MAGEB2 Combined Structure

Abstract

The present invention relates to antibody constructs comprising a domain that binds to a MAGEB2 peptide complexed with HLA and, optionally, another domain that binds to CD3. The present disclosure also provides polynucleotides encoding the antibody constructs, vectors comprising the polynucleotides, and host cells transformed or transfected with the polynucleotides or vectors. The present disclosure also provides methods for producing the antibody constructs of the present disclosure, medical uses of the antibody constructs, and kits comprising the antibody constructs.

Description

Technical Field

[0001] Cross - reference to Related Applications This application claims priority to U.S. Provisional Patent Application No. 63 / 027,148, filed May 19, 2020, and U.S. Provisional Patent Application No. 63 / 128,773, filed December 21, 2020. The above applications are hereby incorporated by reference herein in their entirety for all purposes.

[0002] Reference to Sequence Listing This application includes a sequence listing submitted electronically in ASCII format, which is hereby incorporated by reference herein in its entirety. The ASCII copy created on May 14, 2021, is named A - 2635 - WO - PCT sequence listing.txt and is 7,139,825 bytes in size.

[0003] The field of the present invention relates to compositions and methods related to biopharmaceuticals containing bispecific antibody constructs.

Background Art

[0004] The MAGE (melanoma antigen gene) family contains approximately 60 genes classified into several sub - families. The MAGE - A, - B, and - C sub - families are mainly expressed in the testis and are abnormally expressed in various cancer types. The MAGE - D, - E, - F, - G, - H, - L, and - N sub - families are expressed in a wide variety of tissues. See, e.g., Lee and Potts, J. Mol. Biol., 2018. MAGEB2, one of the MAGE family members, is typically expressed only in normal testis. MAGEB2, which can function to enhance the ubiquitin ligase activity of the RING - type zinc - finger - containing E3 ubiquitin protein ligase, has been found to be abnormally expressed in various human tumors such as lung cancer, breast cancer, and melanoma. Given this abnormal expression, MAGEB2 is a potential target for therapeutic agents.

[0005] While many types of cancer have a variety of available treatments, there is still a high level of unmet medical need, and patients are believed to require improved and effective medications.

[0006] Protein-based drugs play a crucial role in almost all areas of medicine, accounting for a large portion of therapeutic drugs under development and already being marketed. Compared to more traditional small molecule drugs, protein-based drugs can exhibit higher specificity and activity at relatively low concentrations, providing treatment for a wide range of impactful diseases, including various cancers, autoimmune diseases, and metabolic disorders (Roberts, Trends Biotechnol. 2014 Jul;32(7):372-80, Wang, Int J Pharm. 1999 Aug 20;185(2):129-88).

[0007] One type of protein-based drug is bispecific antibodies, typically capable of simultaneously binding to two different types of antigens. Known in several structural forms, they are currently being explored for applications in cancer immunotherapy and drug delivery (Fan, Gaowei; Wang, Zujian; Hao, Mingju; Li, Jinming (2015). "Bispecific antibodies and their applications." Journal of Hematology & Oncology. 8:130).

[0008] Bispecific antibodies can be produced in a variety of different forms. For example, they may be IgG-like, i.e., full-length bispecific antibodies, or non-IgG-like bispecific antibodies that are not full-length antibody constructs. Full-length bispecific antibodies typically retain the usual monoclonal antibody (mAb) structure of two Fab arms and one Fc region, except that two Fab sites bind to different antigens. Non-full-length bispecific antibodies completely lack an Fc region. These include chemically linked Fabs consisting only of Fab regions, as well as various types of bivalent and trivalent single-stranded variable fragments (scFv). There are also fusion proteins that mimic the variable domains of two antibodies. An example of this form is the bispecific T cell engager (BiTE®) (Yang, Fa; Wen, Weihong; Qin, Weijun (2016). "Bispecific Antibodies as a Development Platform for New Concepts and Treatment Strategies". International Journal of Molecular Sciences. 18(1):48).

[0009] Although various cancer treatments exist, patients still need more effective treatments. Therefore, the present invention provides a bispecific antibody construct that targets tumor cells expressing MAGEB2 and CD3, which are expressed by immune effector cells. [Prior art documents] [Non-patent literature]

[0010] [Non-Patent Document 1] Lee and Potts, J. Mol. Biol., 2018 [Non-Patent Document 2] Roberts,Trends Biotechnol.2014 Jul;32(7):372-80 [Non-Patent Document 3] Wang,Int J Pharm.1999 Aug 20;185(2):129-88 [Non-Patent Document 4] Fan,Gaowei;Wang,Zujian;Hao,Mingju;Li,Jinming(2015). “Bispecific antibodies and their applications”.Journal of Hematology&Oncology.8:130) [Non-Patent Document 5] Yang, Fa; Wen, Weihong; Qin, Weijun (2016). “Bispecific Antibodies as a Development Platform for New Concepts and Treatment Strategies”. International Journal of Molecular Sciences. 18(1):48) [Overview of the project] [Means for solving the problem]

[0011] In one embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to the MAGEB2 peptide-HLA complex on the surface of a target cell.

[0012] In another embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to a MAGEB2 peptide-HLA complex on the surface of a target cell and a second domain that binds to human CD3 on the surface of a T cell.

[0013] In further embodiments, the present invention provides isolated antibody constructs that bind to the same epitopes as the antibody constructs provided herein.

[0014] In further embodiments, the present invention provides an isolated antibody construct that binds to an epitope containing the MAGEB2 peptide (e.g., SEQ ID NO: 1), wherein the antibody construct binds to at least one of the following MAGEB2 peptide residues: Asp4, Gly5, Glu6, Glu7, His8, Ser9, or Val10.

[0015] In another embodiment, the present invention provides an isolated antibody construct that binds to an epitope comprising both the MAGEB2 peptide (e.g., SEQ ID NO: 1) and HLA-A2, the antibody construct binding to at least one of the following HLA-A2 residues: Arg65, Lys66, Ala69, Gln72, Thr73, Val76, Lys146, Ala149, Ala150, His151, Glu154, Gln155, Gly16, Arg17, Gly18, Glu19, Pro20, Gln43, Lys68, Ser71, Arg75, Gly79, Thr80, Arg82, Gly83, or Glu89.

[0016] In another embodiment, the present invention provides an isolated antibody construct that binds to an epitope containing both the MAGEB2 peptide (e.g., SEQ ID NO: 1) and HLA-A2 / MHC, wherein the antibody construct binds to at least one of the following MAGEB2 residues: Asp4, Gly5, Glu6, Glu7, His8, Ser9, or Val10, and further binds to at least one of the following HLA-A2 residues: Arg65, Lys66, Ala69, Gln72, Thr73, Val76, Lys146, Ala149, Ala150, His151, Glu154, Gln155, Gly16, Arg17, Gly18, Glu19, Pro20, Gln43, Lys68, Ser71, Arg75, Gly79, Thr80, Arg82, Gly83, or Glu89.

[0017] In one embodiment, the present invention provides an isolated antibody construct that binds to an epitope containing both the MAGEB2 peptide (e.g., SEQ ID NO: 1) and HLA-A2 / MHC, wherein the antibody construct contains at least one of the following heavy chain amino acid residues: Ser30, Ser31, His32, Tyr32, Ala33, Ser52, Gly53, Ser54, Gly56, Gly57, Tyr59, Lys100, Gly101, Val102, His103, Leu104, or Gly105.

[0018] In another embodiment, the present invention provides an antibody construct further comprising at least one of the following light chain amino acid residues: Asn25, Asn26, Gly28, Ser29, Lys30, Ser31, His33, Tyr48, Asp49, Asp50, Asn51, Asp52, Arg53, Asn65, Phe66, Ser66, Gly67, Trp90, Tyr92, Arg93, Leu95, or Gln95. [Brief explanation of the drawing]

[0019] [Figure 1] This figure shows the amino acid sequences of MAGE-B2, MAGE-A4, and MAGE-A8 peptides, highlighting the differences between the three sequences. This figure also shows other similar peptides. Sequence IDs 1-3 and 18461-18469 are disclosed in their respective order of appearance. [Figure 2A] This figure illustrates the expression of the MAGEB2 peptide-HLA-A*02:01 complex in various cancer and normal tissues. The upper part of the left panel shows the median relative MS signal intensity from technically repeated measures, plotted as colored dots for single HLA-A*02-positive normal (left side of the figure) and tumor samples in which the peptide was detected (right side of the figure). The right panel shows the results of absolute quantitative analysis (AbsQuant) of the MAGEB2 peptide in 12 different tumor tissues. Each data point on the logarithmic scale corresponds to a repeated measure from one tumor sample. The median peptide copy number per 67 cells is shown by the red line. [Figure 2B]This figure shows the mRNA expression of MAGEB2 in various tumor tissues. [Figure 2C] This figure shows the mRNA expression of MAGEA4 in various tumor tissues. [Figure 2D] This figure shows the mRNA expression of MAGEB2 in various normal tissues. [Figure 2E] This figure shows the mRNA expression of MAGEA4 in various normal tissues. [Figure 2F] This figure shows the mRNA expression of MAGEB2 in various normal tissues. [Figure 3A] This is a diagram illustrating the visual representation of the preparation of a polyvalent peptide-MHC (pMHC) immunogen used to immunize XenoMouse. [Figure 3B] This figure shows the results of titer analysis of immunized XenoMouse. The y-axis represents the GeoMean multiplier relative to the control peptide (soluble), and the x-axis represents various different immunized mouse groups. [Figure 4A] This figure shows the results of cell-based affinity determination by fluorescence cytometry and nonlinear regression (single-site specific binding) analysis on T2 cells loaded with the MAGEB2 peptide. The MAGEB2 pMHC HLE bispecific antibody constructs tested are as follows (having the sequences shown in Table 22 of this specification). 1=MA_03-E11_AS_CC_x_I2C0_x_scFc_(G2B); 2=MA_09-E2_CC_x_I2C0_x_scFc_(V8T); 3=MA_09-F12_CC_x_I2C0_x_scFc_(D1U); 4=MA_09-G10_CC_x_I2C0_x_scFc_(C4K); 5=MA_09-H7_CC_x_I2C0_x_scFc_(T3S); 6=MA_09-H10_CC_x_I2C0_x_scFc_(B6N); 7=MA_10-B5_CC_x_I2C0_x_scFc_(H6H). [Figure 4B]This figure shows the results of a cytotoxic assay against MAGEB2 pMHC containing the cell line DAN-G. The MAGEB2 pMHC HLE BiTE® molecules tested are as follows (having the sequences shown in Table 22 of this specification): 1=MA_03-G10_AS_CC_x_I2C0_x_scFc_(N3H); 2=MA_98-C7_CC_x_I2C0_x_scFc_(Y8P); 3=MA_03-G11_AS_CC_x_I2C0_x_scFc_(O4R); 4=MA_98-G12_AS_CC_x_I2C0_x_scFc_(Q4Z); and 5=MA_03-E7_AS_CC_x_I2C0_x_scFc_(L6M). [Figure 5A] This figure shows the response of the BiTE® molecule MA_10-B5_CC_x_I2C0_x_scFc_(H6H) to 112 MAGE-B2-like peptides, MAGE-B2, and related controls in two different T cell donors. The y-axis represents %T2Luc survival rate (donor RG1198), and the x-axis represents %T2Luc survival rate (donor 330). [Figure 5B] This figure shows the response of the BiTE® molecule MA_03-G10_AS CC_x_I2C0_x_scFc_(N3H) to 112 MAGE-B2-like peptides, MAGE-B2, and related controls in two different T cell donors. The y-axis represents %T2Luc survival rate (donor RG1198), and the x-axis represents %T2Luc survival rate (donor 330). [Figure 6A] This figure shows the efficacy of the MA_10-B5_CC_x_I2C0_x_scFc_(H6H)BiTE® molecule against MAGE-B2 and various reactive analog peptides. The x-axis represents %T2Luc survival rate, and the x-axis represents log concentration [nM]. [Figure 6B] This figure shows the efficacy of the MA_03-G10_AS CC_x_I2C0_x_scFc_(N3H)BiTE(registered trademark) molecule against MAGE-B2 and various reactive analog peptides. The x-axis represents %T2Luc survival rate, and the x-axis represents log concentration [nM]. [Figure 7]This figure shows the efficacy of two different BiTE® molecules against a full-length 10-mer MAGEB2 peptide compared to a 9-mer deletion variant of the MAGEB2 peptide. [Figure 8] This figure summarizes the TDCC activity of 10 BiTE® molecules in four different cell lines with known MAGEB2 pMHC copies per cell, presented in tabular form. Note that the three-letter names of the BiTE® molecules are abbreviated versions of the complete BiTE® molecular nomenclature as seen in Table 22 of this specification, and the three-letter names are enclosed in parentheses. [Figure 9] This figure shows TDCC plots of three different BiTE® molecules on cell lines with moderately low expression of MAGEB2. The y-axis represents % specific cytotoxicity, and the x-axis represents log BiTE® molecule concentration (pM). Note that the three-letter names of BiTE® molecules are abbreviated versions of the full BiTE® molecular nomenclature as seen in Table 22 of this specification, and the three-letter names are enclosed in parentheses. [Figure 10] This figure shows TDCC plots for three different BiTE® molecules on the DAN-G MAGEB2(-) cell line. The y-axis represents % specific cytotoxicity, and the x-axis represents log BiTE® molecule concentration (pM). Note that the three-letter names of BiTE® molecules are abbreviated versions of the full BiTE® molecular nomenclature as seen in Table 22 of this specification, and the three-letter names are enclosed in parentheses. [Figure 11A] The TDCC plots for two different BiTE® molecules on primary human cells (cardiomyocytes, endothelial cells, and epithelial cells) are shown, with the y-axis representing % specific lysis (cytotoxicity) and the x-axis representing BiTE® molecule concentration (nM). [Figure 11B] This graph shows BiTE® molecule-induced target cell lysis on MAGEB2-negative cells, with the y-axis representing % cytotoxicity and the x-axis representing log BiTE® molecule concentration (pM). Note that the three-letter BiTE® molecule name is an abbreviated version of the complete BiTE® molecule nomenclature as seen in Table 22 of this specification, and the three-letter name is enclosed in parentheses. [Figure 12A] The TDCC plots for various BiTE® molecules on MAGEA4(+) / MAGEB2(-) cell lines (NCI-H1703 and SCaBER) are shown, with the y-axis representing % specific cytotoxicity and the x-axis representing log BiTE® molecule concentration (pM). Note that the three-letter names of BiTE® molecules are abbreviated versions of the complete BiTE® molecular nomenclature as seen in Table 22 of this specification, and the three-letter names are enclosed in parentheses. [Figure 12B] This is a tabular representation of the data from Figure 12A, highlighting the difference in activity between Family 4 molecules and Family 7 molecules, with only Family 7 molecules showing activity against MAGE-A4 expressing cells. [Figure 13A] This diagram shows the structure of the MAGEB2 peptide interaction with MHC. The structure on the left is the protein structure representation of the peptide-MHC complex (pMHC), and the structure on the right is the electrostatic surface representation of the pMHC. [Figure 13B] This diagram shows the structure of the MAGEB2 peptide interaction with other HLA-A*02 serotypes. [Figure 14] This figure shows the overall structure of Family 4 L7E scFv SEQ115734-bound mage-b2 HLA-A at 1.9 Å. [Figure 15] This figure shows the structure of Family 4 L7E scFv SEQ115734-bound MAGE-B2 HLA-A, including the overall structure at 1.9 Å. The left panel highlights the CDR interaction between the MAGEB2 peptide and MHC, while the right panel highlights the light-chain (LC) and heavy-chain (HC) interactions with MHC. [Figure 16] This figure shows the structure of Family 4 L7E scFv SEQ115734 bound to MAGEB2 HLA-A, its overall structure at 1.9 Å, and specific water-mediated interactions. [Figure 17] This figure shows the structures of the H6N scFv heavy chain (HC) interaction with the MAGEB2 peptide, and the H6N scFv light chain (LC) interaction with HLA and the MAGEB2 peptide. [Figure 18]This figure shows the structures of the H6N scFv heavy chain (HC) interaction with the MAGEB2 peptide, and the H6N scFv light chain (LC) interaction with HLA and the MAGEB2 peptide. [Figure 19] This figure shows the structures of the H6N scFv heavy chain (HC) interaction with the MAGEB2 peptide, and the H6N scFv light chain (LC) interaction with HLA and the MAGEB2 peptide. [Figure 20] This figure shows the structures of the H6N scFv heavy chain (HC) interaction with the MAGEB2 peptide, and the H6N scFv light chain (LC) interaction with HLA and the MAGEB2 peptide. [Figure 21] This figure shows the structures of the N3H scFv heavy chain (HC) interaction with the MAGEB2 peptide, and the N3H scFv light chain (LC) interaction with HLA and the MAGEB2 peptide. [Figure 22] This figure shows the structures of the N3H scFv heavy chain (HC) interaction with the MAGEB2 peptide, and the N3H scFv light chain (LC) interaction with HLA and the MAGEB2 peptide. [Figure 23] This figure shows the structures of the N3H scFv heavy chain (HC) interaction with the MAGEB2 peptide, and the N3H scFv light chain (LC) interaction with HLA and the MAGEB2 peptide. [Figure 24] This figure shows the structures of the N3H scFv heavy chain (HC) interaction with the MAGEB2 peptide, and the N3H scFv light chain (LC) interaction with HLA and the MAGEB2 peptide. [Figure 25] This figure summarizes the sequence alignments of three different families of 4BiTE® molecules. Sequence IDs 18470-18472, 556, 556, and 576 are disclosed in their respective order of appearance. [Figure 26] This diagram summarizes the sequence alignments of three different families of 7 BiTE® molecules. Sequence IDs 18473-18478 are disclosed in the order of their appearance. [Figure 27]This figure summarizes the activity, measured by result-IL-2 production, of three different BiTE® molecules that express different HLA-A serotypes and bind to cell lines exogenously loaded with the MAGEB2 peptide of SEQ ID NO: 1. [Figure 28] This figure shows the overall structure of family 7 10B5(H6H) bound to MAGEB2 HLA-A at 2.02 Å. [Figure 29] This figure shows the structure of Family 7 10B5(H6H) in the right panel compared to the structure of Family 4 molecule N3H in the left panel. The different interactions observed and described in Example 10 of this specification are highlighted. [Figure 30] This figure shows a portion of the water-mediated bonding network elucidated by the crystal structures of 10B5(H6H) and MAGEB2 pMHC interactions. Water is represented as black spheres and is further described in Example 10 of this specification. [Figure 31] This figure shows the complementary surface of the heavy chain to the pMHC surface that facilitates this unique recognition. The figure panel shows vacuum electrostatic surfaces pMHC (left panel) and HC CDR (right panel) in grayscale, which are further described in Example 10 of this specification. [Figure 32] This figure is a graph summary of data from an in vivo tumor study in which animals possessing HCT-116 were treated with the N3H molecule at three different time points throughout the course of the study, as further detailed in Example 11 herein. [Figure 33] This figure shows a graphical summary of data from two treatment groups that received different doses in an in vivo tumor study. Animals with HCT-116 were treated with the N3H molecule at three different time points throughout the course of the study, as further detailed in Example 11 herein. The upper panel shows results from the 2.0 mg / kg treatment group, and the lower panel shows results from the 0.2 mg / kg treatment group. [Figure 34] This figure is a graph summary of data from an in vivo tumor study in which animals possessing HCT-116 were treated with the H6H molecule at three different time points throughout the course of the study, as further detailed in Example 11 herein. [Figure 35] This figure shows a graphical summary of data from two treatment groups that received different doses in an in vivo tumor study. Animals with HCT-116 were treated with the H6H molecule at three different time points throughout the course of the study, as further detailed in Example 11 herein. The upper panel shows results from the 2.0 mg / kg treatment group, and the lower panel shows results from the 0.2 mg / kg treatment group. [Figure 36] Figures 32-34 summarize the data from pharmacokinetic (pK) studies performed on samples taken from animal studies, which are further detailed in Example 11 of this specification. The left panel summarizes the pK data from doses of 2.0 mg / kg and 0.2 mg / kg in the N3H molecular treatment group, and the right panel summarizes the pK data from doses of 2.0 mg / kg and 0.2 mg / kg in the H6H molecular treatment group. [Figure 37] This figure contains six different graph summaries of TDCC assays that target cells expressing various different HLAs using the N3H molecule. It demonstrates effective cell targeting and killing of cells expressing HLA A*02:01, HLA A*02:05, HLA A*02:06, and HLA A*02:07, further detailed in Example 13 herein. The y-axis represents % specific cytotoxicity, and the x-axis represents log MAGEB2 peptide concentration (M). [Figure 38] This figure summarizes FACS analysis performed on cells expressing various different HLAs and pulsed with 50 μM and 0.05 μM MAGEB2 peptides (and unpulsed controls). It demonstrates the effective loading of HLA A*02:01, HLA A*02:05, HLA A*02:06, and HLA A*02:07 peptides, which are further detailed in Example 13 herein. [Figure 39]This figure summarizes FACS analysis performed on cells expressing various different HLAs and pulsed with 50 μM and 0.05 μM MAGEB2 peptides (and unpulsed controls), with FACS gating set for D5 cells for each concentration to better illustrate the loading. It demonstrates the effective loading with HLA A*02:01, HLA A*02:05, HLA A*02:06, and HLA A*02:07 peptides, which are further detailed in Example 13 herein. [Figure 40] This diagram summarizes the structure of the in silico modeling workflow for Family 4 molecules. [Figure 41] This figure provides overviews of both the L7E molecular crystal structure and the N3H molecular crystal structure overlaid with the N3H in silico homology model. Further details are provided in Example 14 of this specification. [Figure 42] This figure provides a comparison of the L7E crystal structures and homologous in silico models for WT, L6A, N3H, and Y8P. Further details are provided in Example 14 of this specification. [Figure 43] This figure provides an in silico model comparison between WT and L7E at light chain positions 92, 95, 96, and 97. Further details are provided in Example 14 of this specification. [Figure 44] This figure provides an in silico model comparison between WT and L6A at light chain positions 92, 95, 96, and 97. Further details are provided in Example 14 of this specification. [Figure 45] This figure provides an in silico model comparison between WT and N3H at light chain positions 92, 95, 96, and 97. Further details are provided in Example 14 of this specification. [Figure 46] This figure provides an in silico model comparison between WT and Y8P at light chain positions 92, 95, 96, and 97. Further details are provided in Example 14 of this specification. [Figure 47] This figure provides in silico model predictive alternative residues for light chain positions 92, 95, 96, and 97 of the parent molecule. Further details are provided in Example 14 of this specification. [Figure 48]This figure provides in silico model predictive alternative residues for light chain positions 92, 95, 96, and 97 of the L6A molecule. Further details are provided in Example 14 of this specification. [Figure 49] This figure provides in silico model predictive alternative residues for light chain positions 92, 95, 96, and 97 of the N3H molecule. Further details are provided in Example 14 of this specification. [Figure 50] This figure provides in silico model predictive alternative residues for light chain positions 92, 95, 96, and 97 of the Y8P molecule. Further details are provided in Example 14 of this specification. [Figure 51A] This figure provides a graphical summary of the results for N3H, H6H, and control molecules against tumor organoids using T cell donors. The y-axis represents IFNg concentration in pg / ml units, and the x-axis represents different test molecules. [Figure 51B] This figure provides a graphical summary of the results for N3H, H6H, and control molecules against tumor organoids using T cell donors. The y-axis represents IFNg concentration in pg / ml units, and the x-axis represents different test molecules. [Figure 52A] This figure provides a graphical summary of the results for the N3H molecule in a TDCC assay using Dan-G target cells. The x-axis represents the log BiTE® molecule concentration (nM), and the y-axis represents % cytotoxicity. [Figure 52B] This figure provides a graphical summary of the results for the H6H molecule (Figure 52B) in a TDCC assay using Dan-G targeted cells. The x-axis represents the log BiTE® molecule concentration (nM), and the y-axis represents % cytotoxicity. [Modes for carrying out the invention]

[0020] HLA class I and HLA class II (or MHC class I and MHC class II) molecules on the cell surface present antigens to the immune system. In some cases, endogenously produced proteins such as MAGEB2 are proteolytically cleaved into peptides within the cell and presented on the cell surface via MHC class I molecules. Typically, these peptides are 8-13 amino acid residues long and include "anchor residues" that help bind the peptides to the binding grooves of MHC molecules.

[0021] As described above, MAGEB2 is abnormally expressed in various human cancer types. MAGEB2 is not typically expressed on the cell surface. However, the MAGEB2 peptide is presented on the surface of tumor cells by MHC class I molecules. In particular, the MAGEB2 peptide GVYDGEEHSV (SEQ ID NO: 1) is presented on the surface of tumor cells as a peptide-MHC ("pMHC") complex by MHC class I molecules. See, for example, U.S. Patent Application Publication 2016 / 0250307A1 (U.S. Patent Application No. 14 / 975,952) and U.S. Patent Application Publication 2017 / 0080070A1 (U.S. Patent Application No. 15 / 357,757).

[0022] Given that the MAGEB2 peptide is presented on the surface of cells (e.g., cancer cells), albeit in association with a peptide-MHC complex, it is potentially targetable by specific conjugates such as monoclonal antibodies or bispecific antibody constructs. Generating antibodies, bispecific antibody constructs, or other conjugate constructs that target this peptide-MHC complex presents unique and challenging problems due to MHC expression in almost all cells in the body and the potential for harmful binding to MHC that does not present the MAGEB2 peptide. This is compounded by the fact that the peptide is extremely small compared to the MHC complex. As described herein, the conjugate constructs of the present invention overcome these challenges.

[0023] As used herein, the term “pMAGE-HLA” refers to the MAGEB2 peptide GVYDGEEHSV (SEQ ID NO: 1), the MAGEA4 peptide GVYDGREHTV (SEQ ID NO: 2), and / or the MAGEA8 peptide GLYDGREHSV (SEQ ID NO: 3) when complexed with an HLA (e.g., HLA-A*02:01). Alternatively, as used herein, the term “pMHC” refers to the same thing and may be used interchangeably, or it may be a more general term simply referring to a peptide-MHC(HLA) complex.

[0024] A bispecific antibody construct containing one domain that binds to CD3 expressed on the surface of T cells and another domain that binds to a target protein expressed on target cells directly ligates T cells to target cells, inducing T cell retargeting lysis. This mechanism of action differs from chemotherapy, other types of targeted therapies, and other immunotherapies in that it can act with any CD3-positive T cells independently of costimulatory activation signals (Klinger et al., Immunol Reviews 2016). The presence of pMAGE-HLA on the cell surface of cancer cells such as non-small cell lung cancer, hepatocellular carcinoma, or head and neck cancer provides a basis for targeting these cancers with bispecific antibody constructs that bind to pMAGE-HLA (e.g., the MAGEB2 peptide of SEQ ID NO: 1 complexed with HLA) and CD3.

[0025] In certain embodiments of the present invention, an antibody construct is provided comprising a bispecific antibody construct that specifically targets pMAGE-HLA derived from MAGEB2 associated with malignant tumors. Further advancing the present invention, MAGEB2 may be identified as a gene that is upregulated and / or abnormally expressed in tumors compared to normal tissue expression of MAGEB2. In this regard, it may be shown that the MAGEB2 protein or a specific MAGEB2 peptide is expressed in or presented on the cell surface in a particular tumor according to methods known in the art, such as genetic analysis, immunohistochemistry, or mass spectrometry.

[0026] In one embodiment, the Disclosure provides a method for initiating a T cell-mediated immune response against target cells or tissues of interest. In some embodiments, the method involves administering modified T cells, such as CAR-T cells, containing a nucleic acid encoding a binding domain or antibody construct, e.g., a bispecific antibody, that specifically targets pMAGE-HLA (e.g., MAGEB2 peptide), to a subject requiring such administration. In another embodiment, the modified T cells express a binding domain or antibody construct, e.g., a bispecific antibody, that can be immobilized on the cell surface of the modified T cells or secreted from the modified T cells.

[0027] One specific example of mass spectrometry that can be used to demonstrate MAGEB2 expression by cells is a high-throughput platform based on ultra-high-sensitivity mass spectrometry known as XPRESIDENT®(www.immatics.com / x-president.html). This high-throughput platform identifies HLA-binding peptides presented on tumor cells and has extremely high sensitivity, enabling detection of these HLA-binding peptides at the atomol level. All XPRESIDENT® peptides are derived from natural tumors (20 major cancer indications), including tissues from primary and metastatic biopsies as well as healthy organs (40 of the most important organs in the human body). This target database contains over 2000 tissue samples. Peptides are analyzed and identified by a combination of quantitative HLA-peptidality (mass spectrometry) complemented by quantitative transcriptomics (mRNA sequencing), enabling analysis of differential expression and presentation of these potential drug targets between tumors and normal tissues. All HLA-restricted targets discovered by XPRESIDENT® on any allele have been proven to be present in patient cancer tissue, contrary to predictions made by in silico technology. See also U.S. Patent Nos. 10,545,154, 9,791,443, and 7,811,828.

[0028] The presence of pMAGE-HLA on the cell surface of certain cancer cell lines can also be demonstrated by flow cytometry. Therefore, MAGEB2, and in particular the MAGEB2 peptide GVYDGEEHSV (SEQ ID NO: 1) complexed with HLA, can be identified as an effective target associated with specific cancer types.

[0029] In relation to the present invention, it is a remarkable discovery that the bispecific antibody construct according to the present invention targets the MAGEB2 peptide GVYDGEEHSV (SEQ ID NO: 1), which is presented on the surface of cancer cells by MHC class I molecules with very high specificity, as will be further discussed herein. The absence of non-cancer cell killing by the bispecific antibody construct according to the present invention can also be confirmed in vitro and in vivo.

[0030] In relation to the present invention, preferred bispecific antibody constructs are expected to exhibit not only a favorable ratio of cytotoxicity to affinity, but also sufficient stability properties to facilitate practical handling in formulation, storage, and administration of the constructs. Sufficient stability is characterized by a high monomer content (i.e., non-aggregated and / or non-associated native molecules) after standard preparation, such as at least 65%, more preferably at least 70%, and even more preferably at least 75%, as determined, for example, by preparative size exclusion chromatography (SEC). Furthermore, the turbidity measured at 340 nm as light absorption at a concentration of 2.5 mg / ml should preferably be 0.025 or less, more preferably 0.020 or less, to conclude, for example, that undesirable aggregates are essentially absent. Advantageously, the high monomer content is maintained after freezing / thawing or incubation under stress conditions such as incubation at 37 or 40°C.

[0031] pMAGE-HLA binding molecule The present invention provides an antibody construct comprising a domain that binds to pMAGE-HLA and, optionally, another domain that binds to CD3.

[0032] Table 1 below provides the amino acid sequences of exemplary MAGEB2-binding molecules VH-CDR and VL-CDR. Table 2 below provides the amino acid sequences of the exemplary MAGEB2-binding molecule VH domain and VL domain.

[0033] In certain embodiments, in addition to binding to the MAGEB2 peptide (SEQ ID NO: 1) complexed with an MHC molecule, the binding molecule may also bind to the MAGEA4 peptide (e.g., SEQ ID NO: 2) and / or the MAGEA8 peptide (e.g., SEQ ID NO: 3) complexed with an MHC molecule.

[0034] [Table 1]

[0035] [Table 2]

[0036] In one embodiment, the present invention provides an isolated binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain a) A VH region including CDR-H1 shown in [H6H]SEQ ID NO: 439, CDR-H2 shown in SEQ ID NO: 440, and CDR-H3 shown in SEQ ID NO: 441, and a VL region including CDR-L1 shown in SEQ ID NO: 223, CDR-L2 shown in SEQ ID NO: 224, and CDR-L3 shown in SEQ ID NO: 225; or b) [N3H] A VH region including CDR-H1 shown in SEQ ID NO: 361, CDR-H2 shown in SEQ ID NO: 362, and CDR-H3 shown in SEQ ID NO: 363, and a VL region including CDR-L1 shown in SEQ ID NO: 145, CDR-L2 shown in SEQ ID NO: 146, and CDR-L3 shown in SEQ ID NO: 147; or c) A VH region including CDR-H1 shown in [Y8P]SEQ ID NO: 523, CDR-H2 shown in SEQ ID NO: 524, and CDR-H3 shown in SEQ ID NO: 525, and a VL region including CDR-L1 shown in SEQ ID NO: 307, CDR-L2 shown in SEQ ID NO: 308, and CDR-L3 shown in SEQ ID NO: 309; or d) [L7E] A VH region including CDR-H1 shown in SEQ ID NO: 331, CDR-H2 shown in SEQ ID NO: 332, and CDR-H3 shown in SEQ ID NO: 333, and a VL region including CDR-L1 shown in SEQ ID NO: 115, CDR-L2 shown in SEQ ID NO: 116, and CDR-L3 shown in SEQ ID NO: 117; or e) A VH region including CDR-H1 shown in sequence number 337, CDR-H2 shown in sequence number 338, and CDR-H3 shown in sequence number 339, and a VL region including CDR-L1 shown in sequence number 121, CDR-L2 shown in sequence number 122, and CDR-L3 shown in sequence number 123; or f) The VH region shown in [H6H]Sequence ID 629 and the VL region shown in Sequence ID 628; or g) The VH region shown in [N3H]SEQ ID NO: 577, and the VL region shown in SEQ ID NO: 576; or h) The VH region shown in [Y8P] Sequence ID 684, and the VL region shown in Sequence ID 685; or i) The VH region shown in [L7E] Sequence ID 557 and the VL region shown in Sequence ID 556; or j) The VH region shown in [H6N]Sequence ID 561, and the VL region shown in Sequence ID 560 Includes.

[0037] In one embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain a) A VH region including CDR-H1 shown in [H6H]SEQ ID NO: 439, CDR-H2 shown in SEQ ID NO: 440, and CDR-H3 shown in SEQ ID NO: 441, and a VL region including CDR-L1 shown in SEQ ID NO: 223, CDR-L2 shown in SEQ ID NO: 224, and CDR-L3 shown in SEQ ID NO: 225; or b) [N3H] A VH region including CDR-H1 shown in SEQ ID NO: 361, CDR-H2 shown in SEQ ID NO: 362, and CDR-H3 shown in SEQ ID NO: 363, and a VL region including CDR-L1 shown in SEQ ID NO: 145, CDR-L2 shown in SEQ ID NO: 146, and CDR-L3 shown in SEQ ID NO: 147; or c) A VH region including CDR-H1 shown in [Y8P]SEQ ID NO: 523, CDR-H2 shown in SEQ ID NO: 524, and CDR-H3 shown in SEQ ID NO: 525, and a VL region including CDR-L1 shown in SEQ ID NO: 307, CDR-L2 shown in SEQ ID NO: 308, and CDR-L3 shown in SEQ ID NO: 309; or d) [L7E] A VH region including CDR-H1 shown in SEQ ID NO: 331, CDR-H2 shown in SEQ ID NO: 332, and CDR-H3 shown in SEQ ID NO: 333, and a VL region including CDR-L1 shown in SEQ ID NO: 115, CDR-L2 shown in SEQ ID NO: 116, and CDR-L3 shown in SEQ ID NO: 117; or e) A VH region including CDR-H1 shown in sequence number 337, CDR-H2 shown in sequence number 338, and CDR-H3 shown in sequence number 339, and a VL region including CDR-L1 shown in sequence number 121, CDR-L2 shown in sequence number 122, and CDR-L3 shown in sequence number 123; or f) The VH region shown in [H6H]Sequence ID 629 and the VL region shown in Sequence ID 628; or g) The VH region shown in [N3H]SEQ ID NO: 577, and the VL region shown in SEQ ID NO: 576; or h) The VH region shown in [Y8P] Sequence ID 684, and the VL region shown in Sequence ID 685; or i) The VH region shown in [L7E] Sequence ID 557 and the VL region shown in Sequence ID 556; or j) The VH region shown in [H6N]Sequence ID 561, and the VL region shown in Sequence ID 560 Includes.

[0038] In one embodiment, the present invention provides a T cell receptor (TCR) comprising an isolated binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain a) A VH region including CDR-H1 shown in [H6H]SEQ ID NO: 439, CDR-H2 shown in SEQ ID NO: 440, and CDR-H3 shown in SEQ ID NO: 441, and a VL region including CDR-L1 shown in SEQ ID NO: 223, CDR-L2 shown in SEQ ID NO: 224, and CDR-L3 shown in SEQ ID NO: 225; or b) [N3H] A VH region including CDR-H1 shown in SEQ ID NO: 361, CDR-H2 shown in SEQ ID NO: 362, and CDR-H3 shown in SEQ ID NO: 363, and a VL region including CDR-L1 shown in SEQ ID NO: 145, CDR-L2 shown in SEQ ID NO: 146, and CDR-L3 shown in SEQ ID NO: 147; or c) A VH region including CDR-H1 shown in [Y8P]SEQ ID NO: 523, CDR-H2 shown in SEQ ID NO: 524, and CDR-H3 shown in SEQ ID NO: 525, and a VL region including CDR-L1 shown in SEQ ID NO: 307, CDR-L2 shown in SEQ ID NO: 308, and CDR-L3 shown in SEQ ID NO: 309; or d) [L7E] A VH region including CDR-H1 shown in SEQ ID NO: 331, CDR-H2 shown in SEQ ID NO: 332, and CDR-H3 shown in SEQ ID NO: 333, and a VL region including CDR-L1 shown in SEQ ID NO: 115, CDR-L2 shown in SEQ ID NO: 116, and CDR-L3 shown in SEQ ID NO: 117; or e) A VH region including CDR-H1 shown in sequence number 337, CDR-H2 shown in sequence number 338, and CDR-H3 shown in sequence number 339, and a VL region including CDR-L1 shown in sequence number 121, CDR-L2 shown in sequence number 122, and CDR-L3 shown in sequence number 123; or f) The VH region shown in [H6H]Sequence ID 629 and the VL region shown in Sequence ID 628; or g) The VH region shown in [N3H]SEQ ID NO: 577, and the VL region shown in SEQ ID NO: 576; or h) The VH region shown in [Y8P] Sequence ID 684, and the VL region shown in Sequence ID 685; or i) The VH region shown in [L7E] Sequence ID 557 and the VL region shown in Sequence ID 556; or j) The VH region shown in [H6N]Sequence ID 561, and the VL region shown in Sequence ID 560 Includes.

[0039] In one embodiment, the present invention provides an isolated binding domain that binds to pMAGE-HLA on the surface of a target cell, the binding domain binding to the same epitope as an antibody or antibody construct comprising: a) A VH region including CDR-H1 shown in [H6H]SEQ ID NO: 439, CDR-H2 shown in SEQ ID NO: 440, and CDR-H3 shown in SEQ ID NO: 441, and a VL region including CDR-L1 shown in SEQ ID NO: 223, CDR-L2 shown in SEQ ID NO: 224, and CDR-L3 shown in SEQ ID NO: 225; or b) [N3H] A VH region including CDR-H1 shown in SEQ ID NO: 361, CDR-H2 shown in SEQ ID NO: 362, and CDR-H3 shown in SEQ ID NO: 363, and a VL region including CDR-L1 shown in SEQ ID NO: 145, CDR-L2 shown in SEQ ID NO: 146, and CDR-L3 shown in SEQ ID NO: 147; or c) A VH region including CDR-H1 shown in [Y8P]SEQ ID NO: 523, CDR-H2 shown in SEQ ID NO: 524, and CDR-H3 shown in SEQ ID NO: 525, and a VL region including CDR-L1 shown in SEQ ID NO: 307, CDR-L2 shown in SEQ ID NO: 308, and CDR-L3 shown in SEQ ID NO: 309; or d) [L7E] A VH region including CDR-H1 shown in SEQ ID NO: 331, CDR-H2 shown in SEQ ID NO: 332, and CDR-H3 shown in SEQ ID NO: 333, and a VL region including CDR-L1 shown in SEQ ID NO: 115, CDR-L2 shown in SEQ ID NO: 116, and CDR-L3 shown in SEQ ID NO: 117; or e) A VH region including CDR-H1 shown in sequence number 337, CDR-H2 shown in sequence number 338, and CDR-H3 shown in sequence number 339, and a VL region including CDR-L1 shown in sequence number 121, CDR-L2 shown in sequence number 122, and CDR-L3 shown in sequence number 123; or f) The VH region shown in [H6H]Sequence ID 629 and the VL region shown in Sequence ID 628; or g) The VH region shown in [N3H]SEQ ID NO: 577, and the VL region shown in SEQ ID NO: 576; or h) The VH region shown in [Y8P] Sequence ID 684, and the VL region shown in Sequence ID 685; or i) The VH region shown in [L7E] Sequence ID 557 and the VL region shown in Sequence ID 556; or j) The VH region shown in [H6N]Sequence ID 561, and the VL region shown in Sequence ID 560 Includes.

[0040] In one embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain binds to the same epitope as an antibody or antibody construct comprising the following: a) The VH region including CDR-H1 shown in [H6H]SEQ ID NO: 439, CDR-H2 shown in SEQ ID NO: 440, and CDR-H3 shown in SEQ ID NO: 441, and the VL region including CDR-L1 shown in SEQ ID NO: 223, CDR-L2 shown in SEQ ID NO: 224, and CDR-L3 shown in SEQ ID NO: 225; b) [N3H] The VH region includes CDR-H1 shown in SEQ ID NO: 361, CDR-H2 shown in SEQ ID NO: 362, and CDR-H3 shown in SEQ ID NO: 363, and the VL region includes CDR-L1 shown in SEQ ID NO: 145, CDR-L2 shown in SEQ ID NO: 146, and CDR-L3 shown in SEQ ID NO: 147; c) A VH region including CDR-H1 shown in sequence number 523 of [Y8P], CDR-H2 shown in sequence number 524, and CDR-H3 shown in sequence number 525, and a VL region including CDR-L1 shown in sequence number 307, CDR-L2 shown in sequence number 308, and CDR-L3 shown in sequence number 309; d) [L7E] A VH region including CDR-H1 shown in SEQ ID NO: 331, CDR-H2 shown in SEQ ID NO: 332, and CDR-H3 shown in SEQ ID NO: 333, and a VL region including CDR-L1 shown in SEQ ID NO: 115, CDR-L2 shown in SEQ ID NO: 116, and CDR-L3 shown in SEQ ID NO: 117; e) [H6N] A VH region including CDR-H1 shown in SEQ ID NO: 337, CDR-H2 shown in SEQ ID NO: 338, and CDR-H3 shown in SEQ ID NO: 339, and a VL region including CDR-L1 shown in SEQ ID NO: 121, CDR-L2 shown in SEQ ID NO: 122, and CDR-L3 shown in SEQ ID NO: 123; f) The VH region shown in sequence number 629 of [H6H] and the VL region shown in sequence number 628 of [H6H]; g) The VH region shown in [N3H]Sequence ID 577 and the VL region shown in Sequence ID 576; h) The VH region shown in sequence number 684 of [Y8P] and the VL region shown in sequence number 685; i) The VH region shown in [L7E] Sequence ID 557 and the VL region shown in Sequence ID 556; or j) The VH region shown in [H6N]Sequence ID 561, and the VL region shown in Sequence ID 560 Includes.

[0041] In one embodiment, the present invention provides a T cell receptor (TCR) comprising an isolated binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain binds to the same epitope as an antibody or antibody construct comprising the following: a) A VH region including CDR-H1 shown in [H6H]SEQ ID NO: 439, CDR-H2 shown in SEQ ID NO: 440, and CDR-H3 shown in SEQ ID NO: 441, and a VL region including CDR-L1 shown in SEQ ID NO: 223, CDR-L2 shown in SEQ ID NO: 224, and CDR-L3 shown in SEQ ID NO: 225; or b) [N3H] A VH region including CDR-H1 shown in SEQ ID NO: 361, CDR-H2 shown in SEQ ID NO: 362, and CDR-H3 shown in SEQ ID NO: 363, and a VL region including CDR-L1 shown in SEQ ID NO: 145, CDR-L2 shown in SEQ ID NO: 146, and CDR-L3 shown in SEQ ID NO: 147; or c) A VH region including CDR-H1 shown in [Y8P]SEQ ID NO: 523, CDR-H2 shown in SEQ ID NO: 524, and CDR-H3 shown in SEQ ID NO: 525, and a VL region including CDR-L1 shown in SEQ ID NO: 307, CDR-L2 shown in SEQ ID NO: 308, and CDR-L3 shown in SEQ ID NO: 309; or d) [L7E] A VH region including CDR-H1 shown in SEQ ID NO: 331, CDR-H2 shown in SEQ ID NO: 332, and CDR-H3 shown in SEQ ID NO: 333, and a VL region including CDR-L1 shown in SEQ ID NO: 115, CDR-L2 shown in SEQ ID NO: 116, and CDR-L3 shown in SEQ ID NO: 117; or e) A VH region including CDR-H1 shown in sequence number 337, CDR-H2 shown in sequence number 338, and CDR-H3 shown in sequence number 339, and a VL region including CDR-L1 shown in sequence number 121, CDR-L2 shown in sequence number 122, and CDR-L3 shown in sequence number 123; or f) The VH region shown in [H6H]Sequence ID 629 and the VL region shown in Sequence ID 628; or g) The VH region shown in [N3H]SEQ ID NO: 577, and the VL region shown in SEQ ID NO: 576; or h) The VH region shown in [Y8P] Sequence ID 684, and the VL region shown in Sequence ID 685; or i) The VH region shown in [L7E] Sequence ID 557 and the VL region shown in Sequence ID 556; or j) The VH region shown in [H6N]Sequence ID 561, and the VL region shown in Sequence ID 560 Includes.

[0042] In another embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain binds to the same epitope as an antibody or antibody construct comprising a VH region selected from the group consisting of any VH and / or VL regions as shown in any of the sequences in Table 27 (SEQ ID NOs. 550-697), 35 (SEQ ID NOs. 3083-3118), 43 (SEQ ID NOs. 3979-4010), 51 (SEQ ID NOs. 5023-5118), 57 (SEQ ID NOs. 9351-10066), or 67 (SEQ ID NOs. 16667-16718).

[0043] In another embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of target cells, wherein the binding domain is derived from CDR-H1, C from a VH region selected from the group consisting of any VH regions represented by any of the sequences in Table 24 (SEQ ID NOs: 322-537), 32 (SEQ ID NOs: 2984-3064), 40 (SEQ ID NOs: 3891-3962), 48 (SEQ ID NOs: 4759-4974), 56 (SEQ ID NOs: 7561-9350), or 66 (SEQ ID NOs: 16537-16666). The antibody or antibody construct containing DR-H2 and CDR-H3 binds to the same epitope, and / or the binding domain further comprises CDR-L1, CDR-L2, and CDR-L3 from a VL region selected from the group consisting of any VL regions shown in any of the sequences in Table 23 (SEQ ID NOs: 106-321), 31 (SEQ ID NOs: 2975-3055), 39 (SEQ ID NOs: 3883-3954), 47 (SEQ ID NOs: 4735-4950), 55 (SEQ ID NOs: 7203-8992), or 65 (SEQ ID NOs: 16511-16640).

[0044] In one embodiment, the present invention provides an isolated antibody construct comprising a first binding domain that binds to pMAGE-HLA on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain is a) A VH region including CDR-H1 shown in [H6H]SEQ ID NO: 439, CDR-H2 shown in SEQ ID NO: 440, and CDR-H3 shown in SEQ ID NO: 441, and a VL region including CDR-L1 shown in SEQ ID NO: 223, CDR-L2 shown in SEQ ID NO: 224, and CDR-L3 shown in SEQ ID NO: 225; or b) [N3H] A VH region including CDR-H1 shown in SEQ ID NO: 361, CDR-H2 shown in SEQ ID NO: 362, and CDR-H3 shown in SEQ ID NO: 363, and a VL region including CDR-L1 shown in SEQ ID NO: 145, CDR-L2 shown in SEQ ID NO: 146, and CDR-L3 shown in SEQ ID NO: 147; or c) A VH region including CDR-H1 shown in [Y8P]SEQ ID NO: 523, CDR-H2 shown in SEQ ID NO: 524, and CDR-H3 shown in SEQ ID NO: 525, and a VL region including CDR-L1 shown in SEQ ID NO: 307, CDR-L2 shown in SEQ ID NO: 308, and CDR-L3 shown in SEQ ID NO: 309; or d) [L7E] A VH region including CDR-H1 shown in SEQ ID NO: 331, CDR-H2 shown in SEQ ID NO: 332, and CDR-H3 shown in SEQ ID NO: 333, and a VL region including CDR-L1 shown in SEQ ID NO: 115, CDR-L2 shown in SEQ ID NO: 116, and CDR-L3 shown in SEQ ID NO: 117; or e) A VH region including CDR-H1 shown in sequence number 337, CDR-H2 shown in sequence number 338, and CDR-H3 shown in sequence number 339, and a VL region including CDR-L1 shown in sequence number 121, CDR-L2 shown in sequence number 122, and CDR-L3 shown in sequence number 123; or f) The VH region shown in [H6H]Sequence ID 629 and the VL region shown in Sequence ID 628; or g) The VH region shown in [N3H]SEQ ID NO: 577, and the VL region shown in SEQ ID NO: 576; or h) The VH region shown in [Y8P] Sequence ID 684, and the VL region shown in Sequence ID 685; or i) The VH region shown in [L7E] Sequence ID 557 and the VL region shown in Sequence ID 556; or j) The VH region shown in [H6N]Sequence ID 561, and the VL region shown in Sequence ID 560 The second binding domain includes, k) A VH region including CDR-H1 shown in SEQ ID NO: 547, CDR-H2 shown in SEQ ID NO: 548, and CDR-H3 shown in SEQ ID NO: 549, and a VL region including CDR-L1 shown in SEQ ID NO: 541, CDR-L2 shown in SEQ ID NO: 542, and CDR-L3 shown in SEQ ID NO: 543; or l) The VH region shown in Sequence ID 697 and the VL region shown in Sequence ID 696 Includes.

[0045] In another embodiment, the present invention provides an isolated antibody construct comprising a first binding domain that binds to pMAGE-HLA on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain binds to the same epitope as an antibody or antibody construct comprising the following: a) A VH region including CDR-H1 shown in [H6H]SEQ ID NO: 439, CDR-H2 shown in SEQ ID NO: 440, and CDR-H3 shown in SEQ ID NO: 441, and a VL region including CDR-L1 shown in SEQ ID NO: 223, CDR-L2 shown in SEQ ID NO: 224, and CDR-L3 shown in SEQ ID NO: 225; or b) [N3H] A VH region including CDR-H1 shown in SEQ ID NO: 361, CDR-H2 shown in SEQ ID NO: 362, and CDR-H3 shown in SEQ ID NO: 363, and a VL region including CDR-L1 shown in SEQ ID NO: 145, CDR-L2 shown in SEQ ID NO: 146, and CDR-L3 shown in SEQ ID NO: 147; or c) A VH region including CDR-H1 shown in [Y8P]SEQ ID NO: 523, CDR-H2 shown in SEQ ID NO: 524, and CDR-H3 shown in SEQ ID NO: 525, and a VL region including CDR-L1 shown in SEQ ID NO: 307, CDR-L2 shown in SEQ ID NO: 308, and CDR-L3 shown in SEQ ID NO: 309; or d) [L7E] A VH region including CDR-H1 shown in SEQ ID NO: 331, CDR-H2 shown in SEQ ID NO: 332, and CDR-H3 shown in SEQ ID NO: 333, and a VL region including CDR-L1 shown in SEQ ID NO: 115, CDR-L2 shown in SEQ ID NO: 116, and CDR-L3 shown in SEQ ID NO: 117; or e) A VH region including CDR-H1 shown in sequence number 337, CDR-H2 shown in sequence number 338, and CDR-H3 shown in sequence number 339, and a VL region including CDR-L1 shown in sequence number 121, CDR-L2 shown in sequence number 122, and CDR-L3 shown in sequence number 123; or f) The VH region shown in [H6H]Sequence ID 629 and the VL region shown in Sequence ID 628; or g) The VH region shown in [N3H]SEQ ID NO: 577, and the VL region shown in SEQ ID NO: 576; or h) The VH region shown in [Y8P] Sequence ID 684, and the VL region shown in Sequence ID 685; or i) The VH region shown in [L7E] Sequence ID 557 and the VL region shown in Sequence ID 556; or j) The VH region shown in [H6N]Sequence ID 561, and the VL region shown in Sequence ID 560. The second binding domain is, k) A VH region including CDR-H1 shown in SEQ ID NO: 547, CDR-H2 shown in SEQ ID NO: 548, and CDR-H3 shown in SEQ ID NO: 549, and a VL region including CDR-L1 shown in SEQ ID NO: 541, CDR-L2 shown in SEQ ID NO: 542, and CDR-L3 shown in SEQ ID NO: 543; or l) The VH region shown in Sequence ID 697 and the VL region shown in Sequence ID 696 Includes.

[0046] In another embodiment, the present invention provides an isolated antibody construct comprising a first binding domain that binds to pMAGE-HLA on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain comprises a VH region selected from the group consisting of any VH and / or VL regions as shown in any of the sequences in Table 27 (SEQ ID NOs. 550-697), 35 (SEQ ID NOs. 3083-3118), 43 (SEQ ID NOs. 3979-4010), 51 (SEQ ID NOs. 5023-5118), 57 (SEQ ID NOs. 9351-10066), or 67 (SEQ ID NOs. 16667-16718).

[0047] In another embodiment, the present invention provides an isolated antibody construct comprising a first binding domain that binds to pMAGE-HLA on the surface of target cells and a second binding domain that binds to human CD3 on the surface of T cells, wherein the first binding domain is any VH domain as shown in any of the sequences in Table 24 (SEQ ID NOs: 322-537), 32 (SEQ ID NOs: 2984-3064), 40 (SEQ ID NOs: 3891-3962), 48 (SEQ ID NOs: 4759-4974), 56 (SEQ ID NOs: 7561-9350), or 66 (SEQ ID NOs: 16537-16666). The first binding domain further comprises CDR-H1, CDR-H2, and CDR-H3 from a VH region selected from the group consisting of regions, and / or the first binding domain further comprises CDR-L1, CDR-L2, and CDR-L3 from a VL region selected from the group consisting of any VL regions as shown in any of the sequences in Table 23 (SEQ ID NOs: 106-321), 31 (SEQ ID NOs: 2975-3055), 39 (SEQ ID NOs: 3883-3954), 47 (SEQ ID NOs: 4735-4950), 55 (SEQ ID NOs: 7203-8992), or 65 (SEQ ID NOs: 16511-16640).

[0048] In another embodiment, the present invention provides an isolated antibody construct comprising a first binding domain that binds to pMAGE-HLA on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain binds to the same epitope as an antibody or antibody construct comprising a VH region selected from the group consisting of any VH and / or VL regions as shown in any of the sequences in Table 27 (SEQ ID NOs. 550-697), 35 (SEQ ID NOs. 3083-3118), 43 (SEQ ID NOs. 3979-1010), 51 (SEQ ID NOs. 5023-5118), 57 (SEQ ID NOs. 9351-10066), or 67 (SEQ ID NOs. 16667-16718).

[0049] In another embodiment, the present invention provides an isolated antibody construct comprising a first binding domain that binds to pMAGE-HLA on the surface of target cells and a second binding domain that binds to human CD3 on the surface of T cells, wherein the first binding domain is selected from the group consisting of any VH regions as shown in any of the sequences in Table 24 (SEQ ID NOs: 322-537), 32 (SEQ ID NOs: 2984-3064), 40 (SEQ ID NOs: 3891-3962), 48 (SEQ ID NOs: 4759-4974), 56 (SEQ ID NOs: 7561-9350), or 66 (SEQ ID NOs: 16537-16666). The antibody or antibody construct containing CDR-H1, CDR-H2, and CDR-H3 from the VH region binds to the same epitope, and / or the first binding domain further comprises CDR-L1, CDR-L2, and CDR-L3 from a VL region selected from the group consisting of any VL regions as shown in any of the sequences in Table 23 (SEQ ID NOs: 106-321), 31 (SEQ ID NOs: 2975-3055), 39 (SEQ ID NOs: 3883-3954), 47 (SEQ ID NOs: 4735-4950), 55 (SEQ ID NOs: 7203-8992), or 65 (SEQ ID NOs: 16511-16640).

[0050] In further embodiments, the present invention provides isolated antibody constructs comprising a binding domain that binds to pMAGE-HLA on the surface of target cells, the binding domains being Table 29 (SEQ ID NOs: 1832-2965), 37 (SEQ ID NOs: 3497-3874), 45 (SEQ ID NOs: 4361-4710), 53 (SEQ ID NOs: 5982-6844), 61 (SEQ ID NOs: 13276-16184), 62 (SEQ ID NOs: 13 Includes any consensus sequence of the VH and / or VL and / or CDR and / or framework region as described in 434-16257), 63 (SEQ ID NOs. 13507-16484), 71 (SEQ ID NOs. 16971-17195), 72 (SEQ ID NOs. 16980-17204), 73 (SEQ ID NOs. 16989-17222), 111, 112, 113, 114, 115, 116, or 117. Note that in the consensus sequence listings above (Tables 29, 37, 45, 53, 61, 62, 63, 71, 72, and 73) in this specification, the asterisk (*) in some germline sequences indicates a stop codon within the consensus sequence. Tables 111-117 provide consensus sequences from the above tables in this specification, but in a different format, they provide the numerical position of each amino acid residue within the consensus sequence.

[0051] In certain embodiments, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain comprises VH, comprising the following amino acid residues: Ser30, Ser31, His32, Ala33, Ser52, Gly53, Ser54, Gly56, Gly57, Tyr59, Lys100, Gly101, Val102, His103, Leu104, and Gly105; and VL, comprising the following amino acid residues: Asn25, Asn26, Gly28, Ser29, Lys30, Ser31, His33, Tyr48, Asp49, Asp50, Asn51, Asp52, Asn65, Phe66, Gly67, Trp90, Tyr92, Arg93, and Leu95.

[0052] In certain embodiments, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of target cells, wherein the binding domain comprises at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, at least thirteen, at least fourteen, or fewer of the following amino acid residues: Ser30, Ser31, His32, Ala33, Ser52, Gly53, Ser54, Gly56, Gly57, Tyr59, Lys100, Gly101, Val102, His103, Leu104, and Gly105. The molecule comprises a VH containing at least 15 amino acids, and a VL containing at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, at least thirteen, at least fourteen, at least fifteen, at least sixteen, at least seventeen, or at least eighteen of the following amino acid residues: Asn25, Asn26, Gly28, Ser29, Lys30, Ser31, His33, Tyr48, Asp49, Asp50, Asn51, Asp52, Asn65, Phe66, Gly67, Trp90, Tyr92, Arg93, and Leu95.

[0053] In another embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain comprises VH, comprising the following amino acid residues: Ser30, Ser31, His32, Ala33, Ser52, Gly53, Ser54, Gly56, Gly57, Tyr59, Lys100, Gly101, Val102, His103, Leu104, and Gly105; and VL, comprising the following amino acid residues: Asn25, Asn26, Gly28, Ser29, Lys30, Ser31, His33, Tyr48, Asp49, Asp50, Asn51, Asp52, Arg53, Asn65, Ser66, Gly67, Trp90, Tyr92, and Gln95.

[0054] In another embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain comprises at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, at least thirteen, at least fourteen, or fewer of the following amino acid residues: Ser30, Ser31, His32, Ala33, Ser52, Gly53, Ser54, Gly56, Gly57, Tyr59, Lys100, Gly101, Val102, His103, Leu104, and Gly105. The molecule comprises a VH containing at least 15 amino acids, and a VL containing at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, at least thirteen, at least fourteen, at least fifteen, at least sixteen, at least seventeen, or at least eighteen of the following amino acid residues: Asn25, Asn26, Gly28, Ser29, Lys30, Ser31, His33, Tyr48, Asp49, Asp50, Asn51, Asp52, Arg53, Asn65, Ser66, Gly67, Trp90, Tyr92, and Gln95.

[0055] In another embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain comprises VH, comprising the following amino acid residues: Ser30, Asn31, Trp33, Arg50, Arg52, Arg54, Ser55, Tyr56, Gly57, Thr59, Tyr103, Ser104, Gly105, Ser106, Tyr107, Tyr108, ​​Asn109, Tyr110, Phe111, and Ser112; and VL, comprising the following amino acid residues: Ser30, Ser31, Tyr32, Ala50, Ser67, Thr91, Tyr92, Ser93, Met94, and Phe96.

[0056] In another embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain comprises at least one, at least two, at least three, at least four, at least five, at least six, and at least seven of the following amino acid residues: Ser30, Asn31, Trp33, Arg50, Arg52, Arg54, Ser55, Tyr56, Gly57, Thr59, Tyr103, Ser104, Gly105, Ser106, Tyr107, Tyr108, ​​Asn109, Tyr110, Phe111, and Ser112. VH containing at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 16, at least 17, at least 18, or at least 19 amino acid residues, and VL containing at least 1, at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, or at least 9 of the following amino acid residues: Ser30, Ser31, Tyr32, Ala50, Ser67, Thr91, Tyr92, Ser93, Met94, and Phe96.

[0057] In another embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain comprises a VH derived from one of the VH sequences provided herein and a VL comprising the following amino acid residues: Tyr92, Leu95, Asp96, and Trp97.

[0058] In another embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain comprises a VH derived from one of the VH sequences provided herein and a VL comprising at least one, at least two, or at least three of the following amino acid residues: Tyr92, Leu95, Asp96, and Trp97.

[0059] In another embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain comprises a VH derived from one of the VH sequences provided herein and a VL comprising the following amino acid residues: Tyr92, Gln95, Gln96, and Gln97.

[0060] In another embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain comprises a VH derived from one of the VH sequences provided herein and a VL comprising at least one, at least two, or at least three of the following amino acid residues: Tyr92, Gln95, Gln96, and Gln97.

[0061] In another embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain comprises a VH derived from one of the VH sequences provided herein and a VL comprising the following amino acid residues: Tyr92, Gln95, Arg96, and Gln97.

[0062] In another embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain comprises a VH derived from one of the VH sequences provided herein and a VL comprising at least one, at least two, or at least three of the following amino acid residues: Tyr92, Gln95, Arg96, and Gln97. In another embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain comprises a VH derived from one of the VH sequences provided herein and a VL comprising the following amino acid residues: Tyr92, Leu95, Arg96, and His97.

[0063] In another embodiment, the present invention provides an isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of a target cell, wherein the binding domain comprises a VH derived from one of the VH sequences provided herein and a VL comprising at least one, at least two, or at least three of the following amino acid residues: Tyr92, Leu95, Arg96, and His97.

[0064] In a further embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of target cells and a second binding domain that binds to human CD3 on the surface of T cells, wherein the first binding domain is represented by Table 29 (SEQ ID NOs. 1832-2965), 37 (SEQ ID NOs. 3497-3874), 45 (SEQ ID NOs. 4361-4710), 53 (SEQ ID NOs. 5982-6844), 61 (SEQ ID NOs. 13) Includes any consensus sequence of VH and / or VL and / or CDR and / or framework regions as shown in 276~16184), 62 (sequence codes 13434~16257), 63 (sequence codes 13507~16484), 71 (sequence codes 16971~17195), 72 (sequence codes 16980~17204), 73 (sequence codes 16989~17222), 111, 112, 113, 114, 115, 116 or 117.

[0065] In another embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of target cells and a second binding domain that binds to human CD3 on the surface of T cells, wherein the first binding domain comprises the following amino acid residues: Ser30, Ser31, His32, Ala33, Ser52, Gly53, Ser54, Gly56, Gly57, Tyr59, Lys VH contains 100, Gly101, Val102, His103, Leu104, and Gly105, and VL contains the following amino acid residues: Asn25, Asn26, Gly28, Ser29, Lys30, Ser31, His33, Tyr48, Asp49, Asp50, Asn51, Asp52, Asn65, Phe66, Gly67, Trp90, Tyr92, Arg93, and Leu95.

[0066] In another embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of target cells and a second binding domain that binds to human CD3 on the surface of T cells, wherein the first binding domain comprises at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, and at least one of the following amino acid residues: Ser30, Ser31, His32, Ala33, Ser52, Gly53, Ser54, Gly56, Gly57, Tyr59, Lys100, Gly101, Val102, His103, Leu104, and Gly105. Each comprises a VH containing 13, at least 14, or at least 15 amino acids, and a VL containing at least 1, at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 16, at least 17, or at least 18 of the following amino acid residues: Asn25, Asn26, Gly28, Ser29, Lys30, Ser31, His33, Tyr48, Asp49, Asp50, Asn51, Asp52, Asn65, Phe66, Gly67, Trp90, Tyr92, Arg93, and Leu95.

[0067] In another embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of target cells and a second binding domain that binds to human CD3 on the surface of T cells, wherein the first binding domain comprises the following amino acid residues: Ser30, Ser31, His32, Ala33, Ser52, Gly53, Ser54, Gly56, Gly57, Tyr59, Lys VH contains 100, Gly101, Val102, His103, Leu104, and Gly105, and VL contains the following amino acid residues: Asn25, Asn26, Gly28, Ser29, Lys30, Ser31, His33, Tyr48, Asp49, Asp50, Asn51, Asp52, Arg53, Asn65, Ser66, Gly67, Trp90, Tyr92, and Gln95.

[0068] In another embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of target cells and a second binding domain that binds to human CD3 on the surface of T cells, wherein the first binding domain comprises at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, and at least one of the following amino acid residues: Ser30, Ser31, His32, Ala33, Ser52, Gly53, Ser54, Gly56, Gly57, Tyr59, Lys100, Gly101, Val102, His103, Leu104, and Gly105. Each comprises a VH containing 13, at least 14, or at least 15 amino acids, and a VL containing at least 1, at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 16, at least 17, or at least 18 of the following amino acid residues: Asn25, Asn26, Gly28, Ser29, Lys30, Ser31, His33, Tyr48, Asp49, Asp50, Asn51, Asp52, Arg53, Asn65, Ser66, Gly67, Trp90, Tyr92, and Gln95.

[0069] In another embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of target cells and a second binding domain that binds to human CD3 on the surface of T cells, wherein the first binding domain comprises VH, which includes the following amino acid residues: Ser30, Asn31, Trp33, Arg50, Arg52, Arg54, Ser55, Tyr56, Gly57, Thr59, Tyr103, Ser104, Gly105, Ser106, Tyr107, Tyr108, ​​Asn109, Tyr110, Phe111, and Ser112, and VL, which includes the following amino acid residues: Ser30, Ser31, Tyr32, Ala50, Ser67, Thr91, Tyr92, Ser93, Met94, and Phe96.

[0070] In another embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of target cells and a second binding domain that binds to human CD3 on the surface of T cells, wherein the first binding domain comprises at least one, at least two, at least three, at least four, or at least five of the following amino acid residues: Ser30, Asn31, Trp33, Arg50, Arg52, Arg54, Ser55, Tyr56, Gly57, Thr59, Tyr103, Ser104, Gly105, Ser106, Tyr107, Tyr108, ​​Asn109, Tyr110, Phe111, and Ser112. VH containing at least 6, at least 7, at least 8, at least 9, at least 10, at least 11, at least 12, at least 13, at least 14, at least 15, at least 16, at least 17, at least 18, or at least 19 amino acid residues, and VL containing at least 1, at least 2, at least 3, at least 4, at least 5, at least 6, at least 7, at least 8, or at least 9 of the following amino acid residues: Ser30, Ser31, Tyr32, Ala50, Ser67, Thr91, Tyr92, Ser93, Met94, and Phe96.

[0071] In another embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain comprises a VH derived from one of the VH sequences provided herein and a VL comprising the following amino acid residues: Tyr92, Leu95, Asp96, and Trp97.

[0072] In another embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain comprises a VH derived from one of the VH sequences provided herein and a VL comprising at least one, at least two, or at least three of the following amino acid residues: Tyr92, Leu95, Asp96, and Trp97.

[0073] In another embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain comprises a VH derived from one of the VH sequences provided herein and a VL comprising the following amino acid residues: Tyr92, Gln95, Gln96, and Gln97.

[0074] In another embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain comprises a VH derived from any of the VH sequences provided herein and a VL comprising at least one, at least two, or at least three of the following amino acid residues: Tyr92, Gln95, Gln96, and Gln97.

[0075] In another embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain comprises a VH derived from one of the VH sequences provided herein and a VL comprising the following amino acid residues: Tyr92, Gln95, Arg96, and Gln97.

[0076] In another embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain comprises a VH derived from any of the VH sequences provided herein and a VL comprising at least one, at least two, or at least three of the following amino acid residues: Tyr92, Gln95, Arg96, and Gln97.

[0077] In another embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain comprises a VH derived from one of the VH sequences provided herein and a VL comprising the following amino acid residues: Tyr92, Leu95, Arg96, and His97.

[0078] In another embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain comprises a VH derived from any of the VH sequences provided herein and a VL comprising at least one, at least two, or at least three of the following amino acid residues: Tyr92, Leu95, Arg96, and His97.

[0079] In another specific embodiment, the first domain of the antibody construct is bound to an epitope containing the MAGEB2 peptide (SEQ ID NO: 1), and the antibody construct comprises a VH region containing CDR-H1, CDR-H2, and CDR-H3, wherein CDR-H3 contains lysine at position 2, valine at position 4, and / or histidine at position 5.

[0080] In another specific embodiment, the first domain of the antibody construct is bound to an epitope containing the MAGEB2 peptide (SEQ ID NO: 1), and the antibody construct comprises a VH region containing CDR-H1, CDR-H2, and CDR-H3, wherein CDR-H2 contains glycine at position 5 and / or tyrosine at position 11.

[0081] In another specific embodiment, the first domain of the antibody construct is bound to an epitope containing the MAGEB2 peptide (SEQ ID NO: 1), and the antibody construct comprises a VL region containing CDR-L1, CDR-L2, and CDR-L3, wherein CDR-L3 contains tryptophan at position 3 and / or tyrosine at position 5.

[0082] The complete sequence of an exemplary bispecific binding molecule is shown below. H6H: [ka] N3H: [ka] Y8P: [ka] L7E: [ka] H6N: [ka]

[0083] Further exemplary bispecific binding molecules of the present invention are shown in Tables 88 (SEQ ID NOs: 17390-17398) and 94 (SEQ ID NOs: 18145-18147) herein.

[0084] In one embodiment, the present invention provides an isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell, wherein the antibody construct is a) Sequence ID 73; or b) Sequence ID 47; or c) Sequence ID 101; or d) Sequence ID 37; or e) Sequence ID 39 Includes.

[0085] Affinity / Efficacy The interaction between the binding domain and the epitope or epitope-containing region means that the binding domain exhibits a recognizable affinity for the epitope / epitope-containing region on a specific protein or antigen (wherein both binding domains, if present, specific pMAGE-HLA and CD3, respectively), and, unless otherwise specified, does not exhibit significant reactivity with proteins or antigens other than pMAGE-HLA or CD3. In certain embodiments, only pMAGE-HLA-binding molecules are provided (i.e., the molecules do not bind to other targets such as CD3), and these molecules will exhibit considerable affinity for the pMAGE-HLA target. This affinity can be measured by various techniques known to those skilled in the art, such as surface plasmon resonance assays like the Biacore assay, or cell-based assays.

[0086] "Measurable affinity" is approximately 10 -6 It contains a bond with M(KD) or a bond having a stronger affinity. Preferably, the bond affinity is about 10 -12 ~10 -8 M, 10 -12 ~10 -9 M, 10 -12 ~10 -10 M, 10 -11 ~10-8 M, preferably about 10 -11 ~10 -9 M, the binding is considered specific. Whether the binding domain specifically reacts with or binds to the target can be readily tested, inter alia, by comparing the reaction of the binding domain to the target protein or antigen with the reaction of the binding domain to a protein or antigen other than pMAGE-HLA or CD3. Preferably, the binding domains of the present invention do not bind essentially or substantially to proteins or antigens other than pMAGE-HLA or CD3 (i.e., the first binding domain cannot bind to proteins other than pMAGE-HLA and the second binding domain cannot bind to proteins other than CD3). In certain embodiments, having superior affinity characteristics compared to other HLE formats is an envisioned feature of the antibody constructs according to the present invention. Such superior affinity would, as a result, suggest an extended half-life in vivo. In these embodiments, the longer half-life of the antibody constructs according to the present invention may reduce the dosing period and frequency, which would typically contribute to improved patient compliance. This is of particular importance since the antibody constructs of the present invention are particularly beneficial to cancer patients who are very debilitated or have multiple additional diseases.

[0087] It is notable that the pMAGE-HLA binding molecules of the present invention exhibit surprising efficacy. This efficacy shown by the binding molecules of the present invention is an important attribute for therapeutic molecules, but it can be very difficult to achieve without off-target binding and associated side effects and toxicity. This efficacy may be due to binding to the MAGEB2 peptide of SEQ ID NO: 1, and / or the MAGEA4 peptide of SEQ ID NO: 2, and / or the MAGEA8 peptide of SEQ ID NO: 3, where in each case the peptide is complexed with HLA and present on the surface of target cells. In one embodiment, the efficacy is measured in a cell-based assay such as the TDCC assay described herein, for example in the Examples.

[0088] Therefore, in one embodiment, the pMAGE-HLA binding molecule of the present invention has an EC of less than 250 pM relative to endogenous pMAGE-HLA expressing cells. 50 (For example, cells having HLA-A*02:01 complexed with MAGEB2, MAGEA4, or MAGEA8 peptides). In another embodiment, the pMAGE-HLA binding molecule of the present invention has an EC of less than 200 pM relative to endogenous pMAGE-HLA expressing cells. 50 In another embodiment, the pMAGE-HLA binding molecule of the present invention has an EC of less than 150 pM to endogenous pMAGE-HLA expressing cells. 50 In another embodiment, the pMAGE-HLA binding molecule of the present invention has an EC of less than 100 pM to endogenous pMAGE-HLA expressing cells. 50 In another embodiment, the pMAGE-HLA binding molecule of the present invention has an EC of less than 50 pM to endogenous pMAGE-HLA expressing cells. 50 In another embodiment, the pMAGE-HLA binding molecule of the present invention has an EC of less than 25 pM to endogenous pMAGE-HLA expressing cells. 50 In another embodiment, the pMAGE-HLA binding molecule of the present invention has an EC of less than 10 pM to endogenous pMAGE-HLA expressing cells. 50 In another embodiment, the pMAGE-HLA binding molecule of the present invention has an EC of less than 5 pM to endogenous pMAGE-HLA expressing cells. 50 In another embodiment, the pMAGE-HLA binding molecule of the present invention has an EC of less than 1 pM to endogenous pMAGE-HLA expressing cells. 50 It has.

[0089] specificity The term "not significantly binding" means that, when the antibody construct or binding domain of the present invention is expressed on the surface of a cell, it does not bind to any protein or antigen other than the MAGEB2-HLA peptide complex (pMAGE-HLA) or CD3. For example, an antibody construct or binding domain that shows binding to an HLA that does not present the MAGEB2 peptide (i.e., another peptide from the HLA peptidome), such as HLA-A1*02:01, would not be a desirable antibody construct or binding molecule. The term "HLA peptidome" refers to a pool of peptides that specifically interact with a particular HLA class and may encompass thousands of different peptide sequences. The HLA peptidome includes a diverse range of peptides derived from both normal and abnormal proteins expressed within cells. In another example, an antibody construct or binding domain that shows binding to an HLA, such as HLA-A1*02:01 without any complexing peptide, would also not be a desirable antibody construct or binding molecule.

[0090] In the present invention, the pMAGE-HLA binding molecule exhibits remarkable levels of specificity and selectivity to its targets, as evidenced by the lack of binding to cells expressing HLA, as well as to target-negative cells (e.g., HLA-A*02:01-negative and / or MAGE-B2-negative cells) containing peptides similar to the MAGEB2 peptide GVYDGEEHSV (SEQ ID NO: 1) and / or the MAGEA4 peptide GVYDGREHTV (SEQ ID NO: 2) and / or the MAGEA8 peptide GLYDGREHSV (SEQ ID NO: 3). For example, see the similar peptides shown in Figure 1, which, when complexed with HLA, are not specifically bound by the pMAGE-HLA binding molecule of the present invention.

[0091] Those skilled in the art will be led to expect the exact opposite. Recent publications have investigated the specificity of both bispecific T cell receptors and bispecific antibodies targeting peptide-HLA complexes (see Holland et al., J Clin Invest. 2020). While favorable properties of TCR-based molecules were reported, antibodies binding to peptide-HLA complexes, as reported, were far less specific and exhibited higher levels of cross-reactivity, which would be undesirable in therapeutic molecules.

[0092] This specificity and selectivity are highly desirable for therapeutic molecules, but difficult to achieve, as they limit, reduce, or eliminate off-target binding and any potential associated toxicity or side effects. Unexpectedly, the pMAGE-HLA binding molecule of the present invention achieves this desired specificity and selectivity.

[0093] However, it is also quite surprising that in the present invention, the affinity of various pMAGE-HLA binding molecules does not appear to be decisive in terms of specificity and selectivity. For example, antibodies can have very similar affinities to pMAGE-HLA targets expressed on cells, but exhibit very different levels of specificity and selectivity to MAGEB2-negative cell lines. See, for example, Figure 11B.

[0094] Therefore, in one embodiment, the selectivity of the pMAGE-HLA binding molecule allows for an EC of more than 50 nM against target-negative cells. 50 This yields (e.g., HLA-A*02:01-negative and / or MAGE-B2-negative cells). In another embodiment, the selectivity of the pMAGE-HLA binding molecule allows for EC levels greater than 45 nM against target-negative cells. 50 In another embodiment, the selectivity of the pMAGE-HLA binding molecule allows for an EC of more than 40 nM against target-negative cells. 50 In another embodiment, the selectivity of the pMAGE-HLA binding molecule allows for an EC of more than 35 nM against target-negative cells. 50In another embodiment, the selectivity of the pMAGE-HLA binding molecule allows for an EC of more than 30 nM against target-negative cells. 50 In another embodiment, the selectivity of the pMAGE-HLA binding molecule allows for an EC of more than 25 nM against target-negative cells. 50 In another embodiment, the selectivity of the pMAGE-HLA binding molecule allows for an EC of more than 20 nM against target-negative cells. 50 In another embodiment, the selectivity of the pMAGE-HLA binding molecule allows for an EC of more than 15 nM against target-negative cells. 50 In another embodiment, the selectivity of the pMAGE-HLA binding molecule allows for an EC of more than 10 nM against target-negative cells. 50 This selectivity can be measured by those skilled in the art using the TDCC assay provided herein in the examples.

[0095] Specific binding is thought to be brought about by specific motifs within the binding domain and the amino acid sequence of the antigen. Therefore, binding occurs as a result of their primary, secondary, and / or tertiary structures, as well as as a result of secondary modifications of said structures. Specific interaction between the antigen interaction site and its specific antigen can lead to simple binding of the site to the antigen. Furthermore, specific interaction between the antigen interaction site and its specific antigen can alternatively or additionally lead to signal initiation, for example, by inducing conformational changes in the antigen or oligomerization of the antigen.

[0096] As will be discussed below in this specification regarding structural analysis, the remarkable specificity of the pMAGE-HLA binding molecule of the present invention against difficult targets such as the pMAGE-HLA complex may be due to the specific structural conformation of the molecular paratope, which creates a "hole" that interacts with specific amino acids protruding from the pMAGE-HLA complex.

[0097] The first domain of the antibody construct of the present invention binds to pMAGE-HLA on the surface of target cells. “Target cells” can be any cells expressing or displaying the MAGEB2 peptide on their surface, preferably cells that are part of the human or animal body, such as specific MAGEB2-expressing cancer or tumor cells or cells of MAGEB2-positive neoplasms. In the context of the present invention, the term “on the surface” is understood to mean that the domain of the antibody construct forms a complex with MHC (pMAGE-HLA) and specifically binds to an epitope containing the MAGEB2 peptide (GVYDGEEHSV, SEQ ID NO: 1) presented by MHC molecules on the cell surface. Therefore, the first domain according to the present invention can bind to pMAGE-HLA whether it is naturally expressed by MAGEB2-expressing cells or cell lines, and / or by cells or cell lines transformed or (stable / transiently) transfected with MAGEB2, or whether the peptide is exogenously loaded onto the MHC.

[0098] competitive binding Whether an antibody or antibody construct competes with another given antibody or antibody construct for binding to an antigen (such as pMAGE-HLA) on the surface of target cells can be measured by a competitive assay such as competitive ELISA. Avidin-conjugated microparticles (beads) can also be used. Similar to avidin-coated ELISA plates, each of these beads can be used as a substrate when reacting with biotinylated proteins, and the assay can be performed on it. The antigen is coated onto the beads, and then pre-coated with the first antibody. A secondary antibody is added to confirm any further binding. Readout is performed by flow cytometry. Preferably, the cell-based competitive assay is used using either cells that naturally express MAGEB2 and HLA, or cells that have been stably or transiently transformed with MAGEB2 and / or HLA. In this context, the term "competing for binding" means that, as determined by any one of the assays disclosed above, preferably a cell-based assay, competition occurs between two test antibodies at a rate of at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80%, or at least 90%. Of course, similar analyses can be applied to other targets such as CD3.

[0099] Competitive antibody binding assays include assays that determine the competitive binding of two antibodies / antibody constructs to an antigen bound to a cell surface. A common method aims to detect the binding of two antibodies / antibody constructs A and B to the same antigen on the cell surface and involves the following steps: a) Cells are pre-incubated with antibody / antibody construct A, followed by the addition of labeled antibody / antibody construct B under maximum pressure. Blockage of cell surface antigens is detected by comparing the binding of B with the binding in the absence of A. b) Titration of antibody / antibody construct A in the presence of the lowest maximum amount of labeled antibody / antibody construct B (i.e., adding different amounts) and detection of the effect on binding of B, or c) Co-titration of A and B, i.e., a method that is not affected by the order or relative amounts of adding the antibodies / antibody constructs, in which both antibodies / antibody constructs are incubated together at their maximum concentrations and it is detected whether the total binding is equal to or greater than the binding of A or B alone. It may include.

[0100] When two antibodies / antibody constructs A and B compete for an antigen bound to a cell surface, the antibodies very often compete with each other in a blocking assay, regardless of the order in which they are added. In other words, competition will be detected if the assay is performed in either direction. However, this is not always the case, and under certain circumstances, the order in which antibodies are added or the direction of the assay can affect the signal produced. This may be due to differences in affinity or avidity between potentially competing antibodies / antibody constructs. If the order of addition has a significant effect on the signal produced, and competition is detected in at least one order, then it can be concluded that the two antibodies / antibody constructs compete.

[0101] Epitope amino acid residues The term "epitope" refers to a region on an antigen or specific amino acid residue to which the binding domain of an antibody or immunoglobulin, or a derivative, fragment, or variant of an antibody or immunoglobulin, specifically binds. An epitope is antigenic, and therefore the term epitope may also be referred to herein as an "antigen structure" or "antigen determinant." Thus, the binding domain is an "antigen interaction site." This binding / interaction is also understood to define "specific recognition."

[0102] An "epitope" can be formed by both consecutive or discontinuous amino acids juxtaposed by the tertiary folding of a protein. A "linear epitope" is an epitope that contains a recognized epitope with a consecutive primary amino acid sequence. Linear epitopes typically contain at least three or at least four, more commonly at least five, at least six, or at least seven, for example, about eight to about ten amino acids in their unique sequence.

[0103] In contrast to linear epitopes, "conformational epitopes" are epitopes in which the primary amino acid sequence containing the epitope is not the sole defining component of the recognized epitope (e.g., epitopes in which the primary amino acid sequence is not necessarily recognized by the binding domain). Typically, conformational epitopes contain an increased number of amino acids compared to linear epitopes and include discontinuous amino acid sequences. With regard to the recognition of conformational epitopes, the binding domain paratope recognizes the three-dimensional structure of an antigen, preferably a peptide or protein or a fragment thereof (in the context of this invention, the antigenic structure for one of the binding domains is contained within the target cell surface antigen protein). For example, when a protein molecule folds to form a three-dimensional structure, specific amino acids and / or polypeptide backbone that form a conformational epitope are juxtaposed, allowing the antibody to recognize the epitope. Methods for determining the conformation of an epitope include, but are not limited to, X-ray crystallography, two-dimensional nuclear magnetic resonance (2D-NMR) spectroscopy, and site-specific spin labeling and electron spin resonance (EPR) spectroscopy.

[0104] In one embodiment of the present invention, the epitope comprises the MAGEB2 peptide GVYDGEEHSV (SEQ ID NO: 1) complexed with an HLA molecule on the cell surface. In some embodiments, the antibody construct binds to an epitope comprising only the amino acid residues of the MAGEB2 peptide of SEQ ID NO: 1. In other embodiments, the antibody construct binds to an epitope comprising the MAGEB2 peptide of SEQ ID NO: 1 and comprising at least some amino acid residues of the HLA that is complexed with the MAGEB2 peptide GVYDGEEHSV (SEQ ID NO: 1), i.e., the HLA that is presenting the MAGEB2 peptide GVYDGEEHSV.

[0105] Structural analysis of the interaction between a target polypeptide and, for example, the binding domain of a BiTE® molecule as intended herein, provides amino acid residues of the target epitope involved in the binding interaction. Furthermore, whether an antibody, antibody construct, or binding domain binds to the same pMAGE-HLA epitope on the surface of another given antibody, antibody construct, or binding domain can be measured by different analyses described herein. For example, in certain embodiments, the crystal structure of the target-binding interaction can provide these amino acid residues. In other embodiments, various other analyses can be performed to confirm the amino acid residues involved in the binding interaction, such as in silico modeling and analysis, Xscan, alanine scanning, arginine scanning, epitope mapping, and other techniques known to those skilled in the art.

[0106] In relation to the present invention, where the MAGEB2 peptide or other MAGE peptide sequences are short, for example, GVYDGEEHSV (SEQ ID NO: 1) is only 10 amino acids, a more useful technique for determining the contribution of specific residues of MAGEB2 or MAGEB2-specific peptides to recognition by antibody constructs or binding domains is the technique known to those skilled in the art as X-scan. X-scan is a peptide analysis technique that sequentially mutates each amino acid residue of the analyte to all possible amino acids. In certain cases, this approach can provide significantly enhanced sensitivity and specificity for detecting acceptable or unacceptable substitutions. If most amino acid substitutions are unacceptable, the position is considered essential for binding in X-scan analysis. In alternative embodiments of X-scan analysis, specific anchor residues within the peptide are not mutated.

[0107] The structural analyses presented herein in Examples 6-10 and 14 provide insight into why our pMAGE-HLA binding molecules possess such remarkable specificity. In certain determined molecular structures as shown, the pMAGE-HLA complex appears to have small “grooves” or “cracks” on its surface caused by the protruding conformation of the decameric MAGEB2 peptide bound to the peptide binding groove of HLA, presenting a complementary and highly unexpected prominent surface for specific recognition by the pMAGE-HLA binding agent.

[0108] All of the various MAGEB2 residues (1-10) appear to contribute to the target epitopes of the pMAGE-HLA complex in various ways. Therefore, in one embodiment, the first domain of the antibody construct of the present invention binds to the human MAGEB2 peptide GVYDGEEHSV (SEQ ID NO: 1) complexed with HLA on the surface of a target cell, where at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine or ten amino acid residues of the MAGEB2 peptide Gly1, Val2, Tyr3, Asp4, Gly5, Glu6, Glu7, His8, Ser9 and / or Val10 are essential for the binding of the first domain. In this context, the term "essential for binding" means that specific amino acids are required for the binding of the antibody construct binding domain to the MAGEB2 peptide to occur.

[0109] According to another embodiment, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell (pMAGE-HLA), where at least one of the amino acid residues of the MHC is Arg65, Lys66, Ala69, Gln72, Thr73, Val76, Lys146, Ala149, Ala150, His151, Glu154, Gln155, Gly16, Arg17, Gly18, Glu19, Pro20, Gln43, Lys68, Ser71, Arg75, Gly79, Thr80, Arg82, Gly83, and / or Glu89. One, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, at least thirteen, at least fourteen, at least fifteen, at least sixteen, at least seventeen, at least eighteen, at least nineteen, at least twenty, at least twenty-one, at least twenty-two, at least twenty-three, at least twenty-four, at least twenty-five, at least twenty-six, at least twenty-seven, at least twenty-eight, at least twenty-nine, or at least thirty are essential for the binding of the first domain. In this context, the term "essential for binding" means that specific amino acids are required for the binding of the antibody construct binding domain to the MHC to occur.

[0110] In further embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell (pMAGE-HLA), and to amino acid residues Glu55, Glu58, Tyr59, Asp61, Gly62, Glu63, Arg65, Lys66, Ala69, Ala158, Gly162, Thr163, Glu166, Trp167, Arg170 and also Gln54, Glu55, Gly56, Pro57, Glu58, Tyr59 , at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, at least thirteen, at least fourteen, at least fifteen, at least sixteen, at least seventeen, at least eighteen, at least nineteen, at least twenty, at least twenty-one, at least twenty-two, at least twenty-three, and at least twenty-four of Trp167 and Arg170.

[0111] Paratope amino acid residues The region of the binding domain that binds to an epitope is called a "paratope." Specific binding is thought to be achieved by specific motifs in the amino acid sequence of the binding domain and the antigen. Therefore, binding is achieved as a result of their primary, secondary, and / or tertiary structures, as well as as a result of potential secondary modifications of those structures. As described above, structural analysis of the interaction between the target polypeptide and the binding domain of, for example, the BiTE® molecule intended herein, provides the amino acid residues of the binding molecule paratope involved in the binding interaction with the target epitope. This structural analysis can be performed using the methods already described, such as crystallography or alanine scanning.

[0112] In one embodiment, the binding domain of the antibody construct binds to a human MAGEB2 peptide complexed with HLA on the surface of a target cell (pMAGE-HLA), where the paratope contains at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least 12, at least thirteen, at least fourteen, at least fifteen, at least sixteen, at least seventeen, at least eighteen, at least nineteen, at least ten, at least eleven, at least twelve, at least thirteen, at least fourteen, at least fifteen, at least sixteen, at least seventeen, at least eighteen, at least nineteen, at least twenty Heavy chain: Ser30, Ser31, His32, Tyr32, Ala33, Ser52, Gly53, Ser54, Gly56, Gly57, Tyr59, Lys100, Gly101, Val102, His103, Leu104, or Gly105 Light chain: Asn25, Asn26, Gly28, Ser29, Lys30, Ser31, His33, Tyr48, Asp49, Asp50, Asn51, Asp52, Arg53, Asn65, Phe66, Ser66, Gly67, Trp90, Tyr92, Arg93, Leu95, Gln95

[0113] In other embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell (pMAGE-HLA), where at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, or at least thirteen amino acid residues of the heavy chain, Ser30, Ser31, Ser52, Gly53, Ser54, Gly56, Gly57, Tyr59, Lys100, Val102, His103, Leu104, or Gly105, is essential for the binding of the first domain to the MHC.

[0114] In other embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell (pMAGE-HLA), where at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, or at least twelve amino acid residues of the heavy chain, Ser30, Ser31, His32, Ala33, Ser52, Gly53, Ser54, Lys100, Gly101, Val102, His103, or Leu104, is essential for the binding of the first domain to the MAGEB2 peptide.

[0115] In other embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell (pMAGE-HLA), where at least one, at least two, at least three, at least four, at least five, at least sixteen, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, at least thirteen, at least fourteen, at least fifteen, or at least sixteen heavy chain amino acid residues Trp33, Arg50, Arg52, Ser55, Tyr56, Gly57, Thr59, Tyr103, Gly105, Ser106, Tyr107, Tyr108, ​​Asn109, Tyr110, Phe111, or Ser112 are essential for the binding of the first domain to the MHC.

[0116] In other embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell (pMAGE-HLA), where the light chain amino acid residues are Asn25, Asn26, Gly28, Ser29, Lys30, Ser31, His33, Tyr48, Asp49, Asp50, Asn51, Asp52, Asn65, Phe66, Gly67, Trp90, Tyr92, Arg9 3. At least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, at least thirteen, at least fourteen, at least fifteen, at least sixteen, at least seventeen, at least eighteen, at least nineteen, or at least twenty of the Leu95s are essential for binding to the first domain MHC.

[0117] In other embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell (pMAGE-HLA), where at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, or at least ten light chain amino acid residues Ser30, Ser31, Tyr32, Ala50, Ser67, Thr91, Tyr92, Ser93, Met94, or Phe96 are essential for binding to the first domain MHC.

[0118] In other embodiments, the first domain of the antibody construct of the present invention binds to the human MAGEB2 peptide GVYDGEEHSV (SEQ ID NO: 1) complexed with HLA on the surface of target cells, and the light chain amino acid residue Trp90 is essential for the binding of the first domain to the MAGEB2 peptide.

[0119] In other embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, where at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, or at least thirteen amino acid residues of the heavy chain Ser30, Ser31, Ser52, Gly53, Ser54, Gly56, Gly57, Tyr59, Lys100, Val102, His103, Leu104, and Gly105 is essential for the binding of the first domain to the MHC.

[0120] In other embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, where at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, or at least twelve amino acid residues of the heavy chain, Ser30, Ser31, His32, Ala33, Ser52, Gly53, Ser54, Lys100, Gly101, Val102, His103, or Leu104, is essential for the binding of the first domain to the MAGEB2 peptide.

[0121] In other embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, where at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, or at least twelve amino acid residues of the heavy chain Ser30, Asn31, Arg54, Ser55, Tyr56, Ser104, Gly105, Ser106, or Tyr110 is essential for the binding of the first domain to the MAGEB2 peptide.

[0122] In other embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, where the light chain amino acid residues are Asn25, Asn26, Gly28, Ser29, Lys30, Ser31, His33, Tyr48, Asp49, Asp50, Asn51, Asp52, Arg53, Asn65, Phe66, Gly67, Trp90, Tyr92, Arg93, At least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, at least thirteen, at least fourteen, at least fifteen, at least sixteen, at least seventeen, at least eighteen, at least nineteen, or at least twenty of the Leu95s are essential for binding to the first domain MHC.

[0123] In other embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, and the light chain amino acid residue Trp90 is essential for the binding of the first domain to the MAGEB2 peptide.

[0124] In other embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) that has formed a complex with HLA on the surface of a target cell, where the amino acid residues at positions 92, 95, 96, or 97 of the light chain are essential for the binding of the first domain to the MAGEB2 peptide.

[0125] In another embodiment, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, and the amino acid residue at position 92 of the light chain is Trp, Tyr, Phe, Leu, Arg, Met, Gln, or Glu.

[0126] In certain embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, and the amino acid residue at position 92 of the light chain is Trp, Phe, Tyr, Leu, Glu, Gln, Arg, or Met.

[0127] In certain embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) that has formed a complex with HLA on the surface of a target cell, and the amino acid residue at position 92 of the light chain is Phe, Trp, Ile, Val, or Tyr.

[0128] In certain embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) that has formed a complex with HLA on the surface of a target cell, and the amino acid residue at position 92 of the light chain is Tyr, Trp, Phe, His, or Arg.

[0129] In certain embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) that has formed a complex with HLA on the surface of a target cell, and the amino acid residue at position 92 of the light chain is selected from any of the predicted options shown in Figures 47 to 50 herein.

[0130] In another embodiment, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, and the amino acid residue at position 95 of the light chain is Trp, Tyr, Phe, Arg, Met, Lys, or Gln.

[0131] In another embodiment, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, and the amino acid residue at position 95 of the light chain is Trp, Phe, Met, Tyr, Gln, Arg, His, or Leu.

[0132] In another embodiment, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, and the amino acid residue at position 95 of the light chain is Trp, Arg, Tyr, Leu, Phe, Cys, or Met.

[0133] In certain embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) that has formed a complex with HLA on the surface of a target cell, and the amino acid residue at position 95 of the light chain is selected from any of the predicted options shown in Figures 47 to 50 herein.

[0134] In another embodiment, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, and the amino acid residue at position 96 of the light chain is Met, Asn, His, or Asp.

[0135] In another embodiment, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, and the amino acid residue at position 96 of the light chain is Ile, Met, Leu, Thr, or His.

[0136] In certain embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) that has formed a complex with HLA on the surface of a target cell, and the amino acid residue at position 96 of the light chain is selected from any of the predicted options shown in Figures 47 to 50 herein.

[0137] In another embodiment, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, and the amino acid residue at position 97 of the light chain is Tyr, Phe, Ile, Met, His, Asn, Ala, or Ser.

[0138] In another embodiment, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, and the amino acid residue at position 97 of the light chain is Val, His, or Lys.

[0139] In another embodiment, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, and the amino acid residue at position 97 of the light chain is Phe, Tyr, Leu, Ile, Lys, or Val.

[0140] In certain embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) that has formed a complex with HLA on the surface of a target cell, and the amino acid residue at position 97 of the light chain is selected from any of the predicted options shown in Figures 47 to 50 herein.

[0141] In other embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, where at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, or at least thirteen amino acid residues of the heavy chain, Ser30, Ser31, Tyr32, Ser52, Gly53, Ser54, Gly56, Gly57, Tyr59, Lys100, His103, Leu104, or Gly105, is essential for the binding of the first domain to the MHC.

[0142] In other embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, where at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, or at least twelve amino acid residues of the heavy chain, Ser30, Ser31, Tyr32, Ala33, Ser52, Gly53, Ser54, Lys100, Gly101, Val102, His103, or Leu104, is essential for the binding of the first domain to the MAGEB2 peptide.

[0143] In other embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, where at least one, at least two, at least three, at least four, at least five, at least six, at least seven, at least eight, at least nine, at least ten, at least eleven, at least twelve, at least thirteen, at least four, at least fifteen, at least sixteen, at least seventeen, at least eighteen, or at least nineteen light chain amino acid residues, including Asn25, Asn26, Gly28, Ser29, Lys30, Ser31, His33, Tyr48, Asp49, Asp50, Asn51, Asp52, Arg53, Asn65, Ser66, Gly67, Trp90, Tyr92, or Gln95, is essential for binding to the first domain MHC.

[0144] In other embodiments, the first domain of the antibody construct of the present invention binds to a human MAGEB2 peptide (e.g., SEQ ID NO: 1) complexed with HLA on the surface of a target cell, and at least one and / or at least two amino acid residues of the light chain, Trp90 or Gln95, are essential for the binding of the first domain to the MAGEB2 peptide.

[0145] In another specific embodiment, the first domain of the antibody construct binds to an epitope containing the MAGEB2 peptide (SEQ ID NO: 1), and the antibody construct comprises a VH region containing CDR-H1, CDR-H2, and CDR-H3, wherein CDR-H3 contains lysine at position 2, valine at position 4, and / or histidine at position 5, which are essential for the binding of the first domain to the MAGEB2 peptide.

[0146] In another specific embodiment, the first domain of the antibody construct is bound to an epitope containing the MAGEB2 peptide (SEQ ID NO: 1), and the antibody construct comprises a VH region containing CDR-H1, CDR-H2 and CDR-H3, wherein CDR-H2 contains glycine at position 5 and / or tyrosine at position 11, which are essential for binding to the MAGEB2 peptide of the first domain.

[0147] In another specific embodiment, the first domain of the antibody construct is bound to an epitope containing the MAGEB2 peptide (SEQ ID NO: 1), and the antibody construct comprises a VL region containing CDR-L1, CDR-L2, and CDR-L3, wherein CDR-L3 contains tryptophan at position 3 and / or tyrosine at position 5, which are essential for binding the first domain to the MAGEB2 peptide.

[0148] In this context, the term "essential for binding" means that a specific amino acid or its functional equivalent is required for the binding of the antibody construct's binding domain to the target (e.g., the MAGEB2 peptide complexed with HLA on the surface of target cells).

[0149] Mechanism of CD3 cell killing In embodiments in which the antibody construct includes a second domain that binds to CD3, the mechanism for cell killing is, in most cases, T cell-mediated. T cells, or T lymphocytes, are a type of lymphocyte (a type of white blood cell) that plays a central role in cellular immunity. There are several subsets of T cells, each with a different function. T cells can be distinguished from other lymphocytes, such as B cells and NK cells, by the presence of a T cell receptor (TCR) on their cell surface. The TCR is involved in the recognition of antigens bound to major histocompatibility complex (MHC) molecules and is composed of two different protein chains. In 95% of T cells, the TCR consists of an alpha (α) chain and a beta (β) chain. When the TCR binds to the antigen peptide and MHC (peptide / MHC complex), the T lymphocyte is activated through a series of biochemical events mediated by related enzymes, co-receptors, specialized adapter molecules, and activated or released transcription factors.

[0150] In one embodiment, it is assumed that the second domain of the antibody construct of the present invention binds to an extracellular epitope of human and / or macaque CD3ε chain.

[0151] The CD3 receptor complex is a protein complex composed of four chains. In mammals, this complex contains a CD3γ (gamma) chain, a CD3δ (delta) chain, and two CD3ε (epsilon) chains. These chains associate with the T cell receptor (TCR) and the so-called ζ (zeta) chain to form the T cell receptor CD3 complex, which generates activation signals in T lymphocytes. The CD3γ (gamma), CD3δ (delta), and CD3ε (epsilon) chains are cell surface proteins of the highly related immunoglobulin superfamily, each containing a single extracellular immunoglobulin domain. The intracellular tail of the CD3 molecule contains a single conserved motif, known as the immunoreceptor activation tyrosine motif, or ITAM, which is essential for TCR signaling. The CD3 epsilon molecule is a polypeptide encoded by the CD3E gene located on human chromosome 11. The most preferred CD3 epsilon epitopes are those in the range of amino acid residues 1-27 of the human CD3 epsilon extracellular domain. The antibody construct according to the present invention is expected to exhibit only minimal nonspecific T cell activation, which is typically and advantageously undesirable in specific immunotherapy. This results in a lower risk of side effects.

[0152] Lysis of redirected target cells via T cell recruitment by multispecific, or at least bispecific, antibody constructs, is accompanied by the formation of cytolytic synapses and the delivery of perforin and granzyme. The bound T cells are capable of sequential target cell lysis and are typically unaffected by immune evasion mechanisms that prevent peptide antigen processing and presentation or clonal T cell differentiation (see, e.g., International Publication No. 2007 / 042261). However, even loss of MAGEB2 expression or MHC class I expression by tumor cells could be a possible evasion mechanism.

[0153] CD3 binding The second domain of the antibody construct of the present invention binds to CD3. More preferably, it binds to CD3 on the surface of T cells. It is further conceivable that the second domain binds to human CD3 on the surface of T cells, preferably human CD3. It is also conceivable that the second domain binds to CD3 epsilon. More preferably, it binds to human CD3 epsilon, for example, human CD3 epsilon on the surface of T cells.

[0154] The CD3-binding domain of the antibody construct of the present invention is preferably interspecies-specific in members of the order Mammalia of primates. Interspecies-specific CD3-binding domains are described, for example, in International Publication No. 2008 / 119567. According to one embodiment, in addition to binding to human CD3, the CD3-binding domain will also bind to CD3 of primates including (but not limited to) New World primates (such as marmosets (Callithrix jacchus), cotton-top tamarins (Saguinus Oedipus), or squirrel monkeys (Saimiri sciureus)), Old World primates (such as baboons and macaques), gibbons, and non-human homininae.

[0155] In one embodiment of the present invention, the second domain of the antibody construct binds to human CD3 epsilon (or human CD3 epsilon on the surface of T cells) and common marmoset (Callithrix jacchus) or common squirrel monkey (Saimiri sciureus) CD3 epsilon. The second domain is also envisioned to bind to the extracellular epitope of CD3 epsilon, preferably the extracellular epitope of human CD3 epsilon. The second domain is also envisioned to bind to the extracellular epitope of human and the macaque CD3 epsilon chain. One preferred epitope of CD3 epsilon is contained within amino acid residues 1-27 of the extracellular domain of human CD3 epsilon. More specifically, the epitope includes at least the amino acid sequence Gln-Asp-Gly-Asn-Glu (SEQ ID NO: 18452). The common marmoset (Callithrix jacchus) is a new species of primate belonging to the family Callitrichidae, while the common squirrel monkey (Saimiri sciureus) is a new species of primate belonging to the family Cebidae. Binding agents with these properties are described in detail in International Publication No. 2008 / 119567.

[0156] Antibodies or bispecific antibody constructs against (human) CD3, or specifically against CD3 epsilon, are known in the art, and their CDR, VH, and VL sequences can serve as the basis for the second binding domain of the antibody construct of the present invention. For example, in 1979, Kung et al. reported the development of OKT3 (Ortho Kung T3), the first mAb that recognized CD3 (specifically the epsilon chain of CD3) on human T cells. OKT3 (muromonab) was the first mouse-derived monoclonal antibody made available for therapeutic use in humans. More recent anti-CD3 monoclonal antibodies include otelixizumab (TRX4), teplizumab (MGA031), horalumab, and bicilizumab, all of which target the epsilon chain of CD3. Bispecific antibody constructs against (cancer) targets and CD3 have also been developed and (pre-clinically) tested, and their CD3-binding domains (CDR, VH, VL) can serve as the basis for the second binding domain of the antibody constructs of the present invention. Examples, but not limited to, include blinatumomab, solitomab (MT110, AMG110), catumakisomab, duvortuxizumab, erzumakisomab, mosnetuzumab, FBTA05 (Bi20, TPBs05), CEA-TCB (RG7802, RO6958688), AFM11, and MGD006 (S80880). Other examples of CD3-binding domains are disclosed, for example, in U.S. Patent No. 7,994,289B2, U.S. Patent No. 7,728,114B2, U.S. Patent No. 7,381,803B1, and U.S. Patent No. 6,706,265B1.

[0157] The antibody construct of the present invention is assumed to include a second domain that binds to CD3 on the surface of T cells, comprising a VL region containing CDR-L1 (SEQ ID NO: 541), CDR-L2 (SEQ ID NO: 542), and CDR-L3 (SEQ ID NO: 543), and a VH region containing CDR-H1 (SEQ ID NO: 547), CDR-H2 (SEQ ID NO: 548), and CDR-H3 (SEQ ID NO: 549).

[0158] The antibody construct of the present invention is assumed to have a second domain that binds to CD3 on the surface of T cells, comprising a VL (SEQ ID NO: 696) region and a VH (SEQ ID NO: 697) region.

[0159] In some embodiments, the second domain of the antibody construct of the present invention binds to human CD3 epsilon and / or macaque CD3 epsilon. In preferred embodiments, the second domain further binds to marmoset (Callithrix jacchus), cotton-top tamarin (Saguinus Oedipus), or squirrel monkey (Saimiri sciureus) CD3 epsilon. Both marmosets (Callithrix jacchus) and cotton-top tamarins (Saguinus oedipus) are New World primates belonging to the family Callitrichidae, while squirrel monkeys (Saimiri sciureus) are New World primates belonging to the family Cebidae.

[0160] In some embodiments, the antibody construct of the present invention has a second binding domain that binds to the extracellular epitope of the human and / or macaque CD3 epsilon chain, (a) CDR-L1 as shown in Sequence ID 27 of International Publication No. 2008 / 119567, CDR-L2 as shown in Sequence ID 28 of International Publication No. 2008 / 119567, and CDR-L3 as shown in Sequence ID 29 of International Publication No. 2008 / 119567; (b) CDR-L1 as shown in Sequence ID 117 of International Publication No. 2008 / 119567, CDR-L2 as shown in Sequence ID 118 of International Publication No. 2008 / 119567, and CDR-L3 as shown in Sequence ID 119 of International Publication No. 2008 / 119567; and (c) Includes a VL area containing CDR-L1, CDR-L2, and CDR-L3 selected from CDR-L1 as shown in Sequence ID 153 of International Publication No. 2008 / 119567, CDR-L2 as shown in Sequence ID 154 of International Publication No. 2008 / 119567, and CDR-L3 as shown in Sequence ID 155 of International Publication No. 2008 / 119567.

[0161] In another embodiment of the antibody construct of the present invention, a second domain that binds to the extracellular epitope of the human and / or macaque CD3 epsilon chain is (a) CDR-H1 as shown in Sequence ID 12 of International Publication No. 2008 / 119567, CDR-H2 as shown in Sequence ID 13 of International Publication No. 2008 / 119567, and CDR-H3 as shown in Sequence ID 14 of International Publication No. 2008 / 119567; (b) CDR-H1 as shown in Sequence ID 30 of International Publication No. 2008 / 119567, CDR-H2 as shown in Sequence ID 31 of International Publication No. 2008 / 119567, and CDR-H3 as shown in Sequence ID 32 of International Publication No. 2008 / 119567; (c) CDR-H1 as shown in Sequence ID 48 of International Publication No. 2008 / 119567, CDR-H2 as shown in Sequence ID 49 of International Publication No. 2008 / 119567, and CDR-H3 as shown in Sequence ID 50 of International Publication No. 2008 / 119567; (d) CDR-H1 as shown in Sequence ID 66 of International Publication No. 2008 / 119567, CDR-H2 as shown in Sequence ID 67 of International Publication No. 2008 / 119567, and CDR-H3 as shown in Sequence ID 68 of International Publication No. 2008 / 119567; (e) CDR-H1 as shown in Sequence ID 84 of International Publication No. 2008 / 119567, CDR-H2 as shown in Sequence ID 85 of International Publication No. 2008 / 119567, and CDR-H3 as shown in Sequence ID 86 of International Publication No. 2008 / 119567; (f) CDR-H1 as shown in Sequence ID 102 of International Publication No. 2008 / 119567, CDR-H2 as shown in Sequence ID 103 of International Publication No. 2008 / 119567, and CDR-H3 as shown in Sequence ID 104 of International Publication No. 2008 / 119567; (g) CDR-H1 as shown in Sequence ID 120 of International Publication No. 2008 / 119567, CDR-H2 as shown in Sequence ID 121 of International Publication No. 2008 / 119567, and CDR-H3 as shown in Sequence ID 122 of International Publication No. 2008 / 119567; (h) CDR-H1 as shown in Sequence ID 138 of International Publication No. 2008 / 119567, CDR-H2 as shown in Sequence ID 139 of International Publication No. 2008 / 119567, and CDR-H3 as shown in Sequence ID 140 of International Publication No. 2008 / 119567; (i) CDR-H1 as shown in Sequence ID 156 of International Publication No. 2008 / 119567, CDR-H2 as shown in Sequence ID 157 of International Publication No. 2008 / 119567, and CDR-H3 as shown in Sequence ID 158 of International Publication No. 2008 / 119567; and (j) A VH region containing CDR-H1, CDR-H2, and CDR-H3 selected from CDR-H1 as shown in Sequence ID No. 174 of International Publication No. 2008 / 119567, CDR-H2 as shown in Sequence ID No. 175 of International Publication No. 2008 / 119567, and CDR-H3 as shown in Sequence ID No. 176 of International Publication No. 2008 / 119567.

[0162] In another embodiment of the antibody construct of the present invention, the above 3 groups of VL CDRs are combined with the above 10 groups of VH CDRs within a second binding domain to form a (30) group comprising CDR-L1~3 and CDR-H1~3, respectively.

[0163] With respect to the antibody construct of the present invention, the second domain that binds to CD3 preferably includes a domain selected from the group consisting of VL regions as shown in Sequence IDs 17, 21, 35, 39, 53, 57, 71, 75, 89, 93, 107, 111, 125, 129, 143, 147, 161, 165, 179, or 183 of International Publication No. 2008 / 119567, or as shown in Sequence ID 696 according to the present invention.

[0164] The second domain to bind to CD3 may also preferably include a selection from the group consisting of VH regions as shown in Sequence IDs 15, 19, 33, 37, 51, 55, 69, 73, 87, 91, 105, 109, 123, 127, 141, 145, 159, 163, 177, or 181 of International Publication No. 2008 / 119567, or as shown in Sequence ID 697 according to the present invention.

[0165] More preferably, the antibody construct of the present invention is (a) the VL area as shown in Sequence ID 17 or 21 of International Publication No. 2008 / 119567 and the VH area as shown in Sequence ID 15 or 19 of International Publication No. 2008 / 119567; (b) The VL area as shown in Sequence ID 35 or 39 of International Publication No. 2008 / 119567 and the VH area as shown in Sequence ID 33 or 37 of International Publication No. 2008 / 119567; (c) The VL area as shown in Sequence ID 53 or 57 of International Publication No. 2008 / 119567 and the VH area as shown in Sequence ID 51 or 55 of International Publication No. 2008 / 119567; (d) A VL region as shown in SEQ ID NO: 71 or 75 of WO 2008 / 119567 pamphlet and a VH region as shown in SEQ ID NO: 69 or 73 of WO 2008 / 119567 pamphlet; (e) A VL region as shown in SEQ ID NO: 89 or 93 of WO 2008 / 119567 pamphlet and a VH region as shown in SEQ ID NO: 87 or 91 of WO 2008 / 119567 pamphlet; (f) A VL region as shown in SEQ ID NO: 107 or 111 of WO 2008 / 119567 pamphlet and a VH region as shown in SEQ ID NO: 105 or 109 of WO 2008 / 119567 pamphlet; (g) A VL region as shown in SEQ ID NO: 125 or 129 of WO 2008 / 119567 pamphlet and a VH region as shown in SEQ ID NO: 123 or 127 of WO 2008 / 119567 pamphlet; (h) A VL region as shown in SEQ ID NO: 143 or 147 of WO 2008 / 119567 pamphlet and a VH region as shown in SEQ ID NO: 141 or 145 of WO 2008 / 119567 pamphlet; (i) A VL region as shown in SEQ ID NO: 161 or 165 of WO 2008 / 119567 pamphlet and a VH region as shown in SEQ ID NO: 159 or 163 of WO 2008 / 119567 pamphlet; and (j) A second domain that binds to CD3 and comprises a VL region and a VH region selected from the group consisting of a VL region as shown in SEQ ID NO: 179 or 183 of WO 2008 / 119567 pamphlet and a VH region as shown in SEQ ID NO: 177 or 181 of WO 2008 / 119567 pamphlet.

[0166] A second domain that binds to CD3 and comprises a VL region as shown in SEQ ID NO: 696 and a VH region as shown in SEQ ID NO: 697 is also preferred in relation to the antibody construct of the present invention.

[0167] In other embodiments, the antibody construct of the present invention is characterized by a second domain that binds to CD3, including a VL region and a VH region selected from any of those shown in Table 80 (SEQ ID NOs: 17230-17231), 85 (SEQ ID NOs: 17302-17305), 91 (SEQ ID NOs: 17507-17542), or 97 (SEQ ID NOs: 18184-18195) of this specification.

[0168] In yet another embodiment, the antibody construct of the present invention is characterized by a second domain that binds to CD3, comprising a VL region and a VH region, which comprises a CDR selected from those shown in Table 79 (SEQ ID NOs: 17224-17229), 84 (SEQ ID NOs: 17290-17301), 89 (SEQ ID NOs: 17399-17488), 90 (SEQ ID NOs: 17417-17506), 95 (SEQ ID NOs: 18148-18177), or 96 (SEQ ID NOs: 18154-18183) of this specification.

[0169] In other embodiments, the antibody construct of the present invention is characterized by a second domain that binds to CD3 and contains a sequence selected from the sequences shown in Table 78 (SEQ ID NO: 17223) or 83 (SEQ ID NO: 17288-17289).

[0170] In yet another embodiment, the antibody construct of the present invention is characterized by a second domain that binds to CD3, comprising the sequence disclosed in U.S. Provisional Patent Application No. 63 / 110,545, entitled “CD3-binding polypeptide construct”.

[0171] antibody construct In one embodiment, the present invention provides an antibody construct comprising a first domain that binds to the pMAGE-HLA complex on the surface of a target cell and a second domain that binds to CD3 on the surface of a T cell.

[0172] The term “antibody construct” refers to a molecule whose structure and / or function is based on the structure and / or function of an antibody, for example, a full-length or whole immunoglobulin molecule. An antibody construct can bind to its specific target or antigen and includes variable heavy chain (VH) and / or variable light chain (VL) domains of the antibody or a fragment thereof. Furthermore, the domain that binds to its binding partner according to the present invention is understood herein as the binding domain of the antibody construct according to the present invention. Typically, the binding domain according to the present invention includes the minimum structural requirements of the antibody that enable target binding. These minimum requirements can be defined, for example, by the presence of at least three light chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VL region) and / or three heavy chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VH region), preferably all six CDRs. An alternative approach to defining the minimum structural requirements of an antibody is to define the epitope on which the antibody binds to the protein domain of the target protein, each constituting an epitope region (epitope cluster) within the structure of the specific target, or to define it by reference to a specific antibody that competes with the epitope of the defined antibody. Alternatively, the minimum structural requirements may be defined by the paratope sequence within the binding domain of the antibody. Antibodies on which constructs according to the present invention are based include, for example, monoclonal antibodies, recombinant antibodies, chimeric antibodies, deimmunized antibodies, humanized antibodies, and human antibodies.

[0173] The term "variable" refers to a portion of an antibody or immunoglobulin domain that exhibits variability in its sequence and is involved in determining the specificity and binding affinity of a particular antibody (i.e., a "variable domain"). The pairing of a variable heavy chain (VH) and a variable light chain (VL) both forms an antigen-binding site.

[0174] The variability is not uniformly distributed throughout the antibody's variable domain, but is concentrated in the respective subdomains of the heavy chain and light chain variable regions. These subdomains are called "hypervariable regions" or "complementarity-determining regions" (CDRs). The more conserved (i.e., non-hypervariable) portions of the variable domain are called "framework" regions (FRMs or FRs), which provide a scaffold for six CDRs in three-dimensional space to form the antigen-binding surface. The naturally occurring heavy and light chain variable domains each contain four FRM regions (FR1, FR2, FR3, and FR4), which primarily employ a β-sheet configuration, connected by three hypervariable regions that link β-sheet structures, form loop connections, and in some cases form parts of β-sheet structures. The hypervariable regions of each chain are held together in close proximity by the FRMs and, together with the hypervariable region of the other chain, contribute to the formation of the antigen-binding side (see Kabat et al., loc. cit.).

[0175] The term "CDR" and its multiple counterparts refer to complementarity-determining regions, three of which constitute the binding properties of the light chain variable region (CDR-L1, CDR-L2, and CDR-L3), and three of which constitute the binding properties of the heavy chain variable region (CDR-H1, CDR-H2, and CDR-H3). CDRs contain most of the residues involved in the specific interaction between antibodies and antigens, and therefore contribute to the functional activity of antibody molecules; they are the main determinants of antigen specificity.

[0176] The precise definitional boundaries and lengths of CDRs follow different classification and numbering systems. Therefore, a CDR may be referred to by any other boundary definition, including those of Kabat, Chothia, contact, or the numbering systems described herein. Despite the differing boundaries, each of these systems has some overlap in what constitutes the so-called "hypervariable regions" within the variable sequence. Thus, CDR definitions by these systems may differ in length and boundary regions relative to adjacent framework regions. See, for example, Kabat (an approach based on cross-species sequence variability), Chothia (an approach based on crystallographic studies of antigen-antibody complexes), and / or MacCallum (Kabat et al., loc.cit.; Chothia et al., J.MoI.Biol, 1987, 196:901-917; and MacCallum et al., J.MoI.Biol, 1996, 262:732). Another standard for characterizing the antigen-binding side is the AbM definition used by Oxford Molecular's AbM antibody modeling software. See, for example, Protein Sequence and Structure Analysis of Antibody Variable Domains. In: Antibody Engineering Lab Manual (Ed.: Duebel, S. and Kontermann, R., Springer-Verlag, Heidelberg). Two residue identification techniques can be combined to define hybrid CDRs, as long as they define overlapping but non-identical regions. However, Kabat numbering is preferred.

[0177] Typically, CDRs form loop structures that can be classified as canonical structures. The term "canonical structure" refers to the main chain conformation adopted by the antigen-binding (CDR) loop. Comparative structural studies have revealed that five of the six antigen-binding loops have only a limited repertoire of available conformations. Each canonical structure can be characterized by the twist angle of the polypeptide backbone. Therefore, corresponding loops between antibodies can have very similar three-dimensional structures despite the high amino acid sequence variability in most of the loops (Chothia and Lesk, J.MoI. Biol., 1987, 196:901; Chothia et al., Nature, 1989, 342:877; Martin and Thornton, J.MoI. Biol, 1996, 263:800). Furthermore, there is a relationship between the adopted loop structure and the surrounding amino acid sequence. The conformation of a particular canonical class is determined by the length of the loop and the presence of amino acid residues at key positions within the loop and within the conserved framework (i.e., outside the loop). Therefore, assignment to a particular canonical class can be made based on the presence of these key amino acid residues.

[0178] The term “canonical structure” may also include considerations regarding the linear sequence of the antibody, as cataloged, for example, by Kabat (Kabat et al., loc. cit.). The Kabat numbering scheme(s) is a widely adopted standard for numbering amino acid residues of antibody variable domains in a consistent manner and is the preferred scheme applied in the present invention, as mentioned elsewhere herein. Further structural considerations may also be used to determine the canonical structure of the antibody. For example, these differences not fully reflected by Kabat numbering can be explained by the numbering system of Chothia et al., as well as / or revealed by other techniques, such as crystallography and two- or three-dimensional computational modeling. Thus, a given antibody sequence may be placed in a canonical class, among other things, which allows for the identification of a suitable chassis sequence (e.g., based on the desire to include various canonical structures in the library). The Kabat numbering and structural considerations of antibody amino acid sequences, as described by Chothia et al., loc.cit., and their implications for constructing canonical forms of antibody structures, are documented in the literature. The subunit structures and three-dimensional configurations of different classes of immunoglobulins are well known in the art. For a review of antibody structures, see Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, eds. Harlow et al., 1988.

[0179] In some embodiments, the CDR3 of the light chain, and especially the CDR3 of the heavy chain, can constitute the most important determinants in antigen binding within the light chain variable region and the heavy chain variable region. In some antibody constructs, the heavy chain CDR3 appears to constitute the primary contact region between the antigen and the antibody. In vitro selection schemes that alter only the CDR3 can be used to change the binding properties of the antibody or to determine which residues contribute to antigen binding. Therefore, the CDR3 is often the greatest source of molecular diversity on the antibody-binding side. For example, H3 may be as short as two amino acid residues or may be longer than 26 amino acids.

[0180] The sequences of antibody genes after aggregation and somatic mutation are highly diverse, and it is estimated that these diverse genes encode 10¹⁰ different antibody molecules (Immunoglobulin Genes, 2nd ed., eds. Jonio et al., Academic Press, San Diego, CA, 1995). Thus, the immune system provides a repertoire of immunoglobulins. The term "repertoire" refers to at least one nucleotide sequence that is derived, whole or partially, from at least one sequence encoding at least one immunoglobulin. Sequences can be generated by in vivo rearrangement of the V, D, and J segments of the heavy chain, as well as the V and J segments of the light chain. Alternatively, sequences can be generated from cells in which rearrangement occurs in response, such as in vitro stimulation. Alternatively, some or all of a sequence can be obtained by DNA splicing, nucleotide synthesis, mutagenesis, and other methods (see, for example, U.S. Patent No. 5,565,332). A repertoire may consist of only one sequence, or it may consist of multiple sequences, including sequences in a genetically diverse aggregate.

[0181] The antibody construct of the present invention comprises at least one binding domain. In relation to the present invention, the term “binding domain” characterizes a domain that specifically binds to, specifically interacts with, or recognizes a given target epitope or target region on a target molecule, such as pMAGE-HLA or CD3. The structure and function of the first binding domain (recognizing pMAGE-HLA) and the structure and / or function of the second binding domain (recognizing CD3) are derived from the variable heavy chain (VH) and / or variable light chain (VL) domains of the antibody or a fragment thereof, based on the structure and / or function of the antibody, such as a full-length or complete immunoglobulin molecule. In certain embodiments, the first binding domain comprises the presence of three light chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VL region) and / or three heavy chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VH region). The second binding domain, if present, includes at least three light chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VL region) and / or three heavy chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VH region). In certain embodiments, the first binding domain and / or the second binding domain may be prepared or obtained by phage display or library screening, rather than by transplanting CDR sequences from an existing (monoclonal) antibody onto a scaffold.

[0182] The binding domain of the antibody construct according to the present invention may include, for example, the above-mentioned CDR group. Preferably, these CDRs are included in the framework of the antibody light chain variable region (VL) and the antibody heavy chain variable region (VH), but do not necessarily have to include both. The Fd fragment often has, for example, two VH regions and retains some antigen-binding functions of an intact antigen-binding domain. Further examples of antibody fragment, antibody variant, or binding domain formats include: (1) Fab fragment, a monovalent fragment having VL, VH, CL, and CH1 domains; (2) F(ab')2 fragment, a bivalent fragment having two Fab fragments linked by disulfide crosslinks at the hinge region; (3) Fd fragment having two VH domains and a CH1 domain; (4) Fv fragment having VL and VH domains in a single arm of the antibody; (5) dAb fragment having a VH domain (Ward et al., (1989) Nature 341:544-546); (6) isolated complementarity-determining regions (CDRs); and (7) single-chain Fv (scFv), the latter of which is preferred (e.g., derived from scFV libraries). Non-limiting examples of embodiments of the antibody constructs according to the present invention are described, for example, in International Publication No. 00 / 006605, International Publication No. 2005 / 040220, International Publication No. 2008 / 119567, International Publication No. 2010 / 037838, International Publication No. 2013 / 026837, International Publication No. 2013 / 026833, U.S. Patent Application Publication No. 2014 / 0308285, U.S. Patent Application Publication No. 2014 / 0302037, International Publication No. 2014 / 144722, International Publication No. 2014 / 151910, and International Publication No. 2015 / 048272.

[0183] Furthermore, the definition of "binding domain" or "binding domain" includes fragments of full-length antibodies, such as VH, VHH, VL, (s)dAb, Fv, light chain (VL-CL), Fd(VH-CH1), heavy chain, Fab, Fab', F(ab')2, or "rIgG" (a "half-antibody" consisting of a heavy chain and a light chain). The antibody constructs according to the present invention may also include modified fragments of antibodies, also called antibody variants or antibody derivatives. Examples include, but are not limited to, the following structures: (VH-VL-CH3)2, (scFv-CH3)2, ((scFv)2-CH3+CH3), ((scFv)2-CH3) or (scFv-CH3-scFv)2, multibodies such as triabodies or tetrabodies, and single-domain antibodies such as nanobodies, or single variable-domain antibodies containing only one variable region, which may be VHH, VH, or VL that specifically bind to an antigen or target independently of other variable regions or domains. Further possible forms of antibody constructs according to the present invention include crossovers, maximum entities, hetero-Fc constructs, mono-Fc constructs, and scFc constructs. Examples of these formats are described below herein.

[0184] According to the present invention, the binding domain is in the form of one or more polypeptides. Such polypeptides may comprise a proteinaceous portion and a non-proteinaceous portion (e.g., a chemical linker or chemical crosslinking agent, e.g., glutaraldehyde). Proteins (including fragments thereof, preferably biologically active fragments, and peptides having typically fewer than 30 amino acids) comprise two or more amino acids linked to one another via covalent peptide bonds (resulting in a chain of amino acids).

[0185] As used herein, the term "polypeptide" refers to a group of molecules consisting of more than 30 amino acids. Polypeptides can further form multimers such as dimers, trimers, and higher oligomers, i.e., multimers composed of two or more polypeptide molecules. The polypeptide molecules forming such dimers, trimers, etc. may be the same or different. The corresponding higher-order structures of such multimers are consequently referred to as homo- or hetero-dimers, homo- or hetero-trimers, etc. An example of a heteromultimer is, in its naturally occurring form, an antibody molecule composed of two identical light polypeptide chains and two identical heavy polypeptide chains. The terms "peptide", "polypeptide", and "protein" also refer to naturally modified peptides / polypeptides / proteins where the modification is effected by post-translational modifications such as glycosylation, acetylation, phosphorylation, etc. When referred to herein, "peptide", "polypeptide", or "protein" may be chemically modified such as by pegylation. Such modifications are well known in the art and are described below herein.

[0186] The term "hinge" refers to the IgG hinge region. This region can be identified by analogy using Kabat numbering (see, for example, Kabat positions 223 - 243). Consistent with the above, the minimal requirement for "hinge" is the amino acid residues corresponding to the IgG1 sequence extension of D231 - P243 by Kabat numbering.

[0187] In accordance with this invention, the terms "CH2" and "CH3" refer to constant regions 2 and 3 of the immunoglobulin heavy chain. These regions can also be identified by analogy using Kabat numbering; for example, refer to Kabat positions 244-360 for CH2 and Kabat positions 361-478 for CH3. It is understood that there is some variation among immunoglobulins with respect to the IgG1 Fc region, IgG2 Fc region, IgG3 Fc region, IgG4 Fc region, IgM Fc region, IgA Fc region, IgD Fc region and IgE Fc region (see, for example, Padlan, Molecular Immunology, 31(3), 169-217 (1993)).

[0188] The term "Fc region" refers to the last two heavy chain constant regions of IgA, IgD, and IgG, as well as the last three heavy chain constant regions of IgE and IgM. The Fc region may also contain the flexible hinge N-terminus to these domains. In the case of IgA and IgM, the Fc region may include the J chain. In the case of IgG, the Fc region includes the immunoglobulin domains CH2 and CH3, and the hinge between the first two domains and CH2. While the boundaries of the Fc region of immunoglobulins can vary, an example of a human IgG heavy chain Fc portion containing the functional hinge, CH2 and CH3 domains can be defined, for example, as containing residues D231 (of the hinge domain) to P476 (of the C-terminus of the CH3 domain) or D231 to L476 in IgG4, respectively, with numbering following Kabat.

[0189] The terms “Fc moiety” or “Fc monomer” in relation to the present invention mean a polypeptide comprising at least one domain having the function of the CH2 domain and at least one domain having the function of the CH3 domain of an immunoglobulin molecule. As is evident from the term “Fc monomer,” a polypeptide comprising these CH domains is a “polypeptide monomer.” An Fc monomer may be a polypeptide comprising at least a fragment of the constant region of an immunoglobulin, excluding the first constant region immunoglobulin domain (CH1) of the heavy chain, but maintaining at least a functional portion of one CH2 domain and a functional portion of one CH3 domain, where the CH2 domain is the amino terminus of the CH3 domain.

[0190] In one embodiment of this definition, the Fc monomer may be a polypeptide constant region comprising an Ig-Fc hinge region, a CH2 region, and a portion of the CH3 region, where the hinge region is the amino terminus of the CH2 domain. The hinge region of the present invention is intended to facilitate dimerization. Such an Fc polypeptide molecule can be obtained, for example, by papain digestion of an immunoglobulin region (which, of course, results in a dimer of two Fc polypeptides). In another embodiment of this definition, the Fc monomer may be a polypeptide region comprising a CH2 region and a portion of the CH3 region. Such an Fc polypeptide molecule can be obtained, for example, by pepsin digestion of an immunoglobulin molecule.

[0191] In one embodiment, the polypeptide sequence of the Fc monomer is substantially similar to the Fc polypeptide sequences of the IgG1Fc region, IgG2Fc region, IgG3Fc region, IgG4Fc region, IgM Fc region, IgA Fc region, IgD Fc region, and IgE Fc region. (See, for example, Padlan, Molecular Immunology, 31(3), 169-217 (1993).) There is some variation among immunoglobulins, and for the sake of clarity, the Fc monomer refers to the last two heavy chain constant region immunoglobulin domains of IgA, IgD, and IgG, as well as the last three heavy chain constant region immunoglobulin domains of IgE and IgM. As mentioned, the Fc monomer may also contain a flexible hinge at the N-terminus of these domains. In the case of IgA and IgM, the Fc monomer may contain a J chain. In the case of IgG, the Fc portion includes the immunoglobulin domains CH2 and CH3, and a hinge between the first two domains and CH2. The boundary of the Fc moiety may vary, although an example of a human IgG heavy chain Fc moiety containing a functional hinge may be found. The CH2 and CH3 domains can be defined, for example, as containing residues D231-P476 (corresponding to D234 in Table 3 below) of the hinge domain, and L476 (in the case of IgG4) of the carboxyl terminology of the CH3 domain, respectively, with numbering following Kabat. Two Fc moieties or Fc monomers fused to each other via a peptide linker define a third domain of the antibody construct of the present invention, which may also be defined as an scFc domain.

[0192] In one embodiment of the present invention, the scFc domains disclosed herein, and the Fc monomers fused to each other, are assumed to be included only in the third domain of the antibody construct.

[0193] In some embodiments, the IgG hinge region can be identified by analogy using Kabat numbering as shown in Table 3. Accordingly, for the hinge domain / region of the present invention, the minimum requirement is assumed to be the inclusion of amino acid residues corresponding to the IgG1 sequence extension D231 D234-P243 by Kabat numbering. Similarly, the hinge domain / region of the present invention is assumed to include or consist of the IgG1 hinge sequence DKTHTCPPCP (SEQ ID NO: 4) (modification of the sequence is also assumed if the hinge region corresponding to the extension D234-P243 shown in Table 3 below still promotes dimerization). In preferred embodiments of the present invention, the glycosylation site at Kabat position 314 of the CH2 domain in the third domain of the antibody construct is removed by an N314X substitution, where X is any amino acid except Q. The substitution is preferably an N314G substitution. In a more preferred embodiment, the CH2 domain further comprises the following substitutions (positions according to Kabat) V321C and R309C (these substitutions introduce intradomain cysteine ​​disulfide bridges at Kabat positions 309 and 321).

[0194] [Table 3]

[0195] In further embodiments of the present invention, the hinge domain / region includes or comprises the IgG2 subtype hinge sequence ERKCCVECPPCP (SEQ ID NO: 5), the IgG3 subtype hinge sequence ELKTPLDTTHTCPRCP (SEQ ID NO: 6) or ELKTPLGDTTHTCPRCP (SEQ ID NO: 7), and / or the IgG4 subtype hinge sequence ESKYGPPCPSCP (SEQ ID NO: 8). The IgG1 subtype hinge sequence may be one of the following EPKSCDKTHTCPPCP (SEQ ID NO: 9). Thus, these core hinge regions are also assumed in connection with the present invention.

[0196] The locations and sequences of the IgG CH2 and IgGCD3 domains can be identified by analogy using the Kabat numbering shown in Table 4.

[0197] [Table 4]

[0198] In one embodiment of the present invention, the highlighted bold amino acid residues in the CH3 domain of the first or both Fc monomers are deleted.

[0199] In classical full-length antibodies or immunoglobulins, each light (L) chain is linked to a heavy (H) chain by one covalent disulfide bond, and the two H chains are linked to each other by one or more disulfide bonds depending on the H chain isotype. The heavy chain constant (CH) domain closest to VH is usually called CH1. The constant ("C") domain does not directly participate in antigen binding but exhibits various effector functions such as antibody-dependent cell-mediated cytotoxicity (ADCC) and complement activation (complement-dependent cytotoxicity, CDC). The Fc region of an antibody is the "tail" region of classical antibodies that interacts with cell surface receptors called Fc receptors and several proteins of the complement system. In IgG, IgA, and IgD antibody isotypes, the Fc region consists of two identical protein fragments derived from the second and third constant domains (CH2 and CH3) of the two heavy chains of the antibody. The IgM and IgE Fc regions contain three heavy chain constant domains (CH2, CH3, and CH4) in each polypeptide chain. The Fc region also contains a portion of a so-called "hinge" region held together by one or more disulfide and non-covalent interactions. The Fc region of naturally occurring IgG has a highly conserved N-glycosylation site. Glycosylation of the Fc fragment is essential for Fc receptor-mediated activity.

[0200] In molecules containing the Fc region, antibody-dependent cell-mediated cytotoxicity ("ADCC"), another mechanism of cell killing mediated by the Fc region is considered. ADCC is a mechanism of cell-mediated immune defense in which effector cells of the immune system actively lyse target cells, their membrane surface antigens bound by specific antibodies. ADCC requires immune effector cells, classically known to be natural killer (NK) cells that typically interact with IgG antibodies. However, ADCC can also be mediated by macrophages, neutrophils, and eosinophils. ADCC involves activation of effector cells expressing the Fc receptor by antibodies expressing the Fc region. For example, the most common Fc receptor on the surface of NK cells is called CD16 or FcγRIII. When the Fc receptor binds to the Fc region of IgG, NK cells release cytotoxic factors that cause the death of target cells. Similarly, the Fc receptor on eosinophils (FceRI) recognizes IgE. In contrast, in CDC, the complement system molecule "C1q" binds to the antibody Fc region, and this binding triggers the complement cascade, resulting in the formation of membrane invasion complexes (MACs) on the surface of target cells as a result of classical pathway complement activation. In therapeutic antibodies or antibody constructs, both ADCC and CDC can be modulated by Fc isotype engineering, Fc gene mutations, or modification of the Fc glycosylation profile.

[0201] In some embodiments of the present invention, the CH2 domain of one or preferably each (both) of the third domains of the polypeptide monomer contains an intradomain cysteine ​​disulfide crosslink. As is known in the art, the term “cysteine ​​disulfide crosslink” refers to a functional group having the general structure RSSR. This linkage, also called an SS bond, disulfide crosslink, or cysteine ​​clamp, is obtained by the coupling of two thiol groups of a cysteine ​​residue. In certain embodiments, the cysteine ​​forming the cysteine ​​disulfide crosslink in the mature antibody construct is introduced into the amino acid sequence of the CH2 domain corresponding to 309 and 321 (Kabat numbering). In other embodiments, the cysteine ​​clamp is introduced in other domains of the antibody construct. See, for example, U.S. Patent No. 2016 / 0193295.

[0202] In one embodiment of the present invention, the glycosylation site at Kabat position 314 of the CH2 domain is removed. This removal of the glycosylation site is preferably achieved by an N314X substitution, where X is any amino acid other than Q. The substitution is preferably N314G. In a more preferred embodiment, the CH2 domain further comprises the following substitutions (positions according to Kabat): V321C and R309C (these substitutions introduce intradomain cysteine ​​disulfide crosslinks at Kabat positions 309 and 321).

[0203] The definition of "antibody" according to the present invention includes full-length antibodies, including camel antibodies and other immunoglobulins produced by biotechnological or protein engineering methods or processes. These full-length antibodies may be, for example, monoclonal antibodies, recombinant antibodies, chimeric antibodies, deimmunized antibodies, humanized antibodies and human antibodies, as well as antibodies from other species such as mice, hamsters, rabbits, rats, goats or non-human primates.

[0204] Given that antibody constructs according to some embodiments of the present invention include one domain that binds to pMAGE-HLA and, alternatively, another domain that binds to CD3, these do not exist in nature, and their functions differ significantly from those of naturally occurring products. Therefore, the antibody constructs of the present invention are artificial molecules comprising at least two different binding domains having different specificities.

[0205] As used herein, the terms “single-chain Fv,” “single-chain antibody,” or “scFv” refer to a single polypeptide chain antibody fragment that contains variable regions from both the heavy and light chains but lacks a constant region. Generally, single-chain antibodies further include a polypeptide linker between the VH and VL domains, which enables the formation of a desired structure that allows antigen binding. Single-chain antibodies are discussed in detail by Pluckthun in The Pharmacology of Monoclonal Antibodies, vol. 113, Rosenburg and Moore eds. Springer-Verlag, New York, pp. 269-315 (1994). Various methods for producing single-chain antibodies are known, including, for example, those described in U.S. Patent Nos. 4,694,778 and 5,260,203; International Publication No. 88 / 01649; Bird (1988) Science 242:423-442; Huston et al. (1988) Proc. Natl. Acad. Sci. USA 85:5879-5883; Ward et al. (1989) Nature 334:54454; and Skerra et al. (1988) Science 242:1038-1041. In certain embodiments, single-chain antibodies may also be bispecific, multispecific, human, and / or humanized, and / or synthetic.

[0206] In certain embodiments, either the first, second, or first and second domains may comprise a single-domain antibody, each containing at least a variable domain or CDR of a single-domain antibody. A single-domain antibody contains only one (monomer) antibody variable domain that can selectively bind to a specific antigen independently of other V regions or domains. The first single-domain antibodies are engineered from heavy-chain antibodies found in camelids, which are called VHH fragments. Cartilaginous fish also have heavy-chain antibodies (IgNARs) from which single-domain antibodies called VNAR fragments can be obtained. Another approach is to split dimeric variable domains from common immunoglobulins, e.g., human or rodent, into monomers, thus obtaining VH or VL as single-domain Abs. While most research on single-domain antibodies is currently based on heavy-chain variable domains, nanobodies derived from light chains have also been shown to bind specifically to target epitopes. Examples of single-domain antibodies are called sdAbs, nanobodies, or single variable-domain antibodies.

[0207] Therefore, (single-domain mAb)2 is a monoclonal antibody construct comprising (at least) two single-domain monoclonal antibodies individually selected from the group including VH, VL, VHH, and VNAR. The linker is preferably in the form of a peptide linker. Similarly, "scFv single-domain mAb" is a monoclonal antibody construct comprising at least one of the single-domain antibodies and one of the scFv molecules. Again, the linker is preferably in the form of a peptide linker.

[0208] Furthermore, the definition of the term "antibody construct" includes monovalent, divalent, and polyvalent / multivalent constructs, and therefore bispecific constructs that specifically bind to only two antigenic structures, as well as polyspecific / multispecific constructs that specifically bind to more than two antigenic structures, e.g., three, four, or more antigenic structures, via different binding domains. Furthermore, the definition of the term "antibody construct" includes molecules consisting of only one polypeptide chain, as well as molecules consisting of two or more polypeptide chains, where these chains may be identical (homodimer, homotrimer, or homooligomer) or different (heterodimer, heterotrimer, or heterooligomer). Examples of the antibodies and their variants or derivatives identified above are described, in particular, in Harlow and Lane, Antibodies: A Laboratory Manual, CSHL Press (1988) and Using Antibodies: A Laboratory Manual, CSHL Press (1999), Kontermann and Duebel, Antibody Engineering, Springer, 2nd ed. 2010, and Little, Recombinant Antibodies for Immunotherapy, Cambridge University Press 2009.

[0209] As used herein, the term “bispecificity” refers to an antibody construct that is “at least bispecific,” i.e., an antibody construct comprising at least a first binding domain and a second binding domain, wherein the first binding domain binds to one antigen or target (e.g., the MAGEB2 peptide GVYDGEEHSV in association with HLA, SEQ ID NO: 1) and the second binding domain binds to another antigen or target (e.g., CD3). Thus, the antibody constructs according to the present invention include specificity to at least two different antigens or targets. For example, the first domain preferably does not bind to one or more extracellular epitopes of CD3ε as described herein. The term “target cell surface antigen” refers to an antigenic structure expressed by a cell and present on the cell surface so that the antibody constructs described herein are available. This may be a protein, preferably the extracellular portion of a protein, or a carbohydrate structure, preferably a carbohydrate structure of a protein such as a glycoprotein. It is preferably a tumor antigen, and in more specific embodiments, a peptide tumor antigen presented on the surface of tumor cells by HLA. The term "bispecific antibody construct" in this invention also encompasses multispecific antibody constructs such as triplicate antibody constructs, the latter of which are constructs containing three binding domains or having more than three (e.g., 4, 5...) specificities.

[0210] In certain embodiments, the first and second domains of the antibody construct of the present invention are "bispecific single-strand antibody constructs" that include, but are not limited to, a bispecific "single-strand Fv" (scFv). The two domains of the Fv fragment, VL and VH, are encoded by separate genes, but they can be linked by a synthetic linker (as described above herein) that allows the VL and VH regions to pair up to form a single protein chain that forms a monovalent molecule, using recombination. See, for example, Huston et al. (1988) Proc. Natl. Acad. Sci USA 85:5879-5883). These antibody fragments can be obtained using conventional techniques known to those skilled in the art, and the fragments can be evaluated for function in the same manner as whole or full-length antibodies. Therefore, a single-chain variable fragment (scFv) is a fusion protein of the variable region (VH) of the heavy chain and the variable region (VL) of the light chain of an immunoglobulin, typically linked by a short linker peptide of about 10 to 25 amino acids, preferably about 15 to 20 amino acids. The linker is usually rich in glycine for flexibility and serine or threonine for solubility, and can connect the N-terminus of VH to the C-terminus of VL, or vice versa. This protein retains the specificity of the original immunoglobulin despite the removal of the constant region and the introduction of the linker.

[0211] Bispecific single-chain antibody constructs are known in the art and are described in International Publication No. 99 / 54440, Mack, J. Immunol. (1997), 158, 3965-3970, Mack, PNAS, (1995), 92, 7021-7025, Kufer, Cancer Immunol. Immunother., (1997), 45, 193-197, Loeffler, Blood, (2000), 95, 6, 2098-2103, Bruehl, Immunol., (2001), 166, 2420-2426, and Kipriyanov, J. Mol. Biol., (1999), 293, 41-56. Techniques described for the production of single-chain antibodies (see, in particular, U.S. Patent No. 4,946,778, Kontermann and Duebel (2010), loc.cit., and Little (2009), loc.cit.) can be adapted to produce single-chain antibody constructs that specifically recognize selected targets.

[0212] Divalent (also called bivalent) or bispecific single-strand variable fragments (bi-scFvs or di-scFvs) having format (scFv)2 can be manipulated by linking two scFv molecules (e.g., having a linker as described herein). If these two scFv molecules have the same binding specificity, the resulting (scFv)2 molecule is preferably called divalent (i.e., having two valencies for the same target epitope). If the two scFv molecules have different binding specificities, the resulting (scFv)2 molecule is preferably called bispecific. Linking can be performed by generating a single peptide chain having two VH regions and two VL regions to obtain a tandem scFv (see, for example, Kufer P. et al., (2004) Trends in Biotechnology 22(5):238-244). Another possibility is the creation of scFv molecules with linker peptides (e.g., about 5 amino acids) that are too short for the two variable regions to fold together, forcing the scFv to dimerize. This type is known as a diabody (see, for example, Hollinger, Philipp et al., (July 1993) Proceedings of the National Academy of Sciences of the United States of America 90(14):6444-8).

[0213] Bispecific antibody constructs can be produced by various methods, including hybridoma fusion or Fab' fragment linking. See, for example, Songsivilai & Lachmann, Clin. Exp. Immunol. 79:315-321 (1990). Bispecific antibody constructs can also be produced by other methods that would be apparent to those skilled in the art, including, for example, the use of any of the binding domain sequences provided herein.

[0214] Bispecific antibody-derived molecules, such as BiTE® antibody constructs, are recombinant protein constructs constructed from two flexibly linked antibody-derived binding domains. One binding domain of a BiTE® antibody construct is specific to a selected tumor-associated surface antigen on target cells, and the second binding domain is specific to CD3, a subunit of the T cell receptor complex on T cells. Due to their specific design, BiTE® antibody constructs are uniquely suited to transiently connecting T cells to target cells while simultaneously potently activating the intrinsic cytolytic ability of T cells against target cells. A significant further development of the first generation of BiTE® antibody constructs (see International Publication No. 99 / 54440 and International Publication No. 2005 / 040220) into clinical use as AMG103 and AMG110 was the provision of bispecific antibody constructs that bind to the context-independent epitope at the N-terminus of the CD3 ε chain (International Publication No. 2008 / 119567). The BiTE® antibody constructs that bind to this selected epitope not only exhibit cross-species specificity to human and macaque, or common marmoset (Callithrix jacchus), cotton-top tamarin (Saguinus oedipus), or common squirrel monkey (Saimiri sciureus) CD3ε chains, but also recognize this specific epitope (instead of the previously described epitopes of CD3 conjugates in bispecific T cell engagement molecules), thus not demonstrating the same degree of nonspecific T cell activation as observed with previous generations of T cell engagement antibodies. This reduction in T cell activation is associated with decreased or reduced T cell redistribution in patients, the latter of which has been identified as a side effect, e.g., a risk with passutuximab.

[0215] Antibody constructs, such as those described in International Publication No. 2008 / 119567, are characterized by rapid clearance from the body and can therefore reach a large portion of the body quickly; however, their in vivo application may be limited by their short persistence in vivo. On the other hand, their concentrations in the body can be quickly adapted and fine-tuned. Due to the short in vivo half-life of this small single-chain molecule, long-term administration by continuous intravenous infusion is used to achieve therapeutic effects. However, bispecific antibody constructs with more favorable pharmacokinetic properties, including longer half-lives, are now available. Extending the half-life is generally useful for the in vivo application of immunoglobulins, particularly antibodies, and especially small antibody fragments or constructs, for purposes such as patient compliance.

[0216] In some embodiments, the antibody constructs of the present invention are “in vitro generated antibody constructs.” This term refers to an antibody construct as defined above in which all or part of the variable region (e.g., at least one CDR) is generated by non-immune cell selection, e.g., in vitro phage display, protein chip, or any other method that can be used to test the ability of the candidate sequence to bind to an antigen. In other embodiments, the antibody construct sequence is generated by genomic rearrangement in animal immune cells. Therefore, this term preferably excludes sequences that are generated solely by genomic rearrangement in animal immune cells. It is assumed that the first and / or second domains of the antibody construct are produced or can be obtained by phage display or library screening rather than by transplanting a CDR sequence from an existing (monoclonal) antibody onto a scaffold. However, the use of CDR sequences from monoclonal antibodies prepared as taught in this disclosure is not excluded and will be readily apparent to those skilled in the art.

[0217] A "recombinant antibody" is an antibody produced using recombinant DNA technology or genetic engineering.

[0218] As used herein, the terms “monoclonal antibody” (mAb) or “monoclonal antibody construct” refer to an antibody obtained from a substantially homogeneous population of antibodies; that is, the individual antibodies within the population are identical except for minor, naturally occurring mutations and / or post-translational modifications (e.g., isomerization, amidation) that may be present in small amounts, and refer to an overall polypeptide structure similar to naturally occurring antibodies as understood by those skilled in the art (e.g., IgG with a structure of two heavy chains and two light chains). Monoclonal antibodies are highly specific, in contrast to conventional (polyclonal) antibody preparations which typically contain different antibodies against different antigen sites or epitopes, as they target a single antigen site or epitope on an antigen. In addition to their specificity, monoclonal antibodies have the advantage that they are synthesized by clonal cell culture and are not contaminated by other immunoglobulins. The modifier “monoclonal” indicates the characteristic of an antibody obtained from a substantially homogeneous population of antibodies and should not be interpreted as requiring antibody production by a specific method.

[0219] Any technique that provides antibodies produced by serial cell line culture can be used to prepare monoclonal antibodies. For example, the monoclonal antibodies used may be produced by the hybridoma method first described by Koehler et al., Nature, 256:495 (1975), or by the recombinant DNA method (see, for example, U.S. Patent No. 4,816,567). Further examples of techniques for producing human monoclonal antibodies include the trioma technique, the human B-cell hybridoma technique (Kozbor, Immunology Today 4 (1983), 72), and the EBV-hybridoma technique (Cole et al., Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, Inc. (1985), 77-96).

[0220] Next, hybridomas can be screened using standard methods such as enzyme-linked immunosorbent assay (ELISA) and surface plasmon resonance analysis, e.g., Biacore®, to identify one or more hybridomas that produce antibodies that specifically bind to a particular antigen. Any form of the relevant antigen, e.g., recombinant antigen, naturally occurring form, any variant or fragment thereof, and its antigenic peptide can be used as an immunogen. Surface plasmon resonance, as used in the Biacore system, can be used to enhance the efficiency of phage antibodies that bind to the epitopes of target cell surface antigens (Schier, Human Antibodies Hybridomas 7(1996), 97-105; Malmborg, J.Immunol.Methods 183(1995), 7-13).

[0221] Another exemplary method for producing monoclonal antibodies involves screening protein expression libraries, such as phage display or ribosome display libraries. Phage display is described, for example, in Ladner et al., U.S. Patent No. 5,223,409; Smith (1985), Science 228:1315-1317; Clackson et al., Nature 352:624-628 (1991); and Marks et al., J. Mol. Biol. 222:581-597 (1991).

[0222] In addition to using display libraries, non-human animals, such as rodents (e.g., mice, hamsters, rabbits, or rats), can be immunized using relevant antigens. In one embodiment, the non-human animal contains at least a portion of the human immunoglobulin gene. For example, a large fragment of the human Ig (immunoglobulin) locus can be used to manipulate mouse strains that are deficient in mouse antibody production. Using hybridoma technology, antigen-specific monoclonal antibodies derived from genes with desired specificity can be produced and selected. See, for example, XENOMOUSE®, Green et al. (1994) Nature Genetics 7:13-21, U.S. Patent Application Publication No. 2003-0070185, International Publication Brochure No. 96 / 34096 and International Publication Brochure No. 96 / 33735.

[0223] Monoclonal antibodies can also be obtained from non-human animals and then modified, for example, by humanization, deimmunization, chimerization, etc., using recombinant DNA techniques known in the art. Examples of modified antibody constructs include humanized variants of non-human antibodies, "affinity-mature" antibodies (see, e.g., Hawkins et al. J. Mol. Biol. 254, 889-896 (1992) and Lowman et al., Biochemistry 30, 10832-10837 (1991)) and antibody variants with modified effector function (see, e.g., U.S. Patent No. 5,648,260, Kontermann and Dubel (2010), loc. cit. and Little (2009), loc. cit.).

[0224] In immunology, affinity maturation is the process by which B cells produce antibodies with increased affinity for an antigen during the course of an immune response. Repeated exposure to the same antigen causes the host to continuously produce antibodies with higher affinity. Similar to natural prototypes, in vitro affinity maturation is based on the principles of mutation and selection. In vitro affinity maturation has been successfully used to optimize antibodies, antibody constructs, and antibody fragments. Random mutations within CDRs are introduced using radiation, chemical mutagens, or error-prone PCR. Furthermore, genetic diversity can be increased by chain shuffling. Two or three mutations and selections using display methods such as phage display typically result in antibody fragments with affinities in the low nanomolar range.

[0225] A preferred type of amino acid substitution mutation in antibody constructs involves substituting one or more hypervariable region residues of the parent antibody (e.g., a humanized antibody or a human antibody). Generally, the resulting variants selected for further development have improved biological properties compared to the parent antibody from which they are produced. A simple method for producing such substitution variants involves affinity maturation using phage display. Briefly, several hypervariable region sides (e.g., sides 6-7) are mutated to generate all possible amino acid substitutions on each side. The antibody variants thus produced are presented in a monovalent manner from filamentous phage particles as fusions to the gene III product of M13, packaged within each particle. The phage display variants are then screened for their biological activity (e.g., binding affinity) as disclosed herein. To identify candidate hypervariable region sides for modification, alanine scanning mutagenesis can be performed to identify hypervariable region residues that significantly contribute to antigen binding. Alternatively, or in addition, it may be beneficial to analyze the crystalline structure of the antigen-antibody complex to identify contact points between the binding domain and, for example, pMAGE-HLA. Such contact residues and adjacent residues are candidates for substitution by the techniques detailed herein. Once such variants are generated, a panel of variants can be subjected to the screening described herein, and antibodies exhibiting superior properties in one or more relevant assays can be selected for further development.

[0226] The antibody constructs of the present invention particularly include “chimeric” antibodies (immunoglobulins) in which a portion of the heavy chain and / or light chain is identical or homologous to a corresponding sequence of an antibody derived from a particular species or belonging to a particular antibody class or subclass, and the remainder of the chain is identical or homologous to a corresponding sequence of an antibody derived from another species or belonging to another antibody class or subclass, and to the extent that they exhibit the desired biological activity (U.S. Patent No. 4,816,567; Morrison et al., Proc. Natl. Acad. Sci. USA, 81:6851-6855 (1984)). The chimeric antibodies of interest as used herein include “primatized” antibodies that contain a variable domain antigen-binding sequence and a human constant region sequence derived from a non-human primate (e.g., Old World monkeys, apes, etc.). Various approaches for producing chimeric antibodies are described. For example, see Morrison et al., Proc. Natl. Acad. ScL USA 81:6851, 1985; Takeda et al., Nature 314:452, 1985; Cabilly et al., U.S. Patent No. 4,816,567; Boss et al., U.S. Patent No. 4,816,397; Tanaguchi et al., European Patent No. 0171496; European Patent No. 0173494; and British Patent No. 2177096.

[0227] Antibodies, antibody constructs, antibody fragments, or antibody variants may also be modified by specific deletion of human T cell epitopes (a method called "deimmunization"), for example, by methods disclosed in International Publication No. 98 / 52976 or International Publication No. 00 / 34317. Briefly, the heavy and light chain variable domains of antibodies can be analyzed for MHC class II-binding peptides, which represent potential T cell epitopes (as defined in International Publication No. 98 / 52976 and International Publication No. 00 / 34317). To detect potential T cell epitopes, a computer modeling technique called "peptide threading" can be applied, and furthermore, databases of human MHC class II-binding peptides can be searched for motifs present in VH and VL sequences, as described in International Publication No. 98 / 52976 and International Publication No. 00 / 34317. These motifs bind to any of the 18 major MHC class I DR allotypes and thus constitute potential T cell epitopes. Potential T cell epitopes detected can be eliminated by substituting a small number of amino acid residues in the variable domain, or preferably by single amino acid substitutions. Typically, conservative substitutions are performed. In many cases, but not limited to, amino acids common to the positions in human germline antibody sequences may be used. Human germline sequences are disclosed, for example, in Tomlinson, et al. (1992) J. Mo I. Biol. 227:776-798; Cook, GP et al. (1995) Immunol. Today Vol. 16(5):237-242; and Tomlinson et al. (1995) EMBO J. 14:14:4628-4638. The V BASE directory provides a comprehensive directory of human immunoglobulin variable region sequences (compiled by Tomlinson, LA. et al., MRC Centre for Protein Engineering, Cambridge, UK). These sequences can be used, for example, as a source of human sequences for framework regions and CDRs.For example, the consensus human framework area can also be used, as described in U.S. Patent No. 6,300,064.

[0228] (e.g., Fv, Fab, Fab', F(ab')2, or other antigen-binding subsequences of the antibody) "Humanized" antibodies, antibody constructs, variants, or fragments thereof are antibodies or immunoglobulins of primarily human sequences that include minimal sequences derived from non-human immunoglobulins. In most cases, a humanized antibody is a human immunoglobulin (recipient antibody) in which residues from the hypervariable region (also known as the CDR) of the recipient are replaced by residues from the hypervariable region of a non-human (e.g., rodent) species (donor antibody), e.g., mouse, rat, hamster, or rabbit, having the desired specificity, affinity, and capability. In some examples, Fv framework region (FR) residues of human immunoglobulin are replaced by corresponding non-human residues. Furthermore, as used herein, "humanized antibody" may also include residues not found in either the recipient antibody or the donor antibody. These modifications are made to further improve and optimize the performance of the antibody. Humanized antibodies may also contain at least a portion of the immunoglobulin constant region (Fc), typically at least a portion of the immunoglobulin constant region (Fc) of human immunoglobulin. For further details, see Jones et al., Nature, 321:522-525 (1986); Reichmann et al., Nature, 332:323-329 (1988); and Presta, Curr. Op. Struct. Biol., 2:593-596 (1992).

[0229] Humanized antibodies or fragments thereof can be produced by replacing the sequence of the Fv variable domain, which is not directly involved in antigen binding, with an equivalent sequence derived from the human Fv variable domain. Exemplary methods for producing humanized antibodies or fragments thereof are provided by Morrison (1985) Science 229:1202-1207; Oi et al. (1986) BioTechniques 4:214; and U.S. Patent Nos. 5,585,089; 5,693,761; 5,693,762; 5,859,205; and 6,407,213. These methods involve isolating, manipulating, and expressing a nucleic acid sequence encoding all or part of the immunoglobulin Fv variable domain from at least one heavy or light chain. Such nucleic acids can be obtained from hybridomas producing antibodies against a given target, as described above, as well as from other sources. Next, recombinant DNA encoding a humanized antibody molecule can be cloned into a suitable expression vector.

[0230] Humanized antibodies can also be produced using transgenic animals such as mice that express human heavy chain and light chain genes but cannot express endogenous mouse immunoglobulin heavy chain and light chain genes. Winter describes exemplary CDR grafting methods that may be used to prepare the humanized antibodies described herein (U.S. Patent No. 5,225,539). All of the CDRs of a particular human antibody may be replaced with at least some of the non-human CDRs, or only some of the CDRs may be replaced with non-human CDRs. Only the number of CDRs required for the humanized antibody to bind to a given antigen needs to be replaced.

[0231] Humanized antibodies can be optimized by introducing conservative substitutions, consensus sequence substitutions, germline substitutions, and / or reverse mutations. Such altered immunoglobulin molecules can be prepared by any of several techniques known in the art (e.g., Teng et al., Proc. Natl. Acad. Sci. USA, 80:7308-7312, 1983; Kozbor et al., Immunology Today, 4:7279, 1983; Olsson et al., Meth. Enzymol., 92:3-16, 1982; and European Patent No. 239400).

[0232] The terms “human antibody,” “human antibody construct,” and “human binding domain” include antibodies, antibody constructs, and binding domains having antibody regions such as variable regions and constant regions or domains substantially corresponding to human germline immunoglobulin sequences known in the art, as described, for example, by Kabat et al. (1991) (loc.cit.). The human antibodies, antibody constructs, or binding domains of the present invention may include human germline immunoglobulin sequences (e.g., mutations introduced by random or lateral-specific mutagenesis in vitro or somatic mutation in vivo), such as CDRs, particularly amino acid residues not encoded by CDR3. The human antibodies, antibody constructs, or binding domains may have at least 1, 2, 3, 4, 5 or more positions substituted with amino acid residues not encoded by the human germline immunoglobulin sequence. The definitions of human antibodies, antibody constructs, and binding domains as used herein also intend fully human antibodies that consist only of non-artificial and / or genetically modified human sequences of antibodies that can be induced by using technologies or systems such as Xenomouse. Preferably, the "fully human antibody" does not contain amino acid residues not encoded by a human germline immunoglobulin sequence.

[0233] Preferably, the binding domain that binds to pMAGE-HLA and / or the binding domain that binds to CD3ε is a human binding domain. Antibodies and antibody constructs containing at least one human binding domain avoid some of the problems associated with antibodies or antibody constructs that have non-human regions such as rodent (e.g., mouse, rat, hamster, or rabbit) variable regions and / or constant regions. The presence of such rodent-derived proteins may result in rapid clearance of the antibody or antibody construct or in the patient generating an immune response to the antibody or antibody construct. To avoid the use of rodent-derived antibodies or antibody constructs, human or fully human antibodies / antibody constructs can be produced by introducing human antibody function into rodents so that the rodents produce fully human antibodies.

[0234] The ability to clone and reconstruct megabase-sized human loci in yeast artificial chromosomes (YACs) and introduce them into mouse germlines provides a powerful approach to elucidating the functional components of very large or roughly mapped loci and generating useful models of human diseases. Furthermore, the use of such techniques to replace mouse loci with their human equivalents can provide unique insights into the expression and regulation of human gene products during development, their communication with other systems, and their involvement in disease induction and progression.

[0235] A key practical application of such strategies is the "humanization" of the mouse humoral immune system. Introducing human immunoglobulin (Ig) loci into mice with inactivated endogenous Ig genes provides an opportunity to study the mechanisms underlying programmed antibody expression and aggregation, as well as their role in B cell development. Furthermore, such strategies can provide an ideal source for the production of fully human monoclonal antibodies (mAbs), representing a significant milestone in achieving the prospects of antibody therapy in human diseases. Fully human antibodies or antibody constructs are expected to minimize the immunogenicity and allergic responses inherent to mouse or mouse-derived mAbs, thus enhancing the efficacy and safety of the administered antibody / antibody construct. The use of fully human antibodies or antibody constructs is expected to offer substantial advantages in the treatment of chronic and recurrent human diseases requiring repeated compound administration, such as inflammation, autoimmunity, and cancer.

[0236] One approach to this goal was to manipulate mouse strains lacking mouse antibody production using large fragments of the human Ig locus, anticipating that such mice would produce a large repertoire of human antibodies in the absence of mouse antibodies. Large human Ig fragments would preserve significant variable gene diversity as well as appropriate regulation of antibody production and expression. By leveraging mouse mechanisms for antibody diversification and selection, and the lack of immune tolerance to human proteins, the reproduced human antibody repertoire in these mouse strains should produce high-affinity antibodies against any target antigen, including human antigens. Using hybridoma technology, antigen-specific human mAbs with desired specificity could be easily produced and selected. This general strategy was demonstrated in connection with the creation of the first XenoMouse mouse strain (see Green et al. Nature Genetics 7:13-21 (1994)). The XenoMouse strain was manipulated with YACs containing germline constituent fragments of 245kb and 190kb sizes from the human heavy chain locus and kappa light chain locus, respectively, including core variable region sequences and constant region sequences. Human Ig containing YACs proved compatible with the mouse strain for both antibody rearrangement and expression, and was able to replace the inactivated mouse Ig gene. This was demonstrated by its ability to induce B cell development, produce an adult-like human repertoire of full human antibodies, and produce antigen-specific human mAbs. These results also suggested that the introduction of larger portions of the human Ig locus containing more V genes, additional regulatory elements, and the human Ig constant region could substantially reproduce the complete repertoire characteristic of the human humoral response to infection and immunization. The Green et al. study has recently been expanded to introduce over 80% of the human antibody repertoire through the introduction of megabase-sized germline constituent YAC fragments from the human heavy chain locus and kappa light chain locus, respectively. See Mendez et al. Nature Genetics 15:146-156 (1997) and U.S. Patent Application No. 08 / 759,620.

[0237] The manufacture of XenoMouse® animals is protected by U.S. Patent Applications No. 07 / 466,008, No. 07 / 610,515, No. 07 / 919,297, No. 07 / 922,649, No. 08 / 031,801, No. 08 / 112,848, No. 08 / 234,145, No. 08 / 376,279, No. 08 / 430,938, No. 08 / 464,584, No. 08 / 464,582, No. 08 / 463,191, No. 08 / 462,837, and the same. This is further discussed and described in Japanese Patent Publication Nos. 08 / 486,853, 08 / 486,857, 08 / 486,859, 08 / 462,513, 08 / 724,752, and 08 / 759,620; and U.S. Patent Nos. 6,162,963, 6,150,584, 6,114,598, 6,075,181, and 5,939,598, as well as in Japanese Patent Publication Nos. 3068180B2, 3068506B2, and 3068507B2. See also Mendez et al. Nature Genetics 15:146-156 (1997) and Green and Jakobovits J. Exp. Med. 188:483-495 (1998), European Patent No. 0463151, International Publication No. 94 / 02602, International Publication No. 96 / 34096, International Publication No. 98 / 24893, International Publication No. 00 / 76310 and International Publication No. 03 / 47336.

[0238] Another approach, employed by other companies including GenPharm International, Inc., utilizes the "minilocus" approach. In the minilocus method, the exogenous Ig locus is mimicked by including fragments (individual genes) from the Ig locus. Thus, one or more VH genes, one or more DH genes, one or more JH genes, a mu constant region, and a second constant region (preferably a gamma constant region) are formed into a construct for insertion into an animal. This approach applies to U.S. Patent No. 5,545,807 to Surani et al., and to U.S. Patent No. 5,545,806, 5,625,825, 5,625,126, 5,633,425, 5,661,016, 5,770,429, 5,789,650, 5,814,318, 5,877,397, 5,874,299, and 6,255,458 to Krimpenfort and Berns, U.S. Patent No. 5,591,669 and 6,023,010 to Berns et al. U.S. Patent Nos. 5,612,205; 5,721,367; and 5,789,215 for al., and U.S. Patent Nos. 5,643,763 for Choi and Dunn, and GenPharm This is described in U.S. Patent Applications No. 07 / 574,748, No. 07 / 575,962, No. 07 / 810,279, No. 07 / 853,408, No. 07 / 904,068, No. 07 / 990,860, No. 08 / 053,131, No. 08 / 096,762, No. 08 / 155,301, No. 08 / 161,739, No. 08 / 165,699, and No. 08 / 209,741 filed by International.See also European Patent No. 0546073B1, International Publication No. 92 / 03918, International Publication No. 92 / 22645, International Publication No. 92 / 22647, International Publication No. 92 / 22670, International Publication No. 93 / 12227, International Publication No. 94 / 00569, International Publication No. 94 / 25585, International Publication No. 96 / 14436, International Publication No. 97 / 13852, and International Publication No. 98 / 24884, as well as U.S. Patent No. 5,981,175. See also Taylor et al. (1992), Chen et al. (1993), Tuaillon et al. (1993), Choi et al. (1993), Lonberg et al. (1994), Taylor et al. (1994), Tuaillon et al. (1995), and Fishwild et al. (1996).

[0239] Kirin has also demonstrated the production of human antibodies from mice into which large chromosome fragments or entire chromosomes have been introduced by microcell fusion. See European Patent Applications No. 773288 and No. 843961. Xenerex Biosciences is developing a technology for the potential production of human antibodies. This technology involves reconstituting SCID mice with human lymphocytes, such as B and / or T cells. The mice can then be immunized with an antigen to induce an immune response to the antigen. See U.S. Patents No. 5,476,996; No. 5,698,767; and No. 5,958,765.

[0240] Human anti-mouse antibody (HAMA) responses have led the industry to develop chimeric antibodies or other humanized antibodies. However, certain human anti-chimeric antibody (HACA) responses are expected to be observed, particularly with chronic or multiple-dose use of antibodies. Therefore, to counteract concerns and / or effects of HAMA or HACA responses, it would be desirable to provide antibody constructs containing human-binding domains to pMAGE-HLA and CD3ε.

[0241] In some embodiments, the antibody constructs of the present invention are “isolated” or “substantially pure” antibody constructs. “Isolated” or “substantially pure,” as used to describe the antibody constructs disclosed herein, means an antibody construct identified, separated, and / or recovered from the components of its production environment. Preferably, the antibody construct is unbound or substantially unbound to all other components from its production environment. Contaminating components of its production environment, such as those arising from recombinant transfected cells, are typically materials that interfere with the diagnostic or therapeutic use of polypeptides and may include enzymes, hormones, and other proteinaceous or non-proteinaceous solutes. It is understood that isolated proteins may constitute a wide range of percentage concentrations, depending on the context, e.g., 5% to 99.9% by weight of the total protein content. Polypeptides may be produced at significantly higher concentrations by using inducible or high-expression promoters, thereby producing polypeptides at increased concentration levels. This definition includes the production of antibody constructs in a wide variety of organisms and / or host cells known in the art. In preferred embodiments, the antibody construct is purified to a degree sufficient to obtain at least 15 residues of the N-terminal or internal amino acid sequence by (1) using a spinning cup sequencer, or (2) homogeneous by SDS-PAGE under non-reducing or reducing conditions using Coomassie blue or preferably silver staining. However, typically the isolated antibody construct is prepared by at least one purification step.

[0242] Peptides are short chains of amino acid monomers linked by covalent peptide (amide) bonds. Therefore, peptides are classified into a broad chemical class of biological oligomers and polymers. Amino acids that are part of a peptide or polypeptide chain are called "residues" and can be numbered sequentially. All peptides except cyclic peptides have an N-terminal residue at one end and a C-terminal residue at the other. Oligopeptides consist of only a few amino acids (usually 2 to 20). Polypeptides are longer, continuous, unbranched peptide chains. Peptides are distinguished from proteins based on size, and can be understood, as an arbitrary criterion, as containing approximately 50 or fewer amino acids. Proteins typically consist of one or more polypeptides arranged in a biologically functional manner. While the aspects of laboratory techniques applied to peptides versus polypeptides and proteins differ (e.g., details of electrophoresis, chromatography, etc.), the size boundary distinguishing peptides from polypeptides and proteins is not absolute. Therefore, in relation to this invention, the terms "peptide," "polypeptide," and "protein" may be used interchangeably, and the term "polypeptide" is often preferred.

[0243] Polypeptides can further form macromers such as dimers, trimers, and higher oligomers, which consist of two or more polypeptide molecules. The polypeptide molecules forming such dimers, trimers, etc., may be identical or different. The corresponding higher-order structures of such macromers are consequently called homodimers or heterodimers, homotrimers or heterotrimers, etc. An example of a heteropolymer is an antibody or immunoglobulin molecule, in its naturally occurring form, consisting of two identical light polypeptide chains and two identical heavy polypeptide chains. The terms “peptide,” “polypeptide,” and “protein” also refer to naturally modified peptides / polypeptides / proteins, where modifications are achieved by post-translational modifications such as glycosylation, acetylation, and phosphorylation. Where used herein, “peptide,” “polypeptide,” or “protein” may be chemically modified, such as pegylation. Such modifications are well known in the art and are described below herein.

[0244] Antigen-binding proteins are said to "specifically bind" or "immunospecifically" to their antigens, and they bind to their antigens with a dissociation constant (KD) of ≤10⁻⁷ M, as measured by surface plasma resonance technology (e.g., BIACore, GE-Healthcare Uppsala, Sweden) or kinetic exclusion assay (KinExA, Sapidyne, Boise, Idaho). According to the present invention, antigen-binding proteins specifically bind to pMAGE-HLA (e.g., GVYDGEEHSV (SEQ ID NO: 1) complexed with HLA-A*02:01) and CD3ε, or bind immunospecifically.

[0245] However, due to sequence similarities between homologous proteins in different species, antibody constructs or binding domains that specifically bind to their targets (such as human targets) may cross-react with homologous target molecules from different species (e.g., non-human primates). Therefore, the term "specific / immunospecific binding" may include the binding of antibody constructs or binding domains to epitopes or structurally related epitopes of two or more species.

[0246] Furthermore, for example, due to sequence similarities between different MAGE peptides, the antibody construct or binding domain may cross-react with various peptides. See Figure 1 for sequence comparisons. For example, in some embodiments, the antibody construct or binding domain may specifically bind to pMAGE-HLA, and the peptides are GVYDGEEHSV, SEQ ID NO: 1, GVYDGREHTV, SEQ ID NO: 2, and / or GLYDGREHSV, SEQ ID NO: 3. In other embodiments, the antibody construct or binding domain may specifically bind only to GVYDGEEHSV, SEQ ID NO: 1, in the context of pMAGE-HLA.

[0247] Linker At least two binding domains and variable domains (VH / VL) of the antibody construct of the present invention may or may not contain a peptide linker (spacer peptide). According to the present invention, the term "peptide linker" includes an amino acid sequence in which the amino acid sequences of one (variable and / or binding) domain and another (variable and / or binding) domain of the antibody construct of the present invention are linked to each other. A third domain can also be fused to other domains of the antibody construct of the present invention using a peptide linker. An essential technical feature of such peptide linkers is that they do not contain polymerization activity. Suitable peptide linkers include those described in U.S. Patent Nos. 4,751,180 and 4,935,233 or International Publication No. 88 / 09344. Other domains, modules, or regions (such as half-life extension domains) can also be conjugated to the antibody construct of the present invention using a peptide linker.

[0248] Examples of useful peptide linkers, in addition to those known in the art, are provided herein. Non-limiting examples of linkers include: G4S Linker GGGGS (Sequence No. 10) (G4S) 2-linker GGGGSGGGGS (Sequence ID 11) (G4S) 3-linker GGGGSGGGGSGGGGS (Sequence No. 12) (G4S) 4-linker GGGGSGGGGSGGGGSGGGGS (Sequence No. 13) (G4S) 5-linker GGGGSGGGGSGGGGSGGGGSGGGGS (Sequence No. 14) (G4S) 6-linker GGGGSGGGGSGGGGSGGGGSGGGGSGGGGS (Sequence No. 15) (G4S) 7 Linker GGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGS (Sequence No. 16) (G4S) 8-linker GGGGSGGGGSGGGGSGGGGSGGGGSGGGGSGGGGGSGGGGGS (Sequence No. 17) Peptide linker PGGGGS (SEQ ID NO: 18) Peptide linker PGGDGS (SEQ ID NO: 19) Peptide linker SGGGGS (SEQ ID NO: 20) Peptide linker GGGG (SEQ ID NO: 21)

[0249] In this context, a “short” linker has 2 to 50 amino acids, preferably 3 to 35, 4 to 30, 5 to 25, 6 to 20, or 6 to 17 amino acids. A linker between two variable regions of a single binding domain may have a different length (e.g., longer) than a linker between two binding domains. For example, a linker between two variable regions of a single binding domain may have a length of 7 to 15 amino acids, preferably 9 to 13, while a linker between two binding domains may have a length of 3 to 10 amino acids, preferably 4 to 8. The peptide linker is further assumed to be a glycine / serine linker, such as those shown in SEQ ID NOs: 10 to 17. In some embodiments, the majority of the amino acids in the glycine / serine linker are selected from glycine and serine.

[0250] When a linker is used to fuse a first domain to a second domain, or a first or second domain to a third domain, the linker is preferably of sufficient length and sequence to ensure that each of the first and second domains can independently maintain their different binding specificities. For peptide linkers linking at least two binding domains (or two variable domains) in the antibody construct of the present invention, peptide linkers containing only a small number of amino acid residues, for example, 12 amino acid residues or less, are preferred. Therefore, peptide linkers with 12, 11, 10, 9, 8, 7, 6, or 5 amino acid residues are preferred. Of the assumed peptide linkers having fewer than 5 amino acids, glycemic linkers containing 4, 3, 2, or 1 amino acid are preferred.

[0251] In the context of "peptide linker," the "single amino acid" linker is Gly. Therefore, the peptide linker may consist of a single amino acid, Gly. Another embodiment of the peptide linker is characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e., Gly4Ser (SEQ ID NO: 10) or a polymer thereof, i.e., (Gly4Ser)x, where x is an integer of 1 or more (e.g., 2 or 3) (SEQ ID NO: 18448). The characteristics of such peptide linkers are known in the art and are described, for example, in Dall'Acqua et al. (Biochem. (1998) 37, 9266-9273), Cheadle et al. (Mol Immunol (1992) 29, 21-30) and Raag and Whitlow (FASEB (1995) 9(1), 73-80). Peptide linkers that do not promote secondary structure are preferred. The linking of these domains can be provided, for example, by genetic engineering. Methods for preparing fused and manipulably linked bispecific single-stranded constructs and expressing them in mammalian cells or bacteria are well known in the art (e.g., International Publication No. 99 / 54440 or Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 2001).

[0252] In one embodiment, the peptide linker, into which polypeptide monomers of the third domain ("Fc moieties" or "Fc monomers") are fused with each other, comprises at least 25 amino acid residues (25, 26, 27, 28, 29, 30, etc.). In another embodiment, the peptide linker comprises at least 30 amino acid residues (30, 31, 32, 33, 34, 35, etc.). In further embodiments, the linker comprises up to 40 amino acid residues, up to 35 amino acid residues, or exactly 30 amino acid residues. Specific embodiments of such peptide linkers are characterized by the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e., Gly4Ser (SEQ ID NO: 10) or its polymer, i.e., (Gly4Ser)x, where x is an integer greater than or equal to 5 (e.g., 6, 7, or 8) (SEQ ID NO: 18449). Preferably, the integer is 6 or 7, and more preferably, the integer is 6.

[0253] Exemplary molecular format As described above, the present invention provides embodiments in which the antibody construct is in a format selected from the group consisting of (scFv)2, scFv single-domain mAb, diabody, and oligomers of any of the above formats. The term “in a format” does not preclude the construct from being further modified by binding or fusion to other parts, for example, as described herein.

[0254] Therefore, the antibody construct of the present invention comprises the following in order from the N-terminus to the C-terminus: (a) The first domain; (b) Peptide linker; (c) Second domain; (d) Peptide linker; (e) The first polypeptide monomer of the third domain (including the hinge, CH2 and CH3 domains); (f) peptide linker; and (g) The second polypeptide monomer of the third domain (including the hinge, CH2 and CH3 domains).

[0255] According to one embodiment of the antibody construct of the present invention, the first and / or second domains are in the format of scFv. In scFv, the VH region and the VL region are arranged in the order VH-VL or VL-VH (from N-terminus to C-terminus). It is assumed that the VH region and VL region of the first and / or second binding domain are linked via a linker, preferably a peptide linker. According to one embodiment of the first and / or second domains, the VH region is located at the N-terminus of the linker and the VL region is located at the C-terminus of the linker. In other words, in one embodiment of the first and / or second domains, scFv includes: VH-linker-VL from N-terminus to C-terminus. It is further assumed that the first and second domains of the antibody construct are linked via a linker, preferably a peptide linker. The antibody construct may include domains in the order of first domain-linker-second domain (from N-terminus to C-terminus), for example. The reverse order (second domain - linker - first domain) is also possible.

[0256] According to one embodiment of the present invention, the antibody construct of the present invention is a “single-chain antibody construct.” It is also conceivable that either or both of the first or second binding domains may be in the form of a “single-chain Fv” (scFv). The two domains of the Fv fragment, VL and VH, are encoded by separate genes, but they can be linked by an artificial linker (as described above herein) that allows the VL and VH regions to pair up to form a monovalent molecule as a single protein chain using recombination. See, for example, Huston et al. (1988) Proc. Natl. Acad. Sci USA 85:5879-5883). These antibody fragments are obtained using conventional techniques known to those skilled in the art, and the fragments are evaluated for function in the same manner as full-length antibodies or IgG. Thus, a single-chain variable fragment (scFv) is a fusion protein of the variable region (VH) of the heavy chain and the variable region (VL) of the light chain of an immunoglobulin, usually linked by a short linker peptide. The linker is typically rich in glycine for flexibility, and also rich in serine or even threonine for solubility, and can connect the N-terminus of VH to the C-terminus of VL, or vice versa. This protein retains the specificity of the original immunoglobulin despite the removal of the constant region and the introduction of the linker.

[0257] In certain embodiments of the antibody construct of the present invention, the first and second domains form an antibody construct in a format selected from the group consisting of (scFv)2, scFv single-domain mAb, diabody, and oligomers in any of these formats.

[0258] In certain embodiments, a bispecificity-binding molecule may not be desired, and instead, a more standard antibody format (e.g., an IgG monoclonal antibody) may be the desired molecule. Accordingly, in one embodiment, the present invention provides antibodies comprising VH and VL. In another embodiment, the present invention provides antibodies comprising VH-CDR1, VH-CDR2, VH-CDR3 and VL-CDR1, VL-CDR2, VL-CDR3.

[0259] In one embodiment of the present invention, the first, second, or both of the first and second domains may include a single-domain antibody, a variable domain, or a CDR of at least one single-domain antibody.

[0260] Antibody constructs called "single-domain antibodies" contain a single (monomer) antibody variable region that can selectively bind to a specific antigen, independently of other variable regions. The first single-domain antibodies were engineered from heavy-chain antibodies found in camelids, and these are called VHH fragments. Cartilaginous fish also have heavy-chain antibodies (IgNARs) from which single-domain antibodies called VNAR fragments can be obtained. Another approach is to split the dimeric variable region from a common immunoglobulin into monomers, thus obtaining VH or VL as a single-domain Ab. Most research on single-domain antibodies is currently based on heavy-chain variable regions, but nanobodies derived from light chains have also been shown to specifically bind to target epitopes. Examples of single-domain antibodies are called sdAbs, nanobodies, or single variable-domain antibodies.

[0261] Therefore, (single-domain mAb)2 is a monoclonal antibody construct comprising (at least) two single-domain monoclonal antibody constructs individually selected from the group including VH, VL, VHH, and VNAR. The linker is preferably in the form of a peptide linker. Similarly, "scFv single-domain mAb" is a monoclonal antibody construct comprising at least one of the single-domain antibodies and one of the scFv molecules. Again, the linker is preferably in the form of a peptide linker.

[0262] The antibody constructs of the present invention are expected to have further functions in addition to their function of binding to the target molecules pMAGE-HLA and CD3. In this format, the antibody construct may be a triplicate or multifunctional antibody construct by providing further functions such as targeting target cells via pMAGE-HLA binding, mediating cytotoxic T cell activity via CD3 binding, enhancing or extending serum half-life, a fully functional or modified Fc constant domain mediating ADCC via effector cell recruitment, a therapeutic agent such as labeling (fluorescence, etc.), a toxin, or a radionuclide.

[0263] According to another embodiment, the antibody construct of the present invention comprises a third domain (in addition to the first and second domains) comprising two polypeptide monomers, each containing a hinge, CH2, and CH3 domain, wherein the two polypeptide monomers are fused to each other via a peptide linker. The third domain is envisioned to contain, in N-terminus to C-terminus,: hinge-CH2-CH3-linker-hinge-CH2-CH3. Amino acid sequences that can be used for the third domain are shown in SEQ ID NOs. 22-29. Each of the polypeptide monomers may be selected from the group consisting of SEQ ID NOs. 22-29, or may have an amino acid sequence that is at least 90% identical to those sequences. In another embodiment, the first and second domains of the antibody construct of the present invention are fused to the third domain via a peptide linker.

[0264] Half-life extension Examples of means or domains for extending the serum half-life of the antibody construct of the present invention include peptides, proteins, or protein domains that are fused to or otherwise attached to the antibody construct. The group of peptides, proteins, or protein domains includes peptides that bind to other proteins having a favorable pharmacokinetic profile in the human body, such as serum albumin (see International Publication No. 2009 / 127691). An alternative concept for such half-life-extending peptides includes peptides that bind to the neonatal Fc receptor (FcRn, see International Publication No. 2007 / 098420), which can also be used in the antibody construct of the present invention. The concept of attaching a larger domain or complete protein to a protein includes the fusion of human serum albumin, human serum albumin (see International Publication Nos. 2011 / 051489, 2012 / 059486, 2012 / 150319, 2013 / 135896, 2014 / 072481, and 2013 / 075066) or variants or mutants of that domain, as well as the fusion of the immunoglobulin constant region (Fc domain) and its variants. Such variants of the Fc domain are called Fc-based domains and may be optimized / modified, for example, to enable desired pairing of dimers or multimers, to disable Fc receptor binding (e.g., to evade ADCC or CDC), or for other reasons. A further concept known in the art for extending the half-life of substances or molecules in the human body is the pegylation of those molecules (such as the antibody constructs of the present invention).

[0265] In one embodiment, an antibody construct according to the present invention is linked (e.g., via a peptide bond) to a fusion partner (e.g., a protein, polypeptide, or peptide) for purposes such as extending the serum half-life of the construct. These fusion partners can be selected from human serum albumin ("HSA" or "HALB") and its sequence variants, peptides that bind to HSA, peptides that bind to FcRn ("FcRn BP"), or constructs containing an (antibody-derived) Fc region. Generally, the fusion partner can be linked directly (e.g., via a peptide bond) or via a peptide linker, such as (GGGGS)n (wherein "n" is an integer of 2 or more, e.g., 2, 3, or 4) (SEQ ID NO: 18450). Suitable peptide linkers are provided herein.

[0266] Modification of amino acid sequence Amino acid sequence modifications of antibody constructs described herein are also intended. For example, it may be desirable to improve the binding affinity and / or other biological properties of the antibody construct. Amino acid sequence variants of antibody constructs are prepared by peptide synthesis or by introducing appropriate nucleotide changes into the nucleic acid molecule encoding the antibody construct. All amino acid sequence modifications described below must result in an antibody construct that retains the desired biological activity of the unmodified parent molecule (e.g., binding to pMAGE-HLA and CD3, induction of cytotoxicity against MAGEB2-positive target cells).

[0267] The terms “amino acid” or “amino acid residue” typically refer to amino acids with a definition recognized in the art, such as those selected from the group consisting of alanine (Ala or A), arginine (Arg or R); asparagine (Asn or N); aspartic acid (Asp or D); cysteine ​​(Cys or C); glutamine (Gln or Q); glutamic acid (Glu or E); glycine (Gly or G); histidine (His or H); isoleucine (Ile or I); leucine (Leu or L); lysine (Lys or K); methionine (Met or M); phenylalanine (Phe or F); proline (Pro or P); serine (Ser or S); threonine (Thr or T); tryptophan (Trp or W); tyrosine (Tyr or Y); and valine (Val or V), but modified, synthetic, or rare amino acids may be used as desired. There are essentially four different classes of amino acids, determined by their different side chains. (1) Nonpolar and neutral (uncharged): Ala, Gly, Ile, Leu, Met, Phe, Pro, Val (2) Polar and neutral (uncharged): Asn, Cys (slightly polar), Gln, Ser, Thr, Trp (slightly polar), Tyr (3) Acidity and polarity (negative charge): Asp and Glu (4) Basicity and polarity (positive charge): Arg, His, Lys

[0268] Hydrophobic amino acids can be divided into those with aliphatic and those with aromatic side chains. Phe and Trp (highly hydrophobic), and Tyr and His (lowly hydrophobic) are classified as aromatic amino acids. Strictly speaking, aliphatic means that the side chain contains only hydrogen and carbon atoms. By this strict definition, amino acids with aliphatic side chains are alanine, isoleucine, leucine (also norleucine), proline, and valine. The fact that alanine's side chain is very short means that it is not particularly hydrophobic, and that proline has an unusual geometric shape that gives it a special role in proteins. Methionine, which also contains a sulfur atom, is often convenient to consider as being in the same category as isoleucine, leucine, and valine. The unifying theme is that these amino acids mainly contain non-reactive and flexible side chains. The amino acids alanine, cysteine, glycine, proline, serine, and threonine are often grouped together because they are all small in size. Gly and Pro can influence chain orientation.

[0269] Amino acid modifications include, for example, the deletion of residues from the amino acid sequence of the antibody construct, the insertion of residues into the antibody construct, and / or the substitution of residues within the amino acid sequence of the antibody construct. Any combination of deletions, insertions, and / or substitutions is performed to arrive at the final antibody construct (e.g., binding to pMAGE-HLA and CD3, induction of cytotoxicity against MAGEB2-positive target cells), as long as the final construct possesses the desired characteristics, such as the biological activity of the unmodified parent molecule. Amino acid changes can also alter the post-translational processes of the antibody construct, such as changes in the number or location of glycosylation sites.

[0270] For example, in each CDR (depending on their respective lengths, of course), one, two, three, four, five, or six amino acids may be inserted, deleted, and / or substituted, while in each FR, one, two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, three, four, five, six, seven, eight, nine, ten, eleven, twelve, three, three, four, five, six, six, seven, eight, nine, ten, ten, ten, ten, three, seven, eight, eight, nine, or ten residues) may be inserted, deleted, and / or substituted. Amino acid sequence insertions include, for example, N-terminal and / or C-terminal additions of amino acids in polypeptides ranging in length from one, two, three, four, five, six, seven, eight, nine, or ten residues to more than ten, for example, more than 100 residues, as well as intra-sequence insertions of single or multiple amino acid residues. Insertion variants of antibody constructs of the present invention include the fusion of a polypeptide to the N-terminus or C-terminus of the antibody construct to increase or extend the serum half-life of the antibody construct. Such insertions may occur within the antibody construct, for example, between a first domain and a second domain.

[0271] The most interesting sites for amino acid modification, particularly amino acid substitution, include the hypervariable regions, especially the individual CDRs of the heavy and / or light chains, but FR changes in the heavy and / or light chains are also intended. Substitutions may be conservative substitutions as described herein. Preferably, depending on the length of the CDR or FR, 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acids may be substituted in the CDR, and 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 25 amino acids may be substituted in the framework region (FR). For example, if the CDR sequence contains 6 amino acids, it is conceivable that 1, 2, or 3 of these amino acids may be substituted. Similarly, if the CDR sequence contains 15 amino acids, it is conceivable that 1, 2, 3, 4, 5, or 6 of these amino acids may be substituted.

[0272] As described herein, a useful method for identifying specific residues or regions within an antibody construct that are favorable sites for mutagenesis is a mature technique called "alanine scanning mutagenesis," which is further described, for example, by Cunningham BC and Wells JA (Science. 1989 Jun 2;244(4908):1081-5) and Morrison KL & Weiss GA. (Cur Opin Chem Biol. 2001 Jun;5(3):302-7). Here, residues or groups of residues within an antibody construct are identified (e.g., charged residues such as Arg, His, Lys, Asp, and Glu) and substituted with neutral or nonpolar amino acids (most preferably alanine or polyalanine) via, for example, peptide synthesis or site-directed mutagenesis to affect the interaction between each amino acid and the epitope of the target protein. Alanine scanning is a technique used to determine the contribution of specific residues to the stability or function of a given protein. Alanine is used to mimic the secondary structure preference of many other amino acids, despite being a non-bulky, chemically inert methyl functional group. When size preservation of the mutant residue is required, bulkier amino acids such as valine or leucine may be used. This technique can also be useful in determining whether the side chain of a particular residue plays a significant role in biological activity. Alanine scanning is typically achieved by site-directed mutagenesis or by randomly creating PCR libraries. Furthermore, computational methods have been developed to estimate thermodynamic parameters based on theoretical alanine substitutions. The data can be tested by IR, NMR spectroscopy, mathematical methods, bioassays, etc.

[0273] Next, the amino acid site exhibiting functional sensitivity to the substitution (for example, determined by alanine scanning) can be purified by introducing further or other variants to or from the substitution site. Therefore, while the site or region to which the amino acid sequence mutation is introduced is predetermined, the nature of the mutation itself does not need to be predetermined. For example, to analyze or optimize the performance of a mutation at a given site, alanine scanning or random mutagenesis can be performed on the target codon or region, and the expressed antibody construct variants can be screened for the optimal combination of desired activity. Well-known methods for introducing substitutional mutations at predetermined sites in DNA with known sequences include, for example, M13 primer mutagenesis and PCR mutagenesis. Screening of variants can be performed, for example, using assays for antigen (e.g., pMAGE-HLA or CD3) binding activity and / or cytotoxic activity.

[0274] Generally, when an amino acid is substituted in one, more, or all of the heavy chain and / or light chain CDRs, the resulting “substituted” sequence is assumed to be at least 60% or 65%, more preferably 70% or 75%, even more preferably 80% or 85%, and particularly preferably 90% or 95% identical / homological to the “original” or “parent” CDR sequence. This means that the degree of identity / homology between the original sequence and the substituted sequence depends on the length of the CDR. For example, a CDR having a total of 5 amino acids and containing one amino acid substitution is 80% identical to the “original” or “parent” CDR sequence, while a CDR having a total of 10 amino acids and containing one amino acid substitution is 90% identical to the “original” or “parent” CDR sequence. Therefore, the substituted CDRs of the antibody constructs of the present invention may have different degrees of identity with respect to their original sequences; for example, CDRL1 may have 80% homology, while CDRL3 may have 90% homology. The same considerations apply to the framework domain, as well as the VH and VL domains as a whole.

[0275] A "variant CDR" is a CDR having specific sequence homology, similarity, or identity with the parent CDR of the present invention, and sharing biological function with the parent CDR by at least 60%, 65%, 70%, 75%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the parent CDR, without limiting its specificity and / or activity. Generally, the amino acid homology, similarity, or identity between individual variant CDRs is at least 60% with respect to the parent sequence shown herein, more typically at least 65% or 70%, preferably at least 75% or 80%, more preferably at least 85%, 90%, 91%, 92%, 93%, 94%, and most preferably 95%, 96%, 97%, 98%, 99%, and nearly 100% increased homology, similarity, or identity. The same applies to "variant VH" and "variant VL". According to one embodiment, sequence mutations within "variant VH" and / or "variant VL" do not extend to the CDR. Accordingly, the present invention relates to an antibody construct as defined herein, comprising VH and VL sequences having specific sequence homology (see above) with respect to a specific sequence ("parent" VH and VL) as defined herein, wherein the CDR sequence is 100% identical to the specific CDR sequence ("parent" CDR) as defined herein.

[0276] Preferred substitutions (or replacements) are conservative substitutions. However, any substitution (including non-conservative substitutions or one or more from the “exemplary substitutions” listed in Table 5 below) is permitted, as long as the antibody construct retains its ability to bind to pMAGE-HLA via the first domain and to CD3 or CD3 epsilon via the second domain, and / or its CDR, FR, VH and / or VL sequences have a degree of identity of at least 60% or 65%, more preferably at least 70% or 75%, even more preferably at least 80% or 85%, and particularly preferably at least 90% or 95% with respect to the original sequence or parent sequence.

[0277] Conservative substitutions (also called conservative mutations or conservative substitutions) are amino acid substitutions that change a given amino acid to a different amino acid with similar biochemical properties (e.g., charge, hydrophobicity, size). Conservative substitutions of proteins often have less impact on protein function than non-conservative substitutions. Conservative substitutions are shown in Table 5. Exemplary conservative substitutions are indicated as “exemplary substitutions”. When such substitutions result in changes in biological activity, more substantial changes can be introduced with respect to amino acid classes, as further described herein, and the product can be screened for desired characteristics.

[0278] [Table 5]

[0279] Substantial modification of the biological properties of the antibody constructs of the present invention is achieved by selecting substitutions that have a significantly different effect on (a) the structure of the polypeptide backbone in the region of substitution, e.g., sheet or helical structure, (b) the charge or hydrophobicity of the molecule at the target site, or (c) the maintenance of the bulk of the side chain. Non-conservative substitutions typically involve exchanging one member of one of the above-defined amino acid classes (e.g., polar, neutral, acidic, basic, aliphatic, aromatic, minor, etc.) with another class. Any cysteine ​​residue that does not contribute to maintaining the proper conformation of the antibody construct can generally be substituted with serine to improve the oxidative stability of the antibody construct.

[0280] In addition to the substitutions described above, other substitutions within the CDR that contribute to the joining can be performed. For example, the consensus sequences described in Table 29 (SEQ ID NOs. 1832-2965), 37 (SEQ ID NOs. 3497-3874), 45 (SEQ ID NOs. 4361-4710), 53 (SEQ ID NOs. 5982-6844), 61 (SEQ ID NOs. 13276-16184), 62 (SEQ ID NOs. 13434-16257), 63 (SEQ ID NOs. 13507-16484), 71 (SEQ ID NOs. 16971-17195), 72 (SEQ ID NOs. 16980-17204), 73 (SEQ ID NOs. 16989-17222), 111, 112, 113, 114, 115, 116, or 117 of this specification provide such substitutions. In certain embodiments, deletions led by the provided consensus sequences can also be performed. According to the guidelines provided herein by the consensus sequences shown in Tables 29 (SEQ ID NOs: 1832-2965), 37 (SEQ ID NOs: 3497-3874), 45 (SEQ ID NOs: 4361-4710), 53 (SEQ ID NOs: 5982-6844), 61 (SEQ ID NOs: 13276-16184), 62 (SEQ ID NOs: 13434-16257), 63 (SEQ ID NOs: 13507-16484), 71 (SEQ ID NOs: 16971-17195), 72 (SEQ ID NOs: 16980-17204), 73 (SEQ ID NOs: 16989-17222), 111, 112, 113, 114, 115, 116, or 117, various amino acid substitutions, and their functional equivalents or deletions can be readily performed by those skilled in the art.

[0281] Sequence identity, homology, and / or similarity of amino acid sequences are, but are not limited to, the local sequence identity algorithm of Smith and Waterman, 1981, Adv. Appl. Math. 2:482, the sequence identity alignment algorithm of Needleman and Wunsch (J Mol Biol. 1970 Mar; 48(3): 443-53), the similarity search method of Pearson and Lipman (Proc Natl Acad Sci USA. 1988 Apr; 85(8): 2444-8), computer implementations of these algorithms (GAP, BESTFIT, FASTA, and TFASTA in the Wisconsin Genetics Software Package, Genetics Computer Group, 575 Science Drive, Madison, Wis.), and the Best Fit sequence program described by Develeux et al. (Nucleic Acids Res. 1984 Jan 11; 12(1 Pt) 1):387-95), it is determined, preferably by using default settings or by using standard techniques known in the art, including by inspection. The identity percentage is assumed to be calculated by FastDB based on the following parameters: mismatch penalty 1; gap penalty 1; gap size penalty 0.33; joining penalty 30. See also “Current Methods in Sequence Comparison and Analysis,” Macromolecule Sequencing and Synthesis, Selected Methods and Applications, pp 127-149 (1988), Alan R. Liss, Inc.

[0282] One example of a useful algorithm is PILEUP. PILEUP uses progressive pairwise alignment to create multiple sequence alignments from a group of related sequences. It can also plot a tree showing the clustering relationships used to create the alignments. PILEUP uses a simplified progressive alignment method from Feng and Doolittle (J Mol Evol. 1987;25(4):351-60), and the method is similar to that described by Higgins and Sharp (Comput Appl Biosci. 1989 Apr;5(2):151-3). Useful PILEUP parameters include a default gap weight of 3.00, a default gap length weight of 0.10, and weighted end gaps.

[0283] Another example of a useful algorithm is the BLAST algorithm described in Altschul et al. (J Mol Biol. 1990 Oct 5;215(3):403-10.); Altschul et al., (Nucleic Acids Res. 1997 Sep 1;25(17):3389-402); and Karlin and Altschul (Proc Natl Acad Sci US A. 1993 Jun 15;90(12):5873-7). A particularly useful BLAST program is the WU-Blast-2 program obtained from Altschul et al., (Methods Enzymol. 1996;266:460-80). WU-Blast-2 uses several search parameters, most of which are set to default values. The adjustable parameters are set to the following values: overlap span = 1, overlap portion = 0.125, word threshold (T) = II. The HSP S and HSP S2 parameters are dynamic values, established by the program itself depending on the composition of a particular sequence and the composition of the specific database in which the target sequence is being searched, but the values ​​may be adjusted to increase sensitivity.

[0284] An additional useful algorithm is gapped BLAST, as reported by Altschul et al. (Nucleic Acids Res. 1997 Sep 1;25(17):3389-402). Gapped BLAST uses BLOSUM-62 substitution scores, with the threshold T parameter set to 9, a two-hit method to induce non-gap expansion, a charge gap length k costing 10+k; Xu is set to 16, and Xg is set to 40 for the database search phase of the algorithm and 67 for the output phase. Gapped alignment is caused by scores corresponding to approximately 22 bits.

[0285] In accordance with this specification, the term “percent (%) nucleic acid sequence identity / homology / similarity” with respect to nucleic acid sequences encoding antibody constructs identified herein is defined as the percentage of nucleotide residues in a candidate sequence that are identical to nucleotide residues in the coding sequence of the antibody construct. One method for aligning two sequences and thereby determining their homology is to use the BLASTN module of WU-Blast2 with default parameters set to 1 and 0.125 for overlap span and overlap percentage, respectively. Generally, the homology, similarity, or identity of nucleic acid sequences between the nucleotide sequences encoding individual variant CDRs and the nucleotide sequences shown herein is at least 60%, and more typically, at least 65%, 70%, 75%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%, and increased homology, similarity, or identity of nearly 100%. Hereinafter, the same applies to the nucleic acid sequences encoding "variant VH" and / or "variant VL".

[0286] In one embodiment, the percentage of identity of the antibody construct according to the present invention, or the first and second domains (binding domains) of these antibody constructs, to the human germline is 70% or more, or 75% or more, more preferably 80% or more, or 85% or more, even more preferably 90% or more, most preferably 91% or more, 92% or more, 93% or more, 94% or more, 95% or more, or 96% or more. Identity to human antibody germline gene products is considered an important feature for reducing the risk that therapeutic proteins will induce an immune response to the drug in patients during treatment. Hwang WY and Foote J. (Methods. 2005 May;36(1):3-10) have demonstrated that reducing the non-human portion of drug antibody constructs leads to a reduction in the risk of inducing anti-drug antibodies in patients during treatment. By comparing a comprehensive number of clinically evaluated antibody drugs with their respective immunogenicity data, humanization of the variable region of an antibody / antibody construct tends to result in lower protein immunogenicity (average 5.1% of patients) compared to antibodies / antibody constructs with an unaltered non-human variable region (average 23.59% of patients). Therefore, a higher degree of identity to the human sequence is desirable for protein therapeutics in the form of antibody constructs based on the variable region. For the purpose of determining germline identity, the V region of the VL can be aligned with the amino acid sequences of the human germline V and J segments (http: / / www2.mrc-lmb.cam.ac.uk / vbase / ) using Vector NTI software, as well as the amino acid sequence calculated by dividing identical amino acid residues by the total number of amino acid residues in the VL as a percentage. The same can be done for the VH segment (http: / / www2.mrc-lmb.cam.ac.uk / vbase / ), except that VH CDR3 may be excluded due to its high diversity and the lack of existing human germline VH CDR3 alignment partners. Next, recombinant technology can be used to enhance sequence identity for human antibody germline genes.

[0287] Nucleotides encoding antibody constructs The present invention further provides polynucleotide / nucleic acid molecules encoding the antibody constructs of the present invention. Nucleic acid molecules are biopolymers composed of nucleotides. Polynucleotides are biopolymers composed of 13 or more nucleotide monomers covalently linked in a chain. DNA (cDNA, etc.) and RNA (mRNA, etc.) are examples of polynucleotide / nucleic acid molecules having different biological functions. Nucleotides are organic molecules that function as monomers or subunits of nucleic acid molecules such as DNA or RNA. The nucleic acid molecules or polynucleotides of the present invention may be double-stranded or single-stranded, linear or cyclic. It is envisioned that the nucleic acid molecules or polynucleotides will be contained in a vector. It is further envisioned that such vectors will be contained in host cells. The host cells can express antibody constructs after transformation or transfection using, for example, the vector or polynucleotide / nucleic acid molecule of the present invention. For this purpose, the polynucleotides or nucleic acid molecules are operably linked to a control sequence.

[0288] The genetic code is a set of rules for translating information encoded within genetic material (nucleic acids) into proteins. Biological decoding in living cells is carried out by ribosomes, which use tRNA molecules—which carry amino acids and read three nucleotides from mRNA at once—to link the amino acids in the order specified by the mRNA. This code defines how a sequence of three nucleotides, called a codon, specifies the next amino acid to be added during protein synthesis. With a few exceptions, a three-nucleotide codon in a nucleic acid sequence specifies one amino acid. Because most genes are encoded with the exact same code, this particular code is often referred to as the reference genetic code or standard genetic code.

[0289] Codon degeneracy is the redundancy of the genetic code, expressed as a large number of 3-base pair codon combinations that specify amino acids. Degeneracy arises because there are more codons than amino acids that can be coded. Codons that code for a single amino acid may differ in any of their three positions, but often this difference lies in the second or third position. For example, codons GAA and GAG both code for glutamic acid and exhibit redundancy, but neither specifies any other amino acid and therefore does not exhibit ambiguity. The genetic codes of different organisms may be biased towards using one of several codons that code for the same amino acid more than others, i.e., the frequency of a single codon is higher than would be expected by chance. For example, leucine is specified by six different codons, some of which are rarely used. Codon frequency tables detailing genomic codon usage frequencies for most organisms are available. Recombinant gene technology generally utilizes this effect by employing a technique called codon optimization, in which those codons are used to design polynucleotides preferred by each host cell (e.g., human hamster-derived cells, Escherichia coli cells, or Saccharomyces cerevisiae cells) to increase protein expression, for example. Therefore, it is assumed that the polynucleotide / nucleic acid molecules of this disclosure are codon-optimized. Nevertheless, the polynucleotide / nucleic acid molecules encoding the antibody constructs of the present invention may be designed using any codon encoding a desired amino acid.

[0290] According to one embodiment, the polynucleotide / nucleic acid molecule of the present invention encoding the antibody construct of the present invention may be in the form of a single molecule or two or more distinct molecules. If the antibody construct of the present invention is a single-chain antibody construct, the polynucleotide / nucleic acid molecule encoding such a construct is most likely to also be in the form of a single molecule. However, it is also conceivable that different components of the antibody construct (e.g., different domains, e.g., a domain that binds to pMAGE-HLA, a domain that binds to CD3, and / or further domains such as the antibody constant domain) are located on separate polypeptide chains, in which case the polynucleotide / nucleic acid molecule is most likely to be in the form of two or more distinct molecules.

[0291] The same applies to vectors containing the polynucleotide / nucleic acid molecule of the present invention. If the antibody construct of the present invention is a single-chain antibody construct, one vector may contain the polynucleotide encoding the antibody construct in one location (as a single open reading frame, ORF). One vector may also contain two or more polynucleotide / nucleic acid molecules at separate locations (each having its own ORF), each of which encodes a different component of the antibody construct of the present invention. A vector containing the polynucleotide / nucleic acid molecule of the present invention is envisioned to be in the form of a single vector or two or more separate vectors. In one embodiment, for the purpose of expressing the antibody construct in a host cell, the host cell of the present invention should contain a polynucleotide / nucleic acid molecule encoding the antibody construct, or a vector containing such polynucleotide / nucleic acid molecule as a whole, meaning that all components of the antibody construct (whether encoded as a single molecule or at separate molecules / locations) assemble post-translation to form the biologically active antibody construct of the present invention together.

[0292] The present invention also provides vectors comprising the polynucleotide / nucleic acid molecule of the present invention. A vector is a nucleic acid molecule typically used as a vehicle for transferring (foreign) genetic material into a cell for the purpose of replication and / or expression. The term “vector” includes, but is not limited to, plasmids, viruses, cosmids, and artificial chromosomes. Some vectors are specifically designed for cloning (cloning vectors), while others are designed for protein expression (expression vectors). So-called transcription vectors are primarily used to amplify their inserts. DNA manipulation is typically performed on E. coli vectors that contain the elements necessary for their maintenance in E. coli. However, vectors may also have elements that allow them to be maintained in other organisms such as yeast, plant, or mammalian cells; these vectors are called shuttle vectors. Insertion of a vector into a target cell or host cell is typically called bacterial cell transformation and eukaryotic cell transfection, and insertion of a viral vector is often called transduction.

[0293] Generally, genetically engineered vectors contain an origin of replication, multiple cloning sites, and selection markers. The vector itself is generally a nucleotide sequence, typically a DNA sequence, containing an insert (transgene) and a larger sequence that acts as the vector's "skeleton." The genetic code determines the polypeptide sequence of a given coding region, but other genomic regions can influence when and where these polypeptides are produced. Therefore, modern vectors may include further features beyond the transgene insert and backbone: promoters, genetic markers, antibiotic resistance, reporter genes, targeting sequences, and protein purification tags. Vectors called expression vectors (expression constructs) are specifically for the expression of transgenes in target cells and generally contain regulatory sequences.

[0294] The term "regulatory sequence" refers to a DNA sequence necessary for the expression of an operablely linked coding sequence in a particular host organism. Suitable regulatory sequences for prokaryotes include, for example, promoters, optionally operator sequences, and ribosome-binding sites. Eukaryotic cells are known to utilize promoters, polyadenylation signals, Kozak sequences, and enhancers.

[0295] Nucleic acids are "operably linked" if they have a functional relationship with another nucleic acid sequence. For example, DNA for a pre-sequence or secretion leader is operably linked to DNA for a polypeptide if it is expressed as a protein precursor involved in polypeptide secretion; a promoter or enhancer is operably linked to a coding sequence if it affects the transcription of the sequence; or a ribosome binding site is operably linked to a coding sequence if it is positioned to facilitate translation. Generally, "operably linked" means that the linked nucleotide sequences are contiguous and, in the case of a secretion leader, contiguous and within the reading frame. Enhancers, however, do not need to be contiguous. Linking is done by ligation at a convenient restriction site. If such a site does not exist, synthetic oligonucleotide adapters or linkers are used, according to conventional practice.

[0296] "Transfection" is the process of intentionally introducing nucleic acid molecules or polynucleotides (including vectors) into target cells. This term is primarily used for non-viral methods in eukaryotic cells. Transfection is often used to describe viral-mediated transfer of nucleic acid molecules or polynucleotides. Transfection of animal cells typically involves creating transient pores or "holes" in the cell membrane to allow material uptake. Transfection can be carried out using biological particles (e.g., viral transfection, also called viral transfection), chemical-based methods (e.g., using calcium phosphate, lipofection, Fugene, cationic polymers, nanoparticles), or physical treatments (e.g., electroporation, microinjection, gene gun, cell squeezing, magnetofection, hydrostatic pressure, impulfection, sonication, phototransfection, heat shock).

[0297] The term "transformation" is used to describe the nonviral transfer of nucleic acid molecules or polynucleotides (including vectors) into bacteria and non-animal eukaryotic cells, including plant cells. Therefore, transformation is a genetic modification of a bacterial or non-animal eukaryotic cell resulting from direct uptake from its periphery across the cell membrane and subsequent integration of exogenous genetic material (nucleic acid molecules). Transformation can be induced by artificial means. For transformation to occur, the cell or bacterium must be in a competent state where transformation can occur as a timed response to environmental conditions such as starvation and cell density, and can also be introduced artificially.

[0298] Furthermore, the present invention provides host cells transformed or transfected with the polynucleotide / nucleic acid molecule or vector of the present invention.

[0299] As used herein, the terms “host cell” or “recipient cell” are intended to include any individual cell or cell culture that may or may have been the recipient of the vector, exogenous nucleic acid molecule and / or polynucleotide encoding the antibody construct of the present invention; and / or the antibody construct itself. The introduction of each substance into a cell is carried out by transformation, transfection, etc. (see above). The term “host cell” is also intended to include the offspring or potential offspring of a single cell. In subsequent generations, certain modifications may occur due to spontaneous, accidental, or intentional mutations, or due to environmental influences, and such offspring may not actually be completely identical to the parent cell (morphologically or with respect to the genome or total DNA set), but are still included within the scope of the terms as used herein. Suitable host cells include, but are not limited to, prokaryotic or eukaryotic cells, as well as bacteria (such as Escherichia coli), yeast cells, fungal cells, plant cells, and animal cells, such as insect cells and mammalian cells, such as hamster, mouse, rat, macaque, or human cells.

[0300] In addition to prokaryotes, eukaryotic microorganisms such as filamentous fungi or yeasts are suitable cloning or expression hosts for the antibody constructs of the present invention. Saccharomyces cerevisiae or common baker's yeast are the most commonly used among lower eukaryotic host microorganisms. However, many other genera, species, and strains are generally available and useful in the present invention, for example, Schizosaccharomyces pombe, K. lactis, K. fragilis (ATCC 12424), K. bulgaricus (ATCC 16045), K. wickeramii (ATCC 24178), K. waltii (ATCC 56500), K. drosophilarum (ATCC 16045) Hosts of the genus Kluyveromyces, such as K. thermotolerans and K. marxianus; yarrowia (European Patent No. 402226); Pichia pastoris (European Patent No. 183070); Candida; Trichoderma reesia (European Patent No. 244234); Neurospora crassa; Schwanniomyces occidentalis Hosts include the genus Schwanniomyces (such as Schwanniomyces occidentalis), as well as filamentous fungi such as Neurospora, Penicillium, Tolypocladium, and Aspergillus, such as A. nidulans and A. niger.

[0301] Suitable host cells for the expression of glycosylated antibody constructs are derived from multicellular organisms. Examples of invertebrate cells include plant and insect cells. Numerous baculovirus strains and variants, as well as corresponding acceptable insect host cells derived from hosts such as the armyworm (Spodoptera frugiperda), Aedes aegypti, Aedes albopictus, Drosophila melanogaster, and silkworm (Bombyx mori), have been identified. Various virus strains for transfection, such as the L-1 variant of Autographa californica NPV and the Bm-5 strain of Bombyx mori NPV, are publicly available, and such viruses can be used as the viruses of this specification according to the present invention, particularly for transfection of armyworm (Spodoptera frugiperda) cells.

[0302] Plant cell cultures of cotton, corn, potato, soybean, petunia, tomato, Arabidopsis thaliana, and tobacco can also be used as hosts. Cloning and expression vectors useful for protein production in plant cell cultures are known to those skilled in the art. See, for example, Hiatt et al., Nature (1989) 342:76-78, Owen et al. (1992) Bio / Technology 10:790-794, Artsaenko et al. (1995) The Plant J 8:745-750, and Fecker et al. (1996) Plant Mol Biol 32:979-986.

[0303] However, there is the greatest interest in vertebrate cells, and the proliferation of vertebrate cells under culture conditions has become a standard procedure. Examples of useful mammalian host cell lines include SV40-transformed monkey kidney CV1 cell line (COS-7, ATCC CRL 1651, etc.); human embryonic kidney cell line (293 cells or 293 cells subcloned for growth in suspension culture, Graham et al., J. Gen Virol. 36:59 (1977), etc.); baby hamster kidney cells (BHK, ATCC CCL 10, etc.); Chinese hamster ovary cells / -DHFR (CHO, Urlaub et al., Proc. Natl. Acad. Sci. USA 77:4216 (1980), etc.); mouse Sertoli cells (TM4, Mather, Biol. Reprod. 23:243-251 (1980), etc.); monkey kidney cells (CVI ATCC CCL 70, etc.); African green monkey kidney cells (VERO-76, ATCC CRL These include 1587, etc.; human cervical cancer cells (HELA, ATCC CCL 2, etc.); canine kidney cells (MDCK, ATCC CCL 34, etc.); buffalo rat liver cells (BRL 3A, ATCC CRL 1442, etc.); human lung cells (W138, ATCC CCL 75, etc.); human liver cells (Hep G2, 1413 8065, etc.); mouse mammary tumor cells (MMT 060562, ATCC CCL-51, etc.); TRI cells (Mather et al., Annals NY Acad.Sci. (1982) 383:44-68); MRC 5 cells; FS4 cells; and human hepatocellular carcinoma strains (Hep G2, etc.).

[0304] Manufacturing of antibody constructs In further embodiments, the present invention provides a method for producing an antibody construct of the present invention, comprising culturing host cells of the present invention under conditions that enable the expression of the antibody construct of the present invention, and recovering the produced antibody construct from the culture.

[0305] As used herein, the term “culture” refers to the maintenance, differentiation, growth, proliferation, and / or propagation of cells in vitro under suitable conditions in a culture medium. Cells are grown and maintained in cell growth medium at an appropriate temperature and gas mixture. Culture conditions vary considerably depending on the cell type. Typical growth conditions are a temperature of approximately 37°C, a CO2 concentration of approximately 5%, and a humidity of approximately 95%. The recipe for the growth medium can vary, for example, in terms of pH, the concentration of the carbon source (glucose, etc.), the nature and concentration of growth factors, and the presence of other nutrients (amino acids or vitamins, etc.). Growth factors used to supplement the medium are often derived from animal blood serum, such as fetal bovine serum (FBS), calf serum (FCS), horse serum, and porcine serum. Cells can be grown in suspension or as adherent cultures. Some cell lines have been modified to survive in suspension cultures, and thus can be grown to higher densities than those possible under adherent conditions.

[0306] The term “expression” includes, but is not limited to, any steps involved in the production of the antibody construct of the present invention, including transcription, post-transcriptional modification, translation, folding, post-translational modification, targeting to specific intracellular or extracellular locations, and secretion. The term “recovery” refers to a series of processes intended to isolate the antibody construct from a cell culture. The “recovery” or “purification” process can separate the protein and non-protein portions of a cell culture, ultimately separating the desired antibody construct from all other polypeptides and proteins. The separation process typically utilizes differences in protein size, physicochemical properties, binding affinity, and biological activity. Preparative purification aims to produce relatively large quantities of purified protein for subsequent use, while analytical purification produces relatively small quantities of protein for various research or analytical purposes.

[0307] When recombinant technology is used, the antibody construct can be produced intracellularly in the periplasmic space of the cell membrane or secreted directly into the culture medium. If the antibody construct is produced intracellularly, the first step is to remove host cells or particulate fragments of the lysed fragment, for example, by centrifugation or ultrafiltration. The antibody constructs of the present invention can be produced in bacteria such as Escherichia coli (E. coli). After expression, the construct can be isolated from the bacterial cell paste in a soluble fraction and purified, for example, by affinity chromatography and / or size exclusion. Final purification can be carried out in a manner similar to the process for purifying antibody constructs expressed in mammalian cells and secreted into the culture medium. Carter et al. (Biotechnology (NY) 1992 Feb;10(2):163-7) describe a procedure for isolating antibodies secreted into the periplasmic space of Escherichia coli (E. coli).

[0308] When antibodies are secreted into a culture medium, the supernatant from such an expression system is generally first concentrated using a commercially available protein concentration filter, such as an ultrafiltration unit.

[0309] The antibody constructs of the present invention, prepared from host cells, can be recovered or purified using, for example, hydroxyl apatite chromatography, gel electrophoresis, dialysis, and affinity chromatography. Other techniques for protein purification, such as fractionation with ion exchange columns, mixed-mode ion exchange, HIC, ethanol precipitation, size exclusion chromatography, reverse-phase HPLC, chromatography with silica, chromatography with heparin Sepharose, chromatography with anion or cation exchange resins (e.g., polyaspartate columns), immunoaffinity (e.g., protein A / G / L) chromatography, chromatographic focusing, SDS-PAGE, ultracentrifugation, and ammonium sulfate precipitation, are also available depending on the antibody construct to be recovered.

[0310] A protease inhibitor may be included in any of the aforementioned steps to inhibit protein degradation, and an antibiotic may be included to prevent the growth of contaminants.

[0311] Molecular properties The molecule may exhibit various different properties desired in therapeutic pharmaceuticals, such as stability under various storage conditions, in vivo stability, purity, and other properties that can be assayed.

[0312] Therefore, in further embodiments, the bispecific antibody constructs of the present invention exhibit high monomer yields under standard research-scale conditions, for example, in a standard two-step purification process. Preferably, the monomer yield of the antibody construct according to the present invention is ≥0.25 mg / L of supernatant, more preferably ≥0.5 mg / L, even more preferably ≥1 mg / L, and most preferably ≥3 mg / L of supernatant.

[0313] Similarly, the yield of dimeric antibody construct isoforms, and therefore the monomer percentage of the antibody construct (i.e., monomer:(monomer + dimer)), can be determined. The productivity of monomer and dimeric antibody constructs and the calculated monomer percentage can be obtained, for example, in an SEC purification step of the culture supernatant from standardized study-scale production in roller bottles. In one embodiment, the monomer percentage of the antibody construct is 80% or more, more preferably 85% or more, even more preferably 90% or more, and most preferably 95% or more.

[0314] In one embodiment, the antibody construct has a preferred plasma stability (ratio of EC50 with plasma to EC50 without plasma) of 5 or less, or 4 or less, more preferably 3.5 or less, or 3 or less, even more preferably 2.5 or less, or 2 or less, and most preferably 1.5 or less, or 1 or less. The plasma stability of the antibody construct can be tested by incubating the construct in human plasma at 37°C for 24 hours, followed by determining the EC50 in a 51-chromium-releasing cytotoxicity assay. Effector cells in the cytotoxicity assay can be stimulated with enriched human CD8-positive T cells. Target cells may be, for example, cells transfected with human MAGEB2. The effector-to-target cell (E:T) ratio can be selected as 10:1 or 5:1. The human plasma pool used for this purpose is derived from the blood of a healthy donor collected by an EDTA-coated syringe. Cellular components are removed by centrifugation, the upper plasma phase is collected, and then pooled. As a control, the antibody construct is diluted in RPMI-1640 medium immediately before the cytotoxicity assay. Plasma stability is calculated as the ratio of EC50 (after plasma incubation) to EC50 (control).

[0315] It is even more preferable that the monomer-to-dimer conversion of the antibody construct of the present invention is low. The conversion can be measured under different conditions and analyzed by high-speed size exclusion chromatography. For example, incubation of the monomer isoform of the antibody construct can be carried out in an incubator at 37°C and at a concentration of, for example, 100 μg / ml or 250 μg / ml for 7 days. Under these conditions, it is preferable that the antibody construct of the present invention exhibits a dimer percentage of 5% or less, more preferably 4% or less, even more preferably 3% or less, even more preferably 2.5% or less, even more preferably 2% or less, even more preferably 1.5% or less, most preferably 1% or less, or 0.5% or less, or even 0%.

[0316] Furthermore, it is preferable that the bispecific antibody constructs of the present invention exist with a very low dimerization rate after numerous freeze / thaw cycles. For example, the antibody construct monomer is adjusted to a concentration of, for example, 250 μg / ml in a general formulation buffer, subjected to three freeze / thaw cycles (freezing at -80°C for 30 minutes, followed by thawing at room temperature for 30 minutes), and then high-performance SEC is performed to determine the percentage of the initial monomer antibody construct converted to a dimerized antibody construct. Preferably, the dimerization percentage of the bispecific antibody construct is 5% or less, more preferably 4% or less, even more preferably 3% or less, even more preferably 2.5% or less, even more preferably 2% or less, even more preferably 1.5% or less, most preferably 1% or less, or even more preferably 0.5% or less, after, for example, three freeze / thaw cycles.

[0317] The bispecific antibody constructs of the present invention exhibit preferred thermal stability at aggregation temperatures of preferably 45°C or 50°C or higher, more preferably 52°C or 54°C or higher, even more preferably 56°C or 57°C or higher, and most preferably 58°C or 59°C or higher. The thermal stability parameter can be determined with respect to the antibody aggregation temperature as follows: An antibody solution with a concentration of 250 μg / ml is transferred to a disposable cuvette and placed in a dynamic light scattering (DLS) apparatus. The sample is heated from 40°C to 70°C at a heating rate of 0.5°C / min while maintaining a constant measured range. The increase in the range showing protein melting and aggregation is used to calculate the antibody aggregation temperature.

[0318] Alternatively, the intrinsic biophysical protein stability of the antibody construct can be determined by determining the temperature-melting curve using differential scanning calorimetry (DSC). These experiments are performed using a MicroCal LLC (Northampton, MA, USA) VP-DSC device. Energy uptake of the sample containing the antibody construct is recorded at 20°C to 90°C compared to a sample containing only the formulation buffer. The antibody construct is adjusted to a final concentration of 250 μg / ml in, for example, SEC electrophoresis buffer. The temperature of the entire sample is gradually increased to record each melting curve. Energy uptake is recorded at each temperature T, based on the sample and the formulation buffer. The difference in energy uptake Cp (kcal / mol / °C), obtained by subtracting the baseline from the sample, is plotted against each temperature. The melting temperature is defined as the temperature at the first maximum value of energy uptake.

[0319] The pMAGE-HLAxCD3 bispecific antibody construct of the present invention is also assumed to have a turbidity of 0.2 or less, preferably 0.15 or less, more preferably 0.12 or less, even more preferably 0.1 or less, and most preferably 0.08 or less (measured by OD340 after concentrating the purified monomer antibody construct to 2.5 mg / ml and incubating overnight).

[0320] In further embodiments, the antibody constructs according to the present invention are stable at physiological or slightly lower pH, i.e., about pH 7.4 to 6.0. The more tolerant the antibody construct behaves at non-physiological pH, such as about pH 6.0, the higher the recovery rate of the antibody construct eluted from the ion-exchange column relative to the total amount of loaded protein. The recovery rate of the antibody construct from the ion (e.g., cation) exchange column at about pH 6.0 is preferably 30% or more, more preferably 40% or more, more preferably 50% or more, even more preferably 60% or more, even more preferably 70% or more, even more preferably 80% or more, even more preferably 90% or more, even more preferably 95% or more, and most preferably 99% or more.

[0321] In further embodiments, the antibody construct according to the present invention is stable at acidic pH. The more tolerant the antibody construct behaves at non-physiological pH values ​​such as pH 5.5 (e.g., the pH required for cation exchange chromatography), the higher the recovery rate of the antibody construct eluted from the ion exchange column relative to the total amount of loaded protein. The recovery rate of the antibody construct from the ion (e.g., cation) exchange column at pH 5.5 is preferably 30% or more, more preferably 40% or more, more preferably 50% or more, even more preferably 60% or more, even more preferably 70% or more, even more preferably 80% or more, and most preferably 95% or more. The percentages represent the area under the curve (=AUC) of the major peak.

[0322] The change in potency of a targeted xCD3 antibody construct as a function of pre-incubation of the construct on target cells in the absence of T cells can be measured. If the antibody construct is internalized, it is expected to undergo lysosomal degradation. Therefore, the effective concentration is expected to decrease over time, and thus the apparent potency should also decrease. This effect has been observed with several targets and is a known phenomenon. The antibody construct of the present invention is assumed to not be internalized or not undergo significant internalization by target cells. The rate of internalization can be assayed, for example, as described below: Count T cells and dilute to a concentration of 1 x 10⁵ / ml in assay medium. Count target-positive target cells and seed, for example, 2500 cells per well (cpw). Serially dilute the antibody construct 1:2 at a starting concentration, for example, 100 nM. Add the antibody construct to a culture assay plate to allow incubation for 0, 1, or 2 hours before adding T cells. Next, T cells are seeded at 25,000 cpw (E:T = 10:1), and the assay is incubated at 37°C for 48 hours. Target cell viability is analyzed, for example, using a Steady-Glo® system (25 μl / well). Preferably, the internalization rate (e.g., measured as a decrease in cytotoxicity) is 20% or less, more preferably 15% or less, even more preferably 10% or less, and most preferably 5% or less after 2 hours of (pre)incubation of the antibody construct with target cells.

[0323] Further considerations of the present invention include antibody constructs in which the loss or soluble target does not significantly impair their efficacy or biological activity. This can be measured, for example, by a cytotoxic assay in which the soluble target is added to the assay at progressively increasing concentrations, e.g., 0 nM to 0.3 nM to 0.7 nM to 1 nM to 3 nM to 7 nM to 12 nM. An exemplary E:T value is 10:1. The EC50 value of the tested antibody construct should not significantly increase in the presence of the soluble target.

[0324] The antibody constructs of the present invention are further expected to exhibit therapeutic efficacy manifested as antitumor activity or inhibition of tumor growth. This can be evaluated in studies such as those disclosed in Example 10. In one embodiment, the tumor growth inhibition T / C [%] of the antibody constructs of the present invention is 70 or less, 60 or less, 50 or less, 40 or less, 30 or less, 20 or less, 10 or less, 5 or less, 4 or less, 3 or less, or 2 or less. Modifications or adjustments to specific parameters of these studies (e.g., number of injected tumor cells, injection site, number of transplanted human T cells, amount of antibody construct administered, and timeline) are also expected, but significant and reproducible results can still be obtained.

[0325] Covalent modification of structures Covalent modifications of antibody constructs are also within the scope of the present invention and are generally performed post-translation, though not always. For example, several types of covalent modifications of antibody constructs are introduced into the molecule by reacting specific amino acid residues of the antibody construct with an organic derivatizing agent that can react with selected side chains or N-terminal or C-terminal residues.

[0326] Cysteinyl residues most commonly react with α-haloacetates (and corresponding amines), such as chloroacetic acid or chloroacetamide, to obtain carboxymethyl or carboxyamidemethyl derivatives. Cysteinyl residues are also derivatized by reaction with bromotrifluoroacetone, α-bromo-β-(5-imidazoyl)propionic acid, chloroacetyl phosphate, N-alkylmaleimide, 3-nitro-2-pyridyl disulfide, methyl 2-pyridyl disulfide, p-chloromercrine benzoate, 2-chloromercrine 4-nitrophenol, or chloro-7-nitrobenzo-2-oxa-1,3-diazole.

[0327] Histidyl residues are derivatized by reaction with diethyl pyrocarbonate at pH 5.5-7.0, as this agent is relatively specific to the histidyl side chain. Para-bromophenacyl bromide is also useful, and the reaction is preferably carried out in 0.1 M sodium cacodylate at pH 6.0. Lydinyl and amino-terminal residues are reacted with succinic acid or other carboxylic acid anhydrides. Derivatization with these agents has the effect of reversing the charge of the lydinyl residue. Other suitable reagents for derivatization of α-amino-containing residues include transaminase-catalyzed reactions with imide esters, such as methyl picoline imidate, pyridoxal phosphate, pyridoxal, chloroborohydride, trinitrobenzenesulfonic acid, O-methylisourea, 2,4-pentanedione, and glyoxylate.

[0328] Arginyl residues are modified by reaction with one or more conventional reagents, particularly phenylglyoxal, 2,3-butanedione, 1,2-cyclohexanedione, and ninhydrin. Derivatization of arginine residues requires alkaline conditions due to the high pKa of the guanidine functional group. Furthermore, these reagents can react with lysine groups and arginine epsilon-amino groups.

[0329] Specific modifications of tyrosyl residues can be carried out with particular interest in introducing spectral labeling to tyrosyl residues by reaction with aromatic diazonium compounds or tetranitromethane. Most commonly, N-acetylimidazole and tetranitromethane are used to form O-acetyltyrosyl species and 3-nitro derivatives, respectively. For preparing labeled proteins for use in radioimmunoassays, iodine treatment of tyrosyl residues using 125I or 131I is suitable, followed by the chloramine T method described above.

[0330] The carboxyl side group (aspartyl or glutamyl) is selectively modified by reaction with a carbodiimide (R'-N=C=N--R'), where R and R' are optionally different alkyl groups, for example, 1-cyclohexyl-3-(2-morpholinyl-4-ethyl)carbodiimide or 1-ethyl-3-(4-azonia-4,4-dimethylpentyl)carbodiimide. Furthermore, the aspartyl and glutamyl residues are converted to asparaginyl and glutamyl residues by reaction with ammonium ions.

[0331] Derivatization with difunctional agents is useful for crosslinking water-insoluble support matrices or surfaces for use in various ways with the antibody constructs of the present invention. Commonly used crosslinking agents include, for example, 1,1-bis(diazoacetyl)-2-phenylethane, glutaraldehyde, N-hydroxysuccinimide esters, esters with 4-azidosalicylic acid, homodifunctional imide esters including disuccinimidyl esters such as 3,3'-dithiobis(succinimidylpropionate), and difunctional maleimides such as bis-N-maleimide-1,8-octane. Derivatizing agents such as methyl-3-[(p-azidophenyl)dithio]propioimidate produce photoactivatable intermediates that can form crosslinks in the presence of light. Alternatively, reactive water-insoluble matrices such as cyanide-activated carbohydrates and reactive substrates as described in U.S. Patent Nos. 3,969,287; 3,691,016; 4,195,128; 4,247,642; 4,229,537; and 4,330,440 are used for protein immobilization.

[0332] Glutaminyl and asparaginyl residues are often deamidated to their corresponding glutamyl and aspartyl residues, respectively. Alternatively, these residues are deamidated under weakly acidic conditions. Any form of these residues is within the scope of the present invention.

[0333] Other modifications include hydroxylation of proline and lysine, phosphorylation of hydroxyl groups of ceryl or threonyl residues, methylation of α-amino groups of lysine, arginine, and histidine side chains (TECreighton, Proteins: Structure and Molecular Properties, WH Freeman & Co., San Francisco, 1983, pp. 79-86), acetylation of N-terminal amines, and amidation of any C-terminal carboxyl group.

[0334] Another type of covalent modification of antibody constructs that falls within the scope of the present invention involves altering the glycosylation pattern of a protein. As is known in the art, the glycosylation pattern may depend on the protein sequence (e.g., the presence or absence of certain glycosylated amino acid residues, as discussed below) or on both the host cell or organism in which the protein is produced. Specific expression systems are discussed herein.

[0335] Polypeptide glycosylation is typically either N-linked or O-linked. N-linking refers to the attachment of the carbohydrate moiety to the side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine and asparagine-X-threonine (wherein X is any amino acid except proline) are recognition sequences for the enzymatic attachment of the carbohydrate moiety to the asparagine side chain. Therefore, the presence of either of these tripeptide sequences in a polypeptide creates a potential glycosylation site. O-linked glycosylation refers to the attachment of one of the sugars N-acetylgalactosamine, galactose, or xylose to a hydroxyamino acid, most commonly serine or threonine, although 5-hydroxyproline or 5-hydroxylysine may also be used.

[0336] The addition of glycosylation sites to antibody constructs is conveniently achieved by modifying the amino acid sequence (for N-linked glycosylation sites) to include one or more of the above-described tripeptide sequences. The modification can also be carried out by the addition or substitution of one or more serine or threonine residues to the starting sequence (for O-linked glycosylation sites). For ease of use, the amino acid sequence of the antibody construct is preferably modified at the DNA level, particularly by mutating the polypeptide-encoding DNA with pre-selected bases, so that codons translating to the desired amino acids are produced.

[0337] Another means of increasing the number of carbohydrate moieties on antibody constructs is by chemical or enzymatic binding of glycosides to proteins. These procedures are advantageous in that they do not require the production of proteins in host cells that have the glycosylation capacity for N and O-linked glycosylation. Depending on the binding mode used, sugars can be bound to (a) arginine and histidine, (b) free carboxyl groups, (c) free sulfhydryl groups such as those of cysteine, (d) free hydroxyl groups such as those of serine, threonine, or hydroxyproline, (e) aromatic residues such as those of phenylalanine, tyrosine, or tryptophan, or (f) amide groups of glutamine. These methods are described in International Publication No. 87 / 05330 and Aplin and Wriston, 1981, CRC Crit. Rev. Biochem., pp. 259-306.

[0338] The removal of carbohydrate moieties present on starting antibody constructs can be achieved chemically or enzymatically. Chemical deglycosylation requires exposure of the protein to the compound trifluoromethanesulfonic acid or an equivalent compound. This process results in the cleavage of almost all sugars except for the linking sugars (N-acetylglucosamine or N-acetylgalactosamine), while leaving the polypeptide intact. Chemical deglycosylation has been described by Hakimuddin et al., 1987, Arch. Biochem. Biophys. 259:52, and Edge et al., 1981, Anal. Biochem. 118:131. Enzymatic cleavage of carbohydrate moieties on polypeptides can be achieved by the use of various endo and exoglycosidases, as described by Thotakura et al., 1987, Meth. Enzymol. 138:350. Glycosylation at potential glycosylation sites can be prevented by the use of the compound tunicamycin, as described by Duskin et al., 1982, J. Biol. Chem. 257:3105. Tunicamycin blocks the formation of protein-N-glycosidic bonds.

[0339] Other modifications of antibody constructs are also contemplated herein. For example, another type of covalent modification of antibody constructs includes linking the antibody construct to a variety of non-proteinoid polymers, including but not limited to various polyols such as polyethylene glycol, polypropylene glycol, polyoxyalkylene, or copolymers of polyethylene glycol and polypropylene glycol, in the manner described in U.S. Patent Nos. 4,640,835; 4,496,689; 4,301,144; 4,670,417; 4,791,192 or 4,179,337. Furthermore, as is known in the art, amino acid substitutions can be made at various positions within the antibody construct to facilitate the addition of polymers such as PEG.

[0340] In some embodiments, the covalent modification of the antibody construct of the present invention involves the addition of one or more labels. Labeling groups may be attached to the antibody construct via spacer arms of varying lengths to reduce potential steric hindrance. Various methods for labeling proteins are known in the art and can be used in carrying out the present invention. The terms “label” or “labeling group” refer to any detectable label. Generally, labels are classified into various classes depending on the assay in which they are detected, and examples below include, but are not limited to: a) Radioactive isotopes or radionuclides (for example, 3 H, 14 C, 15 N, 35 S, 89 Zr, 90 Y, 99 Tc, 111 In, 125 I, 131 I) Isotope labels that may be radioactive or heavy isotopes, etc. b) Magnetic labeling (e.g., magnetic particles) c) Redox active moiety d) Optical dyes such as fluorescent groups (e.g., FITC, rhodamine, lanthanide phosphors), chemiluminescent groups, and phosphors that may be either "small molecule" phosphors or protein phosphors (including, but not limited to, chromophores, phosphors, and fluorophores). e) Enzyme groups (e.g., horseradish peroxidase, β-galactosidase, luciferase, alkaline phosphatase) f) Biotinylation group g) A predetermined polypeptide epitope recognized by a secondary reporter (e.g., leucine zipper pair sequence, binding side of a secondary antibody, metal-binding domain, epitope tag, etc.)

[0341] Leucine zipper domains are peptides that promote the oligomerization of the proteins in which they are found. Leucine zippers were first identified in several DNA-binding proteins (Landschulz et al., 1988, Science 240:1759) and have since been found in a variety of different proteins. Known leucine zippers include naturally occurring peptides and their derivatives that dimerize or trimerize. Examples of leucine zipper domains suitable for producing soluble oligomeric proteins are described in PCT application international publication 94 / 10308 brochure, and a leucine zipper derived from lung surfactant protein D (SPD) is described in Hoppe et al., 1994, FEBS Letters 344:191. The use of modified leucine zippers that enable stable trimerization of heterologous proteins fused to them is described in Fanslow et al., 1994, Semin.Immunol.6:267-78.

[0342] The antibody constructs of the present invention may also include additional domains that are useful, for example, for the isolation of molecules or for a fitted pharmacokinetic profile of the molecule. Domains useful for the isolation of antibody constructs can be selected from peptide motifs or secondary introduction portions that can be captured by isolation methods, such as isolation columns. Non-limiting embodiments of such further domains include peptide motifs known as Myc-tags, HAT-tags, HA-tags, TAP-tags, GST-tags, chitin-binding domains (CBD-tags), maltose-binding proteins (MBP-tags), Flag-tags, Strep-tags and their variants (e.g., StrepII-tags) and His-tags. All antibody constructs disclosed herein may include His-tag domains, commonly known as repeats of consecutive His residues, preferably five, more preferably six His residues (hexa-histidine) (SEQ ID NO: 18451), in the amino acid sequence of a molecule. The His-tag may be located, for example, at the N-terminus or C-terminus of the antibody construct, preferably at the C-terminus. Most preferably, the hexa-histidine tag (HHHHHH) (SEQ ID NO: 30) is ligated to the C-terminus of the antibody construct according to the present invention via a peptide bond. Furthermore, the PLGA-PEG-PLGA conjugate system can be combined with the polyhistidine tag for sustained-release application and improved pharmacokinetic profile.

[0343] "Fluorescent labeling" refers to any molecule that can be detected through its inherent fluorescent properties. Suitable fluorescent labels include fluorescein, rhodamine, tetramethylrhodamine, eosin, erythrosine, coumarin, methyl-coumarin, pyrene, malasite green, stilbene, Lucifer yellow, Cascade Blue J, Texas red, IAEDANS, EDANS, BODIPY FL, LC Red 640, Cy5, Cy5.5, LC Red 705, Oregon green, Alexa-Fluor dyes (Alexa Fluor 350, Alexa Fluor 430, Alexa Fluor 488, Alexa Fluor 546, Alexa Fluor 568, Alexa Fluor 594, Alexa Fluor 633, Alexa Fluor 660, Alexa Fluor 680), Cascade Blue, Cascade Yellow, and R-phycoerythrin (PE) (Molecular Probes, Eugene, OR), FITC, rhodamine, and Texas This includes, but is not limited to, Red (Pierce, Rockford, IL), Cy5, Cy5.5, and Cy7 (Amersham Life Science, Pittsburgh, PA). Suitable optical dyes containing fluorophores are listed in the Molecular Probes Handbook by Richard P. Haugland.

[0344] Suitable protein-based fluorescent labels include, but are not limited to, green fluorescent proteins containing GFP from Renilla, Ptilosarcus, and Aequorea species (Chalfie et al., 1994, Science 263:802-805), EGFP (Clontech Laboratories, Inc., Genbank Accession Number U55762), blue fluorescent protein (BFP, Quantum Biotechnologies, Inc. 1801 de Maisonneuve Blvd. West, 8th Floor, Montreal, Quebec, Canada H3H 1J9; Stauber, 1998, Biotechniques 24:462-471; Heim et al., 1996, Curr. Biol. 6:178-182), enhanced yellow fluorescent protein (EYFP, Clontech Laboratories, Inc.), and luciferase (Ichiki et al. al., 1993, J.Immunol. 150:5408-5417), β-galactosidase (Nolan et al.) Examples include al., 1988, Proc. Natl. Acad. Sci. USA 85:2603-2607) and Renilla (International Publication No. 92 / 15673, International Publication No. 95 / 07463, International Publication No. 98 / 14605, International Publication No. 98 / 26277, International Publication No. 99 / 49019, U.S. Patent Nos. 5,292,658; 5,418,155; 5,683,888; 5,741,668; 5,777,079; 5,804,387; 5,874,304; 5,876,995; 5,925,558).

[0345] Pharmaceutical preparations Furthermore, the present invention provides a pharmaceutical composition or formulation comprising the antibody construct of the present invention or an antibody construct produced according to the process of the present invention.

[0346] As used herein, the term “pharmaceutical composition” refers to a composition suitable for administration to a patient, preferably a human patient. Particularly preferred pharmaceutical compositions of the present invention contain one or more antibody constructs of the present invention, preferably in a therapeutically effective amount. Preferably, the pharmaceutical composition further comprises one or more suitable formulations of pharmaceutically effective carriers, stabilizers, excipients, diluents, solubilizers, surfactants, emulsifiers, preservatives and / or adjuvants. The components of the acceptable composition are preferably nontoxic to the recipient at the doses and concentrations employed. Pharmaceutical compositions of the present invention include, but are not limited to, liquid, freeze-dried, and lyophilized compositions.

[0347] The composition may contain a pharmaceutically acceptable carrier. Generally, as used herein, “pharmaceutically acceptable carrier” means any aqueous and non-aqueous solution, sterile solution, solvent, buffer, e.g., phosphate-buffered saline (PBS) solution, water, suspension, emulsion such as oil / water emulsion, various types of wetting agents, liposomes, dispersion media, and coatings suitable for pharmaceutical administration, particularly parenteral administration. The use of such media and agents in pharmaceutical compositions is well known in the art, and compositions containing such carriers can be formulated by well known conventional methods.

[0348] Certain embodiments of the present invention provide a pharmaceutical composition comprising the antibody construct and one or more excipients, for example, those described exemplary in this section and elsewhere in this specification. Excipients can be used in the present invention for a wide range of purposes, such as methods of the present invention for modifying the physical, chemical, or biological properties of a formulation, such as viscosity, and / or improving its efficacy, and / or stabilizing such formulations, as well as methods for preventing degradation and damage caused by stress during, for example, manufacturing, transport, storage, preparation before use, during and after administration. Excipients should generally be used at the lowest effective concentration.

[0349] In certain embodiments, the pharmaceutical composition may include formulation materials for the purpose of modifying, maintaining, or preserving specific properties of the composition, such as pH, molar osmotic pressure, viscosity, clarity, color, isotonicity, odor, sterility, stability, dissolution rate or release rate, adsorption or osmosis (see Remington's Pharmaceutical Sciences, 18th Edition, 1990, Mack Publishing Company). In such embodiments, suitable formulation materials include, but are not limited to, amino acids, antimicrobial agents such as antibacterial and antifungal agents, antioxidants, buffers, buffer systems and buffers (used to maintain the composition at or slightly below physiological pH, typically within a pH range of about 5 to about 8 or 9), non-aqueous solvents, vegetable oils and injectable organic esters, water, aqueous carriers including alcohol / aqueous solutions, emulsions or suspensions. Examples of materials include (physiological saline and buffering media), biodegradable polymers (such as polyester), bulking agents, chelating agents, isotonic agents and absorption retarders, complexing agents, fillers, carbohydrates, (low molecular weight) proteins, polypeptides or protein carriers, preferably of human origin, colorants and flavoring agents, sulfur-containing reducing agents, diluents, emulsifiers, hydrophilic polymers, salt-forming counterions, preservatives, metal complexes, solvents and co-solvents, sugars and sugar alcohols, suspending agents, surfactants or wetting agents, stability enhancers, isotonic enhancers, parenteral delivery vehicles or intravenous delivery vehicles.

[0350] Different components of a pharmaceutical composition may have different effects; for example, amino acids can act as buffers, stabilizers, and / or antioxidants; mannitol can act as fillers and / or isotonic enhancers; and sodium chloride can act as delivery vehicles and / or isotonic enhancers.

[0351] In connection with the present invention, the pharmaceutical composition may comprise (a) an antibody construct as described herein, (b) at least one buffer, (c) at least one sugar, and (d) at least one surfactant, wherein the pH of the pharmaceutical composition is in the range of 3.5 to 6.

[0352] In the above-described composition, the first domain preferably has an isoelectric point (pI) in the range of 4 to 9.5, the second domain has a pI in the range of 8 to 10, preferably 8.5 to 9.0, and the antibody construct optionally includes a third domain comprising two polypeptide monomers, each comprising a hinge, a CH2 domain and a CH3 domain, the two polypeptide monomers being fused to each other via a peptide linker.

[0353] In the above composition, it is further assumed that at least one buffer is present in a concentration range of 5 to 200 mM, more preferably in a concentration range of 10 to 50 mM. It is also assumed that at least one sugar is selected from the group consisting of monosaccharides, disaccharides, cyclic polysaccharides, sugar alcohols, linear branched dextran, or linear unbranched dextran. The disaccharide is also assumed to be selected from the group consisting of sucrose, trehalose, mannitol, sorbitol, and combinations thereof. It is further assumed that the sugar alcohol is sorbitol. It is also assumed that at least one sugar is present in a concentration range of 1 to 15% (m / V), preferably in a concentration range of 9 to 12% (m / V). It is further assumed that the antibody construct is present in a concentration range of 0.1 to 8 mg / ml, preferably 0.2 to 2.5 mg / ml, more preferably 0.25 to 1.0 mg / ml.

[0354] According to one embodiment of the above composition, at least one surfactant is selected from the group consisting of polysorbate 20, polysorbate 40, polysorbate 60, polysorbate 80, poloxamer 188, Pluronic F68, Triton X-100, polyoxyethylene, PEG3350, PEG4000, and combinations thereof. It is further assumed that at least one surfactant is present at a concentration in the range of 0.004 to 0.5% (m / V), preferably in the range of 0.001 to 0.01% (m / V). The pH of the composition is assumed to be in the range of 4.0 to 5.0, preferably in the range of 4.2. The pharmaceutical composition is also assumed to have a molar osmotic pressure concentration in the range of 150 to 500 mOsm. The pharmaceutical composition is further assumed to contain an excipient selected from the group consisting of one or more polyols and one or more amino acids. In relation to the present invention, it is assumed that one or more excipients are present in a concentration range of 0.1 to 15% (w / V).

[0355] The present invention also, (a) Antibody constructs as specified herein, preferably in a concentration range of 0.1 to 8 mg / ml, preferably 0.2 to 2.5 mg / ml, and more preferably 0.25 to 1.0 mg / ml; (b) 10 mM glutamate or acetate; (c) 9% (m / V) sucrose or 6% (m / V) sucrose and 6% (m / V) hydroxypropyl-β-cyclodextrin; (d) 0.01% (m / V) Polysorbate 80 A pharmaceutical composition containing the following is provided: The pH of the liquid pharmaceutical composition is 4.2.

[0356] The compositions of the present invention are expected to include, in addition to the antibody constructs of the present invention as defined herein, further biologically active agents depending on the intended use of the composition. Such agents may include drugs known in the art that act on the gastrointestinal system, drugs that act as cell proliferation inhibitors, drugs that prevent hyperurikemia, drugs that inhibit immune responses, drugs that modulate inflammatory responses, drugs that act on the circulatory system, and / or cytokines. Furthermore, the antibody constructs of the present invention are expected to be used in combination therapy, i.e., in combination with other anticancer drugs.

[0357] In this regard, the pharmaceutical composition of the present invention (including an antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of a target cell and a second domain that binds to CD3 on the surface of a T cell, as described in more detail herein) is envisioned to further comprise a drug, preferably an antibody or antibody construct, that binds to a protein of the immune checkpoint pathway (such as PD-1 or CTLA-4) or a co-stimulatory immune checkpoint receptor (such as 4-1BB). The present invention also refers to combinations of an antibody construct according to the present invention (including an antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of a target cell and a second domain that binds to CD3 on the surface of a T cell, as described in more detail herein) and a drug, preferably an antibody or antibody construct, that binds to a protein of the immune checkpoint pathway (such as PD-1 or CTLA-4) or a co-stimulatory immune checkpoint receptor (such as 4-1BB). Due to the properties of at least two components of the combination, i.e., their pharmaceutically active properties, the combination may also be called a therapeutic combination. In some embodiments, the combination may be in the form of a pharmaceutical composition or a kit. According to one embodiment, a pharmaceutical composition or combination comprises an antibody construct of the present invention and an antibody or antibody construct that binds to PD-1. Anti-PD-1 binding proteins useful for this purpose are described in detail, for example, PCT / US2019 / 013205.

[0358] In certain embodiments, the optimal pharmaceutical composition is determined, for example, according to the intended route of administration, the form of delivery, and the desired dose. See, for example, Remington's Pharmaceutical Sciences cited above. In certain embodiments, such a composition may affect the physical state, stability, in vivo release rate, and in vitro clearance rate of the antibody construct of the present invention. In certain embodiments, the main vehicle or carrier in the pharmaceutical composition may be either aqueous or non-aqueous. For example, suitable vehicles or carriers may be water for injection, physiological saline solution, and may be supplemented with other materials common in parenteral administration compositions. In certain embodiments, a composition containing the antibody construct of the present invention may be prepared for storage by optionally mixing a selected composition having a desired degree of purity with a compounding agent (see, Remington's Pharmaceutical Sciences cited above) in the form of a lyophilized cake or aqueous solution. Furthermore, in certain embodiments, the antibody construct of the present invention may be formulated as a lyophilized product using appropriate excipients.

[0359] When parenteral administration is intended, the therapeutic composition for use in the present invention may be provided in the form of a pyrogen-free, parenterally acceptable aqueous solution containing the desired antibody construct of the present invention in a pharmaceutically acceptable vehicle. A vehicle particularly suitable for parenteral injection is sterile distilled water, in which the antibody construct of the present invention is formulated as a sterile isotonic solution appropriately stored therein. In certain embodiments, the formulation may include a formulation of the desired molecule with an agent that can provide controllability or sustained release of the product that can be delivered via depot injection, or an agent that can enhance the duration of action in circulation. In certain embodiments, the desired antibody construct may be introduced using an implantable drug delivery device.

[0360] Further pharmaceutical compositions, including formulations containing the antibody construct of the present invention in sustained or controlled delivery formulations, will be apparent to those skilled in the art. Techniques for formulating various sustained or controlled delivery methods are known to those skilled in the art. The antibody construct can also be encapsulated in microcapsules prepared in colloidal drug delivery systems or in macroemulsions, for example, by coacervate technology or interfacial polymerization. Such techniques are disclosed in the aforementioned Remington's Pharmaceutical Sciences.

[0361] Pharmaceutical compositions used for in vivo administration are generally provided as sterile formulations. Sterilization can be achieved by filtration through a sterile filtration membrane. When the composition is lyophilized, sterilization using this method can be performed either before or after lyophilization and reconstitution. Parenteral compositions can be stored in lyophilized form or as a solution. Parenteral compositions are generally filled into containers with a sterile access port, such as intravenous solution bags or vials with stoppers that can be penetrated by subcutaneous needles.

[0362] Another aspect of the present invention includes a self-buffering formulation comprising...

Claims

1. An isolated antibody construct comprising a binding domain that binds to pMAGE-HLA on the surface of target cells, wherein the binding domain is a) A VH region including CDR-H1 shown in Sequence ID No. 439, CDR-H2 shown in Sequence ID No. 440, and CDR-H3 shown in Sequence ID No. 441, and a VL region including CDR-L1 shown in Sequence ID No. 223, CDR-L2 shown in Sequence ID No. 224, and CDR-L3 shown in Sequence ID No. 225; or b) The VH region shown in Sequence ID 629 and the VL region shown in Sequence ID 628 Antibody constructs that include this.

2. An isolated antibody construct comprising a first domain that binds to pMAGE-HLA on the surface of target cells and a second domain that binds to human CD3 on the surface of T cells, wherein the first domain is a) A VH region including CDR-H1 shown in Sequence ID No. 439, CDR-H2 shown in Sequence ID No. 440, and CDR-H3 shown in Sequence ID No. 441, and a VL region including CDR-L1 shown in Sequence ID No. 223, CDR-L2 shown in Sequence ID No. 224, and CDR-L3 shown in Sequence ID No. 225; or b) The VH region shown in Sequence ID 629 and the VL region shown in Sequence ID 628 Antibody constructs that include this.

3. The antibody construct according to claim 2, wherein the second domain binds to human CD3 epsilon and common marmoset (Callithrix jacchus) or common squirrel monkey (Saimiri sciureus) CD3 epsilon.

4. The antibody construct comprises a first binding domain and a second binding domain, a) The antibody construct is a single-chain antibody construct, b) The first domain is in scFv format, c) The second domain is in scFv format, d) The first domain and the second domain are linked via a linker, and / or e) The antibody construct includes a domain that provides an extended serum half-life, An antibody construct according to any one of claims 1 to 3.

Images