The UPC: lessons learned from the first 18 months

Between 1 June 2023 and 30 November 2024, there were 585 cases filed before first instance courts and over 42,500 unitary patents have also been registered within the first 18 months. This signals a high level of trust placed in the UPC system.

In its first months, the court’s approach to procedural issues and preliminary measures attracted much attention. Now, 18 months after opening, the UPC has handed down a large number of substantive decisions which are starting to shape the court’s approach to important patent issues, further increasing business confidence. 

Increasing UPC engagement

There has been engagement with the UPC across a wide range of sectors, with the greatest number (38%) of infringement actions filed in International Patent Class (IPC) H, electricity, followed a fair way behind by IPC class A, human necessities (22%) – this being a wide-ranging category, including agriculture, foodstuffs, tobacco, clothing, furniture and some medicines, to name but a few. Actions in class G, physics (16%), have been third most common.

As expected, as well as engaging companies involved in global litigation, the benefits of the UPC – notably the lower costs and speed of the decisions – have successfully attracted many SMEs. 

However, the traditionally patent-rich sectors are taking something of a different approach. While there is an increasing number of UPC disputes in the technology space, including disputes over whether standard-essential patents (SEPs) are being licensed on fair, reasonable and non-discriminatory (FRAND) terms, the picture in life sciences – often the subject of high value patent disputes – has so far been mixed. 

Medical technology – medtech – disputes stand out, with companies involved in global litigation – such as continuous glucose monitoring device manufacturers Abbott and Dexcom, and insulin pump competitors Insulet, Menarini and Eoflow – being active before the UPC. However, the number of pharmaceutical and biotech UPC disputes remains low, with many life sciences innovators having opted all of their patents out of the UPC.

After 18 months, there are still only two innovator versus innovator disputes, notably Sanofi v Amgen in which Amgen’s evolocumab/PCSK9 antibody patent was revoked by the Munich Local Division (LD), and the GSK v Pfizer RSV vaccine dispute which is underway before the Dusseldorf LD. Preliminary injunction applications have been heard in the biosimilar disputes Novartis v Celltrion and Alexion v Amgen & Samsung Bioepis. However, only one traditional originator versus generic medicine dispute, between Sanofi v Accord & Others, in which Pinsent Masons is representing Accord, is underway. The reasons for this hesitancy are likely a combination of commercial risk and uncertainties concerning the UPC’s approach to certain substantive issues. 

Beyond life sciences, however, many businesses seem to have determined that the advantages of the UPC outweigh such uncertainties, and with English now the predominant language in UPC cases (4-page / 336KB PDF), the system is seen as accessible to international businesses.

A speedy resolution

Trust in the system has been buoyed by the efforts of the UPC judges to promote its use, and steps taken to ensure that the overarching aims of the court 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. 

The court aims to hand down first instance decisions within 12 to 14 months of proceedings being initiated, which it has achieved with some of its early substantive decisions, such as with its first substantive decision in an infringement action in the case of Kaldewei v Bette, in which the Dusseldorf LD implicitly confirmed the commitment to handing down a decision within the timeframe by rejecting Bette’s request to retrospectively extend a deadline for submitting documents.  Similarly, despite a number of early procedural skirmishes, in the case of Sanofi v Amgen, the Munich Central Division (CD) handed down its decisions in the consolidated revocation actions – the first decision in revocation proceedings – within approximately 13 months of proceedings being initiated.

German dominance

Whether the UPC will maintain this speed remains to be seen, particularly given the demand placed on the German divisions, notably the Munich LD, at which more than a third of UPC cases have been filed in the first 18 months. This early German dominance is not unexpected, with many of the German national patent judges having moved over to the UPC, and the corresponding ongoing decline in German national patent litigation. 

To ensure a better distribution of the caseload, new judges have been appointed and a second judicial panel established at the Munich LD, and the third seat of the central division – in Milan - became operational on 27 June 2024, sitting alongside seats in Paris and Munich.

As time passes, the speed at which the German divisions can issue decisions may slow, and those wishing to obtain a quick decision may be minded to initiate proceedings at one of the quieter divisions. The number of cases at non-German divisions remains lower but is steadily increasing, ranging from 14 infringement cases at the Paris LD and 12 cases before The Hague LD, down to the Ljubljana LD, which is yet to get off the starting block. The Hague LD is expected to increase in popularity in the coming months.

Opening the door to the UPC

The potential reach of the UPC is undoubtedly one of the most obvious advantages of the court. This increased when Romania joined the UPC on 1 September, bringing the total number of participating countries to 18. However, Ireland’s referendum on the ratification of the UPCA has been delayed, and the UPC Court of Appeal has confirmed that “[o]nly countries that have not only signed but also ratified the UPCA are Contracting Member States”.

The reach of UPC decisions may be impacted by the CJEU Grand Chamber’s decision in the case of BSH Hausgeräte v Electrolux, which is due in 2025. The ruling in that case is expected to clarify whether courts in EU member states – to which the UPC is considered equivalent for jurisdictional purposes – have jurisdiction to rule on the validity of patents in non-EU member states.

As many UPC actions are part of wider global litigation, the existence of national proceedings has led to challenges to the UPC’s jurisdiction. Despite the UPC judges’ evident desire to encourage use of the UPC, on the facts of the matters before them, they have taken a pragmatic approach to jurisdictional issues, with the aims of ensuring fairness and enabling parties to obtain early commercial certainty taking precedence.

The UPC Court of Appeal clarified in the case of Mala v Nokia that UPC proceedings may be put on hold in favour of national proceedings where the national proceedings were brought before or during the UPC’s transitional period. However, with early commercial certainty for parties influencing decisions, the Paris LD declined to halt proceedings before it where a national action involved related but not identical parties and where the UPC was expected to issue a determination first. 

European Patent Office (EPO) opposition proceedings are often run in parallel to national proceedings and the same can be said for UPC proceedings. The UPC has a discretion to stay its proceedings where a decision in the EPO opposition proceedings is expected rapidly. In Bitzer v Carrier, the UPC Court of Appeal found that EPO opposition proceedings scheduled four months after the UPC oral proceedings was not sufficiently rapid to justify granting a stay, reinforcing the UPC’s aim of providing rapid decisions.

With that in mind, the EPO is, on its own volition, taking steps to expedite cases when UPC proceedings have been initiated. Given the importance of balancing the benefits and risks of UPC litigation, further challenges to the court’s jurisdiction are expected, but there seems to be a positive intention for the UPC to forge ahead and issue swift determinations.

The UPC Court of Appeal has recently issued a ruling which potentially opens the door for an increase in UPC actions.  In this decision, AIM Sport v Supponer, the Court of Appeal overturned an earlier ruling of the Helsinki LD and ruled 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 opened, not before. As the UPC now plays a key role in patent litigation in Europe, businesses need to keep their opt-out strategies under review. 

A nuanced approach

The Court of Appeal’s decision in the AIM Sport case is clearly favourable to patent rights holders. However, despite early expectations that the UPC would most likely be a pro-patentee forum, its approach to-date has been far more nuanced.

No clear trend for preliminary injunctions (PIs) has yet emerged, although the UPC’s first 18 months has seen the development of a body of case law concerning PIs, with lower courts applying the Court of Appeal’s test set out in the landmark decision of 10x Genomics v Nanostring and developing clearer guidance as to the availability of PIs, which is providing more certainty to parties.

Similarly, at the 18 month point, no clear trends have emerged concerning whether the UPC is more or less likely to uphold or revoke a patent. The UPC has granted a number of permanent injunctions, including in its first substantive decision which saw the Dusseldorf LD award a permanent injunction to Kaldewei, and the first UPC SEP injunction in the case of Philips v Belkin. The court has also shown a willingness to take a flexible approach to such relief, by, for example, carving out certain products from the scope of the injunction in one of the transcatheter heart valve rulings involving Edwards and Meril, demonstrating the balancing act between protecting the rights of the patent holder and, in this case, prioritising patient safety. However, the risk of valuable patents being revoked across all participating member states has already come to fruition for some rights holders. 

Dexcom issued infringement proceedings against Abbott, only for the court to revoke its CGM patent. Amgen’s patent for evolocumab, a highly valuable monoclonal antibody treatment for high cholesterol, was invalidated. More recently, the Munich CD revoked a 10x Genomics patent (55-page / 2.22MB PDF) which had been asserted against NanoString, on the basis that it lacked novelty. 

Claim construction

After 18 months, the approach to claim construction set out by the UPC Court of Appeal in the 10x Genomics v NanoString case, and based on Article 69 of the European Patent Convention (EPC) and the Protocol of Interpretation, which apply equally to validity assessments, is now applied as a matter of course.  This provides that the claim is the decisive basis for determining the protective scope of the patent. The interpretation of a claim does not depend solely on the strict, literal meaning of the wording used, as the description and the drawings must always be used as explanatory aids for the interpretation of the patent claim, but this does not mean that the patent claim serves only as a guideline and that its subject matter may extend to what, from a consideration of the description and drawings, the patent proprietor has contemplated.

In the Amgen case, the Munich CD noted that the patent specification may represent the patent’s own ‘lexicon’, and where terms used in the patent differ from general usage, “it may therefore be that ultimately the meaning of the terms resulting from the patent specification is authoritative”, which may ultimately lead to less wriggle room over claim interpretation.

The outcome of the pending referral (G1/24) to the EPO’s Enlarged Board of Appeal, where clarification on how and when the description and the drawings in a patent can aid the interpretation of its claims has been requested, may be borne in mind by UPC judges in due course, although the court is not bound to follow EPO case law.

The use of prosecution history in assisting claim interpretation is a point of contention, and the UPC's lack of uniformity on this issue to date perhaps not only reflects the divergent views across Europe but also within the judiciary.  The Dusseldorf LD refused to refer to the prosecution history to aid claim interpretation in the case of SodaStream v Aarke and the PI application in the case of Ortovox v Mammut, in contrast to the Munich LD’s decision in the Philips v Belkin case where, contrary to German national case law, it was permitted, as it was in the PI decision in the case of SES-imagotag v Hanshow on which, on appeal, the Court of Appeal declined to comment. The Hague LD has also declined to refer to the prosecution history to aid claim interpretation in the case of Plant-e v Arkyne on the facts of that case. 

As this issue is increasingly coming before the UPC, the divergent approaches means that it will inevitably come before the UPC Court of Appeal to take a definitive position in the not-too-distant future.

Infringement

The Plant-e v Arkyne case is particularly important because it is the first time the UPC has provided guidance on infringement by equivalents. There are diverging approaches across Europe as to how equivalence should be assessed. Clarification of the UPC’s approach to the so-called ‘doctrine of equivalents’ is important because it can help patent owners achieve meaningful protection to stop products which differ immaterially to their protected original. However, it can make it harder for others to navigate patents with any certainty. 

In this decision, The Hague LD found that infringement claims before the UPC should be assessed in two steps, applying Article 69 of the EPC and the implementing protocol, with the court to first evaluate ‘literal’ infringement of the features of the patent, and, as a second step, if the patent is not judged to have been literally infringed, assess whether there is infringement by equivalence.

This second step seeks to answer the question of whether the patent claims, as perceived by the skilled person and seen in light of the description and drawings, leave room for equivalents. According to The Hague LD, this question is to be answered in the affirmative if four sub-questions, originating from the Dutch national courts, are answered in the affirmative.

However, this is only one decision by one local division of the UPC and does demonstrate an element of local flavour because The Hague LD adopted the approach followed by the Dutch national courts. Further decisions are awaited – for example, infringement by equivalence has been pleaded in the Orthoapnea dispute before the Brussels Local Division. Ultimately, the UPC Court of Appeal will determine whether the test proposed in the Hague LD ruling is ultimately the one universally accepted by the UPC.

Other interesting infringement issues that have arisen leave more questions than answers. The double territoriality requirement for indirect, contributory, infringement was touched upon in the Kaldewei case, where the Dusseldorf LD found infringement based on a narrow interpretation of the requirement – i.e. where the supply and putting the invention into effect both take place in the same UPC country – but left unanswered the question of whether this would be satisfied if the invention was put into effect in a different UPC country to that in which it was supplied.

In the same decision, the court found that to rely on the prior use defence against infringement, a defendant must prove a right of prior use for each of the contracting member states of interest. As Bette’s evidence only related to prior use Germany, the defence failed.

Validity: inventive step

To-date, perhaps the most discussed UPC topic concerns the court’s approach to inventive step – probably the most common ground in which patent validity is challenged in general. The UPC was widely expected to assess inventive step using the EPO’s tried and tested problem-solution approach, which, as a first step, involves identifying the ‘closest piece of prior art’.

In the case of 10x Genomics v NanoString, the UPC Court of Appeal neither approved nor criticised this approach. In the early decisions on the merits, the Düsseldorf LD in the case of Kaldewei v Bette did not explicitly follow the EPO’s test, although the Paris LD in the case of Dexcom v Abbott adopted a similar, but less strict approach.

However, in Sanofi v Amgen, the UPC’s first revocation decision, the Munich CD took a more holistic approach – one that is not an EPO-based problem-solution approach but which is not entirely inconsistent with it either.

The Munich CD said that the first step is to determine a ‘realistic starting point’ in the prior art that would have been of interest to the skilled person considering a similar ‘underlying problem’ as that which the patent claimed to solve. It said this realistic starting point may be one of several and does not need to be the most promising starting point. The court said it then needs to ask whether it would be obvious for the skilled person to arrive at the claimed solution in the patent from the realistic starting point. If the skilled person would be motivated to consider and implement the claimed solution in the patent as a next step, then the patent is obvious.

Following that decision, in Meril v Edwards, the Paris CD acknowledged the EPO’s approach but did not apply it, specifically noting that the EPC does not mandate its application and, as a result, it is not bound by it.

After 18 months of the UPC’s operation, this holistic approach, taking a realistic starting point, has been adopted more widely in subsequent decisions, including in the Plant-e case, although it is not yet universally accepted. It is therefore surely only a matter of time before this matter comes before the Court of Appeal for clarification.

First FRAND decision

The Mannheim LD handed down the UPC’s first FRAND decision, despite the parties having, in principle, reached a settlement in respect of the global dispute. However, at the time the decision was handed down, the settlement agreement had not been finalised or signed, and the UPC claims had not yet been withdrawn. It appears that given the global nature of these disputes and the interplay between UPC proceedings and other actions, including those before the UK courts, the UPC judges took steps to set out what they considered should be the UPC’s stall on FRAND issues. 

The court found that it was bound by the CJEU’s decision in the case of Huawei v ZTE, and interpreted this test in line with standards applied by the German national courts. The Mannheim LD found in favour of the rights holder and dismissed the FRAND defence. However, again, this is only one decision by one local division, and ultimately Court of Appeal clarification will be necessary.

What next?

UPC case law will continue to evolve over the coming months. As the UPC moves closer to the third year of its operation and with an increasing number of first instance decisions, attention will turn to the Court of Appeal. With the automatic right of appeal, the Court of Appeal’s case load is expected to increase rapidly until there is at least some clarity as to the UPC’s approach to some substantive issues. This has led to suggestions that a third Court of Appeal panel may be appointed.

Continuing to further the UPC’s aim of ensuring accessibility to SMEs, the establishment of the UPC’s Patent Mediation and Arbitration Centre is starting to take shape. It is expected to play more of a role in resolving disputes in the future.