How much data does a Patentee need to claim every molecule that hits a target? For antibodies, the Boards of Appeal have long accepted that identifying a new target or epitope can justify a claim to the whole functional genus of antibodies that bind it. The recent decision in T 1355/24 extends that generosity to molecules of any kind. Novartis’s patent claimed a composition comprising “a BRM inhibitor” for treating BRG1-mutant lung and ovarian cancers. The only BRM inhibitor exemplified in the application was a short hairpin RNA (shRNA). The Opposition Division found that the functional definition reached into territory where the patent made no contribution. The Board of Appeal disagreed, holding that the discovery of a synthetic lethal relationship “opens up a new field” and that a claim commensurate with that contribution was justified.
Legal background: Functionally defined therapeutic agents and the reach-through problem
Article 83 EPC requires that a European patent application “disclose the invention in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art”. For a second medical use claim in the purpose-limited product format of Article 54(5) EPC, the therapeutic effect is a functional feature of the claim. Whether that effect is achieved by substantially all embodiments covered by the claim is therefore a question of sufficiency, and it is usually only necessary for the patent to render it credible that the therapeutic agent is suitable for the claimed therapeutic application (T 1959/15, Case Law of the Boards of Appeal, 11th edn, pp. 511-512). Clinical data are not always required. Mechanistic data from which the skilled person can extrapolate to the claimed indication may be enough (T 1601/22).
The complication in T 1355/24 was that the therapeutic agent itself was defined only by its function. The settled law on functional definitions goes back to T 292/85, in which the Board of Appeal held that an invention is sufficiently disclosed if at least one way of carrying it out is clearly indicated, and that the non-availability of some variants of a functionally defined component is immaterial so long as suitable variants are known to the skilled person from the disclosure or the common general knowledge (CLB II.C.5.4, p. 464). There is, however, a limit to this principle. In T 1063/06 the Board of Appeal held that a claim to functionally defined chemical compounds, which were to be found by means of a screening method set out in the description, was a “reach-through” claim directed to future inventions. An Applicant is entitled to protection only for their actual contribution to the art and may not reserve an unexplored field of research (CLB II.C.6.5, p. 494). More recently, in T 3122/19, a claim to FGFR inhibitors for a subgroup of cancer patients was distinguished from T 1063/06 because the pool of candidate compounds was limited and identifying the active ones was not an undue burden.
The Boards of Appeal have been notably relaxed about functional genus claims for antibodies. In T 0835/21 an epitope claim was maintained on the basis of two exemplified antibodies and no sequence information, making antibodies to a known target being routine even if “tedious”. In T 0424/21 a genus of Fc-modified antibodies for use as a medicament was upheld on in vitro data alone. The present case asked whether this principle could be applied even more broadly, beyond the antibody modality class.
Case background: Synthetic lethality in the SWI/SNF complex
The patent in question, EP 2971094 (Novartis), related to biomarkers associated with BRM inhibition. The mammalian SWI/SNF chromatin remodelling complex contains one of two mutually exclusive ATPase subunits, BRG1 (SMARCA4) or its paralog BRM (SMARCA2). Some lung and ovarian cancers carry homozygous loss-of-function mutations in BRG1, and the patent application reported that the inventors found that these BRG1-deficient tumour cells become critically dependent on BRM for survival. Knock down BRM in a cell that has already lost BRG1 and the cell dies, whereas knocking down BRM in a cell with intact BRG1 does little. This is a classic synthetic lethal relationship, familiar from PARP inhibitors in BRCA-mutant cancers.
Claim 8 of the main request on appeal was a second medical use claim to “A composition comprising a BRM inhibitor for use in treatment of squamous cell lung carcinoma, non-small cell lung carcinoma, large cell lung carcinoma, lung adenocarcinoma, and ovarian cancer in a selected patient population”, the population being selected on the basis of a cancer “associated with one or more homozygous loss of function BRG1 mutations”.
BRM (SMARCA2) is currently a clinical target, with clinical programmes selecting patients on the biomarker recited in claim 8. Lilly’s LY4050784 (originated by Foghorn as FHD-909), an oral small molecule that selectively inhibits the ATPase activity of BRM, is in a Phase 1 trial in SMARCA4-mutated solid tumours with non-small cell lung cancer as the primary target population. Prelude Therapeutics has two selective BRM degraders in the clinic, PRT3789 and PRT7732, in SMARCA4-mutated cancers. None of these is an shRNA. All of them are small molecules or degraders of the kind the application mentioned only as “attractive avenues” for future development, and all of them fall within the functional definition of claim 8.
The application as filed demonstrated the synthetic lethal effect using BRM-specific short hairpin RNAs (shRNAs), which reduce the amount of BRM protein by interfering with translation of its mRNA. The shRNAs inhibited proliferation of BRG1-mutant tumour cells in vitro and in a mouse model. Importantly, no other type of BRM inhibitor was exemplified. The application noted that the bromodomain and ATPase domain of BRM “present attractive avenues for the development of targeted inhibitors” (r. 11), and described in general terms a screening assay in which BRG1-deficient cell lines are contacted with candidate compounds (r. 14).
The patent was opposed by a straw-man Opponent, inter alia for insufficiency. The Opposition Division accepted that the contribution of the patent lay in the discovery of which cancers were sensitive to BRM inhibition, but held that the functionally defined feature “BRM inhibitor” “extended to areas where at the date of filing, the patent made no contribution to the art” (r. 8). The patent was maintained on the basis of a narrower auxiliary request. The Patentee appealed alone. The Opponent filed no substantive reply and did not attend. After a preliminary opinion from the Board of Appeal that the broadest version of claim 8 was insufficiently disclosed but the auxiliary version was not, the Patentee withdrew its main request, promoted the auxiliary request and withdrew its request for oral proceedings. The decision was taken in writing, confined by the prohibition of reformatio in peius to the sufficiency of claim 8 (r. 2).
What exactly has to be enabled?
Since claim 8 was a purpose-limited product claim, the Board of Appeal noted that “what has to be enabled in the patent application is that the therapeutic agent (here substantially all types of BRM inhibitors) have the desired therapeutic effect” (r. 9). It was uncontested that the feature “BRM inhibitor” “encompasses any substance that inhibits BRM regardless of its structure, specificity, mode of action and extent of inhibition, since this feature is functionally defined only” (r. 10).
The Board of Appeal was also clear about the limits of the disclosure in the application as filed. The application exemplified only shRNA and disclosed no bromodomain or ATPase inhibitors (r. 13). It also repeatedly emphasised that, given the close structural relationship between BRM and BRG1, “a suitable BRM inhibitor should be selective for BRM, in particular over BRG1 for minimising undesired side effects” (r. 12). One might have thought that an application which itself flags selectivity as a requirement, whilst exemplifying only a modality that is inherently sequence-selective, had set itself a problem for a claim covering non-selective small molecules.
The Board of Appeal did not see it that way. According to the Board of Appeal:
“It is further credible that the therapeutic efficacy of a BRM-specific shRNA in BRG1-deficient tumours serves as a proof-of-principle that BRM inhibitors in general are suitable for the claimed therapeutic application. All of the claimed tumours carry a homozygous loss of function BRG1 mutation. Since the survival of these tumours (in the absence of a functional BRG1) depends critically on the presence of a functional BRM (point 5 above), substantially all types of BRM inhibitors should be therapeutically effective irrespective of the inhibitors’ structure and mode of action.” (r. 15)
The mechanism, in other words, was therefore considered to support the whole claim. If a tumour cell cannot live without functional BRM, then anything that removes functional BRM should kill it, whether by degrading the message or blocking the protein. The Board of Appeal added, pointedly, that “Evidence to the contrary is not on file” (r. 15). There was “also no evidence on file that the skilled person had technical problems in identifying further BRM inhibitors” (r. 16). The application taught a screening assay, and the sequence of BRM was known.
Post-published evidence cuts both ways
Another interesting point from T 1355/24 is the evidence the Board of Appeal relied on to confirm its finding. Two post-published documents were on file: a 2018 medicinal chemistry paper reporting a screen of some 72,000 compounds against the BRM ATPase domain and the identification of small molecule inhibitors with in vitro and in vivo tumour growth inhibition (r. 17), and a 2024 Lilly press release on a BRM inhibitor, LY4050784, reported as effective in the claimed indication (r. 18). The Board of Appeal took both as confirmation that there were no technical obstacles to finding BRM inhibitors.
However, the medicinal chemistry paper was not an unqualified success story. The Board of Appeal itself noted that the compounds “exhibited excellent cellular potency and oral exposure” but that the “ability to fully explore the potential of BRM inhibition on tumor growth in vivo was hampered by dose-limiting tolerability issues”, apparently because they were “dual inhibitors of BRM and BRG1” (r. 17.2). The Board of Appeal’s response was that “while selectivity of BRM inhibition over BRG1 inhibition in cancer treatment is desirable (for the reduction of side effects), the compounds are therapeutically suitable for the claimed purpose” (r. 17.2). Suitability for the therapeutic purpose, in other words, does not require that the compound be a good drug.
The Board of Appeal concluded that “the teaching of the patent application opens up a new field, namely the treatment of a specified patient group by preventing BRG1-deficient tumour cell proliferation mediated by the use of BRM inhibitors. In these circumstances it is justified to allow a claim that is commensurate in scope to its contribution to the art” (r. 19). The decision under appeal was set aside and the case remitted to the Opposition Division with the order to maintain the patent on the basis of the main request, with an adapted description.
Analysis
T 1355/24 confirms that, in Europe, one way of carrying out the invention, together with a generalisable technical concept, may be enough to support a functional definition that is not even limited by modality.
Notably, however, the Board of Appeal did not cite a single previous decision. In T 1959/15, by contrast, a claim to “an inhibitor of soluble epoxide hydrolase” for use in inhibiting cardiomyopathy was revoked for insufficiency. The patent exemplified two structurally similar urea inhibitors. The Board of Appeal in that case observed that inhibitors of the enzyme could be structurally very distinct, including synthetic compounds, antibodies and nucleic acids, and held that two compounds did not make it plausible that all of them would work. Finding the rest would be a matter of trial and error, and an undue burden. The claim in T 1355/24 was supported by less, a single shRNA, and yet survived. It is also worth noting that this was not a contested decision. The Opponent filed nothing of substance on appeal, and the Board of Appeal’s reasoning leans heavily on the absence of evidence to the contrary (r. 15, r. 16). A well-resourced Opponent might well have submitted evidence that selective small molecule inhibition of BRM was a serious scientific challenge at the priority date, as suggested by the tolerability issues mentioned in the post-published literature.
There is also a transatlantic parallel here. A claim to a functionally defined genus of molecules per se would be unlikely to survive written description and enablement in the US after Amgen v Sanofi. However, earlier this year the Federal Circuit in Teva v Eli Lilly upheld a method of treatment claim for any humanised anti-CGRP antagonist antibody, reasoning that the antibodies themselves were not claimed, that all of them would work in the claimed method, and that finding or making all of them would be “more akin to extra credit than a necessary research assignment”. The Board of Appeal in T 1355/24 reaches a similar place by a different route. The BRM inhibitors are not claimed as such, the Board of Appeal thought that the proof of mechanism meant that it was credible that substantially all of them should work, and finding them is routine screening against a known target. The difference is that in Teva v Eli Lilly the claimed genus of antibodies was already well known in the art, whereas in T 1355/24 the only BRM inhibitor in existence at the priority date, so far as the application tells us, was an shRNA. The Board of Appeal was therefore prepared to accept that a therapeutic use claim can cover a whole class of molecules that had, for the most part, yet to be invented.
Author: Rose Hughes
Rose is a biotech and pharmaceutical patent specialist with over a decade of experience in intellectual property. Rose is a patent attorney at Evolve, where she leverages our unique fractional in-house model to provide clients with deep patent law expertise combined with the strategic commercial oversight typically associated with senior in-house counsel.
With a PhD in Immunology from UCL, Rose applies her technical background to complex innovations in biologics, cell and gene therapies, and the rapidly emerging field of AI-assisted drug development. Previously, Rose held the role of Director, Patents at AstraZeneca, where she was responsible for global IP portfolios and IP strategy at every stage of the pharmaceutical pipeline, from platform development and on-market commercialization to SPCs and patent term extensions.
A recognized thought leader in the field, Rose has been a regular contributor to IPKat since 2018, offering practical insights into European patent law developments. She is also a frequent speaker on the epi podcast, a guest lecturer for the Brunel University IP law Postgrad Certificate, and a contributing author to published books A User’s Guide to Intellectual Property in Life Sciences (2021) and Developments and Directions in Intellectual Property Law (2023).
