Court of Appeal tightens patent plausibility and inventive‑step standards in AstraZeneca ruling

On 16  July 2025, the Court of Appeal delivered a pivotal judgment in Generics (UK) Limited & Others v AstraZeneca AB, confirming the High Court’s ruling that AstraZeneca’s patent for dapagliflozin—a sodium‑glucose co‑transporter 2 (SGLT2) inhibitor used in treating type 2 diabetes—was invalid for lack of inventive step and insufficient plausibility and disclosure.

Plausibility under the microscope

At the heart of the dispute lay the question of whether AstraZeneca’s patent plausibly demonstrated that dapagliflozin was effective as a selective SGLT2 inhibitor for treating diabetes. The High Court, and now the Court of Appeal, held it did not.

Lord Justice Arnold reaffirmed the now‑established plausibility threshold from Warner‑Lambert Co LLC v Generics (UK) Ltd. He emphasised that “patents must disclose sufficient information to make the claimed invention plausible at the filing date.” Mere assertions such as “dapagliflozin is an SGLT2 inhibitor useful for the treatment of diabetes” were ruled to be “bare assertions unsupported by any experimental results.”

Reflecting on the high bar required, the Court criticised AstraZeneca’s reliance on unquantified assay language, calling it “some slight of hand” and rejecting the patentee’s attempt to bootstrap assertions into plausibility.

Inventive step: arbitrary selection struck off

The claimants (Generics UK, Teva, Glenmark) contended that AstraZeneca merely chose dapagliflozin arbitrarily from a class of known compounds without showing a technical advance. The Court agreed: the compound was within the preferred class disclosed in prior art (WO 01/27128 A1), which itself provided 18 worked examples including a close structural analogue.

Without data demonstrating a technical benefit, the allegedly innovative move was deemed obvious. The patentee’s lack of evidential distinction between dapagliflozin and the cited prior art undermined any novel technical contribution.

Disclosure and the notional skilled person

The Court reiterated that the patent must disclose a plausible technical effect at the priority date, based solely on the information in the patent and the common general knowledge at that time. It rejected AstraZeneca’s suggestion that post‑filed or unsubmitted data could rescue the plausibility requirement.

The ruling confirmed that both inventive step and sufficiency share a common plausibility foundation: a claim that lacks a plausible demonstration of efficacy fails on both counts. The Court referred to this as rejecting a bifurcated “ab initio plausibility” vs “ab initio implausibility” framework that the European Patent Office’s Enlarged Board of Appeal had been exploring under G 2/21, stating that divergence from UK precedent was “not justified.”

Flavour of EPO vs UK divergence

While the EPO’s G 2/21 decision does attempt to differentiate plausibility between sufficiency and inventive step challenges, the Court stressed that the UK remains firmly bound by Supreme Court authority. As Michael Tappin KC has noted, the UK approach demands “reasonable scientific grounds” underpinning claimed effects at filing, and does not allow retrospective justification.

The Court noted, however, that UK case law remains distinct from the EPO’s evolving standards under G 2/21, and that national alignment was not yet achievable or desirable.

Policy and pharmaceutical patenting impact

This judgment serves as a wake‑up call to pharmaceutical companies: new drug patents must be supported by meaningful pre‑filing data, not mere assertions. We can expect a shift in patenting strategies where experimental results must now be incorporated into the patent application itself, rather than compiled post‑filing.

Technical counsel should also note that inventive‑step objections in pharma cases will increasingly hinge on the plausibility threshold: the skilled person must be able to derive a credible technical contribution from the patent text alone.

What's next?

AstraZeneca sought permission to appeal to the Supreme Court, but the Court of Appeal declined, finding the appeal had “no prospect of success.” Notably, interim injunctions against generic entry were extended by two weeks to afford AstraZeneca time to seek special leave.

Meanwhile, generic manufacturers such as Glenmark remain enjoined from entering the UK market until at least 30 July 2025, pending any Supreme Court intervention.