Humanized or chimeric CD3 antibodies

Abstract

The present invention relates to humanized or chimeric CD3 antibodies. The present invention relates to humanized or chimeric antibodies that bind CD3. The invention further relates to bispecific antibodies, compositions, pharmaceutical compositions, uses of said antibodies in the treatment of disease and methods of treatment.

Description

This application is a divisional application of the patent application for invention with the application date of January 8, 2015, application number 201580013100.0 (PCT / EP2015 / 050276), and invention title of "Humanized or Chimeric CD3 Antibodies". Field of the Invention

[0001] The present invention relates to humanized or chimeric antibodies that bind to human CD3, compositions comprising said humanized or chimeric antibodies, and the use of said humanized or chimeric antibodies in the treatment of diseases. Background of the Invention Cluster of differentiation 3 (CD3) has been known for many years and is thus an object of interest in many respects. In particular, antibodies directed against CD3 or the T-cell receptor complex (of which CD3 is a part) are known. The in vitro characterization of five humanized OKT3 effector function variant antibodies has been described [1].

[0003] In the absence of continuous immunosuppressive drug therapy, treatment with the anti-CD3 monoclonal antibody hOKT3γ1(Ala-Ala) led to improved C-peptide responses and clinical parameters at least 2 years after the onset of type 1 diabetes [2].

[0004] A promising approach to improving targeted antibody therapy is by delivering cytotoxic cells specifically directed against cancer cells expressing the antigen. This concept of using T cells to effectively kill tumor cells has been described [3]. However, the initial clinical studies were rather disappointing, mainly due to low efficacy, severe side effects (cytokine release syndrome), and the immunogenicity of bispecific antibodies [4]. Advances in the design and application of bispecific antibodies have partly overcome the initial hurdles of cytokine release syndrome and improved clinical efficacy, without dose-limiting toxicity [5].

[0005] For example, certain bispecific antibodies that target an antigen on tumor cells with one arm and CD3 on, for example, T cells with the other arm provide Fc receptor binding through the Fc region. Upon binding, a complex of T cells, tumor cells, and effector cells that bind to the Fc region of the antibody is formed, resulting in the killing of tumor cells [4]. Catumaxomab consists of a murine IgG2a / rat IgG2b heterodimer and has been found to be successful in treating cancer-related ascites after intraperitoneal application [6]. However, the mouse / rat hybrid is immunogenic [7] and cannot be used for long-term intravenous therapy in humans. Frequent treatment-related adverse events are attributed to catumaxomab-induced cytokine-release-related symptoms (i.e., fever, nausea, vomiting, chills, tachycardia, and hypotension) [8]-[9], which involve the effector function of the Fc region of catumaxomab. Another antibody is ertumaxomab (HER2xCD3), which induces cytotoxicity in cell lines with low HER2 expression. Ertumaxomab has been in phase II clinical development for metastatic breast cancer

[10] -

[11] .

[0006] CD3 antibodies that cross-react with cynomolgus and / or rhesus macaque CD3 have been described

[12] -

[13] , however, further improvement of the cross-reactive antibodies is needed. Brief Description of the Invention The object of the present invention is to provide humanized or chimeric CD3 antibodies. Thus, in one aspect, the present invention relates to a humanized or chimeric antibody that binds to human CD3, wherein the antibody comprises a binding region that comprises heavy chain variable (VH) region CDR1, CDR2, and CDR3 having the sequences shown in SEQ ID NO: 1, 2, and 3, respectively, and light chain variable (VL) region CDR1, CDR2, and CDR3 having the sequences shown in SEQ ID NO: 4, the sequence GTN, and SEQ ID NO: 5 or SEQ ID NO: 60, respectively.

[0008] In one aspect, the present invention relates to a humanized or chimeric antibody that binds to human CD3, wherein the antibody comprises a binding region that comprises heavy chain variable (VH) region CDR1, CDR2, and CDR3 having the sequences shown in SEQ ID NO: 1, 2, and 3, respectively, and light chain variable (VL) region CDR1, CDR2, and CDR3 having the sequences shown in SEQ ID NO: 4, the sequence GTN, and SEQ ID NO: 5, respectively.

[0009] In another aspect, the present invention relates to a bispecific antibody that comprises a first binding region of the antibody of the present invention and a second binding region that binds to a target different from the first antigen-binding region.

[0010] In another aspect, the invention relates to a nucleic acid construct encoding one or more amino acid sequences of the invention.

[0011] In another aspect, the invention relates to an expression vector comprising (i) a nucleic acid sequence encoding a heavy chain sequence of a humanized or chimeric antibody of the invention, (ii) a nucleic acid sequence encoding a light chain sequence of a humanized or chimeric antibody of the invention, or (iii) both (i) and (ii).

[0012] In another aspect, the invention relates to a host cell comprising the expression vector of the invention.

[0013] In another aspect, the invention relates to a composition comprising an antibody or bispecific antibody of the invention.

[0014] In another aspect, the invention relates to a pharmaceutical composition comprising an antibody or bispecific antibody of the invention and a pharmaceutically acceptable carrier.

[0015] In another aspect, the invention relates to an antibody or bispecific antibody, composition or pharmaceutical composition of the invention for use as a medicament.

[0016] In another aspect, the invention relates to an antibody or bispecific antibody, composition or pharmaceutical composition of the invention for the treatment of a disease.

[0017] In another aspect, the invention relates to a method for treating a disease, comprising administering to a subject in need thereof an antibody or bispecific antibody, composition or pharmaceutical composition of the invention.

[0018] In one aspect, the invention relates to a method for diagnosing a disease characterized by the involvement or accumulation of CD3-expressing cells, comprising administering to a subject a humanized or chimeric antibody, composition or pharmaceutical composition of the invention, optionally wherein the humanized or chimeric antibody is labeled with a detectable reagent.

[0019] In another aspect, the invention relates to a method for producing an antibody or bispecific antibody of the invention, comprising the steps of: a) culturing the host cell of the invention, and b) purifying the antibody from the culture medium.

[0020] In another aspect, the invention relates to a diagnostic composition comprising an antibody or bispecific antibody of the invention.

[0021] In another aspect, the invention relates to a method for detecting the presence of CD3 antigen or CD3-expressing cells in a sample, comprising the steps of: a) contacting the sample with an antibody or bispecific antibody of the invention under conditions that permit the formation of a complex between the antibody or bispecific antibody and CD3, and b) analyzing whether a complex has been formed.

[0022] In another aspect, the present invention relates to a kit for detecting the presence of CD3 antigen or CD3-expressing cells in a sample, comprising i) an antibody or bispecific antibody of the present invention, and ii) instructions for use of the kit.

[0023] In another aspect, the present invention relates to an anti-idiotypic antibody that binds to the antibody of the present invention. Brief Description of the Drawings Figure 1 Showing ( Figure 1 A) Binding curves of monospecific antibody variants of IgG1-huCD3 and ( Figure 1 B) bispecific antibody variant bsIgG1huCD3xHER2 to the human T cell line Jurkat. The data shown are the mean fluorescence intensity (MFI) of a representative experiment as described in Example 2. The table shows the antibody concentration (μg / mL) resulting in half-maximal binding (EC50).

[0025] Figure 2 Showing ( Figure 2 A) Binding curves of monospecific antibody variants of IgG1-huCD3 and ( Figure 2 B) bispecific antibody variant bsIgG1 huCD3xHER2 to the cynomolgus monkey T cell line HSC-F. The data shown are the mean fluorescence intensity (MFI) of a representative experiment as described in Example 2.

[0026] Figure 3 : T cell activation by IgG1-huCD3 antibody variants. Expression of CD69 on T cells from human ( Figure 3 A) and cynomolgus monkey ( Figure 3 B) sources in PBMC cultures was measured by FACS analysis as described in Example 3. These experiments were performed twice, and representative results from one experiment are shown.

[0027] Figure 4 : T-cell proliferation induced by IgG1-huCD3 antibody variants. Human ( Figure 4 A) or cynomolgus monkey ( Figure 4 B) PBMCs were incubated with IgG1-huCD3 antibody variants for 3 days, after which proliferation was measured by cell proliferation ELISA as described in Example 4. Representative results from two independent experiments are shown.

[0028] Figure 5 : Human ( Figure 5 A) and cynomolgus monkey ( Figure 5 B) T cell-mediated cytotoxicity induced by huCD3 antibody variants with non-activated LFLEDA mutations, determined as shown in Example 5. Representative results from two independent experiments performed in duplicate are shown.

[0029] Figure 6 Showing the binding curves of monospecific antibody variants of non-activated IgG1-huCD3( Figure 6 A) and non-activated bispecific antibody variant bsIgG1-huCD3xHER2( Figure 6 B) to the human T cell line Jurkat. The data shown are the mean fluorescence intensity (MFI) of a representative experiment as described in Example 2. The table shows the antibody concentration (μg / mL) (EC50) that results in half-maximal binding.

[0030] Figure 7 Showing the binding curves of monospecific antibody variants of non-activated IgG1-huCD3( Figure 7 A) and non-activated bispecific antibody variant bsIgG1-huCD3xHER2( Figure 7 B) to the cynomolgus monkey T cell line HSC-F. The data shown are the mean fluorescence intensity (MFI) of a representative experiment as described in Example 2. The table shows the antibody concentration (μg / mL) (EC50) that results in half-maximal binding.

[0031] Figure 8 : T cell activation by non-activated monospecific IgG1-huCD3( Figure 8 A and B) or non-activated bispecific bsIgG1-huCD3xHER2 antibody variants( Figure 8 C and D). The expression of CD69 on T cells from human( Figure 8 A and C) and cynomolgus monkey( Figure 8 B and D) sources in PBMC cultures was measured by FACS analysis as described in Example 3. These experiments were performed twice and show representative results from one experiment.

[0032] Figure 9 : T cell proliferation induced by non-activated monospecific IgG1-huCD3( Figure 9 A and B) or non-activated bispecific bsIgG1-huCD3xHER2 antibody variants( Figure 9 C and D). T cell proliferation was measured in human( Figure 9 A and C) or cynomolgus monkey( Figure 9 B and D) PBMCs, which were incubated with various antibody variants for 3 days, after which proliferation was measured by cell proliferation ELISA as described in Example 4. Representative results from two independent experiments are shown.

[0033] Figure 10 : Induction of human by huCD3 antibody variants with non-activated LFLEDA mutationsFigure 10 A) and cynomolgus monkeys ( Figure 10 B) T cell-mediated cytotoxicity, measured as shown in Example 5. Representative results from two independent experiments performed in duplicate are shown.

[0034] Figure 11 : Activation of rhesus monkey T cells by IgG1-huCD3 antibody variants. Expression of CD69 on T cells from rhesus monkey origin in PBMC cultures was measured by FACS analysis as described in Example 6.

[0035] Figure 12 : Activation of T cells by variants of non-activated huCLB-T3 / 4 antibodies. IgG1-huCLB-T3 / 4 variants were titrated against PBMCs. Expression of CD69 on T cells in PBMC cultures was measured by FACS analysis as described in Example 7. Representative examples of 3 experiments are shown.

[0036] Figure 13 : Proliferation of T-cells by variants of non-activated huCLB-T3 / 4 antibodies. PBMCs were incubated with the antibody for 3 days, after which proliferation was measured by cell proliferation ELISA as described in Example 8. Representative results from two independent experiments are shown.

[0037] Figure 14 : In vitro T cell-mediated cytotoxicity induced by antibody variants of non-activated CD3 antibodies. T cell-mediated cytotoxicity was induced by antibody variants (N297Q, LFLE, LFLENQ, LFLEDA, DANQ, LFLEDANQPS Figure 14 A-G]) and measured as shown in Example 9. The average of two experiments performed in duplicate is shown.

[0038] Figure 15 : In vitro T cell-mediated cytotoxicity induced by non-activated huCLB-T3 / 4 variants. T cell-mediated cytotoxicity was induced by antibody variants (LFLEDA LAL Figure 15 A-C] and measured as described in Example 9. The average of one experiment performed in duplicate is shown.

[0039] Figure 16 : Evaluation of the binding of C1q to non-activated huCLB-T3 / 4 antibody variants. Binding of C1q to monospecific IgG1 huCLB-T3 / 4( Figure 16 A-C) and bsIgG1-huCLB-T3 / 4xHER2 (Figure B-D) and their non-activated antibody variants was evaluated by ELISA as described in Example 10. The results in the figure represent n = 2 experiments.

[0040] Figure 17 : Pharmacokinetic (PK) analysis of the non-activated huCLB-T3 / 4 antibody variant compared to the wild-type IgG1-huCLB-T3 / 4 antibody, as described in Example 11. Plasma concentration of human IgG1 plotted against time ( Figure 17 A). Plasma clearance was calculated as described in Example 11 ( Figure 17 B). The horizontal dashed line represents the mean clearance rate of the human IgG1 antibody in SCID mice (10 mL / day / kg).

[0041] Figure 18 : Frequency of positive T cell responses in healthy HLA-typed donors. An SI index ≥ 1.9 in both the proliferation and IL-2 secretion assays was considered a positive response. Humanized A33 was used as a clinical benchmark control antibody, which showed a high level of immunogenicity clinically and routinely induced 20 - 30% T cell responses in the EpiScreen assay. The KLH response was included to check PBMC quality (after thawing).

[0042] Figure 19 : Sequence alignment of the heavy chain (VH) and light chain (VL) variable regions of the humanized CD3 antibody of the present invention.

[0043] Figure 20 : Expression level of the IgG1-huCD3-H1L1 variant in Expi293F supernatant, as measured by Octet RED using an anti-human IgG immobilized sensor.

[0044] Figure 21 : Binding of the IgG1-huCD3-H1L1-LFLEDA mutant to Jurkat cells.

[0045] Figure 22 : Binding of the CD3 bispecific antibody variant to human T cells. Circles represent the binding of the huCD3-H1L1-LFLEDA variant, triangles represent the binding of the huCD3-H1L1-LFLEDA-LK55N variant, and squares represent the binding of the CD3-hmAb286-LFLEDA variant.

[0046] Figure 23 : Cytotoxicity of the huCD3 x HER2 antibody affinity variant against A431 cells (low HER2-expressing cells) treated with CD3 and HER2 affinity variants ( Figure 23 A) and AU565 cells (high HER2-expressing cells) ( Figure 23 B).

[0047] Figure 24: Tumor size reduction was obtained by huCD3-H1L1xHER2-LFLEDA in NOD-SCID mice at 7, 14, 21, and 28 days after tumor inoculation ( Figure 24 A) and at day 29 ( Figure 24 B).

[0048] Figure 25 : Tumor size reduction was obtained by huCD3-H1L1xCD20-LFLEDA in NOD-SCID mice at 7, 14, and 21 days after tumor inoculation ( Figure 25 A) and at day 21 ( Figure 25 B). DETAILED DESCRIPTION OF THE INVENTION In one aspect, the present invention relates to a humanized or chimeric antibody that binds to human CD3, wherein the antibody comprises a binding region that comprises heavy chain variable (VH) region CDR1, CDR2, and CDR3 having the sequences shown in SEQ ID NO: 1, 2, and 3, respectively, and light chain variable (VL) region CDR1, CDR2, and CDR3 having the sequences shown in SEQ ID NO: 4, the sequence GTN, and SEQ ID NO: 5 or SEQ ID NO: 60, respectively.

[0050] In one embodiment, the present invention relates to a humanized or chimeric antibody that binds to human CD3, wherein the antibody comprises a binding region that comprises heavy chain variable (VH) region CDR1, CDR2, and CDR3 having the sequences shown in SEQ ID NO: 1, 2, and 3, respectively, and light chain variable (VL) region CDR1, CDR2, and CDR3 having the sequences shown in SEQ ID NO: 4, the sequence GTN, and SEQ ID NO: 5, respectively.

[0051] As used herein, the term "antibody" refers to an immunoglobulin molecule, a fragment of an immunoglobulin molecule, or a derivative of either, which has the ability to specifically bind an antigen under typical physiological conditions and has a half-life of a significant period of time, such as at least about 30 minutes, at least about 45 minutes, at least about 1 hour, at least about 2 hours, at least about 4 hours, at least about 8 hours, at least about 12 hours, about 24 hours or more, about 48 hours or more; about 3, 4, 5, 6, 7 or more days, etc., or any other relevant functionally defined time (e.g., a time sufficient to induce, promote, enhance, and / or regulate a physiological response related to an antibody that binds an antigen and / or a time when the antibody is sufficient to recruit effector activity). The binding region that interacts with the antigen (or the binding domain, which may also be used herein and the two terms have the same meaning) comprises the variable regions of both the heavy and light chains of the immunoglobulin molecule. The constant region of an antibody (Ab) can mediate the binding of the immunoglobulin to host tissues or factors, which include various cells of the immune system (e.g., effector cells and T cells) and components of the complement system (e.g., C1q, the first complement of the classical pathway of complement activation). As described above, the term antibody as used herein, unless otherwise specified or clearly contradicted by the context, includes fragments of an antibody that retain the ability to specifically interact (e.g., bind) with an antigen. It has been shown that the antigen-binding function of an antibody can be achieved by fragments of a full-length antibody. Examples of binding fragments included within the term "antibody" include (i) Fab’ or Fab fragments, a monovalent fragment consisting of V L 、V H 、C L and C H 1 domains, or a monovalent antibody described in WO2007059782 (Genmab A / S); (ii) F(ab')2 fragments, a bivalent fragment comprising two Fab fragments linked by a disulfide bridge in the hinge region; (iii) Fd fragments consisting essentially of V H and C H 1 domains; and (iv) Fv fragments consisting essentially of the V L and V H domains of a single arm of an antibody. In addition, although the two domains V L and V H of an Fv fragment are encoded by separate genes, they can be joined using recombinant methods by a synthetic linker that enables them to be made into a single protein chain, where V LPairing with the VH region forms a monovalent molecule (referred to as a single-chain antibody or single-chain Fv (scFv), see, for example, Bird et al., Science 242, 423-426 (1988) and Huston et al., PNAS USA 85, 5879-5883 (1988)). Such single-chain antibodies are included within the term antibody unless otherwise indicated or the context clearly dictates. Although such fragments are generally included within the meaning of antibody, they are unique features of the present invention, both jointly and independently, showing different biological properties and functions. These and other useful antibody fragments in the context of the present invention are further discussed herein. It should also be understood that the term antibody, unless otherwise indicated, also includes polyclonal antibodies, monoclonal antibodies (mAbs), chimeric antibodies, and humanized antibodies, and antibody fragments (antigen-binding fragments) that retain the ability to specifically bind an antigen provided by any known technique, such as enzymatic cleavage, peptide synthesis, and recombinant techniques. The antibodies produced can have any isotype.

[0052] As used herein, the terms "immunoglobulin heavy chain", "heavy chain of an immunoglobulin", or "heavy chain" mean one of the chains of an immunoglobulin. A heavy chain typically consists of a heavy chain variable region (referred to herein simply as VH) and a heavy chain constant region (referred to herein simply as CH) that defines the isotype of the immunoglobulin. The heavy chain constant region typically consists of three domains, CH1, CH2, and CH3. The heavy chain constant region may further include a hinge region. As used herein, the term "immunoglobulin" means a class of structurally related glycoproteins that consists of two pairs of polypeptide chains, a pair of low molecular weight light (L) chains and a pair of heavy (H) chains, and all four chains may be interconnected by disulfide bonds. The structure of immunoglobulins has been well characterized (see, for example,

[14] ). Within the structure of an immunoglobulin, the two heavy chains are interconnected by disulfide bonds in a so-called "hinge region". Similar to the heavy chain, each light chain typically consists of several regions: a light chain variable region (referred to herein simply as VL) and a light chain constant region (referred to herein simply as CL). The light chain constant region typically contains one domain, CL. In addition, the VH and VL regions can be further subdivided into regions with hypervariability (or hypervariable regions, which can be highly variable in the sequence and / or form of structurally defined loops), also known as complementarity-determining regions (CDRs), which alternate with more conserved regions called framework regions (FRs). Each VH and VL typically consists of three CDRs and four FRs, arranged in the following order from the amino terminus to the carboxyl terminus: FR1, CDR1, FR2, CDR2, FR3, CDR3, FR4 (see

[15] ). The CDR sequences can be determined by using the methods provided by IMGT

[16] -

[17] .

[0053] As used herein, the term "isotype" refers to an immunoglobulin class (e.g., IgG1, IgG2, IgG3, IgG4, IgD, IgA, IgE, or IgM) or any allotype thereof, such as IgG1m(za) and IgG1m(f) [SEQ ID NO:15] encoded by the heavy chain constant region gene. Thus, in one embodiment, the antibody comprises a heavy chain of an immunoglobulin of the IgG1 type or any of its allotypes. In addition, each heavy chain isotype can be combined with a kappa (κ) or lambda (λ) light chain.

[0054] As used herein, the term "chimeric antibody" refers to an antibody in which the variable region is derived from a non-human species (e.g., derived from a rodent) and the constant region is derived from a different species, such as a human. Chimeric antibodies can be generated by antibody engineering. "Antibody engineering" is a term widely used for different types of antibody modification, and it is a method well known to those skilled in the art. Specifically, chimeric antibodies can be generated by using standard DNA techniques described in

[18] . Thus, chimeric antibodies can be recombinant antibodies that have been genetically or enzymatically engineered. Generating chimeric antibodies is within the knowledge of those skilled in the art, and thus generating the chimeric antibodies of the present invention can be achieved by methods other than those described herein. Chimeric monoclonal antibodies have been developed for therapeutic applications to reduce antibody immunogenicity. They typically can comprise non-human (e.g., murine) variable regions that are specific for the antigen of interest, and human constant heavy and light chain domains. As used in the context of chimeric antibodies, the term "variable region" or "variable domain" refers to the region that contains the CDRs and framework regions of both the heavy and light chains of an immunoglobulin.

[0055] As used herein, the term "humanized antibody" refers to a genetically engineered non-human antibody that comprises human antibody constant domains and a non-human variable domain that has been modified to contain a high level of sequence homology with a human variable domain. This can be achieved by transplanting the six non-human antibody complementarity determining regions (CDRs) that together form the antigen-binding site onto a homologous human receptor framework region (FR) (see

[19] -

[20] ). To fully reconstruct the binding affinity and specificity of the parental antibody, it may be necessary to substitute framework residues of the parental antibody (i.e., the non-human antibody) into the human framework region (backmutation). Structural homology modeling can assist in identifying amino acid residues in the framework region that are important for the binding properties of the antibody. Thus, a humanized antibody can comprise non-human CDR sequences, a substantially human framework region optionally containing one or more amino acid backmutations that are mutations to non-human amino acid sequences, and a complete human constant region. Optionally, additional amino acid modifications that are not necessarily backmutations can be applied to obtain a humanized antibody with preferred characteristics (e.g., affinity and biochemical properties).

[0056] A humanized or chimeric antibody according to any aspect or embodiment of the present invention may be referred to as a "humanized or chimeric CD3 antibody", "the humanized or chimeric antibody of the present invention", "CD3 antibody", or "the CD3 antibody of the present invention", all of which have the same meaning and purpose, unless the context is otherwise contradictory.

[0057] The amino acid sequence of an antibody of non-human origin is different from that of an antibody of human origin, and thus a non-human antibody has the potential to be immunogenic when administered to a human patient. However, despite the non-human origin of the antibody, its CDR regions are responsible for the ability of the antibody to bind its target antigen, and the goal of humanization is to maintain the specificity and binding affinity of the antibody. Therefore, humanization of non-human therapeutic antibodies is performed to minimize their immunogenicity in humans while maintaining the specificity and binding affinity of the non-human origin antibody.

[0058] As used herein, the term "binding region" refers to the region of an antibody that is capable of binding any molecule such as a polypeptide (e.g., present on a cell, bacterium, or virion).

[0059] As used herein, the term "bind" refers to the binding of an antibody to a predetermined antigen or target, which when measured on a BIAcore 3000 instrument using, for example, surface plasmon resonance (SPR) technology with the antigen as the ligand and the antibody as the analyte, typically has a K of about 10 -6 M or less, such as 10 -7 M or less, such as about 10 -8 M or less, such as about 10 -9 M or less, about 10 -10 M or less or about 10 -11 M or even less for the affinity, and binds to the predetermined antigen with an affinity corresponding to at least 10-fold lower, such as at least 100-fold lower, such as at least 1,000-fold lower, such as at least 10,000-fold lower, such as at least 100,000-fold lower than its affinity for non-specific antigens (e.g., BSA, casein) that do not bind the predetermined antigen or closely related antigens. The degree of lower affinity depends on the K D of the antibody, such that when the K D of the antibody is extremely low (i.e., the antibody is highly specific), the degree to which the affinity for the antigen is lower than the affinity for non-specific antigens can be at least 10,000-fold. As used herein, the term "K D " (M) refers to the dissociation equilibrium constant of a specific antibody-antigen interaction. D is extremely low (i.e., the antibody is highly specific), the degree to which the affinity for the antigen is lower than the affinity for non-specific antigens can be at least 10,000-fold. As used herein, the term "K D " (M) refers to the dissociation equilibrium constant of a specific antibody-antigen interaction.

[0060] As used herein, the term "human CD3" refers to the human cluster of differentiation 3 protein, which is part of a T cell co-receptor protein complex and consists of four different chains. CD3 is also present in other species, and thus the term "CD3" can be used herein and is not limited to human CD3, unless the context dictates otherwise. In mammals, the complex contains one CD3γ (gamma) chain (human CD3γ chain Swissprot P09693 or cynomolgus monkey CD3γ Swissprot Q95LI7), one CD3δ (delta) chain (human CD3δ Swissprot P04234 or cynomolgus monkey CD3δ Swissprot Q95LI8), two CD3ε (epsilon) chains (human CD3ε Swissprot P07766; or cynomolgus monkey CD3ε Swissprot Q95LI5), rhesus monkey CD3ε (Swissprot G7NCB9), and one CD3ζ-chain (zeta) chain (human CD3ζ Swissprot P20963, cynomolgus monkey CD3ζ Swissprot Q09TK0). These chains associate with a molecule called the T-cell receptor (TCR) and generate activation signals in T lymphocytes. The TCR and CD3 molecules together form the TCR complex.

[0061] It is within the knowledge of those skilled in the art that the amino acid sequences referred to by the Swissprot numbers include a signal peptide, which is removed after protein translation. Thus, the proteins present on the cell surface, such as CD3, do not include the signal peptide. Specifically, the amino acid sequences listed in Table 1 do not contain such a signal peptide. Such proteins listed in Table 1 may be referred to as "mature proteins". Thus, SEQ ID NO:14 shows the amino acid sequence of mature human CD3δ (delta), SEQ ID NO:13 shows the amino acid sequence of mature human CD3ε (epsilon), SEQ ID NO:21 shows the amino acid sequence of mature cynomolgus monkey CD3ε, and SEQ ID NO:22 shows the amino acid sequence of mature rhesus monkey CD3ε. Thus, as used herein, the term "mature" refers to a protein that does not contain any signal or leader sequence.

[0062] It is well known that signal peptide sequence homology, length, and cleavage site location vary significantly between different proteins. Signal peptides can be determined by different methods, such as SEQ ID NO:13 of the present invention has been determined according to the SignalP application (available from http: / / www.cbs.dtu.dk / services / SignalP / ).

[0063] In a specific embodiment, the humanized or chimeric antibody of the invention binds to the ε chain of CD3, such as the ε chain of human CD3 (SEQ ID NO: 13). In yet another specific embodiment, the humanized or chimeric antibody binds to an epitope within amino acids 1-27 of the N-terminal portion of human CD3ε (epsilon) (SEQ ID NO: 13). In such a specific embodiment, the antibody may even further cross-react with other non-human primate species such as cynomolgus monkeys (cynomolgus monkey CD3ε SEQ ID NO: 21) and / or rhesus monkeys (rhesus monkey CD3ε SEQ ID NO: 22).

[0064] As used herein, the term "cross-react" refers to the ability of an antibody, such as the humanized or chimeric antibody of the invention, to bind to its target on different species. Specifically, the humanized CD3 antibodies exemplified in the examples described herein have the ability to bind to human (Example 2), cynomolgus monkey (Example 2), and rhesus monkey CD3.

[0065] Compared to the original antibody, the antibody of the invention comprising the CDR sequences defined herein and further comprising framework regions may differ in sequences outside the CDR sequences but still retain full binding ability. Accordingly, the invention also relates to antibodies comprising amino acid sequences of variable regions having a certain sequence identity with any of the sequences described herein.

[0066] As used in the context of the present invention, the term "sequence identity" refers to the percentage of identity between two sequences as a function of the number of identical positions shared by the sequences (i.e., % homology = number of identical positions / total number of positions x 100), taking into account the number of gaps and the length of each gap, which are introduced for the best alignment of the two sequences. The percentage of identity between two nucleotide or amino acid sequences can be determined, for example, using the algorithm of E. Meyers and W. Miller

[21] . In addition, the percentage of identity between two amino acid sequences can be determined using the Needleman and Wunsch algorithm

[22] . Multiple alignments are preferably performed using the ClustalW algorithm

[23] (e.g., as used in VectorNTI software version 11.5; Invitrogen Inc.).

[0067] Accordingly, in one embodiment, the VH region has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity with at least one of the amino acid sequences shown in the VH sequences selected from: a) the VH sequence shown in SEQ ID NO: 6; b) the VH sequence shown in SEQ ID NO: 8; c) the VH sequence shown in SEQ ID NO: 7; and d) The VH sequence shown in SEQ ID NO:9.

[0068] In a specific embodiment, the VH region has at least 96% amino acid sequence identity with at least one amino acid sequence shown in the VH sequences selected from the following: a) The VH sequence shown in SEQ ID NO:6; b) The VH sequence shown in SEQ ID NO:8; c) The VH sequence shown in SEQ ID NO:7; and d) The VH sequence shown in SEQ ID NO:9.

[0069] In one embodiment, the VL region has at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with at least one amino acid sequence shown in the VL sequences selected from the following: a) The VL sequence shown in SEQ ID NO:10; b) The VL sequence shown in SEQ ID NO:11; and c) The VL sequence shown in SEQ ID NO:12.

[0070] In a specific embodiment, the VL region has at least 95% amino acid sequence identity with at least one amino acid sequence shown in the VL sequences selected from the following: a) The VL sequence shown in SEQ ID NO:10; b) The VL sequence shown in SEQ ID NO:11; and c) The VL sequence shown in SEQ ID NO:12.

[0071] In one embodiment, the VH region is selected from: a) The VH sequence shown in SEQ ID NO:6; b) The VH sequence shown in SEQ ID NO:8; c) The VH sequence shown in SEQ ID NO:7; and d) The VH sequence shown in SEQ ID NO:9.

[0072] In one embodiment, the VL region is selected from: a) The VL sequence shown in SEQ ID NO:10; b) The VL sequence shown in SEQ ID NO:11; and c) The VL sequence shown in SEQ ID NO:12.

[0073] In one embodiment, only one of the VH or VL sequences is 100% identical to one of the sequences disclosed herein, while the other may have a sequence identity of at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity to one of the sequences disclosed herein.

[0074] In a specific embodiment, the VH sequence has at least 97% amino acid sequence identity to at least one amino acid sequence shown in VH sequences selected from: a) the VH sequence shown in SEQ ID NO:6; b) the VH sequence shown in SEQ ID NO:7; c) the VH sequence shown in SEQ ID NO:8; and d) the VH sequence shown in SEQ ID NO:9; and the VL sequence has at least 95% amino acid sequence identity to at least one amino acid sequence shown in VL sequences selected from: i. the VL sequence shown in SEQ ID NO:10; ii. the VL sequence shown in SEQ ID NO:11; and iii. the VL sequence shown in SEQ ID NO:12.

[0075] In one embodiment, the VH and VL sequences are selected from: a) VH and VL sequences having at least 90% identity to the sequences shown in SEQ ID NO:6 and 10, respectively; SEQ ID NO:7 and 10, respectively; SEQ ID NO:8 and 10, respectively; SEQ ID NO:9 and 10, respectively; SEQ ID NO:6 and 11, respectively; SEQ ID NO:7 and 11, respectively; SEQ ID NO:8 and 11, respectively; SEQ ID NO:9 and 11, respectively; SEQ ID NO:6 and 12, respectively; SEQ ID NO:7 and 12, respectively; SEQ ID NO:8 and 12, respectively; and SEQ ID NO:9 and 12, respectively; b) VH and VL sequences having at least 95% identity to the sequences shown in SEQ ID NO:6 and 10, respectively; SEQ ID NO:7 and 10, respectively; SEQ ID NO:8 and 10, respectively; SEQ ID NO:9 and 10, respectively; SEQ ID NO:6 and 11, respectively; SEQ ID NO:7 and 11, respectively; SEQ ID NO:8 and 11, respectively; SEQ ID NO:9 and 11, respectively; SEQ ID NO:6 and 12, respectively; SEQ ID NO:7 and 12, respectively; SEQ ID NO:8 and 12, respectively; and SEQ ID NO:9 and 12, respectively; c) VH and VL sequences having at least 97% identity to the sequences shown in SEQ ID NO:6 and 10, respectively; SEQ ID NO:7 and 10, respectively; SEQ ID NO:8 and 10, respectively; SEQ ID NO:9 and 10, respectively; SEQ ID NO:6 and 11, respectively; SEQ ID NO:7 and 11, respectively; SEQ ID NO:8 and 11, respectively; SEQ ID NO:9 and 11, respectively; SEQ ID NO:6 and 12, respectively; SEQ ID NO:7 and 12, respectively; SEQ ID NO:8 and 12, respectively; and SEQ ID NO:9 and 12, respectively; d) VH and VL sequences having at least 99% identity with the sequences shown by SEQ ID NO:6 and 10 respectively; 7 and 10 respectively; 8 and 10 respectively; 9 and 10 respectively; 6 and 11 respectively; 7 and 11 respectively; 8 and 11 respectively; 9 and 11 respectively; 6 and 12 respectively; 7 and 12 respectively; 8 and 12 respectively; and 9 and 12 respectively; e) VH and VL sequences having at least 100% identity with the sequences shown by SEQ ID NO:6 and 10 respectively; 7 and 10 respectively; 8 and 10 respectively; 9 and 10 respectively; 6 and 11 respectively; 7 and 11 respectively; 8 and 11 respectively; 9 and 11 respectively; 6 and 12 respectively; 7 and 12 respectively; 8 and 12 respectively; and 9 and 12 respectively; f) A VH sequence having at least 90% identity with the sequence shown by SEQ ID NO:6 and a VL sequence having at least 95% identity with the sequence shown by SEQ ID NO:10, 11 or 12; g) A VH sequence having at least 90% identity with the sequence shown by SEQ ID NO:6 and a VL sequence having at least 97% identity with the sequence shown by SEQ ID NO:10, 11 or 12; h) A VH sequence having at least 90% identity with the sequence shown by SEQ ID NO:6 and a VL sequence having at least 99% identity with the sequence shown by SEQ ID NO:10, 11 or 12; i) A VH sequence having at least 90% identity with the sequence shown by SEQ ID NO:6 and a VL sequence having at least 100% identity with the sequence shown by SEQ ID NO:10, 11 or 12; j) A VH sequence having at least 95% identity with the sequence shown by SEQ ID NO:6 and a VL sequence having at least 90% identity with the sequence shown by SEQ ID NO:10, 11 or 12; k) A VH sequence having at least 95% identity with the sequence shown by SEQ ID NO:6 and a VL sequence having at least 97% identity with the sequence shown by SEQ ID NO:10, 11 or 12; l) A VH sequence having at least 95% identity with the sequence shown by SEQ ID NO:6 and a VL sequence having at least 99% identity with the sequence shown by SEQ ID NO:10, 11 or 12; m) A VH sequence having at least 95% identity with the sequence shown by SEQ ID NO:6 and a VL sequence having at least 100% identity with the sequence shown by SEQ ID NO:10, 11 or 12; n) A VH sequence having at least 97% identity to the sequence set forth in SEQ ID NO:6 and a VL sequence having at least 90% identity to the sequence set forth in SEQ ID NO:10, 11 or 12; o) A VH sequence having at least 97% identity to the sequence set forth in SEQ ID NO:6 and a VL sequence having at least 95% identity to the sequence set forth in SEQ ID NO:10, 11 or 12; p) A VH sequence having at least 97% identity to the sequence set forth in SEQ ID NO:6 and a VL sequence having at least 99% identity to the sequence set forth in SEQ ID NO:10, 11 or 12; q) A VH sequence having at least 97% identity to the sequence set forth in SEQ ID NO:6 and a VL sequence having at least 100% identity to the sequence set forth in SEQ ID NO:10, 11 or 12; r) A VH sequence having at least 99% identity to the sequence set forth in SEQ ID NO:6 and a VL sequence having at least 90% identity to the sequence set forth in SEQ ID NO:10, 11 or 12; s) A VH sequence having at least 99% identity to the sequence set forth in SEQ ID NO:6 and a VL sequence having at least 95% identity to the sequence set forth in SEQ ID NO:10, 11 or 12; t) A VH sequence having at least 99% identity to the sequence set forth in SEQ ID NO:6 and a VL sequence having at least 97% identity to the sequence set forth in SEQ ID NO:10, 11 or 12; u) A VH sequence having at least 99% identity to the sequence set forth in SEQ ID NO:6 and a VL sequence having at least 100% identity to the sequence set forth in SEQ ID NO:10, 11 or 12; v) A VH sequence having at least 100% identity to the sequence set forth in SEQ ID NO:6 and a VL sequence having at least 90% identity to the sequence set forth in SEQ ID NO:10, 11 or 12; x) A VH sequence having at least 100% identity to the sequence set forth in SEQ ID NO:6 and a VL sequence having at least 95% identity to the sequence set forth in SEQ ID NO:10, 11 or 12; y) A VH sequence having at least 100% identity to the sequence set forth in SEQ ID NO:6 and a VL sequence having at least 97% identity to the sequence set forth in SEQ ID NO:10, 11 or 12; z) A VH sequence having at least 100% identity with the sequence shown in SEQ ID NO: 6 and a VL sequence having at least 99% identity with the sequence shown in SEQ ID NO: 10, 11 or 12; aa) A VH sequence having at least 90% identity with the sequence shown in SEQ ID NO: 7 and a VL sequence having at least 95% identity with the sequence shown in SEQ ID NO: 10, 11 or 12; ab) A VH sequence having at least 90% identity with the sequence shown in SEQ ID NO: 7 and a VL sequence having at least 97% identity with the sequence shown in SEQ ID NO: 10, 11 or 12; ac) A VH sequence having at least 90% identity with the sequence shown in SEQ ID NO: 7 and a VL sequence having at least 99% identity with the sequence shown in SEQ ID NO: 10, 11 or 12; ad) A VH sequence having at least 90% identity with the sequence shown in SEQ ID NO: 7 and a VL sequence having at least 100% identity with the sequence shown in SEQ ID NO: 10, 11 or 12; ae) A VH sequence having at least 95% identity with the sequence shown in SEQ ID NO: 7 and a VL sequence having at least 90% identity with the sequence shown in SEQ ID NO: 10, 11 or 12; af) A VH sequence having at least 95% identity with the sequence shown in SEQ ID NO: 7 and a VL sequence having at least 97% identity with the sequence shown in SEQ ID NO: 10, 11 or 12; ag) A VH sequence having at least 95% identity with the sequence shown in SEQ ID NO: 7 and a VL sequence having at least 99% identity with the sequence shown in SEQ ID NO: 10, 11 or 12; ah) A VH sequence having at least 95% identity with the sequence shown in SEQ ID NO: 7 and a VL sequence having at least 100% identity with the sequence shown in SEQ ID NO: 10, 11 or 12; ai) A VH sequence having at least 97% identity with the sequence shown in SEQ ID NO: 7 and a VL sequence having at least 90% identity with the sequence shown in SEQ ID NO: 10, 11 or 12; aj) A VH sequence having at least 97% identity with the sequence shown in SEQ ID NO: 7 and a VL sequence having at least 95% identity with the sequence shown in SEQ ID NO: 10, 11 or 12; ak) a VH sequence having at least 97% identity to the sequence shown in SEQ ID NO:7 and a VL sequence having at least 99% identity to the sequence shown in SEQ ID NO:10, 11 or 12; al) a VH sequence having at least 97% identity to the sequence shown in SEQ ID NO:7 and a VL sequence having at least 100% identity to the sequence shown in SEQ ID NO:10, 11 or 12; am) a VH sequence having at least 99% identity to the sequence shown in SEQ ID NO:7 and a VL sequence having at least 90% identity to the sequence shown in SEQ ID NO:10, 11 or 12; an) a VH sequence having at least 99% identity to the sequence shown in SEQ ID NO:7 and a VL sequence having at least 95% identity to the sequence shown in SEQ ID NO:10, 11 or 12; ao) a VH sequence having at least 99% identity to the sequence shown in SEQ ID NO:7 and a VL sequence having at least 97% identity to the sequence shown in SEQ ID NO:10, 11 or 12; ap) a VH sequence having at least 99% identity to the sequence shown in SEQ ID NO:7 and a VL sequence having at least 100% identity to the sequence shown in SEQ ID NO:10, 11 or 12; aq) a VH sequence having at least 100% identity to the sequence shown in SEQ ID NO:7 and a VL sequence having at least 90% identity to the sequence shown in SEQ ID NO:10, 11 or 12; ar) a VH sequence having at least 100% identity to the sequence shown in SEQ ID NO:7 and a VL sequence having at least 95% identity to the sequence shown in SEQ ID NO:10, 11 or 12; as) a VH sequence having at least 100% identity to the sequence shown in SEQ ID NO:7 and a VL sequence having at least 97% identity to the sequence shown in SEQ ID NO:10, 11 or 12; at) a VH sequence having at least 100% identity to the sequence shown in SEQ ID NO:7 and a VL sequence having at least 99% identity to the sequence shown in SEQ ID NO:10, 11 or 12; ba) a VH sequence having at least 90% identity to the sequence shown in SEQ ID NO:8 and a VL sequence having at least 95% identity to the sequence shown in SEQ ID NO:10, 11 or 12; bb) a VH sequence having at least 90% identity to the sequence shown in SEQ ID NO:8 and a VL sequence having at least 97% identity to the sequence shown in SEQ ID NO:10, 11 or 12; bc) a VH sequence having at least 90% identity to the sequence shown in SEQ ID NO:8 and a VL sequence having at least 99% identity to the sequence shown in SEQ ID NO:10, 11 or 12; bd) a VH sequence having at least 90% identity to the sequence shown in SEQ ID NO:8 and a VL sequence having at least 100% identity to the sequence shown in SEQ ID NO:10, 11 or 12; be) a VH sequence having at least 95% identity to the sequence shown in SEQ ID NO:8 and a VL sequence having at least 90% identity to the sequence shown in SEQ ID NO:10, 11 or 12; bf) a VH sequence having at least 95% identity to the sequence shown in SEQ ID NO:8 and a VL sequence having at least 97% identity to the sequence shown in SEQ ID NO:10, 11 or 12; bg) a VH sequence having at least 95% identity to the sequence shown in SEQ ID NO:8 and a VL sequence having at least 99% identity to the sequence shown in SEQ ID NO:10, 11 or 12; bh) a VH sequence having at least 95% identity to the sequence shown in SEQ ID NO:8 and a VL sequence having at least 100% identity to the sequence shown in SEQ ID NO:10, 11 or 12; bi) a VH sequence having at least 97% identity to the sequence shown in SEQ ID NO:8 and a VL sequence having at least 90% identity to the sequence shown in SEQ ID NO:10, 11 or 12; bj) a VH sequence having at least 97% identity to the sequence shown in SEQ ID NO:8 and a VL sequence having at least 95% identity to the sequence shown in SEQ ID NO:10, 11 or 12; bk) a VH sequence having at least 97% identity to the sequence shown in SEQ ID NO:8 and a VL sequence having at least 99% identity to the sequence shown in SEQ ID NO:10, 11 or 12; bl) a VH sequence having at least 97% identity to the sequence shown in SEQ ID NO:8 and a VL sequence having at least 100% identity to the sequence shown in SEQ ID NO:10, 11 or 12; a VH sequence having at least 99% identity to the sequence set forth in SEQ ID NO:8 and a VL sequence having at least 90% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; bn) a VH sequence having at least 99% identity to the sequence set forth in SEQ ID NO:8 and a VL sequence having at least 95% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; bo) a VH sequence having at least 99% identity to the sequence set forth in SEQ ID NO:8 and a VL sequence having at least 97% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; bp) a VH sequence having at least 99% identity to the sequence set forth in SEQ ID NO:8 and a VL sequence having at least 100% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; bq) a VH sequence having at least 100% identity to the sequence set forth in SEQ ID NO:8 and a VL sequence having at least 90% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; br) a VH sequence having at least 100% identity to the sequence set forth in SEQ ID NO:8 and a VL sequence having at least 95% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; bs) a VH sequence having at least 100% identity to the sequence set forth in SEQ ID NO:8 and a VL sequence having at least 97% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; bt) a VH sequence having at least 100% identity to the sequence set forth in SEQ ID NO:8 and a VL sequence having at least 99% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; ca) a VH sequence having at least 90% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 95% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; cb) a VH sequence having at least 90% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 97% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; cc) a VH sequence having at least 90% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 99% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; cd) a VH sequence having at least 90% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 100% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; ce) a VH sequence having at least 95% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 90% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; cf) a VH sequence having at least 95% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 97% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; cg) a VH sequence having at least 95% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 99% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; ch) a VH sequence having at least 95% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 100% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; ci) a VH sequence having at least 97% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 90% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; cj) a VH sequence having at least 97% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 95% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; ck) a VH sequence having at least 97% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 99% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; cl) a VH sequence having at least 97% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 100% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; cm) a VH sequence having at least 99% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 90% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; cn) a VH sequence having at least 99% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 95% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; a VH sequence having at least 99% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 97% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; cp) a VH sequence having at least 99% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 100% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; cq) a VH sequence having at least 100% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 90% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; cr) a VH sequence having at least 100% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 95% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; cs) a VH sequence having at least 100% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 97% identity to the sequence set forth in SEQ ID NO:10, 11, or 12; and ct) a VH sequence having at least 100% identity to the sequence set forth in SEQ ID NO:9 and a VL sequence having at least 99% identity to the sequence set forth in SEQ ID NO:10, 11, or 12.

[0076] In one embodiment, the binding region comprises a VH and a VL selected from: a) the VH sequence set forth in SEQ ID NO:6, and the VL sequence set forth in SEQ ID NO:10; b) the VH sequence set forth in SEQ ID NO:8, and the VL set forth in SEQ ID NO:10; c) the VH sequence set forth in SEQ ID NO:9, and the VL sequence set forth in SEQ ID NO:10; d) the VH sequence set forth in SEQ ID NO:6, and the VL sequence set forth in SEQ ID NO:11; e) the VH sequence set forth in SEQ ID NO:6, and the VL sequence set forth in SEQ ID NO:12; f) the VH sequence set forth in SEQ ID NO:7, and the VL sequence set forth in SEQ ID NO:10; g) the VH sequence set forth in SEQ ID NO:7, and the VL sequence set forth in SEQ ID NO:11; h) The VH sequence shown in SEQ ID NO:7, and the VL sequence shown in SEQ ID NO:12; i) The VH sequence shown in SEQ ID NO:8, and the VL sequence shown in SEQ ID NO:11; j) The VH sequence shown in SEQ ID NO:8, and the VL sequence shown in SEQ ID NO:12; k) The VH sequence shown in SEQ ID NO:9, and the VL sequence shown in SEQ ID NO:11; and l) The VH sequence shown in SEQ ID NO:9, and the VL sequence shown in SEQ ID NO:12.

[0077] In a specific embodiment, the binding region comprises a VH sequence and a VL sequence selected from the following: a) The VH sequence shown in SEQ ID NO:6, and the VL sequence shown in SEQ ID NO:10; b) The VH sequence shown in SEQ ID NO:8, and the VL sequence shown in SEQ ID NO:10; and c) The VH sequence shown in SEQ ID NO:9, and the VL sequence shown in SEQ ID NO:10.

[0078] The humanized antibodies of the present invention can be generated by comparing the amino acid sequences of the heavy and light chain variable regions with a database of human germline variable region sequences to identify heavy and light chain human sequences with a suitable degree of homology to be used as human variable framework regions. A series of humanized heavy and light chain variable regions can be designed by transplanting, for example, murine CDRs onto the framework regions (identified as described above) and, if necessary, by back-mutating (mutating one or more human amino acid residues at specified positions in the framework region back to non-human amino acids) to identified specific murine residue sequences that may be critical for restoring antibody binding efficacy. Then, using computer technologies such as iTope TM and TCED TM (

[24] ,

[25] and

[26] ), variant sequences with the lowest possible incidence of T cell epitopes are selected.

[0079] In addition, the humanized antibodies of the present invention can also be "deimmunized". Deimmunization may be necessary because the presence of human T cell epitopes in protein sequences such as the humanized antibodies of the present invention can increase the immunogenic risk profile when they have the potential to activate helper T cells. Such activation of helper T cells can be avoided by deimmunization. Deimmunization can be carried out by introducing mutations into the amino acid sequence of the humanized antibody to remove the T cell epitopes without significantly reducing the binding affinity of the antibody.

[0080] Thus, in one embodiment of the present invention, a humanized antibody can be produced by a method comprising the steps of: (i) comparing a non-human full variable heavy chain sequence and / or a full variable light chain sequence with a database of human germline sequences, (ii) selecting the human germline sequence having the highest homology with the non-human sequence to obtain a humanized sequence, (iii) optimizing the humanized sequence by back mutation if desired, and (iv) expressing the sequence in a suitable expression system.

[0081] Thus, a full-length antibody of the present invention can be produced by a method comprising the steps of: (i) comparing a non-human variable heavy chain sequence and a variable light chain sequence with a database of human germline sequences, (ii) selecting the human germline sequence having the highest homology with the non-human sequence, (iii) transplanting the non-human CDRs to the selected human germline to obtain a humanized sequence, (iv) optimizing the humanized sequence by back mutation if desired, (v) identifying the constant heavy chain and light chain sequences, and (vi) expressing the full heavy chain sequence and the full light chain sequence in a suitable expression system. The full-length antibody of the present invention can thus be produced as described in Example 1. It is within the knowledge of those skilled in the art to produce a full-length antibody starting from the CDR sequences or the full variable region sequences. Thus, those skilled in the art will know how to produce the full-length antibody of the present invention.

[0082] As used herein, the term "full heavy chain sequence" refers to a sequence consisting of a variable heavy chain and a constant heavy chain sequence.

[0083] As used herein, the term "full light chain sequence" refers to a sequence consisting of a variable light chain and a constant light chain sequence.

[0084] Back mutations can be introduced by standard DNA mutagenesis. Such standard techniques for DNA mutagenesis are described in

[18] . Alternatively, commercially available kits such as the Quickchange TM Site-Directed Mutagenesis Kit (Stratagene) can be used, or the required back mutations can be introduced by de novo DNA synthesis.

[0085] Thus, in one embodiment, the antibody is a humanized antibody.

[0086] Chimeric antibodies can be produced by replacing all of the constant region sequences of a non-human (e.g., murine) antibody with human-derived constant region sequences. Thus, the entirely non-human variable region sequences are retained in the chimeric antibody. Accordingly, the chimeric antibodies of the present invention can be produced by a method comprising the steps of expressing a non-human variable heavy chain (SEQ ID NO:27), a non-human variable light chain sequence (SEQ ID NO:28), a human constant heavy chain, and a human constant light chain sequence in a suitable expression system, and thereby producing a full-length chimeric antibody. Alternative methods can be used. Such methods of producing chimeric antibodies are within the knowledge of one of ordinary skill in the art, and thus one of ordinary skill in the art will know how to produce the chimeric antibodies of the present invention.

[0087] Accordingly, in one embodiment, the antibody is a chimeric antibody.

[0088] In one embodiment, the antibody is a full-length antibody. As used herein, the term "full-length antibody" refers to an antibody (e.g., a parental or variant antibody) that comprises all of the heavy and light chain constant and variable domains corresponding to those that are naturally present in a wild-type antibody of that isotype.

[0089] In one embodiment, the antibody comprises an Fc region, which comprises first and second immunoglobulin heavy chains.

[0090] As used herein, the term "Fc region" refers to the region that comprises at least a hinge region, a CH2 region, and a CH3 region in the direction from the N-terminus to the C-terminus. The Fc region may further comprise a CH1 region at the N-terminus of the hinge region.

[0091] As used herein, the term "hinge region" refers to the hinge region of an immunoglobulin heavy chain. Thus, for example, the hinge region of a human IgG1 antibody corresponds to amino acids 216-230 according to the Eu numbering described in Kabat.

[0092] Unless otherwise stated or the context is contradictory, the amino acids of the constant region sequences are numbered herein according to the Eu-numbering index (described in

[27] ) and may be referred to as "according to the Eu numbering described in Kabat", "according to the Eu numbering of Kabat", or "according to the Eu numbering system".

[0093] As used herein, the term "CH1 region" or "CH1 domain" refers to the CH1 region of an immunoglobulin heavy chain. Thus, for example, the CH1 region of a human IgG1 antibody corresponds to amino acids 118-215 according to the Eu numbering system. However, the CH1 region may also be any other subtype as described herein.

[0094] As used herein, the term "CH2 region" or "CH2 domain" refers to the CH2 region of an immunoglobulin heavy chain. Thus, for example, the CH2 region of a human IgG1 antibody corresponds to amino acids 231 - 340 according to the Eu numbering system. However, the CH2 region can also be any other subtype as described herein.

[0095] As used herein, the term "CH3 region" or "CH3 domain" refers to the CH3 region of an immunoglobulin heavy chain. Thus, for example, the CH3 region of a human IgG1 antibody corresponds to amino acids 341 - 447 according to the Eu numbering system. However, the CH3 region can also be any other subtype as described herein.

[0096] In one embodiment, the isotype of the immunoglobulin heavy chain is selected from IgG1, IgG2, IgG3, and IgG4. The immunoglobulin heavy chain can be any allotype within each immunoglobulin type, such as IgG1m(f) (SEQ ID NO:15). Thus, in a specific embodiment, the isotype of the immunoglobulin heavy chain is IgG1 or any of its allotypes, such as IgG1m(f) (SEQ ID NO:15).

[0097] When targeting the antigen CD3, which is part of the T - cell receptor (TCR), the T - cell - specific mechanism of cell killing is desirable. Other effector functions, such as complement activation, may be unwanted, and thus reducing effector function is desirable. C1q binding is the first step in the complement cascade and thus serves as an indicator of the complement - dependent cytotoxicity (CDC) ability of an antibody. If the binding of C1q to the antibody can be avoided, then the activation of the complement cascade can also be avoided.

[0098] Thus, in one embodiment, the antibody comprises an Fc region that has been modified such that the binding of C1q to the antibody is reduced by at least 70%, at least 80%, at least 90%, at least 95%, at least 97%, at least 99%, or 100% compared to the wild - type antibody, wherein C1q binding is determined by ELISA.

[0099] As used herein, the term "modified" refers to an amino acid sequence of an Fc region that is not identical to the amino acid sequence of the wild - type Fc region. That is, an amino acid residue at a specified position in the wild - type Fc region has been substituted, deleted, or inserted to alter, for example, the binding site for C1q, the binding site for other effector molecules, or the binding to the Fc receptor (FcR). Such modification of the amino acid sequence can be prepared by substituting one or more amino acids with conservative amino acids, or can be prepared by substituting one or more amino acids with alternative amino acids that are physically and / or functionally similar to the amino acids present in the wild - type. Substitutions can also be prepared by substituting with non - conservative amino acids.

[0100] In the context of the present invention, amino acids can be described as conservative or non-conservative amino acids and can thus be classified accordingly. Amino acid residues can also be classified into types defined by alternative physical and functional properties. Thus, the types of amino acids can be reflected in one or both of the following tables: Conservative type of amino acid residue Acidic residue D and E Basic residue K, R and H Hydrophilic uncharged residue S, T, N and Q Aliphatic uncharged residue G, A, V, L and I Non-polar uncharged residue C, M and P Aromatic residue F, Y and W Alternative physical and functional classification of amino acid residues In the context of the present invention, substitutions in antibodies such as humanized or chimeric antibodies are represented as follows: Original amino acid – position – substituted amino acid; Referring to the well-recognized amino acid nomenclature, the three-letter code or the single-letter code is used, including the codes Xaa and X to indicate any amino acid residue. Thus, the symbol "L234F" or "Leu234Phe" means that the antibody contains a substitution of leucine by phenylalanine at amino acid position 234.

[0101] Substitution of an amino acid at a given position by any other amino acid is represented as follows: Original amino acid – position; or for example "L234".

[0102] For modifications where the original amino acid and / or the substituted amino acid can contain more than one but not all amino acids, the more than one amino acids can be separated by "," or " / ". For example, substitution of leucine by phenylalanine, arginine, lysine or tryptophan at position 234: "Leu234Phe,Arg,Lys,Trp" or "Leu234Phe / Arg / Lys / Trp" or "L234F,R,K,W" or "L234F / R / K / W" or "L234 to F,R,K or W".

[0103] In the context of the present invention, such nomenclatures can be used interchangeably but have the same meaning and purpose.

[0104] In addition, the term "substitution" includes substitution with any one of the other 19 natural amino acids, or other amino acids, such as non-natural amino acids. For example, substitution of the amino acid L at position 234 includes the following various substitutions: 234A, 234C, 234D, 234E, 234F, 234G, 234H, 234I, 234K, 234M, 234N, 234Q, 234R, 234S, 234T, 234V, 234W, 234P, and 234Y. That is, in this way, it is equivalent to naming 234X, where X indicates any amino acid other than the original amino acid. These substitutions can also be referred to as L234A, L234C, etc., or L234A / C / , etc. The naming similarly applies to each and all positions mentioned herein, and specifically includes any one of such substitutions herein.

[0105] The antibodies of the present invention may also comprise deletions of amino acid residues. Such deletions can be denoted as "del", and include, for example, written as L234del. Thus, in such an embodiment, the leucine at position 234 is deleted from the amino acid sequence.

[0106] The terms "amino acid" and "amino acid residue" are used interchangeably herein.

[0107] As used herein, the term "C1q binding" refers to C1q binding to the antibody when the antibody binds to its antigen. The term "binding to its antigen" as used herein refers to the binding of the antibody to its antigen in vivo and in vitro.

[0108] As used herein, the term "reduce" when referring to C1q binding means that the antibodies of the present invention reduce, minimize, or even completely inhibit the ability of C1q to bind to the antibody as compared to the binding of C1q to the wild-type antibody.

[0109] The term "wild-type antibody" as used herein, when used in a comparative assay of the antibodies of the present invention, refers to an antibody that is identical to the antibody being tested except that it is not inactive. In this context, the term "inactive" means that the modified Fc region has reduced or no C1q binding, as determined in Example 10, i.e., when C1q binding is measured by ELISA as determined in Example 4 (i.e., T-cell proliferation is measured in a PBMC-based functional assay) and / or reduced or no Fc-mediated CD69 expression as determined in Example 3 (i.e., Fc-mediated CD69 expression is determined in a PBMC-based functional assay). Thus, a wild-type antibody contains the amino acids that occur naturally in the immunoglobulin heavy chain, i.e., an antibody that does not contain any amino acid modifications that can alter or reduce the ability of the antibody to interact with, for example, C1q, Fc receptors, etc. Thus, such a wild-type antibody will remain an activated antibody capable of binding, for example, C1q. The wild-type antibody and the antibodies of the present invention may contain amino acid modifications that do not affect the ability of the antibody to induce effector functions, in order to prepare the antibody as a bispecific antibody, etc.

[0110] The term "ELISA" as used herein refers to an enzyme-linked immunosorbent assay, which is a test that uses antibodies and a color change to identify a substance. A first specific antibody is attached to the surface of a plate. Thereby, a protein from a sample is added, and the binding to the first specific antibody is tested. A second antibody that binds to the antibody from the sample is added. The second antibody is linked to an enzyme, and in a final step, a substance containing the substrate for the enzyme is added. The subsequent reaction produces a detectable signal, most commonly a color change of the substrate. The concept of the ELISA method is well known in the art, and the various ways of performing an ELISA are considered part of the method for evaluating the antibodies of the present invention. Thus, this explanation should not be construed as limiting, since various forms of ELISA can be performed, such as those described in Example 4.

[0111] In particular, the ability of the antibodies of the present invention to bind C1q can be determined by ELISA, including the steps of: (i) coating the antibody onto a 96-well plate, (ii) adding 3% serum, (iii) adding an anti-human C1q antibody, (iv) developing the plate, and (v) measuring the OD 405 nm. Thus, in one embodiment, the antibody contains an Fc region that has been modified such that the binding of C1q to the antibody is reduced by at least 70%, at least 80%, at least 90%, at least 95%, at least 97% or 100% compared to the wild-type antibody, wherein C1q binding is determined by ELISA, including the steps of: (i) coating the antibody onto a 96-well plate, (ii) adding 3% serum, (iii) adding anti-human C1q, (iv) developing the plate, and (v) measuring the OD 405nm. Thus, in a specific embodiment, the binding of C1q was evaluated as described in Example 10.

[0112] As used herein, the term "Fc receptor" or "FcR" refers to a protein present on the surface of certain cells. FcR binds to the Fc region of an antibody. There are several different types of FcR, which are classified according to the type of antibody they recognize. For example, Fcγ (gamma) receptors bind to IgG-type antibodies.

[0113] As used herein, the term "Fcγ receptor", "Fc gamma receptor" or "FcγR" refers to a group of Fc receptors belonging to the immunoglobulin superfamily and is the most important Fc receptor for inducing phagocytosis of opsonized (coated) microorganisms. This family includes several members, FcγRI (CD64), FcγRIIa (CD32a), FcγRIIb (CD32b), FcγRIIIa (CD16a), FcγRIIIb (CD16b), which differ in antibody affinity due to different molecular structures.

[0114] Fc-mediated effector functions form part of the biological activity of human immunoglobulin G (IgG) molecules. Examples of such effector functions include, for example, antibody-dependent cell-mediated cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC), which are triggered by the binding of various effector molecules to the Fc region. In the context of the present invention, "Fc binding", "Fc receptor binding", "FcR binding", and "binding of the antibody Fc region to an FcR" refer to the binding of the Fc region to an Fc receptor (FcR) or an effector molecule. The terms "FcγR binding" and "FcγRI binding" refer to binding to or binding with the Fc region to an Fc gamma receptor and an Fc gamma receptor I, respectively. When a CD3 antibody binds to a T cell, the wild-type Fc region of the CD3 antibody binds to FcRs present on other cells (e.g., monocytes), which results in non-specific Fc-mediated T cell activation. Such non-specific Fc-mediated T cell activation may be undesirable. T cells can also be activated by targeted or target-specific T cell activation. For the treatment of many indications such as cancer, such targeted T cell activation can be highly desirable. As used herein, the term "targeted T cell activation" refers to the guidance of T cells to specific cells such as tumor cells by using a bispecific antibody that comprises a first binding region that binds to a specific target such as a tumor target on a tumor cell, and a second binding region that binds to a T cell-specific target such as CD3. Thus, the targeting of T cells to specific cells such as tumor cells can be facilitated by using a bispecific antibody in which one of the binding regions binds to CD3 present on the T cell and the other binding region binds to a target-specific antigen, such as a target-specific antigen on a tumor cell. Nevertheless, non-specific Fc-mediated T cell activation may still be present, and thus such unwanted non-specific Fc-mediated T cell activation by Fc-mediated cross-linking should be avoided, and it may be desirable to prepare an Fc region that is inactive with respect to such activity. Thus, the interaction between the inactive Fc region and the FcRs present is blocked. The humanized antibodies of the present invention have proven to be inactive when tested in several different assays (i.e., see Examples 3-5). Another tested CD3 antibody, huCLB-T3 / 4, which contains amino acid modifications in the Fc region, also proved to be inactive when tested in different assays (i.e., see Examples 7-10). The humanized CD3 antibodies of the present invention that contain the amino acid substitutions L234F, L235E, and D265A, as described in the Examples, show low levels of CD69 expression on T cells (Example 3), abolition of Fc-mediated T cell proliferation (Example 4), and no non-specific target killing when in the form of a bispecific antibody (Example 5). Thus, the humanized antibodies of the present invention show excellent results in several assays when compared to wild-type antibodies.

[0115] The antibodies of the invention may comprise modifications of the Fc region. When an antibody comprises such a modification, it may become an inactive or non-activated antibody. As used herein, the terms "inactive", "inactivated" or "non-activated" mean that the antibody is at least unable to bind any Fcγ receptor, induce Fc-mediated FcR cross-linking or FcR-mediated cross-linking of the target antigen by two Fc regions of a single antibody, or unable to bind C1q. Inactivation of the Fc region of a humanized or chimeric CD3 antibody is suitably tested using the monospecific form of the antibody, but the inactivated Fc region so identified can be used with bispecific or other humanized or chimeric multispecific CD3 antibodies.

[0116] Several variants can be constructed to prepare Fc regions of antibodies that are inactive towards interaction with Fc gamma receptors and C1q for therapeutic antibody development. Examples of such variants are described herein.

[0117] Thus, in one embodiment, the antibody comprises an Fc region that has been modified such that, compared to the wild-type antibody, the antibody mediates reduced Fc-mediated T-cell proliferation, reduced by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99% or 100%, wherein the T-cell proliferation is measured in a functional assay based on peripheral blood mononuclear cells (PBMCs).

[0118] As used herein, the term "reduced" means a reduction in activity or expression when compared to a control protein such as an antibody. Specifically, the term "reduced" when referring to T-cell proliferation means the ability of the antibodies of the invention to reduce, minimize or even completely inhibit T-cell proliferation when compared to T-cell proliferation mediated by a wild-type antibody. The ability of an antibody to reduce T-cell proliferation can be evaluated by a functional assay based on PBMCs as described in Examples 4 and 8. In one embodiment, the assay is performed using human PBMCs. In another embodiment, the assay is performed using cynomolgus monkey PBMCs. In yet another embodiment, the assay is performed using rhesus monkey PBMCs. Due to the cross-reactivity of the antibodies of the invention, the PBMC-based assays described herein can be performed using PBMCs from any species to demonstrate a reduction in T-cell proliferation, provided that the PBMCs of the species used are within the cross-reactivity profile of the antibody, such as human, cynomolgus monkey or rhesus monkey.

[0119] As used herein, the term "peripheral blood mononuclear cell (PBMC)-based functional assay" refers to an assay for evaluating the functional characteristics of the antibodies of the present invention, such as the ability of the antibodies to affect T-cell proliferation or CD69 expression, wherein the only cells present are peripheral blood mononuclear cells. Thus, in one embodiment, T-cell proliferation is measured by a method comprising the steps of: incubating PBMCs with an antibody in the range of 1 - 1000 ng / mL for three days at 37 °C in a 5% (vol / vol) CO2 humidified incubator; adding a compound, such as BrdU, which incorporates into the DNA of proliferating cells; incubating for 5 hours, pelleting the cells, drying the cells, optionally storing the cells at 4 °C, coating the cells onto an ELISA plate, incubating with anti-BrdU-peroxidase for 90 minutes at room temperature, developing with 1 mg / mL 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) for approximately 30 minutes, adding 100 μL of 2% oxalic acid to terminate the reaction, and measuring the absorbance at 405 nm in a suitable microplate reader.

[0120] As used herein, the term "proliferation" refers to cell growth in the context of cell division.

[0121] As used herein, the term "BrdU" refers to 5-bromo-2'-deoxyuridine, which is an analogue of thymidine. When BrdU is added to a cell culture for a limited period of time (e.g., 4 hours), it will incorporate into the DNA of proliferating cells. After fixing the cells, anti-BrdU-peroxidase can be used in an ELISA for the detection of incorporated BrdU. BrdU incorporation is thus a measure of proliferation.

[0122] In one embodiment, the antibody comprises an Fc region that has been modified such that, compared to the wild-type antibody, the antibody reduces Fc-mediated CD69 expression by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99% or 100%, wherein the Fc-mediated CD69 expression is determined in a PBMC-based functional assay.

[0123] As used herein, the term "reduction" refers to a reduction in activity or expression as compared to a control protein, such as an antibody. Specifically, the term "reduction", when referring to the expression level of the T cell activation marker CD69, refers to a reduction in the expression level of CD69 as compared to the expression level of CD69 when T cells are bound by a wild-type antibody (provided that both binding regions of the antibody bind CD3). The ability of an antibody to reduce CD69 expression can be evaluated by a PBMC-based functional assay as described in Examples 3 and 7. Thus, in one embodiment, the expression of CD69 is measured by a method comprising the steps of: incubating PBMC with an antibody in the range of 1-1000 ng / mL for 16-24 hours at 37°C in a 5% (vol / vol) CO2 humidified incubator, washing the cells, staining the cells with mouse anti-human CD28-PE and mouse-anti-human CD69-APC antibodies at 4°C, and determining the expression of CD69 on CD28-positive cells by flow cytometry.

[0124] As used herein, the term "CD69" refers to cluster of differentiation 69, a human transmembrane C-type lectin protein encoded by the CD69 gene. Activation of T lymphocytes and natural killer (NK) cells in vivo and in vitro induces the expression of CD69. CD69 functions as a signaling receptor involved in cell activation events including proliferation, as a signaling receptor in lymphocytes including natural killer cells and platelets, and induces specific genes.

[0125] As used herein, the term "peripheral blood mononuclear cell (PBMC)-based functional assay" refers to an assay for evaluating the functional characteristics of the antibodies of the present invention, such as the ability of the antibodies to affect T-cell proliferation or CD69 expression, wherein the only cells present are peripheral blood mononuclear cells. The PBMC-based functional assays described in Examples 3, 4, 5, and 7 comprise the steps of: (i) incubating PBMC with an antibody for about 16-24 hours at 37°C in a 5% (vol / vol) CO2 humidified incubator, (ii) washing the cells, (iii) staining the cells with mouse anti-human CD28-PE and mouse-anti-human CD69-APC antibodies at 4°C, and (iv) when evaluating CD69 expression, determining the expression of CD69 on CD28-positive cells by flow cytometry. Thus, in one embodiment, CD69 expression can be determined as described in Examples 3, 4, 5, or 7.

[0126] Thus, the amino acids that play a major role in the interaction with C1q and Fc Gamma receptors in the Fc region can be modified. Examples of amino acid positions that can be modified include positions L234, L235, and P331. Combinations thereof, such as L234F / L235E / P331S, can cause a significant reduction in binding to human CD64, CD32A, CD16, and C1q.

[0127] Thus, in one embodiment, the amino acids at at least one position corresponding to L234, L235, and P331 can be A, A, and S, respectively ([1],

[28] ). In addition, the L234F and L235E amino acid substitutions can generate an Fc region with abolished interaction with Fc Gamma receptors and C1q (

[29] -

[30] ). Thus, in one embodiment, the amino acids at the positions corresponding to L234 and L235 can be F and E, respectively. The D265A amino acid substitution can reduce binding to all Fc gamma receptors and prevent ADCC (

[31] ). Thus, in one embodiment, the amino acid at the position corresponding to D265 can be A. Binding to C1q can be abolished by mutating the positions D270, K322, P329, and P331. Mutating these positions to any of D270A or K322A or P329A or P331A can render the antibody lacking in CDC activity (

[32] ). Thus, in one embodiment, the amino acids at at least one position corresponding to D270, K322, P329, and P331 can be A, A, A, and A, respectively.

[0128] An alternative method for minimizing the interaction of the Fc region with Fc gamma receptors and C1q is by removing the glycosylation sites of the antibody. Mutating the position N297 to, for example, Q, A, and E removes glycosylation sites that are critical for IgG-Fc gamma receptor interaction. Thus, in one embodiment, the amino acid at the position corresponding to N297 can be G, Q, A, or E (

[33] ). Another alternative method for minimizing the interaction of the Fc region with Fc gamma receptors can be obtained by the following mutations: P238A, A327Q, P329A, or E233P / L234V / L235A / G236del (

[31] ).

[0129] Alternatively, although human IgG2 and IgG4 subclasses are considered to be naturally deficient in their interactions with C1q and Fc gamma receptors, there are still reports of interactions with Fcγ receptors (Fc gamma receptors) (

[34] -

[35] ). Mutations abolishing these residual interactions can be made in these two isotypes, resulting in reduced unwanted side effects associated with FcR binding. For IgG2, these include L234A and G237A, and for IgG4, include L235E. Thus, in one embodiment, the amino acids at the positions corresponding to L234 and G237 in the human IgG2 heavy chain can be A and A, respectively. In one embodiment, the amino acid at the position corresponding to L235 in the human IgG4 heavy chain can be E.

[0130] Other methods for further minimizing the interaction with Fcgamma receptors and C1q in IgG2 antibodies include those described in

[36] and

[37] .

[0131] Regarding the interaction with Fcgamma receptors and complement, the hinge region of the antibody may also be important (

[38] -

[39] ). Thus, mutations or deletions in the hinge region can affect the effector function of the antibody.

[0132] The term "crosslinking" as used herein refers to the indirect bridging of the Fab arms (monovalent or bivalent) of an antibody that binds a target antigen by a cell bearing an FcR that binds the Fc region of the antibody. Thus, an antibody that binds its target antigen on a cell bearing the target antigen can crosslink to another cell expressing an FcR.

[0133] The term "non-specific killing" as used herein refers to the killing of cells through the cytotoxic function of T cells or other effector cells, via tumor target antigen-independent activation of said cells. Thus, non-specific killing means that cells bearing tumor targets can be killed, for example, by cytotoxic T cells, rather than by an antibody that binds the tumor target, for example, by inducing CDC.

[0134] The inventors have shown (see Examples 3-5, 7-10) that a non-activated Fc region can be obtained by modifying one or more of at least five specific amino acid positions in the Fc region.

[0135] Thus, in one embodiment, the antibody comprises first and second immunoglobulin heavy chains, wherein in at least one of said first and second immunoglobulin heavy chains, one or more of the amino acids at positions corresponding to positions L234, L235, D265, N297, and P331 of the human IgG1 heavy chain are not L, L, D, N, and P, respectively.

[0136] In one embodiment, in both the first and second heavy chains, one or more of the amino acids at positions corresponding to positions L234, L235, D265, N297, and P331 of the human IgG1 heavy chain are not L, L, D, N, and P, respectively.

[0137] In another embodiment, in at least one of the first and second heavy chains, one or more of the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are not L, L, and D, respectively, and the amino acids at positions corresponding to N297 and P331 of the human IgG1 heavy chain are N and P, respectively.

[0138] As used herein, the term "amino acid corresponding to a position" refers to the amino acid position numbering of the human IgG1 heavy chain. Unless otherwise specified or the context is contradictory, the amino acids of the constant region sequence are numbered herein according to the Eu numbering index (described in

[27] ). Thus, an amino acid or segment in one sequence that "corresponds to" an amino acid or segment in another sequence is an amino acid or segment that is aligned with the other amino acid or segment using a standard sequence alignment program such as ALIGN, ClustalW or a similar program, typically with default settings, and has at least 50%, at least 80%, at least 90% or at least 95% identity to the human IgG1 heavy chain. It is well known in the art how to align sequences or segments within sequences and thus determine the positions within a sequence that correspond to the amino acid positions of the present invention.

[0139] In the context of the present invention, an amino acid may be defined as described above.

[0140] The term "amino acid is not" or a similar phrase, when referring to an amino acid in the heavy chain, is to be understood to mean that the amino acid is any other amino acid than the specific amino acid mentioned. For example, the amino acid at the position corresponding to L234 of the human IgG1 heavy chain is not L, meaning that the amino acid can be any other natural or non-natural amino acid other than L.

[0141] In one embodiment, in at least one of the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is not D.

[0142] In one embodiment, in at least one of the first and second heavy chains, the amino acid at the position corresponding to D265 of the human IgG1 heavy chain is not D, and the amino acids at the positions corresponding to positions N297 and P331 of the human IgG1 heavy chain are N and P, respectively.

[0143] In one embodiment, in at least one of the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is a hydrophobic or polar amino acid.

[0144] As used herein, the term "hydrophobic" with respect to an amino acid residue refers to an amino acid residue selected from A, C, F, G, H, I, L, M, R, T, V, W and Y. Thus, in one embodiment, in at least one of the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from the amino acids consisting of A, C, F, G, H, I, L, M, R, T, V, W and Y.

[0145] As used herein, the term "polar" with respect to an amino acid residue refers to any amino acid residue selected from C, D, E, H, K, N, Q, R, S, and T. Thus, in one embodiment, in at least one of the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human heavy chain is selected from C, E, H, K, N, Q, R, S, and T.

[0146] In another embodiment, in at least one of the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is an aliphatic uncharged, aromatic, or acidic amino acid.

[0147] As used herein, the term "aliphatic uncharged" with respect to an amino acid residue refers to any amino acid residue selected from A, G, I, L, and V. Thus, in one embodiment, in at least one of the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from A, G, I, L, and V.

[0148] As used herein, the term "aromatic" with respect to an amino acid residue refers to any amino acid residue selected from F, T, and W. Thus, in one embodiment, in at least one of the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from F, T, and W.

[0149] As used herein, the term "acidic" with respect to an amino acid residue refers to any amino acid residue selected from D and E. Thus, in one embodiment, in at least one of the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from D and E.

[0150] In a specific embodiment, in at least one of the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from A, E, F, G, I, L, T, V, and W.

[0151] In one embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is not D.

[0152] In one embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to D265 of the human IgG1 heavy chain is not D, and the amino acids at the positions corresponding to positions N297 and P331 of the human IgG1 heavy chain are N and P, respectively.

[0153] In one embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is a hydrophobic or polar amino acid.

[0154] As used herein, the term "hydrophobic" with respect to an amino acid residue refers to an amino acid residue selected from A, C, F, G, H, I, L, M, R, T, V, W, and Y. Thus, in one embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from the amino acids consisting of A, C, F, G, H, I, L, M, R, T, V, W, and Y.

[0155] As used herein, the term "polar" with respect to an amino acid residue refers to any amino acid residue selected from C, D, E, H, K, N, Q, R, S, and T. Thus, in one embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human heavy chain is selected from C, E, H, K, N, Q, R, S, and T. In one embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from the amino acids consisting of A, C, F, G, H, I, L, M, R, T, V, W, and Y.

[0156] In one embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human heavy chain is selected from C, E, H, K, N, Q, R, S, and T.

[0157] In another embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is an aliphatic uncharged, aromatic, or acidic amino acid.

[0158] As used herein, the term "aliphatic uncharged" with respect to an amino acid residue refers to any amino acid residue selected from A, G, I, L, and V. Thus, in one embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from A, G, I, L, and V.

[0159] As used herein, the term "aromatic" with respect to an amino acid residue refers to any amino acid residue selected from F, T, and W. Thus, in one embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from F, T, and W.

[0160] As used herein, the term "acidic" with respect to an amino acid residue refers to any amino acid residue selected from D and E. Thus, in one embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from D and E.

[0161] In a specific embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from A, E, F, G, I, L, T, V, and W.

[0162] In a further embodiment, in at least one of the first and second heavy chains, the amino acid at the position corresponding to position N297 of the human IgG1 heavy chain is not N.

[0163] In one embodiment, in at least one of the first and second heavy chains, the amino acid at the position corresponding to N297 of the human IgG1 heavy chain is not N, and the amino acid at the position corresponding to position P331 of the human IgG1 heavy chain is P.

[0164] In one embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to position N297 of the human IgG1 heavy chain is not N.

[0165] In one embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to N297 of the human IgG1 heavy chain is not N, and the amino acid at the position corresponding to position P331 of the human IgG1 heavy chain is P.

[0166] In a further embodiment, in at least one of the first and second heavy chains, the amino acids at the positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are not L and L, respectively.

[0167] In one embodiment, in at least one of the first and second heavy chains, the amino acids at the positions corresponding to L234 and L235 of the human IgG1 heavy chain are not L and L, respectively, and the amino acids at the positions corresponding to positions N297 and P331 of the human IgG1 heavy chain are N and P, respectively.

[0168] In one embodiment, in at least one of the first and second heavy chains, the amino acids at the positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are selected from A, C, D, E, F, G, H, I, K, M, N, P, Q, R, S, T, Y, V.

[0169] In one embodiment, in at least one of the first and second heavy chains, the amino acids at the positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are hydrophobic or polar amino acids.

[0170] As used herein, the term "hydrophobic" with respect to an amino acid residue refers to an amino acid residue selected from A, C, F, G, H, I, L, M, R, T, V, W, and Y. Thus, in one embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each independently selected from A, C, F, G, H, I, M, R, T, V, W, and Y.

[0171] As used herein, the term "polar" with respect to an amino acid residue refers to any amino acid residue selected from C, D, E, H, K, N, Q, R, S, and T. Thus, in one embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each independently selected from the group consisting of C, D, E, H, K, N, Q, R, S, and T.

[0172] In a specific embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each independently selected from A, C, D, E, F, G, H, I, K, M, N, Q, R, S, T, V, W, and Y.

[0173] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are not L and L, respectively.

[0174] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are not L and L, respectively, and the amino acids at positions corresponding to positions N297 and P331 of the human IgG1 heavy chain are N and P, respectively.

[0175] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are hydrophobic or polar amino acids.

[0176] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each independently selected from A, C, F, G, H, I, M, R, T, V, W, and Y.

[0177] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each independently selected from the group consisting of C, D, E, H, K, N, Q, R, S, and T.

[0178] In a specific embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each independently selected from A, C, D, E, F, G, H, I, K, M, N, Q, R, S, T, V, W, and Y.

[0179] In another embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are aliphatic uncharged, aromatic, or acidic amino acids.

[0180] As used herein, the term "aliphatic uncharged" with respect to an amino acid residue refers to any amino acid residue selected from A, G, I, L, and V. Thus, in one embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each independently selected from A, G, I, and V.

[0181] As used herein, the term "aromatic" with respect to an amino acid residue refers to any amino acid residue selected from F, T, and W. Thus, in one embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each independently selected from F, T, and W.

[0182] As used herein, the term "acidic" with respect to an amino acid residue refers to any amino acid residue selected from D and E. Thus, in one embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each independently selected from D and E.

[0183] In a specific embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to L234 and L235 are each independently selected from A, D, E, F, G, I, T, V, and W.

[0184] In one embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are F and E, respectively; or A and A.

[0185] In one embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to L234 and L235 of the human IgG1 heavy chain are F and E, respectively; or A and A, and the amino acids at positions corresponding to positions N297 and P331 of the human IgG1 heavy chain are N and P, respectively.

[0186] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are F and E, respectively; or A and A.

[0187] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to L234 and L235 of the human IgG1 heavy chain are F and E, respectively; or A and A, and the amino acids at positions corresponding to positions N297 and P331 of the human IgG1 heavy chain are N and P, respectively.

[0188] In a specific embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are F and E, respectively.

[0189] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are F and E, respectively.

[0190] In one embodiment, in at least one of the first and second heavy chains, at least the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are A and A, respectively.

[0191] In one embodiment, in both the first and second heavy chains, at least the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are A and A, respectively.

[0192] In one embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are not L, L, and D, respectively.

[0193] In one embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to L234, L235, and D265 of the human IgG1 heavy chain are not L, L, and D, respectively, and the amino acids at positions corresponding to positions N297 and P331 of the human IgG1 heavy chain are N and P, respectively.

[0194] In one embodiment, in at least one of the first and second heavy chains, the amino acids corresponding to positions L234 and L235 of the human IgG1 heavy chain are selected from A, C, D, E, F, G, H, I, K, M, N, P, Q, R, S, T, Y, V, and W, and the amino acid corresponding to position D265 is selected from A, C, E, F, G, H, I, K, L, M, N, P, Q, R, S, T, Y, V, and W.

[0195] In one embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are hydrophobic or polar amino acids.

[0196] As used herein, the term "hydrophobic" with respect to an amino acid residue refers to an amino acid residue selected from A, C, F, G, H, I, L, M, R, T, V, W, and Y. Thus, in one embodiment, in at least one of the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from A, C, F, G, H, I, L, M, R, T, V, W, and Y, and the amino acids at the positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each selected from A, C, F, G, H, I, M, R, T, V, W, and Y.

[0197] As used herein, the term "polar" with respect to an amino acid residue refers to any amino acid residue selected from C, D, E, H, K, N, Q, R, S, and T. Thus, in one embodiment, in at least one of the first and second heavy chains, the amino acids at the positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each selected from C, D, E, H, K, N, Q, R, S, and T, and the amino acid at the position corresponding to position D265 of the human heavy chain is selected from C, E, H, K, N, Q, R, S, and T.

[0198] In a specific embodiment, in at least one of the first and second heavy chains, the amino acids at the positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each selected from A, C, D, E, F, G, H, I, K, M, N, Q, R, S, T, V, W, and Y, and the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from A, C, E, F, G, H, I, K, L, M, N, Q, R, S, T, V, W, and Y.

[0199] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to L234, L235, and D265 of the human IgG1 heavy chain are hydrophobic or polar amino acids.

[0200] In one embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from A, C, F, G, H, I, L, M, R, T, V, W, and Y, and the amino acids at the positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each selected from A, C, F, G, H, I, M, R, T, V, W, and Y.

[0201] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each independently selected from C, D, E, H, K, N, Q, R, S, and T, and the amino acid at the position corresponding to position D265 of the human heavy chain is selected from C, E, H, K, N, Q, R, S, and T.

[0202] In a specific embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each independently selected from A, C, D, E, F, G, H, I, K, M, N, Q, R, S, T, V, W, and Y, and the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from A, C, E, F, G, H, I, K, L, M, N, Q, R, S, T, V, W, and Y.

[0203] In another embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are aliphatic uncharged, aromatic, or acidic amino acids.

[0204] As used herein, the term "aliphatic uncharged" with respect to an amino acid residue refers to any amino acid residue selected from A, G, I, L, and V. Thus, in one embodiment, in at least one of the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from A, G, I, L, and V, and the amino acids at the positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each independently selected from A, G, I, and V.

[0205] As used herein, the term "aromatic" with respect to an amino acid residue refers to any amino acid residue selected from F, T, and W. Thus, in one embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are each independently selected from F, T, and W.

[0206] As used herein, the term "acidic" with respect to an amino acid residue refers to any amino acid residue selected from D and E. Thus, in one embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are each independently selected from D and E.

[0207] In a specific embodiment, in at least one of the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from A, E, F, G, I, L, T, V, and W, and the amino acids at the positions corresponding to L234 and L235 are each independently selected from A, D, E, F, G, I, T, V, and W.

[0208] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are not L, L, and D, respectively.

[0209] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are not L, L, and D, respectively, and the amino acids at positions corresponding to positions N297 and P331 of the human IgG1 heavy chain are N and P, respectively.

[0210] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are aliphatic uncharged, aromatic, or acidic amino acids.

[0211] In one embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from A, G, I, L, and V, and the amino acids at the positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are each independently selected from A, G, I, and V.

[0212] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are each independently selected from D and E.

[0213] In a specific embodiment, in both the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is selected from A, E, F, G, I, L, T, V, and W, and the amino acids at the positions corresponding to positions L234 and L235 are each independently selected from A, D, E, F, G, I, T, V, and W.

[0214] In one embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are F, E, and A, respectively; or A, A, and A.

[0215] In one embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are F, E, and A, respectively; or A, A, and A, and the amino acids at positions corresponding to positions N297 and P331 of the human IgG1 heavy chain are N and P, respectively.

[0216] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are F, E, and A, respectively; or A, A, and A.

[0217] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are F, E, and A, respectively; or A, A, and A, and the amino acids at positions corresponding to positions N297 and P331 of the human IgG1 heavy chain are N and P, respectively.

[0218] In a specific embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are F, E, and A, respectively.

[0219] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are F, E, and A, respectively.

[0220] In one embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are A, A, and A, respectively.

[0221] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are A, A, and A, respectively.

[0222] In another embodiment, in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, D265, N297, and P331 of the human IgG1 heavy chain are F, E, A, Q, and S, respectively.

[0223] In one embodiment, in both the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, D265, N297, and P331 of the human IgG1 heavy chain are F, E, A, Q, and S, respectively.

[0224] In a specific embodiment, the antibody of the present invention comprises the VH sequence shown in SEQ ID NO:8, the VL sequence shown in SEQ ID NO:10, and the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are F, E, and A, respectively, in at least one heavy chain.

[0225] In another embodiment, the antibody of the invention comprises the VH sequence shown in SEQ ID NO:8, the VL sequence shown in SEQ ID NO:12, and the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one heavy chain are F, E and A, respectively.

[0226] In another embodiment, the antibody of the invention comprises the VH sequence shown in SEQ ID NO:6, the VL sequence shown in SEQ ID NO:10, and the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one heavy chain are F, E and A, respectively.

[0227] In another embodiment, the antibody of the invention comprises the VH sequence shown in SEQ ID NO:6, the VL sequence shown in SEQ ID NO:12, and the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one heavy chain are F, E and A, respectively.

[0228] In another embodiment, the antibody of the invention comprises the VH sequence shown in SEQ ID NO:9, the VL sequence shown in SEQ ID NO:10, and the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one heavy chain are F, E and A, respectively.

[0229] In another embodiment, the antibody of the invention comprises the VH sequence shown in SEQ ID NO:9, the VL sequence shown in SEQ ID NO:12, and the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one heavy chain are F, E and A, respectively.

[0230] In one aspect, the antibody of the invention comprises the human IgLC2 / IgLC3 constant domain lambda light chain of SEQ ID NO:31.

[0231] In one aspect, the antibodies of the present invention can be modified in the light chain (LC) and / or the heavy chain (HC) to increase the expression level and / or the production yield. In one embodiment, the antibodies of the present invention can be modified in the light chain (LC). Such modifications are known in the art and can be carried out according to the methods described in, for example, Zheng, L., Goddard, J.-P., Baumann, U., & Reymond, J.-L. (2004) Expression improvement and mechanistic study of the retro-Diels-Alderase catalytic antibody 10F11 by site-directed mutagenesis. Journal of Molecular Biology, 341(3), 807–14. doi:10.1016 / j.jmb.2004.06.014.

[0232] In one embodiment, the antibody of the present invention comprises a light chain (LC), wherein the amino acid at the position corresponding to position T41 of the λ light chain of SEQ ID NO:10 is not T.

[0233] In one embodiment, the antibody of the present invention comprises a light chain (LC), wherein the amino acid at the position corresponding to position T41 of the λ light chain of SEQ ID NO:10 is selected from H, I, K, L, Q, R, and V.

[0234] In one embodiment, the antibody of the present invention comprises a light chain (LC), wherein the amino acid at the position corresponding to position T41 of the λ light chain of SEQ ID NO:10 is H, K, or R, such as K.

[0235] In one embodiment, the antibody of the present invention comprises a light chain (LC), wherein the amino acid at the position corresponding to position F10 of the λ light chain of SEQ ID NO:10 is not F, and wherein one or more of the amino acid positions corresponding to positions T41, K55, and L97 of the λ light chain of SEQ ID NO:10 are not T, K, and L, respectively.

[0236] In one embodiment, the antibody of the present invention comprises a light chain (LC), wherein the amino acids at the positions corresponding to positions F10, T41, K55, and L97 of the λ light chain of SEQ ID NO:10 are not F, T, K, and L, respectively.

[0237] In one embodiment, the antibody of the present invention comprises a light chain (LC), wherein the amino acid at the position corresponding to position T41 is selected from H, I, K, L, Q, R or V, such as selected from H, K and R, such as K, and wherein the amino acids at the positions corresponding to positions F10, T41, K55 and L97 of the λ light chain of SEQ ID NO:10 are L, K, N and H, respectively.

[0238] In one embodiment, the antibody of the present invention comprises a light chain (LC), wherein the amino acids at the positions corresponding to positions R23 and A35 are not R and A, respectively.

[0239] In one embodiment, the antibody of the present invention comprises a light chain (LC), wherein the amino acid at the position corresponding to position R23 is selected from A, G, H, K, Q, S and T, such as selected from A and G, and wherein the amino acid at the position corresponding to A35 is selected from I, L, M, P, V, G, F and W, such as selected from I, L, M, P and V.

[0240] In one embodiment, the antibody of the present invention comprises a light chain (LC), wherein the amino acid at the position corresponding to position R23 is A or G, such as A, and the amino acid at the position corresponding to position A35 is P.

[0241] In one embodiment, the antibody of the present invention comprises a light chain (LC), wherein the amino acids at the positions corresponding to positions F10, R23, A35, R47, D71, A82, D83, S86, I87 and F89 of the λ light chain of SEQ ID NO:10 are not F, R, A, R, D, A, D, S, I and F, respectively.

[0242] In one embodiment, the antibody of the present invention comprises a light chain (LC), wherein the amino acid at the position corresponding to position R23 is selected from A, G, H, K, Q, S and T, such as selected from A and G, wherein the amino acid at the position corresponding to A35 is selected from I, L, M, P, V, G, F and W, such as selected from I, L, M, P, and wherein the amino acids at the positions corresponding to positions F10, R47, D71, A82, D83, S86, I87 and F89 of the λ light chain of SEQ ID NO:10 are L, T, G, P, E, A, E and Y, respectively.

[0243] In one embodiment, the antibody of the present invention comprises a light chain (LC), wherein the amino acid at the position corresponding to position R23 is A or G, and wherein the amino acids at the positions corresponding to positions F10, A35, R47, D71, A82, D83, S86, I87 and F89 of the λ light chain of SEQ ID NO:10 are L, P, T, G, P, E, A, E and Y, respectively.

[0244] In one aspect, an antibody of the invention can be modified in the light chain (LC) to increase the affinity of the antibody. Modifications that result in increased antibody affinity are known in the art.

[0245] In one aspect, an antibody of the invention can be modified in the light chain (LC) to decrease the affinity of the antibody. This may be advantageous in certain cases and result in increased efficacy. Specifically, a low-affinity CD3 arm may have an impact on the motility of T cells in the circulation and at the tumor site, leading to better engagement of tumor cells with T cells, see et al., Molecular Immunology 44 (2007). Specifically, this can be used in bispecific forms where the CD3 antibody serves as one of the binding arms. Modifications that result in decreased antibody affinity are known in the art, see for example Webster et al. Int J Cancer Suppl. 1988; 3:13-6.

[0246] In one embodiment, an antibody of the invention comprises a light chain (LC) wherein (i) the amino acid at the position corresponding to position F10 of the λ light chain of SEQ ID NO: 10 is not F, or (ii) the amino acid at the position corresponding to position K55 of the λ light chain of SEQ ID NO: 10 is not K, or (iii) the amino acid at the position corresponding to position F10 of the λ light chain of SEQ ID NO: 10 is not F and the amino acid at the position corresponding to position K55 of the λ light chain of SEQ ID NO: 10 is not K.

[0247] In one embodiment, an antibody of the invention comprises a constant light chain (LC) wherein (i) the amino acid at the position corresponding to position F10 of the λ light chain of SEQ ID NO: 10 is L, or (ii) the amino acid at the position corresponding to position K55 of the λ light chain of SEQ ID NO: 10 is N, or (iii) the amino acid at the position corresponding to position F10 of the λ light chain of SEQ ID NO: 10 is L and the amino acid at the position corresponding to position K55 of the λ light chain of SEQ ID NO: 10 is N.

[0248] In one embodiment, an antibody of the invention comprises a light chain (LC) wherein the amino acids at the positions corresponding to positions F10, T41, K55 and L97 of the λ light chain of SEQ ID NO: 10 are not F, T, K and L, respectively. Such modifications are used to increase both the expression level and decrease the affinity.

[0249] The antibody of the present invention comprises a light chain (LC), wherein the amino acids at positions corresponding to positions F10, T41, K55, and L97 of the λ light chain of SEQ ID NO: 10 are L, K, N, and H, respectively. Such modifications are used to increase both the expression level and reduce the affinity.

[0250] In one embodiment, the antibody of the present invention comprises a heavy chain (HC) and a light chain (LC), wherein the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain of SEQ ID NO: 15 are F, E, and A, respectively, and wherein the amino acid at position F405 of the human IgG1 heavy chain corresponding to SEQ ID NO: 15 is L, and wherein (i) the amino acids at positions corresponding to positions F10, T41, K55, and L97 of the λ light chain of SEQ ID NO: 10 are L, K, N, and H, respectively, or (ii) the amino acid at position T41 of the λ light chain corresponding to SEQ ID NO: 10 is K.

[0251] In one embodiment, the antibody of the present invention comprises a heavy chain (HC) of SEQ ID NO: 25 and a light chain (LC) of SEQ ID NO: 32.

[0252] In one embodiment, the antibody of the present invention comprises a heavy chain (HC) of SEQ ID NO: 25 and a light chain (LC) of SEQ ID NO: 33.

[0253] In one aspect, the present invention relates to a multispecific antibody that comprises at least a first binding region of an antibody of any aspect or embodiment described herein, and one or more binding regions that bind to one or more different targets than the first binding region. Such a multispecific antibody can be a bispecific antibody.

[0254] Thus, in one aspect, the present invention relates to a bispecific antibody that comprises a first binding region of an antibody of any aspect or embodiment described herein, and a second binding region that binds to a different target than the first binding region.

[0255] The term "multispecific antibody" refers to an antibody that is specific for at least two different, e.g., at least three, typically non-overlapping epitopes. Such epitopes can be on the same or different targets. If the epitopes are on different targets, such targets can be on the same cell or different cells or cell types.

[0256] The term "bispecific antibody" refers to an antibody that is specific for at least two different, typically non-overlapping epitopes. Such epitopes can be on the same or different targets. If the epitopes are on different targets, such targets can be on the same cell or different cells or cell types.

[0257] In one embodiment, the bispecific antibody comprises a first and a second heavy chain.

[0258] Embodiments involving modifications to the Fc region and embodiments involving specific amino acid substitutions are expected to be part of any bispecific antibody of the invention. Thus, in one embodiment, at least one of the first and second heavy chains comprises one or more amino acids modified according to the definition of any of the embodiments described herein (such as those described with respect to providing an inactivated Fc region). In one embodiment, both the first and second heavy chains comprise one or more amino acids modified according to the definition of any of the embodiments described herein (such as those described with respect to providing an inactivated Fc region). Thus, the bispecific antibody comprises an Fc region modified according to any aspect or embodiment described herein; or at least one of the first and second heavy chains comprises one or more amino acids modified according to the definition of any aspect or embodiment described herein.

[0259] Thus, in one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:10; and wherein the Fc region has been modified such that the binding of C1q to the antibody is reduced by at least 70%, at least 80%, at least 90%, at least 95%, at least 97% or 100% compared to the wild-type antibody, wherein C1q binding is determined by ELISA.

[0260] In one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:12; and wherein the Fc region has been modified such that the binding of C1q to the antibody is reduced by at least 70%, at least 80%, at least 90%, at least 95%, at least 97% or 100% compared to the wild-type antibody, wherein C1q binding is determined by ELISA.

[0261] In one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:10; and wherein the Fc region has been modified such that the binding of C1q to the antibody is reduced by at least 70%, at least 80%, at least 90%, at least 95%, at least 97% or 100% compared to the wild-type antibody, wherein C1q binding is determined by ELISA.

[0262] In one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:12; and wherein the Fc region has been modified such that the binding of C1q to the antibody is reduced by at least 70%, at least 80%, at least 90%, at least 95%, at least 97% or 100% compared to the wild-type antibody, wherein C1q binding is determined by ELISA.

[0263] In one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:10; and wherein the Fc region has been modified such that the binding of C1q to the antibody is reduced by at least 70%, at least 80%, at least 90%, at least 95%, at least 97% or 100% compared to the wild-type antibody, wherein C1q binding is determined by ELISA.

[0264] In one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:12; and wherein the Fc region has been modified such that the binding of C1q to the antibody is reduced by at least 70%, at least 80%, at least 90%, at least 95%, at least 97% or 100% compared to the wild-type antibody, wherein C1q binding is determined by ELISA.

[0265] In another embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:10; and wherein the Fc region has been modified such that the antibody-mediated Fc-mediated T-cell proliferation is reduced by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99% or 100% compared to the wild-type antibody, wherein the T-cell proliferation is measured by a functional assay based on peripheral blood mononuclear cells (PBMC).

[0266] In one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:12; and wherein the Fc region has been modified such that the antibody-mediated Fc-mediated T-cell proliferation is reduced by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99% or 100% compared to the wild-type antibody, wherein the T-cell proliferation is measured by a functional assay based on peripheral blood mononuclear cells (PBMC).

[0267] In one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:10; and wherein the Fc region has been modified such that, compared to the wild-type antibody, the antibody mediates at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99% or 100% less Fc-mediated T-cell proliferation, wherein the T-cell proliferation is measured using a functional assay based on peripheral blood mononuclear cells (PBMC).

[0268] In one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:12; and wherein the Fc region has been modified such that, compared to the wild-type antibody, the antibody mediates at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99% or 100% less Fc-mediated T-cell proliferation, wherein the T-cell proliferation is measured using a functional assay based on peripheral blood mononuclear cells (PBMC).

[0269] In one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:10; and wherein the Fc region has been modified such that, compared to the wild-type antibody, the antibody mediates at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99% or 100% less Fc-mediated T-cell proliferation, wherein the T-cell proliferation is measured using a functional assay based on peripheral blood mononuclear cells (PBMC).

[0270] In one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:12; and wherein the Fc region has been modified such that, compared to the wild-type antibody, the antibody mediates at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99% or 100% less Fc-mediated T-cell proliferation, wherein the T-cell proliferation is measured using a functional assay based on peripheral blood mononuclear cells (PBMC).

[0271] In another embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:10; and wherein the Fc region has been modified such that, compared to the wild-type antibody, the antibody reduces Fc-mediated CD69 expression by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99% or 100%, wherein the Fc-mediated CD69 expression is measured by a PBMC-based functional assay.

[0272] In one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:12; and wherein the Fc region has been modified such that, compared to the wild-type antibody, the antibody reduces Fc-mediated CD69 expression by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99% or 100%, wherein the Fc-mediated CD69 expression is measured by a PBMC-based functional assay.

[0273] In one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:10; and wherein the Fc region has been modified such that, compared to the wild-type antibody, the antibody reduces Fc-mediated CD69 expression by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99% or 100%, wherein the Fc-mediated CD69 expression is measured by a PBMC-based functional assay.

[0274] In one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:12; and wherein the Fc region has been modified such that, compared to the wild-type antibody, the antibody reduces Fc-mediated CD69 expression by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99% or 100%, wherein the Fc-mediated CD69 expression is measured by a PBMC-based functional assay.

[0275] In one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:10; and wherein the Fc region has been modified such that, compared to the wild-type antibody, the antibody reduces Fc-mediated CD69 expression by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99% or 100%, wherein the Fc-mediated CD69 expression is measured by a PBMC-based functional assay.

[0276] In one embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:12; and wherein the Fc region has been modified such that, compared to the wild-type antibody, the antibody reduces Fc-mediated CD69 expression by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99% or 100%, wherein the Fc-mediated CD69 expression is measured by a PBMC-based functional assay.

[0277] In a specific embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:10; and wherein in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain are F, E and A, respectively.

[0278] In another embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:12; and wherein in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain are F, E and A, respectively.

[0279] In another embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:10; and wherein in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain are F, E and A, respectively.

[0280] In another embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:12; and wherein in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain are F, E and A, respectively.

[0281] In another embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:10; and wherein in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain are F, E and A, respectively.

[0282] In another embodiment, the bispecific antibody comprises a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:12; and wherein in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain are F, E and A, respectively.

[0283] Examples of bispecific antibody molecules useful in the present invention include (i) a single antibody having two arms comprising different antigen-binding regions, (ii) a single-chain antibody specific for two different epitopes, such as two scFvs tandemly linked by an additional peptide linker; (iii) a dual variable domain antibody (DVD-Ig TM ) in which each light and heavy chain comprises two variable domains tandemly linked by a short peptide linker (

[40] ); (iv) a chemically linked bispecific (Fab’)2 fragment; (v) which is a fusion of two single-chain diabodies to produce a tetravalent bispecific antibody having two binding sites for each target antigen; (vi) a flexibody, which is a combination of an scFv and a diabody to produce a multivalent molecule; (vii) the so-called “dock-and-lock” molecule (Dock-and- ) which, based on the “dimerization and docking domain” in protein kinase A, when applied to a Fab, can produce a trivalent bispecific binding protein consisting of two identical Fab fragments linked to different Fab fragments; (viii) the so-called Scorpion molecule which comprises, for example, two scFvs fused to the two ends of a human Fab arm; and (ix) a diabody.

[0284] In one embodiment, the bispecific antibody of the present invention is a diabody, a cross-body, or a bispecific antibody obtained by controlled Fab arm exchange, such as (e.g., as described in

[41] ), such as those described in the present invention.

[0285] Examples of different types of bispecific antibodies include, but are not limited to, (i) IgG-like molecules with complementary CH3 domains to force heterodimerization; (ii) recombinant IgG-like dual-targeting molecules where each side of the molecule contains at least two different Fab fragments or parts of Fab fragments of antibodies; (iii) IgG fusion molecules where a full-length IgG antibody is fused to an additional Fab fragment or part of a Fab fragment; (iv) Fc fusion molecules where a single-chain Fv molecule or a stable diabody is fused to a heavy-chain constant domain, Fc-region, or a part thereof; (v) Fab fusion molecules where different Fab-fragments are fused together and fused to a heavy-chain constant domain, Fc-region, or a part thereof; and (vi) ScFv- and diabody- and heavy-chain antibody-based (e.g., domain antibodies, ), where different single-chain Fv molecules or different diabodies or different heavy-chain antibodies (e.g., domain antibodies, ) are fused to each other or to another protein or carrier molecule fused to a heavy-chain constant domain, Fc-region, or a part thereof.

[0286] Examples of IgG-like molecules with complementary CH3 domain molecules include, but are not limited to (Trion Pharma / Fresenius Biotech,

[42] ), Knobs-into-Holes (Genentech,

[43] ), CrossMAbs (Roche,

[44] ) and electrostatic pairing (Amgen,

[45] -

[46] ; Chugai,

[47] ; Oncomed,

[48] ), LUZ-Y (Genentech), DIG bodies and PIG bodies (Pharmabcine), strand-exchange re-engineered domain bodies (SEED bodies) (EMD Serono,

[49] ), Biclonics (Merus), FcΔAdp (Regeneron,

[50] ), bispecific IgG1 and IgG2 (Pfizer / Rinat,

[51] ), Azymetric scaffolds (Zymeworks / Merck,

[52] ), mAb-Fv (Xencor,

[53] ), bivalent bispecific antibodies (Roche) and molecules (Genmab A / S,

[41] ).

[0287] Examples of recombinant IgG-like dual-targeting molecules include, but are not limited to, dual-targeting (DT)-Ig (GSK / Domantis), tandem diabodies (Genentech), cross-linked Mab (Karmanos Cancer Center), mAb2 (F-Star,

[54] ), Zybodies TM(Zyngenia), the method using a common light chain (Crucell / Merus,

[55] ), κλBodies (NovImmune), and CovX- (CovX / Pfizer).

[0288] Examples of IgG fusion molecules include, but are not limited to, dual variable domain (DVD)-Ig TM (Abbott,

[56] ), dual domain diabody (Unilever; SanofiAventis,

[57] ), IgG-like bispecific (ImClone / Eli Lilly), Ts2Ab (MedImmune / AZ), and BsAb (Zymogenetics), HERCULES (Biogen Idec,

[58] ), scFv fusion (Novartis), scFv fusion (Changzhou Adam Biotech Inc,

[59] ), and TvAb (Roche,

[59] ,

[60] ).

[0289] Examples of Fc fusion molecules include, but are not limited to, ScFv / Fc fusion (Academic Institution), SCORPION (Emergent BioSolutions / Trubion, Zymogenetics / BMS), dual affinity retargeting technology (Fc-DART TM )(MacroGenics,

[62] ,

[63] ), and dual (ScFv)2-Fab (National Research Center for Antibody Medicine–China).

[0290] Examples of Fab fusion bispecific antibodies include, but are not limited to, F(ab)2 (Medarex / AMGEN), dual action or Bis-Fab (Genentech), Dock-and- (DNL) (ImmunoMedics), bivalent bispecific (Biotecnol), and Fab-Fv (UCB-Celltech).

[0291] Examples of ScFv-, diabody-based antibodies, and domain antibodies include, but are not limited to, bispecific T cell engagers (Micromet, Tandem Diabody (Tandab) (Affimed), dual affinity retargeting technology (DART TM)(MacroGenics), single-chain diabodies (Academic), TCR-like antibodies (AIT, ReceptorLogics), human serum albumin ScFv fusions (Merrimack) and COMBODY (EpigenBiotech), dual targeting (Ablynx), dual targeting single domain antibodies of only the heavy chain.

[0292] It is also contemplated that any monospecific antibody that meets the assay conditions described herein can form the basis of a bispecific antibody. That is, a bispecific antibody in which one of the binding regions binds CD3 can be derived from any monospecific CD3 antibody that has been tested in a functional assay and meets the requirements described herein. Such bispecific antibodies can be provided by the method described in

[41] , which is incorporated herein by reference.

[0293] Thus, in a specific embodiment, each of the first and second heavy chains comprises at least a hinge region, CH2 and CH3 regions, wherein at least one amino acid at a position corresponding to a position selected from T366, L368, K370, D399, F405, Y407 and K409 of the human IgG1 heavy chain is replaced in the first heavy chain, and at least one amino acid at a position corresponding to a position selected from T366, L368, K370, D399, F405, Y407 and K409 of the human IgG1 heavy chain is replaced in the second heavy chain, and wherein the first and the second heavy chains are not replaced at the same position. In this context, the term "replacement" means that the amino acid at a specific amino acid position is replaced by another naturally or non-naturally occurring amino acid. Thus, an amino acid "replaced" at a position corresponding to a position in the human IgG1 heavy chain means that the amino acid at the specific position is different from the amino acid naturally present in the IgG1 heavy chain.

[0294] In one embodiment, the amino acid at the position corresponding to K409 of the human IgG1 heavy chain in the first heavy chain is not K, L or M, and optionally the amino acid at the position corresponding to F405 of the human IgG1 heavy chain is F, and at least one amino acid at a position corresponding to a position selected from T366, L368, K370, D399, F405 and Y407 of the human IgG1 heavy chain is replaced in the second heavy chain.

[0295] In one embodiment, the amino acid at the position corresponding to K409 of the human IgG1 heavy chain in the first heavy chain is not K, L or M, and the amino acid at the position corresponding to F405 of the human IgG1 heavy chain in the second heavy chain is not F and optionally the amino acid at the position corresponding to K409 of the human IgG1 heavy chain is K.

[0296] In one embodiment, the amino acid at the position corresponding to F405 of the human IgG1 heavy chain in the first heavy chain is not F, R, or G, and the amino acids at the positions corresponding to the positions selected from T366, L368, K370, D399, Y407, and K409 of the human IgG1 heavy chain in the second heavy chain are replaced.

[0297] In one embodiment, the amino acid at the position corresponding to K409 of the human IgG1 heavy chain in the first heavy chain is not K, L, or M, and the amino acid at the position corresponding to F405 of the human IgG1 heavy chain is not F.

[0298] In a further embodiment, the amino acid at the position corresponding to F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to K409 of the human IgG1 heavy chain in the second heavy chain is R, or vice versa.

[0299] Thus, in one embodiment, the amino acid at the position corresponding to K409 of the human IgG1 heavy chain in the first heavy chain is R, and the amino acid at the position corresponding to F405 of the human IgG1 heavy chain in the second heavy chain is L.

[0300] In a further embodiment, the humanized or chimeric CD3 antibody of the present invention comprises one or more inactivating substitutions disclosed in any of the above embodiments, such as L234F, L235E, and D265A, in at least one of the first and second heavy chains; and the amino acid at the position corresponding to F405 is not F. In one embodiment, the humanized or chimeric CD3 antibody of the present invention comprises one or more inactivating substitutions disclosed in any of the above embodiments, such as L234F, L235E, and D265A, in at least one of the first and second heavy chains; and an additional substitution at the K409 position, such as K409R. Specifically, in one embodiment, the humanized or chimeric CD3 antibody of the present invention comprises one or more inactivating substitutions disclosed in any of the above embodiments, such as L234F, L235E, and D265A, in both the first and second heavy chains; and a substitution at the F405 position, such as F405L. In one embodiment, the humanized or chimeric CD3 antibody of the present invention comprises one or more inactivating substitutions disclosed in any of the above embodiments, such as L234F, L235E, and D265A, in both the first and second heavy chains; and an additional substitution at the K409 position, such as K409R. Such antibodies can be used to generate bispecific antibodies.

[0301] Thus, in a further embodiment, the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E and A respectively, the amino acid at the position corresponding to F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0302] In one embodiment, the amino acids at positions corresponding to positions L234, L235, D265, N297 and P331 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E, A, N and P respectively, the amino acid at the position corresponding to F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0303] In an alternative embodiment, the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E and A respectively, the amino acid at the position corresponding to K409 of the human IgG1 heavy chain in the first heavy chain is R, and the amino acid at the position corresponding to F405 of the human IgG1 heavy chain in the second heavy chain is L.

[0304] In one embodiment, the amino acids at positions corresponding to positions L234, L235, D265, N297 and P331 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E, A, N and P respectively, the amino acid at the position corresponding to K409 of the human IgG1 heavy chain in the first heavy chain is R, and the amino acid at the position corresponding to F405 of the human IgG1 heavy chain in the second heavy chain is L.

[0305] In another embodiment, the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in both the first and second heavy chains are F, E and A respectively, the amino acid at the position corresponding to F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0306] In one embodiment, the amino acids at positions corresponding to positions L234, L235, D265, N297 and P331 of the human IgG1 heavy chain in both the first and second heavy chains are F, E, A, N and P respectively, the amino acid at the position corresponding to F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0307] In an alternative embodiment, the amino acids at positions corresponding to human IgG1 heavy chain positions L234, L235, and D265 in both the first and second heavy chains are F, E, and A, respectively, the amino acid at the position corresponding to human IgG1 heavy chain K409 in the first heavy chain is R, and the amino acid at the position corresponding to human IgG1 heavy chain F405 in the second heavy chain is L.

[0308] In one embodiment, the amino acids at positions corresponding to human IgG1 heavy chain L234, L235, D265, N297, and P331 in both the first and second heavy chains are F, E, A, N, and P, respectively, the amino acid at the position corresponding to human IgG1 heavy chain K409 in the first heavy chain is R, and the amino acid at the position corresponding to human IgG1 heavy chain F405 in the second heavy chain is L.

[0309] As described herein, the recruitment of T cells to specific target cells, such as cancer or tumor cells, provides a method for killing the target cells. The inventors have demonstrated that a bispecific CD3xHER2 antibody containing specific amino acid substitutions L234F, L235E, and D265A in both heavy chains is capable of inducing the killing of AU565 cells, as described in Example 5. T cell-mediated killing can be obtained with a bispecific antibody having a first binding region targeting CD3 and a second binding region targeting another target. Thus, in one embodiment, the first binding region is as described herein for any embodiment of a humanized or chimeric CD3 antibody, and the second binding region binds a different target than the first binding region. It should be understood that when the antibody is a bispecific antibody, at least half of the antibody, i.e., one of the heavy chain and light chain pairs of the antibody, is a humanized or chimeric antibody as described herein. Thus, one half of the bispecific antibody is a humanized or chimeric antibody that binds CD3 according to the present invention and the other half can be a humanized, chimeric, fully non-human, or fully human antibody that binds a second target. Thus, in one embodiment, the antibody comprises a first and second heavy chain, and a first and second light chain, wherein the first heavy chain and the first light chain are humanized or chimeric and are linked by a disulfide bridge to form a first binding region; and the second heavy chain and light chain are fully human and are linked by a disulfide bridge to form a second binding region, wherein the first binding region is as described in any aspect or embodiment herein, and the second binding region binds a different target. In one embodiment, the antibody comprises a first and second heavy chain, and a first and second light chain, wherein the first heavy chain and the first light chain are humanized or chimeric and are linked by a disulfide bridge to form a first binding region; and the second heavy chain and light chain are humanized or chimeric and are linked by a disulfide bridge to form a second binding region, wherein the first binding region is according to the aspect or embodiment described herein, and the second binding region binds a different CD3 epitope than the first binding region.

[0310] Thus, in one embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:10; a second binding region comprising the VH sequence shown in SEQ ID NO:19 and the VL sequence shown in SEQ ID NO:20; wherein in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are F, E, and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0311] In one embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:10; a second binding region comprising the VH sequence shown in SEQ ID NO:29 and the VL sequence shown in SEQ ID NO:30; wherein in at least one of the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are F, E, and A, respectively; and wherein in the first heavy chain, the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0312] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:10; a second binding region comprising the VH sequence shown in SEQ ID NO:19 and the VL sequence shown in SEQ ID NO:20; wherein in both the first and second heavy chains, the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are F, E, and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0313] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:10; a second binding region comprising the VH sequence shown in SEQ ID NO:29 and the VL sequence shown in SEQ ID NO:30; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in both the first and second heavy chains are F, E and A respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0314] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:12; a second binding region comprising the VH sequence shown in SEQ ID NO:19 and the VL sequence shown in SEQ ID NO:20; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E and A respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0315] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:12; a second binding region comprising the VH sequence shown in SEQ ID NO:29 and the VL sequence shown in SEQ ID NO:30; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E and A respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0316] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:12; a second binding region comprising the VH sequence shown in SEQ ID NO:19 and the VL sequence shown in SEQ ID NO:20; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in both the first and second heavy chains are F, E and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0317] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:8 and the VL sequence shown in SEQ ID NO:12; a second binding region comprising the VH sequence shown in SEQ ID NO:29 and the VL sequence shown in SEQ ID NO:30; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in both the first and second heavy chains are F, E and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0318] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:10; a second binding region comprising the VH sequence shown in SEQ ID NO:19 and the VL sequence shown in SEQ ID NO:20; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0319] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:10; a second binding region comprising the VH sequence shown in SEQ ID NO:29 and the VL sequence shown in SEQ ID NO:30; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0320] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:10; a second binding region comprising the VH sequence shown in SEQ ID NO:19 and the VL sequence shown in SEQ ID NO:20; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in both the first and second heavy chains are F, E and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0321] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:10; a second binding region comprising the VH sequence shown in SEQ ID NO:29 and the VL sequence shown in SEQ ID NO:30; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in both the first and second heavy chains are F, E and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0322] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:12; a second binding region comprising the VH sequence shown in SEQ ID NO:19 and the VL sequence shown in SEQ ID NO:20; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E and A respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0323] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:12; a second binding region comprising the VH sequence shown in SEQ ID NO:29 and the VL sequence shown in SEQ ID NO:30; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E and A respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0324] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:12; a second binding region comprising the VH sequence shown in SEQ ID NO:19 and the VL sequence shown in SEQ ID NO:20; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in both the first and second heavy chains are F, E and A respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0325] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:6 and the VL sequence shown in SEQ ID NO:12; a second binding region comprising the VH sequence shown in SEQ ID NO:29 and the VL sequence shown in SEQ ID NO:30; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in both the first and second heavy chains are F, E and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0326] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:10; a second binding region comprising the VH sequence shown in SEQ ID NO:19 and the VL sequence shown in SEQ ID NO:20; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0327] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:10; a second binding region comprising the VH sequence shown in SEQ ID NO:29 and the VL sequence shown in SEQ ID NO:30; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0328] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:10; a second binding region comprising the VH sequence shown in SEQ ID NO:19 and the VL sequence shown in SEQ ID NO:20; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in both the first and second heavy chains are F, E and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0329] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:10; a second binding region comprising the VH sequence shown in SEQ ID NO:29 and the VL sequence shown in SEQ ID NO:30; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in both the first and second heavy chains are F, E and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0330] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:12; a second binding region comprising the VH sequence shown in SEQ ID NO:19 and the VL sequence shown in SEQ ID NO:20; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0331] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:12; a second binding region comprising the VH sequence shown in SEQ ID NO:29 and the VL sequence shown in SEQ ID NO:30; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0332] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:12; a second binding region comprising the VH sequence shown in SEQ ID NO:19 and the VL sequence shown in SEQ ID NO:20; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in both the first and second heavy chains are F, E and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0333] In another embodiment, the bispecific antibody comprises: a first binding region comprising the VH sequence shown in SEQ ID NO:9 and the VL sequence shown in SEQ ID NO:12; a second binding region comprising the VH sequence shown in SEQ ID NO:29 and the VL sequence shown in SEQ ID NO:30; wherein the amino acids at positions corresponding to positions L234, L235 and D265 of the human IgG1 heavy chain in both the first and second heavy chains are F, E and A, respectively; and wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R.

[0334] As used herein, the term "disulfide bridge" refers to a covalent bond between two cysteine residues, i.e., the interaction can also be referred to as a Cys-Cys interaction.

[0335] As used herein, the term "target" refers to a molecule to which the binding region of the antibody of the present invention binds. When used in the context of antibody binding, the term includes any antigen to which the resulting antibody is directed.

[0336] In a specific embodiment, the first heavy chain and the first light chain are humanized or chimeric and are linked by a disulfide bridge to form a first binding region; and the second heavy chain and light chain are fully human and are linked by a disulfide bridge to form a second binding region, wherein the first binding region is in accordance with any aspect or embodiment described herein, and the second binding region binds to a different target; and wherein the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E, and A, respectively.

[0337] In a specific embodiment, the first heavy chain and the first light chain are humanized or chimeric and are linked by a disulfide bridge to form a first binding region; and the second heavy chain and light chain are fully human and are linked by a disulfide bridge to form a second binding region, wherein the first binding region is in accordance with any aspect or embodiment described herein, and the second binding region binds to a different CD3 epitope than the first binding region; and wherein the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain in at least one of the first and second heavy chains are F, E, and A, respectively.

[0338] In a specific embodiment, the first heavy chain and the first light chain are humanized or chimeric and are linked by a disulfide bridge to form a first binding region; and the second heavy chain and light chain are fully human and are linked by a disulfide bridge to form a second binding region, wherein the first binding region is in accordance with any aspect or embodiment described herein, and the second binding region binds to a different target; and wherein the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain in both the first and second heavy chains are F, E, and A, respectively.

[0339] In a specific embodiment, the first heavy chain and the first light chain are humanized or chimeric and are linked by a disulfide bridge to form a first binding region; and the second heavy chain and light chain are fully human and are linked by a disulfide bridge to form a second binding region, wherein the first binding region is in accordance with any aspect or embodiment described herein, and the second binding region binds to a different CD3 epitope than the first binding region; and wherein the amino acids at positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain in both the first and second heavy chains are F, E, and A, respectively.

[0340] In one aspect, the bispecific antibody of the invention comprises a second binding region that binds to human HER2 or human CD20.

[0341] In one embodiment, the bispecific antibody of the invention comprises a second binding region that binds to human CD20.

[0342] In one embodiment, the second binding region may be derived from a full-length monoclonal CD20 antibody, such as the antibodies disclosed in WO2004035607 (Genmab A / S), including ofatumumab (2F2), 11B8, and 7D8; the antibodies disclosed in WO2005103081 (Genmab A / S), including 2C6; AME-133; TRU-015; IMMU-106; ocrelizumab tositumomab and rituximab

[0343] In one embodiment, the second binding region may be derived from a full-length monoclonal CD20 antibody that binds to an epitope on CD20 that does not include or require the amino acid residue alanine at position 170 or proline at position 172, but that includes or requires the amino acid residue asparagine at position 163 and asparagine at position 166.

[0344] In one embodiment, the bispecific antibody of the invention comprises a second binding region that binds to human CD20, the binding region comprising: (i) the VH CDR1 region of SEQ ID NO:34, the VH CDR2 region of SEQ ID NO:35, the VH CDR3 region of SEQ ID NO:36, the VL CDR1 region of SEQ ID NO:37, the VL CDR2 region of DAS, and the VL CDR3 region of SEQ ID NO:38, (ii) the VH CDR1 region of SEQ ID NO:41, the VH CDR2 region of SEQ ID NO:42, the VH CDR3 region of SEQ ID NO:43, the VL CDR1 region of SEQ ID NO:44, the VL CDR2 region of DAS, and the VL CDR3 region of SEQ ID NO:45, (iii) the VH CDR1 region of SEQ ID NO:48, the VH CDR2 region of SEQ ID NO:49, the VH CDR3 region of SEQ ID NO:50, the VL CDR1 region of SEQ ID NO:51, the VL CDR2 region of DAS, and the VL CDR3 region of SEQ ID NO:52, or (iv) the VH CDR1 region of SEQ ID NO:55, the VH CDR2 region of SEQ ID NO:56, the VH CDR3 region of SEQ ID NO:57, the VL CDR1 region of SEQ ID NO:58, the VL CDR2 region of DAS, and the VL CDR3 region of SEQ ID NO:59.

[0345] In one embodiment, the bispecific antibody of the present invention comprises a second binding region that binds to human CD20, the binding region comprising: (i) a VH sequence that has at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO: 29, and a VL sequence that has at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO: 30, (ii) a VH sequence that has at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO: 39, and a VL sequence that has at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO: 40, (iii) a VH sequence that has at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO: 46, and a VL sequence that has at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO: 47, or (iv) a VH sequence that has at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO: 53, and a VL sequence that has at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO: 54.

[0346] In one embodiment, the bispecific antibody of the present invention comprises a second binding region that binds to human CD20, the second binding region comprising: (i) the VH sequence of SEQ ID NO: 29 and the VL sequence of SEQ ID NO: 30, (ii) the VH sequence of SEQ ID NO: 39 and the VL sequence of SEQ ID NO: 40, (iii) the VH sequence of SEQ ID NO: 46 and the VL sequence of SEQ ID NO: 47, or (iv) the VH sequence of SEQ ID NO: 53 and the VL sequence of SEQ ID NO: 54.

[0347] Nucleic acid construct, expression vector and host cell In one aspect, the present invention relates to a nucleic acid construct encoding one or more of the sequences shown in Table 1. Accordingly, the present invention relates to a nucleic acid construct encoding any one of the sequences shown in SEQ ID NO: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25 and 26.

[0348] In a further aspect, the present invention relates to a nucleic acid construct encoding a sequence of a humanized or chimeric CD3 antibody of the present invention, an expression vector comprising the nucleic acid construct of the present invention, a host cell comprising the expression vector, and a method of producing the antibody by culturing the host cell under suitable conditions for producing and optionally recovering the antibody. The humanized CD3 antibody is also referred to as "huCD3".

[0349] In one embodiment, the present invention provides an expression vector comprising (i) a nucleic acid sequence encoding a heavy chain sequence of a humanized or chimeric antibody of the present invention, (ii) a nucleic acid sequence encoding a light chain sequence of a humanized or chimeric antibody of the present invention, or (iii) both (i) and (ii). Accordingly, the expression vector comprises one or more nucleic acid constructs or nucleic acid sequences of any aspect or embodiment described herein.

[0350] In one embodiment, the expression vector of the present invention comprises a nucleic acid sequence encoding one or more of the heavy and light chain CDR sequences selected from SEQ ID NO: 1, 2, 3, 4 and 5 and the sequence GTN.

[0351] In one embodiment, the present invention provides an expression vector comprising a nucleic acid sequence encoding one or more amino acid sequences selected from SEQ ID NO: 6, 7, 8, 9, 10, 11, 12, 19, 20, 27, 28, 29 and 30 or any combination thereof. In another embodiment, the expression vector comprises a nucleic acid sequence encoding the VHCDR3 amino acid sequence shown in SEQ ID NO: 3. In another embodiment, the expression vector comprises a nucleic acid sequence encoding a VH amino acid sequence selected from SEQ ID NO: 6, 7, 8, 9, 19, 27 and 29. In another embodiment, the expression vector comprises a nucleic acid sequence encoding a VL amino acid sequence selected from SEQ ID NO: 10, 11, 12, 20, 28 and 30. In another embodiment, the expression vector comprises a nucleic acid sequence encoding the constant region of a human antibody light chain, a human antibody heavy chain or both. In another embodiment, the present invention provides an expression vector comprising a nucleic acid sequence encoding the amino acid sequences according to SEQ ID NO: 15, 16, 23, 24, 25 and 26.

[0352] In a specific embodiment, the expression vector comprises a nucleic acid sequence encoding a variant of one or more of the above amino acid sequences, said variant having at most 25 amino acid modifications, such as at most 20, such as at most 15, 14, 13, 12 or 11 amino acid modifications, such as 10, 9, 8, 7, 6, 5, 4, 3, 2 or 1 amino acid modification, such as a deletion or insertion, preferably a substitution, such as a conservative or non-conservative substitution, or having at least 80% identity with any of said sequences, such as having at least 85% identity or 90% identity or 95% identity with any of the above amino acid sequences, such as 96% identity or 97% identity or 98% identity or 99% identity. The invention also relates to nucleic acid sequences that are different from the above nucleic acid sequences but encode the same amino acid sequences as the antibodies of the invention due to differences in the genetic code. For example, the nucleic acid sequences may be different, but produce the same amino acid sequences as any of the amino acid sequences described herein. How to identify such additional nucleic acid sequences based on the genetic code is well known to those skilled in the art.

[0353] In a further embodiment, the expression vector further comprises a nucleic acid sequence encoding the constant region of the light chain, heavy chain or both the light chain and heavy chain of an antibody (such as a human antibody).

[0354] Such an expression vector as described above can be used to recombinantly produce the antibodies of the invention.

[0355] In the context of the present invention, the expression vector can be any suitable vector, including chromosomal, non-chromosomal and synthetic nucleic acid vectors (nucleic acid sequences comprising a suitable set of expression control elements). Examples of such vectors include derivatives of SV40, bacterial plasmids, phage DNA, baculoviruses, yeast plasmids, vectors derived from combinations of plasmids and phage DNA, and viral nucleic acid (RNA or DNA) vectors. In one embodiment, the humanized or chimeric CD3 antibody-encoding nucleic acid is contained in a naked DNA or RNA vector, including for example linear expression elements (as described in for example

[64] ), compact nucleic acid vectors (as described in for example

[65] and / or

[66] ), plasmid vectors such as pBR322, pUC 19 / 18 or pUC 118 / 119, "midge" minimal-size nucleic acid vectors (as described in for example

[67] ), or as a precipitated nucleic acid vector construct, such as a CaPO4 - -precipitated construct (as described in for example

[68] ,

[69] ,

[70] and

[71] ). Such nucleic acid vectors and their uses are well known in the art (see for example

[72] and

[73] ).

[0356] In one embodiment, the vector is suitable for expressing a humanized or chimeric CD3 antibody in a bacterial cell. Examples of such vectors include expression vectors such as BlueScript (Stratagene), pIN vectors (

[74] ), pET vectors (Novagen, Madison WI), etc.

[0357] The expression vector may additionally or alternatively be a vector suitable for expression in a yeast system. Any vector suitable for expression in a yeast system can be used. Suitable vectors include, for example, vectors containing constitutive or inducible promoters such as alpha factor, alcohol oxidase, and PGH (reviewed in

[75] and

[76] ).

[0358] The nucleic acid construct and / or vector may also contain a nucleic acid sequence encoding a secretion / localization sequence, which can target a polypeptide (e.g., a nascent polypeptide chain) to the cytoplasmic space or the cell culture medium. Such sequences are known in the art and include secretion leader sequences or signal peptides, organelle targeting sequences (e.g., nuclear localization sequences, ER retention signals, mitochondrial transport sequences, chloroplast transport sequences), membrane localization / anchor sequences (e.g., stop transfer sequences, GPI anchor sequences), etc., which are well known in the art.

[0359] In the expression vectors of the present invention, the humanized or chimeric CD3 antibody-encoding nucleic acid may contain or be associated with any suitable promoter, enhancer, and other expression-promoting elements. Examples of such elements include strong expression promoters (e.g., the human CMV IE promoter / enhancer and RSV, SV40, SL3-3, MMTV, and HIV LTR promoters), efficient poly(A) termination sequences, origins of replication of plasmid products in E. coli, antibiotic resistance genes as selectable markers, and / or suitable cloning sites (e.g., polylinkers). The nucleic acid construct and / or vector may also contain an inducible promoter opposite to a constitutive promoter, such as CMV IE (those skilled in the art will know that these terms are actually descriptors of the degree of gene expression under certain conditions).

[0360] In one embodiment, the humanized or chimeric CD3 antibody-encoding expression vector is located in a host cell or a host animal, and / or is delivered to the host cell or the host animal by a viral vector.

[0361] Such expression vectors can be used for the recombinant production of humanized or chimeric CD3 antibodies.

[0362] In one aspect, the present invention provides a host cell containing the expression vector of the present invention.

[0363] In one aspect, the humanized or chimeric CD3 antibodies of any aspect or embodiment described herein are provided by using recombinant eukaryotic, recombinant prokaryotic, or recombinant microbial host cells that produce the antibodies. Accordingly, the present invention provides recombinant eukaryotic, recombinant prokaryotic, or recombinant microbial host cells that produce a humanized or chimeric CD3 antibody or immunoglobulin as defined herein. Examples of host cells include yeast, bacteria, and mammalian cells such as CHO or HEK-293 cells. For example, in one embodiment, the host cell contains a nucleic acid sequence stably integrated into the cell genome that contains a sequence encoding the expression of a humanized or chimeric CD3 antibody as described herein. In another embodiment, the host cell contains a non-integrated nucleic acid sequence such as a plasmid, cosmid, phagemid, or linear expression element that contains a sequence encoding the expression of a humanized or chimeric CD3 antibody as described herein.

[0364] As used herein, the term "recombinant host cell" (or simply "host cell") means a cell into which an expression vector or nucleic acid construct or sequence has been introduced. It is understood that the term not only refers to the specific cell being tested, but also to the progeny of that cell. Since certain modifications may occur during passage due to mutation or environmental influences, the progeny may in fact be different from the parental cell, but are still included within the scope of the term "host cell" as used herein. Recombinant host cells include, for example: eukaryotic host cells such as CHO cells, HEK-293 cells, PER.C6, NS0 cells, and lymphocytes; and prokaryotic cells such as Escherichia coli; and other eukaryotic hosts such as plant cells and fungi.

[0365] In a further aspect, the present invention relates to a method for producing a humanized or chimeric CD3 antibody of the present invention, the method comprising the steps of: a) culturing the host cell of the present invention as described above, and b) recovering and / or purifying the antibody of the present invention from the culture medium.

[0366] In a further aspect, the nucleotide sequence encoding the sequence of the humanized or chimeric CD3 antibody further encodes a second moiety, such as a therapeutic polypeptide. Exemplary therapeutic polypeptides are described elsewhere herein. In one embodiment, the present invention relates to a method for producing a humanized or chimeric CD3 antibody fusion protein, the method comprising the steps of: a) culturing a host cell that contains an expression vector containing the nucleotide sequence, and b) recovering and / or purifying the humanized or chimeric CD3 antibody fusion protein from the culture medium.

[0367] Composition In one aspect, the present invention provides a composition comprising an antibody or bispecific antibody of any aspect and embodiment described herein.

[0368] In one aspect, the present invention provides a pharmaceutical composition comprising an antibody or bispecific antibody defined in any of the aspects and embodiments described herein and a pharmaceutically acceptable carrier.

[0369] The pharmaceutical composition can be formulated with pharmaceutically acceptable carriers or diluents and any other known adjuvants and excipients according to conventional techniques, such as those described in

[77] .

[0370] The pharmaceutically acceptable carrier or diluent and any other known adjuvants and excipients should be suitable for the humanized or chimeric antibody of the present invention and the selected mode of administration. The suitability of the carrier and other components of the pharmaceutical composition is determined based on the absence of a significant negative impact on the biological properties required for the antigen-binding of the selected compound or pharmaceutical composition of the present invention (e.g., less than a substantial impact (10% or less relative inhibition, 5% or less relative inhibition, etc.)).

[0371] The pharmaceutical composition of the present invention may also include diluents, fillers, salts, buffers, detergents (e.g., nonionic detergents, such as Tween-20 or Tween-80), stabilizers (e.g., sugars or protein-free amino acids), preservatives, tissue fixatives, solubilizers, and / or other materials suitable for inclusion in the pharmaceutical composition.

[0372] The actual dosage level of the active ingredient in the pharmaceutical composition of the present invention can be varied so as to obtain an amount of the active ingredient that is effective in achieving the desired therapeutic response for a particular patient, composition, and mode of administration and that is non-toxic to the patient. The selected dosage level will depend on various pharmacokinetic factors well known in the medical art, including the activity of the specific composition or its amide of the present invention used, the route of administration, the frequency of administration, the excretion rate of the specific compound used, the duration of treatment; other drugs, compounds, and / or materials used in combination with the specific composition used; factors such as the age, sex, weight, medical condition, general health, and previous medical history of the patient being treated, etc.

[0373] The pharmaceutical composition can be administered by any suitable route and manner. Suitable routes for in vivo and in vitro administration of the humanized or chimeric antibody of the present invention are well known in the art and can be selected by one of ordinary skill in the art.

[0374] In one embodiment, the pharmaceutical composition of the present invention is administered parenterally.

[0375] As used herein, the phrases "parenterally administered" and "administered parenterally" mean a route of administration other than enteral and topical administration, typically by injection, and including epidermal, intravenous, intramuscular, intraarterial, intrathecal, intracapsular, intraorbital, intracardiac, intradermal, intraperitoneal, intratendinous, intratracheal, subcutaneous, subepidermal, intraarticular, subcapsular, subarachnoid, intraspinal, intracranial, intrathoracic, epidural, and intrasternal injection and infusion.

[0376] In one embodiment, the pharmaceutical composition is administered by intravenous or subcutaneous injection or infusion.

[0377] Pharmaceutically acceptable carriers include any and all suitable solvents, dispersion media, coating materials, antibacterial and antifungal agents, isotonic agents, antioxidants, and absorption delaying agents, etc., which are physiologically compatible with the humanized or chimeric antibodies of the present invention.

[0378] Examples of suitable aqueous and non-aqueous carriers for the pharmaceutical compositions of the present invention include water, saline, phosphate buffered saline, ethanol, glucose, polyols (such as glycerol, propylene glycol, polyethylene glycol, etc.), and suitable mixtures thereof; vegetable oils such as olive oil, corn oil, peanut oil, cottonseed oil, and sesame oil; carboxymethylcellulose colloidal solutions, tragacanth, and injectable organic esters such as ethyl oleate, and / or various buffers. Other carriers are well known in the pharmaceutical art.

[0379] Pharmaceutically acceptable carriers include sterile aqueous solutions or dispersions and sterile powders for the extemporaneous preparation of sterile injectable solutions or dispersions. The use of such media and reagents for pharmaceutically active substances is known in the art. Unless any conventional medium or reagent is incompatible with the active compound, its use in the pharmaceutical compositions of the present invention is contemplated. When referring to "active compound", the humanized or chimeric antibodies of the present invention are also contemplated.

[0380] Proper fluidity can be maintained, for example, by using coating materials such as lecithin, by maintaining the desired particle size in the case of dispersions, and by using surfactants.

[0381] The pharmaceutical compositions of the present invention may also contain pharmaceutically acceptable antioxidants, such as (1) water-soluble antioxidants, such as ascorbic acid, cysteine hydrochloride, sodium bisulfate, sodium metabisulfite, sodium sulfite, etc.; (2) oil-soluble antioxidants, such as ascorbyl palmitate, butylated hydroxyanisole (BHA), butylated hydroxytoluene (BHT), lecithin, propyl gallate, α-tocopherol, etc.; and (3) metal chelating agents, such as citric acid, ethylenediaminetetraacetic acid (EDTA), sorbitol, tartaric acid, phosphoric acid, etc.

[0382] The pharmaceutical compositions of the present invention may also contain isotonic agents in the composition, such as sugars, polyols, such as mannitol, sorbitol, glycerol, or sodium chloride.

[0383] The pharmaceutical composition of the present invention may further comprise one or more adjuvants suitable for the selected route of administration, such as preservatives, wetting agents, emulsifying agents, dispersing agents, preservatives or buffering agents, which may increase the half-life or efficacy of the pharmaceutical composition. The humanized or chimeric antibodies of the present invention can be prepared with carriers that protect the compound from rapid release, such as controlled-release formulations, including implants, transdermal patches, and microencapsulation delivery systems. Such carriers may include gelatin, glyceryl monostearate, glyceryl distearate, biodegradable biocompatible polymers such as ethylene vinyl acetate, polyanhydrides, polyglycolic acid, collagen, polyorthoesters, and polylactic acid alone or with waxes, or other materials well known in the art. Methods for preparing such formulations are generally known to those skilled in the art (see, for example,

[78] ).

[0384] In one embodiment, the humanized or chimeric antibodies of the present invention can be formulated to ensure proper in vivo distribution. Pharmaceutically acceptable carriers for parenteral administration include sterile aqueous solutions or dispersions and sterile powders for the extemporaneous preparation of sterile injectable solutions or dispersions. For pharmaceutically active substances, the use of such media and reagents is known in the art. Unless any conventional media or reagent is incompatible with the active compound, its use in the pharmaceutical compositions of the present invention is contemplated. Other active or therapeutic compounds may also be incorporated into the compositions.

[0385] The pharmaceutical compositions for injection must generally be sterile and stable under the conditions of preparation and storage. The compositions can be formulated as solutions, microemulsions, liposomes, or other ordered structures suitable for high drug concentrations. The carrier can be an aqueous or non-aqueous solvent or dispersion medium, including, for example, water, ethanol, polyols (such as glycerol, propylene glycol, polyethylene glycol, etc.), and suitable mixtures thereof; vegetable oils such as olive oil, and injectable organic esters such as ethyl oleate. Suitable fluidity can be maintained, for example, by using coating materials such as lecithin, by maintaining the required particle size in the case of dispersions, and by using surfactants. In many cases, it is preferred to include isotonic agents in the compositions, such as sugars, polyols such as glycerol, mannitol, sorbitol, or sodium chloride. Prolonged absorption of injectable compositions can be brought about by including agents that delay absorption, such as monostearates and gelatin, in the compositions. Sterile injectable solutions can be prepared by incorporating the required amount of the active compound into a suitable solvent, together with the required ingredients, for example, one or a combination of the ingredients listed above, followed by sterile microfiltration. Generally, dispersions are prepared by incorporating the active compound into a sterile vehicle containing a basic dispersion matrix and the required other ingredients, such as those listed above. In the case of sterile powders for the preparation of sterile injectable solutions, examples of the preparation methods are vacuum drying and freeze-drying (lyophilization), which yield a powder of the active ingredient plus any other required ingredients from its previously sterile-filtered solution.

[0386] A sterile injectable solution can be prepared by incorporating the required amount of the active compound into a suitable solvent with one or a combination of ingredients listed above as required, followed by sterile microfiltration. Generally, a dispersion is prepared by incorporating the active compound into a sterile vehicle comprising a base dispersion medium and the required other ingredients from those listed above. In the case of a sterile powder for the preparation of a sterile injectable solution, examples of the preparation method are vacuum drying and freeze-drying (lyophilization), which yield a powder of the active ingredient plus any other required ingredients from its previously sterile-filtered solution.

[0387] Therapeutic application In another aspect, the present invention relates to a humanized or chimeric antibody or pharmaceutical composition of the invention as defined in any aspect or embodiment described herein, which is used as a medicament.

[0388] In another aspect, the present invention relates to a humanized or chimeric antibody or pharmaceutical composition of the invention as defined in any aspect or embodiment described herein, which is used for the treatment of diseases.

[0389] The humanized or chimeric antibody or pharmaceutical composition of the present invention can be used for the treatment of any cancer in which the effector mechanisms of cytotoxic T cells are required. For example, the humanized or chimeric antibody can be administered to cells cultured in vitro or ex vivo, or to a human subject in vivo, for example, to treat or prevent conditions such as cancer, inflammatory or autoimmune disorders. The term "subject" as used herein generally refers to a human who responds to the humanized or chimeric antibody or pharmaceutical composition. The subject can include, for example, a human patient having a condition that can be corrected or improved by modulating a target function or by directly or indirectly causing cell death.

[0390] In another aspect, the present invention provides a method for treating or preventing a condition such as cancer, in which the recruitment of T cells will contribute to the treatment or prevention, the method comprising administering a therapeutically effective amount of the humanized or chimeric antibody or pharmaceutical composition of the present invention to a subject in need thereof. The method generally comprises administering to the subject an amount of the humanized or chimeric antibody effective for treating or preventing the condition.

[0391] In a specific aspect, the present invention relates to a method for treating cancer, comprising administering a humanized or chimeric antibody or pharmaceutical composition of the present invention as defined in any aspect and embodiment described herein to a subject in need thereof.

[0392] In another aspect, the invention relates to uses or methods as defined in any aspect or embodiment described herein, wherein the humanized or chimeric antibody is a bispecific antibody that specifically binds both CD3 and a cancer-specific target or a target that is overexpressed in or associated with cancer, such as HER2, CD19, EpCAM, EGFR, CD66e (or CEA, CEACAM5), CD33, EphA2, or MCSP (or HMW-MAA), and wherein the disease is cancer, such as breast cancer, prostate cancer, non-small cell lung cancer, bladder cancer, ovarian cancer, gastric cancer, colorectal cancer, esophageal cancer, and squamous cell carcinoma of the head and neck, cervical cancer, pancreatic cancer, testicular cancer, malignant melanoma, soft tissue cancer (such as synovial sarcoma), indolent or aggressive forms of B-cell lymphoma, chronic lymphocytic leukemia, or acute lymphocytic leukemia.

[0393] The effective dose and dosage regimen of the humanized or chimeric antibody depend on the disease or condition to be treated and can be determined by a person skilled in the art.

[0394] A physician of ordinary skill in the art can readily determine and prescribe the effective amount of the pharmaceutical composition required. For example, the physician can start with a dose of the humanized or chimeric antibody used in the pharmaceutical composition below the level required to achieve the desired therapeutic effect and gradually increase the dose until the desired effect is achieved. In general, a suitable dose of the composition of the invention is the amount of the humanized or chimeric antibody that is the lowest dose effective to produce a therapeutic effect according to a particular dosage regimen. Such an effective dose generally depends on the factors described above.

[0395] For example, an "effective amount" for therapeutic use can be measured by its ability to stabilize disease progression. The ability of a compound to inhibit cancer can be evaluated, for example, in an animal model system that predicts efficacy in human tumors. Alternatively, this property of the composition can be evaluated by in vitro assays known to those skilled in the art to examine the ability of the humanized or chimeric antibody to inhibit cell growth or induce cytotoxicity. A therapeutically effective amount of a therapeutic compound, i.e., the therapeutic humanized or chimeric antibody or pharmaceutical composition of the invention, can reduce tumor size or otherwise improve the symptoms of the subject. A person of ordinary skill in the art will be able to determine such an amount based on factors such as the size of the subject, the severity of the subject's symptoms, and the particular composition or route of administration selected.

[0396] Exemplary non-limiting ranges of therapeutically effective amounts of the humanized or chimeric antibodies of the invention are from about 0.001 - 30 mg / kg, such as about 0.001 - 20 mg / kg, such as about 0.001 - 10 mg / kg, such as about 0.001 - 5 mg / kg, such as about 0.001 - 2 mg / kg, such as about 0.001 - 1 mg / kg, such as about 0.001, about 0.01, about 0.1, about 1, about 5, about 8, about 10, about 12, about 15, about 18 mg / kg.

[0397] Administration can be, for example, intravenous, intramuscular, intraperitoneal or subcutaneous, and for example administered near the target site.

[0398] The dosage regimens in the above-described methods of treatment and uses are adjusted to provide the optimum desired response (e.g., a therapeutic response). For example, a single bolus can be administered, several divided doses can be administered over time, or the dose can be proportionally decreased or increased as indicated by the exigencies of the therapeutic situation.

[0399] In one embodiment, during treatment, for example at predetermined time points, the therapeutic efficacy is monitored.

[0400] If desired, the effective daily dose of the pharmaceutical composition can be administered as two, three, four, five, six or more sub-doses at appropriate time intervals throughout the day, optionally in unit dosage form. In another embodiment, the humanized or chimeric antibody or pharmaceutical composition is administered by slow continuous infusion over a long period, such as more than 24 hours, to minimize unwanted side effects.

[0401] Although it is possible to administer the humanized or chimeric antibodies of the invention alone, it is preferred to administer the humanized or chimeric antibodies as the pharmaceutical composition described above.

[0402] The effective dose of the humanized or chimeric antibodies of the invention can also be administered using once-weekly, once-every-two-week or once-every-three-week dosing schedules. The dosing schedule can be limited, for example, to 8 weeks, 12 weeks or until clinical progression is determined. Alternatively, the effective dose of the humanized or chimeric antibodies of the invention can be administered every other week, every two weeks or every three weeks.

[0403] In one embodiment, the humanized or chimeric antibody can be administered by infusion, in mg / m 2Calculated once-weekly dosing. Such a dose can be obtained, for example, based on the mg / kg dose provided above, according to the following: dose (mg / kg) x 70:1.8. Such dosing can be repeated, for example, 1 - 8 times, such as 3 - 5 times. The dosing can be by continuous infusion for a period of 2 - 24 hours, such as 2 - 12 hours. In one embodiment, the humanized or chimeric antibody can be administered by slow continuous infusion for a long period of time, such as more than 24 hours, to reduce toxic side effects.

[0404] In one embodiment, when administered once a week, the humanized or chimeric antibody can be administered as a calculated once-weekly dose as a fixed dose up to 8 times, such as 4 - 6 times. Such a regimen can be repeated one or more times as needed, for example, after 6 months or 12 months. Such a fixed dose can be, for example, based on the mg / kg dose provided above, with a body weight estimated at 70 kg. The dose can be determined or adjusted by measuring the amount of the humanized or chimeric antibody of the invention in the blood, for example, by obtaining a biological sample after administration and using an anti-idiotypic antibody targeting the binding region of the humanized or chimeric antibody of the invention.

[0405] In one embodiment, the humanized or chimeric antibody can be administered by maintenance therapy, such as once a week for a period of 6 months or more.

[0406] The humanized or chimeric antibody can also be administered prophylactically to reduce the risk of developing cancer, delay the onset of events in cancer progression, and / or reduce the risk of recurrence when cancer remits.

[0407] For ease of administration and uniformity of dosage, the parenteral composition can be formulated in unit dosage form. As used herein, unit dosage form refers to physically discrete units suitable as a single dose for a subject to be treated; each unit contains a predetermined quantity of the active compound calculated to produce the desired therapeutic effect and the required pharmaceutical carrier. The specification of the unit dosage form of the invention is indicated and directly depends on the following aspects: (a) the unique characteristics of the active compound and the particular therapeutic effect to be achieved, and (b) the limitations inherent in compounding such an active compound for treating individuals in the art.

[0408] The humanized or chimeric antibody can also be administered prophylactically to reduce the risk of developing cancer, delay the onset of events in cancer progression, and / or reduce the risk of recurrence when cancer remits. This may be particularly useful in patients in whom it is difficult to localize tumors known to be present due to other biological factors.

[0409] Diagnostic application Using a composition comprising the humanized or chimeric antibodies described herein, the humanized or chimeric antibodies of the present invention can also be used for diagnostic purposes. Accordingly, the present invention provides diagnostic methods and compositions using the humanized or chimeric antibodies described herein. Such methods and compositions can be used for purely diagnostic purposes, such as detecting or identifying a disease, and for monitoring the progress of a therapeutic treatment, monitoring disease progression, evaluating the post-treatment status, monitoring disease recurrence, evaluating the risk of developing a disease, etc.

[0410] In one aspect, the present invention relates to a method for diagnosing a disease characterized by the involvement or accumulation of CD3-expressing cells, comprising administering a humanized or chimeric antibody of the present invention, a composition of the present invention, or a pharmaceutical composition of the present invention to a subject, optionally wherein the humanized or chimeric antibody is labeled with a detectable reagent.

[0411] In one aspect, the humanized or chimeric antibodies of the present invention are used ex vivo, for example, for diagnosing a disease in which cells expressing a specific target and bound by the humanized or chimeric antibody indicate a disease or are involved in the pathogenesis by detecting the level of the target or the level of cells expressing the target of interest on their cell surface in a sample obtained from a patient. This can be achieved, for example, by contacting the test sample, optionally together with a control sample, with the humanized or chimeric antibody of the present invention under conditions that permit binding of the antibody to the target. Complex formation can then be detected (e.g., using ELISA). When using a control sample together with the test sample, the levels of the humanized or chimeric antibody or the antibody-target complex are analyzed in both samples, and a statistically significantly higher level of the humanized or chimeric antibody or the antibody-target complex in the test sample indicates a higher level of the target in the test sample compared to the control sample.

[0412] Examples of conventional immunoassays in which the humanized or chimeric antibodies of the present invention can be used include, but are not limited to, ELISA, RIA, FACS analysis, surface plasmon resonance analysis, chromatography, tissue immunohistochemistry, Western blotting, and / or immunoprecipitation.

[0413] Accordingly, in one embodiment, the present invention relates to a method for diagnosing a disease characterized by the involvement or accumulation of CD3-expressing cells, comprising administering an antibody, bispecific antibody, composition, or pharmaceutical composition of any aspect or embodiment described herein to a subject, optionally wherein the antibody is labeled with a detectable label.

[0414] In one embodiment, the present invention relates to a method for detecting the presence of a target or cells expressing the target in a sample, comprising: - contacting the sample with a humanized or chimeric antibody of the present invention under conditions that permit binding of the humanized or chimeric antibody to the target in the sample; and - analyzing whether a complex is formed. Generally, the sample is a biological sample.

[0415] In one embodiment, the sample is a tissue sample known or suspected to contain a specific target and / or cells expressing the target. For example, in situ detection of target expression can be achieved by removing a histological sample from a patient and providing a humanized or chimeric antibody of the present invention to such a sample. The humanized or chimeric antibody can be provided by applying or coating the humanized or chimeric antibody to the sample, and then detecting it using a suitable tool. Then, it may be possible to detect not only the presence of the target or cells expressing the target, but also the distribution of the target or cells expressing the target in the tissue under examination (e.g., in the case of evaluating the spread of cancer cells). Using the present invention, it will be readily understood by those of ordinary skill in the art that any one of a variety of histological methods (e.g., staining procedures) can be modified to achieve such in situ detection.

[0416] In the above assays, the humanized or chimeric antibody can be labeled with a detectable substance to allow the bound antibody to be detected. Alternatively, the bound (first) specific humanized or chimeric antibody can be detected by labeling with a detectable substance and an antibody that binds to the first specific humanized or chimeric antibody. In addition, in the above assays, a diagnostic composition comprising an antibody or bispecific antibody of any aspect or embodiment described herein can be used. Thus, in one aspect, the present invention relates to a diagnostic composition comprising an antibody or bispecific antibody of any aspect or embodiment described herein.

[0417] The target level in a sample can also be evaluated by a competitive immunoassay using a target standard labeled with a detectable substance and an unlabeled target-specific humanized or chimeric antibody. In this type of assay, a biological sample, the labeled target standard, and the target-specific humanized or chimeric antibody are combined, and the amount of the labeled target standard that binds to the unlabeled target-specific humanized or chimeric antibody is measured. The amount of the target in the biological sample is inversely proportional to the amount of the labeled target standard that binds to the target-specific humanized or chimeric antibody.

[0418] Suitable labels for target-specific humanized or chimeric antibodies, second antibodies, and / or target standards for in vitro diagnostic techniques include, but are not limited to, various enzymes, cofactors, fluorescent materials, luminescent materials, and radioactive materials. Examples of suitable enzymes include horseradish peroxidase, alkaline phosphatase, β-galactosidase, and acetylcholinesterase; examples of suitable cofactor complexes include streptavidin / biotin and avidin / biotin; examples of suitable fluorescent materials include umbelliferone, fluorescein, fluorescein isothiocyanate, rhodamine, dichlorotriazinylamine fluorescein, dansyl chloride, and phycoerythrin; examples of luminescent materials include luminol; and examples of suitable radioactive materials include 125 I, 131 I, 35 S and 3 H.

[0419] In one aspect, the target-specific humanized or chimeric antibodies of the invention are used for in vivo imaging of tissues expressing the target, such as tumors. For in vivo methods, antibody fragments such as (Fab’)2, Fab and Fab’ fragments are particularly advantageous because of their rapid distribution kinetics.

[0420] In vivo imaging can be carried out by any suitable technique. For example, 99 Tc, 131 I, 111 In or other target-specific humanized or chimeric antibodies (such as antibodies or fragments) labeled with γ-ray-emitting isotopes can be used to image the accumulation or distribution of target-specific antibodies in tissues expressing the target, such as tumors, using a γ-scintillation camera (such as an Elscint Apex 409 ECT device), typically using a low-energy high-resolution collimator or a low-energy general collimator. Alternatively, 89 Zr, 76 Br, 18 F or other positron-emitting radionuclide labeling can be used in positron emission tomography (PET) in tumors to image the distribution of target-specific humanized or chimeric antibodies or antibody fragments. Images obtained by using such techniques can be used to evaluate the biodistribution of the target in a patient, mammal or tissue, for example in the case of using the target as a biomarker for the presence of cancer / tumor cells. Variations of this technique can include the use of magnetic resonance imaging (MRI) to improve imaging by γ-camera techniques. Conventional immunoscintigraphy methods and principles are described in, for example,

[79] ,

[80] and

[81] . In addition, such images can also or alternatively be used as a basis for surgical techniques for removing tumors. In addition, such in vivo imaging techniques can allow the identification and localization of tumors in cases where a patient is identified as having a tumor (due to the presence of other biomarkers, metastases, etc.) but the tumor cannot be identified by conventional analytical techniques. All of these methods are features of the invention.

[0421] The in vivo imaging and other diagnostic methods provided by the invention can be particularly used for detecting micrometastases in human patients (e.g., patients who have not been previously diagnosed with cancer or patients in cancer recovery / remission).

[0422] In one embodiment, the present invention provides a method of in vivo imaging, wherein a target-specific humanized or chimeric antibody of the present invention is conjugated to a radiopaque reagent that facilitates detection. The conjugated humanized or chimeric antibody is administered to a host, such as by injection into the bloodstream, and the presence and location of the labeled humanized or chimeric antibody in the host are analyzed. By this technique and any other diagnostic method provided herein, the present invention provides a method for screening for the presence of disease-related cells in a human patient or a biological sample obtained from a human patient and / or a method for evaluating the distribution of a target-specific humanized or chimeric antibody prior to target-specific ADC therapy.

[0423] For diagnostic imaging, a radioisotope can be directly or indirectly bound to a target-specific humanized or chimeric antibody, for example, by using an intermediate functional group. Useful intermediate functional groups include chelating agents such as ethylenediaminetetraacetic acid and diethylenetriaminepentaacetic acid (see, for example,

[82] ).

[0424] In addition to radioisotopes and radiopaque reagents, diagnostic methods can be performed using target-specific antibodies conjugated to dyes (e.g., to biotin-streptavidin complexes), contrast agents, fluorescent compounds or molecules, and enhancers for magnetic resonance imaging (MRI) (e.g., paramagnetic ions) (see, for example,

[83] , which describes MRI techniques and the preparation of antibodies conjugated to MRI enhancers). Such diagnostic / detection reagents can be selected from reagents for MRI and fluorescent compounds. To load a target-specific humanized or chimeric antibody with a radioactive metal or paramagnetic ion, it may be necessary to react it with a reagent having a long tail, to which multiple chelating groups are attached to bind the ion. Such a tail can be a polymer such as polylysine, a glycan, or another derivatized or derivative chain having pendant groups to which chelating groups can be attached, such as porphyrin, polyamine, crown ether, dithiosemicarbazone, polyoxime, and other groups known for this purpose. The chelating agent can be coupled to the target-specific humanized or chimeric antibody using standard chemistry.

[0425] Accordingly, the present invention provides diagnostic target-specific humanized or chimeric antibodies, wherein the target-specific humanized or chimeric antibody is conjugated to a contrast agent (e.g., for magnetic resonance imaging, computed tomography, or ultrasound contrast enhancer) or a radionuclide, which can be, for example, a γ-, β-, α-, Auger electron-, or positron-emitting isotope.

[0426] In one aspect, the present invention relates to a diagnostic composition comprising an antibody or bispecific antibody of the present invention.

[0427] In a further aspect, the present invention relates to a kit for detecting the presence of a target antigen or cells expressing a target in a sample, comprising: - a target-specific humanized or chimeric antibody of the present invention; and - Instructions for use of the kit.

[0428] Thus, in one aspect, the present invention provides a kit for detecting the presence of CD3 antigen or CD3-expressing cells in a sample, comprising the following steps: a) contacting the sample with the antibody or bispecific antibody of the present invention under conditions that permit the formation of a complex between the antibody or bispecific antibody and CD3; and b) analyzing whether a complex is formed.

[0429] In one embodiment, the present invention provides a kit for diagnosing cancer, which comprises a container containing a target-specific humanized or chimeric antibody, and one or more reagents for detecting the binding of the target-specific humanized or chimeric antibody to the target. The reagents can include, for example, a fluorescent tag, an enzyme tag or other detectable tags. The reagents can also include a second or third antibody or reagents for an enzymatic reaction, wherein the enzymatic reaction produces a product that can be visualized. In one embodiment, the present invention provides a diagnostic kit, which comprises one or more target-specific humanized or chimeric antibodies of the present invention in labeled or unlabeled form in a suitable container, reagents for incubation in an indirect assay, and a substrate or derivatized reagent (depending on the nature of the tag) for detection in such an assay. Control reagents and instructions for use can also be included.

[0430] Diagnostic kits can also be provided for use with target-specific humanized or chimeric antibodies, such as labeled target-specific antibodies, for detecting the presence of a target in a tissue sample or host. In such diagnostic kits and in the kits for therapeutic use described elsewhere herein, the target-specific humanized or chimeric antibodies can typically be provided in a container in lyophilized form, either alone or in combination with additional antibodies specific for target cells or peptides. Typically, a pharmaceutically acceptable carrier (e.g., an inert diluent) and / or its components, such as Tris, phosphate or carbonate buffers, stabilizers, preservatives, biocides, inert proteins such as serum albumin, etc. (usually in separate containers for mixing) and additional reagents (also usually in separate containers) can also be included. In certain kits, a second antibody capable of binding to the target-specific humanized or chimeric antibody is also included, which is typically present in a separate container. The second antibody is typically conjugated to a tag and formulated in a manner similar to the target-specific humanized or chimeric antibody of the present invention. Using the methods described above and elsewhere herein, the target-specific humanized or chimeric antibodies can be used to define subsets of cancer / tumor cells and to characterize such cells and associated tumor tissue.

[0431] Anti-idiotypic antibody In a further aspect, the present invention relates to anti-idiotype antibodies that bind to the humanized or chimeric antibodies of the present invention described herein.

[0432] An anti-idiotype (Id) antibody is an antibody that recognizes unique determinants usually associated with the antigen-binding site of an antibody. Anti-Id antibodies can be prepared by immunizing an animal of the same species and genotype as the source of the anti-CD3 monoclonal antibody with the monoclonal antibody against which the anti-Id is to be prepared. The immunized animal can typically recognize and respond to the idiotypic determinants of the immunizing antibody by producing antibodies (anti-Id antibodies) against these idiotypic determinants. Such antibodies are described, for example, in US4,699,880. Such antibodies are an additional feature of the present invention.

[0433] Anti-Id antibodies can also be used as an "immunogen" to induce an immune response in another animal, generating so-called anti-anti-Id antibodies. The anti-anti-Id antibodies can be identical to the original monoclonal antibody that induced the anti-Id antibodies in terms of epitope. Thus, by using antibodies against the idiotypic determinants of a monoclonal antibody, it is possible to identify other clones of antibodies that express the same specificity. By any suitable technique, such as those described elsewhere herein for the CD3-specific antibodies of the present invention, anti-Id antibodies can be altered (thereby generating anti-Id antibody variants) and / or derivatized. For example, a monoclonal anti-Id antibody can be conjugated to a carrier such as keyhole hemocyanin (KLH) and used to immunize BALB / c mice. The sera from these mice typically contain anti-anti-Id antibodies that have binding properties similar to (if not identical to) the original / parental CD3 antibody.

[0434] Sequence Table 1 CDR regions are annotated according to IMGT definitions. Examples

[0435] Example 1 – Generation of Humanized CD3 Antibodies and Non-Activating Antibody Variants Humanization of CD3 antibody Humanization of the murine CD3 antibody (US8,236,308, described herein as IgG1-CD3) was performed by Antitope (Cambridge, UK) using its improved form of germline humanization (CDR-grafting) technology (EP0629240). Using this technology, four different VH chains (SEQ ID NO: 6, 7, 8, and 9) and three different VL chains (SEQ ID NO: 10, 11, and 12) were designed. By combining these four VH and three VL chains, twelve different antibodies were generated. The humanized variants are described herein as huCD3. Thus, the humanized variants of the present invention comprising VH and VL are described, for example, as IgG1-huCD3-H1L1, which means that the specific variant has an IgG1 isotype, is a humanized CD3, and comprises a VH amino acid sequence designated "H1" and defined according to SEQ ID NO: 6 and a VL amino acid sequence designated "L1" and defined according to SEQ ID NO: 10. Thus, H1 refers to the variable heavy chain region VH1, L1 refers to the variable light chain region VL1, and so on.

[0436] Specifically, variants IgG1-huCD3-H1L1 (humanized CD3 comprising the VH1 sequence shown in SEQ ID NO: 6 and the VL1 sequence shown in SEQ ID NO: 10), IgG1-huCD3-H1L2 (humanized CD3 comprising the VH1 sequence shown in SEQ ID NO: 6 and the VL2 sequence shown in SEQ ID NO: 11), IgG1-huCD3-H1L3 (humanized CD3 comprising the VH1 sequence shown in SEQ ID NO: 6 and the VL3 sequence shown in SEQ ID NO: 12), IgG1-huCD3-H3L3 (humanized CD3 comprising the VH3 sequence shown in SEQ ID NO: 8 and the VL3 sequence shown in SEQ ID NO: 12), IgG1-huCD3-H4L1 (humanized CD3 comprising the VH4 sequence shown in SEQ ID NO: 9 and the VL1 sequence shown in SEQ ID NO: 10), IgG1-huCD3-H3L1 (humanized CD3 comprising the VH3 sequence shown in SEQ ID NO: 8 and the VL1 sequence shown in SEQ ID NO: 10), IgG1-huCD3-H3L3 (humanized CD3 comprising the VH3 sequence shown in SEQ ID NO: 8 and the VL3 sequence shown in SEQ ID NO: 12), and IgG1-huCD3-H4L3 (humanized CD3 comprising the VH4 sequence shown in SEQ ID NO: 9 and the VL3 sequence shown in SEQ ID NO: 12) were generated and tested in the examples described herein.

[0437] In some embodiments, an antibody comprising the heavy and light chain variable region sequences of huCLB-T3 / 4 (SEQ ID NO: 17 and 18, respectively) is used as a control antibody (Labrijn et al., PNAS 2013, 110:5145-50) and to validate different non-activating mutation combinations in the Fc region (see Examples 8-10). huCBL-T3 / 4 is a humanized form of the murine CD3 antibody CLB-T3 / 4 (Parren et al., Res Immunol. 1991, 142(9):749-63). The two sequences (SEQ ID NO: 17 and 18) were cloned into the relevant pcDNA3.3 (Invitrogen) expression vector and expressed in HEK293F cells by co-transfection. The resulting control antibody is designated IgG1-huCLB-T3 / 4.

[0438] In some embodiments, an antibody comprising the heavy and light chain variable region sequences of the CD20 antibody 7D8 (SEQ ID NO: 29 corresponding to the VH sequence and SEQ ID NO: 30 corresponding to the VL sequence) is used as a positive control. When used in the context of a positive control, it is designated "IgG1-CD20".

[0439] These IgG1-CD3 (i.e., the chimeric parental CD3 antibody), IgG1-huCD3, and IgG1-huCLB-T3 / 4 antibodies are used in both monospecific and bispecific forms, where the bispecific antibodies are generated as described below.

[0440] HER2 antibody In some embodiments, an antibody against HER2 is used. The VH and VL sequences for this HER2-specific antibody (antibody 169, SEQ ID NO: 19 and 20, respectively) have been previously described (WO2012 / 143524 [Genmab]; Labrijn et al., PNAS 2013, 110:5145-50). This antibody is used in both monospecific and bispecific forms and is designated "IgG1-HER2".

[0441] b12 antibody In some embodiments, the antibody b12, a gp120-specific antibody (Barbas, CF. J Mol Biol. 1993 Apr 5; 230(3):812-23.) is used as a negative control and is designated "IgG1-b12".

[0442] Expression The antibody is expressed as IgG1, κ or IgG1, λ, with or without the non-activating mutations described below and having mutations in the CH3 domain capable of generating bispecific antibodies by the methods described below: IgG1-HER2-K409R, IgG1-b12-K409R, IgG1-CD3-F405L. A plasmid DNA mixture encoding both the heavy and light chains of the antibody was transiently transfected into Freestyle HEK293F cells (Invitrogen, US) using 293fectin (Invitrogen, US) essentially as described by the manufacturer.

[0443] Purification of antibody The culture supernatant was filtered through a 0.2 μm sterile filter and loaded onto a 5 mL MabSelect SuRe column (GE Health Care) and eluted with 0.1 M sodium citrate-NaOH, pH 3. The eluate was immediately neutralized with 2 M Tris-HCl, pH 9 and dialyzed overnight against 12.6 mM NaH2PO4, 140 mM NaCl, pH 7.4 (B. Braun). Alternatively, after purification, the eluate was loaded onto a HiPrep desalting column and the antibody was exchanged into 12.6 mM NaH2PO4, 140 mM NaCl, pH 7.4 (B. Braun) buffer. After dialysis or buffer exchange, the sample was sterile filtered through a 0.2 μm sterile filter. Purity was determined by SDS-PAGE and concentration was measured by absorbance at 280 nm. The purified antibody was stored at 2 - 8 °C.

[0444] Production of bispecific antibody Use Platform technology, namely 2-MEA-induced Fab-arm exchange, as described in WO 2011 / 147986 and Labrijn et al. (Labrijn et al., PNAS 2013, 110:5145-50; Gramer et al., MAbs 2013, 5:962-973), generates bispecific antibodies in vitro. To be able to generate bispecific antibodies by this method, IgG1 molecules with a single mutation in the CH3 domain are generated: the F405L mutation in one parental IgG1 antibody (i.e., the IgG1-CD3 antibody), and the K409R mutation in the other parental IgG1 antibody (i.e., the HER2 or b12 antibody). To generate bispecific antibodies, these two parental antibodies, each at a final concentration of 0.5 mg / mL, are incubated in 500 μL TE with 25 or 75 mM 2-mercaptoethylamine-HCl (2-MEA) at 31 °C for 5 h. The reduction reaction is terminated when the reducing agent 2-MEA is removed by using a PD-10 column (GE-healthcare, product #17-0851-01) equilibrated with 25 mL PBS. Before desalting, 2 mL PBS (B. Braun, product number 3623140) is added to the sample to adjust the volume to 2.5 mL. Elution is performed in 3.5 mL PBS. Samples are collected in an Amicon Ultra centrifugal device (30 kDa MWCO, Millipore, product #UFC803096) and concentrated by centrifugation at 3000 x g for 8 min. The volume is adjusted to 500 μL with PBS (when needed) and the sample is sterile filtered through a 0.2 μm filter (Millex-GV, product #SLGV004SL). The bispecific product is stored at 2-8 °C.

[0445] In an alternative way of obtaining the same bispecific antibody, to generate the bispecific antibody, 100 μg of the two parental antibodies were mixed and incubated with 75 mM 2-mercaptoethylamine-HCl (2-MEA) in a total volume of 400 μL PBS (B. Braun, product number 3623140) at 31 °C for 5 hours. The reduction reaction was terminated when the reducing agent 2-MEA was removed by using an Amicon Ultra 0.5 ml centrifugal device (10 kD MWCO, Millipore, product number UFC501096) and washing 4 times with 400 μl PBS by centrifugation at 13,000 x g for 10 min. The sample was collected in a new tube by inverting the filter and centrifuging at 1000 g for 2 min. The volume was adjusted to 200 μL with PBS (when needed). The absorbance at 280 nm (A280) of the bispecific product was measured to determine the final concentration. HPLC cation exchange chromatography (HPLC-CEX) (described in WO 2013 / 060867) was performed to determine the amount of the bispecific product. The samples were stored at 2-8 °C.

[0446] The generated bispecific antibodies are described hereinafter as "K409R IgG1 backbone" and "F405L IgG1 backbone".

[0447] Non-activated mutation Several antibody variants were generated by substituting one or more amino acids in the Fc region. The non-activated Fc region prevents the antibody from interacting with Fc-receptors present on blood cells (such as monocytes) or from interacting with C1q to activate the classical complement pathway. The reduction of Fc activity was tested in antibody variants containing different combinations of amino acid substitutions in the Fc region. Up to 5 amino acid substitutions were introduced, including the mutations N297Q, L234A, L235A, L234F, L235E, D265A and P331S. Substitutions in one or more of these 5 amino acid positions were introduced into the K409R and / or F405L IgG1 backbone. The following Fc region variants of the huCLB-T3 / 4 antibody were generated: N297Q (referring to the N297Q substitution, called IgG1-huCLB-T3 / 4-N297Q), LFLE (referring to the L234F / L235E substitution, called IgG1-huCLB-T3 / 4-LFLE), LALA (referring to the L234A / L235A substitution, called IgG1-huCLB-T3 / 4-LALA), LFLENQ (referring to the L234F / L235E / N297Q substitution, called IgG1-huCLB-T3 / 4-LFLENQ), LFLEDA (referring to the L234F / L235E / D265A substitution, called IgG1-huCLB-T3 / 4-LFLEDA), DA (referring to the D265A substitution, called IgG1-huCLB-T3 / 4-DA), DAPS (referring to the D265A / P331S substitution, called IgG1-huCLB-T3 / 4-DAPS), DANQ (referring to the D265A / N297Q substitution, called IgG1-huCLB-T3 / 4-DANQ), LFLEPS (referring to the L234F / L235E / P331S substitution, called IgG1-huCLB-T3 / 4-LFLEPS) and LFLEDANQPS (referring to the L234F / L235E / D265A / N297Q / P331S substitution, called IgG1-huCLB-T3 / 4-LFLEDANQPS).

[0448] Specifically, in the IgG1-huCD3 antibody variant, a combination of 3 amino acid substitutions, including the mutations L234F, L235E and D265A and called LFLEDA, was introduced into the K409R and F405L IgG1 backbone to generate an antibody with a non-activated Fc region. The resulting non-activated antibody variant was named with the suffix “-LFLEDA”.

[0449] Example 2 - Binding of humanized CD3 antibodies and their non-activated variants to human and cynomolgus monkey T cell lines expressing CD3 Purified variants of the humanized CD3 (huCD3) antibody and bispecific (bs) IgG1-huCD3 x HER2 molecules (see Example 1), with or without the LFLEDA mutation in the Fc region, were analyzed for their binding to the human T cell line Jurkat (Clone E6-1, TIB-152 TM , LGC Standards GmbH, Wesel, Germany) or the cynomolgus monkey T cell line HSC-F (cat. no. JCRB1164; Health Science Research Resources Bank, Osaka, Japan) by FACS analysis. In addition to the non-activating mutation LFLEDA, the antibody variants contained the F405L or K409R mutation as described in Example 1.

[0450] Cells (1x10 5 cells / well) were incubated with antibody preparations serially diluted (range 5 - 10,000 ng / mL, in 3-fold dilutions) in 100 μL of PBS / 0.1% BSA / 0.02% azide on polystyrene 96-well round bottom plates (Greiner bio-one 650101) at 4 °C for 30 min.

[0451] After washing twice with PBS / 0.1% BSA / 0.02% azide, the cells were incubated with the secondary antibody in 100 μL at 4 °C for 30 min. As the secondary antibody, R-phycoerythrin (PE)-conjugated goat-anti-human IgG F(ab’)2 (109-116-098, Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) diluted 1 / 100 in PBS / 0.1% BSA / 0.02% azide was used for all experiments. Subsequently, the cells were washed twice in PBS / 0.1% BSA / 0.02% azide, resuspended in 150 μL of PBS / 0.1% BSA / 0.02% azide and analyzed on a FACSCantoll (BD Biosciences). Binding curves were analyzed using nonlinear regression (sigmoidal dose response, with variable slope) using GraphPad Prism V5.04 software (GraphPad Software, San Diego, CA, USA).

[0452] Figure 1A shows the binding of IgG1λ-huCD3 variants IgG1-huCD3-H1L1 (SEQ ID NOs: 6 and 10 respectively), IgG1-huCD3-H1L2 (SEQ ID NOs: 6 and 11 respectively), IgG1-huCD3-H1L3 (SEQ ID NOs: 6 and 12 respectively), IgG1-huCD3-H3L3 (SEQ ID NOs: 8 and 12 respectively), and IgG1-huCD3-H4L1 (SEQ ID NOs: 9 and 10 respectively) with wild-type Fc regions to Jurkat cells was observed for all variants, and the binding ability of IgG1-CD3-LFLEDA (the parental CD3 antibody described in Example 1 with a non-activated LFLEDA mutation) and IgG1-huCD3-H3L1-LFLEDA with a non-activated LFLEDA mutation was similar to that of huCD3 variants with wild-type Fc regions. The binding of IgG1-huCLB-T3 / 4 included as a positive control to Jurkat cells was stronger compared to the IgG1-huCD3 variants. No binding was observed for the negative control antibody IgG1-b12.

[0453] Figure 6 A shows that the binding of IgG1-CD3-LFLEDA (the parental CD3 antibody described in Example 1 with a non-activated LFLEDA mutation), IgG1-huCD3-H3L1-LFLEDA, IgG1-huCD3-H3L3-LFLEDA, IgG1-3huCD3-H1L1-LFLEDA, IgG1-huCD3-H1L3-LFLEDA, IgG1-huCD3-H4L1-LFLEDA, and IgG1-huCD3-H4L3-LFLEDA with non-activated LFLEDA mutations to Jurkat cells was similar to the binding ability of huCD3 variants with wild-type Fc regions. The binding of IgG1-huCLB-T3 / 4 included as a positive control to Jurkat cells was stronger at low antibody concentrations compared to the IgG1-huCD3 variants, but similar at higher antibody concentrations. Overall, the humanized CD3 variants retained the binding ability to CD3 as IgG1-CD3 antibodies. No binding was observed for the negative control antibody IgG1-b12.

[0454] Figure 1Panel B shows that the bispecific antibody variants bsIgG1CD3xHER2, bsIgG1CD3xb12-LFLEDA, and bsIgG1huCD3-H3L1xHER2-LFLEDA also bind to Jurkat cells. The maximum binding values of these bispecific antibodies are higher than those of the monospecific antibodies. The EC50 concentrations of the bispecific antibodies are 6 - 10-fold. Again, no binding was observed for the negative control antibody IgG1-b12.

[0455] Figure 6 Panel B shows that the bispecific non-activated Fc antibody variants bsIgG1huCD3-H3L1xHER2-LFLEDA, bsIgG1-huCD3-H3L3xHER2-LFLEDA, bsIgG1-huCD3-H1L1xHER2-LFLEDA, bsIgG1-huCD3-H1L3xHER2-LFLEDA, bsIgG1-huCD3-H4L1xHER2-LFLEDA, and bsIgG1-huCD3-H4L3xHER2-LFLEDA also bind to Jurkat cells. The maximum binding values of these bispecific antibodies are higher than those of the monospecific antibodies. The EC50 concentrations of the bispecific antibodies are 4 - 10-fold. Monovalent binding allows more antibodies to aggregate on the cell surface, and thus, allows higher binding values of the bispecific antibodies. Again, no binding was observed for the negative control antibody IgG1-b12.

[0456] Figure 2 Panel A shows that the binding of IgG1-huCD3 variants IgG1-huCD3-H1L1, IgG1-huCD3-H1L2, IgG1-huCD3-H1L3, IgG1-huCD3-H3L3, and IgG1-huCD3-H4L1, and IgG1-CD3-LFLEDA, IgG1-huCD3-H3L1-LFLEDA with wild-type Fc regions to the cynomolgus monkey T cell line HSC-F is similar. No binding was observed for the control antibody huCLB-T3 / 4 that does not cross-react with cynomolgus monkey CD3, and for the negative control antibody IgG1-b12.

[0457] Figure 7A shows that the binding of IgG1-huCD3 variants IgG1-CD3-LFLEDA, IgG1-huCD3-H3L1-LFLEDA, IgG1-huCD3-H3L3-LFLEDA, IgG1-huCD3-H1L1-LFLEDA, IgG1-huCD3-H1L3-LFLEDA, IgG1-huCD3-H4L1-LFLEDA, and IgG1-huCD3-H4L3-LFLEDA to cynomolgus monkey T cell line HSC-F is similar. No binding was observed for the negative control antibody IgG1-b12.

[0458] Figure 2 B shows that the bispecific antibody variants bsIgG1CD3xHER2 and bsIgG1huCD3-H3L1-LFLEDA also bind to HSC-F cells. The maximum binding values of these bispecific antibodies are higher than those of the monospecific anti-CD3 variants. The EC50 concentration of the bispecific antibodies is 10 - 12 times that of the monospecific anti-CD3 antibodies. Again, no binding was observed for the negative control antibody IgG1-b12.

[0459] Figure 7 B shows that the bispecific non-activated Fc antibody variants bsIgG1huCD3-H3L1xHER2-LFLEDA, bsIgG1-huCD3-H3L3xHER2-LFLEDA, bsIgG1-huCD3-H1L1xHER2-LFLEDA, bsIgG1-huCD3-H1L3xHER2-LFLEDA, bsIgG1-huCD3-H4L1xHER2-LFLEDA, and bsIgG1-huCD3-H4L3xHER2-LFLEDA also bind to HSC-F cells. The maximum binding values of these bispecific antibodies are higher than those of the monospecific anti-CD3 variants. The EC50 concentration of the bispecific antibodies is 3 - 6 times that of the monospecific anti-huCD3 antibodies. Again, no binding was observed for the negative control antibody IgG1-b12.

[0460] Example 3 - T cell activation by humanized CD3 antibody variants CD69 expression is an early marker of T cell activation. Through the binding of the Fc region of an antibody such as the IgG1 Fc region via CD3 expressed on T cells and the Fc receptor expressed on immune cells, a CD3 antibody can mediate the crosslinking of T cells and immune cells. This can result in T cell activation and induction of CD69. Antibody variants containing a non-activated Fc region (LFLEDA mutation) do not bind the Fc receptor. Thus, it is expected that a non-activated CD3 antibody does not induce T cell activation and CD69 expression because the non-activated Fc region does not bind immune cells expressing the Fc receptor and thus cannot crosslink T cells and immune cells.

[0461] After incubation with humanized CD3 (huCD3) variants containing and not containing the LFLEDA mutation in the Fc region, CD69 expression on T cells was evaluated by FACS analysis to determine early activation of T cells. In addition to the non-activated mutation, the LFLEDA variant also contains the F405L or K409R mutation as described in Example 1.

[0462] PBMC were isolated from whole blood or buffy coat by density gradient separation using Leucosep tubes (#227290; Greiner Bio-one, Alphen a / d Rijn, The Netherlands), washed with PBS and resuspended in medium.

[0463] A dose-response series of huCD3 antibody variants, negative control (IgG1-b12), and positive controls (IgE-huCD3 and parental IgG1-CD3) were prepared in medium (range 0.1 - 1,000 ng / mL, 10-fold dilutions) and added to each well of a 96-well round-bottom plate containing human or cynomolgus monkey PBMC. After 16 - 24 hours of incubation, the cells were pelleted by centrifugation, and the supernatant (containing cytokines) was collected and stored at -20 °C. The cells were then washed with PBS / 0.1% BSA / 0.02% azide and stained for 30 minutes at 4 °C with mouse-anti-human CD28-PE (854.222.010; Sanquin, Amsterdam, The Netherlands; T cell marker) and mouse-anti-human CD69-APC antibody (340560; BD Biosciences, Franklin Lakes, NJ) which cross-react with cynomolgus monkey CD28 and CD69, respectively. Unbound antibody was removed by washing twice with PBS / 0.1% BSA / 0.02% azide. The cells were resuspended at 150 μL / well, and CD69 expression on CD28-positive cells was measured on a FACSCanto II (BD Biosciences).

[0464] Figure 3 Show that IgG1-CD3 with wild-type IgG1 Fc region (as described in Example 1) and humanized IgG1-huCD3 variants induce similar levels of CD69 expression on T cells from human Figure 3 A) and cynomolgus monkeys Figure 3 B). Non-activated (LFLEDA) IgG1-CD3-LFLEDA and IgG1-huCD3-H3L1 variants induce low levels of CD69 expression in human T cells. Non-activated IgG1-huCD3 variants do not induce CD69 expression in cynomolgus monkey T cells. Control antibody IgG1-b12 also does not induce CD69 expression in human or cynomolgus monkey T cells.

[0465] Figure 8 Show that non-activated (LFLEDA) IgG1-huCD3-H3L1-LFLEDA, IgG1-huCD3-H3L3-LFLEDA, IgG1-3huCD3-H1L1-LFLEDA, IgG1-huCD3-H1L3-LFLEDA, IgG1-huCD3-H4L1-LFLEDA and IgG1-huCD3-H4L3-LFLEDA variants induce low levels of CD69 expression in human T cells. Figure 8 A and 8B show induction of CD69 expression on T cells from cynomolgus monkeys. The less activation observed with non-activated variants may be due to cross-linking of CD3 molecules resulting from bivalent binding of the CD3 antibody. Such an explanation is supported by the observation that activation is reduced at the highest concentrations, where antibody binding is monovalent. Control antibody IgG1-b12 also does not induce CD69 expression in human or cynomolgus monkey T cells.

[0466] Figure 8 C and 8D show that non-activated bispecific antibody variants bsIgG1-huCD3-H3L1 x HER2-LFLEDA, bsIgG1-huCD3-H3L3 x HER2-LFLEDA, bsIgG1-huCD3-H1L1 x HER2-LFLEDA, bsIgG1-huCD3-H1L3 x HER2-LFLEDA, bsIgG1-huCD3-H4L1 x HER2-LFLEDA and bsIgG1-huCD3-H4L3 x HER2-LFLEDA do not induce CD69 expression in T cells from human Figure 8 C) or cynomolgus monkeys Figure 8 D). However, some induction of CD69 expression was observed at higher antibody concentrations.

[0467] Example 4 - T-cell proliferation induced by humanized CD3 antibody variants The effect of the humanized CD3 (huCD3) antibody variants (described in Example 1) on the proliferation of human and cynomolgus monkey T cells was evaluated by a cell proliferation ELISA kit from Roche Applied Science (Cell Proliferation ELISA, BrdU kit, #11647229001; Roche Applied Science, Mannheim, Germany), which was performed according to the manufacturer's instructions.

[0468] Human or cynomolgus monkey PBMCs isolated from whole blood or buffy coat were incubated in 96-well culture plates with a dilution series (range 0.1 - 1,000 ng / mL, 10-fold dilutions) of the IgG1 huCD3 antibody variants. IgG1-CD3 and IgG1-huCLB-T3 / 4 were included as positive controls and IgG1-b12 as a negative control. After 3 days of incubation with the antibody, BrdU (Roche Applied Science, Mannheim, Germany) was added to the medium, and the plates were incubated for 5 hours. The cells were then pelleted by centrifugation, and the supernatants were collected and stored at -20 °C. The plates were dried and stored at 4 °C until ELISA was performed.

[0469] BrdU incorporation into DNA was determined by ELISA according to the manufacturer's instructions (Roche Applied Science). The cells were fixed to the plates, and then the plates were incubated with an anti-BrdU antibody conjugated to peroxidase for 90 minutes at room temperature. The plates were washed with PBST and detection of binding was performed using ABTS buffer (instead of the TMB solution provided in the kit). Color development was terminated after 30 min by adding 2% oxalic acid to each well. The OD405 nm was then measured on an EL808 ELISA-reader.

[0470] Figure 4 Incubation of PBMCs with the parental IgG1-CD3 and the humanized IgG1-huCD3 variants with wild-type IgG1 Fc regions was shown to induce strong proliferation of human ( Figure 4 A) and cynomolgus monkey ( Figure 4 B) T cells, even at very low concentrations of the antibody. Incubation with the non-activated LFLEDA variant of the IgG1-huCD3 antibody did not induce proliferation of human T cells ( Figure 4 A and 9A) or cynomolgus monkey T cells ( Figure 4 B and 9B). Thus, although the non-activated variants of the IgG1-huCD3 antibody induced low levels of CD69 expression in human T cells (as shown in Example 3), proliferation of human T cells was not induced by these non-activated IgG1-huCD3 variants.

[0471] Figure 9 C and 9D show that the non-activated bispecific antibody variants bsIgG1-huCD3-H3L1 x HER2-LFLEDA, bsIgG1-huCD3-H3L3 x HER2-LFLEDA, bsIgG1-huCD3-H1L1 x HER2-LFLEDA, bsIgG1-huCD3-H1L3xHER2-LFLEDA, bsIgG1-huCD3-H4L1xHER2-LFLEDA, and bsIgG1-huCD3-H4L3xHER2-LFLEDA do not induce the proliferation of T cells isolated from human Figure 9 C) or cynomolgus monkeys Figure 9 D).

[0472] Example 5 - In Vitro T Cell-Mediated Cytotoxicity Induced by Humanized CD3 Antibody Variants Tumor-specific T cell cytotoxicity can be mediated by bispecific antibodies that bind CD3 with one arm and a tumor-specific target such as HER2 with the other arm. Binding of the bispecific antibody to both T cells and tumor cells simultaneously will result in T cell activation and tumor cell-specific cytotoxicity. In this example, T cell-mediated cytotoxicity against HER2-positive tumor cells was evaluated using bispecific antibodies against CD3 (humanized variant) and HER2.

[0473] Thus, AU565 (human breast cancer) cells were cultured in RPMI 1640 supplemented with 10% (vol / vol) heat-inactivated CCS, 1.5 g / L sodium bicarbonate (Lonza), 1 mM sodium pyruvate, 4.5 g / L glucose (Sigma), 50 IU / mL penicillin, and 50 μg / mL streptomycin. The cell line was maintained at 37 °C in a 5% (vol / vol) CO2 humidified incubator. AU565 cells were cultured until near confluence, after which the cells were subjected to trypsin treatment, resuspended in medium, and a single-cell suspension was obtained by passing through a cell filter. 5x10 4 cells were seeded into each well of a 96-well culture plate, and the cells were incubated at 37 °C, 5% CO2 for at least 3 hours to allow attachment to the plate.

[0474] Human or cynomolgus monkey PBMCs were isolated from whole blood or buffy coat. The isolated PBMCs were washed with PBS, resuspended in medium and added to AU565 tumor cells in a 96-well plate at a 1:1 ratio. The percentage of T cells present in the PBMCs was measured by FACS-analysis using mouse anti-human CD3-PerCP (BD, #345766) antibody (for staining T cells), which cross-reacts with cynomolgus monkey CD3. The T cell content in the PBMC population used was typically 50-60%.

[0475] Dilution series (final concentration range 0.001 up to 10,000 ng / mL) of bispecific antibody variants bsIgG1CD3xHER2-LFLEDA, bsIgG1CD3xb12-LFLEDA, bsIgG1huCD3-H3L1xHER2-LFLEDA, bsIgG1-huCD3-H3L3xHER2-LFLEDA, bsIgG1-huCD3-H1L1xHER2-LFLEDA, bsIgG1-huCD3-H1L3xHER2-LFLEDA, bsIgG1-huCD3-H4L1xHER2-LFLEDA and bsIgG1-huCD3-H4L3xHER2-LFLEDA were prepared in medium and added to the plates. IgG1-HER2-LFLEDA and IgG1-b12 were included as controls. Except for non-activating mutations, the LFLEDA antibody variants contain the F405L or K409R mutation for preparation in bispecific form (see Example 1). The plates were incubated at 37 °C, 5% CO2 for 3 days. The cells were incubated with 1 μM staurosporine (#S6942-200, Sigma) and used as a reference for 100% tumor cell kill. The plates were washed twice with PBS and 150 μL medium containing 10% Alamar blue was added to each well. The plates were incubated at 37 °C, 5% CO2 for 4 hours. Absorbance at 590 nm was measured (Envision, Perkin Elmer, Waltham, MA).

[0476] Bispecific CD3xHER2-LFLEDA antibody variants (bsIgG1-huCLB-T3 / 4xHER2-LFLEDA and bsIgG1-CD3xHER2-LFLEDA) induced killing of AU565 cells at low concentrations using human effector cells ( Figure 5 A) or cynomolgus monkey effector cells ( Figure 5 B). The CD3 bispecific control antibody huCLB-T3 / 4xHER2-LFLEDA, which shows no cross-reactivity with cynomolgus monkey CD3, induced killing of AU565 cells only when using human PBMCs ( Figure 5A). Thus, no killing of target cells was observed when cynomolgus effector cells were used in this assay ( Figure 5 B). Incubation with monospecific IGG1-b12 or IgG1-HER2-LFLEDA or bsIgG1-CD3xb12-LFLEDA antibodies did not induce non-specific killing of target cells.

[0477] The bispecific antibody variants bsIgG1 huCD3-H3L1 x HER2-LFLEDA, bsIgG1-huCD3-H3L3xHER2-LFLEDA, bsIgG1-huCD3-H1L1 x HER2-LFLEDA, bsIgG1-huCD3-H1L3 x HER2-LFLEDA, bsIgG1-huCD3-H4L1 x HER2-LFLEDA and bsIgG1-huCD3-H4L3 x HER2-LFLEDA induced killing of AU565 cells when using human effector cells ( Figure 10 A) or cynomolgus effector cells ( Figure 10 B). Incubation with monospecific IgG1-b12 or IgG1-HER2-LFLEDA antibodies did not induce non-specific target cell killing ( Figure 10 A and B). Thus, a humanized CD3 variant containing a non-activated Fc region did not induce non-specific target cell killing, indicating that variants containing a non-activated Fc region can be used to ensure targeted T cell activation and thus avoid non-targeted T cell activation.

[0478] Example 6 – Activation of Rhesus T Cells by Humanized CD3 Antibody Variants After incubation with humanized CD3 (huCD3) antibody variants having a wild-type IgG1 Fc region, CD69 expression on rhesus T cells was evaluated to determine early activation of T cells. Isolation of rhesus PBMC and evaluation of CD69 expression by flow cytometry were performed as described in Example 3.

[0479] Figure 11 It was shown that the humanized CD3 antibody variants IgG1-huCD3-H1L1, IgG1-huCD3-H1L2, IgG1-huCD3-H1L3, IgG1-huCD3-H3L3 and IgG1-huCD3-H4L1 induced CD69 expression on rhesus-derived T cells to levels similar to IgG1-CD3 (as described in Example 1). The negative control antibody IgG1-b12 did not induce CD69 expression in rhesus T cells. Thus, the huCD3 variants of the present invention can be used in experiments involving rhesus CD3. The huCD3 variants cross-react with rhesus CD3.

[0480] Example 7 - T cell activation by non-activated variants of huCLB-T3 / 4 After incubation with IgG1-huCLB-T3 / 4 variants with Fc region mutations (see Example 1), CD69 expression on T cells was evaluated by FACS analysis to determine early activation of T cells.

[0481] PBMC were isolated from whole blood or buffy coat by density gradient separation using Leucosep tubes (#227290; Greiner Bio-one, Alphen a / d Rijn, The Netherlands), washed with PBS and resuspended in medium.

[0482] A dose response series of IgG1-huCLB-T3 / 4 variants, negative control (IgG1-huCLB-T3 / 4-Fab) and positive control (IgE-huCLB-T3 / 4) were prepared in medium (range 1 - 1000 ng / mL, 3-fold dilutions), and added to each well of a 96-well round-bottom plate containing PBMC. After 16 - 24 hours of incubation, the cells were pelleted by centrifugation, and the supernatant (containing cytokines) was collected and stored at -20 °C. The cells were then washed with PBS / 0.1% BSA / 0.02% azide, and stained at 4 °C with mouse-anti-human CD28-PE (854.222.010; Sanquin, Amsterdam, The Netherlands; T cell marker) and mouse-anti-human CD69-APC antibody (340560; BD Biosciences, Franklin Lakes, NJ) for 30 minutes. Unbound antibody was removed by washing twice with PBS / 0.1% BSA / 0.02% azide. The cells were resuspended in 150 μL / well and CD69-expression on CD28-positive cells was measured on a FACS Canto II (BD Biosciences).

[0483] Figure 12A shows high CD69 expression on cells incubated with IgE-huCLB-T3 / 4, IgG1-huCLB-T3 / 4, IgG1-huCLB-T3 / 4-DA, and IgG1-huCLB-T3 / 4-DAPS. Incubation with IgG1-huCLB-T3 / 4-N297Q induced slightly lower CD69 expression levels compared to wild-type IgG1-huCLB-T3 / 4, and incubation with IgG1-huCLB-T3 / 4-LFLE and IgG1-huCLB-T3 / 4-LFLEPS induced CD69 to a lesser extent. Incubation of PBMC with IgG1-CD3 Fab, IgG1-huCLB-T3 / 4-LFLEDA, IgG1-huCLB-T3 / 4-LFLENQ, IgG1-huCLB-T3 / 4-DANQ, and IgG1-huCLB-T3 / 4-LFLEDANQPS antibodies did not induce any expression of CD69 on T cells.

[0484] Figure 12 B shows high CD69 expression on cells incubated with IgE-huCLB-T3 / 4 and IgG1-huCLB-T3 / 4. Incubation with IgG1-huCLB-T3 / 4-LALA induced slightly lower CD69 expression levels compared to wild-type IgG1-huCLB-T3 / 4, and incubation with IgG1-huCLB-T3 / 4-LFLEDA and IgG1-b12 (negative control) did not induce any expression of CD69 on T cells.

[0485] Example 8 – T-cell proliferation by non-activating variants of huCLB-T3 / 4 The effect of huCLB-T3 / 4 variants (described in Example 1) on T-cell proliferation was evaluated using a cell proliferation ELISA kit from Roche Applied Science (Cell Proliferation ELISA, BrdU kit, #11647229001; Roche Applied Science, Mannheim, Germany), which was performed according to the manufacturer's instructions.

[0486] PBMC isolated from whole blood or buffy coat were incubated in 96-well culture plates with a dilution series (range 0.1 - 1000 ng / mL) of IgG1-CD3 variants. IgG1-CD3 and IgE-CD3 were included as positive controls and IgG1-b12 (with the K409R mutation for generating bispecific antibodies) as negative control. After 3 days of incubation with the antibodies, BrdU (Roche Applied Science, Mannheim, Germany) was added to the medium and the plates were incubated for 5 hours. The cells were then pelleted by centrifugation, and the supernatants were collected and stored at -20 °C. The plates were dried and stored at 4 °C until ELISA was performed.

[0487] BrdU incorporation into DNA was determined by ELISA according to the manufacturer's instructions (Cell Proliferation ELISA, BrdU Kit, #11647229001; Roche Applied Science). The cells were fixed to the plates, after which the plates were incubated with anti-BrdU antibody conjugated to peroxidase for 90 minutes at room temperature (RT). The plates were washed with PBST, and binding was measured using ABTS buffer (instead of the TMB solution provided in the kit). Color development was terminated after 30 min by adding 2% oxalic acid to each well. The OD405 nm was then measured on an EL808 ELISA-reader.

[0488] Figure 13 A shows that incubation of PBMC with IgG1-huCLB-T3 / 4, IgG1-huCLB-T3 / 4-DA and IgG1-huCLB-T3 / 4-DAPS induces strong proliferation of T cells, even at very low concentrations of the antibodies. Incubation with IgG1-huCLB-T3 / 4-N297Q induces dose-dependent proliferation, which is comparable to the IgE-huCLB-T3 / 4 positive control. Incubation of PBMC with IgG1-huCLB-T3 / 4-Fab, IgG1-b12-N297Q, IgG1-huCLB-T3 / 4-LFLE, IgG1-huCLB-T3 / 4-LFLEDA, IgG1-huCLB-T3 / 4-LFLENQ, IgG1-huCLB-T3 / 4-LFLEPS, IgG1-huCLB-T3 / 4-DANQ and IgG1-huCLB-T3 / 4-LFLEDANQPS antibodies does not induce proliferation of T cells.

[0489] Figure 13Panel B shows strong proliferation of T cells induced by incubation of PBMC with IgG1-huCLB-T3 / 4, even at very low concentrations of the antibody. Incubation with IgE-huCLB-T3 / 4 (positive control) and IgG1-huCLB-T3 / 4-LALA induced dose-dependent proliferation. Incubation of PBMC with IgG1-huCLB-T3 / 4-LFLEDA did not induce proliferation of T cells.

[0490] Based on the results of Examples 7 and 8, one subgroup of mutants that were considered to be minimally activating was further analyzed.

[0491] Example 9 – In vitro T cell-mediated cytotoxicity induced by non-activating antibody variants huCLB-T3 / 4 AU565 (human breast cancer) cells were incubated in RPMI 1640 supplemented with 10% (vol / vol) heat-inactivated CCS, 1.5 g / L sodium bicarbonate (Lonza), 1 mM sodium pyruvate, 4.5 g / L glucose (Sigma), 50 IU / mL penicillin and 50 μg / mL streptomycin. The cell line was maintained at 37 °C in a 5% (vol / vol) CO2 humidified incubator. AU565 cells were cultured to near confluence. The cells were subjected to trypsin treatment, resuspended in medium and a single cell suspension was obtained by passing through a cell filter. 5x10 4 cells were seeded into each well of a 96-well culture plate, and the cells were incubated at 37 °C, 5% CO2 for at least 3 hours to allow attachment to the plate.

[0492] Peripheral blood mononuclear cells (PBMC) were isolated from the blood of healthy volunteers using Leucosep 30 mL tubes according to the manufacturer's protocol (Greiner Bio-one). The isolated PBMC were washed with PBS, resuspended in medium and added to the AU565 tumor cells in the 96-well plate at a 1:1 ratio. The percentage of T cells present in the PBMC was measured by FACS-analysis using a mouse anti-human CD3-PerCP (BD, #345766) antibody (for staining T cells). The T cell content in the PBMC population used was typically 50 - 60%.

[0493] IgG1-b12, IgG1-huCLB-T3 / 4, IgG1-HER2, and bispecific huCLB-T3 / 4xb12 and huCLB-T3 / 4xHER2 antibodies in dilution series (final concentration range 0.004 - 1000 ng / mL), expressed as different Fc-variants, wild-type, N297Q, LFLE, LALA, LFLENQ, LFLEDA, DANQ, and LFLEDENQPS, were prepared in medium and added to plates. The plates were incubated at 37 °C, 5% CO2 for 3 days. Cells were incubated with 1 μM staurosporine (#S6942-200, Sigma) and used as a reference for 100% tumor cell kill. After incubation, the supernatant was removed and stored at -20 °C. The plates were washed twice with PBS and 150 μL of medium containing 10% Alamar blue was added to each well. The plates were incubated at 37 °C, 5% CO2 for 4 hours. Absorbance at 590 nm was measured (Envision, Perkin Elmer, Waltham, MA).

[0494] Two experiments were performed using PBMC from different donors. In the first experiment, Fc-variants N297Q, LFLE, LFLENQ, LFLEDA, DANQ, and LFLEDANQPS ( Figure 14 A - G) were tested. In the second experiment, Fc-variants LFLEDA and LALA ( Figure 15 A - C) were tested. Antibodies with wild-type Fc-domains were included as a reference in both experiments. Incubation with wild-type monospecific IgG1-huCLB-T3 / 4 or bispecific huCLB-T3 / 4xb12 antibodies induced non-specific killing of target cells ( Figure 14 A - G and 15A - C). Monospecific IgG1-huCLB-T3 / 4 and bsIgG1-huCLB-T3 / 4xb12 variants N297Q ( Figure 14 A - G) and LALA ( Figure 15 A - C) still induced some non-specific target cell killing, although to a lesser extent than the wild-type antibodies tested in the same experiment. Non-specific target cell killing was not induced by any of the other tested IgG1-huCLB-T3 / 4 or bsIgG1-huCLB-T3 / 4xb12 antibodies with non-activating mutations ( Figure 14 A - G and 15A - C).

[0495] All bispecific huCLB-T3 / 4xHER2 antibodies induced dose-dependent killing of AU565 cells with at least comparable efficacy compared to wild-type bispecific huCLB-T3 / 4xHER2 antibodies without non-activating mutations ( Figure 14A-G and 15A-C). Maximum killing occurs at very low concentrations.

[0496] Wild-type or non-activated variants of monospecific b12 or HER2 antibodies do not induce cytotoxicity ( Figure 14 A-G and 15A-C), as expected.

[0497] Example 10 – Evaluation of the binding of C1q to non-activated antibody variants of huCLB-T3 / 4 The interaction of C1q with antibodies bound to target cells is the first step in the classical pathway of complement activation. Since wild-type IgG1 contains the interaction site for C1q, the interaction of C1q with these non-activated IgG1 variants was evaluated by ELISA.

[0498] At 4 °C, dilution series (range 7 - 30,000 ng / mL, 4-fold dilutions) of IgG1-huCLB-T3 / 4, bsIgG1-huCLB-T3 / 4xHER2 and IgG1-CD20 (positive control) and their non-activated antibody variants as described in Example 1 above were coated overnight on 96-well Microlon ELISA plates (Greiner, Germany). The plates were washed and blocked with PBS supplemented with 0.025% Tween 20 and 0.1% gelatin. The plates were sequentially incubated with 3% pooled human serum (Sanquin, product# M0008) for 1 h at 37 °C, with 100 μL / well rabbit anti-human C1q (DAKO, product# A0136, 1 / 4,000) for 1 h at room temperature, and with 100 μL / well swine anti-rabbit IgG-HRP (DAKO, P0399, 1:10,000) as the detection antibody for 1 h at room temperature, with washing between incubations. Detection was performed by adding 1 mg / mL 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS; Roche, Mannheim, Germany) for approximately 30 min. The reaction was terminated by adding 100 μL 2% oxalic acid. Absorbance at 405 nm was measured in a microplate reader (Biotek, Winooski, VT). Log-transformed data were analyzed by fitting a sigmoidal dose-response curve with variable slope using GraphPad Prism software.

[0499] Figure 16A shows that the antibodies IgG1-CD20 and IgG1-huCLB-T3 / 4 with wild-type IgG1 Fc regions show C1q binding. No C1q binding was detected in all evaluated antibody variants with non-activating mutations (N297Q, LFLE, LFLENQ, LFLEDA, DA, DAPS, DANQ, LFLEPS, LFLEDANQPS, LALA).

[0500] Figure 16 B shows that the antibody bsIgG1-huCLB-T3 / 4xHER2 with wild-type IgG1 Fc region shows C1q binding. No C1q binding was detected in all evaluated antibody variants with non-activating mutations (N297Q, LFLE, LFLENQ, LFLEDA, DA, DAPS, DANQ, LFLEPS, LFLEDANQPS, LALA).

[0501] Figure 16 C and Figure 16 D show that the antibodies IgG1-CD20, IgG1-huCLB-T3 / 4, and bsIgG1-huCLB-T3 / 4xHER2 with wild-type IgG1 Fc regions show C1q binding. No C1q binding was detected in the antibody variants with non-activating mutations (LFLEDA and LALA).

[0502] Example 11 – Pharmacokinetic (PK) analysis of non-activating antibody variants The mice in this study were housed in isolation units at the Central Laboratory Animal Facility (Utrecht, The Netherlands) and kept in top-filtered cages with unrestricted access to water and food. All experiments were approved by the Animal Ethics Committee of Utrecht University. C.B-17 SCID mice, 7–10 weeks old (C.B-17 / Icr-Prkdc <scid>Intravenous injection of 100 μg of wild-type antibodies (IgG1-huCLB-T3 / 4, IgG1-HER2, or bsIgG-huCLB-T3 / 4xHER2) or their non-activated variants (LALA, LFLEDA, LFLENQ, DANQ, or LFLEDANQPS) (IcrIcoCrl, Charles-River), using 3 mice / group. At 10 minutes, 4 hours, 1 day, 2 days, 7 days, 14 days, and 21 days after antibody administration, 50 μL of blood samples were collected from the saphenous vein. Blood was collected into heparinized vials and centrifuged at 10,000 x g for 5 minutes. Plasma was stored at -20 °C until antibody concentration determination.

[0503] Human IgG concentration was determined using a total hIgG sandwich ELISA. For this assay, mouse mAb anti-human IgG-kappa clone MH16 (#M1268, CLB Sanquin, The Netherlands), coated at a concentration of 2 μg / mL on 96-well Microlon ELISA plates (Greiner, Germany), was used as the capture antibody. After blocking the plates with PBS supplemented with 0.2% bovine serum albumin, samples serially diluted with ELISA buffer (PBS supplemented with 0.05% Tween 20 and 0.2% bovine serum albumin) were added and incubated for 1 h at room temperature (RT) on a plate shaker. Subsequently, the plates were incubated with goat anti-human IgG immunoglobulin (#109-035-098, Jackson, West Grace, PA), and developed with 2,2'-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS; Roche, Mannheim, Germany). The reaction was terminated after 30 min by adding 2% oxalic acid to each well. Absorbance at 405 nm was measured in a microplate reader (Biotek, Winooski, VT).

[0504] Plasma clearance rate (mL / day / kg) was calculated based on the area under the curve (AUC) according to the following equation: Data analysis was performed using Graphpad prism software.

[0505] Figure 17 A shows that when compared to wild-type antibodies, for antibody variants N297Q, DANQ, LFLENQ, and LFLEDANQPS, plasma human IgG concentrations were lower, indicating a faster clearance rate. For antibody variants LFLEDA and LALA, plasma human IgG concentrations were similar to wild-type antibodies.

[0506] Figure 17 Panel B shows that the plasma clearance rates of the antibody variants N297Q, DANQ, and LFLENQ are 2- to 3-fold that of the wild-type antibody. The clearance rate of the antibody variant LFLEDANQPS is 3- to 5-fold that of the wild-type antibody. The plasma clearance rates of the antibody variants LFLEDA and LALA are similar to that of the wild-type antibody.

[0507] Example 12 – In Vitro Immunogenicity Evaluation of IgG1-LFLEDA Scaffold To determine the potential clinical immunogenicity of the IgG1-LFLEDA-K409R scaffold, Antitope's EpiScreen TM platform was applied to IgG1-HER2-LFLEDA. Briefly, PBMCs were isolated from a panel of 50 HLA-typed healthy donors representing the European and North American populations. After CD8+ T cell depletion, the PBMC populations were cryopreserved and stored individually. Subsequently, the thawed PBMCs were cultured and incubated with either IgG1-HER2-LFLEDA-K409R or one of the control samples (IgG1-HER2 or IgG1-HER2-LFLE-K409R) for 5-8 days. The ability of the samples to induce CD4+ T cell responses was evaluated by measuring cell proliferation ([3 H -thymidine incorporation) and IL-2 production (ELISpot assay). Donors showing responses with a stimulation index (SI; signal / baseline signal) ≥ 1.9 in both assays were considered positive.

[0508] EpiScreen TM analysis showed that for IgG1-HER2-LFLEDA, 4 donors (8%) showed positive CD4+ T cell responses, which was comparable to 4 (8%) and 3 (6%) positive responses for IgG1-HER2 and IgG1-HER2-LFLE, respectively ( Figure 18 ). Thus, IgG1-HER2-LFLEDA-K409R (as well as IgG1-HER2 and IgG1-HER2-LFLE-K409R) shows a low potential for immunogenicity, with a response frequency of less than 10%. The positive control humanized A33 (e.g.,

[84] ) was used as a clinical benchmark control antibody, which shows a high level of immunogenicity clinically and typically induces 20-30% T cell responses in the EpiScreen assay.

[0509] Example 13 – Generation of Mutants to Optimize the Production of Humanized CD3 Antibody Generate huCD3-L1 mutant plasmid Several IgG1-huCD3-L1 LC mutants were generated to improve the expression level of IgG1-huCD3-H1L1 in transient transfection assays, see Table 2. Residue selection was based on: comparison with germline sequences or screening for the presence of rare residues in huCD3-L1, and crystal structures from homologous antibodies. The selected sequences were synthesized at GeneArt (Life Technologies, Germany). p33L encodes the constant domain of the human IgLC2 / IgLC3λ light chain of SEQ ID NO:31. p33G1f encodes the IgG1m(f) heavy chain constant region of SEQ ID NO:15.

[0510] Table 2 Transient expression in Expi293F cells For a single antibody, plasmids encoding the heavy chain (HC) and light chain (LC) were transiently transfected into Freestyle Expi293F cells (Life technologies, USA) using ExpiFectamine 293 (Life technologies). A total of 1.5 μg of HC-encoding plasmid and 1.5 μg of LC-encoding plasmid (Table 2) were diluted in 150 μL of Opti-MEM (Gibco, USA). To prepare the transfection mixture, 8 μL of ExpiFectamine 293 was diluted in 150 μL of Opti-MEM and incubated for 5 minutes at room temperature. Subsequently, the DNA / Opti-MEM and ExpiFectamine 293 / Opti-MEM solutions were mixed, incubated for 20 minutes at room temperature and added to 2.55 mL of Expi293 expression medium containing 7.5x10 6 Expi293F cells and 50 U / mL Pen-Strep. The cells were incubated at 37 °C, 8% CO2 and shaken at 200 rpm. To enhance expression, 15 μL of Boosting Mix 1 and 150 μL of Boosting Mix 2 were added 21 hours post-transfection. The cells were incubated for 4 days, then the supernatant was harvested. The supernatant was centrifuged at 3,000 x g and filtered through a 0.2 μm filter to sterilize. IgG expression levels were measured on an Octet RED (ForteBio, US) using an anti-human IgG sensor (ForteBio, USA).

[0511] IgG concentration analysis For all mutants listed, IgG expression levels were measured individually or in combination. Figure 20 Show the IgG expression levels measured in the supernatant for the antibody sets expressed as listed in Table 2. The IgG1-huCD3-H1L1 antibody was expressed up to 72 μg / mL. For the IgG1-huCD3-H1L1-LT41K mutant, a 4-fold increase in IgG expression (295 μg / mL) was observed. For the IgG1-huCD3-H1L1-LLKNH mutant, which especially includes the T41K mutation, a similar expression level (311 μg / mL) was observed. Compared to IgG1-huCD3-H1L1, when tested individually, other mutations in these constructs did not show increased expression (IgG1-huCD3-H1L1-LF10L, IgG1-huCD3-H1L1-LK55N, IgG1-huCD3-H1L1-LL97H).

[0512] For the combination of R23A and A35P, a second set of mutations that increased expression was observed. Although the IgG1-huCD3-H1L1-LLTGPEAEY form lacking R23A and A35P did not show increased expression (83 μg / mL), the IgG1-huCD3-H1L1-LLAPTGPEAEY containing the additional R23A and A35P mutations did show a 3-fold increase in expression (237 μg / mL). Individually, R23A or A35P did not show increased expression levels (56 and 81 μg / mL, respectively).

[0513] Binding of the Example 14-huCD3-H1L1-LFLEDA variant to Jurkat cells The apparent binding affinity of the huCD3-H1L1-LFLEDA variant to Jurkat cells was tested. The IgG1-CD3-hmAb286 disclosed in US2012038219 was used as a reference antibody.

[0514] Jurkat cells (DSMZ, Brunswick, Germany) were thawed at 37 °C and washed once with 30 mL PBS and centrifuged at 300 xg for 10 minutes at 4 °C. The cell pellet was resuspended in 25 mL PBS / 0.1% BSA / 0.02% azide at a final concentration of 0.8x10 6 cells / mL. 100 μL of this cell suspension (0.8x10 5 The cells / well were transferred to a polystyrene 96-well round bottom plate (Greiner Bio-one, Alphen a / d Rijn, The Netherlands), and incubated with supernatant preparations serially diluted in PBS / 0.1% BSA / 0.02% azide (range 3 - 10,000 ng / mL, 3-fold dilutions) for 30 minutes at 4°C. After washing three times with PBS / 0.1% BSA / 0.02% azide, Jurkat cells were incubated with 50 μL of the secondary antibody for 30 min at 4°C. As the secondary antibody, R-phycoerythrin (PE)-conjugated goat-anti-human IgG F(ab’)2 (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) diluted 1 / 200 in PBS / 0.1% BSA / 0.02% azide was used for all experiments. Next, Jurkat cells were washed twice with PBS / 0.1% BSA / 0.02% azide, resuspended in 150 μL of PBS / 0.1% BSA / 0.02% azide and analyzed on a FACS Cantoll (BD Biosciences). Binding curves were analyzed using non-linear regression (sigmoidal dose response, with variable slope) with GraphPad Prism V5.04 software (GraphPad Software, San Diego, CA, USA).

[0515] Figure 21 The binding curves are shown, and Table 3 shows the EC 50 values. The T41K mutation had no effect on binding affinity, while the K55N and LKNH mutants showed an apparent affinity of 1 / 5 that of wild-type IgG1-huCD3-H1L1-LFLEDA. The affinity of the reference antibody IgG1-CD3-hmAb286-LFLEDA was lower than that of IgG1-huCD3-H1L1-LFLEDA and similar to that of the IgG1-huCD3-H1L1-LFLEDA variant containing the K55N mutation.

[0516] Table 3 Antibody <![CDATA[EC 50 (μg / mL)]]> IgG1-huCD3-H1L1-LFLEDA 0.096 IgG1-huCD3-H1-LFLEDA-LF10L 0.169 IgG1-huCD3-H1-LFLEDA-LT41K 0.096 IgG1-huCD3-H1-LFLEDA-LK55N 0.531 IgG1-huCD3-H1-LFLEDA-LL97H 0.095 IgG1-huCD3-H1-LFLEDA-LLKNH 0.438 IgG1-CD3-hmAb286-LFLEDA 0.397 Example 15 – Binding of Bispecific huCD3xHER2 Affinity Variants to T Cells The binding of purified affinity variants of bispecific (bs) IgG1-huCD3xHER2 molecules with or without the LFLEDA mutation in the Fc region to human T cells isolated from buffy coats (donor number: N001814268806, purchased from Sanquin, Amsterdam, The Netherlands) was analyzed by FACS analysis. According to the manufacturer's instructions, the T cells were isolated by StemSep TM Human T Cell Enrichment Kit (Stem Cell Technologies, Vancouver, Canada). Herceptin-LbH1-mutation was identified from the public database from Bostrom et al. (85, 86), and has a lower affinity for HER2 compared to Herceptin (trastuzumab).

[0517] To evaluate the binding, the isolated T cells (1x10 5 cells / well) were incubated in polystyrene 96-well round bottom plates (Greiner bio-one 650101) with antibody preparations serially diluted (range 6.4 - 10,000 ng / mL, 5-fold dilutions) in 100 μL PBS / 0.1% BSA / 0.02% azide at 4 °C for 30 minutes. After washing twice with PBS / 0.1% BSA / 0.02% azide, the T cells were incubated with the secondary antibody in 50 μL at 4 °C for 30 minutes. As the secondary antibody, R-phycoerythrin (PE)-conjugated goat-anti-human IgG F(ab’)2 (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) diluted 1 / 200 in PBS / 0.1% BSA / 0.02% azide was used for all experiments. Subsequently, the T cells were washed twice with PBS / 0.1% BSA / 0.02% azide, resuspended in 100 μL PBS / 0.1% BSA / 0.02% azide and analyzed on a FACS Cantoll (BD Biosciences). The binding curves were analyzed using nonlinear regression (sigmoidal dose-response, with variable slope) using GraphPad Prism V5.04 software (GraphPad Software, San Diego, CA, USA).

[0518] The binding curve results are shown in Figure 22 and the EC 50 values are shown in Table 4. As can be seen, the maximum difference between the EC 50 values of the CD3 variants is 2-fold.

[0519] Table 4 Example 16 - In Vitro T Cell - Mediated Cytotoxicity Induced by an Affinity Variant of the huCD3 Antibody In this example, T cell - mediated cytotoxicity against HER2 - positive tumor cells was evaluated using bispecific antibodies with different affinities for huCD3 and HER2. The bispecific antibodies were prepared as described above and contained the F405L mutation in the CD3 arm and the K409R mutation in the HER2 arm. A431 and AU565 (human breast cancer) cells with different HER2 expression levels were used for this assay. The HER2 copy number was ∼20,000 for A431 cells and ∼1,000,000 for AU565 cells. AU565 cells were cultured in RPMI 1640 (Lonza, Basel, Switzerland) supplemented with 10% (vol / vol) iron - containing fetal bovine serum (Life Technologies, Germany), 1.5 g / L sodium bicarbonate (Lonza, Germany), 1 mM sodium pyruvate (Lonza, Germany), 4.5 g / L glucose (Sigma), 50 IU / mL penicillin, and 50 μg / mL streptomycin (Lonza, Basel, Switzerland) and having HEPES and L - glutamine. A431 (ATCC, epidermoid carcinoma) cells were cultured in RPMI 1640 (Lonza, Basel, Switzerland) supplemented with 10% (vol / vol) iron - containing fetal bovine serum (Life Technologies, Germany), 50 IU / mL penicillin, and 50 μg / mL streptomycin (Lonza, Basel, Switzerland) and having HEPES and L - glutamine. The cell lines were maintained at 37°C in a 5% (vol / vol) CO2 humidified incubator. AU565 and A431 cells were cultured to near confluence, after which the cells were subjected to trypsin treatment, resuspended in medium, and passed through a cell strainer to obtain a single - cell suspension. The AU565 and A431 cells were seeded at 3x10 4 and 1.6x10 4 cells / well into a 96 - well culture plate and incubated at 37°C, 5% CO2 for at least 3 hours to allow attachment to the plate.

[0520] Human T cells were isolated from buffy coats as described in Example 15. The isolated T cells were washed with PBS, resuspended in medium (RPMI 1640, 10% serum), and added to AU565 at a 1:3 ratio or to A431 cells at a 1:10 ratio in a 96 - well plate.

[0521] Dilution series (final concentration range 0.1 up to 10,000 ng / mL) of bispecific antibody variants bsIgG1-huCD3 / Herceptin-LFLEDA, bsIgG1-huCD3 / Herceptin-LbH1—LFLEDA, bsIgG1-huCD3-H1L1 / HER2-LFLEDA, bisIgG1-huCD3-LK55N / Herceptin-LbH1—LFLEDA, bsIgG1-huCD3-LK55N / HER2-LFLEDA, bsIgG1-huCD3-hmAb286 / Herceptin-LFLEDA, bsIgG1-huCD3-hmAb286 / Herceptin-LbH1—LFLEDA, bsIgG1-CD3-hmAb286 / HER2-LFLEDA and bsIgG1-huCD3-H1L1 / b12-LFLEDA were prepared in medium and added to plates. The plates were incubated at 37 °C, 5% CO2 for 3 days. Cells were incubated with 1 μM staurosporine (Sigma) and used as a reference for 100% tumor cell kill. The plates were washed twice with PBS and 150 μL of medium containing 10% Alamar blue (Life Technologies, Germany) was added to each well. The plates were incubated at 37 °C, 5% CO2 for 4 hours. Absorbance at 590 nm was measured (Envision, Perkin Elmer, Waltham, MA).

[0522] Figure 23 A431 cells treated with CD3 and HER2 affinity variants are shown ( Figure 23 A) and AU565 cells ( Figure 23 B). Lower HER2 affinities combined with high and low CD3 affinities resulted in lower efficacy in both cell lines (shown as open symbols in both figures). Bispecific antibodies with similar high affinities for HER2 (Herceptin and HER2) combined with wild-type huCD3-H1L1, huCD3-LK55N, or CD3-hmAb286 CD3 arms showed similar efficacy against A431 cells. Variations in the affinity of the CD3 arm had no effect on the efficacy against A431 cells. Bispecific antibodies with similar high affinities for HER2 (Herceptin and HER2) combined with the CD3-hmAb286 CD3 arm, huCD3-LK55N arm, and wild-type huCD3-H1L1 arm showed similar efficacy against AU565 cells, with relatively higher efficacy shown for the CD3 hmAb286 CD3 arm. For the CD3 arm, lower binding affinity was more beneficial in AU565 cells with higher HER2 expression compared to A431 cells.

[0523] Example 17 – In Vivo Tumor Killing by huCD3-H1L1xHER2-LFLEDA and huCD3-H1L1xCD20-LFLEDA in NOD-SCID Mice The in vivo anti-tumor efficacy of the bispecific CD3xHER2 antibody bsIgG1-huCD3-H1L1xHER2-LFLEDA was evaluated in a subcutaneous NCI-N87 xenograft model. In this model, human unstimulated PBMCs were used as a source of human T cells and co-injected with tumor cells, similar to the model described by Brischwein et al. (Mol. Immunol. 43 (2006), 1129–1143). In these experiments, 6-11-week-old female NOD-SCID (NOD.CB17-Prkdcscid / NcrCrl) mice were used. Human PBMCs from healthy donors were isolated from buffy coats as described in Example 15. On day 0, a mixture of PBMCs containing 5x10 6 cells and NCI-N87 cells was subcutaneously injected at 200 μL into the right flank of each mouse. Within 1 hour of the injection, the mice were randomly assigned to 7 different treatment groups (n = 4 / donor). Each group was injected intravenously (i.v.) with a single dose of the (bispecific) antibody. The treatment groups are shown in Table 5. Tumor growth was evaluated twice weekly using calipers (PLEXX) until the end-point tumor volume was 1500 mm 3 , until the tumor showed ulceration or until the end of the study.

[0524] Results are shown in Figure 24 As can be seen from Figure 24 As can be seen in A, huCD3-H1L1xHER2-LFLEDA effectively reduced tumor size at all three tested concentrations. Figure 24 B shows the tumor size 29 days after tumor inoculation. Tumor formation was significantly inhibited in mice treated with bsIgG1-huCD3-H1L1xHER2-LFLEDA at doses of 0.05 mg / kg, 0.5 mg / kg, and 5 mg / kg compared to the PBS control group (p < 0.0001, Tukey's multiple comparison test), while treatment with 0.005 mg / kg did not affect tumor formation. Treatment with the control antibody bsIgG1-huCD3-H1L1xb12-LFLEDA at 5 mg / kg did not inhibit tumor formation, while treatment with bsIgG1-b12xHER2-LFLEDA seemed to induce some tumor growth inhibition compared to the PBS control, but this was not significant according to one-way Anova, Tukey's multiple comparison test.

[0525] Table 5 Group Antibody Dose 1 PBS 2 bsIgG1-huCD3-H1L1xHER2-LFLEDA 0.1 μg (=0.005 mg / kg) 3 bsIgG1-huCD3-H1L1xHER2-LFLEDA 1 μg (=0.05 mg / kg) 4 bsIgG1-huCD3-H1L1xHER2-LFLEDA 10 μg (=0.5 mg / kg) 5 bsIgG1-huCD3-H1L1xHER2-LFLEDA 100 μg (=5 mg / kg) 6 bsIgG1-huCD3-H1L1xb12-LFLEDA 100 μg (=5 mg / kg) 7 bsIgG1-b12xHER2-LFLEDA 100 μg (=5 mg / kg) The in vivo anti-tumor efficacy of the bispecific CD20xCD3 antibody huCD3-H1L1xCD20-LFLEDA was evaluated in a subcutaneous Raji xenograft model. In this model, human unstimulated PBMCs were used as a source of human T cells and co-inoculated with tumor cells, similar to the model described by Brischwein et al. (Mol. Immunol. 43 (2006), 1129–1143). Female NOD-SCID (NOD.CB17-Prkdcscid / NcrCrl) mice, 6–11 weeks old, were used. Human PBMCs from healthy donors were isolated from buffy coats, washed, and resuspended in PBS-0.1% BSA. On day 0, a mixture of both PBMCs containing 5x10 6 cells and Raji cells was subcutaneously inoculated into the right flank of each mouse in a volume of 200 μL. Within 1 hour of injection, the mice were sorted into 7 groups (n = 4 / donor) and each group was injected intravenously (i.v.) with a single dose of the (bispecific) antibody. The treatment groups are shown in Table 6. Tumor growth was evaluated twice weekly using calipers (PLEXX) until the end-point tumor volume of 1500 mm 3 , until the tumor showed ulceration or until the end of the study.

[0526] Results are shown in Figure 25 As can be seen from Figure 25 A, at the two higher tested doses of huCD3-H1L1xCD20-LFLEDA (0.5 mg / kg and 0.05 mg / kg), huCD3-H1L1xCD20-LFLEDA effectively reduced tumor size.

[0527] In Figure 25 B, the tumor size of individual mice was indicated on the last day (day 21) when each group was complete. Compared with the PBS control group, tumor formation was significantly inhibited in mice treated with huCD3-H1L1xCD20-LFLEDA at doses of 0.05 mg / kg and 0.5 mg / kg (p-values were p < 0.05 and p < 0.01, respectively, evaluated by one-way Anova, Tukey's multiple comparison test). Treatment with 0.005 mg / kg did not affect tumor formation. Treatment with the control antibody b12xCD20 (bsIgG1-b12xCD20-LFLEDA) at 0.5 mg / kg and 0.05 did not significantly inhibit tumor formation compared with the PBS control (one-way Anova, Tukey's multiple comparison test).

[0528] Table 6 Group Antibody Dose 1 PBS 2 bsIgG1-huCD3-H1L1xCD20-LFLEDA 0.1 μg (=0.005 mg / kg) 3 bsIgG1-huCD3-H1L1xCD20-LFLEDA 1 μg (=0.05 mg / kg) 4 bsIgG1-huCD3-H1L1xCD20-LFLEDA 10 μg (=0.5 mg / kg) 5 bsIgG1-b12xCD20-LFLEDA 1 μg (=0.05 mg / kg) 6 bsIgG1-b12xCD20-LFLEDA 10 μg (=0.5 mg / kg) List of references [1] Xu et al., 2000, Cell Immunol. 200(1):16 - 26 [2] Herold et al., 2005, Diabetes, 54(6):1763 - 9 [3] Staerz et al., 1985, Nature 314:628 - 631 [4] Muller and Kontermann, 2010, BioDrugs 24:89 - 98 [5] Lum and Thakur, 2011, BioDrugs 25:365 - 379 [6] Linke et al., 2010, MAbs 2:129 - 136 [7] Ruf et al., 2010, Br J Clin Pharmacol 69:617 - 625 [8] Bokemeyer et al., 2009, J Clin Oncol(Meeting Abstracts), 3036 [9] Heiss et al., 2010, Int J Cancer 127:2209 - 2221

[10] Jones et al., 2009, Lancet Oncol 10:1179 - 1187

[11] Kiewe et al., 2006, Clin Cancer Res 12:3085 - 3091

[12] WO 2012 / 162067

[13] WO 2008 / 119567

[14] Fundamental Immunology Ch.7, Paul, W editor, Second Edition. Raven Press, N.Y. (1989)

[15] Lefranc MP, et al., 2003, Dev Comp Immunol. Jan; 27(1):55 - 77

[16] Brochet, X. et al., 2008, Nucl. Acids Res. 36, W503 - 508

[17] Giudicelli, V., Brochet, X, Lefranc, M.-P., 2011, Cold Spring Harb Protoc. Jun 1; 2011(6)

[18] Sambrook et al., 1989, Molecular Cloning: A laboratory Manual, New York: Cold Spring Harbor Laboratory Press, Ch.15

[19] WO92 / 22653

[20] EP 0 629 240

[21] E. Meyers and W. Miller, 1988, Comput. Appl. Biosci 4 , 11 - 17

[22] Needleman and Wunsch, 1970, J. Mol. Biol. 48 , 444 - 453

[23] Clustal W algorithm, Thompson, 1994

[24] T cell epitope database, from for example http: / / www.iedb.org /

[25] Perry et al., 2008 Drugs R D 9(6):385 - 396

[26] Bryson et al., 2010, Biodrugs 24(1):1 - 8

[27] Kabat, E.A. et al., 1991, Sequence of proteins of immunological interest. 5th ed. - US Department of Health and Human Services, NIH publication No. 91 - 3242, pp662, 680, 689

[28] Oganesyan et al., 2008, Acta Cryst. (D64): 700 - 4

[29] Canfield et al., 1991, J. Exp. Med. (173): 1483 - 91

[30] Duncan et al., 1988, Nature (332): 738 - 40

[31] Shields et al., 2001, J. Biol. Chem. (276): 6591 - 604

[32] Idusogie EE, et al., 2000, J Immunol. 164: 4178 - 84

[33] Leabman et al., 2013, MAbs;5(6): 896 - 903

[34] Parren et al., 1992, J. Clin Invest. 90: 1537 - 1546

[35] Bruhns et al.,.2009, Blood 113: 3716 - 3725

[36] WO 2011 / 066501

[37] Lightle, S., et al., 2010, Protein Science(19): 753 - 62

[38] Brekke et al., 2006, J Immunol 177: 1129 - 1138

[39] Dall’Acqua WF, et al., 2006, J Immunol 177: 1129 - 1138

[40] Wu et al., 2010, Generation and Characterization of a Dual Variable Domain Immunoglobulin (DVD - Ig TM )Molecule, In: Antibody Engineering, Springer Berlin Heidelberg

[41] WO 2011 / 131746

[42] WO 2002 / 020039

[43] WO 98 / 050431

[44] WO 2011 / 117329

[45] EP 1 870 459

[46] WO 2009 / 089004

[47] US 2010 00155133

[48] WO 2010 / 129304

[49] WO 2007 / 110205

[50] WO 2010 / 015792

[51] WO 2011 / 143545

[52] WO 2012 / 058768

[53] WO 2011 / 028952

[54] WO 2008 / 003116

[55] US 7,262,028

[56] US 7,612,181

[57] WO 2010 / 0226923

[58] US 7,951,918

[59] CN 102250246

[60] WO 2012 / 025525

[61] WO 2012 / 025530

[62] WO 2008 / 157379

[63] WO 2010 / 080538

[64] Sykes and Johnston, 1997, Nat Biotech 17 , 355 - 59

[65] US 6,077,835

[66] WO 2000 / 70087

[67] Schakowski et al., 2001, Mol Ther 3 , 793 - 800

[68] WO 2000 / 46147

[69] Benvenisty and Reshef, 1986, PNAS USA 83 , 9551 - 55

[70] Wigler et al., 1978, Cell 14 ,725

[71] Coraro and Pearson, 1981, Somatic Cell Genetics 7, 603

[72] US 5,589,466

[73] US 5,973,972

[74] Van Heeke & Schuster, 1989, J Biol Chem 264, 5503 - 5509

[75] F. Ausubel et al., 1987, ed. Current Protocols in Molecular Biology, Greene Publishing and Wiley InterScience New York

[76] Grant et al., 1987, Methods in Enzymol 153, 516 - 544

[77] Remington: The Science and Practice of Pharmacy, 19th Edition, Gennaro editor, Mack Publishing Co., Easton, PA, 1995

[78] Sustained and Controlled Release Drug Delivery Systems, J.R. Robinson editor, Marcel Dekker, Inc., New York, 1978

[79] Srivastava (editor), Radiolabeled Monoclonal Antibody For Imaging and Therapy, Plenum Press 1988

[80] Chase, "Medical Applications of Radioisotopes," in Remington's Pharmaceutical Sciences, 18th Edition, Gennaro et al. (editors), pp. 624 - 652, Mack Publishing Co., 1990

[81] Brown, "Clinical Use of Monoclonal Antibodies," in Biotechnology and Pharmacy 227 - 49, Pezzuto et al. (eds.), Chapman & Hall 1993

[82] US 5,057,313

[83] US 6,331,175

[84] Ritter G et al.; 2001, Cancer Res., 61:6851 - 9

[85] Bostrom et al., Science. 2009; 323(5921):1610 - 4

[86] Bostrom et al., PLosOne. 2011; 6(4)e17887.< / scid>

Claims

1. A humanized or chimeric antibody that binds to human CD3, wherein said antibody comprises a binding region that comprises: heavy chain variable (VH) region CDR1, CDR2, and CDR3 having the sequences shown in SEQ ID NOs: 1, 2, and 3, respectively, and light chain variable (VL) region CDR1, CDR2, and CDR3 having the sequence shown in SEQ ID NO: 4, the sequence GTN, and the sequence shown in SEQ ID NO: 5 or SEQ ID NO: 60, respectively.

2. The antibody of claim 1, wherein said VH region has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity to the amino acid sequence shown by a VH sequence selected from: a) the VH sequence shown in SEQ ID NO: 6; b) the VH sequence shown in SEQ ID NO: 8; c) the VH sequence shown in SEQ ID NO: 7; and d) the VH sequence shown in SEQ ID NO:

9.

3. The antibody of any one of the preceding claims, wherein said VL region has at least 90%, at least 95%, at least 97%, or at least 99% amino acid sequence identity to the amino acid sequence shown by a VL sequence selected from: a) the VL sequence shown in SEQ ID NO: 10; b) the VL sequence shown in SEQ ID NO: 11; and c) the VL sequence shown in SEQ ID NO:

12.

4. The antibody of any one of the preceding claims, wherein said VH region is selected from: a) the VH sequence shown in SEQ ID NO: 6; b) the VH sequence shown in SEQ ID NO: 8; c) the VH sequence shown in SEQ ID NO: 7; and d) the VH sequence shown in SEQ ID NO:

9.

5. The antibody of any one of the preceding claims, wherein said VL region is selected from: a) the VL sequence shown in SEQ ID NO: 10; b) the VL sequence shown in SEQ ID NO: 11; and c) the VL sequence shown in SEQ ID NO:

12.

6. The antibody of any one of the preceding claims, wherein said VH and VL regions are selected from: a) the VH sequence shown in SEQ ID NO: 6, and the VL sequence shown in SEQ ID NO: 10; b) the VH sequence shown in SEQ ID NO: 8, and the VL sequence shown in SEQ ID NO: 10; c) the VH sequence shown in SEQ ID NO: 9, and the VL sequence shown in SEQ ID NO: 10; d) the VH sequence shown in SEQ ID NO: 6, and the VL sequence shown in SEQ ID NO: 11; e) the VH sequence shown in SEQ ID NO: 6, and the VL sequence shown in SEQ ID NO: 12; f) the VH sequence shown in SEQ ID NO: 7, and the VL sequence shown in SEQ ID NO: 10; g) the VH sequence shown in SEQ ID NO: 7, and the VL sequence shown in SEQ ID NO: 11; h) The VH sequence shown in SEQ ID NO:7, and the VL sequence shown in SEQ ID NO:12; i) The VH sequence shown in SEQ ID NO:8, and the VL sequence shown in SEQ ID NO:11; j) The VH sequence shown in SEQ ID NO:8, and the VL sequence shown in SEQ ID NO:12; k) The VH sequence shown in SEQ ID NO:9, and the VL sequence shown in SEQ ID NO:11; and l) The VH sequence shown in SEQ ID NO:9, and the VL sequence shown in SEQ ID NO:

12.

7. The antibody of any one of the preceding claims, wherein the binding region comprises a VH sequence and a VL sequence selected from: a) The VH sequence shown in SEQ ID NO:6, and the VL sequence shown in SEQ ID NO:10; b) The VH sequence shown in SEQ ID NO:8, and the VL sequence shown in SEQ ID NO:10; and c) The VH sequence shown in SEQ ID NO:9, and the VL sequence shown in SEQ ID NO:

10.

8. The antibody of any one of the preceding claims, wherein the antibody is a humanized antibody.

9. The antibody of claim 1, wherein the antibody is a chimeric antibody.

10. The antibody of any one of the preceding claims, wherein the antibody is a full-length antibody.

11. The antibody of any one of the preceding claims, wherein the antibody comprises an Fc region, and the Fc region comprises a first and a second immunoglobulin heavy chain.

12. The antibody of any one of the preceding claims, wherein the first and the second heavy chains have an isotype selected from IgG1, IgG2, IgG3, and IgG4.

13. The antibody of any one of the preceding claims, wherein the antibody comprises an Fc region that has been modified such that the binding of C1q to the antibody is reduced by at least 70%, at least 80%, at least 90%, at least 95%, at least 97%, or 100% compared to the wild-type antibody, wherein C1q binding is determined by ELISA.

14. The antibody of any one of the preceding claims, wherein the antibody comprises an Fc region that has been modified such that, compared to the wild-type antibody, the antibody mediates at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99%, or 100% less Fc-mediated T-cell proliferation, wherein the T-cell proliferation is measured in a functional assay based on peripheral blood mononuclear cells (PBMC).

15. The antibody of any one of the preceding claims, wherein the antibody comprises an Fc region that has been modified such that, compared to the wild-type antibody, the antibody reduces Fc-mediated CD69 expression by at least 50%, at least 60%, at least 70%, at least 80%, at least 90%, at least 99%, or 100%, wherein the Fc-mediated CD69 expression is determined in a functional assay based on PBMC.

16. The antibody of any one of the preceding claims, wherein the antibody comprises first and second immunoglobulin heavy chains, and wherein in at least one of the first and second immunoglobulin heavy chains, one or more of the amino acids at positions corresponding to positions L234, L235, D265, N297, and P331 of the human IgG1 heavy chain are not L, L, D, N, and P, respectively.

17. The antibody of claim 16, wherein in at least one of the first and second heavy chains, the amino acid at the position corresponding to position D265 of the human IgG1 heavy chain is not D.

18. The antibody of claim 16, wherein in at least one of the first and second heavy chains, the amino acid at the position corresponding to position N297 of the human IgG1 heavy chain is not N.

19. The antibody of claim 16, wherein in at least one of the first and second heavy chains, the amino acids at the positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are not L and L, respectively.

20. The antibody of any one of claims 16 and 19, wherein in at least one of the first and second heavy chains, the amino acids at the positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are F and E, respectively; or A and A.

21. The antibody of claim 20, wherein in at least one of the first and second heavy chains, the amino acids at the positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are F and E, respectively.

22. The antibody of claim 20, wherein in at least one of the first and second heavy chains, the amino acids at least at the positions corresponding to positions L234 and L235 of the human IgG1 heavy chain are A and A, respectively.

23. The antibody of any one of claims 1-16, wherein in at least one of the first and second heavy chains, the amino acids at the positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are not L, L, and D, respectively.

24. The antibody of claim 23, wherein in at least one of the first and second heavy chains, the amino acids at the positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are F, E, and A, respectively; or A, A, and A.

25. The antibody of claim 24, wherein in at least one of the first and second heavy chains, the amino acids at the positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are F, E, and A, respectively.

26. The antibody of claim 24, wherein in at least one of the first and second heavy chains, the amino acids at the positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain are A, A, and A, respectively.

27. The antibody of claim 16, wherein in at least one of the first and second heavy chains, the amino acids at the positions corresponding to positions L234, L235, D265, N297, and P331 of the human IgG1 heavy chain are F, E, A, Q, and S, respectively.

28. The antibody of any one of the preceding claims, wherein the antibody comprises a light chain (LC), and wherein the amino acid at the position corresponding to position T41 of the λ light chain of SEQ ID NO: 10 is not T.

29. The antibody of claim 28, wherein the amino acid at the position corresponding to position T41 of the λ light chain of SEQ ID NO: 10 is K.

30. The antibody of any one of claims 1-28, wherein the amino acid at the position corresponding to position F10 of the λ light chain of SEQ ID NO: 10 is not F, and wherein one or more amino acid positions corresponding to positions T41, K55 and L97 of the λ light chain of SEQ ID NO: 10 are not T, K and L, respectively.

31. The antibody of claim 30, wherein the amino acids at the positions corresponding to positions F10, T41, K55 and L97 of the λ light chain of SEQ ID NO: 10 are not F, T, K and L, respectively.

32. The antibody of claim 31, wherein the amino acids at the positions corresponding to positions F10, T41, K55 and L97 of the λ light chain of SEQ ID NO: 10 are L, K, N and H, respectively.

33. The antibody of any one of claims 1-28, wherein the amino acids at the positions corresponding to positions R23 and A35 are not R and A, respectively.

34. The antibody of claim 33, wherein the amino acids at the positions corresponding to positions R23 and A35 are A and P, respectively.

35. The antibody of claim 33, wherein the amino acids at the positions corresponding to positions F10, R23, A35, R47, D71, A82, D83, S86, I87 and F89 of the λ light chain of SEQ ID NO: 10 are not F, R, A, R, D, A, D, S, I and F, respectively.

36. The antibody of claim 35, wherein the amino acids at the positions corresponding to positions F10, R23, A35, R47, D71, A82, D83, S86, I87 and F89 of the λ light chain of SEQ ID NO: 10 are L, A, P, T, G, P, E, A, E and Y, respectively.

37. The antibody of any one of claims 1-36, wherein (i) the amino acid at the position corresponding to position F10 of the λ light chain of SEQ ID NO: 10 is not F, or (ii) the amino acid at the position corresponding to position K55 of the λ light chain of SEQ ID NO: 10 is not K, or (iii) the amino acid at the position corresponding to position F10 of the λ light chain of SEQ ID NO: 10 is not F, and the amino acid at the position corresponding to position K55 of the λ light chain of SEQ ID NO: 10 is not K.

38. The antibody of claim 37, wherein (i) the amino acid at the position corresponding to position F10 of the λ light chain of SEQ ID NO: 10 is L, or (ii) the amino acid at the position corresponding to position K55 of the λ light chain of SEQ ID NO: 10 is N, or (iii) The amino acid at the position corresponding to position F10 of the λ light chain of SEQ ID NO:10 is L, and the amino acid at the position corresponding to position K55 of the λ light chain of SEQ ID NO:10 is N.

39. An antibody according to any one of claims 1-38, wherein the antibody comprises a constant heavy chain (HC) and a constant light chain (LC), wherein the positions corresponding to positions L234, L235, and D265 of the human IgG1 heavy chain of SEQ ID NO:15 are F, E, and A, respectively, and wherein the position corresponding to position F405 of the human IgG1 heavy chain of SEQ ID NO:15 is L, and wherein (i) the positions corresponding to positions F10, T41, K55, and L97 of the λ light chain of SEQ ID NO:10 are L, K, N, and H, respectively, or (ii) the position corresponding to position T41 of the λ light chain of SEQ ID NO:10 is K.

40. A bispecific antibody comprising a first binding region of an antibody according to any one of claims 1-39, and a second binding region that binds to a target different from that bound by the first antigen-binding region.

41. The bispecific antibody of claim 40, wherein the antibody comprises a first and a second heavy chain.

42. The bispecific antibody of claim 41, wherein a) the bispecific antibody comprises an Fc region modified according to any one of claims 13-15; or b) at least one of the first and second heavy chains comprises one or more amino acids modified as defined in any one of claims 16-39.

43. The bispecific antibody according to any one of claims 40-42, wherein the first and second heavy chains each comprise at least a hinge region, a CH2, and a CH3 region, wherein at least one amino acid at a position corresponding to a position selected from T366, L368, K370, D399, F405, Y407, and K409 of the human IgG1 heavy chain is substituted in the first heavy chain, and at least one amino acid at a position corresponding to a position selected from T366, L368, K370, D399, F405, Y407, and K409 of the human IgG1 heavy chain is substituted in the second heavy chain, and wherein the first and the second heavy chains are not substituted at the same position.

44. The bispecific antibody of claim 43, wherein the amino acid at the position corresponding to position F405 of the human IgG1 heavy chain in the first heavy chain is L, and the amino acid at the position corresponding to position K409 of the human IgG1 heavy chain in the second heavy chain is R, or vice versa.

45. The bispecific antibody according to any one of claims 40-44, wherein the first binding region is according to any one of claims 1-7, and the second binding region binds to a target different from that bound by the first binding region.

46. The bispecific antibody of claim 45, wherein the second binding region binds to human HER2 or human CD20.

47. The bispecific antibody of claim 46, wherein the second binding region binds to human CD20.

48. The bispecific antibody of claim 47, wherein the second binding region that binds to human CD20 comprises: (i) the VH CDR1 region of SEQ ID NO:34, the VH CDR2 region of SEQ ID NO:35, the VH CDR3 region of SEQ ID NO:36, the VL CDR1 region of SEQ ID NO:37, the VL CDR2 region of DAS, and the VL CDR3 region of SEQ ID NO:38, (ii) the VH CDR1 region of SEQ ID NO:41, the VH CDR2 region of SEQ ID NO:42, the VH CDR3 region of SEQ ID NO:43, the VL CDR1 region of SEQ ID NO:44, the VL CDR2 region of DAS, and the VL CDR3 region of SEQ ID NO:45, (iii) the VH CDR1 region of SEQ ID NO:48, the VH CDR2 region of SEQ ID NO:49, the VH CDR3 region of SEQ ID NO:50, the VL CDR1 region of SEQ ID NO:51, the VL CDR2 region of DAS, and the VL CDR3 region of SEQ ID NO:52, or (iv) the VH CDR1 region of SEQ ID NO:55, the VH CDR2 region of SEQ ID NO:56, the VH CDR3 region of SEQ ID NO:57, the VL CDR1 region of SEQ ID NO:58, the VL CDR2 region of DAS, and the VL CDR3 region of SEQ ID NO:

59.

49. The bispecific antibody of claim 47 or 48, wherein the second binding region that binds to human CD20 comprises: (i) a VH sequence having at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO:29; and a VL sequence having at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO:30, (ii) a VH sequence having at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO:39; and a VL sequence having at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO:40, (iii) a VH sequence having at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO:46; and a VL sequence having at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO:47, or (iv) A VH sequence that has at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO:53; and a VL sequence that has at least 90%, at least 95%, at least 97% or at least 99% amino acid sequence identity with the amino acid sequence shown in SEQ ID NO:

54.

50. The bispecific antibody of claim 49, wherein the second binding region that binds to human CD20 comprises: (i) The VH sequence of SEQ ID NO:29 and the VL sequence of SEQ ID NO:30, (ii) The VH sequence of SEQ ID NO:39 and the VL sequence of SEQ ID NO:40, (iii) The VH sequence of SEQ ID NO:46 and the VL sequence of SEQ ID NO:47, or (iv) The VH sequence of SEQ ID NO:53 and the VL sequence of SEQ ID NO:

54.

51. A nucleic acid construct encoding one or more of the amino acid sequences described in Table 1.

52. An expression vector comprising (i) A nucleic acid sequence encoding the heavy chain sequence of a humanized or chimeric antibody of any one of claims 1-50; (ii) A nucleic acid sequence encoding the light chain sequence of a humanized or chimeric antibody of any one of claims 1-50; or (iii) Both (i) and (ii).

53. A host cell comprising the expression vector of claim 52.

54. The host cell of claim 53, wherein the host cell is a recombinant eukaryotic, recombinant prokaryotic or recombinant microbial host cell.

55. A composition comprising an antibody of any one of claims 1-39 or a bispecific antibody of any one of claims 40-50.

56. A pharmaceutical composition comprising an antibody of any one of claims 1-39 or a bispecific antibody of any one of claims 40-50 and a pharmaceutically acceptable carrier.

57. Use of an antibody of any one of claims 1-39, a bispecific antibody of any one of claims 40-50, a composition of claim 55 or a pharmaceutical composition of claim 56 as a medicament.

58. Use of an antibody of any one of claims 1-39, a bispecific antibody of any one of claims 40-50, a composition of claim 55 or a pharmaceutical composition of claim 56 for the treatment of a disease.

59. A method for treating a disease, comprising administering to a subject in need an antibody of any one of claims 1-39, a bispecific antibody of any one of claims 40-50, a composition of claim 55 or a pharmaceutical composition of claim 56.

60. Use or method of any one of claims 57-59, wherein the disease is cancer, an infectious disease or an autoimmune disease.

61. A method for diagnosing a disease characterized by the involvement or accumulation of CD3-expressing cells, comprising administering to a subject the antibody according to any one of claims 1-39, the bispecific antibody according to any one of claims 40-50, the composition according to claim 55, or the pharmaceutical composition according to claim 56, optionally wherein the antibody or the bispecific antibody is labeled with a detectable reagent.

62. A method for producing the antibody according to any one of claims 1-39 or the bispecific antibody according to any one of claims 40-50, comprising the following steps a) culturing the host cell according to any one of claims 53-54; and b) purifying the antibody from the culture medium.

63. A diagnostic composition comprising the antibody according to any one of claims 1-39 or the bispecific antibody according to any one of claims 40-50.

64. A method for detecting the presence of CD3 antigen or CD3-expressing cells in a sample, comprising the following steps: a) contacting the sample with the antibody according to any one of claims 1-39 or the bispecific antibody according to any one of claims 40-50 under conditions allowing the formation of a complex between the antibody or bispecific antibody and CD3; and b) analyzing whether a complex is formed.

65. A kit for detecting the presence of CD3 antigen or CD3-expressing cells in a sample, comprising i) the antibody according to any one of claims 1-39 or the bispecific antibody according to any one of claims 40-50; and ii) instructions for use of the kit.

66. An anti-idiotype antibody that binds to the antibody according to any one of claims 1-39.

Images