The UPC’s approach to validity in PI proceedings – the scales are balanced

Before the start of the UPC, there was much speculation about how the new court would approach the grant of preliminary injunctions, in particular how it would deal with cases where there is doubt about the validity of the patent, given that the Rules of Procedure give little guidance on this issue and CJEU case law suggests that patents enjoy a presumption of validity from the date of publication.

After 18 months, the message for patent owners as well as defendants in PI proceedings at the UPC, is that all UPC judges, regardless of their location, appear to be making fair and balanced decisions in this regard. This article takes a closer look at some of the UPC case law relating to the standard of patent validity that has been applied by the court when considering whether or not to grant a preliminary injunction, and the guiding principles that emerge from these cases.

In general, a patent proprietor has a statistical chance of success in PI proceedings before the Unified Patent Court of around 50%. Looking into the cases, this statistic reflects the well-balanced approach taken by the Unified Patent Court in PI proceedings so far. While some aspects are still open, for example the circumstances that would constitute an unreasonably long delay detrimental to urgency[1], there are other aspects on which the courts have already issued considerable guidance.

One of the most pressing questions at the opening of the UPC was the criteria that the new court would apply for granting preliminary injunctions when the validity of the patent in suit is challenged by the defendant.

The Rules of Procedure are rather brief in this regard and say in R.211(2):

“In taking its decision the Court may require the applicant to provide reasonable evidence to satisfy the Court with a sufficient degree of certainty that the applicant is entitled to commence proceedings pursuant to Article 47, that the patent in question is valid and that his right is being infringed, or that such infringement is imminent.” (Highlights added by the author).

One might recall that the CJEU, in its decision in Phoenix vs. Harting of April 22, 2022 (Case C-44/21), said that “European patents enjoy a presumption of validity from the date of publication of their grant”. It might therefore have been expected that the UPC would apply a rather low standard, so that validity was generally presumed based on the fact that the patent had been granted. In fact, the case law has shown that the UPC courts go into great detail in their assessment of validity, even if some courts reasonably request that the defendants limit themselves to their three best invalidity attacks in order to preserve the character of PI proceedings as summary proceedings (Munich Local Division Dyson vs. SharkNinja).

Only nine months after the start of the UPC, the Court of Appeal took the opportunity to provide guidance on the standard for a PI being granted or refused (decision of February 26, 2024, UPC: CoA_335/2023, 10xGenomics/Nanostring). The CoA held that to justify a patent as being “PI-compatible”, the court must be convinced that it is more likely than not that the patent is valid. The burden of presentation and proof of invalidity lies with the defendant.

The LD Düsseldorf (order of December 11, 2023, UPC_CFI_452/2023, Ortovox/Mammut, German) found in an ex parte order that there is no requirement that the patent must have survived two-sided validity proceedings to be sufficiently “secure” to be the basis of a preliminary injunction. In this case, the anticipated invalidity attacks that emanated from a nullity action by one of the defendants against the Swiss part of the European patent did not convince the court that there was reasonable doubt that the patent in question was valid. On April 9, 2024, the same LD confirmed that the fact that a patent was battle-tested in two-sided national validity proceedings or opposition proceedings before the EPO is a strong indication of the validity of the patent (GRUR 2024, 932, German, also published here), but it is not an absolute requirement for summary proceedings. Another important indication is given if the validity of a patent has not been challenged for several years following its grant. In September 2025, the CoA agreed with the findings of the first instance court and upheld the PI on appeal (CoA Mammut vs. Ortovox).

The LD Hamburg, in its decision of June 26, 2024 (UPC_CFI_124/2024, Amgen vs. Alexion), also carefully weighed up the criteria that are necessary to consider a patent as a sufficiently secure basis for granting a preliminary injunction. The court said that the UPC has to form its own view on the validity of a patent based on the principles of the EPC. However, it not only has to provide its own evaluation on validity, but also must consider whether it is more likely than not that the EPO will uphold the patent when parallel opposition proceedings are ongoing, because the EPO has the competence to invalidate a patent for the whole UPC territory. Of course, since both the EPO and UPC apply the EPC, there should normally be no differences in the evaluation of validity. Such differences may however occur if, as in this case, the infringement court applies a different claim construction than the EPO did in the grant proceedings. Since the LD Hamburg considered it to be more likely than not that the EPO would revoke the patent, a PI was refused. In this case, the LD Hamburg also considered the third-party observations that had been filed during prosecution, although these mostly dealt with formal aspects of the original patent application rather than arguments that had been central to the defendants’ case. The LD Hamburg made it very clear that it could not blindly follow the decision to grant the patent.

Finally, the recent decision of the LD Milan in Cardo v Asmax (UPC_CFI_643/2024 Cardo vs. Asmax) suggests that the UPC may take into account the volume and quality of the prior art identified by the EPO during prosecution of a patent when assessing validity for the purpose of granting an ex parte PI. In this case, the court took the view that because only three prior art documents had been identified by the EPO during prosecution, and because none of these disclosed or taught key elements of the claimed invention, the patent in question was most likely valid. The court also gave some weight to the fact that a comparable patent had been granted in the United States with essentially the same claim language.

From these decisions, the following principles emerge:

• The burden of presentation and proof of the invalidity of the patent in suit lies with the defendant;
• The fact that the patent was granted is not sufficient per se for a grant of a PI and the UPC will itself evaluate whether the patent is more likely valid or not;
• The UPC has to form its own view on the validity of a patent based on the principles of the EPC, but in the event of parallel opposition proceedings, it also has to consider the likelihood that the EPO will revoke the patent;
• If a patent is battle-tested in two-sided national validity proceedings or opposition proceedings before the EPO, this is a strong indication that the patent is valid, but a battle-tested patent is not an absolute requirement for summary proceedings;
• Third-party observations during prosecution may play a role in the invalidity analysis, if they deal with arguments made by the defendant;
• A low volume of relevant prior art identified by EPO, as well as any parallel granted patent in the US with similar claims, may build confidence in the validity of a patent underlying an ex parte PI request.

Overall, it is good to note for both claimants and defendants that the UPC judges appear to be taking a very balanced approach and are making an effort to deliver fair and balanced decisions, regardless of their geographical location. The decisions of the past 18 months have already shaped many of the principles governing the grant of preliminary injunctions, in particular those relating to the standard of validity that should be applied by the court. The degree of legal certainty resulting from this case law will be welcomed by all stakeholders.

[1] The CoA has recently clarified that when considering urgency, the period to be assessed starts from the day on which the applicant had or should have had such knowledge of the infringement that allowed them to make a promising application for interim measures. Whether or not there has in fact been an unreasonable delay depends on the specific circumstances of the case (CoA Mammut vs. Ortovox).