EPO v UPC: How the EPO’s and UPC’s approaches to validity give us food for thought when drafting and prosecuting patent applications destined for Europe (Part 3)

The Unified Patents Court (UPC) opened its doors on 1 June 2023 and is already having a significant impact on EPO practice. Less than a year after opening, the UPC Court of Appeal (CoA) rendered its decision in 10x Genomics vs. NanoString (UPC_CoA_335/2023), ruling on how claims should be interpreted under the European Patent Convention (EPC). The CoA’s position was at odds with the EPO’s “classical” approach. This inconsistency played an influential role in prompting the EPO to reconsider how claims should be interpreted. In June 2025, the EPO’s Enlarged Board of Appeal (EBoA) (G1/24) brought the EPO’s approach to claim interpretation closer into line with UPC’s, finding “it a most unattractive proposition that the EPO deliberately adopt a contrary practice to that of the tribunals that are downstream of its patents”.

The EBoA in G1/24 made the EPO’s harmonisation intentions very clear. However, harmonisation is not a new concept. When G5/83[1] issued some 46 years ago, the EBoA held that it was incumbent on the EPO to take the decisions of courts and industrial property offices in the EPC Contracting States into consideration. But G5/83 did little to promote harmonisation. However, since G1/24, there has been a palpable change, and we are beginning to see the EPO make significant strides to align its approach on validity with the UPC’s. Conversely, while the UPC is developing its case law independently of the EPO, many aspects of UPC case law have been shaped by the EPO’s approach to validity.

In this four-part series, we explore the similarities and differences between the EPO’s and UPC’s approach to claim interpretation, added matter, novelty, sufficiency and inventive step. As well as influencing litigation strategies, these similarities and differences give us food for thought when drafting, prosecuting and managing patent portfolios destined for Europe.

In the third part to this series, we take a detailed look at inventive step.

Summary

• The UPC and EPO use different frameworks for assessing inventive step.  The EPO favours a problem and solution approach where inventive step is assessed from the closest piece of prior art.  Although multiple starting points may sometimes be relied upon, this is only if they are “equally valid”. The objective technical problem is formulated in terms of the technical effect arising from the features that distinguishing the claimed subject matter from the closest prior art.

• The UPC adopts a holistic approach. As a first step, the objective technical problem is established in terms of what the invention adds to the state of the art, not by looking at the individual features of the claim, but by considering the inventive concept underlying the invention (the technical teaching). This technical teaching is based on the effects that the skilled person understands are achieved upon reading the application. Accordingly, it is advisable to describe and even demonstrate the technical effects of the invention when drafting patent applications destined for Europe. Although, like the EPO, post-filed data may be relied upon to support inventive step, this is only to demonstrate technical effects that are “encompassed by the technical teaching and embodied by the originally disclosed invention”. While it may not be compulsory to explicitly disclose the technical teachings of the invention, the absence of such a disclosure may leave the advantages of the invention open to interpretation, and this may not be in the patentee’s favour.

• Once the objective technical problem is established, the EPO and UPC use similar criteria to assess whether the claimed solution to such a problem is obvious. For example, both the EPO and UPC consider a reasonable expectation of success to be important to a finding of obviousness. The EPO and UPC are broadly aligned in what constitutes a reasonable expectation of success. However, the UPC appears open to taking commercial considerations like costs into account when determining whether or not a skilled person would be directed towards or away from the claimed invention.

Inventive Step

The EPO generally applies the problem and solution approach when deciding whether an invention involves an inventive step. As confirmed by the Enlarged Board of Appeal in G2/21, this involves identifying the closest prior art; identifying the differences between the invention and the closest prior art; defining the objective technical problem in terms of the technical effect linked to these differences; and examining whether a skilled person would have suggested the claimed technical features to obtain the results achieved by the invention. The latter step is an assessment of obviousness, which is influenced by whether a skilled person would have tried the claimed approach with a reasonable expectation of success.

The UPC, on the other hand, has established its own framework on inventive step[2]. While there are similarities between the UPC’s and the EPO’s approaches, there are also important differences. Firstly, rather than identifying the closest prior art, the first step in the UPC’s framework is to establish the object of the invention. This is confusingly termed the “objective problem” but is not the same as the objective technical problem identified in the EPO’s approach. Whereas the EPO formulates the technical problem based on the technical effect arising solely from the technical difference(s) between the claimed subject matter and closest prior art, the objective problem under the UPC’s approach is established in terms of:

“what the invention adds to the state of the art, not by looking at the individual features of the claim, but by comparing the claim as a whole in context of the description and the drawings, thus also considering the inventive concept underlying the invention (the technical teaching), which must be based on the technical effect(s) that the skilled person on the basis of the application understands is (are) achieved with the claimed invention”.

By looking at the contribution that the invention adds to the state of the art rather than individual features, the UPC’s approach is more “holistic” than the EPO’s. However, in line with the EPO, the technical contribution need not result in an improvement, and an inventive step can reside in the provision of a non-obvious alternative[3].

Under the UPC’s approach, the claimed solution is obvious, when at the relevant date the skilled person, starting from a realistic starting point in the state of 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. The UPC defines the relevant field of technology to include what the EPO terms as “related or neighbouring technical fields”[4]. Thus, the relevant field is the field relevant to the objective problem as well as any field where the same or similar problem arises and of which the skilled person must be expected to be aware.

With regards to the “realistic starting point”, the UPC does not require the closest piece of prior art to be identified. Instead, a starting point is realistic if its teaching would have been of interest to a skilled person. This may be the case if the relevant piece of prior art already discloses several features similar to those relevant to the invention as claimed or addresses the same/similar underlying problem as that of the claimed invention. Note that, with the UPC, the number of features in common appears to be given a similar standing to similarity of purpose. In contrast, at the EPO, it is generally accepted that the similarity of purpose between the prior art and the invention has more weight when assessing whether a document is a suitable starting point than merely the number features in common (e.g., T273/92).

As with the EPO, the skilled person has no inventive skills and no imagination. Accordingly, for a finding of obviousness, the skilled person will require a pointer or motivation that directs them to implement a next step in the direction of the claimed invention. The claimed solution is considered obvious if the skilled person would take the next step prompted by the pointer or as a matter of routine and arrive at the claimed invention. This is generally the case when the results of the next step are clearly predictable, or where there is a reasonable expectation of success.

Whether there is a reasonable expectation of success depends on the circumstances of the case. The more unexplored the field of research, the more difficult it is to make predictions about the prospects of success. As with the EPO, a pointer in the prior art may be indicative of an expectation of success and, the stronger the pointer, the lower the threshold for demonstrating that a reasonable expectation of success exists.

In Sanofi v Amgen, the CoA held that envisaged practical or technical difficulties as well as the costs involved in testing may also withhold the skilled person from taking the next step. Those familiar with EPO practice may have been surprised to see the reference to costs as being a consideration, as commercial considerations are not usually persuasive at the EPO. That said, the general premise that difficulties may dissuade a skilled person from taking a step towards the invention seems to be a valid indicator of an inventive step at both the EPO and UPC.

The CoA also indicated that, where a patentee is able to substantiate the existence of uncertainties or practical/technical difficulties, then it is up to the party alleging obviousness to demonstrate that these uncertainties or difficulties would not have prevented a skilled person from having a reasonable expectation of successfully solving the technical problem. In Sanofi v Amgen, the patentee was able to establish that the skilled person could not reasonably have predicted whether the antibody route that underpinned the claimed invention would lead to therapeutically effective treatment of hypercholesterolemia. This prevented the skilled person from having a reasonable expectation of success. Before the priority date, many pharmaceutical companies were already working on the development of similar antibodies. However, the CoA concluded that this did not necessarily show that a skilled person would have had a reasonable expectation that these antibodies would solve the technical problem. Instead, the skilled person may have viewed these developments as an interesting area to explore with a mere hope to succeed.

The UPC’s framework for assessing inventive step is different from the EPO’s. That said, in Meril v Edwards[5], UPC CoA was keen to highlight that this difference in approach should not lead to a difference in outcome if properly applied. The latter may explain why the EPO has not taken steps to adapt its problem and solution approach. Moreover, there is EPO case law to support the notion that its problem and solution approach need not always be applied, although this is rarely relied upon in practice.

The litmus test as to whether the extent to which the UPC’s and EPO’s approaches lead to divergent outcomes may lie in the pending UPC CoA’s decision on the validity of Sanofi’s patent[6] to the combination of cabazitaxel and prednisone/prednisolone for the treatment of prostate cancer. In earlier EPO opposition proceedings, the opponents argued that a skilled person would have had a reasonable expectation of success because the Phase II clinical trial had been skipped in favour of a Phase III clinical trial that had been approved. The TBA disagreed (T136/24). Instead, they held that the expectation of success had to be assessed based on the available facts, including the nature of the active, how far the clinical testing had progressed, and how much was known about the safety and potential toxicity of the active in the indication in question. The TBA referenced an extensive body of EPO case law that focussed on balancing the positive and negative pointers towards success of the claimed solution.  In relation to the Phase III trial, the TBA held that the positive pointers highlighted by the opponent would not have caused the skilled person to have a reasonable expectation of success in terms of the trial’s primary end point, namely an increase in overall survival rate relative to the comparative arm of the trial

When validity was challenged at the UPC, the Mannheim LD disagreed with the TBA’s conclusions. Instead, on essentially the same evidence, the LD found that a skilled person would have had a reasonable expectation of success of solving the problem with the claimed combination in the light of the prior art.  The LD endorsed the approach used by the TBA in weighing positive and negative pointers.  However, they distinguished the concept of a reasonable expectation in the context of inventive step from the question of whether clinical trial would meet its primary endpoint[7]. In this regard, they saw the technical problem of the patent as providing a therapeutic option for treating patients with castration-resistant metastatic prostate cancer who had previously been treated with a docetaxel-based regimen and whose cancer had progressed during or after that treatment. This differed from the primary endpoint of the clinical trial, which was to increase overall survival in patients in the cabazitaxel group compared to the control group. More importantly, LD seemed to place greater emphasis on commercial considerations, noting, for example, that “the sponsor of this trial had not considered it disappointing at the start and at no time until the priority date[8]”.  This, together with the duration of the trial, seemed to convince the LD that there was a reasonable expectation of success insofar as inventive step was concerned.  Thus, the LD contradicted the findings of the TBA and concluded that the patent was obvious.

Interestingly, the LD commented on their divergence with the TBA, but held that the fact that validity had been considered by the EPO did not relieve the UPC of its duty as an independent judicial body to interpret and apply the EPC independently. As such, national and EPO decisions were seen as “elements to be considered by UPC panels” and, while not binding, they “may have a persuasive effect”. It is interesting to compare this language with the EBoA’s view in G1/24 that “it [was] a most unattractive proposition that the EPO deliberately adopt a contrary practice [to the UPC]”. We are waiting to see if the UPC CoA will uphold the LD’s judgment.

Notwithstanding the differences between the UPC’s and EPO’s approach to inventive step, there are nevertheless strong echoes of EPO jurisprudence in the UPC’s framework. In this regard, both the EPO and UPC look at what the skilled person would (and not could) have done. Both the EPO and UPC also consider pointers in the prior art, as well as whether a skilled person would have had a reasonable expectation that the claimed invention would have solved the objective technical problem.  While we have seen the Manheim LD arrive at a different conclusion on “reasonable expectation of success” in relation to the cabazitaxel case, LD adopted a very similar approach in weighing up the positive and negative pointers when assessing whether the next step would have been obvious to a skilled person.

Furthermore, although the “problem” is established differently at the EPO and UPC, both rely on technical effects over the prior art. The UPC CoA has set out that these technical effect(s) are those that the skilled person understands are achieved on the basis of the application. Similarly, at the EPO, G2/21 dictates that a patentee may rely on a technical effect if the skilled person would have derived the effect as being encompassed by the technical teaching and embodied by the originally disclosed invention. Assessing whether this requirement is met is highly fact dependent and different TBAs have approached this assessment in different ways. So far, however, there seems to be some alignment on the broad principles relied upon when assessing whether a technical effect can be relied upon to support inventiveness.

In Sumi Agro v Syngenta [UPC_CoA_523/2024], Sumi Agro argued that Syngenta had not demonstrated a technical effect across the scope of the claim because the examples only demonstrated a technical effect (improved stability) in relation to a subset (63.5 to 68.6%) of the claimed concentration (1 to 95%) of fatty acid. In Sumi Agro’s view, this meant that the technical effect had not be shown to exist across the whole range and, consequently, could not be relied upon by the patentee to demonstrate an inventive step. The CoA, however, held that Sumi Agro had not discharged its burden of proof in this respect. The CoA noted that the specification disclosed explicitly described improved stability in general terms in the specification. Thus, when the examples were viewed in the context of the entire patent specification, there was no reason to believe that the technical effect could not achieved over the whole claimed range. Using the same language as the EBoA in G2/21, the CoA held that this technical effect was “encompassed by the technical teaching and embodied by the originally disclosed invention”.

Neither the UPC’s or EPO’s approach requires the technical effect to be explicitly disclosed. However, it is likely that Syngenta benefited from the general statement of the technical effect in the original application. Together with the examples, this statement was sufficient to allow the technical effect to be relied upon for inventive step. It is impossible to say for certain whether the conclusion would have been different had a general verbal disclosure of the technical effect been absent from the original text. However, risks may arise if no technical effect is explicitly mentioned. Although it may be possible to construe a technical effect from the specification in the absence of one being explicitly disclosed, the technical effect construed may not be as the patentee intended, particularly if there is little discussion or evidence of the benefits of the invention in the original specification. This may put the patentee in a worse position with respect to inventive step, especially since the patentee’s ability to rely on post-filed data may be dependent on whether the effect was “encompassed by the technical teaching and embodied by the originally disclosed invention” as per G2/21 above.

Table 5: Summary of EPO’s and UPC’s approaches to inventive step

[1] See Reasons 6 of G5/83

[2] Sanofi v Amgen [UPC_CoA_528/2024] and Meril v Edwards [UPC_CoA_27/2025]

[3] Meril v Edwards [UPC_CoA_27/2025]

[4] 8.2. Related technical fields (neighbouring fields)

[5] Headnote, reason 5

[6] EP 2493466 B

[7] Page 32, first paragraph of decision

[8] Page 41, penultimate paragraph of the decision

This article was prepared by Hsu Min Chung, Thomas Compton & Robert Scanes.

View Part 1 here

View Part 2 here

Part 4 coming soon.