Amgen v Sanofi and others (UPC_CoA_528/2024, UPC_CoA_529/2024)
Decision date:
25 November 2025
Osborne Clarke summary
- This decision concerned an appeal brought by Amgen relating to a first instance decision by the Munich CD, which found Amgen's patent covering PCSK9 (proprotein convertase subtilisin kexin type 9) inhibitor antibodies invalid for lack of inventive step.
- The Court of Appeal overturned the Munich CD's decision, instead finding Amgen's patent valid
- Amgen and Sanofi/Regeneron both market cholesterol-lowering PCSK9 inhibitor antibody drugs, which are sold under the trade names Repatha® and Praluent®, respectively. This decision forms part of a wider dispute between the parties concerning PCSK9 inhibitors that has been ongoing for the past decade.
- Amgen's patent claims a monoclonal antibody or antigen-binding fragment for use in treating or preventing hypercholesterolaemia or atherosclerotic disease related to elevated serum cholesterol levels, or for reducing the risk of recurrent cardiovascular events, wherein the antibody binds to the catalytic domain of PCSK9 and prevents or reduces PCSK9 binding to the low density lipoprotein receptor (LDLR).
- The Court of Appeal held that when claims are drafted in a "medical use-format", it is an inherent feature that the claimed product must be “therapeutically effective” in a meaningful way. This requires the claimed treatment to cause a "noticeable improvement of the medical condition of the patient suffering from the disease mentioned in the claim".
- The inventive step analysis formed the heart of this appeal. The Court of Appeal set out comprehensive principles for assessing inventive step at the UPC, aligning with the Court of Appeal's decision in Meril v Edwards that was handed down on the same day as this decision.
- Under the holistic approach endorsed by the UPC, the first step is to establish the objective problem from the perspective of the skilled person by determining what the invention adds to the state of the art. This assessment should be made in view of the claim(s) as a whole, in light of the description and the inventive concept. In order to avoid hindsight, the objective problem should not contain pointers to the claimed solution.
- The second step is to identify a realistic starting point in the state of the art in the relevant field of technology. The court reaffirmed that a starting point will be realistic if its teaching would have been of interest to a skilled person who, at the relevant date, wished to solve the objective problem. There can be more than one realistic starting point, but the claimed invention must be inventive starting from each of them.
- Having identified the objective problem, the court will then consider whether the claimed solution is obvious by assessing whether the skilled person would (and not only "could") have arrived at the claimed solution. In the context of a medical use claims, the claimed solution is obvious if the skilled person would have taken the next step in expectation of finding the envisaged solution. This is generally the case when results were clearly predictable or where there was a reasonable expectation of success. The burden of proof lies with the party alleging invalidity of the patent.
- A reasonable expectation of success implies the ability of the skilled person to predict rationally, on the basis of scientific appraisal of the known facts before a research project was started, the successful conclusion of that project within acceptable time limits. Furthermore, the assessment should take into account any practical or technical difficulties as well as the costs involved in testing whether the desired result would be obtained and whether these considerations may have stopped the skilled person from taking that step. On the other hand, the stronger the pointer towards the claimed solution, the lower the threshold for a reasonable expectation of success. Moreover, the fact that other people or teams were working contemporaneously on the same project does not necessarily imply that there was a reasonable expectation of success. It may indicate that it was an interesting are to explore with a "mere hope to succeed".
- Much of the inventive step analysis centred around Lagace, a prior art citation describing experiments investigating the effects of extracellular PCSK9 on cultured liver cells and fibroblast cells. Lagace also described a parabiosis experiment in which the circulatory system of a genetically engineered mouse overexpressing PCSK9 was connected to the circulatory system of a normal mouse, demonstrating that extracellular PCSK9 from the genetically engineered mouse caused removal of LDLR in the normal mouse and increased cholesterol levels in vivo.
- However, this experiment was carried out with PCSK9 at supraphysiological levels. The court concluded that at the priority date the skilled person, starting from Lagace, had a strong incentive to block PCSK9 activity to reduce LDL levels in order to treat hypercholesterolaemia and similar diseases, and that Lagace taught the skilled person that an extracellular pathway of PCSK9 exists.
- Crucially, although Lagace had provided evidence that extracellular PCSK9 could reduce LDLR levels in cultured cells, it also pointed out that the relative contribution of PCSK9's intracellular and extracellular pathways was uncertain under normal and pathological conditions. At the priority date, PCSK9 was considered to function by both intracellular and extracellular pathways, and so the skilled person needed to know whether the extracellular pathway's contribution would be sufficient to result in therapeutically effective treatment. The court found this uncertainty critical. Other publications raised doubts as to whether relevant PCSK9-LDLR binding could occur extracellularly under physiological conditions. For example, Cunningham and Fisher separately reported that PCSK9's binding affinity for LDLR was much higher at acidic pH (inside cells) than neutral pH (in plasma), with LDL particles further reducing binding affinity.
- It was not until April 2009, after the priority date, that another citation provided "initial evidence that PCSK9 functions predominantly as a secreted factor". The Court of Appeal held that suggestions in Lagace about antibodies as potential therapeutic approaches would be understood by the skilled person as conditional upon resolving the existing uncertainties about whether the extracellular pathway was sufficiently relevant for an antibody approach to succeed.
- At the priority date, scientific research in the field of PCSK9's mechanism of action had not reached the stage where the skilled person could reasonably predict that an antibody blocking PCSK9-LDLR interaction would be therapeutically effective. Accordingly, the skilled person could not reasonably predict whether the antibody route would lead to therapeutically effective treatment, preventing a reasonable expectation of success.
- Sanofi/Regeneron were also unable to overturn the Munich CD's findings on sufficiency. They argued that the patent did not disclose the invention sufficiently clearly and completely because the skilled person would be unable to obtain antibodies binding to the catalytic domain without undue burden. Since the catalytic domain cannot be expressed and secreted in isolation, Sanofi/Regeneron contended that the skilled person would need to perform X-ray crystallography to determine whether an antibody bound the catalytic domain, representing an undue burden.
- However, the Court of Appeal found that the patent disclosed several methods to identify whether an antibody binds to the catalytic domain, including X-ray crystallography, domain binding assay, scanning mutagenesis and competition assay, with further methods also known at the priority date. The mere fact that a method (in this case X-ray crystallography) is laborious, time-consuming and challenging does not automatically mean that it constitutes an undue burden, especially when it is the generally accepted standard for obtaining the most accurate results. Moreover, a reasonable amount of trial and error does not prevent an invention from being enabled.
- The Court of Appeal also emphasised that it is not required that each and every conceivable embodiment within functional definition(s) of the claim are enabled. Sanofi/Regeneron had not shown that the skilled person would be unable to obtain further antibodies with the claimed functional properties in a reliable manner with reasonable trial and error and without undue burden.
- The Court of Appeal also rejected Sanofi/Regeneron's added matter arguments, finding clear pointers throughout the application. It noted that literal support is not necessary and that it is sufficient if the skilled person can derive the subject matter from the application as a whole. The court concluded that the preferred features taken together, as well as various links between features in the application, meant that the combination of claimed features was clearly and unambiguously derivable from the application.
- The first instance decision was set aside and Sanofi/Regeneron were ordered to pay Amgen's agreed costs of €1,375,000 for each first instance proceeding (the revocation proceedings and counterclaim for revocation). As there was no agreement on the costs of the appeal, the Court of Appeal ordered Sanofi/Regeneron to bear Amgen's costs of the appeal.
Issue
Curious about how UPC decisions might impact your business? Have questions about the UPC?
Reach out to our patents team for expert guidance and support.