Binding molecules for CD3 and their use

Abstract

To provide novel CD3 binding molecules, e.g., antibodies and multispecific binding molecules that bind to CD3.SOLUTION: The disclosure provides a CD3 binding molecule that specifically binds to CD3, e.g., monospecific binding molecules that binds to CD3 and a multispecific binding molecule (MBM) that binds to CD3 and a tumor-associated antigen, a conjugate comprising a CD3 binding molecule, and a pharmaceutical composition comprising a CD3 binding molecule and a conjugate. Further provided is a method for activating T cells using the CD3 binding molecule, conjugate and the pharmaceutical composition in a subject, e.g., a subject with cancer or autoimmune disease. The disclosure further provides a recombinant host cell modified to express CD3 binding molecule and a method for generating CD3 binding molecules by culturing the host cell under conditions allowing for expression of CD3 binding molecules.SELECTED DRAWING: None

Description

[Technical Field]

[0001] 1. Cross-reference of related applications This application is a PCT application filed on December 4, 2018, PCT / Chinese Patent Application Publication No. This patent application claims priority rights under specification No. 2018 / 119074, and this patent application The entire content is incorporated herein by reference.

[0002] 2. Sequence Listing This application includes a sequence listing submitted electronically in ASCII format, and this sequence listing is entirely This text is incorporated herein by reference. The ASCII copy was created on December 3, 2019. It was named NOV-010WO_SL.txt and had a size of 782,920 bytes. ru.

[0003] 3. Reference All publications, patents, patent applications and other documents cited in this application are the respective publications, patents, patent applications and other documents. Any individual publication, patent, patent application, or other document is invoked by reference for any purpose. To the same extent that it is shown individually, the whole thing is by reference for all purposes. This is incorporated herein by reference. Between one or more of the references incorporated herein by reference and the teachings of this disclosure In the event of any inconsistency, the teachings herein are intended. [Background technology]

[0004] The concepts of bispecific and multispecific antibodies imply that diseases are usually multifactorial, and multiple disease factors This approach addresses the issue, but stems from the idea that efficacy can be enhanced with a single antibody. Protogroup 3 (CD3) is a homozygous chromosome expressed on T cells in relation to the T cell receptor complex (TCR). It is a mer or heterodimer antigen and is necessary for T cell activation. Antibodies against CD3. This clusters CD3 on T cells, thereby loading MHC molecules with peptides. It has been shown that it induces T cell activation in a similar manner to the involvement of the TCR. CD3 antibodies have been proposed as a therapeutic agent involved in T cell activation. Anti-CD3 antibodies are It has been used to treat proliferative disorders such as cancer and autoimmune diseases. Furthermore, CD3 Furthermore, bispecific and multispecific antibodies that can bind to the target antigen express the target antigen. Therapeutic applications have been proposed that include targeting the T cell immune response against the tissues and cells involved. CD19 / CD3 BiTE and blinatumomab are among the currently approved drugs. Heavily specific antibodies are commercially available. However, bispecific and multispecific antibodies are not suitable for biological use. We still face challenges related to internal distribution, inhibitory microenvironment, and antigen loss. Therefore, In this field, there is a need for excellent bispecific and multispecific antibodies. CD3 and Bispecific and multispecific antigen-binding molecules that bind to both the target antigen and the target antigen express the target antigen. This would be useful in therapeutic environments where specific targeting of cells and T cell-mediated cell death are desired. . [Overview of the project] [Problems that the invention aims to solve]

[0005] New CD3-binding molecules are needed, such as antibodies that bind to CD3 and multispecific binding molecules. It is said that... [Means for solving the problem]

[0006] This disclosure relates to a CD3-binding molecule that specifically binds to human CD3, for example, a CD3-binding molecule that specifically binds to human CD3. This invention provides antibodies that bind to a target, antigen-binding fragments thereof, and multispecific molecules.

[0007] In one embodiment, the Disclosure relates to a CD3 antigen-binding domain or antigen-binding module ("AB"). Monospecific CD3-binding molecules containing "M" (e.g., antibodies and their antigen-binding fragments) This provides an exemplary CD3-binding molecule that may have single specificity, as described in Section 7.2 and below. This is described in Embodiments 1 to 456.

[0008] In another embodiment, the present disclosure relates to a multispecific binding molecule ("MBM") containing CD3 ABM. ), for example, to provide bispecific and multispecific antibodies. Therefore, in one embodiment, This disclosure includes at least two distinct antigen-binding domains or ABMs, and is bispecific and This disclosure relates to multispecific antibodies. In some embodiments, this disclosure relates to tumor-associated antigens ("TAA"). ) and other components of the TCR complex on CD3 and / or CD2 or T cells, The present invention provides hemispecific and multispecific binding molecules.

[0009] In certain embodiments, MBM is a bispecific binding molecule ("BBM"). M is the first ABM ("ABM1" or "CD3 ABM") that specifically binds to human CD3. A second ABM that specifically binds to the second antigen ("ABM2"), for example, human T Includes AA (sometimes referred to as "TAA ABM" in this specification). ABM1, AB The terms M2, CD3 ABM, and TAA ABM are used for convenience only, B It is not intended to show any particular form of BM. Using such a multispecific molecule This is used to induce CD3+ effector T cells to the TAA+ site, thereby enabling CD3+ Effector T cells can attack and lyse TAA+ cells and tumors. (Example) The features of MBM are described in sections 7.5 to 7.7 and in specific embodiments 457 to 536 below. It is.

[0010] In certain embodiments, MBM is a triple-specific binding molecule ("TBM"). M is the first ABM ("ABM1" or "CD3 AB") that specifically binds to human CD3. A second antigen, specifically binding to human TAA, is called ABM2. ) and a third antigen, for example, a second human TAA or human CD2, which specifically binds to the third antigen. Includes ABM ("ABM3"): (1) human CD3, (2) TAA, and (3) CD2 The TBM that binds to is, for convenience, referred to in this specification as "Type 1 TBM". (1) Human CD 3, (2) the first TAA (sometimes called "TAA1"), and (3) the second TA The TBM that binds to A (sometimes called "TAA2") is, for convenience, described herein in This is called "Type 2 TBM." Both TAA1 and TAA2 are tumor-associated antigens. The notation of the tumor-associated antigens in this disclosure as TAA1 and TAA2 is optional, and therefore Unless the context should indicate otherwise, any disclosure relating to TAA1 is applicable to TAA2. It is possible to use it, and vice versa.

[0011] In one embodiment, each antigen-binding module of the MBM binds to one or more further antigens. Each of the combined modules binds to its respective target and at the same time to its respective target It is possible to bind them together.

[0012] In MBM, each ABM (except ABM1, which is immunoglobulin-based) is immunoglobulin-based. MBM can be immunoglobulin-based or non-immunoglobulin-based, and therefore MBM is immunoglobulin-based. This may include phosphorus-based ABM, non-immunoglobulin-based ABM, or a combination thereof. Immunoglobulin-based ABMs that can be used in MBMs are described in Section 7.3.1 below. The specific embodiments described in 1 to 469 are non-immunoglobulins that may be used in MBM. Brin-based ABM is described in Section 7.3.2 and in specific embodiments 747-777 below. Further characteristics of exemplary ABMs that bind to components of the TCR complex are described below. Further characteristics of an exemplary ABM that binds to CD2 are described in Section 7.8 below. As described in Section 7.9 and specific embodiments 746-789 below, coupled to TAA Further features of the exemplary ABM are described in Section 7.10 and in specific embodiments 592~ This is described in sections 745 and 790-946.

[0013] The ABM of the MBM (or a portion thereof) is, for example, a short-chain peptide linker or an Fc domain. They can be connected to each other by the following. Method and configuration for connecting ABMs to form MBM The elements are described in Section 7.4 and in specific embodiments 947 to 1155 below.

[0014] MBM has at least two ABMs (i.e., MBM is at least divalent). A can have three or more ABMs. For example, MBM can have three ABMs (that is, (i.e., trivalent), 4 ABMs (i.e., tetravalent), 5 ABMs (i.e., pentavalent) It can have either (i.e., hexavalent) or six ABMs (i.e., hexavalent). One embodiment In this configuration, the MBM can be coupled to at least one ABM, CD3, which can be coupled to the TAA. At least one ABM that can bind and a few that can bind to another antigen It has at least one ABM. Exemplary divalent, trivalent, tetravalent, pentavalent, and hexavalent MBM forms are: As described in Sections 7.5 to 7.7 below and in specific embodiments 477 to 536 and 554 to 590. It is being done.

[0015] This disclosure relates to CD3-binding molecules (e.g., MBMs) (either a single nucleic acid or multiple nucleic acids). Nucleic acids encoding (in the context of) and nucleic acids and CD3-binding molecules (e.g., MBM) express Further provided are recombinant host cells and cell lines modified to do so. Exemplary nucleic acids, host Cells and cell lines are described in Section 7.11 and specific embodiments 1439-1441 below. It is being done.

[0016] This disclosure further provides drug conjugates containing CD3-binding molecules (e.g., MBMs). Yes. Such conjugates may have a non-immunoglobulin domain as part of the ABM. Nevertheless, for convenience, in this specification, the terms "antibody drug conjugate" or "ADC" will be used. This is called "[ADC]". Examples of ADCs are given in Section 7.12 and specific embodiments 1225-126 below. It is described in section 2.

[0017] Pharmaceutical compositions comprising CD3-binding molecules (e.g., MBM) and ADCs are also provided. Examples of finished products are described in Section 7.14 and in specific embodiment 1263 below.

[0018] For example, a CD3-binding molecule for treating a proliferative state in which TAA is expressed (e.g., cancer). Methods using (for example, MBM), ADCs and pharmaceutical compositions are further provided herein. Exemplary methods are described in Section 7.15 and in specific embodiments 1264-1437 below. It is listed.

[0019] This disclosure relates to CD3-binding molecules (e.g., MBM), ADCs, and pharmaceutical compositions to other drugs and Further methods of use in combination with treatments are provided. Exemplary drugs, treatments, and combinations. The treatment methods are described in Section 7.16 and in specific embodiment 1438 below. [Brief explanation of the drawing]

[0020] [Figure 1-1] Figure 1: Exemplary BBM morphology is shown. Figure 1A shows the components of the exemplary BBM morphology shown in Figures 1B-1AH. Not all regions linking the different domains of each strand are shown (for example, the linker linking the VH and VL domains of scFv, and the hinge linking the CH2 and CH3 domains of the Fc domain are omitted). Figures 1B-1F show the bivalent BBM; Figures 1G-1Z show the trivalent BBM; and Figures 1AA-1AH show the tetravalent BBM. In BBM, the variant CD58 domain can substitute for Fab and / or scFv in any of the shown morphologies. [Figure 1-2] (As stated above.) [Figure 1-3] (As stated above.) [Figure 1-4] (As stated above.) [Figure 1-5] (As stated above.) [Figure 1-6] (As stated above.) [Figure 1-7] (As stated above.) [Figure 2-1] Figure 2: Figures 2A-2V show exemplary TBM morphologies. Figure 2A shows the components of the exemplary TBM morphology shown in Figures 2B-2V. Not all regions linking the different domains of each strand are shown (for example, the linker linking the VH and VL domains of scFv, and the hinge linking the CH2 and CH3 domains of Fc are omitted). Figures 2B-2P show trivalent TBMs, Figures 2Q-2S show tetravalent TBMs, Figure 2T shows pentavalent TBMs, and Figures 2U-2V show hexavalent TBMs. In TBMs, the variant CD58 domain can substitute for Fab and / or scFv in any of the shown morphologies. [Figure 2-2] (As stated above.) [Figure 2-3] (As stated above.) [Figure 2-4] (As stated above.) [Figure 2-5] (As stated above.) [Figure 2-6] (As stated above.) [Figure 3-1] Figure 3: Exemplary MBM morphologies are shown. Figure 3A shows the normal IgG morphology, Figure 3B shows the BITE morphology, Figure 3C shows the bispecific morphology, Figure 3D shows the triplicate morphology, and Figure 3E shows the quadruple specific morphology. [Figure 3-2] (As stated above.) [Figure 3-3] (As stated above.) [Figure 3-4] (As stated above.) [Figure 3-5] (As stated above.) [Figure 4-1] Figure 4: Figures 4A-4D show surface plasmon resonance (SPR / Biacore) measurements, indicating the Kd of CD3. Figure 4A: NOV292; Figure 4B: sp34; Figure 4C: NOV123; Figure 4D: sp1c. [Figure 4-2] (As stated above.) [Figure 4-3] (As stated above.) [Figure 4-4] (As stated above.) [Figure 5] This shows the binding of an anti-CD3 antibody to cells transfected with human CD3. [Figure 6] This shows the binding of an anti-CD3 antibody to cells transfected with human CD3. [Figure 7] This shows the binding of an anti-CD3 antibody to cells transfected with human CD3. [Figure 8] This shows the binding of anti-CD3 antibodies to cells transfected with cynomolgus monkey (cyno) CD3. [Figure 9] This shows the binding of anti-CD3 antibodies to cells transfected with cynomolgus monkey (cyno) CD3. [Figure 10]This shows the binding of anti-CD3 antibodies to cells transfected with cynomolgus monkey (cyno) CD3. [Figure 11] This demonstrates CD3 agonist activation by a bispecific antibody in a JNL cell model. [Figure 12] This demonstrates CD3 agonist activation by a bispecific antibody in a JNL cell model. [Figure 13] This demonstrates CD3 agonist activation by a bispecific antibody in a JNL cell model. [Figure 14] This demonstrates CD3 agonist activation by a bispecific antibody in a JNL cell model. [Figure 15] This demonstrates CD3 agonist activation by a bispecific antibody in a JNL cell model. [Figure 16] This study demonstrates the ability of an anti-CD19 / anti-CD3 bispecific antibody to lyse target cells in a redirected T-cell cytotoxicity (RTCC) assay. [Figure 17] This study demonstrates the ability of an anti-CD19 / anti-CD3 bispecific antibody to lyse target cells in a redirected T-cell cytotoxicity (RTCC) assay. [Figure 18] This demonstrates the ability of anti-CD19 / anti-CD3 bispecific antibodies to lyse target cells in an RTCC assay. "OKT3" and "sp34" refer to the positive control bispecific antibody. [Figure 19] This shows the affinity of anti-CD19 / anti-CD3 bispecific antibodies against CD3+ T cells. "OKT3" and "sp34" refer to the positive control bispecific antibodies. [Figure 20] This demonstrates the ability of anti-CD19 / anti-CD3 bispecific antibodies to lyse target cells in an RTCC assay. "OKT3," "sp34," and "H2C" refer to the positive control bispecific antibodies. [Figure 21] This demonstrates the ability of anti-CD19 / anti-CD3 bispecific antibodies to lyse target cells in an RTCC assay. "OKT3," "sp34," and "H2C" refer to the positive control bispecific antibodies. [Figure 22]This demonstrates the ability of anti-CD19 / anti-CD3 bispecific antibodies to lyse target cells in an RTCC assay. "OKT3" and "sp34" refer to the positive control bispecific antibody. [Figure 23] This shows the affinity of anti-CD19 / anti-CD3 bispecific antibodies against CD3+ T cells. "OKT3" and "sp34" refer to the positive control bispecific antibodies. [Modes for carrying out the invention]

[0021] 7.1.Definition As used herein, the following terms are intended to have the following meanings: .

[0022] Antigen-binding module: When used in this specification, "antigen-binding module" or " The term "ABM" refers to the ability to bind to an antigen non-covalently, reversibly, and specifically. This refers to the portion of the MBM of this disclosure. ABM refers to immunoglobulin-based or non-immunoglobulin-based It may be based on. When used in this specification, "ABM1" and "CD3 AB The term "M" (etc.) refers to ABM that specifically binds to CD3, and "ABM2" and The term "TAA ABM" (etc.) refers to ABM that specifically binds to tumor-associated antigens. The terms ABM1 and ABM2 are used for convenience only and do not apply to any specific MBM. It is not intended to show a fixed form.

[0023] Antibody: As used herein, the term "antibody" means non-covalently, reversibly polypeptides of the immunoglobulin family capable of binding to antigens precisely and specifically. It refers to (or a set of polypeptides). For example, a natural IgG type "antibody" is disulfide. It comprises at least two heavy (H) chains and two light (L) chains interconnected by deuterated bonds. It is a tetramer. Each heavy chain consists of a heavy chain variable region (abbreviated as VH herein) and a heavy chain It consists of a constant region. The heavy chain constant region consists of three domains: CH1, CH2, and CH3 It consists of a light chain variable region (abbreviated as VL in this specification) and a light chain. It consists of a chain constant region. The light chain constant region is one domain (CL and in this specification) It consists of (abbreviated). The VH and VL areas are called the framework area (FR). Between these are more conservative regions, scattered throughout, a highly variable region called the complementarity-determining region (CDR). The regions can be further subdivided. Each VH and VL is in the following order: FR1, CDR1, FR2 , arranged from the amino terminus to the carboxyl terminus at CDR2, FR3, CDR3, FR4 It consists of three CDRs and four FRs. The variable regions of the heavy and light chains interact with the antigen. It contains a binding domain for use. The constant region of the antibody is used for various cells of the immune system (e.g., EFF). Injector cells and the first component of the classical complement system (Clq) are present in host tissue or factors. It can mediate the binding of immunoglobulins. The term "antibody" is not limited to mono Clonal antibodies, human antibodies, humanized antibodies, camelized antibodies, chimeric antibodies, bispecific or polyspecific antibodies Heavy specific antibodies and anti-idiotype (anti-Id) antibodies (e.g., anti-Id against the antibodies of this disclosure) d) contains antibodies. Antibodies can be any isotype / class (e.g., IgG, IgE). , IgM, IgD, IgA and IgY) or subclasses (e.g., IgG1, IgG2, It may be IgG3, IgG4, IgA1, or IgA2.

[0024] Both light and heavy chains can be divided into regions of structural and functional homology. "Stationary" and The term "variable" is used functionally. In this regard, light (VL) and heavy (VH) It will be understood that both variable domains in the chain portion determine antigen recognition and specificity. Conversely, the constant domains of the light chain (CL) and heavy chain (CH1, CH2, or CH3) are secreted. It confers important biological characteristics such as transplacental mobility, Fc receptor binding, and complement binding. Therefore, the numbering of constant region domains is based on the number of domains more distal to the antigen-binding site or amino terminus of the antibody. The number increases as it progresses. The N-terminus is a variable region, and the C-terminus is a constant region; The CH3 and CL domains actually contain the carboxyl termini of the heavy and light chains, respectively.

[0025] Antibody fragment: In this specification, the term "antibody fragment" refers to an antibody. The term refers to one or more parts of an antibody. In some embodiments, these parts are antibodies. These are part of the contact domain. In some other embodiments, these parts are described herein In this context, "antigen-binding fragment," "that antigen-binding fragment," and "antigen-binding portion" The ability to bind to an antigen noncovalently, reversibly, and specifically, sometimes referred to as [a specific term]. It is an antigen-binding fragment that is retained. Examples of binding fragments are not limited to those mentioned above. However, single-stranded Fv (scFv), Fab fragment, VL, VH, CL and CH1 domain A monovalent fragment consisting of ; F(ab)2 fragment, with disulfide in the hinge region. A divalent fragment containing two Fab fragments linked by a dysbridge; VH and Fd fragment consisting of CH1 domains; VL and VH domains of a single arm of the antibody Fv fragment consisting of; dAb fragment consisting of VH domain (Ward et al., (1989) Nature 341:544-546); and isolated complementary sex-determining regions (CDRs). Thus, the term "antibody fragment" encompasses proteolytic fragments of an antibody (e.g., Fab and F(ab)2 fragments) as well as

[0026] modified proteins that include one or more portions of an antibody (e.g., scFv). Antibody fragments can be incorporated into single domain antibodies, maxibodies, minibodies, intrabodies, bispecific antibodies, trispecific antibodies, tetraspecific antibodies, v-NAR, and bis-scFv (see, e.g., Hollinger and Hudson, 2005, Nature Biotechnology 23:1126-1136). Antibody fragments can be grafted into

[0027] a scaffold based on polypeptides such as fibronectin type III (Fn3) (see U.S. Patent No. 6,703,199, which describes fibronectin polypeptide monobodies). Antibody fragments can be incorporated into single-chain molecules that include pairs of tandem Fv segments (e.g., VH-CH1-VH-CH1)

[0028] that together with complementary light chain polypeptides (e.g., VL-VC-VL-VC) form pairs of antigen-binding , a moiety that non-covalently, reversibly and specifically binds to an antigen, comprising an antigen-binding fragment and a scaffold portion based on both immunoglobulins and non-immunoglobulins. When used herein, the term "antigen-binding domain" encompasses antibody fragments that non-covalently, reversibly and specifically bind to an antigen.

[0029] Half-antibody: The term "half-antibody" refers to a molecule that contains at least one ABM or ABM chain and can associate with another molecule containing an ABM or ABM chain, for example, by disulfide bridging or molecular interactions (e.g., knob-in-hole interactions between Fc heterodimers). A half-antibody can be composed of one polypeptide chain or two or more polypeptide chains (e.g., the two polypeptide chains of Fab). In a preferred embodiment, the half-antibody contains an Fc region.

[0030] Examples of half-antibodies are molecules that contain the heavy and light chains of an antibody (e.g., an IgG antibody). Another example of a half-antibody is a molecule that contains a first polypeptide comprising a VL domain and a CL domain and a second polypeptide comprising a VH domain, a CH1 domain, a hinge region, a CH2 domain and a CH3 domain, where the VL and VH domains form an ABM. Yet another example of a half-antibody is a polypeptide that contains an scFv domain, a CH2 domain and a CH3 domain.

[0031] A half-antibody can contain two or more ABMs. For example, a half-antibody can contain (in order from the N-terminus to the C-terminus) an scFv domain, a CH2 domain, a CH3 domain and another scFv domain.

[0032] A half-antibody forms a complete ABM when it associates with another ABM chain in another half-antibody. It may also include BM chains.

[0033] Therefore, MBM may contain one, more typically two, or even three or more semiantibodies. The semi-antibody may contain one or more ABMs or ABM chains.

[0034] In some MBMs, the first half-antibody associates with the second half-antibody, for example together Heterodimerization occurs. In other MBMs, the first half-antibody is, for example, a disulfide crosslink. Alternatively, it is covalently bound to the second half-antibody by chemical crosslinking. Furthermore, in other MBMs, The half-antibody of 1 exhibits both covalent and non-covalent interactions, such as disulfide crosslinking and knobs. It associates with the second half-antibody through in-hole interaction.

[0035] The term "half-antibody" is intended solely for explanatory purposes and does not refer to a specific form or production. It does not imply the method of production. "First" half-antibody, "second" half-antibody, "left side" half-antibody, The explanation of half-antibodies, such as "right-side half-antibodies," is for convenience and explanatory purposes only.

[0036] Complementarity Determination Region: As used herein, "complementarity determination region" or "CDR" The term refers to the sequence of amino acids within the antibody variable region that confers antigen specificity and binding affinity. This refers to the three CDRs (e.g., CDR-H1, CDR-H1, CDR-H1) within each heavy chain variable region. DR-H2 and CDR-H3) and three CDRs (CDR-L1, There are CDR-L2 and CDR-L3. The precise amino acid sequence boundary of a given CDR is K abat et al., 1991, “Sequences of Proteins of Immunological Interest”, 5th Ed. Public Health Service, National Institutes of H ealth, Bethesda, MD (the "Kabat" numbering scheme), Al-La zikani et al., 1997, JMB 273:927-948 (the "Chot hia" numbering scheme) and the Immunogenetic (IMGT) numbering (L efranc, 1999, The Immunologist 7:132-136 (1 999); Lefranc et al., 2003, Dev. Comp. Immuno l. 27:55-77 (the "IMGT" numbering scheme), including those described by any of several well-known schemes. For example, in the classical form by the Kabat method, the CDR amino acid residues in the heavy chain variable domain (VH) are numbered 31-3 5 (CDR-H1), 50-65 (CDR-H2) and 95-102 (CDR-H3); and the CDR amino acid residues in the light chain variable domain (VL) are numbered 24-34 (CDR -L1), 50-56 (CDR-L2) and 89-97 (CDR-L3). By the Chothia method, the CDR amino acids in VH are numbered 26-32 (CDR-H1), 52-56 (CDR-H2) and 95-102 (CDR-H3); and the amino acid residues in VL are numbered 26-32 (CDR-L1), 50-52 (CDR-L2) and 91- 96 (CDR-L3). By combining both the Kabat and Chothia CDR definitions, the CDRs are amino acid residues 26-35 (CD ) in human VH, 50-65 (CDR-H2) and 95-102 (CDR-H3), as well as in human ) in human VH, 50-65 (CDR-H2) and 95-102 (CDR-H3), as well as in human R-H1), 50-65 (CDR-H2) and 95-102 (CDR-H3) and in human VL amino acid residues 26-32 (CDR-L1), 50-52 (CDR-L2) and 91- Amino acid residues 24-34 (CDR-L1), 50-56 (CDR-L2), and 8 in VL It consists of 9-97 (CDR-L3). According to the IMGT method, the CDR amino acid residues in VH are Approximately 26-35 (CDR-H1), 51-57 (CDR-H2), and 93-102 (CDR (-H3) is numbered, and the CDR amino acid residues in the VL are approximately 27-32 (CDR-L1). They are numbered as 50-52 (CDR-L2) and 89-97 (CDR-L3) ("K (Numbering according to "abat"). By the IMGT method, the CDR region of the antibody is programmed IM This can be determined using GT / Domain Gap Align.

[0037] Single-stranded Fv or scFv: As used herein, "single-stranded Fv" or "sc The term "Fv" refers to an antibody fragment containing the VH and VL domains of the antibody, and here These domains are present within a single polypeptide chain. Preferably, Fv polypeptide Butide allows scFv to form the desired structure for antigen binding, VH and It further includes polypeptide linkers between VL domains. For an overview of scFv, see P lueckthun in The Pharmacology of Monoclo nal Antibodies,vol.113,Rosenburg and Moo re eds., (1994) Springer-Verlag, New York, p. Please refer to pages 269-315.

[0038] Bispecific antibody: As used herein, the term "bispecific antibody" refers to: Typically, small antigens with two antigen-binding sites are formed by the pairing of scFv chains. Refers to a body fragment. Each scFv contains a light chain variable domain (VL) within the same polypeptide chain. ) contains a heavy chain variable domain (VH) linked to VL (VH-VL, where VH is linked to VL). In contrast, VH and VL are either N-terminus or C-terminus. A linker that enables VH and VL to pair up and form an antigen-binding domain Unlike typical scFv antibodies, bispecific antibodies typically have VH and on the same chain. It includes a linker that is too short to enable matching between VL domains, and thereby, VH and The VL domain is paired with the complementary domain of another chain, forming two antigen-binding sites. Bispecific antibodies are, for example, described in European Patent No. 404,097; International Publication No. 93 / 1 Pamphlet No. 1161; and Hollinger et al., 1993, Proc. Further details can be found in .Natl.Acad.Sci.USA 90:6444-6448. It is being done.

[0039] Fv: The term "Fv" refers to an immunoglobulin that includes a complete target recognition and binding site. This refers to the smallest possible antibody fragment from which it may originate. This region consists of one strongly non-covalently bound heavy It consists of a chain and a dimer of one light chain variable domain (VH-VL dimer). Then, the three CDRs in each variable domain interact and target the surface of the VH-VL dimer. The binding site is defined. In most cases, six CDRs confer target binding specificity to the antibody. However, in some cases, a single variable domain (or only three target-specific CDRs) Even half of Fv (including Fv) may have the ability to recognize and bind to targets. The reference to the VH-VL dimer is not intended to indicate any particular form. Without limitation, VH and VL may be combined in any form as described herein. So, can they form half-antibodies, or do they exist on separate half-antibodies, and separate half-antibodies When antibodies associate, they come together to form an antigen-binding domain, for example, the MBM of this disclosure It can be formed. When present on a single polypeptide chain (e.g., scFv), VH can be located at the N-terminus or C-terminus relative to VL.

[0040] Multiple specificity binding molecules: The terms "multispecificity binding molecules" or "MBM" are used at least as This also refers to a molecule containing two antigen-binding domains, where at least one of the antigen-binding domains One of them is CD3, and at least one antigen-binding domain is specific to TAA. Each of the combined domains independently forms an antibody fragment (e.g., scFv, Fab, nano). Body), ligand or non-antibody-derived binder (e.g., fibronectin, finomer, DARPin) is a possible example. Representative MBMs are shown in Figures 3A-3E. One MBM is , may contain two, three, four or more polypeptide chains.

[0041] VH: The term "VH" is not limited to Fv, scFv, dsFv, or Fab. This refers to the variable region of the antibody's immunoglobulin heavy chain, which includes the heavy chain.

[0042] VL: The term "VL" is not limited to Fv, scFv, dsFv, or Fab. This refers to the variable region of the immunoglobulin light chain, including the light chain itself.

[0043] Operablely linked: The term "operable linked" refers to two or more peptides. Alternatively, it refers to a functional relationship between polypeptide domains or nucleic acid (e.g., DNA) segments. The term "operatably linked" is used in relation to fusion proteins or other polypeptides. The word is formed by linking two or more amino acid segments to produce a functional polypeptide. This means, for example, a separate ABM (or a chain of ABMs) in relation to the MBM of this disclosure. This may be via a peptide linker sequence. Fusion proteins such as the polypeptide chain of the MBM of this disclosure In relation to nucleic acids that code for a substance, "operably linked" means that two nucleic acids are two The amino acid sequence encoded by the nucleic acid is joined in such a way that it remains within the frame. This means that, in relation to transcriptional regulation, this term refers to the function of the transcriptional regulatory sequence on the transcriptional sequence. This refers to a relationship between a promoter or enhancer sequence and the appropriate host cell. It can act on the coding sequence when stimulating or modulating the transcription of the coding sequence in other expression systems. It is connected to.

[0044] Associated: In relation to MBM, the term “associated” refers to two or more polypeptides This refers to the functional relationship between two or more polypeptides. In particular, the term "associated" refers to the functional relationship between two or more polypeptides. This produces functional MBMs such as ABM1 and ABM2 that can bind to their respective targets. To achieve this, for example, by molecular interactions, non-covalently or by one or more disulfides This means that they are covalently associated with each other by crosslinking or chemical crosslinking. Examples of possible meetings within an MBM include (but are not limited to) Fc within an Fc domain. Inter-regional associations (homodimer associations or more preferably as described in Section 7.4.1.5) (heterodimer association), association between VH and VL regions in Fab or Fv, and in Fab Meetings between CH1 and CL are one example.

[0045] ABM chain: Each ABM is a single polypeptide chain (for example, in the case of scFv) It may exist, or by association of two or more polypeptide chains (for example, in the case of Fab). It may be formed. As used herein, the term "ABM chain" refers to a single poly This refers to all or part of the ABM present on the peptide chain. The use of the term "ABM chain" is... These are intended solely for convenience and explanatory purposes and do not imply any particular form or method of manufacture. I haven't done that.

[0046] Host cells or recombinant host cells: The term "host cells" or "recombinant host cells" is used in this context. For example, it refers to cells that have been genetically modified by introducing heterologous nucleic acids. Such terminology is It is understood that this is intended to refer not only to specific target cells, but also to the offspring of such cells. It should be explained that, due to either mutation or environmental influence, a certain modification may occur in later generations. Because they can exist in a certain state, such offspring may not be identical to the parent cell. However, it still falls within the scope of the term "host cell" as used herein. It occurs. Host cells are temporarily expressed on, for example, extrachromosomal xenoexpression vectors or, for example, host cells. By integrating heterologous nucleic acids into the cellular genome, it is possible to stably retain heterologous nucleic acids. The MBM of this disclosure For the purpose of expression, the host cells are preferably monkey kidney cells (COS, e.g., COS -1, COS-7), HEK293, baby hamster kidney (BHK, e.g., BHK21) ), Chinese hamster ovary (CHO), NSO, PerC6, BSC-1, human liver Cellular cancer cells (e.g., Hep G2), SP2 / 0, HeLa, Maidin Darby Bovine kidney (MDBK), myeloma and lymphoma cells or their derivatives and / or modifications These are cell lines of mammalian origin or mammalian-like characteristics, such as mutants. Examples of modified mutants include: Examples include glycan-profile modified and / or site-specific integration site derivatives. ru.

[0047] Sequence Identity: Percent The term "identity" refers to the sequence identity of two or more nucleic acids or polypeptides. A column refers to two or more identical arrays. Two arrays are compared using the following array comparison algorithm. When measured using one of the rhythms or by manual alignment and visual inspection The comparison and alignment are performed for the greatest match on the comparison window or in the specified area. In such cases, the specified percentage of the same amino acid residue or nucleotide (for example, 60% identity across a specified area or, if not specified, across the entire sequence, optional selection Selectively 70%, 71%, 72%, 73%, 74%, 75%, 76%, 77%, 78%, 7 9%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 8 9%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 9 If two sequences have 9% identity, they are "substantially identical." Optionally, Identity is determined by at least approximately 50 nucleotides (or the site of a peptide or polypeptide). In total, over a region of length of at least about 10 amino acids, or more preferably 100 ~500 or 1000 nucleotides or more (or 20, 50, 20 It exists over a region of length (0 or more amino acids).

[0048] In sequence comparison, typically one sequence acts as a reference sequence, and the other sequences are compared to it. The test sequences are compared. When using a sequence comparison algorithm, the test sequence and the reference sequence are compared. The input is entered into the computer, the coordinates of the subarray are specified, and the array algorithm program is applied as needed. Gram parameters are specified. Default program parameters may be used, or Alternative parameters can be specified. Next, the array comparison algorithm is program parameters Based on the data, the sequence identity percentage of the test sequence compared to the reference sequence is calculated. The method for aligning the sequences is well known in the art. Optimal alignment is, for example, in Smith and Waterman, 1970, A. According to the local homology algorithm of dv.Appl.Math.2:482c, Need leman and Wunsch, 1970, J. Mol. Biol. 48:443. The homology alignment algorithm, Pearson and Lipman,19 Searching for similarities between 88, Proc.Nat'l.Acad.Sci.USA 85:2444 These algorithms (the Wisconsin Genetics) are derived from the search method. GAP, BESTFIT, FASTA, and TF in the Software Package ASTA(Genetics Computer Group,575 Science By computer execution by Dr. Madison, WI) or by manual Tests and visual examinations (e.g., Brent et al., 2003, Curren) (See Protocols in Molecular Biology.) This can be done by [method].

[0049] Two examples of algorithms suitable for determining sequence identity and sequence similarity percentages are: These are the BLAST and BLAST 2.0 algorithms, which are, respectively, Alts chul et al.,1977,Nuc.Acids Res.25:3389-3 402; and Altschul et al., 1990, J. Mol. Biol. 21 It is described in 5:403-410. The software for performing BLAST analysis is A National Center for Biotechnology Information (USA) It is publicly available through the Ecology Information Service.

[0050] The percentage of identity between two amino acid sequences is, according to Meyers and Miller. The algorithm from 1988, Comput.Appl.Biosci.4:11-17 was used. It can also be determined by this algorithm, which uses the PAM120 residue weight table and 12 gaps. Using the length penalty and the 4-gap penalty, the ALIGN program (vers It is incorporated into ion 2.0). Furthermore, the identity percentage between two amino acid sequences The source is Needleman and Wunsch, 1970, J. Mol. Biol. 48:444-453) This can be determined using the algorithm, and this algorithm is , either the Blossom 62 matrix or the PAM250 matrix and 1 Gap weights of 6, 14, 12, 10, 8, 6 or 4 and lengths of 1, 2, 3, 4, 5 or 6 The weighting is incorporated into the GAP program in the GCG software package. .

[0051] Conservative Sequence Modification: The term "conservative sequence modification" refers to MBM or its components (for example, Amino acids that do not substantially affect or alter the binding properties (of the ABM or Fc region) This refers to modifications. Such conservative modifications include amino acid substitutions, additions, and deletions. Modifications are part Site-specific mutagenesis and PCR-mediated mutagenesis, and other known standards in the art. Conservative amino acid substitutions can be introduced into the MBM of this disclosure by conventional technology. Conservative amino acid substitutions are amino acid residues This is a substitution with an amino acid residue having a similar side chain. The families of acid residues are defined in the art. These are amino acids with basic side chains (e.g., lysine, arginine, histidine), and acidic side chains. amino acids with chains (e.g., aspartic acid, glutamic acid), having non-charged side chains amino acids (for example, glycine, asparagine, glutamine, serine, threonine, tyrosine) (Syn, cysteine, tryptophan), amino acids with nonpolar side chains (e.g., alanine) (Valine, leucine, isoleucine, proline, phenylalanine, methionine), β- Amino acids with branched side chains (e.g., threonine, valine, isoleucine) and aromatic side amino acids with chains (e.g., tyrosine, phenylalanine, tryptophan, histidine) (n) is an example. Therefore, one or more amino acid residues in the MBM of this disclosure are on the same side Modified MBMs can be substituted with other amino acid residues from the chain family, for example, target Testing for molecular binding and / or effective heterodimerization and / or effector function. It is possible.

[0052] Mutation or modification: In relation to polypeptides used herein, "sudden mutation" The terms "mutation" and "modification" may include the substitution, addition, or deletion of one or more amino acids. .

[0053] Antibody numbering method: In this specification, numbered amino acid residues in the antibody domain References to the EU are based on the EU numbering system unless otherwise specified. This system is based on the EU delman et al.,1969,Proc.Nat'l Acad.Sci.U First devised by SA 63:78-85, and Kabat et al., 1991 ,in Sequences of Proteins of Immunologic al Interest, US Department of Health and Details are provided in Human Services, NIH, USA.

[0054] dsFv: The term "dsFv" refers to a disulfide-stabilized Fv fragment. In dsFv, VH and VL are linked by interdomain disulfide bonds. To generate molecules like this, one amino acid in the VH and VL framework regions Each of these is mutated into cysteine, which then forms a stable interchain disulfide bond. Typically, the 44th position in VH and the 100th position in VL are mutated to cysteine. Brinkmann,2010,Antibody Engineering 181- See 189, DOI:10.1007 / 978-3-642-01147-4_14. I would like to know. The term dsFv is known in the art as dsFv(VH and (VL is a molecule linked by interchain disulfide bonds rather than linker peptides.) scdsFv(VH and VL are linked by a linker and interchain disulfide bonds) It includes both (the molecules that have been extracted).

[0055] VH domain tandem: When used herein, "VH domain (or VH The term "tandem" refers to a series of VH domains consisting of multiples of the same VH domain in an antibody. It refers to the in. Each of the VH domains is a linker, except for the last one at the end of the tandem. - Has its C-terminus ligated to the N-terminus of another VH domain, either present or absent. Tandem It has at least two VH domains, and in certain embodiments of the MBM of this disclosure, 3 It has 4, 5, 6, 7, 8, 9, or 10 VH domains. A tandem VH is A pair with or without a linker that allows them to be made as a single polypeptide chain. Using a rearrangement method (for example, as described in Section 7.4.3), each V in the desired order It can be produced by binding H-domain coding nucleic acids. The first VH-domain in tandem The main N-terminus is defined as the N-terminus of the tandem, while the last VH domain of the tandem The C-terminus of the yin molecule is defined as the C-terminus of the tandem molecule.

[0056] VL domain tandem: When used herein, "VL domain (or VL The term "tandem" refers to a series of VL domains consisting of multiples of the same VL domain of an antibody. It refers to the in. Each of the VL domains is a linker, except for the last one at the end of the tandem. - Has its C-terminus ligated to the N-terminus of another VL, either with or without. Tandems are less common. It has at least two VL domains, and in certain embodiments of the MBM of this disclosure, 3, 4, and 5 , has 6, 7, 8, 9 or 10 VL domains. VL tandems are those that are single Recombination with or without linkers that allows for the creation of polypeptide chains. Using the method (for example, as described in Section 7.4.3), each VL domain in the desired order It can be generated by binding the coding nucleic acid of the tandem. The first VL domain of the tandem The N-terminus is defined as the N-terminus of the tandem, while the C-terminus of the last VL domain of the tandem. The terminus is defined as the C-terminus of the tandem.

[0057] Monovalent: The term "monovalent" as used herein in relation to antigen-binding molecules This refers to an antigen-binding molecule that has a single antigen-binding domain.

[0058] Divalent: The term "divalent" as used herein in relation to antigen-binding molecules This refers to an MBM having two antigen-binding domains, where one antigen-binding domain is C It is D3. The antigen-binding domain may be the same or different. Therefore, the bivalent antigen binding The compound molecule may be single-specific or bispecific. An example of the divalent MBM of this disclosure is shown in Figure 3C. It will be shown in general terms.

[0059] Trivalent: The term "trivalent" as used herein in relation to antigen-binding molecules This refers to an antigen-binding molecule that has three antigen-binding domains. Trivalent MBMs include CD3, TAA and binds specifically to another antigen. The trivalent MBM of this disclosure binds to different antigens respectively. It has at least three antigen-binding domains. An example of the trivalent MBM of this disclosure is shown in Figure 3D. It is shown in a simplified form.

[0060] Tetravalent: As used herein in relation to MBM, the term "tetravalent" means four This refers to an antigen-binding molecule having an antigen-binding domain. The MBM in this disclosure is tetravalent and CD3 , specifically binds to TAA and at least one other antigen. The tetravalent MBM of this disclosure is Generally, it has two antigen-binding domains that bind to the same antigen (preferably TAA) and CD3. It has at least one antigen-binding domain. An example of a tetravalent MBM of this disclosure is shown in Figure 3E. This is roughly shown below.

[0061] Pentavalent: As used herein in relation to MBM, the term "pentavalent" means five This refers to an antigen-binding molecule having an antigen-binding domain. The MBM of this disclosure is quintuplicate. It specifically binds to CD3, TAA, and three other antigens. Therefore, the pentavalent M of this disclosure BM generally consists of (a) two pairs of antigen-binding domains that bind to the same antigen and a third (b) A single antigen-binding domain that binds to an antigen, or (b) three antigen-binding domains that bind to the same antigen. It has either a domain or two antigen-binding domains that bind to different antigens, respectively.

[0062] Hexavalent: As used herein in relation to MBM, the term "hexavalent" means six This refers to an antigen-binding molecule having an antigen-binding domain. The MBM in this disclosure includes CD3, TAA and It specifically binds to at least one other antigen. The hexavalent MBM of this disclosure is generally... Each has three pairs of antigen-binding domains that bind to the same antigen, but they have different forms (for example, T Three antigen-binding domains that bind to AA and at least one antigen-binding domain that binds to CD3 Three antigen-binding domains that bind to the domain or TAA and at least one that binds to CD3 The two antigen-binding domains are within the scope of this disclosure.

[0063] Specifically (or selectively) binds to an antigen or epitope. The term "binding" refers to the binding of alloantigens or other antigens in a heterogeneous population of proteins and other biomolecules. Refers to a binding reaction that determines the presence of epitopes. The binding reaction can be mediated by an antibody or antibody fragment, but it does not have to be mediated. For example, it can be mediated by any type of ABM described in Section 7.3, such as a ligand, DARPin, etc. The ABMs of the present disclosure also typically have a dissociation rate constant (KD) of less than 5×10−9 M, less than 10−9 M, less than 5×10−10 M, less than 10−10 M, less than 5×10−11 M, less than 10−11 M, less than 5×10−12 M, less than 10−12 M, less than 5×10−13 M, less than 10−13 M, less than 5×10−14 M, less than 10−14 M, less than 5×10−15 M, less than 10−15 M, less than 5×10−16 M, less than 10−16 M, and bind to the target antigen with an affinity at least twice higher than its affinity for binding to a non-specific antigen (e.g., HSA). The term "specifically binds" does not exclude cross-reactivity between different species. For example, an antigen-binding module (e.g., an antigen-binding fragment of an antibody) that "specifically binds" to an antigen from one species may also "specifically bind" to the corresponding antigen in one or more other species. Thus, such cross-reactivity between different species itself does not change the classification of the antigen-binding module as a "specific" binder. In certain embodiments, the antigen-binding modules of the present disclosure that specifically bind to human antigens (e.g., ABM1, ABM2, etc.) bind to one or more non-human mammalian species, such as primate species (e.g., cynomolgus). It can be mediated by, but does not have to be mediated by, for example, a ligand, DARPin, etc., any type of ABM described in Section 7.3. The ABMs of the present disclosure also typically have a dissociation rate constant (KD) of less than 5×10−9 M, less than 10−9 M, less than 5×10−10 M, less than 10−10 M, less than 5×10−11 M, less than 10−11 M, less than 5×10−12 M, less than 10−12 M, less than 5×10−13 M, less than 10−13 M, less than 5×10−14 M, less than 10−14 M, less than 5×10−15 M, less than 10−15 M, less than 5×10−16 M, less than 10−16 M, and bind to the target antigen with an affinity at least twice higher than its affinity for binding to a non-specific antigen (e.g., HSA). The term "specifically binds" does not exclude cross-reactivity between different species. For example, an antigen-binding module (e.g., an antigen-binding fragment of an antibody) that "specifically binds" to an antigen from one species may also "specifically bind" to the corresponding antigen in one or more other species. Thus, such cross-reactivity between different species itself does not change the classification of the antigen-binding module as a "specific" binder. In certain embodiments, the antigen-binding modules of the present disclosure that specifically bind to human antigens (e.g., ABM1, ABM2, etc.) bind to one or more non-human mammalian species, such as primate species (e.g., cynomolgus). It can be mediated by any type of ABM described in Section 7.3, such as a ligand, DARPin, etc. The ABMs of the present disclosure also typically have a dissociation rate constant (KD) of less than 5×10−9 M, less than 10−9 M, less than 5×10−10 M, less than 10−10 M, less than 5×10−11 M, less than 10−11 M, less than 5×10−12 M, less than 10−12 M, less than 5×10−13 M, less than 10−1.. Typically, 5×10 , , , , , , , , -5 , -9 , -4 , -8 , -3 , -7 , -2 , -6 , , , , , , , -6 , , -5 , -4 , -3 ,<000001,M less than, 10 -2 M less than, 5×10 -3 M less than, 10 -3 M less than 5×10 -4 M less than, 10 -4 M less than, 5×10<000,M less than, 10 -5 M less than, 5× 10 -6 M less than, 10 -6 M less than, 5×10 -7 M less than, 10 -7 M less than, 5×10 -8 [[ID=, M less than, 10 -8 M less than, 5×10 -9 M less than or 10 -9 M less than dissociation rate constant (KD )(koff / kon) and binds to the target antigen with an affinity at least twice higher than its affinity for binding to a non-specific antigen (e.g., HSA). The term "specifically binds" does not exclude cross-reactivity between different species. For example, an antigen-binding module (e.g., an antigen-binding fragment of an antibody) that "specifically binds" to an antigen from one species may also "specifically bind" to the corresponding antigen in one or more other species. Thus, such cross-reactivity between different species itself does not change the classification of the antigen-binding module as a "specific" binder. In certain embodiments, the antigen-binding modules of the present disclosure that specifically bind to human antigens (e.g., ABM1, ABM2, etc.) bind to one or more non-human mammalian species, such as primate species (e.g., cynomolgus). The term "specifically binds" does not exclude cross-reactivity between different species. For example, an antigen-binding module (e.g., an antigen-binding fragment of an antibody) that "specifically binds" to an antigen from one species may also "specifically bind" to the corresponding antigen in one or more other species. Thus, such cross-reactivity between different species itself does not change the classification of the antigen-binding module as a "specific" binder. In certain embodiments, the antigen-binding modules of the present disclosure that specifically bind to human antigens (e.g., ABM1, ABM2, etc.) bind to one or more non-human mammalian species, such as primate species (e.g., cynomolgus). The term "specifically binds" does not exclude cross-reactivity between different species. For example, an antigen-binding module (e.g., an antigen-binding fragment of an antibody) that "specifically binds" to an antigen from one species may also "specifically bind" to the corresponding antigen in one or more other species. Thus, such cross-reactivity between different species itself does not change the classification of the antigen-binding module as a "specific" binder. In certain embodiments, the antigen-binding modules of the present disclosure that specifically bind to human antigens (e.g., ABM1, ABM2, etc.) bind to one or more non-human mammalian species, such as primate species (e.g., cynomolgus). An antigen-binding module (e.g., an antigen-binding fragment of an antibody) that "specifically binds" to an antigen from one species may also "specifically bind" to the corresponding antigen in one or more other species. Thus, such cross-reactivity between different species itself does not change the classification of the antigen-binding module as a "specific" binder. In certain embodiments, the antigen-binding modules of the present disclosure that specifically bind to human antigens (e.g., ABM1, ABM2, etc.) bind to one or more non-human mammalian species, such as primate species (e.g., cynomolgus). Thus, such cross-reactivity between different species itself does not change the classification of the antigen-binding module as a "specific" binder. In certain embodiments, the antigen-binding modules of the present disclosure that specifically bind to human antigens (e.g., ABM1, ABM2, etc.) bind to one or more non-human mammalian species, such as primate species (e.g., cynomolgus). This cross-reactivity itself does not change the classification of the antigen-binding module as a "specific" binder. In certain embodiments, the antigen-binding modules of the present disclosure that specifically bind to human antigens (e.g., ABM1, ABM2, etc.) bind to one or more non-human mammalian species, such as primate species (e.g., cynomolgus). In certain embodiments, the antigen-binding modules of the present disclosure that specifically bind to human antigens (e.g., ABM1, ABM2, etc.) bind to one or more non-human mammalian species, such as primate species (e.g., cynomolgus). ABM1, ABM2, etc.) bind to one or more non-human mammalian species, such as primate species (e.g., cynomolgus). Monkey (Macaca fascicularis), Rhesus macaque (Macaca mul) This includes one or more of the atta and the pig-tailed macaque (Macaca nemestrina). (and not limited to these) or rodent species, such as the house mouse (Mus muscula) It has interspecies cross-reactivity with (us). In other embodiments, the antigen-binding module of the present disclosure (For example, ABM1, ABM2, etc.) do not exhibit interspecies cross-reactivity.

[0064] Monoclonal antibody: As used herein, "monoclonal antibody" refers to The terminology includes antibodies, antibody fragments, molecules (including MBMs) that originate from the same genetic source. This refers to the polypeptide that is included.

[0065] Humanization: The term "humanized" form of non-human (e.g., mouse) antibodies refers to non-human immunoglobulins. It is a chimeric antibody containing the smallest sequence derived from robulin. In most cases, humanized antibodies are Residues derived from the hypervariable region of the recipient possess the desired specificity, affinity, and ability. Due to the hypervariable region of non-human species (donor antibodies) such as rats, rabbits, or non-human primates It is a human immunoglobulin (recipient antibody) in which the residues have been substituted. The framework region (FR) residues of human immunoglobulins are replaced with corresponding non-human residues. Furthermore, humanized antibodies contain residues not found in recipient antibodies or donor antibodies. These modifications are made to further improve the performance of antibodies. Generally, humanization The antibody contains virtually all of at least one, typically two, variable domains, where, All or virtually all of the hypervariable loops correspond to those of non-human immunoglobulins, and FR All or substantially all are human immunoglobulin lo sequences. Humanized antibodies are, Selectively, at least a portion of the constant region (Fc) of immunoglobulins, typically human immunoglobulins. This includes Brin's work. For further details, see Jones et al., 1986. Nature 321:522-525;Riechmann et al.,1988 Nature 332:323-329; and Presta, 1992, Curr.O See p. Struct. Biol. 2: 593-596. See the following review and its Reference cited within: Vaswani and Hamilton, 1998, An n.Allergy,Asthma&Immunol.1:105-115;Harri s,1995,Biochem.Soc.Transactions 23:1035- 1038;Hurle and Gross,1994,Curr.Op.Biotec See also h.5:428-433.

[0066] Human antibodies: As used herein, the term "human antibodies" refers to a framework. The antibody contains both the ku and CDR regions, which are variable regions derived from human sequences. Furthermore, if the antibody contains a constant region, the constant region is such a human sequence, for example, a human reproductive system Human germline sequences of serial sequences or mutants, or, for example, Knappik et al. As described in 2000, J Mol Biol 296, 57-86, human Also derived from the antibody-containing consensus framework sequence derived from framework sequence analysis. The structure and location of immunoglobulin variable domains, such as CDRs, are known numbering systems. A scheme, for example, the Kabat numbering scheme, the Chothia numbering scheme, or Kab It can be defined using a combination of at and Chothia (for example, Lazikani e t al.,1997,J.Mol.Bio.273:927 948;Kabat e et al., 1991, Sequences of Proteins of Immu. nological interest,5th edit.,NIH Publica tion no.91-3242 USDepartment of Health and Human Services;Chothia et al.,1987, J.Mol.Biol.196:901-917;Chothia et al.,19 See 89, Nature 342:877-883).

[0067] Human antibodies are amino acid residues not encoded by the human sequence (for example, in vitro Introduced by random or site-directed mutations or in vivo somatic mutations. This may include mutations or conservative substitutions to improve stability or production. However, as used herein, the term "human antibody" refers to antibodies from another animal such as a mouse. CDR sequences derived from the germline of mammalian species are grafted onto human framework sequences. It is not intended to contain antibodies.

[0068] Chimeric antibody: The term "chimeric antibody" (or its antigen-binding fragment) is (a) The constant region or a portion thereof, or the antigen-binding site (variable region), is different or modified. The effector function and / or the entire constant region or chimeric antibody to which new properties are imparted. Modified to link to different molecules, such as enzymes, toxins, hormones, growth factors, drugs, etc. (b) The variable region or part thereof is different or altered Antibody molecules (or their antigenic molecules) that have been modified, substituted, or replaced by a variable region having antigen specificity. (composite fragment). For example, mouse antibodies have a constant region that is a composite fragment of human immunoglobulin. It can be modified by substitution in the steady-state region. By substitution with the human steady-state region, The Mela antibody, compared to the original mouse antibody, exhibits reduced antigenicity in humans, while also being an anti- When recognizing the original, its specificity can be preserved.

[0069] Effector function: The term "effector function" usually refers to the combination of effector molecules. mediated by binding via the antibody's domain other than the antigen-binding domain. This refers to the activity of the antibody molecule being mediated. Effector function is, for example, the C1 complement configuration of the antibody. It includes complement-mediated effector functions mediated by the binding of elements. Complement activation is detailed It is important in the opsonization and lysis of cellular pathogens. Complement activation also stimulates the inflammatory response. Furthermore, it may be involved in autoimmune hypersensitivity. The effector function is Fc receptor (FcR) mediated. It also includes an effector function, which involves the binding of the antibody's constant domain to the Fc receptor (FcR). This can sometimes be caused. The binding of antibodies to Fc receptors on the cell surface is coated with the antibody. Phagocytosis and destruction of particles, clearance of immune complexes, and antibody coating by killer cells Lysis of target cells (called antibody-dependent cell-mediated cytotoxicity or ADCC), inflammatory media Many important and diverse functions, including the release of dermatophytes, placental cross-transit, and the control of immunoglobulin production. It triggers a biological response. The effector function of antibodies is such as Fc receptors or complement components. Modify the affinity of the antibody to the effector molecule, for example, by increasing or decreasing it. It can be changed. Binding affinity can generally be altered by modifying the effector molecule binding site. The part is modified, and in this case, the target part is located, and at least a part of this part is suitable Modification by method is appropriate. Modification of the binding site in antibodies for effector molecules. The change does not need to substantially alter the overall coupling affinity, but the effector mechanism is unproductive. It is also conceivable that the geometric properties of the interaction could be modified to invalidate it, as in the case of bonding. It is possible. The effector function also does not directly involve the effector molecular bonding, but the effect It can be modified by modifying parts that are involved in the performance of the ter function in other ways. It is possible to consider other possibilities.

[0070] To recognize: As used herein, the term “recognize” means “to recognize” This refers to ABMs that find a P and interact with it (for example, by binding).

[0071] Epitope: An epitope or antigenic determinant is an antibody or other antigenic antibody as described herein. This is the portion of the antigen recognized by the synthetic domain. Epitopes are linear or three-dimensional structures. It is possible.

[0072] Nucleic acid: The term "nucleic acid" is used herein to be synonymous with the term "polynucleotide." Used in such applications, it contains either single-stranded or double-stranded deoxyribonucleotides or ribonucleotides. This term refers to nucleotides and their polymers. This term can refer to synthetic, natural, and unnatural nucleotides. A known nucleotide that has similar binding properties to a reference nucleic acid and is metabolized in the same way as a reference nucleotide. This includes nucleic acids containing ocidal analogs or modified skeletal residues or ligatures. Examples, though not limited to them, include phosphorothioates, phosphoramidates, and methylphosphonates. Phosphorates, chiral methylphosphonates, 2-O-methylribonucleotides, peptides - Nucleic acid (PNA) is one example.

[0073] Unless otherwise specified, a particular nucleic acid sequence refers not only to the explicitly stated sequence, but also to the nucleic acid itself. This implicitly includes conserved variants of sequences (e.g., degenerate codon substitutions) and complementary sequences. Specifically, as detailed below, degenerate codon substitution involves one or more selected ( The third position of all codons is replaced with a mixed base and / or a deoxyinosine residue. This can be achieved by generating sequences (Batzer et al., (1991)). Nucleic Acid Res.19:5081;Ohtsuka et al., (1985) J. Biol. Chem. 260:2605-2608; and Rossol ini et al.,(1994)Mol.Cell.Probes 8:91-98 ).

[0074] Vector: The term "vector" refers to a device that transports another polynucleotide ligated to it. It is intended to refer to polynucleotide molecules capable of doing so. One type of vector - is a "plasmid," which is a DNA segment that contains further DNA segments. This refers to a circular double-stranded DNA loop that can be used. Another type of vector is a viral vector. Yes, and here, further DNA segments can be ligated into the viral genome. Certain vectors are capable of self-replicating within the host cells into which they are introduced (for example, bacteria). Bacterial vectors having a replication origin and episome mammalian vector. Other vectors (for example) When a non-episome mammalian vector is introduced into a host cell, it integrates into the host cell's genome. It can be incorporated and thereby replicated along the host genome. Furthermore, certain vectors can It is possible to induce the expression of genes that are operablely linked to these vectors. In this specification, the term "recombinant expression vector" (or simply "expression vector") is used. It will be discovered. Generally, expression vectors useful in recombinant DNA technology are plasmid-type. This is often the case. Plasmids are the most commonly used form of vectors, therefore In this specification, the terms "plasmid" and "vector" may be used synonymously. However, However, this disclosure also describes viral vectors that perform equivalent functions (e.g., replication-deficient retroviral vectors). Other forms of expression vectors such as adenoviruses and adeno-associated viruses. It is intended to include

[0075] Associated sequences: Refer to the table (including its sub-sections) for the term "associated sequences" as applicable. This refers to an ABM containing a set of CDRs, VH-VL pairs, or scFv as listed in the table.

[0076] VH-VL or VH-VL pair: Refer to the VH-VL pair, on the same polypeptide chain or on different polypeptide chains. Regardless of which polypeptide chain it is, the terms "VH-VL" and "VH-VL pair" For convenience, unless the context should indicate otherwise, it does not indicate any particular orientation. This is not intended. Therefore, scF containing "VH-VL" or "VH-VL pair" v is, for example, VH and VL in any orientation from the N-terminus of VH to VL or from the N-terminus of VL to VH. It may have a VL domain.

[0077] Polypeptides and proteins: The terms "polypeptide" and "protein" are used in a broader sense. This term is used synonymously herein to refer to polymers of amino acid residues. An amino acid is an artificial chemical mimetic of a corresponding natural amino acid, consisting of one or more amino acid residues. This also applies to polymers, including natural amino acid polymers and non-natural amino acid polymers. Unless otherwise indicated, a given polypeptide sequence implicitly includes its conservedly modified variants. do.

[0078] Subject: The term "subject" includes humans and non-human animals. Non-human animals include all vertebrae. Animals, such as non-human primates, sheep, dogs, cattle, chickens, amphibians, and reptiles, have a large mouth. Includes dairy and non-mammalian animals. Unless otherwise specified, the terms “patient” or “subject” are used. These terms are used synonymously in this specification.

[0079] Cancer: The term "cancer" refers to the uncontrolled (often rapid) growth of abnormal cells. This refers to a disease characterized by [specific characteristics]. Cancer cells spread throughout the body locally or via the bloodstream and lymphatic system. It can spread to other parts of the body. Examples of various cancers are described herein, but are not limited to: , breast cancer, prostate cancer, ovarian cancer, cervical cancer, skin cancer, pancreatic cancer, colorectal cancer, kidney cancer, liver cancer, brain cancer Cancer, adrenal cancer, autonomic ganglion cancer, biliary tract cancer, bone cancer, endometrial cancer, eye cancer, fallopian tube cancer, reproductive organ cancer, colorectal cancer Cancer, meningeal cancer, esophageal cancer, peritoneal cancer, pituitary cancer, penile cancer, placental cancer, pleural cancer, salivary gland cancer, small intestine cancer , gastric cancer, testicular cancer, thymic cancer, thyroid cancer, upper aerodigestive cancer, urinary tract cancer, vaginal cancer, vulvar cancer, lymphoma Examples include any TAA-positive cancer of any of the aforementioned types, such as leukemia and lung cancer. ru.

[0080] Tumor: The term "tumor" is used herein to be synonymous with the term "cancer." For example, both terms encompass solids and liquids, such as diffuse or circulating tumors. When used in this document, the terms "cancer" or "tumor" refer to pre-malignant and malignant cancers and Includes tumors.

[0081] Tumor-associated antigens: The term "tumor-associated antigens" or "TAA" refers to the surface of cancer cells. It is expressed as a whole or as fragments (e.g., MHC / peptide) and against cancer cells. Molecules useful for preferential targeting of drugs (typically proteins, carbohydrates, and lipids) This refers to (or some combination thereof). In one embodiment, TAA is normal cells and Markers expressed by both cancer cells and markers, such as cell line markers, such as those expressed by B cells. This is CD19. In one embodiment, TAA is found in cancer cells compared to normal cells. Overexpression (for example, 1x overexpression, 2x overexpression, 3x overexpression compared to normal cells) The above is an overexpressed cell surface molecule. In one embodiment, TAA is expressed in cancer cells, for example Molecules that are unsuitable compared to molecules expressed in normal cells, including those with deletions, additions, or mutations. It is a cell surface molecule that is synthesized. In one embodiment, TAA is a cell surface molecule of cancer cells. Expressed either entirely or as fragments (e.g., MHC / peptide) on a single plane. , it is not synthesized or expressed on the surface of normal cells. Therefore, the term "TAA" is used. In the art, cancer is sometimes known as tumor-specific antigen ("TSA") It contains antigens specific to the cell.

[0082] To treat, to cure, the act of treating: When used herein, "to treat" and "to cure" The terms “to treat” and “to treat” are derived from the administration of one or more MBMs of this disclosure. , reduction or improvement in the progression, severity and / or duration of a proliferative disorder, or 1 This refers to the improvement of one or more symptoms (preferably one or more identifiable symptoms). In a particular embodiment, In this context, the terms "to treat," "treatment," and "the act of treating" do not necessarily apply to the patient. At least one measurable physical pattern of a proliferative disorder, such as tumor growth, that is not identified This refers to improvement of lameter. Other mechanisms include "to treat," "to cure," and "to treat." The term "thing" means, for example, by stabilizing identifiable symptoms physically, for example, physically Stabilizing parameters inhibits the progression of proliferative disorders, either physiologically or both. It refers to harm. Other means include "to treat," "treatment," and "the act of treating." The term refers to a reduction or stabilization of tumor size or the number of cancerous cells.

[0083] 7.2.CD3 binding molecules In one embodiment, the disclosure includes monospecific and multispecific molecules that bind to human CD3. The present invention provides a CD3-binding molecule. In one embodiment, the CD3-binding molecule has single specificity. It is a binding molecule. For example, a single-specific binding molecule is an antibody or its antigen-binding fragment. For example, antibody fragment, scFv, dsFv, Fv, Fab, scFab, (Fab ')2 or single-domain antibody (SDAB)) may be used. In other embodiments, CD3 The binding molecule is a multispecific (e.g., bispecific) CD3 binding molecule (e.g., bispecific antibody) (Body)

[0084] In one embodiment, the CD3-binding molecule is a chimeric or humanized monoclonal antibody. Chimeric and / or humanized antibodies are antibodies or non-human antibody genes produced on non-human subjects. To modify the expression of antibodies derived from this gene to minimize the immune response by human patients. This is possible. Chimeric antibodies contain a variable region of a non-human animal antibody and a constant region of a human antibody. Such antibodies retain the epitope-binding specificity of the original monoclonal antibody, but when administered to humans... Because it may be less immunogenic when administered, it is more likely to be tolerable by patients. For example, one of the variable regions of the light chain of a mouse antibody (e.g., a mouse monoclonal antibody) all (for example, one, two, or three) and / or one or all of the variable regions of the heavy chain (for example) These include, but are not limited to, one, two, or three, the IgG1 human constant region, etc. It can bind to the human constant region that is not affected. Chimeric monoclonal antibodies are known pairs. It can be produced by substitution DNA technology. For example, the constant region of a non-human antibody molecule can be converted to The gene that codes can be replaced with a gene that codes for a human constant region (Robi nson et al., PCT / US86 / 02269; Akira, et al. al., European Patent Application Nos. 184 and 187; or Taniguchi, M., European (See Patent Application No. 171,496). Furthermore, to generate chimeric antibodies Other suitable technologies that can be used for this purpose include, for example, U.S. Patent No. 4,816,567. Specification, Specification No. 4,978,775; Specification No. 4,975,369; and the same It is described in Specification No. 4,816,397.

[0085] The chimeric or humanized antibodies and their antigen-binding fragments disclosed herein are mouse monoclonal antibodies. It can be prepared based on the sequence of the antibody. It encodes heavy chain and light chain immunoglobulins. The DNA was obtained from the target mouse hybridoma and analyzed using standard molecular biology techniques. It can be modified to include non-mouse (e.g., human) immunoglobulin sequences. For example, to produce chimeric antibodies, the mouse variable region is used in human antibody production using known methods. It can be linked to the steady-state region (for example, Cabilly et al.'s U.S. (See Specification No. 4,816,567). To produce humanized antibodies, known The mouse CDR region can be inserted into the human framework using this method. For example, Winter and U.S. Patent No. 5,225,539 and U.S. Patent No. 5,530,10 Specification No. 1; Specification No. 5,585,089; Specification No. 5,693,762 and the same See Queen et al. in Specification No. 6180370.

[0086] Humanized antibodies are, but are not limited to, CDR implantation (e.g., European Patent No. 239,400). Specification; International Publication No. 91 / 09967; U.S. Patent No. 5,225,539 See the specifications, specifications No. 5,530,101 and No. 5,585,089. (to be used), veneering or resurfacing (for example, European Patent No. 592,106) Detailed Statement and Specification Nos. 519 and 596; Padlan, 1991, Molecular Immunology,28(4 / 5):489-498;Studnicka et al. al.,1994,Protein Engineering,7(6):805-81 4; and Roguska et al., 1994, PNAS, 91:969-973 (See reference), chain shuffling (e.g., U.S. Patent No. 5,565,332) (See details) and, for example, U.S. Patent Application Publication No. 2005 / 0042664 The document, U.S. Patent Application Publication No. 2005 / 0048617, U.S. Patent No. 6,407,2 U.S. Patent No. 13, U.S. Patent No. 5,766,886, International Publication No. 9317105 Tan et al., J.Immunol., 169:1119-25( 2002), Caldas et al., Protein Eng., 13(5):3 53-60(2000), Morea et al.,Methods,20(3):2 67-79(2000), Baca et al., J. Biol. Chem., 272 (16):10678-84(1997), Roguska et al., Prote in Eng.,9(10):895-904(1996), Couto et al. ,Cancer Res.,55(23 Supp):5973s-5977s(199 5), Couto et al., Cancer Res., 55(8):1717-2 2(1995), Sandhu JS, Gene, 150(2):409-10(19 94) and Pedersen et al., J. Mol. Biol., 235(3): Produced using various known techniques, including the techniques disclosed in 959-73 (1994) It is possible to do so. Often, framework residues in the framework region are used to identify antigenic bonds. To change the match, for example, to improve it, the corresponding residue from the CDR donor antibody is substituted. These framework substitutions, for example, conservative substitutions, are known methods, for example, antigen binding. Modeling of the interaction between CDRs and framework residues to identify important framework residues By Delling and sequence comparison to identify abnormal framework residues at specific locations It is identified by [this method]. (For example, Queen et al., U.S. Patent No. 5,585,089) Specification; and Riechmann et al., 1988, Nature, 332:3 (See section 23).

[0087] As provided herein, a humanized antibody or antibody fragment has a framework Non-human immunoglobulin molecules and flames derived entirely or largely from the human germline. May contain one or more CDRs from the 'c' region. For humanization of antibodies or antibody fragments. Several techniques are well known, and involve using rodent CDRs or CDR sequences with the corresponding sequences of human antibodies. Replacing it with, that is, CDR porting (European Patent No. 239,400; International Patent Pamphlet No. 91 / 09967; and U.S. Patent No. 4,816,567; the same Specification No. 6,331,415; Specification No. 5,225,539; Specification No. 5,530,1 (Specifications No. 01; Specifications No. 5,585,089; Specifications No. 6,548,640) More specifically, the method of Winter and his collaborators (Jones et al., Nature) ,321:522-525(1986);Riechmann et al.,Natu re,332:323-327(1988);Verhoeyen et al.,Sc Essentially, to do so in accordance with ience, 239:1534-1536 (1988) This is possible. In such humanized antibodies and antibody fragments, substantially fewer than one The intact human variable domain is replaced by a corresponding sequence from a non-human species. Antibodies often contain certain CDR residues and possibly certain framework (FR) residues. This is a human antibody that is substituted with residues from similar sites in rodent antibodies. Humanization of body fragments involves veneering or resurfacing (European Patent No. 592, 1 Specification No. 06; European Patent No. 519,596; Padlan, 1991, Mole cular Immunology,28(4 / 5):489-498;Studnic ka et al., Protein Engineering, 7(6):805-8 14(1994); and Roguska et al., PNAS, 91:969-97 3 (1994)) or chain shuffling (U.S. Patent No. 5,565,332) This can also be achieved by...

[0088] The selection of both light and heavy human variable domains used to produce humanized antibodies is related to the antigen. This is to reduce the sex. The so-called "best fit" method allows for variable domains of rodent antibodies. The sequence of the in is screened against the entire library of known human variable domain sequences. Next, the human sequence that is closest to the rodent sequence is the human framework (F) of humanized antibodies. (R) recognized as such (Sims et al., J.Immunol., 151.22) 96(1993);Chothia et al., J.Mol.Biol.,196: 901 (1987). Another method involves the whole antibody of a specific subgroup of light or heavy chains. It uses a specific framework derived from census arrays. The same framework is used in several places. It can be used with different humanized antibodies (e.g., Nicholson et al. Mol. I mmun.34(16-17):1157-1165(1997);Carter et al.,Proc.Natl.Acad.Sci.USA,89:4285(1992 ); Presta et al., J. Immunol., 151:2623 (1993 (See reference). In one embodiment, the framework region of the heavy chain variable region, for example Furthermore, all four framework regions are derived from the VH4_4-59 germline sequence. In one embodiment, the framework region is, for example, an amino acid of the corresponding mouse sequence. These include one, two, three, four, or five modifications, such as substitutions, such as conservative substitutions. It can be seen. In one embodiment, the framework region of the light chain variable region, for example, all 4 frames The muwerk region is derived from the VK3_1.25 germline sequence. In one embodiment, The framework area is, for example, one, two, three amino acids from the corresponding mouse sequence. This may include four or five modifications, such as substitutions, such as conservative substitutions.

[0089] In certain embodiments, the CD3-binding molecule is a specific germline heavy chain immunoglobulin gene. Heavy chain variable regions from genes and / or light chain immunoglobulin genes from specific germline light chain immunoglobulin genes It includes a chain variable region. For example, such an antibody is a "product" of a specific germline sequence. Human antibodies containing heavy chain or light chain variable regions derived therefrom or from there It can be. It is either a "product" of or derived from a human germline immunoglobulin sequence. Human antibodies are derived from the amino acid sequence of human germline immunoglobulins. Compared to (using the methods outlined herein), the sequence that is closest to the sequence of a human antibody ( In other words, by selecting the human germline immunoglobulin sequence with the highest % identity, It can be identified as a "product" of a specific human germline immunoglobulin sequence. Human antibodies that are derived from or originate from such antibodies may, for example, be naturally occurring somatic mutations or intentional mutations. To introduce site-directed mutations, the sequence may contain amino acid differences relative to the germline sequence. However, humanized antibodies typically have an amino acid sequence that is incompatible with human germ cell immunity. It is at least 90% identical to the amino acid sequence encoded by the globulin gene, and other Compared with the germline immunoglobulin amino acid sequence of a species (e.g., mouse germline sequence). In this case, the antibody contains amino acid residues that are identified as being derived from a human sequence. In this case, the humanized antibody is derived from the germline immunoglobulin gene in its amino acid sequence. The amino acid sequence encoded by and at least 95, 96, 97, 98 or 99% or less At the very least, they may be 96%, 97%, 98%, or even 99% identical. Typically, a specific type Humanized antibodies derived from germline sequences are produced by human germline immunoglobulin genes. This shows a difference of 10 to 20 amino acids or less from the encoded amino acid sequence (as specified herein). Before the introduction of any skew, pI, and ablation variants; i.e., the present disclosure Before the introduction of the variant, the number of variants is generally small. In certain cases, humanized antibodies are used in the reproductive organs. From the amino acid sequence encoded by the cellular immunoglobulin gene, 5 amino acids or less or 4 amino acids , 3, 2 or 1 amino acid difference may be shown (here as specified in this specification) Before the introduction of any distortion, pI, and ablation variants; i.e., the variants of the present disclosure Before introduction, the number of mutants was generally small.

[0090] In one embodiment, the parent antibody is affinity-matured. A structure-based method is, for example, As described in USSN 11 / 004,590, it is used for humanization and affinity maturation. It is possible. However, it is not limited to Wu et al., 1999, J.Mol.Bi. ol.294:151-162;Baca et al.,1997,J.Biol.C hem.272(16):10678-10684;Rosok et al.,199 6,J.Biol.Chem.271(37):22611-22618;Rader et al.,1998,Proc.Natl.Acad.Sci.USA 95:89 10-8915;Krauss et al.,2003,Protein Engine Selection-based methods, including those described in eering 16(10):753-759. Using this method, the antibody variable region can be humanized and / or affinity matured. The method of synthesis is not limited, but see USSN 09 / 810,510; Tan et al. ,2002,J.Immunol.169:1119-1125;De Pascali This is described in s et al., 2002, J.Immunol. 169:3076-3084. This may include the porting of only a portion of the CDR, including the method used.

[0091] In one embodiment, the CD3-binding molecule includes ABM, which is Fab. The enzyme papain is used to perform proteolytic cleavage or combination of immunoglobulin molecules. It can be generated by substitution expression. The Fab domain is typically bound to the VL domain. It contains a CH1 domain bound to a VH domain that pairs with the CL domain. Wild-type immunity In globulins, the VH domain is paired with the VL domain to form the Fv region, C The H1 domain is paired with the CL domain to further stabilize the binding module. Disulfide bonds between constant domains can further stabilize the Fab domain.

[0092] In one embodiment, the CD3-binding molecule includes ABM, which is scFab. In this state, the antibody domain and linker in the scFab fragment are located from the N-terminus to the C-terminus. In the direction toward the end, one of the following sequences: a) VH-CH1-linker-VL-CL or b) VL -CL-linker-VH-CH1 is present. In some cases, VL-CL-linker-VH -CH1 is used.

[0093] In another embodiment, the antibody domain and linker in the scFab fragment are N From terminal to C-terminus, one of the following sequences: a) VH-CL-linker-VL-CH1 or b) having VL-CH1-linker-VH-CL.

[0094] Selectively, in the scFab fragment, the CL-domain and the CH1 domain In addition to the natural disulfide bond between them, the antibody heavy chain variable domain (VH) and antibody light chain variable The domains (VLs) are located at the following positions: i) heavy chain variable domain position 44 and light chain variable domain position 1 ii) position 00, ii) heavy chain variable domain position 105 and light chain variable domain position 43, or iii) heavy Chain variable domain ranked 101st and light chain variable domain ranked 100th (Kabat EU index) Disulfide stabilization is also achieved by introducing disulfide bonds between (numbered by) the atoms.

[0095] Such further disulfide stabilization of scFab fragments is possible for single-strand Fab fragments. This is achieved by introducing disulfide bonds between the variable domains VH and VL of the lagment. Techniques for introducing non-natural disulfide bridges to stabilize single-strand Fv include, for example, International publication no. 94 / 029350, Rajagopal et al., 1997,Prot.Engin.10:1453-59;Kobayashi et. al.,1998,Nuclear Medicine&Biology,25:387 -393; and Schmidt, et al., 1999, Oncogene 18:1 As described in 711-1721, in one embodiment, the scFab fragment can The optional disulfide bond between the variable domains is formed between the heavy chain variable domain at position 44 and the light chain variable domain. It is between position 100. In one embodiment, the variable domain of the scFab fragment The optional disulfide bond between the two is between the heavy chain variable domain at position 105 and the light chain variable domain at position 4. It is located between third place and third place (numbered according to Kabat's EU index).

[0096] In one embodiment, the CD3-binding molecule includes ABM, which is scFv. The antibody fragment comprises the VH and VL domains of the antibody in a single polypeptide chain. It can be expressed as a true chain polypeptide, and the specificity of intact antibodies derived from it It maintains its properties. Generally, scFv polypeptides are scFv molecules that are in the desired structure for target binding. Further polypeptide linkers between VH and VL domains that enable the formation of structures Includes. Examples of linkers suitable for linking the VH and VL chains of scFV are given in Section 7.4.3. The ABM linkers specified in, for example, the linkers indicated as L1 to L58 It is one of them.

[0097] Unless otherwise specified, when used herein, scFv means, for example, polyp The N-terminus and C-terminus of the plutide may have VL and VH variable regions in either order. scFv may include VL-linker-VH or VH-linker-VL.

[0098] To generate scFv coding nucleic acids, VH and VL coding DNA fragments are used. A linker that codes for one of the linkers described in Section 7.4.3 (amino acids) Create another fragment that codes for the array (Gly4~Ser)3 (sequence number 47), etc. Movable connection, with the VL and VH regions connected by a flexible linker, VH And the VL sequence can be expressed as a continuous single-strand protein (for example) Bird et al., 1988, Science 242:423-426; uston et al.,1988,Proc.Natl.Acad.Sci.USA 85:5879-5883; McCafferty et al., 1990, Nat. See ure 348:552-554.

[0099] CD3-binding molecules include Fv, dsFv, (Fab')2, and single-domain antibodies (SDAB). , VH or VL domain or camelid VHH domain (also called nanobody) This may also include ABM, which is )

[0100] CD3-binding molecules may have a single VH or VL domain that exhibits sufficient affinity for CD3. It may include a single-domain antibody composed of the following. In one embodiment, the single-domain antibody is This is a VHH domain of the camelid family (e.g., Riechmann, 1999, Journal). I of Immunological Methods 231:25-38; International Public Please refer to pamphlet No. 94 / 04678.

[0101] Tables 1A-1J-2 (collectively referred to as "Table 1") show exemplary CDs that may be included in CD3-binding molecules. List the sequences of 3-linked sequences.

[0102] [Table 1]

[0103] [Table 2]

[0104] [Table 3]

[0105] [Table 4]

[0106] [Table 5]

[0107] [Table 6]

[0108] [Table 7]

[0109] [Table 8]

[0110] [Table 9]

[0111] [Table 10]

[0112] Table 11

[0113] Table 12

[0114] Table 13

[0115] Table 14

[0116] Table 15

[0117] Table 16

[0118] Table 17

[0119] Table 18

[0120] Table 19

[0121] Table 20

[0122] Table 21

[0123] Table 22

[0124] Table 23

[0125] Table 24

[0126] Table 25

[0127] Table 26

[0128] Table 27

[0129] Table 28

[0130] Table 29

[0131] Table 30

[0132] Table 31

[0133] Table 32

[0134] Table 33

[0135] Table 34

[0136] Table 35

[0137] Table 36

[0138] Table 37

[0139] Table 38

[0140] Table 39

[0141] Table 40

[0142] Table 41

[0143] Table 42

[0144] Table 43

[0145] Table 44

[0146] Table 45

[0147] Table 46

[0148] Table 47

[0149] Table 48

[0150] Table 49

[0151] Table 50

[0152] Table 51

[0153] Table 52

[0154] Table 53

[0155] Table 54

[0156] Table 55

[0157] Table 56

[0158] Table 57

[0159] Table 58

[0160] Table 59

[0161] Table 60

[0162] Table 61

[0163] Table 62

[0164] Table 63

[0165] Table 64

[0166] Table 65

[0167] Table 66

[0168] Table 67

[0169] Table 68

[0170] Table 69

[0171] Table 70

[0172] Table 71

[0173] Table 72

[0174] Table 73

[0175] [Table 74]

[0176] [Table 75]

[0177] [Table 76]

[0178] [Table 77]

[0179] [Table 78]

[0180] [Table 79]

[0181] [Table 80]

[0182] [Table 81]

[0183] [Table 82]

[0184] Tables 1A-1C show the CD3 binding sequences based on the exemplary CD3 binding molecules described herein. The R consensus sequences are listed below. The group C1 CDR sequences in Table 1A are exemplary CD3 sequences. The binding molecules NOV292, NOV589, NOV567 and the binder name include "sp11a". Examples of CD3-binding molecules include the Kabat CDR sequence, the Chothia CDR sequence, and I Based on MGT CDR sequences and their combinations. Group C2 CDRs in Table 1B. The sequences are exemplary CD3-binding molecules NOV453, NOV229, NOV580, NOV2 21 and Kabat CDR formulation of exemplary CD3 binding molecules containing "sp9a" as the binder name Based on the column, Chothia CDR sequence, IMGT CDR sequence, and combinations thereof The C3 CDR sequence in Table 1C is an example of the CD3-binding molecule NOV123, s Kabat CDR sequences of p10b, NOV110 and NOV832, Chothia It is based on CDR sequences, IMGT CDR sequences, and combinations thereof.

[0185] The specific CDR sequences of the exemplary CD3-binding molecules described in the examples are shown in Table 1B-1 to 1H. -2 is described. Exemplary VH and VL sequences are shown in Tables 1J-1 and 1J-2, respectively. It is listed there.

[0186] In one embodiment, the CD3-binding molecule is C listed in Table 1A, Table 1B, or Table 1C. Contains a heavy chain CDR having one of the amino acid sequences of the DR consensus sequence. In embodiments, the disclosure selects from the heavy chain CDRs listed in Table 1A, Table 1B, or Table 1C. Includes one, two, three or more heavy chain CDRs selected (or instead) It provides a CD3-binding molecule.

[0187] In one embodiment, the CD3-binding molecule is C listed in Table 1A, Table 1B, or Table 1C. Contains a light chain CDR having one of the amino acid sequences of the DR consensus sequence. In embodiments, the disclosure selects from the light chain CDRs listed in Table 1A, Table 1B, or Table 1C. Includes one, two, three or more light chain CDRs selected (or instead) It provides a CD3-binding molecule.

[0188] In one embodiment, the CD3 binding molecule is the CDR-H1 sequence, C, as described in Table 1A. DR-H2 sequence, CDR-H3 sequence, CDR-L1 sequence, CDR-L2 sequence and CDR- Includes L3 sequence.

[0189] In one embodiment, the amino acid shown as X1 in Table 1A is T. In the application, the amino acid shown as X1 in Table 1A is A. In Table 1A, the amino acid indicated as X2 is S. In one embodiment, In Table 1A, the amino acid shown as X2 is R. In one embodiment, Table 1A In this embodiment, the amino acid shown as X3 is N. In one embodiment, in Table 1A, X The amino acid shown as 3 is Y. In one embodiment, the amino acid shown as X3 in Table 1A is The amino acid is Q. In one embodiment, the amino acid shown as X4 in Table 1A is Q. The no acid is H. In one embodiment, the amino acid shown as X4 in Table 1A is It is S. In one embodiment, the amino acid shown as X5 in Table 1A is M. In one embodiment, the amino acid shown as X5 in Table 1A is L. In the embodiment, the amino acid shown as X6 in Table 1A is K. In Table 1A, the amino acid indicated as X6 is R. In one embodiment, In Table 1A, the amino acid indicated as X7 is S. In one embodiment, Table 1A In this embodiment, the amino acid indicated as X7 is K. In one embodiment, in Table 1A, X 55 The amino acid shown is F. In one embodiment, X in Table 1A. 55 and The amino acid shown is Y. In one embodiment, X is shown in Table 1A. 55 It was shown The amino acid is S. In one embodiment, the amino acid shown as X8 in Table 1A is The acid is W. In one embodiment, the amino acid shown as X8 in Table 1A is Y In one embodiment, the amino acid indicated as X8 in Table 1A is S. In one embodiment, the amino acid shown as X8 in Table 1A is T. In terms of form, the amino acid indicated as X9 in Table 1A is W. In Table 1A, the amino acid shown as X9 is Y. In one embodiment, Table In 1A, the amino acid indicated as X9 is S. In one embodiment, Table 1A The amino acid shown as X9 is T. In one embodiment, X1 The amino acid shown as 0 is H. In one embodiment, X in Table 1A. 10 and The amino acid that is used is Y. In one embodiment, X is shown in Table 1A. 11 This is shown The amino acid is S. In one embodiment, X is shown in Table 1A. 11 amino The acid is G. In one embodiment, X is shown in Table 1A. 12 The amino acids shown are, I is in one embodiment, X in Table 1A 12 The amino acid shown is L In one embodiment, X in Table 1A 13 The amino acid shown is V. In the embodiment, X in Table 1A 13 The amino acid shown is G. In terms of morphology, X in Table 1A 14 The amino acid shown is R. In one embodiment In Table 1A, X 14 The amino acid shown is N. In one embodiment, In Table 1A, X 15 The amino acid shown is D. In one embodiment, Table 1 In A, X 15 The amino acid shown is E. In one embodiment, Table 1A IteX 15 The amino acid shown is L. In one embodiment, X in Table 1A. 16 The amino acid shown is G. In one embodiment, X in Table 1A. 16 and The amino acid shown is N. In one embodiment, X is shown in Table 1A. 16 It was shown The amino acid is E. In one embodiment, X in Table 1A. 17 The net is shown as The no acid is R. In one embodiment, X is shown in Table 1A. 17 The amino acids shown are , S. In one embodiment, X in Table 1A. 18 The amino acids shown are in V. Yes. In one embodiment, X in Table 1A 18 The amino acid shown is T. In one embodiment, X in Table 1A 19 The amino acid shown is N. In the application method, X in Table 1A 19The amino acid shown is T. In Table 1A, X 20 The amino acid shown is R. In one embodiment, In Table 1A, X 20 The amino acid shown is L. In one embodiment, Table In 1A, X 21 The amino acid shown is F. In one embodiment, Table 1A Oite X 21 The amino acid shown is E. In one embodiment, in Table 1A X 22 The amino acid shown is S. In one embodiment, X in Table 1A. 22 The amino acid shown is Y. In one embodiment, X in Table 1A. 23 and it was shown The amino acid is S. In one embodiment, X is shown in Table 1A. 23 A is shown as The amino acid is Y. In one embodiment, X is shown in Table 1A. 24 The amino acids shown In one embodiment, X is in Table 1A. 24 The amino acid shown is A In one embodiment, X in Table 1A 25 The amino acid shown is H. In one embodiment, X in Table 1A 25 The amino acid shown is T. In the embodiment, X in Table 1A 26 The amino acid shown is F. In the state, X in Table 1A 26 The amino acid shown is Y. In one embodiment And in Table 1A, X 27 The amino acid shown is W. In one embodiment, In Table 1A, X 27 The amino acid shown is Y.

[0190] In one embodiment, the CD3-binding molecule contains the CDR-H1 sequence C1-1. In the application form, the CD3 binding molecule contains the CDR-H1 sequence C1-2. In one embodiment In this embodiment, the CD3 binding molecule contains the CDR-H1 sequence C1-3. The CD3-binding molecule contains the CDR-H1 sequence C1-4.

[0191] In one embodiment, the CD3-binding molecule contains the CDR-H2 sequence C1-5. In the application form, the CD3 binding molecule contains the CDR-H2 sequence C1-6. In one embodiment In this context, the CD3-binding molecule contains the CDR-H2 sequence C1-7.

[0192] In one embodiment, the CD3-binding molecule contains the CDR-H3 sequence C1-8. In the application form, the CD3-binding molecule contains the CDR-H3 sequence C1-9. In one embodiment In one embodiment, the CD3-binding molecule contains the CDR-H3 sequence C1-10. The CD3-binding molecule contains the CDR-H3 sequence C1-11.

[0193] In one embodiment, the CD3-binding molecule contains the CDR-L1 sequence C1-12. In one embodiment, the CD3-binding molecule contains the CDR-L1 sequence C1-13. In this state, the CD3 binding molecule contains the CDR-L1 sequence C1-14. In one embodiment, In one embodiment, the CD3-binding molecule contains the CDR-L1 sequence C1-15. The CD3-binding molecule contains the CDR-L1 sequence C1-16. In one embodiment, CD3 The binding molecule contains the CDR-L1 sequence C1-17.

[0194] In one embodiment, the CD3-binding molecule contains the CDR-L2 sequence C1-18. In the embodiment, the CD3 binding molecule contains the CDR-L2 sequence C1-19.

[0195] In one embodiment, the CD3-binding molecule contains the CDR-L3 sequence C1-20. In one embodiment, the CD3 binding molecule contains the CDR-L3 sequence C1-21. In this state, the CD3 binding molecule contains the CDR-L3 sequence C1-22. In one embodiment, Furthermore, the CD3-binding molecule contains the CDR-L3 sequence C1-23.

[0196] In one embodiment, the CD3 binding molecule is the CDR-H1 sequence, C, as shown in Table 1B. DR-H2 sequence, CDR-H3 sequence, CDR-L1 sequence, CDR-L2 sequence and CDR- Includes L3 sequence.

[0197] In one embodiment, X in Table 1B 28 The amino acid shown is V. In the embodiment, X in Table 1B 28 The amino acid shown is I. In the state, X in Table 1B 29 The amino acid shown is F. In one embodiment And in Table 1B, X 29 The amino acid shown is Y. In one embodiment, In Table 1B, X 30 The amino acid shown is N. In one embodiment, Table 1B In X 30 The amino acid shown is S. In one embodiment, Table 1B shows teX 31 The amino acid shown is A. In one embodiment, X3 in Table 1B The amino acid shown as 1 is S. In one embodiment, X is shown in Table 1B. 32and The amino acid that is used is T. In one embodiment, X is shown in Table 1B. 32 This is shown The amino acid is K. In one embodiment, X is shown in Table 1B. 33 amino The acid is T. In one embodiment, X is shown in Table 1B. 33 The amino acids shown are, A is the case in one embodiment, X is shown in Table 1B. 34 The amino acid shown is S In one embodiment, X in Table 1B 34 The amino acid shown is R. In the embodiment, X in Table 1B 35 The amino acid shown is N. In terms of morphology, X in Table 1B 35 The amino acid shown is G. In one embodiment In Table 1B, X 36 The amino acid shown is S. In one embodiment, In Table 1B, X 36 The amino acid shown is A. In one embodiment, Table 1 In B, X 37 The amino acid shown is A. In one embodiment, Table 1B IteX 37 The amino acid shown is T. In one embodiment, X in Table 1B. 37 The amino acid shown is S. In one embodiment, X in Table 1B. 38 and The amino acid shown is N. In one embodiment, X is shown in Table 1B. 38 It was shown The amino acid is D. In one embodiment, X is shown in Table 1B. 39 The net is shown as The no acid is N. In one embodiment, X is shown in Table 1B. 39 The amino acids shown are , K. In one embodiment, X in Table 1B. 40 The amino acid shown is D Yes, in one embodiment, X in Table 1B 40 The amino acid shown is N. In one embodiment, X in Table 1B 41 The amino acid shown is H. In the application method, X in Table 1B 41 The amino acid shown is N. In one embodiment In Table 1B, X 42 The amino acid shown is Q. In one embodiment, In Table 1B, X 42 The amino acid shown is E. In one embodiment, Table In 1B, X 43 The amino acid shown is R. In one embodiment, Table 1B Oite X 43 The amino acid shown is S. In one embodiment, Table 1B X 43 The amino acid shown is G. In one embodiment, the CD3 binding molecule is C Contains DR-H1 sequence C2-1. In one embodiment, the CD3 binding molecule is CDR-H Contains sequence C2-2.

[0198] In one embodiment, the CD3-binding molecule contains the CDR-H1 sequence C2-3. In the application form, the CD3-binding molecule contains the CDR-H1 sequence C2-4.

[0199] In one embodiment, the CD3 binding molecule contains the CDR-H2 sequence C2-5. In the application form, the CD3-binding molecule contains the CDR-H2 sequence C2-6. In one embodiment In this context, the CD3-binding molecule contains the CDR-H2 sequence C2-7.

[0200] In certain embodiments, the CD3 binding molecule comprises the CDR-H3 sequence C2-8. In certain embodiments, the CD3 binding molecule comprises the CDR-H3 sequence C2-9.

[0201] In certain embodiments, the CD3 binding molecule comprises the CDR-L1 sequence C2-10. In certain embodiments, the CD3 binding molecule comprises the CDR-L1 sequence C2-11. In certain embodiments the CD3 binding molecule comprises the CDR-L1 sequence C2-12.

[0202] In certain embodiments, the CD3 binding molecule comprises the CDR-L2 sequence C2-13. In certain embodiments, the CD3 binding molecule comprises the CDR-L2 sequence C2-14. In certain embodiments the CD3 binding molecule comprises the CDR-L2 sequence C2-15.

[0203] In certain embodiments, the CD3 binding molecule comprises the CDR-L3 sequence C2-16. In certain embodiments, the CD3 binding molecule comprises the CDR-L3 sequence C2-17.

[0204] In certain embodiments, the CD3 binding molecule comprises the CDR-H1 sequence, CDR-H2 sequence, CDR-H3 sequence, CDR-L1 sequence, CDR-L2 sequence and CDR-L3 sequence set forth in Table 1C.

[0205] In certain embodiments, the amino acid designated as X 44 in Table 1C is G. In certain embodiments, the amino acid designated as X 44 in Table 1C is A. In certain embodiments the amino acid designated as X 45 in Table 1C is H. In certain embodiments the amino acid designated as X 45 in Table 1C is N. In certain embodiments, In Table 1C, X 46 and the amino acid shown is D. In certain embodiments, in Table 1C X 46 and the amino acid shown is G. In certain embodiments, in Table 1C X 47 and the amino acid shown is A. In certain embodiments, in Table 1C, X4 7 and the amino acid shown is G. In certain embodiments, in Table 1C, X 48 and the amino acid shown is N. In certain embodiments, in Table 1C, X 48 and the amino acid shown is K. In certain embodiments, in Table 1C, X 49 and the amino acid shown is V. In certain embodiments, in Table 1C, X 49 and the amino acid shown is A. In certain embodiments, in Table 1C, X 50 and the amino acid shown is N is. In certain embodiments, in Table 1C, X 50 and the amino acid shown is V. In certain embodiments, in Table 1C, X[[ID=�8]] 51 and the amino acid shown is A. In certain embodiments, in Table 1C, X 51 and the amino acid shown is V. In certain embodiments in Table 1C, X 52 and the amino acid shown is Y. In certain embodiments in Table 1C, X 52 and the amino acid shown is F. In certain embodiments, in Table 1 C, X 53 and the amino acid shown is I. In certain embodiments, in Table 1C X 53 and the amino acid shown is V. In certain embodiments, in Table 1C, X 54The amino acid shown is I. In one embodiment, X in Table 1C 54 and The amino acid shown is H.

[0206] In one embodiment, the CD3-binding molecule contains the CDR-H1 sequence C3-1. In the application form, the CD3-binding molecule contains the CDR-H1 sequence C3-2. In one embodiment, In this embodiment, the CD3-binding molecule contains the CDR-H1 sequence C3-3. The CD3-binding molecule contains the CDR-H1 sequence C3-4.

[0207] In one embodiment, the CD3-binding molecule contains the CDR-H2 sequence C3-5. In the application form, the CD3-binding molecule contains the CDR-H2 sequence C3-6. In one embodiment, In this context, the CD3-binding molecule contains the CDR-H2 sequence C3-7.

[0208] In one embodiment, the CD3-binding molecule contains the CDR-H3 sequence C3-8. In the application form, the CD3-binding molecule contains the CDR-H3 sequence C3-9.

[0209] In one embodiment, the CD3-binding molecule contains the CDR-L1 sequence C3-10. In one embodiment, the CD3-binding molecule contains the CDR-L1 sequence C3-11. In this state, the CD3-binding molecule contains the CDR-L1 sequence C3-12.

[0210] In one embodiment, the CD3-binding molecule contains the CDR-L2 sequence C3-13. In the embodiment, the CD3-binding molecule contains the CDR-L2 sequence C3-14.

[0211] In one embodiment, the CD3-binding molecule contains the CDR-L3 sequence C3-15. In the embodiment, the CD3-binding molecule contains the CDR-L3 sequence C3-16.

[0212] In one embodiment, the CD3-binding molecule is CDR-H1, C listed in Table 1D-1. The DR-H2 and CDR-H3 sequences and the corresponding CDR-L1s listed in Table 1D-2, Includes CDR-L2 and CDR-L3 sequences.

[0213] In one embodiment, the CD3-binding molecule is CDR-H1, C listed in Table 1E-1. The DR-H2 and CDR-H3 sequences and the corresponding CDR-L1s listed in Table 1E-2, Includes CDR-L2 and CDR-L3 sequences.

[0214] In one embodiment, the CD3-binding molecule is CDR-H1, C listed in Table 1F-1. The DR-H2 and CDR-H3 sequences and the corresponding CDR-L1s listed in Table 1F-2, Includes CDR-L2 and CDR-L3 sequences.

[0215] In one embodiment, the CD3-binding molecule is CDR-H1, C listed in Table 1G-1. The DR-H2 and CDR-H3 sequences and the corresponding CDR-L1s listed in Table 1G-2, Includes CDR-L2 and CDR-L3 sequences.

[0216] In one embodiment, the CD3-binding molecule is CDR-H1, C listed in Table 1H-1. The DR-H2 and CDR-H3 sequences and the corresponding CDR-L1s listed in Table 1H-2, Includes CDR-L2 and CDR-L3 sequences.

[0217] In one embodiment, the CD3-binding molecule is CDR-H1, C listed in Table 1I-1. The DR-H2 and CDR-H3 sequences and the corresponding CDR-L1s listed in Table 1I-2, Includes CDR-L2 and CDR-L3 sequences.

[0218] In one embodiment, the CD3-binding molecule is shown in Table 1B-1, Table 1C-1, Table 1D-1, Table The CDRs listed in 1E-1, Table 1F-1, Table 1G-1, Table 1H-1, or Table 1I-1 Includes heavy chain CDR having either one amino acid sequence. In certain embodiments, this disclosure Table 1B-1, Table 1C-1, Table 1D-1, Table 1E-1, Table 1F-1, Table 1G-1, Table 1 One, two, three, or any of the heavy chain CDRs selected from H-1 and Table 1I-1. Provides a CD3-binding molecule containing (or consisting of) more than a certain number of heavy chain CDRs. .

[0219] In one embodiment, the CD3 binding molecule is shown in Table 1B-2, Table 1C-2, Table 1D-2, Table The CDRs listed in 1E-2, Table 1F-2, Table 1G-2, Table 1H-2, or Table 1I-2 The light chain CDR having either one amino acid sequence is included. In certain embodiments, the disclosure Table 1B-2, Table 1C-2, Table 1D-2, Table 1E-2, Table 1F-2, Table 1G-2, Table 1 One, two, three, or any of the light chain CDRs selected from H-2 and those listed in Table 1I-2. Provides a CD3-binding molecule containing (or consisting of) more than a certain number of light chain CDRs. .

[0220] Other CD3-binding molecules have mutations, but still have the CDR sequences and CD3 listed in Table 1. In the R region, at least 80, 85, 90, 95, 96, 97, 98, or 99 percent of the same It contains a monogenic amino acid. In one embodiment, such a CD3-binding molecule When compared with the CDR sequences listed in Table 1, in the CDR region, 1, 2, 3, 4 or This includes mutant amino acid sequences in which five or fewer amino acids are mutated.

[0221] In one embodiment, the CD3-binding molecule is any VH and / or V listed in Table 1. Contains VH and / or VL domains having the amino acid sequence of the L domain. Other CD3 binding The molecule contains the VH and / or VL sequences listed in Table 1 and at least 80, 85, 90, 9 VH and containing amino acid sequences having 5, 96, 97, 98, or 99 percent identity. It contains a bi / or VL domain. In one embodiment, the CD3 binding molecule is as shown in Table 1. When compared to the VH and / or VL domains shown in the sequence, the therapeutic activity is substantially the same. While retaining the characteristic, VH and / or have mutations in 1, 2, 3, 4, or 5 or fewer amino acids. This includes VL domains.

[0222] VH and VL sequences (amino acid sequence and nucleotide sequence encoding amino acid sequence) By "mixing and matching," other CD3-binding molecules can be generated. Such "mix-and-match" CD3-binding molecules are known in the art. Testing can be performed using a suitable assay (e.g., the FACS assay described in the examples). When the strands are mixed and matched, the VH sequence from a particular VH / VL pair is Therefore, it should be replaced with a structurally similar VH sequence. VL from a specific VH / VL pairing The sequence should be replaced with a structurally similar VL sequence.

[0223] Therefore, in one embodiment, the present disclosure relates to any of the VH sequences listed in Table 1-J1. The heavy chain variable region (VH) containing an amino acid sequence selected from any one of the following, as described in Table 1-J2. The present invention provides a CD3-binding molecule having a light chain variable region (VL) containing the amino acid sequence.

[0224] CD3-binding molecules are heterologous proteins or polypeptides (or fragments thereof, for example) , at least 10, at least 20, at least 30, at least 40, at least 50 , at least 60, at least 70, at least 80, at least 90 or at least 1 Fusing or chemically conjugating (covalent and non-covalent) a polypeptide of 00 amino acids (This includes both binding.) For example, CD3-binding molecules can detect proteins. It can be fused directly or indirectly to substances such as enzymes or fluorescent proteins. Fusing or binding proteins, polypeptides, or peptides to antibodies or antibody fragments. Methods for this are known, and involve fusing a protein or polypeptide to the CD3 binding molecule of this disclosure. Alternatively, it can be used to conjugate. For example, U.S. Patent No. 5,336, Specification No. 603, Specification No. 5,622,929, Specification No. 5,359,046, Specification No. 5,349,053, Specification No. 5,447,851, and Specification No. 5,112 ,946 specification; European Patent Nos. 307,434 and 367,166; International Pamphlet No. 96 / 04388 and Pamphlet No. 91 / 06570; As hkenazi et al.,1991,Proc.Natl.Acad.Sci.U SA 88:10535-10539;Zheng et al.,1995,J.Im munol.154:5590-5600; and Vil et al., 1992, Pr. See oc.Natl.Acad.Sci.USA 89:11337-11341. sea ​​bream.

[0225] Additional CD3-binding molecules are used for gene shuffling, motif shuffling, and exon shuffling. Shuffling and / or codon shuffling (collectively referred to as "DNA shuffling") It can be generated by the technology called DNA shuffling. The disclosed molecule or fragment thereof (for example, having higher affinity and lower dissociation rate) The activity of the molecule or its fragment can be altered. Generally, U.S. 5 ,605,793 specification, 5,811,238 specification, 5,830,721 Specification No. 5,834,252 and Specification No. 5,837,458; Pa tten et al.,1997,Curr.Opinion Biotechnol .8:724-33;Harayama,1998,Trends Biotechno l.16(2):76-82;Hansson et al.,1999,J.Mol. Biol. 287:265-76; and Lorenzo and Blasco, 199. See 8, Biotechniques 24(2):308-313. The CD3-binding molecules or their fragments described in this document are prone to errors before recombination. Subject to PCR, random nucleotide insertion, or other methods of random mutagenesis. This can be modified by using the CD3 binding molecule fragments described herein. A polynucleotide that encodes is one or more components, motifs, or segments of one or more heterologous molecules. It can be rearranged into sections, parts, domains, fragments, etc.

[0226] Furthermore, the CD3 binding molecule can be fused to marker sequences such as peptides to facilitate purification. This is possible. In one embodiment, the marker amino acid sequence is, in particular, the pQE vector QIAGEN, Inc., 9259 Eton Avenue, Chatswor It is a hexahistidine peptide such as a tag provided by th,CA,91311), and Many of these are commercially available. For example, Gentz ​​et al., 1989, Proc.N As described in atl.Acad.Sci.USA 86:821-824, hexa Histidine provides a convenient purification of fusion proteins. Other peptide tags useful for purification. While not limited to, epitopes derived from influenza hemagglutinin protein Corresponding hemagglutinin ("HA") tag (Wilson et al., 1984 Cel l 37:767) and the "flag" tag are included.

[0227] 7.3. Antigen Binding Module Typically, one or more ABMs in MBMs are immunoglobulin-based antigen-binding domains. For example, sequences of antibody fragments or derivatives. These antibody fragments and derivatives Typically, this includes the CDR of the antibody, as well as larger fragments and their derivatives, such as F This may include ab, scFab, Fv, and scFv.

[0228] Immunoglobulin-based ABMs are used, for example, to improve the properties of MBMs, including ABMs. This may include modifications to framework residues within VH and / or VL. For example, framework Microwave modifications can be performed to reduce the immunogenicity of MBMs. One method for performing framework modification involves modifying one or more framework residues of ABM to the corresponding... This involves "reverse mutation" in the germline sequence. Such residues are in the flame The key sequence can be identified by comparing it with the germline sequence from which ABM originates. To "match" the framework region sequence to the desired germline arrangement, the residues are: For example, site-directed mutagenesis can cause a "reverse mutation" in the corresponding germline sequence. MBMs having such “reverse-mutated” ABMs are included in this disclosure. It is intended to be done.

[0229] Another type of framework qualification is within the framework domain or one or more additional CDs. By mutating one or more residues within the R region, the T cell epitope is removed, thereby This involves reducing the potential immunogenicity of MBMs. This technique is also known as "deimmunization." The details are further described in U.S. Patent Application Publication No. 20030153043 by Carr et al. It is stated.

[0230] ABM can also be modified to have altered glycosylation, for example, This may be useful in increasing the affinity of MBM for one or more of the antigens. Modification can be performed, for example, by altering one or more glycosylation sites within the ABM sequence. This can be done. For example, removal of one or more variable region framework glycosylation sites. This results in the absence of glycosylation at that site by one or more amino acids. Exchange may occur. Such deglycosylation is an antigen This could increase the affinity of MBM to [the subject]. Such methods are, for example, described in the US patent by Co et al. This is described in Specification No. 5,714,350 and Specification No. 6,350,861.

[0231] 7.3.1. Immunoglobulin-based modules 7.3.1.1.Fab In a particular embodiment, ABM is a Fab domain. A Fab domain is papain By using enzymes such as the proteolytic cleavage or recombinant expression of immunoglobulin molecules It can be generated by... The Fab domain is typically a CL domain bound to a VL domain. It includes a CH1 domain bound to the VH domain that is in opposition to it.

[0232] In wild-type immunoglobulin, the VH domain is paired with the VL domain to form the Fv region. The CH1 domain is paired with the CL domain to further stabilize the binding module. The disulfide bond between the two constant domains can further stabilize the Fab domain. ru.

[0233] In MBM, the Fab heterodimerization method is used to create Fab domains that belong to the same ABM. This enables proper meeting and minimizes anomalous pairing of Fab domains belonging to different ABMs. It is advantageous to suppress it. For example, the Fab heterodimerization method shown in Table 2 below can be used. It can be used.

[0234] [Table 83]

[0235] Therefore, in certain embodiments, the appropriate association between two polypeptides of Fab is For example, as described in the international publication brochure No. 2009 / 080251, Fab By exchanging the VL and VH domains with each other, or by exchanging the CH1 and CL domains with each other It is promoted by replacing it with.

[0236] Appropriate Fab pairing involves one or more amino acid modifications in the CH1 domain and one or more A Introduce amino acid modifications into the CL domain of Fab, and / or introduce one or more amino acid modifications into V By introducing one or more amino acid modifications into the H domain and into the VL domain, This can be facilitated. Fab components may interact with each other preferentially over other Fab components. Thus, the amino acids that are modified are typically parts of the VH:VL and CH1:CL boundaries. ru.

[0237] In one embodiment, one or more amino acid modifications are indicated by Kabat numbering of the residues. To that end, the conserved frames of the variable (VH, VL) and steady (CH1, CL) domains Limited to humwork residues. Almagro, 2008, Frontiers In Bioscience 13:1619-1633 includes Kabat, Chothia and A framework definition of residues based on the IMGT numbering scheme is provided.

[0238] In one embodiment, introduced into the VH and CH1 and / or VL and CL domains The modifications are complementary to each other. This complementarity at the heavy-chain and light-chain boundary is due to steric and hydrophobic contact. This can be achieved based on electrostatic / charge interactions or various combinations of interactions. Protein Surface complementarity includes lock-and-key mating, knob-into-hole, protrusions, and cavities. The literature widely describes donors and acceptors, and these are all two. This suggests the properties of structural and chemical matching between interacting surfaces.

[0239] In one embodiment, one or more introduced modifications are novel across the boundaries of Fab components. A hydrogen bond is introduced. In one embodiment, one or more modifications introduced are Fab structures. Introducing a novel salt bridge across the boundaries of constituent elements. An exemplary substitution is described in International Publication No. 2014 / As described in pamphlet No. 150973 and international publication pamphlet No. 2014 / 082179. It is listed.

[0240] In one embodiment, the Fab domain is a 192E substitution in the CH1 domain and C This includes the 114A and 137K substitutions in the L domain, which occur between the CH1 and CL domains. Introducing salt bridges (Golay et al., 2016, J Immunol 196) (See 3199-211).

[0241] In one embodiment, the Fab domain contains 143Q and 188V in the CH1 domain. This includes substitutions and 113T and 176V substitutions in the CL domain, which are CH1 and CL It plays a role in exchanging hydrophobic and polar contact regions between domains (Golay et al. See l., 2016, J Immunol 196:3199-211.

[0242] In one embodiment, the Fab domain facilitates the appropriate assembly of Fab domains. To introduce cross-Fab boundaries, some or all of the VH, CH1, VL, and CL domains This may include modifications (Lewis et al., 2014 Nature Biot (ecology 32:191-198). In one embodiment, 39K, 62E modified The modifier is introduced into the VH domain, and the H172A and F174G modifiers are introduced into the CH1 domain. Then, 1R, 38D, (36F) modifications are introduced into the VL domain, L135Y, S17 In another embodiment, the 6W modification is introduced into the CL domain. It is introduced during the in, and the 38R modification is introduced into the VL domain.

[0243] The Fab domain replaces the natural CH1:CL disulfide bond with a modified disulfide bond. It can also be modified in such a way that it can improve the efficiency of matching Fab components. For example, modified The disulfide bond is located at 126C in the CH1 domain and at 121C in the CL domain. It can be introduced by implementing (Mazor et al., 2015, MAbs) See 7:377-89.

[0244] The Fab domain provides the CH1 domain and CL domain with another domain that facilitates proper assembly. It can also be modified by substituting the main term. For example, Wu et al., 2015. MAbs 7:364-76 describes the CH1 domain of the α T cell receptor as a constant domain. Substitution with and substitution of the CL domain of the T cell receptor with the β domain and 38 By introducing the D modification into the VL domain and the 39K modification into the VH domain, VL and combining further charge-to-charge interactions between VH domains with these domain substitutions It is stated that...

[0245] ABM can contain a single-strand Fab fragment, which is a variable domain of antibody heavy chains. (VH), antibody constant domain 1 (CH1), antibody light chain variable domain (VL), antibody light chain It is a polypeptide comprising a constant domain (CL) and a linker. In one embodiment, The antibody domain and linker are arranged in one of the following orders from N-terminus to C-terminus: a) V H-CH1-linker-VL-CL, b)VL-CL-linker-VH-CH1, c)V Having H-CL-linker-VL-CH1 or d)VL-CH1-linker-VH-CL The linker consists of at least 30 amino acids, preferably 32 to 50 amino acids. It may be a lipeptide. The single-stranded Fab domain is between the CL domain and the CH1 domain. It is stabilized by natural disulfide bonds.

[0246] In one embodiment, the antibody domain and linker in the single-stranded Fab fragment are N From terminal to C-terminus, in one of the following sequences: a) VH-CH1-linker-VL-CL Or b) VL-CL-linker-VH-CH1, more preferably VL-CL-linker- It has VH-CH1.

[0247] In another embodiment, the antibody domain and linker in the single-stranded Fab fragment are From the N-terminus to the C-terminus, in one of the following sequences: a) VH-CL-linker-VL-CH 1) or b) has VL-CH1-linker-VH-CL.

[0248] Selectively, in single-stranded Fab fragments, the CL-domain and the CH1 domain In addition to the natural disulfide bond between them, antibody heavy chain variable domain (VH) and antibody light chain variable The variable domain (VL) is located at the following positions: i) position 44 of the heavy chain variable domain and the light chain variable domain 100th position, ii) heavy chain variable domain position 105 and light chain variable domain position 43, or iii) Heavy chain variable domain position 101 and light chain variable domain position 100 (Kabat EU index) Disulfide stabilization is also achieved by introducing disulfide bonds between the numbers (as indicated by the numbering system).

[0249] Such further disulfide stabilization of single-stranded Fab fragments is a significant development. This is achieved by introducing disulfide bonds between the variable domains VH and VL of the fragment. The technology for introducing non-natural disulfide bridges to stabilize single-strand Fv is, for example, See, International Publication No. 94 / 029350, Rajagopal et al. ,1997,Prot.Engin.10:1453-59;Kobayashi et. al.,1998,Nuclear Medicine&Biology,25:38 7-393; and Schmidt, et al., 1999, Oncogene 18: As described in 1711-1721, one embodiment involves a single-stranded Fab fragment. The optional disulfide bond between the variable domains is between the heavy chain variable domain at position 44 and the light chain variable domain. It is located between the domain and position 100. In one embodiment, the variable of the single-stranded Fab fragment The optional disulfide bond between domains is formed between the heavy chain variable domain at position 105 and the light chain variable domain. It is ranked between 43rd and 43rd (according to Kabat's EU index ranking).

[0250] 7.3.1.2.scFv A single-stranded Fv or "scFv" antibody fragment is a single polypeptide chain containing the V of the antibody. It contains H and VL domains and can be expressed as a single-chain polypeptide, This preserves the specificity of the intact antibody from which it originates. Generally, scFv polypeptides are scF v between the VH and VL domains allows the desired structure for target binding to form. Further comprising a polypeptide linker. Suitable for linking the VH and VL chains of scFV. Examples of linkers are the ABM linkers specified in Section 7.4.3, e.g., L1-L54. It is one of the linkers shown as follows.

[0251] Unless otherwise specified, when used herein, scFv means, for example, polyp The N-terminus and C-terminus of the plutide may have VL and VH variable regions in either order. scFv may include VL-linker-VH or VH-linker-VL.

[0252] To generate scFv coding nucleic acids, VH and VL coding DNA fragments are used. A linker that codes for a linker, for example, one of the ABM linkers listed in Section 7.4.3 (A Operable linking to another fragment encoding a MINOT (Gly4~Ser)3 sequence. With the VL and VH regions connected by a flexible linker, the VH and VL regions are connected. The sequence can be expressed as a continuous single-chain protein (for example, Bir d et al.,1988,Science 242:423-426;Huston et al.,1988,Proc.Natl.Acad.Sci.USA 85:5 879-5883;McCafferty et al.,1990,Nature 3 See 48:552-554.

[0253] 7.3.1.3. Other immunoglobulin-based modules MBM is Fab or scFv, e.g., Fv, dsFv, (Fab')2, single domain N antibody (SDAB), VH or VL domain, or camelid VHH ABMs that have immunoglobulin forms other than domains (also called nanobodies) may also be included. ru.

[0254] ABM consists of a single VH or VL domain that exhibits sufficient affinity for the target. It may be a single-domain antibody. In a particular embodiment, the single-domain antibody is a camel It is a VHH domain (for example, Riechmann, 1999, Journal o f Immunological Methods 231:25-38; International Publication No. 9 Please refer to the 4 / 04678 pamphlet.

[0255] 7.3.2. Non-immunoglobulin-based modules In certain embodiments, one or more ABMs are non-antibody scaffold proteins (designed A DARPin repeat protein (DARPin), avidity polymer (abbreviation for avidity polymer) ), anticarin / lipocalin, centintrin, Kunitz domain, adnexin, affine Phosphorus, aphytin (also known as nonphytin), nottin, pronectin, ba Ligands (including, but not limited to, suppositories, duocalin, and finomers), ligands , derived from receptors, cytokines, or chemokines.

[0256] Non-immunoglobulin scaffolds that can be used in MBM include Mintz and C. rea,2013,Bioprocess International 11(2): Tables 3 and 4, pp. 40-48; Vazquez-Lombardi et al., 2015. Figure 1 of Drug Discovery Today 20(10):1271-83, Table 1 and Figure I; Skrlec et al., 2015, Trends in Biot Listed in Table 1 and Box 2 of echnology 33(7):408-18 This includes Mintz and Crea, 2013, Bioprocess In Tables 3 and 4 of international 11(2):40-48; Vazquez-Lo mbardi et al.,2015,Drug Discovery Today Figure 1, Table 1 and Figure I from 20(10):1271-83; Skrlec et al.,2 015,Trends in Biotechnology 33(7):408-18 The contents of Table 1 and Box 2 (collectively referred to as "Disclosure of Scaffolding") are incorporated herein by reference. In certain embodiments, the disclosure of scaffolding is such that they are disclosed in relation to adnexin. The following is referred to by reference. In another embodiment, the disclosure of scaffolding is that they The disclosures relating to Bimmer are invoked by reference. In another embodiment, the foot The disclosures of the field are invoked by reference with respect to what they disclose with respect to the affiliate body. In yet another embodiment, the disclosure of the scaffolding is such that they are disclosed with respect to antikarin. The details are referred to by reference. In yet another embodiment, the disclosure of the scaffolding is This is referred to by reference in what they disclose regarding DARPin. In terms of form, the disclosure of scaffolding is what they disclose with respect to the Knitz domain. This is incorporated by reference. In yet another embodiment, the disclosure of scaffolding is such that they notching The disclosures relating to this are invoked by reference. In yet another embodiment, the The disclosures of the field are invoked by reference with respect to what they disclose with respect to pronectin. In yet another embodiment, the disclosure of scaffolds is such that they are disclosed with respect to nanophytin. The details are referred to by reference. In yet another embodiment, the disclosure of the scaffolding is The information disclosed by them regarding Affiliin is invoked by reference. In this context, the disclosure of scaffolds refers to what they disclose with respect to adnectin. It is more applicable. In yet another embodiment, the disclosure of scaffolding is that they are in relation to ABM. The disclosures are incorporated by reference. In yet another embodiment, the disclosure of scaffolding These are referred to by reference with respect to what they disclose regarding adheron. In the embodiment, the disclosure of the scaffolding is with respect to what they disclose to affiliates. This is incorporated by reference. In yet another embodiment, the disclosure of the scaffolding is that they are alpha The disclosures relating to the body are incorporated by reference. In yet another embodiment, The disclosure of the scaffolding is what they disclose regarding the armadillo repeat protein. This is incorporated by reference. In yet another embodiment, the disclosure of scaffolding is such that they are attached The disclosures regarding Mer / Tetranectin are referred to by reference. In embodiments, the disclosure of scaffolds is such that they are disclosed with respect to the obody / OB-fold. This is referred to by reference. In yet another embodiment, the disclosure of the scaffolding is Those are referred to with respect to what they disclose with respect to centinlin. In the construction method, the disclosure of scaffolding refers to what they disclose with respect to the repebody. This is invoked by. In yet another embodiment, the disclosure of the scaffolding is that they are anti-karin The disclosures relating to this are invoked by reference. In yet another embodiment, the The disclosures of the field are invoked by reference with respect to what they disclose with respect to the atrim. In yet another embodiment, the disclosure of scaffolds is disclosed in relation to bicyclic peptides. The details are referred to by reference. In yet another embodiment, the disclosure of the scaffolding is The disclosures made by them regarding the cys-knot are referred to by reference. In the construction method, the disclosure of the scaffolding indicates that they are Fn3 scaffolding (adnectin, centrilin, p The disclosures relating to ronectin and Tn3 are incorporated herein by reference.

[0257] In one embodiment, ABM is a designed ankyrin repeat protein ("DAR DARPin may be a highly specific and high-affinity target protein binding agent. These are antibody mimetic proteins that typically exhibit this characteristic. They are typically genetically modified Furthermore, derived from natural ankyrin protein, at least three of these proteins are Typically, it consists of four or five repeating motifs. Their molecular weight is the same as that of four or five repeating DA These are approximately 14 or 18 kDa (kilodaltons) respectively in RPin. An example of DARPin is: For example, see U.S. Patent No. 7,417,130. DARPin coupling module Multispecific binding molecules, including sulfates and immunoglobulin-based binding modules, are, for example, This is disclosed in U.S. Patent Application Publication 2015 / 0030596, Specification A1.

[0258] In another embodiment, the ABM may be an Affibody. The Affibody is an Affibody of the Technology It is well known in the field that staphylococcal protein A Based on a 58-amino acid residue protein domain derived from one of the IgG-binding domains This refers to affinity proteins.

[0259] In another embodiment, ABM may be antikarin. Antikarin is an art This is well known in the field and refers to another antibody mimetic technology, where binding specificity is It is derived from Pokarin. Antikarin is a dual-targeting protein called duocalin and It can be formatted even if it is formatted.

[0260] In another embodiment, the ABM may be a Versabody. The Versabody is an art This is well known in the field and refers to another antibody mimetic technology. These are typical proteins It replaces the hydrophobic core of the material, forming a high disulfide density scaffold, exceeding 15% It is a low molecular weight protein of 3-5 kDa that contains elucysteine.

[0261] Other non-immunoglobulin ABMs include "A" domain oligomers (also known as avimers). (For example, U.S. Patent Application Publication No. 2005 / 0164301, the same) See Specifications No. 2005 / 0048512 and No. 2004 / 017576. (For example, U.S. Patent Application Publication No. 2003 / 017) See Specification No. 0753), VASP polypeptide, bird pancreatic polypeptide (a PP), tetranectin (based on CTLD3), aphyrilin (γB-crystal) Phosphorus / ubiquitin-based), Notchin, SH3 domain, PDZ domain, Ten Damistat, neocardinostatin, protein A domain, lipocalin, transfer Examples include the ferrin and Knitz domains. In one embodiment, AB useful for constructing MBM M is fibronectin as exemplified in International Publication No. 2011 / 130324. Includes the base scaffolding.

[0262] Furthermore, in certain embodiments, the ABM is the ligand-binding domain or ligand of the receptor. It contains a receptor-binding domain. For example, if TAA is an EGF receptor, then ABM3 is It can contain a portion of EGF that binds to EGFR, and if TAA is a PDGF receptor ABM3 may include a portion of the PDGF receptor that binds to PDGF, etc. In this embodiment, ABM1 is a CD2 ligand, particularly as described in Section 7.9.2. This is the CD58 portion. The binding domains of numerous ligand / receptor pairs are part of this technology. It is well known in the field and therefore can be easily selected and adapted for use in MBM. Cut.

[0263] 7.4. Connectors CD3-binding molecules (e.g., MBM) can, in some cases, fuse without a linker. As a protein, ABM or pairs of ABM chains directly linked to each other (e.g., VH- of Fab) It is possible that it may include CH1 or VL-CL components. For example, a CD3-binding molecule. (For example, MBM) includes connector parts that connect individual ABMs or ABM chains. The use of the kuta moiety increases the flexibility of ABM within CD3-binding molecules (e.g., MBM), for example. Therefore, target binding can be improved by reducing steric hindrance. ABM is, for example If, then, Fc domains (each Fc domain represents a pair of associated Fc regions) and / or AB They can be linked to each other via M linkers. The use of Fc domains is typically for optimal antigens. For coupling, the use of a hinge region as a connector for ABM or ABM chains is required. Therefore, the term "connector" is not limited to Fc area, Fc domain, It includes the hinge region and the ABM linker.

[0264] Fc domain (formed by the pairing of two Fc regions), hinge region and ABM region Examples of nka are given in sections 7.4.1, 7.4.2, and 7.4.3, respectively.

[0265] 7.4.1. Fc Domain CD3-binding molecules (e.g., MBM) contain Fc domains derived from any preferred species. Obtain. In one embodiment, the Fc domain is derived from a human Fc domain.

[0266] The Fc domain includes IgA (including subclasses IgA1 and IgA2), IgD, and IgE. IgG (including subclasses IgG1, IgG2, IgG3, and IgG4) and IgM It may be derived from any suitable class of antibodies, including. In one embodiment, the Fc domain is It is derived from IgG1, IgG2, IgG3, or IgG4. In one embodiment, Fc domer The 'in' is derived from IgG1. In one embodiment, the Fc domain is derived from IgG4. ru.

[0267] Each Fc domain contains two polypeptide chains called heavy chain Fc regions. The heavy chain Fc region dimerizes to form an Fc domain. Two within the Fc domain The Fc regions may be the same or different from each other. In natural antibodies, the Fc region is typical Although typologically identical, for the purpose of generating multispecific binding molecules, such as MBMs, Fc The region enables heterodimerization, as described in Section 7.4.1.5 below. It may differ in terms of what is advantageous.

[0268] Typically, each heavy chain Fc region contains or comprises two or three heavy chain constant domains. Yes.

[0269] In natural antibodies, the heavy chain Fc region of IgA, IgD, and IgG consists of two heavy chain constant domains. Composed of ions (CH2 and CH3), the heavy chain Fc region of IgE and IgM consists of three heavy chains It consists of chain-constant domains (CH2, CH3, and CH4). These are Fc domains. Dimerize to produce [the desired result].

[0270] In this disclosure, the heavy chain Fc region is one or more different classes of antibodies, for example, one, two Alternatively, it may contain heavy chain constant domains derived from three different classes.

[0271] In one embodiment, the heavy chain Fc region consists of CH2 and CH3 domains derived from IgG1. Includes.

[0272] In one embodiment, the heavy chain Fc region consists of CH2 and CH3 domains derived from IgG2. Includes.

[0273] In one embodiment, the heavy chain Fc region consists of CH2 and CH3 domains derived from IgG3. Includes.

[0274] In one embodiment, the heavy chain Fc region consists of CH2 and CH3 domains derived from IgG4. Includes.

[0275] In one embodiment, the heavy chain Fc region includes a CH4 domain derived from IgM. The M CH4 domain is typically located at the C-terminus of the CH3 domain.

[0276] In one embodiment, the heavy chain Fc region comprises CH2 and CH3 domains derived from IgG and It contains a CH4 domain derived from IgM.

[0277] Used to generate the heavy chain Fc region for the CD3-binding molecule of this disclosure (e.g., MBM). The heavy chain constant domain for this purpose may include variants of the aforementioned natural constant domain. This will be understood. Such mutants have one or more differences compared to the wild-type constant domain. This may include a mino acid change. For example, the heavy chain Fc region of this disclosure is a wild-type constant domain and It contains at least one constant domain with a different sequence. The mutant constant domain is wild-type constant. It will be understood that this may be longer or shorter than the domain. Preferably, variant The steady-state domain is at least 60% identical or similar to the wild-type steady-state domain. In another example, the mutant constant domain is at least 70% identical or similar. In another example, the mutant constant domains are at least 75% identical or similar. In another example, the mutant constant domain is at least 80% identical or They are similar. In another example, the mutant constant domains are at least 90% identical. Or they are similar. In another example, the mutant constant domains are at least 95% identical. They are either similar or identical. Exemplary Fc variants are shown in sections 7.4.1.1 to 7.4.1.5. It is stated.

[0278] IgM and IgA are naturally occurring in humans as covalent multimers of a common H2L2 antibody unit. It exists naturally. IgM exists as a pentamer when the J chain is incorporated, or the J chain is incorporated. When it is absent, it exists as a hexamer. IgA exists in monomer and dimer forms. The heavy chains of IgM and IgA are divided into a C-terminal constant domain known as the tail. It has an 8-amino acid elongation. The tail portion forms disulfide bonds between heavy chains in the polymer. It contains cysteine ​​residues, which are thought to play an important role in polymerization. The tail portion is It also contains glycosylation sites. In certain embodiments, the CD3 binding molecule of this disclosure (for example) (MBM) does not include the tail.

[0279] The Fc domain incorporated into the CD3-binding molecule of this disclosure (e.g., MBM) is a protein Functional properties of the substance, such as serum half-life, complement binding, Fc receptor binding, and / or antigen-dependent properties. It may include one or more modifications that alter cytotoxicity. Furthermore, the CD3-binding molecule may be chemically modified. (For example, one or more chemical moieties may be bonded to the CD3 bond molecule) or its glyceride To alter cosylation and to change one or more functional properties of the CD3-binding molecule It may be modified for that purpose.

[0280] The effector function of antibody molecules is achieved, for example, by the binding of the complement C1 component to the antibody. This includes complement-mediated effector functions. Complement activation leads to the opsonization of pathogens. It is also important in direct lysis. Furthermore, it mobilizes phagocytic cells to the site of complement activation. And by activating it, it stimulates an inflammatory response. The effector function is Fc receptor It includes a body (FcR)-mediated effector function, which is the antibody to the Fc receptor (FcR). This can be triggered when the constant domain binds. Fc receptor on the surface of effector cells Antigen-antibody complex-mediated crosslinking involves the phagocytosis and destruction of antibody-coated particles, and the clearing of immune complexes. Alan, Lysis of antibody-coated target cells by killer cells (antibody-dependent cell-mediated cell therapy) Damaging (also called ADCC), release of inflammatory mediators, placental transfer and immunoglobulin It triggers many important and diverse biological responses, including the regulation of phosphorus production.

[0281] The Fc region has at least one amino acid residue to alter the effector function. It can be modified by substitution with different amino acid residues. For example, one or more amino acids The Fc region has a modified affinity for the effector ligand, It can be substituted with an amino acid residue. The effector ligand whose affinity for it is modified is For example, it could be an Fc receptor or a C1 component of complement. This method can be, for example, either Also, U.S. Patent No. 5,624,821 and No. 5,648,2 by Winter et al. As described in Specification No. 60, the modified Fc region also modifies the C1q bond and / or This method can reduce or eliminate complement-dependent cytotoxicity (CDC). For example, as described in U.S. Patent No. 6,194,551 by Idusogie et al. The modified Fc region can also alter the ability of the Fc region to bind complements. This technique can be used, for example, This is described in the PCT Publication International Publication No. 94 / 29351 by Bodmer et al. Jefferis et al., 2009, MAbs, 1:332-338. Therefore, as stated, the allotype amino acid residue is not limited to, but I The constant region of the heavy chains of the gG1, IgG2, and IgG3 subclasses, and the light chains of the κ isotype. The steady-state region is one example.

[0282] The Fc region "silences" effector functions, for example, antibody-dependent CD3 mediates cytotoxicity (ADCC) and / or antibody-dependent phagocytosis (ADCP). It can also be modified to reduce or eliminate the ability of the binding molecule. For example, Fc This can be achieved by introducing mutations in the region. Described in the technical field: LALA and N297A (Strohl, 2009, Curr.Opin.Biotechnol.20(6):685-691); and D2 65A(Baudino et al.,2008,J.Immunol.181:66 64-69; Strohl, above). An example of a silent Fc IgG1 antibody is IgG1 The so-called LALA mutation, which includes L234A and L235A mutations in the Fc amino acid sequence. Includes atypical forms. Another example of a silent IgG1 antibody includes a D265A mutation. Rent IgG1 antibody is found in the IgG1 Fc amino acid sequence D265A and P329A This includes so-called DAPA variants, which include mutations. Another silent IgG1 antibody is N29 It contains the 7A mutation, which is aglycosylated / nonglycosylated. This produces cosylated antibodies.

[0283] The Fc region, for example, increases the affinity of CD3-binding molecules to the activated Fcγ receptor. To reduce the affinity of CD3-binding molecules for inhibitory Fcγ receptors, By modifying one or more amino acid residues, antibody-dependent cytotoxicity (ADCC) and / Or the ability of CD3-binding molecules containing an Fc domain that mediates antibody-dependent cell phagocytosis (ADCP). It can be modified to improve. Examples of human activated Fcγ receptors include FcγRIa, F Examples include cγRIIa, FcγRIIIa, and FcγRIIIb, and human inhibitory Fcγ receptors One example of a suitable drug is FcγRIIb. This method is, for example, by Presta. It is described in PCT International Publication No. 00 / 42072 pamphlet. Furthermore, Fc Binding of γRl, FcγRII, FcγRIII, and FcRn in human IgG1 The sites are mapped, and mutants with improved binding are described (Sh ields et al.,J.Biol.Chem.276:6591-6604,2 (See 001). F of monoclonal antibodies such as ADCC / ADCP function improvement. The optimization of the c-mediated effector function is described (Strohl, 2009, Cu rrent Opinion in Biotechnology 20:685-69 See 1). Examples of mutations that can improve ADCC / ADCP function include G23. 6A, S239D, F243L, P247I, D280H, K290S, R292P, S 298A, S298D, S298V, Y300L, V305I, A330L, I332E E333A, K334A, A339D, A339Q, A339T and P396L (all) The location is determined by EU numbering and may be one or more mutations selected from the list.

[0284] The Fc region, for example, typically does not recognize parent CD3-binding molecules such as FcαRI, and is active. Modify one or more amino acids to increase the affinity of the CD3-binding molecule to the sex receptor. This improves the ability of CD3-binding molecules to mediate ADCC and / or ADCP. It can also be modified to allow this. This method is, for example, Borrok et al., 201 This is described in 5, mAbs.7(4):743-751.

[0285] Therefore, in certain embodiments, the CD3-binding molecule of this disclosure is, but is not limited to, F Binding to Fc- receptors such as cRn or leukocyte receptors (see, for example, Section 7.4.1.1) (as described in section 7.4.1.2), binding to the complement (for example, as described in section 7.4.1.2) ), modified disulfide bond structures (for example, as described in Section 7.4.1.3) or Modified glycosylation patterns (for example, as described in Section 7.4.1.4) It may include an Fc domain with modified effector functions. An Fc domain is, for example, This enables heterodimerization, which is the preferential pairing of non-identical Fc regions over identical Fc regions. This modification improves the manufacturability of asymmetric CD3-binding molecules (e.g., MBM). It can also be modified to include. Heterodimerization occurs when different ABMs have different sequences in the Fc region. CD3-binding molecules (e.g., MBM) linked together by Fc domains containing Enables generation. Examples of heterodimerization methods are given in Section 7.4.1.5 (and its subsections). This is illustrated in (Yon).

[0286] Any of the modifications described in Sections 7.4.1.1 to 7.4.1.5 may be used in any preferred manner. Combined by law, it can achieve the desired functional characteristics and / or can be combined with other modifications. Therefore, it will be understood that the properties of CD3-binding molecules (e.g., MBM) can be modified.

[0287] 7.4.1.1. Modified Fc domains with FcR binding The Fc domain of a CD3-binding molecule (e.g., MBM) interacts with the corresponding innate immunoglobulin. By comparison, it may show modified binding to one or more Fc receptors (FcRs). Binding to specific Fc receptors may be increased or decreased. In one embodiment, Fc dormancy The in contains one or more modifications that alter its Fc-receptor binding profile.

[0288] Human cells were selected from FcαR, FcεR, FcγR, FcRn, and glycan receptors. Some cells may express several membrane-bound FcRs. Some cells have a soluble (extracellular domain) It is also possible to express FcR (as investigated by Fridman et al.). , 1993, J Leukocyte Biology 54:504-512). Fc γR is further influenced by its affinity for IgG binding (high / low) and biological effects (activation / inhibition). It can be divided into two categories. Human FcγRI is widely considered to be the only "high affinity" receptor. While some are present, all others are considered to have moderate to low affinity. FcγRIIb is one of them. While it is the only receptor with "inhibitory" functionality due to its intracellular ITIM motif, it is the only receptor that possesses this function. Therefore, it is either "activated" by the ITAM motif or pairs with a common FcγR--γ chain. It is thought that FcγRIIIb is active, but it is filtered via the GPI anchor. It is also unique in that it associates with cells. Overall, humans have six "standard" FcγRs:F cγRI, FcγRIIa, FcγRIIb, FcγRIIc, FcγRIIIa, Fc It expresses γRIIIb. In addition to these sequences, it spreads across these families. Numerous sequence or allotype variants exist. Some of these have important functional consequences. It is known to possess this, and therefore it is considered to be a receptor subtype of its own. This can happen. For example, FcγRIIa H134R FcγRIIb I190T , Fc γRIIIa F158V and FcγRIIIb NA1 FcγRIIIb NA2 , Fcγ RIII SH These include: Each receptor sequence is IgG:IgG1, IgG2, IgG3 and It was shown to have different affinities to four subclasses of IgG4 (Bruh ns, 1993, Blood 113:3716-3725). Other species have FcγR It has several different numbers and functionalities, and the mouse strain has been the most thoroughly tested so far, with four Fcγ R, FcγRI, FcγRIIb, FcγRIII, FcγRIV (Bruhns) (2012, Blood 119:5640-5649). Human FcγRI on cells, Typically, its affinity for IgG1 / IgG3 / IgG4 is (approximately 10 -8 M) and serum Due to these IgG concentrations (approximately 10 mg / ml), monomeric IgG is obtained under normal serum conditions. Therefore, it is considered to be "occupied". Consequently, FcγRI is retained on their surfaces. Cells possessing this "scratch" of their antigenic environment by the bound multispecific IgG. It is considered possible to "lean" or "sample" the IgG subclass. Lower affinity (approximately 10 -5 ~10 -7 Other receptors that have a range of M are usually, It is thought that they are "not occupied." Therefore, low affinity receptors are involved in antibody-mediated immunity. It is inherently sensitive to the detection and subsequent activation of the complex within the antibody immune complex. Increased Fc density increases the functional affinity of binding avidity to low affinity FcγR. This has been demonstrated in vitro using several methods (Shields). et al., 2001, J Biol Chem 276(9):6591-6604 ;Lux et al.,2013,J Immunol 190:4315-4323 ). This is the primary mechanism of action in the use of anti-RhD to treat ITP in humans. It was also suggested that it is one of the codes (Crow, 2008, Transfusion M edicine Reviews 22:103-116).

[0289] Many cell types express multiple types of FcγR, and therefore, cells possessing FcγR... The binding of IgG or antibody immune complexes to cells can result in multiple and complex consequences depending on the biological context. It may have. In short, a cell receives either an activating, inhibiting, or mixed signal. To obtain. This involves phagocytosis (e.g., macrophages and neutrophils), antigen processing (e.g., For example, dendritic cells, decreased IgG production (e.g., B cells), or degranulation (e.g., neutrophils) This can lead to events such as mast cells. Inhibitory signals from FcγRIIb can trigger activity signals. There is data to support the possibility that it may be superior to Naru's (Proulx, 2010, Clinical Immunology 135:422-429).

[0290] FcRn plays an important role in maintaining the long half-life of IgG in the serum of adults and children. The receptor binds to IgG in acidified vesicles (pH < 6.5) and then processes the IgG. It protects the offspring from decomposition and then releases it in the blood at a higher pH of 7.4.

[0291] Unlike leukocyte Fc receptors, FcRn instead exhibits structural similarity to MHC class I molecules. It has a property. It is a membrane-bound chain containing three extracellular domains, with non-covalently bonded β2- It is a heterodimer composed of microglobulin chains. These include carbohydrate chains. One of the main ones is between the CH2 and CH3 domains of Fc, together with β2-microglobulin. It interacts with the site. The interaction is with histidine on positively charged IgG at pH < 6.5. Contains salt crosslinks formed on the residues. At higher pH levels, the His residues lose their positive charge. Loss of this interaction weakens the FcRn-IgG interaction, causing IgG to dissociate.

[0292] In one embodiment, a CD3-binding molecule (e.g., MBM) binds to human FcRn. Includes FC domains.

[0293] In one embodiment, the Fc domain also contains histidine at position 310, preferably at position 435. It has one (e.g., one or two) Fc regions containing histidine residues. This is important for human FcRn binding. In one embodiment, at positions 310 and 435 The histidine residues in this case are native residues, meaning that positions 310 and 435 are modified. No. Instead, one or both of these histidine residues may be present as a result of the modification. .

[0294] A CD3-binding molecule (e.g., MBM) modifies one or more F3 bonds to FcRn. It may include the c region. Modified bonds may be increased or decreased bonds.

[0295] In one embodiment, the CD3-binding molecule (e.g., MBM) is at least one (optional) The Fc region (selectively both) has higher affinity and ability than the corresponding innate immunoglobulin. It contains an Fc domain that has one or more modifications to bind to FcRn at the DITY.

[0296] In one embodiment, the Fc region has a threonine residue at position 250 that is a glutamine residue It is modified by replacing it with (T250Q).

[0297] In one embodiment, the Fc region replaces the methionine residue at position 252 with a tyrosine residue ( It is modified by replacing it with M252Y).

[0298] In one embodiment, the Fc region replaces the serine residue at position 254 with a threonine residue (S It is modified by substitution with 254T).

[0299] In one embodiment, the Fc region has a glutamic acid residue at the threonine residue at position 256. It is modified by substitution with the base (T256E).

[0300] In one embodiment, the Fc region replaces the threonine residue at position 307 with an alanine residue ( It is modified by substituting T307A).

[0301] In one embodiment, the Fc region replaces the threonine residue at position 307 with a proline residue ( It is modified by replacing it with T307P).

[0302] In one embodiment, the Fc region replaces the valine residue at position 308 with a cysteine ​​residue (V It is modified by substitution with 308C).

[0303] In one embodiment, the Fc region has a valine residue at position 308, leaving a phenylalanine residue. It is modified by substitution with base (V308F).

[0304] In one embodiment, the Fc region replaces the valine residue at position 308 with a proline residue (V3 It is modified by replacing it with 08P).

[0305] In one embodiment, the Fc region replaces the glutamine residue at position 311 with an alanine residue ( It is modified by substitution with Q311A).

[0306] In one embodiment, the Fc region replaces the glutamine residue at position 311 with an arginine residue. It is modified by substituting (Q311R).

[0307] In one embodiment, the Fc region replaces the methionine residue at position 428 with a leucine residue ( It is modified by replacing it with M428L).

[0308] In one embodiment, the Fc region has a histidine residue at position 433 replaced by a lysine residue (H It is modified by replacing it with 433K).

[0309] In one embodiment, the Fc region has an asparagine residue at position 434 that is phenylalanine. It is modified by substitution with a nin residue (N434F).

[0310] In one embodiment, the Fc region replaces the asparagine residue at position 434 with a tyrosine residue. It is modified by substituting (N434Y).

[0311] In one embodiment, the Fc region has a methionine residue at position 252 replaced by a tyrosine residue. , the serine residue at position 254 is replaced with a threonine residue, and the threonine residue at position 256 By substituting the group with glutamic acid residues (M252Y / S254T / T256E) It is modified.

[0312] In one embodiment, the Fc region has a valine residue at position 308 replaced with a proline residue, and The asparagine residue at position 434 is replaced with a tyrosine residue (V308P / N434Y). It is modified by doing so.

[0313] In one embodiment, the Fc region has a methionine residue at position 252 replaced by a tyrosine residue. , the serine residue at position 254 is replaced with a threonine residue, and the threonine residue at position 256 is replaced with a threonine residue In the glutamate residue, the histidine residue at position 433 is replaced with a lysine residue at position 434. The asparagine residue in the phenylalanine residue (M252Y / S254T / T256E It is modified by replacing it with / H433K / N434F).

[0314] Any combination of the modifications listed above can alter the FcRn bond. It will be understood.

[0315] In one embodiment, the CD3-binding molecule (e.g., MBM) has one or both Fc regions. However, the Fc domain has lower affinity and avidity than the corresponding natural immunoglobulin. It contains an Fc domain that has one or more modifications to bind to Rn.

[0316] In one embodiment, the Fc region has histidine at position 310 and / or 435. Contains any amino acid residue other than those mentioned above.

[0317] CD3-binding molecules (e.g., MBM) have one or both Fc regions that can be linked to FcγRIIb. It may contain an Fc domain that includes one or more modifications that increase its binding. FcγRIIb It is the only inhibitory receptor in humans and the only Fc receptor found in B cells.

[0318] In one embodiment, the Fc region has a proline residue at position 238 that is aspartic acid It is modified by substitution with the base (P238D).

[0319] In one embodiment, the Fc region is an alanine residue with a glutamic acid residue at position 258. It is modified by replacing it with (E258A).

[0320] In one embodiment, the Fc region replaces the serine residue at position 267 with an alanine residue (S2 It is modified by substitution with 67A).

[0321] In one embodiment, the Fc region has a serine residue at position 267 that is a glutamic acid residue ( It is modified by substituting S267E).

[0322] In one embodiment, the Fc region has a leucine residue at position 328 that is phenylalanine. It is modified by substitution with residue (L328F).

[0323] In one embodiment, the Fc region is an alanine residue with a glutamic acid residue at position 258. Furthermore, the serine residue at position 267 is replaced with an alanine residue (E258A / S267A). It is modified by doing so.

[0324] In one embodiment, the Fc region has a serine residue at position 267 replaced by a glutamic acid residue. Furthermore, the leucine residue at position 328 is replaced with a phenylalanine residue (S267E / L328F It is modified by replacing it with ).

[0325] Any combination of the modifications listed above can increase FcγRIIb binding. This will be understood.

[0326] In one embodiment, a CD3 binding containing an Fc domain exhibiting reduced binding to FcγR is observed. A molecule (e.g., MBM) is provided.

[0327] In one embodiment, the CD3-binding molecule (e.g., MBM) has one or both Fc regions. However, it contains an Fc domain that includes one or more modifications that reduce Fc binding to FcγR.

[0328] The Fc domain may originate from IgG1.

[0329] In one embodiment, the Fc region replaces the leucine residue at position 234 with an alanine residue (L It is modified by substitution with 234A).

[0330] In one embodiment, the Fc region replaces the leucine residue at position 235 with an alanine residue (L It is modified by substitution with 235A).

[0331] In one embodiment, the Fc region replaces the glycine residue at position 236 with an arginine residue ( It is modified by substituting G236R).

[0332] In one embodiment, the Fc region replaces the asparagine residue at position 297 with an alanine residue It is modified by substitution with (N297A) or a glutamine residue (N297Q).

[0333] In one embodiment, the Fc region replaces the serine residue at position 298 with an alanine residue (S2 It is modified by substitution with 98A).

[0334] In one embodiment, the Fc region replaces the leucine residue at position 328 with an arginine residue ( It is modified by replacing it with L328R).

[0335] In one embodiment, the Fc region replaces the leucine residue at position 234 with an alanine residue. Furthermore, the leucine residue at position 235 is replaced with an alanine residue (L234A / L235A). It is modified by...

[0336] In the embodiment, the Fc region has an alanine residue at the phenylalanine residue at position 234. Furthermore, the leucine residue at position 235 is replaced with an alanine residue (F234A / L235A). It is modified by substitution.

[0337] In one embodiment, the Fc region replaces the glycine residue at position 236 with an arginine residue. Furthermore, the leucine residue at position 328 is replaced with an arginine residue (G236R / L328R). It is modified by substitution.

[0338] Any combination of the modifications listed above may reduce the number of FcγR bonds. It will be understood.

[0339] In one embodiment, the CD3-binding molecule of the present disclosure (e.g., MBM) is one or both The Fc region does not affect the binding of Fc to FcγRII, but rather the binding of Fc to FcγRIIIa. It contains an Fc domain that includes one or more modifications that reduce the concentration.

[0340] In one embodiment, the Fc region replaces the serine residue at position 239 with an alanine residue (S2 It is modified by substitution with 39A).

[0341] In one embodiment, the Fc region is defined as an alanine residue with a glutamic acid residue at position 269. It is modified by replacing it with (E269A).

[0342] In one embodiment, the Fc region is an alanine residue with a glutamic acid residue at position 293. It is modified by replacing it with (E293A).

[0343] In one embodiment, the Fc region has a tyrosine residue at position 296 that is phenylalanine. It is modified by substitution with the residue (Y296F).

[0344] In one embodiment, the Fc region has a valine residue at position 303 replaced with an alanine residue (V3 It is modified by substituting with 03A).

[0345] In one embodiment, the Fc region replaces the alanine residue at position 327 with a glycine residue (A It is modified by substituting with 327G).

[0346] In one embodiment, the Fc region has a lysine residue at position 338 replaced by an alanine residue (K3 It is modified by substitution with 38A).

[0347] In one embodiment, the Fc region has an alanine residue at the 376th position of the aspartic acid residue. It is modified by substitution with the base (D376A).

[0348] Any combination of the modifications listed above may reduce the FcγRIIIa bond. It will be understood that...

[0349] Fc region mutants with reduced FcR binding are called "FcγR ablation mutants" or "Fc These are called "γR silencing mutants" or "Fc knockout (FcKO or KO) mutants." It could be discovered. For some therapeutic applications, one or more methods are used to avoid further mechanisms of action. The above or all Fcγ receptors (e.g., FcγR1, FcγRIIa, FcγRIIb, It is desirable to reduce or eliminate the normal binding of the Fc domain to FcγRIIIa). That is, for example, in many embodiments, the use of MBMs that are monovalently bound to CD3 In its use, FcγRIIIa binding is used to eliminate or significantly reduce ADCC activity. Cutting is generally desirable. In one embodiment, the Fc of the MBM described herein At least one region contains one or more Fcγ receptor ablation variants. In the application morphology, both Fc regions contain one or more Fcγ receptor ablation variants. These ablation mutants are shown in Table 3, and each was independently and arbitrarily selected. G236R / L328R, E233P / L234V / L235A / G236del / S2 39K, E233P / L234V / L235A / G236del / S267K, E233 P / L234V / L235A / G236del / S239K / A327G, E233P / L234V / L235A / G236del / S267K / A327G and E233P / L 234V / L235A / G236del ("del" means missing, for example, G236) Ablation mutations selected from a group consisting of (del refers to the deletion of glycine at position 236) Several embodiments utilizing the different body may be included or excluded. Ablation mutants cleave FcγR bonds, but generally cleave FcRn bonds. Please note that this is not the case.

[0350] [Table 84]

[0351] In one embodiment, the CD3-binding molecule of the present disclosure (e.g., MBM) is a first Fc region Includes the region and the second Fc region. In one embodiment, the first Fc region and / or the second The Fc region has the following mutations: E233P, L234V, L235A, G236del and It may include S267K.

[0352] The Fc domain of human IgG1 has the highest binding to the Fcγ receptor, thus forming a heterodimer. If the constant domain (or Fc domain) in the antibody backbone is IgG1, then abrasion Mutant variants can be used.

[0353] Alternatively, or in addition to IgG1 background ablation variants, glycosyl Mutations at position 297, for example, the alanine residue of the asparagine residue at position 297 (N297A Substitution by ) or glutamine residue (N297Q) is, for example, binding to FcγRIIIa. Significant cleavage occurs. Human IgG2 and IgG4 have naturally reduced binding to the Fcγ receptor. Therefore, these skeletons can be used with or without ablation mutants.

[0354] 7.4.1.2. Modified Fc domains with complement binding CD3-binding molecules (e.g., MBM) have one or both Fc regions that bind to complement. It may contain an Fc domain that includes one or more modifications that alter it. The modified complement binding is increased. It may be a bonded or reduced bond.

[0355] In one embodiment, the Fc region is subjected to one or more modifications that reduce its binding to C1q. Includes. The initiation of the classical complement pathway involves the hexamerization of antigen-binding IgG and IgM into the CH2 domain. The process begins with the binding of the C1q protein.

[0356] In one embodiment, the CD3-binding molecule (e.g., MBM) has one or both Fc regions. However, it contains an Fc domain that includes one or more modifications that reduce Fc binding to C1q.

[0357] In one embodiment, the Fc region replaces the leucine residue at position 234 with an alanine residue (L It is modified by substitution with 234A).

[0358] In one embodiment, the Fc region replaces the leucine residue at position 235 with an alanine residue (L It is modified by substitution with 235A).

[0359] In one embodiment, the Fc region is defined as a glutamic acid residue at the 235th position. It is modified by replacing it with (L235E).

[0360] In one embodiment, the Fc region replaces the glycine residue at position 237 with an alanine residue (G It is modified by substitution with 237A).

[0361] In one embodiment, the Fc region has a lysine residue at position 322 replaced by an alanine residue (K3 It is modified by substitution with 22A).

[0362] In one embodiment, the Fc region replaces the proline residue at position 331 with an alanine residue (P It is modified by substitution with 331A).

[0363] In one embodiment, the Fc region has a proline residue at position 331 replaced by a serine residue (P3 It is modified by substitution with 31S).

[0364] In one embodiment, the CD3-binding molecule (e.g., MBM) is derived from IgG4 Fc It includes the domain. IgG4 originally has a lower complement activation profile than IgG1. Furthermore, it also has a weaker bond to FcγR. Therefore, in one embodiment, CD 3-binding molecules (e.g., MBM) contain an IgG4 Fc domain and increase FcγR binding. Includes one or more modifiers that cause something to happen.

[0365] Any combination of the modifications listed above can reduce the number of C1q bonds. It will be understood.

[0366] 7.4.1.3. Fc domains having modified disulfide structures CD3-binding molecules (e.g., MBM) generate and / or remove cysteine ​​residues. It may contain an Fc domain that includes one or more modifications. The cysteine ​​residue is a polypeptide mono By forming disulfide bridges between individual pairs of mers, Fc-based multiple specific bonds are formed. It plays an important role in the spontaneous assembly of composite molecules. Therefore, the number of cysteine ​​residues and / or modify the structure of a CD3-binding molecule (e.g., MBM) by altering its position. This makes it possible to produce proteins with improved therapeutic properties.

[0367] The CD3-binding molecule of this disclosure (e.g., MBM) comprises one or both Fc regions, preferably Both Fc regions may contain an Fc domain with a cysteine ​​residue at position 309. In this state, the cysteine ​​residue at position 309 is generated by modification, for example, IgG In the case of the Fc domain derived from 1, the leucine residue at position 309 is a cysteine ​​residue (L Substituted at 309C), in the case of an Fc domain derived from IgG2, the barrier at position 309 The n residue is replaced with a cysteine ​​residue (V309C).

[0368] In one embodiment, the Fc region replaces the valine residue at position 308 with a cysteine ​​residue (V It is modified by substitution with 308C).

[0369] In one embodiment, two disulfide bonds in the hinge region are in the core hinge array Removal by mutating CPPC (sequence number 9) into SPPS (sequence number 14). It will be done.

[0370] 7.4.1.4. Modified glycosylated Fc domains In certain embodiments, the immunoglobulin contains fewer glycosylation sites than the corresponding immunoglobulin. CD3-binding molecules (e.g., MBM) with the above-mentioned manufacturability are provided. Proteins have a simpler post-translational glycosylation pattern and are therefore simpler. It is cheaper to manufacture.

[0371] In one embodiment, the glycosylation site in the CH2 domain is at position 297. Replace the asparagine residue with an alanine residue (N297A) or a glutamine residue (N297Q). They are removed by replacement. In addition to improved manufacturability, these glycosyl saturates The mutant also reduces the FcγR binding described herein.

[0372] In one embodiment, a low-fucosylated antibody having a reduced amount of fucosyl residues or Modified types of glycosylation, such as antibodies with increased bifurcation GlcNac structures. CD3-binding molecules having such a modified glycosylation pattern can be created. It has been demonstrated that this improves the ADCC ability of antibodies. Such carbohydrate modifications are For example, expressing a CD3-binding molecule in a host cell having a modified glycosylation mechanism This can be achieved by the following. Cells having a modified glycosylation mechanism are in the art. It is described that the CD3 binding molecule is expressed internally, thereby modifying the glyco It can be used as a host cell that produces CD3-binding molecules with sylation. For example, Han European Patent No. 1,176,195 by g et al. describes fucosyltransferase Cell lines possessing a functionally disrupted FUT8 gene that encodes it have been described, and Antibodies expressed in eel cell lines exhibit low fucosylation. In the PCT Publication International Publication No. 03 / 035835 by ta, fucose is A It has a reduced ability to bind sn(297) linked carbohydrates, and in the host cells in question... A mutant CHO cell line, Lecl3 cells, which also exhibits low fucosylation of expressed antibodies, has been described. (Shields et al., 2002, J. Biol. Chem. 27) See 7:26733-26740). International Publication of PCT Publications by Umana et al. Pamphlet No. 99 / 54342 contains information on glycoprotein-modified glycosyltransferases. (For example, β(1,4)-N-acetylglucosaminyltransferase III (GnT III)) has been described as a modified cell line that expresses the modified cell line The antibodies expressed in this way result in increased bifacial GlcNa, which leads to increased ADCC activity of the antibodies. It now shows the c structure (Umana et al., Nat. Biotech.). See also 17:176-180, 1999).

[0373] 7.4.1.5. Fc heterodimerization Unlike innate immunoglobulins, many multispecific molecular forms have non-identical antigen-binding domains. Two operably connected to (or that part, e.g., VH or VH-CH1 of Fab) This involves dimerization between the Fc regions. Insufficient two Fc regions to form an Fc domain. Heterodimerization is always an obstacle to increasing the yield of the desired multispecific molecules. Therefore, it is a difficult problem for purification. Various methods available in this field include, for example, For example, European Patent Application Publication No. 1870459A1; U.S. Patent No. 5,582,996 U.S. Patent No. 5,731,168; U.S. Patent No. 5,910,573 Detailed Statement; U.S. Patent No. 5,932,448; U.S. Patent No. 6,833,441 U.S. Patent No. 7,183,076; U.S. Patent Application Publication No. 2006204493A Specification No. 1; and PCT Publication Number International Publication No. 2009 / 089004A1 Pamphlet As disclosed herein, Fc regions may be present in the CD3-binding molecule (e.g., MBM) of this disclosure. It can be used to promote dimerization of the region.

[0374] This disclosure includes Fc heterodimers, i.e., Fc domains containing different non-identical Fc regions. We provide a CD3-binding molecule (e.g., MBM) using a heterodimerization method. Operablely connected to ABM (or a portion thereof, e.g., VH or VH-CH1 of Fab) Promotes dimerization of the Fc region and operably connects the same ABM or a portion thereof to the Fc region. It reduces dimerization. Typically, each Fc region in the Fc heterodimer is a CH3 group of the antibody. Includes the main. The CH3 domain can be any isotype, as described in the previous section. Rath or subclass, preferably IgG (IgG1, IgG2, IgG3, and IgG4) It originates from the constant region of the antibody class.

[0375] Typically, MBM includes the CH1 domain, CH2 domain, and H3 domain in addition to the CH3 domain. The domain, VH domain, VL domain, CDR and / or antigenic domain as described herein. Includes other antibody fragments such as composite fragments. In one embodiment, two hete A polypolypeptide is a set of two heavy chains that form a bispecific or multispecific molecule. CH3 Heterodimerization of two different heavy chains in a domain produces the desired antibody or antibody-like molecule. On the other hand, homodimerization of the same heavy chain reduces the yield of the desired antibody or molecule. In exemplary embodiments, two or more heteropolypeptide chains include a CH3 domain. and comprising two chains forming any of the multispecific molecular forms described above in this disclosure. In one embodiment, two heteropolypeptide chains containing CH3 domains are unmodified. Compared to chains, the modification method includes modifications that favorably act on the heterodimer association of polypeptides. Various examples are shown in Table 4 and sections 7.4.1.5.1 to 7.4.1.5.8 below. .

[0376] [Table 85]

[0377] [Table 86]

[0378] [Table 87]

[0379] [Table 88]

[0380] [Table 89]

[0381] [Table 90]

[0382] [Table 91]

[0383] [Table 92]

[0384] 7.4.1.5.1. Stereomorphs CD3 binding molecules (e.g., MBM) bind to one or more of the constant domains of the Fc domain. For example, it may include one or more modifications to the CH3 domain. Therefore, the CD3-binding molecule of this disclosure (e.g., MBM) binds to the constant domain of the heavy chain of an antibody, e.g., C3. It contains two polypeptides, each containing either an H2 or a CH3 domain. For example, CD The two heavy chain constant domains of a 3-linked molecule (e.g., MBM), e.g., CH2 or CH3 domain The in includes one or more modifications that enable heterodimer association between the two chains. In one embodiment, In one embodiment, one or more modifications are located on the CH2 domains of the two heavy chains. One or more modifications involve at least two polypeptides of the CD3-binding molecule (e.g., MBM). It will be placed on the CH3 domain.

[0385] One mechanism of Fc heterodimerization is generally known as "knob and hole" or "knob-into These terms are referred to as "hall." For example, Ridgway et al., 19 96,Protein Engineering 9(7):617;Atwell e t al.,1997,J.Mol.Biol.270:26;U.S. Patent No. 8,216, As described in Specification No. 805, there is a greater tendency to form Fc heterodimers than Fc homodimers. This refers to amino acid mutations that produce beneficial steric effects. Knob-in-hole mutations are heterodimers. It can be combined with other methods to improve the process.

[0386] In one embodiment, the first CD3 binding molecule (e.g., MBM) containing a heavy chain constant domain One or more modifications to a polypeptide can generate a "knob" and CD3 binding. One or more modifications to the second polypeptide of a child molecule (e.g., MBM) create a "hole". This process involves the heteracy of polypeptides containing heavy chain constant domains, such as CD3-binding molecules (e.g., MBM). Ronimerization causes the "knob" to link with the "hole" (for example, to interact, for example) The CH2 domain of the first polypeptide interacts with the CH2 domain of the second polypeptide. Either the CH3 domain of the first polypeptide becomes the CH3 domain of the second polypeptide (It is designed to interact with the 'N'. When this term is used herein, "N" is used. "B" is the first polypeptide of a CD3-binding molecule (e.g., MBM) containing a heavy chain constant domain. It protrudes from the boundary of the dot, thus stabilizing the heteropolymer, and thereby, for example, the homopolymer CD3 binding containing a heavy chain constant domain to act more favorably for heteromultimer formation than for general formation. A complementary "hole" at the boundary between a molecule (e.g., MBM) and a second polypeptide. Refers to at least one configurable amino acid side chain. The knob may be present at the original boundary. Alternatively, it can be introduced synthetically (for example, by modifying the nucleic acid encoding the boundary). The preferred import residues for knob formation are generally native amino acid residues. Preferably arginine (R), phenylalanine (F), tyrosine (Y), and tryptophosphate. Selected from phan(W). Most preferred are tryptophan and tyrosine. In a particular embodiment, the original residue for protrusion formation has a small side chain capacity, for example alanine, asparagine, aspartic acid, glycine, serine, threonine, or valine That is the case.

[0387] "Hole" is the second port of a CD3-binding molecule (e.g., MBM) that contains a heavy chain constant domain. It is recessed from the lipeptide boundary and therefore contains a CD3 binding molecule with a heavy chain constant domain. Suitable for the corresponding knob on the adjacent interaction surface of the first polypeptide (for example, MBM). Refers to at least one amino acid side chain that joins. Holes may be present at the original boundary, or It can be introduced synthetically (for example, by modifying the nucleic acid that codes for the boundary). The preferred import residues for the formation of the ligament are usually natural amino acid residues, and are preferably The compound is selected from alanine (A), serine (S), threonine (T), and valine (V). The most preferred are serine, alanine, or threonine. In a preferred embodiment... The original residues for hole formation have a large side chain capacity, such as tyrosine and arginine. It is either phenylalanine or tryptophan.

[0388] In a preferred embodiment, the first CH3 domain is residue 366, 405, or 407 Modified in the above, it generates either a "knob" or a "hole" (as described above) The second CH3 domain, which heterodimerizes with the first CH3 domain, has 366 residues. If the first CH3 domain is modified, residues 407 and 405 become the first CH When modified in 3 domains, residue 394 or residue 407 becomes the first CH3 domain. Modified in such a way that a "knob" or "hole" of the first CH3 domain is complemented by a "hole". When generating "ru" or "nobu", the modification occurs at residue 366.

[0389] In another preferred embodiment, the first CH3 domain is modified at residue 366. The second CH3 domain, which heterodimerizes with the first CH3 domain, is composed of residue 3. Modified in 66, 368 and / or 407, the “knob” of the first CH3 domain or This generates a "hole" or "knob" that is complementary to the "hole". In one embodiment, the first The modification of the CH3 domain involves introducing a tyrosine (Y) residue at position 366. In one embodiment, the first modification for CH3 is T366Y. The first modification to the CH3 domain is the tryptophan (W) residue at position 366. In one embodiment, the first modification for CH3 is T366W. In one embodiment, a heterozygous first CH3 domain modified at position 366 is present along with the heterozygous first CH3 domain. Dimerization, modification of the second CH3 domain (for example, introduced at position 366) For example, modified T366Y or T36 (Including 6W) Modification at position 366, modification at position 368 and modification at position 407 Includes ornamentation. In one embodiment, the modification at position 366 introduces a serine (S) residue. The modification at position 368 introduces alanine (A), and the modification at position 407 introduces valine. (V) is introduced. In one embodiment, the modifications are T366S, L368A and Y40 Contains 7V. In one embodiment, the first CH3 domain of the multispecific molecule is modified T3 The second CH3 domain contains 66Y and heterodimerizes with the first CH3 domain. This includes the modifiers T366S, L368A, and Y407V, or vice versa. One implementation form In this state, the first CH3 domain of the multispecific molecule contains modified T366W, and the first The second CH3 domain, which heterodimerizes with the CH3 domain, is modified T366S. This includes L368A and Y407V, and vice versa.

[0390] Further stereoscopic or "distortion" (e.g., knob-in-hole) modifications are required by the PCT publication number country. International Publication No. 2014 / 145806 (for example, International Publication No. 2014 / 145 Figures 3, 4, and 12 of pamphlet No. 806, PCT Publication Number International Publication No. 2014 / Pamphlet No. 110601 and PCT Publication Number International Publication No. 2016 / 086186 Brochure, International Publication No. 2016 / 086189, International Publication No. 2016 / As described in pamphlet No. 086196 and international publication pamphlet No. 2016 / 182751. These are included, and their entirety is incorporated herein by reference. Examples of KIH variants are L368D and K3 are paired with a second steady chain containing S364K and E357Q modifications. It includes a first steady chain with 70S modification.

[0391] Suitable for use in any of the CD3-binding molecules of this disclosure (e.g., MBM) Other knob-in-hole modification pairs include, for example, International Publication No. 1996 / 027011 PAN Fret and Merchant et al., 1998, Nat. Biotechno Further information can be found in l., 16:677-681.

[0392] In a further embodiment, the CH3 domain is configured to introduce a pair of cysteine ​​residues. Further modifications are possible. While not constrained by theory, disulfide bonds are formed. The introduction of a pair of cysteine ​​residues that can be used is a heterogeneous compound containing a paired CH3 domain. It is thought to provide stability to the merged CD3-binding molecule (e.g., MBM). Morphologically, the first CH3 domain contains cysteine ​​at position 354, and the first C The second CH3 domain, which heterodimerizes with the H3 domain, has a cyanotype at position 349. Contains stain. In one embodiment, the first CH3 domain is a stain at position 354. Stain (e.g., including modified S354C) and tyrosine (Y) at position 366 (for example) If it contains modified T366Y, it heterodimerizes with the first CH3 domain. The second CH3 domain contains cysteine ​​at position 349 (e.g., modified Y349C). ), serine at position 366 (e.g., including modified T366S), and ala at position 368. Nin (e.g., including modified L368A) and valine at position 407 (e.g., modified Y4 (including 07V). In one embodiment, the first CH3 domain is located at position 354. Cysteine ​​(e.g., including modified S354C) and tryptophan at position 366. (W) (e.g., including modified T366W) is heterozygous with the first CH3 domain. The second CH3 domain that dimerizes is the cysteine ​​at position 349 (e.g., modified Y3). (including 49C), serine at position 366 (e.g., including modified T366S), position 368 Alanine at (e.g., including modified L368A) and valine at position 407 (e.g.) (This includes the modified Y407V.)

[0393] Further mechanisms in which its use in heterodimer formation is found are found in Gunasekara n et al.,2010,J.Biol.Chem.285(25):19637 As described, it is sometimes called "electrostatic steering." They are sometimes called "charge pairs". In this embodiment, the formation is heterodimerization. Static electricity is used to direct and distort them. As will be understood by those skilled in the art, these This can affect pI, and consequently purification, and therefore, in some cases, pI mutants may be affected. It can also be considered as such. However, these are produced to force heterodimerization. Because they were not used as purification tools, they are classified as "stereomutants." It is not limited to D221R / P228R / K409R, but D221E / Compatible with P228E / L368E and C220R / E224R / P228R / K409R. This includes C220E / P228E / 368E.

[0394] Further variants include the pI variants outlined herein or those described in U.S. Patent Application Publication No. 2012. Other variants, such as the other stereomutants shown in Figure 37 of Specification No. / 0149876, and optional They can be combined selectively and independently.

[0395] In one embodiment, the stereomutants outlined herein are any pI mutant (or Along with other variants such as Fc variants and FcRn variants, optional in one or both Fc regions It can be selectively and independently incorporated, and independently or optionally, the CD3 coupling It can be included as a child or excluded.

[0396] A list of suitable distortion variants can be found in Table 5, and in many embodiments they have specific utility. Several pairs are shown. In many embodiments, but not limited to, S364K / E 357Q:L368D / K370S;L368D / K370S:S364K;L368E / K370S:S364K;T411T / E360E / Q362E:D401K;L36 8D / K370S:S364K / E357L;andK370S:S364K / E357 A pair of sets containing Q has a specific usage. Regarding nomenclature, "S364K / E357Q The pair ":L368D / K370S" has a double mutant set S364K / in one of its Fc regions. It has E357Q and the other has a double mutant set L368D / K370S. It means...

[0397] [Table 93]

[0398] [Table 94]

[0399] [Table 95]

[0400] [Table 96]

[0401] [Table 97]

[0402] In one embodiment, the CD3-binding molecule includes a first Fc region and a second Fc region. In one embodiment, the first Fc region has the following mutations: L368D and K370S The second Fc region includes the following mutations: S364K and E357Q. In one embodiment In this, the first Fc region contains the following mutations: S364K and E357Q, and the second F The c region contains the following mutations: L368D and K370S.

[0403] 7.4.1.5.2. Alternative knobs and holes: IgG heterodimerization The polypeptide chain of a CD3-binding molecule (e.g., MBM) containing a paired CH3 domain Heterodimerization involves one or more modifications in the CH3 domain derived from the IgG1 antibody class. It can be increased by introducing it. In one embodiment, the modification is a second CH3 domain Includes a K409R modification for one CH3 domain paired with an F405L modification within the molecule. Further modifications include, or instead, 366, 368, 370, 399, 405, 4 These may be located at positions 07 and 409. Preferably, the polypeptide heterozygotes include such modifications. Dimerization occurs under reducing conditions, for example, at 25-37°C, or at 25°C or 37°C for 1-10 hours. For example, 1.5 to 5 hours, for example, 10 to 100 mM 2-MEA over 5 hours (for example) The test is performed using 25, 50, or 100 mM 2-MEA.

[0404] The amino acid substitutions described herein are performed using techniques well known in the art, such as CH3 It can be introduced within a domain (e.g., McPherson, ed., 1991, Dire cted Mutagenesis:a Practical Approach;Ad See Elman et al., 1983, DNA, 2:183.

[0405] IgG heterodimerization methods are described, for example, in the pamphlet International Publication No. 2008 / 119353. , International Publication No. 2011 / 131746 and International Publication No. 2013 / 06 Further details are provided in pamphlet number 0867.

[0406] In any of the embodiments described in this section, the CH3 domain is as defined in 7.4.1. As described in Section 5.1, it can be further modified to introduce pairs of cysteine ​​residues. .

[0407] 7.4.1.5.3. pI (isoelectric point) mutants Generally, as understood by those skilled in the art, pI variants fall into two common categories. -: substances that increase the pI of proteins (basicity change) and substances that decrease the pI of proteins There are some (acidic changes). All combinations of these variants as described herein This can be done: One of the Fc regions does not show a pI that is wild-type or significantly different from wild-type. One variant may be more basic or more acidic. Alternatively, each Fc region may be One becomes more basic, and the other becomes more acidic.

[0408] Exemplary combinations of pI variants are shown in Table 6. These are outlined herein and shown in Table 6. As shown, these changes are compared to IgG1, but all isotypes are different. Similar to sotype hybrids, this method can be used to modify the heavy chain constant domain. If derived from IgG2-4, R133E and R133Q can also be used.

[0409] [Table 98]

[0410] In one embodiment, for example, Figures 1B-1W, 1Y-1AH, 2B-2L and Figure 2 In the N~2V format, the combination of pI mutants is 208D / 295E / 384D / 41 8E / 421D mutant (N208D / Q295E / N38 in relation to human IgG1) One Fc region (negative Fab side) including 4D / Q418E / N421D and positively charged The scFv linker, for example, the second Fc region (explicit) including L36 (described in Section 7.4.3) It has the scFv side. However, as can be understood by those skilled in the art, the first Fc The region includes the CH1 domain, which includes position 208. Therefore, it does not include the CH1 domain. A structure (for example, in the form shown in Figure 2K, for example, CH1 as one of the domains) In the case of MBM that does not utilize domains, the negative pI mutant Fc set is 295E / 384D / 418E / 421D variants (Q295E / N in the case of human IgG1) This may include 384D / Q418E / N421D).

[0411] In one embodiment, the first Fc region has a set of substitutions from Table 6, and the second F The c region is in contact with a charged linker (for example, selected from those described in Section 7.4.3). It will continue.

[0412] In one embodiment, the CD3-binding molecule of the present disclosure comprises a first Fc region and a second Fc region Includes the region. In one embodiment, the first Fc region has the following mutation: N208D, Q Includes 295E, N384D, Q418E and N421D. In one embodiment, the second The Fc region has the following mutations: N208D, Q295E, N384D, Q418E and Includes N421D.

[0413] 7.4.1.5.4. Isotope Variants Furthermore, many embodiments of this disclosure involve transferring from one IgG isotype to another isotype. It depends on the "import" of pI amino acids at specific positions, and therefore is undesirable. Reduce or eliminate the possibility of immunogenicity being introduced into the mutant. Many of these are U.S. patent applications. This is shown in Figure 21 of Publication No. 2014 / 0370013. That is, IgG1 For various reasons, including its high effector function, it is a common isotype of therapeutic antibodies. However, the heavy chain constant region of IgG1 has a higher pI than that of IgG2. (8.10 vs. 7.31). Introducing IgG2 residues into the IgG1 backbone at specific positions. This results in a decrease (or increase) in the pI of the Fc region, and a longer serum half-life. This shows that, for example, IgG1 has glycine at position 137 (pI 5.97), and IgG2 has It contains glutamic acid (pI 3.22); when glutamic acid is imported, as a result This affects the pI of the resulting protein. As explained below, the pI of the mutant antibody is significant. In general, many amino acid substitutions are required to have an effect. However, as explained below... As will be shown, it should be noted that even changes in the IgG2 molecule can extend the serum half-life. sea ​​bream.

[0414] In other embodiments, as will be further described below, the resulting protein To reduce the overall charge state (for example, by changing the pI of amino acids to lower pI) (By changing to amino acids), or to allow for structural modification for stability. Non-isotype amino acids are modified.

[0415] Furthermore, both the heavy and light chain constant domains of the CD3-binding molecule containing two half-antibodies are pI By modifying the antibodies, significant changes can be observed in each half-antibody. The pI of the two half-antibodies is A difference of at least 0.5 indicates that ion exchange chromatography or isoelectric focusing or This allows for separation by other methods that are sensitive to the isoelectric point.

[0416] 7.4.1.5.5. Calculation of pI The pI of a semi-antibody containing the Fc region and ABM or ABM chain is the p of the mutant heavy chain constant domain. I depends on the pI of the mutant heavy chain constant domain and the whole half antibody containing ABM or ABM chain. Therefore, in one embodiment, the change in pI is obtained in U.S. Patent Application Publication No. 2014. Using the chart in Figure 19 of Specification No. / 0370013, based on the mutant heavy chain constant domain The calculation is performed accordingly. As described herein, the half-antibody to be modified is generally a half-antibody. This is determined by the unique pI of each antibody. Alternatively, the pI of each half-antibody can be compared.

[0417] 7.4.1.5.6. pI variants that also confer better FcRn in vivo binding. If the pI variant reduces the pI in the Fc region, it improves serum retention in vivo. It may have further advantages.

[0418] In endosomes, when FcRn binds at pH 6, Fc is sequestered, thus pI mutation The Fc region of the body is thought to contribute to the longer half-life of antigen-binding molecules in vivo. Ghetie and Ward,1997,Immunol Today.18(12 (592-598). Next, the endosomal compartment recycles Fc on the cell surface. It rotates. When the compartment opens into the extracellular space, the pH becomes approximately 7.4, and the blood Fc Release to is induced. In mice, Dall' Acqua et al. found that p Fc variants that show increased FcRn binding at H6 and pH 7.4 actually show decreased serum concentration. It was shown to have the same half-life as wild-type Fc (Dall' Acqua et al. ,2002,J.Immunol.169:5171-5180). F at pH 7.4 The increased affinity of Fc for cRn is thought to hinder the release of Fc into the bloodstream. Therefore, Fc mutations that increase the half-life of Fc in vivo are, ideally, higher Increased FcRn binding at lower pH levels while still allowing Fc release at pH. The amino acid histidine changes its charge state in the pH range of 6.0 to 7.4. Therefore, it is surprising to find His residues at important positions in the Fc / FcRn complex. That's not something that should be done.

[0419] Antibodies with a variable region having a lower isoelectric point may also have a longer serum half-life. This has been suggested (Igawa et al., 2010, PEDS. 23(5):385). -392). However, this mechanism is not yet fully understood. Furthermore, the variable range This varies depending on the antibody. As described herein, pI decreases and half-life is extended. Steady-state region mutants are more modular for improving the pharmacokinetic properties of CD3-binding molecules. This will provide a suitable approach.

[0420] 7.4.1.5.7.Polar crosslinking Heterodimeric polypeptide chain of CD3-binding molecules containing an Fc domain (e.g., MBM) Embodied substances can be increased by introducing modifications based on the principle of "polar crosslinking," and "polar The principle of "crosslinking" is to change the residues at the binding boundary of two polypeptide chains into a heterodimer form. While interacting with residues of similar (or complementary) physical properties, in homodimer form In this context, residues with different physical properties interact with each other. In particular, these modifications result in heterodimers. During formation, polar residues interact with other polar residues, while hydrophobic residues interact with other hydrophobic residues. It is designed to act in a certain way. In contrast, in homodimerization, the residues are polar residues. Modified to interact with aqueous residues. Preferred interaction in heterodimer form. Furthermore, undesirable interactions in the homodimer form are due to the Fc region acting as a homodimer. The goal is to create a stronger tendency to form heterodimers than to form whole bodies.

[0421] In an exemplary embodiment, the above modification is performed on residues 364, 368, and 3 of the CH3 domain. It is generated at one or more of the following positions: 99, 405, 409, and 411.

[0422] In one embodiment, S364L, T366V, L368Q, N399K, F405 One or more modifications selected from S, K409F, and R411K are applied to two CH3 domains. It will be introduced as one. One or more repairs selected from Y407F, K409Q and T411N. Ornamentation may be introduced into the second CH3 domain.

[0423] In another embodiment, S364L, T366V, L368Q, D399K, F405 One or more modifications selected from the group consisting of S, K409F, and T411K are in one CH3 While being introduced within the domain, one selected from Y407F, K409Q, and T411D One or more modifications are introduced into the second CH3 domain.

[0424] In an exemplary embodiment, threonine at position 366 of one CH3 domain The original residue is replaced with valine, while the other CH3 domain has a tyrosine residue at position 407. The original residue is replaced with phenylalanine.

[0425] In another exemplary embodiment, the serine element at position 364 of one CH3 domain While the residue is replaced with leucine, the leucine at position 368 of the same CH3 domain The original residue is replaced with glutamine.

[0426] In yet another exemplary embodiment, at position 405 of one CH3 domain, The original residue of nyalanine is replaced with serine, and the 409th position of this CH3 domain The original residue of din is substituted with phenylalanine, while the 409th position of the other CH3 domain is replaced. The original lysine residue in this compound is replaced with glutamine.

[0427] In yet another exemplary embodiment, at position 399 of one CH3 domain, The original residue of paragate is replaced with lysine, and the same CH3 domain at position 411 is tr The original residue of onin is replaced with lysine, while the other CH3 domain at position 411 The original threonine residue is replaced with aspartic acid.

[0428] The amino acid substitutions described herein are performed using techniques well known in the art, such as CH3 It can be introduced within a domain (e.g., McPherson, ed., 1991, Dire cted Mutagenesis:a Practical Approach;Ad See Elman et al., 1983, DNA, 2:183. Polar crosslinking The methodology is, for example, International Publication No. 2006 / 106905, International Publication No. 200 Pamphlet No. 9 / 089004 and K. Gunasekaran, et al. (20 10)The Journal of Biological Chemistry,2 It is described in 85:19637-19646.

[0429] Further polar crosslinking modifications are, for example, PCT Publication Number International Publication No. 2014 / 145806 (For example, Figure 6 in International Publication No. 2014 / 145806), PCT Publication Numbers International Publication No. 2014 / 110601 Brochure and PCT Publication Numbers by Country International Publication No. 2016 / 086186, International Publication No. 2016 / 086189 Brochure, International Publication No. 2016 / 086196 brochure and International Publication No. 201 This information is included in the 6 / 182751 pamphlet, and the entire contents are included in this statement. This is cited in the book. Examples of polar crosslinking mutants are N208D, Q295E, N384D, and Q41. Includes a constant chain containing 8E and N421D modifications.

[0430] In any of the embodiments described herein, the CH3 domain is defined in Section 7.4.1 As described in Section 5.1, it may be further modified to introduce pairs of cysteine ​​residues. ru.

[0431] Further methods for promoting heterodimerization are, for example, International Publication No. 2016 / 10 Brochure No. 5450, International Publication No. 2016 / 086186, International Publication Pamphlet No. 2016 / 086189, International Publication Pamphlet No. 2016 / 086196 Lett, International Publication No. 2016 / 141378 pamphlet, and International Publication No. 2014 / In brochure No. 145806 and international publication brochure No. 2014 / 110601 It is described. Any of the above methods is a CD3-binding molecule described herein (e.g., It can be used in MBM.

[0432] 7.4.1.5.8. Combinations of heterodimerized mutants and other Fc mutants As will be understood by those skilled in the art, the described heterodimerized mutant (distortion and / or All of them (including pI variants) retain the ability of the Fc domain's Fc region to dimerize. Furthermore, these can be combined arbitrarily, selectively, and independently in any way. All of their mutants can be combined into any heterodimerization configuration.

[0433] In the case of pI mutants, embodiments for finding specific uses are shown in Table 6, but to facilitate purification... In order to do this, the fundamental principle is to change the pI difference between the two Fc regions of the Fc heterodimer. Therefore, other combinations can be generated.

[0434] Furthermore, heterodimerized mutants, distortions, and pIs are all generally used in this specification. As outlined, independently and selectively, Fc ablation variants and Fc mutations are produced. The body is combined with the FcRn mutant.

[0435] In one embodiment, the specific distortion and pI variants for which applications are found in this disclosure The combination is T366S / L368A / Y407V:T366W (optionally selected cross-linked disulfide) Includes Fido, T366S / L368A / Y407V / Y349C:T366W / S35 4C) and one Fc region includes Q295E / N384D / Q418E / N481D Furthermore, other Fc regions are positively charged scFv linkers (if the format includes scFv domains). ) includes. As will be understood by those skilled in the art, the "knob-in-hole" variant has pI To avoid alteration, it can be used in any one of the Fc regions of the Fc heterodimer. .

[0436] In one embodiment, the first and second Fc regions in which the applications of the present disclosure are found are amino Acid substitution includes S364K / E357Q:L368D / K370S, where the first and / or The second Fc region is ablation mutant substitution 233P / L234V / L235A / G It contains 236del / S267K, and the first and / or second Fc region is pI mutant substitution N 208D / Q295E / N384D / Q418E / N421D(pl_(-)_isos Includes teric_A).

[0437] 7.4.2. Hinge Area CD3-binding molecules (e.g., MBMs) can, for example, link an antigen-binding module to the Fc region. This may also include hinge regions. These hinge regions may be natural or modified hinge regions. The region is typically found at the N-terminus of the Fc region.

[0438] The natural hinge region is a hinge region that can normally be found between the Fab and Fc domains in natural antibodies. This is the modified hinge region. The modified hinge region is any hinge that differs in length and / or composition from the natural hinge region. This is the case. Such hinges are found in humans, mice, rats, rabbits, sharks, pigs, and hamsters. - May include hinge regions originating from other species, such as camel, llama, or goat hinge regions. The modified hinge region originates from an antibody of a different class or subclass than that of the heavy chain Fc region. It may include the complete hinge area. Alternatively, the modified hinge area may be a portion or modification of the natural hinge. Each unit in the return may contain repeating units derived from the natural hinge region. Further alternative examples are available. In the natural hinge region, one or more cysteine ​​or other residues such as serine or alanine are present. By converting to a neutral residue, or by converting suitably arranged residues to cysteine ​​residues. It can be modified by such means. The number of cysteine ​​residues in the hinge region This may increase or decrease. This method is based on Bodmer et al.'s U.S. Patent No. 5,677,4 Further details are provided in Specification No. 25. By changing the number of cysteine ​​residues in the hinge region For example, to facilitate the assembly of light and heavy chains, or to increase the stability of CD3-binding molecules. Alternatively, it can be reduced. Other modification hinge regions may be composite overall, and length It can be designed to have desired properties, such as cysteine ​​composition and flexibility.

[0439] Some modified hinge areas are described, for example, in U.S. Patent No. 5,677,425, International Patent No. Pamphlet No. 9915549, International Pamphlet No. 2005003170 International Publication No. 2005003169, International Publication No. 2005003170 Brochure, International Publication No. 9825971, and International Publication No. 2005003 It is already mentioned in pamphlet No. 171.

[0440] Examples of suitable hinge arrangements are shown in Table 7.

[0441] [Table 99]

[0442] [Table 100]

[0443] In one embodiment, the heavy chain Fc region has an intact hinge region at its N-terminus. .

[0444] In one embodiment, the heavy chain Fc region and the hinge region are derived from IgG4, and the hinge region It contains the modified sequence CPPC (sequence number 9). The core hinge region of human IgG4 is sequence C Compared to IgG1 containing PPC (SEQ ID NO: 9), IgG1 containing sequence CPSC (SEQ ID NO: 10) It possesses. The serine residues present in the IgG4 sequence provide increased flexibility in this region. Therefore, a portion of the molecule is attached to the IgG molecule to form interchain disulfides. Rather than cross-linking to other heavy chains, disulfide bonds (intrachain disulfide) are formed within the same protein chain. It forms a (Id). (Angel et al., 1993, Mol lmmunol 30(1):105-108). By changing the serine residue to proline, the same compound as IgG1 is used. Obtaining the sequence enables the complete formation of interchain disulfides in the IgG4 hinge region. This reduces heterogeneity in the purified product. This modified isotype is It is called IgG4P.

[0445] 7.3.3. ABM Linker In certain embodiments, this disclosure relates to a CD3-binding molecule containing at least three ABMs (e.g., For example, provide an MBM, where two or more components of the ABM (e.g., VH of scFv) (and VL), two or more ABM or ABM and non-ABM domains (e.g., Fc region, etc.) The dimerization domains are linked together by peptide linkers. - For example, as described in Section 7.13.2, the drug is linked to a CD3-binding molecule (for example, In contrast to the ADC linker used to bond to MBM, the term "AB" is used herein. It is called "M-Linker".

[0446] Peptide linkers can range from 2 amino acids to 60 or more amino acids, and specific In this embodiment, the peptide linker consists of 3 to 50 amino acids and 4 to 30 amino acids. amino acids, 5-25 amino acids, 10-25 amino acids, or 12-20 amino acids This is within the range. In a particular embodiment, the peptide linker consists of 2 amino acids and 3 amino acids. amino acids, 4 amino acids, 5 amino acids, 6 amino acids, 7 amino acids, 8 amino acids Amino acids, 9 amino acids, 10 amino acids, 11 amino acids, 12 amino acids, 13 amino acids, 14 amino acids, 15 amino acids, 16 amino acids, 17 amino acids amino acids, 18 amino acids, 19 amino acids, 20 amino acids, 21 amino acids , 22 amino acids, 23 amino acids, 24 amino acids, 25 amino acids, 26 amino acids 1 amino acid, 27 amino acids, 28 amino acids, 29 amino acids, 30 amino acids Acid, 31 amino acids, 32 amino acids, 33 amino acids, 34 amino acids, 35 1 amino acid, 36 amino acids, 37 amino acids, 38 amino acids, 39 amino acids 40 amino acids, 41 amino acids, 42 amino acids, 43 amino acids, 4 4 amino acids, 45 amino acids, 46 amino acids, 47 amino acids, 48 ​​amino acids A amino acid is a length of 49 or 50 amino acids.

[0447] A charged and / or flexible linker is particularly preferred.

[0448] Examples of flexible ABM linkers that can be used for CD3-binding molecules (e.g., MBMs) include: , Chen et al.,2013,Adv Drug Deliv Rev.65( 10):1357-1369 and Klein et al., 2014, Protein Engineering,Design&Selection 27(10):325 Examples include those disclosed by -330. Particularly useful flexible linkers are (GGG It is GS)n (sequence number 24) (also called (G4S)n (sequence number 24)). In this embodiment, n is any number from 1 to 10, i.e., 1, 2, 3, 4, 5, 6, 7, Any range bounded by 8, 9, and 10 or any two of the above numbers, for example These include 1-5, 2-5, 3-6, 2-4, 1-4, and others.

[0449] In addition to the ABM linker suitable for use with the CD3-binding molecule (e.g., MBM) of this disclosure, Examples are shown in Table 8 below.

[0450] [Table 101]

[0451] [Table 102]

[0452] [Table 103]

[0453] [Table 104]

[0454] [Table 105]

[0455] In various embodiments, this disclosure relates to CD3-binding molecules containing one or more ABM linkers (e.g.) For example, it provides MBM. Each ABM linker can optionally select from Table 8 above. Two selected amino acids, with a total length of 60 amino acids, preferably 4 to 30 amino acids, 5 to The amino acid length ranges from 25 amino acids, 10 to 25 amino acids, or 12 to 20 amino acids. Obtain. In certain embodiments, the CD3-binding molecule (e.g., MBM) may be 2, 3, or 4. It contains 5 or 6 ABM linkers. The ABM linker is a CD3-binding molecule (for example) It may be on one, two, three, four or more polypeptide chains of MBM. ru.

[0456] 7.5. Bispecific binding molecular morphology An exemplary BBM configuration is shown in Figure 1. Figure 1A shows the BBM configurations shown in Figures 1B to 1AH. The constituent elements of the state are shown: scFv, Fab, scFab, and non-immunoglobulin-based ABM. The Fc domain and its constituent elements are discussed in sections 7.3 and 7.4, respectively. It may have the characteristics described. The components of the BBM configuration shown in Figure 1 are as described in Sections 7.3 and 7. By any of the means described in Section 4 (e.g., direct linking, ABM linker, disulfide), (By filto bonds, knob-in-hole interactions, etc., modified Fc domains) They can be associated with each other. The orientations and associations of the various components shown in Figure 1 are exemplary. It is merely; as will be understood by those skilled in the art, other orientations and associations may be preferable (for example) (For example, as described in sections 7.3 and 7.4).

[0457] The BBM is not limited to the form shown in Figure 1. Other forms that may be used are known to those skilled in the art. For example, see International Publication No. 2014 / 145806 pamphlet; International Publication No. 201 Pamphlet No. 7 / 124002; Liu et al., 2017, Front Im munol.8:38;Brinkmann&Kontermann,2017,mAb s 9:2,182-212; U.S. Patent Application Publication No. 2016 / 0355600; Klein et al., 2016, MAbs 8(6):1010-20; and the United States Please refer to the specification in Patent Application Publication No. 2017 / 0145116.

[0458] 7.5.1. Exemplary divalent BBM BBMs can be bivalent, meaning they have two antigen-binding domains, and One or two of these bind to CD3(ABM1), and one of them becomes the second target antigen (A BM2), for example, binds to CD2 or TAA.

[0459] Exemplary divalent BBM morphologies are shown in Figures 1B-1F.

[0460] As shown in Figures 1B-1D, BBM may contain two half-antibodies, one of which is an AB antibody. The two half-antibodies, one containing M and the other containing one ABM, are paired via the Fc domain. It is.

[0461] In the embodiment shown in Figure 1B, the first (or left-side) half-antibody includes the Fab and Fc regions. The second (or right-side) half-antibody contains the Fab and Fc regions. The first and second half-antibodies are They associate via the Fc region to form an Fc domain.

[0462] In the embodiment shown in Figure 1C, the first (or left-hand) half-antibody comprises the Fab and Fc regions. The second (or right-side) half-antibody contains the scFv and Fc regions. The first and second half-antibodies are They are associated via the Fc region to form an Fc domain.

[0463] In the embodiment shown in Figure 1D, the first (or left-hand) half-antibody includes the scFv and Fc regions. The second (or right-side) half-antibody contains the scFv and Fc regions. (First and second half-antibodies) These are associated via the Fc region to form an Fc domain.

[0464] As shown in Figures 1E-1F, the divalent BBM is linked to one Fc region of the Fc domain. It may include two combined ABMs.

[0465] In the embodiment shown in Figure 1E, the BBM comprises Fab, scFv, and Fc domains. Therefore, scFv is located between the Fab and Fc domains.

[0466] In the embodiment shown in Figure 1F, the BBM (in the "one-arm scFv-mAb" configuration) is Fa b, includes scFv and Fc domains, where Fab is between scFv and Fc domains. It is located in [location].

[0467] In the configurations shown in Figures 1B to 1F, X and Y are either ABM1 or ABM2. It represents one of the following, however BBM includes one ABM1 and one ABM2. Therefore, this disclosure provides a divalent BBM as shown in any one of Figures 1B to 1F, Here, X is ABM1 and Y is ABM2 (for convenience, this form of ABM is called "B (referred to as "1"). Therefore, this disclosure is shown in any one of Figures 1B to 1F. We also provide a divalent BBM such as ABM2 and Y is ABM1 (ABM This form is shown as "B2" for convenience.

[0468] 7.5.2. Exemplary trivalent BBM BBMs can be trivalent, meaning they have three antigen-binding domains, and One or two of these bind to CD3(ABM1), and one or two of these bind to a second target. It binds to antigens (ABM2), such as CD2 or TAA.

[0469] Exemplary trivalent BBM morphologies are shown in Figures 1G-1Z.

[0470] As shown in Figures 1G-1N, 1Q-1W, and 1Y-1Z, BBM uses two half-antibodies. It may contain, one containing two ABMs and the other containing one ABM, and the two half-antibodies, Fc They are being matched via domains.

[0471] In the embodiment shown in Figure 1G, the first (or left-side) half-antibody comprises the Fab and Fc regions. The second (or right-side) half-antibody includes the scFv, Fab, and Fc regions. The semi-antibodies associate via the Fc region to form an Fc domain.

[0472] In the embodiment shown in Figure 1H, the first (or left-side) half-antibody comprises the Fab and Fc regions. The second (or right-side) half-antibody includes the Fab, scFv, and Fc regions. The semi-antibodies associate via the Fc region to form an Fc domain.

[0473] In the embodiment shown in Figure 1I, the first (or left-side) half-antibody includes the scFv and Fc regions. Furthermore, the second (or right-side) half-antibody contains two Fab and Fc regions. Antibodies associate via the Fc region to form an Fc domain.

[0474] In the embodiment shown in Figure 1J, the first (or left-side) half-antibody has two Fav and Fc regions. The first and second semi-antibodies include the Fab and Fc regions. The bodies associate via the Fc region to form an Fc domain.

[0475] In the embodiment shown in Figure 1K, the first (or left-side) half-antibody includes the scFv and Fc regions. The second (or right-side) half-antibody contains two scFv and Fc regions. The semi-antibodies associate via the Fc region to form an Fc domain.

[0476] In the embodiment shown in Figure 1L, the first (or left-side) half-antibody includes the scFv and Fc regions. The second (or right-side) half-antibody contains the scFv, Fab, and Fc regions. The half-antibodies associate via the Fc region to form an Fc domain.

[0477] In the embodiment shown in Figure 1M, the first (or left-hand) half-antibody includes the scFv and Fc regions. The second (or right-side) half-antibody includes the Fab, scFv, and Fc regions. The half-antibodies associate via the Fc region to form an Fc domain.

[0478] In the embodiment shown in Figure 1N, the first (or left-hand) half-antibody is a bispecific antibody-type conjugated dominant. The second (or right-side) half-antibody includes the Fab and Fc regions. The first and second half-antibodies associate via the Fc region to form an Fc domain.

[0479] In the embodiment shown in Figure 1Q, the first (or left-side) half-antibody comprises the Fab and Fc regions. The second (or right-side) half-antibody includes the Fab, Fc region and scFv. The semi-antibodies associate via the Fc region to form an Fc domain.

[0480] In the embodiment shown in Figure 1R, the first (or left-side) half-antibody includes the scFv and Fc regions. The second (or right-side) half-antibody contains the Fab, Fc region and scFv. The half-antibodies associate via the Fc region to form an Fc domain.

[0481] In the embodiment shown in Figure 1S, the first (or left-side) half-antibody includes the scFv and Fc regions. The second (or right-side) half-antibody includes the scFv, Fc region and the second scFv. The first and second half-antibodies associate via the Fc region to form an Fc domain.

[0482] In the embodiment shown in Figure 1T, the first (or left-side) half-antibody comprises the scFv, Fc region and F The first and second half-antibodies include ab, and the second (or right-side) half-antibodies include Fab and Fc regions. The semi-antibodies associate via the Fc region to form an Fc domain.

[0483] In the embodiment shown in Figure 1U, the first (or left-hand) half-antibody has two Fab and Fc regions. The second (or right-side) half-antibody includes non-immunoglobulin-based ABM and Fc regions. The first and second half-antibodies are associated via the Fc region to form an Fc domain.

[0484] In the embodiment shown in Figure 1V, the first (or left-hand) half-antibody consists of Fab, scFv, and Fc The region includes the second (or right-side) hemiantibody, and the non-immunoglobulin-based ABM and Fc region. The region is included. The first and second half-antibodies associate via the Fc region to form an Fc domain. ru.

[0485] In the embodiment shown in Figure 1W, the first (or left-hand) half-antibody comprises the Fab and Fc regions. The second (or right-sided) hemiantibody is scFv, non-immunoglobulin-based ABM, and Fc region. The region is included. The first and second half-antibodies associate via the Fc region to form an Fc domain. ru.

[0486] In the embodiment shown in Figure 1Y, the first (or left-side) half-antibody includes the scFv and Fc regions. The second (or right-side) half-antibody includes the Fab, scFv, and Fc regions. The half-antibodies associate via the Fc region to form an Fc domain.

[0487] In the embodiment shown in Figure 1Z, the first (or left-side) half-antibody comprises the Fab, Fc region and sc The first and second half-antibodies include Fab and the second (or right-side) half-antibodies include Fab and Fc regions. The half-antibodies associate via the Fc region to form an Fc domain.

[0488] Alternatively, as shown in Figures 10 and 1P, the trivalent BBM may contain two half-antibodies. Includes one complete ABM (Fab in Figures 10 and 1P) and a portion of another ABM (one of which is V (H is one half-antibody, the other is VL). When the two half-antibodies pair via the Fc domain, VH and VL then associate to form a complete antigen-binding Fv domain.

[0489] As shown in Figure 1X, the BBM can be single-stranded. The BBM in Figure 1X is a linker. It includes three scFv domains linked via [a specific method / platform].

[0490] In the configurations shown in Figures 1G to 1Z, X, Y, and A are each ABM1 or AB Represents any of M2, where BBM is at least ABM1 and at least one A Contains BM2. Therefore, trivalent MBM contains one or two ABM1 and one or two Contains ABM2. In one embodiment, trivalent BBM contains two ABM1 and one AB It contains M2. In other embodiments, the trivalent BBM consists of one ABM1 and two ABM2. Includes.

[0491] Therefore, in this disclosure, trivalent BBM as shown in any one of Figures 1G to 1Z Provided, where X is ABM1, Y is ABM1, and A is ABM2 (AB This form of M is shown as "T1" for convenience.

[0492] This disclosure further provides trivalent BBMs as shown in any one of Figures 1G to 1Z, Here, X is ABM1, Y is ABM2, and A is ABM1 (this ABM The form is shown as "T2" for convenience.

[0493] This disclosure further provides trivalent BBMs as shown in any one of Figures 1G to 1Z, Here, X is ABM2, Y is ABM1, and A is ABM1 (this ABM The form is shown as "T3" for convenience.

[0494] This disclosure further provides trivalent BBMs as shown in any one of Figures 1G to 1Z, Here, X is ABM1, Y is ABM2, and A is ABM2 (this ABM The form is shown as "T4" for convenience.

[0495] This disclosure further provides trivalent BBMs as shown in any one of Figures 1G to 1Z, Here, X is ABM2, Y is ABM1, and A is ABM2 (this ABM The form is shown as "T5" for convenience.

[0496] This disclosure further provides trivalent BBMs as shown in any one of Figures 1G to 1Z, Here, X is ABM2, Y is ABM2, and A is ABM1 (this ABM The form is shown as "T6" for convenience.

[0497] 7.5.3. Exemplary tetravalent BBM BBMs can be quadrivalent, meaning they have four antigen-binding domains, and One, two, or three of these bind to CD2(ABM1), and one, two, or three of them It binds to a second target antigen (ABM2), such as CD2 or TAA.

[0498] Exemplary tetravalent BBM morphologies are shown in Figures 1AA-1AH.

[0499] As shown in Figures 1AA-1AH, the quadrivalent BBM may contain two half-antibodies, each It contains two complete ABMs, and the two half-antibodies are paired via the Fc domain.

[0500] In the embodiment shown in Figure 1AA, the first (or left-side) half-antibody has Fab, Fc region and s The second (or right-side) half-antibody contains cFv, and the second half-antibody contains Fab, Fc region and scFv. The first and second half-antibodies associate via the Fc region to form an Fc domain.

[0501] In the embodiment shown in Figure 1AB, the first (or left-side) half-antibody consists of Fab, scFv, and F The second (or right-side) half-antibody includes the c region, and the second half-antibody includes the Fab, scFv, and Fc regions. The first and second half-antibodies associate via the Fc region to form an Fc domain.

[0502] In the embodiment shown in Figure 1AC, the first (or left-hand) half-antibody consists of scFv, Fab, and F The second (or right-side) half-antibody includes the c region, and the scFv, Fab, and Fc regions. The first and second half-antibodies associate via the Fc region to form an Fc domain.

[0503] In the embodiment shown in Figure 1AD, the first (or left-side) half-antibody has Fab, Fc region and The Fab contains two Fabs, and the second (or right-side) half-antibody contains the Fab, Fc region and the second Fab. The first and second half-antibodies are associated via the Fc region to form an Fc domain.

[0504] In the embodiment shown in Figure 1AE, the first (or left) half-antibody is scFv, and the second scF The second (or right-side) half-antibody includes the v and Fc regions, and the second scFv, second scFv, and F It includes the c region. The first and second half-antibodies associate via the Fc region to form the Fc domain. To accomplish.

[0505] In the embodiment shown in Figure 1AF, the first (or left-side) half-antibody consists of Fab, scFv, and F The second (or right-side) half-antibody includes the c region, and the second half-antibody includes the Fab, scFv, and Fc regions. The first and second half-antibodies associate via the Fc region to form an Fc domain.

[0506] In the embodiment shown in Figure 1AG, the first (or left-side) half-antibody has Fab, Fc region and s The second (or right-side) half-antibody contains cFv, and includes scFv, Fc region and Fab. The first and second half-antibodies associate via the Fc region to form an Fc domain.

[0507] In the embodiment of Figure 1AH, the first (or left-side) half-antibody has scFv, Fc region and The second (or right-side) half-antibody contains Fab, and includes scFv, Fc region and Fab. The first and second half-antibodies associate via the Fc region to form an Fc domain.

[0508] In the configurations shown in Figures 1AA to 1AH, X, Y, A, and B are not necessarily The order is not necessarily the same, but it represents either ABM1 or ABM2, however BBM is less common. It contains at least one ABM1 and at least one ABM2. Therefore, tetravalent ABM is , including one, two or three ABM1 and one, two or two ABM2. In one embodiment In other embodiments, the tetravalent BBM contains three ABM1 and one ABM2. The tetravalent BBM comprises two ABM1 and two ABM2. In other embodiments, Therefore, tetravalent BBM contains one ABM1 and three ABM2 molecules.

[0509] Therefore, in this disclosure, tetravalent BB is used as shown in any one of Figures 1AA to 1AH. Let M be provided, where X is ABM1 and Y, A and B are each ABM2 ( This form of ABM is, for convenience, referred to as "Tv 1".

[0510] This disclosure further provides tetravalent BBMs as shown in any one of Figures 1AA to 1AH. And here, Y is ABM1, and X, A, and B are each ABM2 (ABM This configuration is referred to as "Tv 2" for convenience.

[0511] This disclosure further provides tetravalent BBMs as shown in any one of Figures 1AA to 1AH. And here, A is ABM1, and X, Y, and B are each ABM2 (ABM This configuration is referred to as "Tv 3" for convenience.

[0512] This disclosure further provides tetravalent BBMs as shown in any one of Figures 1AA to 1AH. And here, B is ABM1, and X, Y, and A are each ABM2 (ABM This configuration is referred to as "Tv 4" for convenience.

[0513] This disclosure further provides tetravalent BBMs as shown in any one of Figures 1AA to 1AH. And here, both X and Y are ABM1, and both A and B are ABM2 ( This form of ABM is, for convenience, referred to as "Tv 5".

[0514] This disclosure further provides tetravalent BBMs as shown in any one of Figures 1AA to 1AH. And here, both X and A are ABM1, and both Y and B are ABM2 ( This form of ABM is, for convenience, referred to as "Tv 6".

[0515] This disclosure further provides tetravalent BBMs as shown in any one of Figures 1AA to 1AH. And here, both X and B are ABM1, and both Y and A are ABM2 ( This form of ABM is, for convenience, referred to as "Tv 7".

[0516] This disclosure further provides tetravalent BBMs as shown in any one of Figures 1AA to 1AH. And here, Y and A are both ABM1, and X and B are both ABM2 ( This form of ABM is, for convenience, referred to as "Tv 8".

[0517] This disclosure further provides tetravalent BBMs as shown in any one of Figures 1AA to 1AH. And here, Y and B are both ABM1, and X and A are both ABM2 ( This form of ABM is, for convenience, referred to as "Tv 9".

[0518] This disclosure further provides tetravalent BBMs as shown in any one of Figures 1AA to 1AH. And here, both A and B are ABM1, and both X and Y are ABM2 ( This form of ABM is, for convenience, referred to as "Tv 10".

[0519] This disclosure further provides tetravalent BBMs as shown in any one of Figures 1AA to 1AH. Here, X, Y, and A are ABM1, and B is ABM2 (ABM This configuration is referred to as "Tv 11" for convenience.

[0520] This disclosure further provides tetravalent BBMs as shown in any one of Figures 1AA to 1AH. Here, X, Y, and B are ABM1, and A is ABM2 (ABM This configuration is referred to as "Tv 12" for convenience.

[0521] This disclosure further provides tetravalent BBMs as shown in any one of Figures 1AA to 1AH. And here, X, A, and B are ABM1 and Y is ABM2 (ABM This configuration is referred to as "Tv 13" for convenience.

[0522] This disclosure further provides tetravalent BBMs as shown in any one of Figures 1AA to 1AH. Here, Y, A, and B are ABM1, and X is ABM2 (ABM This configuration is shown as "Tv 14" for convenience.

[0523] 7.6. Trispecific binding molecular morphology An exemplary TBM configuration is shown in Figure 2. Figure 2A shows the TBM configurations shown in Figures 2B-2V. The components are shown. scFv, Fab, non-immunoglobulin-based ABM and Fc are Each of these components may possess the characteristics described in Sections 7.3 and 7.4. The components of the TBM configuration shown in Figure 2 are the means described in Sections 7.3 and 7.4. By any of the following (e.g., direct bond, ABM linker, disulfide bond, knob-in) They can associate with each other (e.g., via Fc domains modified by hole interactions). See Figure 2. The orientations and associations of the various components shown are illustrative only; to those skilled in the art... As can be understood, other orientations and associations may be preferable (for example, sections 7.3 and 7.4). (As described in the section.)

[0524] TBM is not limited to the form shown in Figure 2. Other forms that can be used are also available. It is publicly known to the public. For example, International Publication No. 2014 / 145806 pamphlet; International Publication Pamphlet No. 2017 / 124002; Liu et al., 2017, Fron t Immunol.8:38;Brinkmann&Kontermann,2017 mAbs 9:2,182-212; U.S. Patent Application Publication No. 2016 / 0355600 Specification; Klein et al., 2016, MAbs 8(6):1010-20; See also U.S. Patent Application Publication No. 2017 / 0145116.

[0525] 7.6.1. Exemplary trivalent TBM TBMs can be trivalent, meaning they have three antigen-binding domains, and One of them binds to CD3, one of those binds to TAA, and one of those binds to CD2. Alternatively, it can be combined with a second TAA.

[0526] Exemplary trivalent TBM morphologies are shown in Figures 2B-2P.

[0527] As shown in Figures 2B-2K and 2N-2P, TBM may contain two half-antibodies. One half-antibody contains two ABMs, and the other contains one ABM, and the two half-antibodies have Fc domains They are facing each other through an intermediary.

[0528] In the embodiment shown in Figure 2B, the first (or left-side) half-antibody includes the scFv and Fc regions. The second (or right-side) half-antibody includes the Fab, scFv, and Fc regions. The half-antibodies associate via the Fc region to form an Fc domain.

[0529] In the embodiment shown in Figure 2C, the first (or left-hand) half-antibody has two Fab and Fc regions. The first and second semi-antibodies include the Fab and Fc regions. The bodies associate via the Fc region to form an Fc domain.

[0530] In the embodiment shown in Figure 2D, the first (or left-hand) half-antibody consists of Fab, scFv, and Fc The first and second half-antibodies include the Fab and Fc regions. The semi-antibodies associate via the Fc region to form an Fc domain.

[0531] In the embodiment shown in Figure 2E, the first (or left-side) half-antibody includes the scFv and Fc regions. Furthermore, the second (or right-side) half-antibody contains two Fab and Fc regions. Antibodies associate via the Fc region to form an Fc domain.

[0532] In the embodiment shown in Figure 2F, the first (or left-side) half-antibody comprises the scFv, Fc region and F The first and second half-antibodies include ab, and the second (or right-side) half-antibodies include Fab and Fc regions. The semi-antibodies associate via the Fc region to form an Fc domain.

[0533] In the embodiment shown in Figure 2G, the first (or left-hand) half-antibody includes the scFv and Fc regions. The second (or right-side) half-antibody includes the Fab, Fc region and scFV. The half-antibodies associate via the Fc region to form an Fc domain.

[0534] In the embodiment shown in Figure 2H, the first (or left-hand) half-antibody has two Fab and Fc regions. The second (or right-side) half-antibody includes non-immunoglobulin-based ABM and Fc regions. The first and second half-antibodies are associated via the Fc region to form an Fc domain.

[0535] In the embodiment shown in Figure 2I, the first (or left-side) half-antibody consists of Fab, scFv, and Fc The region includes the second (or right-side) hemiantibody, and the non-immunoglobulin-based ABM and Fc region. The region is included. The first and second half-antibodies associate via the Fc region to form an Fc domain. ru.

[0536] In the embodiment shown in Figure 2J, the first (or left-side) half-antibody includes the Fab and Fc regions. The second (or right-sided) hemiantibody is scFv, non-immunoglobulin-based ABM, and Fc region. The region is included. The first and second half-antibodies associate via the Fc region to form an Fc domain. ru.

[0537] In the embodiment shown in Figure 2K, the first (or left-hand) half-antibody includes the scFv and Fc regions. The second (or right-side) half-antibody includes the scFv, Fc region and the second scFv. The first and second half-antibodies associate via the Fc region to form an Fc domain.

[0538] In the embodiment shown in Figure 2N, the first (or left-side) half-antibody comprises the Fab, Fc region and sc The first and second half-antibodies include Fv, and the second (or right-side) half-antibodies include Fab and Fc regions. The semi-antibodies associate via the Fc region to form an Fc domain.

[0539] In the embodiment shown in Figure 2O, the first (or left-side) half-antibody comprises the Fab, Fc region and sc The first and second half-antibodies include Fab and the second (or right-side) half-antibodies include Fab and Fc regions. The half-antibodies associate via the Fc region to form an Fc domain.

[0540] In the embodiment shown in Figure 2P, the first (or left-hand) half-antibody is Fab, non-immunoglobulin. The base ABM and Fc region is included, and the second (or right-side) half-antibody contains the scFv and Fc regions. The region is included. The first and second half-antibodies associate via the Fc region to form an Fc domain. ru.

[0541] Alternatively, as shown in Figure 2L, a trivalent TBM may contain two half-antibodies and one complete antibody. Includes all ABM and parts of other ABMs (one being VH, the other VL). As soon as the half-antibody is paired via the Fc domain, VH and VL associate to form a complete antibody. It forms an antigen-binding Fv domain.

[0542] As shown in Figure 2M, the TBM can be single-stranded. The TBM in Figure 1M is a linker. It includes three scFv domains linked via [a specific method / platform].

[0543] In each of the configurations shown in Figures 2B to 2P, the domains indicated as X, Y, and Z Each of the elements is not necessarily in that order, but ABM1, ABM2, or AB This represents M3. In other words, X can be ABM1, ABM2, or ABM3, and Y is ABM1. , can be ABM2 or ABM3, and Z can be ABM1, ABM2 or ABM3, However, TBM includes one ABM1, one ABM2, and one ABM3.

[0544] Therefore, in this disclosure, trivalent TB as shown in any one of Figures 2B to 2P. Let M be given, where X is ABM1, Y is ABM3, and Z is ABM2 ( This form of ABM is shown as "T1" for convenience.

[0545] This disclosure also provides trivalent TBMs as shown in any one of Figures 2B-2P, where X is ABM1, Y is ABM2, and Z is ABM3 (this form of ABM is (For convenience, this will be shown as "T2").

[0546] This disclosure further provides trivalent TBMs as shown in any one of Figures 2B to 2P, Here, X is ABM3, Y is ABM1, and Z is ABM2 (this ABM The form is shown as "T3" for convenience.

[0547] This disclosure further provides trivalent TBMs as shown in any one of Figures 2B-2P And here, X is ABM3, Y is ABM2, and Z is ABM1 (ABM This configuration is shown as "T4" for convenience.

[0548] This disclosure further provides trivalent TBMs as shown in any one of Figures 2B-2P And here, X is ABM2, Y is ABM1, and Z is ABM3 (ABM This configuration is shown as "T5" for convenience.

[0549] This disclosure further provides trivalent TBMs as shown in any one of Figures 2B-2P And here, X is ABM2, Y is ABM3, and Z is ABM1 (ABM This configuration is shown as "T6" for convenience.

[0550] 7.6.2. Exemplary tetravalent TBM TBMs can be quadrivalent, meaning they have four antigen-binding domains. One or two of them bind to CD3, and one or two of those bind to TAA, and One or two of them are combined with CD2 or a second TAA.

[0551] Exemplary tetravalent TBM morphologies are shown in Figures 2Q-2S.

[0552] As shown in Figures 2Q-2S, a tetravalent TBM may contain two half-antibodies, each containing 2 It contains two complete ABMs, and the two half-antibodies are paired via the Fc domain.

[0553] In the embodiment shown in Figure 2Q, the first (or left) half-antibody has the Fab, Fc region and the second The first half-antibody contains the Fab, and the second (or right-side) half-antibody contains the Fab, Fc region and the second Fab. The first and second half-antibodies associate via the Fc region to form an Fc domain.

[0554] In the embodiment shown in Figure 2Q, the first (or left-side) half-antibody comprises the Fab, Fc region and sc The first half-antibody contains Fv, and the second (or right-side) half-antibody contains Fab, Fc region and scFv. The first and second half-antibodies associate via the Fc region to form an Fc domain.

[0555] In the embodiment shown in Figure 2S, the first (or left-side) half-antibody comprises the Fab, Fc region and sc The first half-antibody contains Fv, and the second (or right-side) half-antibody contains scFv, Fc region and Fab. The first and second half-antibodies associate via the Fc region to form an Fc domain.

[0556] In the configurations shown in Figures 2Q to 2S, X, Y, Z, and A are not necessarily the same. The order is not necessarily the same, but it represents ABM1, ABM2, or ABM3, where TBM is , at least one ABM1, at least one ABM2 and at least one ABM3 It includes. Therefore, tetravalent ABM contains 2 for one of CD3, CD2, or a second TAA. It contains two ABMs. In some cases, a tetravalent TBM has two CD3 ABMs.

[0557] Therefore, this disclosure provides a tetravalent TBM as shown in any one of Figures 2Q to 2S. Herein, X, Y, Z and A are CD3, CD2 or the second as shown in Table 9. This is ABM in relation to TAA.

[0558] [Table 106]

[0559] 7.6.3. Exemplary pentavalent TBM TBMs can be pentavalent, meaning they have five antigen-binding domains. One, two, or three of them bind to CD3, and one, two, or three of those bind to TAA. They combine, and one, two, or three of them combine with CD2 or a second TAA.

[0560] An exemplary pentavalent TBM morphology is shown in Figure 2T.

[0561] As shown in Figure 2T, a pentavalent TBM may contain two half-antibodies, one of which is a complete antibody. One half-antibodies contain ABM, the other contains one complete ABM, and the two half-antibodies are separated by the Fc domain. They are confronted.

[0562] In the embodiment shown in Figure 2T, the first (or left-side) half-antibody consists of Fab, scFv, and Fc The second (or right-side) half-antibody includes the Fab, Fc region and scFv. The first and second half-antibodies associate via the Fc region to form an Fc domain.

[0563] In the configuration shown in Figure 2T, X, Y, Z, A, and B are not necessarily in that order. It is not necessarily the introduction, but represents ABM1, ABM2, or ABM3, except that TBM is It includes at least one ABM1, one ABM2, and one ABM3. Therefore, 5 The valence TBM is two values ​​for two of the following: CD3, TAA, and CD2 or the second TAA. Three A's for ABM or CD3, TAA and CD2 or one of the second TAA It may contain BM. In some cases, the pentavalent TBM has two or three CD3 ABMs. In one embodiment, the pentavalent TBM consists of three ABM1, one ABM2 and one A It has BM3.

[0564] Therefore, this disclosure provides a pentavalent TBM as shown in Figure 2T, where X, Y Z, A, and B are CD3, TAA, and CD2 or the second TA, as shown in Table 10. This is ABM for A.

[0565] [Table 107]

[0566] [Table 108]

[0567] [Table 109]

[0568] 7.6.4. Exemplary hexavalent TBM TBMs can be hexavalent, meaning they have six antigen-binding domains. One, two, three, or four of them bind to CD3, and one, two, three, or four of them bind to CD3. One of them combines with the TAA, and one, two, three, or four of them connect to CD2 or the second TAA. Combine.

[0569] Exemplary hexavalent TBM morphologies are shown in Figures 2U-2V.

[0570] As shown in Figures 2U-2V, a pentavalent TBM may contain two half-antibodies, one of which is two One half-antibody contains a complete ABM, the other contains one complete ABM, and the two half-antibodies contain Fc domains. They are confronted through this means.

[0571] In the embodiment shown in Figure 2U, the first (or left) half-antibody is Fab, the second Fab is F The c region and scFv are included, and the second (or right-side) half-antibody is Fab, second Fab, Fc The region and scFv are included. The first and second half-antibodies are associated via the Fc region and Fc It forms the main part.

[0572] In the embodiment shown in Figure 2V, the first (or left) half-antibody is the first Fv, the second Fv, The third Fv and Fc region is included, and the second (or right-side) half-antibody is the first Fv, the second Fv , including a third Fv and Fc region. The first and second half-antibodies are associated via the Fc region. It forms an Fc domain.

[0573] In the configurations shown in Figures 2U to 2V, each of X, Y, Z, A, B, and C is always However, the order is not necessarily the same, but it represents ABM1, ABM2, or ABM3, provided that, TBM includes at least one ABM1, one ABM2, and one ABM3. Therefore, hexavalent TBM is (i) CD3, TAA and CD2 or the second TAA (ii) CD3, TAA and CD2 or 1 of the second TAA (iii) CD3, TAA and CD2 or second TA It can contain four ABMs for one A. For example, hexavalent ABM contains three for CD3. ABM of , two ABMs for TAA and CD2 or one AB for the second TAA It may contain M. As another example, hexavalent ABM has three ABMs against CD3 and TAA. This may include one ABM and two ABMs for CD2 or a second TAA. In some embodiments, the hexavalent TBM has two, three, or four CD3 ABMs. In this embodiment, the hexavalent TBM has three CD3 ABMs. In another embodiment, the hexavalent T The BM has four CD3 ABMs.

[0574] Therefore, in this disclosure, hexavalent TB as shown in any one of Figures 2U to 2V Provide M, where X, Y, Z, A, B and C are as shown in Table 11, CD3, This is ABM for TAA and CD2 or the second TAA.

[0575] [Table 110]

[0576] Table 111

[0577] Table 112

[0578] Table 113

[0579] Table 114

[0580] Table 115

[0581] Table 116

[0582] Table 117

[0583] Table 118

[0584] Table 119

[0585] Table 120

[0586] Table 121

[0587] [Table 122]

[0588] 7.7. Exemplary multispecific binding molecules Exemplary MBM morphologies are shown in Figures 3A-3E. scFv, Fab, non-immunoglobulin The base ABM and Fc are described in sections 7.2 to 7.4, respectively, as they relate to these components. It may have the characteristics described below. The components of the MBM form are described in Sections 7.3 and 7.4. By any of the means (e.g., direct bond, ABM linker, disulfide bond) They associate with each other (e.g., by Fc domains modified by knob-in-hole interactions) The orientations and associations of the various components shown in Figures 3A-3E are exemplary. Not excessive; as will be understood by those skilled in the art, other orientations and associations may be preferable (for example) (as described in sections 7.3 and 7.4).

[0589] MBM is not limited to the forms shown in Figures 3A-3E. Other forms may be used. The condition is publicly known to those skilled in the art. For example, International Publication No. 2014 / 145806; International Publication No. 2017 / 124002 pamphlet; Liu et al., 2017, Front Immunol.8:38;Brinkmann&Kontermann, 2017, mAbs 9:2, 182-212; U.S. Patent Application Publication No. 2016 / 0355 Specification No. 600; Klein et al., 2016, MAbs 8(6):1010 See also -20 and U.S. Patent Application Publication No. 2017 / 0145116.

[0590] 7.7.1. Illustrative MBM MBMs can be bispecific; for example, they have two antigen-binding domains, here One antigen-binding domain binds to CD3, and the other antigen-binding domain binds to TAA. do.

[0591] 7.7.2. Example of trivalent MBM MBMs can be trivalent, for example, they have three antigen-binding domains, where 3 At least one of the antigen-binding domains binds to CD3, and 0 of the three antigen-binding domains ~One binds to CD2, and at least one of the three antigen-binding domains binds to TAA. .

[0592] 7.7.3. Essential Tetravalent MBM MBMs can be tetravalent, for example, they have four antigen-binding domains, where 4 At least one of the antigen-binding domains binds to CD3, and 0 of the four antigen-binding domains ~Two of them bind to CD2, and at least one of the four antigen-binding domains binds to TAA. .

[0593] 7.8. TCR ABM MBM contains ABM that specifically binds to components of the TCR complex. TCR is Highly variable alpha (α) is typically expressed as part of a complex with non-mutant CD3 chain molecules. A disulfide-linked heterodimer consisting of a ) and a beta (β) chain, immobilized on a membrane It is an protein. T cells that express this receptor are called α:β (or αβ) T cells. However, a small number of T cells express different receptors and are affected by variable gamma (γ) and delta (δ) chains. They are formed and called γδ T cells.

[0594] In a preferred embodiment, MBM is CD3, for example, C as shown in Table 1 or Table 19. It contains ABM that specifically binds to the D3 antigen-binding domain.

[0595] 7.8.1.CD3 ABM MBM may contain ABM that specifically binds to CD3. The term "CD3" refers to Three differentiated co-receptors of the T cell receptor (or co-receptor complex or polymorphism of the co-receptor complex) This refers to a group of peptide chains. CD3 proteins may also include mutants. CD3 proteins are This may also include fragments. CD3 proteins also include post-translational modifications of the CD3 amino acid sequence. It is possible. Post-translational modifications include, but are not limited to, N-linking and O-linking glycosylation. It is possible.

[0596] In one embodiment, MBM is an anti-CD3 antibody or its antigen-binding domain AB M may be included. Exemplary anti-CD3 VH and VL and scFV may be used in MBM. The sequences are shown in Tables 1 and 19.

[0597] In one embodiment, the CD3 ABM contains the CDR sequence of NOV292. In one embodiment, the CD3 ABM contains the CDR sequence of NOV123. In one embodiment, the CD3 ABM contains the CDR sequence of NOV453. CD3 ABM contains the CDR sequence of NOV229. In one embodiment, CD3 The ABM contains the CDR sequence of NOV110. In one embodiment, the CD3 ABM is , containing the CDR sequence of NOV832. In one embodiment, the CD3 ABM is NOV It contains 589 CDR sequences. In one embodiment, the CD3 ABM is NOV580 Includes a CDR sequence. In one embodiment, the CD3 ABM contains the CDR sequence of NOV567. Includes a column. In one embodiment, the CD3 ABM includes the CDR sequence of NOV221. .

[0598] MBM is a complete heavy-chain and light-chain variable of either the CD3 sequence shown in Table 1 or Table 19. It may include sequences. In one embodiment, MBM includes the VH and VL sequences of NOV292. Includes CD3 ABM. In one embodiment, MBM includes VH of NOV123 and It includes a CD3 ABM containing a VL sequence. In one embodiment, the MBM is NOV453 It includes a CD3 ABM containing VH and VL sequences. In one embodiment, the MBM is N Includes CD3 ABM containing the VH and VL sequences of OV229. In one embodiment, M The BM includes a CD3 ABM containing the VH and VL sequences of NOV110. In one embodiment, In this context, MBM includes CD3 ABM containing the VH and VL sequences of NOV832. In this embodiment, the MBM is a CD3 ABM containing the VH and VL sequences of NOV589. Includes. In one embodiment, MBM includes CD3 containing the VH and VL sequences of NOV580. Includes ABM. In one embodiment, MBM contains the VH and VL sequences of NOV567 Includes CD3 ABM. In one embodiment, MBM is VH of NOV221 and Contains CD3 ABM with VL sequence.

[0599] 7.8.2. TCR-α / β ABM MBM is an A that specifically binds to the TCR-α chain, TCR-β chain, or TCR-αβ dimer. It may contain BM. Exemplary anti-TCR-α / β antibodies are known in the art. For example, U.S. Patent Application Publication No. 2012 / 0034221; Borst et a l.,1990,Hum Immunol.29(3):175-88(Antibody BMA03 See item 1). VH, VL and Kabat of antibody BMA031. The CDR sequences are shown in Table 12.

[0600] [Table 123]

[0601] In one embodiment, the TCR ABM may contain the CDR sequence of the antibody BMA031. In another embodiment, the TCR ABM contains the VH and VL sequences of the antibody BMA031. It is visible.

[0602] 7.8.3. TCR-γ / δ ABM MBM is an A that specifically binds to TCR-γ chains, TCR-δ chains, or TCR-γδ dimers. It may contain BM. Exemplary anti-TCR-γ / δ antibodies are known in the art. For example, U.S. Patent No. 5,980,892 (accession number HB 9578 as ATC) (This document describes δTCS1 produced by hybridomas deposited by C.) I want to be illuminated.

[0603] 7.9.CD2 ABM 7.9.1. Immunoglobulin-based CD2 ABM In one embodiment, MBM is an anti-CD2 antibody or its antigen-binding domain AB M may be included. Exemplary anti-CD2 antibodies are known in the art (e.g., in the United States). Specification No. 6,849,258, Chinese Patent Application Publication No. 102827281A, United States Patent Application Publication No. 2003 / 0139579 Specification A1 and U.S. Patent No. 5,795,5 (See Specification No. 72). Table 13 lists anti-CD2 antibodies or for use in MBM. This provides exemplary CDR, VH, and VL sequences that may be included in the antigen-binding fragment.

[0604] [Table 124]

[0605] [Table 125]

[0606] In one embodiment, the CD2 ABM is the CD2-1 CDR sequence (sequence number 87~ 92) includes. In one embodiment, CD2 ABM is a heavy chain and light chain of CD2-1. Includes a variant sequence (sequences 93-94). In one embodiment, CD2 ABM is hu Includes heavy and light chain variable sequences of 1CD2-1 (SEQ ID NOs: 95-96). In one embodiment, CD2 ABM is a variable heavy and light chain sequence of hu2CD2-1 (SEQ ID NOs: 97-99). Includes 8).

[0607] In another embodiment, CD2 ABM was introduced on May 16, 2012, by Chinese Culture Collection Committee General Mic A hive deposited with the Robiology Center under accession number CGMCC 6132 The CDR sequence of antibody 9D1 produced by ridoma may be included, as published in the Chinese Patent Application Publication. As described in Patent Publication No. 102827281A. In other embodiments, CD2 ABM was established on June 22, 1999, by American Type Culture Co. Produced by a hybridoma deposited in the collection with accession number PTA-802. The antibody LO-CD2b may contain a CDR sequence, which is the subject of U.S. Patent Application Publication No. 2003 / It is described in specification 0139579 A1. In another embodiment, CD2 ABM was deposited with the ATCC on April 9, 1993, under accession number 69277, and is a U.S. Patent. Specification No. 5,795,572 describes recombinant Escherichia coli (E. coli) It contains the CD2 SFv-Ig CDR sequence produced by the expression of the synthesized construct. obtain.

[0608] In other embodiments, the CD2 ABM may include the VH and VL sequences of antibody 9D1. In other embodiments, the CD2 ABM uses the VH and VL sequences of the antibody LO-CD2b. It may include. In yet another embodiment, CD2 ABM is ATCC accession number 6927 Produced by expression of a cloned construct in recombinant Escherichia coli (E. coli) containing 7. It may contain the VH and VL sequences of CD2 SFv-Ig.

[0609] 7.9.2. CD58-based CD2 ABM In certain embodiments, this disclosure provides an MBM comprising a ligand CD2 ABM. CD2 ABM is a human immunoglobulin that uses CD58, also known as LFA-3, as its natural ligand. It specifically binds to CD2. The CD58 / LFA-3 protein is found on the surface of various cell types. It is an expressed glycoprotein (Dustin et al., 1991, Annu. Re v.Immunol.9:27), APC and It plays a role in mediating T cell interactions (Wallner et al., 1987, (J.Exp.Med.166:923). Therefore, in certain embodiments, CD2 A BM is part of CD58. When used herein, part of CD58 is CD5 At least 70% sequence identity for the CD2 binding portion of 8, e.g., CD2 of CD58 At least 70%, 71%, 72%, 73%, 74%, 75%, and 76% of the bonded portion. 77%, 78%, 79%, 80%, 81%, 82%, 83%, 84%, 85%, 86% 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96% Includes an amino acid sequence with 97%, 98%, or 99% identity. Human CD58 sequence It has the Uniprot identifier P19256. Full-length CD58 contains amino acid residues 30- CD5 containing 123 (i.e., the sequence shown as CD58-4 in Table 14 below) It has been established that 8 fragments are sufficient for binding to CD2. Wang et al. al., 1999, Cell 97:791-803. Therefore, in a particular manner The CD58 portion has at least 70% of the same sequence as amino acids 30-123 of CD58. Unisexuality, for example, at least 70% to 71% of the amino acid sequence indicated as CD58-4 72%, 73%, 74%, 75%, 76%, 77%, 78%, 79%, 80%, 81% 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91% Including 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% identity. It contains an amino acid sequence.

[0610] The interaction between CD58 and CD2 is mapped by X-ray crystallography and molecular modeling. It is being linked. Residues E25, K29, K30, K32, D33, K34, E37, D84 Substitution of K87 (numbering refers to mature polypeptide) reduces binding to CD2. Yes. Ikemizu et al., 1999, Proc. Natl. Acad. Sc i.USA 96:4289-94. Therefore, in a preferred embodiment, CD58 The parts are E25, K29, K30, K32, D33, K34, E37, D84 and K87 It retains wild-type residues.

[0611] In contrast, the following substitutions (numbered to refer to full-length polypeptides) are linked to CD2. The following did not affect the combination: F29S; V37K; V49Q; V86K; T113S; and L121G. Therefore, the CD58 portion is one, two, three, four, five of the aforementioned substitutions. Or it may include all six.

[0612] An example of the CD58 portion is shown in Table 14 below.

[0613] [Table 126]

[0614] 7.9.1. CD48-based CD2 ABM In certain embodiments, the Disclosure provides an MBM including a CD2 ABM which is the CD48 portion. Provided in this specification, the CD48 portion is used in relation to the CD2 bond portion of the CD48. In contrast, at least 70% sequence identity, for example, less than 70% of the CD2 binding portion of CD48 At least 70%, 71%, 72%, 73%, 74%, 75%, 76%, 77%, 78%, 7 9%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 8 9%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or Contains amino acid sequences with 99% identity. The human CD48 sequence is a signal peptide. Unip (containing amino acids 1-26) and GPI anchors (amino acids 221-243) rot identifier P09326(www.uniprot.org / uniprot / P09 326) has. In a particular embodiment, the CD48 portion is Uniprot identifier P09 For an amino acid sequence consisting of 326 amino acids, at least 70% of the sequence Identity (for example, at least 70%, 71%, 72%, 73%, 74%, 75%, 76%) 77%, 78%, 79%, 80%, 81%, 82%, 83%, 84%, 85%, 86% 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96% It contains an amino acid sequence that has 97%, 98%, or 99% identity. Human CD48 is I g-type C2 I domain (amino acids 29-127 of Uniprot identifier P09326) and Ig-like C2 type 2 domain (Amino acids 132- of Uniprot identifier P09326) 212) has. Therefore, in one embodiment, the CD48 portion is Unipro For the amino acid sequence consisting of amino acids 29-212 of t identifier P09326, C2-I For type domains (amino acids 29-127 of Uniprot identifier P09326) and / or Ig-like C2 type 2 domain (Amino acid 132 of Uniprot identifier P09326) For ~212), at least 70% sequence identity (e.g., at least 70%, 71 %, 72%, 73%, 74%, 75%, 76%, 77%, 78%, 79%, 80%, 81 %, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91 (Identity of %, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%) The amino acid sequence is included. The CD48 portion is, in one embodiment, Uniprot identification. It may contain one or more native variants related to the sequence of child P09326. For example, CD48 The portion may include the E102Q substitution. As another example, the CD48 portion is CD-48 isophon A segment or its CD2 coupling portion, for example, having Uniprot identifier P09326-2 It may include an isoform or an amino acid sequence corresponding to its CD2 binding portion.

[0615] 7.10. Tumor-associated antigen ABM MBM includes at least one ABM that specifically binds to tumor-associated antigens (TAAs). For example, BBM may contain ABM2 that specifically binds to TAA, and TBM may contain TA ABM2 ("TAA1") that specifically binds to A and A that specifically binds to a different TAA. It may include MB3 ("TAA2"). Preferably, TAA (or TAA1 and TAA2) In this case, each TAA is a human TAA. Antigens can also be present in normal cells. This may not always be the case. In certain embodiments, TAAs are found to be different in tumor cells compared to normal cells. In other embodiments, TAA is preferentially expressed or upregulated in the cell system. He is a maker.

[0616] Any type of tumor and any type of TAA can be targeted by MBM. This is predicted. Exemplary types of cancer that can be targeted include acute lymphoblastic cancer. Leukemia, acute myeloid leukemia, cholangiocarcinoma, B-cell leukemia, B-cell lymphoma, cholangiocarcinoma, bone cancer, brain Cancer, breast cancer, triple-negative breast cancer, cervical cancer, Burkitt lymphoma, chronic lymphocytic leukemia Diseases, chronic myeloid leukemia, colorectal cancer, endometrial cancer, esophageal cancer, gallbladder cancer, stomach cancer, gastrointestinal cancer, neuropathy Glioma, hairy cell leukemia, head and neck cancer, Hodgkin lymphoma, liver cancer, lung cancer, medullary thyroid carcinoma, Melanoma, multiple myeloma, ovarian cancer, non-Hodgkin lymphoma, pancreatic cancer, prostate cancer, pulmonary ductal cancer, These include kidney cancer, sarcoma, skin cancer, testicular cancer, urothelial carcinoma, and other bladder cancers. However, Those skilled in the art will understand that TAA is known in virtually all types of cancer. Ro.

[0617] Exemplary types of B-cell malignancies that can be targeted include Hodgkin lymphoma and non-Hodgkin lymphoma. Examples include quinn lymphoma (NHL) and multiple myeloma. An example of NHL is diffuse lymphoma. Large B-cell lymphoma (DLBCL), follicular lymphoma, chronic lymphocytic leukemia (CLL) ) / Small lymphocytic lymphoma (SLL), mantle cell lymphoma (MCL), marginal zone lymphoma Burkitt lymphoma, lymphoplasmacytic lymphoma (Waldenstrem macroglobulin tumor) Phosphateemia, hair cell leukemia, primary central nervous system (CNS) lymphoma, primary mediastinal large cell carcinoma Type B-cell lymphoma, mediastinal gray zone lymphoma (MGZL), splenic marginal zone B-cell lymphoma, MA LT type extranodal marginal zone B-cell lymphoma, nodal marginal zone B-cell lymphoma, and primary exudative lymphoma Swelling is one example.

[0618] Examples of MBMs that can be generated (e.g., targeted by ABM2 and / or ABM3) The exemplary TAAs are ABCF1;ACVR1;ACVR1B;ACVR2;ACVR 2B;ACVRL1;ADORA2A;ADRB3;Agrican;AGR2;AICDA ;AIF1;AIG1;AKAP1;AKAP2;ALK;AMH;AMHR2;ANG PT1;ANGPT2;ANGPTL3;ANGPTL4;ANPEP;APC;APO C1; AR; AZGP1 (zinc-a-glycoprotein); B7.1; B7.2; BAD; B AFF; BAG1; BAI1; BCL2; BCL6; BDNF; BLNK; BLR1(M DR15); BlyS; BMP1; BMP2; BMP3B (GDF10); BMP4; B MP6; BMP8; BMPR1A; BMPR1B; BMPR2; BPAG1 (preptin ); BRCA1; C19orf10 (IL27w); C3; C4A; C5; C5R1; cal dherin 17; CANT1; CASP1; CASP4; CAV1; CCBP2 (D6 / J AB61); CCL1 (1-309); CCL11 (eotaxin); CCL13 (MC P-4); CCL15 (MIP-1d); CCL16 (HCC-4); CCL17 (TA RC); CCL18 (PARC); CCL19 (MIP-3b); CCL2 (MCP-1 ); MCAF; CCL20 (MIP-3a); CCL21 (MIP-2); SLC; ex odus-2; CCL22 (MDC / STC-1); CCL23 (MPIF-1); CC L24 (MPIF-2 / eotaxin-2); CCL25 (TECT); CCL26 (e otaxin-3); CCL27 (CTACK / ILC); CCL28; CCL3 (MIP -1a); CCL4 (MIP-1b); CCL5 (RANTES); CCL7 (MCP- 3); CCL8 (mcp-2); CCNA1; CCNA2; CCND1; CCNE1; C CNE2; CCR1 (CKR1 / HM145); CCR2 (mcp-1RB / RA); C CR3 (CKR3 / CMKBR3); CCR4; CCR5 (CMKBR5 / ChemR1 3); CCR6 (CMKBR6 / CKR-L3 / STRL22 / DRY6); CCR7( CKR7 / EBI1); CCR8 (CMKBR8 / TER1 / CKR-L1); CCR9 (GPR-9-6);CCRL1(VSHK1);CCRL2(L-CCR);CD16 4;CD19;CD1C;CD20;CD200;CD-22;CD24;CD28;C D3;CD37;CD38;CD3E;CD3G;CD3Z;CD4;CD32b;CD 40;CD40L;CD44;CD45RB;CD52;CD69;CD72;CD74 ;CD79A;CD79B;CD8;CD80;CD81;CD83;CD86;CD9 7;CD179a;CDH1(E-カドヘリン);CDH10;CDH12;CDH13 ;CDH18;CDH19;CDH20;CDH5;CDH7;CDH8;CDH9;C DK2;CDK3;CDK4;CDK5;CDK6;CDK7;CDK9;CDKN1A (p21Wap1 / Cip1);CDKN1B(p27Kip1);CDKN1C;CD KN2A(p16INK4a);CDKN2B;CDKN2C;CDKN3;CEBPB ;CER1;CHGA;CHGB;Kitchenase;CHST10;CKLFSF2;CKL FSF3;CKLFSF4;CKLFSF5;CKLFSF6;CKLFSF7;CKL FSF8;CLDN3;CLDN6;CLDN7(クローディン-7);CLN3;CL U(クラステリン);CMKLR1;CMKOR1(RDC1);CNR1;COL18 A1;COL1A1;COL4A3;COL6A1;CR2;CRP;CSF1(MC SF);CSF2(GM-CSF);CSF3(GCSF);CTLA4;CTNNB1 (b-カテニン);CTSB(カテプシンB);CX3CL1(SCYD1);CX3C R1(V28);CXCL1(GRO1);CXCL10(IP-10);CXCL11 (1-TAC / IP-9);CXCL12(SDF1);CXCL13;CXCL14; CXCL16;CXCL2(GRO2);CXCL3(GRO3);CXCL5(ENA -78 / LIX);CXCL6(GCP-2);CXCL9(MIG);CXCR3(G PR9 / CKR-L2);CXCR4;CXCR6(TYMSTR / STRL33 / Bo nzo);CYB5;CYC1;CYSLTR1;CGRP;C1q;C1r;C1;C 4a;C4b;C2a;C2b;C3a;C3b;DAB2IP;DES;DKFZp4 51J0118;DNCL1;DPP4;E-selectin;E2F1;ECGF1;ED G1;EFNA1;EFNA3;EFNB2;EGF;EGFR;EGFRvIII;E LAC2;ENG;ENO1;ENO2;ENO3;EPHB4;EPO;ERBB2( Her-2);EREG;ERK8;ESR1;ESR2;F3(TF);Factor VII ;Factor IX;Factor V;Factor VIIa;Factor X;Factor XII;Factor XIII; FADD; FasL; FASN; FCER1A; FCER2; Fc gamma receptor; FCG R3A;FCRL5;FGF;FGF1(aFGF);FGF10;FGF11;FGF 12;FGF12B;FGF13;FGF14;FGF16;FGF17;FGF18; FGF19;FGF2(bFGF);FGF20;FGF21;FGF22;FGF23 ;FGF3(int-2);FGF4(HST);FGF5;FGF6(HST-2); FGF7(KGF);FGF8;FGF9;FGFR3;FIGF(VEGFD);FI L1(EPSILON);FIL1(ZETA);FLJ12584;FLJ25530 FLRT1 (fibronectin); FLT1; folate receptor alpha; folate receptor beta ;FOS;FOSL1(FRA-1);Face GM1;FY(DARC);GABRP (MARKET);GAGEB1;GAGEC1;GALNAC4S-6ST;GATA3 ;GDF5;GFI1;GGT1;GM-CSF;GloboH;GNAS1;GNRH 1;GPNMB;GPR2(CCR10);GPR20;GPR31;GPR44;GP R64;GPR81(FKSG80);GPRC5D;GRCC10(C10);GRP ;GSN(glass);GSTP1;gP(gP)IIb / IIIa;HAV CR1;HAVCR2;HDAC4;HDAC5;HDAC7A;HDAC9;Her2 ;HER3;HGF;HIF1A;HIP1;Hamster and Hematopoietic Progenitors;HLA -A;HLA-DRA;HM74;HMGB1;HMOX1;HMWMAA;HUMCY T2A;ICEBERG;ICOSL;ID2;IFN-a;IFNA1;IFNA2; IFNA4;IFNA5;IFNA6;IFNA7;IFNB1;IFN-γ;IFNW 1;IGBP1;IGF1;IGF1R;IGF2;IGFBP2;IGFBP3;IG FBP6;IL-1;IL-α;IL-1-β;IL10;IL10RA;IL10RB ;IL11;IL11RA;IL-12;IL12A;IL12B;IL12RB1;I L12RB2;IL13;IL13RA1;IL13RA2;IL14;IL15;IL IL15RA;IL16;IL17;IL17B;IL17C;IL17R;IL18;IL 18BP;IL18R1;IL18RAP;IL19;IL1A;IL1B;IL1F1 0;IL1F5;IL1F6;IL1F7;IL1F8;IL1F9;IL1HY1;I L1R1;IL1R2;IL1RAP;IL1RAPL1;IL1RAPL2;IL1R L1;IL1RL2;IL1RN;IL2;IL20;IL20RA;IL21R;IL 22;IL22R;IL22RA2;IL23;IL24;IL25;IL26;IL2 7;IL28A;IL28B;IL29;IL2RA;IL2RB;IL2RG;IL3 ;IL30;IL3RA;IL4;IL4R;IL5;IL5RA;IL6;IL6R; IL6ST(protease130);IL7;IL7R;IL8;IL8RA;IL8R B;IL8RB;IL9;IL9R;ILK;INHA;INHBA;INSL3;IN SL4;IRAQ1;IRAQ2;ITGA1;ITGA2;ITGA3;ITGA6( a6case);ITGAV;ITGB3;ITGB4(b4case);JA G1;JAK1;JAK3;JUN;K6HF;KAI1;KDR;KITLG;KLF 5(GC Box BP);KLF6;KLK10;KLK12;KLK13;KLK1 4;KLK15;KLK3;KLK4;KLK5;KLK6;KLK9;KRT1;KR T19(ツ19);KRT2A;KRTHB6(KR2 Type Type      Type);L -Lingo;LAMAS;LEP(Lingo);Lingo-p75;Lingo-T roy;LRP6;LPS;LTA(TNF-b);LTB;LTB4R(GPR16) ;LTB4R2;LTBR;LY6K;LYPD8;MACMARCKS; mgp;MAP2K7(c-Jun);MDK;MAP2K7(c-Jun);MIB1;MAP2K7; MIF;MIP-2;MKI67(Ki-67);MMP2;MMP9;MS4A1;M SMB;MT3(mucosa-III);MTSS1;MUC1(muco);M YC;MYD88;NCK2;Neurocan;NKG2D;NFKB1;NFKB2;N GF;NGFB(NGF);NGFR;NgR-Lingo;NgR-Nogo66(N ogo);NgR-p75;NgR-Troy;NME1(NM23A);NOX5;N PPB;NR0B1;NR0B2;NR1D1;NR1D2;NR1H2;NR1H3; NR1H4;NRII2;NRII3;NR2C1;NR2C2;NR2E1;NR2E 3;NR2F1;NR2F2;NR2F6;NR3C1;NR3C2;NR4A1;NR 4A2;NR4A3;NR5A1;NR5A2;NR6A1;NRP1;NRP2;NT 5E;NTN4;NY-BR-1;o-acetyl-GD2;ODZ1;OPRD1;OR 51E2;P2RX7;PANX3;PAP;PART1;PATE;PAWR;PCA 3;PCNA;PDGFA;PDGFB;PECAM1;PF4(CXCL4);PGE 2; PGF; PGR; Phosphacan; PIAS2; PIK3CG; PLAC1; Plasmi Nogen activator; PLAU(uPA); PLG; PLXDC1; Polysialic acid; PP BP(CXCL7);PPID;PR1;PRKCQ;PRKD1;PRL;PROC; Protein C; PROK2; PSAP; PSCA; PTAFR; PTEN; PTGS2 ( COX-2);PTN;RAC2(p21Rac2);RAGE;RARB;RGS1; RGS13;RGS3;RNF110(ZNF144);ROBO2;SIO0A2;S CGB1D2 (Lipophyllin B); SCGB2A1 (Mammaglobin 2); SCGB2A 2 (Mammaglobin 1); SCYE1 (Endothelial monocyte-activating cytokine); SDF2; SE RPIN1;SERPIN3;SERPINB5(snail);SERPIN1( PAI-1);SERPINF1;SHBG;SLA2;SLC2A2;SLC33A1 ;SLC34A2;SLC39A6;SLC43A1;SLIT2;SLITRK6;S PP1;SPRR1B(Spr1);ST6GAL1;STAB1;STAT6;STE AP;STEAP2;THE P;TACSTD2;TB4R2;TBX21;TC P10;TDGF1;TEK;TEM1 / CD248;TEM7R;TGFA;TGFB 1;TGFB111;TGFB2;TGFB3;TGFBI;TGFBR1;TGFBR 2;TGFBR3;TH1L;THBS1(THB-1);THBS2;T HBS4;THPO;TIE(Tie-1);TIMP3;protease;TLR10;TL R2;TLR3;TLR4;TLR5;TLR6;TLR7;TLR8;TLR9;TN F;TNF-α;TNFAIP2(B94);TNFAIP3;TNFRSF11A;T NFRSF1A;TNFRSF1B;TNFRSF21;TNFRSF5;TNFRSF 6 (Fas);TNFRSF7;TNFRSF8;TNFRSF9;TNFSF10(TR AIL);TNFSF11(TRANC);TNFSF12(APO3L);TNFS F13(April);TNFSF13B;TNFSF14(WHEM-L);TNFS F15(VEGI);TNFSF18;TNFSF4(OX40β);TNFSF 5(CD40 factor);TNFSF6(FasL);TNFSF7(CD27 factor). );TNFSF8(CD30 factor);TNFSF9(4-1BB factor);TOL LIP; Toll-like receptor; TOP2A (topoisomerase ha); TP53; TPM1 ;TPM2;TRADD;TRAF1;TRAF2;TRAF3;TRAF4;TRAF 5;TRAF6;TREM1;TREM2;TRPC6;TSHR;TSLP;TWEA K; Trombomodulin; Thrombin; UPK2; VEGF; VEGFB; VEGFC; Versican; VHLC5; VLA-4; XCL1 (lymphotactin); XCL2 (SCM) -1b);XCR1(GPRS / CCXCR1);YY1;and ZFPM2 are examples. .

[0619] In one embodiment, the TAA targeted by MBM is ADRB3. In one embodiment, the TAA targeted by MBM is AKAP-4. In one embodiment, the TAA targeted by MBM is ALK. In this context, the TAA targeted by MBM is the androgen receptor. In one embodiment, the TAA targeted by MBM is B7H3. In this embodiment, the TAA targeted by MBM is BCMA. The TAA targeted by MBM is BORIS. In one embodiment, M The TAA targeted by BM is BST2. In one embodiment, MBM Therefore, the TAA that is targeted is cadherin 17. In one embodiment, MBM Therefore, the TAA that is targeted is CAIX. In one embodiment, by MBM The TAA to be targeted is CD171. In one embodiment, it is targeted by MBM. The TAA used is CD179a. In one embodiment, the target is The TAA being targeted is CD19. In one embodiment, it is targeted by MBM. TAA is CD20. In one embodiment, TAA is targeted by MBM This is CD22. In one embodiment, the TAA targeted by MBM is C D24. In one embodiment, the TAA targeted by MBM is CD30 In one embodiment, the TAA targeted by MBM is CD300LF In one embodiment, the TAA targeted by MBM is CD32b. In one embodiment, the TAA targeted by MBM is CD33. In one embodiment, the TAA targeted by MBM is CD38. Morphologically, the TAA targeted by MBM is CD44v6. In this configuration, the TAA targeted by MBM is CD72. In one embodiment, the TAA targeted by MBM is CD79a. The TAA targeted by MBM is CD79b. In one embodiment, M The TAA targeted by BM is CD97. In one embodiment, MBM Therefore, the TAA that is targeted is CEA. In one embodiment, it is targeted by MBM. The TAA that is targeted is CLDN6. In one embodiment, it is targeted by MBM The TAA being targeted is CLEC12A. In one embodiment, the MBM is used to target The TAA being targeted is CLL-1. In one embodiment, it is targeted by MBM. The TAA is CS-1. In one embodiment, the TAA targeted by MBM A is CXORF61. In one embodiment, TA is targeted by MBM. A is cyclin B1. In one embodiment, TA is targeted by MBM. A is CYP1B1. In one embodiment, TAA is targeted by MBM. is EGFR. In one embodiment, the TAA targeted by MBM is E It is GFRvIII. In one embodiment, the TAA targeted by MBM is This is EMR2. In one embodiment, the TAA targeted by MBM is EPC It is AM. In one embodiment, the TAA targeted by MBM is EphA2 In one embodiment, the TAA targeted by MBM is EphB2. In one embodiment, the TAA targeted by MBM is ERBB2. In one embodiment, the TAA targeted by MBM is ERG(TMPRSS2 (ETS fusion gene) In one embodiment, TA targeted by MBM A is ETV6-AML. In one embodiment, T is targeted by MBM. AA is FAP. In one embodiment, TAA targeted by MBM is It is FCAR. In one embodiment, the TAA targeted by MBM is FCR It is L5. In one embodiment, the TAA targeted by MBM is FLT3. Yes. In one embodiment, the TAA targeted by MBM is FLT3. In one embodiment, the TAA targeted by MBM is the folate receptor alpha. In one embodiment, TAA targeted by MBM is folate receptor beta. Yes. In one embodiment, the TAA targeted by MBM is Fos-related antigen 1. In one embodiment, the TAA targeted by MBM is fucosyl GM1 In one embodiment, the TAA targeted by MBM is GD2. In one embodiment, the TAA targeted by MBM is GD2. In one embodiment, the TAA targeted by MBM is GD3. In one embodiment, the TAA targeted by MBM is GloboH. Therefore, the TAA targeted by MBM is GM3. In one embodiment, MB The TAA targeted by M is gp100Tn. In one embodiment, MB The TAA targeted by M is GPC3. In one embodiment, MBM The TAA that is targeted is GPNMB. In one embodiment, by MBM The TAA to be targeted is GPR20. In one embodiment, it is targeted by MBM. The TAA used is GPRC5D. In one embodiment, the target is MBM The TAA being targeted is GPR64. In one embodiment, it is targeted by MBM. The TAA is HAVCR1. In one embodiment, it is targeted by MBM. TAA is HER3. In one embodiment, TAA is targeted by MBM. This is HMWMAA. In one embodiment, the TAA targeted by MBM is , hTERT. In one embodiment, the TAA targeted by MBM is I It is a gf-I receptor. In one embodiment, the TAA targeted by MBM is It is IGLL1. In one embodiment, the TAA targeted by MBM is IL It is -11Ra. In one embodiment, the TAA targeted by MBM is IL It is -13Ra2. In one embodiment, the TAA targeted by MBM is K It is IT. In one embodiment, the TAA targeted by MBM is LAIR1 In one embodiment, the TAA targeted by the MBM is the LCK. In one embodiment, the TAA targeted by MBM is LewisY. In one embodiment, the TAA targeted by MBM is LILRA2. In one embodiment, the TAA targeted by MBM is LMP2. In this configuration, the TAA targeted by MBM is LRP6. And the TAA targeted by MBM is LY6K. In one embodiment, The TAA targeted by MBM is LY75. The TAA targeted by is LYPD8. In one embodiment, by MBM The TAA that is targeted is MAD-CT-1. In one embodiment, MBM Therefore, the TAA that is targeted is MAD-CT-2. In one embodiment, MBM The TAA targeted by is mesothelin. In one embodiment, MBM The TAA that is targeted is ML-IAP. In one embodiment, by MBM The TAA that is targeted is MUC1. In one embodiment, it is targeted by MBM. The TAA that is targeted is MYCN. In one embodiment, it is targeted by MBM. The TAA is NA17. In one embodiment, the TAA targeted by MBM A is NCAM. In one embodiment, the TAA targeted by MBM is It is NKG2D. In one embodiment, the TAA targeted by MBM is NY -BR- It is acetyl-GD2. In one embodiment, the TAA targeted by MBM is , OR51E2. In one embodiment, the TAA targeted by MBM is It is OY-TES1. In one embodiment, the TAA targeted by MBM is It is a p53 variant. In one embodiment, the TAA targeted by MBM is P This is ANX3. In one embodiment, the TAA targeted by MBM is PAX 3. In one embodiment, the TAA targeted by MBM is PAX5. In one embodiment, the TAA targeted by MBM is PDGFR-beta. In one embodiment, the TAA targeted by MBM is PLAC1. In one embodiment, the TAA targeted by MBM is polysialic acid. In one embodiment, the TAA targeted by MBM is PRSS21. In one embodiment, the TAA targeted by MBM is PSCA. In one embodiment, the TAA targeted by MBM is RhoC. In this case, the TAA targeted by MBM is ROR1. Therefore, TAA, which is targeted by MBM, is a sarcoma translocation breakpoint protein. In one embodiment, the TAA targeted by MBM is SART3. In one embodiment, the TAA targeted by the MBM is SLC34A2. In one embodiment, the TAA targeted by the MBM is SLC39A6. In one embodiment, the TAA targeted by MBM is sLe. In this state, the TAA targeted by MBM is SLITRK6. In this state, the TAA targeted by MBM is sperm protein 17. In the implementation configuration, the TAA targeted by MBM is SSEA-4. In one embodiment, the TAA targeted by MBM is SSX2. In one embodiment, the TAA targeted by MBM is TAAG72. In one embodiment, the TAA targeted by MBM is TAARP. In one embodiment, the TAA targeted by MBM is TACSTD2. The TAA targeted by MBM is TEM1 / CD248. In one embodiment, In one embodiment, the TAA targeted by MBM is TEM7R. Therefore, the TAA targeted by MBM is TGS5. In one embodiment, M The TAA targeted by BM is Tie 2. In one embodiment, MBM The TAA targeted by is Tn Ag. In one embodiment, by MBM The TAA that is targeted is TSHR. In one embodiment, it is targeted by MBM. The TAA targeted is tyrosinase. In one embodiment, it is targeted by MBM. The TAA that is targeted is UPK2. In one embodiment, it is targeted by MBM. The TAA is VEGFR2. In one embodiment, it is targeted by MBM. TAA is WT1. In one embodiment, the TAA targeted by MBM is It is XAGE1.

[0620] In one embodiment, the TAA targeted by MBM is BCMA, CD19, CD20, CD22, CD123, CD33, CLL1, CD138 (Sindekan-1, Also known as SDC1, CS1, CD38, CD133, FLT3, CD52 TNFRSF13C (also known as BAFFR: B cell activator receptor), T NF receptor superfamily members 13C, TNFRSF13B (TACI: transmembrane) TNF receptor superpharmaceuticals, also known as activators and CAML interactors Millie member 13B), CXCR4 (CXC motif chemokine receptor 4), PD -L1 (programmed cell death ligand 1), LY9 (also known as CD229, IgG (9), CD200, FCGR2B (also known as CD32b), IgG Receptor IIb Fc fragment), CD21, CD23, CD24, CD40L, CD Selected from 72, CD79a, and CD79b.

[0621] In one embodiment, the TAA targeted by MBM is CD19. In one embodiment, the TAA targeted by MBM is BCMA. In this configuration, the TAA targeted by MBM is CD20. And the TAA targeted by MBM is CD22. In one embodiment, The TAA targeted by MBM is CD123. The TAA targeted by M is CD33. In one embodiment, MBM The TAA that is targeted is CLL1. In one embodiment, it is targeted by MBM. The TAA that is targeted is CD138. In one embodiment, the MBM is targeted The TAA being targeted is CS1. In one embodiment, the T targeted by MBM is CS1. AA is CD38. In one embodiment, the TAA targeted by MBM is , CD133. In one embodiment, the TAA targeted by MBM is F It is LT3. In one embodiment, the TAA targeted by MBM is CD52 In one embodiment, the TAA targeted by MBM is TNFRSF1 3C. In one embodiment, the TAA targeted by MBM is TNFRS It is F13B. In one embodiment, the TAA targeted by MBM is CXC R4. In one embodiment, the TAA targeted by MBM is PD-L1 In one embodiment, the TAA targeted by MBM is LY9. In one embodiment, the TAA targeted by MBM is CD200. In one embodiment, the TAA targeted by MBM is CD21. In this configuration, the TAA targeted by MBM is CD23. And the TAA targeted by MBM is CD24. In one embodiment, The TAA targeted by MBM is CD40L. In one embodiment, MB The TAA targeted by M is CD72. In one embodiment, MBM The TAA that is targeted is CD79a. In one embodiment, by MBM The TAA targeted is CD79b.

[0622] In one embodiment, MBM is a selection of two T's from the TAA described in this section. The target is AA (TAA1 and TAA2).

[0623] In one embodiment, TAA1 is CD19 and TAA2 is CD20. (Or vice versa). In one embodiment, TAA1 is CD19 and TAA2 is CD22 (and vice versa). In one embodiment, TAA1 is CD 19, and TAA2 is CD123 (or vice versa). In one embodiment Therefore, TAA1 is CD19, and TAA2 is BCMA (or vice versa). In one embodiment, TAA1 is CD19 and TAA2 is CD33. (Or vice versa). In one embodiment, TAA1 is CD19, and TAA 2 is CLL1 (and vice versa). In one embodiment, TAA1 is C D19 is CD138, and TAA2 is CD138 (and vice versa). In one embodiment, In this case, TAA1 is CD19, and TAA2 is CS1 (or vice versa). In one embodiment, TAA1 is CD19 and TAA2 is CD38. (Or vice versa). In one embodiment, TAA1 is CD19, and TAA 2 is CD133 (and vice versa). In one embodiment, TAA1 is CD19 is FLT3, and TAA2 is FLT3 (and vice versa). In one embodiment, In this case, TAA1 is CD19, and TAA2 is CD52 (or vice versa). (In one embodiment, TAA1 is CD19 and TAA2 is TNFRSF) It is 13C (and vice versa). In one embodiment, TAA1 is CD19 Yes, TAA2 is TNFRSF13B (and vice versa). In one embodiment, In this case, TAA1 is CD19, and TAA2 is CXCR4 (or vice versa). In one embodiment, TAA1 is CD19 and TAA2 is PD-L1 (And vice versa.) In one embodiment, TAA1 is CD19, TAA2 is LY9 (and vice versa). In one embodiment, TAA1 is CD19 is CD200, and TAA2 is CD200 (and vice versa). In this state, TAA1 is CD19 and TAA2 is FCGR2B (or vice versa). (Similarly). In one embodiment, TAA1 is CD19, and TAA2 is CD It is 21 (or vice versa). In one embodiment, TAA1 is CD19 Therefore, TAA2 is CD23 (and vice versa). In one embodiment, TA A1 is CD19, and TAA2 is CD24 (and vice versa). In the configuration, TAA1 is CD19 and TAA2 is CD40L (or vice versa). (The same applies to the other.) In one embodiment, TAA1 is CD19 and TAA2 is C It is D72 (and vice versa). In one embodiment, TAA1 is CD19 Yes, TAA2 is CD79a (and vice versa). In one embodiment, TAA1 is CD19, and TAA2 is CD79b (and vice versa). In one embodiment, TAA1 is CD20 and TAA2 is CD22 (and (The reverse is also true). In one embodiment, TAA1 is CD20 and TAA2 is , CD123 (or vice versa). In one embodiment, TAA1 is CD In one embodiment, TAA2 is BCMA (and vice versa). Therefore, TAA1 is CD20, and TAA2 is CD33 (and vice versa). In one embodiment, TAA1 is CD20 and TAA2 is CLL1. (Or vice versa). In one embodiment, TAA1 is CD20 and TAA2 is CD138 (and vice versa). In one embodiment, TAA1 is C D20 is CS1, and TAA2 is CS1 (and vice versa). In one embodiment, Therefore, TAA1 is CD20, and TAA2 is CD38 (and vice versa). In one embodiment, TAA1 is CD20 and TAA2 is CD133. (Or vice versa). In one embodiment, TAA1 is CD20, and TAA 2 is FLT3 (and vice versa). In one embodiment, TAA1 is C D20 is CD52, and TAA2 is CD52 (and vice versa). In one embodiment And TAA1 is CD20, and TAA2 is TNFRSF13C (or vice versa). (Similarly). In one embodiment, TAA1 is CD20 and TAA2 is TN It is FRSF13B (and vice versa). In one embodiment, TAA1 is C D20 is CXCR4, and TAA2 is CXCR4 (and vice versa). In one embodiment In this case, TAA1 is CD20, and TAA2 is PD-L1 (or vice versa). In one embodiment, TAA1 is CD20 and TAA2 is LY9. (or vice versa). In one embodiment, TAA1 is CD20, and TA A2 is CD200 (and vice versa). In one embodiment, TAA1 is CD20 is FCGR2B, and TAA2 is FCGR2B (and vice versa). In terms of form, TAA1 is CD20, and TAA2 is CD21 (or vice versa). (It is as follows). In one embodiment, TAA1 is CD20 and TAA2 is CD2 It is 3 (or vice versa). In one embodiment, TAA1 is CD20. In one embodiment, TAA2 is CD24 (and vice versa). 1 is CD20, and TAA2 is CD40L (or vice versa). In the configuration, TAA1 is CD20, and TAA2 is CD72 (or vice versa). (Similarly). In one embodiment, TAA1 is CD20, and TAA2 is CD 79a (and vice versa). In one embodiment, TAA1 is CD20 Yes, TAA2 is CD79b (and vice versa). In one embodiment, TAA1 is CD22, and TAA2 is CD123 (and vice versa). In one embodiment, TAA1 is CD22 and TAA2 is BCMA (and (The reverse is also true). In one embodiment, TAA1 is CD22 and TAA2 is , CD33 (or vice versa). In one embodiment, TAA1 is CD2 In some embodiments, TAA2 is CLL1 (and vice versa). TAA1 is CD22, and TAA2 is CD138 (and vice versa). In one embodiment, TAA1 is CD22 and TAA2 is CS1 (and (The reverse is also true). In one embodiment, TAA1 is CD22 and TAA2 is , CD38 (and vice versa). In one embodiment, TAA1 is CD2 2, and TAA2 is CD133 (and vice versa). In one embodiment, Therefore, TAA1 is CD22, and TAA2 is FLT3 (and vice versa). In one embodiment, TAA1 is CD22 and TAA2 is CD52. (Or vice versa). In one embodiment, TAA1 is CD22 and TAA2 is TNFRSF13C (and vice versa). In one embodiment, TAA 1 is CD22, and TAA2 is TNFRSF13B (or vice versa). In one embodiment, TAA1 is CD22 and TAA2 is CXCR4. (Or vice versa). In one embodiment, TAA1 is CD22, and TAA 2 is PD-L1 (and vice versa). In one embodiment, TAA1 is CD22 is LY9, and TAA2 is LY9 (and vice versa). In one embodiment... And TAA1 is CD22, and TAA2 is CD200 (or vice versa). (In one embodiment, TAA1 is CD22 and TAA2 is FCGR2B) (And vice versa.) In one embodiment, TAA1 is CD22, TAA2 is CD21 (and vice versa). In one embodiment, TAA1 CD22 is CD23, and TAA2 is CD23 (and vice versa). In this state, TAA1 is CD22, and TAA2 is CD24 (and vice versa). (In one embodiment, TAA1 is CD22 and TAA2 is CD40) L (and vice versa). In one embodiment, TAA1 is CD22. In one embodiment, TAA2 is CD72 (and vice versa). 1 is CD22, and TAA2 is CD79a (and vice versa). In the configuration, TAA1 is CD22 and TAA2 is CD79b (or vice versa). (The same applies to the other.) In one embodiment, TAA1 is CD123, and TAA2 is It is BCMA (and vice versa). In one embodiment, TAA1 is CD12 In one embodiment, TAA2 is CD33 (and vice versa). TAA1 is CD123, and TAA2 is CLL1 (and vice versa). In one embodiment, TAA1 is CD123 and TAA2 is CD138. (and vice versa). In one embodiment, TAA1 is CD123, and T AA2 is CS1 (and vice versa). In one embodiment, TAA1 is CD123 is CD38, and TAA2 is CD38 (or vice versa). In one embodiment In this case, TAA1 is CD123, and TAA2 is CD133 (or vice versa). (Apparently so). In one embodiment, TAA1 is CD123 and TAA2 is FL T3 (and vice versa). In one embodiment, TAA1 is CD123 Yes, TAA2 is CD52 (and vice versa). In one embodiment, T AA1 is CD123, TAA2 is TNFRSF13C (or vice versa) In one embodiment, TAA1 is CD123 and TAA2 is TNFR SF13B (and vice versa). In one embodiment, TAA1 is CD1 23 is CXCR4 (and vice versa). In one embodiment And TAA1 is CD123, and TAA2 is PD-L1 (or vice versa). In one embodiment, TAA1 is CD123 and TAA2 is LY9. Yes (and vice versa). In one embodiment, TAA1 is CD123, TAA2 is CD200 (and vice versa). In one embodiment, TAA 1 is CD123, and TAA2 is FCGR2B (and vice versa). In one embodiment, TAA1 is CD123 and TAA2 is CD21 (and (The reverse is also true). In one embodiment, TAA1 is CD123 and TAA2 is CD23 (and vice versa). In one embodiment, TAA1 is CD 123 is CD24, and TAA2 is CD24 (and vice versa). In one embodiment And TAA1 is CD123, and TAA2 is CD40L (or vice versa). In one embodiment, TAA1 is CD123 and TAA2 is CD72. (And vice versa.) In one embodiment, TAA1 is CD123. , TAA2 is CD79a (and vice versa). In one embodiment, TA A1 is CD123, and TAA2 is CD79b (and vice versa). In one embodiment, TAA1 is BCMA and TAA2 is CD33 (or (The reverse is also true.) In one embodiment, TAA1 is BCMA, and TAA2 is CLL1 (and vice versa). In one embodiment, TAA1 is BCMA in Yes, TAA2 is CD138 (and vice versa). In one embodiment, TAA1 is BCMA, and TAA2 is CS1 (and vice versa). In one embodiment, TAA1 is BCMA and TAA2 is CD38 (or vice versa). (The same applies to the other.) In one embodiment, TAA1 is BCMA and TAA2 is C D133 (and vice versa). In one embodiment, TAA1 is BCMA And TAA2 is FLT3 (and vice versa). In one embodiment, TAA1 is BCMA, and TAA2 is CD52 (and vice versa). In this embodiment, TAA1 is BCMA and TAA2 is TNFRSF13C Yes (and vice versa). In one embodiment, TAA1 is BCMA and T AA2 is TNFRSF13B (and vice versa). In one embodiment, TAA1 is BCMA, and TAA2 is CXCR4 (and vice versa). In one embodiment, TAA1 is BCMA and TAA2 is PD-L1. (Or vice versa). In one embodiment, TAA1 is BCMA and TAA2 is LY9 (and vice versa). In one embodiment, TAA1 is BCM A is CD200, and TAA2 is CD200 (and vice versa). In one embodiment, So, TAA1 is BCMA, and TAA2 is FCGR2B (or vice versa). (In one embodiment, TAA1 is BCMA and TAA2 is CD21.) (or vice versa). In one embodiment, TAA1 is BCMA, and TA A2 is CD23 (and vice versa). In one embodiment, TAA1 is BCMA is CD24, and TAA2 is CD24 (and vice versa). In one embodiment In this case, TAA1 is BCMA and TAA2 is CD40L (or vice versa). In one embodiment, TAA1 is BCMA and TAA2 is CD72. Yes (and vice versa). In one embodiment, TAA1 is BCMA and T AA2 is CD79a (and vice versa). In one embodiment, TAA1 BCMA is BCMA, and TAA2 is CD79b (and vice versa). In terms of morphology, TAA1 is CD33, and TAA2 is CLL1 (or vice versa). (It appears that...) In one embodiment, TAA1 is CD33 and TAA2 is CD1 It is 38 (or vice versa). In one embodiment, TAA1 is CD33 Therefore, TAA2 is CS1 (and vice versa). In one embodiment, TAA 1 is CD33, and TAA2 is CD38 (or vice versa). In terms of form, TAA1 is CD33, and TAA2 is CD133 (or vice versa). (Similarly). In one embodiment, TAA1 is CD33 and TAA2 is FL It is T3 (and vice versa). In one embodiment, TAA1 is CD33. Therefore, TAA2 is CD52 (and vice versa). In one embodiment, TA A1 is CD33 and TAA2 is TNFRSF13C (or vice versa). ). In one embodiment, TAA1 is CD33 and TAA2 is TNFRSF1 3B (and vice versa). In one embodiment, TAA1 is CD33. Therefore, TAA2 is CXCR4 (and vice versa). In one embodiment, T AA1 is CD33, and TAA2 is PD-L1 (and vice versa). In one embodiment, TAA1 is CD33 and TAA2 is LY9 (or vice versa). (The same applies to the other.) In one embodiment, TAA1 is CD33 and TAA2 is C D200 (and vice versa). In one embodiment, TAA1 is CD33 And TAA2 is FCGR2B (and vice versa). In one embodiment Therefore, TAA1 is CD33, and TAA2 is CD21 (and vice versa). In one embodiment, TAA1 is CD33 and TAA2 is CD23. (Or vice versa). In one embodiment, TAA1 is CD33 and TAA2 is CD24 (and vice versa). In one embodiment, TAA1 is CD 33, and TAA2 is CD40L (or vice versa). In one embodiment Therefore, TAA1 is CD33, and TAA2 is CD72 (or vice versa). In one embodiment, TAA1 is CD33 and TAA2 is CD79a. (or vice versa). In one embodiment, TAA1 is CD33, and TA A2 is CD79b (and vice versa). In one embodiment, TAA1 is CLL1 is CLL1, and TAA2 is CD138 (and vice versa). A certain implementation In this state, TAA1 is CLL1, and TAA2 is CS1 (or vice versa). In one embodiment, TAA1 is CLL1 and TAA2 is CD38. Yes (and vice versa). In one embodiment, TAA1 is CLL1 and T AA2 is CD133 (and vice versa). In one embodiment, TAA1 CLL1 is CLL1, and TAA2 is FLT3 (and vice versa). In this state, TAA1 is CLL1, and TAA2 is CD52 (and vice versa). (In one embodiment, TAA1 is CLL1 and TAA2 is TNFR It is SF13C (and vice versa). In one embodiment, TAA1 is CLL 1, and TAA2 is TNFRSF13B (and vice versa). A certain implementation In this state, TAA1 is CLL1, and TAA2 is CXCR4 (or vice versa). (Apparently so). In one embodiment, TAA1 is CLL1 and TAA2 is PD- It is L1 (and vice versa). In one embodiment, TAA1 is CLL1 Therefore, TAA2 is LY9 (and vice versa). In one embodiment, TAA 1 is CLL1, and TAA2 is CD200 (and vice versa). In the application configuration, TAA1 is CLL1, and TAA2 is FCGR2B (or The reverse is also true. In one embodiment, TAA1 is CLL1, and TAA2 is CD21 (and vice versa). In one embodiment, TAA1 is CLL1 And TAA2 is CD23 (and vice versa). In one embodiment, TAA1 is CLL1, and TAA2 is CD24 (and vice versa). In one embodiment, TAA1 is CLL1 and TAA2 is CD40L (and (The reverse is also true). In one embodiment, TAA1 is CLL1 and TAA2 is , CD72 (or vice versa). In one embodiment, TAA1 is CLL 1 and TAA2 is CD79a (and vice versa). In one embodiment Therefore, TAA1 is CLL1, and TAA2 is CD79b (or vice versa). In one embodiment, TAA1 is CD138 and TAA2 is CS1. (Or vice versa). In one embodiment, TAA1 is CD138, and TA A2 is CD38 (and vice versa). In one embodiment, TAA1 is CD138 is CD133, and TAA2 is CD133 (and vice versa). In this configuration, TAA1 is CD138, and TAA2 is FLT3 (or vice versa). (It appears that...). In one embodiment, TAA1 is CD138, and TAA2 is CD It is 52 (or vice versa). In one embodiment, TAA1 is CD138 Yes, TAA2 is TNFRSF13C (and vice versa). In one embodiment, In this case, TAA1 is CD138 and TAA2 is TNFRSF13B (or (The reverse is also true). In one embodiment, TAA1 is CD138 and TAA2 is , CXCR4 (and vice versa). In one embodiment, TAA1 is CD 138, and TAA2 is PD-L1 (and vice versa). In one embodiment In this case, TAA1 is CD138, and TAA2 is LY9 (or ...

Claims

1. A CD3-binding molecule that specifically binds to human CD3, (i) Heavy chain variable region including CDR-H1 of SEQ ID NO: 200, CDR-H2 of SEQ ID NO: 201, and CDR-H3 of SEQ ID NO: 202, and (ii) Light chain variable region including CDR-L1 of SEQ ID NO: 216, CDR-L2 of SEQ ID NO: 217, and CDR-L3 of SEQ ID NO: 218 The CD3-binding molecule, which includes the above.

2. (a) A sequence of heavy chain variable region having at least 90%, 95%, 96%, 97%, 98%, or 99% identity with the sequence of Sequence ID No. 209, and a sequence of light chain variable region having at least 90%, 95%, 96%, 97%, 98%, or 99% identity with the sequence of Sequence ID No. 225; or (b) A sequence of a heavy chain variable region having at least 90%, 95%, 96%, 97%, 98%, or 99% identity with the sequence of Sequence ID No. 241, and a sequence of a light chain variable region having at least 90%, 95%, 96%, 97%, 98%, or 99% identity with the sequence of Sequence ID No. 257 A CD3-binding molecule according to claim 1, comprising:

3. (a) the arrangement of the heavy chain variable region of Sequence ID No. 209 and the arrangement of the light chain variable region of Sequence ID No. 225; or (b) Arrangement of the heavy chain variable region of Sequence ID No. 241 and the light chain variable region of Sequence ID No. 257 A CD3-binding molecule according to claim 1, comprising:

4. A CD3-binding molecule according to any one of claims 1 to 3, comprising an antibody, an antibody fragment, scFv, dsFv, Fv, Fab, scFab, or (Fab')2.

5. A CD3 binding molecule according to any one of claims 1 to 4, which is a multispecific binding molecule.

6. A bispecific binding molecule (BBM), wherein the BBM is (a) an antigen-binding module 1 (ABM1) that specifically binds to CD3 and includes the heavy chain and light chain variable region of the CD3-binding molecule described in any one of claims 1 to 3; and (b) Antigen-binding module 2 (ABM2) that specifically binds to tumor-associated antigens (TAAs). The CD3-binding molecule according to claim 5, comprising:

7. The CD3-binding molecule according to claim 6, wherein ABM1 can bind to CD3 at the same time that ABM2 binds to its target molecule.

8. A triple-specific binding molecule (TBM), wherein the TBM is (a) an antigen-binding module 1 (ABM1) that specifically binds to CD3 and includes the heavy chain and light chain variable regions of the CD3-binding molecule described in any one of claims 1 to 3; (b) Antigen-binding module 2 (ABM2) that specifically binds to tumor-associated antigens; and (c) Antigen-binding module 3 (ABM3), (i) Tumor-associated antigens other than the tumor-associated antigen conjugated by ABM2; or (ii) CD2 Antigen-binding module 3 (ABM3) that specifically binds to the antigen. The CD3-binding molecule according to claim 5, comprising:

9. The CD3-binding molecule according to claim 8, wherein ABM1 can bind to CD3 at the same time that ABM2 and ABM3 bind to their target molecules.

10. ABM2 is CD19, BCMA, TSHR, CD171, CS-1, CLL-1, GD3, TnAg, FLT3, CD38, CD44v6, B7H3, KIT, IL-13Ra2, IL-11Ra, PSCA, PRSS21, VEGFR2, LewisY, CD24, PDGFR-beta, SSEA-4, MUC1, EGFR, NCAM, CAIX, LMP2, EphA2, FucosylGM1, sLe, GM3, TGS5, HMWMAA, o-acetyl-GD2, GD2, folate receptor A Rufa, folate receptor beta, TEM1 / CD248, TEM7R, CLDN6, GPRC5D, CXORF61, CD97, CD179a, ALK, polysialic acid, PLAC1, GloboH, NY-BR-1, UPK2, HAVCR1, ADRB3, PANX3, GPR20, LY6K, OR51E2, TAARP, WT1, ETV6-AML, sperm protein 17, XAGE1, Tie2, MAD-CT-1, MAD-CT-2, Fos-related antigen 1, p53 variant, hTERT, sarcoma translocation breakpoint, ML-IAP, ERG (TMPRSS2) ETS fusion gene), NA17, PAX3, androgen receptor, cyclin B1, MYCN, RhoC, CYP1B1, BORIS, SART3, PAX5, OY-TES1, LCK, AKAP-4, SSX2, CD79a, CD79b, CD72, LAIR1, FCAR, LILRA2, CD300LF, CLEC12A, BST2, EMR2, LY75, GPC3, FCRL5, IGLL1, CD20, CD30, ERBB2, ROR 1. A CD3-binding molecule according to any one of claims 6 to 9, which specifically binds to TAA, which is TAAG72, CD22, CD33, gp100Tn, FAP, tyrosinase, EPCAM, CEA, Igf-I receptor, EphB2, mesothelin, cadherin 17, CD32b, EGFRvIII, GPNMB, GPR64, HER3, LRP6, LYPD8, NKG2D, SLC34A2, SLC39A6, SLITRK6, or TACSTD2.

11. The CD3-binding molecule according to claim 8 or 9, wherein ABM2 and ABM3 each specifically bind to TAA ("TAA 1" and "TAA 2" respectively), and TAA 1 and TAA 2 are each independently selected from CD19, CD20, CD22, CD123, BCMA, CD33, CLL1, CD138, CS1, CD38, CD133, FLT3, CD52, TNFRSF13C, TNFRSF13B, CXCR4, PD-L1, LY9, CD200, FCGR2B, CD21, CD23, CD24, CD40L, CD72, CD79a, and CD79b.

12. A conjugate comprising the CD3 binding molecule and the drug according to any one of claims 1 to 11.

13. A pharmaceutical composition comprising a CD3-binding molecule according to any one of claims 1 to 11 or a conjugate according to claim 12, and a pharmaceutically acceptable excipient.

14. A method for activating T cells in a subject requiring such activation, wherein the subject optionally has a proliferative disorder or autoimmune disorder, and a CD3-binding molecule according to any one of claims 1 to 11 for use in the method.

15. One or more nucleic acids encoding a CD3-binding molecule according to any one of claims 1 to 11.

16. A cell modified to express the CD3 binding molecule described in any one of claims 1 to 11.

17. A method for generating a CD3-binding molecule, (a) Culturing the cells described in claim 16 under conditions in which the CD3 binding molecule is expressed; and (b) Recovering the CD3 binding molecule from the cell culture. The method, including the method described above.

Images