These two decisions clarify the approach to be adopted when assessing inventive step before the UPC. While this approach is inspired by the problem-solution approach (PSA) developed by the case law of the Boards of Appeal (BoA) of the European Patent Office (EPO), it nevertheless has certain specific features. Like the PSA, it consists of several successive steps, which can be summarised as follows:

1. Determination of the “objective problem”

The CoA states that this determination is made on the basis of the contribution of the invention in relation to the state of the art. The CoA adopts a “holistic” approach, considering the claim as a whole in the context of the description and drawings, rather than a feature-by-feature assessment, which seems to differentiate it from the PSA. In practice, the CoA bases its formulation of the objective problem on the objective and technical effects expressly described in the patent.

Similar to the practice before the EPO, the CoA specifies that the objective problem selected must not include any element of the claimed solution to avoid an ex post facto reasoning.

2. Determination of the “realistic starting point”

The CoA uses the term “realistic starting point”, previously introduced in the Amgen v. Sanofi/Regeneron decision of first instance. The objective problem determined in 1. makes it possible to determine this starting point, which must present a teaching of interest to a person skilled in the art wishing to solve the objective problem on the priority date of the patent.

The CoA specifies that several realistic starting points may be selected, thus seeming to express a difference from the concept of the closest prior art used before the EPO. The CoA holds that the realistic starting point may address the same or a similar problem as the underlying problem of the patent in question and/or disclose several features similar to those relevant to the invention. This last condition could reflect the CoA’s desire for greater flexibility in the choice of starting point compared to the EPO’s PSA, which is based primarily on a similarity in terms of technical problem rather than on a number of disclosed features of the claim.

3. Inventive step

Similar to EPO practice, the CoA states that the person skilled in the art has no inventive capacity or imagination. According to the CoA, a solution lacks inventive step if the person skilled in the art, prompted by a pointer or as part of a routine operation, would take an additional step from the realistic starting point to arrive at the invention. However, unlike the EPO’s PSA, the objective problem is not necessarily taken into account in this step.

The CoA adds, in the case of Edwards v. Meril, that it is not necessary to demonstrate that the invention brings an improvement over the prior art, as inventive step can be acknowledged even if the claim constitutes a non-obvious alternative to known solutions. In this case, the CoA upheld the decision of the Central Division in Paris to consider a limited version of the patent valid, noting the lack of motivation for a person skilled in the art to arrive at the invention.

In the Amgen v. Sanofi/Regeneron case, which concerns the field of life sciences, the CoA states that a solution is obvious if a person skilled in the art would have taken the additional step to obtain the invention with a reasonable expectation of success. The CoA takes a pragmatic approach to assessing this expectation of success by evaluating the technical or practical difficulties and the costs associated with implementing this step. The CoA provides detailed reasoning and concludes that the patent as granted is valid. This conclusion is in line with that of the EPO Opposition Division in the parallel opposition proceedings, but contrasts with the decision of the Munich Central Division, which revoked the patent.

The CoA’s approach undeniably has similarities with the EPO’s PSA. The CoA also states that the approaches of national courts, on the one hand, and the PSA, on the other, should not generally lead to different results. At the same time, the CoA’s approach has certain specific features, particularly regarding the “objective problem”, which could result in conclusions that differ from those of the BoA. It will be interesting to see whether the BoA will adjust their practice to align with the UPC practice, as seems to be the message conveyed by the EPO’s Enlarged Board of Appeal in its decision G 1/24 , at least with regard to the interpretation of claims. In particular, the outcome of the appeal proceedings before the EPO concerning the patent EP 3 666 797, which is the subject of the Amgen v. Sanofi/Regeneron decision, will be interesting to examine. In this case, the oral proceedings before the Board of Appeal are scheduled for April 2026, and the EPO will therefore have the final say on validity