06 October 2020

Patenting antibodies at the European Patent Office

With antibodies accounting for seven out of the top ten global drugsĀ¹, it is of critical importance that those companies that invest huge sums of money into R&D in this technology space are able to protect their investment from unlawful competition. Whilst the patent system provides a pretty good framework for achieving this, the approaches taken by patent offices in different parts of the world can vary widely, potentially impacting one’s ability to secure optimal patent protection (in terms of territorial scope and/or patent claim scope). Unsurprisingly, therefore, the risk of failing to achieve commercially viable patent protection in a key jurisdiction (plus loss of an important revenue stream) provides a strong incentive to operate at the highest possible patentability threshold. Here, we will focus on the European Patent Office (EPO), and on the key issues you will likely need to address in order to obtain patent protection in Europe for a new antibody therapeutic.

Obtaining Patent Protection for Antibody Subject-matter at the EPO

For antibodies directed at a new target, such as a new antigen structure that has not been targeted before, it may be possible to obtain broad patent protection at the EPO for antibodies that bind specifically to that target. However, these broad patents are becoming less common because it is now unusual to discover such a new target. Most antibody patents, therefore, relate to antibodies that bind to a known target, and where other antibodies binding to the same target have been described.

The EPO allows the patenting of new antibodies, but only when said antibodies also demonstrate an unexpected technical effect (i.e. an inventive step) when compared to antibodies that were known before. Unlike some patent offices around the world, which may concede that a new antibody is inventive because of a unique structure or sequence it possesses when compared with previously known antibodies, a difference in structure or sequence alone is not enough to establish inventive step at the EPO. Moreover, this remains the case irrespective of whether said unique structure or sequence maps to the framework regions or to the complementary determining regions (CDRs) of the antibody. Thus, a new antibody against a known target will only be considered inventive by the EPO if it shows an unexpected property, or if it was unexpected that such an antibody could be produced at all.

A key component of the EPO’s reasoning is that many techniques in the field of antibody production are routine, and that antibodies against a given target can be produced in large numbers without any inventive input being needed. For example, the EPO considers it routine to immunise animals with an antigen, to obtain a large number of different antibodies against that antigen that are produced by the animals, and to screen the resulting antibodies to confirm binding to the antigen of interest. Because the generation of antibodies in such immunisation methods is essentially random, the EPO assumes that essentially all antibodies against that antigen could be found eventually by just routine trial and error experiments, given a sufficient amount of time and resources. As a starting point, therefore, the EPO will assume that any antibody that has been produced against a known target could have been found in a routine way, and so is not inventive. Thus, the burden lies with the applicant of a patent application to convince the EPO otherwise.

The EPO also considers that other techniques in the antibody field, such as humanisation and affinity maturation, are now routine, and again that trial and error would eventually produce any effective humanised or affinity matured variants of a starting antibody. This suggests that over time, it may become increasingly difficult to persuade the EPO that antibody claims are inventive, as more techniques for antibody production, optimisation and selection become routine in the field. Thus, for new antibodies that bind to a known target (particularly if other antibodies against that target are known), the applicant must demonstrate there is something that makes said antibodies surprisingly better than other antibodies that were known to bind to the same antigen (else surprisingly better than would have been expected based on what was known about the target antigen and corresponding antibodies).

In theory, any kind of advantage can be relied upon. For example, this might relate to the way that the antibody binds to its target, such as improved specificity, cross-reactivity, or affinity; it might relate to improved properties of the antibody in vivo or in vitro, such as improved pharmacokinetic properties, low immunogenicity, or improved biological activity; or it might be based on other properties of the antibody which do not relate directly to its binding properties, such as improved storage stability, improved formulation properties or improved expression levels.

In practice, any advantage relied on must be surprising in its own context. For example, for a humanised antibody, the EPO might reasonably expect that it will have reduced immunogenicity compared to an antibody that is not humanised, so such a technical effect alone is unlikely to be enough to confer an inventive step. However, if a humanised antibody were to retain a high affinity for its target antigen, this might reasonably be considered surprising and thus supportive of an inventive step. Conversely, if an asserted technical effect is found to be unsurprising, then it is unlikely the EPO will allow any patent claim to the antibody, irrespective of whether the antibody in question is defined by reference to all six CDR sequences, both variable domain sequences, or even the full amino acid sequence of the antibody molecule. In such scenarios, even a very narrow “picture” claim is unlikely to be awarded.

Show more

Having established the presence of an unexpected technical effect (and thus inventive step), the next question to be considered by the EPO is that of claim scope. Namely, how broad can I claim variants of the antibody?

This assessment flows from analysis of the scientific principles that underpin the asserted unexpected technical effect, and attempts to ensure that the scope of patent claim awarded is commensurate with the level of scientific contribution the invention provides above and beyond the prior art. To put this another way, the EPO’s general approach is the unexpected technical effect relied on to support inventive step must be demonstrated across the full scope of the awarded patent claims. In more detail, it must be at least technically credible that all the antibodies across the scope of your defined claims demonstrate the same unexpected technical effect / advantage. The EPO’s analysis will, therefore, involve considering the properties or features of your antibody that are responsible for that advantage.

For example, you may be able to establish that your antibody is inventive because it has particular binding selectivity for one antigen and not to another. If your inventive step is based on the binding specificity or selectivity of your antibody, then the EPO is likely to take the view that the advantage might reasonably be shared by other antibodies that have the same set of six CDRs, or both variable domain sequences of your antibody. For this type of advantage, the EPO will, therefore, usually insist that you limit your patent claims to antibodies that include all six CDR sequences, or both variable domain sequences, of your antibody. It can be possible to obtain broader protection than this, but to do so, it is likely that you will need evidence that the same advantage is also found for other antibodies that do not have these particular sets of sequences. For example, to obtain a patent that does not require all six CDR sequences to be present, you might need data showing that antibodies with fewer than six CDRs, or antibodies having particular variations in the CDR sequences, will retain the same inventive advantage. The scope of patent claims that you will obtain will depend upon the related antibody sequences that you can persuade the EPO will retain the inventive advantage.

Another common advantage that is used to establish an inventive step at the EPO is improved affinity. The EPO considers that the choice of framework regions, as well as the CDR sequences themselves, may considerably influence antibody affinity. This means that if inventive step is based solely on an antibody having improved affinity for a target, then the EPO is likely to require the framework regions and the CDR sequences to be defined in the patent claim. In practice, this means that you will be asked to limit your patent claims to antibodies having the same heavy and light chain variable region sequences as your antibody. Again, if you wish to obtain broader patent scope, then it is likely that the EPO will require supporting evidence that the improved affinity would be retained with other framework.

The same principles will be applied by the EPO to any advantage that you are relying upon to obtain an inventive step. If the advantage is linked to a particular structure or feature of your antibody, then the EPO is likely to require that structure or feature to be defined in the claims. For example, if your advantage relates to improved effector functions, then the EPO may require particular Fe domain sequences to be recited in the claims. If the advantage relates to a physical property of the antibody, such as its stability or production yield, then the EPO may consider that to be a property of the molecule as a whole, and so require the full antibody sequences to be defined in the claims.

If you need to rely on a technical advantage over other antibodies, then how can you establish that such an advantage exists, and when do you need to provide that information? The EPO considers that it must be derivable from your original patent application that the invention had been made before that application was filed. This does not mean that your patent application needs to provide absolute proof of the advantage. Indeed, it may not be possible to include the ideal comparisons in your application to prove that an advantage exists. For example, the EPO requires that an inventive step is established when compared to what it considers to be the “closest prior art”. In this field, that is likely to be an earlier antibody that binds to the same antigen and that has similar properties. However, you may not know at the time of the patent application being filed what other antibodies may exist to the same target, and you may not be able to determine which antibody the EPO will later consider to be the “closest”. Even if you are aware of earlier publications describing other antibodies, those antibodies may not be publicly available, and so it may not be possible to carry out any direct comparison in order to confirm that an advantage exists.

What the EPO will look for in the patent application is enough information to make it technically plausible that the advantage would be achieved. Your patent application might include data demonstrating particular effects or measuring particular parameters for your antibody, and might, therefore, provide data that could be used for a subsequent comparison with other antibodies, or it might include technical reasons why an effect or advantage can be plausibly derived from the available data.

If you can meet this threshold and persuade the EPO that your advantage was technically plausible from the information in your original patent application, then you may be permitted to rely on additional evidence, not included in the original patent application, to confirm the existence of the advantage. For example, you may be able to submit in vivo data confirming effects that were shown in vitro, or you may be able to submit comparative data confirming that your antibody does show an improvement when compared to particular antibodies that the EPO has selected as the “closest”.

In conclusion, the EPO takes a technical and scientific approach when considering inventions in the antibody field. Every case will be judged on its own facts, but in general, the EPO will start from a number of preconceptions about what could have been done in a routine way, and the burden is likely on you to counter those preconceptions in order to persuade the EPO that your antibody is inventive.

An antibody that is new, and that is effective at binding its desired target, is unlikely to be considered inventive by the EPO unless it also exhibits some kind of unexpected technical effect (e.g. an advantage) when compared to other antibodies against the same target. This is worth considering when you draft a new patent application in this field. Ideally, your patent application will include some data supporting the superior properties of your antibody, or it will at least include a technical rationale to make it credible that your antibody has such an advantage. You should also consider the scope of claim that the EPO is likely to allow based on the advantages that you can establish. If you want to obtain broader claims than are likely to be allowed by default, then you may need to obtain more data before filing the application, to show that your advantage can be obtained with a broader range of antibodies than might otherwise be expected. Most importantly, when preparing a patent specification, you should think ahead to present a tiered range of (plausible) technical effects that may be relied on to provide potential fallback positions during prosecution and beyond.

This article was originally published in the International Biopharmaceutical Industry Journal.


1. Urquhart (2020) Nature Reviews Drug Discovery 19: 228