More Lessons on Validity before the UPC – Part I

It has been a while since I last commented on the case law of the Unified Patent Court (UPC) on validity. To be precise, it was on 22 July 2024 in a blog titled “First Lessons on Validity before the UPC”. Since many more decisions on validity have been issued in the meantime, including the first decisions by the Court of Appeal (CoA)  in main proceedings, it is interesting to see what is to be learned since then. To facilitate a comparison with my 2024 blog, I will again structure this contribution in lessons, using similar headings where appropriate.

Lesson 1: Claim Interpretation – The Description Matters Always, or Does It Not?

a)         Are Compound-for-use Claims Special?

The UPC has meanwhile settled on the language that when interpreting a patent claim, the person does not apply a philological understanding but determines the technical meaning of the terms used with the aid of the description and the drawings. Moreover, it seems to be common ground that the patent description may represent a patent’s own lexicon.

So far, so good, but does the same still hold when the patent claim is a use claim or compound-for-use claim? Having read the decision by the Central Division Munich in the Amgen v. Sanofi case, I had no doubts that it does.

The Central Division regarded “product X for use in the treatment of Y” claims as purpose-limited product claims. In the words of the Central Division (reason 6.10)

Its medical use (the “purpose”) is specified in F.2, i.e. treating or preventing hypercholesterolemia or an atherosclerotic disease related to elevated serum cholesterol levels or in reducing the risk of a recurrent cardiovascular event related to elevated serum cholesterol levels. The product specified in a medical use claim must be objectively suitable for the claimed use; it must be able to be used for the treatment, prevention or reduction as specified in the claim. In this sense, it must be therapeutically effective.

This made sense to me and corresponds to long-standing EPO and (at least) German practice, according to which use claims should be interpreted as including the technical effect as a functional feature, and are accordingly not open to objection under Article 54(1) EPC provided that such technical feature has not previously been made available to the public (G 2/88 Headnote 3, G 5/83 Reason 21). Even more importantly, this is also in conformity with the wording and spirit of Art. 54(4) and (5) EPC2000.

(4) Paragraphs 2 and 3 shall not exclude the patentability of any substance or composition, comprised in the state of the art, for use in a method referred to in Article 53(c), provided that its use for any such method is not comprised in the state of the art.

(5) Paragraphs 2 and 3 shall also not exclude the patentability of any substance or composition referred to in paragraph 4 for any specific use in a method referred to in Article 53(c), provided that such use is not comprised in the state of the art.

Thus, and at the risk of a slight oversimplification, I think one can generally read and interpret a claim of the format “product X for use in treatment of Y” as “product X for the successful use in treatment of Y”.

This still leaves it open what (a) “treatment” exactly means and (b) what “successful” means. For that, I had always thought that the skilled person would still have to look into the description.

However, it now seems that the Court of Appeal in its appeal decision of 25 November 2025 in the Amgen case adopted a much more apodictic approach to claim interpretation: Not only did the CoA conclude that the claimed product must be therapeutically effective, it went further to stipulate, quasi a priori, that it follows from this that the treatment must cause a “noticeable improvement of the medical condition of the patient suffering from the disease”, i.e. the treatment must be “meaningful”. Headnote 3 might even suggest that the CoA does not mind much what is stated in the description, because it ties its interpretation of “effective” directly to the claim format:

2. When the claims are drafted in ‘medical use-format’, it is an inherent claim feature that the claimed product must be objectively suitable for the claimed use, i.e. be therapeutically effective. This requires that the claimed treatment causes a noticeable improvement of the medical condition of the patient suffering from the disease mentioned in the claim, i.e. the treatment must be meaningful.

3. The fact that the skilled person does not derive any minimum required effect from the claim or the description does not lead to another conclusion, since the feature of therapeutic effect does not follow from the claim language that is to be interpreted, but from the use of the medical use claim format.

I must say that I was and still am quite surprised by this interpretation. The CoA has reproduced claim 1 in its decision, which reads:

A monoclonal antibody or an antigen-binding fragment thereof for use in treating or preventing hypercholesterolemia or an atherosclerotic disease related to elevated serum cholesterol levels; or for use in reducing the risk of a recurrent cardiovascular event related to elevated serum cholesterol levels; wherein the monoclonal antibody or the antigen-binding fragment thereof binds to the catalytic domain of a PCSK9 protein of the amino acid sequence of SEQ ID NO: 1, and prevents or reduces the binding of PCSK9 to LDLR.

This claim clearly addresses (a) several diseases (hypercholesterolemia, an atherosclerotic disease related to elevated serum cholesterol levels, recurrent cardiovascular events related to elevated serum cholesterol levels) and (b) three types of uses, i.e. “treating or prevention” or “reducing the risk” (of the recurrent CV events). Unfortunately, this does not seem to me to be reflected in the CoA’s decision, which quite broadly just speaks about “a noticeable improvement of the medical condition (sic!) of the patient suffering (sic!) from the disease (sic!) mentioned in the claim” and a “meaningful treatment”.

This lack of differentiation has consequences, as we will immediately see below.

It can be derived from claim 1 as recited above and seems to have been uncontroversial between the parties that the mechanism of action of the claimed PCSK9 antibodies was to bind to PCSK9 protein, thus preventing or reducing the binding of PCSK9 protein to LDLR (Low Density Lipid Receptor). This reduced binding of PCSK9 protein in turn allows more LDL cholesterol to bind to LDLR and thus results in a lowering of serum cholesterol levels. This raises the question of how much antibody binding and how much lowering of serum cholesterol levels is necessary to achieve any of the effects recited in the claim.

Unfortunately, the judgment does not even begin to discuss this question in the part dealing with claim construction. The CoA simply argues in paragraph 47 of its decision that not any lowering of cholesterol levels suffices; the lowering must “bring about a therapeutic effect” and “cause a noticeable improvement of the medical condition of the patient suffering from the disease (…)”. But claim 1 is clearly not limited to therapy, it also includes “prevention” and “reduction of risk”; it is also not limited to (human) “patients”; and it refers to at least three different diseases.

To me at least, it stands to reason that each of these conditions and types of uses will most probably require a different lowering of serum cholesterol levels, or no such lowering at all, for example if the PCSK9 inhibitor is taken by a patient with normal serum cholesterol levels in a preventive setting, i.e. to inhibit serum cholesterol levels from rising.

The CoA and the CD Munich seem to have been in agreement that the skilled person could not derive any minimum required effect from the claim or the description (CoA, paragraph 48). However, what can be found in the description is at least an indication of the amount of binding of PCSK9 antibody to PCSK9 that the applicant apparently envisaged. The following section of paragraph [0130], which I have slightly edited for better comprehension (“antigen-binding protein” = anti-PCSK9 antibody, “ligand” = PCSK9, “its binding partner” = LDLR), states that:

In assessing the binding and/or specificity of an anti-PCSK9 antibody, the antibody can substantially inhibit binding of PCSK9 to LDLR when an excess of antibody reduces the quantity of LDLR bound to PCSK9 by at least about 1-20, 20-30%, 30-40%, 40-50%, 50-60%, 60-70%, 70-80%, 80-85%, 85-90%, 90-95%, 95-97%, 97-98%, 98-99% or more (as measured in an in vitro competitive binding assay). In some instances, in the case of anti-PCSK9 antibody, such a neutralizing molecule can diminish the ability of PCSK9 to bind the LDLR. In some instances, the neutralizing ability is characterized and/or described via a competition assay. In some instances, the neutralizing ability is described in terms of an IC50 or EC50 value. Antigen Binding Proteins (ABPs) 27B2, 13H1, 13B5 and 3C4 are nonneutralizing ABPs, 3B6, 9C9 and 31A4 are weak neutralizers, and the remaining ABPs in Table 2 are strong neutralizers. The antibodies or antigen binding proteins suitable for the use according to the invention neutralize by binding to PCSK9 and preventing PCSK9 from binding to LDLR (or reducing the ability of PCSK9 to bind to LDLR).

What I take from this paragraph and particularly its last sentence, is that the patent considers that the ability of PCSK9 to bind to LDLR is supposed to be either “prevented” or “reduced”; i.e. some binding of PCSK9 to LDLR may still occur. Even a reduction of the binding (more precisely, the quantity of LDLR bound to PCSK9) by (only) about 1-20% apparently means a “substantial inhibition” of the binding of PCSK9 to LDLR, according to the first sentence of this paragraph.

I don’t know whether this was argued by the parties before the CoA, but it seems to me that a skilled person would not have ignored this passage of the description. Moreover, had the skilled person considered the full breadth of the claimed indications and types of uses, they would – at least in my humble opinion – most probably have arrived at a (much) broader claim construction than the CoA did.

Thus, this part of the CoA decision has left me with the question whether this was an inadvertent oversight by the court or whether the CoA really and deliberately wanted to focus on just one type of treatment (therapeutic) of one disease (hypercholesterolemia, arguably), and if so, why?

I guess that we will have to wait for the next couple of decisions by the UPC to see from where the wind is blowing…

b)         Claim interpretation based on dependent claims

Conversely, the CoA decision provides for desirable clarity on another point of claim interpretation, which was contested between the parties and where the CD Munich had come to a different conclusion.

The CD had held that the skilled person would understand from the patent that the claimed treatment is not limited to a particular lowering of cholesterol levels as long as there is some (measurable) reduction of cholesterol levels in vivo and provided the therapy is safe. The CD considered that this interpretation is also supported by claims 6 and 7 of the patent, which encompass the administration of the claimed antibodies together with at least one other cholesterol-lowering agent, notably statins. From these claims, the CD Munich concluded that also a (very) small cholesterol-lowering effect caused by the claimed antibodies can be “therapeutically effective” in the sense of the claimed treatments, because at least in this case, the statins on their own might cause the desired effect.

The CoA did not agree with this argumentation. Consistent with the case law of the German FCJ (Wärmetauscher, X ZR 114/13 mn. 15), the CoA first noted that the question of whether conclusions can be drawn from the subject matter of a dependent claim and its features when interpreting the main claim, depends on the circumstances of the individual case. If the dependent claim is only adding an additional feature that does not provide a more specific description of the features of the main claim, it generally argues against the possibility of drawing conclusions about the interpretation of the main claim from this dependent claim. Where the interpretation of claim 1 of the patent is concerned, this must be established on the basis of its own claim features only.

Conversely, according to the CoA, there is no ground for interpreting claim 1 in the light of the additional feature in claims 6 and 7 prescribing administration of the claimed antibodies together with at least one other cholesterol-lowering agent. The CoA conceded that this additional feature may lead to a different interpretation of the level of required therapeutic effect of the claimed antibodies where it concerns claims 6 and 7 but held that the use of the antibodies according to claim 1 must be therapeutically effective on its own (paragraphs 43-46).

As stated above, the conclusion of both the CD and the CoA that a substance-for-use claim implies the successful use of the claimed substance for the claimed indication(s) seems completely reasonable, even though I am not convinced that this necessarily implies, in the present case, a certain or even high degree of lowering of cholesterol levels (note that prevention is also claimed!). But if the antibody alone can be (successfully) used for these purposes according to claim 1, should it then not have the same effect also in the context of a combination with another cholesterol-lowering agent as provided in claims 6 and 7? One might be able to use less of the antibody in such a combination, but I see no reason why its fundamental suitability for achieving a meaningful treatment or prophylaxis should be any different in claim 6 or 7. The skilled person will always choose the appropriate amounts of PCSK9 inhibitor to achieve the desired effect(s), depending on the circumstances (e.g. mono-administration or co-administration with another active pharmaceutical ingredient).

c)         Interpretation of means-plus-function claims

The CoA also helpfully clarified another important question of claim interpretation by its order in Abbott v Sibio (UPC_CoA_382/2024). As a general principle of claim interpretation, means-plus-function features must be understood as any feature suitable for carrying out the function. This is also consistent with EPO and earlier national case law and, as such, no big surprise.

d)         How to deal with apparent contradictions between the claim and the description?

In an ideal world, the description should “support” the claims in that it provides the necessary flesh to the bones, no more, no less. In particular, there should be no inconsistencies between the claim and the description. Moreover, the claims must be clear and concise. This is something that EPO examiners are being told to check under Art. 84 EPC. As is stated in the Guidelines for Examination Part F, Chapter IV, 4.3:

Any inconsistency between the description and the claims must be avoided if it casts doubt on the subject-matter for which protection is sought, thereby rendering the claim either unclear or unsupported under Art. 84, second sentence, or objectionable under Art. 84, first sentence. Such inconsistency can take the following forms (…).

But sometimes real life kicks in, an examiner may have had a bad day, there may have been an unusual amount of snow in Munich as early as the end of September, or whatever else may be the reason, it just happens sometimes that the description is not exactly aligned with the claims. Should in such a case the claim be interpreted more broadly, or should the description give way to the “primacy of the claims”? This is of course an old debate, see e.g. here, and many had hoped that the Enlarged Board of Appeal (EBA) would provide an answer thereto in its decision G1/24. However, it did not, at least not beyond stating that the description and drawings shall always be “consulted” to interpret the claims, whatever this means. Referral question 3, which would have been a bit more specific, was rejected as inadmissible. The EBA just preached:

The above considerations highlight the importance of the examining division carrying out a high quality examination of whether a claim fulfils the clarity requirements of Article 84 EPC. The correct response to any unclarity in a claim is amendment.

All well and good, but the UPC and national courts will sometimes also be confronted with patents where this “correct response to any unclarity” has not been carried out. What then? The CoA thankfully provided at least a partial answer in its decision Edwards v Meril (UPC_CoA_27/2025).

Namely, the CoA established that as a general rule, a product or process presented as an embodiment by the patent specification may be considered covered by the patent claims. However, there is room for an exception where the patent as a whole clearly teaches the person skilled in the art that the disclosed embodiment is not claimed, e.g. when it only illustrates a technical specification that is not addressed by the teaching of the patent claim. In the case at stake, the patent in suit described different embodiments. One of them, which was illustrated in Figs. 5-9, clearly had a frame made entirely of hexagonal cells with side struts in accordance with claim 1. “Another embodiment”, which was shown in Figs. 1-4, showed cells consisting of only four angled struts and which were rhomboid, not hexagonal shaped. As claim 1 required the frame of the claimed prosthetic heart valve to be “made up entirely of hexagonal cells”, the CoA rejected Meril’s argument that the frame shown in Figs. 1-4 is a frame according to the claimed invention.

Thus, while the description and the drawings must still always be used as explanatory aids for the interpretation of the patent claim, as is settled case law since the CoA decision 10x Genomics v. Nanostring, UPC_CoA_335/2023, this does not mean that absolutely anything that is designated as an “embodiment” or perhaps even “example” in the description of a patent can be used to broaden the otherwise clear meaning of a claim. As Edwards v. Meril has clarified, “Hexagonal cells” are still required to contain six corners; just four are not enough even when presented as an “embodiment” in the description and drawings.

e)         Claim Interpretation as a Remedy to Correct Errors

The interesting legal dispute between Alexion and Amgen has already been discussed on this blog, so I may leave it there. The following headnote from the CoA’s decision in this case (UPC_CoA_405/2024) may just serve as a reminder about the limits of a correction by way of claim interpretation (after similar attempts for correction by way of an amendment had already failed at the EPO):

1. A linguistic error, a spelling mistake or any other inaccuracy in a patent claim can only be corrected by way of interpretation of the patent claim if the existence of an error and the precise way to correct it are sufficiently certain to the average skilled person on the basis of the patent claim, taking into account the description and the drawings and using common general knowledge.

2. The patent claim must be interpreted from the perspective of the person skilled in the art. The applicant’s assertions during the grant proceedings, and in particular the TBA’s endorsement thereof, can be seen as an indication of the view of the person skilled in the art at the filing date.

Lesson 2: Added Matter is a question of law and will be decided using the EPO’s gold standard

By now, the CoA seems to have firmly settled on the EPO’s gold standard for determining “added matter” or, to be more precise, whether the subject-matter of a patent extends beyond the content of the application as filed (or the earlier application as filed in the case of a divisional). According to this standard, as interpreted by the CoA, the Court must thus first ascertain what the skilled person would derive directly and unambiguously using his common general knowledge and seen objectively and relative to the date of filing, from the whole of the application as filed, whereby implicitly disclosed subject-matter, i.e. matter that is a clear and unambiguous consequence of what is explicitly mentioned, shall also be considered as part of its content (Abbott v Sibio, para. 52; Amgen v Sanofi, para. 54, Meril v. Edwards, para. 77).

By and large, it seems to me that the CoA does not take a photographic approach when it comes to questions of added matter, and I welcome that. The UPC has, however, adopted the concept of an intermediate generalisation from the EPO Case Law. This means that added matter may result from an intermediate generalisation, i.e. the extraction of a specific claim feature from an originally disclosed combination of features (Meril v. Edwards, para. 79). The CoA added:

There will be added matter where the person skilled in the art would not consider the use of omitted features necessary for achieving the overall aim and effect of the invention based on their common general knowledge (UPC_CoA_382/2024, order of 14 February 2025, Abbott v Sibio, paragraph 75).

However, after reading this sentence a couple of times and then re-reading paragraph 75 of Abbott v Sibio, it seems to me that there must be an error here. The amendment considered by the Court resided in extracting a particular feature from a certain context of other features. Thus, added matter should only be found where the person skilled in the art would not consider the use of omitted features (likewise) necessary for achieving the overall aim and effect of the invention based on their common general knowledge. Conversely, where the person skilled in the art considered the other (omitted) features unnecessary for achieving the overall aim and effect of the invention, as was found in Abbott v. Sibio, they may as well be omitted. I hope that this corresponds to that which the CoA intended to express.

Lesson 3: Priority will be determined according to the EPO's "Gold Standard"

The UPC continues to determine the “same invention” criterion of Art. 87 EPC according to the EPO’s Gold Standard, corresponding to the standard for added matter. The decision by the CD Munich in the Amgen v. Sanofi case reported here has meanwhile been applied 1:1 by, inter alia, the Local Division of Hamburg (UPC_CFI_461/2024) and the Local Division of Düsseldorf (UPC_CFI_115/2024).

It is certainly to be welcomed that there is, at least so far, a common understanding on the principles of added matter, priority and novelty (i.e. the unitary concept of disclosure) between the UPC and the EPO. Unfortunately, this is not the case (at least not to the same extent) when it comes to the key question of inventive step. My impression is that sufficiency is also viewed a bit differently by the UPC. I will deal with both of them in part II of this article. Stay tuned.

Transparency notice: 

The author’s firm (Hoffmann Eitle) but not the author personally has been involved in some of the cases discussed above. This article, however, merely provides the author’s own opinion in his personal capacity as a UPC representative and occasional blogger. It does not seek to express the views of Hoffmann Eitle or of any of its clients and was not commissioned by any party in these proceedings. I respect the right of others to differ from the views expressed here and reserve the right to learn more and hold different views in the future. The use of this article in proceedings where I represent a party is not permitted.