PTAB

IPR2025-00605

Xencor Inc v. Merus NV

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Methods and Means for the Production of Ig-Like Molecules
  • Brief Description: The ’859 patent discloses heterodimeric antibodies engineered to improve the correct pairing of two different heavy chains. The technology uses "electrostatic steering" by introducing a positively charged amino acid residue at position 364 of a first human CH3 domain and a negatively charged amino acid residue at position 368 of a second human CH3 domain.

3. Grounds for Unpatentability

Petitioner’s primary contention is that the challenged claims are not entitled to their purported priority date of April 20, 2012. Petitioner argued that the specific combination of a mutation at position 364 on a first chain and a mutation at position 368 on a second chain was not disclosed in any of the priority applications. Therefore, the claims’ effective filing date is May 16, 2023, making several publications, including Desjarlais and Moore, valid prior art under 35 U.S.C. §102.

Ground 1: Anticipation of Claims 1-7 by Desjarlais

  • Prior Art Relied Upon: Desjarlais (Patent 10,472,427).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Desjarlais, which has an effective filing date years before the ’859 patent’s effective filing date, discloses every limitation of claims 1-7. Desjarlais’s Figures 4A and 4B explicitly disclose preferred heterodimerization variants for human IgG1 antibodies, including pairs with a positively charged lysine mutation at position 364 (S364K) on one chain and a negatively charged aspartic acid (L368D) or glutamic acid (L368E) mutation at position 368 on the other chain. This directly anticipates independent claim 1 and dependent claims 2 and 3. Desjarlais further teaches that these variants are used to generate bispecific antibodies (anticipating claim 4), are based on human IgG and IgG1 sequences (anticipating claims 5-6), and can be formulated in pharmaceutical compositions with acceptable carriers (anticipating claim 7).

Ground 2: Anticipation of Claims 1-7 by Moore

  • Prior Art Relied Upon: Moore (a peer-reviewed scientific article published in 2018).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner asserted that Moore, a publicly available scientific article, also anticipates all challenged claims. Moore describes a robust heterodimeric Fc platform (XmAb®) for developing bispecific antibodies. Figure 5 of Moore explicitly illustrates the creation of an “E357Q/S364K-L368D/K370S” variant from native human IgG1. This figure clearly shows a positively charged lysine at position 364 (S364K) on a first chain paired with a negatively charged aspartic acid at position 368 (L368D) on a second chain, meeting the limitations of claims 1-3. Moore also discloses that this platform is used to create various bispecific antibody formats (claim 4), is derived from native human IgG1 (claims 5-6), and its products are used as clinical candidates, which implies formulation in a pharmaceutical composition (claim 7).

Ground 3: Obviousness of Claims 1-7 over Lazar in view of Kannan

  • Prior Art Relied Upon: Lazar (Application # US 2011/0054151) and Kannan (WO 2009/089004).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that Lazar, a patent application publication, discloses using charge pairs at the CH3 domain interface to promote heterodimerization. Lazar’s Table 1 identifies a preferred S364E/L368K variant, which is a "reverse charge pair" (negative at 364, positive at 368) compared to the claimed invention. However, Lazar also discloses a list of potential substitutions for both positions, including positively charged options for 364 (364H, 364R) and negatively charged options for 368 (368E).
    • Motivation to Combine: A POSITA would combine Lazar’s disclosure with the teachings of Kannan. Kannan teaches the general principle of electrostatic steering by creating charge pairs and explicitly suggests "charge reversal" as a strategy for modifying residues at the CH3-CH3 interface to promote heterodimerization. A POSITA reading Lazar’s disclosure of a preferred S364/L368 charge pair would be motivated by Kannan’s teachings to simply swap the charges to arrive at the claimed configuration (e.g., S364K/L368E).
    • Expectation of Success: A POSITA would have a reasonable expectation of success. The modifications involve a finite number of known charged amino acids at specific, identified interface positions. Lazar demonstrates the successful creation of heterodimers using both charge pairs and reverse charge pairs at other positions, confirming the predictability of the technique. Kannan’s method of charge reversal at known interface pairs was a known and successful strategy.

4. Arguments Regarding Discretionary Denial

  • §325(d) - Same or Substantially the Same Art: Petitioner argued denial under §325(d) is inappropriate because the primary references for the anticipation grounds, Desjarlais and Moore, were never considered by the examiner during prosecution. While Lazar was cited on an IDS, its teachings were never substantively analyzed in an office action. The examiner’s prosecution of the ’859 patent family focused on double-patenting and overlooked key teachings from the prior art, such as Kannan's disclosure that its strategy could be extended to uncharged residues.
  • Fintiv Factors: Petitioner argued that the Fintiv factors favor institution. The parallel district court case is in a very early stage with no trial date set, and a stay upon institution is likely. The average time to trial in the venue (D. Del.) is approximately 32.2 months, making it probable that a final written decision would issue first. Investment in the parallel proceeding has been minimal, and there is not yet an overlap of invalidity contentions.

5. Relief Requested

  • Petitioner requests institution of an inter partes review and cancellation of claims 1-7 of the ’859 patent as unpatentable.