UPC Central Division Discards EPO Approach to Obviousness in First Decision on the Merits

On July 16, the Unified Patent Court’s (UPC) Central Division (CD) in Munich revoked Amgen’s patent covering use of PCSK9-binding antibodies for treatment of cardiovascular disease. In its written decision the CD discussed the correct approach to claim interpretation and the assessment of inventive step.

Amgen’s patent, the subject of this decision, is in the same family as the patents that are the subject of the U.S. Supreme Court’s decision in May 2023: here, the first instance division of the UPC revoked the patent on the basis it lacks inventive step. The decision comes out of the first case filed at the UPC and has a complex procedural history not addressed in this article.

Claim Interpretation

On interpretation, the CD considered two UPC Court of Appeal decisions (from appeals in preliminary injunction proceedings) from which it derived the following principles, which it held applied equally to validity as infringement:

• Claims of a patent are the decisive basis for determining the patent’s scope of protection;
• Interpretation must be performed from “the point of view of” the skilled person so as to “determine the technical meaning of the terms used with the aid of the description and drawings”. The skilled person does not “apply a philological understanding”. Put another way, the interpretive exercise is contextual, context being provided by the skilled person’s common general knowledge and the remainder of the patent specification.
• Different features of the claim must be interpreted in the context of the whole claim so as to “deduce [the] technical function these features actually have individually and as a whole”.
• Patent may use its “own lexicon”: the meaning given to terms in the patent specification “is authoritative” even if the “terms used … deviate from general usage…”.
• The interpretive exercise should achieve a result combining adequate protection for the proprietor and legal certainty for third parties.

Aspects of this decision recited the principles already set out by the Court of Appeal (in particular in 10x Genomics v NanoString UPC_CoA_335/2023). However, the holding that the patent may serve as its own lexicon is new in the UPC (though a common concept according to German practice) and of wider significance. Presuming that this approach is followed in future cases, where a patent’s description provides definitions or clear teaching as to what is meant, there may be less room for parties to argue about the correct interpretation.

On the other hand, where there is no such definition or clear teaching, the UPC’s case-law continues to affirm the need to understand the claim language technically and in context of the remainder of the specification. This may encourage parties to submit evidence from experts in an attempt to demonstrate the veracity of their position on this technical meaning, or to influence the judicial panel who may otherwise be inclined to adhere to the technical judge’s view on the common general knowledge context.

Here, the claim was formulated in the “EPC 2000” form for therapeutic products – “product X for use in the treatment of Y” – which replaced the previously-used “Swiss” form. The CD’s holding that the “product specified in medical use claim must be objectively suitable for the claimed use … In this sense, it must be therapeutically effective” may also set the standard for future EPC2000 claims.

Inventive Step

In the EPO, assessment of inventive step involves identification of the “closest prior art” and is the first stage of the well-known “problem-solution approach” (PSA). Various European national courts in countries now participating in the UPC have adopted the PSA into their case law, but not Germany. Presiding Judge Ulrike Vo? for this decision is a German legal judge. Judge Kupecz, the other legal judge, is from the Netherlands, where the PSA is often used but not compulsory.

The CD did not set out any alternative methodology for assessing inventive step, but it explicitly disapproved of the “closest prior art” approach and thereby implicitly rejected the PSA too. This reflects earlier decisions (e.g. of the Dusseldorf Local Division (‘LD’) in Franz Kaldewei v Bette UPC_CFI_7/2023 and the Court of Appeal in 10x Genomics v NanoString) but differs from the July 4 decision of the Paris LD which used a PSA-like approach in Dexcom v Abbott (UPC_CFI_230/2023). This has important ramifications for assessment of inventive step in the UPC compared to the EPO, presuming it is followed.

By discarding the “closest prior art” concept, the CD brushed aside a key and sometimes unsatisfactory aspect of EPO oppositions: arguments focussing on which prior art is the “closest” taking a disproportionally prominent role (even though more recent EPO case law has found lack of inventive step over prior art even where a more promising starting point is available). Here, very similar to German practice, the CD held that the invention may be rendered obvious by any “realistic starting point” – being a piece of prior art with “teaching [that] would have been of interested to the skilled person … seeking to develop a similar product or method … [who] thus has a similar underlying problem as the claimed invention”. If any “realistic starting point” renders the claim obvious, the patent is invalid, although justification bay be needed as to why a particular citation is a realistic starting point.

This “relevant starting point” approach was key to the outcome: the CD’s decision records in 8.12 that, while the parties had disagreed on the “closest” prior art, “it is not actually in dispute between that parties that [prior art citation] Lagace at least is a realistic starting point… This is also the view of the [CD].” Later, at 8.27, the CD noted that “Defendant has not brought forward any concrete arguments as to why Lagace would not be a realistic starting point. It does not follow from the assertion that [alternative prior art citation] Graham is “more realistic” or “closer” that Lagace is not a realistic starting point….”

According to the CD, a claim is obvious if “the skilled person would be motivated … to consider the claimed solution and implement it as a next step … in developing the prior art”. Expected difficulties in taking such steps are relevant. It will be interesting to see how the “next step” concept is applied in future cases (assuming this decision is followed): if this excludes a finding of obviousness where multiple steps would be required to achieve the invention, this may lead to patentees putting forward such “multiple-step” analyses in their argumentation as is sometimes seen in other jurisdictions, such as the UK.

The CD also noted that a technical effect or advantage achieved by the claimed subject matter compared to the prior art may be an indication for an inventive step. A feature that is selected in an arbitrary way out of several possibilities, however, cannot generally contribute to inventive step. Hindsight is of course to be avoided in the inventive step assessment but it “may be allowed to combine prior art” depending on the facts and circumstances.

Discussion

The treatment of claim interpretation in this decision is consistent with prior decisions from the UPC. Additional considerations arising in the present case may have influence on future cases (and potentially on patent prosecution, in the long term) by explicitly recognizing the possibility of a patent acting as its own dictionary even if such self-definition is contrary to established technical understanding. This is interesting also in view of the recent EPO referral to the Enlarged Board of Appeal (G1/24) relating to extent of the requirement to consult the specification in order to interpret claims in particular if the claim language is clear.

Of greater significance is the explicit rejection of the “one closest prior art” approach to assessing inventive step. The CD’s approach is less formulaic than the EPO’s problem solution approach and more holistic – the claim will lack inventive step if it is obvious from any realistic starting point. It would be premature to suggest that this is likely to lead to a higher incidence of patent revocation compared to EPO opposition, but it will require parties to review how their arguments are advanced in that portion of cases where several prior art citations would otherwise compete for being the “closest”. The authors note that the July 20 decision of the CD in Paris in Meril Italy v Edwards commented that the PSA “is not explicitly provided in the EPC and, therefore, does not appear to be mandatory”.