Sandoz v Bristol-Myers Squibb (EWCA, [2023] EWCA Civ 472, 2023)

Date4 May 2023
JurisdictionUnited Kingdom
CourtCourt of Appeal (Civil Division), England and Wales
Case number[2023] EWCA Civ 472 (CA-2022-001183 et 001203)
PartiesSandoz Limited et Teva Pharmaceutical Industries Limited (demanderesses, intimées), Teva UK Limited (défenderesse à la demande reconventionnelle, intimée) v Bristol-Myers Squibb Holdings Ireland Unlimited Company (défenderesse, appelante)
Language of the decisionEN

Text of the decision · Texte officiel, Find Case Law (The National Archives), version XML

Dhenne Avocats · 11 October 2026

Our analysis

Summary

The Court of Appeal upheld the invalidity of Bristol-Myers Squibb’s patent for apixaban (Eliquis), on which an SPC is based. The application did not make it plausible that apixaban was a useful factor Xa inhibitor: it linked no test result to that compound. As the first UK decision to consider the Enlarged Board’s decision in G 2/21, the judgment holds that plausibility also applies to a claim to a single compound and that the test adopted by the Supreme Court majority in Warner-Lambert still governs. Utility confirmed after filing does not save a patent whose application does not link the effect to the claimed compound.

Facts and procedure

BMS was the proprietor of European Patent (UK) No 1 427 415, with a priority date of 21 September 2001, which expired on 16 September 2022 and supports SPC/GB11/042, expiring on 19 May 2026 (para 2). Claim 1 is to apixaban or a salt thereof (para 77). Sandoz and Teva contended that application WO 03/026652 did not make plausible any useful factor Xa inhibitory activity, so that the invention lacked inventive step and was insufficiently disclosed (paras 3 and 4). Meade J held the patent invalid ([2022] EWHC 822 (Pat)); BMS appealed (para 5).

The applicable law

Articles 52(1), 56, 83 and 138(1) EPC, given effect by sections 1(1)(a), 3, 14(3) and 72(1) of the Patents Act 1977, do not mention plausibility, which was developed by case law (para 6). Arnold LJ reviewed the Boards’ case law and UK case law including Warner-Lambert (paras 7 to 42). He took from G 2/21 that the fundamental question is whether the technical effect asserted is derivable by the skilled person from the application as filed read with common general knowledge (para 53).

Question

Does the plausibility requirement apply to a claim to a single compound and, if so, did the application make plausible apixaban’s activity as a factor Xa inhibitor?

Decision

A claim to a single compound is not legally distinguishable from Warner-Lambert: the fundamental principle is that the scope of the monopoly must be justified by the technical contribution, whether the claim is broad or narrow; there is no invention in merely identifying a new compound, only in identifying its utility (para 92). Plausibility therefore applies (para 93), on the standard of the Warner-Lambert majority; the harmonised approach in G 2/21 is in substance much closer to « ab initio plausibility » than to « ab initio implausibility » (para 94). It is not enough that the application encourages the skilled person to carry out simple tests (para 95). The argument that the application impliedly disclosed a nanomolar Ki was a new case not open on appeal (paras 98 and 99). Plausibility being a multi-factorial evaluation, the Court would intervene only for an error of law or principle (para 100). The synthesis of 3 g of apixaban, the largest quantity in the application, admitted explanations other than favourable results (paras 101 to 103). Nothing in the application links the briefly summarised assay results to apixaban: the assertion of activity is speculative, and it is irrelevant that BMS later proved it well founded and limited the claim to apixaban (para 105). The appeal was dismissed (para 106).

Key points for practice

  • Plausibility applies to single compound claims as it does to class and second medical use claims (para 92).
  • For the Court of Appeal, G 2/21 does not justify departing from the Warner-Lambert majority test (paras 91 and 94).
  • The possibility of verifying the effect by simple tests does not replace disclosure (para 95).
  • Practical point: the application must expressly link the compound that will later be claimed to the activity results obtained; general statements about the compounds of the invention are not enough.

Relevance before the UPC

Since the UPC assesses validity against Article 138(1) EPC, this national court reading of G 2/21 may be relied on in arguments on inventive step and sufficiency before it.

Provisions applied

European Patent Convention
Art. 52(1); Art. 56; Art. 83; Art. 138(1)
National law
Patents Act 1977, sections 1(1)(a), 3, 14(3) and 72(1)
Case law cited
G 2/21; T 939/92; T 609/02; T 1329/04; T 578/06; T 488/16; Warner-Lambert v Generics [2018] UKSC 56; Conor Medsystems v Angiotech [2008] UKHL 49; Generics (UK) v Yeda [2013] EWCA Civ 925; Idenix v Gilead [2016] EWCA Civ 1089

Related decisions

Prepared by Dhenne Avocats from the text of the decision (National Archives, Find Case Law), consulted on 11 October 2026. Only the official text is authoritative.

Further reading

All decisions analysed in Pharma Litigation Watch · Pharma Litigation Watch

Dhenne Avocats acts for claimants and defendants in European patent disputes, before the Unified Patent Court and the French courts.