Anti-CDCP1 Antibody

Abstract

The purpose of the present invention is to provide: an anti-hCDCP1 antibody that can be used as an active ingredient in an anticancer agent or the like that has few side effects; and an ADC including the anti-hCDCP1 antibody. Specifically, the present invention is an antibody that binds to human CUB domain-containing protein 1 (CDCP1), and that is characterized by having low binding ability to human CD34-positive cells, or an antigen-binding fragment of said antibody. A representative example of the antibody is one that has a heavy chain CDR1 including the amino acid sequence represented by SEQ ID NO: 25, a heavy chain CDR2 including the amino acid sequence represented by SEQ ID NO: 26, a heavy chain CDR3 including the amino acid sequence represented by SEQ ID NO: 27, a light chain CDR1 including the amino acid sequence represented by SEQ ID NO: 29, a light chain CDR2 including the amino acid sequence represented by SEQ ID NO: 30, and a light chain CDR3 including the amino acid sequence represented by SEQ ID NO: 31.

Description

[Technical field]

[0001] The present invention relates to an anti-CDCP1 antibody, more specifically to an anti-human CDCP1 antibody and a fragment thereof that have low binding affinity to bone marrow hematopoietic stem cells from healthy individuals, and uses thereof. [Background technology]

[0002] Human CDCP1 (cub domain containing protein 1) (hereinafter also referred to as hCDCP1) is a type I cell transmembrane protein consisting of a total length of 836 amino acids and having three CUB domains (complement C1r / C1s, Uegf, Bmp1 domains) (see Non-Patent Document 1). CUB domains are present in many proteins, including BMP1 and C1r, proteins with protease activity such as TMPRSS7, and proteins involved in cell-cell interactions such as LRP3, NRP1, and TLL2, but a unified understanding of their molecular functions is lacking. There are many N-glycosylation sites on the N-terminus of the extracellular domain, and a correlation has been found between the degree of glycosylation and the metastatic state of cancer cells (see Non-Patent Document 2). hCDCP1 is expressed in various cells, but its soluble ligand is unknown, while it has been suggested that it interacts with various membrane proteins such as EGFR (see Non-Patent Document 3). The hCDCP1 protein exhibits a band size of 135 kDa in electrophoresis, but when the amino acid site at positions 368 / 369 or 369 / 370 between the CUB1 and CUB2 domains is cleaved by a serine protease, a 70 kDa fragment is generated (see Non-Patent Document 1).

[0003] hCDCP1 is phosphorylated at tyrosine residues 734, 743, and 762 in the intracellular domain by the action of Src family kinases. Phosphorylated hCDCP1 induces downstream signals by phosphorylating PKCδ, promoting anchorage-independent growth, extracellular matrix degradation, cell migration, and epithelial-mesenchymal transition, leading to cancer cell metastasis. hCDCP1 is also known to interact with various molecules, including EGFR and HER2, to promote cancer cell growth and metastasis.

[0004] It has been reported that hCDCP1 is expressed in various cancer cells and normal tissues. For example, expression of hCDCP1 mRNA and protein has been reported in cancer cells such as prostate cancer, lung cancer, colon cancer, and ovarian cancer, cell lines established therefrom, and normal tissues such as the colon, skin, small intestine, and prostate (see Patent Document 1 and Patent Document 2).

[0005] To date, several antibodies against hCDCP1 (hereinafter sometimes referred to as anti-human CDCP1 or anti-hCDCP1 antibodies) are known. Patent Document 2 discloses an anti-hCDCP1 polyclonal antibody and a method for screening, diagnosing, and treating ovarian cancer using this antibody. Patent Document 1 discloses an anti-hCDCP1 monoclonal antibody (clone name: 25A11). It has been shown that an ADC (antibody-drug-conjugate) antibody obtained by binding saporin to 25A11 exhibits cytotoxic activity against PC3 cancer cell lines in vitro, and that intravenous administration of this ADC antibody exerts a significant tumor growth inhibitory activity. Furthermore, Patent Document 4 discloses multiple anti-hCDCP1 antibodies, and discloses that an anti-hCDCP1 antibody ADCed with PBD (pyrrolobenzodiazepine) exhibits cytotoxic activity against prostate cancer cell lines in vitro, and that an anti-hCDCP1 antibody ADCed with MMAE (monomethyl auristatin E) exhibits antitumor activity in mouse xenograft models transplanted with breast cancer cell lines, colon cancer cell lines or prostate cancer cell lines.

[0006] Furthermore, Patent Document 3 discloses four anti-hCDCP1 monoclonal antibodies (CUB1 antibody, CUB2 antibody, CUB3 antibody, and CUB4 antibody; derived from hybridoma clones CUB1 to 4, respectively). These anti-hCDCP1 monoclonal antibodies inhibit normal CD34 positive (CD34 + ) cells and normal CD133 positive (CD133 + It has been disclosed that the antibody binds to the hCDCP1 protein expressed in .) cells, and can therefore be used to separate and identify hematopoietic stem cells, mesenchymal stem cells, neural stem cells, and the like. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] International Publication No. WO2008 / 133851 [Patent Document 2] European Patent Publication No. 1677875 [Patent Document 3] U.S. Patent Publication No. 7,541,030 [Patent Document 4] International Publication No. WO2018 / 112334 [Non-patent literature]

[0008] [Non-Patent Document 1] Mostageer et al., Oncogene 20:4402-4408 2001. [Non-Patent Document 2] Yang et al., Oncotarget 6:43743-58 2015. [Non-Patent Document 3] He et al., Oncoscience 3:5-8 2016. Summary of the Invention [Problem to be solved by the invention]

[0009] As described above, the expression of hCDCP1 mRNA and hCDCP1 protein has been confirmed in normal tissues / cells and cancer tissues / cells. The CUB1 antibody disclosed in Patent Document 3 inhibits hCDCP1-expressing K562 cells (human chronic myeloid leukemia cells) as well as normal CD34 cells. + / CD38 - It has also been shown to bind strongly to bone marrow cells. Furthermore, the inventors confirmed that the anti-hCDCP1 antibody derived from 25A11 disclosed in Patent Document 2, the CUB4 antibody disclosed in Patent Document 3, and the anti-hCDCP1 antibody (clone name: CUB1) commercially available from BioLegend, bind strongly to various cancer cells expressing hCDCP1 as well as to normal CD34-positive bone marrow cells (see Figures 11, 12, 13A, 15, and 16). The CD34-positive cell population among bone marrow cells contains hematopoietic stem cells that have the ability to regenerate all human blood cells, and CD34 is thought to be one of the cell surface markers of human hematopoietic stem cells.

[0010] The property of these known antibodies to bind strongly to hematopoietic stem cells causes serious problems when these anti-hCDCP1 antibodies are used as antitumor drugs. That is, normal bone marrow cells bound to the anti-hCDCP1 antibodies are attacked by immune functions (e.g., ADCC activity), which may cause functional inhibition and suppression of the tissues to which the cells belong, resulting in serious side effects. Furthermore, when anti-hCDCP1 antibodies are converted into ADCs, the ADC antibodies also harm normal tissues and cells, resulting in more significant and serious adverse effects.

[0011] Therefore, an object of the present invention is to provide an anti-hCDCP1 antibody that is unlikely to induce the above-mentioned serious side effects. [Means for solving the problem]

[0012] The inventors investigated the antigen-binding properties of various anti-hCDCP1 antibodies and succeeded in preparing a novel anti-hCDCP1 antibody that binds to various cancer cells that express hCDCP1, while binding relatively weakly to CD34-positive cells such as hematopoietic stem cells that express hCDCP1. Thus, the present invention relates to an antibody that binds to CDCP1 and is characterized by having low binding affinity to CD34-positive cells, and an antigen-binding fragment thereof. Effect of the Invention

[0013] The present invention provides an anti-hCDCP1 antibody having low binding affinity to CD34-positive cells (e.g., CD34-positive bone marrow cells). The anti-hCDCP1 antibody of the present invention exhibits antitumor activity when converted into an ADC, and is therefore effective as an anticancer agent with few side effects. [Brief description of the drawings]

[0014] [Figure 1] A histogram of gel permeation chromatography of hCDCP1 extracellular domain purified protein is shown. [Diagram 2] The figure shows the results of SDS-PAGE of the purified hCDCP1 extracellular domain protein cleaved with plasmin. [Diagram 3] 1 shows the results of flow cytometry observation of the amount of hCDCP1 protein expression on the cell surface of Ba / F3 cells in which hCDCP1 is forcibly expressed. [Figure 4]This shows the cleavage of cell surface CDCP1 by trypsin treatment of Ba / F3 cells overexpressing hCDCP1. A shows the time course of the mean fluorescence intensity of PE in the cell population measured by flow cytometry. B shows the results of Western blotting for the cleaved hCDCP1 molecule. [Diagram 5] The results of screening for anti-hCDCP1 antibody-producing hybridomas by cell ELISA using PC3 cells and hCDCP1-deficient PC3 cells are shown in A, B, C, and D. The results of cell ELISA using culture supernatants of hybridomas prepared from each immunized animal in Experiment 1, Experiment 2, and Experiment 3 (see Examples) are shown in A, B, C, and D. [Figure 6A] The CDR mutation sites in the sequences used to produce mouse-human chimeric antibodies (mh12A041 series, mh14A025 series) are shown. [Figure 6B] The CDR mutation sites in the sequences used to produce mouse-human chimeric antibodies (mh14A043 series, mh14A063 series) are shown. [Figure 7] The reactivity of mouse-human chimeric antibodies to cells expressing hCDCP1 was observed by flow cytometry. [Figure 8] The reactivity of mouse-human chimeric antibodies to trypsin-treated Ba / F3 cells overexpressing hCDCP1 was observed by flow cytometry. [Figure 9] The reactivity of mouse-human chimeric antibodies to cynomolgus monkey CDCP1 was observed by flow cytometry. [Figure 10A] The reactivity of mouse-human chimeric antibodies to cancer cell lines (SK-MES-1, H358, MDA-MB-231, HCC1143, Capan-2, DLD-1, and OVCAR3) was observed by flow cytometry. [Figure 10B] The reactivity of mouse-human chimeric antibodies to cancer cell lines (SK-OV-3, TFK-1, PC3, and DU145) and primary cultured cells derived from normal tissues (HMEpC and NHEK) was observed by flow cytometry. [Figure 11]The figure shows the results of flow cytometry observation comparing the reactivity of mouse antibodies produced by hybridomas to healthy bone marrow CD34-positive cells at a comparative antibody concentration of 10 μg / mL. [Figure 12] The figures show the results of flow cytometry analysis comparing the reactivity of mouse-human chimeric antibodies against healthy bone marrow CD34-positive cells at a comparative antibody concentration of 10 μg / mL. [Figure 13] A-C show the results of flow cytometry observation of the reactivity of biotinylated mouse-human chimeric antibodies to healthy human bone marrow CD34-positive cells. D shows a comparison of the reactivity of the biotinylated anti-RS virus antibody used in A-C with purified IgG biotinylated protein derived from normal human serum at a comparative antibody concentration of 10 ng / mL to bone marrow CD34-positive cells. [Figure 14] 1 shows the results of observing the reactivity of humanized anti-hCDCP1 antibodies to hCDCP1-expressing Ba / F3 cells by flow cytometry. [Figure 15] 1 shows the results of observing the reactivity of a biotinylated humanized anti-hCDCP1 antibody and a biotinylated control antibody against healthy bone marrow CD34-positive cells at a comparison antibody concentration of 10 ng / mL, by flow cytometry. [Figure 16] The concentration-dependence of the reactivity of a biotinylated humanized antibody to bone marrow CD34-positive cells from healthy subjects was observed by flow cytometry. [Figure 17] The in vitro cytotoxicity of anti-hCDCP1 mouse-human chimeric antibody-drug conjugates with PBD against cancer cell lines and primary cultured cells derived from normal tissues is shown. [Figure 18] 1 shows the in vitro cytotoxic activity of PBD-linked humanized anti-hCDCP1 antibody-drug conjugates against cancer cell lines and primary cultured normal human epidermal keratinocytes. [Figure 19]A and B show the antitumor activity of PBD-conjugated anti-hCDCP1 mouse-human chimeric antibody-drug conjugate in a PC3 cell line xenograft model (scid mouse model). C and D show the antitumor activity of PBD-conjugated anti-hCDCP1 mouse-human chimeric antibody-drug conjugate in a PC3 cell line xenograft model (nude mouse model). [Figure 20] A shows the antitumor activity of a PBD-conjugated humanized anti-hCDCP1 antibody-drug conjugate in a PC3 cell line xenograft model (nude mouse model). B shows the antitumor activity of a PBD-conjugated humanized anti-hCDCP1 antibody-drug conjugate in a colon cancer cell line HCT116 cell line xenograft model. [Figure 21] 1 shows the antitumor activity of MMAE-linked humanized anti-hCDCP1 antibody-drug conjugate in an HCT116 cell line xenograft model. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] The first embodiment of the present invention is an antibody that binds to human CDCP1, + The antibody (hereinafter also referred to as "anti-hCDCP1 antibody of the present invention") or an antigen-binding fragment thereof is characterized by having low binding affinity to hCDCP1 cells. CD34 positive cells are cells that express the CD34 antigen on their cell surface. CD34 is a single-chain transmembrane phosphorylated glycoprotein with a molecular weight of approximately 110 kDa, and has two domains with different structures on the outside of the cell. CD34 is a surface antigen marker for various somatic stem cells, and is expressed on bone marrow-derived hematopoietic stem cells / vascular endothelial progenitor cells, skeletal muscle satellite cells, hair follicle stem cells, adipose tissue mesenchymal stem cells, etc. Examples of CD34-positive cells include hematopoietic stem cells that can differentiate into blood cells. Among hematopoietic stem cells, CD34 is most highly expressed in the most undifferentiated cells, and its expression decreases as the cells differentiate into each cell lineage.

[0016] Here, the "binding ability" of the hCDCP1 antibody of the present invention to CD34-positive cells refers to the ability of the anti-hCDCP1 antibody to bind to any site on CD34-positive cells (binding ability). In the first embodiment, the binding ability of the anti-hCDCP1 antibody to human CD34-positive cells is evaluated relatively by comparing it with the binding ability of non-specific human IgG to CD34-positive cells. Here, "non-specific human IgG" refers to IgG derived from a human that does not have specific reactivity to human CDCP1 protein, specifically a monoclonal antibody known to have no specific reactivity to human CDCP1, or more preferably a mixture of multiple monoclonal antibodies that do not have specific reactivity to CDCP1, or more preferably an IgG mixture purified and extracted from human biological serum by a technique such as affinity chromatography. As a specific evaluation method, when the anti-hCDCP1 antibody exhibits the same binding ability to human CD34-positive cells as nonspecific human IgG under the condition that the anti-hCDCP1 antibody and the nonspecific human IgG are at the same antibody concentration (hereinafter referred to as "comparative antibody concentration") (i.e., under the condition that the concentration of the anti-hCDCP1 antibody and the concentration of the nonspecific human IgG are the same), the binding ability of the anti-hCDCP1 antibody to human CD34-positive cells is evaluated as low. For example, when the anti-hCDCP1 antibody does not exhibit a statistically significantly higher binding ability to human CD34-positive cells compared to nonspecific human IgG, the binding ability may be evaluated as low. Also, when the anti-hCDCP1 antibody exhibits a higher binding ability compared to nonspecific human IgG, the binding ability of the anti-hCDCP1 antibody to human CD34-positive cells is evaluated as high. For example, when the anti-hCDCP1 antibody exhibits a statistically significantly higher binding ability to human CD34-positive cells compared to nonspecific human IgG, the binding ability may be evaluated as high.

[0017] From another perspective, the scope of the present invention includes antibodies that have binding affinity to CD34-positive cells that is lower or similar to that of a certain index antibody. For example, an antibody that has binding affinity to CD34-positive cells that is lower than that of a known antibody, CUB4 (clone name: h12A041VH1A / VL), is included.

[0018] Here, the "comparative antibody concentration" refers to, but is not limited to, a certain specific concentration point, for example, any concentration of 10 ng / mL or more, any concentration of 100 ng / mL or more, any concentration of 1 μg / mL or more, and more preferably any concentration of 10 μg / mL or more. The concentration of 10 ng / ml is the concentration at which CUB4 reacts sufficiently in the test system, i.e., the concentration at which CUB4 binds sufficiently to CD34-positive cells, and therefore this concentration is sufficient for comparative evaluation of binding activity. Furthermore, the binding affinity between the anti-hCDCP1 antibody or nonspecific human IgG and CD34-positive cells can be evaluated by, but is not limited to, for example, flow cytometry analysis, ELISA, RIA, surface plasmon resonance, or the like.

[0019] The anti-hCDCP1 antibody of the present invention can be prepared, for example, as follows: Antibodies that react with human CDCP1 expressed on cancer cells are screened from among antibodies prepared using the whole or a part of the extracellular domain of human CDCP1, or cells expressing the whole or a part of the extracellular domain on their surface as an antigen, and then an antibody that has low reactivity with the CD34-positive cell fraction of healthy bone marrow cells is selected from the screened antibody group, thereby preparing a desired antibody.

[0020] The term "antibody" as used herein is not particularly limited in terms of its preparation method or structure, and includes all "antibodies" that bind to a desired antigen with desired properties, such as monoclonal antibodies, polyclonal antibodies, or nanoantibodies. When the anti-hCDCP1 antibody of the present invention is a polyclonal antibody, it can be prepared, for example, by injecting a mixture of antigen and adjuvant into an animal to be immunized (such as, but not limited to, rabbit, goat, sheep, chicken, guinea pig, mouse, rat, or pig). Usually, the antigen and / or adjuvant are injected subcutaneously or intraperitoneally into the animal to be immunized multiple times. Adjuvants include, but are not limited to, complete Freund's and monophosphoryl lipid A synthetic-trehalose dicorynomycolate (MPL-TMD). After immunization with the antigen, the anti-hCDCP1 antibody can be purified from the serum derived from the immunized animal by a standard method (such as a method using Sepharose carrying Protein A).

[0021] Furthermore, when the anti-hCDCP1 antibody of the present invention is a monoclonal antibody, it can be prepared, for example, as follows. Note that, in this specification, the term "monoclonal" suggests the characteristics of an antibody obtained from a substantially homogeneous antibody population (an antibody population in which the amino acid sequences of the heavy and light chains constituting the antibody are identical), and does not mean that the antibody is prepared by a specific method (for example, the hybridoma method, etc.). Examples of methods for producing monoclonal antibodies include the hybridoma method (Kohler and Milstein, Nature 256 495 1975) or the recombinant method (U.S. Patent No. 4,816,567). Alternatively, the anti-hCDCP1 antibody of the present invention may be isolated from a phage antibody library (e.g., Clackson et al., Nature 352 624-628 1991; Marks et al., J. Mol. Biol. 222 581-597 1991, etc.) or a cell library (e.g., U.S. Patent No. 4214234; Seo et al., Nature Biotech., 23 731-735 2005, etc.). More specifically, when the hybridoma method is used for preparation, the preparation method includes, for example, the following four steps: (i) immunizing an animal with an antigen, (ii) recovering lymphocytes secreting (or potentially secreting) monoclonal antibodies, (iii) fusing the lymphocytes with immortalized cells, and (iv) selecting cells secreting the desired monoclonal antibodies. For example, mice, rats, guinea pigs, hamsters, rabbits, etc. can be selected as the immunized animals. After immunization, lymphocytes obtained from the host animal are fused with an immortalized cell line using a fusing agent such as polyethylene glycol or electrofusion to establish hybridoma cells. For example, rat or mouse myeloma cell lines are used as the fused cells. After cell fusion, the cells are grown in an appropriate medium containing one or more substrates that inhibit the growth or survival of unfused lymphocytes and the immortalized cell line. In a conventional technique, parent cells that lack the enzyme hypoxanthine-guanine phosphoribosyltransferase (HGPRT or HPRT) are used. In this case, hypoxanthine, aminopterin and thymidine are added to a medium (HAT medium) that inhibits the growth of HGPRT-deficient cells and allows the growth of hybridomas. From the hybridomas thus obtained, a hybridoma that produces a desired antibody is selected, and the desired monoclonal antibody can be obtained from the medium in which the selected hybridoma grows, according to a conventional method. The hybridomas thus prepared are cultured in vitro or in vivo in ascites of mice, rats, guinea pigs, hamsters, etc., and the desired antibody can be prepared from the culture supernatant or ascites.

[0022] Nanobodies are polypeptides consisting of the variable domain of the heavy chain of an antibody (VHH). Normally, human antibodies are composed of heavy and light chains, but camelids such as llamas, alpacas, and camels produce single-chain antibodies (heavy-chain antibodies) consisting of only heavy chains. Heavy-chain antibodies can recognize and bind to target antigens, just like regular antibodies consisting of heavy and light chains. The variable domain of a heavy-chain antibody is the smallest unit that has binding affinity to an antigen, and this variable domain fragment is called a "nanobody." Nanobodies are highly heat-resistant, digestion-resistant, and stable at room temperature, and can be easily prepared in large quantities by genetic engineering techniques.

[0023] Nanobodies can be produced, for example, as follows: A camelid animal is immunized with an antigen, and the presence or absence of the desired antibody is detected from the collected serum. cDNA is produced from RNA derived from peripheral blood lymphocytes of the immunized animal in which the desired antibody titer is detected. A DNA fragment encoding VHH is amplified from the obtained cDNA, and this is inserted into a phagemid to prepare a VHH phagemid library. The desired nanobodies can be produced from the produced VHH phagemid library through several rounds of screening.

[0024] The anti-hCDCP1 antibody of the present invention may be a recombinant antibody. Examples of recombinant antibodies include, but are not limited to, human antibodies and chimeric antibodies with human antibodies. Chimeric antibodies are, for example, antibodies in which variable and constant regions derived from different animal species are linked (for example, antibodies in which the variable region of a mouse-derived antibody is linked to a constant region derived from a human) (for example, Proc. Natl. Acad. Sci. USA 81, 6851-6855, (1984)), and can be easily constructed by recombinant technology.

[0025] A humanized antibody is an antibody that has a human-derived sequence in the framework region (FR) and a complementarity determining region (CDR) consisting of a sequence derived from another animal species (e.g., mouse, etc.). A humanized antibody can be produced by first transplanting the CDR from the variable region of an antibody derived from another animal species (here, mouse is used for explanation) into a human antibody variable region, reconstructing the heavy and light chain variable regions, and then linking these humanized reconstructed human antibody variable regions to a human antibody constant region. Methods for producing such humanized antibodies are well known in the art (e.g., Proc. Natl. Acad. Sci. USA, 86: 10029-10033 (1989), etc.).

[0026] An antigen-binding fragment of an antibody of the present invention is a partial region of an antibody of the present invention and refers to an antibody fragment that binds to human CDCP1. Examples of the fragment include Fab, Fab', F(ab')2, Fv (variable fragment of antibody), single-chain antibody (heavy chain, light chain, heavy chain variable region, light chain variable region, nanobody, etc.), scFv (single chain Fv), diabody (scFv dimer), dsFv (disulfide-stabilized Fv), and peptides comprising at least a portion of the CDR of an antibody of the present invention.

[0027] Fab is an antibody fragment having antigen-binding activity, in which about the N-terminal half of the heavy chain and the entire light chain are bound by a disulfide bond, among fragments obtained by treating an antibody molecule with the protease papain. Fab can be produced by treating an antibody molecule with papain to obtain a fragment, or, for example, by constructing an appropriate expression vector into which DNA encoding Fab is inserted, introducing this into an appropriate host cell (e.g., mammalian cells such as CHO cells, yeast cells, insect cells, etc.), and then expressing Fab in the cell.

[0028] F(ab')2 is an antibody fragment that has antigen-binding activity and is slightly larger than the fragments obtained by treating an antibody molecule with the protease pepsin, in which Fab is bonded via a disulfide bond in the hinge region. In addition to obtaining fragments by treating an antibody molecule with pepsin, F(ab')2 can also be produced by bonding Fab via a thioether bond or disulfide bond, and can also be produced by genetic engineering techniques in the same way as Fab.

[0029] Fab' is an antibody fragment having antigen-binding activity, which is obtained by cleaving the disulfide bond in the hinge region of the above-mentioned F(ab')2. Fab' can also be prepared by genetic engineering techniques, like Fab.

[0030] An scFv is an antibody fragment having antigen-binding activity, which is a VH-linker-VL or VL-linker-VH polypeptide in which one heavy chain variable region (VH) and one light chain variable region (VL) are linked using an appropriate peptide linker. The scFv can be produced by genetic engineering techniques using cDNAs encoding the heavy and light chain variable regions of an antibody.

[0031] A diabody is an antibody fragment in which scFv is dimerized and has bivalent antigen-binding activity. The bivalent antigen-binding activity may be the same antigen-binding activity or one of the antigens may be different. A diabody can be produced by genetic engineering techniques by obtaining cDNAs encoding the heavy and light chain variable regions of an antibody and constructing cDNAs encoding scFvs in which the heavy and light chain variable regions are linked by a peptide linker.

[0032] dsFv refers to a polypeptide in which one amino acid residue in each of the heavy chain variable region and the light chain variable region is replaced with a cysteine ​​residue, and the cysteine ​​residues are linked via a disulfide bond. The amino acid residue to be replaced with a cysteine ​​residue can be selected based on the prediction of the three-dimensional structure of the antibody. dsFv can be produced by genetic engineering techniques by obtaining cDNAs encoding the heavy chain variable region and the light chain variable region of an antibody and constructing DNA encoding dsFv.

[0033] The peptide containing CDR is configured to include at least one region of the CDR (CDR1 to 3) of the heavy chain or the light chain. The peptide containing multiple CDRs can be linked directly or via a suitable peptide linker. The peptide containing CDR is prepared by constructing a DNA encoding the CDR of the heavy chain or the light chain of an antibody and inserting it into an expression vector. There is no particular limitation on the type of vector, and it may be appropriately selected depending on the type of host cell to be introduced therein. These can be introduced into a suitable host cell (e.g., mammalian cells such as CHO cells, yeast cells, insect cells, etc.) to be expressed as an antibody, and produced. The peptide containing CDR can also be produced by chemical synthesis methods such as the Fmoc method (fluorenylmethyloxycarbonyl method) and the tBoc method (t-butyloxycarbonyl method).

[0034] A human antibody (fully human antibody) generally has the same structure as a human antibody in the hypervariable region, which is the antigen-binding site of the V region, other parts of the V region, and the constant region. A person skilled in the art can easily prepare a human antibody using known techniques. A human antibody can be obtained, for example, by a method using a human antibody-producing mouse having a human chromosome fragment containing the H-chain and L-chain genes of a human antibody (e.g., Tomizuka et al., Proc. Natl. Acad. Sci. USA, (2000) 97, 722-727, etc.), or by a method of obtaining a phage-display-derived human antibody selected from a human antibody library (e.g., Siriwardena et al., Opthalmology, (2002) 109 (3), 427-431, etc.).

[0035] A multispecific antibody can be constructed using an antigen-binding fragment of the antibody of the present invention. Multispecificity means having binding specificity for two or more antigens, for example, a monoclonal antibody or a protein form containing an antigen-binding fragment having binding specificity for two or more antigens. This can be performed by a person skilled in the art using known techniques. As a method for constructing multispecificity, several techniques have been developed that can be classified into a technique for constructing asymmetric IgG in which two different types of antibody heavy chain molecules are subjected to protein engineering so as to form a heterodimer, a technique for linking antigen-binding fragments obtained from an antibody to each other or to another antibody molecule, and the like. For specific examples of construction methods, the following literature can be referred to, for example. Kontermann, RE, & Brinkmann, U. (2015). Bispecific antibodies. Drug Discovery Today, 20(7), 838-847.

[0036] Examples of the anti-hCDCP1 antibody and antigen-binding fragment thereof of the present invention include an antibody and antigen-binding fragment thereof characterized in that the amino acid sequences of CDRs (complementarity determining regions) 1 to 3 satisfy any of the following (A) to (S): (A) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 25; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:26; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 27; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:29; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 30; and It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:31.

[0037] (B) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 33; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 34; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 35; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:37; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 38, and It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:39.

[0038] (C) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 41; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 42; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 43; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 45; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 46, and It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:47.

[0039] (D) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 49; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:50; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:51; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:53; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:54, and It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:55.

[0040] (E) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 57; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:58; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:59; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 61; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 62, and It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:63.

[0041] (F) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 65; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:66; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:67; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:69; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 70; It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:71.

[0042] (G) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 73; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 74; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 75; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 77; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 78, and It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:79.

[0043] (H) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 81; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 82; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 83; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 85; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 86; and It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:87.

[0044] (I) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 89; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 90; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 91; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:93; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 94, and It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:95.

[0045] (J) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 97; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:98; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 99; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 101; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 102; It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:103.

[0046] (K) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 105; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 106; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 107; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 109; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 110; and It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:111.

[0047] (L) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 49; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:50; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:51; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:53; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:54, and It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:55.

[0048] (M) An antibody that satisfies (L) above has a light chain variable region in which the amino acids at positions 28 and 29 of the light chain and the amino acid at position 102 of the heavy chain according to the Kabat definition are replaced with other amino acids. Examples of other amino acids include, but are not limited to, serine, threonine, alanine, glycine, leucine, isoleucine, valine, phenylalanine, arginine, tryptophan, histidine, methionine, glutamine, glutamic acid, lysine, and tyrosine for position 28 of the light chain, with serine being preferred. Examples of other amino acids for position 29 of the light chain include alanine, aspartic acid, glutamic acid, methionine, arginine, leucine, isoleucine, valine, and lysine, with alanine being preferred. For the 102nd position of the heavy chain, examples include alanine, glutamic acid, serine, threonine, glycine, leucine, isoleucine, valine, phenylalanine, tryptophan, methionine, lysine, histidine, and glutamine, with serine being preferred.

[0049] (N) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 65; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:66; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:67; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:69; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 70; and It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:71.

[0050] (O) The antibody satisfies (N) above and has a heavy chain variable region in which the amino acids at positions 54 and 55 of the heavy chain according to the Kabat definition are replaced with other amino acids. Examples of other amino acids include, but are not limited to, serine, threonine, alanine, glycine, leucine, isoleucine, valine, phenylalanine, arginine, tryptophan, histidine, methionine, glutamine, glutamic acid, lysine, and tyrosine at position 54 of the heavy chain, with serine, phenylalanine, arginine, threonine, and tryptophan being preferred. Examples of other amino acids at position 55 of the heavy chain include alanine, aspartic acid, glutamic acid, methionine, arginine, leucine, isoleucine, valine, and lysine, with alanine, aspartic acid, glutamic acid, leucine, and methionine being preferred.

[0051] (P) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 73; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 74; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 75; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 77; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 78, and It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:79.

[0052] (Q) The antibody satisfies (P) above and has heavy and light chain variable regions in which the amino acids at positions 54 and 55 of the heavy chain and the amino acid at position 33 of the light chain according to the Kabat definition are replaced with other amino acids. Examples of other amino acids include, but are not limited to, serine, threonine, alanine, glycine, leucine, isoleucine, valine, phenylalanine, arginine, tryptophan, histidine, methionine, glutamine, glutamic acid, lysine, and tyrosine for position 54 of the heavy chain, with serine being preferred. For position 55 of the heavy chain, alanine, aspartic acid, glutamic acid, methionine, arginine, leucine, isoleucine, valine, and lysine are preferred, with alanine being preferred. Examples of the 33rd amino acid in the light chain include serine, threonine, alanine, glycine, leucine, isoleucine, valine, phenylalanine, arginine, tryptophan, asparagine, aspartic acid, glutamine, glutamic acid, tyrosine, lysine, and histidine, with leucine being preferred.

[0053] (R) heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 89; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 90; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 91; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:93; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 94, and It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:95.

[0054] (S) The antibody satisfies (R) above and has heavy and light chain variable regions in which positions 52a and 53 of the heavy chain and position 33 of the light chain according to the Kabat definition are replaced with other amino acids. Examples of other amino acids include, but are not limited to, serine, threonine, alanine, glycine, leucine, isoleucine, valine, phenylalanine, arginine, tryptophan, histidine, methionine, glutamine, glutamic acid, lysine, and tyrosine for position 52a of the heavy chain, with serine being preferred. For position 53 of the heavy chain, alanine, aspartic acid, glutamic acid, methionine, arginine, leucine, isoleucine, valine, and lysine are preferred, with alanine being preferred. Examples of the 33rd amino acid in the light chain include serine, threonine, alanine, glycine, leucine, isoleucine, valine, phenylalanine, arginine, tryptophan, asparagine, aspartic acid, glutamine, glutamic acid, tyrosine, lysine, and histidine, with leucine being preferred.

[0055] Further, examples of the anti-hCDCP1 antibody and antigen-binding fragment thereof of the present invention include an antibody or antigen-binding fragment thereof, characterized in that the amino acid sequences of CDR1 to 3 of the heavy chain variable region satisfy any one of the following (T) to (U), and the amino acid sequences of CDR1 to 3 of the light chain variable region satisfy any one of the following (t) to (v). (T) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 49; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:50; and Having a heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 113; (U) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 183; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:50; and Having a heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 113; (t) a light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 117; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:54, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:55; (u) a light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 186; a light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:54, and a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:55. (v) a light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 188; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:54, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:55;

[0056] Furthermore, the anti-hCDCP1 antibodies and antigen-binding fragments thereof of the present invention include SEQ ID NO:24, SEQ ID NO:32, SEQ ID NO:40, SEQ ID NO:48, SEQ ID NO:56, SEQ ID NO:64, SEQ ID NO:72, SEQ ID NO:80, SEQ ID NO:88, SEQ ID NO:96, SEQ ID NO:104, SEQ ID NO:112, SEQ ID NO:114, SEQ ID NO:124, SEQ ID NO:126, SEQ ID NO:128, SEQ ID NO:130, SEQ ID NO:132, SEQ ID NO:134, SEQ ID NO:136, SEQ ID NO:138, SEQ ID NO:140, SEQ ID NO:142, SEQ ID NO:144, 146, SEQ ID NO:150, SEQ ID NO:152, SEQ ID NO:169, SEQ ID NO:170, SEQ ID NO:172, SEQ ID NO:173, SEQ ID NO:177, SEQ ID NO:178, SEQ ID NO:179, SEQ ID NO:182 or SEQ ID NO:184, and any of the heavy chain variable regions comprising the amino acid sequences represented by SEQ ID NO:28, SEQ ID NO:36, SEQ ID NO:44, SEQ ID NO:52, SEQ ID NO:60, SEQ ID NO:68, SEQ ID NO:76, SEQ ID NO:84, SEQ ID NO:92, SEQ ID NO:100, SEQ ID NO:108, SEQ ID NO:116, SEQ ID NO:118, SEQ ID NO: and antibodies having any of the light chain variable regions comprising the amino acid sequences represented by SEQ ID NO:120, SEQ ID NO:122, SEQ ID NO:148, SEQ ID NO:154, SEQ ID NO:155, SEQ ID NO:171, SEQ ID NO:174, SEQ ID NO:175, SEQ ID NO:176, SEQ ID NO:180, SEQ ID NO:181, SEQ ID NO:185 or SEQ ID NO:187, and antibodies having a light chain variable region comprising the amino acid sequence ... Examples of the antibody include antibodies having an amino acid sequence with an amino acid identity of 4% or more, about 85% or more, about 86% or more, about 87% or more, about 88% or more, about 89% or more, more preferably about 90% or more, about 91% or more, about 92% or more, about 93% or more, about 94% or more, about 95% or more, about 96% or more, about 97% or more, about 98% or more, and most preferably about 99% or more, which antibodies bind to human CDCP1 and have low binding affinity to human CD34-positive cells, as well as antigen-binding fragments thereof.

[0057] The second embodiment of the present invention is an antibody that binds to human CDCP1 and has low binding affinity to human CD34-positive cells, and is an antibody (hereinafter also referred to as "competitive antibody of the present invention") that competitively inhibits the binding of the antibody according to the first embodiment (i.e., the anti-hCDCP1 antibody of the present invention) to human CDCP1, or an antigen-binding fragment thereof. The competitive antibody of the present invention can be prepared and obtained by a competitive experiment well known to those skilled in the art. Specifically, when the binding of a first anti-hCDCP1 antibody (an antibody according to the embodiment) to human CDCP1 is competitively inhibited by a second anti-hCDCP1 antibody, it is determined that the first anti-hCDCP1 antibody and the second anti-hCDCP1 antibody are substantially identical or bind to antigen sites in close proximity. When the second anti-hCDCP1 antibody has low binding affinity to CD34-positive cells, the second anti-hCDCP1 antibody is a competitive antibody of the present invention. As a method for such a competitive experiment, for example, a method using a Fab fragment or the like is usually performed in the technical field. See, for example, WO95 / 11317, WO94 / 07922, WO2003 / 064473, WO2008 / 118356 and WO2004 / 046733.

[0058] A third embodiment of the present invention is the antibody according to the first embodiment or the antibody according to the second embodiment, or an antigen-binding fragment thereof, to which a substance having antitumor activity is bound. Targeted cancer therapy can be performed by binding a substance having antitumor activity, such as a drug, to an antibody (such a complex is hereinafter referred to as the "antibody-drug complex of the present invention.") In this case, the substance having antitumor activity includes, but is not limited to, cytotoxic drugs such as anticancer drugs, radioisotopes, and substances that indirectly induce antitumor activity by manipulating the immune system.

[0059] In the third embodiment, a drug showing antitumor activity can be used, and such a complex is called an antibody-drug complex. Examples of drugs showing antitumor activity that can be used include, but are not limited to, tubulin inhibitors and microtubule polymerization inhibitors such as Auristatins (MMAE, MMAF, etc.), Maytansines (DM1, DM4, etc.), Tubulysins, cryptophycins, and rhizoxins, antibiotics such as Calicheamicins, Doxorubicin, and anthracyclines, DNA synthesis inhibitors such as Duocarmycins, PBDs (Benzodiazepines), and IGNs (indolinobenzodiazepines), topoisomerase I inhibitors such as Canptothecin analogs (SN-38, DXd, etc.), RNA polymerase II inhibitors such as Amanitins, and RNA spliceosome inhibitors such as spliceostatins and thailanstatins. In addition, as a drug exhibiting antitumor activity, a compound that is excited by light energy and exhibits toxicity can also be used. Such an antibody-drug conjugate can be administered into the body to bind to tumor cells, and then can be used in a treatment method called photoimmunotherapy (PIT), in which the tumor cells are killed by applying light energy such as near-infrared rays from outside the body. The anti-hCDCP1 antibody of the present invention may also be used as an antibody for photoimmunotherapy. Known compounds that can be used include, but are not limited to, IR700.

[0060] In addition, radioimmunotherapy (RIT) is known in which a radioisotope is bound to an antibody and the radiation emitted by the isotope is used to kill cancer cells. The anti-hCDCP1 antibody of the present invention can also be used as an antibody for radioimmunotherapy. Examples of radioisotopes used in the third embodiment of the present invention include: 131 I, 90 β-ray nuclides such as Y, 213 Bi, 211 At, 225 Ac,223 Ra, 212 Alpha-emitting nuclides such as Pb are known, but are not limited to these.

[0061] In the third embodiment of the present invention, a substance that exerts antitumor activity directly or indirectly by manipulating the physiological action of the tumor itself or tissues other than the tumor, such as the immune system, can be used. Taking the immune system as an example, the immune system elements to be manipulated can be lymphocyte cells such as T cells, B cells, and NK cells, myeloid cells such as monocytes, macrophages, dendritic cells, and granulocytes, or cells other than immune system cells that secrete and present substances that affect these immune system cells. Substances that manipulate these cells can include, but are not limited to, cancer vaccine peptides, cytokines (interleukins, interferons, colony-stimulating factors (CSFs), etc.), hormones, and growth factors (TGF family, FGF family, IGF family, thrombopoietin, erythropoietin, etc.).

[0062] Many methods of chemical modification for binding an antitumor substance to an antibody have been known so far, including the following: covalent bonding to the side chain of a lysine residue, covalent bonding to the side chain of a cysteine ​​residue, site-specific chemical modification of the side chain by introducing a non-natural amino acid into the antibody peptide chain, modification of a specific amino acid sequence in an antibody or modified sugar chain by a specific enzymatic reaction, and modification using an enzyme that performs peptide linkage. In addition, drugs are generally chemically modified to bind to proteins and used as a linker for protein binding. There are many types of chemical linkers, and the pharmacological action of the antibody-drug complex in the body varies greatly depending on the properties of the linker. For example, hydrazone linkers, valine-citrulline linkers, SS bond linkers, pyrophosphate linkers, etc. can be cleaved by enzymes in the body, and the drug can be separated from the antibody, making it possible to prepare an antibody-drug complex that exhibits high antitumor effects. On the other hand, it is also common to use chemical structures that cannot be cleaved in the body as such chemical linkers. An overview of the above methods is described, for example, in the following literature: Tsuchikama and An, (2018). Antibody-drug conjugates: recent advances in conjugation and linker chemistries. Protein and Cell, 9(1), 33-46.

[0063] Drug-antibody ratio (DAR) is a number that indicates how many drug molecules are bound to one antibody molecule in an antibody-drug conjugate. DAR varies depending on the method of chemically binding the drug to the antibody, and is generally between 1 and 8, but it is possible to create an antibody-drug conjugate with a DAR of 9 or more depending on the type of chemical binding.

[0064] A fourth embodiment of the present invention is a pharmaceutical composition for preventing or treating cancer, which comprises the antibody-drug conjugate or antigen-binding fragment thereof of the present invention according to the third embodiment (hereinafter also referred to as the "pharmaceutical composition of the present invention"). The pharmaceutical composition of the present invention may be administered in the form of a pharmaceutical composition containing one or more formulation additives in addition to the antibody-drug conjugate or antigen-binding fragment thereof of the present invention as an active ingredient. The pharmaceutical composition according to this embodiment may also contain other known drugs. When a therapeutic antibody that can be administered to humans is produced using the antibody-drug conjugate or its antigen-binding fragment of the present invention, it is known that chemical modification often occurs after purification of the antibody during the production process. It is recognized in the art that dealing with such chemical modification is important in ensuring the quality and safety of the purified antibody. Specifically, if there is an amino acid sequence motif in the sequence containing the CDR region of the antibody that is susceptible to modifications such as N-glycosylation, protein cleavage, deamidation, racemization, and oxidation, it is expected that the quality of the purified antibody will be reduced. As amino acid sequence motifs that are susceptible to modification, for example, motifs containing asparagine residues and aspartic acid residues are known (e.g., Sydow et al., (2014). Structure-based prediction of asparagine and aspartate degradation sites in antibody variable regions. PLoS ONE, 9(6), etc.), and demonstrating that it is possible to modify these amino acid motifs while maintaining the activity of the antibody will further enhance the usefulness of the antibody.

[0065] The pharmaceutical composition of the present invention may be in the form of an oral or parenteral dosage form, and examples thereof include, but are not limited to, tablets, capsules, granules, powders, syrups, suspensions, suppositories, ointments, creams, gels, patches, inhalants, and injections. These preparations are prepared according to conventional methods. Liquid preparations may be dissolved or suspended in water or other suitable solvents when used. Tablets and granules may be coated by known methods. Injections are prepared by dissolving the antibody or functional fragment thereof in water, but may also be dissolved in physiological saline or glucose solution as necessary, and buffers and preservatives may be added.

[0066] The type of formulation additive used in the production of the pharmaceutical composition of the present invention, the ratio of the formulation additive to the active ingredient, or the production method of the pharmaceutical composition can be appropriately selected by a person skilled in the art depending on the form of the pharmaceutical composition. The formulation additive can be an inorganic or organic substance, or a solid or liquid substance, and can generally be blended in an amount of, for example, 0.1% by weight to 99.9% by weight, 1% by weight to 95.0% by weight, or 1% by weight to 90.0% by weight relative to the weight of the active ingredient. Specific examples of additives for formulations include lactose, glucose, mannitol, dextrin, cyclodextrin, starch, sucrose, magnesium aluminometasilicate, synthetic aluminum silicate, sodium carboxymethylcellulose, hydroxypropyl starch, calcium carboxymethylcellulose, ion exchange resins, methylcellulose, gelatin, gum arabic, hydroxypropylcellulose, hydroxypropylmethylcellulose, polyvinylpyrrolidone, polyvinyl alcohol, light anhydrous silicic acid, magnesium stearate, talc, tragacanth, bentonite, veegum, titanium oxide, sorbitan fatty acid esters, sodium lauryl sulfate, glycerin, fatty acid glycerin esters, purified lanolin, glycerogelatin, polysorbate, macrogol, vegetable oils, wax, liquid paraffin, white petrolatum, fluorocarbons, nonionic surfactants, propylene glycol, and water.

[0067] To prepare a solid preparation for oral administration, the active ingredient is mixed with excipients such as lactose, starch, crystalline cellulose, calcium lactate, or anhydrous silicic acid to prepare a powder, or, if necessary, a binder such as sucrose, hydroxypropylcellulose, or polyvinylpyrrolidone, or a disintegrant such as carboxymethylcellulose or calcium carboxymethylcellulose is added and wet or dry granulated to prepare a granule. To prepare tablets, these powders and granules may be tableted as is, or with the addition of a lubricant such as magnesium stearate or talc. These granules or tablets may be coated with an enteric base such as hydroxypropylmethylcellulose phthalate or methacrylic acid-methyl methacrylate polymer to prepare an enteric coated preparation, or coated with ethylcellulose, carnauba wax, or hardened oil to prepare a sustained release preparation. To prepare capsules, the powders or granules may be filled into hard capsules, or the active ingredient may be dissolved in glycerin, polyethylene glycol, sesame oil, olive oil, or the like, and then coated with gelatin to prepare a soft capsule.

[0068] To prepare an injection, the active ingredient may be dissolved in distilled water for injection, if necessary, together with a pH adjuster such as hydrochloric acid, sodium hydroxide, lactose, lactic acid, sodium, sodium monohydrogen phosphate or sodium dihydrogen phosphate, and an isotonic agent such as sodium chloride or glucose, and then sterile filtered and filled into an ampoule, or the active ingredient may be added with mannitol, dextrin, cyclodextrin, gelatin, etc., and then vacuum lyophilized to prepare an injection that is dissolved when used. Alternatively, the active ingredient may be added with lecithin, polysorbate 80, polyoxyethylene hydrogenated castor oil, etc., and emulsified in water to prepare an emulsion for injection.

[0069] To prepare a rectal preparation, the active ingredient may be dissolved by wetting together with a suppository base such as cacao butter, tri-, di-, or monoglyceride of fatty acid, or polyethylene glycol, and then poured into a mold and cooled; alternatively, the active ingredient may be dissolved in polyethylene glycol or soybean oil, etc., and then coated with a gelatin film or the like.

[0070] The dosage and frequency of administration of the pharmaceutical composition of the present invention are not particularly limited, and can be appropriately selected at the discretion of a physician or pharmacist depending on conditions such as the purpose of preventing and / or treating the worsening or progression of the disease to be treated, the type of disease, and the weight and age of the patient. In general, the daily oral dose for adults is about 0.01 to 1,000 mg (weight of active ingredient), and can be administered once a day or in divided doses, or every few days. When used as an injection, it is desirable to administer 0.001 to 100 mg (weight of active ingredient) per day to adults continuously or intermittently.

[0071] Another embodiment of the pharmaceutical composition of the present invention is a cytotoxic cell such as a T cell expressing the antibody of the present invention or an antigen-binding fragment thereof on the cell surface. Chimeric antigen receptor-expressing T cell (CAR-T) therapy is a therapy in which a fusion gene (chimeric antigen receptor gene) of an antigen-binding site of an antibody and a part of a T cell receptor is expressed in a T cell, which is then transferred into the body of a cancer patient, and the transferred T cell specifically attacks cancer cells to bring about antitumor activity. A gene encoding the antibody of the present invention or an antigen-binding fragment thereof can be used as a component of the chimeric antigen receptor gene to construct an expressing T cell, thereby constructing a CAR-T therapy that specifically attacks tumors expressing human CDCP1 molecules.

[0072] The pharmaceutical composition of the present invention can attack and kill any cancer cell expressing hCDCP1 on the cell surface. Therefore, the cancer to be treated by the pharmaceutical composition of the present invention may be any cancer, and is not particularly limited. Typical examples of cancers include hepatocellular carcinoma, cholangiocarcinoma, renal cell carcinoma, squamous cell carcinoma, basal cell carcinoma, transitional cell carcinoma, adenocarcinoma, malignant gastrinoma, melanoma, fibrosarcoma, myxosarcoma, liposarcoma, leiomyosarcoma, rhabdomyosarcoma, malignant teratoma, angiosarcoma, Kaposi's sarcoma, osteosarcoma, chondrosarcoma, lymphangiosarcoma, malignant meningioma, non-Hodgkin's lymphoma, Hodgkin's lymphoma, leukemia, and brain tumors, malignant tumors such as epithelial cell-derived neoplasms (epithelial carcinomas), basal cell carcinoma, adenocarcinoma, lip cancer, oral cavity cancer, esophageal cancer, small intestine cancer, and gastrointestinal cancer such as stomach cancer, colon cancer, rectal cancer, bladder cancer, pancreatic cancer, ovarian cancer, cervical cancer, lung cancer, breast cancer, skin cancer, and prostate cancer, as well as other known cancers that affect epithelial cells, mesenchymal cells, or blood cells throughout the body.

[0073] A fifth embodiment of the present invention is a method for preventing and / or treating cancer (hereinafter also referred to as "the preventive or therapeutic method of the present invention"), which comprises administering to a patient the pharmaceutical composition of the present invention. Here, "treatment" refers to preventing or alleviating the progression or worsening of the disease in a patient who already has cancer, and is intended to prevent or alleviate the progression or worsening of cancer. "Prevention" means preventing the onset of cancer in those who are at risk of developing cancer that requires treatment, and is a treatment aimed at preventing the onset of cancer in advance. Furthermore, treatment to prevent recurrence of cancer after treatment is also included in "prevention." Furthermore, the subjects of treatment and prevention are not limited to humans, but may also be mammals other than humans, such as mice, rats, dogs, and cats, as well as livestock such as cows, horses, and sheep, and primates such as monkeys, chimpanzees, and gorillas, with humans being particularly preferred.

[0074] In another embodiment of the present invention, a method for diagnosing cancer using the antibody of the present invention is provided. The antibody of the present invention can specifically bind to the human CDCP1 molecule, and by labeling the antibody of the present invention with a fluorescent substance, a radioisotope, an enzyme, or the like, tumors and cancer cells expressing the human CDCP1 molecule, and human CDCP1 molecules or fragments thereof present in the blood can be detected. Examples of detection methods include immunostaining, flow cytometry, Western blotting, ELISA, RIA, CLIA, and PET. Cancer cells in the body can be directly detected, or the expression level of human CDCP1 in a patient sample can be observed, thereby enabling the presence or absence of a primary tumor, the presence or absence of a metastatic tumor, and the like to be evaluated. Furthermore, by diagnosing the expression level of human CDCP1 in a cancer case in advance by a method using the antibody of the present invention, it is possible to predict the therapeutic effect of administering a pharmaceutical composition using the antibody of the present invention. The cancer to be diagnosed may be any type and is not particularly limited. Typical examples of cancers include hepatocellular carcinoma, cholangiocarcinoma, renal cell carcinoma, squamous cell carcinoma, basal cell carcinoma, transitional cell carcinoma, adenocarcinoma, malignant gastrinoma, melanoma, fibrosarcoma, myxosarcoma, liposarcoma, leiomyosarcoma, rhabdomyosarcoma, malignant teratoma, angiosarcoma, Kaposi's sarcoma, osteosarcoma, chondrosarcoma, lymphangiosarcoma, malignant meningioma, non-Hodgkin's lymphoma, Hodgkin's lymphoma, leukemia, and brain tumors, malignant tumors such as epithelial cell-derived neoplasms (epithelial carcinomas), basal cell carcinoma, adenocarcinoma, lip cancer, oral cavity cancer, esophageal cancer, small intestine cancer, and gastrointestinal cancer such as stomach cancer, colon cancer, rectal cancer, bladder cancer, pancreatic cancer, ovarian cancer, cervical cancer, lung cancer, breast cancer, skin cancer, and prostate cancer, as well as other known cancers that affect epithelial, mesenchymal, or blood cells throughout the body.

[0075] The disclosures of all documents cited herein are incorporated by reference in their entirety. Also, throughout this specification, where the words "a," "an," and "the" are used, they shall be understood to refer to the singular only, unless the context clearly indicates otherwise. It is intended to include multiple things. The present invention will be further described below with reference to examples. However, the examples are merely illustrative of embodiments of the present invention and are not intended to limit the scope of the present invention. EXAMPLES

[0076] 1. Expression and purification of recombinant hCDCP1 extracellular domain protein The amino acid sequence of the hCDCP1 protein is registered as isoform 1 under the UniProt registration number Q9H5V8 (SEQ ID NO: 1). A Campath secretion signal was linked to the N-terminus of the 30th to 666th amino acids of this sequence, and a 6xHis tag and a FLAG tag were inserted into the C-terminus to design a sequence (SEQ ID NO: 2). This amino acid sequence was converted into a base sequence based on the mammalian codon table, and a DNA sequence (SEQ ID NO: 3) was synthesized by gene synthesis (GENEWIZ) with a Kozak translation initiation sequence inserted into the 5'-terminus and a translation termination codon inserted into the 3'-terminus. The synthesized DNA was inserted between the restriction enzyme KpnI recognition sequence and PmeI recognition sequence of pEF1 / V5-His A (Thermo Scientific). The expression vector constructed as described above was used below. The expression vector plasmid was transiently transfected into FreeStyle293 cells (Thermo Scientific) using the polyethyleneimine method, and then cultured for 5 days in a 37°C, 5% CO2 incubator. The culture supernatant was collected, filtered through a 0.22 μm filter, and then bound to a HisTrap excel column (GE Healthcare). Next, the protein was eluted with a concentration gradient of 20 mM to 500 mM imidazole in 20 mM phosphate buffer / 300 mM NaCl / pH 7.5 buffer. The eluted fraction was collected in 1 mL portions, and the fractions that showed a band of approximately 100 kDa by SDS-PAGE were collected. This eluted fraction was added with 9 volumes of 20 mM Tris-HCl pH 8.0 and bound to a HiTrap Q (GE Healthcare) column. Elution was performed with a NaCl concentration gradient of 0 to 1 M in 20 mM Tris-HCl pH 8.0 buffer. The eluted fraction was collected in 2.5 mL aliquots, and fractions in which bands were confirmed by SDS-PAGE were collected. Furthermore, the eluted fraction from this ion exchange column was concentrated to approximately 1.7 mL and fractionated using Superdex 200pg 16 / 60 (GE Healthcare) with D-PBS (Nacalai Tesque) as the mobile phase. The elution curve in this purification experiment using gel permeation chromatography is shown in Figure 1. The eluted fraction was collected every 1.0 mL, and the fractions in which bands were confirmed by SDS-PAGE were collected. The collected protein was used as purified recombinant hCDCP1 extracellular domain protein (hereinafter referred to as hCDCP1-ECD protein).

[0077] 2. Plasmin treatment and purification of full-length ECD protein A 25-fold amount of human plasma-derived plasmin protein (Sigma-Aldrich) was added to the hCDCP1-ECD protein and reacted at 4°C for 18 hours. SDS-PAGE confirmed that the hCDCP1-ECD protein was almost completely cleaved by this plasmin treatment. The protein sample was purified using cOmplete His-Tag purification resin (Sigma-Aldrich) and anti-FLAG M2 antibody affinity gel (Sigma-Aldrich), and the results of observing the sample by SDS-PAGE under reducing and non-reducing conditions are shown in Figure 2. The eluted protein was thought to contain two molecular species with different molecular weights. These two molecular species were thought to be two peptide fragments resulting from the cleavage of hCDCP1-ECD, and it was thought that these two fragments were purified while maintaining a bound state in some manner even after cleavage.

[0078] 3. Construction of Ba / F3 cells stably expressing full-length hCDCP1 A sequence was designed in which a Campath secretion signal and a c-Myc tag were linked to the N-terminus of the 30th to 836th amino acids of the UniProt number Q9H5V8 isoform 1 (SEQ ID NO: 4). This amino acid sequence was converted to a base sequence based on the mammalian codon table, and DNA containing this sequence and with a Kozak sequence at the 5' end and a translation termination codon at the 3' end inserted was synthesized by gene synthesis (GENEWIZ; SEQ ID NO: 5). The synthesized DNA was inserted between the KpnI / PmeI recognition sequences of pEF1 / V5-His A (Thermo Scientific) to create the expression vector GS01. Plasmid GS01 was linearized with the restriction enzyme PmeI and diluted to 2 × 10 6 2μg of the gene was transfected into mouse B cell-derived cell line Ba / F3 cells using Nucleofector 2b (Lonza). After gene transfection, the cells were seeded on a 96-well plate and G418 (Nacalai Tesque) was added at a final concentration of 1μg / mL. After 6 days, 24 colonies that showed G418-resistant growth were collected from the wells, and the expression of hCDCP1 was confirmed by flow cytometry using an anti-hCDCP1 antibody (R&D systems; catalog number MAB26662), and the cells derived from the confirmed colonies were subjected to limiting dilution. The proliferated cells were collected, and the expression of hCDCP1 was confirmed by flow cytometry using an anti-hCDCP1 antibody (R&D systems), and finally a cell line derived from a single clone was selected. The results of staining the cells with the same anti-hCDCP1 antibody, anti-Myc tag antibody (Wako Pure Chemical Industries), and PE-labeled anti-mouse IgG secondary antibody are shown in Figure 3. All clones were positive for anti-hCDCP1 and anti-Myc tag antibodies, confirming that the product of the introduced hCDCP1 gene with a Myc tag at the N-terminus was expressed on the cell surface. These cells were used as the Ba / F3-hCDCP1 cell line in the following experiments.

[0079] 4. Generation of ΔN-type molecules by trypsin treatment of Ba / F3-hCDCP1 The hCDCP1-expressing Ba / F3 cell line (hereinafter referred to as Ba / F3-hCDCP1) prepared in the previous section and the original Ba / F3 line were used as a negative control. The cells were washed twice with cold serum-free RPMI 1640 medium (Sigma-Aldrich) and then diluted with 1.25 × 10 7 The cells were suspended in RPMI1640 medium at a concentration of 10 ... Figure 4A shows a graph of the change in the average fluorescence intensity of the cell population based on the fluorescence intensity of PE measured by flow cytometry. The staining intensity of the anti-Myc tag antibody decreased over time only when trypsin / EDTA was added, indicating that the N-terminus of the hCDCP1 molecule on Ba / F3-hCDCP1 was digested by trypsin. On the other hand, the staining intensity of the anti-hCDCP1 antibody did not change, indicating that the number of hCDCP1 molecules recognized by the anti-hCDCP1 antibody used in this experiment was not reduced by trypsin treatment.

[0080] Next, the molecular weight of the truncated hCDCP1 molecule produced by trypsin treatment was observed by Western blotting. 6Each cell was treated with 0.25% trypsin / EDTA diluted 5-fold in RPMI1640 medium or PBS containing 50 nM or 500 nM plasmin (Sigma-Aldrich) at 37°C for 5, 10, or 30 minutes. These cells were collected and washed with HBSS, then lysed in Tris-buffered saline containing 1% TritonX and protease inhibitor cocktail (Thermo Fisher Scientific), protein was quantified using a Micro BCA kit (Thermo Fisher Scientific), and 20 μg of the cells were electrophoresed by SDS-PAGE. At the same time, human prostate cancer-derived cell line PC3 (ATCC), which is known to express truncated hCDCP1, was also lysed in the adherent culture state, or treated with a diluted trypsin / EDTA solution in the same manner as Ba / F3, and then lysed, and SDS-PAGE was performed. Proteins from the electrophoresed gel were transferred to a PVDF membrane and Western blotting was performed using anti-hCDCP1 intracellular domain antibody (Abcam) as the primary detection antibody, alkaline phosphatase-labeled anti-goat IgG antibody (Promega) as the secondary antibody, and NBT / BCIP stock solution (Roche) as the detection reagent. The results of Western blotting are shown in Figure 4B. In Ba / F3-hCDCP1 cells without trypsin treatment, a single band was observed at approximately 130 kDa, whereas in trypsin-treated cells, a single band was observed at approximately 70 kDa. This band size is considered to be at the same position as the band detected in the lysate sample of PC3 cells, suggesting that the molecular species generated by trypsin treatment of Ba / F3 is identical to the cleaved hCDCP1 expressed by PC3. A cleaved protein band was also detected at the same mobility after treatment with 500 nM plasmin. This suggested that the hCDCP1 expressed by Ba / F3-CDCP1 was cleaved by treatment with trypsin and plasmin, leaving the cleaved hCDCP1 molecule on the cells. Using this method, we then screened for antibodies that react with the cleaved hCDCP1 molecule.

[0081] 5. Screening of hybridomas using hCDCP1 knockout cell lines Hybridomas were prepared using the proteins and cells prepared in the previous section, and hybridomas producing antibodies that react specifically with hCDCP1 were screened. To screen for clones that react to hCDCP1 expressed by cancer cells, we generated PC3 cells in which the hCDCP1 molecule was knocked out. The human prostate cancer cell line PC3 was purchased from ATCC. PC3 cells 1×10 6 The CDCP1 CRISPR / Cas9 KO Plasmid(h) (SantaCruz) was introduced into each of the cells using Nucleofector 2b (Lonza). This plasmid contains the GFP gene sequence. After one day, cells that showed positive fluorescence for GFP were single-cell sorted onto a 96-well culture plate using a FACSAriaII (Becton Dickinson). After 7 to 10 days, expression was confirmed by flow cytometry using an anti-hCDCP1 antibody (R&D systems) for clones that showed cell proliferation, and cell clones that did not express hCDCP1 were amplified and used as hCDCP1-KO PC3 cells in the following experiments.

[0082] Monoclonal antibodies against hCDCP1 were produced using lymph node cells isolated from mice immunized with purified hCDCP1 protein or cells expressing hCDCP1. In experiment 1, 50 μg of hCDCP1-ECD protein per mouse was mixed with TiterMaxGold (TiterMax) and administered into the plantar vein. After 7 and 10 days, 10 μg of antigen per mouse was mixed with TiterMaxGold and administered into the plantar vein for booster immunization. In Experiment 2, Ba / F3-hCDCP1 cells were used as the immunizing substance. These cells were trypsinized by the method described above, and then 1 × 10 cells were injected into each mouse. 7The cells were mixed with TiterMaxGold (TiterMax) and administered into the plantar vein. Seven and ten days later, 1 × 10 cells were administered per mouse. 5 Booster immunization was performed by intraperitoneally administering 10 trypsinized Ba / F3-hCDCP1 cells suspended in PBS. In experiment 3, hCDCP1-ECD protein was cleaved with plasmin as described above, and then purified using His-tag and FLAG-tag affinity resins. 50μg of this product was mixed with TiterMaxGold and administered into the plantar vein of each mouse. 7 and 10 days later, 10μg of antigen was mixed with TiterMaxGold and administered into the plantar vein of each mouse for booster immunization.

[0083] In each of the three experiments, popliteal lymph nodes were collected from the mice after immunization, and cell suspensions were prepared. These were then mixed with SP2 / 0-Ag14 myeloma cells in serum-free RPMI1640 (ATCC modified; Thermo Fisher Scientific) and cell fusion was performed using polyethylene glycol (Roche). The fused cells were then cultured in ClonaCell 1000 ml containing Hybridoma enhancing medium (Sigma-Aldrich). TM The cells were suspended in -HY Medium D (STEMCELL) and seeded on plastic petri dishes. Colonies formed 8 to 10 days later were isolated in 96-well plastic plates containing medium (RPMI1640 / 10% FBS / HAT supplement (Thermo Fisher Scientific) / Hybridoma enhancing medium (Sigma-Aldrich)) and expanded. The culture supernatant was used to evaluate binding activity. The hCDCP1 binding of the antibodies produced by the hybridoma was screened by cell ELISA using PC3 cells and hCDCP1-KO PC3 cells. PC3 cells and hCDCP1-KO PC3 cells were seeded at 5000 cells / well on a 384-well plate and cultured overnight at 37°C. After removing the culture medium, 25μL of hybridoma culture supernatant or antibody solution diluted with medium was added per well and reacted at 4°C for 1 hour. After removing the culture supernatant and washing with PBS, secondary antibody diluted with medium was added and reacted at 4°C for 1 hour. After washing three times with PBS and completely removing the washing solution, substrate solution (ELISA POD substrate TMB Kit, Nacalai Tesque) was reacted, stopped after 10 minutes with 1N sulfuric acid, and absorbance at 450 nm was measured. The results of cell ELISA using the culture supernatants of hybridomas prepared from immunized animals in Experiments 1, 2, and 3 are shown in Figures 5A, 5B, 5C, and 5D. The values ​​in the figures indicate the values ​​obtained by subtracting the values ​​obtained by reacting with hybridoma culture medium containing no antibody from the measured values ​​of each clone. These results indicate that hybridomas producing antibodies that react with PC3 cells, which are cancer cells expressing hCDCP1, were obtained from immunized mice using hCDCP1 purified protein or forced expression cells. In addition, the results in Figures 5A, 5B, and 5C indicate that the antibodies produced by the hybridomas shown in these figures do not react with hCDCP1-KO PC3 cells, indicating that these antibodies indeed react specifically with hCDCP1 expressed in PC3.

[0084] 6. Antibody Sequence Analysis Regarding the hybridoma producing the antibody that specifically reacts with hCDCP1 expressed in PC3 cells described in the previous section, the sequence of the immunoglobulin gene variable region was analyzed as follows. Hybridoma cells were expanded and cultured, and RNA was extracted and cDNA was synthesized using reverse transcriptase using the SuperPrep II Cell Lysis & RT Kit (Toyobo Co., Ltd.). Antibody genes were amplified from the synthesized cDNA by PCR. Primers that recognize the upstream of the variable region and the downstream of the constant region were used for both the heavy and light chains. The primer sequences used are as follows: For heavy chain 5' amplification: 5'MsVHE (SEQ ID NO: 6) For heavy chain 3' amplification: 3'Cg1_outer (SEQ ID NO: 7), 3'Cg2c_outer (SEQ ID NO: 8), 3'Cg2b_outer (SEQ ID NO: 9), 3'Cg3_outer (SEQ ID NO: 10), 3'mIgG2a_CH (SEQ ID NO: 11) For κ light chain 5' amplification: 5'L-Vk_3 (SEQ ID NO: 12), 5'L-Vk_4 (SEQ ID NO: 13), 5'L-Vk_5 (SEQ ID NO: 14), 5'L-Vk_6 (SEQ ID NO: 15), 5'L-Vk_689 (SEQ ID NO: 16), 5'L-Vk_14 (SEQ ID NO: 17), 5'L-Vk_19 (SEQ ID NO: 18), 5'L-Vk_20 (SEQ ID NO: 19) For κ light chain 3' amplification: 3'mCk (SEQ ID NO: 20) For 5' amplification of lambda light chain: 5'mVλ1 / 2 (SEQ ID NO: 21), 5'mVλx (SEQ ID NO: 22) For 3' amplification of λ light chain: 3'mCλ_outer (SEQ ID NO: 23) The resulting DNA fragment was cloned into the pcDNA3.4 vector using a TOPO TA cloning kit (Thermo Scientific), and the DNA sequence was analyzed. The CDR regions of the resulting antibody sequences were determined according to the method of Kabat et al. (Sequences of Proteins of Immunological Interests, Fifth edition, NIH Publication No. 91-3242, US Department of Health and Human Services, 1991). The antibody sequences of the analyzed clones and their CDR region sequences are summarized in Table 1 below. [Table 1]

[0085] 7. Preparation of Mouse-Human Chimeric Antibodies Based on the antibody sequences obtained in the previous section, chimeric antibodies (hereinafter referred to as mouse-human chimeric antibodies) were constructed using mouse heavy and light chain variable regions and human IgG1 heavy and κ light chain constant regions. Mouse-human chimeric antibodies were constructed for six of the antibodies sequenced in the previous section: 12A041, 14A025, 14A043, 14A055, 14A063, and 14A091. For each of the antibody sequences 12A041, 14A025, 14A043, and 14A063, the same amino acid sequences as those analyzed in the previous section were prepared, and amino acid modifications in the CDR regions were also attempted. The correspondence between the introduction sites of sequence modifications, the amino acids after modification, and the names of the antibody sequences after modification are shown in Figures 6A and 6B. Modifications were performed on cysteines that are not thought to form SS bonds in the CDR regions, and on sequences that are likely to undergo cleavage and oxidation reactions in protein solvents. In addition, it was thought that the affinity for the antigen would be improved for mh12A041HCori by mutating the serine residue at Kabat number 32 to a phenylalanine residue, and for mh12A041LCori by mutating the histidine residue at Kabat number 27d to a tyrosine residue, or by simultaneously mutating the serine residue at number 27a and the asparagine residue at number 28 to an asparagine residue and an aspartic acid residue, respectively, and therefore sequences containing these mutations (mh12A041HCoriA, mh12A041LCoriA, mh12A041LCoriB) were designed. The antibody sequences with CDR sequence modifications and the sequences from which they are derived are shown in Table 2 below. [Table 2]

[0086] For the sequences in Table 2, the base sequences were designed based on the amino acid sequences of the antibody molecules, and expression vectors were synthesized by gene synthesis. In addition, for the 14A055 heavy chain variable region (sequence number 80), the 14A055 light chain variable region (sequence number 84), the 14A091 heavy chain variable region (sequence number 96), and the 14A091 light chain variable region (sequence number 100), expression vectors were prepared by amplifying the variable region DNA by PCR from the sequence subcloned into the pcDNA3.4 vector in the previous section and subcloning it. The vectors used were pFUSE-CHIg-hG1 vector (Invivogen) for the heavy chain variable region sequences of antibody groups other than the mh12A041 and mh14A025 antibody groups, pFUSE-CHIOME-HC vector for the heavy chain variable region sequences of the mh12A041 and mh14A025 antibody groups, and pFUSE2-CLIg-hk vector (Invivogen) for the light chain variable region sequences of all antibody groups. The amino acid sequences of the antibody constant regions of these three vectors are shown in SEQ ID NO: 157, SEQ ID NO: 158, and SEQ ID NO: 159, respectively. The constructed expression vectors were expressed in Expi293 cells (Thermo Fisher Scientific). At this time, for each of the antibody groups mh12A041, mh14A025, mh14A043, and mh14A063, heavy and light chain vectors were combined in the following pattern for expression.

[0087] mh12A041: HCori / LCori, HCori / LCv1, HCori / LCv2, HCv1 / LCori, HCv1 / LCv1, HCv1 / LCv2, HCori / LCoriA, HCori / LCoriB, HCoriA / LCori, HCoriA / LCoriA, HCoriA / LCoriB mh14A025: HCori / LCori, HCv1 / LCori, HCv2 / LCori, HCv5 / LCori, HCv7 / LCori, HCv8 / LCori, HCv15 / LCori, HCv17 / LCori, HCv19 / LCori, HCv21 / LCori, HCv24 / LCori mh14A043: HCori / LCori, HCori / LCv1, HCv1 / LCori, HCv1 / LCv1, HCv2 / LCori, HCv2 / LCv1 mh14A063: HCori / LCori, HCori / LCv1, HCv1 / LCori, HCv1 / LCv1, HCv2 / LCori, HCv2 / LCv1

[0088] The antibody protein concentration secreted into the culture supernatant was measured by the AlphaLISA method, and the antibodies were diluted to various concentrations based on the measured values ​​and bound to Ba / F3-hCDCP1 cells. The reactivity of each antibody was evaluated by reacting with a PE-labeled anti-human IgG antibody (Becton Dickinson) as a secondary antibody and measuring the fluorescence of PE by flow cytometry. The results are shown in Figure 7. As shown in the figure, all of the antibodies evaluated retained their reactivity with hCDCP1.

[0089] 8. Reactivity of antibodies to trypsin-treated hCDCP1-expressing cells To examine whether the obtained anti-hCDCP1 mouse-human chimeric antibody binds to the N-terminus of the serine protease cleavage site of the hCDCP1 molecule or to the transmembrane domain side, we examined the reactivity of the antibody against hCDCP1-expressing Ba / F3 cells cleaved by trypsin treatment using the method described in the previous section. The anti-hCDCP1 mouse-human chimeric antibody clones evaluated are as follows: mh12A041HCori / LCori, mh14A025HCori / LCori, mh14A043HCori / LCori, mh14A055, mh14A063HCori / LCori, mh14A091 Ba / F3-hCDCP1 cells were treated in RPMI1640 medium containing 0.05% trypsin at 37℃ for 30 minutes, and then washed twice with RPMI1640 / 10% FBS. In the treated Ba / F3-CDCP1 cells, the disappearance of the full-length form of hCDCP1 and the presence of only the truncated form were confirmed by Western blotting as described in the previous section. The cells were incubated with anti-hCDCP1 mouse-human chimeric antibody at 4℃ for 30 minutes in a dilution series from 5μg / mL to 5ng / mL. As a negative control, the anti-RS virus antibody described in the following section was used as a nonspecific human IgG antibody in a similar dilution series and incubated at 4℃ for 30 minutes. Next, PE-labeled anti-human IgG Fc antibody (Southern Biotech) was applied as a secondary antibody, and after washing, the fluorescence intensity was observed using a FACSCantoII (Becton Dickinson). The results are shown in Figure 8. These results indicate that all six evaluated antibodies react to cells with hCDCP1 molecules cleaved by trypsin treatment to the same extent as cells without trypsin treatment, that is, these antibodies bind to the transmembrane region side of the hCDCP1 molecule relative to the serine protease cleavage site.

[0090] 9. Reactivity of antibodies to cynomolgus monkey CDCP1 molecule A sequence was designed in which a Campath secretion signal and a c-Myc tag were linked to the N-terminus of the 30th to 836th amino acids of the CDCP1 protein (isoform X1; NCBI Reference Sequence: XP_005546930.1; SEQ ID NO: 160) of the cynomolgus monkey (Macaca fascicularis). This amino acid sequence was converted to a base sequence based on the mammalian codon table, and DNA containing this sequence and with a Kozak translation initiation sequence at the 5' end and a translation termination codon at the 3' end was synthesized by gene synthesis (GENEWIZ; SEQ ID NO: 162). The synthesized DNA was ligated to the KpnI / PmeI sites of pEF1 / V5-His A (Thermo Fisher Scientific) to create an expression vector GS02. Plasmids GS01 (prepared in the previous section) and GS02 were linearized with the restriction enzyme PmeI and diluted to 2 × 10 6 Two micrograms of the gene were introduced into each CHO-K1 cell using Nucleofector (Lonza). After gene introduction, the cells were cultured for three days in the presence of 400 μg / mL hygromycin, after which the cells were detached and dispersed, stained with anti-hCDCP1 antibody (R&D systems), and positive cells were subjected to single-cell sorting using FACS AriaII (Becton Dickinson). The cell clones that proliferated after separation were further stained with anti-hCDCP1 antibody (R&D systems) to isolate clones expressing hCDCP1 and cynomolgus monkey CDCP1. These clones were expanded and used below as hCDCP1-expressing CHO-K1 cells and cynomolgus monkey CDCP1-expressing CHO-K1 cells. hCDCP1 expressing CHO-K1 cells and cynomolgus monkey CDCP1 expressing CHO-K1 cells were detached and dispersed with 0.25% trypsin / EDTA (Nacalai Tesque), and then reacted with each of the mouse-human chimeric antibodies mh12A041HCori / LCori, mh14A025HCori / LCori, mh14A043HCori / LCori, mh14A055, mh14A063HCori / LCori, and mh14A091 at various concentrations, and binding was observed by flow cytometry (Figure 9). As a result, all of these six antibodies reacted with cynomolgus monkey CDCP1, and in particular, the five antibodies other than mh14A055 reacted to human and cynomolgus monkey CDCP1 to the same extent.

[0091] 10. Antibody reactivity to human cancer cells To investigate the reactivity of the anti-hCDCP1 antibodies to each human cancer cell line, the binding reactivity of the antibodies was evaluated by flow cytometry. The human cancer cell lines used and their sources are as follows: Prostate cancer: PC3 (ATCC), DU145 (RIKEN BioResource Center) Lung cancer: H358 (ATCC), SK-MES-1 (ATCC) Breast cancer: MDA-MB-231 (ATCC), HCC1143 (ATCC) Bile duct cancer: TFK-1 (RIKEN BioResource Center) Ovarian cancer: SK-OV3 (ATCC), OVCAR3 (ATCC) Pancreatic cancer: Capan-2 (ATCC) Colorectal cancer: DLD-1 (JCRB) Normal cells: Human mammary epithelial cells (HMEpC) (Cell Applications), normal human epidermal keratinocytes (NHEK) (PromoCell) Each human cancer cell line and normal tissue-derived cells were cultured in a 10 cm petri dish in a culture medium appropriate for each cell. The cells in the petri dish were detached and suspended using 0.25% trypsin / EDTA (Nacalai Tesque), then washed with FCM buffer, and each antibody was diluted with FCM buffer to the concentrations shown in the figure and reacted for 30 minutes at 4°C. After washing twice with FCM buffer, the cells were stained with a PE-labeled mouse anti-human IgG antibody (Becton Dickinson) as the secondary antibody. Flow cytometry analysis was performed using a FACSCantoII (Becton Dickinson). The results are shown in Figure 10A and Figure 10B. All of the candidate clones showed reactivity against cancer cells, as well as NHEK and HMEpC.

[0092] 11. Reactivity of mouse hybridoma-derived antibodies to bone marrow cells Mononuclear cells derived from bone marrow of healthy subjects were purchased from AllCells, STEMCELL Technologies, or Lonza. After thawing, the frozen cell stock was washed with IMDM medium (Thermo Fisher Scientific) containing 2% FBS, and then treated with FcR blocking reagent (Miltenyi) at room temperature for 15 minutes. A primary antibody sample with a final concentration of 10 μg / mL was added and reacted at 4 ° C for 30 minutes. As the primary antibody sample, an antibody sample purified from the culture supernatant of the hybridoma clone obtained by the screening in the previous section using protein G sepharose fast flow (GE Healthcare) was used. As the positive control antibody, anti-hCDCP1 antibody clone CUB1 (BioLegend) was used. Non-specific mouse IgG (monoclonal antibody clone 2E12) from MBL was used. After washing twice with FCM buffer, PE-labeled anti-mouse IgG antibody (Becton Dickinson) was reacted at 4 ° C for 30 minutes. After further washing twice with FCM buffer, APC-labeled anti-human CD34 antibody (BioLegend) was reacted at 4 ° C for 30 minutes. After washing twice with FCM buffer, the cells were suspended in FCM buffer containing 7-AAD (Becton Dickinson) and observed by flow cytometry using a FACSCantoII (Becton Dickinson). The reactivity of the evaluated antibodies is shown in Figure 11. Only CD34-APC positive live cells were gated from the flow cytometry data, and a histogram of their PE fluorescence values ​​was created. As shown in Figure 11, the CUB1 antibody binds more strongly to the CD34 positive cell fraction of bone marrow mononuclear cells than non-specific mouse IgG, while the reactivity of the following antibody clones listed in Figure 11 is sufficiently low, and even at a comparative antibody concentration of 10 μg / mL, it is comparable to non-specific mouse IgG, and the reactivity to the CD34 positive cell fraction is so weak that it cannot be detected. Clone: ​​01A2C5, 12A033, 12A038, 12A041, 14A025, 14C013

[0093] 12.Reactivity of mouse-human chimeric antibodies to bone marrow cells The same healthy bone marrow-derived mononuclear cells as in the previous section were used. After thawing, the frozen cell stock was washed with FCM buffer, and then FcR blocking reagent (Miltenyi) was added and treated at room temperature for 15 minutes. The primary antibody sample was added at a final concentration of 10 μg / mL and reacted at 4°C for 30 minutes. The mouse-human chimeric antibody prepared in section 7 was purified using rProtein A Sepharose Fast Flow (GE Healthcare). The anti-hCDCP1 antibody clone CUB1 (BioLegend) was used as the positive control antibody. The anti-RS virus antibody described in the following section was used as the non-CDCP1 specific human IgG (hIgG). After washing twice with FCM buffer, the cells were reacted with PE-labeled anti-human IgG antibody (SantaCruz) or, for the CUB1 antibody, PE-labeled anti-mouse IgG antibody (Becton Dickinson) at 4°C for 30 minutes. After further washing twice with FCM buffer, the cells were reacted with APC-labeled anti-human CD34 antibody (BioLegend) for 30 minutes at 4° C. After washing twice with FCM buffer, the cells were suspended in FCM buffer containing 7-AAD (Becton Dickinson) and observed with a FACSCantoII (Becton Dickinson). The reactivity of the evaluated antibodies is shown in Figure 12. Only CD34-positive live cells were gated from the flow cytometry data, and a histogram of their PE fluorescence values ​​was created. As shown in Figure 12, the CUB1 antibody bound more strongly to the CD34-positive cell fraction of bone marrow mononuclear cells compared to the cell group to which no primary antibody was added, whereas the reactivity of the following antibody clones listed in Figure 11 was sufficiently low, and even when the comparative antibody concentration was 10 μg / mL, the staining intensity was equivalent to that of cells to which no antibody was added, and reactivity to the CD34-positive cell fraction was not detectable. Clones: mh12A041HCori / LCori, mh14A025HCori / LCori, mh14A043HCori / LCori, mh14A055, mh14A063HCori / LCori, mh14A091

[0094] 13.Reactivity of biotinylated mouse-human chimeric antibodies to bone marrow cells To examine the reactivity of the following antibodies to bone marrow mononuclear cells from healthy subjects in detail using a more sensitive method, we directly biotinylated the mouse-human chimeric antibodies and used them to perform flow cytometry on bone marrow mononuclear cells. mh12A041HCori / LCori, mh14A025HCori / LCori, mh14A043HCori / LCori, mh14A063HCori / LCori The purified mouse-human chimeric antibody was biotinylated using the EZ-Link Micro NHS-PEG4-Biotinylation Kit (Thermo Scientific). The 25A11 antibody, CUB4 antibody, and anti-RS virus antibody used as controls were prepared based on the sequences listed below. 25A11 antibody: heavy chain variable region sequence as set forth in SEQ ID NO: 20 of International Publication No. WO 2008 / 133851, and light chain variable region sequence as set forth in SEQ ID NO: 4. The heavy chain variable region and light chain variable region produced in the present application are shown in SEQ ID NO: 163 and SEQ ID NO: 164, respectively. CUB4 antibody: heavy chain variable region sequence as set forth in SEQ ID NO: 3 and light chain variable region sequence as set forth in SEQ ID NO: 15 of US Patent US 8394928. The heavy chain variable region of the antibody produced in the present application is shown in SEQ ID NO: 165, and the light chain variable region is shown in SEQ ID NO: 166. Anti-RS virus antibody: sequence registered in the PBD database under registration number 2hwz. The heavy chain variable region is from base 1 to base 120 of 2hwz_H, and the light chain variable region is from base 1 to base 106 of 2hwz_L. The heavy chain variable region of the antibody produced in the present application is shown in SEQ ID NO: 167, and the light chain variable region is shown in SEQ ID NO: 168. The base sequences encoding these were genetically synthesized, and the heavy chain variable region was cloned into the pFUSE-CHIg-hG1 vector, and the light chain variable region was cloned into the pFUSE2-CLIg-hk vector. They were then expressed in Expi293 cells and purified with protein A Sepharose to prepare antibody samples. Hereinafter, anti-RS virus antibodies were used as nonspecific human IgG (hIgG). As controls, purified IgG protein derived from normal human serum (Sigma-Aldrich) and anti-hCDCP1 antibody clone CUB1 (BioLegend) were also biotinylated in the same manner and subjected to the experiment. The biotinylation value of each antibody was measured using a Biotin Quantification Kit (Thermo Scientific) and confirmed to be as shown in Table 3 below. [Table 3]

[0095] The bone marrow mononuclear cells derived from healthy subjects were the same as those used in the previous section. After thawing, the frozen cell stock was washed with FCM buffer or IMDM medium containing 2% FBS, and then FcR blocking reagent (Miltenyi) was added and treated at room temperature for 15 minutes. The primary antibody sample was added and reacted at 4°C for 30 minutes. After washing twice with FCM buffer, the cells were reacted with a mixture of PE-labeled goat anti-human IgG antibody (Southern Biotechnology), APC-labeled anti-human CD34 antibody (BioLegend), and PE-Cy7-labeled anti-human CD45 antibody (BioLegend) for 30 minutes at 4°C. After washing twice with FCM buffer, the cells were suspended in FCM buffer containing 7-AAD (Becton Dickinson) and observed with a FACSCantoII (Becton Dickinson). The reactivity to bone marrow cells from this experiment is shown in Figure 13. In this figure, only live cells that were both CD34 / CD45 positive were extracted from the observed results, and their average PE fluorescence intensity was calculated, and the relationship with the antibody concentration added was shown. As can be seen from this figure, while the biotinylated antibodies CUB1, CUB4, and 25A11 showed sufficient reactivity to bone marrow CD34-positive cells even at low concentrations, the reactivity of the biotinylated antibodies of the four antibodies, mh12A041HCori / LCori, mh14A025HCori / LCori, mh14A043HCori / LCori, and mh14A063HCori / LCori, to CD34-positive cells was significantly lower. In particular, at the comparative antibody concentration of 10 ng / mL, the biotinylated antibodies CUB1, CUB4, and 25A11 showed strong reactivity, while the biotinylated antibodies of the four antibodies, mh12A041HCori / LCori, mh14A025HCori / LCori, mh14A043HCori / LCori, and mh14A063HCori / LCori, showed weak reactivity, almost the same level as biotinylated non-specific human IgG. In addition, Fig. 13D shows a comparison of the reactivity of the anti-RS virus antibody biotinylated antibody used as the non-specific human IgG in Figs. 13A-C with the purified IgG biotinylated protein derived from normal human serum against bone marrow CD34-positive cells at a comparative antibody concentration of 10 ng / mL. Measurements were performed on three independent samples, and Fig. 13D shows the average value and standard error of the PE mean fluorescence intensity. This shows that the reactivity of the anti-hCDCP1 antibody group described in this invention is comparable to that of the IgG protein derived from human serum.

[0096] 14. Antibody reactivity to humanized and overexpressed cells Next, the following mouse anti-hCDCP1 antibody sequences were humanized: mh12A041HCv1, mh12A041LCv1, mh14A043HCv2, mh14A043LCv1, mh14A063HCv1, mh14A063LCori. Humanization was performed by CDR grafting from the sequences of each antibody variable region. The humanized sequences were designed based on the method described in the following paper: Tsurushita et al., 2005. Design of humanized antibodies: From anti-Tac to Zenapax. Methods 36:69-83. First, a three-dimensional molecular model of a mouse antibody was created by a standard method. Next, based on this molecular model, residues considered to be important for the formation of the CDR structure and residues considered to be essential for the reaction with the antigen were estimated in the amino acid sequence of the framework region. In parallel, sequences highly homologous to the heavy chain variable region and light chain variable region of each anti-hCDCP1 antibody were searched for in the cDNA sequence database of the human antibody heavy chain variable region and light chain variable region. After that, sequences were designed by linking the sequences of the framework part of the searched human antibody sequences with the CDR sequences of each anti-hCDCP1 antibody, and the sequences of residues considered to be essential for the formation of the CDR structure or the reaction with the antigen were further transplanted thereto to design a humanized antibody sequence. The designed sequences are shown in Table 4 below. The CDR sequences of the humanized antibody sequences are the same as those of the original mouse antibody, and are the same sequences as the above-mentioned sequence numbers. [Table 4]

[0097] It was believed that the affinity for the antigen would be improved for mh12A041HCv1 by mutating the serine residue at Kabat number 32 to a phenylalanine residue, and for mh12A041LCv1 by mutating the histidine residue at Kabat number 27d to a tyrosine residue, or by simultaneously mutating the serine residue at number 27a and the serine residue at number 28 to an asparagine residue and an aspartic acid residue, respectively. Therefore, the sequences shown in Table 5 below, which include these mutations, were created. [Table 5]

[0098] The DNA sequences encoding the designed variable region amino acid sequences were synthesized by gene synthesis. The heavy chain variable region sequence DNA was cloned into pFUSE-CHIg-hg1, a vector containing the human IgG1 constant region, after connecting with a human antibody secretory signal peptide or a human IL-2 secretory signal peptide. The light chain variable region sequence DNA was cloned into pFUSE2-CLIg-hk, a vector containing the human Igκ constant region, after connecting with a human antibody secretory signal peptide or a human IL-2 secretory signal peptide. The cloned plasmid was transfected into Expi293 cells using Expifectamine, and the antibody was expressed in the culture medium. The heavy and light chain vectors were combined in the following pattern. h12A041:VH1 / VL, VH2 / VL, VH1 / VLA, VH2 / VLA, VH1 / VLB, VH2 / VLB, VH1A / VL, VH2A / VL, VH1A / VLA, VH2A / VLA, VH1A / VLB, VH2A / VLB h14A043:VH1 / VL1, VH1 / VL2, VH1 / VL3, VH2 / VL1, VH2 / VL2, VH2 / VL3 h14A063:VH2 / VL1, VH2 / VL2, VH3 / VL1, VH3 / VL2, VH4 / VL1, VH4 / VL2 The culture medium was collected and purified by a filter, and the antibody protein concentration in the culture medium was measured by the AlphaLISA method. The reactivity of the constructed humanized anti-hCDCP1 antibody protein and the mouse-human chimeric antibody of the original clone for comparison was examined by flow cytometry. Ba / F3-hCDCP1 cells were used as target cells. PE-labeled mouse anti-human IgG antibody (Becton Dickinson) was used as the secondary antibody. As is clear from the results of this experiment shown in Figure 14, all of the designed humanized antibodies showed reactivity equal to or higher than that of the original mouse-human chimeric antibody. This indicates that the constructed humanized anti-hCDCP1 antibody group fully retains reactivity to hCDCP1.

[0099] 15.Reactivity of humanized antibodies to bone marrow cells Purified antibody proteins with the following heavy and light chain combinations were directly biotinylated using the same method as described in the previous section. h12A041:VH1 / VL, VH1 / VLA, VH1 / VLB, VH1A / VL h14A043:VH1 / VL1, VH1 / VL2, VH1 / VL3, VH2 / VL1, VH2 / VL2, VH2 / VL3 h14A063:VH4 / VL1, VH4 / VL2 As comparative antibodies, the 25A11 antibody, CUB4 antibody, and CUB1 antibody prepared in Section 13, an anti-RS virus antibody as a nonspecific human IgG antibody, and a directly biotinylated purified IgG protein derived from normal human serum were used. The biotinylation valence of each antibody was measured using a Biotin Quantification Kit (Thermo Scientific) and was as shown in Table 6 below. [Table 6] Using these antibodies, we observed the reactivity to healthy bone marrow-derived mononuclear cells by flow cytometry using the same method as described in the previous section. Figure 15 shows a histogram of the PE fluorescence intensity of the CD34 / CD45 positive live cell population after binding the following biotinylated antibodies to healthy bone marrow-derived mononuclear cells at a concentration of 10 ng / mL and observing them by flow cytometry. 25A11-biotin, CUB4-biotin, CUB1-biotin, h12A041VH1 / VLA-biotin, h14A043VH1 / VL1-biotin, h14A063VH4 / VL2-biotin, human serum IgG-biotin These results demonstrated that the antibody described in the present invention had extremely low reactivity to CD34 positive cells in bone marrow compared to the comparative antibody, and had almost no reactivity, similar to IgG in human serum. Further, the same experiment was carried out for the antibody group shown in Table 6, varying the antibody concentration. From the flow cytometry measurement results, the concentration dependence of the average fluorescence intensity of PE in CD34-positive live cells is shown in a log-log graph in FIG. 16. As a result, while the CUB4 antibody exhibited significant binding to CD34-positive cells in the bone marrow, all of the antibodies shown in Table 6 were sufficiently low in reactivity to the positive control antibody, and at least at a comparison antibody concentration of 10 ng / mL, the reactivity was comparable to that of the nonspecific human IgG antibody observed as the negative control. From this, it was considered that the antibodies described in the present invention have low reactivity to hematopoietic stem cells, and therefore modified antibodies with antitumor activity derived from these antibodies also have low cytotoxicity to bone marrow cells.

[0100] 16. Generation of PBD-conjugated antibodies For the mouse-human chimeric antibody prepared in Section 7, the humanized antibody prepared in Section 14, and the anti-RS virus antibody described above as a non-specific human IgG, we directly bound the purified antibodies to the anticancer drug PBD (pyrrolobenzodiazepine) to prepare antibody-drug conjugates and evaluated their cytotoxicity. mh12A041HCori / LCori, mh14A025HCori / LCori, mh14A043HCori / LCori, mh14A063HCori / LCori h12A041VH1 / VLA, h14A043VH1 / VL1, h14A063VH4 / VL2 After cleaving the disulfide bond of 2 mg of purified antibody with TCEP (Tris(2-carboxyethyl)phosphine Hydrochloride), maleimide-PEG4-Val-Ala-PBD was bound by maleimide-cysteine ​​addition reaction. Unreacted maleimide-PEG4-Val-Ala-PBD was then removed using a Sephadex G50 column, and the buffer was replaced with PBS pH 7.2. The names of the PBD-conjugated antibodies prepared are as follows. The names are shown together with the number of PBD molecules bound per antibody molecule (DAR), measured by the ratio of absorbance at 333 nm to absorbance at 280 nm. Mouse-human chimeric antibody PBD conjugate antibody: mh12A041-PBD (DAR 2.1); mh14A025-PBD (DAR 1.7); mh14A043-PBD (DAR 1.8); mh14A063-PBD (DAR 2.0) Humanized antibody PBD conjugate antibody: h12A041-PBD (DAR 3.8); h14A043-PBD (DAR 3.6); h14A063-PBD (DAR 3.4) Anti-RS virus antibody PBD conjugate antibody: hIgG-PBD (DAR 2.2)

[0101] 17. Cytotoxic activity of mouse-human chimeric antibody PBD conjugate antibody against cancer cells The cytotoxic activity of the PBD-conjugated antibody of the mouse-human chimeric antibody prepared in the previous section against cultured cells was evaluated by the following method. The cancer cell lines and primary cultured cells derived from normal tissues were the same as those described in the previous section. Cells were cultured in appropriate medium in plastic dishes. Cells were detached by trypsinization and seeded at 1,000 or 2,000 cells per well in a 96-well flat-bottom plate. After incubation at 37°C until the next day, PBD-conjugated antibodies diluted to various concentrations with medium were added and incubated in a 37°C incubator for 3 days (DU145), 5 days (PC3), or 7 days (other cells). Cell viability was then measured using Cell Titer Glo (Promega). For each concentration of PBD-conjugated antibody, two wells were treated under the same conditions, and the average value of the two wells was used. The four anti-hCDCP1 mouse-human chimeric antibody PBD conjugate antibodies tested showed strong cytotoxicity against lung cancer, breast cancer, ovarian cancer, prostate cancer, colon cancer, pancreatic cancer, and bile duct cancer cell lines. They also showed cytotoxicity against human mammary epithelial cells (HMEpC) and normal human epidermal keratinocytes (NHEK) at high concentrations, but the degree of cytotoxicity was weaker than the cytotoxicity against the cancer cell lines. Figure 17A shows the 50% inhibitory concentration (IC50) calculated from the experimental results, and Figure 17B shows a graph of this. As can be seen from these figures, the cytotoxic activity of each antibody PBD conjugate against NHEK and HMEpC was weaker than the cytotoxicity against the cancer cell lines.

[0102] 18. Cytotoxic activity of humanized antibody-PBD conjugate antibodies against cancer cells The cytotoxic activity of the humanized antibody-PBD conjugate antibody prepared in Section 16 against cultured cells was evaluated by the same method as in Section 17. The cell lines used were PC3, DU145, HCT116, MDA-MB-231, H358, SK-OV-3, and normal human epidermal keratinocytes (NHEK), and were the same as those described in Section 10. The cells were detached by trypsinization and seeded at 1000 or 2000 cells per well in a 96-well flat-bottom plate. After culturing at 37°C until the next day, PBD-conjugated antibodies diluted to various concentrations with medium were added and incubated in a 37°C incubator for 3 days (DU145), 5 days (PC3), or 7 days (other cells). The cell viability was then measured using Cell Titer Glo (Promega). For each concentration of PBD-conjugated antibodies, two wells were treated under the same conditions, and the measured value was the average of the two wells. The results are shown in Figure 18. The three humanized anti-hCDCP1 antibody PBD conjugates tested showed strong cytotoxicity against PC3, DU145, HCT116, MDA-MB-231, and H358 cells. The cytotoxicity against SK-OV-3 cells was weaker than these. Cytotoxicity was also shown against normal human epidermal keratinocytes (NHEK) at high concentrations, but the degree of cytotoxicity was weaker than that against cancer cell lines. The 50% inhibitory concentration (IC50) was calculated from the experimental results and is shown in the table below. [Table 7] As can be seen from Table 7, the cytotoxic activity of each antibody-PBD conjugate against NHEK was weaker than the cytotoxic activity against the cancer cell lines.

[0103] 19. Antitumor activity of mouse-human chimeric antibody PBD conjugates against xenograft models Next, the in vivo antitumor activity of the anti-CDCP1 mouse-human chimeric antibody PBD conjugate antibody was evaluated by the following method. Mice used were 6-7 week-old female scid mice (C.B17 / Icr-scidJcl; CLEA Japan) or nude mice (BALB / cAJcl-nu / nu; CLEA Japan). PC3 cells were purchased from ATCC. PC3 cells were cultured in a medium containing Ham's F-12K (Wako Pure Chemical Industries) supplemented with 7% FBS and 10 μg / mL Gentamicin (Thermo Scientific). The cells were detached with 0.25% trypsin / 0.02% EDTA (Thermo Scientific), washed with PBS, and then cultured at 5 × 10 cells per mouse. 6 The cells were mixed 1:1 with Matrigel growth factor reduced, phenol red-free (Corning) and implanted subcutaneously into the right flank of mice. Tumors grow and the average tumor volume reaches 100 mm 3 When the tumor volume exceeded 100 μg / g, the mice were randomized and grouped (8 mice per group) based on the tumor volume of each individual. On the same day or the next day of grouping, the test drug was administered once via the tail vein at 10 μL / g of mouse body weight. Measurement of tumor size and body weight was started 3-4 days after cell transplantation and was performed twice a week thereafter. The short and long diameters of the tumor were measured using a caliper, and the tumor size was calculated as (short diameter mm). 2 This was calculated as ×(long diameter mm)×3.14 / 6. 19A and B show the results of an experiment using a scid mouse model. Both mh12A041-PBD and mh14A025-PBD at 1 mg / kg and 0.3 mg / kg significantly suppressed tumor growth compared to the group administered the same concentration of nonspecific hIgG-PBD. The results of an experiment using a nude mouse model are shown in Figures 19C and D. Both the 1 mg / kg and 0.3 mg / kg groups of mh14A043-PBD and mh14A063-PBD significantly suppressed tumor growth compared to the group administered the same concentration of nonspecific hIgG-PBD.

[0104] 20. Antitumor activity of humanized antibody-PBD conjugate antibodies against xenograft models Next, the in vivo antitumor activity of the anti-CDCP1 humanized antibody PBD conjugate antibody was evaluated by the following method. Nude mice (BALB / cAJcl-nu / nu; Nippon Clea) female 6-7 weeks old were used. PC3 cells and HCT116 cells were purchased from ATCC. PC3 cells were cultured in a medium containing Ham's F-12K (Wako Pure Chemical Industries) supplemented with 7% FBS and 10 μg / mL Gentamicin (Thermo Scientific). HCT116 cells were cultured in a medium containing McCoy's 5A (Thermo Scientific) supplemented with 10% FBS and 1% penicillin / streptomycin (Nacalai Tesque). The cells were detached with 0.25% trypsin / 0.02% EDTA (Thermo Scientific), washed with PBS, and then cultured at 5 × 10 cells per mouse. 6 The cells were mixed 1:1 with Matrigel growth factor reduced, phenol red-free (Corning) and implanted subcutaneously into the right flank of mice. Tumors grow and the average tumor volume reaches 100 mm 3 When the tumor volume exceeded 100 mg / kg, the mice were randomized and grouped (8 mice per group) based on the tumor volume of each individual. On the same day as grouping, the test drug was administered once via the tail vein at 10 μL / g of mouse body weight. Measurement of tumor size and body weight was performed in the same manner as in item 19. Figure 20A shows the results of an experiment using a xenograft model with PC3 cells. A single dose of 1 mg / kg of h12A041-PBD, h14A043-PBD, and h14A063-PBD significantly suppressed tumor growth compared to the group administered the same concentration of nonspecific hIgG-PBD. Figure 20B shows the results of an experiment using a xenograft model with HCT116 cells. A single dose of 0.3 mg / kg of h12A041-PBD, h14A043-PBD, and h14A063-PBD significantly suppressed tumor growth compared to the group administered the same concentration of nonspecific hIgG-PBD.

[0105] 21. Preparation of MMAE-conjugated antibodies For the humanized antibody produced in Section 14 below, an antibody-drug conjugate was produced by directly binding the purified antibody to the anticancer drug MMAE (monomethyl auristatin E), and its cytotoxicity was evaluated. h14A043VH1 / VL1, h14A063VH4 / VL2 After disulfide bonds were cleaved with TCEP (Tris(2-carboxyethyl)phosphine Hydrochloride) for approximately 30 mg of purified antibody, maleimidocaproyl-valine-citruline-p-aminobenzyloxycarbonyl-monomethyl auristatin E (MC-vc-PAB-MMAE) was conjugated by maleimide-cysteine ​​addition reaction. Unreacted MC-vc-PAB-MMAE was then removed using a Sephadex G50 column, and the buffer was replaced with PBS pH 7.2. The number of MMAE molecules bound per antibody molecule was measured by the ratio of absorbance at 248 nm to absorbance at 280 nm, and was approximately 3.9 for h14A043VH1 / VL1 and approximately 4.0 for h14A063VH4 / VL2. The names of the MMAE-conjugated antibodies produced are as follows: h14A043-MMAE;h14A063-MMAE

[0106] 22. Antitumor activity of humanized MMAE-conjugated antibodies against xenograft models Next, the in vivo antitumor activity of the humanized antibody MMAE conjugate antibody was evaluated by the following method. Nude mice (BALB / cAJcl-nu / nu; Japan CLEA) female mice aged 6 to 7 weeks were used. HCT116 cells purchased from ATCC were used. HCT116 cells were cultured in the same manner as in Section 20. The cells were detached with 0.25% trypsin / 0.02% EDTA (Thermo Scientific), washed with PBS, and then cultured at 5 × 10 cells per mouse. 6 The cells were mixed 1:1 with Matrigel growth factor reduced, phenol red-free (Corning) and implanted subcutaneously into the right flank of mice. Tumors grow and the average tumor volume reaches 100 mm 3 When the tumor volume exceeded 100%, each individual was randomized and grouped (8 mice per group) based on the tumor volume. The test drug and the vehicle phosphate buffered saline (PBS) as a negative control were administered twice a week for a total of four times, starting with the day of grouping as the first administration day. Administration was performed via tail vein at 10 μL / g of mouse body weight. Tumor size and body weight were measured using the same method as in Section 19. The experimental results are shown in Figure 21. Both h14A043-PBD and h14A063-PBD significantly suppressed tumor growth after four doses of 5 mg / kg compared to the PBS-administered group. [Industrial Applicability]

[0107] The antibodies and antigen-binding fragments thereof provided by the present invention are believed to play an important role in providing cancer prevention or treatment, and in the development of cancer prevention or treatment agents, etc., and therefore the present invention is expected to be utilized in the medical field, pharmaceutical field, etc.

Claims

1. An antibody or antigen-binding fragment thereof that binds to human CDCP1 (CUB domain-containing protein 1), characterized in that the amino acid sequences of CDRs (complementarity determining regions) 1 to 3 satisfy any of the following (A) to (K). (A) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 25; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:26; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:27; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:29; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 30, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 31; (B) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 33; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 34; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 35; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:37; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:38, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 39; (C) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 41; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:42; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:43; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:45; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:46, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:47; (D) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 49; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:50; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:51; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:53; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:54, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:55; (E) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:57; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:58; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:59; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:61; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:62, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:63; (F) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 65; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:66; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:67; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:69; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 70; Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:71; (G) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 73; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:74; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:75; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:77; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:78, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:79; (H) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 81; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:82; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:83; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:85; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:86, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 87; (I) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 89; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:90; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:91; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:93; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:94, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:95; (J) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 97; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:98; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:99; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 101; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 102; Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 103; (K) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 105; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 106; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 107; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 109; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 110; and It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:

111.

2. An antibody or antigen-binding fragment thereof that binds to human CDCP1 (CUB domain-containing protein 1), characterized in that the amino acid sequences of CDRs (complementarity determining regions) 1 to 3 satisfy either (L) or (M) below. (L) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 49; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:50; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:51; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:53; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:54, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:55; (M) An antibody that satisfies (L) above, which has heavy and light chain variable regions in which the amino acid at position 102 in the heavy chain and the amino acids at positions 28 and 29 in the light chain, as defined by the Kabat definition, are substituted with other amino acids.

3. An antibody or antigen-binding fragment thereof that binds to human CDCP1 (CUB domain-containing protein 1), characterized in that the amino acid sequences of CDRs (complementarity determining regions) 1 to 3 satisfy either (N) or (O) below. (N) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 65; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:66; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:67; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:69; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 70, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:71; (O) An antibody that satisfies (N) above, which has a heavy chain variable region in which the amino acids at positions 54 and 55 of the heavy chain according to the Kabat definition are substituted with other amino acids.

4. An antibody or antigen-binding fragment thereof that binds to human CDCP1 (CUB domain-containing protein 1), characterized in that the amino acid sequences of CDRs (complementarity determining regions) 1 to 3 satisfy either (P) or (Q) below. (P) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 73; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:74; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:75; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:77; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:78, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:79; (Q) An antibody that satisfies (P) above, which has heavy and light chain variable regions in which the amino acids at positions 54 and 55 in the heavy chain and the amino acid at position 33 in the light chain, as defined by the Kabat definition, are substituted with other amino acids.

5. An antibody or antigen-binding fragment thereof that binds to human CDCP1 (CUB domain-containing protein 1), characterized in that the amino acid sequences of CDRs (complementarity determining regions) 1 to 3 satisfy either (R) or (S) below. (R) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 89; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:90; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:91; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:93; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:94, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:95; (S) An antibody that satisfies the above (R), which has heavy and light chain variable regions in which the amino acids at positions 52a and 53 in the heavy chain and at position 33 in the light chain, as defined by the Kabat definition, are substituted with other amino acids.

6. An antibody or antigen-binding fragment thereof that binds to human CDCP1 (CUB domain-containing protein 1), characterized in that the amino acid sequences of CDR1 to 3 of a heavy chain variable region satisfy any one of the following (T) to (U), and the amino acid sequences of CDR1 to 3 of a light chain variable region satisfy any one of the following (t) to (v). (T) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 49; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:50; and Having a heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 113; (U) a heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 183; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:50; and Having a heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO: 113; (T) a light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 117; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:54, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:55; (u) a light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 186; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:54, and Having a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:55; (v) a light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 188; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:54, and It has a light chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:

55.

7. A heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:73; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 147; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:75; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO: 149; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:78, and An antibody or antigen-binding fragment thereof that binds to human CDCP1 (CUB domain-containing protein 1), characterized in having a light chain CDR3 comprising the amino acid sequence represented by sequence number 79.

8. A heavy chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:89; A heavy chain CDR2 comprising the amino acid sequence represented by SEQ ID NO: 151; A heavy chain CDR3 comprising the amino acid sequence represented by SEQ ID NO:91; A light chain CDR1 comprising the amino acid sequence represented by SEQ ID NO:93; A light chain CDR2 comprising the amino acid sequence represented by SEQ ID NO:94, and An antibody or antigen-binding fragment thereof that binds to human CDCP1 (CUB domain-containing protein 1), characterized in having a light chain CDR3 containing the amino acid sequence represented by sequence number 95.

9. An antibody or antigen-binding fragment thereof according to any one of claims 1 to 8, characterized in that it is a humanized antibody.

10. 10. The antibody or antigen-binding fragment thereof according to claim 1, which is bound to a substance having antitumor activity.

11. The antibody or antigen-binding fragment thereof according to claim 10 , wherein the substance having antitumor activity is an anticancer drug.

12. Fab, Fab', F(ab') 2 12. An antigen-binding fragment according to any one of claims 1 to 11, characterized in that it is an Fv, a single chain antibody, an scFv, an scFv dimer or a dsFv.

13. A pharmaceutical composition comprising an antibody or antigen-binding fragment thereof according to any one of claims 1 to 12.

Images