PTAB

IPR2017-01488

Pfizer Inc v. Genentech Inc

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Method for making humanized antibodies
  • Brief Description: The ’213 patent describes humanized antibodies and methods for making them by substituting specific amino acid residues in the framework regions (FR) of a human antibody with corresponding residues from a non-human (e.g., mouse) antibody. The goal is to reduce immunogenicity in human patients while maintaining the antibody's antigen-binding affinity.

3. Grounds for Unpatentability

Ground 1: Anticipation by Kurrle - Claims 1-2, 25, 29, 63, 66-67, 71-72, 75-76, and 80-81 are anticipated by Kurrle.

  • Prior Art Relied Upon: Kurrle (EP Publication Number 0403156).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued Kurrle, published in 1990, disclosed a detailed method for humanizing a mouse monoclonal antibody (BMA 031) that rendered it "essentially a human antibody with a much lower immunogenicity." Kurrle taught incorporating non-human Complementarity Determining Regions (CDRs) into a human framework and then substituting selected framework amino acids necessary for antigen binding back to the murine version. Petitioner asserted that Kurrle explicitly disclosed making several FR substitutions that fall within the Markush groups of the challenged claims, including at positions 4L and 69H, which anticipates independent claim 1. Dependent claims were met because Kurrle taught substituting the residue found at the corresponding location of the parent non-human antibody (claim 2) and created a full antibody with the humanized domain (claim 29). The property of reduced immunogenicity (claim 63) was an explicitly stated goal and result in Kurrle.

Ground 2: Anticipation by Queen 1990 - Claims 1-2, 4, 29, 62-64, and 80-81 are anticipated by Queen 1990.

  • Prior Art Relied Upon: Queen 1990 (International Publication No. WO 1990/07861).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner contended that Queen 1990 provided a universal roadmap for humanizing antibodies, including four explicit criteria for selecting FR residues for substitution. Criterion III of Queen 1990 specifically taught substituting residues in positions "immediately adjacent to one or more of the 3 CDR's" with the corresponding donor (mouse) amino acid. Using the standard Kabat numbering system, Petitioner argued that a person of ordinary skill in the art (POSITA) would have understood that this teaching inevitably led to substituting claimed residues such as 98L and 36H, thereby anticipating claim 1. Queen 1990 also taught using a "consensus framework" (anticipating claim 4) and that the resulting humanized variable domains would be incorporated into intact antibodies for therapeutic use (anticipating claim 29).

Ground 3: Obviousness over Queen 1990 and Kurrle - Claims 1-2, 4, 25, 29, 62-64, 66-67, 69, 71-72, 75-76, 78, and 80-81 are obvious over Queen 1990 and Kurrle.

  • Prior Art Relied Upon: Queen 1990 (WO 1990/07861) and Kurrle (EP Publication Number 0403156).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Queen 1990 provided the general principles and a detailed "roadmap" for humanization, while Kurrle provided a specific, successful example of applying a highly similar roadmap. The combination of Queen 1990's general framework with Kurrle's specific substitutions (e.g., 4L, 69H, 71H, 73H, 76H) disclosed all the residues required by the challenged independent claims.
    • Motivation to Combine: A POSITA would combine these references because they addressed the exact same problem of reducing antibody immunogenicity while preserving affinity. Queen 1990 provided the foundational strategy, and Kurrle, published less than six months later, provided a practical, successful application. Petitioner noted the references were cited in each other's patent family search reports, indicating their close technical relationship.
    • Expectation of Success: The combination provided a strong expectation of success because Queen 1990's explicit goal was to produce "substantially non-immunogenic" antibodies that "retain substantially the same affinity," and Kurrle demonstrated that a similar method achieved this precise result.

Ground 4: Obviousness over Queen 1990 and Hudziak (Humanizing the 4D5 Antibody) - Claims 30-31, 33, and 42 are obvious over Queen 1990 in view of Hudziak.

  • Prior Art Relied Upon: Queen 1990 (WO 1990/07861) and Hudziak (a 1989 journal article on the p185HER2 monoclonal antibody).

  • Core Argument for this Ground:

    • Prior Art Mapping: This ground focused on claims specific to an antibody that binds p185HER2. Hudziak identified p185HER2 as a prime therapeutic target for breast cancer and disclosed a specific high-affinity mouse antibody, 4D5, that inhibited tumor cell proliferation. Hudziak explicitly suggested that 4D5 should be humanized for chronic therapy. Queen 1990 provided the well-known, state-of-the-art method for performing such humanization.
    • Motivation to Combine: A POSITA would be motivated to apply the general humanization method of Queen 1990 to the specific, promising 4D5 antibody disclosed in Hudziak to create a viable cancer therapeutic. Hudziak's call to create "humanized versions" of 4D5 provided an explicit motivation.
    • Expectation of Success: There was a high expectation of success in humanizing 4D5 using Queen 1990's roadmap, as this was the standard and logical next step for developing any promising murine therapeutic antibody at the time.

  • Additional Grounds: Petitioner asserted additional obviousness challenges based on combinations including Queen 1990, Kurrle, Furey, Chothia & Lesk, and Chothia 1985. These grounds argued that a POSITA would have been motivated to consult additional prior art (Furey, Chothia) that identified other specific residues (e.g., 66L, 78H, 93H) as being structurally important for maintaining CDR conformation or antigen contact, making their substitution an obvious design choice when following the general humanization protocols of Queen 1990 and Kurrle.

4. Key Claim Construction Positions

  • "a humanized antibody" / "a humanized antibody variable domain": Petitioner argued these preamble phrases should be construed broadly to mean any antibody or fragment made more human-like, not a single specific antibody.
  • Markush Group of Residues: Petitioner contended that the recited lists of amino acid residues for substitution are a Markush group, meaning the members are functionally equivalent for the purpose of the invention. Therefore, the broadest reasonable interpretation is that any of the recited residues can be substituted in any given antibody to meet the claim limitation.
  • "numbering system set forth in Kabat": Petitioner asserted this encompasses the amino acid position numbering and the CDR/FR boundary designations from both the Kabat 1987 and Kabat 1991 database references cited in the patent.

5. Relief Requested

  • Petitioner requests institution of IPR and cancellation of claims 1-2, 4, 12, 25, 29-31, 33, 42, 60, 62-67, 69, and 71-81 as unpatentable under 35 U.S.C. §102 and/or §103.