Bispecific antibody constructs for CDH3 and CD3

Abstract

The present invention relates to a bispecific antibody construct comprising a first human binding domain that binds to human CDH3 on the surface of a target cell and a second binding domain that binds to human CDS on the surface of a T cell. Furthermore, the present invention provides a vector comprising a polynucleotide encoding the antibody construct, said polynucleotide, and a host cell that is transformed or transfected with said polynucleotide or the vector. Furthermore, the present invention provides a method for producing the antibody construct of the present invention, said antibody construct for medical use, and a kit comprising said antibody construct.

Description

Technology Field

[0001] The present invention relates to a bispecific antibody construct comprising a first human binding domain that binds to human CDH3 on the surface of a target cell and a second binding domain that binds to human CD3 on the surface of a T cell. Furthermore, the present invention provides a vector comprising a polynucleotide encoding the antibody construct, said polynucleotide, and a host cell that is transformed or transfected with said polynucleotide or the vector. Furthermore, the present invention provides a method for producing the antibody construct of the present invention, said antibody construct for medical use, and a kit comprising said antibody construct. Background Technology

[0002] The cadherin superfamily comprises over 100 members in humans, including the so-called classical cadherins P-cadherin, E-cadherin, N-cadherin, and R-cadherin, each of which is active in individual tissue sets (Takeichi M. Development, 102:639-55(1988); van Roy F., Nature Rev., V14:121-134 (2014)). Cadherins play an essential role in the homeostasis of various tissues, as well as in the development of adult tissues and organs (Conacci-Sorrell M, et al., J Clin Invest, 109:987-91, (2002)). Cadherins are transmembrane glycoproteins that regulate cell-cell junction processes by forming calcium-dependent junctions and by converting mechanical stimuli into electrochemical activity, a process referred to as mechanotransduction (Gumbiner J. Cell. Biol., 148:399-404 (2000); Yagi, et al., Genes Dev., 14:1169-1180 (2000, Parades et al. Biochimica et Biophysica Acta 1826, 297-311 (2012)).

[0003] Placental cadherin (P-cadherin), also known as calcium-dependent cell-cell adhesion protein 3 (CDH3), is a 118 kDa protein with a large extracellular domain (ECD) of about 800 amino acids. The CDH3 ECD contains five cadherin repeats, which are designated herein as (extracellular) domain 1-5 / Dom1-Dom5 / D1-D5 and each consists of about 110 amino acids. The proteins with the highest sequence homology to P-cadherin are E-cadherin, with 53% sequence homology, and N-cadherin, with 39% homology.

[0004] P-cadherin expression levels are considered to be low in healthy adult individuals and limited to the basal or lower layers of stratified epithelium, including prostate, skin, and mammary myoepithelial cells (References [Takeichi M. J Cell Biol 103:2649-58, (1986) and Shimoyama Y, et al., Cancer Res, 49:2128-33(1989)]). Although not lethal in gene knockout mice, loss of P-cadherin function has been shown to be associated with developmental fusion as well as mammary epithelial hyperplasia and dysplasia (GL Radice et al. J. Cell Biol. 139: 1025-1032 (1997)).

[0005] In contrast to low gene expression in healthy individuals, the expression of P-cadherin is upregulated in some diseases, including immune diseases such as Crohn's disease and colitis (Hardy, et al., Gut 50:513-519 (2002)). Additionally, P-cadherin is considered to play a significant role in the invasion-inducing properties of cancer cells (Furukawa, et al., Microscopy Res. Technique 38 (4):343-352 (1997), Parades et al. Clin Cancer Res. 11(16), 5869-5877 (2005), Parades et al. Biochimica et Biophysica Acta 1826, 297-311 (2012)). Upregulation of P-cadherin has been described in various tumors, including colorectal, pulmonary (NSCLC), breast (triple-negative), pancreatic, head and neck, thyroid, cervical, ovarian, and gastric cancers (Milic et al. Cancer Res 68: (19) 7760-7768 (2008)., Imai et al. Clin Cancer Res 14(20) 6487-6495 (2008), Paredes et al. Clin Cancer Res 11 (16) 5869-5877 (2005), Dasgupta et al. Oral Oncology 42, 306-316 (2006), Jarzab et al. Cancer Res; 65: (4) 1587-1597 (2005)., Patel et al. Int. J. Cancer: 106, 172-177 (2003), Kim et al. Human Pathology 41, 877-885 (2010)). Furthermore, increased expression of P-cadherin has been observed to correlate with lower survival rates in patients across various cancer types (Sun L et al. Am J Pathol. 2011; 179:380-90; Gamallo, Modern Pathology, 14:650-654, (2001); Stefansson, et al., J.Clin. Oncol. 22(7):1242-1252 (2004), Parades et al. Cancer Res. 64, 8309-8317 (2004)), Taniuchi K et al. Cancer Res. 2005; 65:3092-9; Paredes J et al. Clin Cancer Res. 2005; 11:5869-77; Hardy RG et al. Gut. 2002; 50:513-9; Peralta Soler A et al. Cancer. 1999; 86:1263-72).

[0006] After cardiovascular disease, neoplasms rank second among non-communicable diseases as a cause of death, causing approximately 8.3 million cases worldwide in 2013 (GBD 2013 Lancet 2015; 385: 117-71). The absolute number of cancer cases has increased by 45.6% since 1990, which can be attributed to the fact that the global population is growing, people are aging, and the prevalence of established risk factors (e.g., smoking, overweight, inactivity) is also increasing (GBD 2013 Lancet 2015; 385: 117-71, Torre LA et al. CA Cancer J Clin. 2015; 65:87-108). In humans, lung cancer is the leading cause of death with approximately 1.1 million established cases annually, followed by liver and gastric cancer. In women, breast cancer is the leading cause of death with 500,000 cases annually, followed by lung and colorectal cancer. It is estimated that 14 million new cases of cancer occurred in 2012, led by lung cancer with approximately 1.2 million cases in men and breast cancer with approximately 1.7 million cases in women (Torre LA et al. CA Cancer J Clin. 2015; 65:87-108). Cancer collectively constitutes a significant burden on global society, and there is a great need for treatment options to combat this disease.

[0007] Conventional strategies for treating cancer involve surgery, followed by chemotherapy, radiation therapy, or, more recently, targeted therapy or a combination thereof (e.g., NCCN guidelines for oncology). Chemotherapy agents include nucleotide analogs such as 5-fluorouracil (5-FU), DNA damaging agents such as oxaliplatin, and topoisomerase inhibitors such as irinotecan or microtubule inhibitors such as docetaxel, all of which cause inhibition of tumor cell proliferation. Targeted therapies include small molecule compounds that selectively inhibit mutated oncogenic kinases, such as vemurafenib, a BRAF V600E selective compound approved for the treatment of melanoma (Garbe C. et al Recent Results Cancer Res. 2014; 201:215-25), or anaplastic lymphoma kinase (ALK) inhibitors crizotinib and ceritinib, approved for the treatment of lung cancer (Pall G. Curr Opin Oncol. 2015; 27:118-24). Additionally, there are antibody-based compounds such as cetuximab and panitumumab, which recognize and inactivate the epidermal growth factor receptor (EGFR) and are approved for the treatment of KRAS wild-type colon cancer (Tol J, Punt CJ. Clin Ther. 2010; 32:437-53), or trastuzumab, which recognizes Her2 and is approved for the treatment of breast cancer (Ahmed S. et al. Breast Cancer. 2015; 22:101-16).

[0008] Despite research into a wide variety of therapeutic interventions, treatment for this heterogeneous disease is not available except in sporadic cases. The heterogeneity of individual tumors leads to the fact that only a limited number of patients respond to specific therapies, and furthermore, that resistance to therapies develops during tumor progression (Jamal-Hanjani M. et al. Clin Cancer Res. 2015; 21:1258-1266). Finally, while newly developed targeted therapies have been shown to be better tolerated, drug side effects could also lead to treatment discontinuation or blockade ( http: / / www.cancer.net / navigating-cancer-care / how-cancer-treated / chemotherapy / side-effects-chemotherapy ; Sun G. C. et al. Anticancer Agents Med Chem. 2015 Mar 17. [Epub ahead of print]).

[0009] Also, for this reason, there is still a high medical need for the development of new drugs.

[0010] Overexpression of CDH3 in tumor cells provides the basis for a new approach to treating cancer by using bispecific antibodies that recognize CDH3, which overexpresses tumor cells, and kill them through the orientation of cytotoxic T cells that recognize the CD3 binding moiety of the bispecific antibody.

[0011] Expression analyses performed at the RNA level or at the protein level using immunohistochemistry demonstrate that CDH3 is expressed at high and detectable levels in a wide variety of different types of cancer, including colon cancer (Milic et al., Kita et al, Imai et al., http: / / www.proteinatlas.org / ), lung cancer including non-small cell and small cell lung cancer (Imai et al, Human Protein Atlas), breast cancer, preferably triple-negative cancer (Perou et al, Paredes et al, Turashvili et al, Imai et al.), pancreatic cancer (Imai et al., Taniuchi et al), head and neck cancer including squamous cell carcinoma of the tongue (Dasgupta et al; data itself is not publicly disclosed), thyroid cancer (Jarzab et al, Rocha et al., Human Protein Atlas), cervical cancer (Imai et al., Han et al.), ovarian cancer, preferably stage II or higher ovarian cancer (Patel et al human proeion atlas), and stomach (Kim et al; Imai et al). Includes endometrial cancer (Sugiyama Y et al. Clin Cancer Res. 2003; 9:5589-600; Human Protein Map), cholangiocarcinoma (Baek S et al. Anat Cell Biol. 2010; 43:110-7; Imai et al), bladder cancer (Imai et al, data itself not published), prostate cancer (Imai et al; data itself not published), testicular cancer (Imai et al), soft tissue sarcoma (Imai et al), esophageal cancer (data itself not published), and kidney cancer (data itself not published).

[0012] Future analyses may subsequently demonstrate elevated expression in additional tumors or subtypes thereof that may also be suitable for this therapy. Fotouhi et al. demonstrated, for example, reduced methylation of the CDH3 promoter in small intestinal neuroendocrine tumors that can cause elevated CDH3 protein expression (Fotouhi O et al. Epigenetics. 2014; 9:987-97). A positive correlation was observed between elevated CDH3 expression and miR-205 (a microRNA identified as a specific marker for lung squamous cell carcinoma) (Huang W et al. Chin Med J (Engl). 2014; 127:272-8).

[0013] In contrast to the elevated expression of CDH3 in tumor tissues, its expression in normal tissues is low or undetectable (Milic et al., Imai et al, Taniuchi et al, Rocha et al, Dasgupta et al, Han et al, Patel., Kim et al, Sugiyma et al, Jarzab et al., human protein atlas). Taken together, this provides evidence that CDH3 is a suitable target for the proposed dual-specificity antibody approach.

[0014] Various antibody bindings to P-cadherin have been described in the following published patents and have been most thoroughly characterized: WO9919477, WO02 / 097395, WO04 / 110345, WO06 / 114704, WO07 / 102525, WO2010 / 001585, WO2010126137; WO2010054007, WO2011056997, WO2011080796, WO2011071541.

[0015] Some antibodies known in the art block the function of CDH3, e.g., the anti-CDH3 antibody PF-03732010, by interfering with its adhesion properties, for example (Zhang CC et al. Clin Cancer Res. 2010; 16:5177-88). In vitro, this antibody causes the destruction of 3D spheres due to changes in intracellular signaling, such as the dissociation of β-catenin, at nanomolar concentrations, but does not inhibit proliferation. Long-term survival, as well as in vivo inhibition of tumor growth and metastasis formation, was observed in preclinical models at a dose of 10 mg / kg and in a manner dependent on the expression level of CDH3. By mechanism, it was possible to rule out the possibility that the in vivo antitumor effect is further mediated by the antibody-dependent cytotoxicity (ADCC) process of cells, which entails apoptosis caused by the binding of cytotoxic T cells to the Fc domain of immunoglobulin. This mechanism of action has been attributed to some patented anti-CDH3 antibodies. Blocking P-cadherin activity has been described as an approach to inhibit the growth of P-cadherin-expressing tumors, but the results achieved frequently remain unsatisfactory because they delay tumor growth or, at best, induce tumor retention, but do not eliminate the tumor.

[0016] To enhance their inhibitory effects on tumor growth, antibodies are conjugated to cytotoxic agents or cytostatic agents (e.g., chemotherapy drugs, toxins, radioisotopes, etc.) to form antibody-drug conjugates (ADCs), or more broadly, immunoconjugates. Immunoconjugates allow for the targeted delivery of drug moiety to tumors and their intracellular accumulation. The efficacy of these immunoconjugates for tumor cell death strongly depends on various parameters, such as the mode of action of each immunoconjugate, as well as the internalization behavior of target molecules, which are typically located on the cell membrane and possess extracellular domains. Overall, binding to immunoconjugates has been shown to significantly enhance the antitumor efficacy of various antibodies.

[0017] More recently, the binding of bispecific molecules to T cells as well as target molecules has yielded promising results that avoid many of the aforementioned problems. These drug molecules do not rely on the complete blockade of target function or the proliferative status of target cells, because they utilize the highly effective natural killer effect of T cells. An example of such a bispecific molecule is a bispecific anti-target x anti-CD3 short-chain antibody construct previously shown to mediate T cell-associated death in target cells with very high efficiency (see, e.g., blinatumomab).

[0018] To identify bispecific scFv molecules that are promising drug substances capable of ultimately curing patients, various distinct criteria must be met. The combination and strictness of these criteria make the occurrence of molecules that satisfy them an extremely rare and unpredictable event. The most obvious criterion is the efficacy of the drug candidate.

[0019] The efficiency of these molecules depends not only on the target binding domain but also on the T cell recruitment portion of the bispecific molecule. In particular, regarding the target binding portion of the bispecific scFv molecule, it is unpredictable because it varies very strongly depending on the characteristics of the target molecule, including its tertiary as well as quaternary structure. Among the efficiency parameters, the binding kinetics between the standard and the bispecific binding molecule, as well as the precise binding region called the binding epitope, play an important role.

[0020] Anti-CDH3 monoclonal antibodies (which is generally true for any other monoclonal antibody) act by the highly specific recognition of their target molecules. They recognize only a single site or epitope on their target CDH3 molecule. Additionally, many antibodies have been found to exert their functions in a species-specific manner. However, this species-specificity, which is unique not only to CDH3 monoclonal antibodies (and their fragments) but also to monoclonal antibodies in general, poses a significant obstacle to their development as therapeutic agents for the treatment of human diseases. To obtain market approval, any new drug candidate must pass rigorous testing. Testing is further divided into preclinical and clinical phases, with the latter conducted in human patients and the former in animals. The goal of preclinical testing is to demonstrate that the drug candidate possesses the intended activity and, most importantly, is safe. Only when safety in animals and the potential efficacy of the drug candidate are established in preclinical testing will the drug candidate be likely to be approved for clinical trials in humans by the respective regulatory agencies. Drug candidates may be tested for safety in animals in the following three ways: (i) in appropriate species, i.e., species capable of recognizing the ortholog antigen; (ii) in genetically modified animals containing human antigens; and (iii) by using surrogates for the drug candidate capable of binding to the ortholog antigen present in the animal. The limitation of genetically modified animals is that this technique is typically limited to rodents. There are significant physiological differences between rodents and humans, and safety results cannot be easily inferred for humans. The limitation of surrogates for drug candidates is the differing composition of facts compared to the actual drug candidate, and the animals often used are rodents with the limitations discussed above. Therefore, preclinical data generated in rodents have predictive power for drug candidates.

[0021] The approach of choice for safety testing is the use of chimpanzees due to their appropriate species, preferably primates, and genetic similarity. However, chimpanzees are considered an endangered species, and due to their human-like characteristics, the use of these animals for drug safety testing is banned in Europe and highly restricted elsewhere.

[0022] The T-cell-engaging bispecific short-chain antibodies described in this technology have great therapeutic potential for the treatment of malignant tumor diseases; however, most of these bispecific molecules are limited in that they are species-specific and recognize only human antigens, possibly primates, i.e., macaque counterparts. Furthermore, most of the aforementioned bispecific molecules are additionally limited in that, although they can recognize human and primate homologs, they cannot perform their intended functions across species boundaries, for example, to prevent the exertion of T-cell-mediated cytotoxicity.

[0023] Since there is still a need to have additional available options for the treatment of various types of cancer disclosed herein, means and methods for solving this problem are provided herein in the form of a bispecific antibody construct in which one binding domain is directed toward CDH3 and a second binding domain is directed toward CD3 on T cells.

[0024] The present invention provides a bispecific antibody comprising, in a first aspect, a first preferably human binding domain that binds to an epitope cluster of human CDH3 on the surface of a target cell, and a second preferably human binding domain that binds to human CD3 on the surface of a T cell, wherein the epitope cluster of human CDH3 is contained within amino acid positions 291 to 363 (SEQ ID NO. 36).

[0025] In a preferred embodiment of the present invention, the bispecific antibody construct is characterized in that the first binding domain also binds to macaque CDH3, preferably Philippine monkey CDH3.

[0026] This species-specific functionality implies that the same molecule can be used in preclinical studies as well as in clinical studies in humans. This results in highly similar outcomes in animal studies and significantly increased predictive power compared to species-specific surrogate molecules. Since the CD3 and CDH3 binding domains of the CDH3xCD3 bispecific antibody construct of the present invention are both cross-species specific and functional—that is, reactive to human and macaque antigens that exert similar T-cell-mediated cytotoxic effects—these binding domains can be used for preclinical evaluation of safety, activity, and / or pharmacokinetic profiles in primates and as a drug in humans (in the same form). In a preferred embodiment, it will be understood that the cross-species specificity of the first binding domain and the second binding domain of the antibody construct of the present invention is identical.

[0027] Considering the above, the need to construct a surrogate CDH3xCD3 bispecific antibody construct for testing in phylogenetically distant species is eliminated. As a result, the same molecule intended for administration to humans in clinical trials, as well as in subsequent market approvals and therapeutic drug administrations, can be used in animal clinical trials. The ability to use the same molecule for preclinical animal trials, as in subsequent administration to humans, effectively eliminates, or at least significantly reduces, the risk that data obtained from preclinical animal trials may limit applicability to human cases. In short, obtaining preclinical safety data in animals using the same molecule actually administered to humans greatly guarantees the applicability of the data to human-related scenarios. In contrast, in the conventional approach of using surrogate molecules, said surrogate molecule must be molecularly compatible with the animal test system used for preclinical safety evaluation. Therefore, the molecule to be used in human therapy may actually differ in sequence and may also differ structurally from the surrogate molecule used in preclinical trials in pharmacokinetic parameters and / or biological activity, and as a result, data obtained from preclinical animal trials have limited applicability to human cases. The use of a surrogate molecule requires the structure, generation, purification, and characterization of a completely new construct. This results in additional development costs and time required to obtain the corresponding molecule. Taken together, the surrogate molecule must be developed separately from the actual drug to be used in human therapy, and thus two lines of development must be carried out for the two molecules. Therefore, the main advantage of the present invention, preferably human, CDH3xCD3 bispecific antibody construct exhibiting cross-species specificity and functionality (i.e., responsiveness) as described herein, is that the same molecule used for therapeutic agents in humans and in preclinical animal trials can be used.

[0028] With the cross-species specific CDH3xCD3 bispecific antibody construct of the present invention, there is no longer a need to adapt test animals, such as genetically modified animals, for drug candidates intended for administration to humans. The cross-species specific and reactive CDH3xCD3 bispecific antibody construct of the present invention, which exhibits cross-species specificity and reactivity according to the use and method of the present invention, can be used directly for preclinical testing in non-chimpanzee primates, such as macaques, without any genetic manipulation of the animals. As is well known to those skilled in the art, the approach of adapting test animals for drug candidates always carries the risk that results obtained in preclinical safety tests are less representative and predictive due to variations in the animals. For example, in genetically modified animals, proteins encoded by the transplanted gene are often highly overexpressed. Therefore, data obtained regarding the biological activity of antibodies against these protein antigens may be limited to their predictive values ​​for humans, where the protein is expressed at much lower and more physiological levels.

[0029] An additional advantage of the preferably human, CDH3xCD3 bispecific antibody construct of the present invention is the ability to extract multiple blood samples when used, for example, as part of animal preclinical testing during pharmacokinetic animal studies. Multiple blood extractions can be obtained much more easily from lower animals, such as non-chimpanzee primates, than from mice. Extraction of multiple blood samples allows for the sequential testing of blood parameters to determine the biological effects induced by the CDH3xCD3 bispecific antibody construct of the present invention. Furthermore, extraction of multiple blood samples enables researchers, as set forth herein, to evaluate the pharmacokinetic profile of the preferably human, CDH3xCD3 bispecific antibody construct of the present invention. Additionally, potential side effects that may be induced by the CDH3xCD3 bispecific antibody construct of the present invention, as reflected in blood parameters, can be measured in different blood samples extracted during the course of administration of said antibody.

[0030] The advantages of the CDH3xCD3 dual-specificity antibody construct of the present invention, as described herein and exhibiting cross-species specificity, can be briefly summarized as follows:

[0031] Initially, the CDH3xCD3 bispecific antibody construct of the present invention, as set forth herein, used in preclinical trials is identical to that used in human therapy. Therefore, there is no longer a need to develop two independent molecules that may differ in pharmacokinetic properties and biological activity. This is advantageous, for example, in that pharmacokinetic results are more directly deliverable and applicable to a human setting than, for example, through conventional surrogate approaches.

[0032] Secondly, the use of the CDH3xCD3 bispecific antibody construct of the present invention, as described herein for the manufacture of a treatment in humans, is less expensive and less labor-intensive than surrogate approaches.

[0033] Thirdly, the CDH3xCD3 dual-specificity antibody construct of the present invention as described herein can limit the risk of potential species differences between primates and humans by being used for preclinical testing in a series of different primate species as well as in a single primate species.

[0034] Fourth, chimpanzees as a species at risk for animal testing can be avoided if desired.

[0035] Fifth, multiple blood samples can be performed for extensive pharmacokinetic studies.

[0036] Sixth, due to the human origin of the antibody composition according to a preferred embodiment of the present invention, the generation of an immune response to the binding molecule is minimized when administered to a human patient. The induction of an immune response by an antibody specific to the drug candidate is carried out by a mouse from a non-human species, for example, one that causes the generation of a human-anti-mouse antibody (HAMA) against a therapeutic molecule of murine origin.

[0037] The therapeutic use of the CDH3xCD3 bispecific antibody construct of the present invention provides a novel and inventive therapeutic approach for cancer, preferably solid tumors, more preferably carcinomas, and other cancer indications listed below. As illustrated in the following examples, the CDH3xCD3 bispecific antibody construct of the present invention provides an advantageous tool for killing CDH3-expressing human cancer cells. Furthermore, the cytotoxic activity of the CDH3xCD3 bispecific antibody construct of the present invention provides an advantageous mode of action compared to classical (single-specific) IgG molecules or ADCs.

[0038] To characterize drug candidates in animal models, cross-species reactivity to appropriate animal species is essential, and the inter-species affinity gap must be maintained within a 10-fold affinity difference to ensure a predictive animal model setup. Macaque monkeys (and particularly cynomolgus) are considered the most suitable species for efficacy and toxicity testing, possessing the highest predictive values. To analyze whether antibody constructs meet efficiency criteria, each cytotoxicity assay must be developed and quantified.

[0039] To prevent harmful side effects, it must be ensured that the target binding portion of the bispecific antibody construct specifically binds to P. cadherin and does not specifically bind to its closest homologs described above or any other proteins present in the body.

[0040] The antibody compositions of the present invention preferably do not bind to CHD3 extracellular domain D1 (positions 108 to 215 of SEQ ID NO. 1), they preferably do not bind to CHD3 extracellular domain D4 (positions 441 to 546 of SEQ ID NO. 1), and they preferably do not bind to CHD3 extracellular domain D5 (positions 547 to 650 of SEQ ID NO. 1).

[0041] According to the most preferred embodiment of the present invention, the bispecific antibody composition is characterized in that the first binding domain binds to an epitope contained within amino acid positions 291 to 327 of human CDH3 (SEQ ID NO. 34). The antibody that specifically binds to amino acid positions 291 to 327 (SEQ ID NO. 34) preferably does not bind to the extracellular domain D3 of CHD3 of human CDH3 (positions 328 to 440 of SEQ ID NO. 1).

[0042] A preferred antibody construct according to the present invention may also be represented as a bispecific antibody construct comprising a first preferably human binding domain that binds to an epitope of human CDH3 on the surface of a target cell and a second preferably human binding domain that binds to human CD3 on the surface of a T cell, wherein the antibody construct binds to the same epitope as an antibody named CDH3-11, CDH3-12, CDH3-13, or CDH3-14, i.e., an antibody comprising the following, or competes for binding to CDH3 with the antibody named as above:

[0043] The VH region as presented in SEQ ID NO. 155 and the VL region as presented in SEQ ID NO. 156,

[0044] The VH region as presented in SEQ ID NO. 165 and the VL region as presented in SEQ ID NO. 166,

[0045] A VH region as presented in SEQ ID NO. 175 and a VL region as presented in SEQ ID NO. 176; or

[0046] VH region as presented in SEQ ID NO. 185 and VL region as presented in SEQ ID NO. 186.

[0047] In another aspect of the present invention, a bispecific antibody construct is characterized in that the first binding domain binds to an epitope contained within amino acid positions 328 to 363 of human CDH3 (SEQ ID NO. 35).

[0048] A preferred antibody composition of the present invention may also be represented as a bispecific antibody composition comprising a first preferably human binding domain that binds to an epitope of human CDH3 on the surface of a target cell and a second preferably human binding domain that binds to human CD3 on the surface of a T cell, wherein the antibody composition binds to the same epitope as an antibody named CDH3-24, i.e., an antibody comprising a VH region as shown in SEQ ID NO. 285 and a VL region as shown in SEQ ID NO. 286, or competes with said antibody for binding to CDH3.

[0049] Another preferred antibody construct according to the present invention may also be represented as a bispecific antibody construct comprising a first preferably human binding domain that binds to an epitope of human CDH3 on the surface of a target cell and a second preferably human binding domain that binds to human CD3 on the surface of a T cell, wherein the antibody construct binds to the same epitope as an antibody named CDH3-25, CDH3-26, or CDH3-27, i.e., an antibody comprising the following, or competes for binding to CDH3 with the antibody named as above:

[0050] VH region as depicted in SEQ ID NO. 295 and VL region as depicted in SEQ ID NO. 296,

[0051] The VH region as depicted in SEQ ID NO. 305 and the VL region as depicted in SEQ ID NO. 306, or

[0052] VH region as depicted in SEQ ID NO. 315 and VL region as depicted in SEQ ID NO. 316.

[0053] Whether or not an antibody construct competes with another given antibody construct can be measured in a competitive assay, such as a competitive ELISA or a cell-based competitive assay. Avidin-binding microparticles (beads) may also be used. Similar to avidin-coated ELISA plates, when reacted with biotinylated proteins, each of these beads can be used as a substrate on which the assay can be performed. The antigen is coated onto the beads and then pre-coated with a first antibody. A second antibody is added, and any additional binding is determined. Readings are performed via flow cytometry.

[0054] In another embodiment of the present invention, the bispecific antibody construct is characterized in that the first binding domain binds to an epitope contained within amino acid positions 328 to 363 of human CDH3 (SEQ ID NO. 35) and to an epitope contained within amino acid positions 404 to 440 of human CDH3 (SEQ ID NO. 390).

[0055] Furthermore, the antibody composition of the present invention preferably does not bind to an epitope contained within amino acid positions 216 to 252 or 253 to 290 of human CDH3 as presented in SEQ ID NO. 1.

[0056] Furthermore, the antibody composition of the present invention preferably does not bind to an epitope contained within amino acid positions 364 to 403 of human CDH3 as shown in SEQ ID NO. 1.

[0057] Antibody compositions that specifically bind to epitopes contained within amino acid positions 328 to 363 (SEQ No. 35) preferably do not bind to the extracellular domain D2 of human CDH3 CHD3 (positions 216 to 327 of SEQ No. 1).

[0058] One advantage of the present invention is to provide a bispecific antibody construct comprising a binding domain that binds to CD3 and a binding domain capable of binding to CDH3, wherein the binding domain exhibits cross-species specificity for human and macaque CDH3. Unexpectedly, it was discovered that the CDH3xCD3 bispecific antibody construct of the present invention not only specifically binds to human and macaque CDH3 homologs of CDH3 and CD3, but also exerts T cell-mediated cytotoxicity in human and macaque CDH3 assay systems. Advantageously, the present invention provides a CDH3xCD3 bispecific antibody construct that exhibits T cell-mediated cytotoxicity in humans and macaques. This advantage is achieved by a CDH3xCD3 bispecific antibody construct that binds to an epitope cluster contained within amino acid positions 291 to 363 (SEQ ID NO. 36) of human CDH3.

[0059] It should be noted that the singular form as used herein includes a plurality of objects unless otherwise clearly indicated in the context. Accordingly, for example, a reference to "reagent" includes one or more of these different reagents, and a reference to "method" includes a reference to an equivalent step and method known to those skilled in the art that is a modification or substitution of the method described herein.

[0060] Unless otherwise indicated, the term “at least” preceding a series of components should be understood to refer to all elements in such series. Those skilled in the art may recognize or identify numerous equivalents to the specific embodiments of the invention described herein, or merely through routine experimentation. Such equivalents are intended to be included by the invention.

[0061] The term "and / or" as used anywhere in this specification includes the meanings of "and," "or," and "any or any other combination of elements connected by said term."

[0062] As used herein, the terms "about" or "approximately" mean within ±20%, preferably within ±15%, more preferably within ±10%, and most preferably within ±5% of a given value or range.

[0063] Throughout this specification and the following claims, unless otherwise required by the context, the word “comprise” and its variations, e.g., “comprises” and “comprising” shall be understood to indicate the inclusion of any mentioned integer or step or group of integers, but not to exclude any other integer or step or group of integers or steps. As used herein, the term “comprising” may occasionally be replaced by the terms “containing” or “including” when used in conjunction with the term “having”.

[0064] As used herein, "consisted of" excludes any component, step, or element not embodied in the claim components. As used herein, "essentially constituted of" does not exclude any material or step that does not substantially affect the basic and novel features of the claim.

[0065] In each example of this specification, any of the terms “comprising,” “essentially made,” and “made” may be replaced with one of the other two terms.

[0066] The term "antibody construct" refers to a molecule whose structure and / or function is based on the structure and / or function of an antibody, for example, a full-length or whole immunoglobulin molecule. Thus, the antibody construct can bind to its specific target or antigen. Furthermore, the antibody construct according to the present invention comprises a minimum structural requirement of the antibody that allows for target binding. This minimum requirement may be indicated, for example, preferably by the presence of at least three light chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VL region) and / or three heavy chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VH region) of all six CDRs. Antibodies based on the construct according to the present invention include, for example, monoclonal, recombinant, chimeric, deiminized, humanized, and human antibodies.

[0067] Full-length or whole antibodies, including camelid antibodies and other immunoglobulin antibodies produced by biotechnological or protein manipulation methods or processes, fall within the definition of "antibody compositions" according to the present invention. These full-length antibodies may be, for example, monoclonal, recombinant, chimeric, deimimized, humanized, and human antibodies. Additionally, fragments of full-length antibodies, such as VH, VHH, VL, (s)dAb, Fv, Fd, Fab, Fab', F(ab')2, or "r IgG" ("half-antibody"), fall within the definition of "antibody compositions." The antibody construct according to the present invention may also be a modified fragment of an antibody, also called an antibody variant, such as scFv, di-scFv or bi(s)-scFv, scFv-Fc, scFv-zipper, scFab, Fab2, Fab3, a diabody, a single-strand diabody, a tandem diabody (Tandab's), a tandem di-scFv, a tandem tri-scFv, or a "minibody" exemplified by the following structures: (VH-VL-CH3)2, (scFv-CH3)2, ((scFv)2-CH3 + CH3), ((scFv)2-CH3) or (scFv-CH3-scFv)2, a multibody, such as a triabody or tetrabody, and a single-domain antibody, such as VHH, VH or which specifically binds to an antigen or epitope independently of other V regions or domains. A nanobody or a single variable domain antibody containing only one variable domain, which may be VL.

[0068] Furthermore, the definition of the term "antibody construct" includes monovalent, divalent, and multivalent / multivalent constructs, and thus includes not only monospecific constructs that specifically bind to only one antigenic structure, but also bispecific and multispecific / multispecific constructs that specifically bind to more than one antigenic structure, e.g., two, three or more, through distinct binding domains. In addition, the definition of the term "antibody construct" includes not only molecules consisting of only one polypeptide chain, but also molecules consisting of more than one polypeptide chain, wherein these chains may be identical (homodimer, homotrimer, or homooligomer) or different (heterodimer, heterotrimer, or heterooligomer). Examples of the antibodies and variants identified above, as well as their variants or derivatives, are described in particular in the literature [Harlow and Lane, Antibodies a laboratory manual, CSHL Press (1988) and Using Antibodies: a laboratory manual, CSHL Press (1999), Kontermann and Dubel, Antibody Engineering, Springer, 2nd ed. 2010 and Little, Recombinant Antibodies for Immunotherapy, Cambridge University Press 2009].

[0069] The antibody constructs of the present invention are preferably "antibody constructs generated in vitro." This term refers to antibody constructs according to the above definition in which all or part of the variable region (e.g., at least one CDR) is generated from non-immune cell selection, e.g., in vitro phage displays, protein chips, or any other method in which candidate sequences can be tested for their ability to bind to an antigen. Accordingly, this term preferably excludes sequences generated solely by genomic rearrangement within immune cells in animals. "Recombinant antibody" is an antibody produced through the use of recombinant DNA techniques or genetic engineering.

[0070] As used herein, the terms “monoclonal antibody” (mAb) or “monoclonal antibody composition” refer to antibodies obtained from substantially homologous antibodies, that is, that individual antibodies comprising a population are identical except for possible naturally occurring mutations and / or post-translational modifications (e.g., isomerization, amideation) that may be present in small amounts. Monoclonal antibodies are directed toward a single antigenic site or a determinant on an antigen, in contrast to conventional (polyclonal) antibody preparations which typically contain different antibodies directed toward different determinants (or epitopes), and are highly specific. In addition to their specificity, monoclonal antibodies are advantageous in that they are synthesized by hybridoma cultures and are therefore not contaminated by other immunoglobulins. The variant term “monoclonal” indicates the characteristic of antibodies obtained from a substantially homologous population of antibodies and should not be interpreted as requiring antibody production by any specific method.

[0071] For the production of monoclonal antibodies, any technique providing antibodies produced by continuous cell line cultures may be used. For example, the monoclonal antibody to be used is [Koehler et al.This can be accomplished by a hybridoma method first described in [Nature, 256: 495 (1975)] or by a recombinant DNA method (see, e.g., U.S. Patent No. 4,816,567). Examples of additional techniques for generating human monoclonal antibodies include the trioma technique, the human B-cell hybridoma technique (Kozbor, Immunology Today 4 (1983), 72), and the EBV-hybridoma technique (Cole et al., Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, Inc. (1985), 77-96).

[0072] Subsequently, hybridomas are screened using standard methods, such as enzyme-linked immunosorbent assay (ELISA) and surface plasmon resonance (BIACORE (trademark)) analysis, to identify one or more hybridomas that produce antibodies specifically binding to a specified antigen. Any form of the appropriate antigen may be used as an immunogen, e.g., recombinant antigen, naturally derived form, any variant or fragment thereof, as well as its antigenic peptide. Surface plasmon resonance used in the BIACORE system may be used to increase the efficiency of phage antibodies binding to epitopes of target antigens, such as CDH3 or CD3 epsilon (Schier, Human Antibodies Hybridomas 7 (1996), 97-105; Malmborg, J. Immunol. Methods 183 (1995), 7-13).

[0073] Another exemplary method for preparing monoclonal antibodies comprises the step of screening a protein expression library, e.g., a phage display or lysosome display library. Phage displays are, e.g., U.S. Patent No. 5,223,409 of Ladner et al.; literature [Smith (1985) Science 228:1315-1317, Clackson et al. , Nature, 352: 624-628 (1991) and Marks et al. It is described in J. Mol. Biol., 222: 581-597 (1991).

[0074] In addition to the use of the display library, suitable antigens may be used to immunize non-human animals, e.g., rodents (e.g., mice, hamsters, rabbits, or rats). In one embodiment, the non-human animal comprises at least a portion of the human immunoglobulin gene. For example, mouse strain deficiency may be engineered to generate mouse antibodies possessing a large fragment of the human Ig (immunoglobulin) locus. Using hybridoma techniques, antigen-specific monoclonal antibodies derived from genes possessing the desired specificity may be generated and selected. For example, XENOMOUSE (trademark), literature [Green et al.

[1994] Nature Genetics 7:13-21, U.S. Patent No. 2003-0070185, WO 96 / 34096, and WO 96 / 33735.

[0075] Monoclonal antibodies can also be obtained from non-human animals and subsequently modified, e.g., humanized, deiminized, or chimeric, using recombinant DNA techniques known in the art. Examples of modified antibody constructs include humanized variants of non-human antibodies and "affinity-mature" antibodies (e.g., Hawkins et al. J. Mol. Biol. 254, 889-896 (1992) and Lowman et al[See Biochemistry 30, 10832-10837 (1991)]) and antibody mutants having modified effector function(s) (e.g., U.S. Patent No. 5,648,260, Kontermann and Dubel (2010), loc . cit . and Little (2009), loc . cit Includes .)

[0076] In immunology, affinity maturation is the process by which B cells produce antibodies with increased affinity for an antigen during the course of an immune response. Through repeated exposure to the same antigen, the host will successfully produce antibodies with greater affinity. As with natural phenotypes, in vitro affinity maturation is based on the principles of mutation and selection. In vitro affinity maturation has been successfully used to optimize antibodies, antibody constructs, and antibody fragments. Random maturation within the CDR is introduced using radiation, chemical mutagens, or error-induced PCR. Additionally, genetic diversity can be increased by chain shuffling. Two or three rounds of maturation and selection using display methods, such as phage display, typically result in antibody fragments with affinities in the low nanomolar range.

[0077] A preferred type of amino acid substitution variant of the antibody construct involves substituting one or more hypervariable domain residues of the parent antibody (e.g., humanized or human antibody). Generally, the obtained variant(s) selected for further generation will have improved biological properties compared to the parent antibody from which they are generated. A convenient method for generating these substituent variants involves affinity maturation using phage display. Briefly, several hypervariable domain sites (e.g., 6 to 7 sites) are mutated to generate all possible amino acid substitutions at each site. The antibody variants thus generated are presented in a monovalent manner from the silicate phage particles as fusions to the gene III product of M13 packaged within each particle. Subsequently, the phage-displayed variants are screened for their biological activities (e.g., binding affinity) as disclosed herein. To identify candidate hypervariable domain sites for modification, alanine scanning mutagenesis may be performed to identify hypervariable domain residues that significantly contribute to antigen binding. Alternatively, or additionally, it may be advantageous to analyze the crystal structure of the antigen-antibody complex to identify invariant sites between the binding domain and, for example, human CDH3. These contact residues and neighboring residues are candidates for substitution according to the techniques described herein. Once such variants are generated, a panel of variants is screened as described herein, and antibodies exhibiting superior characteristics in one or more appropriate analyses may be selected for further study.

[0078] The monoclonal antibodies and antibody compositions of the present invention specifically comprise “chimeric” antibodies (immunoglobulins) in which a portion of the heavy chain and / or light chain is identical or homologous to a corresponding sequence in an antibody derived from a specific species or belongs to a specific antibody class or subclass, whereas, insofar as they exhibit the intended biological activity, the remainder of the chain(s) is identical or homologous to a corresponding sequence in an antibody derived from another species or belongs to a fragment of such antibody as well as to a different antibody class or subclass (U.S. Patent No. 4,816,567; Morrison et al., Proc. Natl. Acad. Sci. USA, 81: 6851-6855 (1984)). The chimeric antibodies of interest herein comprise “primate-ized” antibodies comprising variable domain antigen-binding sequences derived from non-human primates (e.g., Old World monkeys, apes, etc.) and human invariant domain sequences. Various approaches for preparing chimeric antibodies have been described. For example, literature [Morrison et al. , Proc. Natl. Acad. ScL USA 81:6851, 1985; Takeda et al. [Nature 314:452, 1985], U.S. Patent No. 4,816,567 by Cabilly et al.; U.S. Patent No. 4,816,397 by Boss et al.; European Patent No. 0171496 by Tanaguchi et al.; European Patent No. 0173494; and British Patent No. 2177096.

[0079] Antibodies, antibody constructs, or antibody fragments may also be modified by specific deletions of human T cell epitopes (a method referred to as “deimmunization”) by methods disclosed, for example, in WO 98 / 52976 or WO 00 / 34317. Briefly, the heavy and light chain variable domains of antibodies may be analyzed for peptides that bind to MHC class II; these peptides exhibit potential T cell epitopes (as shown in WO 98 / 52976 and WO 00 / 34317). For the detection of potential T cell epitopes, a computer modeling approach referred to as “peptide threading” may be applied, and additionally, a database of human MHC class Il binding peptides may be searched for motifs present in VH and VL sequences as described in WO 98 / 52976 and WO 00 / 34317. These motifs bind to any 18 major MHC class Il DR homologous types and thus constitute potential T cell epitopes. Detected potential T cell epitopes can be eliminated by substituting a small number of amino acid residues within the variable domain, or preferably by a single amino acid substitution. Typically, conservative substitutions are performed. Often, but not exclusively, conventional amino acids may be used for positions within human germline antibody sequences. Human germline sequences are, for example, the literature [Tomlinson, et al. (1992) J. MoI. Biol. 227:776-798; Cook, G.P. et al. (1995) Immunol. Today Vol. 16 (5): 237-242; and Tomlinson et al. (1995) EMBO J. 14: 14:4628-4638] are disclosed. The V BASE directory provides a comprehensive directory of human immunoglobulin variable region sequences (reference [Tomlinson, LA. et al.(Edited by MRC Centre for Protein Engineering, Cambridge, UK). These sequences may be used as sources of human sequences, for example, for framework regions and CDRs. Common human framework regions may also be used, for example, as described in U.S. Patent No. 6,300,064.

[0080] "Humanized" antibodies, antibody constructs, or fragments thereof (e.g., Fv, Fab, Fab', F(ab')2, or other antigen-binding subsequences of antibodies) are antibodies or immunoglobulins of most human sequences containing minimal sequence(s) derived from non-human immunoglobulins. In most cases, the humanized antibody is a human immunoglobulin (recipient antibody) in which residues from the recipient's hypervariable region (CDR) are replaced by residues from the hypervariable region of a non-human (e.g., rodent) species (donor antibody), such as mouse, rat, hamster, or rabbit, having the desired specificity, affinity, and ability. In some examples, residues from the Fv framework region (FR) of the human immunoglobulin are replaced by corresponding non-human residues. Furthermore, "humanized antibodies" as used herein may also include residues that are not expressed in either the recipient antibody or the donor antibody. These modifications are made to further purify and optimize antibody performance. Humanized antibodies may also include at least a portion of the immunoglobulin constant region (Fc), typically of human immunoglobulin. For further detailed description, see the literature [Jones et al. , Nature, 321: 522-525 (1986); Reichmann et al. See , Nature, 332: 323-329 (1988); and Presta, Curr. Op. Struct. Biol., 2: 593-596 (1992)].

[0081] Humanized antibodies or fragments thereof can be generated by replacing the sequence of an Fv variable domain that is not directly involved in antigen binding with a sequence equivalent to the human Fv variable domain. An exemplary method for generating humanized antibodies or fragments thereof is [Morrison (1985) Science 229:1202-1207; Oi et al.

[1986] BioTechniques 4:214; and provided by U.S. Patent No. 5,585,089; U.S. Patent No. 5,693,761; U.S. Patent No. 5,693,762; U.S. Patent No. 5,859,205; and U.S. Patent No. 6,407,213. The method comprises the steps of isolating a nucleic acid sequence encoding all or part of an immunoglobulin Fv variable domain from at least one of a heavy chain or a light chain, manipulating it, and expressing it. Such nucleic acid may be obtained from other sources as well as from a hybridoma that generates an antibody against a predetermined target as described above. Recombinant DNA encoding a humanized antibody molecule may then be cloned into a suitable expression vector.

[0082] Humanized antibodies may also be produced using genetically modified animals, such as mice, that express human heavy and light chain genes but cannot express endogenous mouse immunoglobulin heavy and light chain genes. Winter describes an exemplary CDR conjugation method that may be used to produce the humanized antibodies described in the specification (U.S. Patent No. 5,225,539). All CDRs of a particular human antibody may be replaced with at least some of non-human CDRs, or only some of the CDRs may be replaced with non-human CDRs. This requires only replacing the number of CDRs required for the binding of the humanized antibody with a predetermined antigen.

[0083] Humanized antibodies can be optimized by introducing conservative substitutions, co-sequence substitutions, germline substitutions, and / or reverse mutations. These modified immunoglobulin molecules can be achieved by any of the techniques known in the art (e.g., the literature [Teng et al. , Proc. Natl. Acad. Sci. USA, 80: 7308-7312, 1983; Kozbor et al ., Immunology Today, 4: 7279, 1983; Olsson et al. [ , Meth. Enzymol., 92: 3-16, 1982] and European Patent No. 239 400).

[0084] The terms "human antibody," "human antibody construct," and "human binding domain" refer to antibodies, antibody constructs, and binding domains having antibody regions, e.g., variable and constant regions, or, e.g., the literature [Kabat et al. (1991) ( loc . cit It comprises a domain substantially corresponding to a human germline immunoglobulin sequence known in the art, including that described in []. The human antibody, antibody construct, or binding domain of the present invention may comprise amino acid residues that are not encoded by the human germline immunoglobulin sequence (e.g., mutations introduced by in vitro random or site-specific mutagenesis or by in vivo somatic mutation) in, for example, in CDR, and particularly in CDR3. The human antibody, antibody construct, or binding domain may have at least one, two, three, four, or five or more positions replaced by amino acid residues that are not encoded by the human germline immunoglobulin sequence. The definitions of the human antibody, antibody construct, and binding domain as used herein also consider a fully human antibody comprising only non-artificially and / or genetically modified human sequence of the antibody, which may be induced by using techniques or systems such as a xenomouse.

[0085] In some embodiments, the antibody construct of the present invention is an “isolated” or “substantially pure” antibody construct. When used to describe the antibody construct disclosed herein, “isolated” or “substantially pure” means an antibody construct isolated, separated, and / or recovered from components of its production environment. Preferably, the antibody construct has no or substantially no association with any other components from its production environment. Contaminating components of its production environment, such as those resulting from recombinant transfected cells, are typically substances that do not interfere with the diagnostic or therapeutic use of polypeptides and may include enzymes, hormones, and other proteinaceous or non-proteinaceous solutes. The antibody construct may constitute, for example, at least about 5% by weight, or at least about 50% by weight, of the total protein in a given sample. It is understood that the isolated protein may, depending on the circumstances, constitute 5% to 99.9% by weight of the total protein content. Polypeptides may be produced at significantly higher concentrations through the use of inducible promoters or high-expression promoters so that they are produced at increased concentration levels. The definition includes the production of antibody constructs in a wide variety of organisms and / or host cells known in the art. In a preferred embodiment, the antibody construct may be purified for homogeneity by (1) using a spinning cup sequenator to a degree sufficient to obtain at least 15 residues of the N-terminal or internal amino acid sequence, or by (2) SDS-PAGE under non-reducing or reducing conditions using Coomassie blue or, preferably, silver staining. However, usually, the isolated antibody construct will be prepared by at least one purification step.

[0086] The term "binding domain" in relation to the present invention characterizes a domain that (specifically) binds to, interacts with, and recognizes a given target epitope or a given target site on the target molecules (antigens) CDH3 and CD3, respectively. The structure and function of the first binding domain (recognizing CDH3), and preferably also the structure and / or function of the second binding domain (CD3), are based, for example, on the structure and / or function of the full-length or whole immunoglobulin molecule. According to the present invention, the first binding domain is characterized by the presence of three light chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VL region) and three heavy chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VH region). The second binding domain preferably also includes the minimum structural requirements of the antibody that allow target binding. More preferably, the second binding domain comprises at least three light chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VL region) and / or three heavy chain CDRs (i.e., CDR1, CDR2, and CDR3 in the VH region). It is thought that the first binding domain and / or the second binding domain may be generated or obtained by a phage display or library screening method rather than by conjugating CDR sequences from a pre-existing (monoclonal) antibody into a scaffold.

[0087] According to the present invention, the binding domain is preferably in the form of a polypeptide. Such polypeptides may comprise a protein portion and a non-protein portion (e.g., a chemical linker or chemical crosslinking agent, e.g., glutaraldehyde). Proteins (fragments thereof having fewer than 30 amino acids, preferably biologically active fragments and peptides) comprise two or more amino acids joined together by covalent peptide bonds (resulting in an amino acid chain). As used herein, the term “polypeptide” describes a group of molecules composed of more than 30 amino acids. Polypeptides may further form multimers, e.g., dimers, trimers, and higher-order oligomers (i.e., composed of more than one polypeptide molecule). The polypeptide molecules forming these dimers, trimers, etc. may be identical or non-identical. The corresponding higher-order structures of these multimers are consequently referred to as homo- or heterodimers, homo- or heterotrimers, etc. An example of a heteropolymer is an antibody molecule composed of two identical light-chain polypeptides and two identical heavy-chain polypeptides in their natural form. The terms “peptide,” “polypeptide,” and “protein” also refer to naturally modified peptides / polypeptides / proteins, wherein the modification is achieved by post-translational modifications such as glycosylation, acetylation, phosphorylation, etc. When referred to herein, “peptide,” “polypeptide,” or “protein” may also be chemically modified, e.g., pegylated. Such modifications are well known in the art and are described below in this specification.

[0088] As mentioned above, the binding domain may typically include an antibody light chain variable region (VL) and an antibody heavy chain variable region (VH); however, it does not include both. Fd fragments, for example, have two VH regions and often retain some antigen-binding functions of the intact antigen-binding domain. Examples of (modified) antigen-binding antibody fragments are (1) a monovalent fragment having Fab fragments, VL, VH, CL, and CH1 domains; (2) an F(ab')2 fragment (a divalent fragment having two Fab fragments connected by a disulfide bridge at the hinge region); (3) an Fd fragment having two VH and CH1 domains; (4) an Fv fragment having VL and VH domains of a single arm of the antibody; and (5) a dAb fragment having a VH domain (Ward et al., (1989) Nature 341:544-546). (6) an isolated complementarity determining region (CDR), and (7) a single-stranded Fv (scFv), the latter being preferred (e.g., derived from a scFV library). Examples of embodiments of antibody compositions according to the present invention are described in, for example, WO 00 / 006605, WO 2005 / 040220, WO 2008 / 119567, WO 2010 / 037838, WO 2013 / 026837, WO 2013 / 026833, US 2014 / 0308285, US 2014 / 0302037, WO 2014 / 144722, WO 2014 / 151910 and WO 2015 / 048272.

[0089] Preferably, the binding domain that binds to CDH3 and / or the binding domain that binds to CD3 is a human binding domain. Antibodies and antibody constructs containing at least one human binding domain avoid some of the problems associated with non-human, e.g., rodent (e.g., murine, rat, hamster, or rabbit) antibodies or antibody constructs having variable and / or constant regions. The presence of such rodent-derived proteins can cause rapid clearance of the antibody or antibody construct or cause the generation of an immune response to the antibody or antibody construct by the patient. To avoid the use of rodent-derived antibodies or antibody constructs, human or full human antibodies / antibody constructs can be generated by introducing human antibody functions into rodents, and thus the rodents produce full human antibodies.

[0090] The ability to clone and reconstruct meganucleotide-sized human loci within YACs and introduce them into mouse germlines provides a powerful approach that not only elucidates the functional components of very large or coarsely mapped loci but also generates useful models of human diseases. Furthermore, the use of these techniques for substituting mouse loci with their human equivalents has been able to provide unique insights during the study into the expression and regulation of human gene products, their communication with other systems, and their implication in the induction and progression of diseases.

[0091] A significant practical application of this strategy is the "humanization" of the mouse humoral immune system. The introduction of human immunoglobulin (Ig) loci into mice with inactivated endogenous Ig genes has provided the basis for studying their roles in B-cell development as well as their fundamentally programmed expression and antibody assembly. Furthermore, this strategy has been able to provide an ideal source for the generation of fully human monoclonal antibodies (mAbs) (a significant event in meeting the prospects for antibody therapy in human diseases). Fully human antibodies or antibody constructs are expected to minimize the immunogenicity and allergic reactions inherent to mice or mouse-derivative mAbs, thereby increasing the efficacy and safety of the administered antibodies / antibody constructs. The use of fully human antibodies or antibody constructs can be expected to provide substantial benefits in the treatment of chronic and recurrent human diseases requiring repeated compound administration, such as inflammation, autoimmunity, and cancer.

[0092] One approach to this goal was to manipulate the deficiency of mouse strains within mouse antibodies using a large fragment of a human Ig locus, based on the prediction that these mice would generate a vast repertoire of human antibodies in the absence of mouse antibodies. The large human Ig fragment would preserve not only vastly variable genetic diversity but also the appropriate regulation of antibody production and expression. By studying the mouse mechanisms for antibody diversification and selection and the lack of immunological tolerance to human proteins, the human antibody repertoire regenerated in these mouse strains should yield high-affinity antibodies against any antigen of interest, including human antigens. Using hybridoma techniques, antigen-specific human mAbs with desired specificity could be easily generated and selected. This general strategy was demonstrated in connection with the generation of the first xenomouse strain (see reference [Green et al. Nature Genetics 7:13-21 (1994)]). Xenomouse strains were engineered using yeast artificial chromosomes (YACs) containing 245kb and 190kb germline stereotype fragments of human heavy chain loci and kappa light chain loci, respectively, contained within core variable and constant region sequences. Human Ig containing YACs proved to be compatible with the mouse system for both antibody rearrangement and expression, and were capable of substituting inactivated mouse Ig genes. This was demonstrated by their ability to induce B cell development to generate an adult-like human repertoire of complete human antibodies and to produce antigen-specific human mAbs. These results also suggested that the introduction of a larger portion of human Ig loci containing a greater number of V genes, additional regulatory elements, and human Ig constant regions could encapsulate a substantially complete repertoire characteristic of human humoral responses to infection and immunization.The work of Green et al. has recently been extended to the introduction of over approximately 80% of the human antibody repertoire through the introduction of meganucleotide-sized germline stereoconfigured YAC fragments of human heavy chain loci and kappa light chain loci, respectively. Literature [Mendez. et al. See Nature Genetics 15:146-156 (1997) and U.S. Patent Application No. 08 / 759,620.

[0093] The creation of the xenomouse mouse is U.S. Patent Application No. 07 / 466,008, No. 07 / 610,515, No. 07 / 919,297, No. 07 / 922,649, No. 08 / 031,801, No. 08 / 112,848, No. 08 / 234,145, No. 08 / 376,279, No. 08 / 430,938, No. 08 / 464,584, No. 08 / 464,582, No. 08 / 463,191, No. 08 / 462,837, No. 08 / 486,853, No. 08 / 486,857, No. 08 / 486,859, No. 08 / 462,513, No. 08 / 724,752 and Further discussed and explained in U.S. Patent Applications No. 08 / 759,620; and U.S. Patent Applications No. 6,162,963; No. 6,150,584; No. 6,114,598; No. 6,075,181, and No. 5,939,598 and Japanese Patent Applications No. 3,068,180 B2, No. 3,068,506 B2, and No. 3,068,507 B2. Reference [Mendez et al. See Nature Genetics 15:146-156 (1997) and Green and Jakobovits J. Exp. Med. 188:483-495 (1998)], European Patent No. 0 463 151 B1, WO 94 / 02602, WO 96 / 34096, WO 98 / 24893, WO 00 / 76310 and WO 03 / 47336.

[0094] In an alternative approach, others, including GenPharm International, Inc., have utilized a "minilocus" approach. In the minilocus approach, an exogenous Ig locus is mimicked through the inclusion of fragments (individual genes) from the Ig locus. Thus, one or more VH genes, one or more genes, one or more DH genes, one or more JH genes, a mu constant region, and a second constant region (preferably a gamma constant region) are formed as constructs for insertion into animals. This approach is described in U.S. Patent No. 5,545,807 and U.S. Patents No. 5,545,806; No. 5,625,825; No. 5,625,126; No. 5,633,425; No. 5,661,016; No. 5,770,429 of Surani et al. No. 5,789,650; No. 5,814,318; No. 5,877,397; No. 5,874,299; and No. 6,255,458 (Lonberg and Kay, respectively), U.S. Patent No. 5,591,669 and No. 6,023,010 (Krimpenfort and Berns), U.S. Patent No. 5,612,205; No. 5,721,367; and is described in No. 5,789,215 (Berns et al.), and U.S. Patent No. 5,643,763 (Choi and Dunn), and GenPharm International U.S. Patent Applications No. 07 / 574,748, No. 07 / 575,962, No. 07 / 810,279, No. 07 / 853,408, No. 07 / 904,068, No. 07 / 990,860, No. 08 / 053,131, No. 08 / 096,762, No. 08 / 155,301, No. 08 / 161,739, No. 08 / 165,699, and No. 08 / 209,741.See also European Patent No. 0 546 073 B1, WO 92 / 03918, WO 92 / 22645, WO 92 / 22647, WO 92 / 22670, WO 93 / 12227, WO 94 / 00569, WO 94 / 25585, WO 96 / 14436, WO 97 / 13852, and WO 98 / 24884 and U.S. Patent No. 5,981,175. Additional literature [Taylor. et al. (1992), Chen et al . (1993), Tuaillon et al . (1993), Choi et al . (1993), Lonberg et al. (1994), Taylor et al. (1994), and Tuaillon et al. (1995), Fishwild et al. See (1996)].

[0095] Kirin has also demonstrated the generation of human antibodies from mice into which microcell fusions, large fragments of chromosomes, or whole chromosomes have been introduced. See European Patent Applications No. 773,288 and No. 843,961. Xenerex Biosciences is developing a technique for the potential generation of human antibodies. In this technique, SCID mice are reconstituted with human lymphoid cells, e.g., B and / or T cells. The mice are then immunized with an antigen, after which an immune response to the antigen can be generated. See U.S. Patents No. 5,476,996; No. 5,698,767; and No. 5,958,765.

[0096] Human anti-mouse antibody (HAMA) reactions have led to an industry of manufacturing chimeric or other humanized antibodies. However, certain human anti-chimeric antibody (HACA) reactions will be observed, particularly with chronic or multiple dose use of the antibody. Therefore, to mitigate concerns and / or effects regarding HAMA or HACA reactions, it would be desirable to provide antibody constructs containing a fully human binding domain to CDH3 and a fully human binding domain to CD3.

[0097] The terms “(specifically) binding,” “(specifically) recognizing,” “(specifically) being sent,” and “(specifically) reacting” according to the present invention mean that the binding domain interacts with or specifically interacts with one or more, preferably at least two, more preferably at least three, and most preferably at least four amino acids of an epitope located on a target protein or antigen (CDH3 / CD3).

[0098] The term “epitope” refers to a binding domain, such as a site on an antigen to which an antibody or immunoglobulin, or a derivative or fragment of an antibody or immunoglobulin, specifically binds. The “epitope” is antigenic, and therefore the term epitope is sometimes referred to herein as an “antigen structure” or “antigen determinant.” Thus, the binding domain is an “antigen interaction site.” Such binding / interaction is also understood to represent “specific recognition.”

[0099] An "epitope" can be formed by continuous or discontinuous amino acids concatenated by the tertiary folding of a protein. A "linear epitope" is an epitope that includes an epitope in which the primary sequence of amino acids is recognized. A linear epitope typically contains at least 3 or at least 4, more commonly, at least 5 or at least 6 or at least 7, for example, about 8 to about 10 amino acids in a unique sequence, and may also be longer, containing at least 15 or 20 amino acids, at least 25 or 30 amino acids, or much more.

[0100] In contrast to linear epitopes, a "stereolocated epitope" is an epitope in which the primary sequence of amino acids containing the epitope is not the sole determining component of the recognized epitope (e.g., an epitope in which the primary sequence of amino acids is not necessarily recognized by the binding domain). Typically, stereolocated epitopes contain an increased number of amino acids compared to linear epitopes. Regarding the recognition of stereolocated epitopes, the binding domain recognizes the three-dimensional structure of the antigen, preferably a peptide or protein or a fragment thereof (in the context of the present invention, an antigen for one of the binding domains is contained within the CDH3 protein). For example, when a protein molecule is folded to form a three-dimensional structure, specific amino acids and / or polypeptide backbones forming the stereolocated epitope may be conjugated to allow the antibody to recognize the epitope. Methods for determining epitope formation include, but are not limited to, X-ray crystallography, two-dimensional nuclear magnetic resonance (2D-NMR) spectroscopy, site-designated spin labeling, and electron paramagnetic resonance (EPR) spectroscopy.

[0101] The provided example describes an additional method for characterizing a given binding domain, including testing whether the given binding domain binds to one or more epitope(s) of a given protein in a specific CDH3.

[0102] The term “epitope cluster” as used herein refers to an epitope located on a defined continuous stretch of an antigen. An epitope cluster may comprise one or more epitopes. An antibody construct may also bind to an epitope within an epitope cluster and additionally bind to additional epitopes outside this cluster, which may correspond to discontinuous epitopes. Discontinuous epitopes are typically characterized by comprising non-adjacent amino acid stretches of the antigen. For example, the antibody construct may bind to extracellular subdomains D3A and D3C but not to D3B. The concept of “epitope clustering” is also used in characterizing the features of the antibody construct of the present invention. The epitope clusters and epitopes defined in the extracellular domain of CDH3 in relation to the present invention are described above and illustrated in FIG. 1.

[0103] The extracellular domains (D1 to D5) or their subdomains (A, B, C) in the human CDH3 protein are exchanged with the respective extracellular domains (D1 to D5) or their subdomains (A, B, C) of the non-human and non-primate (e.g., chicken or mouse) CDH3 antigens (resulting in a composition containing human CDH3, wherein one human extracellular domain or its subdomain is replaced by its relative non-human extracellular domain or its subdomain), and a reduction in binding domain binding will occur. The reduction is preferably at least 10%, 20%, 30%, 40%, 50% compared to each epitope cluster in the human CDH3 protein; more preferably at least 60%, 70%, 80%, 90%, 95%, or even 100%, thereby setting the binding to each extracellular domain (D1 to D5) or its subdomain in the human CDH3 protein to 100%. It is thought that the aforementioned human CDH3 / non-human CDH3 chimera is expressed in CHO cells. Additionally, it is thought that the human CDH3 / non-human CDH3 chimera fuses with different membrane-binding proteins, such as the transmembrane domain and / or cytoplasmic domain of EpCAM.

[0104] A method for testing such loss of binding resulting from exchange with each extracellular domain (D1–D5) or its subdomain of the non-human (e.g., mouse, but other similar, rat, hamster, rabbit, chicken, etc. may also be possible) CDH3 antigen is described in Example 2. An additional method for determining the contribution of specific residues of the target antigen to recognition by antibody constructs or binding domains is alanine scanning (see, for example, [Morrison KL & Weiss GA. Cur Opin Chem Biol. 2001 Jun;5(3):302-7]), in which each residue to be analyzed is replaced, for example, by site-specific mutagenesis. Although alanine mimics the secondary structural criteria possessed by many other amino acids, it is used due to its non-large, chemically inactive, methyl functional group. Sometimes, large amino acids, such as valine or leucine, may be used when a size conversion of the mutant residue is desired. Alanine scanning is a maturation technique that has been used for a long time.

[0105] The interaction between the binding domain and the epitope or epitope cluster indicates that the binding domain exhibits a recognizable affinity for the epitope or epitope cluster on a specific protein or antigen (here, CDH3 and CD3, respectively), and generally does not exhibit significant reactivity with proteins or antigens other than CDH3 or CD3. The "recognizable affinity" is approximately 10 -6 It includes a bond with M (KD) or a stronger affinity. Preferably, the bond has a bond affinity of about 10 -12 to 10 -8 M, 10 -12 to 10 -9 M, 10 -12 to 10 -10 M, 10 -11 to 10 -8 M, preferably about 10 -11 to 10 -9It is considered specific when M. Whether the binding domain specifically reacts with or binds to a target can be easily tested, in particular, by comparing the reaction of the binding domain with the target protein or antigen with the reaction of the binding domain with a protein or antigen other than CDH3 or CD3. Preferably, the binding domain of the present invention does not essentially or substantially bind to a protein or antigen other than CDH3 or CD3 (i.e., the first binding domain cannot bind to a protein other than CDH3, and the second binding domain cannot bind to a protein other than CD3).

[0106] The terms “essentially / substantially not binding” or “unable to bind” are defined such that the binding domain of the present invention does not bind to a protein or antigen other than CDH3 or CD3, that is, does not exhibit reactivity with a protein or antigen other than CDH3 or CD3 of more than 30%, preferably 20% or less, more preferably 10% or less, particularly preferably 9%, 8%, 7%, 6% or 5% or less, thereby setting the binding to CDH3 or CD3 to be 100%, respectively.

[0107] Specific binding is believed to be achieved by specific motifs within the amino acid sequences of the binding domain and the antigen. Thus, binding is achieved as a result of their primary, secondary, and / or tertiary structures, as well as secondary modifications of said structures. Specific interaction between an antigen-interaction site and its specific antigen can result in simple binding of said site to the antigen. Furthermore, specific interaction between an antigen-interaction site and its specific antigen can alternatively or additionally result in signal initiation, for example, due to the induction of stereolocation modifications of the antigen, oligomerization of the antigen, etc.

[0108] In another aspect, the present invention provides a bispecific antibody construct comprising a first preferably human binding domain that binds to human CDH3 on the surface of a target cell and a second preferably human binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain comprises a VH region comprising CDR-H1, CDR-H2, and CDR-H3 selected from the group consisting of the following, and a VL region comprising CDR-L1, CDR-L2, and CDR-L3:

[0109] a) CDR-H1 as presented in SEQ ID NO. 149, CDR-H2 as presented in SEQ ID NO. 150, CDR-H3 as presented in SEQ ID NO. 151, CDR-L1 as presented in SEQ ID NO. 152, CDR-L2 as presented in SEQ ID NO. 153, and CDR-L3 as presented in SEQ ID NO. 154;

[0110] b) CDR-H1 as presented in SEQ ID NO. 159, CDR-H2 as presented in SEQ ID NO. 160, CDR-H3 as presented in SEQ ID NO. 161, CDR-L1 as presented in SEQ ID NO. 162, CDR-L2 as presented in SEQ ID NO. 163, and CDR-L3 as presented in SEQ ID NO. 164;

[0111] c) CDR-H1 as presented in SEQ ID NO. 169, CDR-H2 as presented in SEQ ID NO. 170, CDR-H3 as presented in SEQ ID NO. 171, CDR-L1 as presented in SEQ ID NO. 172, CDR-L2 as presented in SEQ ID NO. 173, and CDR-L3 as presented in SEQ ID NO. 174;

[0112] d) CDR-H1 as presented in SEQ ID NO. 179, CDR-H2 as presented in SEQ ID NO. 180, CDR-H3 as presented in SEQ ID NO. 181, CDR-L1 as presented in SEQ ID NO. 182, CDR-L2 as presented in SEQ ID NO. 183, and CDR-L3 as presented in SEQ ID NO. 184;

[0113] e) CDR-H1 as presented in SEQ ID NO. 189, CDR-H2 as presented in SEQ ID NO. 190, CDR-H3 as presented in SEQ ID NO. 191, CDR-L1 as presented in SEQ ID NO. 192, CDR-L2 as presented in SEQ ID NO. 193, and CDR-L3 as presented in SEQ ID NO. 194;

[0114] f) CDR-H1 as presented in SEQ ID NO. 199, CDR-H2 as presented in SEQ ID NO. 200, CDR-H3 as presented in SEQ ID NO. 201, CDR-L1 as presented in SEQ ID NO. 202, CDR-L2 as presented in SEQ ID NO. 203, and CDR-L3 as presented in SEQ ID NO. 204;

[0115] g) CDR-H1 as presented in SEQ ID NO. 209, CDR-H2 as presented in SEQ ID NO. 210, CDR-H3 as presented in SEQ ID NO. 211, CDR-L1 as presented in SEQ ID NO. 212, CDR-L2 as presented in SEQ ID NO. 213, and CDR-L3 as presented in SEQ ID NO. 214;

[0116] h) CDR-H1 as presented in SEQ ID NO. 219, CDR-H2 as presented in SEQ ID NO. 220, CDR-H3 as presented in SEQ ID NO. 221, CDR-L1 as presented in SEQ ID NO. 222, CDR-L2 as presented in SEQ ID NO. 223, and CDR-L3 as presented in SEQ ID NO. 224;

[0117] i) CDR-H1 as presented in SEQ ID NO. 229, CDR-H2 as presented in SEQ ID NO. 230, CDR-H3 as presented in SEQ ID NO. 231, CDR-L1 as presented in SEQ ID NO. 232, CDR-L2 as presented in SEQ ID NO. 233, and CDR-L3 as presented in SEQ ID NO. 234; and

[0118] j) CDR-H1 as presented in SEQ ID NO. 239, CDR-H2 as presented in SEQ ID NO. 240, CDR-H3 as presented in SEQ ID NO. 241, CDR-L1 as presented in SEQ ID NO. 242, CDR-L2 as presented in SEQ ID NO. 243, and CDR-L3 as presented in SEQ ID NO. 244.

[0119] In another aspect, the present invention provides a bispecific antibody construct comprising a first preferably human binding domain that binds to human CDH3 on the surface of a target cell and a second preferably human binding domain that binds to human CD3 on the surface of a T cell, wherein the first binding domain comprises a VH region comprising CDR-H1, CDR-H2, and CDR-H3 selected from the group consisting of the following, and a VL region comprising CDR-L1, CDR-L2, and CDR-L3:

[0120] a) CDR-H1 as presented in SEQ ID NO. 279, CDR-H2 as presented in SEQ ID NO. 280, CDR-H3 as presented in SEQ ID NO. 281, CDR-L1 as presented in SEQ ID NO. 282, CDR-L2 as presented in SEQ ID NO. 283, and CDR-L3 as presented in SEQ ID NO. 284;

[0121] b) CDR-H1 as presented in SEQ ID NO. 289, CDR-H2 as presented in SEQ ID NO. 290, CDR-H3 as presented in SEQ ID NO. 291, CDR-L1 as presented in SEQ ID NO. 292, CDR-L2 as presented in SEQ ID NO. 293, and CDR-L3 as presented in SEQ ID NO. 294;

[0122] c) CDR-H1 as presented in SEQ ID NO. 299, CDR-H2 as presented in SEQ ID NO. 300, CDR-H3 as presented in SEQ ID NO. 301, CDR-L1 as presented in SEQ ID NO. 302, CDR-L2 as presented in SEQ ID NO. 303, and CDR-L3 as presented in SEQ ID NO. 304;

[0123] d) CDR-H1 as presented in SEQ ID NO. 309, CDR-H2 as presented in SEQ ID NO. 310, CDR-H3 as presented in SEQ ID NO. 311, CDR-L1 as presented in SEQ ID NO. 312, CDR-L2 as presented in SEQ ID NO. 313, and CDR-L3 as presented in SEQ ID NO. 314;

[0124] e) CDR-H1 as presented in SEQ ID NO. 319, CDR-H2 as presented in SEQ ID NO. 320, CDR-H3 as presented in SEQ ID NO. 321, CDR-L1 as presented in SEQ ID NO. 322, CDR-L2 as presented in SEQ ID NO. 323, and CDR-L3 as presented in SEQ ID NO. 324;

[0125] f) CDR-H1 as presented in SEQ ID NO. 329, CDR-H2 as presented in SEQ ID NO. 330, CDR-H3 as presented in SEQ ID NO. 331, CDR-L1 as presented in SEQ ID NO. 332, CDR-L2 as presented in SEQ ID NO. 333, and CDR-L3 as presented in SEQ ID NO. 334;

[0126] g) CDR-H1 as presented in SEQ ID NO. 339, CDR-H2 as presented in SEQ ID NO. 340, CDR-H3 as presented in SEQ ID NO. 341, CDR-L1 as presented in SEQ ID NO. 342, CDR-L2 as presented in SEQ ID NO. 343, and CDR-L3 as presented in SEQ ID NO. 344; and

[0127] h) CDR-H1 as presented in SEQ ID NO. 349, CDR-H2 as presented in SEQ ID NO. 350, CDR-H3 as presented in SEQ ID NO. 351, CDR-L1 as presented in SEQ ID NO. 352, CDR-L2 as presented in SEQ ID NO. 353, and CDR-L3 as presented in SEQ ID NO. 354.

[0128] The term "variable" refers to a portion of an antibody or immunoglobulin domain (i.e., "variable domain(s)") that indicates variability in their sequence and is involved in determining the specificity and binding affinity of a particular antibody. A pair of variable heavy chains (VH) and variable light chains (VL) together form a single antigen-binding site. The CH domain closest to the VH is denoted as CH1. Each light (L) chain is connected by a single covalent disulfide bond, whereas the two H chains are connected to each other by one or more disulfide bonds depending on the H chain isotype.

[0129] Variability is not uniformly distributed throughout the entire variable domain of the antibody; instead, it is concentrated in the subdomains of each heavy and light chain variable region. These subdomains are referred to as "hypervariable regions" or "complementary determination regions" (CDRs). The more conserved (i.e., non-hypervariable) portion of the variable domain is called the "framework" region (FRM or FR), which forms an antigen-binding surface by providing a scaffold for the six CDRs in three-dimensional space. The variable domains of naturally derived heavy and light chains each contain four FRM regions (FR1, FR2, FR3, and FR4) that primarily adopt a β-sheet configuration connected by three hypervariable regions, forming loop links and, in some cases, parts of the β-sheet structure. The hypervariable region in each chain is retained by the FRM proximal to and along with the hypervariable region from the other chain, contributing to the formation of the antigen-binding site (Reference [Kabat et al. , loc . cit .] See reference). The constant domain is not directly involved in antigen binding, but exhibits various effector functions, e.g., antibody-dependent, cell-mediated cytotoxicity and complement activation.

[0130] The term "CDR" and its plural "CDR" refer to complementarity determining regions, three of which constitute the binding features of the light chain variable regions (CDR-L1, CDR-L2, and CDR-L3), and three of which constitute the features of the heavy chain variable regions (CDR-H1, CDR-H2, and CDR-H3). The CDRs contain most of the residues that cause specific interactions between the antibody and the antigen, and thus contribute to the functional activity of the antibody molecule: they are the main determinants of antigen specificity.

[0131] The precise definitions of CDR boundaries and lengths are subject to different classification and numbering systems. Accordingly, a CDR may be referred to by any other boundary definition including Kabat, Chothia, Tact, or the numbering systems described herein. Despite the differences in boundaries, each of these systems has some degree of overlap with constituting the so-called "hypervariable region" within variable sequences. Therefore, CDR definitions according to these systems may differ in length and boundary area relative to adjacent framework regions. For example, Kabat (an approach based on cross-species sequence variability), Chothia (an approach based on crystallographic studies of antigen-antibody complexes), and / or MacCallum (Kabat et al ., loc . cit .; Chothia et al ., J. MoI. Biol, 1987, 196: 901-917; MacCallum et al See ., J. MoI. Biol, 1996, 262: 732). Another standard for characterizing antigen binding sites is the AbM definition used by Oxford Molecular AbM antibody modeling software. For example, see the literature [Protein Sequence and Structure Analysis of Antibody Variable Domains. In: Antibody Engineering Lab Manual (Ed.: Duebel, S. and Kontermann, R., Springer-Verlag, Heidelberg)]. Two residue identification techniques can be combined to determine hybrid CDRs, to the extent that they define overlapping regions but not identical regions. However, numbering according to the so-called Kavat system is preferred.

[0132] Typically, CDRs form loop structures that can be classified as normal structures. The term "normal structure" refers to the major chain stereolocations adopted by the antigen-binding (CDR) loop. From similar structural studies, it was found that five out of six antigen-binding loops possess only a limited repertoire of available stereolocations. Each normal structure can be characterized by the torsion angle of the polypeptide backbone. Therefore, corresponding loops between antibodies can have very similar three-dimensional structures despite high amino acid sequence variability in most loops (Chothia and Lesk, J. MoI. Biol., 1987, 196: 901; Chothia et al ., Nature, 1989, 342: 877; Martin and Thornton, J. MoI. Biol, 1996, 263: 800). Furthermore, there is a relationship between the adopted loop structure and the surrounding amino acid sequence. The stereolocation of a specific normality is determined by the length of the loop and amino acid residues present at important positions not only within the loop but also within the conserved framework (i.e., outside the loop). Therefore, placement for a specific normality can be based on the presence of these important amino acid residues.

[0133] The term "normal structure" also refers to the linear sequence of an antibody, for example, that listed by Kabat (Kabat et alConsiderations regarding ., loc. cit.) may be included. The Kavat numbering scheme (system) is a standard adopted in various ways for numbering amino acid residues of antibody variable domains in a consistent manner, and is a preferred scheme applied to the present invention as also mentioned elsewhere in this specification. Additional structural considerations may also be used to determine the normal structure of the antibody. For example, the differences not fully reflected by Kavat numbering may be described by numbering systems such as Chothia and / or represented by other techniques, such as crystallography and two- or three-dimensional computer modeling. Thus, a given antibody sequence may be placed in a normal classification that allows for the identification of an appropriate classification sequence, particularly (e.g., based on the need to include various normal structures within a library). Literature [Chothia et al ., loc . cit The involvement of antibody amino acid sequences and structural considerations, such as the kavat numbering and the understanding of the regular patterns of antibody structures, as described in [...], is described in the literature. The subunit structures and three-dimensional configurations of different classes of immunoglobulins are well known in the art. For the examination of antibody structures, the literature [Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, eds. Harlow et al See ., 1988].

[0134] CDR3 in the light chain and, in particular, CDR3 in the heavy chain can constitute the most important determinants of antigen binding within the variable regions of the light and heavy chains. In some antibody constructs, heavy chain CDR3 appears to constitute the primary area between the antigen and the antibody. In vitro selection schemes in which only CDR3 is modified can be used to alter the binding properties of antibodies or to determine which residues contribute to antigen binding. Therefore, CDR3 is typically the largest source of molecular diversity within the antibody-binding site. For example, H3 can be as short as two amino acid residues or more than 26 amino acids.

[0135] After assembly and somatic mutation, antibody gene sequences are highly altered, and these altered genes are 10 10 It is presumed to encode two different antibody molecules (Immunoglobulin Genes, 2 nd ed., eds. Jonio et al., Academic Press, San Diego, CA, 1995). Thus, the immune system provides a repertoire of immunoglobulins. The term “repertoire” refers to at least one nucleotide sequence derived wholly or partially from at least one sequence encoding at least one immunoglobulin. The sequence(s) may be generated by in vivo rearrangement of the V, D, and J segments of the heavy chain and the V and J segments of the light chain. Alternatively, the sequence(s) may be generated from cells in response to rearrangement, for example, occurring under in vitro stimulation. Alternatively, some or all of the sequence(s) may be obtained by DNA splicing, nucleotide synthesis, mutagenesis, and other methods; see, for example, U.S. Patent No. 5,565,332. The repertoire may include only one sequence or may include multiple sequences, including one of genetically diverse collections.

[0136] In one embodiment, the first binding domain of the antibody construct of the present invention comprises a VH region selected from the group consisting of VH regions as presented in SEQ ID NO. 155, SEQ ID NO. 165, SEQ ID NO. 175, SEQ ID NO. 185, SEQ ID NO. 195, SEQ ID NO. 205, SEQ ID NO. 215, SEQ ID NO. 225, SEQ ID NO. 235 and SEQ ID NO. 245.

[0137] In a further embodiment of the antibody composition of the present invention, the first binding domain comprises a VL region selected from the group consisting of VL regions as presented in SEQ ID NO. 156, SEQ ID NO. 166, SEQ ID NO. 176, SEQ ID NO. 186, SEQ ID NO. 196, SEQ ID NO. 206, SEQ ID NO. 216, SEQ ID NO. 226, SEQ ID NO. 236 and SEQ ID NO. 246.

[0138] In another embodiment of the antibody composition of the present invention, the first binding domain comprises a VH region and a VL region selected from the group consisting of pairs of VH regions and VL regions as presented in SEQ ID NO. 155+156, SEQ ID NO. 165+166, SEQ ID NO. 175+176, SEQ ID NO. 185+186, SEQ ID NO. 195+196, SEQ ID NO. 205+206, SEQ ID NO. 215+216, SEQ ID NO. 225+226, SEQ ID NO. 235+236 and SEQ ID NO. 245+246.

[0139] In another embodiment, the first binding domain of the antibody construct of the present invention comprises a VH region selected from the group consisting of VH regions as presented in SEQ ID NO. 285, SEQ ID NO. 295, SEQ ID NO. 305, SEQ ID NO. 315, SEQ ID NO. 325, SEQ ID NO. 335, SEQ ID NO. 345, and SEQ ID NO. 355.

[0140] In a further embodiment of the antibody composition of the present invention, the first binding domain comprises a VL region selected from the group consisting of VL regions as presented in SEQ ID NO. 286, SEQ ID NO. 296, SEQ ID NO. 306, SEQ ID NO. 316, SEQ ID NO. 326, SEQ ID NO. 336, SEQ ID NO. 346 and SEQ ID NO. 356.

[0141] In another embodiment of the antibody composition of the present invention, the first binding domain comprises a VH region and a VL region selected from the group consisting of pairs of VH regions and VL regions as presented in SEQ ID NO. 285+286, SEQ ID NO. 295+296, SEQ ID NO. 305+306, SEQ ID NO. 315+316, SEQ ID NO. 325+326, SEQ ID NO. 335+336, SEQ ID NO. 345+346, and SEQ ID NO. 355+356.

[0142] The term “double specificity” as used herein refers to an antibody construct that is “at least double specific,” i.e., comprises at least a first binding domain and a second binding domain, wherein the first binding domain binds to one antigen or target (here: CDH3) and the second binding domain binds to another antigen or target (here: CD3). Thus, the antibody construct according to the present invention comprises specificity for at least two different antigens or targets. The term “double specificity antibody construct” of the present invention also includes a multispecificity antibody construct, e.g., a triple specificity antibody construct, the latter comprising three binding domains or a construct having more than three (e.g., four, five...) specificities.

[0143] Considering that the antibody constructs according to the present invention are (at least) bispecific, they do not occur naturally and are significantly different from naturally occurring products. A "bispecific" antibody construct or immunoglobulin is an artificial hybrid antibody or immunoglobulin having at least two distinct binding sites having different specificities. Bispecific antibodies can be produced by various methods including the fusion of hybridomas or the linkage of Fab' fragments. For example, see the literature [Songsivilai & Lachmann, Clin. Exp. Immunol. 79:315-321 (1990)].

[0144] At least two binding domains and variable domains of the antibody construct of the present invention may or may not include a peptide linker (spacer peptide). The term “peptide linker” defines, according to the present invention, an amino acid sequence in which one (variable and / or binding) domain and another (variable and / or binding) domain of the antibody construct of the present invention are linked to each other. An essential technical feature of such a peptide linker is that it does not possess any polymeric activity. Suitable peptide linkers include those described in U.S. Patents No. 4,751,180 and 4,935,233 or WO 88 / 09344.

[0145] Where a linker is used, the linker preferably has a length and sequence sufficient to ensure that each of the first and second domains can independently possess their differential binding specificities. For a peptide linker connecting at least two binding domains (or two variable domains) in the antibody composition of the present invention, said peptide linker comprising only a small number of amino acid residues, for example, fewer than 12 amino acid residues, is preferred. Thus, a peptide linker having 12, 11, 10, 9, 8, 7, 6, or 5 amino acid residues is preferred. A peptide linker thought to have fewer than 5 amino acids comprises 4, 3, 2, or 1 amino acid(s), and is preferably a Gly-rich linker. A "single" amino acid particularly preferred with respect to said "peptide linker" is Gly. Thus, said peptide linker may consist of a single amino acid Gly. Another preferred embodiment of the peptide linker features the amino acid sequence Gly-Gly-Gly-Gly-Ser, i.e., Gly4Ser (SEQ No. 393), or a polymer thereof, i.e., (Gly4Ser)x, where x is an integer greater than or equal to 1. The feature of the peptide linker comprising the absence of a secondary structure promoting is known in the art and is described, for example, in the literature [Dall'Acqua et al. (Biochem. (1998) 37, 9266-9273), Cheadle et al. (Mol Immunol (1992) 29, 21-30) and Raag and Whitlow (FASEB (1995) 9(1), 73-80)]. Additionally, a peptide linker that does not promote any secondary structure is preferred. The binding of the domains to each other may be provided by a gene organization, for example, as described in the examples.Methods for preparing fused and operationally linked bispecific single-strand constructs and expressing them in mammalian cells or bacteria are well known in the art (e.g., WO 99 / 54440 or literature [Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 2001]).

[0146] Accordingly, the present invention provides preferred embodiments, wherein the antibody construct is a form selected from the group consisting of (scFv)2, scFv-single-domain mAb, any of the aforementioned forms of diabosids and oligomers. The term "form" does not exclude the fact that the construct may be further modified, for example by attachment or fusion to another moiety, as described herein.

[0147] According to a particularly preferred embodiment, the antibody construct of the present invention is a “bispecific single-strand antibody construct,” more preferably a bispecific “single-strand Fv (scFv).” Although the two domains of the Fv fragment, VL and VH, are encoded for distinct genes, they can be combined using a recombinant method by a synthetic linker that can produce a single protein chain in which the VL region and the VH region are paired to form a monovalent molecule; see, for example, the literature [Huston et al. (1988) Proc. Natl. Acad. Sci USA 85:5879-5883]). These antibody fragments are obtained using conventional techniques known to those skilled in the art, and the fragments are evaluated for function in the same manner as whole or full-length antibodies. Therefore, a single-strand variable fragment (scFv) is a fusion protein of the variable regions of the heavy chain (VH) and light chain (VL) of an immunoglobulin, connected by a short linker peptide of about 10 to about 25 amino acids, preferably about 15 to 20 amino acids. The linker is usually rich in glycine for flexibility as well as serine or threonine for solubility, and can connect the N-terminus of the VH to the C-terminus of the VL, or vice versa. This protein retains the specificity of the original immunoglobulin despite the removal of the constant region and the introduction of the linker.

[0148] Bispecific single-strand molecules are known in the art and are described in WO 99 / 54440 [Mack, J. Immunol. (1997), 158, 3965-3970; Mack, PNAS, (1995), 92, 7021-7025; Kufer, Cancer Immunol. Immunother., (1997), 45, 193-197; Loffler, Blood, (2000), 95, 6, 2098-2103; Bruuhl, Immunol., (2001), 166, 2420-2426; Kipriyanov, J. Mol. Biol., (1999), 293, 41-56]. Techniques described for the generation of short-chain antibodies (particularly, U.S. Patent No. 4,946,778, literature [Kontermann and Dubel (2010), loc . cit. and Little (2009), loc . cit .] (see reference) may be suitable for generating short-chain antibody constructs that specifically recognize selected target(s).

[0149] Bivalent (also called divalent) or bispecific single-strand variable fragments (bi-scFv or di-scFv having the form (scFv)2 can be fabricated by linking two scFv molecules. If these two scFv molecules have the same binding specificity, the resulting (scFv)2 molecule can preferably be called divalent (i.e., having two valencies for the same target epitope). If the two scFv molecules have different binding specificities, the resulting (scFv)2 molecule can preferably be called bispecific. Linking is performed by creating a single peptide chain having two VH regions and two VL regions to obtain a tandem scFv (e.g., the literature [Kufer P. et alSee ., (2004) Trends in Biotechnology 22(5):238-244]). Another possibility is the generation of scFv molecules possessing a linker peptide that causes the scFv to dimerize, where the two variable regions are too short (e.g., about 5 amino acids) to fold together. This type is known as a diabody (e.g., in the literature [Hollinger, Philipp et al. , (July 1993) Proceedings of the National Academy of Sciences of the United States of America 90 (14): 6444-8.].

[0150] According to a further preferred embodiment of the antibody composition of the present invention, the heavy chain (VH) and light chain (VL) of the binding domain (binding to or to the target antigen CDH3) are not directly connected through the peptide linker as described above, but the binding domain is formed as described for the diabody. Thus, the VH of the CD3 binding domain can be fused to the VL of the CDH3 binding domain through the peptide linker, and the VH of the CDH3 binding domain is fused to the VL of the CD3 binding domain through this peptide linker.

[0151] A single-domain antibody comprises only one (monomer) antibody variable domain capable of selectively binding to a specific antigen independently of other V regions or domains. A first single-domain antibody is engineered from a heavy-chain antibody found in camels, and these are V H It is called the H fragment. Cartilaginous fish also V NARHeavy chain antibodies (IgNARs) can be obtained from single-domain antibodies referred to as fragments. An alternative approach is to split the dimeric variable domain from common immunoglobulins, for example from humans or rodents, into monomers, thereby obtaining VH or VL as a single-domain Ab. While most research on single-domain antibodies is currently based on heavy-chain variable domains, nanobodies derived from light chains have also been shown to bind specifically to target epitopes. Examples of single-domain antibodies are referred to as sdAbs, nanobodies, or single-variable-domain antibodies.

[0152] Therefore, (single domain mAb)2 is VH, VL, V H H and V NAR It is a monoclonal antibody construct composed of (at least) two single-domain monoclonal antibodies individually selected from the group including. The linker is preferably in the form of a peptide linker. Similarly, the "scFv-single-domain mAb" is a monoclonal antibody construct composed of at least one single domain as described above and one scFv molecule as described above. Additionally, the linker is preferably in the form of a peptide linker.

[0153] In one embodiment, the first binding domain comprises an amino acid sequence selected from the group consisting of the corresponding sequences as presented in SEQ ID NO. 157, SEQ ID NO. 167, SEQ ID NO. 177, SEQ ID NO. 187, SEQ ID NO. 197, SEQ ID NO. 207, SEQ ID NO. 217, SEQ ID NO. 227, SEQ ID NO. 237 and SEQ ID NO. 247.

[0154] In another embodiment, the first binding domain comprises an amino acid sequence selected from the group consisting of the corresponding sequences as presented in SEQ ID NO. 287, SEQ ID NO. 297, SEQ ID NO. 307, SEQ ID NO. 317, SEQ ID NO. 327, SEQ ID NO. 337, SEQ ID NO. 347 and SEQ ID NO. 357.

[0155] In addition, it is thought that the antibody construct of the present invention has additional functions in addition to its function of binding to the target molecule CDH3 and CD3. In this form, the antibody construct is a trifunctional or multifunctional antibody construct by providing additional functions such as a fully functional Fc constant domain that targets target cells through binding to CDH3, mediates cytotoxic T cell activation through binding to CD3, and mediates cytotoxicity of antibody-dependent cells through additional functions such as the recruitment of effector cells such as NK cells, labeling (fluorescence, etc.), therapeutic agents such as toxins or radionuclides and / or means to enhance serum half-life.

[0156] Examples of means for extending the serum half-life of antibody constructs of the present invention include peptides, proteins, or protein domains fused to or otherwise attached to the antibody construct. Groups of peptides, proteins, or protein domains include peptide binding to other proteins having a desirable pharmacokinetic profile in the human body, such as serum albumin (see WO 2009 / 127691). One example of these is shown in SEQ ID NO. 437. An alternative concept of such half-life extension (HLE) peptides includes peptide binding to the neonatal Fc receptor (FcRn, see WO 2007 / 098420), which may also be used in some constructs of the present invention. The concept of attaching a larger domain of a protein or complete protein includes, for example, fusions of human serum albumin, variants or mutants of human serum albumin (see WO 2011 / 051489, WO 2012 / 059486, WO 2012 / 150319, WO 2013 / 135896, WO 2014 / 072481, WO 2013 / 075066) or fusions of the domains thereof as well as the constant region of immunoglobulin (Fc domain) and variants thereof. These variants of the Fc domain may be optimized / modified to allow for coupling for the purpose of dimers or polymers, or to eliminate Fc receptor binding (e.g., Fcg receptor) for other reasons. If the HLE molecules described above consist of only a single polypeptide chain, they have the advantage that (i) there is no need for two separate expression systems, and (ii) they can be isolated to have high purity due to the absence of a "dummy chain." An additional concept known in the art for extending the half-life of small protein compounds in the human body is the pegylation of the said compound, e.g., the antibody composition of the present invention.

[0157] In a preferred embodiment, a bispecific antibody construct according to the present invention may be linked (e.g., via a peptide bond) to a fusion partner (e.g., a protein, polypeptide, or peptide) for the purpose of extending the serum half-life of the construct. These fusion partners may be selected from human serum albumin (“HSA” or “HALB”) as well as sequence variants thereof, peptide bonds to HSA, peptide bonds to FcRn (“FcRn BP”), or constructs comprising an Fc region (derived from the antibody). Exemplary sequences of these fusion partners are presented in SEQ ID NOs 406 to 421 and 437 to 444. Generally, the fusion partners are linked directly (e.g., via a peptide bond) or via a peptide linker, e.g., (GGGGS) n (Here, "n" is an integer of 2 or more, e.g., 2 or 3 or 4) may be linked to the N-terminus or C-terminus of the bispecific antibody construct according to the present invention. Suitable peptide linkers are presented in SEQ ID NOs 392 to 400.

[0158] An antibody construct designated as CDH3-13 (the full-length sequence of a bispecific molecule as presented in SEQ ID NO. 178) was linked within a frame by the selection of a fusion protein or fusion peptide (e.g., SEQ ID NOs 437 to 444) for the purpose of extending the serum half-life of the construct. The sequences of each of these fusion constructs are presented in SEQ ID NOs 379 to 389. The sequences of these fusion counterparts to the "naked" bispecific antibody construct may likewise be linked to any other antibody construct disclosed herein (as a C- or N-terminus in the method it corresponds to and which it is shown for CDH3-13).

[0159] Accordingly, it is considered that the bispecific antibody construct of the present invention further comprises SEQ ID NO. 437 or albumin, preferably human albumin or a variant thereof (having improved properties, e.g., affinity for the FcRn receptor and extended plasma half-life), most preferably a polypeptide as presented in albumin as depicted in SEQ ID NO. 443 or 444. These moiety is preferably fused within the frame for the C-terminus of the bispecific antibody construct.

[0160] Example 15 further demonstrates an unexpected advantage that the bispecific antibody construct of the present invention involves a C-terminal fusion of albumin. For the corresponding bispecific T cell engaging molecules comprising a binding domain specific to the CD3ε chain of human and non-silver monkey, fluffy tamarin, or squirrel monkey, the epitope is a portion of the amino acid sequence included in the group consisting of SEQ ID NOs 2, 4, 6, or 8 of WO 2008 / 119567, and at least the amino acid sequence Gln-Asp-Gly-Asn-Glu, and it has been observed that the corresponding molecule exhibited T cell cytotoxicity—when used at very high concentrations—even in the absence of target cells. This high concentration issue can be appropriately addressed for a specific administration route or in combination with a specific target situation and the required compound concentration. Preferred examples of such a second binding domain that binds to human CD3 on the T cell surface are described herein below and are presented in SEQ ID NOs 445 to 537. When serum albumin is fused to the C-terminus of this bispecific construct, T cell cytotoxicity is avoided, see Fig. 13. Without being bound by theory, T cell activation in the presence of high concentrations of T cells engaging with the bispecific antibody construct and in the absence of target cells can be explained by the dimerization or multimerization of the antibody construct through the CD3 binding domain. Such dimerization or multimerization is sterically impaired by the fusion of albumin or a variant thereof to the C-terminus of the antibody construct, while maintaining the characteristics of the antibody construct for its mode of T cell engagement.

[0161] Accordingly, a preferred antibody composition according to the present invention comprises N- to C-terminal order:

[0162] A first binding domain having an amino acid sequence selected from the group consisting of SEQ ID NO. 157, SEQ ID NO. 167, SEQ ID NO. 177, SEQ ID NO. 187, SEQ ID NO. 197, SEQ ID NO. 207, SEQ ID NO. 217, SEQ ID NO. 227, SEQ ID NO. 237, SEQ ID NO. 247, SEQ ID NO. 287, SEQ ID NO. 297, SEQ ID NO. 307, SEQ ID NO. 317, SEQ ID NO. 327, SEQ ID NO. 337, SEQ ID NO. 347, and SEQ ID NO. 357;

[0163] A peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs 392 to 400;

[0164] A second binding domain having an amino acid sequence selected from the group consisting of SEQ ID NO 453, SEQ ID NO 462, SEQ ID NO 471, SEQ ID NO 480, SEQ ID NO 489, SEQ ID NO 498, SEQ ID NO 507, SEQ ID NO 516, SEQ ID NO 525, SEQ ID NO 534, and SEQ ID NO 537; and

[0165] Optionally, a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs 392 to 400; and

[0166] Polypeptide having an amino acid sequence selected from the group consisting of SEQ ID NOs 437, 443 and 444.

[0167] According to another preferred embodiment, the bispecific antibody construct of the present invention comprises a third domain comprising two polypeptide monomers (in addition to the two binding domains), each comprising a hinge, CH2, and CH3 domain, wherein the two polypeptides (or polypeptide monomers) are fused together via a peptide linker. Preferably, the third domain is included in the N-to-C-terminal order: hinge-CH2-CH3-linker-hinge-CH2-CH3. Preferred amino acid sequences for the third domain are presented in SEQ ID NOs 414 to 421. Each of the polypeptide monomers is preferably selected from the group consisting of SEQ ID NOs 406 to 413, or has an amino acid sequence that is at least 90% identical to the sequences. In another preferred embodiment, the first binding domain and the second binding domain of the bispecific antibody construct of the present invention are fused to a third domain through a peptide linker selected, for example, from any one of SEQ ID NOs 392 to 400, preferably from any one of SEQ ID NOs 392, 393, 395, 396, 397, 399 and 400.

[0168] In relation to the present invention, "hinge" is an IgG hinge region. This region can be similarly identified using Kavat numbering, see Kavat positions 223 to 243. In relation to the above, the minimum requirement for "hinge" is an amino acid residue corresponding to the IgG1 sequence elongation of D231 to P243 according to Kavat numbering. The terms CH2 and CH3 refer to immunoglobulin heavy chain constant regions 2 and 3. These regions can likewise be similarly identified using Kavat numbering, see Kavat positions 244 to 360 for CH2 and Kavat positions 361 to 478 for CH3. It is understood that there are some modifications among immunoglobulins regarding the IgG1 Fc region, IgG2 Fc region, IgG3 Fc region, IgG4 Fc region, IgM Fc region, IgA Fc region, IgD Fc region, and IgE Fc region (see, for example, the literature [Padlan, Molecular Immunology, 31(3), 169-217 (1993)]). The term Fc monomer refers to the last two heavy chain constant regions of IgA, IgD, and IgG, and the last three heavy chain constant regions of IgE and IgM. Fc monomers may also include a flexible hinge at the N-terminus for these domains. For IgA and IgM, Fc monomers may include a J chain. For IgG, the Fc portion includes immunoglobulin domains CH2 and CH3 and a hinge between the first two domains and CH2. Although the boundaries of the Fc portion of the immunoglobulin may vary, examples of the human IgG heavy chain Fc portion containing functional hinge, CH2 and CH3 domains may be defined to include residue D231 (of the hinge domain) for P476 (of the C-terminal of the CH3 domain) or D231 for L476 for IgG4, respectively, with numbering according to the kavat.

[0169] Accordingly, the antibody composition of the present invention may include N- to C-terminals in order:

[0170] (a) First binding domain;

[0171] (b) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs 393, 399 and 400;

[0172] (c) Second binding domain;

[0173] (d) a peptide linker having an amino acid sequence selected from the group consisting of sequence numbers 392, 393, 395, 396, 397, 399 and 400;

[0174] (e) a first polypeptide monomer of a third domain (including hinge, CH2 and CH3 domains);

[0175] (f) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs 402, 403, 404 and 405; and

[0176] (g) A second polypeptide monomer of a third domain (including hinge, CH2 and CH3 domains).

[0177] In addition, it is preferable that the antibody compositions of the present invention be included in an N- to C-terminal order:

[0178] (a) a first binding domain having an amino acid sequence selected from the group consisting of SEQ ID NO 157, SEQ ID NO 167, SEQ ID NO 177, SEQ ID NO 187, SEQ ID NO 197, SEQ ID NO 207, SEQ ID NO 217, SEQ ID NO 227, SEQ ID NO 237, SEQ ID NO 247, SEQ ID NO 287, SEQ ID NO 297, SEQ ID NO 307, SEQ ID NO 317, SEQ ID NO 327, SEQ ID NO 337, SEQ ID NO 347 and SEQ ID NO 357;

[0179] (b) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs 393, 399 and 400;

[0180] (c) a second binding domain having an amino acid sequence selected from the group consisting of SEQ ID NO 453, SEQ ID NO 462, SEQ ID NO 471, SEQ ID NO 480, SEQ ID NO 489, SEQ ID NO 498, SEQ ID NO 507, SEQ ID NO 516, SEQ ID NO 525, SEQ ID NO 534 and SEQ ID NO 537;

[0181] (d) a peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs 392, 393, 395, 396, 397, 399 and 400; and

[0182] (e) A third domain having an amino acid sequence selected from the group consisting of SEQ ID NOs 414 to 421.

[0183] Accordingly, in a preferred embodiment, the antibody composition of the present invention comprises or is composed of a polypeptide selected from the group consisting of those presented in SEQ ID NO. 422, SEQ ID NO. 423, SEQ ID NO. 424, SEQ ID NO. 425, SEQ ID NO. 426, SEQ ID NO. 427, SEQ ID NO. 428, SEQ ID NO. 429, SEQ ID NO. 430, SEQ ID NO. 431, SEQ ID NO. 432, SEQ ID NO. 433, SEQ ID NO. 434 and SEQ ID NO. 435.

[0184] In addition, Example 15 is mentioned, which demonstrates one of the Fc constructs described above and their antitumor activities in a mouse xenograft model.

[0185] Covalent modifications of antibody constructs are also included within the scope of the present invention and are generally, but not always, performed after translation. For example, some types of covalent modifications of antibody constructs are introduced into the molecule by reacting specific amino acid residues of the antibody construct with an organic derivatizing agent capable of reacting with selected side chains or N- or C-terminal residues.

[0186] The cysteineyl residue is most commonly reacted with α-haloacetates (and corresponding amines), such as chloroacetic acid or chloroacetamide, to provide carboxymethyl or carboxyamidomethyl derivatives. The cysteineyl residue is also derivatized by reaction with bromotrifluoroacetone, α-bromo-β-(5-imidozoyl)propionic acid, chloroacetyl phosphate, N-alkylmaleimide, 3-nitro-2-pyridyl disulfide, methyl 2-pyridyl disulfide, p-chloromercurybenzoate, 2-chloromercury-4-nitrophenol, or chloro-7-nitrobenzo-2-oxa-1,3-diazole.

[0187] Histidyl residues are derivatized by reaction with diethylpyrocarbonate at pH 5.5 to 7.0 because this agent is relatively specific to the histidyl side chain. Para-bromophenacyl bromide is also useful, and the reaction is preferably carried out in 0.1 M sodium cacodilate at pH 6.0. Lysinyl and amino-terminal residues are reacted with succinic acid or other carboxylic acid anhydrides. Derivatization by these agents has the effect of reversing the charge of the lysinyl residue. Other suitable reagents for alpha-amino-containing residues include transaminase-catalyzed reactions with imidoesters, such as methyl picolineimidate; pyridoxal phosphate; pyridoxal; chloroborohydride; trinitrobenzenesulfonic acid; O-methylisourea; 2,4-pentanedione; and glyoxylates.

[0188] The arginyl residue is modified by reaction with one of phenylglyoxal, 2,3-butanedione, 1,2-cyclohexanedione, and ninhydrin, or with some common reagents. The derivatization of the arginine residue needs to be carried out under alkaline conditions due to the high pKa of the guanidine functional group. Furthermore, these reagents can react with the arginine epsilon-amino group as well as the lysine group.

[0189] Specific modifications of tyrosyl residues may be made with particular interest in introducing spectral labels into the tyrosyl residues via reaction with aromatic diazonium compounds or tetranitromethane. Most commonly, N-acetylimidazole and tetranitromethane are used to form O-acetyl tyrosyl species and 3-nitro derivatives, respectively. Tyrosyl residues are used to prepare labeled proteins for use in radioimmunoassays. 125 I or 131 It is iodized using I, and the chloramine T method described above is suitable.

[0190] The carboxyl side group (aspartyl or glutamyl) is selectively modified by reaction with a carbodiimide (R'―N=C=N--R'), where R and R' are selectively different alkyl groups, e.g., 1-cyclohexyl-3-(2-morpholinyl-4-ethyl)carbodiimide or 1-ethyl-3-(4-azonia-4,4-dimethylpentyl)carbodiimide. Furthermore, the aspartyl and glutamyl residues are converted into asparagineyl and glutamineyl residues by reaction with an ammonium ion.

[0191] Derivatization with a bifunctional agent is useful for crosslinking antibody constructs of the present invention onto water-insoluble support substrates or surfaces for use in various methods. Commonly used crosslinking agents include, for example, 1,1-bis(diazoacetyl)-2-phenylethane, glutaraldehyde, N-hydroxysuccinimide esters, for example, esters with 4-azidosalicylic acid, disuccinimidyl esters, such as 3,3'-dithiobis(succinimidylpropionate), and homodifunctional imidoesters comprising bis-N-maleimido-1,8-octane, which are bifunctional maleimides. Derivatizing agents, such as methyl-3-[(p-azidophenyl)dithio]propioimidate, yield photoactive intermediates capable of forming crosslinks in the presence of light. Alternatively, reactive water-insoluble matrices, such as cyanogen bromide-activated carbohydrates and reactive substrates as described in U.S. Patents No. 3,969,287; No. 3,691,016; No. 4,195,128; No. 4,247,642; No. 4,229,537; and No. 4,330,440 are used for protein fixation.

[0192] The glutamine-1 and asparagine-1 residues are frequently deamidated into the corresponding glutamyl and aspartyl residues, respectively. Alternatively, these residues are deamided under weakly acidic conditions. One of these forms of residues falls within the scope of the present invention.

[0193] Other modifications include hydroxylation of proline and lysine, phosphorylation of the hydroxyl group of a ceryl or threonyl residue, methylation of the α-amino group of the lysine, arginine, and histidine side chains (TE Creighton, Proteins: Structure and Molecular Properties, WH Freeman & Co., San Francisco, 1983, pp. 79-86), acetylation of the N-terminal amine, and amideation of any C-terminal carboxyl group.

[0194] Other types of covalent modifications of antibody constructs included within the scope of the present invention include altering the glycosylation pattern of a protein. As is known in the art, the glycosylation pattern may depend on the sequence of the protein (e.g., the presence or absence of specific glycosylated amino acid residues discussed below) or on the host cell or organism in which the protein is produced. Specific expression systems are discussed below.

[0195] Glycosylation of polypeptides is typically N-linked or O-linked. N-linking refers to the attachment of a carbohydrate moiety to the side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine and asparagine-X-threonine (where X is any amino acid excluding proline) are recognition sequences for the enzymatic attachment of a carbohydrate moiety to the asparagine side chain. Therefore, the presence of one of these tripeptide sequences within a polypeptide creates a potential glycosylation site. O-linked glycosylation refers to the attachment of a hydroxyamino acid of the sugar N-acetylgalactosamine, galactose, or xylose—most commonly to serine or threonine—although 5-hydroxyproline or 5-hydroxylysine may also be used.

[0196] The addition of a glycosylation site to the antibody composition is conveniently performed by modifying the amino acid sequence so that it contains one or more of the tripeptide sequences described above (for the N-linked glycosylation site). The modification may also be made by the addition or substitution thereof of one or more serine or threonine residues to the starting sequence (for the O-linked glycosylation site). For ease, the amino acid sequence of the antibody composition is preferably modified through a change at the DNA level, specifically by mutating the DNA encoding the polypeptide at a pre-selected base to generate a codon that is translated to the target amino acid.

[0197] Another means of increasing the number of carbohydrate moiety units in antibody compositions is by chemical or enzymatic binding of glycosides to proteins. These procedures are advantageous in that they do not require the generation of host cell proteins capable of glycosylation for N- and O-linked glycosylation. Depending on the binding method used, the sugar(s) may be attached to (a) arginine and histidine, (b) free carboxyl groups, (c) free sulfhydryl groups, e.g., the sulfhydryl group of cysteine, (d) free hydroxyl groups, e.g., the hydroxyl group of serine, threonine, or hydroxyproline, (e) aromatic residues, e.g., the aromatic residues of phenylalanine, tyrosine, or tryptophan, or (f) the amide group of glutamine. These methods are described in WO 87 / 05330 and the literature [Aplin and Wriston, 1981, CRC Crit. Rev. Biochem., pp. It is listed in [259-306].

[0198] The removal of carbohydrate moiety present on the starting antibody construct can be performed chemically or enzymatically. Chemical deglycosylation requires the exposure of the protein to the compound trifluoromethanesulfonic acid, or an equivalent compound. This treatment results in the cleavage of most or all sugars, excluding the linking sugar (N-acetylglucosamine or N-acetylgalactosamine), while leaving the polypeptide intact. Chemical deglycosylation [Hakimuddin et al. , 1987, Arch. Biochem . Biophys . 259:52 and Edge et al. , 1981, Anal. Biochem . [118:131] is described. Enzymatic cleavage of carbohydrate moiety on polypeptides is described in the literature [Thotakura et al.This can be achieved by the use of various endo- and exo-glycosidases as described by [Duskin, 1987, Meth. Enzymol. 138:350]. Glycosylation at potential glycosylation sites can be achieved through the literature [Duskin et al. This can be prevented by the use of the compound tunicamycin as described in [J. Biol. Chem. 257:3105]. Tunicamycin blocks the formation of protein-N-glycoside bonds.

[0199] Other variations of antibody constructs are also conceived herein. For example, other types of covalent variations of antibody constructs include linking the antibody construct to various non-protein polymers, such as polyols, e.g., polyethylene glycol, polypropylene glycol, polyoxyalkylene, or copolymers of polyethylene glycol and polypropylene glycol, in a manner set forth in U.S. Patents No. 4,640,835; No. 4,496,689; No. 4,301,144; No. 4,670,417; No. 4,791,192 or No. 4,179,337, but not limited to these. Additionally, as is known in the art, amino acid substitutions may be made at various positions within the antibody construct to facilitate the addition of polymers, such as PEG.

[0200] In some embodiments, covalent modifications of the antibody constructs of the present invention include the addition of one or more labels. Labels may be bound to the antibody constructs via spacer arms of various lengths to reduce potential steric hindrance. Various methods of label proteins are known in the art and may be used to carry out the present invention. The terms “label” or “labeler” refer to any detectable label. Generally, labels fall into various classifications depending on the analysis in which they are detected—the following examples include, but are not limited to:

[0201] a) radioactive or heavy isotopes, e.g., isotope labels that may be radioactive isotopes or radionuclides (e.g., 3 H, 14 C, 15 N, 35 S, 89 Zr, 90 Y, 99 Tc, 111 In, 125 I, 131 I)

[0202] b) Magnetic markers (e.g., magnetic particles)

[0203] c) Redox active moiety

[0204] d) optical dyes (including, but not limited to, chromophores, phosphors, and fluorophores), such as fluorescent groups (e.g., FITC, rhodamine, lanthanide phosphors), chemiluminescent groups, and fluorophores that may be one of "small molecule" fluorescence or protein fluorescence.

[0205] e) Enzymatic group (e.g., mustard peroxidase, β-galactosidase, luciferase, alkaline phosphatase)

[0206] f) Biotinylated cells

[0207] g) A predetermined polypeptide epitope is recognized by a secondary reporter (e.g., leucine zipper pair sequence, binding site for a secondary antibody, metal binding domain, epitope tag, etc.).

[0208] "Fluorescent label" refers to any molecule that can be detected through its unique fluorescent properties. Suitable fluorescent labels are fluorescein, rhodamine, tetramethylrhodamine, eosin, erythrosine, coumarin, methyl-coumarin, pyrene, malachite green, stilbene, Lucifer yellow, Cascade Blue J, Texas red, IAEDANS, EDANS, BODIPY FL, LC Red 640, Cy 5, Cy 5.5, LC Red 705, Oregon green, and Alexa-Fluor dyes (Alexa-Fluor 350, Alexa-Fluor 430, Alexa-Fluor 488, Alexa-Fluor 546, Alexa-Fluor 568, Alexa-Fluor 594, Alexa-Fluor 633, Alexa Fluor 660, Alexa Fluor 680), Cascade Blue, Cascade Yellow, and R-Pycoerythrin (PE) (Molecular Probes, Eugen, Oregon), FITC, Rhodamine, and Texas Red (Pierce, Rockford, Illinois), Cy5, Cy5.5, and Cy7 (Amersham Life Science, Pittsburgh, Pennsylvania), but are not limited to these. Suitable optical dyes containing fluorescent dyes are described in the literature by Richard P. Haugland [Molecular Probes Handbook].

[0209] Suitable proteinaceous fluorescent labels are also green fluorescent proteins (Chalfie), including the Renilla, Ptilosarcus, or Aequorea species of GFP. et al. , 1994, Science263:802-805), EGFP (Clontech Laboratories, Inc., Genbank Accession Number U55762), blue fluorescent protein (BFP, Quantum Biotechnologies, Inc. 1801 de Maisonneuve Blvd. West, 8th Floor, Montreal, Quebec, Canada H3H 1J9; Stauber, 1998, Biotechniques 24:462-471; Heim et al. , 1996, Curr . Biol . 6:178-182), Enhanced Yellow Fluorescent Protein (EYFP, Clontech Laboratories, Inc.), Luciferase (Ichiki et al. , 1993, J. Immunol . 150:5408-5417), β-galactosidase (Nolan et al. , 1988, Proc . Natl. Academic . Sci . USA . 85:2603-2607) and Renilla (WO92 / 15673, WO95 / 07463, WO98 / 14605, WO98 / 26277, WO99 / 49019, U.S. Patents No. 5,292,658; No. 5,418,155; No. 5,683,888; No. 5,741,668; No. 5,777,079; No. 5,804,387; No. 5,874,304; No. 5,876,995; No. 5,925,558), including but not limited to these.

[0210] Leucine zipper domains are peptides that promote the oligomerization of the proteins in which they were discovered. Leucine zippers were originally identified in several DNA-binding proteins (Landschulz et al. , 1988, Science (240:1759), it has been found in various different proteins. Among the known leucine zippers are naturally derived peptides and their derivatives that undergo dimerization or trimerization. An example of a leucine zipper domain suitable for producing soluble oligomeric proteins is described in PCT application WO 94 / 10308, and a leucine zipper derived from pulmonary surfactant protein D (SPD) is described in the literature [Hoppe et al. , 1994, FEBS Letters It is described in [344:191]. The use of a modified leucine zipper that allows for stable trimerization of fused heterogeneous proteins is described in the literature [Fanslow et al. It is described in [Semin. Immunol. 6:267-78], 1994. In one approach, a recombinant fusion protein containing a CDH3 antibody fragment or a derivative fused to a leucine zipper peptide is expressed in suitable host cells, and the soluble oligomeric CDH3 antibody fragment or derivative formed is recovered from the culture supernatant.

[0211] The antibody compositions of the present invention may also include additional domains, for example, that aid in the isolation of the molecule or relate to a configured pharmacokinetic profile of the molecule. The domains that aid in the isolation of the antibody composition may be selected from peptide motifs or secondarily introduced moieties that can be captured in an isolation method, such as an isolation column. Non-limiting embodiments of such additional domains include peptide motifs known as Myc-tag, HAT-tag, HA-tag, TAP-tag, GST-tag, chitin-binding domain (CBD-tag), maltose-binding protein (MBP-tag), FLAGG-tag, STREP-tag and variants thereof (e.g., STREP II-tag), and His-tag. All antibody constructs disclosed herein, characterized by identified CDRs, preferably comprise a His-tag domain generally known as a repeat of consecutive His residues within the amino acid sequence of the molecule, preferably 5, and more preferably 6 His residues (hexahistidine, see SEQ ID NO. 436). The His-tag may be located, for example, at the N- or C-terminus of the antibody construct, and preferably at the C-terminus. Most preferably, the hexahistidine tag (HHHHHH) is connected via a peptide bond to the C-terminus of the antibody construct according to the present invention.

[0212] The first binding domain of the antibody construct of the present invention binds to human CDH3 on the surface of a target cell. The amino acid sequence of human CDH3 is denoted by SEQ ID NO. 1. In the context of the present invention, the term “on the surface” is understood to mean that the binding domain specifically binds to an epitope or epitope cluster contained within the CDH3 extracellular domain (CDH3 ECD). Accordingly, the first binding domain according to the present invention preferably binds to CDH3 when it is expressed by a naturally expressing cell or cell line and / or by a cell or cell line that is transformed (or stably / transiently) transfected with CDH3. In a preferred embodiment, when CDH3 is used as a “target” or “ligand” molecule in an in vitro assay, e.g., ViaCore or Scatchard, the first binding domain binds to CDH3. "Target cell" may be any prokaryotic or eukaryotic cell expressing CDH3 on its surface, and preferably, the target cell is a cell that is part of the human or animal body, such as a tumor or cancer cell.

[0213] The term "CDH3 ECD" refers to a form of CDH3 that essentially lacks the transmembrane and cytoplasmic domains of CDH3. It will be understood by those skilled in the art that the transmembrane domain identified for the CDH3 polypeptide of the present invention is identified by criteria routinely used in techniques for identifying the corresponding type of hydrophobic domain. Although the exact boundaries of the transmembrane domain may vary, they are likely to be about 5 or fewer amino acids from one of the ends of the domain specifically mentioned herein. A preferred human CDH3 ECD is shown in SEQ ID NO. 3.

[0214] The affinity of the first binding domain to human CDH3 is preferably ≤15 nM, more preferably ≤10 nM, much more preferably ≤5 nM, much more preferably ≤1 nM, much more preferably ≤0.5 nM, much more preferably ≤0.1 nM, and most preferably ≤0.05 nM. The affinity can be measured, for example, in a viacor analysis or a sketched analysis, as described in the examples. Other methods for determining affinity are well known to those skilled in the art.

[0215] T cells, or T lymphocytes, are a type of lymphocyte (the very type of white blood cell) that plays a crucial role in cell-mediated immunity. There are several subsets of T cells, each with distinct functions. T cells can be distinguished from other lymphocytes, such as B cells and NK cells, by the presence of a T cell receptor (TCR) on their cell surface. The TCR results in a recognition antigen bound to a major histocompatibility complex (MHC) molecule and consists of two distinct protein chains. In 95% of T cells, the TCR is composed of alpha (α) and beta (β) chains. When the TCR engages with an antigen peptide and MHC (peptide / MHC complex), the T lymphocyte is activated through a series of biochemical events mediated by associated enzymes, co-receptors, specialized adapter molecules, and activating or released transcription factors.

[0216] The CD3 receptor complex is a protein complex composed of four chains. In mammals, the complex contains a CD3γ (gamma) chain, a CD3δ (delta) chain, and two CD3ε (epsilon) chains. These chains associate with the T cell receptor (TCR) and the so-called ζ (zeta) chain to form the T cell receptor CD3 complex, which generates an activation signal in T lymphocytes. The CD3γ (gamma), CD3δ (delta), and CD3ε (epsilon) chains are highly related cell surface proteins of the immunoglobulin superfamily containing a single extracellular immunoglobulin domain. The intracellular tail of the CD3 molecule contains a single conserved motif known as the immunoreceptor tyrosine-based activation motif, or ITAM for short, which is essential for the signaling capacity of the TCR. The CD3 epsilon molecule is located on chromosome 11 in humans. CD3E It is a polypeptide encoded by a gene. The most desirable CD3 binding epitope corresponds to amino acid residues 1 to 27 of the human CD3 epsilon extracellular domain.

[0217] The redirected lysis of target cells through the recruitment of T cells by multispecific, or at least bispecific, antibody constructs entails the formation of lysis synapses and the delivery of perforins and granzymes. The coupled T cells can perform sequential target cell lysis and are not affected by immune evasion mechanisms that interfere with peptide antigen processing and presentation or clonal T cell differentiation; see, for example, WO 2007 / 042261.

[0218] Cytotoxicity mediated by CDH3 / CD3 bispecific antibody constructs can be measured by various methods. Effector cells may be, for example, stimulated, abundant (human) CD8-positive T cells or unstimulated (human) peripheral blood monocytes (PBMCs). If the target cells are of macaque origin, express macaque CDH3, or are transfected with it, the effector cells must also be of macaque origin, e.g., a macaque T cell line, e.g., 4119LnPx. The target cells must express CDH3, e.g., human or macaque CDH3 (at least its extracellular domain). The target cells may be cell lines (e.g., CHO) stably or transiently transfected with CDH3, e.g., human or macaque CDH3. Alternatively, the target cells may be CDH3-positive natural-expressing cell lines. Typically, EC50 values ​​are expected to be lower for target cell lines expressing higher levels of CDH3 on the cell surface. The effector-to-target cell (E:T) ratio is typically about 10:1, but may also vary. The cytotoxic activity of CDH3 / CD3 bispecific antibody constructs can be measured by 51-chromium release assay (incubation time of about 18 hours) or FACS-based cytotoxicity assay (incubation time of about 48 hours). Variations in incubation time (cytotoxic response) are also possible. Other methods for measuring cytotoxicity are well known to those skilled in the art and include MTT or MTS assays, ATP-based assays including bioluminescence assays, sulforodamine B (SRB) assays, WST assays, colony formation assays, and ECIS techniques.

[0219] The cytotoxic activity mediated by the CDH3 / CD3 bispecific antibody construct of the present invention is preferably measured in a cell-based cytotoxicity assay. It can also be measured in a 51-chromium release assay. This corresponds to the EC10 of the maximum effective concentration (the concentration of the antibody construct that induces a cytotoxic response between the reference and the maximum value).50 Expressed as a value. Preferably, the EC of the CDH3xCD3 bispecific antibody construct. 50 The value is ≤5000 pg / ml or ≤4000 pg / ml, more preferably ≤3000 pg / ml or ≤2000 pg / ml, much more preferably ≤1000 pg / ml or ≤500 pg / ml, much more preferably ≤400 pg / ml or ≤300 pg / ml, much more preferably ≤200 pg / ml, much more preferably ≤100 pg / ml, much more preferably ≤50 pg / ml, much more preferably ≤20 pg / ml or ≤10 pg / ml, and most preferably ≤5 pg / ml.

[0220] The EC given above 50 Values ​​may be measured in different analyses. Those skilled in the art recognize that when stimulated / enriched CD8+ T cells are used as effector cells, EC50 values ​​may be expected to be lower compared to unstimulated PBMCs. Furthermore, when target cells express a very large number of target antigens compared to low-target-expression rats, EC50 values ​​may be expected to be lower. For example, when stimulated / enriched human CD8+ T cells are used as effector cells (where one of the CDH3 transfected cells, e.g., CHO cells, or a CDH3-positive natural-expressing cell line, e.g., A431, is used as the target cell), the EC of the CDH3xCD3 bispecific antibody construct 50 The value is preferably ≤1000 pg / mL, more preferably ≤500 pg / mL, much more preferably ≤250 pg / mL, much more preferably ≤100 pg / mL, much more preferably ≤50 pg / mL, much more preferably ≤10 pg / mL, and most preferably ≤5 pg / mL. When human PBMCs are used as effector cells, the EC of the CDH3xCD3 bispecific antibody construct 50The value is preferably ≤5000 pg / mL or ≤4000 pg / mL (particularly when the target cell is a CDH3-positive natural expressor cell line, e.g., A431), more preferably ≤2000 pg / mL (particularly when the target cell is a CDH3-transfected cell, e.g., CHO cell), more preferably ≤1000 pg / mL or ≤500 pg / mL, much more preferably ≤200 pg / mL, much more preferably ≤150 pg / mL, much more preferably ≤100 pg / mL, and most preferably ≤50 pg / mL or lower. When a macaque T cell line, e.g., LnPx4119, is used as the effector cell and a macaque CDH3-transfected cell line, e.g., CHO cell, is used as the target cell line, the EC of the CDH3xCD3 bispecific antibody construct 50 The value is preferably ≤2000 pg / ml or ≤1500 pg / ml, more preferably ≤1000 pg / ml or ≤500 pg / ml, much more preferably ≤300 pg / ml or ≤250 pg / ml, much more preferably ≤100 pg / ml, and most preferably ≤50 pg / ml.

[0221] Preferably, the CDH3 / CD3 bispecific antibody construct of the present invention does not induce or mediate lysis, or essentially does not induce or mediate lysis of CDH3-negative cells, e.g., CHO cells. The terms “does not induce lysis,” “essentially does not induce lysis,” “does not mediate lysis,” or “essentially does not mediate lysis” mean that the antibody construct of the present invention does not induce or mediate lysis of more than 30%, preferably 20% or less, more preferably 10% or less, particularly preferably 9%, 8%, 7%, 6%, or 5% or less of CDH3-negative cells, thereby establishing 100% lysis of CDH3-positive cell lines, e.g., A431. This applies to concentrations of the antibody construct up to 500 nM. Those skilled in the art know how to measure cell lysis without further effort. Furthermore, this specification teaches specific instructions for a method of measuring cell lysis.

[0222] The difference in cytotoxic activity between the monomeric and dimeric isoforms of individual CDH3 / CD3 bispecific antibody constructs is referred to as the "efficacy gap." This efficacy gap, for example, the EC of the monomeric and dimeric forms of the molecule 50 It can be calculated as a ratio between. The efficacy gap of the CDH3 / CD3 bispecific antibody construct of the present invention is preferably ≤ 5, more preferably ≤ 4, much more preferably ≤ 3, much more preferably ≤ 2, and most preferably ≤ 1. For example, the efficacy gap for CDH3-13 was determined to be 0.5, the efficacy gap for CDH3-13xCD3-HLE(Fc) was determined to be 0.9, the efficacy gap for CDH3-25 was determined to be 0.7, and the efficacy gap for CDH3-25xCD3-HALB was likewise determined to be 0.7.

[0223] The first and / or second (or any additional) binding domain(s) of the antibody construct of the present invention are preferably cross-species specific to members of the order Mammalia of primates. Cross-species specific CD3 binding domains are described, for example, in WO 2008 / 119567. According to one embodiment, in addition to binding to human CDH3 and human CD3, respectively, the first binding domain and / or second binding domain are also New World primates (e.g., marmosets ( Callithrix jacchus ), downy tamarin ( Saguinus Oedipus ) or squirrel monkey( Saimiri sciureus It will bind to CDH3 / CD3 of primates including (but not limited to) Old World primates (such as baboons and macaques), gibbons, and non-human hominidae.

[0224] In one aspect of the present invention, the first binding domain binds to human CDH3, and further binds to macaque CDH3, e.g., CDH3 of Philippine monkey (SEQ No. 5), and more preferably to macaque CDH3 ECD. The affinity of the first binding domain for macaque CDH3 is preferably ≤15 nM, more preferably ≤10 nM, much more preferably ≤5 nM, much more preferably ≤1 nM, much more preferably ≤0.5 nM, much more preferably ≤0.1 nM, and most preferably ≤0.05 nM or even ≤0.01 nM.

[0225] Preferably, the affinity gap of the antibody construct according to the present invention for conjugating macaque CDH3 to human CDH3 [ma CDH3:hu CDH3] (as determined, e.g., by ViaCore or by Scatchard analysis, see Examples) is 0.1 to 10, more preferably 0.2 to 5, much more preferably 0.3 to 2.5, much more preferably 0.4 to 2, and most preferably 0.5 to 1.

[0226] In one embodiment of the antibody composition of the present invention, the second binding domain binds to human and silk monkey, downy-haired tamarin, or squirrel monkey CD3 epsilon. Preferably, the second binding domain binds to an extracellular epitope of these CD3 epsilon chains. It is also thought that the second binding domain binds to an extracellular epitope of human and macaca CD3 epsilon chains. The most preferred epitope of CD3 epsilon is contained within amino acid residues 1 to 27 of the human CD3 epsilon extracellular domain. Much more specifically, the epitope contains at least the amino acid sequence Gln-Asp-Gly-Asn-Glu. Both silk monkeys and downy-haired tamarins are kalitricidae ( Callitrichidae While they are New World primates belonging to the family ), squirrel monkeys are pygmy ( Cebidae It is a New World primate belonging to the family.

[0227] It is particularly preferable that for the antibody construct of the present invention, the second binding domain that binds to human CD3 on the surface of a T cell comprises a VL region including CDR-L1, CDR-L2, and CDR-L3 selected from the following:

[0228] (a) CDR-L1 as presented in sequence number 27 of WO 2008 / 119567, CDR-L2 as presented in sequence number 28 of WO 2008 / 119567, and CDR-L3 as presented in sequence number 29 of WO 2008 / 119567;

[0229] (b) CDR-L1 presented at sequence number 117 of WO 2008 / 119567, CDR-L2 presented at sequence number 118 of WO 2008 / 119567, and CDR-L3 presented at sequence number 119 of WO 2008 / 119567; and

[0230] (c) CDR-L1 presented at sequence number 153 of WO 2008 / 119567, CDR-L2 presented at sequence number 154 of WO 2008 / 119567, and CDR-L3 presented at sequence number 155 of WO 2008 / 119567.

[0231] In an alternatively preferred embodiment of the antibody construct of the present invention, a second binding domain that binds to human CD3 on a T cell surface comprises a VH region comprising CDR-H1, CDR-H2, and CDR-H3 selected from the following:

[0232] (a) CDR-H1 presented at sequence number 12 of WO 2008 / 119567, CDR-H2 presented at sequence number 13 of WO 2008 / 119567, and CDR-H3 presented at sequence number 14 of WO 2008 / 119567;

[0233] (b) CDR-H1 presented at sequence number 30 of WO 2008 / 119567, CDR-H2 presented at sequence number 31 of WO 2008 / 119567, and CDR-H3 presented at sequence number 32 of WO 2008 / 119567;

[0234] (c) CDR-H1 presented at sequence number 48 of WO 2008 / 119567, CDR-H2 presented at sequence number 49 of WO 2008 / 119567, and CDR-H3 presented at sequence number 50 of WO 2008 / 119567;

[0235] (d) CDR-H1 presented at sequence number 66 of WO 2008 / 119567, CDR-H2 presented at sequence number 67 of WO 2008 / 119567, and CDR-H3 presented at sequence number 68 of WO 2008 / 119567;

[0236] (e) CDR-H1 presented at sequence number 84 of WO 2008 / 119567, CDR-H2 presented at sequence number 85 of WO 2008 / 119567, and CDR-H3 presented at sequence number 86 of WO 2008 / 119567;

[0237] (f) CDR-H1 presented at sequence number 102 of WO 2008 / 119567, CDR-H2 presented at sequence number 103 of WO 2008 / 119567, and CDR-H3 presented at sequence number 104 of WO 2008 / 119567;

[0238] (g) CDR-H1 presented at sequence number 120 of WO 2008 / 119567, CDR-H2 presented at sequence number 121 of WO 2008 / 119567, and CDR-H3 presented at sequence number 122 of WO 2008 / 119567;

[0239] (g) CDR-H1 presented at sequence number 138 of WO 2008 / 119567, CDR-H2 presented at sequence number 139 of WO 2008 / 119567, and CDR-H3 presented at sequence number 140 of WO 2008 / 119567;

[0240] (i) CDR-H1 presented at sequence number 156 of WO 2008 / 119567, CDR-H2 presented at sequence number 157 of WO 2008 / 119567, and CDR-H3 presented at sequence number 158 of WO 2008 / 119567;

[0241] (j) CDR-H1 presented at sequence number 174 of WO 2008 / 119567, CDR-H2 presented at sequence number 175 of WO 2008 / 119567, and CDR-H3 presented at sequence number 176 of WO 2008 / 119567.

[0242] Additionally, for the antibody composition of the present invention, the second binding domain that binds to human CD3 on the surface of a T cell preferably comprises a VL region selected from the group consisting of VL regions as presented in SEQ ID NOs 35, 39, 125, 129, 161 or 165 of WO 2008 / 119567.

[0243] Alternatively, the second binding domain that binds to human CD3 on the surface of a T cell preferably comprises a VH region selected from the group consisting of VH regions as presented in sequence numbers 15, 19, 33, 37, 51, 55, 69, 73, 87, 91, 105, 109, 123, 127, 141, 145, 159, 163, 177, or 181 of WO 2008 / 119567.

[0244] More preferably, the antibody construct of the present invention features a second binding domain that binds to human CD3 on the surface of a T cell comprising a VL region and a VH region selected from the group consisting of:

[0245] (a) the VL region presented in sequence number 17 or 21 of WO 2008 / 119567 and the VH region presented in sequence number 15 or 19 of WO 2008 / 119567;

[0246] (b) the VL region presented in sequence number 35 or 39 of WO 2008 / 119567 and the VH region presented in sequence number 33 or 37 of WO 2008 / 119567;

[0247] (c) VL region presented in SEQ ID NO. 53 or 57 of WO 2008 / 119567 and VH region presented in SEQ ID NO. 51 or 55 of WO 2008 / 119567;

[0248] (d) VL region presented in sequence number 71 or 75 of WO 2008 / 119567 and VH region presented in sequence number 69 or 73 of WO 2008 / 119567;

[0249] (e) the VL region presented in sequence number 89 or 93 of WO 2008 / 119567 and the VH region presented in sequence number 87 or 91 of WO 2008 / 119567;

[0250] (f) VL region presented in sequence number 107 or 111 of WO 2008 / 119567 and VH region presented in sequence number 105 or 109 of WO 2008 / 119567;

[0251] (g) VL region as presented in sequence number 125 or 129 of WO 2008 / 119567 and VH region as presented in sequence number 123 or 127 of WO 2008 / 119567;

[0252] (h) VL region as presented in SEQ ID NO. 143 or 147 of WO 2008 / 119567 and VH region as presented in SEQ ID NO. 141 or 145 of WO 2008 / 119567;

[0253] (i) the VL region presented in sequence number 161 or 165 of WO 2008 / 119567 and the VH region presented in sequence number 159 or 163 of WO 2008 / 119567;

[0254] (j) VL region presented in sequence number 179 or 183 of WO 2008 / 119567 and VH region presented in sequence number 177 or 181 of WO 2008 / 119567.

[0255] The binding domain disclosed in WO 2008 / 119567, which binds to human CD3, is also illustrated in the present SEQ ID NOs 445 to 537.

[0256] According to a preferred embodiment of the antibody composition of the present invention, the binding domain and, in particular, the second binding domain (binding to human CD3 on the surface of a T cell) have the following format: the pair of VH and VL regions is in the format of a short-chain antibody (scFv). The VH and VL regions are arranged in the order of VH-VL or VL-VH. It is preferable that the VH region be positioned at the N-terminus of the linker sequence and the VL region be positioned at the C-terminus of the linker sequence.

[0257] A preferred embodiment of the antibody construct of the present invention described above features a second binding domain that binds to human CD3 on the surface of a T cell, comprising an amino acid sequence selected from the group consisting of SEQ ID NOs 23, 25, 41, 43, 59, 61, 77, 79, 95, 97, 113, 115, 131, 133, 149, 151, 167, 169, 185, or 187 of WO 2008 / 119567.

[0258] In one embodiment of the present invention, the antibody composition has an amino acid sequence selected from the group consisting of the corresponding sequences presented in SEQ ID NO. 158, SEQ ID NO. 168, SEQ ID NO. 178, SEQ ID NO. 188, SEQ ID NO. 198, SEQ ID NO. 208, SEQ ID NO. 218, SEQ ID NO. 228, SEQ ID NO. 238 and SEQ ID NO. 248.

[0259] In another embodiment of the present invention, the antibody composition has an amino acid sequence selected from the group consisting of the corresponding sequences as presented in SEQ ID NO. 379, SEQ ID NO. 380, SEQ ID NO. 381, SEQ ID NO. 382, ​​SEQ ID NO. 383, SEQ ID NO. 384, SEQ ID NO. 385, SEQ ID NO. 386, SEQ ID NO. 387, SEQ ID NO. 388 and SEQ ID NO. 389.

[0260] In another embodiment of the present invention, the antibody composition has an amino acid sequence selected from the group consisting of the corresponding sequences as presented in SEQ ID NO. 288, SEQ ID NO. 298, SEQ ID NO. 308, SEQ ID NO. 318, SEQ ID NO. 328, SEQ ID NO. 338, SEQ ID NO. 348 and SEQ ID NO. 358.

[0261] Amino acid sequence modifications of the antibody constructs described herein are also conceivable. For example, it may be desirable to improve the binding affinity and / or other biological properties of the antibody construct. Amino acid sequence variants of the antibody construct are prepared by introducing appropriate nucleotide changes into the antibody construct nucleic acid or by peptide synthesis. All amino acid modifications described below must result in an antibody construct that still retains the intended biological activity (binding to CDH3 and to CD3) of the unmodified parent molecule.

[0262] The term “amino acid” or “amino acid residue” typically refers to an amino acid having its recognized definition in the art, such as an amino acid selected from the group consisting of alanine (Ala or A); arginine (Arg or R); asparagine (Asn or N); asphaltic acid (Asp or D); cysteine ​​(Cys or C); glutamine (GIn or Q); glutamic acid (GIu or E); glycine (GIy or G); histidine (His or H); isoleucine (He or I); leucine (Leu or L); lysine (Lys or K); methionine (Met or M); phenylalanine (Phe or F); proline (Pro or P); serine (Ser or S); threonine (Thr or T); tryptophan (Trp or W); tyrosine (Tyr or Y); and valine (VaI or V), although modified, synthetic, or rare amino acids may be used if desired. Generally, amino acids are grouped into those having nonpolar side chains (e.g., Ala, Cys, He, Leu, Met, Phe, Pro, VaI); negatively charged side chains (e.g., Asp, GIu); positively charged side chains (e.g., Arg, His, Lys); or non-charged polar side chains (e.g., Asn, Cys, GIn, GIy, His, Met, Phe, Ser, Thr, Trp, and Tyr).

[0263] Amino acid modifications include, for example, deletions and / or insertions and / or substitutions of residues within the amino acid sequence of the antibody construct. Any combination of deletions, insertions, and substitutions is made to arrive at a final construct having the desired characteristics. Amino acid changes may also alter post-translational processes of the antibody construct, for example, by changing the number or location of glycosylation sites.

[0264] For example, 1, 2, 3, 4, 5, or 6 amino acids may be inserted or deleted in each CDR (of course, depending on their length), while 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 25 amino acids may be inserted or deleted in each FR. Preferably, amino acid sequence insertion includes the insertion of single or multiple amino acid residues into the sequence, as well as amino- and / or carboxyl-terminal fusions with a length ranging from 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 residues for polypeptides containing 100 or more residues. The insertion variant of the antibody construct of the present invention comprises a fusion of an enzyme to the N-terminus or C-terminus of the antibody construct or a fusion to a polypeptide that increases the serum half-life of the antibody construct.

[0265] The sites of greatest interest for substituted mutagenicity include the CDR of the heavy chain and / or light chain, particularly the hypervariable region, but changes to the FR of the heavy chain and / or light chain are also conceivable. The substitutions are preferably conservative substitutions as described herein. Preferably, 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acids may be substituted in the CDR, while 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 25 amino acids may be substituted in the framework region (FR) depending on the length of the CDR or FR. For example, if the CDR sequence contains 6 amino acids, it is conceivable that 1, 2, or 3 of these amino acids are substituted. Similarly, if the CDR sequence contains 15 amino acids, it is thought that 1, 2, 3, 4, 5, or 6 of these amino acids are substituted.

[0266] A useful method for identifying specific residues or regions of an antibody construct that are desirable sites for mutagenesis is referred to as “alanine scanin mutagenesis” as described in the literature [Cunningham and Wells in Science, 244: 1081-1085 (1989)]. In this specification, residues or groups of target residues within the antibody construct are identified (e.g., charged residues, e.g., arg, asp, his, lys, and glu) and replaced by neutral or negatively charged amino acids (most preferably alanine or polyalanine) to affect the interaction between the amino acid and the epitope.

[0267] Subsequently, the corresponding amino acid position demonstrating functional sensitivity to the substitution is improved by introducing additional or different variants at or for the substitution site. Thus, while the site or region for introducing the amino acid sequence is predetermined, the mutation characteristics themselves do not need to be predetermined. For example, to analyze or optimize mutation performance at a given site, alanine scanning or random mutagenesis can be performed at a target codon or region, and variants of the expressed antibody construct are selected for the optimal combination of desired activities. Techniques for inducing substitution mutations at predetermined sites within DNA having known sequences, e.g., M13 primer mutagenesis and PCR mutagenesis, are well known. Selection of mutants is performed using the analysis of antigen-binding activity, e.g., CDH3 or CD3 binding.

[0268] Generally, amino acids are substituted in one or more or all of the CDRs of the heavy chain and / or light chain, and the resulting "substituted" sequence is preferably at least 60%, more preferably 65%, much more preferably 70%, particularly preferably 75%, and even more particularly preferably 80% identical to the "original" CDR sequence. This means that the length of the CDR depends to the same degree as the "substituted" sequence. For example, a CDR having five amino acids is preferably 80% identical to its substituted sequence in order to have at least one substituted amino acid. Thus, the CDRs of antibody constructs may have different degrees of identity with their substituted sequences, for example, CDRL1 may have 80% while CDRL3 may have 90%.

[0269] A preferred substitution (or replacement) is a conservative substitution. However, any substitution (a non-conservative substitution or one or more from the “exemplary substitutions” listed in Table 1 below) is considered, provided that the antibody construct retains its ability to bind through the first binding domain and to bind to CD3 or CD3 epsilon through the second binding domain, and / or its CDR is identical to the subsequently substituted sequence (at least 60%, more preferably 65%, much more preferably 70%, particularly preferably 75%, more particularly preferably 80% identical to the “original” CDR sequence).

[0270] Conservative substitutions are shown in Table 1 under the heading "Desirable substitutions." If such substitutions result in a change in biological activity, more substantial changes, such as those designated as "Exemplary substitutions" in Table 1 or further described below for amino acid classifications, may be introduced and the product selected for the desired characteristics.

[0271] 아미노산 치환 본래 예시적인 치환 바람직한 치환 Wing (A) val, leu, ile wal Arg (R) lys, gln, asn lys Asn (N) gln, his, asp, lys, arg gln Asp (D) glu, asn glu Cys (C) to be, wing be Gln (Q) ass, glu donkey Glu (E) asp, gln asp Gly (G) To the to the His (H) asn, gln, lys, arg arg Ile (I) leu, val, met, ala, phe leu Leu (L) 노르류신, ile, val, met, ala ile Lys (K) arg, gln, asn arg Met (M) leu, phe, ile leu Phe (F) leu, val, ile, ala, tyr tyr Pro (P) To the to the Ser (S) Thr thr Thr (T) Ser ser Trp (W) tyr, phe tyr Tyr (Y) trp, phe, thr, ser phe Val (V) ile, leu, met, phe, ala leu

[0272] Substantial modifications to the biological properties of the antibody constructs of the present invention are made by selecting substitutions that differ significantly in (a) the structure of the polypeptide backbone at the substitution site, for example, as a sheet or helical stereolocule, (b) the charge or hydrophobicity of the molecule at the target site, or (c) their effect on maintaining the bulk of the side chain. Naturally derived residues are divided into groups based on common side chain characteristics: (1) hydrophobic: norleucine, met, ala, val, leu, ile; (2) neutral hydrophilic: cys, ser, thr; (3) acidic: asp, glu; (4) basic: asn, gin, his, lys, arg; (5) residues affecting orientation: gly, pro; and (6) aromatic: trp, tyr, phe.

[0273] Non-conservative substitution will involve exchanging a member of one of these classes for another. Any cysteine ​​residue not required to maintain the proper stereochemical orientation of the antibody construct may generally be substituted with serine to improve the oxidative stability of the molecule and to prevent abnormal cross-linking. Conversely, cysteine ​​bond(s) may be added to the antibody to improve its stability (particularly when the antibody is an antibody fragment, e.g., an Fv fragment).

[0274] Regarding amino acid sequences, sequence identity and / or similarity are [Smith and Waterman, 1981, Adv . Appl . Math. Local sequence identity algorithm of 2:482], literature [Needleman and Wunsch, 1970, J. Mol . Biol . Sequence identity alignment algorithm of 48:443], literature [Pearson and Lipman, 1988, Proc . Nat. Acad . Sci . USA . Search for similarity methods of [85:2444], computerized execution of these algorithms (GAP, BESTFIT, FASTA, and TFASTA in the Wisconsin Genetics Software Package of the Genetics Computer Group, located at 575 Madison Science Drive, Wisconsin), preferably using default settings or by examination, literature [Devereux et al. , 1984, Nucl Acid Res.It is determined by using standard techniques known in the art, including but not limited to Best Fit sequence programs as described in [12:387-395]. Preferably, the identity percentage is calculated by FastDB based on the following parameters: mismatch penalty 1; gap penalty 1; gap size penalty 0.33; and linkage penalty 30, ["Current Methods in Sequence Comparison and Analysis," Macromolecule Sequencing and Synthesis, Selected Methods and Applications, pp 127-149 (1988), Alan R. Liss, Inc].

[0275] An example of a useful algorithm is PILEUP. PILEUP generates multiple sequence alignments from groups of related sequences using progressive, pairwise alignment. It can also plot a tree representing the clustering relationships used to generate the alignments. PILEUP is [Feng & Doolittle, 1987, J. Mol . Evol . It utilizes a simplification of the progressive alignment method of [35:351-360]; the said method uses [Higgins and Sharp, 1989, CABIOS Similar to that described in [5:151-153]. Useful PILEUP parameters including a default gap weight of 3.00, a default gap length weight of 0.10, and a weighted end gap.

[0276] Other examples of useful algorithms can be found in the literature [Altschul et al. , 1990, J. Mol . Biol . 215:403-410; Altschul et al. , 1997, Nucleic Acids Res.25:3389-3402; and Karin et al. , 1993, Proc . Natl . Acad . Sci . USA . It is a BLAST algorithm described in [90:5873-5787]. A particularly useful BLAST program is [Altschul et al. , 1996, Methods in Enzymology This is the WU-BLAST-2 program obtained from [266:460-480]. WU-BLAST-2 uses several search parameters that are mostly set to default values. The adjustable parameters are set to the following values: duplicate extension=1, duplicate fraction=0.125, word threshold (T)=II. The HSP S and HSP S2 parameters are dynamic values, and while the sequence of interest is searched, the values ​​are established by the program itself based on the composition of the specific sequence and the composition of the specific database, which can be adjusted to increase sensitivity.

[0277] Additional useful algorithms are in the literature [Altschul et al. , 1993, Nucl Acids Res. This is a gapped BLAST as reported by [25:3389-3402]. The gapped BLAST uses a BLOSUM-62 permutation score; a threshold T parameter set to 9; a 2-hit method to trigger ungapped extension, a charge gap length of k with a cost of 10+k; Xu set to 16 for the database search step, and Xg set to 40, and 67 for the output step of the algorithm. Gapped alignment is triggered by a score corresponding to approximately 22 bits.

[0278] Generally, the amino acid homology, similarity, or identity between individual variant CDRs is at least 60% for the sequences presented herein, more typically preferably at least 65% or 70%, more preferably at least 75% or 80%, much more preferably at least 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, and nearly 100% increased homology or identity. In a similar manner, for the nucleic acid sequences of the binding proteins identified herein, the “nucleic acid sequence identity percentage (%)” is defined as the percentage of nucleotide residues in candidate sequences that are identical to nucleotide residues in the coding sequence of the antibody construct. The specific method utilizes the BLASTN module of WU-BLAST-2 set to default parameters, with the duplicate expansion and duplicate fraction set to 1 and 0.125, respectively.

[0279] Generally, the nucleic acid sequence homology, similarity, or identity between the nucleotide sequence encoding an individual variant CDR and the nucleotide sequence described herein is at least 60%, and more typically, preferably at least 65%, 70%, 75%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%, and has increased homology or identity of nearly 100%. Accordingly, the “variant CDR” has homology, similarity, or identity embodied with respect to the parent CDR of the present invention and shares biological functions comprising, but not limited to, at least 60%, 65%, 70%, 75%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the specificity and / or activity of the parent CDR.

[0280] In one embodiment, the bispecific antibody construct of the present invention exhibits a high monomer yield under standard study-scale conditions, for example, in a standard two-step purification process. Preferably, the monomer yield of the antibody construct according to the present invention is ≥ 0.25 mg / L supernatant, more preferably ≥ 0.5 mg / L, much more preferably ≥ 1 mg / L, and most preferably ≥ 3 mg / L supernatant.

[0281] Likewise, the yield of the dimeric antibody construct isoform and the resulting monomer percentage of the antibody construct (i.e., monomer : (monomer + dimer)) can be determined. The productivity of the monomeric and dimeric antibody constructs and the calculated monomer percentage can be obtained, for example, from the SEC purification step of the culture supernatant from standardized study-scale production in a roller bottle. In one embodiment, the monomer percentage of the antibody construct is ≥ 80%, more preferably ≥ 85%, much more preferably ≥ 90%, and most preferably ≥ 95%.

[0282] In additional embodiments, the percentage of identity of the antibody construct according to the present invention with respect to the human germline is ≥ 70% or ≥ 75%, more preferably ≥ 80% or ≥ 85%, much more preferably ≥ 90%, and most preferably ≥ 95%. See Example 7. Identity with respect to the human antibody germline gene product is considered to be an important feature that reduces the risk of therapeutic proteins inducing an immune response to the drug in patients during treatment. Hwang & Foote (“Immunogenicity of engineered antibodies”; Methods 36 (2005) 3-10) demonstrate that a reduction in the non-human portion of a drug antibody construct leads to a reduction in the risk of inducing anti-drug antibodies in patients during treatment. By comparing a thorough number of clinically evaluated antibody drugs with their respective immunogenicity data, they show a tendency to make proteins less immunogenic (average 5.1% of patients) than antibodies carrying the unmodified non-human V region (average 23.59% of patients). Therefore, a higher degree of identity with respect to human sequences is desirable for V-region-based protein therapies in the form of antibody constructs. For this purpose of determining germline identity, the V-region of VL can be aligned with the amino acid sequences of human germline V and J segments using Vector NTI software (http: / / vbase.mrc-cpe.cam.ac.uk / ), and the amino acid sequences are calculated as a percentage by dividing the number of identical amino acid residues by the total number of amino acid residues in VL. It can be identical to the VH segment, except that VH CDR3 may be identical due to its high diversity and the lack of existing human germline VH CDR3 alignment status (http: / / vbase.mrc-cpe.cam.ac.uk / ). Subsequently, recombination techniques may be used to increase sequence identity with respect to human antibody germline genes.

[0283] In one embodiment, the antibody construct has a preferred plasma stability (ratio of EC50 to EC50 w / o plasma) of ≤ 5, more preferably ≤ 4 or ≤ 3.5, much more preferably ≤ 3 or ≤ 2.5, and most preferably ≤ 2 or ≤ 1.5 or ≤ 1. The plasma stability of the antibody construct can be tested by determining the EC50 in a 51-chromium-releasing cytotoxicity assay following incubation of the construct in human plasma at 37°C for 24 hours. In the cytotoxicity assay, the effector cells may be stimulated abundant human CD8-positive T cells. The target cells may be, for example, CHO cells transfected with human CDH3. The effector-to-target cell (E:T) ratio may be selected as 10:1. The human plasma pool used for this purpose is derived from the blood of a healthy donor collected by an EDTA-coated syringe. Cellular components are removed by centrifugation, and the supernatant plasma phase is collected and subsequently pooled. As a control, the antibody construct is diluted immediately before the cytotoxicity assay in RPMI-1640 medium. Plasma stability is calculated as the ratio of EC50 (after plasma incubation) to EC50 (control). See Example 11.

[0284] It is desirable that the conversion of the monomer to the dimer of the antibody construct of the present invention be low. The conversion can be measured under different conditions and analyzed by high-performance size exclusion chromatography. For example, the incubation of the monomer isoform of the antibody construct can be carried out for 7 days at 37°C and, for example, at a concentration of 100 µg / ml or 250 µg / ml in an incubator. Under these conditions, it is desirable that the antibody construct of the present invention exhibit a dimer percentage of ≤5%, more preferably ≤4%, much more preferably ≤3%, much more preferably ≤2.5%, much more preferably ≤2%, much more preferably ≤1.5%, and most preferably ≤1%. See Example 9.

[0285] In addition, it is desirable that the bispecific antibody construct of the present invention provides a very low dimer conversion after a number of freeze / thaw cycles. For example, the antibody construct monomer is controlled to a concentration of 250 μg / ml in a general formulation buffer, and three freeze / thaw cycles (freezing at -80°C for 30 minutes followed by thawing at room temperature for 30 minutes) are followed by high-performance SEC to determine the percentage of the initial monomer antibody construct converted into a dimer antibody construct. Preferably, the dimer percentage of the bispecific antibody construct is, for example, ≤5%, more preferably ≤4%, much more preferably ≤3%, much more preferably ≤2.5%, much more preferably ≤2%, much more preferably ≤1.5%, and most preferably ≤1% after three freeze / thaw cycles.

[0286] The bispecific antibody construct of the present invention exhibits desirable thermal stability with an agglutination temperature of preferably greater than 50°C or greater than 52°C, more preferably greater than 54°C or greater than 55°C, much more preferably greater than 56°C or greater than 57°C, and most preferably greater than 58°C or greater than 59°C. The thermal stability parameter can be determined for the antibody agglutination temperature as follows: An antibody solution at a concentration of 250 µg / mL is transferred to a disposable cuvette and placed in a dynamic light scattering device. The sample is heated from 40°C to 70°C at a heating rate of 0.5°C / min with a constant acquisition of the measurement radius. The increase in the radius indicating the melting of the protein and agglutination is used to calculate the antibody agglutination temperature. See Example 10.

[0287] Alternatively, the temperature melting curve is determined by differential scanning calorimetry (DSC) to determine the intrinsic biophysical protein stability of the antibody construct. These experiments are performed using a VP-DSC instrument from MicroCal LLC (located in Northampton, Massachusetts, USA). The energy uptake of samples containing the antibody construct is reported to be between 20°C and 90°C compared to samples containing only the formulation buffer. The antibody construct is adjusted to a final concentration of, for example, 250 µg / mL in SEC execution buffer. For the recording of each melting curve, the overall sample temperature is increased in steps. At each temperature, the sample's T energy uptake and the formulation buffer reference are recorded. The difference between the sample's energy uptake Cp (kcal / mol / °C) and the reference is plotted for each temperature. The melting temperature is defined as the temperature at which the energy uptake reaches its first maximum.

[0288] Furthermore, it is considered that the CDH3xCD3 bispecific antibody of the present invention does not cross-react (i.e., does not bind) with human CDH3 paralogs CDH1, CDH2, CDH4, and CDH5. Furthermore, it is considered that the CDH3xCD3 bispecific antibody of the present invention does not cross-react (i.e., does not bind) with macaque / cyano CDH3 paralogs CDH1, CDH2, CDH4, and CDH5. See Example 6.

[0289] The CDH3xCD3 bispecific antibody of the present invention is also thought to have a turbidity of ≤ 0.1, most preferably ≤ 0.05 (measured by OD340 after concentration of the monomeric antibody purified to 2.5 mg / mL and overnight incubation). See Example 12.

[0290] In a further embodiment, the antibody construct according to the present invention is stable at an acidic pH. The better it tolerates, the more the antibody construct behaves at a non-physiological pH, e.g., pH 5.5 (e.g., the pH required to perform cation exchange chromatography), and the higher the recovery of the antibody construct eluted from the ion exchange column relative to the total amount of loaded protein. At pH 5.5, the recovery of the antibody construct from the ion (e.g., cation) exchange column is preferably ≥ 30%, more preferably ≥ 40%, even more preferably ≥ 50%, much more preferably ≥ 60%, much more preferably ≥ 70%, much more preferably ≥ 80%, and most preferably ≥ 90%.

[0291] Furthermore, it is thought that the bispecific antibody composition of the present invention exhibits therapeutic efficacy or anti-tumor activity. This can be evaluated, for example, in a study as disclosed in Example 14.

[0292] Those skilled in the art will know how to modify or adapt specific parameters of this study, such as the number of tumor cells injected, the injection site, the number of human T cells transplanted, the amount of bispecific antibody construct to be administered, and the timetable, while still achieving meaningful and reproducible results. Preferably, the tumor growth inhibition T / C[%] is ≤ 70 or ≤ 60, more preferably ≤ 50 or ≤ 40, much more preferably ≤ 30 or ≤ 20, and most preferably ≤ 10 or ≤ 5 or even ≤ 2.5.

[0293] The present invention further provides a polynucleotide / nucleic acid molecule encoding the antibody construct of the present invention.

[0294] Polynucleotides are biopolymers composed of 13 or more nucleotide monomers covalently bonded within a chain. DNA (e.g., cDNA) and RNA (e.g., mRNA) are examples of polynucleotides having distinct biological functions. Nucleotides are organic molecules that act as monomers or subunits of nucleic acid molecules, such as DNA or RNA. Nucleic acid molecules, molecules, or polynucleotides may be double-stranded and single-stranded, linear and circular. Preferably, they are contained in a vector included in a host cell. The host cell may express an antibody construct, for example, after transformation or transfection by the vector or polynucleotide of the present invention. For this purpose, the polynucleotide or nucleic acid molecule is operably linked to a control sequence.

[0295] The genetic code is a set of rules governing how information encoded within genetic material (nucleic acid) is translated into proteins. In living cells, biological decoding is performed by ribosomes, which use tRNA molecules to link amino acids in the sequence specified by the mRNA, allowing the mRNA to read three nucleotides at a time. The code determines how sequences of these nucleotide triplets, called codons, specify which amino acid is added next during protein synthesis. With some exceptions, three-nucleotide codons within a nucleic acid sequence specify a single amino acid. Because a vast number of genes are encoded by exactly the same code, this particular code is often referred to as the regular or standard genetic code. While the genetic code determines the protein sequence for a given coded region, other genomic regions can influence when and where these proteins are produced.

[0296] Furthermore, the present invention provides a vector comprising the polynucleotide / nucleic acid molecule of the present invention.

[0297] A vector is a nucleic acid molecule used as a vehicle to deliver (foreign) genetic material into a cell. The term "vector" includes, but is not limited to, plasmids, viruses, cosmids, and artificial chromosomes. Generally, engineered vectors include a replication origin, a multiple cloning site, and selectable markers. The vector itself is typically a nucleotide sequence, usually a DNA sequence, containing an insert (transplant gene) and a larger sequence that acts as the vector's "backbone." Modern vectors may include additional features beyond the transplant gene insert and backbone: promoters, gene markers, antibiotic resistance, reporter genes, target sequences, and protein purification tags. Vectors referred to as expression vectors (expression constructs) are specifically intended for the expression of the transplant gene within target cells and generally contain control sequences.

[0298] The term "control sequence" refers to a DNA sequence essential for the expression of a coding sequence operably linked in a specific host organism. Control sequences suitable for prokaryotes include, for example, promoters, optionally operator sequences, and ribosome binding sites. Eukaryotic cells are known to utilize promoters, polyadenylation signals, and enhancers.

[0299] Nucleic acids are "operably linked" when they are placed in a functional relationship with other nucleic acid sequences. For example, DNA for a whole sequence or secretory reader is operably linked to DNA for a polypeptide if it is expressed as a whole protein participating in the secretion of the polypeptide; a promoter or enhancer is operably linked to a coding sequence if it influences the transcription of the sequence; or a ribosome binding site is operably linked to a coding sequence if it is positioned to facilitate translation. Generally, "operably linked" means that the linked DNA sequences are continuous, and in the case of a secretory reader, continuous and reader-like. However, enhancers must not be adjacent. Linkage is performed by ligation at convenient restriction sites. If such sites are absent, synthetic oligonucleotide adapters or linkers are used according to standard practice.

[0300] "Transfection" is the process of intentionally introducing nucleic acid molecules or polynucleotides (including vectors) into target cells. The term is mostly used for non-viral methods within eukaryotic cells. Transfection is often used to describe the virus-mediated delivery of nucleic acid molecules or polynucleotides. Transfection in animal cells typically involves open, transient pores or "holes" within the cell membrane that allow for the absorption of substances. Transfection can be performed using calcium phosphate, by electroporation, by cell squeezing, or by mixing cationic lipids with substances to create liposomes that fuse with the cell membrane and deposit their internal loads.

[0301] The term "transformation" is used to describe the nonviral transfer of nucleic acid molecules or polynucleotides (including vectors) into bacteria and into non-animal eukaryotic cells, including plant cells. Thus, transformation is the genetic alteration of bacteria or non-animal eukaryotic cells resulting from direct uptake through the cell membrane(s) from around the cell, and the subsequent incorporation of exogenous genetic material (nucleic acid molecules). Transformation can be achieved by artificial means. For such transformation, the cell or bacteria must be in a state of competence, which may occur as a time-limited response to environmental conditions such as starvation and cell density.

[0302] In addition, the present invention provides a host cell transformed or transfected by a polynucleotide / nucleic acid molecule or by a vector of the present invention.

[0303] The terms “host cell” or “recipient cell” as used herein are intended to include any individual cell or cell culture that may or was a recipient of a vector encoding the antibody construct of the present invention, an exogenous nucleic acid molecule, and a polynucleotide; and / or may be a recipient of the antibody construct itself. The introduction of each substance into the cell is carried out by morphological transformation, transfection, etc. The term “host cell” is also intended to include the offspring or potential offspring of a single cell. Because specific modifications may occur in subsequent generations due to natural, accidental, or intentional mutations, or due to environmental influences, such offspring may not actually be completely identical to the parent cell (in morphology, genome, or total DNA complement), but are still included within the category of terms as used herein. Suitable host cells include prokaryotic or eukaryotic cells, and also include bacteria, yeast cells, fungal cells, plant cells, and animal cells, such as insect cells and mammalian cells, such as murine, rat, macaque, or human, but are not limited to these.

[0304] The antibody construct of the present invention can be produced in bacteria. After expression, the antibody construct of the present invention is E. coli (in a soluble fraction) E.　 coli It is isolated from the cell paste and can be purified, for example, through affinity chromatography and / or size exclusion. Final purification can be performed, for example, in a manner similar to the process of purifying antibodies expressed in CHO cells.

[0305] In addition to prokaryotes, eukaryotic microorganisms, such as fungi or yeasts, are suitable cloning or expression hosts for the antibody constructs of the present invention. Saccharomyces cerevisiae ( Saccharomyces cerevisiae) or conventional bread yeast is the most commonly used among lower eukaryotic host microorganisms. However, numerous other genera, species, and strains, such as Schizoscaromyces pombe ( Schizosaccharomyces pombe ), Kluyveromyces host, for example, Kluyveromyces lactis ( K.　 lactis ) , Cluiberomyces fragilis ( K.　 fragilis )(ATCC 12424), Cluiberomyces bulgaricus( K. bulgaricus )(ATCC 16045), Cluiberomyces wickerami( K. wickeramii )(ATCC 24178), Cluiberomyces walti( K.　 waltii )(ATCC 56500), Cluiberomyces drosophyllum( K.　 drosophilarum )(ATCC 36906), Cluiberomyces thermotolerans( K.　 thermotolerans ) and Cluyveromyces Maxianus( K. marxianus ); yarrowia (European Patent No. 402 226); Pikia pastoris ( Pichia pastoris )(European Patent No. 183 070); Candida; Trichoderma ricia( Trichoderma reesia )(European Patent No. 244 234); Neurospora crassa( Neurospora crassa ); Schwanniomyces, for example, Schwanniomyces occidentalis ( Schwanniomyces occidentalis ); and fungi, such as Neurospora, Penicillium, Tolypocladium, and Aspergillus hosts, such as Aspergillus nidulans ( A.　 nidulans ) and Aspergillus niger( A. niger) is commonly available and is useful in this specification.

[0306] Suitable host cells for the expression of the glycosylated antibody constructs of the present invention are derived from multicellular organisms. Examples of invertebrate cells include plant and insect cells. Hosts, such as Spodoptera prugifera ( Spodoptera frugiperda )(larva), Aedes aegypti( Aedes aegypti )(mosquito), Aedes albopictus( Aedes albopictus )(mosquito), yellow fruit fly( Drosophila melanogaster )(fruit fly) and silkworm moth( Bombyx mori Numerous baculovirus strains and variants from ) and corresponding accepting insect host cells were identified. Various virus strains for transfection, e.g., Autograppa Califonica ( Autographa californica ) L-1 variants of NPV and the Bm-5 strain of silkworm NPV are publicly available, and these viruses, in particular Spodopteri prugifera ( Spodoptera frugiperda For the transfection of ) cells, it can be used as the virus of the present specification according to the present invention.

[0307] Plant cell cultures of cotton, corn, potatoes, soybeans, petunias, tomatoes, Arabidopsis, and tobacco can also be used as hosts. Cloning and expression vectors useful for the production of proteins within plant cell cultures are known to those skilled in the art. For example, the literature [Hiatt et al ., Nature (1989) 342: 76-78, Owen et al . (1992) Bio / Technology 10: 790-794, Artsaenko. et al . (1995) The Plant J 8: 745-750, and Fecker. et alSee . (1996) Plant Mol Biol 32: 979-986].

[0308] However, the greatest interest lies in vertebrate cells, and the proliferation of vertebrate cells in cultures (tissue cultures) has become a routine procedure. Examples of useful mammalian host cell lines include the monkey kidney CV1 line transformed by SV40 (COS-7, ATCC CRL 1651); the human embryonic kidney line (293 or 293 cells subcloned for growth in suspension cultures, literature [Graham et al., J. Gen Virol. 36 : 59 (1977)]); the young hamster kidney cell (BHK, ATCC CCL 10); and the Chinese hamster ovary cell / -DHFR (CHO, Urlaub). et al ., Proc. Natl. Acad. Sci. USA 77: 4216 (1980)); Mouse Sertoli cells (TM4, Mather, Biol. Reprod. 23: 243-251 (1980)); Monkey kidney cells (CVI ATCC CCL 70); African green monkey kidney cells (VERO-76, ATCC CRL1587); Human cervical carcinoma cells (HELA, ATCC CCL 2); Dog kidney cells (MDCK, ATCC CCL 34); Buffalo rat liver cells (BRL 3A, ATCC CRL 1442); Human lung cells (W138, ATCC CCL 75); Human liver cells (Hep G2,1413 8065); Mouse mammary tumor (MMT 060562, ATCC CCL5 1); TRI cells (Mather et al ., Annals N. Y Acad. Sci. (1982) 383: 44-68); MRC 5 cells; FS4 cells; and human liver cell line (Hep G2).

[0309] In a further embodiment, the present invention provides a method for generating an antibody construct of the present invention, the method comprising the steps of culturing a host cell of the present invention under conditions allowing the expression of the antibody construct of the present invention and recovering the generated antibody construct from the culture.

[0310] As used herein, the term “culturing” refers to the in vitro maintenance, differentiation, growth, proliferation, and / or reproduction of cells under suitable conditions in a medium. The term “expression” includes any step accompanying the production of the antibody constructs of the present invention, including but not limited to transcription, post-transcriptional modification, translation, post-translational modification, and secretion.

[0311] When using recombinant techniques, antibody constructs can be generated intracellularly, in the peripheral cytoplasm, or secreted directly into the medium. If antibody constructs are generated intracellularly, as a first step, particulate fragments—namely, either the host cell or the lysed fragment—are removed, for example, by centrifugation or ultrafiltration. [Carter et al[Bio / Technology 10: 163-167 (1992)] describes a procedure for isolating antibodies secreted into the pericytic space of E. coli. Briefly, the cell paste is thawed over 30 minutes in the presence of sodium acetate (pH 3.5), EDTA, and phenylmethylsulfonylfluoride (PMSF). Cell debris can be removed by centrifugation. If antibodies are secreted into the medium, the supernatant from this expression system is first concentrated using a commercially available protein concentration filter, e.g., an Amicon or Millipore Pellicon ultrafiltration unit. A protease inhibitor, e.g. PMSF, may be included in any of the aforementioned steps to inhibit protein degradation and to prevent the growth of accidental contaminants.

[0312] Antibody constructs of the present invention prepared from host cells may be recovered or purified, for example, using hydroxyapatite chromatography, gel electrophoresis, dialysis, and affinity chromatography. Other techniques for protein purification, such as fractionation on an ion exchange column, ethanol precipitation, reverse-phase HPLC, chromatography on silica, chromatography on heparin Sepharose (trademark), chromatography on anion or cation exchange resin (e.g., polyaspartic acid column), chromato-focusing, SDS-PAGE, and ammonium sulfate precipitation, may also be used depending on the antibody to be recovered. If the antibody construct of the present invention contains a CH3 domain, Bakerbond ABX resin (JT Baker, Phillipsburg, New Jersey) is useful for purification.

[0313] Affinity chromatography is a desirable purification technique. The substrate to which the affinity ligand is attached is most often agarose, but other substrates are also available. Mechanically stable substrates, such as controlled pore glass or poly(styrenedivinyl)benzene, allow for faster flow rates and shorter processing times than can be achieved with agarose.

[0314] Furthermore, the present invention provides a pharmaceutical composition comprising an antibody construct of the present invention or an antibody construct produced according to the method of the present invention.

[0315] As used herein, the term “pharmaceutical composition” relates to a composition suitable for administration to a patient, preferably a human patient. A particularly preferred pharmaceutical composition of the present invention comprises one or more antibody composition(s) of the present invention, preferably in a therapeutically effective amount. Preferably, the pharmaceutical composition further comprises suitable formulations of one or more (pharmaceutically effective) carriers, stabilizers, excipients, diluents, solubilizers, surfactants, emulsifiers, preservatives, and / or adjuvants. Acceptable components of the composition are preferably non-toxic to the recipient at the dosage and concentration used. The pharmaceutical composition of the present invention comprises, but is not limited to, liquid, cryo-dried, and lyophilized compositions.

[0316] The compositions of the present invention may comprise pharmaceutically acceptable carriers. Generally, as used herein, "pharmaceutically acceptable carrier" means any and all aqueous and non-aqueous solutions, sterile solutions, solvents, buffers, e.g., phosphate-buffered saline (PBS) solutions, water, suspensions, emulsions, e.g., oil / water emulsions, various types of wetting agents, liposomes, dispersion media, and coatings suitable for pharmaceutical administration, particularly parenteral administration. The use of such media and formulations in pharmaceutical compositions is well known in the art, and compositions comprising such carriers can be formulated by well-known conventional methods.

[0317] Specific embodiments provide a pharmaceutical composition comprising an antibody composition of the present invention and one or more additional excipients, such as those exemplarily described in this section and elsewhere in this specification. The excipients may be used for a wide variety of purposes in this regard, such as controlling the physical, chemical, or biological properties of a formulation, such as controlling viscosity and / or improving efficacy, and / or stabilizing such formulation, and for the methods of the present invention and for processes regarding degradation and destruction caused by stress occurring, for example, during manufacturing, transport, storage, pre-use formulation, and administration.

[0318] In certain embodiments, the pharmaceutical composition may contain a formulation material for the purpose of modifying, maintaining, or preserving, for example, pH, osmotic concentration, viscosity, clarity, color, isotonicity, odor, sterility, stability, dissolution or release rate, or absorption or penetration of the composition (see [REMINGTON'S PHARMACEUTICAL SCIENCES, 18" Edition, (AR Genrmo, ed.), 1990, Mack Publishing Company]). In these embodiments, suitable formulation materials may include, but are not limited to, the following:

[0319] Amino acids, e.g., glycine, alanine, glutamine, asparagine, threonine, proline, 2-phenylalanine, charged amino acids, preferably lysine, lysine acetate, arginine, glutamate and / or histidine

[0320] Antimicrobial agents, such as antibacterial and antifungal agents

[0321] Antioxidants, e.g., ascorbic acid, methionine, sodium sulfate, or sodium bisulfite;

[0322] Buffers, buffer systems, and buffer formulations used to maintain a composition at physiological pH or slightly lower pH, typically in a pH range of about 5 to about 8 or 9; examples of buffers are borates, bicarbonates, Tris-HCl, citrates, phosphates or other organic acids, succinates, phosphates, histidine, and acetates; for example, Tris buffers with a pH of about 7.0 to 8.5 or acetate buffers with a pH of about 4.0 to 5.5;

[0323] Non-aqueous solvents, e.g., propylene glycol, polyethylene glycol, vegetable oils, e.g., olive oil, and injectable organic esters, e.g., ethyl oleate;

[0324] Water, an aqueous carrier including alcohol / aqueous solution, an emulsion or suspension including saline solution and a buffer medium;

[0325] Biodegradable polymers, e.g., polyester;

[0326] Extenders, e.g., mannitol or glycine;

[0327] Chelating agents, e.g., ethylenediamine tetraacetic acid (EDTA);

[0328] Isotonic and absorption retardant;

[0329] Complexing agents, e.g., caffeine, polyvinylpyrrolidone, beta-cyclodextrin, or hydroxypropyl-beta-cyclodextrin)

[0330] filter;

[0331] Monosaccharides; disaccharides; and other carbohydrates (e.g., glucose, mannose, or dextrin); the carbohydrates may be non-reducing sugars, preferably trehalose, sucrose, octasulfate, sorbitol, or xylitol;

[0332] (Low molecular weight) protein, polypeptide, or proteinaceous carrier, e.g., human or bovine serum albumin, gelatin, or immunoglobulin (preferably human-derived);

[0333] Colorants and flavorings;

[0334] Sulfur-containing reducing agents, e.g., glutathione, thioxane, sodium thioglycolate, thioglycerol, [alpha]-monothioglycerol, and sodium thiosulfate.

[0335] diluent;

[0336] Emulsifier;

[0337] hydrophilic polymers, e.g., polyvinylpyrrolidone)

[0338] Salt-forming counterions, e.g., sodium;

[0339] Preservatives, e.g., antimicrobial agents, antioxidants, chelating agents, inert gases, etc.; examples include benzalkonium chloride, benzoic acid, salicylic acid, thimerosal, phenethyl alcohol, methylparaben, propylparaben, chlorhexidine, sorbic acid, or hydrogen peroxide;

[0340] Metal complexes, e.g., Zn-protein complexes;

[0341] Solvent and co-solvent (e.g., glycerin, propylene glycol, or polyethylene glycol);

[0342] Sugars and sugar alcohols, e.g., trehalose, sucrose, octasulfate, mannitol, sorbitol or xylitol, stachyose, mannose, sorbose, xylose, ribose, myoinisitose, galactose, lactitol, ribitol, myoinisitol, galactitol, glycerol, cyclitol (e.g., inositol), polyethylene glycol; and polyglycols;

[0343] Suspension;

[0344] Surfactants or wetting agents, e.g., Pluronics, PEG, sorbitan esters, polysorbates, e.g., polysorbate 20, polysorbate, Triton, tromethamine, lecithin, cholesterol, tiloxapal; the surfactant may preferably be a detergent with a molecular weight >1.2 KD and / or preferably a polyether with a molecular weight greater than 3 KD; non-limiting examples of preferred detergents are Tween 20, Tween 40, Tween 60, Tween 80, and Tween 85; and non-limiting examples of preferred polyethers are PEG 3000, PEG 3350, PEG 4000, and PEG 5000;

[0345] Stability enhancers, e.g., sucrose or sorbitol;

[0346] Isotonic enhancers, e.g., alkaline earth metal halides, preferably sodium chloride or potassium chloride, mannitol, sorbitol;

[0347] Parenteral delivery vehicle comprising sodium chloride solution, Ringer's dextrose, dextrose and sodium chloride, lactic acid Ringer's, or fixative oil;

[0348] Intravenous delivery vehicles containing fluids and nutritional supplements, and electrolyte supplements (e.g., Ringer's dextrose-based).

[0349] It is obvious to those skilled in the art that different components of a pharmaceutical composition (e.g., those listed above) may have different effects, for example, amino acids may act as buffers, stabilizers, and / or antioxidants; mannitol may act as bulking agents and / or isotonic enhancers; and sodium chloride may act as a delivery vehicle and / or isotonic enhancer, etc.

[0350] It is considered that the compositions of the present invention may comprise additional biological active agents in addition to the polypeptides of the present invention as defined herein, depending on the intended use of the composition. Such agents may be agents such as drugs acting on the gastrointestinal system, drugs acting as cell proliferation inhibitors, drugs for preventing hyperuricemia, drugs inhibiting immune responses (e.g., corticosteroids), drugs regulating inflammatory responses, drugs acting on the circulatory system, and / or cytokines known in the art. It is also considered that the antibody compositions of the present invention are applied in co-therapy, that is, in combination with other anticancer drugs.

[0351] In certain embodiments, the optimal pharmaceutical composition will be determined by those skilled in the art, for example, based on the intended route of administration, mode of delivery, and intended dosage. For example, refer to the literature [REMINGTON'S PHARMACEUTICAL SCIENCES] above. In certain embodiments, such compositions may affect the physical state, stability, in vivo release rate, and in vivo clearance rate of the antibody compositions of the present invention. In certain embodiments, the major vehicle or carrier in the pharmaceutical composition may be aqueous or non-aqueous in nature. For example, a suitable vehicle or carrier may be water for injection, physiological saline solution, or artificial cerebrospinal fluid, which may be possible to supplement with other ordinary substances in the composition for parenteral administration. Neutral buffered saline or saline mixed with serum albumin are additional exemplary vehicles. In certain embodiments, the antibody construct composition of the present invention may be prepared for storage by mixing a selected composition having a desired degree of purity in the form of a freeze-dried cake or an aqueous solution with a selective formulation agent (REMINGTON'S PHARMACEUTICAL SCIENCES, see above). Additionally, in certain embodiments, the antibody construct of the present invention may be formulated as a freeze-dried product using a suitable excipient such as sucrose.

[0352] When parenteral administration is envisioned, the therapeutic composition for use in the present invention may be provided in the form of a pyrogen-free, parenterally acceptable aqueous solution comprising the antibody construct of the present invention in a pharmaceutically acceptable vehicle. In particular, a suitable vehicle for parenteral injection is sterile distilled water in which the antibody construct of the present invention is formulated as a sterile, isotonic solution in which it is appropriately preserved. In certain embodiments, the preparation may involve the formulation of the molecule of the present invention by an agent such as injectable microspheres, biodegradable particles, polymeric compounds (e.g., polylactic acid or polyglycolic acid), beads, or liposomes, which can provide controlled or sustained release of the product that can be delivered via depot injection. In certain embodiments, hyaluronic acid having the effect of promoting sustained duration in circulation may also be used. In certain embodiments, an implantable drug delivery device may be used to introduce the antibody construct of the present invention.

[0353] Further pharmaceutical compositions comprising formulations containing the antibody construct of the present invention in sustained or controlled delivery / release formulations will be apparent to those skilled in the art. Techniques for formulating various other sustained or controlled delivery means, such as liposome carriers, biodegradable microparticles or porous beads, and depot injections, are also known to those skilled in the art. For example, see International Patent Application PCT / US93 / 00829 describing the controlled release of porous polymer microparticles for the delivery of pharmaceutical compositions. Sustained-release formulations may comprise a semipermeable polymer in the form of molded particles, for example, films or microcapsules. The sustained-release substrates are polyesters, hydrogels, polylactide (as disclosed in U.S. Patent No. 3,773,919 and European Patent Publication No. 058481), copolymers of L-glutamic acid and gamma-ethyl-L-glutamate (Sidman et al., 1983, Biopolymers 2:547-556), and poly(2-hydroxyethyl-methacrylate) (Langer et al It may include ., 1981, J. Biomed. Mater. Res. 15:167-277 and Langer, 1982, Chem. Tech. 12:98-105), ethylene vinyl acetate (Langer et al., 1981, see above) or poly-D(-)-3-hydroxybutyric acid (European Patent Application Publication No. 133,988). The sustained-release composition may also include liposomes that can be prepared by any of the methods known in the art. For example, see [Eppstein et al., 1985, Proc. Natl. Acad. Sci. USA 82:3688-3692]; European Patent Application Publication No. 036,676; European Patent No. 088,046 and European Patent No. 143,949.

[0354] Antibody constructs may also be captured in colloidal drug delivery systems (e.g., liposomes, albumin microspheres, microemulsions, nanoparticles, and nanocapsules) or macroemulsions, for example, in microcapsules prepared by coacervation techniques or interfacial polymerization (e.g., hydroxymethylcellulose or gelatin-microcapsules and poly(methyl methacrylate) microcapsules, respectively). Such techniques are disclosed in the literature [Remington's Pharmaceutical Sciences, 16th edition, Oslo, A., Ed., (1980)].

[0355] Pharmaceutical compositions used for in vivo administration are typically provided as sterile preparations. Sterilization can be achieved by filtration through a sterile filter membrane. When the composition is lyophilized, sterilization using this method can be performed before or after lyophilization and reconstitution. Compositions for parenteral administration may be stored in lyophilized form or in solution. Parenteral compositions are generally placed in containers having sterile access ports, such as intravenous solution bags or vials with caps that can be punctured by a subcutaneous syringe needle.

[0356] Another aspect of the present invention comprises a self-buffered antibody formulation of the present invention that can be used as a pharmaceutical composition as described in international patent application WO 06138181A2 (PCT / US2006 / 022599). Various descriptions of protein stabilization and formulation materials and methods useful thereto, e.g., literature [Arakawa et al., "Solvent interactions in pharmaceutical formulations," Pharm Res. 8(3): 285-91 (1991); Kendrick et al., "Physical stabilization of proteins in aqueous solution" in: RATIONAL DESIGN OF STABLE PROTEIN FORMULATIONS: THEORY AND PRACTICE, Carpenter and Manning, eds. Pharmaceutical Biotechnology. 13: 61-84 (2002), and Randolph et al., "Surfactant-protein interactions", Pharm Biotechnol. [13: 159-75 (2002)] is available, and in particular, with reference to the parts suitable for excipients and processes of self-buffering protein formulations according to the present invention, particularly with respect to protein pharmaceutical products and methods for veterinary and / or human medical uses.

[0357] For example, salts may be used according to specific embodiments of the present invention to control the ionic strength and / or isotonicity of the formulation and / or to improve the solubility and / or physical stability of proteins or other components of the composition according to the present invention. As is well known, ions can stabilize the natural state of a protein by binding to charged residues on the protein surface and by shielding charged and polar groups in the protein and by reducing their electrostatic interactions, such as attraction and repulsion. Ions can also stabilize the denatured state of a protein, particularly by binding to denatured peptide bonds (--CONH) of the protein. Furthermore, ionic interactions with charged and polar groups within the protein can also reduce intermolecular electrostatic interactions, thereby preventing or reducing protein aggregation and insolubility.

[0358] Ionic species differ significantly in their effects on proteins. According to the present invention, a number of categorical rankings of ions and their effects on proteins that can be used to formulate pharmaceutical compositions have been developed. One example is the Hofmeister series, which ranks ionic and polar nonionic solutes by their effect on the stereostabilization of proteins in solution. Solute stabilization is referred to as "kosmotropic." Solute destabilization is referred to as "chaotropic." Kosmotropes are typically used at high concentrations (e.g., more than 1 mol of ammonium sulfate) to precipitate proteins from solutes ("salting out"). Chaotropes are typically used to denature and / or solubilize proteins ("salting out"). The relative effectiveness of ions for "salting out" and "salting out" determines their position in the Hofmeister series.

[0359] Free amino acids may be used in the antibody composition formulations of the present invention as extenders, stabilizers, and antioxidants, as well as for other standard uses, according to various embodiments of the present invention. Lysine, proline, serine, and alanine may be used to stabilize proteins within the formulation. Glycine is useful in freeze-drying to ensure accurate cake structure and properties. Arginine may be useful to inhibit protein aggregation in both liquid and freeze-drying formulations. Methionine is useful as an antioxidant.

[0360] Polyols include sugars, e.g., mannitol, sucrose, and sorbitol, polyhydric alcohols, e.g., glycerol and propylene glycol, and, for the purposes of discussion herein, polyethylene glycol (PEG) and related substances. Polyols are cosmotropic. They are stabilizers useful in both liquid and lyophilized formulations to protect proteins from physical and chemical degradation processes. Polyols are also useful for controlling the isotonicity of formulations. Among the polyols useful in optional embodiments of the present invention, mannitol is typically used to ensure the structural stability of the cake within the lyophilized formulation. This ensures structural stability for the cake. It is generally used in conjunction with lyophilized protective agents, e.g., sucrose. Sorbitol and sucrose are among the preferred formulations for controlling isotonicity and as stabilizers for protection against freeze-thaw stress during transport, or for the preparation of bulk during the manufacturing process. Reducing agents (containing free aldehyde or ketone groups), e.g., glucose and lactose, can cause sugar chemistry reactions on surface lysine and arginine residues. Therefore, they are generally not preferred polyols for use according to the present invention. Additionally, sugars forming these species, such as sucrose hydrolyzed into fructose and glucose under acidic conditions, and consequently undergoing saccharification, are also not preferred polyols of the present invention in this regard. PEG is useful for stabilizing proteins and as a cryoprotectant, and in this regard, it can be used in the present invention.

[0361] In embodiments of the antibody composition of the present invention, the formulation further comprises a surfactant. Protein molecules are susceptible to absorption and denaturation on surfaces, and as a result, may be prone to aggregation at air-liquid, solid-liquid, and liquid-liquid interfaces. These generally act on a scale inversely proportional to the protein concentration. These harmful interactions are generally on a scale inversely proportional to the protein concentration and are typically exacerbated by physical agitation, such as that generated during the transport and processing of the product. Surfactants are routinely used to prevent, minimize, or reduce surface absorption. In this regard, surfactants useful in the present invention include polysorbate 20, polysorbate 80, other fatty acid esters of sorbitan polyethoxylate, and poloxamer 188. Surfactants are also commonly used to control protein stereolocating stability. In this regard, the use of surfactants is protein-specific, as any given surfactant typically stabilizes some proteins and destabilizes others.

[0362] Polysorbates are susceptible to oxidative degradation and, as is often the case, contain sufficient amounts of peroxides that cause oxidation of protein residue side chains, particularly methionine. Consequently, polysorbates must be used with caution and, when used, at their lowest effective concentrations. In this regard, polysorbates exemplify that excipients should be used at their lowest effective concentrations.

[0363] In embodiments of the antibody composition of the present invention, the formulation further comprises one or more antioxidants. To a certain extent, harmful oxidation of the protein can be prevented in the pharmaceutical formulation by maintaining appropriate levels of ambient oxygen and temperature and by avoiding exposure to light. Antioxidant excipients may likewise be used to prevent oxidative degradation of the protein. Among the antioxidants useful in this regard are reducing agents, oxygen / free radical scavengers, and chelating agents. Antioxidants for use in the therapeutic protein formulation according to the present invention are preferably water-soluble and maintain their activity throughout the shelf life of the product. EDTA is a preferred antioxidant according to the present invention in this regard. Antioxidants can damage proteins. For example, reducing agents, such as glutathione, can, in particular, break intramolecular disulfide bonds. Therefore, in particular, antioxidants for use in the present invention are selected to eliminate or sufficiently reduce the very possibility of damaging the protein within the formulation.

[0364] The formulation according to the present invention is a protein cofactor and may include metal ions essential for forming protein coordination complexes, such as zinc, which is essential for forming certain insulin suspensions. Metal ions may also inhibit some processes of protein degradation. However, metal ions also catalyze physical and chemical processes of protein degradation. Magnesium ions (10 to 120 mM) may be used to inhibit the isomerization of aspartic acid to isoaspartic acid. Ca +2 Ions (up to 100 mM) can increase the stability of human deoxyribonuclease. However, Mg +2 , Mn +2 and Zn +2 can destabilize rhDNase. Similarly, Ca +2 and Sr +2 It can stabilize Factor VIII, and this is Mg +2, Mn +2 and Zn +2 , Cu +2 and Fe +2 It can be destabilized by, and its aggregation is Al +3 It can be increased by ions.

[0365] In embodiments of the antibody composition of the present invention, the formulation further comprises one or more preservatives. Preservatives are essential when developing multi-dose parenteral formulations involving more than one extraction from the same container. Their primary function is to inhibit microbial growth and ensure the sterility of the product throughout its shelf life or use period. Commonly used preservatives include benzyl alcohol, phenol, and m-cresol. Although preservatives have a long history of use with small molecule parenterals, the development of protein formulations containing preservatives can be challenging. Preservatives almost always have a destabilizing effect (aggregation) on proteins, which has been a major factor limiting their use in multi-dose protein formulations. To date, most protein drugs have been formulated only as single-dose products. However, when multi-dose formulations are available, they offer added advantages that enable patient convenience and increased marketability. Human growth hormone (hGH) is a good example where the development of a preserved formulation is more convenient and leads to the commercialization of a multi-dose injection pen presentation. At least four such pen devices containing preserved formulations of hGH are currently available on the market. Norditropin (liquid, Novo Nordisk), Neurotropin AQ (liquid, Genentech), and Genotropin (lyophilized dual-chamber cartridge, Pharmacia & Upjohn) contain phenol, whereas Somatrop (Eli Lilly) is formulated with m-cresol. Several aspects need to be considered during the development of formulation and preserved dosage forms. The effective preservation concentration in the drug product must be optimized. This requires testing the given preservative in the dosage form within a concentration range that confers antimicrobial efficacy without compromising protein stability.

[0366] As might be expected, the development of liquid formulations containing preservatives is more challenging than that of lyophilized formulations. Lyophilized products can be lyophilized without preservatives and can be reconstituted with preservatives containing diluents at use. This reduces the time the preservative is in contact with the protein and minimizes the associated stability risks. With liquid formulations, the efficacy and stability of the preservative must be maintained over the entire shelf life of the product (about 18 to 24 months). An important point to note is that the efficacy of the preservative must be demonstrated in the final formulation containing the active drug and all excipient components.

[0367] The antibody constructs disclosed herein may also be formulated as immunoliposomes. "Liposomes" are small vesicles composed of various types of lipids, phospholipids, and / or surfactants useful for the delivery of drugs to mammals. Liposome components are typically arranged in a duplex formation similar to the lipid arrangement of biological membranes. Liposomes containing antibody constructs [Epstein et al ., Proc. Natl. Acad. Sci. USA, 82: 3688 (1985); Hwang et al [Proc. Natl Acad. Sci. USA, 77: 4030 (1980)]; U.S. Patents No. 4,485,045 and No. 4,544,545; and methods known in the art, such as those described in W0 97 / 38731. Liposomes having an improved cycle time are disclosed in U.S. Patent No. 5,013,556. Particularly useful liposomes can be produced by a reverse-phase evaporation method using a lipid composition comprising phosphatidylcholine, cholesterol, and PEG-derived phosphatidylethanolamine (PEG-PE). The liposomes are extracted through a filter of a defined pore size to obtain liposomes having a desired diameter. The Fab fragment of the antibody construct of the present invention is produced through a disulfide exchange reaction [Martin et al.It can be conjugated to liposomes as described in [J. Biol. Chem. 257: 286-288 (1982)]. Chemotherapy agents are selectively contained within the liposomes. Literature [Gabizon et al See . J. National Cancer Inst. 81 (19) 1484 (1989)].

[0368] Once a pharmaceutical composition is formulated, it may be stored in sterile vials as a solution, suspension, gel, emulsion, solid, crystal, or as a dehydrated or lyophilized powder. Such formulations may be stored in a form ready for use or in a form that is reconstituted before administration (e.g., lyophilized).

[0369] The biological activity of the pharmaceutical compositions defined herein is, for example, in the following examples, WO 99 / 54440 or the literature [Schlereth et alIt may be determined by cytotoxicity analysis as described in [Cancer Immunol. Immunother. 20 (2005), 1-12)]. As used herein, “efficacy” or “in vivo efficacy” refers to the response to a therapy using, for example, standardized NCI response criteria, with respect to the pharmaceutical composition of the present invention. Success or in vivo efficacy of a therapy using the pharmaceutical composition of the present invention refers to the effectiveness of the composition for its intended purpose, that is, the ability of the composition to produce its intended effect, that is, the depletion of pathological cells, e.g., tumor cells. In vivo efficacy may be monitored by established standard methods for each disease entity, including, but not limited to, leukocyte counting, differentiation, fluorescence-activated cell sorting, and bone marrow aspiration. Additionally, various disease-specific clinical chemistry parameters and other established standard methods may be used. Furthermore, computed tomography, X-rays, and nuclear magnetic resonance imaging (e.g., National Cancer Institute—regarding standard-based response assessment [Cheson BD, Horning SJ, Coiffier B, Shipp MA, Fisher RI, Connors JM, Lister TA, Vose J, Grillo-Lopez A, Hagenbeek A, Cabanillas F, Klippensten D, Hiddemann W, Castellino R, Harris NL, Armitage JO, Carter W, Hoppe R, Canellos GP. Report of an international workshop to standardize response criteria for non-Hodgkin's lymphomas. NCI Sponsored International Working Group. J Clin Oncol.1999 Apr;17(4):1244]), positron emission tomography scanning, leukocyte counting, differentiation, fluorescence-activated cell sorting, bone marrow aspiration, lymph node biopsy / histology and various lymphoma-specific clinical chemistry parameters (e.g., lactate dehydrogenase) and other established standard methods may be used.

[0370] Another major challenge in the development of drugs, such as the pharmaceutical composition of the present invention, is the predictable control of pharmacokinetic properties. To this end, the pharmacokinetic profile of a drug candidate—that is, the profile of pharmacokinetic parameters that influence the ability of a specific drug to treat a given pathological condition—can be established. The pharmacokinetic parameters of a drug that influence its ability to treat a specific disease entity include, but are not limited to, half-life, volume of distribution, hepatic first-pass metabolism, and blood serum binding. The efficacy of a given drug formulation can be influenced by each of the aforementioned parameters.

[0371] "Half-life" refers to the time it takes for 50% of an administered drug to be eliminated through biological processes, such as metabolism and excretion. "Hepatic first-pass metabolism" refers to the tendency of a drug to be metabolized upon its first contact with the liver, that is, during its first passage through the liver. "Volume of distribution" refers to the extent of the drug's retention throughout various compartments of the body, such as intracellular and extracellular spaces, tissues, and organs, and the distribution of the drug within these compartments. "Blood serum binding degree" refers to the tendency of a drug to interact with and bind to blood serum proteins, such as albumin, which, for example, cause a decrease or loss of the drug's biological activity.

[0372] Pharmacokinetic parameters also include bioavailability, time lag (Tlag), Tmax, absorption rate, and Cmax for a given amount of drug administered. "Bioavailability" refers to the amount of drug in the blood compartment. "Time lag" refers to the time delay between the administration of the drug and its detection and measurement in blood or plasma. "Tmax" is the time after the maximum blood concentration of the drug is reached, and "Cmax" is the maximum blood concentration obtained by a given drug. The time to reach the blood or tissue concentration of the drug required for a biological effect is influenced by all parameters. The pharmacokinetic parameters of bispecific antibody constructs exhibiting cross-species specificity, which can be determined in preclinical animal studies in non-chimpanzee primates as outlined above, are also, for example, [Schlereth et al. It is presented in the publication by [Cancer Immunol. Immunother. 20 (2005), 1-12)].

[0373] In one embodiment, the present invention provides an antibody construct of the present invention or an antibody construct produced according to the method of the present invention for use in the prevention, treatment, or improvement of cancer.

[0374] The formulations described herein are useful as pharmaceutical compositions for the treatment, improvement, and / or prevention of pathological medical conditions as described herein in patients requiring treatment, improvement, and / or prevention of pathological medical conditions. The term “treatment” refers to therapeutic treatment and preventive or protective measures. Treatment includes the application or administration of the formulation to a body, isolated tissue, or cells, symptoms of a disease / disorder, or predisposition to a disease / disorder from a patient having the disease / disorder, for the purpose of curing, healing, alleviating, relieving, altering, correcting, improving, or affecting the disease, symptoms of a disease, or predisposition to a disease / disorder.

[0375] As used herein, the term "improvement" refers to any improvement in the disease state of a patient having one or more of the types of (metastatic) tumors or cancers as embodied herein, by administration of an antibody composition according to the present invention to a subject requiring improvement in the disease state. Such improvement may also manifest as a delay or cessation of the patient's (metastatic) tumor or cancer. As used herein, the term "prevention" refers to the avoidance of onset or recurrence of a patient having one or more of the types of (metastatic) tumors or cancers as embodied herein, by administration of an antibody composition according to the present invention to a subject requiring avoidance of onset or recurrence.

[0376] The term "disease" refers to any pathological condition that is advantageous to treat with the antibody compositions or pharmaceutical compositions described herein. This includes chronic and acute disorders or diseases comprising the relevant pathological conditions that cause mammals to develop a predisposition to the disease.

[0377] A "neoplasm" is an abnormal growth of tissue, but it does not usually always form a mass. Also, when it does form a mass, it is typically referred to as a "tumor." Neoplasms or tumors can be benign, potentially malignant (precancerous), or malignant. Malignant neoplasms are typically referred to as cancer. They usually invade or destroy surrounding tissues and form metastases; that is, they can spread to other parts, tissues, or organs of the body. Therefore, the term "metastatic cancer" includes metastasis to tissues or organs other than the original tumor. Lymphoma and leukemia are lymphoid neoplasms. For the purposes of this invention, they are also included by the terms "tumor" or "cancer."

[0378] In a preferred embodiment, the present invention provides an antibody construct of the present invention or an antibody construct produced according to the method of the present invention for use in the prevention, treatment, or improvement of cancer, wherein the cancer is a lung carcinoma, head and neck carcinoma, primary or secondary CNS tumor, primary or secondary brain tumor, primary CNS lymphoma, spinal tumor, brainstem glioma, glioblastoma, pituitary adenoma, adrenocortical carcinoma, esophageal carcinoma, colon cancer, breast cancer, ovarian cancer, NSCLC (non-small cell lung cancer), SCLC (small cell lung cancer), endometrial cancer, cervical cancer, uterine cancer, transitional cell carcinoma, bone cancer, pancreatic cancer, skin cancer, cutaneous or intraocular melanoma, liver cancer, cholangiocarcinoma, gallbladder cancer, kidney cancer, rectal cancer, cancer of the anus, gastric cancer, gastric (stomach, colorectal, and duodenum) cancer, small intestine cancer, biliary tract cancer, urethral cancer, renal cell carcinoma, endometrial carcinoma, thyroid cancer, testicular cancer, cutaneous squamous cell carcinoma. It is selected from the group consisting of melanoma, gastric cancer, prostate cancer, bladder cancer, osteosarcoma, mesothelioma, Hodgkin's disease, non-Hodgkin lymphoma, chronic or acute leukemia, chronic myeloid leukemia, lymphocytic lymphoma, multiple myeloma, fibrosarcoma, neuroblastoma, retinoblastoma, and soft tissue sarcoma, and metastatic cancer diseases derived from any of the aforementioned. The (metastatic) cancer is preferably a P-cadherin positive or P-cadherin expressing cancer.

[0379] In a further preferred embodiment, the present invention provides an antibody construct of the present invention or an antibody construct produced according to the method of the present invention for use in the prevention, treatment or improvement of cancer, wherein the cancer is (metastatic) squamous cell carcinoma.

[0380] The present invention also provides a method for improving the treatment and prevention of (metastatic) tumors or cancer, comprising the step of administering an antibody construct of the present invention or an antibody construct produced according to the method of the present invention to a subject requiring improvement in the treatment and prevention of (metastatic) tumors or cancer.

[0381] In a preferred embodiment, the present invention provides a method for treating or improving a tumor or cancer, comprising the step of administering an antibody construct of the present invention or an antibody construct produced according to the method of the present invention to a subject requiring treatment or improvement of a tumor or cancer, wherein the cancer is a lung carcinoma, head and neck carcinoma, primary or secondary CNS tumor, primary or secondary brain tumor, primary CNS lymphoma, spinal tumor, brainstem glioma, pituitary adenoma, adrenocortical carcinoma, esophageal carcinoma, colon cancer, breast cancer, ovarian cancer, NSCLC (non-small cell lung cancer), SCLC (small cell lung cancer), endometrial cancer, cervical cancer, uterine cancer, transitional cell carcinoma, bone cancer, pancreatic cancer, skin cancer, cutaneous or intraocular melanoma, liver cancer, cholangiocarcinoma, gallbladder cancer, kidney cancer, rectal cancer, cancer of the anus, gastric cancer, gastric (stomach, colorectal, and duodenum) cancer, small intestine cancer, biliary tract cancer, urethral cancer, renal cell carcinoma, endometrial carcinoma, thyroid cancer. It is selected from the group consisting of testicular cancer, cutaneous squamous cell carcinoma, melanoma, gastric cancer, prostate cancer, bladder cancer, osteosarcoma, mesothelioma, Hodgkin's disease, non-Hodgkin lymphoma, chronic or acute leukemia, chronic myeloid leukemia, lymphocytic lymphoma, multiple myeloma, fibrosarcoma, neuroblastoma, retinoblastoma, and soft tissue sarcoma, and any metastatic cancer diseases derived from the aforementioned.

[0382] In a further preferred embodiment, the present invention provides a method for treating or improving a tumor or cancer or a metastatic tumor or cancer, wherein the cancer is (metastatic) squamous cell carcinoma.

[0383] The terms "subjects requiring treatment" or "subjects requiring therapy" include not only subjects who already possess a disability but also subjects for whom a disability is to be prevented. "Subjects requiring treatment" or "patients" include human and other mammalian subjects receiving either preventive or therapeutic treatment.

[0384] The antibody compositions of the present invention will generally be designed for specific routes and methods of administration, specific dosages and frequencies of administration, and specific treatments for specific diseases, particularly within a range of bioavailability and persistence. The materials of the composition are preferably formulated at an acceptable concentration for the site of administration.

[0385] Accordingly, formulations and compositions can be designed according to the present invention for delivery by any suitable route of administration. In connection with the present invention, routes of administration include, but are not limited to, the following.

[0386] Local routes (e.g., epidermis, inhalation, nasal cavity, eyes, ear / aural, vagina, mucous membranes);

[0387] Intestinal pathways (e.g., oral, gastrointestinal, sublingual, sublabial, buccal, rectal); and

[0388] Parenteral routes (e.g., intravenous, intra-arterial, intraosseous, intramuscular, intracerebral, intraventricular, epidural, intrathecal, subcutaneous, intraperitoneal, extraamniotic, intra-articular, intracardiac, intradermal, intralesional, intrauterine, intravesical, intravesical, intravitreal, transcutaneous, intranasal, via mucosa, intrasynovial, intracavernous).

[0389] The pharmaceutical compositions and antibody compositions of the present invention are particularly useful for parenteral administration, e.g., subcutaneous or intravenous delivery, by injection, e.g., bolus injection, or by infusion, e.g., continuous infusion. The pharmaceutical compositions may be administered using medical devices. Examples of medical devices for administering the pharmaceutical compositions are U.S. Patents No. 4,475,196; No. 4,439,196; No. 4,447,224; No. 4,447,233; No. 4,486,194; No. 4,487,603; No. 4,596,556; No. 4,790,824; No. 4,941,880; No. 5,064,413; No. 5,312,335; No. 5,312,335; No. 5,383,851. and is listed in No. 5,399,163.

[0390] In particular, the present invention provides continuous administration of a suitable composition. As a non-limiting example, uninterrupted or substantially uninterrupted, i.e., continuous administration, may be realized by a small pump system worn by the patient to measure the inflow of the therapeutic agent into the patient's body. A pharmaceutical composition comprising the antibody composition of the present invention may be administered by using said pump system. Such pump systems are generally known in the art and typically rely on the periodic exchange of a cartridge containing the therapeutic agent to be infused. When the cartridge is exchanged in such a pump system, a temporary interruption of another uninterrupted flow of the therapeutic agent into the patient's body may continue to occur. In such cases, the administration step before cartridge replacement and the step after cartridge replacement will still be considered within the meaning of the pharmaceutical means of the present invention, and the method of the present invention constitutes a single "uninterrupted administration" of such therapeutic agent.

[0391] Continuous or uninterrupted administration of the antibody composition of the present invention may be intravenous or subcutaneous by a fluid delivery device or a small pump system comprising a fluid drive mechanism for discharging fluid out of the reservoir and an actuating mechanism for operating the drive mechanism. A pump system for subcutaneous administration may include a needle or cannula for penetrating the patient's skin and delivering a suitable composition into the patient's body. The pump system may be fixed or attached directly to the patient's skin independently of a vein, artery, or blood vessel, thereby allowing direct contact between the pump system and the patient's skin. The pump system may be attached to the patient's skin for 24 hours to several days. The pump system may have a small size by means of a reservoir for a small volume. As a non-limiting example, the volume of the reservoir for the suitable pharmaceutical composition to be administered may be 0.1 to 50 ml.

[0392] Continuous administration can also be transdermal by wearing a patch on the skin, and the interval is replaced. Those skilled in the art recognize patch systems for drug delivery suitable for this purpose. Transdermal administration is particularly well-suited for uninterrupted administration, because the exchange of a first depletion patch can be advantageously performed, for example, simultaneously with a new second patch on the skin surface immediately adjacent to the first depletion patch and just before the removal of the first depletion patch. Problems of flow interruption or power cell failure do not occur.

[0393] If a pharmaceutical composition has been lyophilized, the lyophilized material is first reconstituted in a suitable liquid before administration. The lyophilized material may be reconstituted, for example, in bacteriostatic water for injection (BWFI), physiological saline, phosphate-buffered saline (PBS), or in the same formulation in which the protein was present prior to lyophilization.

[0394] The composition of the present invention may be administered to subjects in dose escalation studies by administering increased doses of the antibody composition of the present invention exhibiting cross-species specificity described herein for non-chimpanzee primates, e.g., macaques, at suitable doses that can be administered. As indicated above, the antibody composition of the present invention exhibiting cross-species specifically described herein may be advantageously used as a drug in humans in the same form as in preclinical studies in non-chimpanzee primates. The dosing regimen will be determined by the attending physician and clinical factors. As is well known in the medical field, the dosage for any single patient depends on a number of factors, including the patient's size, body surface area, age, the specific compound administered, gender, time and route of administration, general health status, and other drugs administered concurrently.

[0395] The terms "effective dose" or "effective dosage" are defined as an amount sufficient to achieve the intended effect, or at least partially achieve it. The term "therapeutic effective dose" is defined as an amount sufficient to cure the disease and its complications, or at least partially prevent them, in patients who already have the disease. The effective dose or dosage for this use will depend on the condition to be treated (indication), the antibody composition to be delivered, the therapeutic content and purpose, the severity of the disease, prior therapy, the patient's clinical history and response to the therapeutic agent, the route of administration, the patient's size (body weight, body surface area, or organ size) and / or condition (age and general health status), and the general state of the patient's own immune system. The appropriate dose may be adjusted at the discretion of the attending physician so that it can be administered to the patient in a single dose or over a series of doses, and to obtain the optimal therapeutic effect.

[0396] A typical dosage may be in the range of about 0.1 μg / kg to about 30 mg / kg or more, depending on the factors mentioned above. In a specific embodiment, the dosage may be in the range of 1.0 μg / kg to about 20 mg / kg, optionally 10 μg / kg to about 10 mg / kg or 100 μg / kg to about 5 mg / kg.

[0397] A therapeutically effective dose of the antibody construct of the present invention preferably results in a reduction in the severity of disease symptoms, an increase in the frequency or duration of disease-asymptomatic periods, or the prevention of damage or disability caused by disease distress. To treat CDH3-expressing tumors, a therapeutically effective dose of the antibody construct of the present invention, e.g., an anti-CDH3 / anti-CD3 antibody construct, preferably inhibits cell growth or tumor growth by at least about 20%, at least about 40%, at least about 50%, at least about 60%, at least about 70%, at least about 80%, or at least about 90% compared to untreated patients. The ability of the compound to inhibit tumor growth can be evaluated in animal models that predict efficacy in human tumors.

[0398] The pharmaceutical composition may be administered as a sole therapeutic agent or, if necessary, in combination with additional therapies, such as anticancer therapy, such as other proteinaceous and non-proteinaceous drugs. These drugs may be administered separately before or after the administration of the antibody composition, either simultaneously with the composition comprising the antibody composition of the present invention as defined herein, or at a predetermined time interval and dose.

[0399] As used herein, the term “effective and non-toxic dose” refers to an acceptable dose of the antibody composition of the present invention that is sufficiently high to cause depletion of pathological cells, tumor elimination, tumor shrinkage, or disease stabilization without or essentially without major toxic effects. Such effective and non-toxic dose may be determined, for example, by dose escalation studies described in the art, and must be below the dose that induces severe adverse effects (dose-limiting toxicity, DLT).

[0400] The term “toxicity” as used herein refers to the toxic effects of a drug that manifest in adverse events or severe adverse events. These adverse effects may generally refer to a lack of tolerance to the drug and / or a lack of local tolerance after administration. Toxicity may also include teratogenic or carcinogenic effects caused by the drug.

[0401] As used herein, the terms “safety,” “in vivo safety,” or “tolerability” define the administration of a drug that does not directly induce severe adverse events after administration (local tolerance) and during longer-term application of the drug. “Safety,” “in vivo safety,” or “tolerability” may be evaluated, for example, at regular intervals during treatment and subsequent periods. Measurements include clinical evaluations, for example, organ enlargement and screening of experimental abnormalities. Clinical evaluations may be performed, and deviations from normal are recorded / encoded according to NCI-CTC and / or MedDRA standards. Organ signs may include criteria such as those presented in the literature [Common Terminology Criteria for adverse events v3.0 (CTCAE)], e.g., allergy / immunology, blood / bone marrow, cardiac arrhythmia, agglutination, etc. Parameters that may be tested include, for example, hematology, clinical chemistry, coagulation profiles, and urinalysis, as well as testing of other body fluids, e.g., serum, plasma, lymph or cerebrospinal fluid, liquids, etc. Accordingly, safety can be evaluated by measuring experimental parameters and recording adverse events, for example, through physical examination, imaging techniques (i.e., ultrasound, X-ray, CT scan, magnetic resonance imaging (MRI)), other measurements by technical devices (i.e., electrocardiogram), and vital signs. For example, adverse events in non-chimpanzee primates in use according to the use and method of the present invention may be tested by histopathological and / or histochemical methods.

[0402] The above term is also mentioned, for example, in the literature [Preclinical safety evaluation of biotechnology-derived pharmaceuticals S6; ICH Harmonised Tripartite Guideline; ICH Steering Committee meeting on July 16, 1997].

[0403] In a further embodiment, the present invention provides a kit comprising an antibody construct of the present invention, an antibody construct produced according to the method of the present invention, a vector of the present invention and / or a host cell of the present invention.

[0404] In relation to the present invention, the term “kit” means two or more components packaged together in a container, vessel, or other (one of which corresponds to the antibody construct, pharmaceutical composition, vector, or host cell of the present invention). Accordingly, a kit may be described as a set of products and / or apparatus sufficient to achieve a specific objective that can be marketed as a single unit.

[0405] The kit may comprise one or more containers (e.g., vials, ampoules, containers, syringes, bottles, bags) of any suitable shape, size, and material (preferably waterproof, e.g., plastic or glass) containing the antibody construct or pharmaceutical composition of the present invention in an appropriate dosage for administration (see above). The kit may further accommodate directions for use (e.g., leaflets or instruction manuals), means for administering the antibody construct of the present invention, e.g., syringes, pumps, infusers, etc., means for reconstituting the antibody construct of the present invention and / or means for diluting the antibody construct of the present invention.

[0406] The present invention also provides a kit for a single-dose administration unit. The kit of the present invention may also accommodate a first container comprising a dried / lyophilized antibody construct and a second container comprising an aqueous formulation. In a specific embodiment of the present invention, a kit is provided for accommodating single-chamber and multi-chamber pre-filled syringes (e.g., liquid syringes and cryo-syringes). Brief explanation of the drawing

[0407] Present the drawing: Fig. 1: Schematic representation of the five extracellular domains D1 to D5 of human CDH3, murine CDH3, and one exemplary chimeric CDH3 (here: D1B murine). The five domains are shown again below, each further divided into three subdomains. Description of an exemplary FACS signal ("epitope clustering" in Example 2) is shown in the lower right. Fig. 2: Sequence alignment of human CDH3 and murine CDH3 and exemplary display of different domains: signal peptide, propeptide, extracellular domains D1 to D5, transmembrane domain, and cytoplasmic domain. Murine sequence exchanges of the five extracellular domains were introduced into the human CDH3 backbone (see Example 1), and then the chimeric construct was used for epitope clustering (epitope mapping) (see Example 2). Fig. 3: Human and murine CDH3 expressed on the surface of CHO cells as shown by flow cytometry, as well as 20 chimeric human-murine CDH3 constructs (5 extracellular domains (ECDs) and 3 subdomains for each ECD). The expression of human wild-type CDH3, murine wild-type CDH3, and chimeric CDH3 constructs on CHO cells was confirmed by a murine cross-reactive monoclonal mouse IgG1 anti-human CDH3 antibody. Conjugated monoclonal antibodies were detected using anti-mouse IgG Fcg-PE (1:100, 50 µl; Jackson Immunoresearch # 115-116-071). D1*): Hu CDH3 D1 mu-CHO. Fig. 4: Epitope mapping of CDH3 constructs. For examples of binding molecules specific to different epitope clusters / extracellular sub-domains as detected by epitope mapping of chimeric CDH3 constructs, see Example 2. Fig. 4a : D1B binder. Fig. 4b : D2C binder. Fig. 4c : D3A binder. Fig. 5: FACS binding analysis of 5 µg / ml purified bispecific antibody monomers for the indicated cell lines. See also Example 5. Detection of CDH3xCD3 bispecific antibody binding was performed using an internal mouse antibody specific to the CD3 binding portion of the bispecific antibody, followed by goat anti-mouse Fcγ-PE. The negative control was the only buffer, followed by the detection antibody. Figs. 5a and 5b : CDH3xCD3 bispecific binder ( Fig. 5a : Epitope cluster / extracellular sub-domain D2C; Fig. 5b : Epitope cluster / extracellular sub-domain D3A) was analyzed for their binding to human CDH3-transfected CHO cells, human CD3 on the human T cell line HPB-all, cynoCDH3-transfected CHO cells, cynoCD3 on the cynoCD3-expressing T cell line HSC-F, human CDH3-positive cell line A431, and murine CDH3-transfected CHO cells (negative control). Binding was detected in all cases except the negative control. FIGS. 5C and FIGS. 5D : CDH3xCD3 bispecific binder ( Fig. 5c : Epitope cluster / extracellular sub-domain D2C; Fig. 5d : Epitope clusters / extracellular sub-domain D3A) were analyzed for their binding to human CDH3 paralogs CDH1, CDH2, CDH4, and CDH5. No binding to the paralogs was detected. dhfr - / - Binding to CHO cells (negative control) was not detected. Fig. 6: 18 hours 51Cytotoxic activity of stimulated human CD8+ T cells against human CDH3-transfected CHO cells in the presence of a CDH3xCD3 bispecific antibody as measured by chromium release assay. Effector cells: Stimulated, abundant human CD8+ T cells. Target cells: Human CDH3-transfected CHO cells. Effector-to-target cell (E:T) ratio: 10:1. Epitope cluster / extracellular sub-domain D2C ( Fig. 6A ) and antibodies specific to D3A ( Fig. 6B ). Fig. 7: In the presence of CDH3xCD3 bispecific antibodies and 18 hours 51 Cytotoxic activity of stimulated human CD8+ T cells against human CDH3-positive ophthalmomorphic carcinoma cell line A431 in chromium release assay. Effector cells: Stimulated, abundant human CD8+ T cells. Target cells: Human A431 cells. Effector-to-target (E:T) ratio: 10:1. Epitope cluster / extracellular sub-domain D2C ( Fig. 7A ) and antibodies specific to D3A ( Fig. 7B ). Fig. 8: Cytotoxic activity of unstimulated human PBMCs against human CDH3-transfected CHO cells in the presence of a CDH3xCD3 bispecific antibody as measured by a 48-hour FACS-based cytotoxicity assay. Effector cells: Unstimulated human PBMCs (CD14- / CD56-). Target cells: Human CDH3-transfected CHO cells. Effector-to-target cell (E:T) ratio: 10:1. Epitope cluster / extracellular sub-domain D2C ( Fig. 8A ) and antibodies specific to D3A ( Fig. 8B ). Fig. 9: Cytotoxic activity of unstimulated human PBMCs against human CDH3-positive ophthalmocarcinoma cell line A431 in the presence of CDH3xCD3 bispecific antibodies as measured by a 48-hour FACS-based cytotoxicity assay. Effector cells: Unstimulated human PBMCs (CD14- / CD56-). Target cells: Human A431 cells. Effector-to-target cell (E:T) ratio: 10:1. Epitope cluster / extracellular sub-domain D2C ( Fig. 9A ) and antibodies specific to D3A ( Fig. 9B ). Fig. 10: Cytotoxic activity of macaque T cells against macaque CDH3-transfected CHO cells in the presence of CDH3xCD3 bispecific antibodies as measured by a 48-hour FACS-based cytotoxicity assay. Effector cells: Macaque CD3-positive T cell line LnPx4119. Target cells: Macaque CDH3-transfected CHO cells. Effector to target cell (E:T) ratio: 10:1. Epitope cluster / extracellular sub-domain D2C ( Fig. 10a ), D3A( Fig. 10b ) and D1B( Fig. 10c Antibody specific to ). Fig. 11: Antitumor activity of a CDH3xCD3 bispecific antibody against epitope cluster / extracellular sub-domain D2C (CDH3-13) in a human tumor xenograft model (see Example 14). The antibody prevents the formation of A-431 tumors in a dose-dependent manner in the presence of human PBMCs. High tumor volume at the start of measurement (Day 5) due to the large volume of the cell mixture injected on Day 1. ** p<0.01; *** p<0.0001. Fig. 12: Antitumor activity of CDH3xCD3 bispecific and half-life extended (HLE) antibodies against epitope cluster / extracellular sub-domain D2C (CDH3-13) in a human tumor xenograft model (see Example 15). The HLE antibody prevents the formation of human HCT-116 tumors in a dose-dependent manner in the presence of human PBMCs. Figure 12A shows the overall results, but Figure 12B shows the results obtained for higher antibody concentrations (Group 2), distinguished between responding animals (7 / 10) and non-responding animals (3 / 10). Fig. 13: T cell activation in the absence of target cells by a bispecific antibody construct in the absence (upper panel) and presence (lower panel) of albumin fusion at the C-terminus of the construct. Specific details for implementing the invention

[0408] Examples :

[0409] The following examples illustrate the invention. These examples should not be construed as limiting the scope of the invention. The invention is limited only by the claims.

[0410] Examples 1

[0411] Wild type and chimera CDH3 expressing CHO Cell generation

[0412] For the construction of the chimeric molecules used for epitope mapping, the sequences of each of the five extracellular domains Dom1 to Dom5 (or D1 to D5) of human CDH3 and their sub-domains (A, B, and C) were replaced with the corresponding murine sequences. The following 20 molecules were generated; see also Figures 1 and 2:

[0413] Hu CDH3 / Dom1 mu(aa 108-215) Sequence No. 13

[0414] o Hu CDH3 / Dom1A mu(aa 108-143) Sequence No. 14

[0415] o Hu CDH3 / Dom1B mu(aa 144-179) Sequence No. 15

[0416] o Hu CDH3 / Dom1C mu(aa 180-215) Sequence No. 16

[0417] Hu CDH3 / Dom2 mu(aa 216-327) Sequence No. 17

[0418] o Hu CDH3 / Dom2A mu(aa 216-252) Sequence No. 18

[0419] o Hu CDH3 / Dom2B mu(aa 253-290) Sequence No. 19

[0420] o Hu CDH3 / Dom2C mu(aa 291-327) Sequence No. 20

[0421] Hu CDH3 / Dom3 mu(aa 328-440) Sequence No. 21

[0422] o Hu CDH3 / Dom3A mu(aa 328-363) Sequence No. 22

[0423] o Hu CDH3 / Dom3B mu(aa 364-403) Sequence No. 23

[0424] o Hu CDH3 / Dom3C mu(aa 404-440) Sequence No. 24

[0425] Hu CDH3 / Dom4 mu(aa 441-546) Sequence No. 25

[0426] o Hu CDH3 / Dom4A mu(aa 441-474) Sequence No. 26

[0427] o Hu CDH3 / Dom4B mu(aa 475-511) Sequence No. 27

[0428] o Hu CDH3 / Dom4C mu(aa 512-546) Sequence No. 28

[0429] Hu CDH3 / Dom5 mu(aa 547-650) Sequence No. 29

[0430] o Hu CDH3 / Dom5A mu(aa 547-581) Sequence No. 30

[0431] o Hu CDH3 / Dom5B mu(aa 582-616) Sequence No. 31

[0432] o Hu CDH3 / Dom5C mu(aa 617-650) Sequence No. 32

[0433] The above list indicates the positions of different domains (Dom1 to Dom5) within the amino acid sequence of human CDH3 as illustrated in SEQ ID NO. 1, as well as their respective subdomains A through C. For example, subdomain D1A is located at amino acid positions 108 through 143 of SEQ ID NO. 1. This applies similarly to all other domains listed above.

[0434] For expression in CHO cells, the coding sequence of the chimeric extracellular domain described above is followed by the coding sequence of an artificial Ser / Gly-linker within the frame, followed by a domain derived from the transmembrane / intracellular domain of human EpCAM (amino acids 266 to 314 of the sequence as disclosed in GenBank accession number NM_002354). All chimeric constructs included an N-terminal signal sequence (signal peptide) and a propeptide.

[0435] To generate CHO cells expressing human, cynomolgus macaque ("cyno"), mouse, and human / mouse chimera CDH3, the respective coding sequences of human CDH3 (SEQ No. 2, also see GenBank accession No. NM_001793), cyno CDH3 (SEQ No. 6), mouse CDH3 (SEQ No. 10, also see GenBank accession No. NM_001037809), and 20 human-mouse CDH3 chimeras (see above) were cloned into a plasmid designated as pEF-DHFR (pEF-DHFR is [Raum et al.Listed in Cancer Immunol Immunother 50 (2001) 141-150). The coding sequence of macaque CDH3 was obtained by standard cloning using a cynomolgus spleen cDNA library (BioChain) that hybridizes to the non-translating or conserved sequence regions of the human CDH3 mRNA transcript (NM_001793) and human sequence-specific oligonucleotides (5' GGCCCGCCGTCGCGGCAGC 3' (SEQ No. 538); 5' CTCCTTCTCCAGGTTTGCTGGC 3' (SEQ No. 539); 5' AACTGAGACCCCTTGGAGATGC 3' (SEQ No. 540); 5' TAGTCGTCCTCCCCGCCACC 3' (SEQ No. 541); 5' GGAGGGTGGGACAAACACAGG 3' (SEQ No. 542); 5' ACGTTGAAGTGACCAACGAGGC 3' (SEQ No. 543)). Sequence analysis revealed amino acid sequence similarity of the core extracellular domain compared to the Rhesus CDH3 GenBank sequence (JU473826, JU473827). All cloning procedures were performed according to standard protocols (Sambrook, Molecular Cloning; A Laboratory Manual, 3rd edition, Cold Spring Harbour Laboratory Press, Cold Spring Harbour, New York (2001)). For each construct, the corresponding plasmid was transfected into DHFR-deficient CHO cells for eukaryotic expression as described in the literature [Kaufman RJ (1990) Methods Enzymol. 185, 537-566].

[0436] The expression of CDH3 (human, murine, and chimeric constructs) on CHO cells was confirmed by FACS analysis using a murine cross-reactive monoclonal mouse IgG1 anti-human CDH3 antibody. The conjugated monoclonal antibody was detected using anti-mouse IgG Fcg-PE. As a negative control, cells were incubated with PBS / 2% FCS instead of the first antibody. Samples were measured by flow cytometry. The results are shown in Figure 3. The expression of human and cyno-CDH3 in CHO cells (see Example 5) was detected by PE-conjugated R&D 861-P.

[0437] Examples 2

[0438] Murin scFv -Short story Epitope Clustering

[0439] Cells transfected with human or murine CDH3, or with chimeric human / mouse CDH3 molecules (see Example 1), were stained with crude, undiluted peripheral cytoplasmic extract containing scFv γ-binding to human / macaque CDH3. The bound scFv molecules were detected with mouse monoclonal anti-FLAG-M2 antibody (1 µg / mL; 50 µl in PBS / 2% FCS; Sigma F1804) followed by anti-mouse IgG Fcg-PE (1:100, 50 µl; Jackson Immunoresearch # 115-116-071). All antibodies were diluted in PBS using 2% FCS. As a negative control, cells were incubated with PBS / 2% FCS instead of the peripheral cytoplasmic extract. Samples were measured by flow cytometry. The results are shown in Figure 4.

[0440] Specifically, Fig. 4a shows a binder that recognizes the extracellular domain D1 of human CDH3, and more precisely, the sub-domain D1B (loss of FACS signal in each chimeric CDH3 construct). The binder named CDH3-6 refers to a potential binder for CDH3-4. The binder named CDH3-10 is a parenteral binder for CDH3-1, CDH3-2, and CDH3-3. Fig. 4b shows a binder that recognizes the extracellular domain D2 of human CDH3, and more precisely, the sub-domain D2C. Note that the binder named CDH3-21 is a parenteral binder for CDH3-11, CDH3-12, and CDH3-14. The binder named CDH3-23 is a parenteral binder for CDH3-13. Finally, FIG. 4c shows a binder that recognizes the extracellular domain D3 of human CDH3, and more precisely, the sub-domain D3A. It is noted that the binder CDH3-32 further binds to the sub-domain D3C. The binder named CDH3-32 is a parenteral binder for CDH3-25, CDH3-26, and CDH3-27. The binder named CDH3-33 is a parenteral binder for CDH3-24. The term "parenteral binder" implies that these binders are further developed to produce or obtain a binder optimized in this regard.

[0441] Epitope clustering analysis was similarly performed on binders CDH3-11, CDH3-12, CDH3-13, and CDH3-14, and they were found to recognize the extracellular domain D2 of human CDH3, and more precisely, the sub-domain D2C (data not shown). Furthermore, the same analysis was performed on binders CDH3-24, CDH3-25, CDH3-26, and CDH3-27. These binders recognized the extracellular domain D3 of human CDH3, and more precisely, the sub-domain D3A (data not shown).

[0442] Examples 3

[0443] Human and Cynomolgus CDH3 Antibodies against of affinity Viacore Decision based on

[0444] To determine the CDH3 target binding of the antibody of the present invention, a viacore analysis experiment was performed using recombinant CDH3 (human and cyno CDH3, respectively) fusion proteins together with human albumin (HALB).

[0445] Specifically, the CM5 sensor chip (GE Healthcare) was immobilized using approximately 600 to 800 RU of acetate buffer pH 4.5, respectively, according to the manufacturer's manual. CDH3xCD3 bispecific antibody samples were loaded at five concentrations: 50 nM, 25 nM, 12.5 nM, 6.25 nM, and 3.13 nM, diluted in HBS-EP execution buffer (GE Healthcare). The flow rate was 30 µl / min for 3 minutes, followed by the application of HBS-EP execution buffer at a flow rate of 30 µl / ml for 8 to 20 minutes. Chip regeneration was performed using a 10 mM glycine and 10 mM NaCl pH 1.5 solution. Data sets were analyzed using BiaEval Software. Generally, two independent experiments were conducted.

[0446] The CDH3xCD3 bispecific antibody according to the present invention exhibited high affinity for human CDH3 in the 1-digit nanomolar range. Binding to macaque CDH3 was balanced and also exhibited affinity in a similar range. The affinity values ​​as well as the calculated affinity gap are shown in Table 2. CDH3-25 x F12q-HALB and CDH3-13- xI2C-HLE(Fc) were measured in separate analyses, respectively, and are presented as follows:

[0447] KD(hu) 31.8 ± 1.9 nM, KD(cyno) 40.9 ± 3.89 nM and affinity gap 1.29, and

[0448] KD(hu) 12.95 ± 0.5 nM, KD(cyno) 12.35 ± 0.35 nM, and affinity gap 0.95, respectively.

[0449] Affinity of CDH3xCD3 bispecific antibodies for human and macaque CDH3 as determined by Viacore analysis, as well as the calculated interspecies affinity gap. CDH3xCD3 bispecific antibodies KD hu CDH3 [nM] KD Cyno CDH3 [nM] Affinity gap Cyno / hu CDH3-11 7.5 6.2 0.83 CDH3-12 8.8 7.5 0.85 CDH3-13 7.4 6.0 0.81 CDH3-14 9.5 8.1 0.85 CDH3-24 6.1 5.2 0.85 CDH3-25 7.9 6.7 0.85 CDH3-26 8.8 7.5 0.85 CDH3-27 8.7 7.6 0.87

[0450] Examples 4

[0451] Human and on target antigen-positive cells Macaque CDH3 Korea CDH3xCD3 Scatchard-based analysis of bispecific antibody affinity and interspecies affinity gap's decision

[0452] The affinity of CDH3xCD3 bispecific antibodies for CHO cells transfected with human or macaque CDH3 was determined by Scatchard analysis as the most reliable method to measure the potential affinity gap between human and macaque CDH3. For Scatchard analysis, saturation binding experiments were performed using a monovalent detection system to accurately determine the monovalent binding of CDH3xCD3 bispecific antibodies.

[0453] 2 x 10⁴ cells of each cell line (recombinant human CDH3-expressing CHO cell line, recombinant macaque CDH3-expressing CHO cell line) were incubated with 50 µl of each triple serial dilution of CDH3xCD3 bispecific antibody (12 times diluted at 1:2) starting at 10 to 20 nM (until saturation was reached), and then incubated at 4°C for 16 hours under stirring and one residue wash step. Subsequently, the cells were incubated with 30 µl of CD3xALEXA488 conjugate solution for another 1 hour. After a single washing step, cells were resuspended in 150 µl of FACS buffer containing 3.5% formaldehyde and incubated for an additional 15 minutes. After centrifugation, the cells were resuspended in FACS buffer and analyzed using a FACS Canto II machine and FACS Diva software. Data were generated from two independent sets of experiments (each used in triplicate). Maximum binding (Bmax) was inferred by calculating each Scatchard analysis. The concentration of the CDH3xCD3 bispecific antibody at half maximum binding was determined by reflecting each KD. The values ​​from the triplicate measurements were plotted as a hyperbola and as an S-shaped curve showing the appropriate concentration range from minimum to optimal binding.

[0454] The values ​​shown in Table 3 were derived from two independent experiments per CDH3xCD3 bispecific antibody. Cell-based scatched analysis confirmed that the CDH3xCD3 bispecific antibody of the present invention has a nanomolar to subnanomolar affinity for human CDH3 and a small cyno-human CDH3 affinity gap of approximately 1. CDH3-25 x F12q-HALB and CDH3-13 x I2C-HLE(Fc) were measured in separate analyses, respectively, and are presented as follows.

[0455] Cell-based affinity (hu) 0.37 ± 0.11 nM, cell-based affinity (cyno) 0.35 ± 0.08 nM and affinity gap 0.95, and

[0456] Cell-based affinity (hu) 0.32 ± 0.003 nM, cell-based affinity (cyno) 0.5 ± 0.09 nM, and affinity gap 1.56, respectively.

[0457] Calculated affinity gap KD. Affinity (KD) of CDH3xCD3 bispecific antibodies as determined by cell-based scatched assays with macaque CDH3 / KD human CDH3. Antibodies were measured in two independent experiments (using three duplicates each). CDH3xCD3 bispecific antibodies Cell-based affinity hu CDH -3 [nM] Cell-based affinity Mac CDH -3 [nM] Affinity gap KDmac / KDhu CDH -3 CDH3-11 1.74 ± 0.37 1.98 ± 0.67 1.14 CDH3-12 0.40 ± 0.27 0.52 ± 0.21 1.30 CDH3-13 0.42 ± 0.35 0.41 ± 0.28 0.98 CDH3-14 0.49 ± 0.14 0.41 ± 0.03 0.84 CDH3-24 0.09 ± 0.03 0.15 ± 0.07 1.67 CDH3-25 0.12 ± 0.01 0.11 ± 0.03 0.96 CDH3-26 0.17 ± 0.01 0.16 ± 0.04 0.94 CDH3-27 0.20 ± 0.10 0.23 ± 0.03 1.15

[0458] Examples 5

[0459] Dual specificity binding and interspecies cross-reactivity

[0460] To confirm binding to human CDH3 and CD3 and to cyano CDH3 and CD3, bispecific antibodies were tested by flow cytometry using the following.

[0461] CHO cells transfected with human and cyno CDH3, respectively,

[0462] Human CDH3-positive foreskin carcinoma cell line A431,

[0463] CD3-expressing human T-cell leukemia cell line HPB-all (DSMZ, Braunschweig, ACC483), and

[0464] Cynomolgus CD3-expressing T cell line HSC-F.

[0465] In addition, murine CDH3-transfected CHO cells were used as a negative control.

[0466] For flow cytometry analysis, each cell line of 200,000 cells was incubated on ice for 30 minutes with 50 µl of purified bispecific antibody at a concentration of 5 µg / ml. After washing the cells twice in PBS / 2% FCS, binding of the construct was detected using an internal mouse antibody specific to the CD3 binding site. After washing, the bound mouse antibody was detected using goat anti-mouse Fcγ-PE. Samples were measured by flow cytometry.

[0467] The results are shown in Figures 5a and 5b. After staining CHO cells transfected with human CDH3 and cyano CDH3 with the CDH3xCD3 bispecific antibody of the present invention, they were also stained with human CDH3-positive pyloric carcinoma cell line A431 (natural expresser) as well as human and cyano T cell-expressed CD3. Furthermore, there was no staining of negative control cells (murine CDH3-transfected CHO).

[0468] Examples 6

[0469] Human and Macaque In the paralog Confirmation of the joint member for Korea

[0470] Human and macaque CDH3 paralogs (CDH1 = E-cadherin, CDH2 = N-cadherin, CDH4 = R-cadherin, and CDH5 = VE-cadherin) dhfr - / - The cells were stably transfected into CHO cells. Protein expression was confirmed by FACS analysis using antibodies specific to each paralog. The antibodies were R&D MAB18381 (for CDH1), eBioscience 12-3259-41 (for CDH2), R&D Systems polyclonal AF2217 (for CDH4), and BD Bioscience # 555661 (for CDH5).

[0471] Identify the sequences of the paralogs as used in this embodiment in the sequence list (Sequence Nos. 41-44). They can also be found at the following GenBank accession numbers:

[0472] NM_004360 [Human CDH1]

[0473] NM_001792 [Human CDH2]

[0474] NM_001794 [Human CDH4]

[0475] NM_001795 [Human CDH5]

[0476] Flow cytometry was performed as described in Example 5. The results are shown in Figures 5c and 5d. The analysis confirmed that none of the CDH3xCD3 bispecific antibodies of the present invention cross-reacted with the human CDH3 paralog tested.

[0477] It was also confirmed that the antibody of the present invention does not cross-react with the macaque CDH3 paralogs CDH1, CDH2, CDH4, and CDH5 (data not shown). The expression of macaque paralogs on CHO cells was confirmed for human paralogs by the same antibody described above. The sequences of the macaque paralogs as used in this example are identified in the sequence list (Sequence Nos. 45 to 48). They can also be found at the following GenBank accession numbers:

[0478] XM_002802516 [Macaque CDH1]

[0479] JU321883 [Macaque CDH2]

[0480] XM_002802511 [Macaque CDH5]

[0481] Additionally, the macaque CDH4 sequence was obtained from the Ensembl Genome Browser (ENSMMUT00000017252) and fused with the human CDH4 signaling peptide at the in-frame N-terminus (amino acids 1 to 19).

[0482] Examples 7:

[0483] Identity regarding the human reproductive lineage

[0484] To analyze the identity / similarity of antibody sequences with respect to human antibody germline genes, the CDH3 binders of the present invention were aligned as follows: a complete VL containing all CDRs was aligned; and a complete VH containing CDRs 1 and 2 but excluding CDR3 was aligned with respect to human antibody germline genes (Vbase). Further details can be found in the specification of this application. The results are shown in Table 4 below:

[0485] CDH3xCD3 bispecific antibodies In the human reproductive system Korean VH and VL's sameness[ % ] CDH3-11 92.7 CDH3-12 93.6 CDH3-13 87.8 CDH3-14 94.1 CDH3-24 89.4 CDH3-25 89.8 CDH3-26 89.8 CDH3-27 90.7

[0486] Examples 8

[0487] cytotoxic activity

[0488] The efficacy of the CDH3xCD3 bispecific antibody of the present invention in converting effector T cells against CDH3-expressing target cells was analyzed in five in vitro cytotoxicity assays:

[0489] The efficacy of the CDH3xCD3 bispecific antibody in converting human CD8+ effector T cells stimulated against human CDH3-transfected CHO cells was measured in an 18-hour 51-chromium release assay. Refer to Figure 6 for the results.

[0490] The efficacy of the CDH3xCD3 bispecific antibody in converting stimulated human CD8+ effector T cells against CDH3-positive human carcinoma cell line A431 was measured in an 18-hour 51-chromium release assay. Refer to Figure 7 for the results.

[0491] The efficacy of the CDH3xCD3 bispecific antibody in converting T cells in unstimulated human PBMCs to human CDH3-transfected CHO cells was measured in a 48-hour FACS-based cytotoxicity assay. Refer to Figure 8 for the results.

[0492] The efficacy of the CDH3xCD3 bispecific antibody in converting T cells in human PBMCs unstimulated against CDH3-positive human carcinoma cell line A431 was measured in a 48-hour FACS-based cytotoxicity assay. Refer to Figure 9 for the results.

[0493] To confirm whether cross-reactive CDH3xCD3 bispecific antibodies can convert macaque T cells against macaque CDH3-transfected CHO cells, a 48-hour FACS-based cytotoxicity assay was performed using macaque T cell lines as effector T cells. Refer to Figure 10 for the results.

[0494] Examples 8.1

[0495] Chromium release analysis using stimulated human T cells

[0496] CD8 + T cell-rich stimulated T cells were obtained as described below. Petri dishes (145 mm diameter, Greiner bio-one GmbH, Kremsmünster) were coated with a commercially available anti-CD3 specific antibody (OKT3, Orthoclone) at a final concentration of 1 µg / ml for 1 hour at 37°C. Unbound proteins were removed by a single wash step using PBS. 3 to 5 x 10⁶ cells were placed in a pre-coated Petri dish in 120 ml of RPMI 1640 with stabilized glutamine / 10% FCS / IL-2 20 U / ml (Proleukin (registered trademark), Chiron). 7Canine human PBMCs were added and stimulated for 2 days. On the 3rd day, cells were collected and washed once with RPMI 1640. IL-2 was added to a final concentration of 20 U / mL, and the cells were cultured again for 1 day in the same cell culture medium as above. CD8 using Dynal-Beads according to the manufacturer's protocol + Cytotoxic T lymphocytes (CTLs) CD4 + T cells and CD56 + Concentrated by the depletion of NK cells.

[0497] After washing cyano-CDH3- or human CDH3-transfected CHO target cells twice with PBS, 11.1 MBq in a final volume of 100 µl RPMI using 50% FCS at 37°C for 60 minutes. 51Cr was used for labeling. Subsequently, the labeled target cells were washed three times with 5 mL of RPMI and then used for cytotoxicity analysis. The analysis was performed in 96-well plates with a total volume of 200 µL of RPMI supplemented with a 10:1 E:T ratio. Purified bispecific antibodies at starting concentrations of 0.01 to 1 µg / mL and 3-fold dilutions thereof were used. The incubation time for the analysis was 18 hours. Cytotoxicity was determined as the relative difference in released chromium in the supernatant relative to the difference between maximum lysis (addition of Triton-X) and spontaneous lysis (no effector cells). All measurements were performed in four duplicates. Chromium activity in the supernatant was measured using a Wizard 3'' gamma counter (Perkin Elmer Life Sciences GmbH, Cologne, Germany). The results were analyzed using Prism 5 for Windows (Version 5.0, GraphPad Software Inc., San Diego, California). The EC50 values ​​calculated by the analysis program from the S-shaped dose-response curve were used to compare cytotoxic activity.

[0498] Examples 8.2

[0499] human being CDH3 -Transfected CHO The efficacy of converting human effector T cells stimulated against cells

[0500] 51-chromium using human CDH3 as target cells and CHO cells transfected with stimulated human CD8+ T cells as effector cells ( 51 The cytotoxic activity of the CDH3xCD3 dual-specific antibody according to the present invention was analyzed in a Cr)-release cytotoxicity assay. The experiment was performed as described in Example 8.1.

[0501] The results are shown in Figure 6 and Table 5. The CDH3xCD3 bispecific antibody exhibited strong cytotoxic activity against human CDH3-transfected CHO cells even below the 1-digit picomolar range. The claimed antibody, specific to the epitope cluster corresponding to positions 291 to 363 of human CDH3, provides a favorable epitope-activation relationship supporting strong bispecific antibody-mediated cytotoxic activity. CDH3-25 x F12q-HALB and CDH3-13 xI2C-HLE(Fc) were measured in separate analyses, respectively, and were found to have EC50 values ​​of 3.6 pM and 8.9 pM, respectively.

[0502] 51-chromium using human CDH3 as target cells and CHO cells transfected with stimulated human CD8 T cells as effector cells ( 51 The EC50 value [pg / mL] of the CDH3xCD3 bispecific antibody was analyzed in the Cr) release cytotoxicity assay. CDH3xCD3 bispecific antibodies EC50 [ pg / ㎖] CDH3-11 76 CDH3-12 370 CDH3-13 7.2 CDH3-14 138 CDH3-24 14 CDH3-25 4.9 CDH3-26 21 CDH3-27 61

[0503] Examples 8.3

[0504] CDH3 - Hermaphroditic humans carcinoma The efficacy of converting human effector T cells stimulated for lineage A431

[0505] 51-chromium using CDH3-positive human foreskin carcinoma cell line A431 as a source of target cells, and stimulated human CD8+ T cells as effector cells ( 51 The cytotoxic activity of the CDH3xCD3 bispecific antibody was analyzed in a Cr)-release cytotoxicity assay. The analysis was performed as described in Example 8.1.

[0506] According to the results of the 51-chromium release assay by stimulated enriched human CD8+ T lymphocytes as effector cells and human CDH3-transfected CHO cells as target cells, the CDH3xCD3 bispecific antibody of the present invention also exhibits potent cytotoxic activity against natural expressor target cells (Fig. 7 and Table 6). CDH3-25 x F12q-HALB and CDH3-13 xI2C-HLE (Fc) were measured in separate assays, respectively, and were found to have EC50 values ​​of 1.2 pM and 16 pM, respectively.

[0507] 18-hour 51-chromium using CDH3-positive human carcinoma cell line A431 as a source of target cells, and stimulated enriched human CD8 T cells as effector cells ( 51 The EC50 value [pg / mL] of the CDH3xCD3 bispecific antibody was analyzed in the Cr) release cytotoxicity assay. CDH3xCD3 bispecific antibodies EC50 [ pg / ㎖] CDH3-11 47 CDH3-12 362 CDH3-13 62 CDH3-14 385 CDH3-24 5.8 CDH3-25 4.6 CDH3-26 19 CDH3-27 31

[0508] Examples 8.4

[0509] Non-stimulated human being to PBMC by FACS --based cytotoxicity analysis

[0510] Isolation of effector cells

[0511] Human peripheral blood monocytes (PBMCs) were prepared by Ficoll density gradient centrifugation from concentrated lymphocyte preparations (soft layer) and byproducts from blood banks collecting blood for transfusion. The soft layer was supplied by a local blood bank, and the PBMCs were prepared on the same day as blood collection. After Ficoll density centrifugation and extensive washing with Dulbecco PBS (Gibco), remaining red blood cells were removed from the PBMCs by incubation with erythrocyte lysis buffer (155 mM NH4Cl, 10 mM KHCO3, 100 μM EDTA). Platelets were removed from the supernatant during centrifugation of the PBMCs at 100 x g. The remaining lymphocytes mainly consist of B and T lymphocytes, NK cells, and monocytes. PBMC was maintained in a culture at 37°C / 5% CO2 in RPMI medium (Gibco) containing 10% FCS (Gibco).

[0512] CD14 + and CD56 + Cell depletion

[0513] CD14 +Human CD14 microbeads (Milteny Biotec, MACS, #130-050-201) were used for cell depletion, and human CD56 microbeads (MACS, #130-050-401) were used for NK cell depletion. PBMCs were counted and centrifuged at 300 xg for 10 minutes at room temperature. After discarding the supernatant, the cell pellet was immersed in MACS isolation buffer [80 µl / 10 7 Canine cells were resuspended in PBS (Invitrogen, #20012-043), 0.5% (v / v) FBS (Gibco, #10270-106), and 2 mM EDTA (Sigma-Aldrich, #E-6511). CD14 micro beads and CD56 micro beads (20 µl / 10 7 After adding the cells, the cells were incubated at 4 to 8°C for 15 minutes. The cells were then treated with MACS isolation buffer (1 to 2 ml / 10 7 The cells were washed with (see above). After centrifugation (see above), the supernatant was discarded, and the cells were washed with MACS isolation buffer (500 µl / 10 8 They were resuspended in dog cells. Subsequently, CD14 / CD56 negative cells were isolated using an LS column (Miltenyi Biotec, #130-042-401). PBMC w / o CD14+ / CD56+ cells were cultured in RPMI complete medium, i.e. RPMI1640 (Biochrome AG, #FG1215), supplemented until necessary with 10% FBS (Biochrome AG, #S0115), 1x non-essential amino acids (Biochrome AG, #K0293), 10 mM Hepes buffer (Biochrome AG, #L1613), 1 mM sodium pyruvate (Biochrome AG, #L0473), and 100 U / mL penicillin / streptomycin (Biochrome AG, #A2213) at 37°C in an incubator.

[0514] Target cell labeling

[0515] For the analysis of cell lysis in flow cytometry, fluorescent dye DiOC 18 Human CDH3- or macaque CDH3-transfected CHO cells were labeled as target cells using (DiO) (Molecular Probe, #V22886) and distinguished from effector cells. Briefly, after collecting cells, they were washed once with PBS, 2% (v / v) FBS, and membrane dye DiO (5 µl / 10 6 10 in PBS containing dog cells 6 The cell count was adjusted to 1 cell / mL. After incubation at 37°C for 3 minutes, the cells were washed twice in complete RPMI medium, and the cell count was 1.25 x 10⁶ 5 The cell volume was adjusted to 1 cell / mL. Cell viability was determined using 0.5% (v / v) isotonic eosin G solution (Roth, #45380).

[0516] Flow cytometry Basic analysis

[0517] This assay was designed to quantify the lysis of cyano or human CDH3-transfected CHO cells in the presence of serial dilutions of CDH3 bispecific antibodies. Equal volumes of DiO-labeled target cells and effector cells (i.e., PBMC w / o CD14) +Cells were mixed to result in an E:T cell ratio of 10:1. 160 µl of this suspension was transferred to each well of a 96-well plate. 40 µl of serial dilutions of CDH3xCD3 bispecific antibody and negative control bispecific antibody (a CD3-based bispecific antibody that recognizes an inappropriate target antigen) or RPMI complete medium as an additional negative control were added. The bispecific antibody-mediated cytotoxicity reaction was carried out for 48 hours in a 7% CO2 humidified incubator. Subsequently, cells were transferred to a new 96-well plate, and the loss of target cell membrane integrity was monitored by adding propidium iodide (PI) at a final concentration of 1 µg / ml. PI is a membrane-impermeable dye that is normally excluded from viable cells, whereas it takes up apoptotic cells, which can be identified by fluorescence emission.

[0518] Samples were measured by flow cytometry on a FACSCanto II instrument and then analyzed using FACSDiva software (both from Becton Dickinson). Target cells were identified as DiO-positive cells. PI-negative target cells were classified as live target cells. The percentage of cytotoxicity was calculated according to the following formula:

[0519]

[0520] Using GraphPad Prism 5 software (GraphPad software based in San Diego), the percentage of cytotoxicity was plotted against the corresponding bispecific antibody concentration. To evaluate the s-shaped dose-response curve with a fixed slope, the dose-response curve was analyzed using a 4-parameter logistic regression model, and the EC50 value was calculated.

[0521] Examples 8.5

[0522] human being CDH3 -Transfected CHO About cells Non-stimulated human being PBMC The efficacy of conversion

[0523] The cytotoxic activity of the CDH3xCD3 bispecific antibody was analyzed using FACS-based cytotoxicity assays with CHO cells transfected with human CDH3 as the target cell and unstimulated human PBMC as the effector cell. The analysis was performed as described in Example 8.4 above.

[0524] Figure 8 and Table 7 show the results of a FACS-based cytotoxicity assay using unstimulated human PBMCs as effector cells and human CDH3-transfected CHO cells as target cells. CDH3-25 x F12q-HALB and CDH3-13 xI2C-HLE (Fc) were measured in separate assays and were found to have EC50 values ​​of 4.6 pM and 5.6 pM, respectively.

[0525] EC50 value [pg / mL] of CDH3xCD3 bispecific antibody as measured in a 48-hour FACS-based cytotoxicity assay using unstimulated human PBMCs as effector cells and CHOs transfected with human CDH3 as target cells. CDH3xCD3 bispecific antibodies EC50 [ pg / ㎖] CDH3-11 462 CDH3-12 1021 CDH3-13 129 CDH3-14 1885 CDH3-24 19 CDH3-25 10 CDH3-26 47 CDH3-27 61

[0526] Examples 8.6

[0527] CDH3 - Hermaphroditic humans carcinoma Regarding System A431 Non-stimulated human being PBMC The efficacy of conversion

[0528] Furthermore, the cytotoxic activity of the CDH3xCD3 bispecific antibody was analyzed in a FACS-based cytotoxicity assay using CDH3-positive human foreskin carcinoma cell line A431 as the source of target cells and unstimulated human PBMCs as effector cells. The analysis was performed as described in Example 8.4 above. The results are shown in Figure 9 and Table 8. CDH3-25 x F12q-HALB and CDH3-13 xI2C-HLE (Fc) were measured in separate assays, respectively, and were found to have EC50 values ​​of 2.3 pM and 32 pM, respectively.

[0529] The EC50 value [pg / mL] of the CDH3xCD3 bispecific antibody was measured in a 48-hour FACS-based cytotoxicity assay using unstimulated human PBMCs as effector cells and human A431 cell line as the source of target cells. CDH3xCD3 bispecific antibodies EC50 [ pg / ㎖] CDH3-11 389 CDH3-12 3141 CDH3-13 83 CDH3-14 4842 CDH3-24 15 CDH3-25 9.2 CDH3-26 35 CDH3-27 41

[0530] As expected, EC50 values ​​were generally higher in cytotoxicity assays using unstimulated PBMCs as effector cells compared to cytotoxicity assays using stimulated human CD8+ T cells.

[0531] Examples 8.7

[0532] Macaque CDH3 -expression CHO About cells Macaque The efficacy of converting T cells

[0533] Finally, the cytotoxic activity of the CDH3xCD3 bispecific antibody was analyzed in a FACS-based cytotoxicity assay using macaque (cyno) CDH3 as the target cell and CHO cells transfected with a macaque T cell line as the effector cells. Macaque T cell line 4119LnPx (Knappe et al. Blood 95:3256-61 (2000)) was used as the source of effector cells. Flow cytometry-based analysis of the target cell labeling and cytotoxic activity of macaque CDH3-transfected CHO cells was performed as described above.

[0534] The results are shown in Fig. 10 and Table 9. Macaque T cells from cell line 4119LnPx were induced to efficiently kill macaque CDH3-transfected CHO cells by the CDH3xCD3 bispecific antibodies of the present invention, namely antibodies that bind to epitope clusters of human CDH3 corresponding to positions 291 to 363 of human CDH3 and including neighboring subdomains D2C (positions 291 to 327) and D3A (positions 328 to 363). The antibodies are strongly presented in this analysis with 2-digit to very low 4-digit pg / ml EC50 values, which confirms that these antibodies are highly active in the macaque system.

[0535] Another group of anti-CDH3 antibodies was identified during epitope clustering that binds to the extracellular domain D1, and more specifically, to the sub-domain D1B of human CDH3 (see Example 2). Unexpectedly, the CDH3xCD3 bispecific antibodies of this group proved to exhibit very weak cytotoxic activity against macaque CDH3-transfected CHO cells (although they showed strong cytotoxic activity against CHO cells transfected with human CDH3) (Fig. 10c and Table 9). The antibodies of this group showed significantly weaker efficacy in terms of EC50 values ​​at very high 4-digits and even in the 5-digit pg / ml range. CDH3-25 x F12q-HALB and CDH3-13 xI2C-HLE(Fc) were measured in separate analyses, respectively, and were found to have EC50 values ​​of 1.4 pM and 4.0 pM, respectively.

[0536] Therefore, the CDH3xCD3 antibody of the present invention, which binds to the epitope cluster of CDH3 corresponding to positions 291 to 363, is about 5 to nearly 1000 times stronger in the macaque system than the antibody that binds to the extracellular domain D1, more specifically, to the CDH3 sub-domain D1B.

[0537] EC50 values ​​[pg / mL] of CDH3xCD3 bispecific antibodies binding to CDH3 epitope clusters corresponding to positions 291 to 363 (rows 1 to 8) and EC50 values ​​[pg / mL] of CDH3xCD3 bispecific antibodies binding to CDH3 epitope clusters / sub-domain D1B (rows 9 to 12) measured in a 48-hour FACS-based cytotoxicity assay using macaque T cell line 4119LnPx as effector cells and CHO cells transfected with macaque CDH3 as target cells. CDH3xCD3 bispecific antibodies EC50 [ pg / ㎖] 1 CDH3-11 1183 2 CDH3-12 1325 3 CDH3-13 257 4 CDH3-14 1015 5 CDH3-24 32 6 CDH3-25 14 7 CDH3-26 37 8 CDH3-27 98 9 CDH3-1 13055 10 CDH3-2 12862 11 CDH3-3 8390 12 CDH3-4 8795

[0538] Examples 9

[0539] (i) 3 freeze / thaw cycles and (ii) 7 days at 250 µg / ml Incubation Later, the monomer into dimers conversion

[0540] Different stress conditions were applied to the dual-specificity CDH3xCD3 antibody monomers, followed by high-performance SEC, to determine the percentage of initial monomer antibodies converted into antibody dimers.

[0541] (i) 15 µg of monomeric antibody was prepared at a concentration of 250 µg / ml using a standard formulation buffer, then frozen at -80°C for 30 minutes and thawed at room temperature for 30 minutes. After three freeze / thaw cycles, the dimer content was determined by HP-SEC.

[0542] (ii) 15 µg of monomeric antibody was adjusted to a concentration of 250 µg / ml using a standard formulation buffer and then incubated at 37°C for 7 days. The dimer content was determined by HP-SEC.

[0543] Acta equipped with an A905 autosampler ( A high-resolution SEC column TSK Gel G3000 SWXL (Tosoh, Tokyo, Japan) was connected to a Purifier 10 FPLC (GE Life Sciences). The column equilibrium and run buffer consisted of 100 mM KH2PO4–200 mM Na2SO4, and the pH was adjusted to 6.6. Antibody solution (15 µg protein) was applied to the equilibrium column, and elution was performed at a flow rate of 0.75 mL / min at a maximum pressure of 7 MPa. The entire run was monitored at optical absorbances of 280, 254, and 210 nm. Analysis was performed by peak integration of the 210 nm signal recorded in the Akta Unicorn software run evaluation sheet. The dimer content was calculated by dividing the area of ​​the dimer peak by the total area of ​​the monomer + dimer peaks.

[0544] The results are shown in Table 10 below. After 3 freeze / thaw cycles as well as 7 days of incubation at 37°C, the binding of the CDH3xCD3 bispecific antibody to the epitope cluster / extracellular sub-domain D2C yielded a dimer percentage ≤1%, more precisely a dimer percentage 0.0%, which is considered very good. The dimer conversion rate of the CDH3xCD3 bispecific antibody to the epitope cluster / extracellular sub-domain D3A reached 2% or less, more precisely a value of 0.2 to 1.8, which is considered good. CDH3-25 x F12q-HALB and CDH3-13 x I2C-HLE(Fc) were measured in separate analyses, and were found to have percentages of dimers after 3 freeze / thaw cycles of 1.1 and 0.84, respectively, and percentages of dimers after 7 days of incubation of 0.0 (both were HLE products).

[0545] Percentage of monomers for dimeric CDH3xCD3 bispecific antibodies as determined by high-performance size exclusion chromatography (HP-SEC). CDH3xCD3 antibodies After 3 freezing / thawing cycles dimer percentage 7 days Incubation after dimer percentage CDH3-11 0.00 0.00 CDH3-12 0.00 0.00 CDH3-13 0.00 0.00 CDH3-14 0.00 0.00 CDH3-24 1.01 0.20 CDH3-25 1.31 0.60 CDH3-26 0.82 1.80 CDH3-27 1.69 1.50

[0546] Examples 10

[0547] Thermal stability

[0548] The antibody agglutination temperature was determined as follows: 40 µl of antibody solution at 250 µg / ml was transferred to a disposable cuvette and placed in a Wyatt Dynamic Light Scattering device DynaPro Nanostar (Wyatt). The sample was heated from 40°C to 70°C at a heating rate of 0.5°C / min with a constant acquisition of the measurement radius. The antibody agglutination temperature was calculated using a software package that transmits the increase in radius indicating the melting of proteins and agglutination to the DLS device.

[0549] All tested CDH3xCD3 bispecific antibodies of the present invention exhibited very desirable thermal stability at agglutination temperatures above 54°C, as shown in Table 11 below. CDH3-25 x F12q-HALB was measured in a separate assay and was found to have thermal stability of 56.3°C.

[0550] Thermal stability of bispecific antibodies as determined by DLS (Dynamic Light Scattering) CDH3xCD3 bispecific antibodies Thermal stability (DLS ℃ Aggregation) CDH3-11 59.8 CDH3-12 55.9 CDH3-13 59.6 CDH3-14 59.6 CDH3-24 55.1 CDH3-25 55.4 CDH3-26 54.9 CDH3-27 54.1

[0551] Examples 11

[0552] In human plasma for 24 hours Incubation Subsequent stability

[0553] Bispecific antibodies purified from a human plasma pool were incubated in a 1:5 ratio at 37°C for 24 to 96 hours at a final concentration of 2 to 20 µg / ml. After plasma incubation, the antibodies were compared in a 51-chromium release assay (as described in Example 8.1 Chromium Release Assay using stimulated human T cells) using human T cells and CDH3-transfected CHO cells at starting concentrations of 0.01 to 0.1 µg / ml and an effector-to-target cell (E:T) ratio of 10:1. Freshly thawed, unincubated bispecific antibodies were included as a control. The results are shown in Table 12 below. All test antibodies had a desirable plasma stability (EC 4 or less). 50 Plasma / EC 50 The control group, and even the antibody binding group to D3A, had plasma stability of 3 or less. CDH3-25 x F12q-HALB and CDH3-13 x I2C-HLE (Fc) were measured in separate analyses, respectively, and are presented as follows.

[0554] EC50 w / plasma 4.4 pM, EC50 w / o plasma 3.6 pM, and plasma-to-control ratio 1.2, and

[0555] EC50 w / plasma 3.4 pM, EC50 w / o plasma 8.9 pM, and plasma-to-control ratio 0.4, respectively.

[0556] EC50 values ​​with and without plasma incubation and calculated plasma / control values CDH3xCD3 bispecific antibodies EC 50 [pg / ㎖] w / Plasma w / o plasma Plasma vs. Control Group Ratio (EC 50 Plasma / EC 50 Control group) CDH3 -11 47 76 0.6 CDH3 -12 296 370 0.8 CDH3 -13 25 7.2 3.5 CDH3 -14 291 138 2.1 CDH3 -24 19 14 1.4 CDH3 -25 9.9 4.9 2.0 CDH3 -26 24 21 1.1 CDH3 -27 129 61 2.1

[0557] Examples 12

[0558] Turbidity at an antibody concentration of 2500 µg / ml

[0559] 1 mL of purified monomeric antibody solution at 250 µg / mL was concentrated to 2500 µg / mL using a spin concentration unit. After storage at 5°C for 16 hours, the turbidity of the antibody solution was determined by measuring optical absorption at OD340 nm against a standard formulation buffer. The results are shown in Table 13 below. All test antibodies have a highly desirable turbidity of 0.05 or less. CDH3-25 x F12q-HALB and CDH3-13 x I2C-HLE(Fc) were measured in separate analyses, respectively, and were found to have turbidities of 0.066 and 0.026 at 2500 µg / mL, respectively.

[0560] Antibody turbidity after overnight concentration to 2.5 mg / mL CDH3xCD3 bispecific antibodies Turbidity at 2500 µg / ml CDH3-11 0.035 CDH3-12 0.025 CDH3-13 0.030 CDH3-14 0.025 CDH3-24 0.019 CDH3-25 0.026 CDH3-26 0.028 CDH3-27 0.022

[0561] Examples 14

[0562] In a human tumor xenograft model CDH3xCD3 Therapeutic efficacy of bispecific antibodies

[0563] On day 1 of the study, human pyloric carcinoma cell line A-431 and newly isolated human PBMC cells were co-injected subcutaneously into the right dorsal flank of female NOD / SCID mice (E:T cell ratio 1:2). Mice in vehicle control 1 (n=5) did not receive effector cells and were used as non-transplant controls compared to vehicle control 2 (n=10, received effector cells) to monitor the effect of PBMC on tumor growth in the absence of antibodies.

[0564] Mice were treated with 0.5 mg / kg / day (Group 3, n=10), 0.05 mg / kg / day (Group 4, n=10), or 0.005 mg / kg / day (Group 5, n=10) of a CDH3xCD3 bispecific antibody that specifically binds to the extracellular CDH3 sub-domain D2C by intravenous bolus injection daily for 10 consecutive days, starting approximately 2 hours after tumor cell injection on Day 1.

[0565] After measuring the tumor with a caliper during the study, progression was evaluated by comparing the tumor volume (TV) between groups. Tumor growth inhibition T / C[%] was determined by calculating the TV as T / C% = 100 x (median TV of the analysis group) / (median TV of control group 2).

[0566] The results are shown in Figure 11. Treatment of mice with the CDH3xCD3 bispecific antibody at a dose level of 0.5 mg / kg / day resulted in the complete inhibition of tumor formation, and none of the animals developed tumors by the end of the study (day 40).

[0567] Examples 15

[0568] In the HCT-116 tumor model CDH3xCD3 double singularity HLE antibody's Anti-tumor active

[0569] Analysis was performed in female NOD / SCID mice subcutaneously injected with human HCT-116 colon carcinoma cells. Effector cells were expanded in vitro, and human CD3 + T cells were activated (Day 12). Treatment was initiated when the tumor reached a volume of approximately 200 mm³ (Day 17). The control group was the q5d vehicle treatment group by T cells. Antibodies containing SEQ ID NO. 425 were administered via intravenous bolus injection every 5 days (q5d) at concentrations of 5 mg / kg / administration (Group 2) and 0.5 mg / kg / administration (Group 3). The results are shown in Figs. 12A and 12B. In particular, Fig. 12B distinguishes the results for responding animals (7 / 10) and non-responding animals (3 / 10). Although the reason for non-response in 30% of the animals is unclear, Fig. 12B indicates that administration of the bispecific construct with an extended half-life at a concentration of 5 mg / kg to responding animals caused the cessation of tumor growth starting at the moment of antibody administration.

[0570] Examples 16

[0571] T cells Seonghwa analyze

[0572] PBMCs isolated from healthy human donors were cultured for 48 hours while increasing the concentration of CDH3-13xI2C or CDH3-13xI2C-HALB dual-specificity antibody constructs (serial dilutions from 0.001 pM to 20 μM). The expression of CD69, an activation marker for CD4+ and CD8+ T cells, was determined by immunostaining, flow cytometry, and antigen-specific conjugate mAbs. The results are shown in Fig. 13 and discussed above in this specification.

[0573] Examples 17

[0574] Cyno-pharmacokinetic study of CDH3 x CD3 constructs with extended half-life

[0575] Female cynomolgus monkeys ate 0.015 mg / kg of half-life extended CDH3 x CD3 in a buffer The bispecific antibody constructs (dose volume 1 ml / kg) were administered via IV infusion over 60 minutes. The three HLE variants tested were P156, HALB, and HALB variant 1 (see SEQ ID NOs 437, 443, and 444), with one fused to the C-terminus of each construct. In the cynomodel, the half-lives of these HLE constructs were calculated based on their blood plasma concentrations analyzed at time 4 (P156) and HALB variant 1) between 96 hours after administration and the end of the study. As a result, the HLE construct induced by P156 showed a half-life of 57 hours, the HLE construct induced by HALB showed 63 to 85 hours, and the construct induced by HALB variant 1 showed 68 hours. These PK characteristics suggest once-weekly IV administration in humans.

[0576]

[0577]

[0578]

[0579]

[0580]

[0581]

[0582]

[0583]

[0584]

[0585]

[0586]

[0587]

[0588]

[0589]

[0590]

[0591]

[0592]

[0593]

[0594]

[0595]

[0596]

[0597]

[0598]

[0599]

[0600]

[0601]

[0602]

[0603]

[0604]

[0605]

[0606]

[0607]

[0608]

[0609]

[0610]

[0611]

[0612]

[0613]

[0614]

[0615]

[0616]

[0617]

[0618]

[0619]

[0620]

[0621]

[0622]

[0623]

[0624]

[0625]

[0626]

[0627]

[0628]

[0629]

[0630]

[0631]

[0632]

Claims

Claim 1 A bispecific short-chain antibody construct comprising a first human binding domain that binds to an epitope cluster of human CDH3 on the surface of a target cell and also binds to macaque CDH3, and a second binding domain that binds to human CD3 on the surface of a T cell; wherein the epitope cluster of human CDH3 is contained within amino acid positions 291 to 363 of human CDH3 as presented in SEQ ID NO. 1 (SEQ ID NO. 36); and the first and second binding domains have the following format: a pair of a VH region and a VL region in the format of a short-chain antibody (scFv); The redirected lysis of target cells through the recruitment of T cells by the above antibody construct entails the formation of lysis synapses and the delivery of perforin and granzyme, wherein the first binding domain binds to an epitope contained within amino acid positions 291 to 327 (SEQ ID 34) of human CDH3 as presented in SEQ ID 1, and a) CDR-H1 as presented in SEQ ID 149, CDR-H2 as presented in SEQ ID 150, CDR-H3 as presented in SEQ ID 151, CDR-L1 as presented in SEQ ID 152, CDR-L2 as presented in SEQ ID 153, and CDR-L3 as presented in SEQ ID 154; b) CDR-H1 as presented in SEQ ID NO. 159, CDR-H2 as presented in SEQ ID NO. 160, CDR-H3 as presented in SEQ ID NO. 161, CDR-L1 as presented in SEQ ID NO. 162, CDR-L2 as presented in SEQ ID NO. 163, and CDR-L3 as presented in SEQ ID NO. 164; c) CDR-H1 as presented in SEQ ID NO. 169, CDR-H2 as presented in SEQ ID NO. 170, CDR-H3 as presented in SEQ ID NO. 171, CDR-L1 as presented in SEQ ID NO. 172, CDR-L2 as presented in SEQ ID NO. 173, and CDR-L3 as presented in SEQ ID NO. 174;and d) comprising a VH region including CDR-H1, CDR-H2 and CDR-H3 and a VL region including CDR-L1, CDR-L2 and CDR-L3, selected from the group consisting of CDR-H1 as presented in SEQ ID NO. 179, CDR-H2 as presented in SEQ ID NO. 180, CDR-H3 as presented in SEQ ID NO. 181, CDR-L1 as presented in SEQ ID NO. 182, CDR-L2 as presented in SEQ ID NO. 183 and CDR-L3 as presented in SEQ ID NO. 184, or the first binding domain binds to an epitope contained within amino acid positions 328 to 363 (SEQ ID NO. 35) of human CDH3 as presented in SEQ ID NO. 1, and the first binding domain also binds to amino acid positions 404 to 440 (SEQ ID NO. 390) binds to an epitope included within, and the first binding domain is e) CDR-H1 as presented in SEQ ID NO. 279, CDR-H2 as presented in SEQ ID NO. 280, CDR-H3 as presented in SEQ ID NO. 281, CDR-L1 as presented in SEQ ID NO. 282, CDR-L2 as presented in SEQ ID NO. 283, and CDR-L3 as presented in SEQ ID NO. 284; f) CDR-H1 as presented in SEQ ID NO. 289, CDR-H2 as presented in SEQ ID NO. 290, CDR-H3 as presented in SEQ ID NO. 291, CDR-L1 as presented in SEQ ID NO. 292, CDR-L2 as presented in SEQ ID NO. 293, and CDR-L3 as presented in SEQ ID NO. 294; g) CDR-H1 as presented in SEQ ID NO. 299, CDR-H2 as presented in SEQ ID NO. 300, CDR-H3 as presented in SEQ ID NO. 301, CDR-L1 as presented in SEQ ID NO. 302, CDR-L2 as presented in SEQ ID NO. 303, and CDR-L3 as presented in SEQ ID NO. 304;and h) a bispecific single-chain antibody construct comprising a VH region comprising CDR-H1, CDR-H2 and CDR-H3 and a VL region comprising CDR-L1, CDR-L2 and CDR-L3, selected from the group consisting of CDR-H1 as presented in SEQ ID NO. 309, CDR-H2 as presented in SEQ ID NO. 310, CDR-H3 as presented in SEQ ID NO. 311, CDR-L1 as presented in SEQ ID NO. 312, CDR-L2 as presented in SEQ ID NO. 313 and CDR-L3 as presented in SEQ ID NO. 314.; Claim 2 A bispecific single-chain antibody construct according to claim 1, wherein the first binding domain comprises a VH region selected from the group consisting of VH regions as presented in SEQ ID NO. 155, SEQ ID NO. 165, and SEQ ID NO. 175, and / or the first binding domain comprises a VL region selected from the group consisting of VL regions as presented in SEQ ID NO. 156 and SEQ ID NO.

176. Claim 3 A bispecific single-chain antibody construct according to claim 1, wherein the first binding domain comprises a VH region selected from the group consisting of VH regions presented in SEQ ID NO. 295 and SEQ ID NO. 305 and / or the first binding domain comprises a VL region selected from the group consisting of VL regions as presented in SEQ ID NO. 296 and SEQ ID NO.

306. Claim 4 A bispecific single-chain antibody construct according to claim 2, wherein the first binding domain comprises a VH region and a VL region selected from the group consisting of pairs of VH regions and VL regions as presented in SEQ ID NOs 155+156 and SEQ ID NOs 175+176. Claim 5 A bispecific single-chain antibody construct, wherein the first binding domain comprises a VH region and a VL region selected from the group consisting of pairs of VH regions and VL regions as presented in SEQ ID NOs 295+296 and SEQ ID NOs 305+306. Claim 6 A bispecific single-chain antibody construct according to claim 1, wherein the first binding domain comprises an amino acid sequence selected from the group consisting of those presented in SEQ ID NO. 157 and SEQ ID NO.

177. Claim 7 A bispecific single-chain antibody construct according to claim 1, wherein the first binding domain comprises an amino acid sequence selected from the group consisting of those presented in SEQ ID NO. 297 and SEQ ID NO.

307. Claim 8 A bispecific single-chain antibody construct according to claim 1, wherein the second binding domain binds to human and silk monkey (Callithrix jacchus), downy-headed tamarin (Saguinus Oedipus) or squirrel monkey (Saimiri sciureus) CD3 epsilon. Claim 9 A bispecific single-chain antibody construct according to claim 1, comprising an amino acid sequence selected from the group consisting of those presented in SEQ ID NO. 158 and SEQ ID NO. 178, or comprising an amino acid sequence selected from the group consisting of those presented in SEQ ID NO. 379, SEQ ID NO. 380, SEQ ID NO. 381, SEQ ID NO. 382, ​​SEQ ID NO. 383, SEQ ID NO. 384, SEQ ID NO. 385, SEQ ID NO. 386, SEQ ID NO. 387, SEQ ID NO. 388, SEQ ID NO. 389, SEQ ID NO. 422, SEQ ID NO. 423, SEQ ID NO. 424, SEQ ID NO. 425, SEQ ID NO. 426, and SEQ ID NO.

427. Claim 10 A bispecific single-chain antibody construct according to claim 1, comprising an amino acid sequence selected from the group consisting of those presented in SEQ ID NO. 298 and SEQ ID NO.

308. Claim 11 A polynucleotide encoding a bispecific single-chain antibody construct as described in any one of claims 1 to 10. Claim 12 A vector for transforming or transfecting a host cell, comprising a polynucleotide as described in paragraph 11. Claim 13 Isolated host cells for the expression of a bispecific short-chain antibody construct, which are transformed or transfected with a polynucleotide as described in claim 11, or transformed or transfected with a vector for transforming or transfecting a host cell comprising said polynucleotide. Claim 14 A method for generating a bispecific short-chain antibody construct according to any one of claims 1 to 10, wherein the method comprises, in vitro, isolating a host cell transformed or transfected with a polynucleotide encoding the bispecific short-chain antibody construct under conditions for expressing the antibody construct as described in any one of claims 1 to 10 ( in vitro A method comprising the steps of culturing in ), and recovering the generated bispecific antibody construct from said culture. Claim 15 A pharmaceutical composition for the prevention, treatment, or improvement of a tumor or cancer, comprising a bispecific short-chain antibody construct according to any one of claims 1 to 10, or comprising a bispecific short-chain antibody construct produced according to a method for producing said bispecific short-chain antibody construct, wherein the method comprises the steps of: culturing a host cell transformed or transfected with a polynucleotide encoding said bispecific short-chain antibody construct under conditions for expressing an antibody construct as described in any one of claims 1 to 10; and recovering said produced bispecific antibody construct from said culture. Claim 16 In paragraph 15, the above cancers are lung carcinoma, head and neck carcinoma, primary or secondary CNS tumor, primary or secondary brain tumor, primary CNS lymphoma, spinal tumor, brainstem glioma, pituitary adenoma, adrenocortical carcinoma, esophageal carcinoma, colon cancer, breast cancer, ovarian cancer, NSCLC (non-small cell lung cancer), SCLC (small cell lung cancer), endometrial cancer, cervical cancer, uterine cancer, transitional cell carcinoma, bone cancer, pancreatic cancer, skin cancer, cutaneous or intraocular melanoma, liver cancer, cholangiocarcinoma, gallbladder cancer, kidney cancer, rectal cancer, cancer of the anal region, stomach cancer, gastric cancer, colorectal cancer, duodenal cancer, small intestine cancer, biliary tract cancer, urethral cancer, renal cell carcinoma, endometrial carcinoma, thyroid cancer, testicular cancer, cutaneous squamous cell carcinoma, melanoma, prostate cancer, bladder cancer, osteosarcoma, mesothelioma, Hodgkin's disease, non-Hodgkin lymphoma, chronic or acute A pharmaceutical composition selected from the group consisting of leukemia, chronic myeloid leukemia, lymphocytic lymphoma, multiple myeloma, fibrosarcoma, neuroblastoma, retinoblastoma, soft tissue sarcoma, and squamous cell carcinoma. Claim 17 delete Claim 18 delete Claim 19 delete Claim 20 delete Claim 21 delete Claim 22 delete Claim 23 delete Claim 24 delete Claim 25 delete Claim 26 delete Claim 27 delete Claim 28 delete Claim 29 delete Claim 30 delete Claim 31 delete Claim 32 delete

Images