Background facts

Bayer was the owner of patents relating to rivaroxaban, an active ingredient used to treat thromboembolic disorders. Bayer markets rivaroxaban in the UK under the brand name Xarelto. The compound patent for rivaroxaban was covered by a supplementary protection certificate (SPC), which expired on 1 April 2024.

The present proceedings concerned a Bayer follow-on patent (EP (UK) 1 845 961), which covered the use of rivaroxaban for manufacture of a medicament for the treatment of a thromboembolic disorder. The claims of this patent were limited to use of a rapid release tablet, and administration no more than once daily for at least five days.

Sandoz and other generic manufacturers (the Defendants) sought to revoke the patent, alleging that it was obvious over several groups of prior art citations, including the “Harder” group of prior art, that it was insufficient and that it added matter. Bayer had counterclaimed for threatened infringement of the patent, but the infringement issues were ultimately conceded, such that the trial concerned validity issues only.

The groups of prior art citations included posters produced by Bayer and used by its researchers to present their findings at conferences, which were shown to the public before the priority date but were no longer available at the priority date.

The first instance trial on obviousness was heard on 9, 12-16 and 21-22 February 2024.

Key arguments on obviousness

Bayer argued that the prior art differed from the claimed invention in four key ways: (i) it did not identify rivaroxaban, (ii) it lacked any mention of a rapid-release form or (iii) a once-daily dose for at least five consecutive days, and (iv) the prior art concerned the treatment of healthy phase I patients as opposed to patients vulnerable to thromboembolism.

While there was no perceived technical barrier to the invention, Bayer argued that the skilled team would have been held back by the ethical issue of whether the administration of once-daily doses would have posed a real risk of harm to patients vulnerable to thromboembolism in phase II trials, and so would not have had a reasonable expectation of success. Bayer said that the skilled team would not know where the therapeutic window lay. In particular, if dosed incorrectly, it could result in failure to clot (leading to excessive bleeding) on the one hand, or the development of serious thromboembolism on the other.

The Defendants argued that the phase I data of the prior art would be sufficient incentive for a skilled team to conduct a phase II study of a daily dosing regimen of rapid-release form rivaroxaban with a reasonable expectation of success, such that the claimed invention was obvious.

Interim injunction

Shortly before the 1 April expiry date for the compound SPC, and between the conclusion of the first instance trial and the handing down of judgment, Bayer made a successful application for an interim injunction to prevent the supply of generic rivaroxaban for the once daily administration.

In granting the injunction, the High Court emphasised that the injunction only related to a short period (9-10 days between the expiry of the compound SPC and the handing down of the judgment on validity), and that in short cases such as this, maintaining the status quo of the market is particularly important. ([2024] EWHC 711 (Pat))

First instance decision

In its judgment of 12 April 2024, the court held that the patent was invalid for lack of inventive step over the Harder prior art group.

The court found that a skilled team, reading each group of prior art at the priority date, would have recognised that they reported results of phase I clinical trials. The skilled team would have identified rivaroxaban in the prior art, either by asking for and obtaining the data from Bayer or by requesting a data specialist to obtain it from published sources, where they would have found the relevant information.

On expectation of success, the real question was whether the skilled team would have applied to an external ethics committee for permission to conduct phase II clinical trials with a once-daily dose, with a reasonable expectation that the committee would grant permission. The court considered that the skilled team would have sought approval to conduct such phase II trials and that certain statements in the Harder prior art would have encouraged the skilled team that once-daily administration would be both safe and effective.

The court rejected arguments that the patent was insufficient, and the added matter arguments fell away as a result of the finding on insufficiency. ([2021] EWHC 796 (Pat))

Appeal decision

Bayer’s appeal was heard shortly after the first instance decision on an expedited basis, with the Court of Appeal giving judgment on 24 May 2024. The Court of Appeal upheld the decision of the trial judge that the patent was invalid for obviousness.

As obviousness involves a multi-factorial evaluation, an appellate court is not justified in intervening unless there has been an error of law or principle. The Court of Appeal noted that Bayer faced further difficulties on this appeal, namely that (i) the judge’s decision contained a "careful, detailed and nuanced" summary of the evidence and arguments, (ii) the judge’s decision was based on preferring the Defendants’ evidence to the evidence of Bayer’s experts and (iii) that the judge’s preference for the Defendants’ evidence was “not surprising” on the facts.

Bayer’s first ground of appeal was that the judge had erred in law or principle by failing to apply the correct obviousness standard. In particular, Bayer argued that the judge’s finding that there was no perceived technical barrier to the claimed invention was inconsistent with his findings that the skilled team would not know where the therapeutic window was. The Court of Appeal rejected this argument as being based on a “misreading” of the judgment.

Bayer also argued that the judge’s focus on the question of whether the skilled team would have thought it was worth applying to the relevant ethics committee for permission to conduct a Phase II trial (a point which neither side had argued) showed that the obviousness assessment was flawed. The Court of Appeal disagreed, finding that the judge had correctly identified the key legal criterion (whether the invention would have been obvious to try, with a reasonable expectation of success), and that the reference to the need for ethics committee approval did not undermine the judge’s assessment of obviousness.

A further ground of appeal was that the judge had failed to grapple with the key question of whether the skilled team would have a reasonable expectation (based on the Harder prior art group) that a once daily dose of 30mg of rivaroxaban would have had a therapeutically useful effect for 24 hours. The Court of Appeal dismissed this ground on the basis that Bayer’s submissions did not identify any errors of principle in the judge’s approach.

The Court of Appeal also dismissed the argument that the judge had placed weight on Bayer’s decision not to adduce evidence relating to the invention story in the UK, when they had done so in other jurisdictions. The Court held that no weight had been placed on this finding. ([2024] EWCA Civ 562)

Practical points to note

The High Court commented on the instruction of experts working in teams, namely, that where real life experts would collaborate in the pursuit of a project, corresponding expert witnesses should consult one another in a broadly similar way. With this in mind, the court highlighted the fact that one side’s experts had only seen a draft of the other’s report shortly before filing. While had not resulted in any significant evidential issues in this case, a case on other facts could be different.

The Court of Appeal’s postscript to its judgment concerning timing of proceedings should also be noted. The Court of Appeal stated that the Defendants had known from October 2021 (when the Technical Board of Appeal had upheld the validity of the patent) that they had two and a half years to clear the way for the once daily administration of rivaroxaban, but had only brought proceedings a year later in October 2022. Bayer had then had to make swift applications for interim injunctive relief, expedition and permission to appeal following the expiry of the compound SPC and the first instance judgment. This was “not a sensible use of the resources either of the parties or of the courts”. The Court of Appeal’s postscript is a reminder that parties seeking to clear the way should start proceedings at an early stage, having regard to potential appeal timings, and keeping the court informed of key dates.