Antigen-binding proteins containing two Fc domains and uses thereof

Abstract

The present invention provides a fusion protein with a novel antibody structure, comprising one antigen-binding site and two Fc domains. Although such a novel antibody has a molecular weight similar to that of human IgG, the antibody structure allows the Fc domain to be present on cell surface antigens up to four times more than natural human antibodies. Thus, the fusion protein has increased affinity for Fcγ receptors and has increased effector function. Therefore, the fusion protein with the novel antibody format can be used as a novel antibody platform.

Description

[Technical field]

[0001] The present invention relates to a novel antibody format having an antigen-binding site that specifically binds to a cancer surface antigen and two Fc domains. [Background technology]

[0002] Antibody-based therapeutics and Fc fusion proteins are a group of clinically important drugs for patients with cancer, immune disorders, infectious diseases, and inflammatory diseases. ADCC (antibody-dependent cell-mediated cytotoxicity), ADCP (antibody-dependent cellular phagocytosis), and CDC (complement-dependent cytotoxicity), induced by the interaction between the antibody Fc domain and the innate immune system, play important roles in alleviating or treating the symptoms of diseases.

[0003] Attempts are being made to maintain the bivalency of antibodies and improve effector function by increasing the number of Fc domains (Claudio Sustmann et al., MAbs. 2019; Dennis R Goulet et al., Proteins, 2020). Although these platforms have shown improved binding ability to Fcγ receptors and ADCC, it is difficult to obtain homogeneous antibodies due to the complexity of production and purification. Attempts are also currently underway to improve effector function by connecting antibody Fc domains in series or constructing multiple Fc domains (US Patent Application Publication No. 2020 / 0040084). In this case, there is a disadvantage that tissue permeability may be significantly reduced due to an increase in the size or molecular weight of the antibody. Summary of the Invention [Problem to be solved by the invention]

[0004] The inventors therefore investigated to improve antibody function and simultaneously solve the problems of existing antibody formats that are designed to contain multiple Fc domains in tandem as described above. [Means for solving the problem]

[0005] As a result, the inventors have developed a new, improved antibody format that allows the Fc domain to be up to four times more present on cell surface antigens compared to native human antibodies, even though the antibody has a molecular weight similar to that of native human immunoglobulin G (IgG) (approximately 150 kDa).

[0006] In one embodiment of the invention there is provided a fusion protein comprising an antigen binding site, a first Fc domain or variant thereof linked to a first linkage position of the antigen binding site, and a second Fc domain or variant thereof linked to a second linkage position of the antigen binding site.

[0007] In another aspect of the present invention, there is provided a fusion protein comprising two antigen-binding sites linked in tandem (two antigen-binding regions in tandem), a first Fc domain or a variant thereof linked to a first linking position of the two antigen-binding sites in tandem, and a second Fc domain or a variant thereof linked to a second linking position of the two antigen-binding sites in tandem. According to one embodiment, each of the two antigen-binding sites constituting the two antigen-binding sites in tandem may be a sequence comprising a CDR sequence or a variable region capable of binding to different epitopes of the same antigen or different antigens, or a sequence consisting of a variable region.

[0008] In the fusion protein described in the present disclosure, according to one embodiment, the antigen-binding site may be a sequence including the CDR sequence or variable region of an antibody, or a sequence consisting of the variable region. Thus, the antigen-binding site may comprise a first peptide consisting of or comprising the light chain CDR sequence or light chain variable region of an antibody, and a second peptide consisting of or comprising the heavy chain CDR sequence or heavy chain variable region of an antibody. Each of the first Fc domain and the second Fc domain may be a dimer consisting of two peptide sequences. The first peptide of the antigen-binding site binds to the first Fc domain or a variant thereof, and the second peptide of the antigen-binding site binds to the second Fc domain or a variant thereof.

[0009] In the fusion protein described in the present disclosure, according to another embodiment, the first Fc domain and the second Fc domain may be linked to each other via a covalent bond, a non-covalent bond, or a linker, or may not be linked to each other. In a preferred embodiment, the first Fc domain and the second Fc domain are not linked to each other.

[0010] According to embodiments of the present disclosure, the Fc domain may be an Fc domain of a wild-type immunoglobulin or may include modifications to modulate the reactivity of the Fc domain to Fcγ receptors (FcγR), ADCC, or to minimize undesired multimer formation of the Fc domain, e.g., amino acid substitutions. The Fc domain includes CH2 and CH3 regions, and may include a CH4 region and / or a hinge region, and the Fc domain should be taken to include Fc domain fragments that exhibit the function of an Fc domain.

[0011] According to embodiments of the present disclosure, the antigen-binding site and the Fc domain or variant thereof may be joined directly or through a linker. For example, the antigen-binding site and the Fc domain or variant thereof may be connected between the N-terminus and C-terminus, between the N-terminus and N-terminus, or between the C-terminus and C-terminus of each peptide molecule, with or without a linker.

[0012] According to an embodiment of the present disclosure, when the antigen-binding site and the Fc domain or a variant thereof are joined via a linker, the linker may be a commonly used peptide linker. For example, the linker may be a peptide consisting of 1 to 70 amino acid residues, 2 to 60 amino acid residues, 2 to 50 amino acid residues, 2 to 40 amino acid residues, 2 to 30 amino acid residues, 3 to 50 amino acid residues, 3 to 40 amino acid residues, 3 to 30 amino acid residues, 2 to 28 amino acid residues, 2 to 26 amino acid residues, 2 to 24 amino acid residues, 2 to 22 amino acid residues, 2 to 20 amino acid residues, 2 to 18 amino acid residues, 2 to 16 amino acid residues, 2 to 14 amino acid residues, 2 to 12 amino acid residues, or 2 to 10 amino acid residues. The connection between the first Fc domain and the antigen-binding site, the connection between the second Fc domain and the antigen-binding site, or both may be achieved via a linker.

[0013] In another aspect of the present invention, there is provided a pharmaceutical composition for preventing or treating cancer, comprising the fusion protein as an active ingredient.

[0014] In another aspect of the invention, there are provided nucleotides encoding the fusion proteins, vectors comprising the nucleotides, and transformed cells into which the vectors have been introduced.

[0015] In another aspect of the invention, there is provided a method for treating or preventing cancer, comprising administering to a subject a fusion protein.

[0016] In another aspect of the invention, there is provided a use of the fusion protein for the treatment of cancer. Effect of the Invention

[0017] Unlike wild-type antibodies, the fusion protein with the novel antibody format of the present invention comprises one or two antigen-binding sites and two Fc domains. The two Fc domains are not directly linked to each other, but are independently linked to two different polypeptide chains that constitute the antigen-binding site, respectively. Even though this novel antibody format has a size and molecular weight similar to that of human IgG, the novel antibody format may allow the Fc domain to be present on cell surface antigens up to four times more than natural human antibodies. Due to these properties, the fusion protein with the novel antibody format may have improved affinity (avidity) for Fcγ receptors and induce improved effector functions. Therefore, the fusion protein with the novel antibody format may be utilized for various purposes by replacing conventional antibodies. [Brief description of the drawings]

[0018] [Figure 1a] FIG. 1 is a schematic diagram of a native human immunoglobulin (IgG). [Figure 1b] FIG. 1 is a schematic diagram of a novel engineered antibody format. [Figure 1c] FIG. 1 is a schematic diagram showing a cancer cell with its tumor antigen bound in a monovalent manner to an antigen-specific human immunoglobulin (IgG). [Figure 1d] FIG. 1 is a schematic diagram showing a cancer cell with its tumor antigen bound to an antigen-specific human immunoglobulin (IgG) in a monovalent or bivalent manner. [Figure 1e] FIG. 1 is a schematic diagram showing a cancer cell with its tumor antigen bound in a bivalent manner to an antigen-specific human immunoglobulin (IgG). [Figure 1f] FIG. 1 is a schematic diagram showing a cancer cell with its tumor antigen bound to an antigen-specific novel engineered antibody. [Figure 2a] FIG. 1 is a schematic diagram of a novel monovalent antibody format (WT) with two Fc domains. [Figure 2b]FIG. 1 is a schematic diagram of an antibody (M1) in which the VH Q105C and VL A43C amino acid substitutions have been introduced into a novel monovalent antibody format with two Fc domains. [Figure 2c] FIG. 1 is a schematic diagram of an antibody (M2) in which CH1 F122C and CL S121C amino acid substitutions have been introduced into a novel monovalent antibody format with two Fc domains. [Figure 2d] FIG. 1 is a schematic diagram of an antibody (M3) in which the VH G44C and VL Q100C amino acid substitutions have been introduced into a novel monovalent antibody format with two Fc domains. [Figure 3a] 1 is a diagram of sequence information showing the location of Cys substitutions in the VH-CH1 domain of Trastuzumab. The WT sequence is SEQ ID NO:8, the Mutant 1 sequence is SEQ ID NO:10, the Mutant 2 sequence is SEQ ID NO:12, and the Mutant 3 sequence is SEQ ID NO:14. [Figure 3b] 1 is a diagram of sequence information showing the location of Cys substitutions in the VL-CL domain of Trastuzumab. The WT sequence is SEQ ID NO:9, the Mutant 1 sequence is SEQ ID NO:11, the Mutant 2 sequence is SEQ ID NO:13, and the Mutant 3 sequence is SEQ ID NO:15. [Figure 4] FIG. 1 illustrates the results obtained by SDS-PAGE analysis of WT, M1, M2, and M3. [Figure 5a] FIG. 1 illustrates the results obtained by size-exclusion chromatography analysis of WT. [Figure 5b] FIG. 1 illustrates the results obtained by size exclusion chromatography analysis of M1. [Figure 5c] FIG. 1 illustrates the results obtained by size exclusion chromatography analysis of M2. [Figure 5d] FIG. 1 illustrates the results obtained by size exclusion chromatography analysis of M3. [Figure 6a] FIG. 1 is a schematic diagram of an antibody (M3) in which the CL domain and hinge region are linked by a 15-mer peptide. [Figure 6b]FIG. 1 is a schematic diagram of an antibody (V1) in which the CL domain and hinge region are linked by a 10-mer peptide. [Figure 6c] FIG. 1 is a schematic diagram of an antibody (V2) in which the CL domain and hinge region are linked by a 5-mer peptide. [Figure 6d] Schematic diagram of an antibody (V3) in which the CL domain and hinge region are directly linked without a linker. [Figure 7] FIG. 1 illustrates the results obtained by SDS-PAGE analysis of M3, V1, V2, and V3. [Figure 8a] FIG. 1 is a schematic representation of the fragments produced when H01 and P01 are cleaved with papain. [Figure 8b] FIG. 1 is a schematic diagram of fragments generated by papain cleavage of H01 and P01 when the disulfide bond in the hinge region is abnormally formed. [Figure 8c] FIG. 1 illustrates the results obtained by SDS-PAGE analysis of H01 papain cleavage products. [Figure 8d] FIG. 1 illustrates the results obtained by SDS-PAGE analysis of P01 papain cleavage products. [Figure 9] FIG. 1 illustrates the results obtained by SDS-PAGE analysis of H01wt and H01. [Figure 10a] FIG. 1 is a schematic diagram of H01Fv1 having an Fv-(Fc)2 structure. [Figure 10b] FIG. 1 is a schematic diagram of H01Fv2 having an Fv-(Fc)2 structure. [Figure 10c] FIG. 1 is a schematic diagram of H01Fv3 having an Fv-(Fc)2 structure. [Figure 10d] FIG. 1 is a schematic diagram of H01Fv4 having an Fv-(Fc)2 structure. [Figure 10e] FIG. 1 is a schematic diagram of H01Fv5 having an Fv-(Fc)2 structure. [Figure 10f] FIG. 1 is a schematic diagram of H01Fv6 having an Fv-(Fc)2 structure. [Figure 10g]FIG. 1 is a schematic diagram of H01Fv7 having an Fv-(Fc)2 structure. [Figure 10h] 1 is a table showing the mutation positions in the Fv-(Fc)2 structure. [Figure 11] FIG. 1 illustrates the results obtained by SDS-PAGE analysis of Fv-(Fc)2 structure after Protein A purification. [Figure 12a] FIG. 1 illustrates the results obtained by SEC analysis of Fv-(Fc)2 structure after Protein A purification. [Figure 12b] FIG. 1 illustrates the results obtained by SEC analysis of Fv-(Fc)2 structure after Protein A purification. [Figure 12c] FIG. 1 illustrates the results obtained by SEC analysis of Fv-(Fc)2 structure after Protein A purification. [Figure 12d] FIG. 1 illustrates the results obtained by SEC analysis of Fv-(Fc)2 structure after Protein A purification. [Figure 12e] FIG. 1 illustrates the results obtained by SEC analysis of Fv-(Fc)2 structure after Protein A purification. [Figure 12f] FIG. 1 illustrates the results obtained by SEC analysis of Fv-(Fc)2 structure after Protein A purification. [Figure 12g] FIG. 1 illustrates the results obtained by SEC analysis of Fv-(Fc)2 structure after Protein A purification. [Figure 13a] FIG. 1 illustrates the analysis of sensorgram data for binding of purified Fv-(Fc)2 construct, H01Fv1, to human HER2. [Figure 13b] FIG. 1 illustrates the analysis of sensorgram data for binding of purified Fv-(Fc)2 construct, H01Fv2, to human HER2. [Figure 13c] FIG. 1 illustrates the analysis of sensorgram data for binding of purified Fv-(Fc)2 construct, H01Fv4, to human HER2. [Figure 13d]FIG. 1 illustrates the analysis of sensorgram data for binding of purified Fv-(Fc)2 construct, H01Fv5, to human HER2. [Figure 13e] FIG. 1 illustrates the analysis of sensorgram data for binding of purified Fv-(Fc)2 construct, H01Fv6, to human HER2. [Figure 13f] FIG. 1 illustrates the analysis of sensorgram data for binding of purified Fv-(Fc)2 construct, H01Fv7, to human HER2. [Figure 14] FIG. 1 illustrates differential scanning fluorescence spectroscopy analysis of the melting temperatures of H01, P01, Trastuzumab, and Pertuzumab. [Figure 15] FIG. 11. Biolayer interferometry analysis of sensorgram data showing competitive binding of H01 and P01. [Figure 16a] FIG. 1 is a schematic diagram showing the binding mode of H01 in combination with P01 to HER2 in HER2-positive cancer cells. [Figure 16b] FIG. 1 is a schematic diagram showing the binding mode of trastuzumab in combination with pertuzumab to HER2 in HER2-positive cancer cells. [Figure 17a] FIG. 1 illustrates flow cytometry analysis of the amount of Fc domains present on the surface of NCI-N87 gastric cancer cell line upon treatment with 50 nM anti-HER2 antibody. [Figure 17b] FIG. 1 illustrates flow cytometry analysis of the amount of Fc domains present on the surface of the BT474 breast cancer cell line upon treatment with 50 nM anti-HER2 antibody. [Figure 17c] FIG. 1 illustrates flow cytometry analysis of the amount of Fc domains present on the surface of the SK-OV3 ovarian cancer cell line upon treatment with 50 nM anti-HER2 antibody. [Figure 17d] FIG. 1 illustrates flow cytometry analysis of the amount of Fc domains present on the surface of the SNU-1 gastric cancer cell line upon treatment with 50 nM anti-HER2 antibody. [Figure 17e]FIG. 1 illustrates flow cytometry analysis of the amount of Fc domains present on the surface of the SNU-5 gastric cancer cell line upon treatment with 50 nM anti-HER2 antibody. [Figure 18a] FIG. 1 illustrates flow cytometry analysis of the amount of Fc domains present on the surface of NCI-N87 gastric cancer cell lines upon treatment with anti-HER2 antibodies at the indicated concentrations. [Figure 18b] FIG. 1 illustrates flow cytometry analysis of the amount of Fc domains present on the surface of the BT474 breast cancer cell line upon treatment with the indicated concentrations of anti-HER2 antibodies. [Figure 18c] FIG. 1 illustrates flow cytometry analysis of the amount of Fc domains present on the surface of the SK-OV3 ovarian cancer cell line upon treatment with the indicated concentrations of anti-HER2 antibodies. [Figure 18d] FIG. 1 illustrates flow cytometry analysis of the amount of Fc domains present on the surface of the SNU-1 gastric cancer cell line upon treatment with the indicated concentrations of anti-HER2 antibodies. [Figure 18e] FIG. 1 illustrates flow cytometry analysis of the amount of Fc domains present on the surface of the SNU-5 gastric cancer cell line upon treatment with the indicated concentrations of anti-HER2 antibodies. [Figure 19a] FIG. 1 is a schematic diagram of H01DE4 in which S239D and I332E mutations were introduced into H01. [Figure 19b] FIG. 1 is a schematic diagram of P01DE4, in which S239D and I332E mutations were introduced into P01. [Figure 20a] Sensorgram binding profiles of H01, H01DE4, P01, and P01DE4 to human HER2. [Figure 20b] Sensorgram binding profiles of H01, H01DE4, P01, and P01DE4 to human HER2. [Figure 20c] Sensorgram binding profiles of H01, H01DE4, P01, and P01DE4 to human HER2. [Figure 20d]Sensorgram binding profiles of H01, H01DE4, P01, and P01DE4 to human HER2. [Figure 21a] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 1. [Figure 21b] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 1. [Figure 21c] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 1. [Figure 21d] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 1. [Figure 21e] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 1. [Fig. 21f] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 1. [Fig. 21g] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 1. [Fig. 21h] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 1. [Figure 22a]Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 2A (131R isoform). [Figure 22b] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 2A (131R isoform). [Figure 22c] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 2A (131R isoform). [Fig. 22d] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 2A (131R isoform). [Figure 22e] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 2A (131R isoform). [Fig. 22f] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 2A (131R isoform). [Fig. 22g] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 2A (131R isoform). [Fig. 22h] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 2A (131R isoform). [Figure 23a] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 3A (176V isoform). [Figure 23b] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 3A (176V isoform). [Figure 23c] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 3A (176V isoform). [Figure 23d] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 3A (176V isoform). [Figure 23e] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 3A (176V isoform). [Fig. 23f] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 3A (176V isoform). [Figure 23g] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 3A (176V isoform). [Figure 23h] Sensorgram binding profiles of H01, P01, H01DE4, P01DE4, human IgG1, trastuzumab, pertuzumab, and marjetuximab to Fcγ receptor 3A (176V isoform). [Figure 24a] 1 is a graph showing blood antibody concentrations over time when H01, P01, trastuzumab, and pertuzumab were intravenously administered at 10 mg / kg to Sprague-Dawley rats. [Figure 24b] FIG. 24 is a table illustrating PK parameters calculated from FIG. 24a when H01, P01, trastuzumab, and pertuzumab were administered intravenously at 10 mg / kg to Sprague-Dawley rats. [Diagram 25] FIG. 1 is a schematic diagram of HP501, a HER2 biparatopic engineered antibody. [Figure 26] FIG. 1 is a schematic diagram of HER2 biparatopic engineered antibodies, HP501-HP516. [Figure 27] FIG. 1 illustrates size exclusion chromatography analysis of HER2 biparatopic engineered antibodies HP501-HP516. [Figure 28a] FIG. 10. Biolayer interferometry analysis of sensorgram data for HP503 binding to HER2. [Figure 28b] FIG. 10. Biolayer interferometry analysis of sensorgram data for HP507 binding to HER2. [Figure 28c] FIG. 10. Biolayer interferometry analysis of sensorgram data for binding of HP511 to HER2. [Fig. 28d] FIG. 10: Biolayer interferometry analysis of sensorgram data for binding of HP515 to HER2 protein using biolayer interferometry analysis. [Figure 29a] FIG. 13. Biolayer interferometry analysis of sensorgram data for HP503 binding to Fcγ receptor 1, Fcγ receptor 2A (131R isoform), and Fcγ receptor 3A (176V isoform). [Figure 29b]FIG. 13. Biolayer interferometry analysis of sensorgram data for HP507 binding to Fcγ receptor 1, Fcγ receptor 2A (131R isoform), and Fcγ receptor 3A (176V isoform). [Figure 29c] FIG. 13. Biolayer interferometry analysis of sensorgram data for HP511 binding to Fcγ receptor 1, Fcγ receptor 2A (131R isoform), and Fcγ receptor 3A (176V isoform). [Figure 29d] FIG. 13. Biolayer interferometry analysis of sensorgram data for HP515 binding to Fcγ receptor 1, Fcγ receptor 2A (131R isoform), and Fcγ receptor 3A (176V isoform). [Figure 30a] FIG. 13. Biolayer interferometry analysis of sensorgram data for binding of HP503, HP507, HP511, and HP515 to the neonatal Fc receptor (FcRn). [Figure 30b] FIG. 10: Biolayer interferometry analysis of sensorgram data for binding of human IgG1, trastuzumab, pertuzumab, and marjetuximab to the neonatal Fc receptor (FcRn). [Figure 31a] FIG. 1 illustrates the CDC activity of anti-HER2 antibodies in the BT474 breast cancer cell line. [Figure 31b] FIG. 1 illustrates the CDC activity of anti-HER2 antibodies in the NCI-N87 gastric cancer cell line. [Figure 32a] FIG. 1 illustrates the ADCC activity of anti-HER2 antibodies in NCI-N87 gastric cancer cell lines. [Figure 32b] FIG. 1 illustrates the ADCC activity of anti-HER2 antibodies in MDA-MB-453 breast cancer cell lines. [Fig. 32c] FIG. 1 illustrates the ADCC activity of anti-HER2 antibodies in the SNU-601 gastric cancer cell line. [Fig. 32d] FIG. 1 illustrates the ADCC activity of anti-HER2 antibodies in the SNU-5 gastric cancer cell line. [Figure 33a]FIG. 1 illustrates the anti-tumor activity of anti-HER2 antibodies in a CB-17 SCID mouse model of tumor xenografts of the SNU-5 gastric cancer cell line. [Figure 33b] FIG. 1 illustrates the anti-tumor activity of anti-HER2 antibodies in a Balb / c-nude mouse model of tumor xenografts of the SNU-5 gastric cancer cell line. [Figure 33c] 1 is a table showing the results of Example 15. [Diagram 34] FIG. 1 illustrates the anti-tumor activity of anti-HER2 antibodies in a mouse model of tumor xenografts of the SNU-601 gastric cancer cell line. [Diagram 35] FIG. 1 illustrates the anti-tumor activity of anti-HER2 antibodies in a mouse model of tumor xenografts of the NCI-N87 gastric cancer cell line. [Diagram 36] 1 is a diagram of a vector for expressing human HER2 protein in mammalian cells. [Figure 37] FIG. 1 illustrates flow cytometry quantification of HER2 expression in a CT26 mouse colon cancer cell line clone expressing human HER2 (CT26-HER2). [Figure 38] FIG. 1 illustrates the stability of human HER2 expression in the CT26-HER2 clone. [Figure 39] FIG. 1 illustrates the relative expression of human HER2 in the CT26-HER2 cell line (clone name: #2-60) compared to human cancer cell lines, and shows that H01 allows a greater amount of Fc domain to bind to the surface of CT26-HER2 cells compared to trastuzumab. [Diagram 40] FIG. 1 illustrates the anti-tumor activity of anti-HER2 antibodies in a syngeneic CT26-HER2 mouse tumor model. [Figure 41a] FIG. 1 is a schematic diagram of a monovalent engineered mAb according to one embodiment. [Figure 41b] FIG. 1 is a schematic diagram of a biparatopic engineered mAb according to one embodiment. [Diagram 42]42A-4C illustrate sensorgram data for the binding of a fusion protein to a target, according to one embodiment. Specifically, FIG. 42(A) illustrates sensorgram data for the binding of GPM01, a monovalent engineered mAb targeting GPC-3, to human GPC-3. FIG. 42(B) illustrates sensorgram data for the binding of GPM02, a monovalent engineered mAb targeting GPC-3, to human GPC-3. FIG. 42(C) illustrates sensorgram data for the binding of GPM04, a monovalent engineered mAb targeting GPC-3, to human GPC-3. FIG. 42(D) illustrates sensorgram data for the binding of GPB01, a biparatopic engineered mAb targeting GPC-3, to human GPC-3. FIG. 42(E) illustrates sensorgram data for the binding of GPB03, a biparatopic engineered mAb targeting GPC-3, to human GPC-3. Figure 42(F) illustrates sensorgram data for the binding of GPB04, a biparatopic engineered mAb targeting GPC-3, to human GPC-3. Figure 42(G) illustrates sensorgram data for the binding of GPB06, a biparatopic engineered mAb targeting GPC-3, to human GPC-3. [Diagram 43] FIG. 1 illustrates flow cytometry analysis of the amount of Fc domains present on the surface of the HepG2 liver cancer cell line upon treatment with 100 nM GPC-3 antibody. [Diagram 44] FIG. 1 illustrates an SDS-PAGE analysis showing inhibition of AKT phosphorylation in PC-3 prostate cancer cell lines upon treatment with 50 nM EphA2 antibody. [Diagram 45] FIG. 1 illustrates flow cytometry analysis of the amount of Fc domains present on the surface of the PC-3 prostate cancer cell line upon treatment with 100 nM EphA2 antibody. [Figure 46]FIG. 1 illustrates biolayer interferometry analysis of sensorgram data for binding of MET-targeting monovalent engineered mAbs, MEM01 and MEM06, to human MET. [Figure 47] FIG. 1 illustrates flow cytometry analysis of the amount of Fc domains present on the surface of MKN45 gastric cancer cell lines upon treatment with the indicated concentrations of MET antibody. [Figure 48] FIG. 1 illustrates flow cytometry analysis of the amount of Fc domains present on the surface of SNU5 gastric cancer cell lines upon treatment with the indicated concentrations of antibodies. [Figure 49] FIG. 1 illustrates biolayer interferometry analysis of sensorgram data for binding of EGFR-targeting monovalent engineered mAbs to human EGFR. [Figure 50] FIG. 1 illustrates biolayer interferometry analysis of sensorgram data for binding of CD33-targeting monovalent engineered mAbs 33-1, 33-2, and 33-3, and CD33-targeting biparatopic engineered mAbs 33-4, 33-5, 33-6, and 33-7 to human CD33. [Figure 51] FIG. 1 illustrates biolayer interferometry analysis of sensorgram data for binding of CEACAM5-targeting monovalent engineered mAbs to human CEACAM5. [Figure 52] 52A-52C illustrate biolayer interferometry analysis of sensorgram data for binding of a fusion protein according to one embodiment to a target. Specifically, FIG. 52(A) illustrates sensorgram data for binding of T01, a monovalent engineered mAb targeting TROP2, to human TROP2. FIG. 52(B) illustrates sensorgram data for binding of MSM01, a monovalent engineered mAb targeting mesothelin, to human mesothelin. FIG. 52(C) illustrates sensorgram data for binding of LIM01, a monovalent engineered mAb targeting LIV-1, to human LIV-1. A: T01, B: MSM01, C: LIM01. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] Definition of Terms As used herein, the term "fusion protein with two Fc's" or "antibody with two Fc's" refers to a fusion protein in which two Fc domains are independently attached to two polypeptide chains that constitute the antigen-binding site. The two polypeptide chains that constitute the antigen-binding site may be different from each other. For example, one of the two polypeptide chains that constitute the antigen-binding site may be a sequence that includes or consists of an antibody light chain CDR sequence or a light chain variable region, or may be an scFv, and the other may be a sequence that includes or consists of an antibody heavy chain CDR sequence or a heavy chain variable region, or may be an scFv. In one embodiment, the fusion protein with two Fc regions may include a sequence of a "humanized" form of a non-human antibody, which is a chimeric antibody that includes a human immunoglobulin that includes a natural CDR. In addition, the fusion protein may include a "fully human antibody" or a portion of a "human antibody". In addition, in one embodiment, the multispecific fusion protein or the antigen-binding domain may be a "monoclonal antibody" or a portion thereof.

[0020] As used herein, the term "antibody" refers to a substance that specifically binds to an antigen and causes an antigen-antibody reaction. In addition, an antibody is also referred to as an immunoglobulin. The antibody may refer to any one selected from IgG, IgE, IgM, IgD, and IgA, and may be a subclass of IgG, such as IgG1, IgG2, IgG3, IgG4, IgA1, or IgA2. In addition, the antibody may be an agonist antibody or an antagonist antibody.

[0021] As used herein, the term "Fab" or "Fab region" refers to the region of an antibody that binds to an antigen. Conventional IgG generally contains two Fab regions. Each Fab region typically consists of one variable region and one constant region of each heavy and light chain. Specifically, the variable and constant regions of the heavy chain in the Fab region are the VH and CH1 regions, and the variable and constant regions of the light chain in the Fab region are the VL and CL regions. The VH, CH1, VL, and CL of the Fab region may be arranged in various ways to confer antigen binding ability according to the present disclosure, including CrossMab Fab technology, in which the VH and VL have a permuted arrangement relative to each other.

[0022] As used herein, the term "heavy chain" refers to a polypeptide chain of about 50 kDa to about 70 kDa, where the N-terminal portion contains a variable region of at least about 120 to 130 amino acids, and the C-terminal portion contains a constant region. The constant region may be one of five types: alpha (α), delta (δ), epsilon (ε), gamma (γ), and mu (μ), where α, δ, and γ contain about 450 amino acids, and μ and ε contain about 550 amino acids.

[0023] As used herein, the term "light chain" refers to a polypeptide chain of about 25 kDa, where the N-terminal portion contains a variable region of at least about 100 to about 110 amino acids, and the C-terminal portion contains a constant region. There are two types of light chain constant domains: kappa (κ) or lambda (λ). In addition, the constant region of the light chain is referred to as "CL". The heavy chain C domains (CH domains) are numbered from the N-terminus to the C-terminus (e.g., CH1, CH2, CH3, etc.). The CL and CH1 regions of any of these antibody classes may be used in the present disclosure. In certain embodiments, the CL and CH1 regions provided herein are of the IgG type (e.g., IgG1).

[0024] As used herein, the term "Fc" or "Fc region" refers to the C-terminal region of an immunoglobulin heavy chain, including native Fc regions, recombinant Fc regions, and variant Fc regions. Thus, Fc refers to the last two constant region immunoglobulin domains of IgA, IgD, and IgG, the last three constant region immunoglobulin domains of IgE and IgM, and the hinge present at the N-terminus of these domains. For IgA and IgM, Fc may include the J chain. For IgG, Fc includes immunoglobulin domains Cy2 (CH2) and Cy3 (CH3), and the hinge between Cy1 and Cy2. Although the interface of the Fc region may vary, the human IgG heavy chain Fc region is generally defined as including residues C226, P230, or A231 at the C-terminus, with numbering according to the EU index. As used herein, "Fc polypeptide" or "Fc-derived polypeptide" refers to a polypeptide that includes all or a portion of an Fc region. In one embodiment, the variant Fc region may be in a form in which at least one amino acid, for example about 1 to about 10 amino acids, or about 1 to about 5 amino acids, are substituted compared to a native sequence Fc region. In addition, the variant Fc region may have at least about 80% homology, at least about 90% homology, or at least about 95% homology to the native sequence Fc region.

[0025] As used herein, the term "Fv" or "Fv fragment" or "Fv region" is a polypeptide comprising the VL and VH domains of a single antibody.

[0026] As used herein, the term "single-chain Fv" or "scFv" refers to an antibody fragment comprising the VH and VL domains of an antibody within a single polypeptide chain.

[0027] As used herein, the term "variable region" refers to the region of an antibody that comprises one or more immunoglobulin domains encoded by either the VL (including Vkappa (VK) and Vlambda (VL)) and / or VH genes that constitute the light (including kappa and lambda) and heavy chain immunoglobulin domain loci, respectively. A light or heavy chain variable region (VL or VH) consists of a "framework" or "FR" region that comprises three hypervariable regions called "complementarity determining regions" or "CDRs". As used herein, the term "antigen" refers to a structure that can selectively bind to an antibody. Target antigens may be polypeptides, carbohydrates, nucleic acids, lipids, haptens, or other naturally occurring or synthetic compounds. In particular, antigens are polypeptides and may be proteins present on or in cells.

[0028] As used herein, the term "epitope" refers to an antigenic determinant, which is the part on an antigen to which an antibody or polypeptide binds. A protein epitope may include amino acid residues that are directly involved in binding as well as amino acid residues that are effectively blocked by a specific antigen-binding antibody or peptide. An epitope is the simplest form or smallest structural region of a complex antigen molecule that can bind to an antibody or receptor. Epitopes may be linear or structural / conformational.

[0029] As used herein, the term "vector" refers to a material for transporting or expressing nucleic acid sequences, including nucleic acid sequences encoding the multispecific fusion proteins (e.g., antibodies) described herein. Specifically, vectors include expression vectors, plasmids, phage vectors, viral vectors, episomes, and artificial chromosomes.

[0030] As used herein, the term "polynucleotide," also referred to as "nucleic acid," refers to a polymer of nucleotides of any length. Specifically, a polynucleotide may be DNA or RNA. Antibodies containing two Fc domains In one embodiment of the present invention, there is provided an antibody comprising multiple Fc domains, characterized in that the ratio of antigen-binding sites and Fc domains is 1: 2 or 2: 2. Specifically, the antibody may be a fusion protein comprising an antigen-binding site, a first Fc domain or a variant thereof, and a second Fc domain or a variant thereof.

[0031] Here, the antigen-binding site may consist of two different polypeptide chains, each of which may in addition be linked to a first Fc domain or a variant thereof, and a second Fc domain or a variant thereof.

[0032] Here, in one embodiment of the fusion protein, when the antigen-binding site includes a Fab, the fusion protein may be a fusion protein in which two Fc domains are linked to the C-terminus of the CH1 region of the heavy chain and the C-terminus of the constant region of the light chain, respectively. In addition, the Fc domains and the Fab may be linked via a peptide linker.

[0033] In addition, in one embodiment of the fusion protein, when the antigen-binding site is an Fv, the fusion protein may be a fusion protein in which two Fc domains are linked to the C-terminus of the heavy chain variable region and the C-terminus of the light chain variable region, respectively. In addition, the Fc domains and the Fv may be linked through a peptide linker.

[0034] This novel antibody format or structure has a molecular weight similar to that of human IgG. In addition, the fusion protein can have an antigen binding affinity equivalent to that of a human IgG-based antibody. However, the antibody format allows the Fc domain to be up to four times more present on cell surface antigens compared to native human antibodies. Due to these characteristics, the fusion protein can have increased affinity for Fcγ receptors and increased effector function compared to wild-type antibodies. Although each Fc domain attached to the fusion protein can have a similar level of Fc receptor (Fcγ receptor and FcRn) binding affinity as the Fc domain of an IgG-based antibody, due to avidity effects, the apparent binding affinity (apparent affinity) of the fusion protein to the Fc receptor (Fcγ receptor and FcRn) can be significantly increased compared to a human IgG antibody. In addition, the fusion protein has a similar level of thermal stability to that of an IgG-based antibody.

[0035] Specifically, the fusion protein may be a fusion protein comprising: (a) an antigen-binding site consisting of a first polypeptide comprising at least one complementarity determining region (CDR) sequence and a second polypeptide comprising at least one complementarity determining region (CDR) sequence, wherein the first and second polypeptides form a dimer, and wherein the antigen-binding site can specifically bind to a target antigen; (b) a first Fc domain or a variant thereof, which is a dimer consisting of two polypeptide sequences, one of which is joined to the first polypeptide of the antigen-binding site; and (c) a second Fc domain or a variant thereof, which is a dimer consisting of two polypeptide sequences, one of which is joined to the second polypeptide of the antigen-binding site.

[0036] Here, the first polypeptide of the antigen-binding site may comprise CDR1, CDR2, and CDR3 of an antibody heavy chain, and the second polypeptide of the antigen-binding site may comprise CDR1, CDR2, and CDR3 of an antibody light chain. In addition, the first polypeptide of the antigen-binding site may further comprise the CH1 region of an antibody heavy chain and / or the second polypeptide of the antigen-binding site may further comprise the constant region of an antibody light chain.

[0037] The specific structure of the fusion protein is described in more detail below.

[0038] antigen binding site Here, the antigen-binding site can specifically bind to a protein expressed on the cell surface. Specifically, the antigen-binding site can specifically bind to a cancer antigen.

[0039] In one embodiment, the antigen binding site is selected from the group consisting of PD-L1, EGFR, EGFRvIII, BCMA, CD22, CD25, CD30, CD33, CD37, CD38, CD52, CD56, CD123, c-Met (MET), DLL3, DR4, DR5, GD2, nectin-4, RANKL, SLAMF7, Trop-2, LIV-1, claudin 18.2, IL13α2, CD3, HER2, HER3, It can specifically bind to any one selected from the group consisting of FGFR2, FGFR3, GPC3, ROR1, Folα, CD20, CD19, CTLA-4, VEGFR, NCAM1, ICAM-1, ICAM-2, CEACAM5, CEACAM6, carcinoembryonic antigen (CEA), CA-125, alpha-fetoprotein (AFP), MUC-1, MUC-16, PSMA, PSCA, epithelial tumor antigen (ETA), melanoma-associated antigen (MAGE), immature laminin receptor, TAG-72, HPV E6 / E7, BING-4, calcium-activated chloride channel 2, cyclin-B1, 9D7, Ep-CAM, EphA2, EphA3, mesothelin, SAP-1, survivin, and virus-derived antigens.

[0040] The second antigen-binding site may also specifically bind to any one of the antigens selected from the above group. According to one embodiment, the antigen that the first antigen-binding site binds to may be different from the antigen that the second antigen-binding site binds to. For example, the first antigen-binding site may comprise a sequence that specifically binds to HER2, and the second antigen-binding site may comprise a sequence that specifically binds to EGFR. In another embodiment, the first antigen-binding site may comprise a sequence that specifically binds to one epitope of an antigen, and the second antigen-binding site may comprise a sequence that specifically binds to a different epitope of the same antigen.

[0041] Specific examples of antigen-binding sites Here, the antigen-binding site may include a variable region that specifically binds to an antigen. Specifically, the variable region may be selected from the group consisting of cetuximab, panitumumab, necitumumab, imgatuzumab, depatuxizumab, losatuxizumab, etevritamab, AMG-595, atezolizumab, avelumab, durvalumab, trastuzumab, pertuzumab, onartuzumab, emibetuzumab, terisotuzumab, datopotamab, sacituzumab, rovalpituzumab, tarulata, and the like. The antibody may include, but is not limited to, the heavy chain variable region and the light chain variable region of any one of the antibodies selected from the group consisting of mab, belantamab, ladiratuzumab, codrituzumab, aprutumab, bemarituzumab, bofatamab, ramucirumab, rituximab, obinutuzumab, daratumumab, and 1C1 (clone name).

[0042] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to EGFR. For example, the antigen-binding site may include a heavy chain variable region that includes H-CDR1 represented by SEQ ID NO: 175, H-CDR2 represented by SEQ ID NO: 176, and H-CDR3 represented by SEQ ID NO: 177 of cetuximab, and may include a light chain variable region that includes L-CDR1 represented by SEQ ID NO: 178, L-CDR2 represented by SEQ ID NO: 179, and L-CDR3 represented by SEQ ID NO: 180. As another example, the antigen-binding site may include a heavy chain variable region that includes H-CDR1 represented by SEQ ID NO: 181, H-CDR2 represented by SEQ ID NO: 182, and H-CDR3 represented by SEQ ID NO: 183 of panitumumab, and may include a light chain variable region that includes L-CDR1 represented by SEQ ID NO: 184, L-CDR2 represented by SEQ ID NO: 185, and L-CDR3 represented by SEQ ID NO: 186. As another example, the antigen-binding site may comprise a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 187, H-CDR2 represented by SEQ ID NO: 188, and H-CDR3 represented by SEQ ID NO: 189 of necitumumab, and a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 190, L-CDR2 represented by SEQ ID NO: 191, and L-CDR3 represented by SEQ ID NO: 192. In addition, the antigen-binding site may comprise a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 193, H-CDR2 represented by SEQ ID NO: 194, and H-CDR3 represented by SEQ ID NO: 195 of imgatuzumab, and a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 196, L-CDR2 represented by SEQ ID NO: 197, and L-CDR3 represented by SEQ ID NO: 198. In addition, the antigen-binding site may comprise a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 199, H-CDR2 represented by SEQ ID NO: 200, and H-CDR3 represented by SEQ ID NO: 201 of depatuxizumab, and may comprise a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 202, L-CDR2 represented by SEQ ID NO: 203, and L-CDR3 represented by SEQ ID NO: 204.In addition, the antigen-binding site may comprise a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 199, H-CDR2 represented by SEQ ID NO: 205, and H-CDR3 represented by SEQ ID NO: 206 of rosatuxizumab, and may comprise a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 202, L-CDR2 represented by SEQ ID NO: 203, and L-CDR3 represented by SEQ ID NO: 204.

[0043] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to EGFRvIII. For example, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 207, H-CDR2 represented by SEQ ID NO: 208, and H-CDR3 represented by SEQ ID NO: 209 of eteburitamab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 210, L-CDR2 represented by SEQ ID NO: 211, and L-CDR3 represented by SEQ ID NO: 212. In addition, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 213, H-CDR2 represented by SEQ ID NO: 214, and H-CDR3 represented by SEQ ID NO: 215 of AMG-595, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 210, L-CDR2 represented by SEQ ID NO: 216, and L-CDR3 represented by SEQ ID NO: 217.

[0044] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to PD-L1. For example, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 218, H-CDR2 represented by SEQ ID NO: 219, and H-CDR3 represented by SEQ ID NO: 220 of atezolizumab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 221, L-CDR2 represented by SEQ ID NO: 222, and L-CDR3 represented by SEQ ID NO: 223. In addition, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 224, H-CDR2 represented by SEQ ID NO: 225, and H-CDR3 represented by SEQ ID NO: 226 of avelumab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 227, L-CDR2 represented by SEQ ID NO: 228, and L-CDR3 represented by SEQ ID NO: 229. In addition, the antigen-binding site may comprise a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 230, H-CDR2 represented by SEQ ID NO: 231, and H-CDR3 represented by SEQ ID NO: 232 of durvalumab, and may comprise a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 233, L-CDR2 represented by SEQ ID NO: 234, and L-CDR3 represented by SEQ ID NO: 235.

[0045] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to HER2. For example, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 21, H-CDR2 represented by SEQ ID NO: 22, and H-CDR3 represented by SEQ ID NO: 23 of trastuzumab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 24, L-CDR2 represented by SEQ ID NO: 25, and L-CDR3 represented by SEQ ID NO: 26. In addition, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 33, H-CDR2 represented by SEQ ID NO: 34, and H-CDR3 represented by SEQ ID NO: 35 of pertuzumab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 36, L-CDR2 represented by SEQ ID NO: 37, and L-CDR3 represented by SEQ ID NO: 38.

[0046] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to c-Met. For example, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 236, H-CDR2 represented by SEQ ID NO: 237, and H-CDR3 represented by SEQ ID NO: 238 of onartuzumab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 239, L-CDR2 represented by SEQ ID NO: 240, and L-CDR3 represented by SEQ ID NO: 241. In addition, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 242, H-CDR2 represented by SEQ ID NO: 243, and H-CDR3 represented by SEQ ID NO: 244 of emibetuzumab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 245, L-CDR2 represented by SEQ ID NO: 246, and L-CDR3 represented by SEQ ID NO: 247. In addition, the antigen-binding site may comprise a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 248, H-CDR2 represented by SEQ ID NO: 249, and H-CDR3 represented by SEQ ID NO: 250 of terisotuzumab, and may comprise a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 251, L-CDR2 represented by SEQ ID NO: 252, and L-CDR3 represented by SEQ ID NO: 253.

[0047] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to Trop-2. For example, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 254, H-CDR2 represented by SEQ ID NO: 255, and H-CDR3 represented by SEQ ID NO: 256 of datopotamab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 257, L-CDR2 represented by SEQ ID NO: 258, and L-CDR3 represented by SEQ ID NO: 259. In addition, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 260, H-CDR2 represented by SEQ ID NO: 261, and H-CDR3 represented by SEQ ID NO: 262 of sacituzumab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 263, L-CDR2 represented by SEQ ID NO: 264, and L-CDR3 represented by SEQ ID NO: 265.

[0048] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to DLL3. For example, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 266, H-CDR2 represented by SEQ ID NO: 267, and H-CDR3 represented by SEQ ID NO: 268 of rovalpituzumab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 269, L-CDR2 represented by SEQ ID NO: 270, and L-CDR3 represented by SEQ ID NO: 271. In addition, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 272, H-CDR2 represented by SEQ ID NO: 273, and H-CDR3 represented by SEQ ID NO: 274 of tarlatamab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 275, L-CDR2 represented by SEQ ID NO: 276, and L-CDR3 represented by SEQ ID NO: 277.

[0049] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to BCMA. For example, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 278, H-CDR2 represented by SEQ ID NO: 279, and H-CDR3 represented by SEQ ID NO: 280 of belantamab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 281, L-CDR2 represented by SEQ ID NO: 282, and L-CDR3 represented by SEQ ID NO: 283.

[0050] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to LIV-1. For example, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 284, H-CDR2 represented by SEQ ID NO: 285, and H-CDR3 represented by SEQ ID NO: 286 of lajiratuzumab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 287, L-CDR2 represented by SEQ ID NO: 288, and L-CDR3 represented by SEQ ID NO: 289.

[0051] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to GPC-3. For example, the antigen-binding site may include a heavy chain variable region that includes H-CDR1 represented by SEQ ID NO: 99, H-CDR2 represented by SEQ ID NO: 100, and H-CDR3 represented by SEQ ID NO: 101 of codrituzumab, and may include a light chain variable region that includes L-CDR1 represented by SEQ ID NO: 102, L-CDR2 represented by SEQ ID NO: 103, and L-CDR3 represented by SEQ ID NO: 104.

[0052] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to FGFR2. For example, the antigen-binding site may include a heavy chain variable region that includes H-CDR1 represented by SEQ ID NO: 290, H-CDR2 represented by SEQ ID NO: 291, and H-CDR3 represented by SEQ ID NO: 292 of apurutumab, and may include a light chain variable region that includes L-CDR1 represented by SEQ ID NO: 293, L-CDR2 represented by SEQ ID NO: 294, and L-CDR3 represented by SEQ ID NO: 295. In addition, the antigen-binding site may include a heavy chain variable region that includes H-CDR1 represented by SEQ ID NO: 296, H-CDR2 represented by SEQ ID NO: 297, and H-CDR3 represented by SEQ ID NO: 298 of bemarituzumab, and may include a light chain variable region that includes L-CDR1 represented by SEQ ID NO: 299, L-CDR2 represented by SEQ ID NO: 300, and L-CDR3 represented by SEQ ID NO: 301.

[0053] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to FGFR3. For example, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 302, H-CDR2 represented by SEQ ID NO: 303, and H-CDR3 represented by SEQ ID NO: 304 of vofatamab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 305, L-CDR2 represented by SEQ ID NO: 306, and L-CDR3 represented by SEQ ID NO: 307.

[0054] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to VEGFR2. For example, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 308, H-CDR2 represented by SEQ ID NO: 309, and H-CDR3 represented by SEQ ID NO: 310 of ramucirumab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 311, L-CDR2 represented by SEQ ID NO: 312, and L-CDR3 represented by SEQ ID NO: 313.

[0055] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to CD20. For example, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 314, H-CDR2 represented by SEQ ID NO: 315, and H-CDR3 represented by SEQ ID NO: 316 of rituximab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 317, L-CDR2 represented by SEQ ID NO: 318, and L-CDR3 represented by SEQ ID NO: 319. For example, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 320, H-CDR2 represented by SEQ ID NO: 321, and H-CDR3 represented by SEQ ID NO: 322 of obinutuzumab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 323, L-CDR2 represented by SEQ ID NO: 324, and L-CDR3 represented by SEQ ID NO: 325.

[0056] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to CD38. For example, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 326, H-CDR2 represented by SEQ ID NO: 327, and H-CDR3 represented by SEQ ID NO: 328 of daratumumab, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 329, L-CDR2 represented by SEQ ID NO: 330, and L-CDR3 represented by SEQ ID NO: 331.

[0057] As a specific example of the present invention, the example may include an antigen-binding site that specifically binds to EphA2. For example, the antigen-binding site may include a heavy chain variable region comprising H-CDR1 represented by SEQ ID NO: 157, H-CDR2 represented by SEQ ID NO: 158, and H-CDR3 represented by SEQ ID NO: 159 of 1C1, and may include a light chain variable region comprising L-CDR1 represented by SEQ ID NO: 160, L-CDR2 represented by SEQ ID NO: 161, and L-CDR3 represented by SEQ ID NO: 162.

[0058] Table 1 below shows the CDR sequences of non-limiting exemplary antibodies with anti-cancer efficacy that may be used in specific embodiments of the present invention.

[0059] [Table 1] JPEG2024539509000003.jpg198149 JPEG2024539509000004.jpg198149 JPEG2024539509000005.jpg214149

[0060] antigen Examples of antigens to which the antigen-binding sites described herein can specifically bind include, but are not limited to, the following substances:

[0061] "Epidermal Growth Factor Receptor (EGFR)": EGFR is a cell membrane receptor that regulates cell growth, division, survival, and death. In various cancers, expression of EGFR increases in tumor tissue. Tumor tissues with increased EGFR are known to be aggressive, metastatic, and highly resistant to anti-cancer drugs. In one embodiment, the substance that inhibits EGFR may be, but is not limited to, cetuximab, panitumumab, necitumumab, imgatuzumab, depatuxizumab, or rosatuxizumab.

[0062] "Epidermal growth factor receptor variant 3 (EGFRvIII)": EGFRvIII is a mutation in which exons 2 to 7 of EGFR are deleted. EGFRvIII is mainly reported in glioblastoma multiforme, and most patients with EGFRvIII-positive mutations have poor prognosis. In one embodiment, the substance that inhibits EGFRvIII may be, but is not limited to, cetuximab, panitumumab, necitumumab, imgatuzumab, depatuxizumab, rosatuxizumab, eteburitamab, or AMG-595.

[0063] "Programmed Death-Ligand 1 (PD-L1)": PD-L1 is a protein that is overexpressed on the surface of cancer cells. PD-L1 is a major cancer-specific antigen that plays an important role in inducing T cell exhaustion and apoptosis in cancer cells and in obtaining immune tolerance. In one embodiment, the PD-L1-targeting anti-cancer antibody may be, but is not limited to, atezolizumab, avelumab, and durvalumab.

[0064] "HER-2 / neu (human epidermal growth factor receptor 2)": HER-2 regulates cell proliferation through activation of PI3K / AkT. HER-2 is known to be overexpressed in metastatic breast cancer and ovarian cancer, etc., and to induce resistance to anticancer drugs. The HER2 / neu targeted anticancer drug may be, but is not limited to, trastuzumab or pertuzumab.

[0065] "c-Met (mesenchymal epithelial transition factor)": c-Met is a hepatocyte growth factor receptor. Amplification or mutation of c-Met is frequently reported in cancer cells and is known to promote tumor growth, metastasis, and malignancy. Specifically, inhibitors of the protein may be, but are not limited to, onartuzumab, emibetuzumab, or terisotuzumab.

[0066] "Trop-2 (tumor-associated calcium signal transducer 2)": Trop-2 is a cellular glycoprotein involved in cancer cell growth and proliferation. Trop-2 is known to be specifically overexpressed in non-small cell lung cancer and breast cancer. Specifically, antibodies targeting the protein may be, but are not limited to, datopotamab or sacituzumab.

[0067] "DLL3 (Delta-Like Ligand 3)": DLL3 is a major cancer target antigen known to be expressed in approximately 85% of small cell lung cancer patients. Specifically, the antibody targeting the protein may be, but is not limited to, rovalpituzumab or talutamab.

[0068] "BCMA (B-cell maturation antigen)": BCMA is a key factor in the survival and proliferation of myeloma cells and is a clinically proven target for the treatment of multiple myeloma. Specifically, but not limited to, an antibody targeting the protein may be belantamab.

[0069] "LIV-1 (Zinc Transporter ZIP6)": LIV-1 is a highly cancer-specific antigen that is overexpressed in breast cancer. Specifically, the antibody that targets this protein may be, but is not limited to, lazirituzumab.

[0070] "GPC-3 (Glypican-3)": GPC-3 is a highly cancer-specific antigen that is specifically overexpressed in liver cancer.Specifically, the antibody that targets this protein can be, but is not limited to, codrituzumab.

[0071] "FGFR (fibroblast growth factor receptor)": FGFR is a receptor for fibroblast growth factor (FGF) that regulates various biological processes including cell growth, differentiation, and migration. FGFR genes are easily mutated, and these variants are commonly observed in breast cancer, uterine cancer, ovarian cancer, cervical cancer, etc. Four FGFR genes are composed of seven signaling receptors, of which FGFR2 and FGFR3 are highly cancer-specific antigens. Antibodies targeting FGFR2 or FGFR3 may be, but are not limited to, apurutumab, bemarituzumab, or vofatamab.

[0072] "Vascular endothelial growth factor receptor (VEGFR)": VEGFR is a cell membrane receptor for vascular endothelial growth factor, which induces angiogenesis. VEGFR inhibitors inhibit tumor growth and metastasis by inhibiting angiogenesis. In one embodiment, the VEGFR inhibitor can be, but is not limited to, ramucirumab.

[0073] "CD20 (B-lymphocyte antigen CD20)": CD20 is a protein expressed on the surface of B cells that is used as a target protein for the treatment of B-cell lymphoma. CD20 inhibitors can be, but are not limited to, rituximab or obinutuzumab.

[0074] "CD38 (cluster of differentiation 38)": CD38 is a protein that acts as a signal transduction receptor in immune cells and regulates cell proliferation and death. An inhibitor targeting the protein may be, but is not limited to, daratumumab.

[0075] "EphA2 (EPH receptor A2)": EphA2 is overexpressed in cancer cells and has a significant impact on the growth and metastasis of cancer cells. The antibody targeting this protein may be, but is not limited to, 1C1.

[0076] An antigen binding site which specifically binds to the antigens exemplified above may comprise the CDR sequences exemplified below. EGFR: 1) a VH region (SEQ ID NO: 334) comprising the amino acid sequences of SEQ ID NO: 175 (VH-CDR1), SEQ ID NO: 176 (VH-CDR2), and SEQ ID NO: 177 (VH-CDR3), and a VL region (SEQ ID NO: 335) comprising the amino acid sequences of SEQ ID NO: 178 (VL-CDR1), SEQ ID NO: 179 (VL-CDR2), and SEQ ID NO: 180 (VL-CDR3); 2) a VH region (SEQ ID NO: 336) comprising the amino acid sequences of SEQ ID NO: 181 (VH-CDR1), SEQ ID NO: 182 (VH-CDR2), and SEQ ID NO: 183 (VH-CDR3), and a VL region (SEQ ID NO: 337) comprising the amino acid sequences of SEQ ID NO: 184 (VL-CDR1), SEQ ID NO: 185 (VL-CDR2), and SEQ ID NO: 186 (VL-CDR3); 3) a VH region (SEQ ID NO: 338) comprising the amino acid sequences of SEQ ID NO: 187 (VH-CDR1), SEQ ID NO: 188 (VH-CDR2), and SEQ ID NO: 189 (VH-CDR3), and a VL region (SEQ ID NO: 339) comprising the amino acid sequences of SEQ ID NO: 190 (VL-CDR1), SEQ ID NO: 191 (VL-CDR2), and SEQ ID NO: 192 (VL-CDR3); 4) a VH region (SEQ ID NO: 340) comprising the amino acid sequences of SEQ ID NO: 193 (VH-CDR1), SEQ ID NO: 194 (VH-CDR2), and SEQ ID NO: 195 (VH-CDR3), and a VL region (SEQ ID NO: 341) comprising the amino acid sequences of SEQ ID NO: 196 (VL-CDR1), SEQ ID NO: 197 (VL-CDR2), and SEQ ID NO: 198 (VL-CDR3); 5) a VH region (SEQ ID NO: 342) comprising the amino acid sequences of SEQ ID NO: 199 (VH-CDR1), SEQ ID NO: 200 (VH-CDR2), and SEQ ID NO: 201 (VH-CDR3), and a VL region (SEQ ID NO: 343) comprising the amino acid sequences of SEQ ID NO: 202 (VL-CDR1), SEQ ID NO: 203 (VL-CDR2), and SEQ ID NO: 204 (VL-CDR3); 6) a VH region (SEQ ID NO: 344) comprising the amino acid sequences of SEQ ID NO: 199 (VH-CDR1), SEQ ID NO: 205 (VH-CDR2), and SEQ ID NO: 206 (VH-CDR3), and a VL region (SEQ ID NO: 345) comprising the amino acid sequences of SEQ ID NO: 202 (VL-CDR1), SEQ ID NO: 203 (VL-CDR2), and SEQ ID NO: 204 (VL-CDR3);

[0077] EGFRvIII: 7) a VH region (SEQ ID NO: 346) comprising the amino acid sequences of SEQ ID NO: 207 (VH-CDR1), SEQ ID NO: 208 (VH-CDR2), and SEQ ID NO: 209 (VH-CDR3), and a VL region (SEQ ID NO: 347) comprising the amino acid sequences of SEQ ID NO: 210 (VL-CDR1), SEQ ID NO: 211 (VL-CDR2), and SEQ ID NO: 212 (VL-CDR3); 8) a VH region (SEQ ID NO: 348) comprising the amino acid sequences of SEQ ID NO: 213 (VH-CDR1), SEQ ID NO: 214 (VH-CDR2), and SEQ ID NO: 215 (VH-CDR3), and a VL region (SEQ ID NO: 349) comprising the amino acid sequences of SEQ ID NO: 210 (VL-CDR1), SEQ ID NO: 216 (VL-CDR2), and SEQ ID NO: 217 (VL-CDR3);

[0078] PD-L1: 9) a VH region (SEQ ID NO: 350) comprising the amino acid sequences of SEQ ID NO: 218 (VH-CDR1), SEQ ID NO: 219 (VH-CDR2), and SEQ ID NO: 220 (VH-CDR3), and a VL region (SEQ ID NO: 351) comprising the amino acid sequences of SEQ ID NO: 221 (VL-CDR1), SEQ ID NO: 222 (VL-CDR2), and SEQ ID NO: 223 (VL-CDR3); 10) a VH region (SEQ ID NO: 352) comprising the amino acid sequences of SEQ ID NO: 224 (VH-CDR1), SEQ ID NO: 225 (VH-CDR2), and SEQ ID NO: 226 (VH-CDR3), and a VL region (SEQ ID NO: 353) comprising the amino acid sequences of SEQ ID NO: 227 (VL-CDR1), SEQ ID NO: 228 (VL-CDR2), and SEQ ID NO: 229 (VL-CDR3); 11) a VH region (SEQ ID NO: 354) comprising the amino acid sequences of SEQ ID NO: 230 (VH-CDR1), SEQ ID NO: 231 (VH-CDR2), and SEQ ID NO: 232 (VH-CDR3), and a VL region (SEQ ID NO: 355) comprising the amino acid sequences of SEQ ID NO: 233 (VL-CDR1), SEQ ID NO: 234 (VL-CDR2), and SEQ ID NO: 235 (VL-CDR3);

[0079] HER2: 12) a VH region (SEQ ID NO: 356) comprising the amino acid sequences of SEQ ID NO: 21 (VH-CDR1), SEQ ID NO: 22 (VH-CDR2), and SEQ ID NO: 23 (VH-CDR3), and a VL region (SEQ ID NO: 357) comprising the amino acid sequences of SEQ ID NO: 24 (VL-CDR1), SEQ ID NO: 25 (VL-CDR2), and SEQ ID NO: 26 (VL-CDR3); 13) a VH region (SEQ ID NO: 27) comprising the amino acid sequences of SEQ ID NO: 33 (VH-CDR1), SEQ ID NO: 34 (VH-CDR2), and SEQ ID NO: 35 (VH-CDR3), and a VL region (SEQ ID NO: 28) comprising the amino acid sequences of SEQ ID NO: 36 (VL-CDR1), SEQ ID NO: 37 (VL-CDR2), and SEQ ID NO: 38 (VL-CDR3);

[0080] c-Met: 14) a VH region (SEQ ID NO: 358) comprising the amino acid sequences of SEQ ID NO: 236 (VH-CDR1), SEQ ID NO: 237 (VH-CDR2), and SEQ ID NO: 238 (VH-CDR3), and a VL region (SEQ ID NO: 359) comprising the amino acid sequences of SEQ ID NO: 239 (VL-CDR1), SEQ ID NO: 240 (VL-CDR2), and SEQ ID NO: 241 (VL-CDR3); 15) a VH region (SEQ ID NO: 360) comprising the amino acid sequences of SEQ ID NO: 242 (VH-CDR1), SEQ ID NO: 243 (VH-CDR2), and SEQ ID NO: 244 (VH-CDR3), and a VL region (SEQ ID NO: 361) comprising the amino acid sequences of SEQ ID NO: 245 (VL-CDR1), SEQ ID NO: 246 (VL-CDR2), and SEQ ID NO: 247 (VL-CDR3); 16) a VH region (SEQ ID NO: 362) comprising the amino acid sequences of SEQ ID NO: 248 (VH-CDR1), SEQ ID NO: 249 (VH-CDR2), and SEQ ID NO: 250 (VH-CDR3), and a VL region (SEQ ID NO: 363) comprising the amino acid sequences of SEQ ID NO: 251 (VL-CDR1), SEQ ID NO: 252 (VL-CDR2), and SEQ ID NO: 253 (VL-CDR3);

[0081] Trop 2: 17) a VH region (SEQ ID NO: 364) comprising the amino acid sequences of SEQ ID NO: 254 (VH-CDR1), SEQ ID NO: 255 (VH-CDR2), and SEQ ID NO: 256 (VH-CDR3), and a VL region (SEQ ID NO: 365) comprising the amino acid sequences of SEQ ID NO: 257 (VL-CDR1), SEQ ID NO: 258 (VL-CDR2), and SEQ ID NO: 259 (VL-CDR3); 18) a VH region (SEQ ID NO: 366) comprising the amino acid sequences of SEQ ID NO: 260 (VH-CDR1), SEQ ID NO: 261 (VH-CDR2), and SEQ ID NO: 262 (VH-CDR3), and a VL region (SEQ ID NO: 367) comprising the amino acid sequences of SEQ ID NO: 263 (VL-CDR1), SEQ ID NO: 264 (VL-CDR2), and SEQ ID NO: 265 (VL-CDR3);

[0082] DLL3: 19) a VH region (SEQ ID NO: 368) comprising the amino acid sequences of SEQ ID NO: 266 (VH-CDR1), SEQ ID NO: 267 (VH-CDR2), and SEQ ID NO: 268 (VH-CDR3), and a VL region (SEQ ID NO: 369) comprising the amino acid sequences of SEQ ID NO: 269 (VL-CDR1), SEQ ID NO: 270 (VL-CDR2), and SEQ ID NO: 271 (VL-CDR3); 20) a VH region (SEQ ID NO: 370) comprising the amino acid sequences of SEQ ID NO: 272 (VH-CDR1), SEQ ID NO: 273 (VH-CDR2), and SEQ ID NO: 274 (VH-CDR3), and a VL region (SEQ ID NO: 371) comprising the amino acid sequences of SEQ ID NO: 275 (VL-CDR1), SEQ ID NO: 276 (VL-CDR2), and SEQ ID NO: 277 (VL-CDR3);

[0083] BCMA: 21) a VH region (SEQ ID NO: 372) comprising the amino acid sequences of SEQ ID NO: 278 (VH-CDR1), SEQ ID NO: 279 (VH-CDR2), and SEQ ID NO: 280 (VH-CDR3), and a VL region (SEQ ID NO: 373) comprising the amino acid sequences of SEQ ID NO: 281 (VL-CDR1), SEQ ID NO: 282 (VL-CDR2), and SEQ ID NO: 283 (VL-CDR3);

[0084] LIV-1: 22) a VH region (SEQ ID NO: 374) comprising the amino acid sequences of SEQ ID NO: 284 (VH-CDR1), SEQ ID NO: 285 (VH-CDR2), and SEQ ID NO: 286 (VH-CDR3), and a VL region (SEQ ID NO: 375) comprising the amino acid sequences of SEQ ID NO: 287 (VL-CDR1), SEQ ID NO: 288 (VL-CDR2), and SEQ ID NO: 289 (VL-CDR3);

[0085] GPC-3: 23) a VH region (SEQ ID NO: 87) comprising the amino acid sequences of SEQ ID NO: 99 (VH-CDR1), SEQ ID NO: 100 (VH-CDR2), and SEQ ID NO: 101 (VH-CDR3), and a VL region (SEQ ID NO: 88) comprising the amino acid sequences of SEQ ID NO: 102 (VL-CDR1), SEQ ID NO: 103 (VL-CDR2), and SEQ ID NO: 104 (VL-CDR3);

[0086] FGFR2: 24) a VH region (SEQ ID NO: 376) comprising the amino acid sequences of SEQ ID NO: 290 (VH-CDR1), SEQ ID NO: 291 (VH-CDR2), and SEQ ID NO: 292 (VH-CDR3), and a VL region (SEQ ID NO: 377) comprising the amino acid sequences of SEQ ID NO: 293 (VL-CDR1), SEQ ID NO: 294 (VL-CDR2), and SEQ ID NO: 295 (VL-CDR3); 25) a VH region (SEQ ID NO: 378) comprising the amino acid sequences of SEQ ID NO: 296 (VH-CDR1), SEQ ID NO: 297 (VH-CDR2), and SEQ ID NO: 298 (VH-CDR3), and a VL region (SEQ ID NO: 379) comprising the amino acid sequences of SEQ ID NO: 299 (VL-CDR1), SEQ ID NO: 300 (VL-CDR2), and SEQ ID NO: 301 (VL-CDR3);

[0087] FGFR3: 26) A VH region (SEQ ID NO: 380) comprising the amino acid sequences of SEQ ID NO: 302 (VH-CDR1), SEQ ID NO: 303 (VH-CDR2), and SEQ ID NO: 304 (VH-CDR3), and a VL region (SEQ ID NO: 381) comprising the amino acid sequences of SEQ ID NO: 305 (VL-CDR1), SEQ ID NO: 306 (VL-CDR2), and SEQ ID NO: 307 (VL-CDR3);

[0088] VEGFR2: 27) a VH region (SEQ ID NO: 382) comprising the amino acid sequences of SEQ ID NO: 308 (VH-CDR1), SEQ ID NO: 309 (VH-CDR2), and SEQ ID NO: 310 (VH-CDR3), and a VL region (SEQ ID NO: 383) comprising the amino acid sequences of SEQ ID NO: 311 (VL-CDR1), SEQ ID NO: 312 (VL-CDR2), and SEQ ID NO: 313 (VL-CDR3);

[0089] CD20: 28) A VH region (SEQ ID NO: 384) comprising the amino acid sequences of SEQ ID NO: 314 (VH-CDR1), SEQ ID NO: 315 (VH-CDR2), and SEQ ID NO: 316 (VH-CDR3), and a VL region (SEQ ID NO: 385) comprising the amino acid sequences of SEQ ID NO: 317 (VL-CDR1), SEQ ID NO: 318 (VL-CDR2), and SEQ ID NO: 319 (VL-CDR3); 29) A VH region (SEQ ID NO: 386) comprising the amino acid sequences of SEQ ID NO: 320 (VH-CDR1), SEQ ID NO: 321 (VH-CDR2), and SEQ ID NO: 322 (VH-CDR3), and a VL region (SEQ ID NO: 387) comprising the amino acid sequences of SEQ ID NO: 323 (VL-CDR1), SEQ ID NO: 324 (VL-CDR2), and SEQ ID NO: 325 (VL-CDR3);

[0090] CD38: 30) a VH region (SEQ ID NO: 388) comprising the amino acid sequences of SEQ ID NO: 326 (VH-CDR1), SEQ ID NO: 327 (VH-CDR2), and SEQ ID NO: 328 (VH-CDR3), and a VL region (SEQ ID NO: 389) comprising the amino acid sequences of SEQ ID NO: 329 (VL-CDR1), SEQ ID NO: 330 (VL-CDR2), and SEQ ID NO: 331 (VL-CDR3);

[0091] EphA2: 31) A VH region (SEQ ID NO: 143) comprising the amino acid sequences of SEQ ID NO: 157 (VH-CDR1), SEQ ID NO: 158 (VH-CDR2), and SEQ ID NO: 159 (VH-CDR3), and a VL region (SEQ ID NO: 145) comprising the amino acid sequences of SEQ ID NO: 160 (VL-CDR1), SEQ ID NO: 161 (VL-CDR2), and SEQ ID NO: 162 (VL-CDR3).

[0092] Table 2 below shows the nucleotide and polypeptide sequences of exemplary signal sequences for efficient expression of fusion proteins according to various embodiments. When the above antibodies are expressed in mammalian cells, SEQ ID NO: 333 may be used as a signal sequence, but is not limited thereto.

[0093] [Table 2]

[0094] Table 3 below shows the variable region polypeptide sequences of anti-cancer antibodies that are set forth as antigen binding sites for the various fusion proteins described herein. Fusion proteins according to exemplary embodiments may comprise or consist of these variable region polypeptides.

[0095] [Table 3] JPEG2024539509000008.jpg207149 JPEG2024539509000009.jpg208149 JPEG2024539509000010.jpg207149 JPEG2024539509000011.jpg156149

[0096] Table 4 below shows the nucleotide sequences encoding the variable regions of the anti-cancer antibodies described as the antigen binding sites of the various fusion proteins described herein.

[0097] [Table 4] JPEG2024539509000013.jpg194149 JPEG2024539509000014.jpg194149 JPEG2024539509000015.jpg194149 JPEG2024539509000016.jpg194149 JPEG2024539509000017.jpg194149 JPEG2024539509000018.jpg194149 JPEG2024539509000019.jpg195149 JPEG2024539509000020.jpg194149 JPEG2024539509000021.jpg195149 JPEG2024539509000022.jpg138149

[0098] Fc region or fragment thereof wherein the first Fc domain and the second Fc domain described above may each be an Fc region of an immunoglobulin, the Fc region of the immunoglobulin may be an Fc domain variant as well as a wild-type Fc domain, the Fc region may be an IgG, IgA, IgE, IgD, or IgM Fc region.

[0099] As used herein, the term "Fc domain variant" may refer to a form that differs from the wild-type c domain in terms of glycosylation pattern, has a higher level of a specific glycan species compared to the wild-type Fc domain, a lower level of a specific glycan species compared to the wild-type Fc domain, or a deglycosylated form. In addition, aglycosylated Fc domains are included in the term. Fc domains or variants thereof can be adapted to have a regulated number of sialic acids, fucosylation, or other types of glycosylation by modulation of culture conditions or genetic engineering of the host cell.

[0100] In addition, the glycosylation of the Fc domain of the immunoglobulin can be modified by conventional methods such as chemical methods, enzymatic methods, and genetic engineering methods using microorganisms. In addition, the Fc domain variant may be in a mixed form of the Fc region of each of the immunoglobulins IgG, IgA, IgE, IgD, or IgM. In addition, the Fc domain variant may be in a form in which some amino acids in the Fc domain are replaced with other amino acids.

[0101] The "amino acid" introduced by substitution and / or addition may be any one selected from the group consisting of lysine (K), alanine (A), arginine (R), asparagine (N), aspartic acid (D), cysteine ​​(C), glutamine (Q), glutamic acid (E), glycine (G), histidine (H), isoleucine (I), leucine (L), methionine (M), phenylalanine (F), proline (P), serine (S), threonine (T), tryptophan (W), tyrosine (Y), and valine (V).

[0102] In one embodiment, the variant of the Fc region may be in a form in which amino acids 239 and / or 332 in the CH2 region are substituted with other amino acids (see Kabat numbering system). Specifically, S239 may be substituted with an amino acid other than S, specifically, S239 may be substituted with S239D. Additionally, I332 may be substituted with an amino acid other than I, specifically, I332 may be substituted with I332E.

[0103] In addition, the Fc region may comprise a knob variant or structure, or a hole variant or structure.

[0104] As used herein, the term "knob-into-hole" refers to an Fc heterodimerization strategy to generate antibodies that specifically bind to different regions, such as bispecific, multispecific, or heterodimeric antibodies. In general, this technique involves introducing a knob mutation at the interface of a first polypeptide (e.g., a first CH3 domain of a first antibody heavy chain) and a corresponding hole mutation at the interface of a second polypeptide (e.g., a second CH3 domain of a second antibody heavy chain) such that the knob can fit into the hole to promote heterodimer formation and prevent homodimer formation.

[0105] "Knob" variants are constructed by replacing a small amino acid side chain from the interface of a first polypeptide (e.g., the first CH3 domain of a first antibody heavy chain) with a larger side chain (e.g., arginine, phenylalanine, tyrosine, or tryptophan). Complementary "hole" variants of the same or similar size as the knob are created by replacing a large amino acid side chain in the interface of a second polypeptide (e.g., the second CH3 domain of a second antibody heavy chain) with a smaller side chain (e.g., alanine, serine, valine, or threonine). Knobs and holes may be created by altering the nucleic acid encoding the polypeptide, for example, by site-directed mutagenesis or by peptide synthesis.

[0106] Examples of variants of Fc regions that promote the formation of heterodimers may include those described in WO2014084607 and WO2018059502, etc. The disclosures of WO2014084607 and WO2018059502 are incorporated herein by reference. WO2014084607 may include, for example, (a-1) a tryptophan (W) substituted at Lys409 of one CH3 domain that interacts with a valine (V) substituted at Asp399 and a threonine (T) substituted at Phe405 of the other CH3 domain; and (a-2) a serine (S) substituted at Tyr349 of one CH3 domain that interacts with a tryptophan (W) substituted at Glu357 of the other CH3 domain, and in addition and (b-2) a glutamic acid (E) substituted at Gln347 and a glutamic acid substituted at Lys360 of one CH3 domain, interacting with an arginine (R) substituted at Gln347 of the other CH3 domain, where the positions of amino acid residues are according to the EU index. WO 2018059502 discloses, for example, a) to e) each of: a) L351G, L351Y, L351V, L351P, L351D, L351E, L351K, or L351W; b) T366L, T366P, T366W, or T366V; c) D399C, D399N, D399I, D399G, D399 d) Y407L, Y407A, Y407P, Y407F, Y407T, or Y407H; and e) one or more mutations selected from K409C, K409P, K409S, K409F, K409V, K409Q, or K409R, where amino acid residue positions are according to the EU index.

[0107] Fusion protein structure The fusion proteins have the following structural formulas (I), (II), (III), and (IV), respectively: N'-X-(L1)nA-C' (I); N'-Y-(L2)mB-C' (II); N'-C-C' (III); and N'-D-C' (IV) and In the structural formulas (I), (II), (III), and (IV), N' is the N-terminus of each polypeptide, C' is the C-terminus of each polypeptide; - means concatenation, A, B, C, and D are monomeric polypeptide sequences of an Fc domain, each comprising an immunoglobulin CH2 and CH3 region, and optionally further comprising a CH4 and / or hinge sequence; A dimerizes with either C or D to form a first Fc domain (b); B dimerizes with the remaining one of C or D to form a second Fc domain (c); L1 and L2 are each a peptide linker; n and m are each independently 0 or 1; X comprises a heavy or light chain variable region of an antibody that specifically binds to an antigen; Y comprises a light chain variable region or a heavy chain variable region of an antibody that specifically binds to an antigen; X and Y pair with each other to form an antigen-binding site (a) that specifically binds to an antigen, The polypeptides (I), (II), (III), and (IV) can be associated into a fusion protein comprising one antigen-binding site and two Fc domains.

[0108] Specifically, X is a first polypeptide sequence of an antigen-binding site comprising a heavy chain CDR1, CDR2, and CDR3 sequence of an antibody that specifically binds to a first antigen, or a heavy chain variable region of an antibody that specifically binds to the first antigen; Y is a second polypeptide sequence of an antigen-binding site comprising a light chain CDR1, CDR2, and CDR3 sequence of an antibody that specifically binds to the first antigen, or a light chain variable region of an antibody that specifically binds to the first antigen; X and Y pair with each other to form an antigen-binding site (a) that specifically binds to the antigen.

[0109] According to one embodiment, the CH3 region may be mutated to minimize interactions between A and B, and between C and D, and to promote the formation of heterodimeric Fc between A and C, and between B and D. Specifically, the Fc domain monomer may comprise a knob variant or a hole variant that promotes the formation of Fc heterodimers (heterodimeric Fc); or the Fc domain monomer may comprise a variant that promotes the formation of heterodimers by an electrostatic steering mechanism.

[0110] According to one embodiment, X in structural formula (I) may further comprise a heavy chain CH1 region and / or Y in structural formula (II) may further comprise a light chain constant region. In addition, the fusion protein may have the following structural formulas (I'), (II'), (III), and (IV): N'-VD1-(L3)pX-(L1)nA-C' (I'); N'-VD2-(L4)qY-(L2)mB-C' (II'); N'-C-C' (III); and N'-D-C' (IV) and In the structural formulas (I'), (II'), (III), and (IV), N' is the N-terminus of the polypeptide chain; C' is the C-terminus of the polypeptide chain; - means concatenation, A, B, C, and D are monomeric polypeptide sequences of Fc domains each comprising an immunoglobulin CH2 and CH3 region, and optionally further comprising a CH4 and / or hinge sequence, where A dimerizes with either C or D to form a first Fc domain (b) and B dimerizes with the remaining one of C or D to form a second Fc domain (c); L1, L2, L3, and L4 are each a peptide linker; n, m, p, and q are each independently 0 or 1; VD1 consists of a heavy or light chain variable region of an antibody that specifically binds to an antigen, or CDR1, CDR2, and CDR3 of the antibody heavy or light chain; VD2 consists of a light or heavy chain variable region of an antibody that specifically binds to an antigen, or CDR1, CDR2, and CDR3 of an antibody heavy or light chain; VD1 and VD2 pair with each other to form a second antibody variable region that specifically binds to a second antigen; X comprises a heavy or light chain variable region of an antibody that specifically binds to an antigen, or CDR1, CDR2, and CDR3 of an antibody heavy or light chain; Y comprises an antibody light or heavy chain variable region that specifically binds to an antigen, or CDR1, CDR2, and CDR3 of an antibody heavy or light chain; X and Y pair with each other to form a first antibody variable region that specifically binds to a first antigen; VD1-(L3)pX forms a first polypeptide sequence of the antigen-binding site (a) and VD2-(L4)qY forms a second polypeptide sequence of the antigen-binding site (a).

[0111] According to one embodiment, the CH3 region may be mutated to minimize interactions between A and B, and between C and D, and to promote the formation of heterodimeric Fc between A and C, and between B and D. Specifically, the Fc domain monomer may comprise a knob variant or a hole variant that promotes the formation of Fc heterodimers (heterodimeric Fc); or the Fc domain monomer may comprise a variant that promotes the formation of heterodimers by an electrostatic steering mechanism.

[0112] According to one embodiment, the heavy chain variable region may further comprise a heavy chain CH1 region. In addition, the light chain variable region may further comprise a light chain constant region.

[0113] In the structures of the fusion proteins described herein, the bond between X and Y can be achieved i) through a disulfide bond formed by Cys present in the CH1 and light chain constant region, ii) through a disulfide bond formed by Cys present in the heavy chain variable region and the light chain variable region, or iii) through a disulfide bond formed by Cys present in the CH1 and light chain constant region and Cys present in the heavy chain variable region and the light chain variable region.

[0114] Specifically, the bond between X and Y can be formed by a disulfide bond existing between CH1233 and CL214 based on the Kabat numbering system. In addition, X and Y can further include Cys by amino acid substitution. Examples of such variants can include mutations in the variable region, specifically, mutations at 105C of VH and 43C of VL, or mutations at 44C of VH and 100C of VL, based on the Kabat numbering system. In one embodiment, the mutations can be Q105C of VH and A43C of VL. In addition, in one embodiment, the mutations can be G44C of VH and Q100C of VL. In addition, examples of variants in the constant region can include mutations at 122C of CH1 and 121C of CL, based on the Kabat numbering system. In one embodiment, the mutations can be F122C of CH1 and S121C of CL.

[0115] Linkers and hinges The hinge is a hinge region derived from an immunoglobulin. In one embodiment, the antibody hinge region is an IgG hinge region. The IgG hinge region provided herein may be selected from, for example, antibody hinge regions of various IgG subtypes. The table below lists exemplary IgG subtypes with core hinge sequences that may be included in the flexible peptide region provided herein. In addition, at least one Cys may be present in the hinge. Specifically, one, two, or three Cys may be present in the hinge.

[0116] [Table 5]

[0117] The hinge may be modified to delete disulfide bonds or to introduce additional disulfide bonds.

[0118] In addition, each of the linkers L1 and L2 may contain 1 to about 70 amino acids. According to one exemplary embodiment, each of L1 and L2 may contain about 5 to about 60 amino acids, about 10 to about 50 amino acids, about 15 to about 40 amino acids, or about 20 to about 30 amino acids. According to another exemplary embodiment, for example, each of L1 and L2 may be a peptide consisting of 1 to 70 amino acid residues, 2 to 60 amino acid residues, 2 to 50 amino acid residues, 2 to 40 amino acid residues, 2 to 30 amino acid residues, 3 to 50 amino acid residues, 3 to 40 amino acid residues, 3 to 30 amino acid residues, 2 to 28 amino acid residues, 2 to 26 amino acid residues, 2 to 24 amino acid residues, 2 to 22 amino acid residues, 2 to 20 amino acid residues, 2 to 18 amino acid residues, 2 to 16 amino acid residues, 2 to 14 amino acid residues, 2 to 12 amino acid residues, or 2 to 10 amino acid residues. Specifically, L1 and L2 may include, but are not limited to, the amino acid sequence of (G4S)o (wherein o is an integer of 1 to 5) in Table 6 below. In addition, L1 and L2 may have different amino acid sequences. In addition, L1 and L2 may include at least one Cys. In addition, a disulfide bond may be formed through the Cys present in L1 and L2.

[0119] [Table 6] In addition, each of the linkers L3 and L4 may contain 1 to about 30 amino acids. According to one exemplary embodiment, each of L3 and L4 may contain about 5 to about 25 amino acids, about 10 to about 20 amino acids, or about 15 amino acids. According to another exemplary embodiment, each of L3 and L4 may be a peptide consisting of 2 to 30 amino acid residues, 2 to 25 amino acid residues, 2 to 20 amino acid residues, 2 to 15 amino acid residues, 3 to 30 amino acid residues, 2 to 28 amino acid residues, 2 to 26 amino acid residues, 2 to 24 amino acid residues, 2 to 22 amino acid residues, 2 to 20 amino acid residues, 2 to 18 amino acid residues, 2 to 16 amino acid residues, 2 to 14 amino acid residues, 2 to 12 amino acid residues, or 2 to 10 amino acid residues. Specifically, L3 and L4 may include, but are not limited to, the amino acid sequence of (G4S)o in Table 6 above, where o is an integer of 1 to 5. In addition, L3 and L4 may have different amino acid sequences.

[0120] Specific examples of fusion proteins Fusion protein containing one antigen-binding site and two Fc i) Fusion proteins in which the antigen-binding site is a Fab As shown in FIG. 2a, the fusion protein comprises polypeptides of structural formulas (I), (II), (III), and (IV), where X is a heavy chain variable region and further comprises CH1, and Y is a light chain variable region and comprises a light chain constant region. In addition, X and Y are attached to each other by Cys in the CH1 structure and the light chain variable region to form a Fab structure. Wherein n is 0, CH1 is directly linked to the hinge, m is 1, and L2 comprises a peptide linker. Wherein A and C are attached to each other to form a first Fc domain, and B and D are attached to each other to form a second Fc domain. In addition, the CH3 region of A comprises a hole variant, and the CH3 region of C comprises a knob variant. In addition, the CH3 region of B comprises a hole variant, and the CH3 region of D comprises a knob variant. In addition, the antigen binding site, antigen, hinge, linker, and Fc are as described above.

[0121] As shown in FIG. 2b, the fusion protein comprises polypeptides of structural formulas (I), (II), (III), and (IV), where X is a heavy chain variable region and further comprises CH1, and Y is a light chain variable region and comprises a light chain constant region. In addition, X and Y are attached to each other by Cys in the CH1 structure and the light chain variable region to form a Fab structure. Wherein n is 0, CH1 is directly linked to the hinge, m is 1, and L2 comprises a peptide linker. Wherein A and C are attached to each other to form a first Fc domain, and B and D are attached to each other to form a second Fc domain. In addition, the CH3 region of A comprises a hole variant, and the CH3 region of C comprises a knob variant. In addition, the CH3 region of B comprises a hole variant, and the CH3 region of D comprises a knob variant. In addition, X comprises a 105C mutation, and Y comprises a 43C mutation, and a disulfide bond between Cys is formed. In addition, the antigen binding site, antigen, hinge, linker, and Fc are as described above.

[0122] As shown in FIG. 2c, the fusion protein comprises polypeptides of structural formulas (I), (II), (III), and (IV), where X is a heavy chain variable region and further comprises CH1, and Y is a light chain variable region and comprises a light chain constant region. In addition, X and Y are attached to each other by Cys in the CH1 structure and the light chain variable region to form a Fab structure. Wherein n is 0, CH1 is directly linked to the hinge, m is 1, and L2 comprises a peptide linker. Wherein A and C are attached to each other to form a first Fc domain, and B and D are attached to each other to form a second Fc domain. In addition, the CH3 region of A comprises a hole variant, and the CH3 region of C comprises a knob variant. In addition, the CH3 region of B comprises a hole variant, and the CH3 region of D comprises a knob variant. In addition, CH1 comprises a 122C mutation, and the light chain constant region comprises a 121C mutation, and a disulfide bond between Cys is formed. In addition, the antigen binding site, antigen, hinge, linker, and Fc are as described above.

[0123] As shown in FIG. 2d, the fusion protein comprises polypeptides of structural formulas (I), (II), (III), and (IV), where X is a heavy chain variable region and further comprises CH1, and Y is a light chain variable region and comprises a light chain constant region. In addition, X and Y are attached to each other by Cys in the CH1 structure and the light chain variable region to form a Fab structure. Wherein n is 0, CH1 is directly linked to the hinge, m is 1, and L2 comprises a peptide linker. Wherein A and C are attached to each other to form a first Fc domain, and B and D are attached to each other to form a second Fc domain. In addition, the CH3 region of A comprises a hole variant, and the CH3 region of C comprises a knob variant. In addition, the CH3 region of B comprises a hole variant, and the CH3 region of D comprises a knob variant. In addition, X comprises a 44C mutation, and Y comprises a 100C mutation, and a disulfide bond between Cys is formed. In addition, the antigen binding site, antigen, hinge, linker, and Fc are as described above.

[0124] As shown in Figures 6a-6d, the fusion protein comprises polypeptides of structural formulas (I), (II), (III), and (IV), where X is a heavy chain variable region and further comprises CH1, and Y is a light chain variable region and comprises a light chain constant region. In addition, X and Y are attached to each other by Cys in the CH1 structure and the light chain variable region to form a Fab structure. Wherein n is 0 and CH1 is directly linked to the hinge. Wherein A and C are attached to each other to form a first Fc domain, and B and D are attached to each other to form a second Fc domain. In addition, the CH3 region of A comprises a hole variant, and the CH3 region of C comprises a knob variant. In addition, the CH3 region of B comprises a hole variant, and the CH3 region of D comprises a knob variant. In addition, X comprises a 44C mutation, and Y comprises a 100C mutation, and a disulfide bond between Cys is formed. wherein m is 0 and the light chain variable region may be directly linked to the hinge (Figure 6d). Additionally, m is 1 and L2 may comprise a 15-mer peptide linker (Figure 6a), a 10-mer peptide linker (Figure 6b), or a 5-mer peptide linker (Figure 6c). Additionally, the antigen binding site, antigen, hinge, linker, and Fc are as described above.

[0125] As shown in FIG. 19a, the fusion protein comprises polypeptides of structural formulas (I), (II), (III), and (IV), where X is a heavy chain variable region and further comprises CH1, and Y is a light chain variable region and comprises a light chain constant region. In addition, X and Y are attached to each other by Cys in the CH1 structure and the light chain variable region to form a Fab structure, where n is 0, CH1 is directly linked to the hinge, m is 1, and L2 comprises a peptide linker. Wherein A and C are attached to each other to form a first Fc domain, and B and D are attached to each other to form a second Fc domain. In addition, the CH3 region of A comprises a hole variant, and the CH3 region of C comprises a knob variant. In addition, the CH3 region of B comprises a hole variant, and the CH3 region of D comprises a knob variant. In addition, X comprises a 44C mutation, and Y comprises a 100C mutation, and a disulfide bond between Cys is formed. wherein all CH2s of A, B, C, and D contain the 239D and 332E mutations, in addition, the antigen binding site, antigen, hinge, linker, and Fc are as described above.

[0126] ii) Fusion proteins in which the antigen-binding site is an Fv As shown in Figure 10, the fusion protein comprises polypeptides of structural formulas (I), (II), (III), and (IV), where X and Y are attached to each other by at least one Cys present in X and Y to form an Fv structure, where A and C are attached to each other to form a first Fc domain, and B and D are attached to each other to form a second Fc domain. In addition, the CH3 region of A comprises a whole variant and the CH3 region of C comprises a knob variant. In addition, the CH3 region of B comprises a whole variant and the CH3 region of D comprises a knob variant. X may comprise a 44C, 105C, 122C, 44C / 105C, 44C / 122C, 105C / 126C, or 44C / 105C / 126C mutation, and Y may comprise a 100C, 43C, 121C, 100C / 43C, 100C / 121C, 43C / 121C, or 100C / 43C / 121C mutation. In addition, a disulfide bond may be formed by the Cys present in L1 and L2. In addition, the antigen binding site, antigen, hinge, linker, and Fc are as described above.

[0127] Fusion protein containing two antigen-binding sites and two Fc iii) Fusion proteins in which the antigen-binding site is a Fab As shown in FIG. 25, the fusion protein comprises polypeptides of structural formulas (I'), (II'), (III), and (IV), where X is a heavy chain variable region and further comprises CH1, and Y is a light chain variable region and comprises a light chain constant region. In addition, X and Y are attached to each other by Cys in the CH1 structure and the light chain variable region to form a Fab structure, where n is 0, CH1 is directly linked to the hinge, m is 1, and L2 comprises a peptide linker. Wherein A and C are attached to each other to form a first Fc domain, and B and D are attached to each other to form a second Fc domain. In addition, the CH3 region of A comprises a hole variant, and the CH3 region of C comprises a knob variant. In addition, the CH3 region of B comprises a hole variant, and the CH3 region of D comprises a knob variant. In addition, VD1 in structural formula (I') is a heavy chain variable region, and VD2 in structural formula (II') is a light chain variable region, and VD1 and VD2 pair with each other to form Fv. In addition, p and q are each 1, and L3 and L4 are peptide linkers. As a non-limiting example, X and Y can pair with each other to form the variable region of pertuzumab, and VD1 and VD2 can pair with each other to form the variable region of trastuzumab. In addition, the antigen binding site, antigen, hinge, linker, and Fc are as described above.

[0128] As shown in Figure 26, the peptide linkers L3 and L4 in structural formulas (I') and (II') may be of various lengths. In addition, the first antigen-binding site formed by the pairing between X and Y, and the second antigen-binding site formed by the pairing between VD1 and VD2 may be the same or different. In addition, L1 and L2 may also include various peptide linkers.

[0129] Use of fusion proteins In another aspect of the present invention, there is provided a pharmaceutical composition for preventing or treating cancer, comprising the fusion protein as an active ingredient.

[0130] Here, the cancer may be any one selected from the group consisting of gastric cancer, liver cancer, lung cancer, colon cancer, breast cancer, prostate cancer, skin cancer, bone cancer, multiple myeloma, glioma, ovarian cancer, pancreatic cancer, cervical cancer, thyroid cancer, laryngeal cancer, acute myeloid leukemia, chronic myeloid leukemia, acute lymphoblastic leukemia, chronic lymphoblastic leukemia, brain tumor, neuroblastoma, retinoblastoma, head and neck cancer, salivary gland cancer, and lymphoma.

[0131] Polynucleotides encoding fusion proteins In another aspect of the invention, there are provided polynucleotides encoding the polypeptides of structural formula (I), (II), (III), and / or (IV).

[0132] In another aspect of the invention, there are provided polynucleotides encoding the polypeptides of structural formula (I'), (II'), (III), and / or (IV).

[0133] The polynucleotide may further comprise a nucleic acid encoding a signal sequence or leader sequence. As used herein, the term "signal sequence" refers to a signal peptide that directs the secretion of a target protein. The signal peptide is translated and then cleaved in the host cell. Specifically, a signal sequence is an amino acid sequence that initiates the translocation of a protein across the endoplasmic reticulum (ER) membrane.

[0134] Signal sequences are well known in the art for their characteristics. Such signal sequences typically contain 16-30 amino acid residues, but may contain more or fewer than such amino acid residues. A typical signal peptide consists of three regions: an N-terminal region, a central hydrophobic region, and a more polar C-terminal region. The central hydrophobic region contains 4-12 hydrophobic residues that immobilize the signal sequence during translocation of the immature polypeptide through the membrane lipid bilayer.

[0135] After initiation, the signal sequence is cleaved in the lumen of the ER by a cellular enzyme commonly known as a signal peptidase. Here, the signal sequence may be a secretory signal sequence of tPa (tissue plasminogen activator), HSV gD (signal sequence of herpes simplex virus glycoprotein D), or growth hormone. Preferably, a secretory signal sequence used in higher eukaryotic cells, including mammals, may be used. In addition, as the signal sequence, a wild-type signal sequence may be used, or a signal sequence substituted with a codon having a high expression frequency in the host cell may be used.

[0136] Vector carrying a polynucleotide In another aspect of the present invention, a vector comprising a polynucleotide is provided. The vector may comprise a polynucleotide encoding a polypeptide of structural formula (I), (II), (III), and / or (IV). Additionally, the vector may comprise a polynucleotide encoding a polypeptide of structural formula (I'), (II'), (III), and / or (IV).

[0137] A vector can be introduced into a host cell to be recombined and can be inserted into the genome of the host cell. Alternatively, a vector is understood as a nucleic acid means that includes a polynucleotide sequence that can replicate autonomously as an episome. Vectors include linear nucleic acids, plasmids, phagemids, cosmids, RNA vectors, viral vectors, and analogs thereof. Examples of viral vectors include, but are not limited to, retroviruses, adenoviruses, and adeno-associated viruses.

[0138] Specifically, the vector may include plasmid DNA, phage DNA, etc.; and commercially developed plasmids (pUC18, pBAD, pIDTSAMRT-AMP, etc.), E. coli-derived plasmids (pYG601BR322, pBR325, pUC118, pUC119, etc.), Bacillus subtilis-derived plasmids (pUB110, pTP5, etc.), yeast-derived plasmids (YEp13, YEp24, YCp50, etc.), phage DNA (Charon4A, Charon21A, EMBL3, EMBL4, λgt10, λgt11, λZAP, etc.), animal virus vectors (retrovirus, adenovirus, vaccinia virus, etc.), insect virus vectors (baculovirus, etc.). Since vectors exhibit different protein expression levels and modifications depending on the host cell, it is preferable to select and use the host cell that is most suitable for the purpose.

[0139] As used herein, the term "gene expression" or "expression" of a target protein is understood to mean the transcription of a DNA sequence, the translation of an mRNA transcript, and the secretion of a fusion protein product or a fragment thereof. A useful expression vector may be RcCMV (Invitrogen, Carlsbad) or a variant thereof. The expression vector may include a human cytomegalovirus (CMV) promoter to promote continuous transcription of the target gene in mammalian cells, and a bovine growth hormone polyadenylation signal sequence to increase the stability level of the RNA after transcription.

[0140] Transformed cells expressing the fusion protein In another embodiment of the present invention, a transformed cell capable of expressing the gene is provided. Specifically, the transformed cell may be a cell into which a vector has been introduced.

[0141] Host cells for transformed cells may include, but are not limited to, prokaryotic cells, eukaryotic cells, and cells of mammalian, plant, insect, fungal, or cellular origin. As an example of a prokaryotic cell, E. coli may be used. In addition, as an example of a eukaryotic cell, yeast may be used. In addition, for mammalian cells, CHO cells, F2N cells, CSO cells, BHK cells, Bowes melanoma cells, HeLa cells, 911 cells, AT1080 cells, A549 cells, HEK293 cells, HEK293T cells, etc. may be used. However, mammalian cells are not limited thereto, and any cell known to those skilled in the art to be usable as a mammalian host cell may be used.

[0142] In addition, for the introduction of expression vectors into host cells, CaCl2 precipitation, the Hanahan method, the efficiency of which is increased by the use of a reducing agent such as dimethyl sulfoxide (DMSO) in CaCl2 precipitation, electroporation, calcium phosphate precipitation, protoplast fusion, agitation using silicon carbide fibers, Agrobacterium-mediated transformation, transformation using PEG, dextran sulfate-, Lipofectamine-, and desiccation / inhibition-mediated transformation, etc. may be used.

[0143] As described above, for optimization of the properties of the fusion protein as a therapeutic agent or for any other purpose, the glycosylation pattern of the fusion protein (e.g., sialic acid, fucosylation, glycosylation) can be adjusted by manipulating the glycosylation-related genes possessed by the host cell by methods known to those skilled in the art.

[0144] Methods for Producing Fusion Proteins In another aspect of the invention, there is provided a method for producing a fusion protein comprising an antigen binding site, a first Fc domain or variant thereof, and a second Fc domain or variant thereof, the method comprising the steps of i) culturing the transformed cell; and ii) recovering the produced fusion protein.

[0145] The step of culturing the transformed cells may be carried out using methods well known in the art. In particular, the culturing may be carried out in a batch process or continuously in a fed-batch or repeated fed-batch process.

[0146] Compositions or preparations containing fusion proteins In another aspect of the invention, there is provided a pharmaceutical composition comprising the fusion protein as an active ingredient.

[0147] The pharmaceutical composition may be used for the prevention or treatment of cancer, such as any one type of cancer selected from the group consisting of gastric cancer, liver cancer, lung cancer, colon cancer, breast cancer, prostate cancer, gallbladder cancer, bladder cancer, kidney cancer, esophageal cancer, skin cancer, rectal cancer, osteosarcoma, multiple myeloma, glioma, ovarian cancer, pancreatic cancer, cervical cancer, endometrial cancer, thyroid cancer, laryngeal cancer, testicular cancer, mesothelioma, acute myeloid leukemia, chronic myeloid leukemia, acute lymphoblastic leukemia, chronic lymphoblastic leukemia, brain tumor, neuroblastoma, retinoblastoma, head and neck cancer, salivary gland cancer, and lymphoma.

[0148] The preferred dosage of the pharmaceutical composition varies depending on the condition and weight of the patient, the severity of the disease, the form of the drug, the route and duration of administration, and can be appropriately selected by those skilled in the art. In the pharmaceutical composition for treating or preventing tumors of the present invention, the active ingredient may be contained in any amount (effective amount) depending on the application, dosage form, purpose of formulation, etc., so long as the active ingredient can exhibit therapeutic activity against tumors or can exhibit therapeutic effect against cancer in particular. The conventional effective amount of the active ingredient will be determined within the range of 0.001% by weight to 20.0% by weight based on the total weight of the composition. Here, the term "effective amount" refers to the amount of the active ingredient that can induce the effect of improving or treating the disease condition, especially the effect of improving or treating the cancer condition. Such an effective amount can be determined experimentally within the scope of the common knowledge of those skilled in the art.

[0149] As used herein, the term "treatment" may be used to mean both therapeutic and prophylactic treatment. Here, prevention may be used to mean that the condition or disease of the subject is alleviated or relieved. In one embodiment, the term "treatment" includes both application or any form of administration to treat disease in mammals, including humans. In addition, the term includes the meaning of inhibiting or slowing down the progression of disease; restoring or repairing impaired or lost functions so that the disease is partially or completely alleviated; stimulating inefficient processes; or alleviating severe disease.

[0150] Pharmacokinetic parameters such as bioavailability and underlying parameters such as clearance rate can also affect efficacy. Thus, "improved efficacy" (e.g., improved efficacy) can result from improved pharmacokinetic parameters and improved efficacy, which can be measured by comparing parameters such as clearance rate and tumor treatment or improvement in test animals or human subjects.

[0151] As used herein, the term "therapeutically effective amount" or "pharmacologically effective amount" refers to an amount of a compound or composition that is effective to prevent or treat the disease in question, sufficient to treat the disease at a reasonable benefit / risk ratio applicable to any medical treatment and without causing adverse effects. The level of the effective amount can be determined according to factors including the patient's health condition, the type and severity of the disease, the activity of the drug, the patient's sensitivity to the drug, the mode of administration, the number of administrations, the route of administration and the excretion rate, the duration of treatment, the drugs that are combined or used simultaneously, and other factors well known in the medical field. In one embodiment, a therapeutically effective amount refers to an amount of drug that is effective to treat cancer.

[0152] Here, the pharmaceutical composition may further include a pharma- ceutically acceptable carrier. The pharma- ceutically acceptable carrier may be any carrier, so long as the carrier is a non-toxic material suitable for delivery to a patient. Distilled water, alcohol, fat, wax, and inert solids may be included as carriers. Pharmaceutically acceptable adjuvants (buffers, dispersants) may also be included in the pharmaceutical composition.

[0153] Specifically, by including a pharma- ceutical composition in addition to an active ingredient, a parenteral preparation can be prepared according to the route of administration using a conventional method known in the art. Here, the term "pharma-ceutical acceptable" means that the carrier is not more toxic than the subject to which it is applied (prescribed) and can be applied without inhibiting the activity of the active ingredient.

[0154] When the pharmaceutical composition is prepared into a parenteral preparation, it can be prepared into the form of injection, transdermal patch, nasal aspirate, or suppository using a suitable carrier according to the method known in the art. When prepared into injection, sterile water, ethanol, polyol such as glycerol or propylene glycol, or mixtures thereof can be used as a suitable carrier; isotonic solutions such as Ringer's solution, phosphate buffered saline (PBS) containing triethanolamine or sterile water for injection, and 5% dextrose can be preferably used. The formulation of pharmaceutical compositions is known in the art, and specific reference can be made to Remington's Pharmaceutical Sciences (19th Edition, 1995), etc. This document is considered to be a part of this specification.

[0155] The preferred dosage of the pharmaceutical composition may range from 0.01 μg / kg to 10 g / kg, or from 0.01 mg / kg to 1 g / kg per day, depending on the patient's condition, weight, sex, age, severity of the patient, and route of administration. The dosage may be administered once a day, or may be divided into several doses per day. Such dosages should not be construed as limiting the scope of the invention in any aspect.

[0156] The subjects to which the pharmaceutical composition may be applied (prescribed) are mammals, including dogs, cats, humans, etc., with humans being particularly preferred. In addition to the active ingredient, the pharmaceutical composition of the present invention may further comprise any compound or natural extract known to have a therapeutic effect against tumors.

[0157] Fusion Protein Therapy In another aspect of the invention, there is provided a method for treating or preventing cancer comprising administering to a subject a fusion protein comprising an antigen-binding site, a first Fc domain or variant thereof, and a second Fc domain or variant thereof.

[0158] In another aspect of the invention there is provided a use of a fusion protein comprising an antigen-binding site, a first Fc domain or a variant thereof, and a second Fc domain or a variant thereof for the treatment of cancer.

[0159] Here, the subject may be an individual suffering from cancer. In addition, the subject may be a mammal, preferably a human.

[0160] The route of administration, dosage, and frequency of administration of the fusion protein may vary depending on the patient's condition and the presence or absence of side effects, and therefore the fusion protein may be administered to the subject by various means and amounts. The optimal administration method, dosage, and frequency of administration can be selected within an appropriate range by those skilled in the art. In addition, the fusion protein may be administered in combination with other drugs or physiologically active substances known to have a therapeutic effect on the disease to be treated, or may be formulated in the form of a combined preparation with other drugs. EXAMPLES

[0161] Hereinafter, the present invention will be described in more detail by the following examples. However, the following examples are merely illustrative of the present invention, and the scope of the present invention is not limited to these examples.

[0162] Example 1. Design, preparation, and analysis of a novel antibody with two Fc domains Naturally occurring human immunoglobulin G (IgG) consists of two fragment antigen-binding (Fab) regions and one fragment crystallizable (Fc) region (Figure 1a). Human IgG binds target antigens in a monovalent or bivalent manner, and in some cases has 0.5 to 1 Fc region per target antigen (Figures 1c, 1d, and 1e).

[0163] The aim of the present invention is to improve effector function by increasing the amount of Fc region present per antigen, while having a molecular weight similar to that of an antibody and a uniform composition. Therefore, the inventors have designed a novel antibody format with two Fc regions, with a molecular weight similar to that of a natural human IgG antibody (approximately 150 kDa) (Figure 1b). This format binds to cancer cell target antigens and allows the Fc region to be present on the cancer cell surface up to four times more than conventional antibodies (Figures 1b and 1f).

[0164] To implement the novel antibody format mentioned above, trastuzumab was used as a template (Figure 2a). To improve the pairing between the VH-CH1 and VL-CL regions of the trastuzumab Fab region, specific amino acids were replaced with cysteines to introduce artificial disulfide bonds (Figures 2b, 2c, and 2d).

[0165] Fabs were designed in which glutamine (Q) at position 105 of the heavy chain and alanine (A) at position 43 of the light chain were replaced with cysteine, phenylalanine (F) at position 122 of the heavy chain and serine (S) at position 121 of the light chain were replaced with cysteine, and glycine (G) at position 44 of the heavy chain and glutamine (Q) at position 100 of the light chain were replaced with cysteine, and these Fabs were designated mutant 1, mutant 2, and mutant 3, respectively (Figures 2b, 2c, 2d, 3a, and 3b). Based on the above, the Fab-(Fc)2 structure using trastuzumab as a template was designated wild type (WT) (Figure 2a), and the Fab-(Fc)2 structures having Fabs corresponding to mutants 1 to 3 were designated M1, M2, and M3, respectively (Figures 2b, 2c, 2d, 3a, and 3b).

[0166] The position of the amino acids constituting the antibody follows the Kabat numbering system. To minimize unwanted Fc-related by-products, the knob-into-hole mutation technique (Merchant et al., Nat. Biotechnol. 1998) was applied to the Fc domain of human immunoglobulin G1 (IgG1) (SEQ ID NO: 3) to design Fc polypeptides with knob mutations (S354C and T366W; SEQ ID NO: 4) and Fc polypeptides with hole mutations (Y349C, T366S, L368A, and Y407V; SEQ ID NO: 5). To provide additional flexibility between the CL domain and the hinge region, a (G4S)3 linker was introduced (SEQ ID NO: 6; Figure 2). Expression of WT was performed by co-transfection of vectors capable of expressing polypeptides corresponding to Fc-hole (SEQ ID NO: 7), TraH-WT-knob (SEQ ID NO: 8), and TraL-WT-knob (SEQ ID NO: 9) into the EXPICHO-S™ (Gibco, A29127) cell line.

[0167] M1 consists of the Fc-hole (SEQ ID NO: 7), TraH-Q105C-knob (SEQ ID NO: 10) and TraL-A43C-knob (SEQ ID NO: 11), M2 consists of the Fc-hole (SEQ ID NO: 7), TraH-F122C-knob (SEQ ID NO: 12) and TraL-S121C-knob (SEQ ID NO: 13) and M3 consists of the Fc-hole (SEQ ID NO: 7), TraH-G44C-knob (SEQ ID NO: 14) and TraL-Q100C-knob (SEQ ID NO: 15). All M1, M2 and M3 were expressed in the EXPICHO-S™ (Gibco, A29127) cell line. The M1, M2, and M3 were purified using AKTA pure 25 (Cytiva) or AKTA avant 150 (Cytiva) protein isolation and purification system equipped with CAPTURESELECT™ CH1-XL prepacked column (Thermo Scientific, 494346205) purification column, the purified products were further subjected to affinity chromatography using KappaSelect resin (Cytiva, 17545801), and the samples were concentrated using Amicon Ultra-15 centrifugal filter units (Merck millipore). For the final purified products, the absorbance of the samples at 280 nm was measured using a NanoDrop One trace spectrophotometer (Thermo Fisher Scientific), and the concentration was quantified based on the intrinsic extinction coefficient and molecular weight of the samples.

[0168] The purified products were analyzed using sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and size exclusion chromatography (SEC) (Figures 4 and 5a-5d). Bio-rad electrophoresis gels and systems were used for SDS-PAGE analysis, samples were analyzed under non-reducing conditions, and the size of each band was identified using Coomassie brilliant blue staining (Figure 4). WT, M1, M2, and M3 were identified at approximately 150 kDa, and monomers were identified at approximately 75 kDa in WT, which does not have an additional disulfide bond introduced at the Fab interface (Figure 4). Similarly, for M1 and M2, trace amounts of monomers were identified at 75 kDa. For M3, almost no monomers were identified, probably because most of the monomers were readily paired due to disulfide bond formation (Figure 4).

[0169] For size exclusion chromatography analysis, an ALLIANCE® HPLC-e2695 separation module (Waters, 2695) equipped with an Agilent Bio SEC-3 HPLC column (Agilent, 5190-2511) was used. The analysis showed that the main product was identified with a retention time of 8.6-8.8 min (Figures 5a, 5b, 5c, and 5d).

[0170] Based on the M3 structure, the effect of the linker connecting the CL domain and the hinge region on the structural integrity of the antibody was analyzed. M3 had a 15-mer polypeptide linker consisting of (G4S)3, V1 (SEQ ID NOs: 7, 14, and 16) and V2 (SEQ ID NOs: 7, 14, and 17) had polypeptide linkers of (G4S)2 and G4S, respectively, and V3 (SEQ ID NOs: 7, 14, and 18) directly linked the CL domain and the hinge region without a linker (Figure 6). By SDS-PAGE analysis, a major product was identified at approximately 150 kDa, and no by-products were identified (Figure 7). It was found that the presence or absence of a linker between the CL domain and the hinge region had no significant effect on the formation of by-products.

[0171] Based on these results, it was found that when VH44 and VL100 of Fab were replaced with their respective cysteines, the Fab-(Fc)2 structure was stably formed.

[0172] [Table 7] JPEG2024539509000026.jpg209149 JPEG2024539509000027.jpg177149 JPEG2024539509000028.jpg177149 JPEG2024539509000029.jpg119149

[0173] [Table 8] JPEG2024539509000031.jpg209149 JPEG2024539509000032.jpg209149 JPEG2024539509000033.jpg208149 JPEG2024539509000034.jpg208149 JPEG2024539509000035.jpg198149 JPEG2024539509000036.jpg208149 JPEG2024539509000037.jpg208149 JPEG2024539509000038.jpg208149 JPEG2024539509000039.jpg208149 JPEG2024539509000040.jpg208149 JPEG2024539509000041.jpg171149

[0174] Table 9 below shows the polypeptide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of H01. Table 10 below shows the nucleotide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of H01.

[0175] [Table 9]

[0176] [Table 10]

[0177] Table 11 below shows the H01 heavy and light chain CDR sequences.

[0178] [Table 11]

[0179] Example 2. Preparation of a novel antibody with two Fc domains using pertuzumab as a template M3 is characterized by a (Trastuzumab Fab)-(Fc)2 structure with the mutations VH G44C and VL Q100C, hereafter referred to as H01. Similarly, a (Pertuzumab Fab)-(Fc)2 structure with the mutations VH G44C and VL Q100C, based on the VH and VL regions of Pertuzumab (SEQ ID NOs: 27 and 28), hereafter referred to as P01. For P01, expression vectors containing sequences corresponding to Fc-hole (SEQ ID NO: 7), PerH-G44C-knob (SEQ ID NO: 29) and PerL-Q100C-knob (SEQ ID NO: 30) were co-transfected into EXPICHO-S™ (Gibco, A29127) and purification and analysis were carried out in the same manner as described in Example 1.

[0180] [Table 12]

[0181] [Table 13] JPEG2024539509000047.jpg208149 JPEG2024539509000048.jpg55149

[0182] Table 14 below shows the polypeptide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of P01. Table 15 below shows the nucleotide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of P01.

[0183] [Table 14]

[0184] [Table 15]

[0185] Table 16 below shows the CDR sequences in the heavy and light chains of P01.

[0186] [Table 16]

[0187] Example 3. Analysis of H01 and P01 by-products by papain digestion Papain recognizes a specific sequence in the hinge region and induces antibody digestion. In the case of the Fab-(Fc)2 structure, when papain digestion is performed, the Fab-(Fc)2 structure is cleaved into a Fab portion of approximately 49.3 kDa and two Fc domains of approximately 50.4 kDa (Figure 8a). However, when an abnormal disulfide bond is formed in the hinge region, unwanted interchain disulfide bond by-products can be observed (Figure 8b). In this case, a Fab fragment of approximately 49.3 kDa and an abnormal (Fc)2 product of approximately 100.7 kDa can be observed (Figure 8b).

[0188] To verify this, papain digestion of H01 and P01 was performed. Papain (Sigma, P3125) was used by diluting papain to 0.1 mg / mL in digestion buffer (20 mM EDTA + 10 mM Cys-HCl in PBS pH 7.4). 200 μg of H01 and P01 were digested for 2 hours at 37° C., and then SDS-PAGE was performed. As a result of SDS-PAGE performed under non-reducing conditions, no abnormal (Fc)2 at about 100 kDa was identified (FIGS. 8c and 8d).

[0189] Example 4. Analysis of physical properties of H01wt In H01, due to the knob-into-hole mutations, four Fc monomers associate into two Fc dimers, resulting in the structure shown in FIG. 6a. To analyze the effect of the knob-into-hole mutations on the formation of the H01 structure, the Fc hole monomer polypeptide (SEQ ID NO: 7) was replaced with a polypeptide corresponding to the wild-type IgG1 Fc monomer (SEQ ID NO: 39) (Table 17). The two knob polypeptides (SEQ ID NOs: 14 and 15) that make up H01 were also replaced with polypeptides corresponding to the wild-type IgG1 Fc monomer (SEQ ID NOs: 40 and 41) (Table 17). This new antibody format, consisting of two wtFc polypeptides (SEQ ID NO: 39), one TraH-G44C-wtFc polypeptide (SEQ ID NO: 40), and one TraL-Q100C-wtFc polypeptide (SEQ ID NO: 41), is referred to as H01wt (FIG. 9).

[0190] Expression vectors containing sequences corresponding to TraH-wtFc (SEQ ID NO: 39), TraH-G44C-wtFc (SEQ ID NO: 40), and TraL-Q100C-wtFc (SEQ ID NO: 41) were cotransfected into EXPICHO-S™ (Gibco, A29127), and purification and analysis were performed in the same manner as described in Example 1. SDS-PAGE analysis under non-reducing conditions (NR) identified a small amount of H01wt at approximately 150 kDa, and most of H01wt was expressed as a by-product with an abnormal structure (Figure 9). In the case of H01 with knob-into-hole mutations in the Fc region, each polypeptide efficiently assembled into a product identified at approximately 150 kDa (Figure 9).

[0191] [Table 17]

[0192] Table 18 below shows the nucleotide sequences encoding the wtFc of HO1wt, TraH-G44C-wtFc, and TraL-Q100C-wtFc.

[0193] [Table 18] JPEG2024539509000054.jpg208149 JPEG2024539509000055.jpg60149

[0194] Example 5. Characterization of H01Fv variants A schematic diagram of Fv-(Fc)2, in which two Fc domains are fused in parallel to an antibody Fv fragment, is shown in Figures 10a-10g. Fv consists of a VH domain and a VL domain. To improve domain interactions at the domain interface, disulfide bonds were artificially formed by substituting amino acids at specific positions with cysteines (Figures 10a-10h, Table 19).

[0195] [Table 19]

[0196] Table 20 shows the polypeptide sequences that constitute H01Fv1 to H01Fv7.

[0197] [Table 20] JPEG2024539509000058.jpg193149 JPEG2024539509000059.jpg193149 JPEG2024539509000060.jpg145149

[0198] Table 21 shows the nucleotide sequences encoding the polypeptides that constitute H01Fv1 to H01Fv7.

[0199] [Table 21] JPEG2024539509000062.jpg209149 JPEG2024539509000063.jpg208149 JPEG2024539509000064.jpg198149 JPEG2024539509000065.jpg208149 JPEG2024539509000066.jpg208149 JPEG2024539509000067.jpg209149 JPEG2024539509000068.jpg208149 JPEG2024539509000069.jpg209149 JPEG2024539509000070.jpg208149 JPEG2024539509000071.jpg66149

[0200] When expression vectors containing sequences corresponding to Fc-hole-RF (SEQ ID NO: 789), H01Fv1-HC (SEQ ID NO: 790), and H01Fv1-LC (SEQ ID NO: 791) were co-transfected into EXPICHO-S™ (Gibco, A29127), H01Fv1 was formed (Figure 10, Tables 20 and 21). H01Fv1 was then purified by affinity chromatography using MABSELECT™ Prism A (Cytiva, 17549853). The Fc-hole polypeptide (SEQ ID NO: 7) can form Fc-hole / Fc-hole dimers. To remove the Fc-hole / Fc-hole dimer, H435R and Y436F mutations were introduced into the Fc-hole polypeptide sequence (SEQ ID NO: 7) to generate the Fc-hole-RF polypeptide (SEQ ID NO: 789) (Figure 10, Tables 20 and 21). This Fc-hole-RF polypeptide prevents the Fc-hole / Fc-hole dimer from binding to Protein A resin and removes the mispaired Fc-hole / Fc-hole dimer (Figures 10 and 11). SDS-PAGE analysis identified a major product at approximately 130 kDa under non-reducing conditions (NR) (Figure 11) and the monomeric purity was determined by SEC (Figure 12). H01Fv3 exists mostly in an unpaired form of approximately 65 kDa, and the monomeric purity of the fully assembled form was determined to be 14.66% (Figures 11 and 12c). H01Fv1, H01Fv2, H01Fv4, H01Fv5, H01Fv6, and H01Fv7 were found to have monomeric purities of 71.24%, 61.25%, 68.55%, 73.05%, 67.73%, and 79.33%, respectively (Figure 12). The binding characteristics of H01Fv1, H01Fv2, H01Fv4, H01Fv5, H01Fv6, and H01Fv7 were analyzed using Biolayer Interferometry (BLI) Octet Red 96e (Sartorius) (Figure 13).Human HER2 recombinant protein (R&D systems, 1129-ER) was loaded onto an anti-pentaHIS (HIS1K) biosensor (Sartorius, 18-5120), and the binding constants of H01Fv1, H01Fv2, H01Fv4, H01Fv5, H01Fv6, and H01Fv7 were then calculated (Figure 13, Table 22).

[0201] [Table 22]

[0202] Example 6. Thermal stability analysis Thermostability analysis was performed using the PROTEIN THERMAL SHIFT™ dye kit (Applied biosystems, 4461146) according to the manufacturer's manual. Briefly, 5 μL of reaction buffer and 2.5 μL of dye included in the kit were mixed with 5 μg of trastuzumab, pertuzumab, H01, or P01, and the final volume was adjusted to 20 μL using PBS.

[0203] The mixture was incubated at 20°C for 30 seconds in a C1000 thermal cycler (Bio-Rad, 1841000) equipped with a CFX96 optical reaction module (Bio-Rad, 1845096), and the fluorescence intensity of the plate was measured while the temperature was increased from 20°C to 95°C at 1°C / min. The reaction was stopped after incubation at 95°C for 30 seconds. After the reaction, the median relative fluorescence unit (RFU) value was calculated and the melting temperature (T m ) were analyzed. m1 The values ​​were found to be 68, 68, 66, and 66° C., and T m2 The T values ​​were found to be 81, 79, 83, and 83°C, indicating that H01 and P01 have T values ​​similar to those of commercialized therapeutic antibodies. m It was shown that the α-amino acid sequence was significantly increased in the α-amino acid sequence (Figure 14, Table 23).

[0204] [Table 23]

[0205] Example 7. Identification of competitive binding of H01 and P01 To determine whether H01 and P01 bind to different epitopes or compete for binding, biolayer interferometry (BLI) with Octet Red 96e (Sartorius) was used.

[0206] Human HER2 recombinant protein (R&D systems, 1129-ER) was loaded onto an anti-penta-HIS (HIS1K) biosensor (Sartorius, 18-5120). 100 nM human IgG1 (Bio X cell, BE0297) or 100 nM H01 or 100 nM trastuzumab was first bound to each biosensor carrying HER2 antigen, followed by 100 nM human IgG1, 100 nM P01, or 100 nM pertuzumab to determine whether the antibodies competitively bound (Figure 15). The binding signals (nm shift from baseline) measured at equilibrium after completion of Her2 recombinant protein loading were 0.620, 0.625, and 0.672 nm, respectively (Figure 15, Table 24).

[0207] The first and second analytes were bound sequentially with binding and dissociation times of 900 seconds. When the human IgG1 antibody was bound sequentially, it did not bind to HER2 (Figure 15, Table 24). Sequential binding of H01 and P01, or trastuzumab and pertuzumab was observed, indicating that the antibodies bind to different epitopes (Figure 15, Table 24). The binding signals (nm shift from baseline) of H01+P01 and trastuzumab+pertuzumab on the HER2-loaded sensor were measured to be 1.477 nm (=y-axis value at 4140 s-y-axis value at 1260 s) and 0.923 nm (=y-axis value at 4140 s-y-axis value at 1260 s), respectively. The binding signal (nm shift from baseline) tends to increase as more protein binds to the biosensor surface, thus indicating that for the same amount of HER2, H01 and P01 induce a greater amount of antibody binding than Trastuzumab and Pertuzumab (Figure 15, Table 24).

[0208] [Table 24]

[0209] Example 8. Quantitation of Fc loading When H01 and P01 are treated in combination, a total of 16 Fc domains bind to the four HER2 antigens present on the surface of cancer cells (Figure 16a). When trastuzumab and pertuzumab are treated in combination, eight Fc domains bind to the four HER2 antigens present on the surface of cancer cells when the binding is in a monovalent manner, and fewer Fc domains can bind when the binding is in a bivalent manner (Figure 16b). The combination of H01 and P01 should result in increased Fc loading on the surface of HER2-positive cancer cells than the combination of trastuzumab and pertuzumab, which should lead to stronger effector functions.

[0210] NCI-N87, BT474, SK-OV-3, SNU1, and SNU5 cancer cell lines used to quantify Fc-bearing of antibodies on the surface of HER2-expressing cells were cultured in RPMI-1640 + 10% FBS medium. Cancer cell lines were treated with 50 nM human IgG1 (Bio X cell, BE0297), 50 nM trastuzumab (TRA), or 50 nM trastuzumab + 50 nM pertuzumab (TRA + PER), 50 nM H01, 50 nM H01 + 50 nM P01 antibodies for 30 minutes at 4°C in a 96-well plate. The cancer cell lines were then treated with Alexa 488 fluorescently conjugated anti-human IgG Fcγ Fab antibody (Jackson ImmunoResearch, 109-547-008), and the antibody Fc bound to the cells was quantified using a flow cytometer (BD biosciences, FACSverse) (Figures 17a-17e, Table 25). In the five cancer cell lines analyzed, the fluorescence intensity of the 50 nM H01 alone group was found to be higher than that of the 50 nM trastuzumab + 50 nM pertuzumab (TRA + PER) combination group (Figures 17a-17e, Table 25). When 50 nM P01 was treated in combination with 50 nM H01, an additive increase in Fc loading on the cancer cell surface was observed in NCI-N87, BT474, SK-OV-3, SNU1, and SNU5 cancer cells compared to treatment with 50 nM H01 alone (Figures 17a-17e, Table 25).

[0211] [Table 25]

[0212] To determine the saturation concentration of antibody binding to the cell surface, each test antibody was bound to a final concentration of 20, 50, and 100 nM, and subsequent sampling steps were performed in the same manner as the experimental conditions above (Figures 18a-18e). The saturation concentration of antibody in NCI-N87, BT474, SK-OV-3, SNU1, and SNU5 cancer cell lines was found to be 50 nM (Figures 18a-18e). Treatment with 50 nM H01 alone was found to result in more Fc loading on the surface of the five cell lines than treatment with 50 nM trastuzumab and 50 nM pertuzumab in combination (Figures 18a-18e).

[0213] Example 9. Analysis of antibody binding affinity to HER2 S239D and I332E mutations in antibody Fc domains improve the affinity of antibodies for Fcγ receptors, which leads to improved effector function (Greg A. Lazar et al., PNAS, 2006). H01DE4 and P01DE4 were designed by introducing S239D and I332E mutations into the H01 and P01 Fc domains (Figures 19a and 19b). H01DE4 was prepared by cotransfection of vectors capable of expressing polypeptides corresponding to Fc-hole-S239D-I332E (Table 26, SEQ ID NO: 42), TraH-G44C-knob-S239D-I332E (Table 26, SEQ ID NO: 43), and TraL-Q100C-knob-S239D-I332E (Table 26, SEQ ID NO: 44) into EXPICHO-S™ (Gibco, A29127). P01DE4 was prepared by cotransfection of vectors capable of expressing polypeptides corresponding to Fc-hole-S239D-I332E (Table 26, Table 27, and SEQ ID NO: 42), PerH-G44C-knob-S239D-I332E (Table 26, Table 27, and SEQ ID NO: 45), and PerL-Q100C-knob-S239D-I332E (Table 26, Table 27, and SEQ ID NO: 46) into EXPICHO-S™ (Gibco, A29127).

[0214] [Table 26] JPEG2024539509000077.jpg129149

[0215] [Table 27] JPEG2024539509000079.jpg208149 JPEG2024539509000080.jpg208149 JPEG2024539509000081.jpg208149 JPEG2024539509000082.jpg34149

[0216] Binding characteristics were analyzed at 25 °C using Biolayer Interferometry (BLI), Octet Red 96e (Sartorius). The buffer used for the analysis was 10x kinetics buffer (Sartorius, 18-1042) diluted in PBS pH 7.4 (Gibco, 10010) and the assay plate was stirred at 1,000 rpm. Human HER2 recombinant protein (R&D systems, 1129-ER) was loaded onto an anti-penta-HIS (HIS1K) biosensor (Sartorius, 18-5120). The binding rate constant (K a To measure the dissociation rate constant (K), 1-32 nM of H01, H01DE4, P01, and P01DE4 were bound to the supported antigen for 300 s, followed by dissociation for 600 s in kinetics buffer. d ) was determined. a and K d Values ​​were determined by a 1:1 binding model in Octet analysis software (Sartorius) and the equilibrium dissociation constants (K D ) values ​​were determined (Figures 20a-20d, Table 28).

[0217] [Table 28]

[0218] Example 10. Analysis of antibody binding affinity to Fcγ receptors The binding constants of each antibody to Fcγ receptors at 25°C were analyzed using Octet Red 96e (Sartorius).Anti-PentaHIS (HIS1K) biosensor (Sartorius, 18-5120) was used, and human FcγRI (R&D systems, 1257-FC) or human FcγIIA (R&D systems, 1330-CD) or human FcγRIIIA (176V isoform, R&D systems, 4325-FC) containing His tags were loaded onto the biosensor.

[0219] The association rate constants (K a ) and dissociation rate constant (K d ) was determined. a and K d Values ​​were calculated using a 1:1 binding model in the Octet analysis software (Sartorius) and the equilibrium dissociation constants (K D ) values ​​were determined (Figures 21a-21h, Figures 22a-22h, Figures 23a-23h, Table 29).

[0220] [Table 29]

[0221] Example 11. Pharmacokinetic (PK) Analysis in Rats H01, P01, trastuzumab, and pertuzumab were administered at 10 mg / kg via intravenous (iv) route to 7-week-old male Sprague-Dawley rats (ORIENT BIO INC.). Blood samples were collected at 5 min, 15 min, 30 min, 1 h, 2 h, 4 h, 8 h, 1 day, 2 days, 3 days, 7 days, 10 days, 14 days, 21 days, 28 days, 35 days, and 42 days after administration. Then, serum only was separated from the blood for analysis. Antibody concentrations in serum were measured by ELISA.

[0222] Briefly, 96-well ELISA plates (Corning, 3590) were coated with human HER2 recombinant protein (R&D systems, 1129-ER) and stored overnight at 4°C, and serum obtained at each time point was appropriately diluted and allowed to bind to the coated human HER2. Peroxidase-conjugated anti-human Fab goat antibody (Invitrogen, 31482) was used to detect H01, P01, trastuzumab, and pertuzumab.

[0223] Standard samples of H01, P01, trastuzumab, and pertuzumab were prepared and the concentration of the analyte at each time point was quantified based on a standard curve generated from natural rat serum containing various concentrations of antibody standards. The half-lives of intravenously administered H01, P01, trastuzumab, and pertuzumab were determined to be approximately 11.8 days, 14.2 days, 7.3 days, and 11.6 days, respectively (Figures 24a and 24b, Table 30). The engineered novel antibodies H01 and P01 were found to have similar PK parameters to the parent humanized antibodies trastuzumab and pertuzumab.

[0224] [Table 30]

[0225] Example 12. Design, preparation, and analysis of a novel antibody that recognizes two epitopes of HER2 To construct an antibody that recognizes two epitopes of HER2 protein, biparatopic antibody HP501 was designed by connecting the V domains of trastuzumab and pertuzumab via a linker (Figure 25). To minimize the decrease in binding affinity caused by the interference between different V domains, linkers with variable lengths connecting the V domains were tested. At the same time, in an attempt to improve the physical integrity of the antibody, 16 variants were designed in which Cys substitution mutations that can form disulfide bonds in the V domain were introduced (Figure 26, Table 31). According to the Kabat numbering system, mutations were only introduced at position 44 for the heavy chain and position 100 for the light chain (Table 31). In Table 31, VH linkers having 6 amino acid residues are designated as VH-S-linkers, VH linkers having 13 amino acid residues are designated as VH-L-linkers, VL linkers having 6 amino acid residues are designated as VL-S-linkers, and VL linkers having 13 amino acid residues are designated as VL-L-linkers.

[0226] [Table 31]

[0227] For HP501, an expression vector consisting of sequences corresponding to Fc-hole (SEQ ID NO: 7), TH-S-PH-knob (SEQ ID NO: 51), and TL-S-PL-knob (SEQ ID NO: 59) was co-transfected into EXPICHO-S™ (Gibco, A29127) (Tables 31, 32, 33, and 34), and purification and analysis were performed in the same manner as described in Example 1. Expression, purification, and analysis were performed for HP502 to HP516 in the same manner as mentioned above, and the composition of the expression vectors is detailed in Tables 32 and 34.

[0228] [Table 32] JPEG2024539509000088.jpg203149 JPEG2024539509000089.jpg209149 JPEG2024539509000090.jpg209149 JPEG2024539509000091.jpg209149 JPEG2024539509000092.jpg198149

[0229]

Table 33

[0230] [Table 34]

[0231] Purity was analyzed by size exclusion chromatography (Figure 27, Table 35) in the same manner as described in Example 1. Analysis showed that HP503, HP507, HP511, and HP515, in which the first V domain is wild type and the second V domain is a Cys-substituted variant (VH 44C, VL 100C), had excellent purity (Figure 27, Table 35).

[0232] [Table 35]

[0233] The binding constants of HP501, HP502, HP503, HP504, HP505, HP506, HP507, HP508, HP509, HP510, HP511, HP512, HP513, HP514, HP515, and HP516 to the D2 and D4 regions of the HER2 protein were determined using Octet Red 96e (Sartorius). To analyze the binding constants of 16 antibodies to the D2 region, human HER2 recombinant protein (R&D systems, 1129-ER) was loaded onto an anti-pentaHIS (HIS1K) biosensor (Sartorius, 18-5120) and then saturated with 100 nM trastuzumab, which targets the D4 region. The 16 antibodies were then added at a concentration of 100 nM to the binding reaction (300 seconds) and dissociation reaction (600 seconds), and the affinity of the antibodies to the D2 region was calculated (Table 36). To analyze the binding constants of the 16 antibodies to the D4 region, human HER2 recombinant protein (R&D systems, 1129-ER) was loaded onto an anti-penta HIS (HIS1K) biosensor (Sartorius, 18-5120), and then saturated with 100 nM pertuzumab targeting the D2 region. The 16 antibodies were then added at a concentration of 100 nM to the binding reaction (300 seconds) and dissociation reaction (600 seconds), and the affinity of the antibodies to the D4 region was calculated (Table 36). The binding constants of HP507, HP511, and HP515 to the D2 region were 2.285, 3.267, and 2.012 nM, respectively, demonstrating superior binding affinity to the D2 region compared to other clones (Table 36). HP503 had a binding constant of 8.098 nM to the D2 region, indicating a relatively low binding ability to the D2 region compared to HP507, HP511, and HP515. However, when the binding constants to the D4 region were measured, the binding constants of HP503, HP507, HP511, and HP515 were found to be 0.181, 0.228, 0.162, and 0.227 nM, respectively, demonstrating strong binding affinity (Table 36).

[0234] [Table 36]

[0235] The binding constants of HP503, HP507, HP511, and HP515 to the HER2 extracellular domain (ECD) were measured using Octet Red 96e (Sartorius). Human HER2 recombinant protein (R&D systems, 1129-ER) was loaded onto an anti-pentaHIS (HIS1K) biosensor (Sartorius, 18-5120). Concentrations of 0.25, 0.5, 1, 2, 4, or 8 nM of HP503 or HP507 or HP511 or HP515 were added to the binding reaction (600 s) and dissociation reaction (1,800 s) on the HER2 protein-loaded sensor, and the binding constants were calculated (Figures 28a-28d, Table 37). HP507, HP511, and HP515 were found to have dissociation constants (<1.0E-07 1 / s) above the instrument's detection limits under the following analytical conditions (Figures 28a-28d, Table 37).

[0236] [Table 37]

[0237] The measured binding constants of HP503, HP507, HP511, and HP515 to Fcγ receptors were analyzed using Octet Red 96e (Sartorius) in the same manner as described in Example 10 (Figures 29a-29d, Table 38). The analysis showed that HP503, HP507, HP511, and HP515 have superior binding affinity to FcγRI (CD64), FcγRIIA (CD32A, 131R), and FcγRIIIA (CD16A, 176V) compared to human IgG1, trastuzumab, pertuzumab, and marjetuximab (Figures 21a-21h, Figures 22a-22h, Figures 23a-23h, Figures 29a-29d, and Tables 29 and 38).

[0238] [Table 38]

[0239] The binding constants of HP503, HP507, HP511, and HP515 to the neonatal Fc receptor (FcRn) were measured using Octet Red 96e (Sartorius). HP503, HP507, HP511, HP515, human IgG1 (Bio X cell, BE0297), trastuzumab, pertuzumab, and marjetuximab were loaded onto an anti-human Fab-CH1 2nd generation (FAB2G) biosensor (Sartorius, 18-5126), and the binding and dissociation times were set to be 120 seconds, respectively (Figures 30a-30b, Table 39). For the analysis, a kinetics buffer (Sartorius, 18-1105) was used at pH 6.0.

[0240] [Table 39]

[0241] Example 13. Analysis of complement-dependent cytotoxicity For the analysis of complement-dependent cytotoxicity, BT474 (HER2 3+; high) breast cancer cell line and NCI-N87 (HER2 3+; high) gastric cancer cell line were used. Cells were diluted in cell culture medium and dispensed at 10,000 cells per well in a 96-well plate. Cells, antibodies, and human serum (Sigma, H4522) were reacted at a volume ratio of 1:1:1, respectively.

[0242] To identify dose-response relationships, human IgG1, trastuzumab (TRA), trastuzumab + pertuzumab (TRA + PER), H01, and H01 + P01 were serially diluted 6 times in 2-fold increments from an initial concentration of 1200 nM, and reactions were performed starting at 400 nM (plus another 3-fold dilution if dispensed). Human serum was diluted and dispensed into culture medium to a final concentration of 25%, and the mixture of cells, antibodies, and human serum was incubated for 5 hours in a humidified incubator at 37°C and 5% (v / v) CO2.

[0243] Cell Titer Glo-Reagent (Promega, G9243), pre-dissolved at 4°C, was dispensed into each well in an equal volume of mixed culture medium, and then cell lysis was induced using a plate shaker (Allsheng, MX100-4A) with agitation at 500 rpm for 2 min. To stabilize the luminescence signal, the mixture was incubated at room temperature for 10 min and then analyzed using a plate reader instrument (Envision; PerkinElmer, 2105-0010) (Figures 31a-31b). Complement-dependent cytotoxicity activity (CDC activity) was calculated as follows: <Equation 1> CDC activity (%) = 100 x [1 - (luminescence with experimental antibody / luminescence without antibody)]

[0244] Human IgG1, trastuzumab (TRA), trastuzumab + pertuzumab (TRA + PER), and H01 did not induce CDC responses in BT474 and NCI-N87 cell lines (Figures 31a-31b). CDC was shown to be induced in both cell lines only when H01 was treated in combination with P01 (Figures 31a-31b).

[0245] Example 14. Analysis of antibody-dependent cellular cytotoxicity NCI-N87 (HER2 3+; high), MDA-MB-453 (HER2 2+; medium), SNU-601 (HER2 1+; low), and SNU-5 (HER2 1+; low) cancer cell lines were used for antibody-dependent cell-mediated cytotoxicity assays (Figures 32a, 32b, 32c, and 32d). Each cancer cell line was cultured at 1.0×10 4 Each antibody was then diluted in culture medium to an appropriate concentration and treated. Peripheral blood mononuclear cells (PBMCs) isolated on the same day were used as effector cells, with 1.5 × 10 5 PBMCs were treated with 15x the number of target cells (E:T ratio = 15:1) by treating with 15x the number of PBMCs / well.

[0246] After treatment, the cells were incubated for 18 hours in a humidified incubator at 37°C and 5% (v / v) CO2, and then cytotoxicity was measured using a cytotoxicity detection kit (LDH) (Roche, 11644793001) (Figures 32a, 32b, 32c, and 32d). Cytotoxicity was calculated using the following formula for antibody-dependent cellular cytotoxicity: <Equation 2> Cytotoxicity (%) = [(test release - spontaneous release) / (maximum release - spontaneous release)] x 100

[0247] H01 showed superior cytotoxicity at lower concentrations compared to trastuzumab in NCI-N87 (HER2 3+; high) and MDA-MB-453 (HER2 2+; medium) cancer cell lines (Figures 32a and 32b). Cytotoxicity analysis in SNU-601 (HER2 1+; low) and SNU-5 (HER2 1+; low) cancer cell lines H01 showed superior cytotoxicity of H01 compared to that of trastuzumab (Figures 32c and 32d).

[0248] Example 15. Efficacy evaluation in xenograft mouse models Efficacy in a SNU-5 (HER2 1+; low) gastric cancer cell line-derived xenograft model was evaluated in 6-week-old female SCID mice (CB-17 / NcrKoat-Prkdc scid The SNU-5 cancer cell line was evaluated using a 1×10 7 The cells were diluted in PBS at 100 cells / 100 μL and mixed 1:1 with MATRIGEL® Growth Factor Reduced (GFR) Basement Membrane Matrix (Corning, 354230), and 100 μL of the mixture was implanted subcutaneously in the right flank and tumor growth was monitored. The mean tumor volume was approximately 107 mm. 3Mice were regrouped as follows: 1) PBS (vehicle), 5 mg / kg H01, 5 mg / kg H01 + 5 mg / kg P01, 5 mg / kg HP507, 5 mg / kg trastuzumab, 5 mg / kg trastuzumab + 5 mg / kg pertuzumab were administered intravenously (IV) once a week for a total of 6 weeks (Figure 33a). Analysis showed that H01 alone exhibited superior antitumor activity compared to trastuzumab and trastuzumab + pertuzumab (Figure 33a). H01 + P01 induced improved antitumor activity compared to H01 alone, with HP507 shown to induce the most potent antitumor activity in the SNU-5 gastric cancer xenograft model (Figure 33a).

[0249] In addition, efficacy in a SNU-5 (HER2 1+; low) gastric cancer cell line-derived xenograft model was evaluated using 6-week-old female BALB / c-nu mice (ORIENT BIO INC.) (Figure 33b). The SNU-5 cancer cell line was incubated at 1 x 10 7 The cells were diluted in PBS at 100 μL / cell and mixed with Matrigel® growth factor reduced (GFR) basement membrane matrix (Corning, 354230) at a 1:1 ratio, and 100 μL of the mixture was subcutaneously implanted in the right flank and tumor growth was observed. The average tumor volume was approximately 122 mm 3 Mice were regrouped as follows: 1 mg / kg trastuzumab, 1 mg / kg pertuzumab, 0.5 mg / kg trastuzumab + 0.5 mg / kg pertuzumab, 1 mg / kg H01, 1 mg / kg P01, and 0.5 mg / kg H01 + 0.5 mg / kg P01 were administered intraperitoneally (IP) twice a week for a total of 6 weeks (Figure 33b). Analysis showed that H01, P01, and H01 + P01 induced superior antitumor activity compared to trastuzumab, pertuzumab, and trastuzumab + pertuzumab (Figure 33b).

[0250] Efficacy in a SNU-601 (HER2 1+; low) gastric cancer cell line-derived xenograft model was evaluated in 6-week-old female SCID mice (CB-17 / NcrKoat-Prkdc scid The SNU-5 cancer cell line was evaluated using 1 × 10 7 The cells were diluted in PBS at 100 μL / cell and mixed with Matrigel® growth factor reduced (GFR) basement membrane matrix (Corning, 354230) at a 1:1 ratio, and 100 μL of the mixture was subcutaneously implanted in the right flank and tumor growth was observed. The average tumor volume was approximately 142 mm 3 Mice were regrouped as follows: PBS (vehicle), 5 mg / kg H01, 5 mg / kg trastuzumab, and 5 mg / kg trastuzumab + 5 mg / kg pertuzumab were administered intraperitoneally (IP) twice a week for a total of 6 weeks (Figure 34). Since there are no antibodies in the blood of SCID mice, all mice were administered 50 mg / kg intravenous immunoglobulin (IVIG; LIV-r, SK Plasma) twice a week for 6 weeks to simulate a real human blood environment (Figure 34). In the SNU-601 gastric cancer xenograft model, H01 alone induced the most prominent antitumor activity (Figure 34).

[0251] Efficacy in an NCI-N87 (HER2 3+; high) gastric cancer cell line-derived xenograft model was evaluated in 6-week-old female SCID mice (CB-17 / NcrKoat-Prkdc scid The NCI-N87 cancer cell line was evaluated using 5 × 10 6 The cells were diluted in PBS at 100 μL / cell and mixed with Matrigel® growth factor reduced (GFR) basement membrane matrix (Corning, 354230) at a 1:1 ratio, and 100 μL of the mixture was subcutaneously implanted in the right flank and tumor growth was observed. The average tumor volume was approximately 146 mm 3Mice were regrouped as follows: 1) PBS (vehicle), 0.2 mg / kg H01, 5 mg / kg H01, 0.2 mg / kg trastuzumab, and 5 mg / kg trastuzumab were administered intraperitoneally (IP) twice a week for a total of 6 weeks (Figure 35). Since there are no antibodies in the blood of SCID mice, all mice were administered 50 mg / kg intravenous immunoglobulin (IVIG; LIV-r, SK Plasma) twice a week for 6 weeks to simulate a real human blood environment (Figure 35). Analysis showed that when the antibody was administered at 5 mg / kg and 0.2 mg / kg, H01 induced superior antitumor activity compared to trastuzumab (Figure 35).

[0252] Example 16. Evaluation of antitumor activity in the CT26-HER2 syngeneic mouse model The nucleotide sequence (SEQ ID NO:566, Table 40) encoding the human HER2 protein (SEQ ID NO:567, Table 40) was cloned into a protein expression vector (ORIGENE, PS100020) containing a neomycin resistance gene to construct the human HER2 expression vector pCMV6-AC-hHER2 (Figure 36, Table 40).

[0253] [Table 40] JPEG2024539509000119.jpg203149 JPEG2024539509000120.jpg43149

[0254] Mouse colon-derived CT26 cancer cells were transfected with pCMV6-AC-hHER2 human HER2 expression vector using Lipofectamine 2000 transfection reagent (Invitrogen, 11668-019). The transfected cells were incubated in culture medium containing 1 mg / mL G418 (Invivogen, ant-gn-5) for 14 days to select only cells transfected with pCMV6-AC-hHER2 human HER2 expression vector. The top 3% clones in terms of HER2 expression were sorted into 96-well plates (ThermoFisher, 167008) with one cell per well using a SH800S cell sorter (SONY). Selection was performed by incubation in G418-containing medium for 21 days, and a total of eight CT26 mouse colon cancer cell line clones expressing human HER2 were obtained, and human HER2 expression in these cells was monitored by flow cytometer (BD Biosciences, FACSverse) analysis after staining with anti-human HER2-BV421 (BD, 744811) (Figure 37, Table 41).

[0255] [Table 41]

[0256] After 6 passages in a G418-free environment for 20 days, cells were stained with anti-human HER2-BV421 (BD, 744811) to measure the level of human HER2 expression. The level of human HER2 expression was shown to be not reduced in cells grown without G418 compared to cells grown with G418 (Figure 38, Table 42).

[0257] [Table 42]

[0258] To compare cell surface Fc loading between parental CT26, CT26-HER2 cell line (clone #2-60), and human cancer cell lines (SNU5, SNU601, and NCI-N87), each cell was bound to 100 nM human IgG1 (Bio X cell, BE0297), 100 nM trastuzumab (TRA), and 100 nM H01 antibody in a 96-well v-bottom plate (Corning, 3363) for 30 minutes at 4° C. The antibodies were then treated with Alexa 488 fluorescent conjugated anti-human IgG Fcγ Fab antibody (Jackson ImmunoResearch, 109-547-008), and Fc loading on the cells was quantified using a flow cytometer (FIG. 39, Table 43). The CT26-HER2 cell line (clone #2-60) was shown to express human HER2 at a level similar to SNU5 (Figure 39, Table 43). In addition, in the CT26-HER2 cell line, treatment with 100 nM H01 was shown to result in increased Fc bearing on the cell surface compared to that of 100 nM trastuzumab (TRA).

[0259] [Table 43]

[0260] Efficacy in the CT26-HER2 (clone #2-60) syngeneic mouse model was evaluated using 6-week-old female Balb / c mice (ORIENT BIO INC.). PBS (vehicle), 5 mg / kg trastuzumab, and 5 mg / kg H01 were administered intraperitoneally (IP) twice a week for a total of 2 weeks (Figure 40). Analysis showed that H01 induced superior antitumor activity compared to trastuzumab (Figure 40).

[0261] Example 17. Design, preparation, and analysis of novel antibodies targeting glypican-3 (GPC3) The variant light and heavy chain polypeptide sequences of the antibody that specifically recognizes the glypican-3 (GPC-3) protein are shown in Table 44. For GPM01, an expression vector consisting of sequences corresponding to Fc-hole (SEQ ID NO: 7), GPM01 HC (SEQ ID NO: 67), and GPM01 LC (SEQ ID NO: 68) was co-transfected into EXPICHO-S™ (Gibco, A29127) (Figure 41a, Table 44), and purification and analysis were performed in the same manner as described in Example 1. Expression, purification, and analysis were performed for GPM02, GPM04, GPB01, GPB03, GPB04, and GPB06 in the same manner as mentioned above (Figures 41a-41b, Table 44). GPM01, GPM02, and GPM04 bind monovalently to different epitopes of the antigen and have a structure consisting of two Fc domains (Figure 41a). GPB01, GPB03, GPB04, and GPB06 have a structure in which the variable regions of GPM01, GPM02, and GPM04 are linked by a polypeptide linker (sequence number 48, sequence number 50), bind to GPC-3 biparatopic, and have two Fc domains (Figure 41b).

[0262] [Table 44] JPEG2024539509000125.jpg209149 JPEG2024539509000126.jpg209149 JPEG2024539509000127.jpg193149 JPEG2024539509000128.jpg150149

[0263] [Table 45] JPEG2024539509000130.jpg208149 JPEG2024539509000131.jpg208149 JPEG2024539509000132.jpg208149 JPEG2024539509000133.jpg208149 JPEG2024539509000134.jpg208149 JPEG2024539509000135.jpg208149 JPEG2024539509000136.jpg208149 JPEG2024539509000137.jpg208149 JPEG2024539509000138.jpg208149 JPEG2024539509000139.jpg208149 JPEG2024539509000140.jpg208149 JPEG2024539509000141.jpg208149 JPEG2024539509000142.jpg208149 JPEG2024539509000143.jpg23149

[0264] Table 46 below shows the polypeptide sequences of the heavy chain variable region (VH) and the light chain variable region (VL) of the engineered antibody that targets GPC-3. Table 47 below shows the nucleotide sequences of the heavy chain variable region (VH) and the light chain variable region (VL) of the engineered antibody that targets GPC-3.

[0265] [Table 46] JPEG2024539509000145.jpg195149

[0266] [Table 47] JPEG2024539509000147.jpg203149 JPEG2024539509000148.jpg202149 JPEG2024539509000149.jpg90149

[0267] Table 48 below shows the heavy and light chain CDR sequences of engineered antibodies targeting GPC-3.

[0268] [Table 48] JPEG2024539509000151.jpg153149

[0269] The GPC-3 protein binding constants of GPM01, GPM02, GPM04, GPB01, GPB03, GPB04, and GPB06 were determined using Octet Red 96e (Sartorius). To analyze the binding constants of the seven antibodies, human GPC-3 recombinant protein (Sino Biologicals, 10088-H08H) was loaded onto an anti-pentaHIS (HIS1K) biosensor (Sartorius, 18-5120). The seven antibodies were then added at various concentrations to the binding reaction (300 seconds) and dissociation reaction (1,200 seconds), and the affinity of the antibodies to GPC-3 was calculated (Figure 42, Table 49). Table 49 below illustrates the binding constants of engineered antibodies targeting GPC-3.

[0270] [Table 49]

[0271] HepG2 liver cancer cell line was used to quantify Fc loading on the surface of GPC-3 expressing cells. 100 nM human IgG1, GPM02, GPB01, GPB03, and GC33 were bound to HepG2 cell line at 4° C. for 30 minutes, and Fc loading was quantified using Alexa 488 fluorescent conjugated anti-human IgG Fcγ Fab antibody (Jackson ImmunoResearch, 109-547-008) (FIG. 43). GC33, a humanized antibody targeting GPC-3, was used as a positive control (Nakano et al., U.S. Patent No. 7,919,086). Higher Fc loading on the surface of cancer cells was shown to be induced by treatment of GPM02, GPB01, and GPB03 compared to GC33 (FIG. 43).

[0272] Example 18. Design, preparation, and analysis of antibody constructs targeting EPH receptor A2 (EphA2) The variant light and heavy chain polypeptide sequences of antibodies that specifically bind to EPH receptor A2 (EphA2) protein are shown in Table 46. For EPB01, an expression vector consisting of sequences corresponding to Fc-hole (SEQ ID NO: 7), EPB01 HC (SEQ ID NO: 111), and EPB01 LC (SEQ ID NO: 112) was co-transfected into EXPICHO-S™ (Gibco, A29127) (Figure 41b, Table 50), and purification and analysis were performed in the same manner as described in Example 1. Expression, purification, and analysis were performed for EPB02, EPB03, EPB04, EPB05, EPB06, EPB07, EPB08, EPB09, EPB10, EPB11, and EPB12 in the same manner as mentioned above (Table 50). The 12 antibodies have a structure in which variable regions that bind to two different epitopes of EphA2 are linked by a peptide linker (sequence number 48, sequence number 50), bind to EphA2 biparatopic, and have two Fc domains (Figure 41b, Table 50).

[0273] [Table 50] JPEG2024539509000154.jpg200149 JPEG2024539509000155.jpg201149 JPEG2024539509000156.jpg203149 JPEG2024539509000157.jpg203149 JPEG2024539509000158.jpg203149 JPEG2024539509000159.jpg44149

[0274]

Table 51

[0275] Table 52 below shows the polypeptide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of engineered antibodies that target EphA2. Table 53 below shows the nucleotide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of engineered antibodies that target EphA2.

[0276] [Table 52] JPEG2024539509000180.jpg199149 JPEG2024539509000181.jpg161149

[0277] [Table 53] JPEG2024539509000183.jpg203149 JPEG2024539509000184.jpg204149 JPEG2024539509000185.jpg204149 JPEG2024539509000186.jpg202149 JPEG2024539509000187.jpg205149 JPEG2024539509000188.jpg204149 JPEG2024539509000189.jpg113149

[0278] Table 54 below shows the heavy and light chain CDR sequences of engineered antibodies that target EphA2.

[0279] [Table 54] JPEG2024539509000191.jpg204149 JPEG2024539509000192.jpg202149 JPEG2024539509000193.jpg152149

[0280] The EphA2 protein binding constants of EPB01, EPB02, EPB03, EPB04, EPB05, EPB06, EPB07, EPB08, EPB09, EPB10, EPB11, and EPB12 were determined using Octet Red 96e (Sartorius). To analyze the binding constants of the 12 biparatopic antibodies, human EphA2 recombinant protein (Sino Biologicals, 13926-H08H) was loaded onto an anti-pentaHIS (HIS1K) biosensor (Sartorius, 18-5120), and then the 12 antibodies were added at various concentrations for binding (300 seconds) and dissociation (1,200 seconds) reactions. Based on the above, the affinity for EphA2 was calculated (Table 55). Table 55 below illustrates the results obtained by analyzing the binding constants of engineered antibodies targeting EphA2.

[0281] [Table 55]

[0282] The ability of the antibodies to inhibit EphA2 signaling was analyzed using the PC-3 prostate cancer cell line. PC-3 cancer cell lysates treated with each antibody at a concentration of 50 nM for 30 minutes were analyzed by Western blot. 1C1 humanized antibody targeting human EphA2 was used as a positive control and was prepared based on the sequence published in the literature (Kinch et al., US Patent Publication No. 20090304721). Akt rabbit mAb (Cell Signaling Technology, 9272), phospho-Akt (Ser473) (D9E) XP® rabbit mAb (Cell Signaling Technology, 4060), and β-actin (13E5) rabbit mAb (Cell Signaling Technology, 4970) were used as primary antibodies for the analysis, and anti-rabbit IgG, HRP-linked antibody (Cell Signaling Technology, 7074) was used as the secondary antibody. In analysis of AKT signaling pathway inhibition, EPB02, EPB03, and EPB05 showed similar levels of inhibition as the positive control 1C1 (FIG. 44).

[0283] PC-3 prostate cancer cell line was used to quantify Fc loading on the surface of EphA2 expressing cells. 100 nM antibody was bound to PC-3 cell line for 30 minutes at 4° C., and Fc loading was quantified using Alexa 488 fluorescent conjugated anti-human IgG Fcγ Fab antibody (Jackson ImmunoResearch, 109-547-008) (FIG. 44). Higher Fc loading on the surface of cancer cells was shown to be induced by treatment with EPB02, EPB03, EPB05, EPB06, EPB07, and EPB10 compared to treatment with 1C1 humanized antibody targeting EphA2 (FIG. 45).

[0284] Example 19. Design, preparation, and analysis of antibody constructs targeting MET The variant light and heavy chain polypeptide sequences of antibodies that specifically bind to MET protein are shown in Table 56. For MEM01, expression vectors consisting of sequences corresponding to Fc-hole (SEQ ID NO: 7), MEM01 HC (SEQ ID NO: 568), and MEM01 LC (SEQ ID NO: 569) were co-transfected into EXPICHO-S™ (Gibco, A29127) (Figure 41a, Table 56) and purification and analysis were performed in the same manner as described in Example 1. Expression, purification, and analysis were performed for MEM06 in the same manner as mentioned above (Table 56, Table 57).

[0285] [Table 56] JPEG2024539509000196.jpg60149

[0286] [Table 57] JPEG2024539509000198.jpg208149 JPEG2024539509000199.jpg208149 JPEG2024539509000200.jpg124149

[0287] Table 58 below shows the polypeptide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of engineered antibodies that target MET. Table 59 below shows the nucleotide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of engineered antibodies that target MET.

[0288] [Table 58]

[0289] [Table 59]

[0290] Table 60 below shows the heavy and light chain CDR sequences of engineered antibodies that target MET.

[0291] [Table 60]

[0292] The MET protein binding constants of MEM01 and MEM06 were determined using Octet Red 96e (Sartorius). To analyze the binding constants of the antibodies, the antibodies were loaded onto an anti-human Fab-CH1 second generation (FAB2G) biosensor (Sartorius, 18-5125). Human MET recombinant protein (Sino Biologicals, 10692-H08H) was then added at various concentrations to the binding reaction (300 seconds) and dissociation reaction (600 seconds) (Figure 46), and the affinity of the antibodies to MET was calculated (Figure 46, Table 61). Table 61 below illustrates the binding constants of engineered antibodies targeting MET.

[0293] [Table 61]

[0294] MKN45 and SNU-5 gastric cancer cell lines were used to quantify Fc loading on the surface of MET-expressing cells. Human IgG1 control, onartuzumab (produced in a CHO cell line), emibetuzumab, MEM01, and MEM06 antibodies were bound to the cell lines for 30 minutes at 4° C., and Fc loading was quantified using an Alexa 488 fluorescent conjugated anti-human IgG Fcγ Fab antibody (Jackson ImmunoResearch, 109-547-008) (FIGS. 47 and 48). High Fc loading on the surface of MET-expressing cancer cells was shown to be induced by treatment with MEM01 compared to onartuzumab and emibetuzumab, which target MET (FIGS. 47 and 48). Treatment with MEM06 induced higher Fc loading on the surface of MET-expressing cancer cells compared to emibetuzumab, but similar levels when compared to onartuzumab (FIGS. 47 and 48).

[0295] Example 20. Design, preparation, and analysis of antibody constructs targeting EGFR The variant light and heavy chain polypeptide sequences of antibodies that specifically bind to EGFR protein are shown in Table 62. For EGM01, expression vectors consisting of sequences corresponding to Fc-hole (SEQ ID NO: 7), EGM01 HC (SEQ ID NO: 590), and EGF01 LC (SEQ ID NO: 591) were co-transfected into EXPICHO-S™ (Gibco, A29127) (Figure 41a, Table 62) and purification and analysis were performed in the same manner as described in Example 1. Expression, purification, and analysis were performed for EGM02, EGM03, EGM04, EGM05, and EGM06 in the same manner as mentioned above (Table 62).

[0296] [Table 62] JPEG2024539509000206.jpg209149 JPEG2024539509000207.jpg188149

[0297] Table 63 below shows the heavy and light chain nucleotide sequences of engineered antibodies that target EGFR.

[0298] [Table 63] JPEG2024539509000209.jpg208149 JPEG2024539509000210.jpg208149 JPEG2024539509000211.jpg208149 JPEG2024539509000212.jpg208149 JPEG2024539509000213.jpg208149 JPEG2024539509000214.jpg198149 JPEG2024539509000215.jpg176149 JPEG2024539509000216.jpg182149

[0299] Table 64 below shows the polypeptide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of engineered antibodies that target EGFR. Table 65 below shows the nucleotide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of engineered antibodies that target EGFR.

[0300] [Table 64]

[0301] [Table 65] JPEG2024539509000219.jpg113149

[0302] Table 66 below shows the heavy and light chain CDR sequences of engineered antibodies that target EGFR.

[0303] [Table 66]

[0304] The EGFR protein binding constants of EGM01-EGM05 were determined using Octet Red 96e (Sartorius). To analyze the binding constants of the antibodies, the antibodies were loaded onto an anti-human Fab-CH1 second generation (FAB2G) biosensor (Sartorius, 18-5125). Human EGFR recombinant protein (Sino Biologicals, 10692-H08H) was then added at various concentrations for the binding reaction (300 seconds) and dissociation reaction (600 seconds) (Figure 49), and the affinity of the antibodies to EGFR was calculated (Figure 49, Table 67). Table 67 below illustrates the binding constants of engineered antibodies targeting EGFR.

[0305] [Table 67]

[0306] Example 21. Design, preparation, and analysis of novel antibodies targeting CD33 The variant light and heavy chain polypeptide sequences of the antibody specifically recognizing the CD33 protein are shown in Table 68. For GPM01, an expression vector consisting of sequences corresponding to Fc-hole (SEQ ID NO: 7), 33-1 HC (SEQ ID NO: 636), and 33-1 LC (SEQ ID NO: 637) was co-transfected into EXPICHO-S™ (Gibco, A29127) (Figure 41a, Table 64), and purification and analysis were performed in the same manner as described in Example 1. Expression, purification, and analysis were performed for 33-2, 33-3, 33-4, 33-5, 33-6, and 33-7 in the same manner as mentioned above (Figures 41a-41b, and Tables 68 and 69). 33-1, 33-2, and 33-3 bind different epitopes of the antigen monovalently and have a structure consisting of two Fc domains (Figure 41a). 33-4, 33-5, 33-6, and 33-7 have a structure in which the variable regions of the CD33 antibody are linked by a polypeptide linker (sequence number 48, sequence number 50), bind to CD33 biparatopic, and have two Fc domains (Figure 41b).

[0307] [Table 68] JPEG2024539509000223.jpg201149 JPEG2024539509000224.jpg201149 JPEG2024539509000225.jpg43149

[0308] [Table 69] JPEG2024539509000227.jpg202149 JPEG2024539509000228.jpg201149 JPEG2024539509000229.jpg202149 JPEG2024539509000230.jpg202149 JPEG2024539509000231.jpg201149 JPEG2024539509000232.jpg202149 JPEG2024539509000233.jpg202149 JPEG2024539509000234.jpg200149 JPEG2024539509000235.jpg87149

[0309] Table 70 below shows the polypeptide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of engineered antibodies that target CD33. Table 71 below shows the nucleotide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of engineered antibodies that target CD33.

[0310] [Table 70] JPEG2024539509000237.jpg33149

[0311] [Table 71] JPEG2024539509000239.jpg203149 JPEG2024539509000240.jpg204149 JPEG2024539509000241.jpg73149

[0312] Table 72 below shows the heavy and light chain CDR sequences of engineered antibodies that target CD33.

[0313] [Table 72] JPEG2024539509000243.jpg153149

[0314] The CD33 protein binding constants of 33-1, 33-2, 33-3, 33-4, 33-5, 33-6, and 33-7 were determined using Octet Red 96e (Sartorius). To analyze the binding constants of the seven antibodies, human CD33 recombinant protein (Sino Biologicals, 12238-H08H) was loaded onto an anti-penta HIS (HIS1K) biosensor (Sartorius, 18-5120). The seven antibodies were then added at various concentrations to the binding reaction (600 seconds) and dissociation reaction (1,200 seconds), and the affinity of the antibodies to CD33 was calculated (Figure 50, Table 73). Table 73 below illustrates the binding constants of engineered antibodies targeting CD33.

[0315] [Table 73]

[0316] Example 22. Design, preparation, and analysis of antibody constructs targeting CEACAM5 The light and heavy chain variant polypeptide sequences of antibodies that specifically bind to the CEACAM5 protein are shown in Table 74. For CEA01, expression vectors consisting of sequences corresponding to Fc-hole (SEQ ID NO: 7), CEA01 HC (SEQ ID NO: 590), and CEA01 LC (SEQ ID NO: 591) were co-transfected into EXPICHO-S™ (Gibco, A29127) (Figure 41a, Table 74), and purification and analysis were performed in the same manner as described in Example 1. Expression, purification, and analysis were performed for CEA02, CEA03, and CEA04 in the same manner as mentioned above (Table 74).

[0317] [Table 74] JPEG2024539509000246.jpg133149

[0318] Table 75 below shows the heavy and light chain nucleotide sequences of engineered antibodies that target CEACAM5.

[0319] [Table 75] JPEG2024539509000248.jpg201149 JPEG2024539509000249.jpg202149 JPEG2024539509000250.jpg202149 JPEG2024539509000251.jpg152149

[0320] Table 76 below shows the polypeptide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of engineered antibodies that target CEACAM5. Table 77 below shows the nucleotide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of engineered antibodies that target CEACAM5.

[0321] [Table 76]

[0322] [Table 77] JPEG2024539509000254.jpg188149

[0323] Table 78 below shows the heavy and light chain CDR sequences of engineered antibodies that target CEACAM5.

[0324] [Table 78]

[0325] To analyze the binding constant, human CEACAM5 recombinant protein (Sino Biologicals, 11077-H08H) was loaded onto an anti-penta-HIS (HIS1K) biosensor (Sartorius, 18-5120). Antibodies were then added at various concentrations to the binding reaction (600 seconds) and dissociation reaction (1,200 seconds), and the affinity of the antibodies to human CEACAM5 was calculated (Figure 51, Table 79). Table 79 below illustrates the binding constants of engineered antibodies targeting human CEACAM5.

[0326] [Table 79]

[0327] Example 23. Design, preparation, and analysis of antibody constructs targeting TROP2, mesothelin, or LIV-1 The variant light and heavy chain polypeptide sequences of antibody T01, which specifically binds to TROP2 protein, are shown in Table 80. For T01, an expression vector consisting of sequences corresponding to Fc-hole (SEQ ID NO: 7), T01 HC (SEQ ID NO: 753), and T01 LC (SEQ ID NO: 754) was co-transfected into EXPICHO-S™ (Gibco, A29127) ( FIG. 41 a, Table 80), and purification and analysis were performed in the same manner as described in Example 1. The variant light and heavy chain polypeptide sequences of antibody MSM01, which specifically binds to mesothelin protein, are shown in Table 80. For MSM01, an expression vector consisting of sequences corresponding to Fc-hole (SEQ ID NO: 7), MSM01 HC (SEQ ID NO: 755), and MSM01 LC (SEQ ID NO: 756) was co-transfected into EXPICHO-S™ (Gibco, A29127) (Figure 41a, Table 80), and purification and analysis were performed in the same manner as described in Example 1. Variant light and heavy chain polypeptide sequences of antibody LIM01 that specifically binds to LIV-1 protein are shown in Table 80. For LIM01, an expression vector consisting of sequences corresponding to Fc-hole (SEQ ID NO: 7), LIM01 HC (SEQ ID NO: 757), and LIM01 LC (SEQ ID NO: 758) was co-transfected into EXPICHO-S™ (Gibco, A29127) (Figure 41a, Table 80), and purification and analysis were performed in the same manner as described in Example 1.

[0328] [Table 80] JPEG2024539509000258.jpg193149

[0329] Table 81 below shows the heavy and light chain nucleotide sequences of engineered antibodies that target TROP2, mesothelin, or LIV-1.

[0330] [Table 81] JPEG2024539509000260.jpg197149 JPEG2024539509000261.jpg197149 JPEG2024539509000262.jpg110149

[0331] Table 82 below shows the polypeptide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of engineered antibodies targeting TROP2, mesothelin, or LIV-1. Table 83 below shows the nucleotide sequences of the heavy chain variable region (VH) and light chain variable region (VL) of engineered antibodies targeting TROP2, mesothelin, or LIV-1.

[0332] [Table 82]

[0333] [Table 83] JPEG2024539509000265.jpg166149

[0334] Table 84 below shows the heavy and light chain CDR sequences of engineered antibodies targeting TROP2, mesothelin, or LIV-1.

[0335] [Table 84]

[0336] To analyze the binding constants, human TROP2 recombinant protein (Sino Biologicals, 10428-H08H), human mesothelin recombinant protein (Sino Biologicals, 13128-H08H), or human LIV-1 recombinant protein (Acro biosystems, LV1-H5223) were loaded onto an anti-penta-HIS (HIS1K) biosensor (Sartorius, 18-5120). The antibodies were then added at various concentrations to the binding and dissociation reactions, and the affinity of each antibody for the human antigen was calculated (Figure 52, Table 85). Table 85 below illustrates the binding constants of engineered antibodies targeting TROP2, mesothelin, or LIV-1.

[0337] [Table 85]

Claims

1. (a) an antigen-binding site consisting of a first polypeptide comprising at least one complementarity-determining region (CDR) sequence and a second polypeptide comprising at least one complementarity-determining region (CDR) sequence, wherein the first polypeptide and the second polypeptide form a dimer, and the antigen-binding site is capable of specifically binding to a target antigen; (b) a first Fc domain or variant thereof that is a dimer consisting of two polypeptide sequences, one of which is bound to the first polypeptide of the antigen-binding site; and (c) a second Fc domain or variant thereof that is a dimer consisting of two polypeptide sequences, one of which is bound to the second polypeptide in the antigen-binding site; A fusion protein comprising:

2. 2. The fusion protein of claim 1, wherein the first polypeptide of the antigen-binding site comprises CDR1, CDR2, and CDR3 of an antibody heavy chain, and the second polypeptide of the antigen-binding site comprises CDR1, CDR2, and CDR3 of an antibody light chain.

3. 3. The fusion protein of claim 2, wherein the first polypeptide of the antigen-binding site further comprises a CH1 region of an antibody heavy chain and / or the second polypeptide of the antigen-binding site further comprises a constant region of an antibody light chain.

4. The fusion protein of claim 1 , wherein the antigen-binding site specifically binds to a protein expressed on a cell surface.

5. The fusion protein of claim 1 , wherein the antigen-binding site specifically binds to a cancer antigen.

6. The fusion protein of claim 1, which induces improved anti-tumor activity compared to IgG-based antibodies of conventional structure having the same antigen-binding site.

7. the antigen-binding site is selected from the group consisting of PD-L1, EGFR, EGFRvIII, BCMA, CD22, CD25, CD30, CD33, CD37, CD38, CD52, CD56, CD123, c-Met, DLL3, DR4, DR5, GD2, nectin-4, RANKL, SLAMF7, Trop-2, LIV-1, claudin 18.2, IL13α2, CD3, HER2, HER3, FGFR2, FGFR3, The fusion protein of claim 1, which specifically binds to any one selected from the group consisting of GPC3, ROR1, Folα, CD20, CD19, CTLA-4, VEGFR, NCAM1, ICAM-1, ICAM-2, CEACAM5, CEACAM6, carcinoembryonic antigen (CEA), CA-125, alpha-fetoprotein (AFP), MUC-1, MUC-16, PSMA, PSCA, epithelial tumor antigen (ETA), melanoma-associated antigen (MAGE), immature laminin receptor, TAG-72, HPV E6 / E7, BING-4, calcium-activated chloride channel 2, cyclin-B1, 9D7, Ep-CAM, EphA2, EphA3, mesothelin, SAP-1, survivin, and virus-derived antigens.

8. 2. The fusion protein of claim 1, wherein the first Fc domain and the second Fc domain are each a wild-type Fc domain or an Fc domain variant.

9. The method of claim 8, wherein the first Fc domain and the second Fc domain are each an Fc region of an immunoglobulin; 9. The fusion protein of claim 8, wherein the Fc region is an IgG, IgA, IgE, IgD, or IgM Fc region, or a variant thereof.

10. the first Fc domain variant and the second Fc domain variant each independently comprise a knob variant or a whole variant that promotes the formation of an Fc heterodimer (heterodimeric Fc); and / or wherein the first Fc domain variant and the second Fc domain variant each independently comprise a variant that promotes heterodimer formation by an electrostatic steering mechanism. The fusion protein of claim 8.

11. The following structural formulas (I), (II), (III), and (IV): N'-X-(L1)n-A-C' (I); N'-Y-(L2)m-B-C' (II); N'-C-C' (III); and N'-D-C' (IV) comprising a polypeptide of In the structural formulas (I), (II), (III), and (IV), N' is the N-terminus of each polypeptide; C' is the C-terminus of each polypeptide; - indicates a connection, A, B, C, and D are monomeric polypeptide sequences of Fc domains, each comprising an immunoglobulin CH2 and CH3 region, and optionally further comprising a CH4 and / or hinge sequence, wherein A dimerizes with either C or D to form said first Fc domain (b), and B dimerizes with the remaining one of C or D to form said second Fc domain (c); L1 and L2 are each a peptide linker; n and m are each independently 0 or 1; X is a first polypeptide sequence of the antigen-binding site comprising the heavy chain CDR1, CDR2, and CDR3 sequences of an antibody that specifically binds to a first antigen, or a heavy chain variable region of an antibody that specifically binds to a first antigen; Y is a second polypeptide sequence of the antigen-binding site comprising the light chain CDR1, CDR2, and CDR3 sequences of an antibody that specifically binds to a first antigen, or a light chain variable region of an antibody that specifically binds to a first antigen; X and Y pair with each other to form the antigen-binding site (a) that specifically binds to an antigen; The fusion protein of claim 1.

12. The fusion protein of claim 11, wherein X in structural formula (I) further comprises a heavy chain CH1 region and / or Y in structural formula (II) further comprises a light chain constant region.

13. The following structural formulae (I'), (II'), (III), and (IV): N'-VD1-(L3)p-X-(L1)n-A-C' (I'); N'-VD2-(L4)q-Y-(L2)m-B-C' (II'); N'-C-C' (III); and N'-D-C' (IV) comprising a polypeptide of wherein in the structural formulae (I'), (II'), (III), and (IV): N' is the N-terminus of each polypeptide; C' is the C-terminus of each polypeptide; - indicates a connection, A, B, C, and D are monomeric polypeptide sequences of Fc domains, each comprising an immunoglobulin CH2 and CH3 region, and optionally further comprising a CH4 and / or hinge sequence, wherein A dimerizes with either C or D to form said first Fc domain (b), and B dimerizes with the remaining one of C or D to form said second Fc domain (c); L1, L2, L3, and L4 are each a peptide linker; n, m, p, and q are each 0 or 1; VD1 consists of a heavy or light chain variable region of an antibody that specifically binds to an antigen, or CDR1, CDR2, and CDR3 of an antibody heavy or light chain; VD2 consists of an antibody light or heavy chain variable region that specifically binds to an antigen, or CDR1, CDR2, and CDR3 of an antibody heavy or light chain; VD1 and VD2 pair with each other to form a second antibody variable region that specifically binds to a second antigen; X comprises a heavy or light chain variable region of an antibody that specifically binds to an antigen, or CDR1, CDR2, and CDR3 of an antibody heavy or light chain; Y comprises an antibody light or heavy chain variable region that specifically binds to an antigen, or CDR1, CDR2, and CDR3 of an antibody heavy or light chain; X and Y pair with each other to form a first antibody variable region that specifically binds to a first antigen; VD1-(L3)p-X forms a first polypeptide sequence of the antigen-binding site (a), and VD2-(L4)q-Y forms a second polypeptide sequence of the antigen-binding site (a). The fusion protein of claim 1.

14. 14. The fusion protein of claim 13, wherein the heavy chain variable region further comprises a heavy chain CH1 region and the light chain variable region further comprises a light chain constant region.

15. the Fc domain monomers comprise knob or whole variants that promote the formation of Fc heterodimers (heterodimeric Fc); or the Fc domain monomers include variants that promote heterodimer formation by an electrostatic steering mechanism; A fusion protein according to claim 11 or 13.

16. The bond between X and Y is i) through a disulfide bond formed by Cys present in the CH1 and light chain constant regions; ii) through disulfide bonds formed by Cys present in the heavy and light chain variable regions; or iii) through disulfide bonds formed by Cys present in the CH1 and light chain constant regions, and Cys present in the heavy and light chain variable regions; The fusion protein according to claim 12 or 14, which is achieved.

17. The bond between X and Y is CH according to the Kabat numbering system. １ In addition to the disulfide bond that exists between 233 and CL214, i) the disulfide bond between VH105 and VL43; ii) the disulfide bond between VH44 and VL100; or iii) CH １ Disulfide bond between CL122 and CL121 15. The fusion protein of claim 12 or 14, further comprising:

18. A pharmaceutical composition for preventing or treating cancer, comprising the fusion protein of claim 1 as an active ingredient.

19. 19. The pharmaceutical composition of claim 18, wherein the cancer is any one selected from the group consisting of gastric cancer, liver cancer, lung cancer, colon cancer, breast cancer, prostate cancer, gallbladder cancer, bladder cancer, kidney cancer, esophageal cancer, skin cancer, rectal cancer, osteosarcoma, multiple myeloma, glioma, ovarian cancer, pancreatic cancer, cervical cancer, endometrial cancer, thyroid cancer, laryngeal cancer, testicular cancer, mesothelioma, acute myeloid leukemia, chronic myeloid leukemia, acute lymphoblastic leukemia, chronic lymphoblastic leukemia, brain tumor, neuroblastoma, retinoblastoma, head and neck cancer, salivary gland cancer, and lymphoma.

20. A transformed cell that expresses the fusion protein of claim 1.

21. 10. A method for treating or preventing cancer, comprising administering to a non-human subject in need thereof the fusion protein of claim 1.

22. Use of the fusion protein of claim 1 for the treatment of cancer in a non-human subject.

23. 10. Use of the fusion protein of claim 1 for the manufacture of a medicament for treating cancer.