There are several ways to claim antibodies in patent applications: these range from purely functional definitions based on the antibody’s binding affinity, through to defining the complete heavy and light chain amino acid sequences of the antibodies.

Increasingly, the Patent Offices are requiring more structural (i.e. sequence) information in the patent claims as it becomes more recognised that small changes to the antibody’s sequence can have profound effects on its properties.

 

Antibodies defined by antigens

Traditionally, the European Patent Office (EPO) has granted claims of the following format, particularly if the protein antigen itself satisfies the criteria for patentability:

An antibody which binds specifically to protein X.

In such claims, the antibody is being defined indirectly, that is, by reference to the antigen to which it binds. Care needs to be taken, however, to ensure that such claims do not inherently cover known antibodies, particularly if the protein is part of a family of well-known proteins (e.g. GPCRs) and antibodies against other proteins are already known. In such circumstances, a claim of the following format should be considered:

An antibody which binds to protein X, but not to protein Y.

(where protein X is the novel protein and protein Y is a known one having epitopes in common with the novel protein.)

Antibodies defined by epitopes

In cases where the antigen to which the antibody binds is already known and some antibodies which bind to that antigen have already been publicly disclosed, a general claim to all antibodies to that antigen will lack novelty.

However, broad claims to antibodies that are directed to specific epitopes on that antigen might still be possible (assuming that the known antibodies are not directed to those epitopes).

Epitopes may be defined by reference to a specific monoclonal antibody which binds to that epitope or by reference to the continuous or discontinuous amino acid sequence of the epitope in the antigen.

Antibodies defined by sequence

With the advent of phage-display libraries and readily-available DNA sequencing methods, antibodies are now often defined by reference to specific amino acid or nucleic acid sequences of one or more of the CDRs which form the antigen-binding sites of the antibody.

If sufficient supporting data is available, variants of the CDRs may also be claimed in order to try to avoid the patent claims being limited merely to the precise sequences of the CDRs. Antibodies may also be defined by their complete VH and VL chain sequences, although patent claims based on such sequences are of rather narrow scope.

It should be noted, however, that if an antibody is known (for example, an antibody that is produced by a known hybridoma), then merely determining the sequence of that antibody will not render that antibody new − it is still the same chemical entity. However, specific fragments of that known antibody might still be patentable. Claiming antibodies in the form of an antibody-drug conjugate (ADC) is an additional way to achieve novelty over a previously-known (non-conjugated) antibody. Antibody patent applications will often also include claims to expression vectors comprising nucleic acid molecules which encode the new antibodies and host cells which produce such antibodies.