Inventive step at the UPC: Court of Appeal sets definitive test
The Court of Appeal has adopted a holistic assessment of inventive step and has moved away from the EPO’s problem solution approach.
Assessment of inventive step as confirmed by the UPC Court of Appeal
Two landmark decisions have been issued by the first and second panels of the Court of Appeal on the same day, 25 November 2025: Meril v Edwards (Edwards) and Amgen v Sanofi (Amgen).
The decisions have clearly been coordinated and agreed upon between the two panels of the Court of Appeal, with sections of the text common between the two decisions. In both decisions, the Court of Appeal uses the headnotes to provide guidance on various principles to be applied at the UPC. Headnotes 8 to 12 of Edwards and headnotes 11 to 16 of Amgen set out the same inventive step approach, which the Court of Appeal explicitly states is the approach of the UPC.
The assessment of inventive step set out by the Court of Appeal can be summarised as follows:
- Define the skilled person and their common general knowledge at the effective date.
- Establish the “objective problem” of the invention from the perspective of the skilled person at the effective date.
- Assess whether the skilled person, starting from a “realistic starting point” in the art in the relevant field of technology, wishing to solve the objective problem, would (and not only: could) have arrived at the claimed solution.
Perhaps owing to the different facts before the Court of Appeal in each case, the two decisions provide further headnotes on inventive step that are unique to each decision. Some comments on these additional headnotes follow below.
Importantly, the Court of Appeal clarifies in headnote 13 of Edwards that “it is not necessary to show improvement of the technical teaching as defined by the patent claims over the prior art” for an inventive step to be present, and confirms that inventive step may also be present if the patent claims “a non-obvious alternative” to solutions known in the prior art.
In Amgen, the Court of Appeal provides additional guidelines in headnotes 17 to 22 on the assessment of reasonable expectation of success, which was crucial in this case. We discuss this in more detail in our report of this decision (see UPC Court of Appeal reverses First Instance decision: Amgen’s patent ruled valid). Notably, the Court of Appeal confirms that the assessment of reasonable expectation of success depends upon the specific facts, assessed on a case-by-case basis. The Court of Appeal also imposes a high burden for the party asserting invalidity of the patent on this basis. This party bears the burden of proof that:
- results were clearly predictable or the skilled person would have reasonably expected success; and
- “sufficiently substantiate[d] uncertainties and/or practical or technical difficulties” brought forward by the patentee would not prevent a skilled person from having a reasonable expectation of success.
Applying this approach to the assessment of inventive step, the Court of Appeal concludes that both Edwards’ patent EP3646825 and Amgen’s patent EP3666797 are valid, upholding the Paris Central Division’s decision to maintain Edwards’ patent in amended form and overturning the decision of the Munich Central Division revoking Amgen’s patent for lack of inventive step.
Comparison with other approaches to the assessment of inventive step
In both decisions, the Court of Appeal refers to the different approaches for assessing inventive step used by the various EPC countries, including the EPO’s problem solution approach and the approaches used by jurisdictions such as the UK and Germany that are often referred to as more holistic.
The approach set by the UPC Court of Appeal possesses some similarities to the EPO’s problem solution approach, but there are marked differences.
The most notable difference is structural. The UPC’s approach set out in Edwards and Amgen first establishes the objective problem and then determines a realistic starting point in the prior art. By contrast, the EPO’s problem solution approach first determines the closest prior art and then establishes the objective technical problem based on isolated, distinguishing features of the claim compared with this closest prior art (which may be a specific embodiment in a document).
This change in the structure of the approach directly affects the establishment of the problem to be solved. In the UPC’s approach, the objective problem is formulated by establishing what the invention contributes to the state of the art, considering the claim in the context of the application and the inventive concept underlying the invention. Thus, the objective problem is derived from the patent itself in isolation of the prior art, thereby aligning the established problem with the teachings of the patent. By contrast, in the EPO’s problem solution approach, the objective technical problem is established in the second step in view of the differences with the closest prior art determined in the first step. This approach formulates a different objective technical problem for each closest prior art document and sometimes results in a problem that is not contemplated in the patent itself. Accordingly, the structural differences in the approaches could, in certain cases, impact on the formulated problem.
A further difference is in the determination of a realistic starting point as opposed to the closest prior art. The Court of Appeal confirms previous UPC case law in this respect (see Inventive step at the UPC, two years on) in that a realistic starting point would have been “of interest” to a skilled person wishing to solve the objective problem. Such a document could disclose “several features similar to those relevant to the invention” and/or address “the same or a similar underlying problem”. The Court of Appeal acknowledges that there may be more than one realistic starting point. The claimed invention must involve an inventive step when starting from each of these starting points.
The final step is the most similar to the EPO’s problem solution approach. Indeed, the Court of Appeal in Amgen specifically emphasises the term “would” in the third stage of the assessment. In line with the could-would approach applied by the EPO, the question is whether the skilled person would have arrived at the claimed solution when wishing to solve the objective problem from a realistic starting point: it is not sufficient that the skilled person simply could have done so. The Court of Appeal expressly states that “the skilled person has no inventive skills and no imagination and requires a pointer or motivation” that “directs it to implement a next step in the direction of the claimed invention”. This aligns with the EPO’s assessment of how a skilled person proceeds from the starting point.
Overall, the UPC’s approach to inventive step set out in Edwards and Amgen is more similar to the holistic approaches used by jurisdictions such as the UK and Germany than to the EPO’s problem solution approach. This may be a welcome development to UPC representatives in these countries.
Final comments
To date, a mix of approaches to the assessment of inventive step have been applied by the First Instance Divisions of the UPC. This has included the EPO’s problem solution approach and the approach applied by the German Federal Court of Justice (see, for example, UPC_CFI_501/2023, UPC_CFI_1/2023, UPC_CFI_315/2023, and UPC_CFI_189/2024 & UPC_CFI_434/2024). Such divergence of approach by the first instance divisions has led to significant uncertainty in, and much discussion of, the approach to inventive step which would ultimately be adopted by the UPC. The clarity the Court of Appeal provides on this issue in these coordinated decisions will be welcomed by patent practitioners and users of the UPC system.
Interestingly, the Court of Appeal comments in both decisions that all of the different approaches to inventive step used by the various EPC countries “are merely guidelines to assist in the establishment of inventive step” and “when properly applied, should and generally do lead to the same conclusion”. This implicitly acknowledges that the use of different assessments of inventive step may not always result in the same outcome, which is an uncomfortable situation for any UPC user to deal with, since EPO proceedings pre- and post-grant will likely continue to employ the problem solution approach.
We wait to see whether the different approaches to the assessment of inventive step will lead to the same outcomes in parallel UPC and EPO proceedings as suggested by the UPC Court of Appeal. Certainly, in the case of Amgen, the UPC Court of Appeal’s application of the holistic approach to inventive step has aligned the outcome with that of the EPO Opposition Division in the parallel opposition proceedings. However, we await the final chapter of this long running saga, with oral proceedings in the pending EPO appeal T 0716/25 scheduled for April 2026.
Related articles
- UPC Court of Appeal reverses First Instance decision: Amgen’s patent ruled valid.
- Inventive step at the UPC, two years on, 12 June 2025.
Case details at a glance
Jurisdiction: UPC
Decision level: Court of Appeal (Luxembourg)
Parties: Amgen Inc v Sanofi-Aventis Deutschland GmbH, Sanofi-Aventis Groupe and Sanofi Winthrop Industrie SA
Citation: UPC_CoA_528/2024
Date: 25 November 2025
Decision: dycip.com/upc-coa-528-2024
Jurisdiction: UPC
Decision level: Court of Appeal (Luxembourg)
Parties: Meril Italy Srl, Meril GmbH, Meril Life Sciences Pvt Ltd, and Edwards Lifesciences Corporation
Citation: UPC_CoA_464/2024, UPC_CoA_457/2024, UPC_CoA_458/2024, UPC_CoA_530/2024, UPC_CoA_532/2024, UPC_CoA_533/2024, UPC_CoA_21/2025, UPC_CoA_27/2025
Date: 25 November 2025
Decision: dycip.com/upc-coa-meril-edwards
