Irides: Weekly patent litigation update

This edition features updates from: The European Patent Office (EPO), Ireland, the Netherlands and the Unified Patent Court (UPC).

The Irides Weekly Update is our round-up of patent litigation news highlights from around the world.
 

EPO

The Technical Board of Appeal provides guidance on obviousness over clinical trial protocols.

On 15 September 2025, the Technical Board of Appeal (TBA) published its decision upholding Sanofi’s patent EP 2 493 466. The patent claims the use of cabazitaxel in combination with prednisone or prednisolone to treat prostate cancer in patients with castration resistant metastatic prostate cancer (mCRPC) which had previously progressed during or after a docetaxel based treatment regimen. Cabazitaxel is marketed as Jevtana®. 

The patent was opposed by 13 opponents and was upheld by the Opposition Division (OD) on 25 January 2024.  On appeal to the TBA the opponents ran a number of attacks, including added matter, insufficiency, loss of priority, lack of novelty, and lack of inventive step. The TBA rejected all of these attacks. The decision is of particular interest for the TBA’s approach to inventive step where the prior art is a protocol for a clinical trial investigating the claimed treatment effect.

The combination of cabazitaxel and prednisone was investigated in the phase III TROPIC trial. The opponents’ preferred starting point was the treatment arm of this trial. The key question is whether this arm provided the skilled person with a reasonable expectation of success. Success in this context meant meeting the primary endpoint of the trial, which was an improvement in overall survival compared to the comparator.

The opponents argued that there was an established body of case law that a reasonable expectation of success was generally implied by the fact that the trial had been authorized, unless there was a negative pointer in the prior art. The TBA provided an extensive analysis of the clinical trial case law to conclude that no such general presumption existed. Instead, each case should be decided on its own particular facts. 

As to those facts, the opponents argued that there was a reasonable expectation of success because:

1. cabazitaxel was in the same class as the known treatment docetaxel;
2. existing data was promising;
3. the fact that a phase III trial is being carried out indicates progression along a successful development path;
4. the approving authorities must have had access to additional, unpublished data to justify their approval;
5. the fact that the study was nearing completion suggested it had not been terminated early for futility; and 
6. the fact that the trial sponsor had invested money into the trial. 

The TBA did not find that any of these pointed towards an expectation of success. Although cabazitaxel had been designed to overcome resistance to docetaxel, a successful outcome was not guaranteed. Existing data were minimal and in other indications and so gave no information about prospects of success in mCRPC. In fact, there was no phase II trial in mCRPC and so cabazitaxel was not progressing along the successful development path. Even if the approving authorities had access to unpublished data, those data were not available to the skilled person and so irrelevant to the inventive step assessment. Without evidence on when and how any futility assessments should be conducted, no conclusions could be drawn. Finally, the investment choices of the trial sponsor without detailed information on the rationale underlying those choices could not be determinative. 
 

Ireland

Approved judgment published from Irish High Court finding Bayer’s rivaroxaban patent invalid.

Readers will be aware of the ongoing international litigation relating to Bayer’s patent, EP 1 845 961 (EP 961). EP 961 claims a once-daily dosing regimen for Bayer’s rivaroxaban product, a factor Xa inhibitor. In June 2025, following a revocation action filed by Sandoz and Rowex, the Irish High Court revoked the Irish designation of EP 961, finding it invalid for lack of inventive step. Following some debate in relation to confidentiality of secondary evidence, an approved copy of this judgment has recently been published, along with a judgment related to confidentiality. 

Merits judgment

In finding that EP 961 lacked inventive step, the Court ruled that two prior art documents, which set out phase I data from various studies, could be read together due to the fact that one document was expressly referenced in another. One of these documents, termed ‘Harder’, provided an express suggestion that rivaroxaban might be suitable for a once-daily dosing regimen. 

Bayer’s evidence that the skilled person would not have been motivated to test once-daily administration of rivaroxaban based on Harder was deemed unconvincing. Bayer relied upon evidence from experts who had also provided evidence in the UK proceedings, but this evidence was criticised by the Irish Court. The Court asserted that phase III trials would be required to identify a correlation with clinical effect, so the skilled person would not exclude the results of assay studies in the absence of this correlation. Further, phase I trials were not powered to show statistical significance, so the skilled person would not dismiss the results on this basis.

The Court also considered that secondary evidence (published post-priority) was relevant in this case (see below). This evidence consisted of documents relating to the development of once-daily dosing of rivaroxaban by Bayer. The Court concluded that these documents supported Sandoz’s experts’ position that the assay in Harder was the motivation to progress once daily dosing. These secondary documents therefore further undermined Bayer’s experts’ criticisms of the Harder study data. 

Finally, the court considered decisions issued in other jurisdictions and noted that all final decisions (issued in the UK, Australia and Switzerland) had found the patent invalid. Most Courts that had found the patent valid concluded that no weight should be placed on the Harder study due to the lack of correlation with clinical effect. The same conclusion could not be reached based upon the evidence before the Irish Court.

Confidentiality judgment

Bayer had argued that certain secondary documents (obtained by Sandoz via discovery) relating to the development of once-daily dosing of rivaroxaban, and the relevant portions of the judgment which cited these documents, should be kept confidential given that they related to Bayer’s internal development processes. 

The Court assessed whether Bayer had established a real risk of harm to its commercial interests in not protecting this information. The Court noted that the development process of EP 961 had already been extensively publicly discussed by Bayer in numerous journal articles, and Bayer did not assert confidentiality over some other documents at trial that related to its internal development processes.

Bayer’s arguments that the secondary documents were of a type that are “typically regarded as valuable” were deemed too general. Bayer had failed to identify anything within the documents which revealed or disclosed something more than had already been revealed, or something different than what any person within the industry would expect. Therefore, there was nothing in the body of documents which presented a risk of harm to Bayer and they were not deemed to be confidential.
 

The Netherlands

Hague District Court upholds Janssens Ustekinumab second medical use patent.

On 10 September, the Hague District Court handed down its judgment upholding Janssen's second medical use patent, EP 3 883 606 (EP 606), covering its product Stelara®. The Court dismissed Samsung Bioepis’ (SB) claims that EP 606 was anticipated and obvious over the prior art. 

EP 606 relates to the treatment of patients with moderately to severely active ulcerative colitis (UC) with the antibody ustekinumab with a defined dosage regimen. At the priority date (20 November 2018), the first-line treatment for moderate to severe UC and Crohn's Disease (another inflammatory bowel disease) (CD) consisted of administration of corticosteroids (immunosuppressants) to control symptoms. However, while effective for this purpose, long-term administration of corticosteroids can lead to serious side effects and patients may become dependent on them over time.

The dispute between Janssen and SB focused on the last integer of claim 1, as underlined (Integer 2.5).

Claim 1 : "An antibody for use in a method of treating moderately to severe active UC…wherein the method comprises (a) intravenously administering the antibody at week 0 of the treatment; (b) subcutaneously administering the antibody… at week 8 of the treatment and in a maintenance dose every 8 weeks or every 12 weeks after the treatment at week 8, wherein the subject is in corticosteroid-free clinical remission at least 44 weeks after week 0".

Entitlement to priority

The arguments around entitlement to priority focused on what Integer 2.5 meant for the timing of Corticosteroid-Free Clinical Remission (CFCR). Did Integer 2.5 mean it could be achieved long before week 44, as long as the patient was still in remission at week 44, or did it also include a situation where early CFCR was achieved, even if the patient had relapsed by week 44? SB argued that the latter was not disclosed in the priority document. The Court preferred the former interpretation, thereby dismissing SB’s entitlement to priority challenge.

Novelty

SB argued that Integer 2.5 was anticipated by several pieces of prior art, including Janssen's protocol for the phase III clinical trial that established the efficacy of ustekinumab as a treatment for UC (the Protocol), from which they asserted Integer 2.5 followed as an inevitable result.

The Court disagreed. For a prior art document to be novelty destroying, all features must be directly and unambiguously disclosed (whether explicitly or implicitly) in that document. While the Protocol described the design of the phase III trial, the dosing regime and the fact that CFCR would be assessed at week 44 of the maintenance phase - it did not unambiguously disclose that patients would achieve CFCR within 44 weeks after starting the maintenance dose.

The Court dismissed the remaining cited pieces of prior art on the same grounds that SB applied the wrong novelty standard for second medical use claims.

Inventive Step

SB argued that claim 1 was obvious in light of the Protocol by itself, as well as in combination with the (i) CGK and (ii) with the other cited prior art.

The Court identified that the key question when assessing the inventiveness of claim 1 was whether the skilled person after reviewing the prior art would have carried out the research "with a reasonable expectation of success". The most interesting commentary came from the Court's assessment of the standard of reasonable success when faced with prior art that is a clinical trial protocol or an announcement of a clinical/phase III trial. SB argued that the skilled person would have had a reasonable expectation of success based on the Protocol because Janssen had not sufficiently demonstrated an expectation of failure. The Court disagreed with this argument, and referencing statements from a 2024 TBA decision (T 1941/21) and from the parallel UK High Court proceedings, held that a reasonable expectation of success cannot solely be derived from the announcement of a clinical trial (even if it is a phase III study). Rather, a weighing up of the particular circumstances needs to be considered in each case.

SB also argued lack of inventive step based on the Protocol combined with the skilled person's CGK about the successful treatment of a different disease CD with the dosage regimen claimed in EP 606. The Court was quick to dismiss this argument, holding that the skilled person would not equate the treatment of the two diseases due to (i) the known difference in the signalling pathway in CD compared to UC; (ii) the difference in clinical profile and disease progression in UC and CD patients; (iii) and the knowledge that a number of other drugs used for CD were known to be not effective in UC (and vice versa).

For these reasons, the Court found EP 606 to be novel and inventive over the prior art.

SB has announced that it has appealed the decision.
 

UPC

Court of Appeal provides guidance on request for unitary effect
(UPC_CoA_796/2025). 

On 16 September 2025, the Court of Appeal dismissed an application to annul a decision of the EPO to reject a request for unitary effect under r.97 RoP. In doing so, the Court provided guidance on the procedure for challenging decisions from the EPO with respect to requests for unitary effect and the need for strict adherence to deadlines associated with correcting deficiencies in the original request. 

Following the grant of the patent at issue, Bodycap, CNRS and Université de Rennes (the Co-proprietors) had filed a request for unitary effect at the EPO. On 27 January 2025, the EPO invited the Co-proprietors to correct a deficiency in the request, namely the particulars of one of the Co-proprietors did not match up with the details on the EP register (there were differences in the University’s name and address). Pursuant to r.7(3) of the Rules relating to Unitary Patent Protection (UPR), the Co-proprietors had a non-extendable period of one month to correct the deficiencies. The Co-proprietors filed their response on 3 March 2025, thereby missing the deadline and causing the EPO to reject the request for unitary effect.

On 8 July 2025, the Co-proprietors lodged an application at the Milan Central Division to reverse the EPO’s decision by way of an interlocutory revision (r.91 RoP) or, in the alternative, an annulment of the decision to reject the request for unitary effect (r.97 RoP). On 6 August 2025, the Court of First Instance rejected the application, citing the exception in r.85.(2) RoP which means that interlocutory revision is not available in the context of an application under r.97 RoP. As to the annulment, the Court agreed that the request for unitary effect had been deficient (with respect to the particulars of the Co-proprietor) and the appellants had failed to rectify the request within the non-extendable one month period (previously reported here).

On 26 August 2025, the Co-proprietors lodged an appeal seeking to overturn the first instance order, arguing, inter alia, that interlocutory revision by the EPO is only excluded in the context of expedited actions; given that the parties had not requested expedition, interlocutory revision by the EPO should be possible. In reaching its decision, the Court held that an application under r.97 RoP constitutes an expedited action due to the timings set out in the rules and the pace at which the application must move forwards (e.g. “as soon as practicable” and three-week deadlines). Therefore, the application to annul the decision falls within the exception in r.85(2) RoP and interlocutory revision by the EPO is excluded. On the application to annul, the Court noted that failure to observe a time limit can, under certain circumstances, be subject to a request for a re-establishment of rights (r.22 UPR). However, r.22(6) UPR includes an express exclusion for the one-month period in r.7(3) UPR. Accordingly, compliance with this deadline is mandatory and an applicant who fails to do so cannot have their rights reinstated. The Court therefore dismissed the appeal.
 

UPC

UPC orders Huawei to disclose licence agreements 
(UPC_CFI_247/2025).

On 16 September 2025, the Mannheim Local Division issued an order requiring Huawei to disclose certain licences in its FRAND and SEP litigation with MediaTek.

Huawei objected to the disclosure of two of these licence agreements on the grounds that (i) they were not relevant to the smartphone technology in suit and therefore not relevant for a comparable licence exercise; and (ii) because Huawei had not been able to obtain approval to disclose these licences from the respective licensees as required by the confidentiality terms of the licences.

The Court ordered Huawei to disclose the two licences, in addition to the licence agreements mentioned in its pleadings. In reaching this decision, the Court clarified that it has discretion under r.190 RoP to order the production of evidence. It also noted that, when assessing such an application, it will take into account the stage of the proceedings.

The Court ordered disclosure of the two licences on the basis that there was no evidence of other equally suitable licences being available. The Court further noted that disclosure of the licences would enable a proper comparative exercise to be undertaken, which would assist the Court (and Mediatek) in determining whether Huawei’s offer is FRAND compliant. However, the Court did allow Huawei to redact those parts of the licences on which it did not intend to rely.