Below, Prof. Willem Hoyng provides his unfiltered views on the decisions that were published on the website of the Unified Patent Court (“UPC”) last week. His comments offer a unique insight into the UPC’s case law, as he chairs the Advisory Board of the UPC and participated in drafting the Rules of Procedure of the UPC.
Interested in more of this? Stay tuned and subscribe here for weekly updates.
On the go, multi-tasking or just prefer to listen? “Willem Hoyng’s UPC Unfiltered AI Podcast” – your weekly, AI-generated podcast discussing Willem Hoyng’s commentary on UPC case law of last week, offers a convenient alternative. Listen on Spotify or Apple Podcasts.
10 July 2025
Local Division Hamburg, Nera v Xiaomi
UPC_CFI_173/2024; UPC_CFI_424/2024
Infringement and revocation action
Note: This decision was not published in the week of 10 July 2025. The UPC publishes its decisions under the decision date only, instead of (also) on the publication date. I suggested several times to (also) do that, since otherwise a daily or weekly reader of the latest decisions may miss decisions that are published later. This is what happened with this decision.
Facts
- Nera (a licensing company) sues Xiaomi for infringement of its patents with respect to the mobile phones sold by Xiaomi in Germany and The Netherlands.
- Nera had bought the patent (EP 2 642 632).
- The defendants filed a counterclaim for revocation of the patent in all countries of the UPC for which it was in force, arguing extension of subject matter (Article 123(2) EPC), lack of novelty and lack of inventive step.
- The defendants also argued that the claimant was not entitled to sue and that they did not infringe.
- The claimant filed 66 auxiliary requests.
The Court
- Claimant has standing to sue. R. 8.4 RoP states that the registered owner of the patent should be considered as the patentee. The burden of proof that this information is incorrect, is on the defendant.
- The Court has international jurisdiction with respect to the counterclaim. The fact that not all claims are invoked in the infringement case does not prevent an invalidity attack with respect to the entire patent.
- The claimant is entitled, in response to the defence with respect to the application to amend the patent, to combine auxiliary requests which have already been discussed in depth in the application to amend the patent, without having to ask for approval in accordance with R. 30.2 RoP.
- The number of auxiliary requests is not objectionable. They are considered reasonable in number.
- The Court repeats the principles of claim interpretation of as laid down by the Court of Appeal (in Nanostring v 10XGenomics, UPC_CoA_335/2023) and discusses the meaning of the claim features. The fact that a claim feature (in the English language patent) is incorrectly translated in German and French is not important if the meaning is clear in the light of what is stated in the description.
- With respect to added matter the Court agrees with the defendants that the claim of the patent extends beyond the original application because claim feature 2.1 and also claim feature 3 cannot be inferred from the original application. In feature 2.1, the word “magnetic” in “magnetic substrate” in the original claim has been dropped, resulting in an intermediate generalization. The same is true for feature 3, in which also a limitation has been removed.
- The Court states:
“It is true that not every deviation between the wording of the original application documents and the requested version of the patent to be granted or granted constitutes a prohibited amendment under Article 123(2) EPC. This is because the purpose of Art. 123 EPC is not to preserve the wording of the original application documents, but the technical teachings contained therein (see EPO 20 September 2007 – T 1269/06, Beck RS 2007, 30 688 833, para. 2). Therefore, all changes that leave the information content of the original application documents untouched are unobjectionable. Standardization of the terms used or the replacement of cumbersome but unambiguous expressions by more easily understandable expressions with the same meaning therefore do not infringe Article 123(2) EPC. The requested version does not have to have a literal basis in the original application documents or be within the scope of customary technical expressions; it is sufficient if its technical teaching was originally disclosed (Benkard EPC/Sendrowski, 4th ed. 2023, EPC Art. 123 para. 140, 141).”
and
“by omitting the feature “magnetic” from the wording of the claim, an object is claimed which is more general than the feature complex from which the feature was removed (cf. Benkard EPC/Sendrowski, 4th ed. 2023, EPC Art. 123 para. 173).” - Auxiliary requests 1 and 2 resolve the added matter problem.
- The auxiliary requests 1 and 2 are novel. The Court considers as follows (in my translation):
“For an invention to be considered prior art (Art. 54(1) EPC), it must be clearly, directly and unambiguously found in a single prior art and in its existing form. It must be identical in its essential features, in the same form, with the same arrangement and the same features (LK Munich, 31 July 2024 – UPC_CFI_233/2023, ACT_547520/2023, p. 21). For novelty to be lacking, each feature of the claimed subject-matter must be directly and unambiguously derivable from a single prior art document. This question must be answered from the perspective of the notional skilled person, taking into account his general knowledge at the time of publication of the cited document, if it is prior art under Art. 54(2) EPC (LD Düsseldorf, 28 January 2025 -UPC_CFI_335/2023, ACT_578607/2023, p. 46; LD Hamburg, 30 April 2025 – UPC_CFI_278/2023, ACT 561734/2024,p. 26 under C. I. 1.).
For the assessment of novelty, it is not relevant which problem is solved by a prior art document as long as the problem is not a feature of the claim or is interpreted as such. What is relevant is a feature-by-feature comparison of a claim with the content of a prior art document which shows that all features are disclosed in combination by this prior art document. The decisive factor is whether a prior art document discloses a composition which contains all the elements required for coverage by the claim. For example, if such a composition is described in individualised form in an example of a prior art document, this is sufficient to deny novelty (LD Düsseldorf, 28 January 2025 – UPC_CFI_335/2023, ACT_578607/2023, p. 46; CD Munich, 17 October 2024 -UPC_CFI_252/2023, ACT_551180/2023, p. 33).” - With respect to D5, on which the defendants rely for lack of novelty of claim 1, the Court notes (in my translation):
“It is not absolutely necessary for a prior art document to address the same problem as the patent in order to be prejudicial to novelty. However, the completely divergent problem solution of D5 prevents the skilled man from referring to D5.” - With respect to inventive step, the Court does not refer to the problem-solution-approach, but cites in the first place the text of Article 56 EPC and continues:
“A possible starting point is realistic if it’s teaching would have been of interest to a person skilled in the art who, at the priority date of the patent in suit, wish to develop a product or process similar to that disclosed in the prior art and thus having a similar problem as the claimed invention. In order to assess inventive step the claimed subject matter, as understood, must be compared to the prior art and it must be seen whether it would have been obvious that he arrives at the claimed solution on the basis of a disclosure in the prior art which is regarded as a realistic starting point in view of the underlying problem.” - The Court concludes that the inventive step attack fails.
- The Court concludes that the devices of Xiaomi do (at least) not fulfil claim element 3 as added to claim 1 in the auxiliary request, so there is no infringement.
Comment
- This case is not one which should make patent attorneys, the European Patent Office (“EPO”) and the litigators proud! It contains wise lessons for all these professionals:
a. First of all the patent attorney who drafted the application. They did not do much more than apparently describing and claiming precisely the device as actually conceived by the inventor in all its details, never thinking about the fact that the inventive idea could also be applied in devices with other or different details irrelevant for embodying the invention.
b. During prosecution, the patent attorney clearly broadened the scope of the claims compared to the originally filed claims, without properly looking if there was basis in the description.
c. During examination, the patent attorney surprisingly got away with this broadening, because the examiner apparently did not notice it, while a very simple comparison with the first claim of the application and the new first claim should have immediately drawn his attention to this.
d. Before you start litigation, thorough preparation is extremely important. Do not rely on the work of the EPO but do a new (more creative) prior art search and check critically the entire prosecution file. In all my years as a litigator, there was always more relevant prior art than what was found by the EPO. In this case, such preparation should have resulted in the finding of the extension of subject matter, as it was so blatant. If you then not decide to go through a central limitation procedure, you should prepare your auxiliary requests and ask yourself the question if there would be still infringement. In this case, the action failed because there was clearly no infringement of the language added in the auxiliary request. - I think 66 auxiliary requests is not a reasonable number. The drafters of the Rules of Procedure had in mind a number of about 5 or 6. They wanted to avoid what happens at the EPO, and force the patentee to choose among the different possibilities in order to keep the litigation (also for the defendant) within reasonable boundaries.
- Above at points 7 and 10, I included (in translation) the considerations of the Court with respect to added matter and novelty. I always thought that the criteria (the so-called “gold standard”) were the same, but I read in these points slightly different tests with references to German text books (why are these so relevant?). I have the impression that under these tests it is more difficult to argue lack of novelty than lack of sufficient basis, while in my opinion it should be the same test. I am all in favor for a reasonable and not unrealistically strict application of the gold standard, to avoid invalidating patents for good inventions based on unrealistic considerations with respect to the views of the skilled person.
- Above at point 10, I included the Court’s statements with respect to novelty and prior art document D5. I have always learned that we have an absolute novelty requirement, and that also when no skilled person would or even could have “found” a certain prior art device which is exactly the same as the patented device (or a publication in which such device is described), the claim for that device is invalid. (In the past there have been Patent Acts which stated that prior art which was more than one hundred years old was no longer relevant!) The purpose for which such device is used is totally irrelevant except when that purpose means that the device must have different physical properties than the patented device. The latter is clearly the case here, as the Court convincingly shows, but the first sentence could have been more clear: A different purpose as such is irrelevant.
- The Court states that the teaching of a prior art document is what the skilled person learns from it at the time of publication. I know that this is in conformity with the EPO case law and the law in most UPC countries, but what happens if that teaching changes between the time the prior art is published and the priority date of the patent? My view is that what is relevant is what the skilled person reads in it at the priority date (which is also far easier to establish).
- My remarks with respect to German textbooks is not meant to say that they are of no value. Of course, literature is important, but in several UPC countries courts do not cite literature, as in the end it is (in the UPC) the Court of Appeal or the Court of Justice of the European Union (“CJEU”) that establishes the law, and not mr. X or ms. Y. Moreover, the UPC is an international Court, and although certain representatives seem to think that the UPC will simply follow German law, it seems unwise to reinforce this by referring only to German literature. After all, in other UPC countries there may also be relevant literature. As most judges do not master all languages let alone read all the literature, it is better to stop referring to literature and follow the tradition of the UPC countries which do not cite literature.
22 September 2025
Local Division Düsseldorf, Otec v Steros
Facts
- On 22 September 2025, Otec applied for an order for inspection and evidentiary seizure during an exhibition at the booth of Steros.
- No opposition had been filed with respect to the patent of Otec, but in a case with respect to a different machine of Steros, Steros had filed a revocation claim.
- On its website, Steros offers DLyte machines of the PRO500-Serie, PRO500 Carbide and PRO500 Hybrid.
- At the exhibition in Hannover, Steros shows a DLyte PRO500 Automated Cell.
- After an evidentiary seizure during an exhibition in Cologne with respect to the DLyte Compact machine, Otec has started proceedings on 12 June 2025.
- Otec states that it could not inspect the PRO500 machines at the exhibition in Cologne because the products were not exhibited. Otec further states an anonymous test purchase of a PRO500 machine (of over €100.000) is impossible.
The Court