A new edition of the EPO Guidelines for Examination ('the guidelines') came into force on 1 March 2021. Relevant to life sciences, this edition includes a new subsection detailing EPO practice with respect to the interpretation of terms relating to amino or nucleic acid sequences, as well as a new section on the examination of claims to antibodies.

Terms in relation to amino or nucleic acid sequences (F-IV, 4.24)

In the field of biotechnology, claims often include nucleotide or amino acid sequences that are defined in terms of their percentage identity or percentage similarity (amino acid sequences only) to a specified reference sequence. The new guidelines now provide written guidance on how these different terms are interpreted.

Firstly, they confirm that when an amino acid or nucleic acid sequence is defined by using percentage sequence identity language, this is determined by the number of identical residues over a defined length in a given alignment. If no algorithm or calculation method for determining the percentage of identity is defined, the broadest interpretation will be applied using any reasonable method known at the relevant filing date.

In addition, it is confirmed that amino acid sequences can also be defined by a degree of similarity, expressed as a percentage of similarity. Similarity is considered broader than identity, as it allows conservative substitutions of amino acid residues having similar physicochemical properties over a defined length of a given alignment. The percentage of similarity is determinable only if a similarity-scoring matrix is defined. If no such matrix is defined, a claim referring to a sequence displaying a percentage of similarity to a recited sequence is considered to cover any sequence fulfilling the similarity requirement as determined with any reasonable similarity-scoring matrix known at the relevant filing date.

Finally, for amino acid sequences, the guidelines now state that if a percentage of homology is used by the applicant as the only feature to distinguish the subject matter of a claim from the prior art, its use is objected to under Article 84 EPC unless the determination or calculation of the percentage of homology is clearly defined in the application as filed. However, for nucleic acid sequences, homology percentage and identity percentage are usually considered to have the same meaning.

Antibodies (G-II, 5.6)

A new section, summarising EPO practice and case law on how an antibody may be defined (e.g. by sequence, functional feature, or epitope), and the inventive step requirement for new antibodies binding to a known antigen, has also been included in the 2021 guidelines. 

Claiming antibodies

They also outline how conventional antibodies, recombinant antibody derivatives (such as antibody fragments, bispecific antibodies or antibody fusions) or new antibody formats (such as heavy-chain-only antibodies) can be defined in a claim by reference to one or more of the following features:

(a) their own structure (amino acid sequences);
(b) nucleic acid sequences encoding the antibody;
(c) reference to the target antigen;
(d) target antigen and further functional features;
(e) functional and structural features;
(f) the production process;
(g) the epitope; and
(h) the hybridoma producing the antibody.

Definition by structure of the antibody

Conventional antibodies defined by their structure are to be defined by at least six complementary-determining regions (CDRs). CDRs, when not defined by their specific sequence, must be defined according to a numbering scheme, for example, chosen from that of Kabat, Chothia or IMGT. If the claim has fewer than six, it will be objected to under Article 84 EPC because it lacks an essential technical feature. An exception applies only if it is experimentally shown that one or more of the six CDRs do not interact with the target epitope or if it concerns a specific antibody format allowing for epitope recognition by fewer CDRs (such as the heavy-chain-only antibodies which have an antigen-binding region with only three CDRs).

Definition by reference to the target antigen

An antibody can be functionally defined by the antigen it binds to, as long as the antigen is clearly defined in the claims. If the antigen is defined by a protein sequence, no sequence variability and no open language (e.g. 'comprising') can be used in the definition of the antigen. Examples of accepted antigen-defined antibody claim wording are:

(i) antibody binding to X;
(ii) anti-X antibody;
(iii) antibody reacting with X;
(iv) antibody specific for antigen X; or
(v) antibody binding to antigen X consisting of the sequence defined by SEQ ID NO:Y.

Definition by target antigen and further functional features

Antibody claims can be further characterised by functional features defining further properties of the antibodies; for example, the binding affinity, neutralising properties, induction of apoptosis, internalisation of receptors, inhibition or activation of receptors. The burden of proof of novelty resides with the applicant and it has to be carefully assessed whether the application provides an enabling disclosure across the whole scope claimed and whether the functional definition allows the skilled person to clearly determine the limits of the claim.

Definition by functional and structural features

Antibodies can also be defined by both functional properties and structural features. It is possible to claim an antibody characterised by the sequences of both variable domains or CDRs with less than 100% sequence identity when combined with a clear functional feature.

Definition by the production process

Antibodies can be defined by the process of their production; for example, either by the immunisation protocol of a non-human animal with a well-characterised antigen or by the specific cell line used to produce them. However, if the process is based on an antigen comprising a sequence less than 100% identical to a defined sequence, it does not fulfil the requirements of Article 84 EPC because the use of variants renders the scope of the antibodies obtained by the immunisation process unclear.

Definition by the epitope

An antibody may be defined by its epitope. If the epitope is a 'linear epitope', whereby the antibody interacts with continuous amino acids on the antigen, it needs to be defined as a clearly limited fragment using closed wording, such as 'an epitope consisting of'. If the epitope is 'non-linear' or 'discontinuous', whereby the antibody interacts with multiple, distinct segments from the primary amino-acid sequence of the antigen, the specific amino acid residues of the epitope need to be clearly identified.

Inventive step

The new guidelines now state that a novel, further antibody binding to a known antigen does not involve an inventive step unless a surprising technical effect is shown by the application, such as an improved affinity, an improved therapeutic activity, a reduced toxicity or immunogenicity, an unexpected species cross-reactivity, or a new type of antibody format with proven binding activity.

If inventive step relies on an improved property in comparison to the antibodies of the prior art (which must be enabled), the main characteristics of the method for determining the property must also be indicated in the claim or by reference to the description. Notably, in the case of binding affinity, the structural requirements for conventional antibodies inherently reflecting this affinity must typically comprise the six CDRs and the framework regions because the framework regions also can influence the affinity.

An inventive step can also be acknowledged if the application overcomes technical difficulties in producing or manufacturing the claimed antibodies.

Originally Published by Mathys & Squire, March 2021

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.