The Unified Patent Court (UPC) has established a “more likely than not” standard for assessing patent validity in provisional measure proceedings, as set out in NanoString v 10x Genomics. This test becomes particularly complex when an EPO opposition is pending in parallel. Recent decisions, including those from the Local Division Hamburg in Alexion v Samsung Bioepis/Amgen, confirm that the UPC must consider not only its own case law but also the likelihood of the EPO revoking the patent. This dual-track approach means the stricter standard—whether from the UPC or the EPO—will apply, potentially making it harder for patentees to secure provisional measures. While this may disadvantage patentees, the “more likely than not” test remains more favourable than national standards in many UPC countries. The hope is that, over time, UPC and EPO standards will converge, simplifying the assessment of validity in such cases.
17 February 2026 by Simmons + Simmons LLP
The Court of Appeal set the test for provisional measures concerning preliminary injunctions already at an early stage in NanoString v 10x Genomics on 26 February 2024 (CoA_335/2023). In this decision, the Court of Appeal also formulated the standard of the validity test for the patent in dispute which led to the “more likely than not that the patent is not valid” formula. This test of probabilities of likely invalidity or invalidity constantly has been applied in the following decisions by the UPC.
The grounds on which a patent has to be revoked are set forth in Art. 100 EPC. The UPC operates in the same legal framework than the EPO when assessing validity of a European Patent. Applying the same legislative provisions by different authorities not necessarily leads to the same result. (Slightly) different approaches have been emerged by the UPC for the inventive step compared to the EPO. With regard to added matter one could question whether the decisions of the UPC and the EPO are completely aligned, even if the gold standard for added matter is referred to by both.
Such nuances in the application of the rules become particularly relevant when an EPO opposition and a UPC revocation action are running in parallel. In proceedings on the merits Art. 33.10 UPCA and Rule 295 (a) RoP provide some guidance how the relationship between such parallel actions should be dealt with and the Court of Appeal provided initial guidance in the application of these provisions in its decision Carrier Corp. v Bitzer of 28 May 2024 (CoA_22/2024).
Art. 33.10 UPCA and Rule 295 (a) RoP are, however, not applicable in provisional measure proceedings. It also is very much questionable whether provisional measure proceedings are suitable for a stay when an EPO opposition is pending, even if a decision in such proceedings may be expected to be given rapidly. This means that the UPC seized with an application for provisional measures while an EPO opposition is pending, must rule on the application and apply the more likely than not test. The UPC cannot wait for an EPO decision by means of a stay.
The UPC seized with the application for provisional measures in first instance has to assess the validity of the patent in dispute based on the existing (and in the future developing) UPC case law for the grounds of invalidity of a European Patent. This is an obvious starting point. The question, however, is, whether the UPC can stop there, or more specifically whether the UPC also does not need to examine the more likely than not test of validity of the patent in suit based on the case law and practice of the EPO concerning the same grounds of invalidity.
The Local Division Hamburg dealt with this question in the two (parallel) cases Alexion v Samsung Bioepis and Alexion v Amgen (CFI_123/2024 and CFI_124/2024), decision delivered on 26 June 2024 (the decisions are identical on the questions relevant here). First, the court stressed that the UPC has to form its own view of the validity of the patent in dispute also in preliminary measure proceedings. The court, however, soon comes to the conclusion that it cannot base its decision solely on its own opinion of validity if an opposition at the EPO against the patent in suit has been filed. In this case, the UPC also has to consider the likelihood of a decision invalidating the patent by the EPO, as the EPO has the competence to invalidate a patent for the EPC countries that cover the UPC territory as a whole. The Local Division Hamburg correctly notes that neither the UPCA nor the Rules of Procedure contain a provision such as Art. 33.10 UPCA. The only hint is Rule 209.2 (a) RoP which underlines the relevance of a positive decision by the EPO in an opposition procedure or by national courts on validity. When assessing the probability of validity, the UPC also needs to take into account the likelihood of an invalidity decision of the patent in suit by the EPO. In the case at hand, the Local Division Hamburg concluded that it is reasonable likely that the EPO will revoke the patent in suit due to lack of sufficient disclosure (Art. 83 EPC).
These decisions of the Local Division Hamburg went on appeal. The Court of Appeal rejected the appeal of the patentee by decisions of 20 December 2024 (CoA_402/2024 and CoA_405/2024). The Court of Appeal did not come back to the question of the standard to be applied by the UPC concerning the likelihood of validity of a patent in provisional measure proceedings. The Court of Appeal rather followed a different claim interpretation than the Local Division Hamburg and rejected the appeal for reasons of non-infringement. It would have been interesting to see which position the Court of Appeal would have taken on the test for the likelihood of validity in provisional measure proceedings when an EPO opposition is pending. Interestingly, the Court of Appeal refers at various places in its decisions to the parallel TBA decisions of the EPO concerning other divisional patents of the patent in suit in the context of claim interpretation.
A careful analysis of the legal framework supports the conclusion of the Local Division Hamburg. The EPO has the authority to invalidate a patent for the EPC countries if the conditions of patentability are not met. This invalidity decision is binding and cannot be overruled by a national court or the UPC. Of course, this principle also applies to the UPC. A decision to invalidate a European patent for the UPC territory cannot be overruled by the EPO. However, the fact that the EPO can invalidate a patent with binding effect (and overrule the UPC), points into the right direction. In provisional measure proceedings where it concerns an initial assessment based on probabilities of validity, the UPC must take parallel opposition proceedings and decisions or opinions rendered in said proceedings into account. This, by the way, also concerns decisions of the EPO about divisional patents to the patent in suit (as in the Alexion v Samsung Bioepis/Amgen cases). Moreover and probably even more importantly, the UPC also has to assess the likelihood that the European patent will be invalidated in the opposition procedure based on the grounds of invalidity as interpreted and applied by the EPO in its case law. The Local Division Hamburg has formulated this quite accurately in its decisions:
“This is because the court cannot assume a secure legal status by way of preliminary measures if a revocation of a patent by an organisation appointed to make a decision appears likely.”
This in principle leads to a dual test: validity pursuant to UPC and EPO case law. This is particularly relevant when the EPO seems to apply a stricter standard of the patentability requirements as the UPC. A possible inroad most likely relates to the prohibition of added matter and sufficiency of disclosure. If based on the strict case law of the EPO for added matter the invalidity of the patent in the parallel EPO procedure seems more likely, no provisional measure can be ordered, even if the UPC based on probabilities could still see a basis for the patent claim in the application as filed. In practice, the dual test means that the UPC when assessing the probability of validity in principle needs to apply the stricter practice, either of the UPC or the EPO, with regard to the requirements of patentability.
As far as we know, this dual‑track ‘validity’ approach developed by the Local Division Hamburg has not been explicitly revisited or overturned in later UPC case law on provisional measures in the presence of parallel EPO opposition proceedings.
The necessity of a dual test may have more disadvantages for the patentee than for a potential infringer as it leads to the application of the stricter test. This may be less critical, as in the context of provisional measures the UPC is still assessing probabilities and the test “more likely than not” is quite favourable for the patentee compared to national standards in most UPC countries, as for example Germany. Of course, it would be desirable that any differences in the approach of the patentability requirements by the UPC and the EPO will diminish, ideally completely disappear. Then the dual test will turn out to be a single test against the requirements of patentability based on the joint case law of the UPC and the EPO.