The patentability of mutated proteases has been considered by the Australian Patent Office in Novozymes A/S v Genencor International, Inc.  APO 15.
Australian Patent Application No. 2002353178 to Genencor International, Inc. concerned “Proteases producing an altered immunological response and methods of making the same” (“the Genencor Application”). The Opponent opposed the grant of the patent on the grounds of manner of manufacture, novelty, inventive step, utility, clarity and fair basis.
At acceptance, Claim 1 of the Genencor Application related to a variant protease having altered immunological properties in humans as a result of a substitution at any one or more of 103 residues. During the course of the Opposition, the list of residues was reduced to just seven (7) specific residues.
The Opponent identified two documents that listed many potential mutation sites, including the seven specific residues. While there was no doubt that the specific residues were mentioned in the prior art documents, it was the level of disclosure that was relevant to the question of novelty. Genencor argued that the documents were simply “shopping lists” of possible sites that represented “mere paper anticipations” and that there was no disclosure in either documents that would lead the skilled person to choose the specifically claimed residues.
The Hearing Officer found that the first citation, while disclosing many potential sites for mutation, failed to suggest that the relevant mutations had been made, nor that they represented any particularly preferred subset of the mutations disclosed. In particular, she found that the citation “does not provide an adequate disclosure of the specific variants to anticipate the opposed claims.”
The Hearing Officer also relied on an earlier Federal Court decision which stated that the Court, armed with the evidence of the skilled addressee as to terms of art and the nature and extent of the disclosure in the prior art document, must determine whether the prior disclosure is sufficient to enable the skilled addressee to perceive, understand and, where appropriate, apply the prior disclosure necessarily to obtain the invention. Accordingly, although the reverse infringement test is useful for determining novelty, it is not the only test and in this case, it was the lack of sufficiency of the disclosure in the prior art that was considered to be most important.
The second citation also included a “shopping list” of potential mutation sites. However, this document was not directed to altering the immunologic properties of a protease and did not refer to this feature. Therefore, the second citation did not disclose all the essential features of the Claims.
Inventive step can be difficult to determine, and in this case, the experts for the two parties identified different problems to be solved. However, the Hearing Officer decided that the invention was directed to not just mutants of proteases that produce an altered immunological response (the Opponent’s position), but rather, to “clinically relevant” mutants (the Applicant’s position). This was consistent with the requirement of the Claims that the altered immunological response was elicited “in a human”. To this end, the Hearing Officer turned to the evidence of the experts insofar as it related to human testing of mutant proteases.
Genencor used antibodies from sera obtained from humans sensitised to the protease of interest to identify short peptides, derived from the protease, which might contain B-cell epitopes. In contrast, the methodology described in the prior art relied on antibodies generated in rodents and rabbits. A significant amount of the expert evidence addressed the relative merits of the use of human versus animal sera to identify epitopes on proteins of interest. Both experts agreed that while rodents can be good models for predicting B-cell epitopes in humans, the rodent antibody response will not always exactly mirror the response in humans.
Thus, it is clear from the evidence that the person skilled in the art would understand that a B-cell epitope predicted in a rodent model will not always constitute a B-cell epitope in a human. If a B-cell epitope predicted in a rodent model is not recognised by the human immune system, substituting its amino acids to make it less immunogenic will have no effect on the immunological response that the variant epitope elicits in a human. Therefore, the person skilled in the art would understand that variants generated on the basis of a B-cell epitope predicted by a rodent model, will not necessarily produce an altered immunological response in humans.
The Opponent provided no submissions and led no evidence to establish that the person skilled in the art would proceed, as a matter of routine, from the rodent-identified B-cell epitopes to identify human-relevant epitopes, as is required by Claim 1.
Furthermore, the evidence did not establish that the person skilled in the art, having read the prior art, would have prepared the specific claimed variants as a matter of routine. Consequently, the Opponent did not establish that the opposed Claims lack an inventive step in light of this document.
The Opponent failed to prove any of its Grounds of Opposition.