Legal background: The novelty of non-medical use claims versus process claims
The distinction between second non-medical use claims and process claims is crucial for the assessment of novelty. Use claims, e.g. use of compound X for use in a (non-medical) use Y, are an acceptable claim format in Europe. However, specific case law applies to novelty assessment of such claims, as compared to process claims (IPKat). Second medical use claims are dealt with by different provisions.
![]() |
Antibody manufacture |
In other words, according to G 2/88 and G 6/88, a new use based on a newly discovered technical effect of a known compound can be novel even if the physical activity required to carry out the use was already known. For example, the use of compound as e.g. a fungicidal plant spray, may be considered novel even if the use of the compound as a general spray for plants was already known. Whilst the fungicidal use and effect of the compound could be considered inherent in the previous use of spraying plants, the new use is considered to include the fungicidal technical effect as a technical feature of the claim, rendering the claim novel (T 231/85).
Non-medical use claims do not permit patent protection for known products
The EBA in G 2/88 made it clear that its findings on second non-medical use claims applied specifically to use claims that do not result in a product (r. 5.1). For process claims, Article 64(2) EPC states that "[i]f the subject-matter of the European patent is a process, the protection conferred by the patent shall extend to the products directly obtained by such process". Given that, according to G 2/88, the use considered therein are not those resulting in a product, they should also not fall within the scope of Article 64(2) EPC. As such, second non-medical use claims cannot be used as a back-door method for achieving patent protection for an already known product, even if a new functional feature of the product has been identified.
Case background: Antibody manufacture
T 1913/21 related to Biogen's European patent EP 2707383 for methods of preventing and eliminating the formation of trisulfide bonds in proteins, e.g. antibodies, during manufacture. Unwanted trisulfide bonds in antibodies are generally undesirable modifications that may inadvertently affect the antibody's desired function. Claim 1 on appeal was as follows:
"Use of an inhibitor of cysteine degradation for reducing the formation of trisulfide bonds in proteins, wherein the inhibitor of cysteine degradation is selected from pyruvate [...], wherein the use comprises: culturing cells expressing said proteins in the presence of an effective amount of the inhibitor of cysteine degradation, whereby trisulfide linkage formation in said proteins is reduced [...]"
Claim construction: Use claim or process claim?
Following opposition by Roche, the patent was found to be novel and inventive. The Opposition Division construed claim 1 as a second non-medical use claim directed to e.g. pyruvate for reducing trisulfide bond formation in proteins, wherein the technical effect of pyruvate in reducing trisulfide linkage formation was considered a limiting functional feature of the claim. On appeal, the central issue was whether the claim should be construed as a second non-medical use claim (as argued by the Patentee) or as a process claim for the production of proteins (as argued by the Opponents).
On appeal, the Board of Appeal disagreed with OD's analysis. The Board of Appeal particularly found that the claim should be regarded as a process claim for the production of proteins, not as a non-medical use claim in the sense of G 2/88 and G 6/88. The Board noted that:
"[T]he claimed subject-matter comprises carrying out process steps, which result in the production of a product. Following the principles laid down in G 2/88 and G 6/88 [...] the claimed subject-matter must be regarded as a process for the production of proteins with a 'reduced' number of trisulfide linkages, regardless of the fact that the claim is drafted as the 'use' of a chemical compound." (r. 20)
The Board noted that the technical effect stated in the claim ("reducing the formation of trisulfide bonds in proteins") could not occur except as part of a process for producing proteins:
"[W]here an invention relates to a new technical effect of a physical entity that can only occur as part of a process for the production or manufacture of a product, such that this effect is inextricably linked to and cannot occur in isolation from the production process, a claim directed to that 'use' of the physical entity to achieve that effect must be regarded as directed to the production process per se." (r. 23)
Having construed the claim as a process claim, the Board of Appeal assessed the claims novelty accordingly. The prior art documents disclosed processes for the production of recombinant proteins in which cells were cultured in the presence of pyruvate. The technical effect of "reducing the formation of trisulfide bonds" was not explicitly mentioned in these documents. However, the Board found that because the claimed subject-matter was a process for the production of a product, the stated purpose or effect was not a limiting feature. As a result, the Board concluded that claim 1 lacked novelty.
Conflicting case law?
The Patentee requested a referral to the EBA, arguing that the Board of Appeal's approach to claim construction represented a deviation from G 2/88. However, the Board of Appeal rejected this request on the basis that the teaching of G 2/88 was limited to true use claims, and that it had therefore properly applied the decision. The Board of Appeal also cited a number of previous decisions, including T 1855/06, T 304/08, T 1179/07, T 684/02, that had also highlighted the limited applicability of G 2/88 and G 6/88.
Interestingly, the Board of Appeal nonetheless acknowledged divergent case law on the issue. In its reasoning, the Board of Appeal explicitly disagreed with decisions T 0062/02 and T 1011/04, which had interpreted similar claims in the sense of G 2/88. Addressing this disagreement head-on, the Board of Appeal in the present case explained that T 0062/02 and T 1011/04 contradicted the established case law. A critical failing of these earlier decisions, according to the Board of Appeal, was their lack of reasoning regarding the distinction between use claims and process claims under Article 64(2) EPC. For the Board of Appeal, these previous decisions did not address why the technical effect recited in the claims should be considered a functional feature when that effect related to a property of the product rather than to the use itself.
Final thoughts
The decision in T 1913/21 provides further clarification on the distinction in the novelty assessment between non-medical use claims and process claims. In its decision, the Board of Appeal made it clear that the format of a claim, and particularly whether or not it includes the word "use", is not determinative. What matters is the technical subject-matter claimed.
This case thus confirms that the mere drafting of a claim as a "use" claim cannot circumvent the fact that when the claim defines a production process for a product, the claim will be considered a process claim. Therefore, even when a technical effect of a compound is inextricably linked to a production process, the EPO is still likely to treat such a claim as a process claim and not a use claim, regardless of the claim format.
Antibody manufacturing patents can be valuable if they protect a generally applicable method that is adopted by the industry. However, process patents also have significant drawbacks, given that competitors may be able to easily design around process claims or use the method without this being readily apparent to the patentee. The patentee in the present case faced the even more fundamental problem that the antibody manufacturing process they were seeking to protect involved a compound already commonly used in antibody manufacture. The method was therefore considered inherently disclosed by the prior art methods, despite the identification of a new technical effect of the compound and the "use" format of the claim.
Further reading

Fundamentally this case serves to highlight, once again, that G 2/88 is a bad and intractable decision and we are still grappling with its implications over 35 years later. And yet no Board seems willing to step up to the plate and ask the EBA to have another go. Why is that, do readers think?
ReplyDeleteA few comments:
(1) if claims are in substance considered "process" and not "use" claims, this obviously has implications for their scope vis-a-vis the protection conferred on the direct product of that process, regardless of how they are written. This is well known.
But now imagine a situation where claims of the type which were at issue here are interpreted in substance as process claims, and then found novel and inventive by the EPO (i.e. in situations where the prior art did not disclose all of the integers of the claim "but for" the novel technical effect). Article 64(2) then says that the direct product of the process is protected, notwithstanding their formatting as "use" claims. Would that be respected by the courts in an infringement action, hinging as it does upon the EPO's interpretation?
(2) does this decision highlight a tension between the Enlarged Board's case law on second medical indications on the one hand (see the Case Law book, I.C.8.1.1 quoting G 5/83: "whether an activity is claimed as a method of carrying out the activity ... or as the use of a thing for a stated purpose ... is a matter of preference. For the Enlarged Board of Appeal there is no difference of substance") and the special treatment given to G 2/88-style use claims on the other hand?
(3) it is worth noting that it was only one of the two opponents who argued that this was in substance a process claim - the other opponent did not raise novelty as a ground of opposition in the appeal, and in fact supported the Patentee's request for referral of questions to the EBA.
(4) did the Board get it right in denying a referral to the EBA in this case? Part (iii) of the order in G 2/88 says "A claim to the use of a known compound for a particular purpose [...] should be interpreted as including that technical effect as a functional technical feature [...] provided that such technical feature has not previously been made available to the public." It does not say "unless the claim is in substance to be interpreted as a process claim". It's arguable that the order in G 2/88 focuses principally on legal form - "is it written as a use claim or not" - rather than the substance of the steps recited in the claim (as long as the alleged technical effect was not previously disclosed).
(5) Is the distinction between "processes that make a product vs uses that do not" as clean as the Board here suggests? One could argue that at least the fungicide example discussed in G 2/88 falls into the same category as the claims in this case - the "product" being a plant with a lower degree of fungal infection than if the fungicide had not been used. Would the fungicide in G 2/88 still be considered novel today in light of T 1913/21, and if so, is that merely because the fungicide "use" claim in question did not include any steps defining a process and/or defining the characteristics of an end product? Again, is this a question of legal form ("does the claim recite certain features or not"), or one of technical substance ("how does the alleged new use manifest itself")?
This decision was taken by Board 3.3.04, the same Board which declined to refer questions to the Enlarged Board in T 56/21 despite acknowledging divergent case law there too.
ReplyDeleteCurious.