DCT

1:21-cv-01807

Harbour Antibodies BV v. Teneobio Inc

Log In

I. Executive Summary and Procedural Information

  • Parties & Counsel:
  • Case Identification: 1:21-cv-01807, D. Del., 02/04/2025
  • Venue Allegations: Venue is alleged to be proper in the District of Delaware because both Defendants are incorporated in the State of Delaware and, therefore, reside in the judicial district.
  • Core Dispute: Plaintiffs allege that Defendants' UniRat® platform for antibody discovery, which uses transgenic rats, infringes four patents related to methods for producing fully human heavy chain-only antibodies in transgenic rodents.
  • Technical Context: The technology involves genetically engineering rodents to produce heavy chain-only antibodies (HCAbs), a class of smaller, simpler antibodies with potential advantages for developing complex therapeutic drugs.
  • Key Procedural History: The complaint alleges that Defendant Amgen previously entered into a license agreement with Plaintiff HBAB, effective in 2011, which covered patent applications that later issued as two of the patents-in-suit. Amgen allegedly terminated this agreement in November 2017. Separately, Plaintiffs allegedly notified Defendant Teneobio of potential infringement of two of the patents-in-suit in October 2017. Amgen acquired Teneobio in October 2021.

Case Timeline

Date Event
2004-07-22 Priority Date for all Asserted Patents
2011-12-08 Amgen enters license agreement with Plaintiff HBAB
~2015 Teneobio begins developing UniRat® platform
2016-05-24 U.S. Patent No. 9,346,877 Issues
2016-05-31 U.S. Patent No. 9,353,179 Issues
2017-10-06 Plaintiffs contact Teneobio regarding infringement of '877 and '179 patents
2017-11 Amgen terminates license agreement
2021-02-02 U.S. Patent No. 10,906,970 Issues
2021-05-04 U.S. Patent No. 10,993,420 Issues
2021-10-19 Amgen completes acquisition of Teneobio
2025-02-04 Complaint Filed

II. Technology and Patent(s)-in-Suit Analysis

U.S. Patent No. 9,346,877, "Binding Molecules", issued May 24, 2016

The Invention Explained

  • Problem Addressed: Conventional therapeutic antibodies are complex molecules composed of two heavy and two light protein chains (H2L2), making their manufacture complex and costly, particularly for "bi-specific" formats that target multiple antigens (U.S. Patent No. 10,906,970, col. 1:56-2:4, col. 2:23-30). While heavy chain-only antibodies (HCAbs) exist in nature (e.g., in camels), a method was needed to efficiently produce soluble, high-affinity human HCAbs suitable for therapeutic use (Compl. ¶47).
  • The Patented Solution: The invention provides a method to generate HCAbs in transgenic rodents. The rodent's genome is engineered to include a heterologous heavy chain locus that crucially lacks a functional CH1 domain—the part of the heavy chain that normally binds to the light chain (Compl. ¶71; '970 Patent, col. 3:51-54). When this transgenic rodent is challenged with an antigen, its immune system produces soluble antibodies consisting only of heavy chains. These antigen-specific HCAbs undergo natural affinity maturation within the rodent and can then be isolated for therapeutic development ('970 Patent, Abstract, Fig. 6). The complaint includes a diagram illustrating the structure of a heavy chain-only antibody, which consists of two heavy chains without the associated light chains typical of conventional antibodies (Compl. p. 10).
  • Technical Importance: This technology enables the in vivo generation of affinity-matured human HCAbs, which possess desirable biophysical properties like stability and solubility and can be more easily developed into next-generation therapeutics like bi-specific or multi-specific antibodies (Compl. ¶55).

Key Claims at a Glance

  • The complaint asserts infringement of at least independent claim 1 (Compl. ¶70).
  • Essential elements of Claim 1 include:
    • A method for producing a soluble, antigen-specific VH binding domain.
    • (a) Immunizing a transgenic rodent that expresses a heterologous VH heavy chain locus with specific genetic features: at least one VH, five D, and one J gene segment; a constant region that does not encode a functional CH1 domain; and where the locus is capable of VDJ recombination and forming a soluble, heavy chain-only antibody.
    • (b) Cloning the VH locus from an antibody-producing cell of the rodent after affinity maturation.
    • (c) Producing the soluble, VH binding domain from the cloned cell line. (Compl. ¶71).

U.S. Patent No. 9,353,179, "Binding Molecules", issued May 31, 2016

The Invention Explained

  • The technology disclosed in the U.S. Patent No. 9,353,179 ('179 Patent) is substantially the same as that of the '877 Patent, arising from the same family of patent applications and addressing the same technical problem (Compl. ¶¶ 32, 49-51).

Key Claims at a Glance

  • The complaint asserts infringement of at least independent claim 1 (Compl. ¶83).
  • Essential elements of Claim 1 include:
    • A method for producing soluble, antigen-specific heavy chain only antibodies.
    • (a) Immunizing a transgenic rodent with a heterologous VH locus containing the same genetic features as required by claim 1 of the '877 Patent.
    • (b) Cloning a soluble heavy chain only antibody from an antibody-producing cell of the rodent after affinity maturation.
    • (c) Producing the soluble, antigen-specific heavy chain only antibody from the clone. (Compl. ¶84).
  • A primary distinction between claim 1 of the '179 Patent and claim 1 of the '877 Patent is the final product: the '179 Patent claims a method of producing the full "heavy chain only antibody," which includes constant regions, whereas the '877 Patent claims a method of producing the "VH binding domain," which is the variable region responsible for antigen recognition.

U.S. Patent No. 10,906,970, "Methods of Making Heavy Chain Only Antibodies Using Transgenic Animals", issued February 2, 2021

  • Technology Synopsis: The '970 Patent describes a method for producing soluble, antigen-specific HCAbs by immunizing a transgenic rodent. The claims specify that the rodent's engineered VH heavy chain locus must comprise "from 20 to 40 D gene segments," a more specific requirement than the "at least five" D gene segments recited in the earlier '877 and '179 patents (Compl. ¶97; '970 Patent, Claim 1).
  • Asserted Claims: Independent claim 1 (Compl. ¶97).
  • Accused Features: The complaint alleges that the UniRat® platform utilizes a heavy chain locus made up of "two to 40 D gene segments," falling within the range claimed by the '970 Patent (Compl. ¶100(c)).

U.S. Patent No. 10,993,420, "Production of Heavy Chain Only Antibodies in Transgenic Mammals", issued May 4, 2021

  • Technology Synopsis: The U.S. Patent No. 10,993,420 ('420 Patent) claims a method for producing a VH heavy chain-only antibody in a transgenic rat or mouse. The method requires expressing a transgene comprising a heterologous human VH locus that includes specific VH3 gene segments (e.g., VH3-66 or VH3-9) and lacks all subclasses of human VH gene segments. A key element is that the mammal's endogenous heavy and light chain loci have been silenced to ensure production of only the desired HCAbs (Compl. ¶110).
  • Asserted Claims: Independent claim 1 (Compl. ¶110).
  • Accused Features: The accused UniRat® platform allegedly uses transgenic rats in which endogenous rat Ig expression has been silenced. The complaint further alleges, citing a scientific publication and a reproduced figure, that different UniRat® strains express the specific human VH3-9 and VH3-66 gene segments required by the claim (Compl. ¶¶ 113(g), 113(k); Compl. p. 42).

III. The Accused Instrumentality

  • Product Identification: The accused instrumentalities are Defendants' "UniRat®" antibody discovery platform and associated activities, services, and products derived from it, including UniAb™, UniDab™, and specific therapeutic candidates like TNB-383B (Compl. ¶70).
  • Functionality and Market Context: The UniRat® platform is described as a system that uses genetically engineered rats to generate "chimeric HCAbs with fully human VH domains" in response to an antigen challenge (Compl. ¶74(b)). The platform is based on immunizing these transgenic rats, which allegedly have an engineered immunoglobulin locus containing human variable gene segments linked to rat constant regions that lack the CH1 domain required for light chain pairing (Compl. ¶¶ 74(c)-(d)). The platform's commercial importance is highlighted by Amgen's acquisition of Teneobio for over $2.5 billion and the licensing of the platform to third parties for antibody production (Compl. ¶¶ 4, 61).

IV. Analysis of Infringement Allegations

U.S. Patent No. 9,346,877 Infringement Allegations

Claim Element (from Independent Claim 1) Alleged Infringing Functionality Complaint Citation Patent Citation
(a) immunising a transgenic rodent expressing a heterologous VH heavy chain locus with an antigen wherein: Defendants' platform is based on the immunization of "genetically engineered rats expressing heavy chain only antibodies." ¶74(b) col. 3:33-36
(i) the VH heavy chain locus comprises a variable region comprising at least one VH gene segment, at least five D gene segment... The accused platform's locus allegedly comprises "one or more human V gene segments" and "two to 40 D gene segments." ¶74(c) col. 9:48-51
(ii) each constant region does not encode a functional CH1 domain; The UniRat® platform allegedly uses rat constant regions with CH1 domains deleted and produces molecules that "lack CH1 domains." ¶74(d) col. 3:51-54
(iii) a VH gene segment, a D gene segment and a J gene segment are capable of recombining to form a VDJ coding sequence; The accused platform's V gene segments are described as being "capable of recombining with a D gene segment, a J gene segment... to generate a heavy chain-only antibody." ¶74(e) col. 3:30-33
(iv) the recombined VH heavy chain locus... is capable of forming a soluble, heavy chain-only antibody... The UniRat® platform allegedly "result[s] in production of high affinity antigen-specific heavy chain only antibodies." ¶74(a) col. 3:51-54
(b) cloning a VH locus resulting from recombination... from an antibody-producing cell of said immunised transgenic rodent after affinity maturation... The platform is alleged to involve a locus that, following recombination and "affinity maturation, encode[s] a heavy chain variable (VH) region" which can be cloned. ¶74(g) col. 4:14-19
(c) producing said soluble, VH binding domain from the clone of step (b). The resulting antibodies are allegedly produced via recombinant DNA technology by expressing the nucleic acid encoding the cloned locus. ¶74(h) col. 4:46-49

U.S. Patent No. 9,353,179 Infringement Allegations

Claim Element (from Independent Claim 1) Alleged Infringing Functionality Complaint Citation Patent Citation
(a) immunising a transgenic rodent expressing a heterologous VH heavy chain locus with an antigen wherein... [sub-elements i-iv are substantively identical to '877 patent] The allegations for step (a) and its sub-parts are identical to those asserted for the '877 Patent, as described in Section IV above. ¶¶ 87(b)-(f) col. 3:30-54
(b) cloning a soluble heavy chain only antibody from an antibody-producing cell of said immunised transgenic rodent after affinity maturation... The platform is alleged to involve a locus that encodes a heavy chain variable region after affinity maturation, and the resulting antibodies are cloned. ¶87(g) col. 4:14-19
(c) producing said soluble, antigen specific heavy chain only antibody from the clone of step (b). The final heavy chain-only antibodies are allegedly produced via recombinant DNA technology by expressing the nucleic acid encoding the cloned antibody. ¶87(h) col. 4:46-49

Identified Points of Contention

  • Scope Questions: A potential issue for claim construction may be the scope of "transgenic rodent." The patents-in-suit extensively describe work in mice, whereas the accused UniRat® platform utilizes rats (Compl. ¶74(a)). The litigation may address whether the claims, read in light of the specification, should be interpreted to cover both.
  • Technical Questions: Claim 1 of the '877 and '179 patents requires a locus with "at least five D gene segments." The complaint alleges the accused platform uses "two to 40 D gene segments" (Compl. ¶¶ 74(c), 87(c)). This raises an evidentiary question as to whether all commercial versions of the UniRat® platform necessarily meet the "at least five" limitation, or if some configurations may fall outside this claimed range.
  • A significant technical point is evidenced by a figure included in the complaint, which allegedly depicts the genetic maps for two UniRat® strains (HC27 and HC31) and details the specific human V gene segments used in each (Compl. p. 42). This visual evidence directly supports the allegation that Defendants' platform uses specific gene segments as required by claim 1 of the '420 patent (Compl. ¶113(g)).

V. Key Claim Terms for Construction

  • The Term: "functional CH1 domain" (from claim 1 of the '877, '179, and '970 patents)

    • Context and Importance: The absence of this domain is the core technical feature enabling the production of heavy chain-only antibodies. The definition of "functional" will be critical, as the dispute will turn on whether the accused platform's constant regions, which are "deleted for CH1," lack a "functional" CH1 domain as claimed (Compl. ¶74(d)).
    • Intrinsic Evidence for Interpretation:
      • Evidence for a Broader Interpretation: Language in the related '970 Patent specification states that the engineered heavy chain is "unable to associate with an immunoglobulin light chain," suggesting that "non-functional" means any CH1 domain that cannot perform this specific binding role ('970 Patent, col. 3:44-45).
      • Evidence for a Narrower Interpretation: The specification describes the CH1 region's role as facilitating interaction with the light chain ('970 Patent, col. 3:35-38). A party might argue that a modified CH1 region could be considered "functional" if it retains other biological activities (e.g., structural stability, regulatory signaling) even if its ability to bind a light chain is impaired.

  • The Term: "rodent" (from claim 1 of the '877, '179, and '970 patents)

    • Context and Importance: Practitioners may focus on this term because the patents' specifications heavily describe work in mice, while the accused platform uses rats. The question for the court will be whether the term should be limited by the examples in the specification.
    • Intrinsic Evidence for Interpretation:
      • Evidence for a Broader Interpretation: The plain and ordinary meaning of "rodent" is a well-understood scientific classification that includes both mice and rats. The claims use the general term "rodent" without limitation.
      • Evidence for a Narrower Interpretation: The specification of the related '970 Patent repeatedly refers to "transgenic mice" as the preferred embodiment ('970 Patent, col. 3:23-24, 3:34-35). A party could argue that the invention as described and enabled is limited to mice.

VI. Other Allegations

  • Indirect Infringement: The complaint alleges inducement on the basis that Defendants provide the UniRat® platform to customers and instruct them on how to use it to perform the patented methods (Compl. ¶¶ 75, 88). It further alleges contributory infringement, stating that the UniRat® platform (specifically, the transgenic rats) is a material component especially made for practicing the invention and is not a staple article of commerce (Compl. ¶¶ 77, 90).
  • Willful Infringement: Willfulness is alleged based on both pre-suit and post-suit knowledge. The complaint alleges Teneobio had actual knowledge of the '877 and '179 patents since at least October 6, 2017, via a direct notice letter (Compl. ¶78). It alleges Amgen had knowledge since 2016 based on a prior 2011 license agreement that identified the underlying patent applications (Compl. ¶¶ 67, 78). Knowledge of all patents is also alleged based on Amgen's acquisition of Teneobio in October 2021 (Compl. ¶78).

VII. Analyst’s Conclusion: Key Questions for the Case

  • A core issue will be one of definitional scope: How will key claim terms like "functional CH1 domain" and "transgenic rodent" be construed? The viability of the infringement claims depends on whether the accused UniRat® platform, which uses rats and constant regions "deleted for CH1," falls within the scope of patent claims developed using mice and claiming the absence of a "functional" domain.
  • A key evidentiary question will be one of quantitative infringement: The asserted patents recite specific numerical ranges for the gene segments in the engineered locus (e.g., "at least five D gene segments," "from 20 to 40 D gene segments"). The case may turn on factual evidence establishing the precise genetic makeup of every accused UniRat® configuration to determine if they meet these quantitative limitations.
  • A central issue for damages will be one of intent: Given the detailed allegations of a prior licensing history with Amgen and a specific notice letter to Teneobio, the court will have to determine whether Defendants' continued activities after acquiring knowledge of the patents were sufficiently egregious to constitute willful infringement and justify enhanced damages.