is an effective SGLT2 inhibitor for treating diabetes.
- The assay description does not identify dapagliflozin as the inhibitor tested, and the references to dapagliflozin as an SGLT2 inhibitor are bare assertions unsupported by data.
- The skilled team would consult WO 128 and the cited Tanabe Seiyaku papers, which contain no data supporting dapagliflozin’s activity and indeed suggest reasons to doubt it.
- Warner-Lambert remains binding law for sufficiency of medical use claims, and the standard of plausibility applies equally to inventive step and sufficiency in this context.
- The Patent is an arbitrary selection from the genus disclosed in WO 128, lacking any technical contribution or improved property over the prior art, and mere plausibility is insufficient to establish inventiveness.
Table of Precedents Cited
| Precedent |
Rule or Principle Cited For |
Application by the Court |
| Generics (U.K.) Ltd v Warner-Lambert Co LLC [2018] UKSC 56 |
Standard of "ab initio plausibility" in sufficiency of disclosure for medical use patents |
Applied as binding authority; the Patent must make it plausible that the invention achieves the claimed therapeutic effect. |
| Sandoz Ltd v Bristol-Myers Squibb Holdings Ireland Unlimited Company [2023] EWCA Civ 472 |
Application of plausibility standard to inventive step and sufficiency for single chemical compounds |
Followed and reaffirmed; the Patent must make plausible the utility of the compound; no difference in standard for claims to compound per se and use claims. |
| G 2/21 [2023] OJ EPO A85 (Enlarged Board of Appeal) |
Permissibility of relying on post-published evidence and interpretation of plausibility in inventive step assessment |
Interpreted as requiring the technical effect to be derivable from the application as filed; distinction between sufficiency and inventive step noted but no departure from plausibility principle. |
| Dr Reddy’s Laboratories (UK) Ltd v Eli Lilly & Co Ltd [2009] EWCA Civ 1362 |
Distinction between arbitrary and inventive selection from prior art genus |
Applied to find that a selection is inventive only if it makes a real technical advance over the prior art. |
| Generics (UK) Ltd v Yeda Research & Development Co Ltd [2013] EWCA Civ 925 |
Clarification of technical advance requirement in arbitrary selection cases |
Applied to confirm that a technical contribution must be plausible and not merely asserted. |
| Catnic Components Ltd v Hill & Smith Ltd [1982] RPC 183 |
Principle of patent interpretation as a unilateral statement by the patentee |
Applied to emphasize that patents are technical documents to be read by the skilled person with common general knowledge. |
| Akebia Therapeutics Inc v FibroGen Inc [2020] EWHC 866 (Pat) |
Context- and fact-dependent approach to what documents the skilled person would consult |
Applied to determine that the skilled team would consult prior art documents referenced in the Patent for plausibility assessment. |
| Actavis UK Ltd v Merck & Co Inc [2008] EWCA Civ 444 |
Exception allowing departure from Court of Appeal precedent if inconsistent with EPO Boards of Appeal jurisprudence |
Considered but not accepted as grounds to depart from Sandoz v BMS in this case. |
| Human Genome Sciences Inc v Eli Lilly and Co [2011] UKSC 51 |
Importance of harmonised approach with EPO Boards of Appeal |
Noted in context of arguments for following EPO jurisprudence but did not justify departure from established UK case law. |
| Pharmacia Corp v Merck & Co Inc [2001] EWCA Civ 1610 |
Identification of a new chemical compound alone is not inventive |
Applied to confirm that utility must be shown for patentability. |
| Idenix Pharmaceuticals Inc v Gilead Sciences Inc [2016] EWCA Civ 1089 |
Requirement of utility for patentability of chemical compounds |
Followed to emphasize that mere identification of a compound is insufficient without demonstrating utility. |
Court's Reasoning and Analysis
The Court began by reaffirming the binding nature of the plausibility standard established in Warner-Lambert and applied in Sandoz v BMS, requiring that a patent application, read with the common general knowledge, must make it plausible that the claimed invention achieves the asserted technical effect. The Court rejected the Appellant’s contention that G 2/21 justified a lower plausibility standard for claims to a compound per se compared to use claims, noting that the Enlarged Board of Appeal distinguished sufficiency and inventive step but did not prescribe different plausibility thresholds that would justify departing from established UK law.
In interpreting the Patent, the Court agreed with the trial judge that the assay description did not disclose that dapagliflozin had been tested, as it generically referred to an "inhibitor" without identifying the compound. The absence of any assay results or a priori reasoning to support dapagliflozin’s activity meant the Patent contained only bare assertions. The Court found the Appellant’s reliance on structural similarity to phlorizin insufficient to render the claim plausible, given expert evidence that small structural differences could result in large differences in activity.
The Court accepted the trial judge’s finding that the skilled team would consult the prior art document WO 128 and the Tanabe Seiyaku papers referenced therein, which contained no data supporting dapagliflozin’s activity and in fact suggested reasons for doubt. This reinforced the conclusion that the Patent did not make it plausible that dapagliflozin was an effective SGLT2 inhibitor for treating diabetes.
Regarding arbitrary selection, the Court upheld the principle that a selection from a prior art genus is only inventive if it makes a technical contribution by showing a new or improved property. Mere plausibility or assertion is insufficient. The Patent did not demonstrate any property of dapagliflozin that distinguished it from compounds disclosed in WO 128, including a closely related example. Therefore, the Patent was an arbitrary selection lacking a technical contribution.
The Court also addressed the Appellant’s arguments seeking to reinterpret the Patent’s disclosure or to apply a lower plausibility standard, rejecting them as inconsistent with binding precedent and unsupported by the evidence. The Court found no basis to depart from the trial judge’s factual findings or the established legal framework.
Holding and Implications
The Court DISMISSED THE APPEAL, affirming the trial judge’s decision that the Patent claims were invalid for lack of inventive step and insufficient disclosure.
The direct effect of this decision is that the Patent and the associated Supplementary Protection Certificates are invalid, preventing the Defendant from enforcing exclusivity over dapagliflozin based on this Patent. The Court emphasized that its decision was based strictly on the disclosure in the Patent and the applicable legal standards, without regard to subsequent commercial success or data not disclosed at the priority date. No new legal precedent was established; rather, the Court reaffirmed existing principles concerning plausibility, inventive step, sufficiency of disclosure, and arbitrary selection in patent law.