Patenting Antibodies: The epitope claim is dead, long live the epitope claim

Antibodies may be defined in a patent claim by their amino acid sequence or by the sequence of the target (epitope) to which the antibody binds. Historically, epitope claims were relatively common in the field of antibody patents, as they represented a way of broadly protecting any antibody that bound the same epitope. As the field of therapeutic monoclonal antibodies has grown, epitope claims have become progressively more difficult to obtain. Patent offices around the world now commonly insist that a claimed antibody is defined by its structural characteristics and not just by its function. The recent US case law in Amgen v Sanofi particularly has been hailed as signalling the death of the epitope claim. In Europe, by contrast, the case law on epitope claims is patchy. Nonetheless, recent first instance decisions indicate that the EPO remains comfortable with the broad claim format of epitope claims. 

Legal background: Death of the epitope claim?

Antibody

In the US, the stringent enablement and written description requirements have made it very difficult (if not quite impossible) to achieve grant of epitope claims. Most recently, the  decision of the Federal Circuit in Amgen v. Sanofi (No. 20-1074, Fed. Cir. 2021) confirmed the high patentability hurdle for epitope claims. The claims in question were directed to a genus of anti-PCSK9 antibodies defined only by their target epitope. The Fed. Circuit found that this claim placed too high a burden on the skilled person attempting to practise the invention. The argument was that the skilled person would have to screen millions of antibody candidates to determine which antibodies fell under the functional definition of the claim. The decision has been interpreted as not theoretically ruling out epitope claims, but of raising the bar to enablement so high as to make them almost practically impossible to obtain (patentlyo).

A lack of EPO Board of Appeal case law on epitope claims

The EPO case law appears, on the face of it, to be more lenient than the US with regards to epitope claims. The latest version of the EPO Guidelines for Examination (which included for the first time a section detected to antibodies (IPKat)), confirmed that antibodies may be defined by their antigen (G-II, 5.6.1.2), and particularly by their target epitope (G-II-5.6.1.6). However, as far as this Kat is aware, there is currently a dearth of Board of Appeal case law on the sufficiency bar for epitope claims (Case Law of the Boards of Appeal, II-C-7.3). In the EPO appeal case corresponding to the Federal Circuit Amgen v. Sanofi decision, Amgen amended the epitope claim so that the sequence of the PCSK9 antibody was also defined (T 0845/19). The Board of Appeal thus did not provide a final decision on the validity of the epitope claim. 

The epitope claim lives on (at EPO first instance) 

Whilst there is a lack of Board of Appeal case law on epitope claims, there are a number of examples of epitope claims surviving opposition in Europe. EP1793855, for example, was granted with an impressively broad claim to an antibody that binds to a specific epitope within human alpha-synuclein for use in therapy. The Opponents in the case argued that the disclosure of the patent did not fulfil its side of the patent bargain, in that it did not provide a sufficient enough disclosure to justify the broad monopoly granted. Particularly, the Opponent argued that there would be an undue burden placed on the skilled person carrying out the invention across the full scope of the claims, and that not all antibodies embraced by the claim would exert the desired therapeutic effect. The Opposition Division did not agree. Particularly, whilst the OD did agree that there was no guarantee that any antibody binding the epitope would have the claimed effect, the OD nonetheless found that the examples provided in the patent provided an expectation of success. The OD placed emphasis on the lack of any data submitted by the Opponent contradicting the claims of the granted patent. In other words, the patent did not lack ab initio plausibility (for more on the different types of plausibility, see IPKat: Enlarged Board of Appeal to tackle the "Humpty Dumpty-ish" plausibility question (G2/21)). 

EP2812443 as granted claimed an antibody that binds a specific epitope on the antigen CD47, and thereby inhibits the function of CD47. The specification disclosed example anti-CD47 antibodies falling under the scope of the claims. The Opponent argued that the claimed antibodies were insufficiently disclosed because generating such antibodies would require an unduly burdensome research programme. The Opponent submitted that the method for determining  the epitope of candidate CD47 antibodies, X-ray crystallography, was particularly challenging. At oral proceedings, the Opposition Division rejected this argument, and maintained the patent. 

These two cases show that, whilst the epitope claim may be more-or-less dead in the US, it does seem to live on in Europe, at least for now. It is expected that one of these cases, or another similar case, will eventually reach the Boards of Appeal. We can therefore hope that confirmation of the sufficiency bar for epitope claims in Europe may be forthcoming in the next few years. 

Final thoughts: "Routine experimentation" versus "Undue burden"

With all the focus on epitope claims following the Amgen v Sanofi decision in the US, there can be tendency to forget about the challenge of protecting an antibody in the more common case where an antibody target or epitope is already known. In such cases, a broad epitope claim is ruled out for lacking novelty. The majority of antibody patents are thus necessarily limited by the sequence of the antibody itself, and not its target. However, in Europe, there can be particular difficulty in convincing the patent office that even a specifically claimed antibody sequence is inventive. The EPO will often take the view that the generation of a new antibody against a known target falls under the category of routine experimentation, unless an unexpected property of the new antibody can be demonstrated in view of the closest prior art (II-5.6.2). The availability of epitope claims in Europe may thus belie the more common challenge of inventive step to antibody patentability in Europe. 

Are the positions of the USPTO and EPO consistent with respect to how easy it is to generate new antibodies for a target? One might argue that if working across the scope of an antibody genus claim would be an undue burden, selecting an alternative species from within the genus should be considered non-obvious. Although, there is of course no requirement for alignment between jurisdictions on antibody patentability. Consequently, as with diagnostic methods inventions (IPKat), applicants face distinct and sometimes contradictory challenges to the patentability of antibody inventions across different jurisdictions. It is thus worth remembering that the purported demise of the epitope claim in the US is only a snap-shot of these challenges. 

Further reading

1 June 2021: Reference to a hybridoma does not limit a product-by-process antibody claim (T 0032/17)

25 Oct 2021: Enlarged Board of Appeal to tackle the "Humpty Dumpty-ish" plausibility question (G2/21)