Lessons from dapagliflozin in the UK: status and effects of “hold the ring” injunctions

Background and chronology to the dapagliflozin interim injunctions

Teva (for whom Pinsent Masons acted), Glenmark and Viatris sought to revoke European Patent (UK) No. 1 506 211 and two SPCs based on this patent held by AstraZeneca covering AstraZeneca’s SGLT2 inhibitor dapagliflozin product, Forxiga®, which is used in the treatment of Type 2 diabetes. The purpose of these revocation actions was to “clear the way” of AstraZeneca’s patent rights, thus allowing the launch of generic dapagliflozin.

On 28 April 2025 the High Court issued its decision finding the patent and the SPCs covering Forxiga® invalid. A more detailed review of this judgment is covered by JUVE Patent here.

AstraZeneca consequently sought interim “hold the ring” injunctions against these parties (as well as Sandoz) to maintain the status quo of its dapagliflozin monopoly, preventing generic launch, until determination of the appeal. These interim injunctions were granted by HHJ Hacon on 28 May 2025.

On 16 July 2025, the CoA rejected AstraZeneca’s appeal against the invalidity finding, maintaining the first instance decision, on all grounds. A detailed review of this judgment is covered by JUVE Patent here.

AstraZeneca then sought permission to appeal the invalidity finding to the Supreme Court. Pending this determination by the Supreme Court, the CoA was asked to allow AstraZeneca’s application for another injunction to hold the ring against the generics. That hold the ring injunction was to last until the determination of AstraZeneca’s application for permission to appeal to the Supreme Court, and if granted, pending the invalidity appeal judgment of the Supreme Court, which the parties agreed could take between 15-23 months. Although the CoA declined this request on the basis that AstraZeneca had no real prospect of success in its invalidity appeal, it nevertheless granted a further 14-day injunction allowing AstraZeneca to seek the same injunctive relief directly from the Supreme Court. The Supreme Court then granted a further injunction to hold the ring against the generics pending determination of AstraZeneca’s application for permission to appeal the CoA’s invalidity ruling. Permission was subsequently refused three days later, and the injunction was finally lifted.

Impact of the dapagliflozin interim injunctions

Notwithstanding strong resistance from the generic parties, AstraZeneca was successful in securing a series of short-term interim injunctions, each respectively “holding the ring” pending further determinations of its appeals. Although each individual injunction was granted for only a short period of time, when taken together they kept the generic companies off the market for just over three months (i.e. from the first instance decision on 28 April 2025 through to the rejection of AstraZeneca’s application for permission to appeal to the UK Supreme Court on 31 July 2025). In the pharmaceutical space, such a time period can have a significant commercial and financial impact: based on annual sales of over £400 million per annum (see paragraph 13 of the CoA decision in AstraZeneca AB & Others v Generics (UK) Limited & Others [2025] EWCA Civ 924), sales for the patentee in this three month period therefore generated approximately £100 million in additional revenue. The generic companies involved in the dapagliflozin dispute were meanwhile prevented from making any sales until the final appeal had been refused, at which time the market was open to all competitors and the opportunity for first mover advantage was lost.

This case demonstrates the potential for patentees effectively to extend their period of exclusive market control by seeking one or more hold the ring injunctions, an approach which is possible under the current framework even in circumstances where a patent has been found invalid at first instance and the decision later upheld by the CoA. While potentially value-generating for patentees, this scenario is disadvantageous to would-be generic entrants who have invested in clearing the way and to third parties, such as the NHS, who continue to pay significantly higher drug prices whilst the injunction is ongoing. The dapagliflozin case is therefore illustrative of the multiple interests at stake when deciding hold the ring injunctions.

Given the considerable risks to generic interests posed by the current UK interim relief framework, following the dapagliflozin case generic litigants may seek to advocate for the Supreme Court to further consider the UK’s approach to hold the ring injunctions in light of how the pharmaceutical market has developed and fairness for all parties. Specifically, generics may raise questions as to whether the case law strikes an optimum balance between the competing interests of patentees and generics following an invalidity determination, particularly as relates to enabling timely market entry.

Injunctions pending appeal following a first instance merits judgement: the current approach

The courts’ jurisdiction to grant interim relief pending appeal where a patent has been invalidated at first instance was recognised by the CoA in Novartis v Hospira [2013] EWCA Civ 583 with Floyd LJ setting out five principles, which can be paraphrased as three factors to be considered as follows:

• whether the court is satisfied the appeal has a real prospect of success;
• all relevant circumstances following judgment which weigh on the decision to grant interim relief (e.g., the period of time before any appeal is likely to be heard and the balance of hardship to each party if an injunction is refused or granted); and
• that matters should be arranged so that the court is best able to do justice between the parties once the appeal has been heard.

In the present case, the CoA held that AstraZeneca’s application for interim relief should be refused as its application for permission to appeal to the Supreme Court concerning the invalidity determination had no real prospect of success. On this basis, the CoA held that AstraZeneca did not meet the first limb of the Novartis v Hospira test. As such, the CoA also determined that it was unnecessary to consider arguments on the adequacy of damages or the overall balance of the risk of injustice. Thus, had the CoA’s decision been based on the Novartis v Hospira test alone, the injunction would have been lifted allowing the generics to enter the market and damages against the generics pursued if the Supreme Court had ultimately reached a different conclusion.

Instead, despite its definitive conclusion on the merits of the appeal, the CoA considered that it would be just and convenient to grant a 14-day injunction to enable AstraZeneca to make a new injunction application to the Supreme Court. In doing so, the CoA effectively allowed AstraZeneca to extend its period of market exclusivity, arguably lowering the threshold required for a patentee to secure further injunctive relief. This is particularly surprising because (i) the invalidity arguments had been heard twice and upheld on appeal, (ii) the CoA had stated that, in its view, the appeal had no real prospects of success, and (iii) the threshold for the Supreme Court granting permission to appeal is high and should relate to a point of law of general public importance. None of these criteria were met in this case, yet the CoA favoured maintaining the status quo.

From the CoA’s decision, it appears that the Courts continue to place great weight on the third limb of the Novartis test seeking to enable the higher court to best do justice between the parties in the matter. But if the CoA will always opt to maintain the status quo in order that the Supreme Court may consider the appeal, one has to ask whether limb (1) holds any weight at all. Going forward, generic litigants may raise questions as to whether the Courts’ current approach in fact allows justice to be done between the parties, or whether other relevant factors should be given greater weight in this assessment, including whether the Courts should place more emphasis on the invalidity status of the patent at this stage.

Practical effects of the current approach

In assessing whether an injunction should be granted before a first instance decision on validity, unlike some European courts including, notably, the UPC, the English court does not carry out a mini trial assessing infringement and validity.

Once a first instance judgment has been delivered, the first limb of Novartis v Hospira recognises, to a limited degree, that validity has been evaluated by the trial judge. However, even where, as in this case, the invalidity of the patent has been resoundingly upheld on appeal and the first limb of Novartis is dismissed, a further hold the ring injunction was still granted by the Supreme Court.

The relative ease with which patentees are able to obtain hold the ring injunctions, particularly based on invalid patents, creates two key consequences which should be considered in any evaluation of the current framework and the balance of interests it strikes.

First, the approach in the dapagliflozin case of granting discretionary relief on the basis that doing so is just and convenient without further judicial reasoning runs the risk that such injunctions will be requested and ordered as a matter of course in the future. This would greatly reduce the attractiveness of the UK as a forum for generics or biosimilars seeking to clear the way, likely producing significant negative economic impacts.

Second, if such hold the ring injunctions by the Court of Appeal are awarded almost by default to allow the Supreme Court the opportunity to consider whether it wants to take the case, some patentees may be encouraged to appeal to the Supreme Court to extend their monopoly period, regardless of the merits of their appeal. The Supreme Court will then be burdened with determining a greater number of urgent injunction requests, particularly in pharmaceutical patent disputes. This is not the Supreme Court’s function.

Nor is the Supreme Court particularly well placed to deal with matters of evidence on which such injunctions may turn. Given the limited time the Supreme Court was provided in the dapagliflozin case to decide permission to appeal, or otherwise grant a further injunction, it is likely the Court will be inclined to continue to maintain the status quo in similar cases until such time as it can properly address the application for permission to appeal. Further, a greater number of urgent requests for injunctions to the Supreme Court may even work against patentees by putting increased pressure on the Supreme Court to decide permission to appeal applications quickly, perhaps making it more likely that they may be dismissed. Broad adoption of dapagliflozin-type hold the ring injunctions therefore carries meaningful risks which should be carefully considered by litigants and the Courts.

Conclusion

Although the Supreme Court declined to consider the issue of injunctions holding the ring following its decision to reject AstraZeneca’s application for permission to appeal, we expect to see this issue raised in future injunction proceedings, especially post a first instance invalidity decision. This has the potential to significantly delay future generic product launches in the UK, and to produce other potential consequences discussed above. As it stands, therefore, generic and biosimilar manufacturers should expect to navigate multiple applications for hold the ring injunctions. For both generics and patentees, consideration of hold the ring injunctions and the status of broader judicial adoption of the dapagliflozin precedent is strongly advised for businesses and their legal teams. Navigation of these issues may prove critical to successful outcomes for all parties.

If you are interested in legal advice to assist with navigating generic pharmaceutical market entry or issues relating to interim injunctive relief, please contact the Pinsent Masons European patent litigation team through tracey.roberts@pinsentmasons.com or sarah.taylor@pinsentmasons.com.