Two years of the UPC: why the patent court is growing in influence

Made up of a system of first-instance local, regional and central divisions, as well as an appeal court, the UPC is a dedicated judicial system for litigating unitary patents as well as existing European patents that have not been specifically opted-out of the UPC’s jurisdiction.

Features of the UPC system that patent owners have welcomed include the potential to obtain injunctions and damages across a broad geographic area, which is particularly topical following the UPC’s recent decision to award relief beyond its borders, in that case covering the UK. All users of the system – including those defending infringement claims or challenging a patent’s validity – have welcomed the speed and efficiency of UPC divisions in issuing decisions. As a result, the UPC has become central to the way businesses across all sectors develop and operate global patent litigation strategies.

One of the most prominent developments in the UPC’s second year is the number of disputes that have settled, including the multi-jurisdictional medical technology disputes between Abbott and Dexcom, and 10x Genomics and Bruker, formerly NanoString ,which shaped much of the early UPC case law. This underscores the value the court places on cooperation between the parties, as well as its role in encouraging settlement discussions in multi-jurisdictional proceedings.

Below, we examine some of the prevalent issues the UPC has dealt with, helping to inform why the court has growing influence.

Increasing certainty offered by robust Court of Appeal rulings

As anticipated, the UPC Court of Appeal’s (CoA) activity increased throughout year two, with its growing bank of rulings bringing certainty for businesses over UPC procedures and the interpretation of important questions of patent law. Such rulings are expected to spur an increase in UPC actions going forward. 

For example, in the case of Fives v REEL, the CoA confirmed that the UPC has jurisdiction to decide on acts of infringement committed before 1 June 2023, provided that the European patent in issue had not lapsed before that date.

The CoA also ruled, in AIM Sport v Supponer, that businesses that have opted European patents out of being litigated before the UPC are only blocked from withdrawing that opt-out if national litigation concerning those patents was initiated after the UPC began operating, not before. It has further confirmed that the UPC is competent to decide on infringing acts which occurred between the date of the opt-out and the date of withdrawal.

Pragmatic case management decisions and flexible relief

Trust in the UPC has been engendered in year two of the UPC by continuing efforts to ensure that its overarching aims are met. 

One of the UPC’s primary aims is to speed up patent proceedings in Europe, enabling parties to reach commercial certainty at an earlier stage than might be achieved with multiple national actions. In this regard, the CoA has set a high bar for businesses that want UPC proceedings to be put on hold when parallel opposition proceedings before the European Patent Office (EPO) are pending – it has rejected multiple ‘stay’ applications, emphasising the UPC’s determination to proceed with its own actions.

The UPC has also made pragmatic case management decisions which prioritise the aim of issuing first instance decisions within 12 to 14 months of proceedings being initiated. The UPC’s decisions suggest the court is very reluctant to make orders that would significantly impact a case’s timetable. Such decisions have highlighted the front-loaded nature of UPC proceedings, emphasising too the need for parties to plead their full case up-front and taking a strict approach to the introduction of new facts and evidence unless they could not reasonably have been foreseen at an earlier stage.

The UPC has also shown that it is willing to take a flexible approach to relief. For example, in the dispute between Edwards and Meril concerning prosthetic heart valves, the Munich local division (LD) carved out certain products from the scope of an injunction it granted, demonstrating the balancing act between protecting a patentee’s rights and, in that case, prioritising patient safety.

Extending beyond the UPC’s borders

The geographic reach of the UPC’s decisions was an issue that came to the fore in the court’s second year.

In a decision with profound implications for national and UPC proceedings, the Court of Justice of the EU (CJEU) ruling, in the case of BSH Hausgeräte v Electrolux, reinforced the view of the Düsseldorf LD in an earlier decision that, where a defendant is based in a UPC country, the UPC can determine claims of infringement of European patents in every country in which those patents apply – including, in the circumstances of the case in Düsseldorf, third countries such as the UK.

The CJEU’s ruling was subsequently applied by the Paris LD,which held that, where an alleged infringer is based in a UPC country, in that case France, the UPC may decide on infringement for all parts of the European patent – including those parts that apply in non-UPC EU countries like Spain, Lugano Convention countries like Switzerland, and third countries, like the UK. The Paris LD went further and determined a legal basis for it to determine infringement by a Swiss co-defendant because of the presence of the French ‘anchor defendant’ in a UPC country. 

The concept of anchor defendant has since been considered in more detail by The Hague LD, in a case involving Moderna and Genevant. It determined that it had jurisdiction over the Spanish, Polish and Norwegian co-defendants because their activities were sufficiently connected with the Dutch anchor defendant.

Following these rulings, there was a wave of applications to broaden the geographic scope of claims before the UPC, and at the beginning of the court’s third year, the UPC’s long-arm jurisdiction became a reality as the Mannheim LD became the first UPC division to issue an injunction that extends to the UK – a decision that has significant implications for businesses. 

Preliminary relief: some clarification from the CoA

Although no clear trend regarding the granting of applications for preliminary injunctions (PIs) has yet emerged – the UPC has granted 45% of the PIs sought – they are an attractive proposition for applicants seeking relief across the UPC territory. 

In the first year of the UPC, the CoA set a legal test for the grant of PIs by the UPC. In its second year, that test was applied by lower UPC courts. Further decisions of the CoA have provided clarity as to the hurdles that must be overcome for a PI to be granted. 

Under the CoA’s test, the court must be satisfied with a “sufficient degree of certainty” that the patent in issue is valid and infringed, before a PI can be granted. The CoA said this means that it must be at least predominantly probable the applicant is entitled to initiate proceedings and that the patent will be infringed. It went on to confirm that a “sufficiently certain” conviction is lacking if the court considers it predominantly probable that the patent is not valid.

Unlike some national courts, the UPC will therefore conduct detailed infringement and validity assessments at an early stage in the context of PI applications. Where parallel EPO opposition proceedings are ongoing, the UPC should consider both the likelihood of invalidity based on its own independent assessment, as well as the likelihood of an EPO invalidity decision.

PIs can also be granted in circumstances where there has been no patent infringement if the UPC nevertheless assesses that infringement is imminent. 

The meaning of ‘imminent infringement’ is of particular interest to generic and biosimilar manufacturers, which routinely undertake activities prior to launching products in anticipation of the expiry of existing patent rights. In year two of the UPC, the Düsseldorf LD confirmed that all acts aimed at preparing a rival, allegedly infringing, product for launch must be completed in order for an infringement to be considered imminent. In a separate case, the Lisbon LD held that the process of obtaining a marketing authorisation and regulatory actions required by specific member states are merely administrative steps which will not amount to imminent infringement.

A PI applicant must act promptly once they become aware of the allegedly infringing activity to avoid an unreasonable delay in bringing the application. The CoA has so far declined to specify an acceptable timeframe. However, it has clarified that the period for evaluating claims of unreasonable delay begins from when a patent holder becomes aware of a violation or should reasonably have known about it. 

While each application will be assessed on its own facts, a ‘safe harbour’ of two months has been accepted by the Munich LD in, for example, the cases of Dyson v SharkNinja and Syngenta v Sumi Agro, but this approach has not yet been universally adopted across the UPC system – for example, in 10x Genomics v Curio Bioscience, the Düsseldorf LD considered a one month deadline for action was appropriate. Further, like any rule, there are exceptions, and the facts of the cases may lead to the court taking a more flexible approach to unreasonable delay.

The CoA has also considered the potential for harm for the parties arising out of the grant or refusal of a PI. It has noted in the case of Mammut v Ortovox that the presumption that there will be potential financial loss arising from patent infringement does not require evidence of irreparable damage – a credible threat of such damage is sufficient, according to the CoA, to justify the grant of a PI.

Harmonisation of substantive patent laws

Claim construction

It is now well established in UPC cases that the protective scope of a patent is to be determined through an assessment of the patent claims – these being the claims included in patent applications that go to the heart of what the invention is and what it achieves. How patent claims are worded is important because they determine whether the invention is considered to meet the legal tests for obtaining patent protection – including whether it is new, not obvious, sufficiently described and has an industrial application.

In this regard, UPC case law has confirmed that the interpretation of a claim does not depend solely on the strict, literal meaning of the wording used – the description and any drawings must always be used as explanatory aids for the interpretation of the patent claims, which nevertheless remain more than mere guidelines for determining the protective scope of a patent.

The UPC’s approach has been endorsed by the EPO’s Enlarged Board of Appeal, which emphasised the need for harmonisation with the UPC and national courts. 

However, there remains a lack of clarity over whether the way in which patent holders have prosecuted patents – how they have engaged with patent offices in relation to patent applications – can be used to assist claim interpretation. Both The Hague LD, in Plant-e Knowledge v Arkyne, and the Düsseldorf LD, in SodaStream v Aarke, have refused to refer to the prosecution history to aid claim interpretation, but the Munich LD has, in Philips v Belkin, contrary to German case law. 

So far, the prosecution history question has only been considered by the CoA in the context of a decision over provisional measures, where it only had to consider whether it was more likely than not that the claims are invalid rather than reach a definitive view on the matter. However, in that case, between Alexion and Amgen, it did rely on statements made during patent prosecution to interpret the relevant claim, doing so from the perspective of the skilled addressee of the patent. 

Infringement

Interesting developments concerning infringement also punctuated year two of the UPC.

In the case of SodaStream v Aarke, the Düsseldorf LD considered whether the Gillette defence could be cited to counter claims of infringement. The Gillette defence was established in 1913 by the English courts and provides that if the product or process used by an alleged infringer falls within the scope of a patent claim, but that product or process would have been known or obvious at the priority date of the patent, that claim must be invalid for lack of novelty or inventive step. 

The Düsseldorf LD found that as Aarke had raised no counterclaim for invalidity, the Gillette defence to patent infringement was not available in this case. However, this does not necessarily mean that it will not be available in other cases where both infringement and validity are in issue. 

The availability of the Gillette defence arose before The Hague LD in the case of Plant-e v Arkyne, which was the first time the UPC considered whether there had been infringement by equivalence. The so-called doctrine of equivalents provides, in essence, that if a product or process is not substantially different from the elements of a patented invention, it can be considered infringing.

In the course of reaching its decision in the case, The Hague LD confirmed the possibility of raising a Gillette defence to claims of infringement by equivalence, which stands in contrast to the findings of the Düsseldorf LD in the SodaStream case. The Plant-e dispute has now settled so will not come before the CoA.

Subsequently, the Mannheim LD said that the UPC will need to develop its own test and case law on the doctrine of equivalents, considering that this is the only way it will be able to achieve its overarching aim of harmonising case law. 

Also in relation to infringement, the Düsseldorf LD earlier this year handed down the UPC’s first decision on second medical use claims. Second medical use patents are typically used to protect the use of a known compound or substance for a new medical purpose, and can therefore be highly valuable for the repurposing of known medicines for new diseases.

The Düsseldorf LD’s decision contains useful learnings on how the UPC may deal with second medical use infringement actions. It considered that, to establish a finding of infringement of a second medical use claim, it must be shown that: the alleged infringer placed, or offered to place, the medical product on the market in such a way that it leads to, or may lead to, the claimed therapeutic use according to the patent; and that the alleged infringer knew, or should reasonably have known, of such use.

Validity: inventive step

Patents can only be obtained for inventions if they are new, not obvious, and have an industrial application. To pass the obviousness test, applicants must be able to show that their new product or process delivers an inventive step – i.e. there is a substantive technical advance that someone skilled in the relevant area would not be able to work out easily using their own know-how and information available to them.

Different patent offices and courts assess whether there has been an inventive step in different ways. How the UPC assesses inventive step has been a keenly watched issue – including during year two of its operation.

Though not bound to do so, the UPC was expected to adopt the European Patent Office’s ‘problem-solution approach’ for the purposes of assessing inventive step in patent validity cases. This involves, as a first step, identifying the ‘closest piece of prior art’. However, an alternative approach, of taking a realistic starting point, has emerged. First proposed by the Munich central division in a case in which a patent owned by Amgen for a cholesterol drug was revoked, the ‘realistic starting point’ approach has been implemented more widely than the problem-solution approach.

That said, there has been divergence – in the dispute between Edwards and Meril, the Munich LD ruled decisively in favour of the UPC adopting the problem-solution approach when assessing inventive step, to ensure legal certainty and alignment with the EPO’s established case law. In the Tesla v Avago case, it went on to say that as neither test is enshrined in the European Patent Convention, both can be applied to assess inventive step, adding that in the majority of cases it should lead to the same result. 

It is surely only a matter of time before the matter comes before the CoA for clarification.

What next?

The UPC is now well-established, with litigants becoming more inclined to bring multi-jurisdictional disputes before it. This will undoubtedly be helped by the roll out of its new case management system in the second half of 2025,aimed at improving functionality and accessibility. 

However, many substantive questions remain unanswered. 

The UPC’s first damages award and injunction extending to a non-UPC territory will lead to a greater appreciation of how the UPC may interact with non-UPC territories and will inform wider litigation strategies. How far the UPC’s jurisdiction will extend remains up for debate.

While, to date, only German LDs have ruled on SEP/FRAND issues, there have been recent examples of SEP/FRAND-related filings before The Hague LD. How a non-German division will determine such issues will be of major interest to owners of technology patents and implementers of such patents.

We expect the number of pharmaceutical and biotech disputes before the UPC will gradually increase, not least because the transitional period for the court – after which European patents will automatically fall subject to the UPC’s jurisdiction – is expected to end in 2030, without being extended. Pharmaceutical and biotech businesses might be advised to engage with the UPC sooner rather than later to gain experience and help influence case law. 

We also expect the role of the UPC’s Patent Mediation and Arbitration Centre (PMAC), which is scheduled to open in early 2026, to be monitored with interest.