Background
Jevtana® (cabazitaxel) in combination with prednisone or prednisolone is approved for the treatment of patients with hormone-refractory metastatic prostate cancer (mCRPC) that have previously been treated with a docetaxel-containing regimen.
Sanofi’s patent EP2493466 was granted with second medical use claims encompassing the approved indication.
The patent was maintained as granted at EPO Opposition. The Opposition decision was subsequently appealed. In June 2025 the EPO Boards of Appeal (TBA) also upheld the patent as granted. Decision T 0136/24 includes a detailed discussion of inventive step, in particular the concept of “reasonable expectation of success” based on prior art disclosing a clinical trial protocol.
At the UPC, Sanofi launched infringement actions against Stada, Dr Reddy’s, Zentiva and Accord. Each defendant counterclaimed for revocation [1]. In December 2025 the Munich Local Division (LD) of the UPC revoked the patent. The LD Decision (UPC_CFI_146/2024, 496/2024, 147/2024, 374/2024, 148/2024 and 503/2024) also discussed inventive step in detail and concluded that the granted claims did not possess an inventive step.
This article compares how inventive step was evaluated by the EPO and UPC and explores how they reached different decisions.
Key prior art
The prior art included preclinical and clinical studies involving the testing of cabazitaxel:
Opponents’/defendants’ position
The opponents/defendants argued that, in accordance with EPO, French and German case law, where the prior art disclosed that a clinical study with the same active agent(s) for the same therapeutic indication had been proposed or was underway, a “reasonable expectation of success” was implied by the mere fact that the study had been authorized – unless there was evidence dissuading the skilled person from following the prior art clinical study (for example, a ‘negative’ pointer). The claims should thus be found lacking in inventive step.
The opponents argued that there were no such negative pointers in the prior art. Moreover, they argued that, starting from the TROPIC study protocol, there were several ‘positive’ pointers in the prior art that would have led to a reasonable expectation of success. These included:
Patentee’s position
The patentee argued that, in accordance with EPO case law, the effect of the disclosure of a clinical trial protocol on inventive step should be analysed according to the specifics of each case. They argued that the TROPIC study was merely a description of a clinical trial which does not disclose any effect of treatment, whereas the invention provides a treatment which prolonged the life of patients. The TROPIC study protocol would have, at most, provided the skilled person with a hope of succeeding.
They argued inter alia that the preclinical, phase I and phase II data were not strong enough to create a reasonable expectation of success.
EPO inventive step assessment
The TBA applied the problem-solution approach and starting from the treatment arm of the TROPIC study, the TBA formulated the problem as “to put into practice the effective treatment of prostate cancer with cabazitaxel in co-administration with prednisone in patients with mCRPC who have been previously treated with a docetaxel-based regimen and who have prostate cancer that progressed during or after that treatment“.
The solution offered by the patent was considered non-obvious. The TBA considered that the skilled person’s expectation of success had to be considered in the context of the TROPIC study’s primary endpoint, which was overall survival. As such, the skilled person would need to have a reasonable expectation that the treatment arm of the TROPIC study would lead to improved overall survival for there to be a lack of inventive step.
Overall, the TBA held that the positive pointers referred to by the opponents would not have given rise to a reasonable expectation of success. The phase I data was considered “sparse and incomplete”. Further, they held that, as overall survival is linked to the type of cancer and stage of disease progression, the phase II clinical study with breast cancer patients could not be considered a positive pointer. The fact that the TROPIC study was nearing completion at the priority date, without information on the parameters selected for monitoring, was considered neither a positive nor a negative pointer.
The claims were held inventive.
UPC inventive step assessment
The LD applied the test endorsed in the recent UPC Court of Appeal Amgen vs Sanofi decision (UPC_CoA_528-2024, 529/2024). This decision applied a ‘holistic’ approach to assess inventive step.
The LD formulated the objective problem as “to provide a therapeutic option for treating patients suffering of castration resistant metastatic prostate cancer who have been previously treated with docetaxel-based regimen and have prostate cancer that progressed during or after that treatment “.
The LD considered inventive step starting from the treatment arm of the TROPIC study.
Starting from the TROPIC study protocol, the LD evaluated all the various indicators and arrived at an overall assessment. Whilst acknowledging the limitations of the prior art phase I and phase II trial data, the LD considered that these data nevertheless demonstrated that cabazitaxel is safe and tolerable in patients. Collectively, the data were considered to give rise to a reasonable expectation of success. Moreover, the fact that the TROPIC study was nearing completion at the priority date, also provided a reasonable expectation of success.
The claims were held to lack an inventive step.
Points of agreement between the EPO and UPC
Both the TBA and LD endorsed the patentee’s view that the impact of a disclosure of a clinical trial protocol on inventive step should be evaluated according to the specifics of each case. They rejected the suggestion that approval of a clinical trial implicitly gives rise to a reasonable expectation of success.
Both decisions also endorsed an approach in which both positive and negative pointers are evaluated and an overall assessment is provided.
Divergence between the EPO and UPC
A critical point of distinction between these decisions is the threshold applied to evaluate “reasonable expectation of success” starting from the TROPIC clinical trial protocol. The TBA considered the skilled person would require a reasonable expectation of successfully achieving the primary endpoint of the trial i.e., improved overall survival. The LD disagreed and considered the skilled person would require a reasonable expectation of achieving either increased overall survival or a palliative effect. As such, a more stringent threshold for “success” was applied by the TBA.
This divergence cascaded through to the assessment of the positive pointers, as illustrated by the TBA’s and LD’s comments on the significance of the phase I and II data in the prior art. In particular, the TBA considered these data were not positive pointers because they provided no insights on overall survival in the claimed patient group, whereas the LD held that collectively they were a positive pointer because they were indicative that cabazitaxel would be safe and tolerated in the claimed therapeutic indication.
Summary
Both decisions confirm that the impact of clinical trial protocol prior art on inventive step must be evaluated on a case-by-case basis at the EPO and UPC. That said, these divergent decisions clearly demonstrate that the question of “reasonable expectation of success” can be decided differently when two different forums are presented with essentially the same case.
[1] Sanofi and Accord subsequently settled prior to the oral hearing.
Meet our UPC team.
