UPC Unfiltered, by Willem Hoyng – UPC decisions week 03, 2026

18 December 2025 (late published)
Central Division Paris, IMI v Belparts

UPC_CFI_104/2025 ; UPC_CFI_364/2025 

Security for costs

Facts

1. This case started with a revocation action. Belparts filed a counterclaim for infringement.
2. The interim conference was held for the full panel. During the interim conference, security for costs was discussed.
3. In the counterclaim for infringement, IMI asked to order Belparts to provide security for € 500.000, arguing that Belparts is in a bad economic situation. 

The Court

1. The Court cites Art. 69(4) UPCA and R. 158.1 RoP about security for costs, and refers to the principles laid down by the Court of Appeal in the case Chint v Jingoa (CoA_431/2025).
2. The Court considers that since Belparts is established in Belgium, there is no risk with respect to enforcement.
3. IMI has not been able to establish that Belparts would not be able to pay costs.
4. The Court leaves open the question if IMI as defendant in the counterclaim can ask for security.
5. The Court dismisses the request.

The Court

1. A well-reasoned logical decision.
2. The open question if a defendant in a counterclaim for infringement can ask security should, in my opinion, be answered positively. This is different from a counterclaim for revocation lodged by a defendant who is sued for infringement. In the latter case the filing of the counterclaim is in fact acting as defendant, because without filing a counterclaim one cannot defend against infringement with the argument that the patent is invalid.

19 December 2025 (late published)
Local Division Mannheim, Centripetal v Palo Alto 

UPC_CFI_134/2025; UPC_CFI_660/2024 

Infringement / revocation

Facts

1. The main non-infringement defence in this case is the argument that the attacked embodiments were not configured to operate in accordance with the specific way on which the claimant based its alleged infringement, and that with respect to a certain claim feature it would be non-sensical to configure the embodiment in that way.
2. The counterclaim for revocation was based on insufficient disclosure, lack of novelty and lack of inventive step.

The Court

1. The Court finds no infringement, because patent is revoked. The application to amend the patent is unfounded.
2. The Court does not have to decide about allowance of the additional invalidity arguments put forward in the reply to the statement of defence, because the counterclaim for revocation itself results already in revocation.
3. The Court follows the usual decision making process:
   1. it first discusses the scope of the patent (network security);
   2. it discusses the disputed claim features relevant for the decision. The Court gives a broader interpretation to certain claim features than claimant.
4. Thereafter, the Court discusses the counterclaim for revocation, concluding that the patent lacks novelty over a document referred to as “Hawk”.
5. The application to amend is unfounded, which means that the Court does not have to decide if filing 32 auxiliary requests is reasonable and admissible. The Court discusses the different auxiliary requests and holds them not novel or not inventive.
6. With respect of the assessment of inventive step, the Court refers to the Court of Appeal’s decision in Amgen v Sanofi and Merel v Edwards.

Comment

1. A well-reasoned and in my opinion convincing decision, which hinges for a large part on the failed effort of the patentee to give the claims a limited meaning in order to escape on the one hand invalidity while on the other hand being able to argue infringement.
2. I do not understand why the Court says that it does not have to decide whether or not 32 auxiliary requests is a reasonable number because the application to amend is unfounded. The Court deals, as far as I understand the decision, with (only) 12 auxiliary requests. Apparently, this also covers the remaining 20?
3. Also in this case, the claimant (an Irish company), which filed a lawsuit against a US company, should not have filed in this busy German Division but could (and in my opinion should) have filed in for instance the Paris Local Division.
4. I hope that, after reading the decisions of this week, the users of the system now realize that it is not true that German Divisions are more patent-friendly than other Divisions. Representatives bear a responsibility not to overload German Divisions as it will lead to decisions which take (much) longer than 12-14 months, while elsewhere the same quality decision can be obtained within the timeframe promised to the users of the UPC. 

22 December 2025 (late published)
Local Division Düsseldorf, Atlas v Vantiva

UPC_CFI_407/2024; UPC_CFI_686/2024 

UPC_CFI_314/2024; UPC_CFI_568/2024 

Withdrawal

Facts

The parties settled the case. They both filed a withdrawal request and asked for a return of the fees paid for the claim for infringement and the counterclaim for revocation.

The Court

The Court decided accordingly.

Comment

1. The busy Düsseldorf Division was relieved from a case which a responsible representative should have never filed in the busy Düsseldorf Division as the case had hardly anything to do with Germany: the plaintiff was a US company and the defendant a French company. So the most logical place to sue was the Paris Local Division in English, and not in German which even the claimant itself does not speak!
2. The complicated settlement system (for which R. 265 RoP was not even written) should be replaced by a simple letter to the Registry, which can also deal with the straightforward requests for return of fees. Precious court time is saved and it is cheaper for the users of the system. 

29 December 2025 (late published)
Local Division Düsseldorf, Canon v Katun

UPC_CFI_351/2024; UPC_CFI_595/2024 

Costs of simultaneous translation

Facts

The case is in English. The Japanese claimant asks for simultaneous translation of the oral argument in Japanese.

The Court

The Court refers to a decision of the Local Division in The Hague and applies a two stage test:

1. Is simultaneous appropriate?
   Answer: yes, it is important that the people of Canon can follow the oral argument.
2. Should these be procedural costs or own costs?
   Answer: own costs.

Comment

This is a logical decision. I do not know if all UPC Divisions are equipped with hearing rooms in which simultaneous translation without disturbing the proceeding is possible, but if not, the Registry should arrange for such a set up and not allow a so-called “fluster” interpreter. In my experience it is very annoying if, while you are arguing a case, you at the same time hear very softly the Japanese translation. Especially if you are blessed with a very good hearing which has not been harmed by visiting rock concerts or bands like The Who!