The recent Court of Appeal judgment in Generics (UK) Ltd. & Ors v AstraZeneca AB [2025] reaffirms that plausibility is a standalone requirement for patentability in the UK and that the applicable standard for assessing plausibility is that laid down in Warner Lambert v Generics [2018] (pregabalin) and Sandoz v BMS [2023] (apixaban). In particular, the Court considered that the test laid down by the EPO’s Enlarged Board of Appeal in G 2/21 does not justify a change to the UK courts’ approach to plausibility.
In this article, we review the background and then some of the key points of the case, before turning to look at the practical implications this judgment has for patent drafting, IP strategy, and due diligence. The case serves as a useful reminder that, without experimental data or convincing reasoning in the application as filed supporting that a technical effect can be achieved with the claimed subject matter, it is unlikely that the threshold for plausibility in the UK will be met. Applicants should be mindful of this higher standard in the UK relative to the EPO, where post-filed data can be relied on for inventive step, subject to the conditions laid down in G 2/21.
Background: the plausibility hurdle in the UK and Europe
Since 2018, the test for plausibility in the UK has been dictated by the Supreme Court’s Warner Lambert principles: that a bare assertion of a technical effect is not enough, and it must be demonstrated in the application as filed that there is a reasonable prospect the attainment of a technical effect is true (e.g. by experimental data or a technical rationale).
The EPO introduced a significantly softer stance in 2023 in its G 2/21 decision, disowning the term “plausibility” altogether. The G 2/21 test dictates that a technical effect relied upon for inventive step must merely be “encompassed by the technical teaching” and “embodied by the same originally disclosed invention” as the application as filed. Some EPO Boards of Appeal have since interpreted this as meaning a bare assertion is enough (see our previous article here for a digest of recent decisions confirming this approach). In general, the EPO case law suggests that the threshold for meeting these requirements is a low one, in stark contrast with the UK and the persistent requirement to demonstrate plausibility.
The appealed UK decision
In the UK decision under appeal, Generics sought a declaration of invalidity for AstraZeneca’s European (UK) Patent 1506211 and two related supplementary protection certificates (SPCs). Two claims were of relevance to the proceedings: claim 2, directed to the compound “dapagliflozin” per se, and claim 15, which was a Swiss form claim directed to use of dapagliflozin in the preparation of a medicament for treating diabetes.
In his judgment, Dr Michael Tappin KC (sitting as a Deputy Judge of the High Court) agreed with Generics that the application did not render it plausible that dapagliflozin achieved the technical effect put forward by AstraZeneca (namely, that dapagliflozin was an SGLT2 inhibitor and was useful for treating diabetes). The main rationale here was that the patent merely labelled dapagliflozin as an “SGLT2 inhibitor”, and described an assay to determine SGLT2 activity, but did not include any experimental data nor any suggestion that dapagliflozin itself had been tested. This was determined not to be enough to render the technical effect plausible.
Since the technical effect was not rendered plausible, dapagliflozin could not be held to make any technical contribution to the art, and as such was merely an obvious, arbitrary selection from a broad class of compounds disclosed in an earlier application. The patent and related SPCs were declared invalid for lack of inventive step and insufficiency.
The appeal
As part of their appeal, AstraZeneca asked the Court to reconsider the applicable test for plausibility. In particular, AstraZeneca argued that, in view of G 2/21, Warner-Lambert sets too high a standard for plausibility. They further argued that the Court should depart from Sandoz v BMS and apply the lower threshold of “ab initio implausibility” (i.e. assumed plausibility unless there is legitimate reason to doubt the purported technical effect) in relation to inventive step.
It is perhaps not surprising that AstraZeneca would wish for plausibility to be assessed with an approach more closely aligned to that of the EPO’s in G 2/21. At the EPO, an applicant can use post-published data to support a technical effect relied on for inventive step, subject to the conditions laid down in G 2/21 (see above). This was relevant to AstraZeneca’s case because post-published data confirmed that dapagliflozin was indeed an SGLT2 inhibitor, and in fact had achieved commercial success as a diabetes treatment.
However, after reviewing decisions of the EPO Boards of Appeal and the Courts of other EPC Contracting States, the UK Court of Appeal concluded that no settled view had yet been reached on how G 2/21 is to be interpreted and applied. In the Court’s view, a departure from Sandoz v BMS could not be justified:
127. The conclusion I draw from this brief review of the case law of courts in other EPC Contracting States is that we are some way from unanimity as to what test G 2/21 lays down and how to apply it. There appears to be an emerging consensus that speculative patents are not permitted, but little agreement as to what constitutes a speculative patent.
128. Accordingly, neither the jurisprudence of the Boards of Appeal nor the case law of courts of other Contracting States justifies a departure from Sandoz v BMS any more than G 2/21 itself does.
As a result, there is no change to the UK courts’ approach to assessing plausibility, and the applicable standards remain Warner Lambert v Generics and Sandoz v BMS.
AstraZeneca is likely to seek leave to appeal to the Supreme Court as a next (and final) step in this case. If leave to appeal is granted, the Supreme Court would have the opportunity to soften the UK’s approach to plausibility as set out in Warner Lambert, perhaps moving to a standard closer to the EPO’s.
Practice points
While there is no change to the UK courts’ approach, there are practical considerations arising from this decision which should be kept in mind.
1. Patent drafting
From a UK perspective, this decision suggests it is critical that applications include something beyond a mere assertion that a particular technical effect is achieved. Where the claims concern therapeutics, clinical data may not be available at the drafting stage, so consider what preclinical, in vitro or in silico data could be included to render a technical effect plausible. Remember, it isn’t necessary to definitely prove that a product works for a specific technical effect (e.g. therapeutic effect), but there must be something in the application that would make the skilled person reasonably believe that it would work (e.g. technical rationale).
In the absence of data, it is possible to rely on a priori reasoning to render a technical effect plausible (e.g. with reference to the prior art). However, one needs to tread carefully between making a technical effect plausible, while not making it seem as though the invention was obvious.
2. Due diligence
When it comes to due diligence on a patent portfolio, the strengths of the data should be carefully considered. Is there data at all? Is the data commensurate with the scope of the claim? Simply asserting that claimed subject matter exhibits a particular technical effect, without providing experimental results, is unlikely to meet the current plausibility threshold in the UK (and may also be problematic in EPO opposition). Such claims may be vulnerable to inventive step and/or sufficiency attacks.
3. UPC
The relatively new Unified Patent Court (UPC) has not yet addressed the issue of plausibility. While the UPC does diverge from the EPO’s existing body of case law in some areas (e.g. on the problem-solution approach to inventive step), the UPC is largely harmonised with Enlarged Board of Appeal decisions, for example on priority (following G 1/22 and G 2/22), and more recently on claim interpretation following G 1/24. Assuming the UPC does choose to follow the G 2/21 approach, we may finally get some much needed clarity on exactly how this test is to be applied. Time will tell how this will unfold and for now we expect plausibility arguments to feature in post-grant challenges before the UPC.
Our expertise and commercially-minded approach means we can advise on everything from the nitty-gritty details of which experiments need to be carried out to support a claim across its scope, to big-picture advice on how to embed your IP strategy into your wider research and development plan (i.e. when do I file, bearing all of this in mind). If you would like to know more, please do get in touch with Jess Duncombe, Joseph Etherington, James Snaith, or your usual K&S advisor.