2024 Week 29
This week saw the first UPC decision on the merits issued in the long-running patent battle between Amgen and Sanofi/Regeneron. The decision is succinct and well-reasoned, featuring a high level of technical detail. As expected, the judges at the UPC are committed to resolving complex patent disputes through careful consideration of the technical facts, similar to what we are accustomed to in EPO opposition proceedings.
What caught our eye was how the decision steers clear of the EPO’s problem and solution approach to the assessment of inventive step, instead adopting an approach that shares certain elements with the German approach to evaluating inventive step. We are now seeing the first few validity decisions from the UPC on the merits and the snapshot view is that there is not yet clear certainty regarding the UPC’s approach to inventive step. We are actively monitoring these issues so that we can combine our extensive EPO opposition experience with detailed understanding of the developing practice of the UPC.
Background
The patent under decision was Amgen’s EP 3666797 B1, relating to anti-PCSK9 antibodies for use in the treatment of diseases related to high cholesterol. “Bad cholesterol” (“LDL”) is removed from the blood by the LDL receptor (LDLR) in the liver. PCSK9 binds LDLR and causes its degradation, preventing removal of “bad cholesterol” from the blood. By targeting PCSK9, the claimed antibodies regulate levels of LDLR and, therefore, “bad cholesterol”.
Amgen’s patent covers its product Repatha®. Sanofi and Regeneron also have an antibody against PCSK9 on the market, under the trade name Praluent®. Reflecting the importance of these patents, Sanofi and Amgen both filed actions (for revocation and infringement, respectively) on the first day the UPC ‘opened its doors’. (This was not long after the invalidation of a related US patent in a Supreme Court decision we reported last year.)
Inventive step before the UPC to date
For pretty much every case where validity is raised, and where the parties need a court to adjudicate on their dispute, inventive step is a key issue. So, as we commented recently, the European IP profession is watching the UPC closely to see how it will consider inventive step. European patent attorneys are well-versed in the EPO’s problem and solution approach, along with the approach taken by their own national courts. Some national European approaches, including those taken in the UK and Germany, are rather different to the EPO’s.
There is no requirement on the UPC to follow the EPO’s inventive step approach, but we have seen flashes of it. Different UPC divisions are taking different approaches. For example, the problem and solution approach was clearly used by the Paris Local Division in DexCom v Abbott, but not explicitly followed by the Düsseldorf Local Division in Franz Kaldewei v Bette. The UPC Court of Appeal has not yet given a clear direction. For example, in Nanostring Technologies v 10x Genomics (which were summary proceedings for a preliminary injunction) it neither endorsed nor criticised the problem and solution approach and did not give its own methodology.
The decision on EP 3666797 B1 marks the first decision on validity by a UPC Central Division and very clearly does not follow the “problem and solution” approach. As we explain below, it adopts elements seen in the assessment of inventive step by the German national courts. This highlights that the UPC could still diverge from the approach most familiar to UPC representatives across Europe.
Where to begin?
The decision states that the first step when assessing inventive step is to identify “a realistic starting point”, choosing prior art document Lagace as such a starting point. The author of this paper was interested in finding out more about the mechanism by which PCSK9 reduces the number of LDLRs and neither party disputed that it was a realistic starting point. The decision then asserts that, once a realistic starting point has been identified, there is no need to examine in detail whether another starting point is “more promising” or “closer”. Even if there were a “closer” starting point (as argued by Amgen), it would not mean that Lagace were not a “realistic” starting point. All the division needs to do is assess whether the claimed subject matter is “inventive over any realistic starting point”.
Not having to select a closest prior art will sound familiar to those acquainted with the German approach to assessing inventive step. According to the German Federal Court of Justice, whether a specific state of the art serves as an appropriate starting point for the skilled person does not hinge on whether the document is classified as the closest state of the art. This classification, from an ex post perspective, is neither sufficient nor necessary.
At first glance, this seems to be a significant contrast to the EPO’s practice. In our experience, it is quite challenging to convince the EPO that there are two (or more) equally appropriate starting points. However, more recent EPO case law clarifies that the term closest prior art is to be understood in the sense of an appropriate starting point. Accordingly, if there are several workable routes to arriving at the invention (for example, starting from different starting points), inventive step must be assessed relative to all of these possible routes before an inventive step can be acknowledged. If the invention is obvious in respect of just one of these routes, then an inventive step is lacking. Recognising that an invention may be arrived at in an obvious way from a piece of prior art that is originally assumed to be less relevant, but not from the piece of prior art that is assumed to be most relevant, EPO case law states that it may be necessary to repeat the “problem and solution” approach for any prior art that is a suitable starting point. In such cases, the competent bodies at the EPO have sometimes even refrained from repeating the “problem and solution” approach for another route if they already found the invention to be obvious in respect of just one of these routes, similar to the approach taken by the UPC Central Division in this instance.
Amgen was pushing for Graham to be closest prior art because Graham chose an antisense oligonucleotide (ASO) for targeting PCSK9. Lagace, however, refers to the development of antibodies to block PCSK9’s interaction with the LDLR inhibitors as an additional approach to be explored if PCSK9 functions as a secreted receptor. EPO practitioners will easily see how choosing either one of these two documents leads to the formulation of objective technical problems that differ from each other and can result in different outcomes of the obviousness assessment.
So what’s the problem?
The UPC Central Division considered the “underlying problem” of the claimed subject-matter based on the patent description as a whole: the provision of a treatment for hypercholesterolemia (or other conditions related to elevated serum cholesterol levels) using PCSK9 as a target. Again, such a holistic approach to the task that the skilled person was facing is reminiscent of the German way of assessing inventive step.
This contrasts to the EPO’s problem and solution approach, in which an objective technical problem is meticulously formulated based on the differences between the chosen closest prior art and the claimed subject-matter. However, it's worth noting that in this case, the technical problem posed by the Court appears very similar to what one would derive using the problem-and-solution approach starting from Lagace.
Obviously ...
The decision then assesses whether the claimed solution to this underlying problem was obvious, in a robust and (to our eye) rather patentee-unfriendly way. The UPC Central Division considered that a skilled person aiming to provide the treatment would, as a next step, have pursued the development of anti-PCSK9 antibodies, as suggested in Lagace. Unsurprisingly to those familiar with antibody patents, methods for generating and selection such antibodies were dismissed as “routine” and as not requiring any inventive skill. This led to the conclusion that claim 1 of the patent as granted was invalid due to a lack of inventive step and, ultimately, to the revocation of the patent.
Next steps
Opposition proceedings against the same patent continue before the EPO, even after the revocation of the patent in all 17 states covered by the UPC. The EPO opposition division have not yet opined on the ongoing opposition against the same patent. We look forward to seeing if their approach to inventive step is any more patentee-friendly.
Of particular interest will be whether the decision of the UPC has any influence on the EPO opposition division’s reasoning.
As we reported in relation to DexCom v Abbott, the UPC has reached a different conclusion on inventive step compared with an EPO opposition division before, even when deciding this issue with the same parties and relying on the same prior art.
It may not be surprising that different parts of the UPC, operating with judges having different backgrounds, are assessing inventive step using their national experience. It is highly likely that this decision will be appealed and the essential role of the UPC Court of Appeal will be to gradually impose a uniform approach to inventive step. We are keeping a close eye on this and will report again.
Lucy is an associate patent attorney in our Life Sciences Patent Team. She has a B.A. degree in Natural Sciences from the University of Cambridge where she specialised in Pathology. Her undergraduate research project focused on the down regulation of antigen presentation during human cytomegalovirus infection and how this impacted T cell responses to infected cells.
Email: lucy.white@mewburn.com
