Cldn18.2-targeting antibody, bispecific antibody and use thereof

Abstract

This invention discloses an antibody targeting CLDN18.2, a bispecific antibody, and their applications. The antibody targeting CLDN18.2 is a single-domain heavy chain antibody, exhibiting strong affinity for tumor cells endogenously expressing CLDN18.2 and inducing strong endocytic activity. The bispecific antibody targets both CLDN18.2 and CD3, retaining the binding interaction between Fc and FcRn. Preferably, a mutated Fc reduces binding to FcgR, thereby reducing nonspecific T cell activation caused by FcgR cross-linking. The activity at the CD3 terminus is optimized, thereby reducing the release of common cytokines in CRS, such as IL-6 and TNFα.

Description

Antibodies targeting CLDN18.2, bispecific antibodies and their applications This invention relates to the field of biomedicine, and more particularly to an antibody targeting CLDN18.2, a bispecific antibody, and its uses. Cancer is one of the leading causes of death worldwide. According to a 2018 report by the World Health Organization (WHO), approximately 18.07 million new cancer cases are diagnosed annually, resulting in about 9.55 million cancer deaths each year. The WHO estimates that stomach cancer is the fifth most commonly diagnosed cancer worldwide. Stomach cancer is the third leading cause of cancer-related death (for men) and the fourth leading cause (for women). Globally, one million new cases of stomach cancer are diagnosed each year, and in the United States, approximately 35% of newly diagnosed stomach cancer patients have metastatic stomach cancer. The 5-year survival rate for advanced stomach cancer is 5%, with a median survival of approximately 6 months. First-line treatment for metastatic / recurrent stomach cancer falls into two categories: ① HER2-neu positive patients receive trastuzumab combined with chemotherapy; ② HER2-neu negative patients can only be treated with chemotherapy alone, with poor efficacy (Front Pharmacol. 2018 Sep 13; 9: 404). Claudin18 (CLDN18) splice variant 1 (CLD18A1, i.e., CLDN18.1) has GenBank accession numbers NP_057453 and NM016369, and splice variant 2 (CLD18A2, i.e., CLDN18.2) has GenBank accession numbers NM_001002026 and NP_001002026. These are internal transmembrane proteins with a molecular weight of approximately 27.9 / 27.72 kDa. Claudins are internal membrane proteins located in the tight junctions of epithelium and endothelium. Two other major tight-binding protein families are occludins and junctional adhesion molecules (JAMs). Claudins are essential components of tight junctions and play important roles in maintaining epithelial cell polarity, controlling paracellular diffusion, and regulating cell growth and differentiation. It is speculated that claudins are almost inaccessible to antibodies in well-structured epithelium, but become exposed in tumor cells. Claudin molecules cross the cell membrane four times, with both the N-terminus and C-terminus located in the cytoplasm. Human CLDN18.2 (Claudin 18.2) is a transmembrane protein, consisting of 261 amino acids, of which 1-23 are signal peptides. It has two extracellular regions: Extracellular loop 1 (ECL1), containing approximately 55 amino acids following the signal peptide, and ECL2, containing 23 amino acids. CLDN18.1 (Claudin 18.1) and CLDN18.2 differ in the first 21 amino acids of the N-terminus, including the first TM and loop 1 (ECL1), while their C-terminal primary protein sequences are identical. The ECL1 regions of human CLDN18.2 and human CLDN18.1 are very similar, and their ECL2 regions are completely identical. Therefore, the development of antibodies targeting the human CLDN18.2 protein requires the search for antibodies targeting the ECL1 region or spatial structure of the human CLDN18.2 protein. This makes this work more difficult. CLDN18.1 is selectively expressed in normal lung and gastric epithelium (Mol Cell Biol. 2001 Nov;21(21):7380-90.). The expression of CLDN18.2 in normal tissues is highly restricted, limited to differentiated cells of the gastric epithelium, and not present in the gastric stem cell region. However, it is strongly expressed in several cancer types, including gastric, esophageal, pancreatic, and lung tumors, as well as human cancer cell lines. The molecular weight of this protein differs between some cancers and adjacent normal tissues. The higher molecular weight protein observed in healthy tissues can be converted to the same molecular weight observed in cancer by treating tissue lysates with the deglycosylation compound PNGaseF. This suggests that the protein is less N-glycosylated in cancer compared to its normal tissue counterpart. This structural difference is likely to result in altered epitopes.The classic N-glycosylation motif is found in the amino acid at position 116 of the D3 domain of this molecular ring (CN 103509110 B). Currently, only Claudiximab (IMAB362) antibody (see WO 2014 / 146672) is in phase II and III clinical trials for human CLDN18.2 monoclonal antibodies. IMAB362 can induce ADCC (antibody-dependent cell-mediated cytotoxicity) and CDC (complement-dependent cytotoxicity) effects, and mediate tumor killing. IMAB362 has shown encouraging efficacy in phase I and II clinical trials for the treatment of advanced gastric and esophageal cancer (Eur J Cancer. 2018 Sep;100:17-26). However, IMAB362 is a human-mouse chimeric antibody, which carries the risk of immunogenicity and has relatively low affinity. Due to the unmet medical needs of many malignant tumors, there is a need for other CLDN18.2 antibodies with more desirable pharmaceutical characteristics. Therefore, there is a lack of effective antibodies targeting human CLDN18.2 protein, especially fully human monoclonal antibodies, as well as monoclonal antibodies with better cell-binding activity. Currently, Amgen's AMG910 is a bispecific antibody targeting CLDN18.2×CD3 that is in clinical development. AMG910 can induce TDCC (T-cell-dependent cellular cytotoxicity) to mediate tumor killing. However, existing antibodies may suffer from problems such as short half-life, poor efficacy, and cytokine release syndrome (CRS). Therefore, there is an urgent need to develop a safer and more effective bispecific antibody that targets both human CLDN18.2 and CD3 and can bind to CLDN18.2 and CD3 in rhesus monkeys. To address the technical problem of the lack of safe and effective monoclonal antibodies targeting human CLDN18.2 and bispecific antibodies that simultaneously target human CLDN18.2 and CD3 and can bind to CLDN18.2 and CD3 in rhesus macaques, this invention provides a monoclonal antibody targeting CLDN18.2, a bispecific antibody targeting CLDN18.2 and CD3, and their applications. To address the aforementioned technical problems, a first aspect of the present invention provides an antibody targeting CLDN18.2, comprising a heavy chain variable region, wherein the heavy chain variable region comprises HCDR1, HCDR2, and HCDR3; wherein HCDR1 comprises an amino acid sequence as shown in any one of SEQ ID NO: 16-18, HCDR2 comprises an amino acid sequence as shown in any one of SEQ ID NO: 42-46 and SEQ ID NO: 48-54, and HCDR3 comprises an amino acid sequence as shown in any one of SEQ ID NO: 77-82. In a preferred embodiment of the present invention, HCDR1, HCDR2, and HCDR3 respectively comprise amino acid sequences as shown in SEQ ID NO: 16, SEQ ID NO: 42, and SEQ ID NO: 77; or HCDR1, HCDR2, and HCDR3 respectively comprise amino acid sequences as shown in SEQ ID NO: 16, SEQ ID NO: 43, and SEQ ID NO: 78; or HCDR1, HCDR2, and HCDR3 respectively comprise amino acid sequences as shown in SEQ ID NO: 16, SEQ ID NO: 44, and SEQ ID NO: 79; or HCDR1, HCDR2, and HCDR3 respectively comprise amino acid sequences as shown in SEQ ID NO: 17, SEQ ID NO: 45, and SEQ ID NO: 80; or HCDR1, HCDR2, and HCDR3 respectively comprise amino acid sequences as shown in SEQ ID NO: 18, SEQ ID NO: 43, and SEQ ID NO: 77. The amino acid sequences shown in SEQ ID NO: 80; or HCDR1, HCDR2, and HCDR3 respectively contain amino acid sequences as shown in SEQ ID NO: 18, SEQ ID NO: 43, and SEQ ID NO: 78; or HCDR1, HCDR2, and HCDR3 respectively contain amino acid sequences as shown in SEQ ID NO: 16, SEQ ID NO: 43, and SEQ ID NO: 81; or HCDR1, HCDR2, and HCDR3 respectively contain amino acid sequences as shown in SEQ ID NO: 16, SEQ ID NO: 46, and SEQ ID NO: 82; or HCDR1, HCDR2, and HCDR3 respectively contain amino acid sequences as shown in SEQ ID NO: 16, SEQ ID NO: 48, and SEQ ID NO: 78; or HCDR1, HCDR2, and HCDR3 respectively contain amino acid sequences as shown in SEQ ID NO: 16, SEQ ID NO: 49, and SEQ ID NO: 81. The amino acid sequence shown in SEQ ID NO: 78; or HCDR1, HCDR2, and HCDR3 respectively contain the amino acid sequences shown in SEQ ID NO: 16, SEQ ID NO: 50, and SEQ ID NO: 78; or HCDR1, HCDR2, and HCDR3 respectively contain the amino acid sequences shown in SEQ ID NO: 16, SEQ ID NO: 51, and SEQ ID NO: 78; or HCDR1, HCDR2, and HCDR3 respectively contain the amino acid sequences shown in SEQ ID NO: 16, SEQ ID NO: 52, and SEQ ID NO: 78.Alternatively, HCDR1, HCDR2, and HCDR3 may each contain amino acid sequences as shown in SEQ ID NO: 16, SEQ ID NO: 53, and SEQ ID NO: 79, respectively; or HCDR1, HCDR2, and HCDR3 may each contain amino acid sequences as shown in SEQ ID NO: 16, SEQ ID NO: 54, and SEQ ID NO: 78, respectively. The amino acid sequence combinations of HCDR1, HCDR2, and HCDR3 mentioned above are detailed in Table a below. [Table a] HCDR combinations of antibodies targeting CLDN18.2 In a preferred embodiment of the present invention, the heavy chain variable region further includes a frame region, wherein HFR1 in the heavy chain variable region frame region contains an amino acid sequence as shown in SEQ ID NO: 6 or 7, HFR2 contains an amino acid sequence as shown in any of SEQ ID NO: 28-34, HFR3 contains an amino acid sequence as shown in any of SEQ ID NO: 63-68, and HFR4 contains an amino acid sequence as shown in any of SEQ ID NO: 84 and 86-89. In a preferred embodiment of the present invention, the heavy chain variable region comprises an amino acid sequence as shown in any of SEQ ID NO: 150-157 and SEQ ID NO: 159-165, as detailed in Table b below. [Table b] Heavy chain variable region of antibody targeting CLDN18.2 In a preferred embodiment of the present invention, the antibody further includes a heavy chain constant region. More preferably, the heavy chain constant region is selected from hIgG1, hIgG2, hIgG3, or hIgG4 or variants thereof. Even more preferably, the heavy chain constant region is hIgG1. In a preferred embodiment of the present invention, the antibody is a full-length antibody, Fab, Fab', or F(ab'). 2. Fv, scFv, bispecific antibodies, multispecific antibodies, heavy chain antibodies or single-domain antibodies, or monoclonal antibodies or multiclonal antibodies prepared from the above antibodies. In a preferred embodiment of the present invention, the antibody is a single-domain antibody, which comprises the amino acid sequences shown in any one of SEQ ID NO: 182-189 and SEQ ID NO: 191-197, as detailed in Table c below. [Table c] Heavy chain of antibody targeting CLDN18.2 In this invention, the "Fab fragment" consists of a light chain, a heavy chain (CH1 region), and a variable region. The heavy chain of the Fab molecule cannot form disulfide bonds with another heavy chain molecule. The "Fc" region contains two heavy chain fragments, each containing the CH1 and CH2 domains of the antibody. The two heavy chain fragments are held together by two or more disulfide bonds and by the hydrophobic interaction of the CH3 domain. The "Fab' fragment" contains a light chain and a portion of a heavy chain containing the VH domain, the CH1 domain, and the region between the CH1 and CH2 domains, thereby allowing interchain disulfide bonds to form between the two heavy chains of the two Fab' fragments to form F(ab'). 2 molecules. "F(ab')" The '2' segment contains two light chains and two heavy chains comprising a constant region between the CH1 and CH2 domains, thereby forming interchain disulfide bonds between the two heavy chains. Therefore, F(ab') The 2-fragment consists of two Fab' fragments held together by disulfide bonds between the two heavy chains. The term "Fv" refers to an antibody fragment consisting of the VL and VH domains of a single arm of the antibody, but lacks the constant region. In this invention, the scFv (single chain antibody fragment) can be a conventional single chain antibody in the art, comprising a heavy chain variable region, a light chain variable region, and a short peptide of 15-20 amino acids. The VL and VH domains enable them to pair as linkers to form monovalent molecules of a single polypeptide chain [see, for example, Bird et al., Science 242:423-426 (1988) and Huston et al., Proc. Natl. Acad. Sci. USA 85:5879-5883 (1988)]. Such scFv molecules can have a general structure: NH2-VL-linker-VH-COOH or NH2-VH-linker-VL-COOH. Suitable prior art linkers consist of repeating G... Composed of a 4S amino acid sequence or a variant thereof. For example, a sequence having an amino acid sequence (G... 4S) 4 or (G) 4S) 3 connectors, but variations thereof may also be used. The term "multispecific antibody," used in its broadest sense, encompasses antibodies that exhibit multi-epitope specificity. Such multispecific antibodies include, but are not limited to: antibodies comprising a heavy chain variable region (VH) and a light chain variable region (VL), wherein the VH-VL unit exhibits multi-epitope specificity; antibodies having two or more VL and VH regions, each VH-VL unit binding to a different target or a different epitope of the same target; antibodies having two or more single variable regions, each single variable region binding to a different target or a different epitope of the same target; full-length antibodies, antibody fragments, bispecific antibodies, and triabodies, antibody fragments covalently or non-covalently linked together, etc. The antibodies of this invention include monoclonal antibodies. The monoclonal antibody, mAb, or Ab described in this invention refers to an antibody obtained from a single colony cell line, and the cell line is not limited to eukaryotic, prokaryotic, or bacteriophage colony cell lines. In this invention, the "heavy chain antibody" refers to an antibody that contains only one heavy chain variable region (VHH) and two conventional CH2 and CH3 regions, also known as HCAbs. In this invention, the "single-domain antibody," also known as a "nanobody," refers to a VHH structure selected from heavy chain antibodies, which is the smallest known unit capable of binding to a target antigen. To address the aforementioned technical problems, a second aspect of the present invention provides a bispecific antibody comprising a first protein functional region and a second protein functional region, wherein the first protein functional region targets CD3 and the second protein functional region targets CLDN18.2; the first protein functional region is in Fab form and the second protein functional region is in VH form, preferably comprising 2 or 3 VHs; when the second protein functional region comprises 3 VHs and the 3 VHs are tandemly linked, the first protein functional region and the second protein functional region are respectively connected to an Fc double strand; when the second protein functional region comprises 2 VHs, the 2 VHs are tandemly linked, and the first protein functional region and the second protein functional region are respectively connected to an Fc double strand; when the second protein functional region comprises 3 VHs and one of the 3 VHs is connected to the first protein functional region, the other two VHs are tandemly linked, and the first protein functional region and the two VHs of the tandemly linked second protein functional region are respectively connected to an Fc double strand; or, the first protein functional region is in Fab form and the second protein functional region is in HCAb form; Alternatively, the first protein functional region is in the form of Fab, and the second protein functional region is in the form of VH-HCAb, preferably the second protein functional region includes a total of 4 VHs. In this invention, the terms "first" and "second" in the first and second protein functional regions have no practical meaning; they are merely used to distinguish antigen-binding domains targeting different targets. A protein functional region may contain multiple antigen-binding domains of the same or different forms. Antigen-binding domains of different protein functional regions may be connected together in an operable manner, while different antigen-binding domains of the same protein functional region may not be connected to each other. For example, in this invention, the first protein functional region may be an antigen-binding domain targeting CD3, and the second protein functional region may be an antigen-binding domain targeting CLDN18.2, wherein the antigen-binding domain is... In a preferred embodiment of the present invention, the second protein functional region includes a heavy chain variable region, the heavy chain variable region including HCDR1, HCDR2 and HCDR3; wherein, HCDR1 includes an amino acid sequence as shown in any of SEQ ID NO: 16-18, HCDR2 includes an amino acid sequence as shown in any of SEQ ID NO: 42-46 and SEQ ID NO: 48-54, and HCDR3 includes an amino acid sequence as shown in any of SEQ ID NO: 77-82. More preferably, HCDR1, HCDR2, and HCDR3 respectively comprise amino acid sequences as shown in SEQ ID NO: 16, SEQ ID NO: 42, and SEQ ID NO: 77; or HCDR1, HCDR2, and HCDR3 respectively comprise amino acid sequences as shown in SEQ ID NO: 16, SEQ ID NO: 43, and SEQ ID NO: 78; or HCDR1, HCDR2, and HCDR3 respectively comprise amino acid sequences as shown in SEQ ID NO: 16, SEQ ID NO: 44, and SEQ ID NO: 79; or HCDR1, HCDR2, and HCDR3 respectively comprise amino acid sequences as shown in SEQ ID NO: 16, SEQ ID NO: 48, and SEQ ID NO: 78; or HCDR1, HCDR2, and HCDR3 respectively comprise amino acid sequences as shown in SEQ ID NO: 16, SEQ ID NO: 49, and SEQ ID NO: 77. The amino acid sequences shown in SEQ ID NO: 78; or HCDR1, HCDR2, and HCDR3 respectively contain the amino acid sequences shown in SEQ ID NO: 16, SEQ ID NO: 50, and SEQ ID NO: 78; or HCDR1, HCDR2, and HCDR3 respectively contain the amino acid sequences shown in SEQ ID NO: 16, SEQ ID NO: 51, and SEQ ID NO: 78; or HCDR1, HCDR2, and HCDR3 respectively contain the amino acid sequences shown in SEQ ID NO: 16, SEQ ID NO: 52, and SEQ ID NO: 78; or HCDR1, HCDR2, and HCDR3 respectively contain the amino acid sequences shown in SEQ ID NO: 16, SEQ ID NO: 53, and SEQ ID NO: 79; or HCDR1, HCDR2, and HCDR3 respectively contain the amino acid sequences shown in SEQ ID NO: 16, SEQ ID NO: 54, and SEQ ID NO: 78, as detailed in Table d below. [Table d] HCDR combination of the second protein functional region of bispecific antibodies More preferably, the heavy chain variable region includes any of the amino acid sequences shown in SEQ ID NO: 150-152 and SEQ ID NO: 159-165, all of which are listed in Table b. In one specific embodiment of the present invention, the first protein functional region includes a light chain variable region and a heavy chain variable region. The heavy chain variable region includes HCDR1, HCDR2, and HCDR3 as shown in SEQ ID NO: 11, SEQ ID NO: 38, and SEQ ID NO: 72, respectively. The light chain variable region includes LCDR1, LCDR2, and LCDR3 as shown in SEQ ID NO: 101, SEQ ID NO: 116, and SEQ ID NO: 131, respectively. Preferably, the heavy chain variable region comprises an amino acid sequence as shown in SEQ ID NO: 149 or SEQ ID NO: 144, and the light chain comprises an amino acid sequence as shown in SEQ ID NO: 168. The amino acid sequences of the first protein functional region mentioned above are shown in Table e below. [Table e] Amino acid sequences related to the first protein functional region of bispecific antibodies Preferably, the bispecific antibody comprises three polypeptide chains in the following form: the two N-termini of Fc are respectively linked to Fab and VH; preferably, the first polypeptide chain of the bispecific antibody is as follows: VH CLDN18.2- VH CLDN18.2 -Hinge-CH2-CH3 or VH CLDN18.2 -VH CLDN18.2 -VH CLDN18.2 -Hinge-CH2-CH3 is shown, and the second polypeptide chain is as follows: VH CD3 As shown in the diagram -CH1-hinge-CH2-CH3, the third polypeptide chain is as follows: VL CD3 -CL; or, one C-terminus of HCAb is linked to VH or VL of Fab; preferably, the first polypeptide chain of the bispecific antibody is as follows: VH CLDN18.2 -Hinge-CH2-CH3 is shown, and the second polypeptide chain is as follows: VH CLDN18.2 -Hinge-CH2-CH3-VH CD3 As shown in -CH1, the third polypeptide chain is as follows: VL CD3 -CL as shown; or, the first polypeptide chain of the bispecific antibody as shown in formula: VH CLDN18.2-Hinge-CH2-CH3 is shown, and the second polypeptide chain is as follows: VH CLDN18.2 -Hinge-CH2-CH3-VL CD3 As shown in -CL, the third polypeptide chain is as follows: VH CD3 -CH1 as shown; or, the first polypeptide chain of the bispecific antibody as shown in formula: VH CLDN18.2- VH CLDN18.2 -Hinge-CH2-CH3 is shown, and the second polypeptide chain is as follows: VH CLDN18.2- VH CLDN18.2 -Hinge-CH2-CH3-VH CD3 As shown in -CH1, the third polypeptide chain is as follows: VL CD3 -CL as shown; or, the first polypeptide chain of the bispecific antibody as shown in formula: VH CLDN18.2- VH CLDN18.2 -Hinge-CH2-CH3 is shown, and the second polypeptide chain is as follows: VH CLDN18.2- VH CLDN18.2 -Hinge-CH2-CH3-VL CD3 As shown in -CL, the third polypeptide chain is as follows: VH CD3 -CH1 is shown; or, the N end of the Fab heavy chain is connected to a VH CLDN18.2 The heavy chain C-terminus is connected to one N-terminus of Fc, and the C-terminus of VH is connected to the other N-terminus of Fc; preferably, the first polypeptide chain of the bispecific antibody is as follows: VH CLDN18.2- VH CLDN18.2 -Hinge-CH2-CH3 is shown, and the second polypeptide chain is as follows: VH CLDN18.2 -VH CD3 As shown in the diagram -CH1-hinge-CH2-CH3, the third polypeptide chain is as follows: VL CD3 -CL is shown. More preferably, different functional units are operatively linked by linker peptides, such as VH, CH2-CH3, and VL, wherein the linker peptide preferably comprises any of the amino acid sequences shown in SEQ ID NOs: 244-248, and more preferably comprises the sequence shown in SEQ ID NO: 246, as detailed in Table f below. [Table f] Linker peptide sequences In one specific embodiment of the present invention, the two N-termini of Fc are respectively connected to Fab and VH; preferably, the first polypeptide chain of the bispecific antibody is as follows: VH CLDN18.2- Linker peptide-VH CLDN18.2 -Hinge-CH2-CH3 or VH CLDN18.2 -Linker peptide-VH CLDN18.2 -Linker peptide-VH CLDN18.2 -Hinge-CH2-CH3 is shown, and the second polypeptide chain is as follows: VH CD3 As shown in the diagram -CH1-hinge-CH2-CH3, the third polypeptide chain is as follows: VL CD3 -CL is shown in Figure 4, and specific examples are shown in structures (1) and (7); or, one C-terminus of HCAb is linked to VH or VL of Fab; preferably, the first polypeptide chain of the bispecific antibody is as follows: VH CLDN18.2 -Hinge-CH2-CH3 is shown, and the second polypeptide chain is as follows: VH CLDN18.2 -Hingle-CH2-CH3-Linking peptide-VH CD3 As shown in -CH1, the third polypeptide chain is as follows: VL CD3 -CL as shown; or, the first polypeptide chain of the bispecific antibody as shown in formula: VH CLDN18.2 -Hinge-CH2-CH3 is shown, and the second polypeptide chain is as follows: VH CLDN18.2 -Hingle-CH2-CH3-Linking peptide-VL CD3 As shown in -CL, the third polypeptide chain is as follows: VH CD3-CH1 is shown, and specific examples are shown in Figure 4 structures (2) and (3); or, one C-terminus of VH-HCAb is linked to VH or VL of Fab; preferably, the first polypeptide chain of the bispecific antibody is as follows: VH CLDN18.2- Linker peptide-VH CLDN18.2 -Hinge-CH2-CH3 is shown, and the second polypeptide chain is as follows: VH CLDN18.2- Linker peptide-VH CLDN18.2 -Hingle-CH2-CH3-Linking peptide-VH CD3 As shown in -CH1, the third polypeptide chain is as follows: VL CD3 -CL as shown; or, the first polypeptide chain of the bispecific antibody as shown in formula: VH CLDN18.2- Linker peptide-VH CLDN18.2 -Hinge-CH2-CH3 is shown, and the second polypeptide chain is as follows: VH CLDN18.2- Linker peptide-VH CLDN18.2 -Hingle-CH2-CH3-Linking peptide-VL CD3 As shown in -CL, the third polypeptide chain is as follows: VH CD3 -CH1 is shown in Figure 4, and specific examples are shown in structures (4) and (5); or, the N end of the heavy chain of Fab is connected to a VH CLDN18.2 The heavy chain C-terminus is connected to one N-terminus of Fc, and the C-terminus of VH is connected to the other N-terminus of Fc; preferably, the first polypeptide chain of the bispecific antibody is as follows: VH CLDN18.2- Linker peptide-VH CLDN18.2 -Hinge-CH2-CH3 is shown, and the second polypeptide chain is as follows: VH CLDN18.2 -Linker peptide-VH CD3 As shown in the diagram -CH1-hinge-CH2-CH3, the third polypeptide chain is as follows: VL CD3 -CL is shown in Figure 4, and a specific example is shown in structure (6). In one specific embodiment of the present invention, the first polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 214, the second polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 213, and the third polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 219, the second polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 213, and the third polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 220, the second polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 213, and the third polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 221, the second polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 213, and the third polypeptide chain comprises the amino acid sequence shown in SEQ ID NO: 200. The amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 222, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 213, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 223, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 213, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 224, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 213, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 225, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 213, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200. The amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 226, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 213, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 227, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 228, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200.Alternatively, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 227, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 230, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 229; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 219, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 231, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 219, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 232, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 229; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 219, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 233, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 229. The amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 234, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 213, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 219, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 209, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 221, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 209, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 236, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 235, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200. The amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 236, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 237, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 238, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 235, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200.Alternatively, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 239, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 235, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 240, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 235, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 241, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 235, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 242, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 235, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200. The amino acid sequence shown in SEQ ID NO: 200; or, the first polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 243, the second polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 235, and the third polypeptide chain contains the amino acid sequence shown in SEQ ID NO: 200. The sequence information in the above specific embodiments is shown in Table g below. [Table g] Bispecific antibody polypeptide chain sequence information To address the aforementioned technical problems, a third aspect of the present invention provides an isolated nucleic acid that encodes an antibody as described in the first aspect of the present invention or a bispecific antibody as described in the second aspect of the present invention. The method for preparing the nucleic acid is a conventional method in the art. Preferably, it includes the following steps: obtaining the nucleic acid molecule encoding the antibody by gene selection technology, or obtaining the nucleic acid molecule encoding the antibody by artificial full-sequence synthesis. Those skilled in the art will understand that the base sequence encoding the amino acid sequence of the aforementioned antibody can be appropriately substituted, deleted, altered, inserted, or added to provide a homologue of a polynucleotide. The homologue of the polynucleotide in this invention can be prepared by substituting, deleting, or adding one or more bases of the gene encoding the antibody sequence within the range of maintaining antibody activity. To address the aforementioned technical problems, a fourth aspect of the present invention provides a recombinant expression vector comprising the isolated nucleic acid as described in the third aspect of the present invention. The recombinant expression vector can be obtained by conventional methods in the art, i.e., constructed by linking the nucleic acid molecule described in the present invention to various expression vectors. The expression vector can be any conventional vector in the art, as long as it can accommodate the aforementioned nucleic acid molecule. Preferably, the recombinant expression vector is a plastid, myxosome, bacteriophage, or viral vector, and the viral vector is preferably a retroviral vector, lentiviral vector, adenovirus vector, or adeno-associated virus vector. To address the aforementioned technical problems, a fifth aspect of the present invention provides a transformant comprising a recombinant expression vector as described in the fourth aspect of the present invention within a host cell; preferably, the host cell is... E. coli TG1, BL21 cells, or CHO-K1 cells. The preparation method of the recombinant expression transformant can be a conventional preparation method in the art, such as transforming the above-mentioned recombinant expression vector into host cells. The host cell can be any conventional host cell in the art, as long as it can stably and spontaneously replicate the above-mentioned recombinant expression vector and effectively express the nucleic acid it carries. Preferably, the host cell is... E. coli TG1 or BL21 cells (expressing single-chain antibodies or Fab antibodies), or CHO-K1 cells (expressing full-length IgG antibodies). Transforming the aforementioned recombinant transplastomers into host cells yields the preferred recombinant transfectants of this invention. The transformation method is a conventional method in the art, preferably a chemical transformation, heat shock, or electroporation. In this invention, the antibody targeting CLDN18.2 can be used to prepare a chimeric antigen receptor (CAR) and thus modify it onto cells such as T cells or NK cells. Therefore, a sixth aspect of this invention provides a chimeric antigen receptor (CAR) comprising an antibody as described in the first aspect of this invention or a bispecific antibody as described in the second aspect of this invention. For example, the chimeric antigen receptor may include the following structures: (a) an extracellular binding domain scFv that specifically recognizes CLDN18.2; (b) a hinge domain; (c) a transmembrane domain; (d) a co-stimulatory intracellular domain; and (e) a signaling domain; wherein the extracellular binding domain includes an antibody targeting CLDN18.2 as described in the first aspect of the invention. To address the aforementioned technical problems, a seventh aspect of the present invention provides a genetically modified cell comprising an antibody as described in the first aspect of the present invention or a bispecific antibody as described in the second aspect of the present invention. Preferably, the genetically modified cell is a eukaryotic cell, more preferably an isolated human cell; even more preferably, it is an immune cell such as a T cell or an NK cell. To address the aforementioned technical problems, an eighth aspect of the present invention provides a method for preparing a bispecific antibody, comprising culturing a transformant as described in the fifth aspect of the present invention, and obtaining the antibody or the bispecific antibody from the culture. To address the aforementioned technical problems, a ninth aspect of the present invention provides an antibody-drug conjugate (ADC) comprising a cytotoxic agent and an antibody as described in the first aspect of the present invention or a bispecific antibody as described in the second aspect of the present invention; preferably, the cytotoxic agent is MMAF or MMAE. The preparation method of the antibody-drug conjugate can be conventional in the art, and preferably the preparation method described in Doronina, 2006, Bioconjugate Chem. 17, 114-124. Preferably, the preparation method produces an antibody-drug conjugate with a minimum low coupling fraction (LCF) of less than 10%. The antibody-drug conjugate can exist in any physical form known in the art, preferably as a clear solution. To address the aforementioned technical problems, a tenth aspect of the present invention provides a pharmaceutical composition comprising an antibody as described in the first aspect of the present invention or a bispecific antibody as described in the second aspect of the present invention, and a pharmaceutically acceptable carrier. Preferably, the pharmaceutical composition further comprises one or more of the following groups: hormone preparations, targeted small molecule preparations, proteasome inhibitors, imaging agents, diagnostic agents, chemotherapeutic agents, oncolytic drugs, cytotoxic agents, interferons, activators of co-stimulatory molecules, inhibitors of inhibitory molecules, and vaccines. To address the aforementioned technical problems, the eleventh aspect of this invention provides the use of an antibody as described in the first aspect of this invention, a bispecific antibody as described in the second aspect of this invention, or a pharmaceutical composition as described in the tenth aspect of this invention in the preparation of a medicament for preventing and treating diseases related to CD3 and / or CLDN18.2; preferably, the disease is cancer, and more preferably, breast cancer, ovarian cancer, endometrial cancer, kidney cancer, melanoma, lung cancer, gastric cancer, liver cancer, esophageal cancer, cervical cancer, head and neck tumors, bile duct cancer, gallbladder cancer, bladder cancer, sarcoma, or colorectal cancer; more preferably, the cancer is breast cancer, ovarian cancer, endometrial cancer, kidney cancer, or bile duct cancer; even more preferably, the cancer is breast cancer. To address the aforementioned technical problems, a twelfth aspect of the present invention provides a kit comprising an antibody as described in the first aspect of the present invention, or a bispecific antibody as described in the second aspect of the present invention, a chimeric antigen receptor as described in the sixth aspect of the present invention, a genetically modified cell as described in the seventh aspect of the present invention, or an antibody-drug conjugate as described in the ninth aspect of the present invention, or a pharmaceutical composition as described in the tenth aspect of the present invention; preferably, the kit further comprises (i) a means for dispensing the antibody or its antigen-binding fragment, or the antibody-drug conjugate, or the pharmaceutical composition; and / or (ii) instructions for use. To address the aforementioned technical problems, a thirteenth aspect of the present invention provides a kit comprising kit A and kit B, wherein: kit A contains an antibody as described in the first aspect of the present invention or a bispecific antibody as described in the second aspect of the present invention, a chimeric antigen receptor as described in the sixth aspect of the present invention, a genetically modified cell as described in the seventh aspect of the present invention, an antibody-drug conjugate as described in the ninth aspect of the present invention, and / or a pharmaceutical composition as described in the tenth aspect of the present invention; kit B contains other antitumor antibodies or a pharmaceutical composition containing said other antitumor antibodies, and / or one or more of the group consisting of hormone preparations, targeted small molecule preparations, proteasome inhibitors, imaging agents, diagnostic agents, chemotherapeutic agents, oncolytic drugs, cytotoxic agents, interferons, activators of co-stimulatory molecules, inhibitors of inhibitory molecules, and vaccines. The medicine box A and medicine box B can be used simultaneously, or medicine box A can be used first and then medicine box B, or medicine box B can be used first and then medicine box A, depending on the actual needs of the specific application. To address the aforementioned technical problems, the fourteenth aspect of the present invention provides a method for diagnosing, treating, and / or preventing CLDN18.2-mediated diseases or conditions, the method comprising administering to a patient in need a therapeutically effective amount of an antibody as described in the first aspect of the present invention, or a bispecific antibody as described in the second aspect of the present invention, a chimeric antigen receptor as described in the sixth aspect of the present invention, an antibody-drug conjugate as described in the ninth aspect of the present invention, or a pharmaceutical composition as described in the tenth aspect of the present invention, or treating a patient in need using a kit as described in the thirteenth aspect of the present invention. Preferably, the disease or condition is a tumor, more preferably a CLDN18.2 positive tumor, and even more preferably gastric cancer, esophageal cancer, lung cancer, ovarian cancer, melanoma, kidney cancer, breast cancer, colorectal cancer, liver cancer, pancreatic cancer, bladder cancer, head and neck cancer, bronchial cancer, glioma and / or leukemia. The "CLDN18.2 positive" mentioned in this invention refers to the overexpression of the CLDN18.2 protein, such as the CLDN18.2 positive cell line NUGC4_D8; conversely, it is called "CLDN18.2 negative". To address the aforementioned technical problems, the fifteenth aspect of this invention provides a method for immunoassay or determination of CLDN18.2, comprising using an antibody as described in the first aspect of this invention, a bispecific antibody as described in the second aspect of this invention, a chimeric antigen receptor as described in the sixth aspect of this invention, an antibody-drug conjugate as described in the ninth aspect of this invention, or a pharmaceutical composition as described in the tenth aspect of this invention. Preferably, the assay is for non-diagnostic and / or therapeutic purposes. To address the aforementioned technical problems, the sixteenth aspect of the present invention provides a combination therapy comprising administering to a patient in need an antibody as described in the first aspect of the present invention or a bispecific antibody as described in the second aspect of the present invention, a chimeric antigen receptor as described in the sixth aspect of the present invention, an antibody-drug conjugate as described in the ninth aspect of the present invention, or a pharmaceutical composition as described in the tenth aspect of the present invention, and a second therapeutic agent; the second therapeutic agent preferably comprises other antitumor antibodies or a pharmaceutical composition comprising said other antitumor antibodies, and / or one or more of the group consisting of hormone preparations, targeted small molecule preparations, proteasome inhibitors, imaging agents, diagnostic agents, chemotherapeutic agents, oncolytic drugs, cytotoxic agents, interferons, activators of co-stimulatory molecules, inhibitors of inhibitory molecules, and vaccines. In this application, the amino acid sequences of the CDRs listed above are all as shown in the Chothia definition rule (the sequences in the claims of this invention are also as shown in the Chothia definition rule). However, those skilled in the art will know that there are various methods in the art to define the CDR of an antibody, such as the Kabat definition rule based on sequence variability (see, Kabat et al., Protein Sequences in Immunology, 5th Edition, National Institutes of Health, Bethesda, Maryland (1991)) and the Chothia definition rule based on the location of the structural loop region (see JMol Biol 273:927-48, 1997). In this application, the Combined definition rule, which includes both the Kabat and Chothia definitions, can also be used to determine the amino acid residues in the variable domain sequence. The Combined definition rule combines the ranges of the Kabat and Chothia definitions, and thus takes a larger range, as detailed in Table 1. Those skilled in the art will understand that, unless otherwise specified, the terms "CDR" and "complementarity-determining region" for a given antibody or its region (e.g., variable region) should be understood to encompass complementarity-determining regions defined as described in any of the known schemes described herein. While the scope of protection claimed in this application is based on the sequence defined by the Chothia definition rule, amino acid sequences corresponding to other CDR definition rules should also fall within the scope of protection of this invention. [Table 1] Method for defining antibody CDR in this application (see http: / / bioinf.org.uk / abs / ) In this context, Laa-Lbb can refer to the amino acid sequence from position aa to position bb, starting from the N-terminus of the antibody light chain; Haa-Hbb can refer to the amino acid sequence from position aa to position bb, starting from the N-terminus of the antibody heavy chain. For example, L24-L34 can refer to the amino acid sequence from position 24 to position 34, following the Chothia coding rules, starting from the N-terminus of the antibody light chain; H26-H32 can refer to the amino acid sequence from position 26 to position 32, following the Chothia coding rules, starting from the N-terminus of the antibody heavy chain. Those skilled in the art will know that when using Chothia to encode CDRs, there are insertion sites at some positions. In this invention, unless otherwise stated, the scientific and technical terms used herein have the meanings commonly understood by those skilled in the art. Furthermore, the cell culture, molecular genetics, nucleic acid chemistry, and immunology laboratory procedures used herein are all standard procedures widely used in their respective fields. To better understand this invention, definitions and explanations of relevant terms are provided below. The three-letter and single-letter codes for amino acids used in this invention are as known to those skilled in the art, or as described in J. Biol. Chem, 243, p3558 (1968). As used herein, the terms "comprising" or "including" are intended to mean that a composition and method includes the described elements but excludes other elements, but, depending on the context, also includes the case of "consisting of". The term "CLDN18.2" includes isotypes, mammalian (e.g., human) CLDN18.2, species homologs of human CLDN18.2, and analogs containing at least one common epitope with CLDN18.2. The amino acid sequence of CLDN18.2 (e.g., human CLDN18.2) is known in the art, as shown in the NCBI database. The term "CLDN18.1" includes isotypes, mammalian (e.g., human) CLDN18.1, species homologs of human CLDN18.1, and analogs containing at least one common epitope with CLDN18.1. The amino acid sequence of CLDN18.1 (e.g., human CLDN18.1) is known in the art, as shown in the NCBI database. The term "epitope" refers to a portion of an antigen (e.g., human CLDN18.2) that specifically interacts with an antibody molecule. The term "competition" in this invention refers to the ability of an antibody molecule to interfere with the binding of an anti-CLDN18.2 antibody molecule to its target (e.g., human CLDN18.2). Interference with binding can be direct or indirect (e.g., through allosteric regulation of the antibody molecule or the target). The extent to which an antibody molecule can interfere with the binding of another antibody molecule to its target can be determined using competitive binding assays (e.g., FACS, ELISA, or BIACORE assays). The term "antibody" as used in this invention includes immunoglobulins, which are tetrapeptide chains composed of two identical heavy chains and two identical light chains linked by interchain disulfide bonds. The composition and sequence of amino acids in the constant regions of the heavy chains of immunoglobulins differ, thus their antigenicity also differs. Accordingly, immunoglobulins can be classified into five classes, or isotypes of immunoglobulins: IgM, IgD, IgG, IgA, and IgE, with their corresponding heavy chains being µ, δ, γ, α, and ε chains, respectively. Within the same class of Ig, based on differences in the amino acid composition of their hinge regions and the number and position of disulfide bonds in their heavy chains, different subclasses can be further divided; for example, IgG can be divided into IgG1, IgG2, IgG3, and IgG4. Light chains are classified as κ chains or λ chains based on differences in their constant regions. Each of the five classes of Ig can have either a κ chain or a λ chain. In this invention, the variable region of the antibody light chain may further include a constant region of the light chain, wherein the constant region of the light chain comprises human κ, λ chains or variants thereof. In this invention, the variable region of the antibody heavy chain may further include a constant region of the heavy chain, wherein the constant region of the heavy chain comprises human IgG1, 2, 3, 4 or variants thereof. The sequence of approximately 110 amino acids near the N-terminus of both the antibody heavy and light chains varies considerably and is known as the variable region (V region). The remaining amino acid sequences near the C-terminus are relatively stable and are known as the constant region (C region). The variable region includes three highly variable regions (HVR) and four relatively conserved backbone regions (FWR). The three highly variable regions determine the antibody's specificity and are also known as complementarity-determining regions (CDR). Each light chain variable region (VL) and heavy chain variable region (VH) consists of three CDR regions and four FWR regions, arranged in the following order from the amino terminus to the carboxyl terminus: FWR1, CDR1, FWR2, CDR2, FWR3, CDR3, FWR4. The three CDR regions of the light chain refer to LCDR1, LCDR2, and LCDR3; the three CDR regions of the heavy chain refer to HCDR1, HCDR2, and HCDR3. Within the light and heavy chains, variable and constant regions are linked by "J" regions of approximately 12 or more amino acids, while the heavy chain also contains "D" regions of approximately 3 or more amino acids. Each heavy chain consists of a heavy chain variable region (VH) and a heavy chain constant region (CH). The heavy chain constant region consists of three domains (CH1, CH2, and CH3). Each light chain consists of a light chain variable region (VL) and a light chain constant region (CL). The light chain constant region consists of one domain, CL. The constant region of an antibody mediates the binding of immunoglobulins to host tissues or factors, including various cells of the immune system (e.g., effector cells) and the first component (C1q) of the classical complement system. The VH and VL regions can be further subdivided into highly variable regions [called complementation-determining regions (CDRs)], interspersed with more conserved regions called framework regions (FWRs). Each VH and VL consists of three CDRs and four FWRs arranged in the following order: FWR1, CDR1, FWR2, CDR2, FWR3, CDR3, and FWR4, from the amino terminus to the carboxyl terminus. The variable regions (VH and VL) corresponding to each heavy / light chain form antibody binding sites. In particular, the heavy chain may also contain more than three CDRs, such as six, nine, or twelve. For example, in the bispecific antibody of this invention, the heavy chain may be the N-terminus of the heavy chain of an IgG antibody linked to the ScFv of another antibody; in this case, the heavy chain contains nine CDRs. The term "human antibody" includes antibodies having variable and constant regions of human immunoglobulin sequences. Human antibodies of the present invention may include amino acid residues not encoded by human immunoglobulin sequences (e.g., mutations introduced by random or site-specific mutagenesis in vitro or by somatic mutations in vivo). However, the term "human antibody" does not include antibodies in which a CDR sequence derived from another mammalian species (such as a mouse) has been grafted onto a human backbone sequence (i.e., "humanized antibody"). As used herein, the term "specificity" in relation to antibodies means an antibody that recognizes a specific antigen but substantially does not recognize or bind to other molecules in a sample. For example, an antibody that specifically binds to an antigen from one species may also bind to that antigen from one or more species. However, this interspecies cross-reactivity itself does not change the antibody's classification based on specificity. In another instance, an antibody that specifically binds to an antigen may also bind to different pairs of gene forms of that antigen. However, this cross-reactivity itself does not change the antibody's classification based on specificity. In some cases, the terms "specificity" or "specific binding" may be used to refer to the interaction of an antibody, protein, or peptide with a second chemical substance, meaning that the interaction depends on the presence of a specific structure on the chemical substance (e.g., an antigenic determinant or epitope); for example, an antibody generally recognizes and binds to a specific protein structure, rather than the protein itself. If an antibody is specific for epitope "A," then in a reaction containing labeled "A" and an antibody, the presence of a molecule containing epitope A (or free, unlabeled A) will reduce the amount of labeled A bound to the antibody. As used herein, the term "chimeric antigen receptor" or "CAR" refers to a domain comprising an extracellular binding domain (extracellular binding domain), a hinge domain, a transmembrane domain (transmembrane region), and a polypeptide that facilitates cytoplasmic signaling to the binding domain (i.e., intracellular signaling domain). The hinge domain can be considered as part of the structure that provides flexibility to the extracellular antigen-binding region. The intracellular signaling domain refers to a protein that transmits information into the cell via a defined signaling pathway by generating a second messenger to regulate cellular activity, or a protein that acts as an effector in response to such a messenger, generating information that can promote the immune effector function of CAR-receptor cells (e.g., CAR-T cells). The intracellular signaling domain includes the signaling domain and may also include co-stimulatory intracellular domains derived from co-stimulatory molecules. "Identity," "variant sequence," and "mutation" refer to the sequence similarity between two polynucleotide sequences or two polypeptides. When positions in two compared sequences are occupied by the same base or amino acid monomer subunit—for example, if every position in two DNA molecules is occupied by adenine—then the molecules are homologous at that position. The percentage of identity between two sequences is a function of the number of matching or homologous positions shared by the two sequences divided by the number of compared positions × 100. For example, in optimal sequence alignment, if 6 out of 10 positions in two sequences match or are homologous, then the two sequences are 60% homologous. Generally, comparisons are made when the highest percentage of identity is obtained by aligning the two sequences. "Optimization" refers to a mutation that maintains or improves the binding of the antibody to the antigen; in this invention, it refers to a mutation that maintains, preserves, or improves the binding to CLDN18.2. The terms "polypeptide," "peptide," and "protein" (if single-chain) are used interchangeably in this invention. The terms "nucleic acid," "nucleic acid sequence," "nucleotide sequence," or "polynucleotide sequence" and "polynucleotide" are used interchangeably. The term "mutation" includes the substitution, addition, and / or deletion of amino acids or nucleotides. "Amino acid substitution" and "conserved amino acid substitution" refer to the substitution of an amino acid residue with another amino acid residue and the substitution of an amino acid residue with a similar side chain, respectively. The term "lentivirus" as used in this article refers to the genus *Retroviridae*. Lentivirals are unique among retroviruses in that they can infect non-dividing cells; they can deliver significant amounts of genetic information into the host cell's DNA, making them one of the most efficient gene delivery vectors. HIV, SIV, and FIV are examples of lentiviruses. Lentiviral vectors provide a means to achieve significant levels of gene transfer in vivo. As used herein, the term "vector" refers to a component containing isolated nucleic acids and capable of delivering those isolated nucleic acids into the cell. Many vectors are known in the art, including, but not limited to, linear polynucleotides, polynucleotides associated with ionic or amphiphilic compounds, plastids, and viruses. Therefore, the term "vector" includes autonomously replicating plastids or viruses. The term should also be interpreted to include non-plastid and non-viral compounds that facilitate the transfer of nucleic acids into cells, such as polylysine compounds, liposomes, etc. Examples of viral vectors include, but are not limited to, adenovirus vectors, adeno-associated virus vectors, and retroviral vectors. The terms "cell" and "cell line" used in this invention are used interchangeably, and all such names include descendants. The term "host cell" refers to a cell that can be used to introduce the vector, including but not limited to prokaryotic cells such as Escherichia coli, fungal cells such as yeast cells, or animal cells such as fibroblasts, CHO cells, COS cells, NSO cells, HeLa cells, BHK cells, HEK 293 cells, or human cells. The term "transfection" refers to the introduction of exogenous nucleic acids into eukaryotic cells. Transfection can be achieved by a variety of methods known in the art, including calcium phosphate-DNA coprecipitation, DEAE-dextran-mediated transfection, polybrene-mediated transfection, electroporation, microinjection, liposome fusion, lipid transfection, protoplast fusion, retroviral infection, and biolistics. The term "immune cell" refers to cells that can elicit an immune response. "Immune cell" and its other grammatical forms can refer to immune cells from any source. "Immune cells" include, for example, leukocytes (white blood cells), lymphocytes (T cells, B cells, natural killer (NK) cells), and bone marrow-derived cells (neutrophils, eosinophils, basophils, monocytes, macrophages, dendritic cells), derived from hematopoietic stem cells (HSCs) produced in the bone marrow. The term "immune cell" can also refer to human or non-human cells. As used herein, the term "T cell" refers to a type of lymphocyte that matures in the thymus. T cells play a crucial role in cell-mediated immunity and differ from other lymphocytes (such as B cells) in that they possess T cell receptors on their cell surface. "T cells" encompass all types of immune cells expressing CD3, including helper T cells (CD4+ cells), cytotoxic T cells (CD8+ cells), natural killer T cells, regulatory T cells (Tregs), and γ-δ T cells. "Cytotoxic cells" include CD8+ T cells, natural killer (NK) cells, and neutrophils, which mediate cytotoxic responses. As used herein, the term "NK cell" refers to a type of lymphocyte that originates in the bone marrow and plays a vital role in the innate immune system. NK cells provide a rapid immune response against virally infected cells, tumor cells, or other stress cells, even in the absence of antibodies and the major histocompatibility complex on their cell surface. For example, immune cells can be derived from blood, such as autologous T cells, allogeneic T cells, autologous NK cells, and allogeneic NK cells, or they can be derived from cell lines, such as NK cell lines prepared by EBV virus infection, NK cells induced from embryonic stem cells and iPSCs, and the NK92 cell line. "As needed," "any," "any," or "any one" means that the event or situation described below may but does not have to occur, and the description includes the possibility or absence of the event or situation. For example, "as needed, includes one antibody heavy chain variable region" means that the antibody heavy chain variable region of a particular sequence may but does not have to be present. As used in this invention, "an" and "a" are used to refer to one or more grammatical objects. Unless explicitly stated otherwise, the term "or" is used in this invention to mean the term "and / or" and is used interchangeably with it. "About" and "approximately" should generally mean the acceptable degree of error of the measured quantity given the nature or precision of the measurement. Exemplary error degrees are generally in the range of 10% and more generally in the range of 5%. The methods and compositions disclosed in this invention cover polypeptides and nucleic acids having a specified sequence, a variant sequence, or a sequence substantially the same as or similar thereto, for example, a sequence that is at least 85%, 90%, 95%, 99%, or more identical to the specified sequence. In the case of amino acid sequences, the term "substantially identical" is used in this invention to refer to the first amino acid sequence. As used in this article, the term EC 50 The half-maximal effect concentration refers to the concentration that produces a 50% maximum effect. As used in this invention, the terms "antibody-drug conjugate" or "ADC" are used interchangeably. Auristatin is a fully synthetic drug, and its chemical structure is relatively easy to modify in order to optimize its physical properties and drug characteristics. Auristatin derivatives used for antibody conjugation mainly include monomethylaurestatin E (MMAE) and monomethylaurestatin F (MMAF). The former is a synthetic pentapeptide derived from the natural microtubule polymerase inhibitor dolastatin-10, synthesized by adding a 2-amino-1-phenylpropyl-1-ol to the C-terminus. MMAE exhibits less than one nanomolar inhibitory activity against various human tumor cell lines. To reduce the cytotoxic activity of MMAE itself, MMAF adds a phenylalanine to the C-terminus of dolastatin-10. Because of the introduction of a carboxyl group, MMAF has poor cell membrane permeability, thus significantly reducing its biological activity against cells. However, its inhibitory activity against cells is greatly enhanced after conjugation with an antibody (US 7750116). In some embodiments, the antibody-cytotoxic drug conjugate or its pharmaceutically acceptable salt or solvent compound comprises the antibody of the present invention coupled with one or more maytansine alkaloid molecules. Maytansine alkaloids are mitotic inhibitors that do not function by inhibiting tubulin polymerization. Maytansine was initially isolated from the East African shrub *Maytenus serrata* (US Patent No. 3,896,111). Subsequently, it was discovered that certain microorganisms also produce maytansine alkaloids, such as maytanol and C-3 maytanol ester (US Patent No. 4,151,042). Maytansine alkaloid drug modules are attractive drug modules in antibody-drug conjugates because they: (i) are relatively easy to prepare by fermentation or chemical modification or derivatization of fermentation products; (ii) are easily derivatized with functional groups suitable for coupling to antibodies via non-disulfide linkers; (iii) are stable in plasma; and (iv) are effective against a variety of tumor cell lines. Maytansine compounds suitable for use as maytansine alkaloid drug modules are well known in the art and can be isolated from natural sources by known methods or produced using genetic engineering techniques (see Yu et al. (2002) PNAS99:7968-7973). Maytansine and maytansine analogues can also be synthesized by known methods. Exemplary embodiments of maytansine alkaloid drug modules include DM1, DM3, and DM4, as disclosed herein. The methods, compositions, and combination therapies described in this invention can be combined with other active agents or therapeutic modalities. The methods include administering the anti-CLDN18.2 antibody molecule described in this invention to an object in an amount effective in treating or preventing disease (e.g., cancer), and, as needed, combining it with one or more inhibitors of PD-1, PD-L1, PD-L2, LAG-3, CTLA-4, Tim-3 antibodies (immunotherapy) or other tumor therapeutic antibodies, Her-2, EGFR, VEGF, VEGFR antibodies, etc., and ADCs (such as T-DM1), bispecific antibodies, chemotherapy drugs, etc., and also include administering the anti-CLDN18.2 antibody molecule, additional active agents, or all of them in an amount or dose that is higher than, lower than, or equal to the amount or dose of each active agent used alone (e.g., as a monotherapy). Anti-CLDN18.2 antibody, additional active agent or total dose or dosage is lower than the amount or dosage of each active agent used alone (e.g., as a monotherapy) (e.g., at least 20%, at least 30%, at least 40% or at least 50%). Furthermore, as described in the embodiments of the present invention, anti-CLDN18.2 antibodies and drug conjugates of CLDN18.2 antibodies can bind to CLDN18.2 to induce apoptosis of target cells (tumor cells), inhibit tumor cell growth, and increase the ADCC and CDC killing effects of effector cells on tumor cells in vivo, thereby achieving the purpose of treating cancer patients. Therefore, in some embodiments, the anti-CLDN18.2 antibodies and drug conjugates of CLDN18.2 antibodies described in the present invention exhibit the anti-tumor effect of the antibodies of the present invention and the method of inhibiting tumor cell growth through these mechanisms, including administering a therapeutically effective amount of the anti-CLDN18.2 antibodies and drug conjugates of CLDN18.2 antibodies described in the present invention to a subject. This method is applicable to in vivo treatment of cancer. To obtain targeted and specific therapeutic effects, anti-CLDN18.2 antibody molecules can be administered together with other antibodies. When CLDN18.2 antibodies and drug conjugates of CLDN18.2 antibodies are administered in combination with one or more active agents, the combination can be administered in any order or simultaneously to cancer types, especially to tumor patients with high CLDN18.2 expression. In some aspects, the method provides treatment of an object (e.g., reduction or relief) of hyperplastic symptoms or diseases (e.g., cancer). The method comprises administering, alone or in combination with other active agents or therapeutic agents, one or more of the anti-CLDN18.2 antibodies or pharmaceutical conjugates of CLDN18.2 antibodies described in this invention to the object. Anti-CLDN18.2 antibody molecules, used alone or in combination with another immunomodulatory agent (e.g., anti-LAG-3, anti-Tim-3, anti-PD-1, or anti-PD-L1, anti-CTLA-4 antibody molecules), can be used to treat gastric cancer, pancreatic cancer, lung cancer, esophageal cancer, ovarian cancer, etc. Anti-CLDN18.2 antibody molecules can be administered in combination with one or more of the following: immune-based strategies; targeted drugs (e.g., VEGF inhibitors such as monoclonal antibodies against VEGF); VEGF tyrosine kinase inhibitors such as sunitinib, sorafenib, and apatinib; RNAi inhibitors; or inhibitors of downstream mediators of VEGF signaling, such as inhibitors of mammalian target of rapamycin (mTOR). As used in this invention, the terms "cancer," "cancer," and "cancer patient" are intended to include all types of cancerous growths or tumorigenic processes, metastatic tissues, or malignantly transformed cells, tissues, or organs, regardless of their histopathological type or stage of invasiveness. Examples include, but are not limited to, solid tumors, hematologic malignancies, soft tissue tumors, and metastatic lesions. Non-limiting examples of cancers suitable for treatment using the antibody targeting CLDN18.2 disclosed in this invention include gastric cancer, esophageal cancer, lung cancer, melanoma, kidney cancer, breast cancer, colorectal cancer, liver cancer, pancreatic cancer, bladder cancer, glioma and / or leukemia, or their metastatic lesions. Based on common knowledge in the field, the above-mentioned preferred conditions can be combined arbitrarily to obtain various preferred embodiments of the present invention. The reagents and raw materials used in this invention are all commercially available. The positive and progressive effects of this invention are as follows: 1. This invention describes a CLDN18.2 HCAb antibody with excellent affinity, specificity, and endocytic activity. The antibody is a novel fully human antibody containing only a "heavy chain," with a molecular weight only about half that of traditional IgG antibodies. Due to the absence of a light chain, this antibody can be used for the development of bispecific antibodies, avoiding common problems in bispecific antibody development such as light chain mismatch and heterodimerization, and also has the potential to be developed into an ADC. In some preferred embodiments, compared with IMAB362 analogs, the HCAb antibody has a stronger affinity for tumor cells endogenously expressing CLDN18.2 and can induce stronger endocytic activity. 2. This invention also describes a CLDN18.2×CD3 bispecific antibody with excellent in vitro TDCC activity and in vivo efficacy. The bispecific antibody of this invention has specific binding activity to CLDN18.2 and exhibits a stronger killing effect on tumor cells than Amgen's patented bispecific antibody analogs. In some preferred embodiments, the bispecific antibody carries a human Fc fragment, retaining the binding interaction between Fc and FcRn. Preferably, the Fc fragment is mutated to reduce binding with FcgR, thereby reducing nonspecific T cell activation caused by FcgR cross-linking. Optimized CD3 end activity reduces the release of cytokines commonly found in CRS, such as IL-6 and TNF-α. The CLDN18.2 end employs a tandem VHH configuration, avoiding common light-heavy chain mismatch issues in bispecific antibody development, retaining excellent hydrophilicity, and increasing selectivity for tumor cells exhibiting high CLDN18.2 expression. It demonstrates good in vivo stability, a long in vivo half-life, and strong in vivo antitumor activity. The present invention is further illustrated below by way of embodiments, but the invention is not limited to the scope of the embodiments described herein. Experimental methods in the following embodiments, unless otherwise specified, are performed according to conventional methods and conditions, or as selected in the product manual. Example 1 Preparation of expression vectors, stable cell lines, and mouse immunization 1.1 Preparation of expression vectors for immunizing mice The method for preparing the human CLDN18.2 expression vector for fully humanized transgenic mice was as follows: The cDNA sequence encoding human CLDN18.2 (Uniprot ID P56856-iso2) was synthesized, and the coding sequence of the human CLDN18.2 gene was selected and colonized into pCAGGS plastids (YOUBIO, VT1076) by enzyme digestion. 1.2 Preparation of stable transgenic cell lines The construction of stable HEK293 (ATCC, Cat#: CRL-1573) cell lines expressing human CLDN18.1 or CLDN18.2 was specifically as follows: HEK293 cells were transfected with plastids encoding human CLDN18.1 (GenScript, OHu29174D) or CLDN18.2 (GenScript, OHu03374D) to generate stable cell lines expressing human CLDN18.1 or CLDN18.2. The expression of CLDN18.1 and CLDN18.2 was detected by fluorescence-activated cell classification (FACS). Specifically, 20,000 transfected cells were plated in each well of a 96-well plate, and then a commercially available rabbit anti-human CLDN18 antibody (LifeSpan Bio, LS-C168812-400) was added. After incubating at 4°C for 1 hour, the cells were washed twice with PBS, and then AF-680-conjugated goat anti-rabbit IgG secondary antibody (Invitrogen, A21109) was added. After incubating at 4°C for 1 hour, the cells were washed three times with PBS, and then the cell fluorescence intensity was monitored using an FACS instrument (IntelliCytiQue Plus BR). 1.3 Mouse Immunization The human CLDN18.2 expression vector and HEK293 cells expressing human CLDN18.2 (HEK293 / hCLDN18.2 cells) were used to immunize fully humanized transgenic mice (Harbour HCAB 1.0 mice, commercially available mice purchased from Harbour BioMed). The human CLDN18.2 expression vector and gold powder were used to make gene gun bullets, and the mice were immunized at multiple sites on the abdomen using the gene gun. 50 μg of the expression vector DNA was used for each immunization, with a 2-week interval between each immunization. After three immunizations, each mouse was immunized with HEK293 / hCLDN18.2 cells, with each immunization consisting of 4 × 10⁶ cells. 6 Two immunizations were administered to mice at two-week intervals, followed by blood collection to determine the titer. The binding affinity of mouse serum was measured using FACS with CHOK1 cells (kyinno, KC-1180) expressing human CLDN18.2. Based on the titer results, mice were selected for screening of HCAB monoclonal antibodies. Three days prior to screening, mice received a booster immunization with HEK293 / hCLDN18.2 cells, with each mouse immunized with 4 × 10⁶ cells. 6 Each cell. Example 2 CLDN18.2 HCAB Generation and screening of single-domain antibodies Mice with high CLDN18.2 antibody titers in their serum obtained in Example 1 were selected. Spleens were removed from these mice, and B cells were isolated. CD138 (BD, 558626) positive plasma cells were sorted using the BD FACS AriaIII Cell Sorter. CLDN18.2 (CHOK1 / hCLDN18.2, kyinno, KC-1180) positive B cell populations were enriched using magnetic beads (Thermofisher, 11206D). RNA was extracted from B cells, and cDNA was reverse transcribed using the SuperScript IV First-Strand synthesis system (Invitrogen, 18091200). The human VH gene was then amplified by PCR using specific primers. PCR primers: 5'- GGTGTCCAGTGTSAGGTGCAGCTG-3' (SEQ ID NO: 249) 5'- AATCCCTGGGCACTGAAGAGACGGTGACC-3' (SEQ ID NO: 250) The amplified VH gene fragment was constructed into the mammalian cell phenotype pCAG vector encoding the Fc domain sequence of the human IgG1 antibody heavy chain. The constructed plastids were transfected into mammalian host cells HEK293 (ATCC, CRL-1573) to obtain the expression supernatant of HCAb antibody. Primary screening was performed using mirrorballs with CHOK1 / hCLDN18.2 cells expressing human CLDN18.2. Positive strains were selected for secondary screening. Secondary screening was performed using FACS with HEK293 / hCLDN18.1 and HEK293 / hCLDN18.2 cells. 295 monoclonal antibodies specifically binding to human CLDN18.2 were obtained. Simultaneously, conventional sequencing methods were used to obtain the nucleotide sequences encoding the variable domains of the antibody molecules and the corresponding amino acid sequences. After removing repetitive sequences, 211 fully human CLDN18.2 HCAb monoclonal antibodies with unique sequences specifically binding to human CLDN18.2 were obtained. Based on the initial and secondary screening results, the top 54 antibodies were selected for recombinant expression. The purified monoclonal antibodies were further screened using flow cytometry to assess their binding affinity to tumor cells endogeneously expressing human CLDN18.2. The top 8 antibody sequences were selected as candidate molecules (see Table 3). Those skilled in the art will know that various methods can be used to define the CDR of an antibody, such as the Kabat definition rule based on sequence variability (see Kabat et al., Protein Sequence in Immunology, 5th Edition, National Institutes of Health, Bethesda, Maryland (1991)) and the Chothia definition rule based on the location of the structural loop region (see JMol Biol 273:927-48, 1997). In this application, the Combined definition rule, which includes both the Kabat and Chothia definitions, can also be used to determine the amino acid residues in the variable domain sequence. The Combined definition rule combines the scope of the Kabat and Chothia definitions, and thus takes a larger scope, as detailed in Table 1 of the Summary of the Invention. The germline gene analysis and PTM site analysis obtained after sequencing in this embodiment are shown in Table 2 below. The antigen-binding protein sequence number information is shown in Table 3 below. [Table 2] Germline gene analysis and post-translational modification site (PTM) analysis of HCAb antibody sequence [Table 3] Serial Numbers of HCAbs Anti-CLDN18.2 Example 3 CLDN18.2 HCAB Post-translational modification site removal of antibodies Both PR004533 and PR004536 have isomerization and deacetylation sites in the CDR2 region of their heavy chains. Post-translational modification sites in the CDR2 region were randomly mutated using PCR to target the four amino acids NS and DG. The PCR products were electroporated into *E. coli* to construct random mutant libraries for the four amino acid sites. The mutant libraries were screened using mirrorballs in CHOK1 / hCLDN18.1 and CHOK1 / hCLDN18.2 cells to select and sequence molecules that specifically bind to CLDN18.2. The molecular sequence numbering information is shown in Table 4 below. [Table 4] Sequence numbers of some PR004533 and PR004536 mutant molecules. Example 4 full length CLDN18.2 HCAb 4.1 Preparation and Characterization of Single-Domain Antibodies (Recombinant HCAb Single-Domain Antibodies) After obtaining the heavy chain variable domain sequence encoding the HCAb single-domain antibody molecule, conventional recombinant DNA technology can be used to fuse the heavy chain variable domain sequence with the corresponding human antibody heavy chain constant domain sequence to obtain a recombinant HCAb single-domain antibody molecule. In this embodiment, the antibody heavy chain variable domain sequence (VH) is synthesized and selected into a mammalian cell expression plasmid vector encoding the human IgG1 antibody heavy chain constant domain sequence to encode the full-length sequence of the HCAb single-domain antibody. In this embodiment, since the variable domain sequence of the monoclonal antibody molecule obtained from immunized Harbour HCAb mice is a human antibody sequence, this embodiment also obtains a fully human anti-CLDN18.2 recombinant HCAb single-domain antibody. Transfecting mammalian host cells (such as Chinese hamster ovary cells, CHO) with plastids encoding recombinant HCAb single-domain antibodies, and then using conventional recombinant protein expression and purification techniques, yields the corresponding purified recombinant antibody. Specifically, ExpiCHO-S™ cells (Gibco, A29127) were expanded in ExpiCHO™ Expression Medium (Gibco, A2910001). Before transient transfection, the cell concentration was adjusted to 3–4 × 10⁻⁶ cells / year. 6 Cells / ml, at 37°C, 8% CO2 2. Incubate in a shaker for 24 hours until the cell concentration reaches 7-10 × 10⁻⁶. 6 Cells / ml, then diluted to 6 × 10⁻⁶ 6 Prepare 10 ml of cultured cells / ml. Dissolve 8 µg of plasmids encoding the HCAb single-domain antibody (plastmid to cell ratio of 0.8 µg : 1 ml) in 0.4 ml of OptiPRO™ SFM medium (Gibco, 12309019) and sterilize by filtration through a 0.22 µm filter. Add 32 µl of ExpiFectamine™ CHO reagent (Gibco, A29129) to 0.37 ml of OptiPRO™ SFM medium (Gibco, 12309019). Immediately add the ExpiFectamine™ CHO reagent solution slowly to the plasmid solution, invert to mix, and while shaking the culture flask, slowly add the plasmid and transfection reagent mixture dropwise. Incubate at 37°C with 8% CO2. 2. Culture in a shaker for 8-9 days. Observe cell viability after 8 days. Collect the culture, centrifuge at 3,300 g for 10 minutes, and collect the supernatant; then centrifuge the supernatant at high speed to remove impurities. Equilibrate the gravity column (Bio-Rad, #7311550) containing MabSelect™ (GE Healthcare Life Science, 71-5020-91 AE) with PBS (pH 7.4), and wash with 2-5 column volumes. Pass the supernatant sample through the column. Wash the column with 5-10 column volumes of PBS. Elute the target protein with 0.1 M glycine at pH 3.5, then adjust to neutral with Tris-HCl at pH 8.0. Finally, concentrate the solution using an ultrafiltration tube (Millipore, UFC901024) and replace the buffer with PBS to obtain the purified antibody solution. Then determine the concentration using NanoDrop (Thermo Scientific™ NanoDrop™ One), aliquot, and store for later use. 4.2 Antibody characterization using SEC-HPLC, HIC-HPLC, and DSF. Take appropriate amounts of the purified samples and load them onto an analytical SEC column (TSKgel G3000SWxl, HPLC instrument model: Agilent 1260 Infinity II) to determine the purity of the samples. The mobile phase used was 1×PBS, pH 7.4 (Sangon Biotech, E607016), room temperature, flow rate 1.0 ml / min, sample concentration 1 mg / ml, injection volume 20 μl, and detection wavelength 280 nm. After acquisition, ChemStation software was used to integrate the chromatograms and calculate relevant data to generate an analysis report, showing the residence time of different components in the sample. Take appropriate amounts of the purified samples and load them onto an analytical HIC column (TSKge1 Buty1-NPR 4.6*35, HPLC instrument model: Agilent 1260 Infinity II) to determine the purity and hydrophobicity of the samples. This method involves loading the sample from 100% mobile phase A (20 mM PB, 1.8 M (NH4)) within 16 minutes. 2SO A linear gradient composition was established from 4% (pH 6.0) to 100% mobile phase B (20 mM PB, pH 6.0). The flow rate was set at 0.7 ml / min, the sample concentration at 1 mg / ml, the injection volume at 20 μl, and the detection wavelength at 280 nm. After acquisition, the chromatograms were integrated and relevant data were calculated using ChemStation software to generate an analysis report, which showed the residence time of different components in the sample. This embodiment utilizes Differential Scanning Fluorescence (DSF) to determine the thermal denaturation temperature (Tm) of protein molecules. 10 μg of protein was added to a 96-well PCR plate (Thermo, AB-0700 / W), followed by 2 μl of 100× diluted SYPROTM dye (Invitrogen, 2008138), and then buffer was added to a final volume of 40 μl per well. The PCR plate was sealed and placed in a Bio-Rad CFX96 PCR System. It was first incubated at 25°C for 5 min, then gradually increased from 25°C to 95°C in increments of 0.2°C / 0.2 min. At the end of the test, the temperature was lowered to 25°C. The FRET scan mode was used, and data analysis was performed using Bio-Rad CFX Maestro software to calculate the Tm of the samples. The results of the characterization are shown in Table 5 below. [Table 5] Characterization of anti-CLDN18.2 HCAB antibody Example 5 CLDN18.2 HCAB antibody-cell binding affinity Antibody binding affinity was assessed using FACS with HEK293 cells expressing human CLDN18.1, HEK293 cells expressing human CLDN18.2, endogeneic NUGC4_D8 cells expressing human CLDN18.2, and SNU601 cells (CBP60507). The NUGC4_D8 subculture was obtained using NUGC4 cells (JCRB0834) selected via limiting dilution. The binding affinity assay was performed as follows: cells were centrifuged at 300 g for 5 minutes and then resuspended in FACS buffer (PBS containing 2% FBS). The cell density was adjusted to 10-1. 6Cells / ml: Add 50 μl of cell suspension to each well of a 96-well plate. Antibodies were diluted to different concentrations with FACS buffer, and 50 μl of antibody dilution was added to each well of the 96-well plate. After incubating at 4°C for 2 hours, the plate was washed twice with FACS buffer. Then, FACS buffer (Jackson, 109-605-098) containing APC-conjugated goat anti-human IgG secondary antibody was added. After incubating at 4°C for 1 hour, the plate was washed twice with FACS buffer. Cells were resuspended in fixative, and cell fluorescence was monitored using an FACS analyzer (ACEA NovoCyte). PR000400, an IMAB362 analog (the IMAB362 analog was prepared in-house; see WO 2014 / 146672. Its variable region is identical to IMAB362, with only a few amino acids differing in the constant region), served as a positive control for CLDN18.2 binding. PR004549, a CL-1xI2C scFc analog (the CL-1xI2C scFc analog was prepared in-house; see WO 2020025792 A1. Its variable region is identical to CL-1xI2C scFc, with only a few amino acids differing in the constant region), served as a positive control for CLDN18.2 binding. PR002725 antibody served as a positive control for CLDN18.1 binding (see CN2020 / 118650, Tables 6, 7, and 8). Iso hIgG1 (CrownBio, C0001-4) antibody served as a negative control. Figures 1(ab), Tables 9 and 10 show the binding affinity of the antibodies to NUGC4_D8 cells with endogeneous CLDN18.2 expression. The antibodies were tested to bind to NUGC4_D8 cells in a dose-dependent manner. Results showed that antibodies PR004533, PR004949, PR004950, PR004952, PR004953, PR007242, PR007243, PR007244, PR007245, PR007246, and PR007248 exhibited higher binding affinity to NUGC4_D8 cells with endogeneous CLDN18.2 expression compared to PR000400. Figure 1c and Table 11 show the binding affinity of the antibodies to SNU601 cells with endogeneous CLDN18.2 expression. The antibodies were tested to bind to SNU601 cells in a dose-dependent manner. The results showed that antibodies PR004227, PR004533, PR004536, PR004540, PR004949, PR004950, and PR004952 exhibited higher binding affinity to SNU601 cells overexpressing CLDN18.2 compared to PR000400. Figure 1d and Table 12 show the binding affinity of the antibodies to HEK293 cells overexpressing human CLDN18.2 (HEK293 / hCLDN18.2). Figure 1(ef) shows the binding affinity of the antibodies to HEK293 cells overexpressing human CLDN18.1 (HEK293 / hCLDN18.1). The binding affinity of the antibodies to HEK293 / hCLDN18.1 cells was low. Based on the above results, it can be inferred that the detection antibody system binds to ECL1 (Extracellular loop 1) of the human CLDN18.2 protein, rather than ECL2. [Table 6] Description of control antibody molecules [Table 7] Sequence Numbering Table of Control Antibodies (Monoclonal Antibodies) [Table 8] Sequence Numbering Table of Control Antibodies (Bilateral Antibodies) [Table 9] Binding affinity of CLDN18.2 antibody (CDR has post-translational modification sites) to NUGC4_D8 cells (see Figure 1.a) [Table 10] Binding affinity of PR004533 and PR004536 mutant antibodies to NUGC4_D8 cells (see Figure 1.b) [Table 11] Binding affinity of CLDN18.2 antibody to SNU601 cells [Table 12] Binding affinity of CLDN18.2 antibody to HEK293 / hCLDN18.2 cells Example 6 CLDN18.2 HCAb Competitive binding activity of antibodies This embodiment aims to investigate the epitope region of human CLDN18.2 antigen binding to an anti-human CLDN18.2 HCAb monoclonal antibody. A cellular-level competitive binding assay was performed using HEK293 / hCLDN18.2 cells expressing human CLDN18.2. In short, biotinylation kits (ThermoFisher, A35358) were used to biotinylate anti-human CLDN18.2 antibodies PR000400 and PR004549 according to the manufacturer's instructions. In 96-well V plates (Corning, 3894), 50 µl of biotinylated anti-human CLDN18.2 antibody PR000400 or PR004549 was thoroughly mixed with 50 µl of the corresponding serially diluted unbiotinylated anti-human CLDN18.2 antibody. Then, 3 × 10⁻⁶ ppm of the antibody was added. 6 HEK293 / hCLDN18.2 cell suspension overexpressing human CLDN18.2 was incubated at 4°C for 2 hours with 50 µl / well of pre-chilled FACS buffer (2% FBS in PBS). Cells were washed twice with 200 µl / well of pre-chilled FACS buffer and centrifuged at 500 g at 4°C for 5 minutes, discarding the supernatant. After two washes, a secondary fluorescence antibody (BD, 554060, final concentration 1 µg / ml) was added and incubated at 4°C in the dark for 1 hour. Cells were washed twice with 200 µl / well of pre-chilled FACS buffer (2% FBS in PBS) and centrifuged at 500 g at 4°C for 5 minutes, discarding the supernatant. Finally, cells were resuspended in 200 µl / well of pre-chilled FACS buffer, and fluorescence information values ​​were read using a BD FACS CANTOII reader. The inhibition rate was calculated using the formula: Inhibition rate (%) = (AB) / A * 100 (Note: A: fluorescence information after the biotinylated antibody interacts with ISO hIgG1 (Crownbio, C0001-4), B: fluorescence information after the biotinylated antibody interacts with the non-biotinylated antibody). As shown in Figure 2(ab) and Table 13, the HCAb antibodies against CLDN18.2 of this invention can block the binding of PR000400 or PR004549 to human CLDN18.2, and the detected antibody blocking ability increases positively with antibody concentration, with an inhibition rate of >80%, indicating that the detected HCAb, PR000400, and PR004549 have very similar epitopes. The detection antibody has low affinity for HEK293 / hCLDN18.1 cells. Based on the above results, it can be inferred that the detection antibody system binds to ECL1 (Extracellular loop 1) of human CLDN18.2 protein rather than ECL2. [Table 13] Competitive binding affinity of CLDN18.2 antibody to PR000400 and PR004549 Example 7 CLDN18.2 HCAB Antibody endocytosis activity The CellTiter-Glo chemiluminescence immunoassay kit (Promega, G7573) was used to detect the cytotoxicity induced by co-culturing NUGC4_D8 cells with an antibody conjugated with MMAF and an anti-human IgG antibody (Moradec, Cat#: AH-102-AF). NUGC4_D8 cells were centrifuged at 300 g for 5 minutes, then resuspended in culture medium (RPMI 1640 + 10% FBS) and the cell density was adjusted to 2 x 10⁻⁶ cells / cm². 4Cells / ml: Add 50 μl of cell suspension to each well of a 96-well plate and incubate overnight at 37°C. Dilute the antibody to different concentrations with culture medium, and add 25 μl of antibody dilution buffer to each well of a 96-well plate. Dilute the MMAF-conjugated anti-human IgG antibody with culture medium, and add 25 μl of antibody dilution buffer to each well of a 96-well plate, resulting in a final concentration of 6.6 nM. Incubate the cells and antibodies at 37°C for 3 days. Incubate the 96-well plate at room temperature for 30 minutes, and add 100 μl / well of room temperature CellTiter-Glo chromogenic solution. Then incubate the samples at room temperature in the dark for 10 minutes. Read the plate using a PE Enspire. Cell viability (%) = [(luminescent sample) / (luminescent mock control)] × 100. PR000400 (IMAB362 analogue) was used as a positive control, and Iso hIgG1 (CrownBio, C0001-4) antibody was used as a negative control. Figure 3 (ab) shows the survival rate of target cells. The co-culture of the antibody with the MMAF-conjugated anti-human IgG antibody demonstrated a dose-dependent cytotoxic effect in NUGC4_D8 cells compared to PR000400. Example 8 CLDN18.2×CD3 Structure and design of bispecific antibodies The selected anti-CLDN8.2 and anti-CD3 antibodies were used to prepare bispecific antibodies. The preparation of the CLDN18.2×CD3 bispecific antibody allows for the simultaneous binding of two targets: one end recognizes CLDN18.2, specifically expressed on the surface of tumor cells, while the other end binds to CD3 molecules on T cells. Once the CLDN18.2×CD3 bispecific antibody molecule binds to the surface of tumor cells, it can recruit and activate T cells near the tumor cells, thereby killing the tumor cells. As shown in Figure 4, structure (1) is a "2 + 1" Fab-Fc-Dual VH asymmetric structure molecule; for the "2 + 1" asymmetric structure molecule, the structure involves three protein chains, which respectively contain the heavy chain and light chain of the corresponding anti-CD3 antibody, and the Dual VH polypeptide chain of the anti-CLDN18.2 antibody; structures (2) and (3) are "2 + 1" HCAb-Fc-Fab asymmetric structure molecules, which involve three protein chains, which respectively contain the heavy chain and light chain of the corresponding anti-CD3 antibody, and the HCAb polypeptide chain of the anti-CLDN18.2 antibody. Structure (4) is a "4 + 1" VH-VH_HC-Fab asymmetric structure molecule, which involves three protein chains, which respectively contain the heavy chain and light chain of the corresponding anti-CD3 antibody, and the tetravalent VH polypeptide chain of the anti-CLDN18.2 antibody. Structure (5) is a "4 + 1" VH-VH_LC-Fab asymmetric structure molecule, which involves three protein chains, which respectively contain the heavy chain and light chain of the corresponding anti-CD3 antibody, and the tetravalent VH polypeptide chain of the anti-CLDN18.2 antibody. Structure (6) is a "3 + 1" VH_HC-Fab-Fc-Dual VH asymmetric structure molecule. This structure involves three protein chains, which respectively contain the heavy and light chains of the monovalent VH anti-CD3 antibody linked to the N-terminus of the anti-CLDN18.2 antibody, and the bivalent VH polypeptide chain of the anti-CLDN18.2 antibody. Structure (7) is a “3 + 1” Fab-Fc-VH-VH-VH asymmetric structure molecule, which involves three protein chains, which respectively contain the heavy chain and light chain of the corresponding anti-CD3 antibody, and the trivalent VH polypeptide chain of the anti-CLDN18.2 antibody. Structure (8) is a "1 + 1" Fab-Fc-scFv asymmetric structure molecule, which involves two protein chains, which respectively contain the heavy chain and light chain of the corresponding anti-CD3 antibody, and the scFv polypeptide chain of anti-Lysozyme or anti-CLDN18.1 antibody. To minimize the formation of byproducts with mismatched heavy chains (e.g., mismatched heavy chains in anti-CD3 antibodies), a mutated heterodimer Fc region carrying a "knob-hole" mutation and modified disulfide bonds was used, as described in WO 2009080251 and WO 2009080252. The CLDN18.2×CD3 bispecific antibody has an IgG1 Fc region carrying L234A, L235A, or L234A, L235A, and G237A (according to EU index number) mutations at CH3 of the Fc region. Each bispecific antibody was generated by co-transfection with three or four different mammalian expression vectors, encoding: 1) the corresponding anti-CLDN18.2 antibody heavy chain carrying a "Hole" mutation in the Fc region to produce a heterodimer antibody, with the CH3 of the Fc region carrying L234A, L235A, or L234A, L235A, and G237A mutations. 2) The heavy chain of the corresponding anti-CD3 antibody carries a "knob" mutation in the Fc region to produce a heterodimer antibody. The CH3 region of the Fc region carries L234A, L235A, or L234A, L235A, and G237A mutations. 3) The light chain of the corresponding anti-CD3 antibody. The "knob" mutation in the human IgG1 Fc region consists of T366W, and the "Hole" mutation consists of T366S, L368A, and Y407V. In addition, a disulfide bond can be formed between S354C of the "knob" Fc region and Y349C of the "Hole" region to increase stability and heterodimer antibody yield. Specific information about the CLDN18.2 x CD3 bispecific antibody molecule constructed in this invention is shown in Tables 14, 15, and 16. [Table 14] CLDN18.2 x CD3 bispecific antibody molecules constructed by structures (1) to (8) in this embodiment [Table 15] Serial Number of the CLDN18.2 x CD3 Bispecific Antibody of the Present Invention [Table 16] Sequence Numbering Table of CDRs of the Antigen-Binding Domain of the CLDN18.2 x CD3 Bispecific Antibody Molecule of the Present Invention Example 9 CLDN18.2×CD3 Preparation and Characterization of Bispecific Antibodies 9.1 Preparation of Recombinant CLDN18.2×CD3 Bispecific Antibody Multiple plastids encoding the recombinant CLDN18.2×CD3 bispecific antibody were transfected into mammalian host cells (e.g., Chinese hamster ovary cells, CHO) at a specific ratio. The purified recombinant antibody was obtained using standard recombinant protein expression and purification techniques. Specifically, ExpiCHO-S™ cells (Gibco, A29127) were expanded in ExpiCHO™ Expression Medium (Gibco, A2910001). Before transient transfection, the cell concentration was adjusted to 3–4 × 10⁻⁶ cells / year. 6 Cells / ml, at 37°C, 8% CO2 2. Incubate in a shaker for 24 hours until the cell concentration reaches 7-10 × 10⁻⁶. 6 Cells / ml, then diluted to 6 × 10⁻⁶ 6Prepare 10 ml of cultured cells / ml. Dissolve 8 µg of plasmids encoding the recombinant CLDN18.2×CD3 bispecific antibody (plastid-to-cell ratio of 0.8 µg : 1 ml) in 0.4 ml of OptiPRO™ SFM medium (Gibco, 12309019) and sterilize by filtration through a 0.22 µm filter. Add 32 µl of ExpiFectamine™ CHO reagent (Gibco, A29129) to 0.37 ml of OptiPRO™ SFM medium (Gibco, 12309019). Immediately add the ExpiFectamine™ CHO reagent solution slowly to the plasmid solution, invert to mix, and while shaking the culture flask, slowly add the plasmid and transfection reagent mixture dropwise. Incubate at 37°C with 8% CO2. 2. Culture in a shaker for 8-9 days. Observe cell viability after 8 days. Collect the culture, centrifuge at 3300 g for 10 minutes, and collect the supernatant; then centrifuge the supernatant at high speed to remove impurities. Equilibrate the gravity column (Bio-Rad, 7311550) containing MabSelect™ (GE Healthcare Life Science, 71-5020-91 AE) with PBS (pH 7.4), and wash with 2-5 column volumes. Pass the supernatant sample through the column. Wash the column with 5-10 column volumes of PBS. Elute the target protein with 0.1 M glycine at pH 3.5, then adjust to neutral with Tris-HCl at pH 8.0. Finally, concentrate the solution using an ultrafiltration tube (Millipore, UFC901024) and replace the buffer with PBS to obtain the purified antibody solution. Then determine the concentration using NanoDrop (Thermo Scientific™ NanoDrop™ One), aliquot, and store for later use. 9.2 Antibody characterization using SEC-HPLC, HIC-HPLC, and DSF. Take appropriate amounts of the purified samples and load them onto an analytical SEC column (TSKgel G3000SWxl, HPLC instrument model: Agilent 1260 Infinity II) to determine the purity of the samples. The mobile phase used was 1×PBS, pH 7.4 (Sangon Biotech, E607016), room temperature, flow rate 1.0 ml / min, sample concentration 1 mg / ml, injection volume 20 μl, and detection wavelength 280 nm. After acquisition, ChemStation software was used to integrate the chromatograms and calculate relevant data to generate an analysis report, showing the residence time of different components in the sample. Take appropriate amounts of the purified samples and load them onto an analytical HIC column TSKge1 Buty1-NPR 4.6*35 (HPLC instrument model: Agilent 1260 Infinity II) to determine the purity and hydrophobicity of the samples. This method involves loading the sample from 100% mobile phase A (20 mM PB, 1.8 M (NH4)) within 16 minutes. 2SO A linear gradient composition was established from 4% (pH 6.0) to 100% mobile phase B (20 mM PB, pH 6.0). The flow rate was set at 0.7 ml / min, the sample concentration at 1 mg / ml, the injection volume at 20 μl, and the detection wavelength at 280 nm. After acquisition, the chromatograms were integrated and relevant data were calculated using ChemStation software to generate an analysis report, which showed the residence time of different components in the sample. This embodiment utilizes Differential Scanning Fluorescence (DSF) to determine the thermal denaturation temperature (Tm) of protein molecules. 10 μg of protein was added to a 96-well PCR plate (Thermo, AB-0700 / W), followed by 2 μl of 100× diluted SYPROTM dye (Invitrogen, 2008138), and then buffer was added to a final volume of 40 μl per well. The PCR plate was sealed and placed in a Bio-Rad CFX96 PCR System. It was first incubated at 25°C for 5 min, then gradually increased from 25°C to 95°C in increments of 0.2°C / 0.2 min. At the end of the test, the temperature was lowered to 25°C. The FRET scan mode was used, and data analysis was performed using Bio-Rad CFX Maestro software to calculate the Tm of the samples. The results of the characterization are shown in Table 17. [Table 17] Characterization of CLDN18.2×CD3 Bispecific Antibody Example 10 CLDN18.2×CD3 Bispecific antibody binding affinity to cells Antibody binding affinity was detected using FACS. The cells used in the assay included HEK293 / hCLDN18.2, HEK293 / hCLDN18.1, NUGC4_D8, and Jurkat cells. The specific procedure for binding affinity detection was as follows: cells were centrifuged at 300 g for 5 minutes, and then resuspended in FACS buffer (PBS containing 2% FBS). The cell density was adjusted to 10-1. 6Cells / ml: 50 μl of cell suspension was added to each well of a 96-well plate. Antibodies were diluted to different concentrations with FACS buffer, and 50 μl of antibody dilution was added to each well of the 96-well plate. After incubation at 4°C for 2 hours, the plate was washed twice with FACS buffer. Then, FACS buffer containing APC-conjugated goat anti-human IgG secondary antibody (final concentration 1.5 μg / ml, Jackson, 109-605-098) was added. After incubation at 4°C for 1 hour, the plate was washed twice with FACS buffer. Cells were resuspended in fixative, and cell fluorescence was monitored using an FACS instrument (ACEA NovoCyte). Figure 5 and Table 18 show the binding affinity of the antibodies to cells. The antibodies were found to bind to cells exhibiting CLDN18.2 and CD3 expression, but not to cells exhibiting CLDN18.1 expression. [Table 18] Binding affinity of CLDN18.2×CD3 bispecific antibody to cells. Note: Blank spaces in the table indicate that the molecule was not tested for affinity with HEK293 / hCLDN18.2 cells; NA indicates that the molecule could not fit EC in the affinity test of HEK293 / hCLDN18.2 cells. 50 . Example 11 CLDN18.2×CD3 Bispecific antibodies TDCC active Using CytoTox 96 ® A non-radioactive cytotoxicity assay kit (Promega, G1780) was used to detect the activity of the CLDN18.2×CD3 bispecific antibody against NUGC4_D4, SNU620, IM95, and HEK293 / hCLDN 18.1-induced TDCC effects. T cells were isolated from human PBMCs using a human total T cell isolation kit (Miltenyi, 130-096-535). Human T cells and target cells were resuspended in culture medium (RPMI 1640 + 5% FBS). The target cell density was adjusted to 3 × 10⁶ cells / year. 5 Cells / ml, T cell density adjusted to 1.2 × 10⁻⁶ 6 / ml, add 50 μl of each of the two cell types to the wells of a 96-well plate (effect-to-target ratio 4:1). Dilute the test antibody to different concentrations with culture medium (RPMI 1640 + 5% FBS) and add 50 μl to the wells of a 96-well plate. Incubate the samples at 37°C for 48 hours, then add 10x Triton-X 100 lysis buffer (RPMI 1640 + 5% FBS + 10% Triton-X 100) to the target cell maximum LDH release control well and the volume correction control well, mix well, and incubate at 37°C for 0.5 hours. Centrifuge the 96-well plate at 400 g for 4 minutes, take 50 μl of supernatant, and then add LDH chromogenic solution at a concentration of 50 μl / well. After the mixture is incubated at room temperature in the dark for 20 minutes, the plate is read on an MD StakMax (OD). 490 PR004549 was used as a positive control for CLDN18.2 target cells, PR004313 as a positive control for CLDN18.1 target cells, and PR004312 antibody as a negative control. Results were calculated by first subtracting the culture medium background control well reading from the readings of the experimental wells, the target cell spontaneous LDH release control wells, and the effector cell spontaneous LDH release control wells; then, the target cell maximum LDH release control well reading was subtracted from the volume correction control well reading. TDCC activity (%) = (experimental well correction reading - effector cell spontaneous LDH release control well correction reading - target cell spontaneous LDH release control well correction reading) / (target cell maximum LDH release control well correction reading - target cell spontaneous LDH release control well correction reading) × 100. Figure 6 shows the TDCC activity of the detection antibody; detailed values ​​are shown in Table 19. In NUGC4_D8 cells with high CLDN18.2 endogeneity, the detection antibody induced stronger or comparable TDCC activity than PR004549. In IM95 cells with low CLDN18.2 expression, antibody-induced TDCC activity was weaker than that PR004549. SNU620 has a mutation at amino acid position 149 of CLDN18.2, changing from methionine to leucine, representing a mutant CLDN18.2 in some gastric cancer patients; antibody in this cell induced TDCC activity stronger than or comparable to PR004549. Antibody could not induce TDCC in HEK293 / hCLDN 18.1 cells. [Table 19] TDCC activity of CLDN18.2×CD3 bispecific antibody Example 12 use BLI Methods for detecting antibodies and human antibodies Fc Affinity of receptor proteins Biomembrane interferometry (BLI) was used to analyze the binding kinetics between proteins and antibodies using an Octet Red 96e (Fortebio) molecular interaction analyzer. The instrument was set to 1000 rpm. A 10-fold kinetic buffer (ForteBio, Cat#18-1105) was diluted 1-fold for affinity testing and sample dilution. FAB2G sensors (Fortebio, 18-5125) were equilibrated in the test buffer for 10 minutes, and then used to capture antibodies (PR002725, PR005397, PR006384, PR005411, PR006292, PR006023, PR006293) at a capture height of 1 nm. After equilibration in the buffer for 120 seconds, the FAB2G sensors bound to 2-fold serially diluted human Fc receptor proteins. Protein concentrations and binding / dissociation times are shown in Table 20. Finally, the FAB2G sensor was regenerated by immersing it in a 10 mM glycine-hydrochloric acid pH 1.5 solution to elute the protein bound to the sensor. The affinity test between the antibody and FcRn was performed under two conditions: buffer pH 6.0 and 7.4. For PR004549, the capture sensor was ProL (Fortebio, 18-5085). Using Octet Data Analysis software (Fortebio, version 11.0), data analysis was performed with a 0 nM reference well. Reference subtraction was subtracted, and the "1:1 Global fitting" method was selected for data fitting to calculate the kinetic parameters of protein-antibody binding, obtaining the kon (1 / Ms) value, kdis (1 / s) value, and KD (M) value. For fast binding and fast dissociation interactions, the "steady state" method was used to fit the data to obtain the KD (M) value. The affinity between the antibody and human Fc receptor protein is shown in Table 21. [Table 20] Fc receptor protein information and experimental parameters *For antibody PR002725, the CD64 test concentration was 40-1.25 nM [Table 21] Affinity of antibody to human Fc receptor protein NA 1 Antibody does not bind to protein NA 2The antibody binds to the receptor protein in a fast binding and dissociation manner. Steady-state fitting is used, and no KON / KDIS value is specified. Example 13 use BLI Methods for detecting antibodies against different species CD3 Protein affinity Biomembrane interference (BLI) technology was used to analyze the binding kinetics between proteins and antibodies using an Octet Red 96e (Fortebio) molecular interaction analyzer. Human CD3E (Acro, CDE-H5223), rhesus macaque CD3E (Acro, CDE-C5226), mouse CD3E (Acro, CDE-M5256), and biotin (Thermo Scientific, A39257) were mixed at a molar ratio of 1:3 and incubated overnight at 4°C to remove excess biotin, yielding biotinylated CD3E. The instrument was set to a rotation speed of 1000 rpm, and 10-fold kinetic buffer (ForteBio, Cat#18-1105) was diluted to 1-fold for affinity testing and sample dilution. SA sensors (Fortebio, 18-5019) arranged in a row were equilibrated in test buffer for 10 minutes. Then, the SA sensors were used to capture biotinylated or biotinylated CD3 molecules at a capture height of 0.2 nm. After equilibrating the captured CD3 SA sensors in buffer for 120 seconds, they were bound to antibodies diluted two-fold (see Table 22). Binding and dissociation times were set at 180 seconds and 300 seconds, respectively. Finally, the SA sensors were regenerated by immersing them in a 10 mM glycine-hydrochloric acid solution (pH 1.5) to elute the antibodies bound to the sensors. The antibody anti-CD3e 48-2B (Santa Cruz Biotechnology, SC-1174) was a murine positive antibody. When using Octet Data Analysis software (Fortebio, version 11.0) for data analysis, 0 nM was used as the reference well. Reference subtraction was subtracted, and the "1:1 Global Fitting" method was selected for data fitting to calculate the kinetic parameters of protein-antibody binding, obtaining kon (1 / Ms), kdis (1 / s), and KD (M) values. For fast binding and dissociation interactions, the "steady state" method was used to fit the data, obtaining the KD (M) value. The affinity of the antibody for CD3 proteins of different species is shown in Table 23. PR002199 is derived from Teneobio's patent WO 2018052503, an anti-BCMA (TNB308902) × CD3 (TNB_F2B) bispecific antibody. PR004931 is derived from Roche's patent WO 2017055389 A1, an anti-CEA × CD3 bispecific antibody. [Table 22] CD3 receptor protein information and experimental parameters [Table 23] Affinity of antibodies to CD3 proteins of different species NA 1 Antibody does not bind to protein. Example 14 CLDN18.2×CD3 Bispecific antibodies against humans CLDN18.2 Binding affinity of paralogous proteins The binding affinity of the antibody to the human CLDN18.2 paraline homolog was detected using a flow cytometer (ACEA NovoCyte). Human CLDN18.2 paraline family genes CLDN1, CLDN2, CLDN3, CLDN4, CLDN6, and CLDN9 were transiently transferred into HEK293 cells. Plastid information is shown in Table 24. The specific procedure for binding affinity detection was as follows: cells were centrifuged at 300 g for 5 minutes and then resuspended in FACS buffer (PBS containing 2% FBS). The cell density was adjusted to 102 6 Cells / ml: Add 50 μl of cell suspension to each well of a 96-well plate. Dilute the antibody to 60 nM with FACS buffer and add 50 μl of antibody dilution to each well of the 96-well plate. After incubating at 4°C for 2 hours, wash the plate twice with FACS buffer. Then, add FACS buffer containing secondary antibody. After incubating at 4°C for 1 hour, wash the plate twice with FACS buffer. Resuspend the cells in fixative and monitor cell fluorescence using a flow cytometer. Information on positive control antibody, negative control antibody, and secondary antibody is provided in Table 25. PR005080 strain 1A2 antibody (prepared in-house, see EP 3567053 A1) was used as a positive control for CLDN2 binding. The results of flow cytometry analysis are shown in Figure 7. The results indicate that PR006384 and PR006292 do not specifically bind to the human CLDN18.2 paralogous protein family CLDN1, CLDN2, CLDN3, CLDN4, CLDN6, and CLDN9. [Table 24] Phenoplastin information of CLDN18.2 paralogous genes [Table 25] Information on Antibody Detection by Flow Cell Analysis Example 15 CLDN18.2×CD3 Affinity of bispecific antibodies to target protein of stone crab and macaque Antibody binding affinity was detected using FACS. The cells tested included HEK293 cells overexpressing CLDN18.2 (HEK293 / cynoCLDN18.2), HEK293 cells overexpressing CLDN18.1 (HEK293 / cynoCLDN18.1), and CD3-positive T cells from lizard monkeys. CD3-positive T cells were isolated from lizard monkey PBMCs using a non-human primate CD3 cell isolation kit (Miltenyi, 130-092-012). The binding affinity assay was performed as follows: cells were centrifuged at 300 g for 5 minutes and then resuspended in FACS buffer (PBS containing 2% FBS). The cell density was adjusted to 10-1. 6 Cells / ml: 50 μl of cell suspension was added to each well of a 96-well plate. Antibody was diluted to different concentrations with FACS buffer, and 50 μl of antibody dilution was added to each well of the 96-well plate. After incubation at 4°C for 2 hours, the plate was washed twice with FACS buffer. Then, FACS buffer containing APC-conjugated goat anti-human IgG secondary antibody (final concentration 1.5 μg / ml, Jackson, 109-605-098) was added. After incubation at 4°C for 1 hour, the plate was washed twice with FACS buffer. Cells were resuspended in fixative, and cell fluorescence was monitored using a FACS analyzer (ACEA NovoCyte). Figure 8 and Table 26 show the binding affinity of the antibody to HEK293 cells overexpressing CLDN18.1 and CLDN18.2 of the rhesus macaque, and CD3-positive T cells in PBMCs of the rhesus macaque. The antibodies tested could bind to cells exhibiting CLDN18.2 and CD3 expression, but not to cells exhibiting CLDN18.1 expression. [Table 26] Binding affinity of CLDN18.2×CD3 bispecific antibody to cells. Example 16 CLDN18.2×CD3 Bispecific antibodies against stone crab macaques CLDN18.2 Expression cells TDCC active Using CytoTox 96 ®A non-radioactive cytotoxicity assay kit (Promega, G1780) was used to detect the activity of the CLDN18.2×CD3 bispecific antibody against the TDCC effect induced by HEK293 / cynoCLDN18.2. CD3-positive T cells were isolated from PBMCs of *Rhesus macaque* using a non-human primate CD3 cell isolation kit (Miltenyi, 130-092-012). *Rhesus macaque* T cells and target cells were resuspended in culture medium (RPMI 1640 + 5% FBS). The target cell density was adjusted to 3 × 10⁶ cells / year. 5 Cells / ml, T cell density adjusted to 1.2 × 10⁻⁶ 6 / ml, add 50 μl of each of the two cell types to the wells of a 96-well plate (effect-to-target ratio 4:1). Dilute the test antibody to different concentrations with culture medium (RPMI 1640 + 5% FBS) and add 50 μl to the wells of a 96-well plate. Incubate the samples at 37°C for 24 hours, then add 10x Triton-X 100 lysis buffer (RPMI 1640 + 5% FBS + 10% Triton-X 100) to the target cell maximum LDH release control well and the volume correction control well, mix well, and incubate at 37°C for 0.5 hours. Centrifuge the 96-well plate at 400 g for 4 minutes, take 50 μl of supernatant, and then add LDH chromogenic solution at a concentration of 50 μl / well. After the mixture is incubated at room temperature in the dark for 20 minutes, the plate is read on an MD StakMax (OD). 490 The results were calculated by first calculating the correction readings, subtracting the culture medium background control reading from the readings of the experimental wells, the target cell spontaneous LDH release control wells, and the effector cell spontaneous LDH release control wells; then, subtracting the volume correction control well reading from the target cell maximum LDH release control well reading. TDCC activity (%) = (experimental well correction reading - effector cell spontaneous LDH release control well correction reading - target cell spontaneous LDH release control well correction reading) / (target cell maximum LDH release control well correction reading - target cell spontaneous LDH release control well correction reading) × 100. Figure 9 shows the TDCC activity induced by the antibody against HEK293 / cynoCLDN18.2, with specific values ​​shown in Table 27. [Table 27] TDCC activity of CLDN18.2×CD3 bispecific antibody Example 17 In vitro cytokine release assay Human PBMCs were used to detect the in vitro release of interferons induced by the CLDN18.2×CD3 bispecific antibody, thereby predicting antibody safety. PBMCs and the antibody were co-incubated, or PBMCs, the antibody, and HEK293 / hCLDN18.1 were co-incubated. PBMCs and HEK293 / hCLDN18.1 were resuspended in medium (RPMI 1640 + 10% FBS). The density of HEK293 / hCLDN18.1 was adjusted to 1.5 × 10⁻⁶. 6 Cells / ml, PBMC cell density adjusted to 2 × 10⁻⁶ 6 / ml. Add 100 μl of HEK293 / hCLDN18.1 cells and 200 μl of PBMC to the wells of a 48-well plate. Dilute the test antibody to different concentrations with culture medium (RPMI 1640 + 10% FBS) and add 100 μl to each well of the 48-well plate, for a final volume of 400 μl. LPS (Sigma, L6529) was used as a positive control. Iso hIgG1 (CrownBio, C0001-4) antibody was used as a negative control. The samples were incubated at 37°C for 24 hours. The supernatant was centrifuged at 300 g for 10 minutes, and 300 μl of the supernatant was collected. The concentrations of IL-6 (Invitrogen, 88-7066) and TNF-α (Invitrogen, 88-7346) in the supernatant were quantified by ELISA. Figure 10 shows the in vitro induced cytokine release. In the absence of CLDN18.2 target cells, PR004549 induced higher levels of IL-6 and TNF-α release than PR006292, indicating that PR006292 has better safety. Example 18 CLDN18.2×CD3 Bispecific antibodies ADCC active Using CytoTox 96 ®A non-radioactive cytotoxicity assay kit (Promega, G1780) was used to detect the activity of the CLDN18.2×CD3 bispecific antibody against Jurkat cells and the ADCC effect induced by HEK293 / hCLDN 18.2. Human PBMCs were centrifuged at 300 g for 10 min and cultured overnight in medium (RPMI 1640 + 10% FBS). NK cells were isolated from human PBMCs using a human NK cell isolation kit (Miltenyi, 130-092-657). Jurkat cells were centrifuged at 300 g for 5 min, and human NK cells were centrifuged at 300 g for 10 min, then resuspended in medium (RPMI 1640 + 5% FBS). The target cell density was adjusted to 3 × 10⁶ cells / year. 5 Cells / ml, NK cell density adjusted to 1.8 × 10⁻⁶ 6 / 50 μl of each of the two cell types was added to the wells of a 96-well plate (effect-to-target ratio 6:1). The antibody to be tested was diluted to different concentrations with culture medium (RPMI 1640 + 5% FBS) and 50 μl was added to the wells of the 96-well plate. The samples were incubated at 37°C for 5 hours. Then, 10x Triton-X 100 lysis buffer (RPMI 1640 + 5% FBS + 10% Triton-X 100) was added to the target cell maximum LDH release control well and the volume correction control well, mixed thoroughly, and incubated at 37°C for 0.5 hours. The 96-well plate was centrifuged at 300 g for 5 minutes, and 50 μl of supernatant was collected. Then, LDH chromogenic solution was added at a concentration of 50 μl / well. After the mixture was incubated at room temperature in the dark for 20 minutes, the plate was read on an MD StakMax (OD). 490 PR003767 was used as a positive control, and Iso hIgG1 (CrownBio, C0001-4) antibody was used as a negative control. Results were calculated by first calculating corrected readings: subtracting the culture medium background control reading from the readings of the experimental wells, target cell spontaneous LDH release control wells, and effector cell spontaneous LDH release control wells; then subtracting the volume correction control well reading from the target cell maximum LDH release control well reading. ADCC activity (%) = (Experimental well corrected reading - Effector cell spontaneous LDH release control well corrected reading - Target cell spontaneous LDH release control well corrected reading) / (Target cell maximum LDH release control well corrected reading - Target cell spontaneous LDH release control well corrected reading) × 100. Figure 11(a) shows the ADCC activity of the antibody against Jurkat. PR006292 and PR004549 failed to induce ADCC effects in Jurkat cells. The activity of the CLDN18.2×CD3 bispecific antibody against NUGC4_D8 cells induced ADCC was detected using NK92 / CD16a cells. NUGC4_D8 cells and NK92 / CD16a cells were resuspended in RPMI 1640 + 5% FBS. The target cell density was adjusted to 3 × 10⁻⁶ cells / year. 5 Cells / ml, NK92 / CD16a cell density adjusted to 1.8 × 10⁻⁶. 6 / 50 μl of each of the two cell types was added to the wells of a 96-well plate (effect-to-target ratio 6:1). The test antibodies were diluted to different concentrations with culture medium (RPMI 1640 + 5% FBS) and 50 μl was added to the wells of the 96-well plate. The samples were incubated at 37°C for 5 hours. PR003197 was used as a positive control. Figure 11(b) shows the ADCC activity of the antibodies against NUGC4_D8 cells. PR006292 and PR004549 failed to induce ADCC effects against NUGC4_D8 cells. Example 19 CLDN18.2×CD3 Bispecific antibodies CDC active The ability of the CLDN18.2 antibody to induce a CDC effect in HEK293 / hCLDN18.2 and Jurkat cells was detected using a CellTiter-Glo chemiluminescence immunoassay kit (Promega, G7573). Target cells were centrifuged at 300 g for 5 minutes and then resuspended in RPMI 1640 serum-free medium. The target cell density was adjusted to 2 × 10⁶ cells / year. 5Cells / ml: Add 25 μl of cell suspension to each well of a 96-well plate. Dilute the antibody to different concentrations with serum-free medium, and add 25 μl of antibody dilution to each well of a 96-well plate. Add 50 μl of normal human serum (Access cell culture, 515) to a final concentration of 50%, and incubate the resulting mixture at 37°C for 24 hours. Incubate the 96-well plate at room temperature for 30 minutes, and add 100 μl / well of room temperature CellTiter-Glo chromogenic solution. Then incubate the samples at room temperature in the dark for 10 minutes. Read the plate using a PE Enspire. CDC activity (%) = [1 - (luminescent sample) / (luminescent mock control)] × 100. IMAB362 analogue was used as a positive control, and Iso hIgG1 (CrownBio, C0001-4) antibody was used as a negative control. Figure 12 shows the CDC activity of PR006292 antibody against Jurkat cells and HEK293 cells overexpressing human CLDN18.2. PR006292 induced a stronger CDC effect in HEK293 / hCLDN18.2 cells than PR004549, while no CDC activity was observed in Jurkat cells. The specific CDC activities of the CLDN18.2×CD3 bispecific antibody are shown in Table 28. [Table 28] CDC Activity of CLDN18.2×CD3 Bispecific Antibody Example 20 CL ​​DN18.2×CD3 Competitive binding activity of bispecific antibodies This embodiment aims to investigate the epitope region of human CLDN18.2 antigen binding to bispecific antibodies against human CLDN18.2. Competitive binding experiments at the cellular level were performed using HEK293 / hCLDN18.2 cells overexpressing human CLDN18.2. In short, biotinylation kits (ThermoFisher, A35358) were used to biotinylate anti-human CLDN18.2 antibodies PR000400 and PR004549 according to the manufacturer's instructions. 2 × 10⁻⁶ ppm of the anti-human CLDN18.2 antibodies were added to a 96-well V plate (Corning, 3894). 6Add 50 µl / well of HEK293T / hCLDN18.2 cell suspension overexpressing human CLDN18.2, followed by 25 µl of biotinylated anti-human CLDN18.2 antibody PR000400 or PR004549 and mix thoroughly. Incubate at 4°C for 30 min. Add 25 µl of the corresponding serially diluted unbiotinylated anti-human CLDN18.2 antibody and mix thoroughly. Incubate at 4°C for 1 h. Wash cells twice with 200 µl / well of pre-chilled FACS buffer (2% BSA in DPBS), centrifuge at 500 g, 4°C for 5 min, and discard the supernatant. Add fluorescent secondary antibody (Jackson ImmunoResearch, 016-540-084, 1:500) and incubate at 4°C in the dark for 1 h. Cells were washed twice with 200 µL / well pre-chilled FACS buffer (2% BSA in DPBS), centrifuged at 500 g, 4°C for 5 min, and the supernatant was discarded. Finally, cells were resuspended in 200 µL / well pre-chilled FACS buffer, and fluorescence information was read using ACEA_NovoCyte. The inhibition rate was calculated using the formula: Inhibition rate (%) = (AB) / A * 100 (Note: A: Fluorescence information after interaction of biotinylated antibody with ISO (hIgG1) (Crownbio, c0001-4); B: Fluorescence information after interaction of biotinylated antibody with non-biotinylated antibody). As shown in Figure 13 and Table 29, the bispecific antibodies against CLDN18.2 of this invention can block the binding of PR000400 or PR004549 to human CLDN18.2, and the detected antibody blocking ability increases positively with antibody concentration, indicating that these antibodies have very similar epitopes to PR000400 and PR004549. The detection antibody has low affinity for HEK293 / hCLDN18.1 cells. Based on the above results, it can be inferred that the detection antibody system binds to ECL1 (Extracellular loop 1) of human CLDN18.2 protein rather than ECL2. [Table 29] Competitive binding affinity of CLDN18.2 antibody to antibodies PR000400 and PR004549 Example 21 CLDN18.2×CD3 Pharmacokinetics of Bispecific Antibodies Pharmacokinetic studies were conducted using BALB / c nude mice. Six female BALB / c nude mice weighing 18–22 grams were selected and administered the antibody drug intravenously at a dose of 5 mg / kg. Whole blood was collected from one group of three mice before administration and at 15 minutes, 24 hours (day 1), day 4, and day 10 after administration. Whole blood was collected from another group of three mice before administration and at 5 hours, day 2, day 7, and day 14 after administration. The whole blood was allowed to stand for 30 minutes to coagulate, then centrifuged, and the separated serum samples were frozen at -80°C until analysis. The drug concentration in mouse serum was quantitatively determined using an ELISA method. The total ELISA method used goat anti-human Fc multi-strain antibodies to capture antibodies containing human Fc in mouse serum, followed by the addition of HRP-labeled goat anti-human Fc secondary antibody for detection. The CLDN18.2 binding domain ELISA method (Free X method) uses the CLDN18.2 protein to capture antibodies containing the CLDN18.2 binding domain in mouse serum, and then adds HRP-labeled goat anti-human Fc secondary antibody for detection. Phoenix WinNonlin software version 8.2 was used, and a non-compartmental (NCA) model was selected to analyze blood drug concentration data to evaluate its pharmacokinetic parameters. The pharmacokinetics of PR006292, PR006384, and PR004549 are shown in Figure 14, and the pharmacokinetic parameters are shown in Table 30. PR006292 and PR006384 showed better stability than PR004549 in mice, with longer half-lives and higher drug exposure rates. [Table 30] Pharmacokinetic parameters of PR006292, PR006384, and PR004549 *AUC (%) = AUC / AUC ( 總檢測方法 ; 平均值 ) * 100 Example 22 CLDN18.2×CD3 In vivo pharmacodynamic studies of bispecific antibodies NUGC4_D8 Tumor model In vivo pharmacodynamic studies were conducted using a NUGC4_D8 tumor model derived from the reconstructed human PBMC immune system in NCG mice. The specific method was as follows: on the day of cell inoculation, each NCG mouse was subcutaneously inoculated with NUGC4_D8 tumor cells and PBMCs. When the average tumor volume of each group of mice reached 90 mm... 3 The drugs were administered in groups, with a total of one dose given via tail vein injection. After the start of administration, body weight and tumor volume were measured twice weekly. Tumor volume was calculated as follows: Tumor volume (mm²) 3 = 0.5 × Tumor long diameter × Tumor short diameter 2 Data analysis was performed using t-tests. The in vivo antitumor effects of PR005397, PR005411, and PR004549 are shown in Figure 15(a). Specifically, the average tumor volume of the Iso hIgG1 control mice was 1897 mm on day 25 post-inoculation. 3 The average tumor volume in the test drug PR004549 (0.2 mg / kg) treatment group was 104 mm on day 25 post-inoculation. 3 Compared to the Iso hIgG1 control group, there was a significant difference (p < 0.0001), with a tumor inhibition rate (TGI) of 94.48%. In the test drug PR004549 (0.04 mg / kg) treatment group, the mean tumor volume on day 25 post-inoculation was 538 mm. 3 Compared to the Iso hIgG1 control group, there was a significant difference (p < 0.0001), with a tumor inhibition rate (TGI) of 71.61%. In the test drug PR005411 (0.2 mg / kg) treatment group, the average tumor volume on day 25 after inoculation was 30 mm. 3 Compared to the Iso hIgG1 control group, there was a significant difference (p < 0.0001), with a tumor inhibition rate (TGI) of 98.39%. In the test drug PR005411 (0.04 mg / kg) treatment group, the average tumor volume on day 25 after inoculation was 263 mm. 3 Compared to the Iso hIgG1 control group, there was a significant difference (p < 0.0001), with a tumor inhibition rate (TGI) of 86.1%. In the test drug PR005397 (0.04 mg / kg) treatment group, the average tumor volume on day 25 after inoculation was 327 mm. 3Compared with the Iso hIgG1 control group, there was a significant difference (p < 0.0001), with a tumor inhibition rate (TGI) of 82.75%. Throughout the treatment period, the animals showed good drug tolerance, and no serious weight loss or death occurred. The in vivo antitumor effects of PR005397 and PR005411 were superior to those of PR004549. The in vivo antitumor effects of PR006292, PR006293, PR006384, and PR004549 are shown in Figure 15(b). Specifically, the average tumor volume of the Iso hIgG1 control group mice was 1355 mm on day 25 post-inoculation. 3 The average tumor volume in the test drug PR004549 (0.2 mg / kg) treatment group was 408 mm on day 25 post-inoculation. 3 Compared to the Iso hIgG1 control group, there was a significant difference (p = 0.0001), with a tumor inhibition rate (TGI) of 69.83%. In the test drug PR004549 (0.04 mg / kg) treatment group, the mean tumor volume on day 25 post-inoculation was 743 mm. 3 Compared to the Iso hIgG1 control group, there was a significant difference (p = 0.0037), with a tumor inhibition rate (TGI) of 45.15%. In the test drug PR006293 (0.2 mg / kg) treatment group, the average tumor volume on day 25 after inoculation was 39 mm. 3 Compared to the Iso hIgG1 control group, there was a significant difference (p < 0.0001), with a tumor inhibition rate (TGI) of 97.06%. In the test drug PR006293 (0.04 mg / kg) treatment group, the average tumor volume on day 25 after inoculation was 190 mm. 3 Compared to the Iso hIgG1 control group, there was a significant difference (p < 0.0001), with a tumor inhibition rate (TGI) of 85.96%. In the test drug PR006384 (0.2 mg / kg) treatment group, the average tumor volume on day 25 after inoculation was 81 mm. 3 Compared to the Iso hIgG1 control group, there was a significant difference (p < 0.0001), with a tumor inhibition rate (TGI) of 94%. In the test drug PR006384 (0.04 mg / kg) treatment group, the average tumor volume on day 25 after inoculation was 752 mm. 3Compared to the Iso hIgG1 control group, there was a significant difference (p = 0.0071), with a tumor inhibition rate (TGI) of 44.47%. In the test drug PR006292 (0.04 mg / kg) treatment group, the average tumor volume on day 25 after inoculation was 580 mm. 3 Compared to the Iso hIgG1 control group, there was a significant difference (p = 0.0006), with a tumor inhibition rate (TGI) of 57.15%. Throughout the treatment period, the animals exhibited good drug tolerance, with no severe weight loss or death. The in vivo antitumor effects of PR006292, PR006293, and PR006384 were superior to those of PR004549. SNU620 PBMC Tumor model In vivo pharmacodynamic studies were conducted using an SNU620 tumor model derived from the reconstructed human PBMC immune system in NCG mice. The specific method was as follows: on the day of cell inoculation, each NCG mouse was subcutaneously inoculated with SNU620 tumor cells. When the average tumor volume of each group of mice reached 70 mm... 3 The patients were administered the medication in groups, with a total of four doses given via tail vein injection. After the start of administration, body weight and tumor volume were measured twice weekly. Tumor volume was calculated as follows: Tumor volume (mm²) 3 = 0.5 × Tumor long diameter × Tumor short diameter 2 Data analysis was performed using t-tests. The in vivo antitumor effects of PR006292 and PR004549 are shown in Figure 15(c). Specifically, the average tumor volume of the Iso hIgG1 control mice was 847 mm on day 36 post-inoculation. 3 The average tumor volume in the test drug PR006292 (0.2 mg / kg) treatment group was 131 mm on day 36 post-inoculation. 3 Compared to the Iso hIgG1 control group, there was a significant difference (p = 0.0076), with a tumor inhibition rate (TGI) of 84.53%. In the test drug PR006292 (0.04 mg / kg) treatment group, the mean tumor volume on day 36 post-inoculation was 505 mm. 3There was no significant difference compared to the Iso hIgG1 control group (p = 0.3856), and the tumor inhibition rate (TGI) was 40.36%. In the test drug PR004549 (0.2 mg / kg) treatment group, the mean tumor volume on day 36 post-inoculation was 858 mm. 3 Compared to the Iso hIgG1 control group, there was a significant difference (p > 0.9999), with a tumor inhibition rate (TGI) of -1.28%. In the test drug PR004549 (0.04 mg / kg) treatment group, the mean tumor volume on day 36 post-inoculation was 844 mm. 3 Compared with the Iso hIgG1 control group, there was a significant difference (p > 0.9999), with a tumor inhibition rate (TGI) of 0.39%. Throughout the treatment period, the animals showed good drug tolerance, and no serious weight loss or death occurred. HuP-T4 PBMC Tumor model In vivo pharmacodynamic studies were conducted using a HuP-T4 tumor model derived from the reconstructed human PBMC immune system in NCG mice. The specific method was as follows: On the day of cell inoculation, each NCG mouse was subcutaneously inoculated with HuP-T4 tumor cells. When the average tumor volume of each group of mice reached 130 mm... 3 The patients were administered the medication in groups, with a total of four doses given via tail vein injection. After the start of administration, body weight and tumor volume were measured twice weekly. Tumor volume was calculated as follows: Tumor volume (mm²) 3 = 0.5 × Tumor long diameter × Tumor short diameter 2 Data analysis was performed using t-tests. The in vivo antitumor effect of PR006292 is shown in Figure 15(d). Specifically, the average tumor volume of the Iso hIgG1 control mice was 1059 mm on day 36 post-inoculation. 3 The average tumor volume in the test drug PR006292 (0.5 mg / kg) treatment group was 129 mm on day 36 post-inoculation. 3 The tumor inhibition rate (TGI) was 87.75% compared to the Iso hIgG1 control group (p = 0.0022). Throughout the treatment period, the animals showed good drug tolerance and no serious weight loss or death occurred. In vivo cytokine release assay in mice In vivo cytokine storm studies were conducted using NCG mice to reconstruct the human PBMC immune system. The specific method was as follows: each NCG mouse was intravenously injected with 2 x 10⁻⁶ NCG mice. 7 Individual PBMCs were administered PR006292, PR004549, and a control IgG1 antibody intravenously the following day. Blood samples were collected at hour 0 (before administration), 4 hours after administration, and 24 hours after administration. Serum levels of multiple intercellular substances, including IFN-γ, IL-10, IL-12p70, IL-13, IL-1β, IL-2, IL-4, IL-6, IL-8, and TNF-α, were measured using the MSD method. Data analysis was performed using a t-test. Figure 15(e) shows the presence of detectable interferons in mouse serum 4 hours after antibody injection. The results indicate that PR006292 induced lower release of interferons such as IFN-γ, IL-2, and TNF-α compared to the control antibody PR004549, suggesting better safety. none [Figures 1a-f] show the binding affinity of HCAB antibodies to (ab) NUGC4_D8, (c) SNU601, (d) HEK293 / hCLDN18.2 and (ef) HEK293 / hCLDN18.1 cells. [Figures 2a-b] show the competitive binding activity of HCAB antibody against (a) PR000400 and (b) PR004549 in HEK293 / hCLDN18.2 cells. [Figures 3a-b] show the survival rate of target cells when the antibody is co-cultured with anti-human IgG antibody coupled with MMAF. [Figures 4a-h] show the structure of the CLDN18.2×CD3 bispecific antibody. [Figures 5a-s] show the binding affinity of the CLDN18.2×CD3 bispecific antibody to (ac) HEK293 / hCLDN18.2, (di) NUGC4_D8, (jo) Jurkat and (ps) HEK293 / hCLDN18.1 cells. [Figures 6a-w] show the expression level of CLDN18.2 in (a) IM95 cells and the TDCC activity of bispecific antibodies against (bk) NUGC4_D8, (Lo) IM95, (ps) HEK293 / hCLDN18.1, and (tw) SNU620 cells. [Figure 7] shows the binding affinity of the CLDN18.2×CD3 bispecific antibody to cells overexpressing the human CLDN18.2 paralogous protein family CLDN1, CLDN2, CLDN3, CLDN4, CLDN6 and CLDN9. [Figures 8a-c] show the binding affinity of the CLDN18.2×CD3 bispecific antibody to (a) HEK293 / cynoCLDN18.1, (b) HEK293 / cynoCLDN18.2, and (c) CD3+ T cells of rock crab macaques. [Figure 9] shows the TDCC activity of CLDN18.2×CD3 bispecific antibody against HEK293 / cynoCLDN18.2 cells via cyno T cells. [Figure 10a-b] shows the release of cytokines induced in vitro by the CLDN18.2×CD3 bispecific antibody. (a) TNFα (b) IL-6. [Figure 11a-b] shows the ADCC activity of CLDN18.2×CD3 bispecific antibody against (a) Jurkat and (b) NUGC4_D8 cells. [Figure 12a-b] shows the CDC activity of the CLDN18.2×CD3 bispecific antibody against (a) HEK293 / hCLDN18.2 and (b) Jurkat cells. [Figure 13a-b] shows the competitive binding activity of the CLDN18.2×CD3 bispecific antibody against (a) PR000400 and (b) PR004549 in HEK293 / hCLDN18.2 cells. [Figure 14] shows the pharmacokinetics of the CLDN18.2×CD3 bispecific antibody. [Figures 15a-e] show the in vivo pharmacodynamic studies of the CLDN18.2×CD3 bispecific antibody in tumor models (ab) NUGC4_D8 PBMC, (c) SNU620 PBMC and (d) HuP-T4 PBMC, and (e) in vivo cytokine storm studies of the antibody. none TW202334215A_111129765_SEQL.xml

Claims

1. An antibody comprising a single chain targeting CLDN18.2, comprising a heavy chain variable region including HCDR1, HCDR2, and HCDR3; wherein, The HCDR1 contains an amino acid sequence as shown in SEQ ID NO: 16, the HCDR2 contains an amino acid sequence as shown in SEQ ID NO: 51, and the HCDR3 contains an amino acid sequence as shown in SEQ ID NO:

78.

2. The antibody of claim 1, wherein the heavy chain variable region further comprises a frame region, wherein in the heavy chain variable region frame region, HFR1 comprises an amino acid sequence as shown in SEQ ID NO: 6, HFR2 comprises an amino acid sequence as shown in SEQ ID NO: 29, HFR3 comprises an amino acid sequence as shown in SEQ ID NO: 64, and HFR4 comprises an amino acid sequence as shown in SEQ ID NO:

86.

3. The antibody of claim 2, wherein the heavy chain variable region comprises an amino acid sequence as shown in SEQ ID NO:

162.

4. The antibody as claimed in claim 1, wherein the antibody further comprises a heavy chain constant region.

5. A bispecific antibody comprising a first protein functional region targeting CD3 and a second protein functional region comprising a single chain targeting CLDN18.2, wherein the second protein functional region comprises a heavy chain variable region comprising HCDR1, HCDR2 and HCDR3; wherein HCDR1, HCDR2 and HCDR3 comprise amino acid sequences as shown in SEQ ID NO: 16, SEQ ID NO: 51 and SEQ ID NO: 78, respectively.

6. The bispecific antibody of claim 5, wherein the first protein functional region comprises a light chain variable region and a heavy chain variable region, the light chain variable region comprising LCDR1, LCDR2 and LCDR3 as shown in SEQ ID NO: 101, SEQ ID NO: 116 and SEQ ID NO: 131 respectively, and the heavy chain variable region comprising HCDR1, HCDR2 and HCDR3 as shown in SEQ ID NO: 11, SEQ ID NO: 38 and SEQ ID NO: 72 respectively.

7. Bispecific antibodies as requested in item 5 or 6, wherein, This bispecific antibody contains three polypeptide chains, as follows: The first polypeptide chain is shown as: VHCLDN18.2-linker peptide-VHCLDN18.2-hinge-CH2-CH3; the second polypeptide chain is shown as: VHCD3-CH1-hinge-CH2-CH3; and the third polypeptide chain is shown as: VLCD3-CL.

8. The bispecific antibody of claim 7, wherein the first polypeptide chain comprises an amino acid sequence as shown in SEQ ID NO: 236, the second polypeptide chain comprises an amino acid sequence as shown in SEQ ID NO: 235, and the third polypeptide chain comprises an amino acid sequence as shown in SEQ ID NO:

200.

9. The bispecific antibody of claim 5, wherein the heavy chain variable region comprises an amino acid sequence as shown in SEQ ID NO:

162.

10. The bispecific antibody of claim 6, wherein the CD3-targeting functional region comprises a heavy chain variable region and a light chain variable region, the heavy chain variable region comprising an amino acid sequence as shown in SEQ ID NO: 149, and the light chain variable region comprising an amino acid sequence as shown in SEQ ID NO:

168.

11. A pharmaceutical composition comprising an antibody as claimed in any one of claims 1 to 4 or a bispecific antibody as claimed in any one of claims 5 to 10, and a pharmaceutically acceptable carrier.

12. The pharmaceutical composition of claim 11, wherein the pharmaceutical composition further comprises one or more of the group consisting of hormone preparations, targeted small molecule preparations, proteasome inhibitors, imaging agents, diagnostic agents, chemotherapeutic agents, oncolytic drugs, cytotoxic agents, interferons, activators of co-stimulatory molecules, inhibitors of inhibitory molecules, and vaccines.

13. A kit for the treatment of CLDN18.2-mediated diseases or conditions, comprising kit A and kit B, wherein: Kit A contains an antibody as claimed in any of claims 1 to 4 or a bispecific antibody as claimed in any of claims 5 to 10; Kit B contains other antitumor antibodies or pharmaceutical compositions containing said other antitumor antibodies, and / or is one or more of the group consisting of hormone preparations, targeted small molecule preparations, proteasome inhibitors, imaging agents, diagnostic agents, chemotherapeutic agents, oncolytic drugs, cytotoxic agents, interferons, activators of co-stimulatory molecules, inhibitors of inhibitory molecules, and vaccines.

14. Use of an antibody as claimed in any one of claims 1 to 4, a bispecific antibody as claimed in any one of claims 5 to 10, a pharmaceutical composition as claimed in claim 11 or 12, or a kit as claimed in claim 13, for the preparation of a medicament for cancer therapy.

15. Use of an antibody as claimed in any one of claims 1 to 4, a bispecific antibody as claimed in any one of claims 5 to 10, a pharmaceutical composition as claimed in claim 11 or 12, or a kit as claimed in claim 13, for the preparation of a medicament for the treatment and / or prevention of CLDN18.2 positive tumors.

16. As claimed in claim 15, wherein the tumor is selected from gastric cancer, esophageal cancer, lung cancer, ovarian cancer, endometrial cancer, cervical cancer, melanoma, kidney cancer, breast cancer, colorectal cancer, liver cancer, pancreatic cancer, bile duct cancer, gallbladder cancer, sarcoma, bladder cancer, head and neck cancer, bronchial cancer, glioma and / or leukemia.

17. As claimed in claim 16, wherein the tumor is breast cancer, ovarian cancer, endometrial cancer, kidney cancer, or bile duct cancer.

Images