Navigating Australian antibody patent protection strategies 12 min read
Many of the top selling drugs in Australia are antibodies, all part of a global market estimated to reach US$140 billion this year. Recently, the US FDA (Food and Drug Administration) approved the first interchangeable biosimilar antibody, Cyltezo (adalimumab-adbm) - biosimilar to and interchangeable with its reference product Humira (adalimumab). With the Australian Government also looking to support the use of biosimilars, robust patent protection for antibody-related inventions is more important than ever.
In the US and Europe, broad antibody claims have become harder to obtain because of stringent patentability requirements and evolving case law. The current European Patent Office (EPO) Guidelines for Examination specifically covers a series of antibody patentability issues often raised during examination. While Australia's Patent Manual of Practice and Procedure (the Patent Manual) provides some guidance on this matter, there is little judicial consideration of how Australia's patent law relates to therapeutic antibodies.
Notably, section 40 of Australia's Patents Act 1990 was amended under the 'Raising the Bar' Act in 2012 to align with European patentability requirements of support and enablement (only recently considered in the Federal Court). It’s now clear that biopharmaceutical companies face similar hurdles in Australia in obtaining broad claims to antibodies and their therapeutic uses. Here, we consider some of the main challenges and strategies for claiming antibodies in Australia.
While Australia has examination guidelines for antibodies, it doesn’t yet have a comprehensive body of case law to guide the prosecution of antibody-related inventions.
Obtaining broad antibody claims is becoming more difficult. In protecting antibodies, consider how they are defined in the specification (by structure, function, antigen/epitope, or a combination of these), and remember it’s up to the applicant to differentiate the claimed antibodies from prior art antibodies.
Keep in mind the detail needed to support and enable the antibody across the claims’ scope. Make sure any unexpected effects are well described in the specification, ideally with experimental data.
When assessing a claim to an antibody, an Australian examiner will consider many technical aspects of antibody technology routine. In particular: antibody structure-function relationships, methods for producing, identifying or selecting antibodies, and methods for modifying antibodies including humanisation and affinity maturation.
This has consequences for inventive step.
Generally, a claimed invention involves an inventive step if it’s not obvious to the skilled person having regard to the common general knowledge and prior art. Mere structural non-obviousness won’t be enough. So if a novel antibody binds the same antigen as known antibodies, an inventive step isn’t automatically acknowledged simply because the claimed antibody’s sequence is different. This is because the generation of antibodies by known methods is considered obvious to the skilled person.
Instead the claimed antibody must be shown to have a surprising effect or advantage – such as an improved affinity or activity, reduced toxicity or immunogenicity, or improved pharmacokinetics. However, if the advantage could be expected (for instance, reduced immunogenicity because of humanisation), it won’t be enough to establish an inventive step even if the claim is quite narrow.
However, the routine nature of antibody technology also has its benefits.
Because generating an antibody is routine, if the target or epitope is new then a broad claim to 'An antibody that specifically binds the new epitope' will be enabled and supported by minimal disclosure in the specification.
For antibodies directed to known antigens, both the structural elements and surprising effect will need to be shown by data for the claimed antibody to be fully supported and enabled. If an inventive step is based on an unexpected advantage, the independent claims will at least need to cover this advantage.
When assessing support and enablement, Australian examiners will look at the disclosures of the specification, including a priori reasoning, and consider whether a skilled person would find the claimed invention 'plausible' and could perform the invention across the whole breadth of the claims (having regard to the common general knowledge, and without undue burden or the need for further experimentation). This is a strict requirement and Australia has adopted a similar position to Europe. Because of this, it can be difficult to obtain broad antibody claims without significant data in the specification.
For antibodies with an improved property compared to prior art antibodies, the method for determining the property should be disclosed. If the advantage relates to binding affinity, it’s generally required to define the six complementarity-determining regions (CDRs) and the framework regions. Although neither the courts nor the Australian Patent Office (APO) has advised on the level of experimental data needed to enable a claim to an antibody (defined functionally, structurally or both), an obvious strategy is to include as much data as possible. Eg include binding assays which employ antibodies encompassed by the scope of the claims.
Below, we outline some of the main strategies to claim antibodies.
In a typical scenario where antibodies are raised against known targets, an antibody must usually be claimed by reference to its CDR sequences. A conventional antibody has six CDRs - three located on the variable domain of the light chain (VL) and three located on the variable domain of the heavy chain (VH). These are interspersed by framework regions.
Gone are the days it’s possible to define an antibody by a single heavy chain and light chain CDR, primarily CDR3, and argue antibody binding specificity is predominantly the result of CDR3 as well as the functionality of single domain antibodies. After the 'Raising the Bar' amendments, Australian examiners require that all six CDRs are defined to claim an antibody by its sequence. The Patent Manual recognises this is a relatively narrow claim. However, like in Europe, the default position is that all six CDRs are needed to provide specific binding to the target antigen. In some instances, it might be possible to avoid reciting all six CDRs. But this is only if the data shows one or more of the CDRs are not involved with antigen binding (ie do not interact with the target), or a particular antibody format allows epitope recognition by fewer CDRs (eg, heavy chain only antibodies).
A claim to an antibody that specifically binds epitope A and comprises 'one or more' of six CDRs is generally considered to lack enablement as it represents an undue burden. Meaning, a skilled person would need to generate all the antibodies that bind epitope A and then carry out the extra step of sequencing them to work out those within the claimed subset.
One way to pursue broader claims is to include a percentage sequence identity to the antibody’s structure, either alone or in combination with a functional limitation. Antibody claims defined by the combination of structural and functional features may allow for some sequence variability. An example claim format would be 'An antibody comprising SEQ ID NOs: 1 and 2 or a sequence having at least 90% sequence identity to SEQ ID NOs: 1 and 2, where the antibody is capable of binding to antigen X'. In Evolva SA  APO 57 (Evolva), the APO found claims to a polypeptide (enzyme) having at least 90% sequence homology limited by the structure-function relationship did not give rise to an exceedingly large number of polypeptides, and so was sufficiently enabled by the specification. Although Evolva relates to an enzyme, the APO has allowed claims which define antibodies having 90% sequence identity to heavy and light chain variable regions. It’s helpful to have disclosures in the specification to show that sequence variability doesn’t affect the structure-function relationship.
In principle, an antibody may be claimed exclusively by reference to its functional features such as binding affinity, neutralising properties, inhibition or activation of receptors. However, functional definitions in claims are likely to encounter novelty issues. This is because any prior art that discloses an antibody directed to the same antigen and displays similar functional properties is assumed to have these same properties (unless shown otherwise). Further, a claim defining an antibody solely by its functional features will likely be viewed as covering 'any and all' antibodies having those features - including antibodies the disclosure in the specification has not enabled, and would likely lead to support and enablement objections.
Where circumstances allow, claiming antibodies by reference to previously unknown targets (antigen or epitope) can be used to achieve broad claims. For example, 'An antibody that specifically binds epitope A'. This broad claim type is only allowed where the target is previously unknown. As noted already, under Australian practice, such a claim would be fully enabled if the specification discloses the epitope and shows that antibodies can be raised against it. This is because raising a genus of antibodies that bind to the epitope is considered 'a principle of general application' (can be performed without undue burden) and so is enabled over the whole scope of the claim. According to the Patent Manual, the specification only needs to disclose one such antibody.
The APO also allows claims which define an antibody by its ability to compete with a reference antibody. For example, 'An antibody that specifically binds to antigen X and competes with reference antibody Y for binding to antigen X'. For this approach, the specification needs to describe the methods used to determine competition and any relevant binding threshold. To avoid enablement issues, the scope of the claims should be supported by at least some exemplified antibodies.
Of course, there are many ways to claim an antibody. This can be by a combination of functional and structural features, by the hybridomas from which the antibodies are produced, the methods of producing the antibodies, by reference to post-translational modifications, and the use of the antibodies for the treatment of a disease or general classes of diseases.
With regard to claiming bispecific antibodies, chimeric antibodies, and antibody fragments (eg Fab, Fv, and scFv), the general issues outlined for conventional antibodies commonly apply. As with many other inventions, data and the common general knowledge at the time of filing will inform whether an applicant may obtain claims with the desired scope and features.
When it comes to the inventive step, objections commonly arise because there’s a lack of:
- evidence (eg data or information) in the specification that any functional property or advantage of the antibody is not a mere alternative;
- evidence showing the treatment of a disease or general classes of diseases by the claimed antibody is 'plausible' (this is particularly the case if the antibody is directed to a novel target; and/or
- evidence that competing antibodies bind the same epitope and therefore the claimed antibody is merely an alternative to the reference antibody.
Detailed submissions that set out a contrary view to the examiner with reference to the specification’s data might persuade the examiner that the claimed antibody is inventive. However, post filing data can also be used to show inventiveness - for example, to show favourable pharmacokinetics disclosed in the specification is available to all antibodies within the claims’ scope. It can also establish the claimed antibodies share a function (eg a particular binding affinity). More generally, post filing data may be used to demonstrate that the claimed antibodies provide an advantage over the antibodies of the prior art.
From a support and enablement perspective, an Australian examiner will often object to an antibody claim unless it’s defined by its six CDRs, regardless of whether the claim recites additional features. What’s more, claims relating to a use of an antibody (eg, 'when used', method of treatment or Swiss-style claims) to treat general classes of conditions and diseases like 'cancer' commonly attract scrutiny, and an Australian examiner will raise objections against these claims by default. Like an inventive step objection, submissions may be filed with reference to the disclosures in the specification (eg a priori reasoning), the common general knowledge, and post filing data. In Australia, post filing data can help establish support and enablement if the data confirms what was originally disclosed in the specification.