Bispecific antibody constructs for cdh3 and cd3

The present invention relates to bispecific antibody constructs comprising a first human binding domain that binds to human CDH3 on the surface of target cells and a second binding domain that binds to human CD3 on the surface of T cells. In addition, the present invention provides polynucleotides encoding antibody constructs, vectors containing polynucleotides and host cells transformed with or transfected with polynucleotides or vectors. In addition, the present invention provides a method of producing the antibody construct of the present invention, medical use of the antibody construct and a kit comprising the antibody construct.

Introduction: The cadherin superfamily includes more than 100 members in humans, including the so-called classical cadherins P-cadherin, E-cadherin, N-cadherin and R-cadherin, each of which is an individual tissue set. It is active in (Takeichi M. Development, 102: 639-55 (1988), van Roy F., Nature Rev., V14: 121-134 (2014)). Cadherins play an essential role in homeostasis of various tissues, among other things, in addition to 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 adhesion processes by forming calcium-dependent bonds and by converting mechanical stimuli into electrochemical activity, a process called 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)).

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 approximately 800 amino acids. The CDH3 ECD contains five cadherin repeats, which are referred to herein as (extracellular) domains 1-5 / Dom1-Dom5 / D1-D5, each consisting 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.

Expression levels of P-cadherin are low in healthy adult individuals and are thought to be confined to stratified epithelium, including the prostate and skin, and the basal or inferior layers of myoepithelial cells in the breast (Takeichi MJ Cell Biol 103: 2649). -58, (1986) and Shimoyama Y, et al., Cancer Res, 49: 2128-33 (1989)). Loss of P-cadherin function in gene knockout mice has been shown to be associated with developmental disorders, as well as mammary epithelial hyperplasia and dysplasia, although not lethal (GL Radice et al. J. Cell Biol. 139). 1025-1032 (1997)).

In contrast to low gene expression levels in healthy individuals, P-cadherin expression is upregulated in the context of several diseases, including immune disorders such as Crohn's disease and colitis (Hardy, et al., Gut 50: 513-519 (2002)). In addition, P-cadherin is thought to play an important role in promoting cancer cell infiltration (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 relation to a variety of tumors, including cancers of the rectum, lung (NSCLC), breast (triple negative), pancreas, head and neck, thyroid, neck, ovary and stomach. Tekita (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)). In addition, increased expression of P-cadherin was observed to correlate with poor survival in patients in 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).

Following cardiovascular disease, neoplasms are the second leading cause of death in the non-communicable disease category, with approximately 8.3 million deaths worldwide in 2013 (GBD 2013 Lancet 2015; 385: 117-71). The absolute number of cancer cases has increased by 45.6% since 1990, which means that the world population is growing and people are getting older and established risk factors (eg smoking, overweight, physical inactivity). It may be due to the fact that the prevalence of cancer is also increasing (GBD 2013 Lancet 2015; 385: 117-71, Torre LA et al. CA Cancer J Clin. 2015; 65: 87-108). In men, lung cancer is the leading cause of death with an estimated number of cases of about 1.1 million per year, followed by liver and stomach cancers. In women, breast cancer is the leading cause of death with about 500,000 cases annually, followed by lung and colon cancers. Approximately 14 million new cancer cases will occur worldwide in 2012, following approximately 1.2 million cases of lung cancer in men and 1.7 million cases of breast cancer in women. Estimated (Torre LA et al. CA Cancer J Clin. 2015; 65: 87-108). In short, cancer constitutes a heavy burden in societies around the world, so treatment options are needed to combat the disease.

Common strategies for treating cancer involve surgery followed by chemotherapy, radiation therapy or, more recently, targeted therapy or a combination thereof (eg, NCCN guidelines for oncology). Chemotherapeutic agents include nucleotide analogs such as 5-flurouracil (5-FU), DNA damaging agents such as oxaliplatin and topoisomerase inhibitors such as irinotecan or microtubule inhibitors such as docetaxel. All of these result in inhibition of tumor cell growth. Targeted therapies include, for example, small molecule compounds that selectively inhibit mutant carcinogenic kinases, such as the BRAF V600E selective compound vemurafenib (Garbe C. et al Recent Results Cancer Res. 2014; 201: 215-25) ( Crizotinib and Ceritinib (Pall G. Curr Opin), anaplastic lymphoma linase (ALK) inhibitors approved for the treatment of melanoma) or lung cancer Oncol.2015; 27: 118-24). In addition, antibody-based compounds such as cetuximab and panitumumab (recognizing and inactivating epidermal growth factor receptor (EGFR), approved for the treatment of KRAS wild-type colon cancer) (Tol J, Punt CJ.Clin Ther.2010; 32: 437-53), or trastuzumab (recognizing Her2, approved for breast cancer therapy (Ahmed S. et al. Breast Cancer. 2015; 22: 101-16) exist.

Despite the development of a wide range of therapeutic interventions, there is no cure available for this heterogeneous disease, except in sporadic cases. Heterogeneity of individual tumors leads to the fact that only a limited number of patients respond to a particular therapy and develop resistance to therapeutic agents during tumor progression (Jamal-Hanjani M. et al Clin). Cancer Res. 2015; 21: 1258-1266). Finally, side effects of the drug can also lead to discontinuation or discontinuation of treatment, but newly developed targeted therapies appear to be better tolerated (http://www.cancer.net/). navigating-cancer-care / how-cancer-treated / chemotherapy / side-effects-chemotherapy; Sun GC et al. Anticancer Agents Med Chem.2015 Mar 17. [Electronic publication before publication]).

For these reasons, there is still a high medical need for the development of new drugs.

Overexpression of CDH3 in tumor cells recognizes bispecific antibodies, ie CDH3 overexpressing tumor cells, and redirection of cytotoxic T cells recognized by the CD3 binding portion of this antibody to the tumor cells. It provides the basis for new methods of treating cancer by using killing antibodies.

Expression analysis performed either at the RNA level or at the protein level, primarily using immunohistochemistry, demonstrates that CDH3 is expressed at highly and well-detectable levels in a wide variety of different cancer types. These cancer types include colon cancer (Milic et al., Kita et al, Imai et al., Http://www.proteinatlas.org/), non-small cell cancer and small cell lung cancer. Including 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 on the tongue (Dasgupta et al; unpublished data from the inventors), thyroid cancer (Jarzab et al, Rocha et al., Human protein atlas), neck Cancer (Imai et al., Han et al.), Preferably stage II and ovarian cancer (Patel et al human proeion atlas), gastric cancer (Kim et al; Imai et al), endometrial cancer (Sugiyama Y et al.). Clin Cancer Res. 2003; 9: 5589-600; human protein atlas), bile duct cancer (Baek S et al. Anat Cell Biol. 2010; 43: 110-7; Imai et al), bladder cancer (Imai et al) , Inventors'unpublished data), Prostatic cancer (Imai et al; Inventors' unpublished data), Testis cancer (Imai et al), Soft tissue sarcoma (Imai et al), Esophageal cancer (Invention) (Unpublished data of the authors) and kidney cancer (unpublished data of the inventors).

Future analysis may demonstrate increased expression in additional tumors or their subtypes, but these tumors may also be associated with such therapies. Fotouhi et al. Demonstrated reduced methylation of the CDH3 promoter in small intestinal neuroendocrine tumors, for example, which may result in increased CDH3 protein expression (Fotouhi O et al. Epigenetics. 2014; 9: 987- 97). A positive correlation was observed between increased expression of CDH3 and increased expression of miR-205, a microRNA that appears to be a specific marker for lung squamous cell carcinoma (Huang W et al. Chin Med J). (Engl). 2014; 127: 272-8).

In contrast to increased 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 good target for the proposed bispecific antibody approach.

Various antibodies that bind to P-cadherin have been described and most of them well characterized are published in the following patents: WO9919477, WO02 / 097395, WO04 / 110345, WO06 / 114704, WO07 / 102525, WO2010 / 001585, WO2010126137; WO2010054007, WO2011056997, WO2011080796, WO2011071541.

Some of the antibodies known in the art block the function of CDH3, for example by interfering with its adhesive properties, such as the anti-CDH3 antibody PF-03732010 (Zhang CC et al. Clin Cancer Res.2010; 16: 5177-88). In vitro, this antibody resulted in destruction of 3D spheroids with intracellular signaling changes such as dissociation of β-catenin at nanomolar concentrations, but did not inhibit proliferation. In vivo inhibition of long-term survival in addition to tumor growth and metastasis formation was observed in a preclinical model at a dose of 10 mg / kg in a CDH3 expression-dependent manner. Mechanically, it cannot be ruled out that the antitumor effect in vivo is further mediated by the process of antibody-dependent cellular cytotoxicity (ADCC), which is due to the binding of cytotoxic T cells to the Fc domain of immunoglobulins. Accompanied by cell death. Such a mechanism of action was due to some of the claimed anti-CDH3 antibodies. Blocking of P-cadherin activity has been described as a technique for inhibiting the growth of P-cadherin-expressing tumors, but the results achieved often remain unsatisfactory because they are tumor growth. This is because it delays the tumor, or at best induces tumor quiescence, but does not eliminate the tumor.

To improve their inhibitory effect on tumor growth, antibodies are complexed with cytotoxic or cell proliferation inhibitors (eg, chemotherapeutic agents, toxins, radioisotopes, etc.) to form antibody-drug conjugates (eg, chemotherapeutic agents, toxins, radioisotopes, etc.). ADC), or in a broader sense, can result in immune complexes. Immune complexes allow targeted delivery of drug moieties to tumors and their intracellular accumulation. With respect to tumor cell death, the potency of such immune complexes varies, such as the mechanism of action of each immune complex, in addition to the internalizing behavior of target molecules normally located on the cell membrane with extracellular domains. Strongly dependent on parameters. In general, coupling to immune complexes has been shown to significantly enhance the antitumor efficacy of various antibodies.

More recently, bispecific molecules that bind to T cells in addition to the target molecule have shown promising results, avoiding many of the drawbacks mentioned above. These drug molecules take advantage of the highly effective natural killing effect of T cells and are therefore independent of completely blocking the function of the target or depending on the proliferative state of the target cell. An example of such a bispecific molecule is a bispecific antitarget x anti-CD3 single-chain antibody construct, which has so far been able to mediate T cell-related killing of target cells with very high efficiency. It is shown in. See, for example, blinatumomab.

A number of different criteria must be met in order to determine that the bispecific scFv molecule is the appropriate drug substance that will ultimately serve the treatment of the patient. The combination of these criteria and stringency make the development of molecules that meet these needs extremely rare and unpredictable. The most obvious criterion is the efficiency of drug candidates.

The efficiency of such molecules depends on multiple parameters with respect to the target binding domain of the bispecific molecule, as well as with respect to the T cell recruitment moiety. In particular, the target-binding moieties of bispecific scFv molecules are unpredictable, because they are very dependent on the properties of the molecule, including its tertiary and quaternary structure. Because it fluctuates. Among the parameters that determine efficiency, the exact binding region, the so-called binding epitope, plays a major role in addition to the binding rate between the target molecule and the bispecific binding molecule.

Anti-CDH3 monoclonal antibodies function by recognizing their target molecules very specifically, as is generally the case with any other monoclonal antibody. This antibody recognizes only a single site or epitope on their target CDH3 molecule. In addition, many antibodies have been found to exert their function in a species-specific manner. However, this species specificity is unique not only to CDH3 monoclonal antibodies (and fragments thereof), but also to common monoclonal antibodies, which are critical in developing them as therapeutic agents for treating human diseases. Obstacle. To obtain marketing approval, any new candidate drug must pass rigorous testing. The study is subdivided into preclinical and clinical phases, the latter being performed in human patients, while the former being performed in animals. The purpose of preclinical studies is to prove that a drug candidate has the desired activity and, most importantly, that the candidate is safe. This drug candidate is most likely to be approved for human clinical trials by each regulatory agency only if the safety and potential efficacy of the drug candidate in animals has been established in preclinical studies. Drug candidates are prepared in three ways for animal safety: (i) related species, i.e. species in which the drug candidate can recognize the ortholog antigen, and (ii) in transgenic animals containing the human antigen. , And (iii) can be tested by using a substitute drug candidate that can bind to the ortholog antigen present in the animal. The limitation of transgenic 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 applied to humans. The limitation of drug candidate substitutes is that they differ in substance composition compared to actual drug candidates, and in many cases the animal used is a rodent with restrictions as described above. Therefore, preclinical data generated in rodents is a prediction for drug candidates.

A preferred method of safety testing is the use of chimpanzees due to related species, preferably primates and genetic similarities. However, chimpanzees are considered endangered and the use of such animals for drug safety testing due to their human-like nature is banned in Europe and other It is also severely restricted in the area.

Although the T cell-induced bispecific single-chain antibodies described in the art have great therapeutic potential for the treatment of malignant diseases, most of these bispecific molecules are species-specific. It is limited in that it can recognize human antigens and can only recognize primates, that is, macaque-compatible species. In addition, the majority of bispecific molecules fail to perform their desired function across species boundaries, resulting in recognition of human and primate homologues, eg, T cell-mediated cytotoxicity. It is further limited in that it may not be able to be demonstrated.

Detailed Description of the Invention The present application has one binding domain for CDH3, as it is still necessary to have additional options available for the treatment of the various cancer types disclosed herein. And in the form of bispecific antibody constructs having a second binding domain for CD3 on T cells, means and methods for solving this problem are provided.

In the first aspect, the invention preferably comprises a first human binding domain that binds to an epitope cluster of human CDH3 on the surface of a target cell and preferably a second that binds to human CD3 on the surface of a T cell. Providing a bispecific antibody construct containing a human binding domain, an epitope cluster of human CDH3 is contained within amino acids 291 to 363 (SEQ ID NO: 36) of human CDH3.

In a preferred embodiment of the invention, the bispecific antibody construct is characterized in that the first binding domain also binds to macaque CDH3, preferably cynomolgus monkey (Macaca fascicularis) CDH3.

This cross-species functionality means that the same molecule can be used in preclinical animal studies as well as in human clinical trials. This provides well-comparable results compared to species-specific substitute molecules and greatly increases the predictive power of animal studies. The CD3 and CDH3 binding domains of the CDH3 x CD3 bispecific antibody constructs of the present invention are both interspecific and functional, i.e., for T cell-mediated cytotoxicity in response to human and macaque antigens. As it exerts comparable effects, it is for preclinical evaluation of the safety, activity and / or pharmacokinetic profile of these binding domains in primates, and in the same form, both as a drug in humans. Can be used. In a preferred embodiment, it will be understood that the interspecificity of the first and second binding domains in the antibody construct of the present invention is identical.

Considering the above, there is no need to construct a substituted CDH3 × CD3 bispecific antibody construct for testing in species that are phylogenetically distant (from humans). As a result, the same molecule can be used in preclinical animal studies as well as intended to be administered to humans after commercial approval and in therapeutic drug administration in addition to clinical trials. The ability to use the same molecules used in subsequent administrations to humans in preclinical animal studies substantially risks limiting the applicability of the data obtained in preclinical animal studies to human cases. Eliminate, or at least significantly reduce. In short, obtaining preclinical safety data in animals using the same molecules that will actually be administered to humans contributes to ensuring the applicability of the data to human-related scenarios. In contrast, in conventional approaches using substitute molecules, the substitute molecule must be molecularly adapted to the animal testing system used for preclinical safety assessment. Therefore, the molecules actually used in human therapy may be different in sequence and structure from the substitute molecules used in preclinical studies in terms of pharmacokinetic parameters and / or biological activity, resulting in The applicability / migration of data obtained in preclinical animal studies to human cases is limited. The use of substitute molecules requires the construction, production, purification and characterization of completely new constructs. This requires extra development costs and time to obtain the molecule. That is, the substitute must be developed separately in addition to the actual drug used in human therapy, and as a result two strains of development must be carried out for the two molecules. Therefore, the main advantage of the preferably human CDH3 × CD3 bispecific antibody constructs of the invention that exhibit interspecific specificity and functionality (ie, reactivity) as described herein is that the same molecule It can be used as a therapeutic agent in humans and in preclinical animal studies.

It is no longer necessary to adapt test animals to drug candidates intended for administration to humans, such as the production of transgenic animals, by means of the interspecific CDH3 × CD3 bispecific antibody constructs of the present invention. By the use and method of the present invention, the preferably human CDH3 × CD3 bispecific antibody construct of the present invention, which exhibits interspecificity and reactivity, is a non-chimpanzee such as macaque without any genetic manipulation of animals. Can be used directly for preclinical trials in primates. As is well known to those skilled in the art, method n for adapting test animals to drug candidates risks that the results obtained in preclinical safety studies are not very typical and unpredictable for humans due to animal modification. Always get involved. For example, in transgenic animals, the protein encoded by the transgene is often highly overexpressed. Therefore, the data obtained on the biological activity of antibodies against this protein antigen may be much lower in protein and may limit their predictions for humans expressed at higher physiological levels.

A further advantage of the preferably human CDH3 × CD3 bispecific antibody constructs of the present invention is the ability to extract multiple blood samples, for example, when used as part of preclinical studies in the course of pharmacokinetic animal studies. Is. Multiple blood extracts can be obtained much more easily in non-chimpanzee primates than in lower animals such as mice. Extraction of multiple blood samples allows continuous testing of blood parameters to determine the biological effects induced by the CDH3 × CD3 bispecific antibody constructs of the present invention. In addition, extraction of multiple blood samples allows researchers to assess the pharmacokinetic profile of preferably human CDH3 × CD3 bispecific antibody constructs of the invention as defined herein. In addition, potential side effects may be induced by the CDH3 × CD3 bispecific antibody constructs of the invention reflected in blood parameters and measured in different blood samples extracted during the antibody administration process. obtain.

The advantages of the CDH3 × CD3 bispecific antibody constructs of the invention as defined herein, which exhibit interspecific specificity, may be summarized briefly as follows.

First, the CDH3 × CD3 bispecific antibody constructs of the invention, as defined herein, used in preclinical studies are the same as those used in human therapy. Therefore, it is no longer necessary to develop two independent molecules that may differ in pharmacokinetic properties and biological activity. This is extremely advantageous in that, for example, the pharmacokinetic results are more directly transferable and applicable to the human setting than, for example, conventional substitute methods.

Second, the use of CDH3 × CD3 bispecific antibody constructs of the invention as defined herein for preparing therapeutic agents for humans is less expensive than alternative methods and is less expensive. And it is not labor-intensive.

Third, the CDH3 × CD3 bispecific antibody constructs of the invention as defined herein are not only preclinical studies in one primate species, but also preclinical studies in a series of different primate species. It can also be used to limit the risk of species differences between primates and humans.

Fourth, chimpanzees are an endangered species, so animal testing can be avoided if desired.

Fifth, multiple blood samples can be extracted for extensive pharmacokinetic testing.

Sixth, because the antibody construct according to the preferred embodiment of the invention is of human origin, the development of an immune response to the binding molecule is minimized when administered to a human patient. Induction of immune response by antibodies specific for drug candidates derived from non-human species, such as mice, which result in the development of human anti-mouse antibodies (HAMA) against therapeutic molecules of mouse origin is eliminated. ..

Therapeutic use of the CDH3 × CD3 bispecific antibody constructs of the present invention is a therapeutic method of novel invention for cancer, preferably solid tumors, more preferably carcinomas and other cancer indications as listed below. I will provide a. As shown in the Examples below, the CDH3 × CD3 bispecific antibody constructs of the present invention provide an advantageous tool for killing CDH3-expressing human cancer cells. Moreover, the cytotoxic activity of the CDH3 × CD3 bispecific antibody constructs of the present invention provides a favorable mechanism of action as compared to classical (unispecific) IgG molecules or ADCs.

Cross-reactivity to related animal species is essential for characterization of drug candidates in animal models, and interspecific affinity gaps are maintained at less than 10-fold affinity differences to ensure predictive animal model setup. Should be. Macaque monkeys (and especially cynomolgus monkeys) are considered to be among the most relevant species with the highest predictors of efficacy and toxicity testing. Each cytotoxicity assay must be developed and adapted to analyze whether the antibody construct meets efficiency criteria.

To prevent harmful side effects, the target binding moiety of the bispecific antibody construct specifically binds to P-cadherin and does not bind to its closest homolog or any other protein present in the body. Must be guaranteed.

The antibody constructs of the present invention do not preferably bind to CHD3 extracellular domain D1 (positions 108-215 of SEQ ID NO: 1), and they preferably bind to CHD3 extracellular domain D4 (positions 441-546 of SEQ ID NO: 1). Do not bind, and they do not preferably bind to CHD3 extracellular domain D5 (SEQ ID NO: 1 at positions 547-650).

According to the most preferred embodiment of the invention, the bispecific antibody construct binds the first binding domain to an epitope contained within amino acids 291 to 327 (SEQ ID NO: 34) of human CDH3. Features. Antibodies that specifically bind to amino acids positions 291 to 327 (SEQ ID NO: 34) do not preferably bind to the CHD3 extracellular domain D3 of human CDH3 (positions 328 to 440 of SEQ ID NO: 1).

Preferred antibody constructs according to the invention also include a preferably first human binding domain that binds to an epitope of human CDH3 on the surface of target cells and a preferably second human binding domain that binds to human CD3 on the surface of T cells. Can be defined as a bispecific antibody construct, the antibody construct is an antibody referred to as CDH3-11, CDH3-12, CDH3-13 or CDH3-14, ie, the VH region and sequence set forth in SEQ ID NO: 155. Number: VL region shown in 156, · VH region shown in SEQ ID NO: 165 and VL region shown in SEQ ID NO: 166, · VH region shown in SEQ ID NO: 175 and VL region shown in SEQ ID NO: 176 Or Binds to the same epitope as the antibody containing the VH region set forth in SEQ ID NO: 185 and the VL region set forth in SEQ ID NO: 186, or competes for binding to CDH3.

In another aspect of the invention, the bispecific antibody construct is characterized in that the first binding domain binds to an epitope contained within amino acids 328-363 (SEQ ID NO: 35) of human CDH3. ..

Preferred antibody constructs according to the invention also comprise a preferably first human binding domain that binds to an epitope of human CDH3 on the surface of target cells and a preferably second human binding domain that binds to human CD3 on the surface of T cells. Can be defined as a bispecific antibody construct, the antibody construct is the same as an antibody referred to as CDH3-24, that is, an antibody containing the VH region set forth in SEQ ID NO: 285 and the VL region set forth in SEQ ID NO: 286. It binds to an epitope or competes with it for binding to CDH3.

Another preferred antibody construct according to the invention is also a preferably first human binding domain that binds to an epitope of human CDH3 on the surface of target cells and a preferably second human binding domain that binds to human CD3 on the surface of T cells. Can be defined as a bispecific antibody construct comprising, the antibody construct is an antibody referred to as CDH3-25, CDH3-26 or CDH3-27, ie, the VH region shown in SEQ ID NO: 295 and SEQ ID NO:: The VL region shown in 296, the VH region shown in SEQ ID NO: 305 and the VL region shown in SEQ ID NO: 306, or the VH region shown in SEQ ID NO: 315 and the VL region shown in SEQ ID NO: 316. It binds to or competes with the same epitope as the containing antibody for binding to CDH3.

Whether an antibody construct competes with another given antibody construct for binding can be measured in a competing assay, such as a competing ELISA or cell-based competing assay. Avidin-bound microparticles (beads) can also be used. Similar to an avidin-coated ELISA plate, each of these beads can be used as a substrate on which the assay can be performed when reacting with a biotinylated protein. The antigen is coated on the beads and then precoated with the first antibody. A second antibody is added to determine any further binding. The measurements are generated by flow cytometry.

In another embodiment of the invention, the bispecific antibody construct is an epitope whose first binding domain is contained within amino acids 328-363 (SEQ ID NO: 35) of human CDH3 and amino acid 404 of human CDH3. It is characterized in that it binds to an epitope contained in position ~ 440 (SEQ ID NO: 390).

Furthermore, the antibody constructs of the present invention do not preferably bind to epitopes contained within amino acids 216-252 or 253-290 of human CDH3 set forth in SEQ ID NO: 1.

Furthermore, the antibody construct of the present invention does not preferably bind to an epitope contained within amino acids 364 to 403 of human CDH3 shown in SEQ ID NO: 1.

Antibody constructs that specifically bind to the epitope contained within amino acids positions 328-363 (SEQ ID NO: 35) do not preferably bind to the CHD3 extracellular domain D2 of human CDH3 (positions 216-327 of SEQ ID NO: 1). ..

One advantage of the present invention is the provision of bispecific antibody constructs containing binding domains that bind to CD3 and binding domains that can bind to CDH3, whereas both binding domains are human and macaque. CDH3 shows interspecific specificity. Unexpectedly, the CDH3 × CD3 bispecific antibody constructs of the present invention not only specifically bind to the human and macaque CDH3 homologues of CDH3 and CD3, but also T cell-mediated cells in the human and macaque CDH3 assay system. It was also found to be injured. Advantageously, the present invention provides CDH3 × CD3 bispecific antibody constructs that exhibit T cell-mediated cytotoxicity in humans and macaques. This advantage is achieved by a CDH3 × CD3 bispecific antibody construct that binds to an epitope cluster contained within amino acids 291 to 363 (SEQ ID NO: 36) of human CDH3.<u style="single">[Invention 1001]</u><u style="single"> A bispecific 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 a second binding domain that binds to human CD3 on the surface of a T cell. , The bispecific antibody construct, wherein the epitope cluster of human CDH3 is contained within amino acids 291 to 363 (SEQ ID NO: 36) of human CDH3.</u><u style="single">[Invention 1002]</u><u style="single"> The antibody construct of the present invention 1001 in which the first binding domain binds to an epitope contained in amino acids 291 to 327 (SEQ ID NO: 34) of human CDH3.</u><u style="single">[Invention 1003]</u><u style="single"> The antibody construct of the present invention 1001 in which the first binding domain binds to an epitope contained within amino acids 328 to 363 (SEQ ID NO: 35) of human CDH3.</u><u style="single">[Invention 1004]</u><u style="single"> The antibody construct of the present invention 1003, wherein the first binding domain also binds to an epitope contained in amino acids 404 to 440 (SEQ ID NO: 390) of human CDH3.</u><u style="single">[Invention 1005]</u><u style="single"> An antibody construct of any of the preceding inventions, wherein the first binding domain also binds to macaque CDH3, preferably cynomolgus monkey (Macaca fascicularis) CDH3.</u><u style="single">[Invention 1006]</u><u style="single"> The first binding domain is</u><u style="single">a) SEQ ID NO: 149, CDR-H1, SEQ ID NO: 150, CDR-H2, SEQ ID NO: 151, CDR-H3, SEQ ID NO: 152, CDR-L1, SEQ ID NO: 153 CDR-L2 shown in and SEQ ID NO: 154: CDR-L3;</u><u style="single">b) CDR-H1 shown in SEQ ID NO: 159, CDR-H2 shown in SEQ ID NO: 160, CDR-H3 shown in SEQ ID NO: 161; CDR-L1 shown in SEQ ID NO: 162, SEQ ID NO: 163 CDR-L2 shown in and SEQ ID NO: 164 shown in CDR-L3;</u><u style="single">c) CDR-H1 shown in SEQ ID NO: 169, CDR-H2 shown in SEQ ID NO: 170, CDR-H3 shown in SEQ ID NO: 171; CDR-L1 shown in SEQ ID NO: 172, SEQ ID NO: 173. CDR-L2 shown in and SEQ ID NO: 174: CDR-L3;</u><u style="single">d) CDR-H1 shown in SEQ ID NO: 179, CDR-H2 shown in SEQ ID NO: 180, CDR-H3 shown in SEQ ID NO: 181 and CDR-L1 shown in SEQ ID NO: 182, SEQ ID NO: 183. CDR-L2 shown in and SEQ ID NO: 184;</u><u style="single">e) CDR-H1 shown in SEQ ID NO: 189, CDR-H2 shown in SEQ ID NO: 190, CDR-H3 shown in SEQ ID NO: 191 and CDR-L1 shown in SEQ ID NO: 192, SEQ ID NO: 193. CDR-L2 shown in and SEQ ID NO: 194: CDR-L3;</u><u style="single">f) SEQ ID NO: 199, CDR-H1, SEQ ID NO: 200, CDR-H2, SEQ ID NO: 201, CDR-H3, SEQ ID NO: 202, CDR-L1, SEQ ID NO: 203 CDR-L2 shown in and SEQ ID NO: 204;</u><u style="single">g) SEQ ID NO: 209, CDR-H1, SEQ ID NO: 210, CDR-H2, SEQ ID NO: 211, CDR-H3, SEQ ID NO: 212, CDR-L1, SEQ ID NO: 213 CDR-L2 shown in and SEQ ID NO: 214;</u><u style="single">h) CDR-H1 shown in SEQ ID NO: 219, CDR-H2 shown in SEQ ID NO: 220, CDR-H3 shown in SEQ ID NO: 221; CDR-L1 shown in SEQ ID NO: 222, SEQ ID NO: 223. CDR-L2 shown in and SEQ ID NO: 224 shown in CDR-L3;</u><u style="single">i) SEQ ID NO: 229, CDR-H1, SEQ ID NO: 230, CDR-H2, SEQ ID NO: 231, CDR-H3, SEQ ID NO: 232, CDR-L1, SEQ ID NO: 233 CDR-L2 shown in and SEQ ID NO: 234; and CDR-L3;</u><u style="single">j) SEQ ID NO: 239, CDR-H1, SEQ ID NO: 240, CDR-H2, SEQ ID NO: 241, CDR-H3, SEQ ID NO: 242, CDR-L1, SEQ ID NO: 243 CDR-L2 shown in and SEQ ID NO: 244 shown in CDR-L3</u><u style="single">Includes the VH region containing CDR-H1, CDR-H2 and CDR-H3 and the VL region containing CDR-L1, CDR-L2 and CDR-L3 selected from the group consisting of:</u><u style="single">The antibody construct of any of 1001, 1002 and 1005 of the present invention.</u><u style="single">[Invention 1007]</u><u style="single"> The first binding domain is</u><u style="single">a) CDR-H1 shown in SEQ ID NO: 279, CDR-H2 shown in SEQ ID NO: 280, CDR-H3 shown in SEQ ID NO: 281; CDR-L1 shown in SEQ ID NO: 282, SEQ ID NO: 283 CDR-L2 shown in and SEQ ID NO: 284;</u><u style="single">b) SEQ ID NO: 289, CDR-H1, SEQ ID NO: 290, CDR-H2, SEQ ID NO: 291, CDR-H3, SEQ ID NO: 292, CDR-L1, SEQ ID NO: 293 CDR-L2 shown in and SEQ ID NO: 294;</u><u style="single">c) CDR-H1 shown in SEQ ID NO: 299, CDR-H2 shown in SEQ ID NO: 300, CDR-H3 shown in SEQ ID NO: 301, CDR-L1 shown in SEQ ID NO: 302, SEQ ID NO: 303 CDR-L2 shown in and SEQ ID NO: 304;</u><u style="single">d) SEQ ID NO: 309, CDR-H1, SEQ ID NO: 310, CDR-H2, SEQ ID NO: 311, CDR-H3, SEQ ID NO: 312, CDR-L1, SEQ ID NO: 313 CDR-L2 shown in and SEQ ID NO: 314 in CDR-L3;</u><u style="single">e) CDR-H1 shown in SEQ ID NO: 319, CDR-H2 shown in SEQ ID NO: 320, CDR-H3 shown in SEQ ID NO: 321, CDR-L1 shown in SEQ ID NO: 322, SEQ ID NO: 323. CDR-L2 shown in and SEQ ID NO: CDR-L3 shown in 324;</u><u style="single">f) SEQ ID NO: 329, CDR-H1, SEQ ID NO: 330, CDR-H2, SEQ ID NO: 331, CDR-H3, SEQ ID NO: 332, CDR-L1, SEQ ID NO: 333 CDR-L2 shown in and SEQ ID NO: 334 shown in CDR-L3;</u><u style="single">g) CDR-H1 shown in SEQ ID NO: 339, CDR-H2 shown in SEQ ID NO: 340, CDR-H3 shown in SEQ ID NO: 341, CDR-L1 shown in SEQ ID NO: 342, SEQ ID NO: 343. CDR-L2 shown in and SEQ ID NO: 344; and CDR-L3;</u><u style="single">h) SEQ ID NO: CDR-H1, SEQ ID NO: 349, CDR-H2, SEQ ID NO: 351, CDR-H3, SEQ ID NO: 352, CDR-L1, SEQ ID NO: 353 CDR-L2 shown in and SEQ ID NO: 354 shown in CDR-L3</u><u style="single">Includes the VH region containing CDR-H1, CDR-H2 and CDR-H3 and the VL region containing CDR-L1, CDR-L2 and CDR-L3 selected from the group consisting of:</u><u style="single">The antibody construct of any of 1001, 1003, 1004 and 1005 of the present invention.</u><u style="single">[Invention 1008]</u><u style="single"> The first binding domain is 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 the antibody construct of the invention 1006 comprising a VH region selected from the group consisting of the VH region set forth in SEQ ID NO: 245.</u><u style="single">[Invention 1009]</u><u style="single"> The first binding domain is shown 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. The antibody construct of the present invention 1007 comprising a VH region selected from the group consisting of VH regions.</u><u style="single">[Invention 1010]</u><u style="single"> The first binding domain is 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 the antibody construct of the invention 1006 or 1008 comprising a VL region selected from the group consisting of the VL region set forth in SEQ ID NO: 246.</u><u style="single">[Invention 1011]</u><u style="single"> The first binding domain is shown 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. The antibody construct of the present invention 1007 or 1009 comprising a VL region selected from the group consisting of VL regions.</u><u style="single">[Invention 1012]</u><u style="single"> The first binding domain is 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: The VH region and VL region selected from the group consisting of the pair of VH region and VL region shown in No .: 215 + 216, SEQ ID NO: 225 + 226, SEQ ID NO: 235 + 236, and SEQ ID NO: 245 + 246. The antibody construct of any of the invention 1006, 1008 or 1010, including.</u><u style="single">[Invention 1013]</u><u style="single"> The first binding domain is 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: An antibody of any of 1007, 1009 or 1011 of the present invention comprising a VH region and a VL region selected from the group consisting of a pair of VH region and VL region set forth in No .: 345 + 346 and SEQ ID NO: 355 + 356. Construct.</u><u style="single">[Invention 1014]</u><u style="single"> (scFv)</u>_{<u style="single">2</u>}<u style="single">, ScFv-An antibody construct of any of the preceding inventions, which is a form selected from the group consisting of single domain mAbs, diabodies and oligomers of the form described above.</u><u style="single">[Invention 1015]</u><u style="single"> The first binding domain is 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:: The antibody construct of any of the invention 1006, 1008, 1010, 1012 or 1014, comprising an amino acid sequence selected from the group consisting of 237 and the sequence set forth in SEQ ID NO: 247.</u><u style="single">[Invention 1016]</u><u style="single"> The first binding domain is shown 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. The antibody construct of any of the invention 1007, 1009, 1011, 1013 or 1014, comprising an amino acid sequence selected from the group consisting of sequences.</u><u style="single">[Invention 1017]</u><u style="single"> An antibody construct of any of the preceding inventions, wherein the second binding domain binds to human and CD3 epsilon of common marmosets (Callithrix jacchus), Wataboushi tamarins (Saguinus Oedipus) or common squirrel monkeys (Saimiri sciureus).</u><u style="single">[Invention 1018]</u><u style="single"> 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 The antibody construct of any of the present invention 1006, 1008, 1010, 1012, 1014, 1015 or 1017, comprising an amino acid sequence selected from the group consisting of the sequences shown in.</u><u style="single">[Invention 1019]</u><u style="single"> Selected from the group consisting of the sequences shown 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. The antibody construct of any of the present inventions 1007, 1009, 1011, 1013, 1014, 1016 or 1017, comprising the amino acid sequence.</u><u style="single">[Invention 1020]</u><u style="single"> 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, Contains an amino acid sequence selected from the group consisting of the sequences shown in 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. The antibody construct of any of 1006, 1008, 1010, 1012, 1014, 1015, 1017 or 1018 of the present invention.</u><u style="single">[Invention 1021]</u><u style="single"> A polynucleotide encoding an antibody construct as defined in any of the preceding inventions.</u><u style="single">[Invention 1022]</u><u style="single"> A vector containing the polynucleotide defined in 1021 of the present invention.</u><u style="single">[Invention 1023]</u><u style="single"> Transformed with the polynucleotide defined in the invention 1021 or the vector defined in the invention 1022 or transfected with the polynucleotide defined in the invention 1021 or the vector defined in the invention 1022. Host cell.</u><u style="single">[1024 of the present invention]</u><u style="single"> A method for producing an antibody construct according to any one of the present inventions 1001 to 1020, which is a host defined in the present invention 1023 under conditions that allow expression of the antibody construct defined in any one of the present inventions 1001 to 1020. The method described above comprising culturing cells and recovering the antibody construct produced from the culture.</u><u style="single">[Invention 1025]</u><u style="single"> A pharmaceutical composition comprising an antibody construct according to any one of 1001 to 1020 of the present invention or an antibody construct produced by the method of 1024 of the present invention.</u><u style="single">[Invention 1026]</u><u style="single"> An antibody construct according to any one of 1001 to 1020 of the present invention or an antibody construct produced by the method of 1024 of the present invention for use in the prevention, treatment or amelioration of a tumor or cancer.</u><u style="single">[Invention 1027]</u><u style="single"> The cancers include lung cancer, head and neck cancer, primary or secondary CNS tumor, primary or secondary brain tumor, primary CNS lymphoma, spinal axis tumor, brain stem glioma, pituitary adenoma, and adrenal cortex. , Esophageal cancer, colon cancer, breast cancer, ovarian cancer, NSCLC (non-small cell lung cancer), SCLC (small cell lung cancer), endometrial cancer, cervical cancer, uterine cancer, transition epithelial cancer, bone Cancer, pancreatic cancer, skin cancer, skin or intraocular melanoma, liver cancer, bile duct cancer, bile sac cancer, kidney cancer, rectal cancer, anal cancer, gastric cancer, gastrointestinal (stomach,) Colon-rectal and duodenal) cancer, small intestine cancer, biliary tract cancer, urinary tract cancer, renal cell carcinoma, endometrial carcinoma, thyroid cancer, testis cancer, cutaneous squamous cell carcinoma, melanoma , Gastric cancer, prostate cancer, bladder cancer, osteosarcoma, mesenteric tumor, Hodgkin's disease, non-hodgkin lymphoma, chronic or acute leukemia, chronic myeloid leukemia, lymphocytic lymphoma, multiple myeloma, fibrosarcoma, neuroblast The antibody construct of the present invention 1026 selected from the group consisting of cell tumors, retinoblastomas, and soft sarcomas.</u><u style="single">[Invention 1028]</u><u style="single"> The antibody construct of the present invention 1027, wherein the cancer is a squamous cell carcinoma.</u><u style="single">[Invention 1029]</u><u style="single"> An antibody construct according to any one of 1001 to 1020 of the present invention or an antibody construct produced by the method of 1024 of the present invention is administered to a subject who is a method of treating, preventing or ameliorating a tumor or cancer and needs it. The method comprising steps.</u><u style="single">[Invention 1030]</u><u style="single"> The cancers include lung cancer, head and neck cancer, primary or secondary CNS tumor, primary or secondary brain tumor, primary CNS lymphoma, spinal axis tumor, brain stem glioma, pituitary adenoma, and adrenal cortex. , Esophageal cancer, colon cancer, breast cancer, ovarian cancer, NSCLC (non-small cell lung cancer), SCLC (small cell lung cancer), endometrial cancer, cervical cancer, uterine cancer, transition epithelial cancer, bone Cancer, pancreatic cancer, skin cancer, skin or intraocular melanoma, liver cancer, bile duct cancer, bile sac cancer, kidney cancer, rectal cancer, anal cancer, gastric cancer, gastrointestinal (stomach, stomach, Colon-rectal and duodenal) cancer, small intestine cancer, biliary tract cancer, urinary tract cancer, renal cell carcinoma, endometrial carcinoma, thyroid cancer, testis cancer, cutaneous squamous cell carcinoma, melanoma , Gastric cancer, prostate cancer, bladder cancer, osteosarcoma, mesenteric tumor, Hodgkin's disease, non-hodgkin lymphoma, chronic or acute leukemia, chronic myeloid leukemia, lymphocytic lymphoma, multiple myeloma, fibrosarcoma, neuroblast The method of the present invention 1029 selected from the group consisting of cell tumors, retinoblastomas, and soft sarcoma.</u><u style="single">[Invention 1031]</u><u style="single"> The method of the present invention 1030, wherein the cancer is a squamous cell carcinoma.</u><u style="single">[Invention 1032]</u><u style="single"> The antibody construct of any of 1001 to 1020 of the present invention, the antibody construct produced by the method of 1024 of the present invention, the vector defined in 1022 of the present invention, and / or the host cell defined in 1023 of the present invention.</u><u style="single">Including, kit.</u>

As used herein, the singular forms "one (a)", "one (an)", and "that (the)" may be plural unless otherwise explicitly stated in their content. It should be noted that the referent is included. Thus, for example, a reference to a "reagent" comprises one or more of these different reagents, and a reference to a "method" may be modified to, or with, the method described herein. Includes references to equivalent steps and methods known to those of skill in the art that can be replaced.

Unless otherwise indicated, the term "at least" preceding a set of elements should be understood to refer to the entire set of elements. One of ordinary skill in the art will be able to recognize or confirm many equivalents of the particular embodiments of the invention described herein by simply using routine experiments. Such equivalents are intended to be included by the present invention.

The term "and / or" as used herein means "and", "or" and "all or any other combination of elements connected by this term". include.

The terms "about" or "approximately" as used herein are within ± 20%, preferably within ± 15%, more preferably within ± 10%, most preferably within a given value or range. It means within ± 5%.

Throughout this specification and subsequent claims, the word "comprise", as well as "comprises" and "comprising", etc., unless the context should be interpreted in a different sense. It will be understood that the variant of is suggesting the inclusion of the integers or steps or integers or groups of steps described, but not the exclusion of any other integers or steps or integers or groups of steps. As used herein, the term "comprising" is the term "containing" or "including," or as used herein, sometimes ". Can be replaced by the term "having".

As used herein, "consisting of" excludes any element, step, or component not specified in the claim element. As used herein, "becomes essential" does not exclude materials or steps that do not substantially affect the basic and novel features of the claims.

In each case herein, any of the terms "contains," "consisting essentially of," and "consisting of" may be replaced with either of the other two terms.

The term "antibody construct" refers to a molecule whose structure and / or function is based on, for example, the structure and / or function of a full-length or total immunoglobulin molecule. Thus, the antibody construct can bind to its specific target or antigen. In addition, antibody constructs according to the invention include minimal structural requirements for antibodies that allow target binding. This minimum requirement is, for example, at least three light chain CDRs (ie, CDR1, CDR2 and CDR3 in the VL region) and / or three heavy chain CDRs (ie, CDR1, CDR2 and CDR3 in the VH region), preferably six. It may be defined by the presence of all CDRs. Antibodies on which the constructs according to the invention are based include, for example, monoclonal antibodies, recombinant antibodies, chimeric antibodies, deimmunized antibodies, humanized antibodies and human antibodies.

Within the definition of "antibody construct" according to the present invention are full-length or total antibodies, including camel antibodies and other immunoglobulin antibodies produced by biotechnology or protein engineering methods or processes. These full-length antibodies may be, for example, monoclonal antibodies, recombinant antibodies, chimeric antibodies, deimmunized antibodies, humanized antibodies and human antibodies. Also, within the definition of "antibody construct" are fragments of full-length antibodies such as VH, VHH, VL, (s) dAb, Fv, Fd, Fab, Fab', F (ab') 2 or "rIgG" ( "Half antibody") and so on. The antibody constructs according to the invention are also referred to as antibody variants, such as modified antibody fragments such as scFv, di-scFv or bi (s) -scFv, scFv-Fc, scFv-zipper, scFab, Fab2, Fab3, diabody. , Single chain diabody, tandem diabody (Tandab's), tandem di-scFv, tandem tri-scFv, "minibody" exemplified by the following structure: (VH-VL-CH3)₂, (ScFv-CH3)₂, ((ScFv)₂-CH3 + CH3), ((scFv)₂-CH3) or (scFv-CH3-scFv)₂, Multibody, such as triabodies or tetrabodies, and single domain antibodies, such as nanobodies, or single variable domain antibodies that contain only one variable domain, ie antigens independent of other V regions or domains. Alternatively, it may be an antibody that may be a VHH, VH or VL that specifically binds to an epitope.

In addition, the definition of the term "antigen construct" includes monovalent, bivalent and multivalent / multivalent constructs, and thus monospecific constructs that specifically bind to only one antigenic structure, plus different bindings. Includes bispecific and multispecific / multispecific constructs that specifically bind to more than one antigenic structure, eg, two, three or more, depending on the domain. Furthermore, the definition of the term "antibody construct" includes a molecule consisting of only one polypeptide chain and a molecule consisting of two or more polypeptide chains, that is, the same chain (homotrimer, homo). Includes molecules that can be either trimeric or homo-oligomer) or different (heterodimer, heterotrimer or heterooligomer). Examples of antibodies and variants or derivatives thereof identified above are, among other things, Harlow and Lane, Antibodies a laboratory manual, CSHL Press (1988) and Using Antibodies: a laboratory manual, CSHL Press (1999), Kontermann and Dubel, Antibody. Described in Engineering, Springer, 2nd ed. 2010 and Little, Recombinant Antibodies for Immunotherapy, Cambridge University Press 2009.

The antibody construct of the present invention is preferably an "in vitro generated antibody construct". The term refers to antibody constructs as defined above, where all or part of the variable region (eg, at least one CDR) is non-immune cell selection, eg, in vitro phage display, protein chip or candidate sequence. It is produced by any other method that can be tested for their ability to bind antigens. Therefore, the term preferably excludes sequences produced only by genomic rearrangement in animal immune cells. A "recombinant antibody" is an antibody produced by the use of recombinant DNA technology or genetic engineering.

The term "monoclonal antibody" (mAb) or monoclonal antibody construct, as used herein, refers to an antibody obtained from a substantially homogeneous population of antibodies, i.e., the individual antibodies contained in the population are identical. Excludes possible naturally occurring mutations and / or post-translational modifications (eg, isomerization, amidation) that may be present in small amounts. Monoclonal antibodies are highly specific and are single on the antigen, as opposed to conventional (polyclonal) antibody preparations that typically contain different antibodies directed against different determinants (or epitopes). Directed to the antigenic site or determinant. 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 modifier "monoclonal" indicates the characteristics of the antibody as obtained from a substantially homogeneous population of antibodies and should not be construed as requiring the production of the antibody by any particular method.

For the preparation of monoclonal antibodies, any technique that provides antibodies produced by continuous cell line cultures can be used. For example, the monoclonal antibody used may be made by the hybridoma method first described by Koehler et al., Nature, 256: 495 (1975), or by the recombinant DNA method (eg, US Pat. No. 4,816,567). (See specification). Examples of additional techniques for producing human monoclonal antibodies include trioma technology, human B cell hybridoma technology (Kozbor, Immunology Today 4 (1983), 72) and EBV-hybridoma technology (Cole et al., Monoclonal Antibodies and Cancer Therapy, Alan R. Liss, Inc. (1985), 77-96).

Hybridomas then use standard methods, such as enzyme-linked immunosorbent assay (ELISA) and surface plasmon resonance (BIACORE ) analysis, to produce antibodies that specifically bind to a particular antigen. It can be screened to identify one or more hybridomas. Any form of related antigen may be used as an immunogen, for example, these are recombinant antigens, naturally occurring forms, any variants or fragments thereof, plus their antigenic peptides. Surface plasmon resonance, such as that used in the BIAcore system, can be used to increase the efficiency of phage antibodies that bind to target antigens, such as epitopes such as CDH3 or CD3 epsilon (Schier, Human Antibodies Hybridomas 7 (1996), 97). -105; Malmborg, J.Immunol.Methods 183 (1995), 7-13).

Another exemplary method of making a monoclonal antibody includes screening a protein expression library, such as a phage display or ribosome display library. Phage displays are described, for example, in Ladner et al., US Pat. No. 5,223,409; Smith (1985) Science 228: 1315-1317, Clackson et al., Nature, 352: 624-628 (1991) and Marks et al. , J. Mol. Biol., 222: 581-597 (1991).

In addition to the use of display libraries, related antigens can be used to immunize non-human animals such as rodents (such as 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 strains deficient in mouse antibody production can be manipulated with large fragments of the human Ig (immunoglobulin) locus. Hybridoma techniques may be used to produce and select antigen-specific monoclonal antibodies derived from genes with the desired specificity. See, for example, XENOMOUSE , Green et al. (1994) Nature Genetics 7: 13-21, US2003-0070185, WO96 / 34096, and WO96 / 33735.

Monoclonal antibodies can also be obtained from non-human animals and can be modified using recombinant DNA techniques known in the art, such as humanization, deimmunization, chimerization and the like. Examples of modified antibody constructs are humanized variants of non-human antibodies, "affinity maturation" antibodies (eg, Hawkins et al. J. Mol. Biol. 254, 889-896 (1992) and Lowman et al., Biochemistry 30, 10832-10837 (1991)) and antibody variants with altered effector function (s) (eg, US Pat. No. 5,648,260, Kontermann and Dubel (2010) above, and Little (s) above. See 2009)).

In immunology, affinity maturation is the process by which B cells produce antibodies with increased affinity for antigens during the course of an immune response. Repeated exposure to the same antigen will result in the host producing antibodies with continuously increased affinity. Like the natural prototype, in vitro affinity maturation is based on the principles of maturation and selection. In vitro affinity maturation has been successfully used to optimize antibodies, antibody constructs, and antibody fragments. Random mutations in the CDRs are introduced using radiation, chemical mutagens or error-prone PCR. In addition, genetic diversity can be increased by chain shuffling. Usually, two or three rounds of mutation and selection using a display method such as phage display yields antibody fragments with affinities in the low nanomolar concentration range.

A preferred type of amino acid substitution change in an antibody construct comprises substituting one or more hypervariable region residues of the parent antibody (eg, humanized or human antibody). In general, the resulting variants (s) selected for further development will have improved biological properties compared to the parent antibody from which they were produced. A convenient method for producing such substitution variants involves affinity maturation using a phage display. Briefly, several hypervariable region sites (eg, 6-7 sites) are mutated to produce all possible amino acid substitutions at each site. The antibody variant thus generated is presented in monovalent form from the filamentous phage particles as a fusion with the Gene III product of M13 packaged within each particle. The phage-displayed variants are then screened for their biological activity (eg, binding affinity) as disclosed herein. To identify candidate hypervariable region sites for modification, alanine scanning mutagenesis can be performed to identify hypervariable region residues that significantly contribute to antigen binding. Alternatively or additionally, it may be beneficial to analyze the crystal structure of the antigen-antibody complex to identify the point of contact between the binding domain and, for example, human CDH3. Such contact and flanking residues are candidates for substitution by the techniques detailed herein. Once such variants are generated, a panel of variants is screened as described herein and antibodies with superior properties in one or more related assays may be selected for further development. ..

While the monoclonal antibodies and antibody constructs of the invention are identical or homologous to the corresponding sequences of heavy chains and / or parts of the light chains of antibodies derived from a particular species or antibodies belonging to a particular antibody class or subclass. , A "chimeric" antibody (immunoglobulin), in which the balance of the chain (s) is identical or homologous to the corresponding sequence of an antibody derived from another species or an antibody belonging to another antibody class or subclass, and such. These fragments are specifically included as long as the antibody fragments exhibit the desired biological activity (US Pat. No. 4,816,567; Morrison et al., Proc. Natl. Acad. Sci. USA, 81: 6851-6855 (1984). )). Chimeric antibodies of interest herein include "primated" antibodies that include variable domain antigen binding sequences and human constant region sequences from non-human primates (eg, Old World monkeys, apes, etc.). .. Various techniques for making chimeric antibodies have been described. For example, Morrison et al., Proc.Natl.Acad.ScL USA 81: 6851, 1985; Takeda et See al., Nature 314: 452,1985, Variable et al., US Pat. No. 4,816,567; Boss et al., US Pat. No. 4,816,397; Tanaguchi et al., EP0171496; EP0173494; and GB2177096. matter.

Antibodies, antibody constructs or antibody fragments may also be modified by specific deletion of human T cell epitopes (a method called "deimmunization"), eg, by the methods disclosed in WO98 / 52976 or WO00 / 34317. .. Briefly, the heavy and light chain variable domains of an antibody can be analyzed for peptides that bind to MHC class II; these peptides are potentially (defined in WO98 / 52976 and WO00 / 34317). Corresponds to the T cell epitope. As described in WO98 / 52976 and WO00 / 34317, a computer modeling technique called "peptide threading" can be applied for the detection of potential T cell epitopes, in addition to human MHC class II binding. A database of peptides can be searched for motifs present in VH and VL sequences. These motifs are 18 major MHC class II It binds to any of the DR allotypes and therefore constitutes a potential T cell epitope. Detected potential T cell epitopes can be removed by substituting a small number of amino acid residues in the variable domain, or preferably by a single amino acid substitution. Typically, conservative substitutions are made. In many cases, but not limited to, amino acids common to positions in the human germline antibody sequence may be used. Human germline sequences are, for example, Tomlinson, et al. (1992) J.MoI.Biol. 227: 776-798; Cook, GP et al. (1995) Immunol. Today Vol.16 (5): 237-242. ; And Tomlinson et al. (1995) EMBO J. 14:14: 4628-4638. The V BASE directory is (Tomlinson, LA.et al. MRC Center for Protein) Provides a comprehensive directory of human immunoglobulin variable region sequences (compiled by Engineering, Cambridge, UK). These sequences can be used, for example, as a source of human sequences for framework regions and CDRs. The consensus human framework area can also be used, for example, as described in US Pat. No. 6,300,064.

"Humanized" antibodies, antibody constructs or fragments thereof (such as Fv, Fab, Fab', F (ab') 2 or other antigen-binding partial sequences of an antibody) are mostly human sequences of antibodies or immunoglobulins. Yes, they contain (a) the smallest sequence (s) derived from non-human immunoglobulins. For the most part, humanized antibodies are non-human (eg, rodent) species, eg, mice, in which the residues of the recipient's hypervariable region (also CDR) have the desired specificity, affinity, and ability. , A human immunoglobulin (recipient antibody) that has been replaced by a residue (donor antibody) in the hypervariable region, such as rat, hamster or rabbit. In some cases, the Fv framework region (FR) residue of human immunoglobulin is replaced by the corresponding non-human residue. In addition, a "humanized antibody" may also include residues not found in either the recipient antibody or the donor antibody as used herein. These modifications are made to further improve and optimize the performance of the antibody. Humanized antibodies may also contain at least a portion of an immunoglobulin constant region (Fc), typically a human immunoglobulin constant region. For more details, see Jones et al., Nature, 321: 522-525 (1986); Reichmann et al., Nature, 332: 323-329 (1988); and Presta, Curr. Op. Struct. Biol., 2 See: 593-596 (1992).

Humanized antibodies or fragments thereof can be generated by replacing the sequences of Fv variable domains that are not directly involved in antigen binding with equivalent sequences from human Fv variable domains. Illustrative methods of producing humanized antibodies or fragments thereof are described by Morrison (1985) Science 229: 1202-1207; by Oi et al. (1986) BioTechniques 4: 214; and US 5,585,089; US 5,693,761; Provided by US5,693,762; US5,859,205; and US6,407,213. These methods involve isolating, manipulating, and expressing 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. Such nucleic acids may be obtained from hybridomas that produce antibodies against a given target as described above, as well as from other sources. Recombinant DNA encoding a humanized antibody molecule can then be cloned into a suitable expression vector.

Humanized antibodies are also produced using transgenic animals, such as animals that express human heavy and light chain genes, such as mice, but cannot express endogenous mouse immunoglobulin heavy and light chain genes. May be done. Winter describes an exemplary CDR transplantation method that may be used to prepare the humanized antibodies described herein (US Pat. No. 5,225,539). All CDRs in a particular human antibody may be replaced by at least some of the non-human CDRs, or only some of the CDRs may be replaced by non-human CDRs. It is only necessary to replace as many CDRs as needed to bind the humanized antibody to a given antigen.

Humanized antibodies can be optimized by the introduction of conservative substitutions, consensus sequence substitutions, germline substitutions and / or reversion mutations. Immunoglobulin molecules thus altered can be made by any of several techniques known in the art (eg, 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 EP 239400).

The terms "human antibody," "human antibody construct," and "human binding domain" are used, for example, in Kabat et. It has antibody regions, such as variable and constant regions or domains, that substantially correspond to human germline immunoglobulin sequences known in the art, including those described by al. (1991) (above). Includes antibodies, antibody constructs and binding domains. The human antibodies, antibody constructs or binding domains of the invention are, for example, amino acid residues not encoded by human germline immunoglobulin sequences in CDR, especially in CDR3 (eg, by random or site-specific mutagenesis in vitro, or. Mutations introduced by somatic mutations in vivo) may be included. Human antibodies, antibody constructs or binding domains have at least one, two, three, four, five or more positions replaced by amino acid residues not encoded by the human germline immunoglobulin sequence. Can be done. Definitions of human antibodies, antibody constructs and binding domains also assume fully human antibodies as used herein, such as those that can be derived by using techniques or systems such as Xenomouse. Contains only non-artificial and / or genetically modified human sequences of antibodies.

In some embodiments, the antibody constructs of the invention are "isolated" or "substantially pure" antibody constructs. "Isolated" or "substantially pure", when used to describe the antibody constructs disclosed herein, have been identified, isolated and / or recovered from components of their production environment. Means an antibody construct. Preferably, the antibody construct is unrelated or substantially unrelated to all other components from its production environment. Contaminating components of its production environment, such as those resulting from recombinant transfected cells, are materials that typically interfere with the diagnostic or therapeutic use of polypeptides, such as enzymes, hormones, and other proteins. May contain sex or non-proteinaceous solutes. The antibody construct may, for example, make up at least about 5% by weight, or at least about 50% by weight, of total protein in a given sample. It is understood that the isolated protein may constitute 5% to 99.9% by weight of the total protein content, depending on the circumstances. The polypeptide may be made at significantly higher concentrations by the use of an inducible promoter or a highly expressed promoter, resulting in higher concentration levels. This definition includes the production of antibody constructs in a wide variety of organisms and / or host cells known in the art. In a preferred embodiment, the antibody construct is (1) to the extent sufficient to obtain at least 15 residues of the N-terminal or internal amino acid sequence by use of a spinning cup sequencer, or (2) Coomassie blue, or preferably. Will be purified by SDS-PAGE under non-reducing or reducing conditions using silver staining until uniform. However, usually the isolated antibody construct will be prepared by at least one purification step.

The term "binding domain", in the context of the present invention, binds (specifically) to / interacts with a given target epitope or given target site on each of the target molecules (antigens) CDH3 and CD3. Characterize the domain that acts / recognizes it. The structure and function of the first binding domain (recognizing CDH3), and preferably the structure and / or function of the second binding domain (CD3), are also the structure and / or function of the antibody, eg, full-length or total immunoglobulin molecule. Based on function. According to the present invention, the first binding domain contains the presence of three light chain CDRs (ie, CDR1, CDR2 and CDR3 in the VL region) and three heavy chain CDRs (ie, CDR1, CDR2 and CDR3 in the VH region). It is a feature. The second binding domain also preferably comprises the minimum structural requirements of the antibody to allow target binding. More preferably, the second binding domain comprises at least three light chain CDRs (ie, CDR1, CDR2 and CDR3 in the VL region) and / or three heavy chain CDRs (ie, CDR1, CDR2 and CDR3 in the VH region). .. The first and / or second binding domains may be made or available by phage display or library screening methods rather than transplanting CDR sequences from existing (monoclonal) antibodies into the scaffold. is assumed.

According to the present invention, the binding domain is preferably in the form of a polypeptide. Such polypeptides may contain proteinaceous and non-proteinaceous moieties such as chemical linkers or chemical cross-linking agents, such as glutaraldehyde. A protein, including fragments thereof, preferably biologically active fragments, and peptides usually having less than 30 amino acids, comprises two or more amino acids linked to each other by a shared peptide bond (which results in an amino acid chain). .. The term "polypeptide" as used herein describes a group of molecules and usually consists of more than 30 amino acids. Polypeptides may further form multimers, such as dimers, trimer or higher oligomers, etc., i.e. consist of two or more polypeptide molecules. The polypeptide molecules that form such dimers, trimers, and the like may be the same or non-identical. As a result, the corresponding higher-order structures of such multimers are referred to as homodimers or heterodimers, homotrimers or heterotrimers, and the like. An example of a heteromultimer is an antibody molecule, which, in its naturally occurring form, consists of two identical light polypeptide chains and two identical heavy polypeptide chains. The terms "peptide", "polypeptide" and "protein" also refer to naturally modified peptides / polypeptides / proteins, where the modification is post-translational, such as, for example, glycosylation, acetylation, phosphorylation, etc. Achieved by modification. The "peptide", "polypeptide" or "protein" may also be chemically modified, eg PEGylated, as referred to herein. Such modifications are well known in the art and are described herein below.

As mentioned above, the binding domain may typically include the antibody light chain variable region (VL) and the antibody heavy chain variable region (VH); but it does not have to. The Fd fragment has, for example, two VH regions and often retains some antigen-binding function of the intact antigen-binding domain. Examples of (modified) antigen-binding antibody fragments include (1) Fab fragments, i.e. monovalent fragments with VL, VH, CL and CH1 domains; (2) F (ab') 2 fragments, i.e. hinges. A divalent fragment with two Fab fragments linked by disulfide bridges in the region; (3) an Fd fragment with two VH and CH1 domains; (4) an Fv fragment with a single arm VL and VH domain of the antibody, (5) dAb fragment with VH domain (Ward et al., (1989) Nature 341: 544-546); (6) Isolating complementarity determining regions (CDRs), and (7) single-chain Fv (scFv), preferably the latter (eg, derived from the scFV library). Examples of embodiments of antibody constructs according to the invention include, for example, WO00 / 006605, WO2005 / 040220, WO2008 / 119567, WO2010 / 037838, WO2013 / 026837, WO2013 / 026833, US2014 / 0308285, US2014 / 0302037, WO2014 / 144722, It is described in WO2014 / 151910 and WO2015 / 048272.

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 that include at least one human binding domain are associated with antibodies or antibody constructs that have variable and / or constant regions such as non-humans, such as rodents (eg, mice, rats, hamsters or rabbits). Avoid some of the problems. The presence of such rodent-derived proteins can result in rapid clearance of the antibody or antibody construct, or can result in the generation of an immune response by the patient to the antibody or antibody construct. To avoid the use of rodent-derived antibodies or antibody constructs, human or fully human antibody / antibody constructs should introduce human antibody function into rodents such that rodents produce fully human antibodies. Can be generated by.

The ability to clone and reconstruct megabase-sized human loci in YAC and to introduce them into the mouse germline is powerful in revealing the functional components of very large or coarsely mapped loci. In addition to the method, it provides a powerful method to generate a useful model of human disease. In addition, the use of such techniques to replace mouse loci with their human equivalents is used for the expression and regulation of developing human gene products, their transmission with other systems, and the induction and progression of disease. It can provide unique insights into their involvement.

An important practical application of such a strategy is the "humanization" of the mouse humoral immune system. Introducing the human immunoglobulin (Ig) locus into mice in which the endogenous Ig gene has been inactivated tests the underlying mechanisms of their role in B cell development, in addition to antibody programming and assembly. Opportunity to do is provided. In addition, such strategies may provide an ideal source for the production of fully human monoclonal antibodies (mAbs), which is an important milestone in achieving assurance of antibody therapy in human disease. be. Fully human antibodies or antibody constructs are expected to minimize the immunogenicity and allergic response inherent in mice or mouse derivatized mAbs, thereby increasing the efficacy and safety of the antibodies / antibody constructs administered. NS. The use of fully human antibodies or antibody constructs can be expected to provide significant benefits in the treatment of chronic and recurrent human diseases requiring repeated compound administration, such as inflammation, autoimmunity, and cancer.

One approach to this goal is to use mouse strains deficient in mouse antibody production in anticipation that such mice will produce a broad repertoire of human antibodies in the absence of mouse antibodies. It was to manipulate with a large fragment of the locus. Large human Ig fragments preserve a wide range of variable gene diversity, as well as proper regulation of antibody production and expression. By utilizing mouse mechanisms for antibody diversification and selection and lack of immune tolerance to human proteins, the human antibody repertoire reproduced in these mouse strains is for any antigen of interest, including human antigens. It should produce high affinity antibodies. Using hybridoma technology, antigen-specific human mAbs with the desired specificity can be readily produced and selected. This general strategy has been demonstrated in connection with the generation of the first XenoMouse mouse strain (Green et al. Nature Genetics). See 7: 13-21 (1994)). The XenoMouse strain was engineered on a yeast artificial chromosome (YAC) containing germline constituents 245 kb and 190 kb in size, respectively, at the human heavy chain locus and kappa light chain locus containing core variable and constant region sequences. .. Human Ig-containing YAC was demonstrated to be compatible with mouse strains for both antibody rearrangement and expression and could be replaced with the inactivated mouse Ig gene. This was demonstrated by their ability to induce B cell development, produce an adult-like human repertoire of fully human antibodies, and produce antigen-specific human mAbs. These results also characterize the human humoral response to infection and immunization by introducing the majority of human Ig loci containing more V genes, additional regulatory elements, and human Ig constant regions. It was suggested that a substantially complete repertoire may be repeated. The study by Green et al. Has recently been extended to the introduction of approximately over 80% of the human antibody repertoire with the introduction of megabase-sized germline-constituting YAC fragments at the human heavy chain and kappa light chain loci, respectively. Mendez et al. Nature See Genetics 15: 146-156 (1997) and US Patent Application No. 08 / 759,620.

The production of XenoMouse mice is described in US Patent Application No. 07 / 466,008, 07 / 610,515, 07 / 919,297, 07 / 922,649, 08 / 031,801, No. 08 / 112,848, 08 / 234,145, 08 / 376,279, 08 / 430,938, 08 / 464,584, 08 / 464,582, 08 / 463,191, 08 / 462,837, 08 / 486,853, 08 / 486,857, 08 / 486,859, 08 / 462,513, 08 / 724,752 Description and 08 / 759,620; and US Pat. No. 6,162,963; No. 6,150,584; No. 6,114,598; No. 6,075,181 and No. 5,939,598 and Japanese Patent No. 3068180B2. Further discussed and illustrated in the specification, specification 3068506B2, and specification 3068507B2. Mendez et al.Nature Genetics 15: 146-156 (1997) and Green and Jakobovits J.Exp.Med.188: 483-495 (1998), EP0463151B1, WO94 / 02602, WO96 / 34096, WO98 / 24893, WO00 / 76310, and See also WO 03/47336.

In alternative methods, others, including GenPharm International, Inc., used the "mini-locus" method. In the mini-locus technique, the extrinsic Ig locus is mimicked by including fragments (individual genes) from the Ig locus. Thus, one or more VH genes, one or more DH genes, one or more JH genes, a mu constant region, and a second constant region (preferably a gamma constant region) are constructs for insertion into animals. Is formed into. This technique is described by Surani et al., U.S. Pat. No. 5,545,807, and Lonberg and Kay, U.S. Pat. Nos. 5,545,806; 5,625,825; 5,625,126; 5,633,425; 5,661,016, respectively. No. 5,770,429; No. 5,789,650; No. 5,814,318; No. 5,877,397; No. 5,874,299; and No. 6,255,458, Krimpenfort and Berns US Pat. Nos. 5,591,669 and 6,023.010, Berns et al. US Pat. Nos. 5,612,205; 5,721,367; and 5,789,215; and Choi and Dunn US Pat. 5,643,763, as well as GenPharm International US Patent Application Nos. 07 / 574,748, 07 / 575,962, 07 / 810,279, 07 / 853,408, 07 / 904,068, 07 / 990,860 Specification, 08 / 053,131, 08 / 096,762, 08 / 155,301, 08 / 161,739, 08 / 165,699, 08 / 209,741 It is described in. EP0546073B1, WO92 / 03918, WO92 / 22645, WO92 / 22647, WO92 / 22670, WO93 / 12227, WO94 / 00569, WO94 / 25585, WO96 / 14436, WO97 / 13852, and WO98 / 24884 and U.S. Pat. No. 5,981,175. See also. In addition, Taylor et al. (1992), Chen et al. (1993), Tuaillon et al. (1993), Choi et al. (1993), Lonberg et al. (1994), Taylor et al. See al. (1994), and Tuaillon et al. (1995), Fishwild et al. (1996).

Kirin also demonstrated the production of human antibodies from mice into which large chromosome fragments or whole chromosomes were introduced by microcell fusion. See European Patent Application No. 773288 and No. 843961. Xenerex Biosciences is developing techniques for the possible production of human antibodies. In this technique, SCID mice are reconstituted with human lymph cells, such as B cells and / or T cells. Mice can then be immunized with the antigen, resulting in an immune response against that antigen. See U.S. Pat. Nos. 5,476,996; 5,698,767; and 5,958,765.

The human anti-mouse antibody (HAMA) response has contributed to the industry's preparation of chimeric or otherwise humanized antibodies. However, it is expected that certain human anti-chimeric antibody (HACA) responses will be observed, especially in the use of chronic or multiple doses of the antibody. Therefore, it is desirable to provide antibody constructs comprising a fully human binding domain for CDH3 and a fully human binding domain for CD3 in order to reduce the concerns and / or effects of the HAMA or HACA response.

The terms "to (specifically) bind to", "to (specifically) recognize", "to (specifically) target", and "to (specifically) react with" are used in the book. According to the invention, the binding domain interacts with one or more, preferably at least two, more preferably at least three, most preferably at least four amino acids of the epitope located on the target protein or antigen (CDH3 / CD3). It means to do or interact specifically.

The term "epitope" refers to a site on an antigen to which a binding domain, such as an antibody or immunoglobulin or a derivative or fragment of an antibody or immunoglobulin, specifically binds. "Epitope" is antigenic and therefore the term epitope is sometimes also referred to herein as "antigenic structure" or "antigenic determinant". Therefore, the binding domain is the "antigen interaction site". This binding / interaction is also understood to define "specific cognition."

An "epitope" can be formed by both continuous amino acids or discontinuous amino acids in close proximity by three-dimensional folding of a protein. A "linear epitope" is an epitope whose primary amino acid sequence contains a recognized epitope. Linear epitopes typically contain at least 3 or at least 4, more usually at least 5 or at least 6 or at least 7, in a unique sequence, eg, about 8 to about 10 amino acids. It may also be longer and contain at least 15 or 20 amino acids, at least 25 or 30 amino acids, or more.

A "three-dimensional epitope" is an epitope in which the primary sequence of an amino acid containing the epitope is not the only defining component of the recognized epitope, as opposed to a linear epitope (eg, the primary sequence of an amino acid is not necessarily dependent on the binding domain. It is an epitope that is not always recognized). Typically, conformational epitopes contain a large number of amino acids compared to linear epitopes. With respect to the recognition of tertiary structure epitopes, the binding domain recognizes the three-dimensional structure of an antigen, preferably a peptide or protein or fragment thereof (in the context of the present invention, an antigen for one of the binding domains is contained within the CDH3 protein. Is). For example, when a protein molecule is folded to form a three-dimensional structure, the specific amino acid and / or polypeptide skeletons that form the tertiary structure epitope are in close proximity, which allows the antibody to recognize the epitope. .. Methods for determining the conformation of an epitope include X-ray crystallography, two-dimensional nuclear magnetic resonance (2D-NMR) spectroscopy, and site-specific spin label and electron paramagnetic resonance (EPR) spectroscopy. Not limited to.

The examples provided describe a further method of characterizing a given binding domain, which indicates whether a given binding domain binds to a given protein, in particular one or more epitopes (s) of CDH3. Including the test.

As used herein, the term "epitope cluster" refers to an epitope located in a defined continuous stretch of antigen. Epitope clusters can contain one, two or more epitopes. The antibody construct can also bind to additional epitopes other than this cluster in addition to the epitope within the epitope cluster, which can correspond to discontinuous epitopes. Discontinuous epitopes are usually characterized by including amino acid stretches of non-contiguous antigens. For example, antibody constructs can bind to the extracellular subdomains D3A and D3C, but not D3B. The concept of "epitope clustering" is also used in characterizing the antibody constructs of the present invention. The epitope clusters and epitopes defined in the context of the present invention in the extracellular domain of CDH3 are described above and are shown in FIG.

The extracellular domains (D1 to D5) or their subdomains (A, B, C) in the human CDH3 protein are the respective extracellular domains (D1 to) of the non-human and non-primate (eg, chicken or mouse) CDH3 antigens. D5) or its subdomains (A, B, C) are exchanged (causing a construct containing human CDH3, one human extracellular domain or its subdomain is its counterpart, the non-human extracellular domain or its counterpart. If it is replaced by a subdomain), there will be a decrease in the connectivity of the binding domain. This reduction is preferably at least by setting the binding of the human CDH3 protein to each extracellular domain (D1-D5) or its subdomain to 100% compared to each epitope cluster of the human CDH3 protein. 10%, 20%, 30%, 40%, 50%; more preferably at least 60%, 70%, 80%, 90%, 95% or even 100%. It is assumed that the human CDH3 / non-human CDH3 chimera described above is expressed in CHO cells. It is also envisioned that the human CDH3 / non-human CDH3 chimera will be fused with a different membrane binding protein, eg, a transmembrane domain such as EpCAM and / or a cytoplasmic domain.

Non-humans (eg, mice, but other animals such as rats, hamsters, rabbits, chickens, etc. may also be considered) with each extracellular domain (D1-D5) of the CDH3 antigen or its subdomains. A method of testing this loss of binding due to exchange is described in Example 2. A further method of determining the contribution of a particular residue of a target antigen to recognition by an antibody construct or binding domain is alanine scanning (eg, Morrison KL &amp; Weiss GA.Cur Opin Chem Biol.2001 Jun; 5 (3): 302-7. ), Each residue analyzed is replaced with alanine, for example, by site-directed mutagenesis. The reason alanine is used is that it is not bulky and is chemically inert, but nevertheless has a methyl functional group that mimics the secondary structure references found in many other amino acids. Because it has. Occasionally, bulky amino acids, such as valine or leucine, can be used when conservation of the size of the mutant residue is desired. Alanine scanning is a mature technique that has been used for many years.

The interaction between the binding domain and the epitope or epitope cluster is that the binding domain exhibits a significant affinity for the epitope or epitope cluster on a particular protein or antigen (here, CDH3 and CD3, respectively). It generally suggests that it does not show significant reactivity with proteins or antigens other than CDH3 or CD3. "Remarkable affinity" is about 10^-6Includes binding with M (KD) or stronger affinity. Preferably, the binding has a binding affinity of about 10^-12~10^-8M, 10^-12~10^-9M, 10^-12~10^-10M, 10^-11~10^-8M, preferably about 10^-11~10^-9It is considered to be specific when it is M. Whether or not the binding domain reacts specifically with or binds to the target specifically compares 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. This can be easily tested. Preferably, the binding domain of the invention is essentially or substantially non-binding to a protein or antigen other than CDH3 or CD3 (ie, the first binding domain is unable to bind to a protein other than CDH3, and the first The 2-binding domain cannot bind to proteins other than CD3).

The terms "essentially / substantially non-binding" or "cannot bind" mean that the binding domain of the invention does not bind to a protein or antigen other than CDH3 or CD3, i.e. a protein other than CDH3 or CD3. Alternatively, by setting the binding to each of CDH3 or CD3 to 100% with respect to the antigen, the reactivity is not more than 30%, and the reactivity is preferably 20% or less, more preferably 10% or less, particularly. It means that it is preferably 9%, 8%, 7%, 6% or 5% or less.

Specific binding is believed to be brought about by specific motifs in the binding domain and the amino acid sequence of the antigen. Therefore, binding is achieved as a result of their primary, secondary and / or tertiary structure, as well as the secondary modification of this structure. Specific interactions between an antigen interaction site and that particular antigen may result in a simple binding of that site to the antigen. In addition, specific interactions between the antigen interaction site and its particular antigen may, as an alternative or addition, result in the initiation of a signal, eg, by inducing changes in the conformation of the antigen, by oligomerization of the antigen, etc. ..

In another aspect, the invention comprises a dual comprising a first human binding domain that binds to human CDH3 on the surface of target cells and preferably a second human binding domain that binds to human CD3 on the surface of T cells. Providing a specific antibody construct, the first binding domain is a) CDR-H1 shown in SEQ ID NO: 149, CDR-H2 shown in SEQ ID NO: 150, CDR-H3 shown in SEQ ID NO: 151, sequence. CDR-L1 shown in number: 152, CDR-L2 shown in SEQ ID NO: 153 and CDR-L3 shown in SEQ ID NO: 154; b) CDR-H1 shown in SEQ ID NO: 159, SEQ ID NO: 160 CDR-H2 shown, CDR-H3 shown in SEQ ID NO: 161; CDR-L1 shown in SEQ ID NO: 162, CDR-L2 shown in SEQ ID NO: 163 and CDR-L3 shown in SEQ ID NO: 164; c) CDR-H1 shown in SEQ ID NO: 169, CDR-H2 shown in SEQ ID NO: 170, CDR-H3 shown in SEQ ID NO: 171; CDR-L1 shown in SEQ ID NO: 172, SEQ ID NO: 173. CDR-L2 shown in SEQ ID NO: and CDR-L3; d shown in SEQ ID NO: 174, CDR-H1 shown in SEQ ID NO: 179, CDR-H2 shown in SEQ ID NO: 180, CDR shown in SEQ ID NO: 181. -H3, CDR-L1 shown in SEQ ID NO: 182, CDR-L2 shown in SEQ ID NO: 183 and CDR-L3; e) SEQ ID NO: 189 shown in CDR-H1, SEQ ID NO: CDR-H2 shown in number: 190, CDR-H3 shown in SEQ ID NO: 191; CDR-L1 shown in SEQ ID NO: 192, CDR-L2 shown in SEQ ID NO: 193 and SEQ ID NO: 194. CDR-L3; f) CDR-H1 shown in SEQ ID NO: 199, CDR-H2 shown in SEQ ID NO: 200, CDR-H3 shown in SEQ ID NO: 201, CDR-L1 shown in SEQ ID NO: 202, CDR-L2 shown in SEQ ID NO: 203 and CDR-L3 shown in SEQ ID NO: 204;g) CDR-H1 shown in SEQ ID NO: 209, CDR-H2 shown in SEQ ID NO: 210, CDR-H3 shown in SEQ ID NO: 211, CDR-L1 shown in SEQ ID NO: 212, SEQ ID NO: 213. CDR-L2 shown in SEQ ID NO: 214 and CDR-L3; h shown in SEQ ID NO: 214, CDR-H1 shown in SEQ ID NO: 219, CDR-H2 shown in SEQ ID NO: 220, CDR shown in SEQ ID NO: 221. -H3, CDR-L1 shown in SEQ ID NO: 222, CDR-L2 shown in SEQ ID NO: 223 and CDR-L3; i shown in SEQ ID NO: 224; CDR-H1 shown in SEQ ID NO: 229, SEQ ID NO: CDR-H2 shown in number: 230, CDR-H3 shown in SEQ ID NO: 231; CDR-L1 shown in SEQ ID NO: 232, CDR-L2 shown in SEQ ID NO: 233 and SEQ ID NO: 234. CDR-L3; and j) CDR-H1 shown in SEQ ID NO: 239, CDR-H2 shown in SEQ ID NO: 240, CDR-H3 shown in SEQ ID NO: 241 and CDR-L1 shown in SEQ ID NO: 242. , VH region containing CDR-H1, CDR-H2 and CDR-H3 selected from the group consisting of CDR-L2 shown in SEQ ID NO: 243 and CDR-L3 shown in SEQ ID NO: 244, and CDR-L1, CDR. -Includes VL region containing L2 and CDR-L3.CDR-H3 shown in 231, CDR-L1 shown in SEQ ID NO: 232, CDR-L2 shown in SEQ ID NO: 233 and CDR-L3 shown in SEQ ID NO: 234; and j) SEQ ID NO: 239. CDR-H1, CDR-H2 shown in SEQ ID NO: 240, CDR-H3 shown in SEQ ID NO: 241, CDR-L1 shown in SEQ ID NO: 242, CDR-L2 and sequence shown in SEQ ID NO: 243. Includes the VH region containing CDR-H1, CDR-H2 and CDR-H3 selected from the group consisting of CDR-L3 shown in number: 244 and the VL region containing CDR-L1, CDR-L2 and CDR-L3.CDR-H3 shown in 231, CDR-L1 shown in SEQ ID NO: 232, CDR-L2 shown in SEQ ID NO: 233 and CDR-L3 shown in SEQ ID NO: 234; and j) SEQ ID NO: 239. CDR-H1, CDR-H2 shown in SEQ ID NO: 240, CDR-H3 shown in SEQ ID NO: 241, CDR-L1 shown in SEQ ID NO: 242, CDR-L2 and sequence shown in SEQ ID NO: 243. Includes the VH region containing CDR-H1, CDR-H2 and CDR-H3 selected from the group consisting of CDR-L3 shown in number: 244 and the VL region containing CDR-L1, CDR-L2 and CDR-L3.

In another aspect, the invention comprises a dual comprising a first human binding domain that binds to human CDH3 on the surface of target cells and preferably a second human binding domain that binds to human CD3 on the surface of T cells. Providing a specific antibody construct, the first binding domain is a) CDR-H1 set forth in SEQ ID NO: 279, CDR-H2 set forth in SEQ ID NO: 280, CDR-H3 set forth in SEQ ID NO: 281. Number: CDR-L1 shown in 282, CDR-L2 shown in SEQ ID NO: 283 and CDR-L3 shown in SEQ ID NO: 284; b) CDR-H1 shown in SEQ ID NO: 289, SEQ ID NO: 290 CDR-H2 shown, CDR-H3 shown in SEQ ID NO: 291; CDR-L1 shown in SEQ ID NO: 292, CDR-L2 shown in SEQ ID NO: 293 and CDR-L3 shown in SEQ ID NO: 294; c) CDR-H1 shown in SEQ ID NO: 299, CDR-H2 shown in SEQ ID NO: 300, CDR-H3 shown in SEQ ID NO: 301, CDR-L1 shown in SEQ ID NO: 302, SEQ ID NO: 303. CDR-L2 shown in SEQ ID NO: and CDR-L3; d shown in SEQ ID NO: 304, CDR-H1 shown in SEQ ID NO: 309, CDR-H2 shown in SEQ ID NO: 310, CDR shown in SEQ ID NO: 311. -H3, CDR-L1 shown in SEQ ID NO: 312, CDR-L2 shown in SEQ ID NO: 313 and CDR-L3; e) SEQ ID NO: 319 shown in CDR-H1, SEQ ID NO: CDR-H2 shown in number: 320, CDR-H3 shown in SEQ ID NO: 321, CDR-L1 shown in SEQ ID NO: 322, CDR-L2 shown in SEQ ID NO: 323 and SEQ ID NO: 324. CDR-L3; f) CDR-H1 shown in SEQ ID NO: 329, CDR-H2 shown in SEQ ID NO: 330, CDR-H3 shown in SEQ ID NO: 331, CDR-L1 shown in SEQ ID NO: 332, CDR-L2 shown in SEQ ID NO: 333 and CDR-L3 shown in SEQ ID NO: 334;g) CDR-H1 shown in SEQ ID NO: 339, CDR-H2 shown in SEQ ID NO: 340, CDR-H3 shown in SEQ ID NO: 341, CDR-L1 shown in SEQ ID NO: 342, SEQ ID NO: 343. CDR-L2 shown in SEQ ID NO: CDR-L3 shown in SEQ ID NO: 344; and h) CDR-H1 shown in SEQ ID NO: 349, CDR-H2 shown in SEQ ID NO: 350, and SEQ ID NO: 351. CDR-H1 and CDR-H2 selected from the group consisting of CDR-H3, CDR-L1 shown in SEQ ID NO: 352, CDR-L2 shown in SEQ ID NO: 353 and CDR-L3 shown in SEQ ID NO: 354. And VH region containing CDR-H3 and VL region containing CDR-L1, CDR-L2 and CDR-L3.

The term "variable" refers to a portion of an antibody or immunoglobulin domain that exhibits variability in those sequences and is involved in determining the specificity and binding affinity of a particular antibody (ie, "variable domain". ) ). The pairing of variable heavy chain (VH) and variable light chain (VL) both form a single antigen binding site. The CH domain most proximal to VH is represented as CH1. Each light (L) chain is linked to a heavy (H) chain by one covalent disulfide bond, while the two H chains are linked to each other by one or more disulfide bonds, depending on the H chain isotype.

The variability is not uniformly distributed throughout the variable domain of the antibody; it is concentrated in each subdomain of the heavy chain variable region and the light chain variable region. These subdomains are called "hypervariable regions" or "complementarity determining regions" (CDRs). The more conservative (ie, non-supervariable) portion of the variable domain is called the "framework" region (FRM or FR), the skeleton for the six CDRs to form the antigen-binding surface in three-dimensional space. I will provide a. The naturally occurring heavy and light chain variable domains each contain four FRM regions (FR1, FR2, FR3, and FR4), mostly in a beta-sheet arrangement with three hypervariable regions linked to it. , These are linked to the β-sheet structure to form loops that, in some cases, form part of it. The hypervariable regions of each strand are brought together by FRM and, together with the hypervariable regions of the other strand, contribute to the formation of antigen-binding sites (see Kabat et al. Above). .. The constant domain is not directly involved in antigen binding, but exhibits various effector functions such as antibody dependence, cell-mediated cytotoxicity and complement activation.

The term "CDR" and its plurals refer to complementarity determining regions, three of which constitute the binding properties of the light chain variable regions (CDR-L1, CDR-L2 and CDR-L3) and three of which are heavy chains. It constitutes the coupling characteristics of the variable regions (CDR-H1, CDR-H2 and CDR-H3). They are the major determinants of antigen specificity because CDRs contain most of the residues involved in the specific interaction of the antibody with the antigen and therefore contribute to the functional activity of the antibody molecule.

The exact definition of CDR boundaries and lengths depends on different classification and numbering systems. Therefore, CDRs may be referred to by Kabat, Chothia, contact or any other boundary definition, including the numbering system described herein. Despite the different boundaries, each of these systems overlaps to some extent with respect to the parts that make up the so-called "hypervariable regions" within the variable array. Therefore, the CDR definitions by these systems may differ in length and boundary area with respect to adjacent framework areas. For example, Kabat (a method based on interspecific sequence variability), Chothia (a method based on crystallographic testing of antigen-antibody complexes), and / or MacCallum (Kabat et al .; Chothia et al., J., supra). See MoI.Biol, 1987,196: 901-917; and MacCallum et al., J.MoI.Biol, 1996,262: 732). Yet another standard for characterizing antigen binding sites is Oxford. AbM definition used by Molecular's AbM antibody modeling software. See, for example, the Protein Sequence and Structure Analysis of Antibody Variable Domains.In: Antibody Engineering Lab Manual (Ed .: Duebel, S. and Kontermann, R., Springer-Verlag, Heidelberg). To the extent that two residue identification techniques define overlapping but not identical regions, they can be combined to define a hybrid CDR. However, numbering according to the so-called Kabat system is preferred.

Typically, the CDRs form a loop structure that can be classified as a canonical structure. The term "canonical structure" refers to the conformation of the main chain taken by the antigen binding (CDR) loop. Comparative structural studies have found that 5 of the 6 antigen-binding loops have only a limited repertoire of available conformations. Each canonical structure can be characterized by the helix angle of the polypeptide backbone. Therefore, the corresponding loops between antibodies may have very similar three-dimensional structure, despite the high degree of amino acid sequence variability in most of the loops (Chothia and Lesk, J.MoI.Biol., 1987,196: 901; Chothia et al., Nature, 1989,342: 877; Martin and Thornton, J.MoI.Biol, 1996, 263: 800). Furthermore, there is a relationship between the loop structure taken and the amino acid sequence around it. The conformation of a particular canonical class is determined by the length of the loop as well as the amino acid residues present at key positions within the loop and within the conservative framework (ie, outside the loop). Therefore, assignments to specific canonical classes can be made based on the presence of these major amino acid residues.

The term "canonical structure" also refers to, for example, Kabat (Kabat et above). It may include consideration of the linear sequence of antibodies, as classified by al.). The Kabat numbering scheme (system) is a widely adopted standard for consistently numbering amino acid residues in antibody variable domains, as referred to elsewhere herein. This is the preferred scheme applied to the present invention. Further structural considerations can also be used to determine the canonical structure of the antibody. For example, these differences that are not fully reflected by Kabat numbering can be explained by the numbering system of Chothia et al. And / or revealed by other techniques such as crystallography and 2D or 3D computer modeling. .. Therefore, a given antibody sequence may be specifically classified into a canonical class that allows the identification of a suitable chassis sequence (eg, based on the desire to include various canonical structures in the library). Kabat numbering of antibody amino acid sequences and structural considerations as described by Chothia et al. Above and their suggestions for interpreting the canonical aspects of antibody structure are described in the literature. The subunit structure and three-dimensional arrangement of different classes of immunoglobulins are well known in the art. For an overview of antibody structure, see Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory, eds. Harlow et al., 1988.

Light chain CDR3, and especially heavy chain CDR3, may constitute the most important determinants of antigen binding within the light chain variable region and heavy chain variable region. In some antibody constructs, heavy chain CDR3 appears to constitute the major contact area between the antigen and the antibody. In vitro selection schemes that alter only CDR3 can be used to alter the binding properties of an antibody or to determine which residues contribute to antigen binding. Therefore, CDR3 is typically the largest source of molecular diversity within the antibody binding site. H3 can be, for example, as short as two amino acid residues, or greater than 26 amino acids.

The sequences of antibody genes are highly altered after assembly and somatic mutations, and these altered genes are 10¹⁰Presumed to encode a number of different antibody molecules (Immunoglobulin Genes, 2)^nd ed., eds. Jonio et al., Academic Press, San Diego, CA, 1995). Therefore, the immune system provides a repertoire of immunoglobulins. The term "repertoire" refers to at least one nucleotide sequence derived entirely or partially from at least one sequence encoding at least one immunoglobulin. Sequences (s) may be generated by in vivo rearrangement in the heavy chain V, D, and J segments, as well as the light chain V and J segments. Alternatively, the sequence (s) can be generated from cells in response to reorganization, eg, in vitro stimuli. Alternatively, some or all of the sequence (s) may be obtained by DNA splicing, nucleotide synthesis, mutagenesis, and other methods, see, eg, US Pat. No. 5,565,332. The repertoire may contain only one sequence or may contain multiple sequences, including sequences in genetically diverse groups.

In one embodiment, the first binding domain of the antibody construct of the invention is 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:: Includes the VH region selected from the group consisting of the VH regions shown in 215, SEQ ID NO: 225, SEQ ID NO: 235, and SEQ ID NO: 245.

In a further embodiment of the antibody construct of the invention, the first binding domain is 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:: Includes a VL region selected from the group consisting of the VL regions shown in 216, SEQ ID NO: 226, SEQ ID NO: 236, and SEQ ID NO: 246.

In another embodiment of the antibody construct of the invention, the first binding domain is SEQ ID NO: 155 + 156, SEQ ID NO: 165 + 166, SEQ ID NO: 175 + 176, SEQ ID NO: 185 + 186, SEQ ID NO: 195. Consists of a pair of VH and VL regions shown in +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. Includes VH and VL regions selected from the group.

In another embodiment, the first binding domain of the antibody construct of the invention is 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: Includes a VH region selected from the group consisting of the VH regions shown in: 345 and SEQ ID NO: 355.

In a further embodiment of the antibody construct of the invention, the first binding domain is 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:: Includes a VL region selected from the group consisting of the VL regions shown in 346 and SEQ ID NO: 356.

In another embodiment of the antibody construct of the invention, the first binding domain is SEQ ID NO: 285 + 286, SEQ ID NO: 295 + 296, SEQ ID NO: 305 + 306, SEQ ID NO: 315 + 316, SEQ ID NO: 325. Includes VH and VL regions selected from the group consisting of pairs of VH and VL regions shown in +326, SEQ ID NO: 335 + 336, SEQ ID NO: 345 + 346, and SEQ ID NO: 355 + 356.

The term "bispecific" as used herein refers to an antibody construct that is "at least bispecific". That is, it comprises at least a first binding domain and a second binding domain, in which case the first binding domain binds to one antigen or target (here, CDH3) and the second binding domain is another antigen or target (here). Here, it is bound to CD3). Thus, antibody constructs according to the invention contain specificity for at least two different antigens or targets. The term "bispecific antibody construct" of the present invention is also a multispecific antibody construct, such as a trispecific antibody construct, the latter comprising three binding domains or more than three (eg,). Includes constructs with specificity (4, 5, etc.).

If the antibody constructs according to the invention are (at least) bispecific, they are not naturally occurring and they are significantly different from naturally occurring products. Thus, a "bispecific" antibody construct or immunoglobulin is an artificial hybrid antibody or immunoglobulin having at least two different binding sites with different specificities. Bispecific antibodies can be produced by a variety of methods, including fusion of hybridomas or ligation of Fab'fragments. See, for example, Songsivilai &amp; Lachmann, Clin.Exp.Immunol. 79: 315-321 (1990).

At least two binding domains and variable domains of the antibody constructs of the invention may or may not include a peptide linker (spacer peptide). The term "peptide linker" refers to an amino acid sequence in which the amino acid sequences of one (variable and / or bound) domain and another (variable and / or bound) domain in the antibody construct of the present invention are linked to each other in accordance with the present invention. Define. An essential technical feature of such a peptide linker is that it does not contain any polymerization activity. Some suitable peptide linkers are described in US Pat. Nos. 4,751,180 and 4,935,233 or WO 88/09344.

When a linker is used, it is preferably of sufficient length and sequence to ensure that each of the first and second domains can retain their differential binding specificity independently of each other. Linker. For peptide linkers that link at least two binding domains (or two variable domains) in the antibody constructs of the invention, these peptide linkers contain only a few amino acid residues, eg, no more than 12 amino acid residues. Is preferable. Therefore, peptide linkers with 12, 11, 10, 9, 8, 7, 6 or 5 amino acid residues are preferred. Possible peptide linkers with less than 5 amino acids include 4, 3, 2 or 1 amino acid (s), with Gly-rich linkers being preferred. A particularly preferred "single" amino acid in the context of the "peptide linker" is Gly. Therefore, the peptide linker may consist of the single amino acid Gly. Another preferred embodiment of the peptide linker is the amino acid sequence Gly-Gly-Gly-Gly-Ser, ie Gly.₄Ser (SEQ ID NO: 393), or a polymer thereof, ie (Gly)₄Ser) It is characterized by x, and in the formula, x is an integer of 1 or more. Peptide linker characteristics include the absence of promotion of secondary structure and are known in the art, eg, Dall'Acqua et al. (Biochem. (1998) 37,9266-9273), Cheadle et al. ( It is described in Mol Immunol (1992) 29, 21-30) and Raag and Whitlow (FASEB (1995) 9 (1), 73-80). Also, peptide linkers that do not promote any secondary structure are preferred. Domain interconnection can be provided, for example, by genetic engineering as described in the Examples. Methods of preparing fused and functionally linked bispecific single-stranded constructs and expressing them in mammalian cells or bacteria are well known in the art (eg, WO99 / 54440 or Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, 2001).

Therefore, in the present invention, the antibody construct is (scFv).₂, ScFv-Provides a preferred embodiment in which the form is selected from the group consisting of single domain mAbs, diabodies and oligomers of any of the forms described above. The term "formal" does not preclude that the construct can be further modified, for example, by binding or fusing to other parts as described herein.

According to a particularly preferred embodiment, the antibody construct of the present invention is a "bispecific single chain antibody construct", more preferably a bispecific "single chain Fv" (svFv). The two domains of the Fv fragment, VL and VH, are encoded by separate genes, but they use recombinant methods to pair the VL and VH regions to form a monovalent molecule. They can be linked by synthetic linkers that can make them as protein chains. For example, Huston et al. (1988) Proc. Natl. Acad. Sci USA 85: 5879-5883). These antibody fragments are obtained using prior art known to those of skill in the art, and the fragments are evaluated for function in the same manner as whole or full length antibodies. Thus, a single chain variable fragment (scFv) is a fusion protein of the heavy chain variable region (VH) of the immunoglobulin and the light chain variable region (VL), which are usually about 10 to about 25 amino acids, preferably about 10 to about 25 amino acids. Is linked by a short linker peptide of about 15-20 amino acids. Linkers are usually rich in glycine for flexibility, plus serine or threonine for solubility, and the N-terminus of VH can be linked to the C-terminus of VL, or vice versa. It can be. This protein retains the original immunoglobulin specificity despite removal of constant regions and introduction of a linker.

Bispecific single-stranded molecules are known in the art and are WO99 / 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, Bruhl, Immunol., (2001), 166,2420-2426, Kipriyanov, It is described in J. Mol. Biol., (1999), 293, 41-56. The techniques described for the production of single chain antibodies (see, among others, US Pat. No. 4,946,778, Kontermann and Dubel (2010) above and Little (2009) above) have been selected targets. Yes) can be adapted to produce single chain antibody constructs that specifically recognize.

Bivalent (also called divalent) or bispecific single-chain variable fragment (form (scFv))₂Bi-scFv or di-scFv) with can be manipulated by linking two scFv molecules. If these two scFv molecules have the same binding specificity, they are obtained (scFv).₂The molecule will preferably be referred to as divalent (ie, it has a valence of 2 for the same target epitope). Obtained when two scFv molecules have different binding specificities (scFv)₂The molecule will preferably be referred to as bispecificity. Linkage is carried out by creating a single peptide chain with two VH regions and two VL regions, which can produce tandem scFv (eg, Kufer P. et al., (2004) Trends in Biotechnology 22 (eg). 5): See 238-244). Another possibility is the creation of scFv molecules using linker peptides that are too short (eg, about 5 amino acids) to fold two variable regions together and dimerize scFv. This species is known as the diabody (see, eg, Hollinger, Philipp et al., (July 1993) Proceedings of the National Academy of Sciences of the United States of America 90 (14): 6444-8. ).

According to a more preferred embodiment of the antibody construct of the present invention, the heavy chain (VH) and light chain (VL) of the binding domain (which binds to either the target antigen CDH3 or CD3) is driven by a peptide linker as described above. Rather than being directly linked, a binding domain is formed for the diabody as described. Thus, the VH of the CD3 binding domain may be fused to the VL of the CDH3 binding domain by a peptide linker, and the VH of the CDH3 binding domain is fused to the VL of the CD3 binding domain by such a peptide linker.

A single domain antibody contains only one (monomeric) antibody variable domain that can selectively bind to a particular antigen, independent of other V regions or domains. The first single-domain antibodies were engineered from heavy chain antibodies found in Camelids, which are Vs._HCalled the H fragment. Cartilaginous fish also have heavy chain antibodies (IgNARs) from which V_NARA single domain antibody called a fragment can be obtained. An alternative approach is, for example, to split a common immunoglobulin dimeric variable domain from humans or rodents into monomers, thereby obtaining VH or VL as a single domain Ab. Although most studies on single domain antibodies are currently based on heavy chain variable domains, Nanobodies derived from light chains have also been shown to specifically bind to target epitopes. Examples of single domain antibodies are referred to as sdAbs, Nanobodies or single variable domain antibodies.

Therefore, (single domain mAb)₂Is a monoclonal antibody construct composed of (at least) two single domain monoclonal antibodies, VH, VL, V_HH and V_NARAre individually selected from the group containing. The linker is preferably in the form of a peptide linker. Similarly, a "scFv-single domain mAb" is a monoclonal antibody construct composed of at least one single domain antibody as described above and one scFv molecule as described above. Again, the linker is preferably in the form of a peptide linker.

In one embodiment, the first binding domain is 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, which comprises an amino acid sequence selected from the group consisting of these sequences.

In another embodiment, the first binding domain is 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: Includes an amino acid sequence selected from the group consisting of these sequences shown in: 357.

The antibody constructs of the present invention are also expected to have additional functions in addition to their function of binding to the target molecules CDH3 and CD3. In this form, antibody constructs target target cells by binding to CDH3, mediate cytotoxic T cell activity by binding to CD3, and further function, eg, recruitment of effector cells such as NK cells. By providing fully functional Fc constant domains that mediate antibody-dependent cytotoxicity, labels (such as fluorescence), therapeutic agents such as toxins or radioactive nuclei, and / or means of enhancing serum half-life. It is a functional or multifunctional antibody construct.

Examples of means for extending the serum half-life of antibody constructs of the invention include peptides, proteins or domains of proteins, which are fused to or otherwise bound to antibody constructs. The group of peptides, proteins or protein domains contains peptides that bind to other proteins, such as serum albumin, with a preferred pharmacokinetic profile in the human body (see WO2009 / 127691). An example of these is shown in SEQ ID NO: 437. Such alternative half-life (HLE) peptides include peptides that bind to neonatal Fc receptors (FcRn, see WO2007 / 098420), which are also used in some of the constructs of the invention. NS. Concepts for binding larger domains or complete proteins of proteins include, for example, human serum albumin, variants or variants of human serum albumin (WO2011 / 051489, WO2012 / 059486, WO2012 / 150319, WO2013 / 135896, WO2014 / 072481, WO2013. In addition to the fusion of (see / 075066) or their domains, it involves the fusion of the constant region of albumin (Fc domain) and its variants. Such variants of the Fc domain disable Fc receptor binding (eg, Fcγ receptors) to allow the desired pairing of dimers or multimers, or for other reasons. May be optimized / modified for. If the above HLE molecules consist of only one single polypeptide chain, they (i) do not require two separate expression systems, and (ii) are of high purity due to the absence of "dummy chains". Has the advantage that it can be isolated. A further concept known in the art for extending the half-life of small protein compounds in the human body is PEGylation of these compounds, such as the antibody constructs of the present invention.

In a preferred embodiment, the bispecific antibody construct according to the invention may be linked with a fusion partner (eg, protein or polypeptide or peptide, etc.), eg, for the purpose of extending the serum half-life of the construct. By peptide bond). These fusion partners are human serum albumin (HSA or HALB) plus their sequence variants, HSA-binding peptides, FcRn-binding peptides (FcRn BP), or (antibody-derived) Fc regions. Can be selected from constructs containing. Illustrative sequences of these fusion partners are shown in SEQ ID NOs: 406-421 and 437-444. Generally, the fusion partner is either directly (eg, by peptide bond) or a peptide linker, eg (GGGGS)._n(In the formula, "n" is an integer greater than or equal to 2 such as 2 or 3 or 4) and may be linked to the N-terminus or C-terminus of the bispecific antibody construct according to the invention. .. Suitable peptide linkers are set forth in SEQ ID NOs: 392-400.

The antibody construct, referred to as CDH3-13 (full-length sequence of bispecific molecule set forth in SEQ ID NO: 178), is a selection of fusion proteins or fusion peptides aimed at prolonging the serum half-life of the construct. For example, it was concatenated in-frame with SEQ ID NO: 437-444). The sequence of each of these fusion constructs is shown in SEQ ID NOs: 379-389. Sequences of these fusion partners for "naked" bispecific antibody constructs may also be linked to any other antibody construct disclosed herein (corresponding to the C-terminus or N-terminus). And in the way shown for CDH3-13).

Therefore, the bispecific antibody constructs of the present invention have improved properties, such as affinity for the polypeptide or albumin set forth in SEQ ID NO: 437, preferably human albumin or a variant thereof (eg, affinity and extension for the FcRn receptor). (Has plasma half-life, etc.), most preferably further containing albumin set forth in SEQ ID NO: 443 or 444. These moieties are preferably fused in-frame to the C-terminus of the bispecific antibody construct.

Example 15 further demonstrates the unexpected benefits associated with C-terminal fusion of albumin with the bispecific antibody constructs of the invention. Human and common marmosets (Callithrix jacchus), cotton-top tamarins (Saguinus oedipus) or common squirrel monkeys (Saimiri) For these bispecific T cell-inducing molecules containing a binding domain specific for the CD3ε chain of sciureus), the epitope is included in the group consisting of SEQ ID NO: 2, 4, 6, or 8 of WO2008 / 119567. T It was observed to show cell damage. Such high concentration problems may be associated with specific routes of administration or in combination with specific targeting and required compound concentrations. Preferred examples of such a second binding domain that binds to human CD3 on the surface of T cells are described herein below and are set out in SEQ ID NOs: 445-537. When serum albumin is fused to the C-terminus of such a bispecific construct, T cell injury is avoided, see Figure 13. Although not intended to be bound by theory, activation of T cells in the presence of high concentrations of T cell-induced bispecific antibody constructs and in the absence of target cells is an antibody by CD3 binding domain. It may be explained by dimerization or multimerization of constructs. Such dimerization or multimerization is sterically impaired by fusion of albumin or a variant thereof with the C-terminus of the antibody construct, while maintaining the characteristics of the antibody construct for its T cell-inducing mechanism of action. ..

Therefore, the preferred antibody constructs according to the invention are, in order from N-terminal to C-terminal: 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, First binding domain having an amino acid sequence selected from the group consisting of SEQ ID NO: 347 and SEQ ID NO: 357; Peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NO: 392 to 400; : 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 Sequence Second binding domain having an amino acid sequence selected from the group consisting of number: 537; and · Peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NO: 392 ~ 400; and · SEQ ID NO: 437, Contains a polypeptide having an amino acid sequence selected from the group consisting of 443 and 444.

According to another preferred embodiment, the bispecific antibody construct of the invention comprises a third domain containing two polypeptide monomers (in addition to the two binding domains), each containing hinges, CH2 and Containing the CH3 domain, the two polypeptides (or polypeptide monomers) are fused together by a peptide linker. Preferably, this third domain comprises N-terminal to C-terminal: hinge-CH2-CH3-linker-hinge-CH2-CH3. The preferred amino acid sequence of the third domain is shown in SEQ ID NOs: 414-421. Each of the polypeptide monomers is preferably selected from the group consisting of SEQ ID NOs: 406 to 413, or preferably has an amino acid sequence that is at least 90% identical to these sequences. In another preferred embodiment, the first and second binding domains of the bispecific antibody constructs of the invention are preferably SEQ ID NO: 392, rather than, for example, the group consisting of any one of SEQ ID NOs: 392-400. , 393, 395, 396, 397, 399, and 400 are fused to the third domain by a peptide linker selected from the group.

According to the present invention, the "hinge" is an IgG hinge region. This region can be identified by analogy using Kabat numbering. See Kabat at positions 223 to 243. According to the above, the minimum requirement for a "hinge" is the amino acid residue corresponding to the IgG1 sequence stretch of D231-P243 by Kabat numbering. The terms CH2 and CH3 refer to immunoglobulin heavy chain constant regions 2 and 3. These regions can also be identified by analogy using Kabat numbering. See Kabat's 244-360 positions for CH2 and Kabat's 361-478 positions for CH3. It is understood that there are some changes between immunoglobulins in terms of their 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 (eg,). Padlan, Molecular See 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. The Fc monomer may also include a flexible hinge N-terminus to these domains. For IgA and IgM, the Fc monomer may contain a J chain. For IgG, the Fc portion contains the immunoglobulin domains CH2 and CH3 as well as the hinge between the first two domains and CH2. Although the boundaries of the Fc moiety of immunoglobulins may vary, examples of human IgG heavy chain Fc moieties containing functional hinges, CH2 and CH3 domains, for example, for IgG4, are each residue D231 (of the hinge domain). It can be defined as containing ~ P476 (at the C-terminus of the CH3 domain), or D231 ~ L476, in which case the numbering follows Kabat.

Therefore, the antibody construct of the present invention is a peptide linker having an amino acid sequence selected from the group consisting of (a) first binding domain; (b) SEQ ID NO: 393, 399 and 400 in the order of N-terminal to C-terminal. (C) Second binding domain; (d) Peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NO: 392, 393, 395, 396, 397, 399, and 400; (e) Third domain ( First polypeptide monomer (including the terminus, CH2 and CH3 domains); (f) Peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 402, 403, 404 and 405; and (g) th. It may contain a second polypeptide monomer of 3 domains, including the hinge, CH2 and CH3 domains.

The antibody constructs of the present invention are arranged in the order of N-terminal to C-terminal (a) 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: Number: 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, Sequence First binding domain having an amino acid sequence selected from the group consisting of No .: 347 and SEQ ID NO: 357; (b) Peptide linker having an amino acid sequence selected from the group consisting of SEQ ID NOs: 393, 399 and 400; (c) 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: A second binding domain having an amino acid sequence selected from the group consisting of: 534 and SEQ ID NO: 537; (d) Selected from the group consisting of SEQ ID NOs: 392, 393, 395, 396, 397, 399, and 400. It is also preferable to include a peptide linker having an amino acid sequence; and (e) a third domain having an amino acid sequence selected from the group consisting of SEQ ID NOs: 414 to 421.

Thus, in a preferred embodiment, the antibody constructs of the invention are 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: Does it contain a polypeptide selected from the group consisting of number: 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? , Or consists of it.

See also Example 15 showing its antitumor activity in one of the above Fc constructs and in a mouse xenograft model.

Covalent modifications of antibody constructs are also included within the scope of the invention and are generally, but not necessarily, post-translational. For example, some types of covalent modifications of antibody constructs react with organic derivatizers capable of reacting specific amino acid residues of antibody constructs with selected side chains or N-terminal or C-terminal residues. It is introduced into the molecule by letting it.

Most commonly, the cystenyl residue is reacted with α-haloacetate (and the corresponding amine), such as chloroacetic acid or chloroacetamide, to give a derivative of carboxymethyl or carboxamide methyl. Cistenyl residues are also bromotrifluoroacetone, α-bromo-β- (5-imidezoyl) propionic acid, chloroacetylphosphate, N-alkylmaleimide, 3-nitro-2-pyridyldisulfide, methyl2-pyridyldisulfide, Derivatized by reaction with p-chloromerkuri benzoate, 2-chloromerkuri-4-nitrophenol, or chloro-7-nitrobenzo-2-oxa-1,3-diazol.

The histidyl residue is induced by reaction with diethylpyrocarbonate at pH 5.5-7.0 because the drug is relatively specific for the histidyl side chain. Para-bromophenacil bromide is also useful; the reaction is preferably carried out in 0.1 M sodium cacodylate at pH 6.0. Ridinyl and amino-terminal residues are reacted with succinic acid or other carboxylic acid anhydrides. Derivatization with these agents has the effect of reversing the charge of lysinyl residues. Other suitable reagents for derivatizing alpha-amino-containing residues include imide esters such as methyl picolinimide; pyridoxal phosphate; pyridoxal; chloroborohydride; trinitrobenzenesulfonic acid; O-methylisourea; 2 , 4-Pentandion; and transaminase-catalyzed reactants with glyoxyphosphate.

Arginyl residues are modified by reaction with one or several conventional reagents, among others phenylglyoxal, 2,3-butandione, 1,2-cyclohexanedione, and ninhydrin. Derivatization of arginine residues requires that the reaction be carried out under alkaline conditions due to the high pKa of the guanidine functional group. In addition, these reagents may react with the arginine epsilon amino group in addition to the lysine group.

Specific modification of the tyrosyl residue may be carried out by reaction with an aromatic diazonium compound or tetranitromethane, with particular interest in introducing the spectral label into the tyrosyl residue. Most commonly, N-acetylimidazole and tetranitromethane are used to form O-acetyltyrosyl species and 3-nitro derivatives, respectively. Tyrosine residues are suitable for preparing labeled proteins for use in the chloramine-T method of the radioimmunoassay described above.¹²⁵I or¹³¹Iodized using I.

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

Derivatization with a bifunctional agent is useful for cross-linking the antibody constructs of the invention to a water-insoluble support matrix or surface for use in a variety of methods. Commonly used cross-linking agents include, for example, 1,1-bis (diazoacetyl) -2-phenylethane, glutaraldehyde, N-hydroxysuccinimide ester, for example, an ester with 4-azidosalicylic acid, homobifunctional. Seximide esters include, for example, succinimide esters such as 3,3'-dithiobis (succinimidyl propionate), and bifunctional maleimides such as bis-N-maleimide-1,8-octane. Derivatizing agents such as methyl-3-[(p-azidophenyl) dithio] propioimidate produce photoactivated intermediates capable of forming crosslinks in the presence of light. Alternatively, in a reactive water-insoluble matrix, eg, US Pat. Nos. 3,969,287; 3,691,016; 4,195,128; 4,247,642; 4,229,537; and 4,330,440. Cyanogen bromide activated carbohydrates and reaction substrates as described are used for protein immobilization.

Glutaminyl and asparaginyl residues are frequently deamidated to the corresponding glutamil and aspartyl residues, respectively. Alternatively, these residues are deamidated under weakly acidic conditions. Any form of these residues is within the scope of the invention.

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

Another type of covalent modification of antibody constructs within the scope of the invention involves altering the glycosylation pattern of proteins. As is known in the art, the glycosylation pattern depends on both the sequence of the protein (eg, the presence or absence of specific glycosylated amino acid residues discussed below), or the host cell or organism from which the protein is produced. Can be. Specific expression systems are discussed below.

Glycosylation of polypeptides is typically either N-linked or O-linked. The N-linked type refers to the binding of the carbohydrate moiety to the asparagine residue side chain. The tripeptide sequences asparagine-X-serine and asparagine-X-threonine (where X is any amino acid except proline) are recognition sequences for the enzymatic binding of the carbohydrate moiety to the asparagine side chain. .. Therefore, the presence of any of these tripeptide sequences in a polypeptide forms a potential glycosylation site. O-linked glycosylation refers to the binding of one of the sugars N-acetylgalactosamine, galactose, or xylose to a hydroxyamino acid, most commonly serine or threonine, with 5-hydroxyproline or 5-hydroxylysine. May be used.

Addition of a glycosylation site to the antibody construct may be conveniently performed by altering the sequence so that the amino acid sequence contains one or more of the above tripeptide sequences (of the N-linked glycosylation site). .. This change may also be made by the addition or substitution of one or more serine or threonine residues to the starting sequence (at the O-linked glycosylation site). To facilitate, the amino acid sequence of the antibody construct is a DNA encoding a polypeptide with preselected bases such that changes at the DNA level produce codons that will be translated specifically into the desired amino acid. It is preferably changed by mutating.

Another means of increasing the number of carbohydrate moieties on the antibody construct is by chemical or enzyme coupling of glycosides with proteins. These procedures are advantageous in that they do not require protein production in host cells capable of glycosylation for N-linked and O-linked glycosylation. Depending on the coupling method used, the sugars (s) are (a) arginine and histidine, (b) free carboxyl groups, (c) free sulfhydryl groups such as cysteine groups, (d) serine, threonine, Alternatively, it may be attached to a free hydroxyl group such as a hydroxyproline group, an aromatic residue such as (e) a phenylalanine, tyrosine, or tryptophan group, or (f) an amide group of glutamine. These methods are described in WO 87/05330 and in Aplin and Wriston, 1981, CRC Crit. Rev. Biochem., Pp. 259-306.

Removal of the carbohydrate moiety present on the starting antibody construct may be performed chemically or enzymatically. Chemical deglycosylation requires exposure of the compound to trifluoromethanesulfonic acid, or an equivalent compound. This treatment leaves the polypeptide intact, resulting in cleavage of almost or all sugars except the sugar to which it binds (N-acetylglucosamine or N-acetylgalactosamine). Chemical deglycosylation is described by Hakimuddin et al., 1987, Arch. Biochem. Biophys. 259: 52, and Edge et al., 1981, Anal. Biochem. 118: 131. Enzymatic cleavage of the carbohydrate moiety on the polypeptide can be achieved by the use of various endoglycosidases and exoglycosidases, as described by Thotakura et al., 1987, Meth. Enzymol. 138: 350. Glycosylation at potential glycosylation sites is Duskin et. It may be prevented by the use of the compound tunicamycin, as described by al., 1982, J. Biol. Chem. 257: 3105. Tunicamycin blocks the formation of protein-N-glycosidic bonds.

Other modifications of the antibody construct are also envisioned herein. For example, another type of covalent modification of the antibody construct transfers the antibody construct to various non-proteinaceous polymers, eg, but not limited to, various polyols such as polyethylene glycol, polypropylene glycol, polyoxyalkylene, or polyethylene glycol. And polymers containing polypropylene glycol copolymers, etc., as described in US Pat. Nos. 4,640,835; 4,496,689; 4,301,144; 4,670,417; 4,791,192 or 4,179,337. Including concatenating by the methods that have been done. In addition, 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, for example, PEG.

In some embodiments, the covalent modification of the antibody construct of the invention comprises the addition of one or more labels. The labeling group may be attached to the antibody construct by spacer arms of various lengths to reduce potential steric hindrance. Various methods for labeling proteins are known in the art and can be used in practicing the present invention. The term "label" or "labeling group" refers to any detectable label. In general, labels are classified into various classes depending on the assay in which they will be detected, eg: a) isotope labels that may be radioisotopes or heavy isotopes, eg, radioisotopes. Body or radionuclide, etc. (eg,³H,¹⁴C,¹⁵N,³⁵S,⁸⁹Zr,⁹⁰Y,⁹⁹Tc,¹¹¹In,¹²⁵I,¹³¹I) b) Magnetic labels (eg, magnetic particles) c) Oxidation-reduction active moieties d) Optical dyes (including, but not limited to, chromophores, phosphors and fluorophores), eg, fluorescent groups (eg, FITC). , Rhodamine, lanthanidrin chromophore), chemical luminescent groups, and fluorophores that can be either "small molecule" or proteinaceous fluorophores e) Enzyme groups (eg, horseradish peroxidase, β-galactosidase, luciferase) , Alkaline phosphatase) f) Biochromophore g) Contains certain polypeptide epitopes recognized by secondary reporters (eg, leucine zipper pair sequences, secondary antibody binding sites, metal binding domains, epitope tags, etc.) Not limited to these.

"Fluorescent label" means any molecule that may be detected due to its unique fluorescent properties. Suitable fluorescent labels include fluorescein, rhodamine, tetramethylrhodamine, eosin, erythrosin, coumarin, methyl-coumarin, pyrene, malakite green, stillben, lucifer yellow, Cascade BlueJ, Texas Red, IAEDANS, EDANS, BODIPY FL, LC Red640. , Cy5, Cy5.5, LC Red705, Oregon Green, Alexa-Fluor Dyes (Alexa Fluor350, Alexa Fluor430, Alexa Fluor488, Alexa Fluor546, Alexa Fluor568, Alexa Fluor594, Alexa Fluor633, Alexa Fluor660, Alexa Fluor680), Cascade Blue, Cascade Yellow and R-Phicoerythrin (PE) (Molecular Probes, Eugene, Oregon), FITC, Rhodamine, and Texas Red (Pierce, Rockford, Illinois), Cy5, Cy5.5, Cy7 (Amersham Life Science, Pittsburgh, PA), but are not limited to these. Suitable optical dyes, including fluorophores, are described in Richard P. Haugland's Molecular Probes Handbook.

Suitable proteinaceous fluorescent labels include GFP (Chalfie et al., 1994, Science 263: 802-805), EGFP (Clontech Laboratories, Inc., Genbank Accession No. U55762) of Renilla, Petitrosarcus, or Equorea species. Green Fluorescent Protein, 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.Acad.Sci.USA85: 2603-2607) and Renilla (WO92 / 15673, WO95 / 07463, WO98 / 14605, WO98 / 26277, WO99 / 49019, US Pat. No. 5,292,658; Also listed: 5,418,155; 5,683,888; 5,741,668; 5,777,079; 5,804,387; 5,874,304; 5,876,995; 5,925,558). However, it is not limited to these.

Leucine zipper domains are peptides that promote the oligomerization of the proteins in which they are found. Leucine zippers were originally identified in several DNA-binding proteins (Landschulz et al., 1988, Science 240: 1759) and have since been found in a variety of different proteins. Among the known leucine zippers are naturally occurring peptides and derivatives thereof that dimerize or trimerize. Examples of leucine zipper domains suitable for making soluble oligomeric proteins are described in PCT application WO94 / 10308, where leucine zippers derived from pulmonary surfactant protein D (SPD) are popped et al., 1994, FEBS. It is described in Letters 344: 191. The use of modified leucine zippers, which allows for stable trimerization of heterologous proteins fused to leucine zippers, is described by Fanslow et. Al., 1994, Semin. Immunol. 6: 267-78. In one approach, a recombinant fusion protein containing a CDH3 antibody fragment or derivative fused to a leucine zipper peptide is expressed in a suitable host cell and the soluble oligomeric CDH3 antibody fragment or derivative formed is recovered from the culture supernatant. NS.

The antibody constructs of the invention may also contain additional domains, which are useful, for example, in isolating the molecule or are related to the adapted pharmacokinetic profile of the molecule. Domains useful for isolation of antibody constructs may be selected from peptide motifs or secondary introduction moieties, which can be captured by isolation methods, such as isolation columns. Non-limiting embodiments of such additional domains include Myc tags, HAT tags, HA tags, TAP tags, GST tags, chitin-binding domains (CBD tags), maltose-binding proteins (MBP tags), Flag tags, Strep tags. And its variants (eg, StrepII tags) and peptide motifs known as His tags. All antibody constructs disclosed herein, characterized by the identified CDRs, preferably contain a His tag domain, which is a contiguous His residue, preferably five, more in the amino acid sequence of the molecule. Preferably commonly known as a repeat of 6 His residues (hexahistidine, see SEQ ID NO: 436). The His tag may be located, for example, at the N-terminus or C-terminus of the antibody construct, preferably at the C-terminus. Most preferably, the hexahistidine tag (HHHHHH) is linked to the C-terminus of the antibody construct according to the invention by a peptide bond.

The first binding domain of the antibody construct of the present invention binds to human CDH3 on the surface of target cells. The amino acid sequence of human CDH3 is represented by SEQ ID NO: 1. The term "superficially" is understood to mean that the binding domain specifically binds to an epitope or epitope cluster contained within the CDH3 extracellular domain (CDH3 ECD) in the context of the present invention. Thus, the first binding domain according to the invention is expressed by naturally expressed cells or cell lines and / or by cells or cell lines transformed or (stable / transiently) transfected with CDH3. It preferably binds to CDH3. In a preferred embodiment, the first binding domain also binds to CDH3 when it is used as a "target" or "ligand" molecule in an in vitro binding assay, such as BIAcore or Scatchard. A "target cell" can be any prokaryotic or eukaryotic cell that expresses CDH3 on its surface; preferably the target cell is a cell that is part of the human or animal body, such as a tumor or cancer. Such as cells.

The term "CDH3 ECD" refers to the form of CDH3 that is essentially free of the transmembrane and cytoplasmic domains of CDH3. Those skilled in the art will appreciate that the transmembrane domain identified for the CDH3 polypeptide of the invention will be identified according to routinely used criteria in the art for identifying that type of hydrophobic domain. Will be understood. The exact boundaries of the transmembrane domain may vary, but are most likely to be no more than about 5 amino acids at the end of any of the domains specifically mentioned herein. Preferred human CDH3 ECDs are shown in SEQ ID NO: 3.

The affinity of the first binding domain for human CDH3 is preferably 15 nM, more preferably 10 nM, even more preferably 5 nM, even more preferably 1 nM, even more preferably 0.5 nM, even more preferably . 0.1 nM, most preferably 0.05 nM. Affinity can be measured, for example, in the BIAcore assay or in the Scatchard assay, as described in the Examples. Other methods of determining affinity are well known to those of skill in the art.

T cells or T lymphocytes are a type of lymphocyte that plays a central role in cell-mediated immunity (a type of white blood cell itself). There are several subsets of T cells, each with a different function. T cells can be distinguished from other lymphocytes, such as B cells and NK cells, by the presence of T cell receptors (TCRs) on the cell surface. The TCR is involved in the recognition of antigens bound to major histocompatibility complex (MHC) molecules and is composed of two different protein chains. In 95% of T cells, the TCR consists of alpha (α) and beta (β) chains. When TCRs are linked to antigenic peptides and MHC (peptide / MHC complexes), T lymphocytes are a series of live mediated by related enzymes, co-receptors, specialized adapter molecules, and activated or released transcription factors. Activated by chemical events.

The CD3 receptor complex is a protein complex, composed of four strands. 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 so-called ζ (zeta) chains to form the T cell receptor CD3 complex and generate activation signals in T lymphocytes. The CD3γ (gamma), CD3δ (delta), and CD3ε (epsilon) chains are highly relevant cell surface proteins of the immunoglobulin superfamily that contain a single extracellular immunoglobulin domain. The intracellular tail of the CD3 molecule contains an immunoreceptor tyrosine-based activation motif or a single conserved motif known as ITAM for short, which is essential for TCR signaling. The CD3 epsilon molecule is a polypeptide encoded by the CD3E gene located on chromosome 11 in humans. The most preferred CD3 binding epitope corresponds to amino acid residues 1-27 of the human CD3 epsilon extracellular domain.

Redirected lysis of target cells by recruitment of T cells by multispecific, at least bispecific antibody constructs, involves cytolytic synapse formation and delivery of perforins and granzymes. Linked T cells can undergo a series of target cell lysis and are unaffected by peptide antigen processing and immune avoidance mechanisms that interfere with presentation or clonal T cell differentiation. See, for example, WO2007 / 042261.

Cytotoxicity mediated by CDH3 / CD3 bispecific antibody constructs can be measured in a variety of ways. Effector cells can be, for example, stimulated concentrated (human) CD8-positive T cells or unstimulated (human) peripheral blood mononuclear cells (PBMC). If the target cell is of macaque origin, or expresses macaque CDH3, or is transfected with macaque CDH3, the effector cells should also be of macaque origin, eg, macaque T cell line, eg 4119LnPx. Is. Target cells should express CDH3, eg, human or macaque CDH3 (at least the extracellular domain). The target cell can be a cell line (such as CHO) that has been stably or transiently transfected with CDH3, eg, human or macaque CDH3. Alternatively, the target cell can be a CDH3-positive naturally expressed cell line. EC50 levels are usually expected to be lower in target cell lines that express higher levels of CDH3 on the cell surface. The effector cell to target cell (E: T) ratio is usually about 10: 1, but can fluctuate. The cytotoxic activity of CDH3 / CD3 bispecific antibody constructs can be measured in a 51 chromium release assay (approximately 18 hours incubation time) or in a FACS-based cytotoxicity assay (approximately 48 hours incubation time). It is also possible to change the incubation time of the assay (cytotoxic reaction). Other methods of measuring cell damage are well known to those skilled in the art and include MTT or MTS assays, ATP-based assays, such as bioluminescence assay, sulforhodamine B (SRB) assay, WST assay, clonogenic assay and Includes ECIS technology.

Cytotoxic activity mediated by the CDH3 / CD3 bispecific antibody constructs of the present invention is preferably measured in cell-based cytotoxicity assays. This activity may also be measured in the 51 chromium release assay. Activity is EC₅₀Expressed as a value, this corresponds to the half effect concentration (the concentration of antibody construct that induces a cytotoxic response between baseline and maximum). Preferably, the EC of the CDH3 × CD3 bispecific antibody construct₅₀The values are 5000 pg / ml or 4000 pg / ml, more preferably 3000 pg / ml or 2000 pg / ml, even more preferably 1000 pg / ml or 500 pg / ml, even more preferably 400 pg / ml or 300 pg / ml, even more preferably 200 pg / ml, even more preferably 100 pg / ml, even more preferably 50 pg / ml, even more preferably 20 pg / ml or 10 pg / ml, most preferably 5 pg. / ml.

Given EC above₅₀Values can be measured in different assays. Those skilled in the art recognize that EC50 values can be expected to be lower when stimulated / enriched CD8 + T cells are used as effector cells compared to unstimulated PBMCs. In addition, EC50 values can be expected to be lower when the target cells express a large number of target antigens compared to low target expression rats. For example, when stimulated / enriched human CD8 + T cells are used as effector cells (and when either CDH3 transfect cells such as CHO cells or CDH3-positive naturally expressed cell lines such as A431 are used as target cells). , CDH3 x CD3 bispecific antibody construct EC₅₀The values are preferably 1000 pg / ml, more preferably 500 pg / ml, even more preferably 250 pg / ml, even more preferably 100 pg / ml, even more preferably 50 pg / ml, even more preferably . 10 pg / ml, most preferably 5 pg / ml. EC of CDH3 × CD3 bispecific antibody constructs when human PBMCs are used as effector cells₅₀The value is preferably 5000 pg / ml or 4000 pg / ml (especially when the target cell is a CDH3-positive naturally expressing cell line such as A431), more preferably 2000 pg / ml (especially when the target cell is a CHO cell or the like). CDH3 transfected cells), more preferably 1000 pg / ml or 500 pg / ml, even more preferably 200 pg / ml, even more preferably 150 pg / ml, even more preferably 100 pg / ml, Most preferably 50 pg / ml or less. If a Macaku T cell line such as LnPx4119 is used as the effector cell and a Macaku CDH3 transfected cell line such as CHO cells is used as the target cell line, the EC of the CDH3 × CD3 bispecific antibody construct₅₀The values are preferably 2000 pg / ml or 1500 pg / ml, more preferably 1000 pg / ml or 500 pg / ml, even more preferably 300 pg / ml or 250 pg / ml, even more preferably 100 pg / ml. , Most preferably 50 pg / ml.

Preferably, the CDH3 / CD3 bispecific antibody constructs of the invention do not induce / mediate lysis of CDH3-negative cells such as CHO cells, or essentially do not induce / mediate lysis. The terms "do not induce lysis", "essentially do not induce lysis", "do not mediate lysis" or "do not essentially mediate lysis" mean 100% lysis of CDH3-positive cell lines such as A431. By setting, the antibody constructs of the invention do not induce or mediate lysis of more than 30% of CDH3-negative cells, preferably not greater than 20%, more preferably not greater than 10%, particularly preferably 9%. Means not to exceed 8%, 7%, 6% or 5%. This is typically applied to antibody construct concentrations up to 500 nM. Those of skill in the art know how to measure cytolysis with less effort. In addition, the present specification teaches specific instructions on how to measure cytolysis.

The difference in cytotoxic activity between the monomeric and dimeric isoforms of individual CDH3 / CD3 bispecific antibody constructs is referred to as the "potency gap." This potency gap is, for example, EC in the monomeric form of the molecule.₅₀EC of value and dimer form₅₀It can be calculated as a ratio to a value. The potency gap of the CDH3 / CD3 bispecific antibody construct of the present invention is preferably 5, more preferably 4, even more preferably 3, even more preferably 2, and most preferably 1. As an example, the potency gap of CDH3-13 was determined to be 0.5, the potency gap of CDH3-13 × CD3-HLE (Fc) was determined to be 0.9, and the potency gap of CDH3-25 was 0.7. The efficacy gap for CDH3-25 x CD3-HALB was also determined to be 0.7.

The first and / or second (or any additional) binding domains (s) of the antibody constructs of the invention are preferably interspecific for members of the primate mammalian order. Interspecific CD3 binding domains are described, for example, in WO2008 / 119567. According to one embodiment, the first and / or second binding domains, in addition to binding to human CDH3 and human CD3, respectively, include new world primates (such as common marmosets, cotton boushi tamarins or common squirrel monkeys), and the Old World. It will also bind to primates CDH3 / CD3, including (but not limited to) primates (such as hihi and macaques), tamarins, and non-human apes.

In one aspect of the invention, the first binding domain binds to human CDH3 and more preferably to macaque CDH3, such as cynomolgus monkey CDH3 (SEQ ID NO: 5), more preferably macaque CDH3 ECD. The affinity of the first binding domain for macaque CDH3 is preferably 15nM, more preferably 10nM, even more preferably 5nM, even more preferably 1nM, even more preferably 0.5nM, even more preferably 0.1 nM, most preferably 0.05 nM or even 0.01 nM.

Preferably, the affinity gap of the antibody construct according to the invention for binding macaque CDH3 to human CDH3 [maCDH3: huCDH3] (eg, as determined by BiaCore or by Scatchard analysis, Examples). (See) is 0.1 to 10, more preferably 0.2 to 5, even more preferably 0.3 to 2.5, even more preferably 0.4 to 2, and most preferably 0.5 to 1.

In one embodiment of the antibody construct of the invention, the second binding domain binds to CD3 epsilon of human and common marmosets, cotton-top tamarins or common squirrel monkeys. Preferably, the second binding domain binds to the extracellular epitope of these CD3 epsilon chains. The second binding domain is also expected to bind to extracellular epitopes of human and macaque CD3 epsilon chains. The most preferred epitope of CD3 epsilon is contained within amino acid residues 1-27 of the human CD3 epsilon extracellular domain. More specifically, the epitope comprises at least the amino acid sequence Gln-Asp-Gly-Asn-Glu. Common marmosets and cotton-top tamarins are both New World primates belonging to the family of Callitrichidae, while common squirrel monkeys are New World primates belonging to the family Cebidae.

The second binding domain that binds to human CD3 on the surface of T cells is (a) CDR-L1 shown in SEQ ID NO: 27 of WO2008 / 119567, CDR-L2 shown in SEQ ID NO: 28 of WO2008 / 119567, and CDR-L3 shown in WO2008 / 119567 SEQ ID NO: 29; (b) CDR-L1 shown in WO2008 / 119567 SEQ ID NO: 117, CDR-L2 and WO2008 / shown in WO2008 / 119567 SEQ ID NO: 118. SEQ ID NO: 119567: CDR-L3 shown in 119; and (c) CDR-L1 shown in WO2008 / 119567 SEQ ID NO: 153, CDR-L2 and WO2008 / 119567 shown in WO2008 / 119567 SEQ ID NO: 154. It is particularly preferred for the antibody constructs of the invention to include a VL region containing CDR-L1, CDR-L2 and CDR-L3 selected from CDR-L3 set forth in SEQ ID NO: 155.

In an alternative preferred embodiment of the antibody construct of the invention, the second binding domain that binds to human CD3 on the surface of T cells is (a) CDR-H1, WO2008 / set forth in SEQ ID NO: 12 of WO2008 / 119567. SEQ ID NO: 119567, CDR-H2 and WO2008 / 119567, SEQ ID NO: 14; (b) WO2008 / 119567, SEQ ID NO: 30: CDR-H1, WO2008 / 119567 SEQ ID NO: 31 for CDR-H2 and WO2008 / 119567 SEQ ID NO: 32; (c) WO2008 / 119567 SEQ ID NO: 48 for CDR-H1, WO2008 / 119567 SEQ ID NO: CDR-H2 and WO2008 / 119567 shown in: 49: CDR-H3 shown in 50; (d) CDR-H1 shown in WO2008 / 119567: SEQ ID NO: 67 CDR-H2 and WO2008 / 119567 shown in SEQ ID NO: 68; (e) WO2008 / 119567 SEQ ID NO: 84 shown in CDR-H1 and WO2008 / 119567 SEQ ID NO: 85. CDR-H2 and WO2008 / 119567 SEQ ID NO: 86, CDR-H3; (f) WO2008 / 119567 SEQ ID NO: 102, CDR-H1, WO2008 / 119567 SEQ ID NO: 103. -CDR-H3 shown in SEQ ID NO: 104 of H2 and WO2008 / 119567; (g) CDR-H1 shown in SEQ ID NO: 120 of WO2008 / 119567, CDR-H2 shown in SEQ ID NO: 121 of WO2008 / 119567 And WO2008 / 119567 SEQ ID NO:: CDR-H3 shown in 122; (h) WO2008 / 119567 SEQ ID NO:: CDR-H1 shown in 138, WO2008 / 119567 SEQ ID NO::SEQ ID NO: CDR-H2 and WO2008 / 119567 shown in 139: CDR-H3 shown in 140; (i) SEQ ID NO: CDR-H1 shown in WO2008 / 119567: SEQ ID NO: 157 CDR-H2 and WO2008 / 119567 shown in SEQ ID NO: 158; and (j) WO2008 / 119567 SEQ ID NO: 174 shown in CDR-H1, WO2008 / 119567 SEQ ID NO: 175. Contains the VH region containing CDR-H1, CDR-H2 and CDR-H3 selected from CDR-H3 shown in SEQ ID NO: 176 of CDR-H2 and WO2008 / 119567.

The second binding domain that binds to human CD3 on the surface of T cells is selected from the group consisting of the VL regions set forth in WO2008 / 119567 SEQ ID NO: 35, 39, 125, 129, 161 or 165. It is more preferred for the antibody constructs of the present invention to include.

The second binding domain that binds to human CD3 on the surface of T cells is WO2008 / 119567 SEQ ID NO: 15, 19, 33, 37, 51, 55, 69, 73, 87, 91, 105, 109, 123, Instead, it is preferred to include a VH region selected from the group consisting of the VH regions shown in 127, 141, 145, 159, 163, 177 or 181.

More preferably, the antibody constructs of the invention are (a) the VL region set forth in WO2008 / 119567 SEQ ID NO: 17 or 21 and the VH region set forth in WO2008 / 119567 SEQ ID NO: 15 or 19; (b) WO2008. SEQ ID NO: 35 or 39 of / 119567 and VH region shown by SEQ ID NO: 33 or 37 of WO2008 / 119567; (c) VL region and VL region shown by SEQ ID NO: 53 or 57 of WO2008 / 119567. WO2008 / 119567 SEQ ID NO: 51 or 55 VH region; (d) WO2008 / 119567 SEQ ID NO: 71 or 75 VL region and WO2008 / 119567 SEQ ID NO: 69 or 73 VH region (e) WO2008 / 119567 SEQ ID NO: 89 or 93 in the VL region and WO2008 / 119567 SEQ ID NO: 87 or 91 in the VH region; (f) WO2008 / 119567 SEQ ID NO: 107 or 111 VL region shown and SEQ ID NO: WO2008 / 119567: VH region shown in 105 or 109; (g) SEQ ID NO: 125 or 129 and SEQ ID NO: VL region and WO2008 / 119567: 123 or 127 VH region shown in; (h) WO2008 / 119567 SEQ ID NO: 143 or 147 in VL region and WO2008 / 119567 SEQ ID NO: 141 or 145 in VH region; (i) WO2008 / 119567 SEQ ID NO: : VL region shown in 161 or 165 and SEQ ID NO: WO2008 / 119567: VH region shown in 159 or 163; and (j) SEQ ID NO: 179 or 183 in VL region and WO2008 / 119567. SEQ ID NO:It features a second binding domain that binds to human CD3 on the surface of T cells, including a VL region and a VH region selected from the group consisting of the VH regions shown in 177 or 181.

The above binding domains that bind to human CD3 and are disclosed in WO2008 / 119567 are also set forth in this SEQ ID NO: 445-537.

According to a preferred embodiment of the antibody construct of the present invention, the binding domain and in particular the second binding domain (which binds to human CD3 on the surface of T cells) has the following form: a pair of VH and VL regions: , Is in the form of a single chain antibody (scFv). The VH and VL regions are arranged in the order of VH-VL or VL-VH. The VH region is preferably located at the N-terminus of the linker sequence and the VL region is preferably located at the C-terminus of the linker sequence.

Preferred embodiments of the antibody constructs of the invention are WO2008 / 119567 SEQ ID NOs: 23, 25, 41, 43, 59, 61, 77, 79, 95, 97, 113, 115, 131, 133, 149, 151. It is characterized by a second binding domain that binds to human CD3 on the surface of T cells, which comprises an amino acid sequence selected from the group consisting of, 167, 169, 185 or 187.

In one embodiment of the invention, the antibody construct is 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:: It has an amino acid sequence selected from the group consisting of these sequences set forth in 228, SEQ ID NO: 238, and SEQ ID NO: 248.

In another embodiment of the invention, the antibody construct is 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: It has an amino acid sequence selected from the group consisting of these sequences shown in: 386, SEQ ID NO: 387, SEQ ID NO: 388 and SEQ ID NO: 389.

In another embodiment of the invention, the antibody construct is 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: It has an amino acid sequence selected from the group consisting of these sequences shown in number: 358.

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

The term "amino acid" or "amino acid residue" refers to an amino acid having a definition recognized in the art, such as alanin (Ala or A); arginine (Arg or R); aspartic acid (Asn or N); Aspartic acid (Asp or D); Cysteine (Cys or C); Glutamine (Gln or Q); Glutamic acid (Glu or E); Glycin (Gly or G); Histidine (His or H); Isoleucine (He or I); Leucine (Leu or L); Lysine (Lys or K); Methionin (Met or M); Phenylalanine (Phe or F); Proline (Pro or P); Serin (Ser or S); Treonin (Thr or T); Tryptophan (Trp or W); typically refers to amino acids selected from the group consisting of tyrosine (Tyr or Y); and valine (Val or V), but if modified, synthesized, or rare amino acids are desired. May be used. In general, amino acids are non-polar side chains (eg Ala, Cys, He, Leu, Met, Phe, Pro, Val); negatively charged side chains (eg Asp, Glu); positively charged side chains (eg Arg, His, Lys); or can be grouped into those with uncharged polar side chains (eg, Asn, Cys, Gln, Gly, His, Met, Phe, Ser, Thr, Trp, and Tyr).

Amino acid modifications include, for example, deletion of residues in the amino acid sequence of the antibody construct and / or insertion into residues, and / or substitution of residues. Any combination of deletions, insertions, and substitutions is made to reach the final construct, provided that the final construct has the desired characteristics. Amino acid changes, such as changes in the number or position of glycosylation sites, may also alter the post-translational process of antibody constructs.

For example, 1, 2, 3, 4, 5, or 6 amino acids may be inserted or deleted in each of the CDRs (depending on their length, of course), but 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 of the FRs. Preferably, the amino acid sequence inserts are amino and / / in the length range from 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10 residues to a polypeptide containing 100 or more residues. Alternatively, carboxyl-terminal fusion, plus intra-sequence insertion of single or multiple amino acid residues can be mentioned. Insertion variants of the antibody constructs of the invention include fusions of the antibody constructs of the enzyme to the N-terminus or C-terminus or to polypeptides that increase the serum half-life of the antibody constructs.

The sites of greatest concern for induction of substitution mutations include CDRs of heavy and / or light chains, especially hypervariable regions, but FR changes in heavy and / or light chains are also envisioned. The substitution is preferably a conservative substitution as described herein. Preferably, 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acids may be substituted in the CDR, depending on the length of the CDR or FR, but 1, 2, 3, 4 , 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, or 25 amino acids may be substituted in the framework region (FR). For example, if the CDR sequence contains 6 amino acids, it is assumed that one, two or three of these amino acids will be replaced. Similarly, if the CDR sequence contains 15 amino acids, it is assumed that one, two, three, four, five or six of these amino acids will be replaced.

A useful method for identifying specific residues or regions of antibody constructs at preferred positions for mutagenesis is "alanine scanning mutagenesis" as described by Cunningham and Wells in Science, 244: 1081-1085 (1989). Is called. The method identifies a target residue residue or group of target residues within the antibody construct (eg, charged residues such as arg, asp, his, lys, and glu) and is neutral or charged. It is replaced by an amino acid (most preferably alanine or polyalanine) and affects the interaction of the amino acid with the epitope.

These amino acid positions that demonstrate functional susceptibility to substitutions are then modified by introducing additional or other variants at or at the site of substitution. Therefore, the site or region into which the amino acid sequence change is introduced is determined in advance, but the nature of the mutation itself does not need to be determined in advance. For example, alanine scanning or random mutagenesis may be performed at the target codon or region to analyze or optimize the performance of the mutation at a given site, and the antibody construct variant expressed will be optimal for the desired activity. The combination is screened. Techniques for making substitution mutations at predetermined sites in DNA having a known sequence are well known, such as M13 primer mutagenesis and PCR mutagenesis. Screening for variants is performed using an assay for antigen binding activity such as CDH3 or CD3 binding.

In general, when an amino acid is replaced by one or all of the heavy and / or light chain CDRs, the "substituted" sequence obtained at that time is at least 60% of the "original" CDR sequence. It is more preferably 65%, even more preferably 70%, particularly preferably 75%, and even more preferably 80% identical. This means that how much the sequence is identical to the "replaced" sequence depends on the length of the CDR. For example, a CDR with 5 amino acids is preferably 80% identical to its substituted sequence because it has at least one substituted amino acid. Thus, the CDRs of antibody constructs may have a different degree of identity with their substituted sequences, for example, CDRL1 may have 80%, while CDRL3 may have 90%.

A preferred substitution (or substitution) is a conservative substitution. However, any substitutions (including non-conservative substitutions or one or more of the "exemplary substitutions" listed in Table 1 below) will cause the antibody construct to bind to the first binding domain and the second. As long as it retains its ability to bind CD3 or CD3 epsilon by the binding domain, and / or its CDR has identity with the sequence to be substituted at that time (at least 60% of the "original" CDR sequence, more preferably. Is assumed to be 65%, even more preferably 70%, particularly preferably 75%, and even more preferably 80% identical).

Conservative substitutions are shown in Table 1 under the title "Preferable substitutions". If such substitutions result in changes in biological activity, more substantive changes referred to in Table 1 as "exemplary substitutions" or as further described below in relation to the amino acid class. It may be introduced and the product may be screened for the desired characteristics.

(Table 1) Amino acid substitution<img file="JP6907124B2_D0001.tif" />

Substantial modifications in the biological properties of the antibody constructs of the invention are: (a) the structure of the polypeptide skeleton in the region of substitution, eg, a sheet or helix conformation, (b) the charge or hydrophobicity of the molecule at the target site. It is done by selecting substitutions that differ significantly in their effect on sex, or (c) maintenance of side chain bulk. Naturally occurring residues are based on common side chain properties: (1) Hydrophobicity: Norleucine, met, ala, val, leu, ile; (2) Neutral hydrophilicity: cys, ser, thr; (3) Acidity: asp, glu; (4) Basicity: asn, gin, his, lys, arg; (5) Residues affecting chain orientation: gly, pro; and (6) Aroma: trp, It is classified into tyr and phe.

Non-conservative permutations would require exchanging one member of one of these classes for another. Any cysteine residue that is not involved in maintaining the proper conformation of the antibody construct may generally be replaced with serine to improve the oxidative stability of the molecule and prevent abnormal cross-linking. Conversely, a cysteine bond (s) may be added to the antibody to improve its stability (especially if the antibody is an antibody fragment, such as an Fv fragment).

For amino acid sequences, sequence identity and / or sequence similarity is a standard technique known in the art, such as Smith and Waterman, 1981, Adv.Appl.Math.2: 482 Local Sequence Identity Algorithm, Needleman. And Wunsch, 1970, J.Mol.Biol.48: 443 Sequence Identity Alignment Algorithm, Pearson and Lipman, 1988, Proc.Nat.Acad.Sci.USA 85:2444 Search for Similar Methods, These Algorithms (Wisconsin Genetics) Software Package, Genetics Computer Group, Science Drive 575, Computerized Implementation of GAP, BESTFIT, FASTA, and TFASTA) in Madison, Wisconsin, as described by Devereux et al., 1984, Nucl. Acid Res. 12: 387-395. Best Determined by using techniques including, but not limited to, Fit sequence programs, preferably using default settings or by inspection. Preferably, the percent identity is calculated by FastDB based on the following parameters: mismatch penalty = 1; gap penalty = 1; gap size penalty = 0.33; and binding penalty = 30, "Current Methods in Sequence Comparison and Analysis". , Macromolecule Sequencing and Synthesis, Selected Methods and Applications, pp127-149 (1988), Alan R. Liss, Inc.

An example of a useful algorithm is PILEUP. PILEUP uses progressive pairwise alignment to create multiple sequence alignments from a set of related sequences. PILEUP can also plot a tree showing the clustering relationships used to create the alignment. PILEUP uses a simplification of the progressive alignment method of Feng &amp; Doolittle, 1987, J.Mol.E vol.35: 351-360. This method is similar to the method described by Higgins and Sharp, 1989, CABIOS 5: 151-153. Useful PILEUP parameters include default gap weight = 3.00, default gap length weight = 0.10, and weighted end gap.

Other examples of useful algorithms are 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.USA90: 5873-5787, BLAST algorithm. A particularly useful BLAST program is the WU-BLAST-2 program obtained from Altschul et al., 1996, Methods in Enzymology 266: 460-480. WU-BLAST-2 uses several search parameters, most of which are set to default values. Adjustable parameters are set with the following values: Overlap span = 1, Overlap fraction = 0.125, Word threshold (T) = II. HSP S-parameters and HSP Although the S2 parameter is a dynamic value, it is established by the program itself depending on the composition of the particular sequence and composition in the particular database in which the sequence of interest is being searched; May be adjusted.

A more useful algorithm is gapped BLAST as reported by Altschul et al., 1993, Nucl. Acids Res. 25: 3389-3402. BLAST with gaps, BLOSUM-62 substitution score; threshold T parameter set to 9; 2-hit method to initiate gapless elongation, load 10 + k cost on gap length k; set to 16. Use Xu, as well as Xg set to 40 at the database search stage and 67 at the algorithm output stage. Gap alignment is initiated by a score corresponding to approximately 22 bits.

In general, amino acid homology, similarity, or identity between individual variant CDRs is at least 60% relative to the sequences shown herein, and more typically at least 65 homology or identity. % Or 70%, more preferably at least 75% or 80%, even more preferably at least 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, It is preferably increased to 99%, and almost 100%. In a similar manner, the "% (%) nucleic acid sequence identity" for the nucleic acid sequence of the binding protein identified herein is the same as the nucleotide residue in the coding sequence of the antibody construct, the nucleotide residue in the candidate sequence. Defined as the percentage of the group. The specific method utilizes the BLASTN module of WU-BLAST-2 set to the default parameters and sets the overlap span and overlap fraction to 1 and 0.125, respectively.

In general, the nucleic acid sequence homology, similarity, or identity between the nucleotide sequences encoding the individual variant CDRs and the nucleotide sequences shown herein is at least 60%, more typically homology. Gender or identity 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 preferably almost 100%. Thus, a "variant CDR" is one that has a particular homology, similarity, or identity to the parent CDR of the invention and is at least 60%, 65%, of the specificity and / or activity of the parent CDR. 70%, 75%, 80%, 81%, 82%, 83%, 84%, 85%, 86%, 87%, 88%, 89%, 90%, 91%, 92%, 93%, 94% , 95%, 96%, 97%, 98%, or 99%, but share biological functions not limited to these.

In one embodiment, the bispecific antibody constructs of the invention exhibit high monomer yields under standard research scale conditions, eg, in a standard two-step purification process. Preferably, the monomer yield of the antibody construct according to the invention is 0.25 mg / L supernatant, more preferably 0.5 mg / L, even more preferably 1 mg / L, most preferably 3 mg / L supernatant. Is.

Similarly, the yield of the dimeric antibody construct isoform, and thus the monomer percentage of the antibody construct (ie, monomer: (monomer + dimer)) can be determined. The productivity of monomeric and dimeric antibody constructs and the calculated monomeric percentage can be obtained, for example, by the SEC purification step of the culture supernatant from standardized research scale production in roller bottles. In one embodiment, the monomer percentage of the antibody construct is 80%, more preferably 85%, even more preferably 90%, most preferably 95%.

In a further embodiment, the percentage of identity of the antibody construct according to the invention to the human germline is 70% or 75%, more preferably 80% or 85%, even more preferably 90%, most preferably. Preferably 95%. See Example 7. Identity to human antibody germline gene products is believed to be an important feature for therapeutic proteins to reduce the risk of inducing an immune response to a drug within a patient during treatment. Hwang &amp; Foote ("Immunogenicity of engineered antibodies"; Methods 36 (2005) 3-10) demonstrate that reduction of the non-human portion of the drug antibody construct results in a reduced risk of inducing anti-drug antibodies within the patient during treatment. By comparing a vast number of clinically evaluated antibody drugs and their immunogenicity data, the humanization of the V region of the antibody is an antibody with an unchanged non-human V region (23.59 in patients on average). %) Tends to reduce protein immunogenicity (on average 5.1% of patients). Therefore, a higher degree of identity to the human sequence is desirable for V-region-based protein therapeutics in the form of antibody constructs. For this purpose of determining germline identity, the V region of VL is a human germline V segment and J segment (http://vbase.mrc-cpe.cam.ac.ac) using Vector NTI software. It can be aligned with the amino acid sequence of .uk /), and the amino acid sequence can be calculated as a percentage by dividing the same amino acid residue by the total number of amino acid residues in VL. Also for the VH segment (http://vbase.mrc-cpe.cam.ac.uk/), VH CDR3 can be the same, except that it may be excluded due to its high diversity and lack of existing human germline VH CDR3 alignment partners. Recombinant techniques can then be used to increase sequence identity to human antibody germline genes.

In one embodiment, the antibody construct has a favorable plasma stability of 5 (ratio of EC50 to EC50 in the absence of plasma), more preferably 4 or 3.5, even more preferably. Is 3 or 2.5, most preferably 2 or 1.5 or 1. Plasma stability of antibody constructs can be tested by construct incubation in human plasma at 37 ° C for 24 hours, followed by EC50 determination in a 51 chromium-releasing cytotoxic assay. The effector cells in the cytotoxic assay can be stimulus-enriched human CD8-positive T cells. The target cell can be, for example, a human CDH3 transfected CHO cell. The effector cell to target cell (E: T) ratio can be selected as 10: 1. The human plasma pool used for this purpose is obtained from healthy donor blood collected by an EDTA-coated syringe. Cellular components are removed by centrifugation and the upper plasma phase is collected and then pooled. As a control, the antibody construct is diluted in RPMI-1640 medium immediately prior to the cytotoxicity assay. Plasma stability is calculated as the ratio of EC50 (after plasma incubation) to EC50 (control). See Example 11.

The conversion of the antibody construct of the present invention from a monomer to a dimer is preferably low. Conversion rates can be measured under different conditions and analyzed by high performance size exclusion chromatography. For example, incubation of the monomeric isoforms of the antibody construct can be performed in an incubator at 37 ° C. for 7 days, eg, at a concentration of 100 μg / ml or 250 μg / ml. Under these conditions, the antibody constructs of the invention preferably exhibit a dimer percentage of 5%, more preferably 4%, even more preferably 3%, even more preferably 2.5%. Even more preferably 2%, even more preferably 1.5%, and most preferably 1%. See Example 9.

It is also preferred that the bispecific antibody constructs of the invention are present at very low dimeric conversion rates after multiple freeze / thaw cycles. For example, antibody construct monomers are adjusted to a concentration of 250 μg / ml in common formulation buffer and 3 freeze / thaw cycles (freeze at -80 ° C for 30 minutes, followed by 30 minutes at room temperature). Thaw) followed by high performance SEC to determine the percentage of initial monomeric antibody construct, which was converted to dimeric antibody construct. Preferably, the dimer percentage of the bispecific antibody construct is, for example, 5% after 3 freeze / thaw cycles, more preferably 4%, even more preferably 3%, even more preferably. Is 2.5%, even more preferably 2%, even more preferably 1.5%, and most preferably 1%.

The bispecific antibody constructs of the present invention preferably exhibit good thermal stability at aggregation temperatures greater than 50 ° C or greater than 52 ° C, more preferably greater than 54 ° C or greater than 55 ° C, and even more preferably. Is greater than 56 ° C or greater than 57 ° C, most preferably greater than 58 ° C or greater than 59 ° C. Thermal stability parameters can be determined with respect to antibody aggregation temperature as follows: The antibody solution at a concentration of 250 μg / ml is transferred into a single-use cuvette and placed in a dynamic light scattering device. The sample is heated to 40 ° C to 70 ° C at a heating rate of 0.5 ° C / min, and the measured radius is obtained constantly. The increase in radius indicating protein melting and aggregation is used to calculate antibody aggregation temperature. See Example 10.

Alternatively, the temperature melting curve can be determined by differential scanning calorimetry (DSC) to determine the inherent biophysical protein stability of the antibody construct. These experiments are performed using MicroCal LLC (Northampton, Mass., USA) VP-DSC devices. The energy uptake of the sample containing the antibody construct is recorded from 20 ° C to 90 ° C compared to the sample containing only the product buffer. The antibody construct is adjusted to a final concentration of 250 μg / ml, for example, in SEC running buffer. The overall sample temperature is raised stepwise to record each melting curve. At each temperature T, record the energy uptake of the sample and formulation buffer references. The difference in energy uptake Cp (kcal / mole / ° C) minus the reference from the sample is plotted for each temperature. Melting temperature is defined as the temperature at which energy uptake is initially maximized.

It is further envisioned that the CDH3 x CD3 bispecific antibodies of the invention do not cross-react (ie, do not bind) with CDH1, CDH2, CDH4, and CDH5 of the human CDH3 paralog. Furthermore, it is assumed that the CDH3 × CD3 bispecific antibodies of the invention do not cross-react (ie, do not bind) with CDH1, CDH2, CDH4, and CDH5 of the macaque / cynoCDH3 paralog. See Example 6.

The CDH3 × CD3 bispecific antibodies of the invention are also measured by OD340 after turbidity of 0.1, most preferably 0.05 (purified monomeric antibody concentrated to 2.5 mg / ml and incubated overnight). If), it is also assumed to have. See Example 12.

In a further embodiment, the antibody construct according to the invention is stable at acidic pH. The more resistant the antibody construct is to a non-physiological pH, such as pH 5.5 (eg, the pH required to perform cation exchange chromatography), the more the ion exchange column with respect to the total amount of loaded protein. The recovery rate of antibody constructs eluted from is high. The recovery of antibody constructs from an ion (eg, cation) exchange column at pH 5.5 is preferably 30%, more preferably 40%, more preferably 50%, even more preferably 60%. , Even more preferably 70%, even more preferably 80%, and most preferably 90%.

It is further envisioned that the bispecific antibody constructs of the present invention exhibit therapeutic efficacy or antitumor activity. This can be evaluated, for example, in the test disclosed in Example 14.

Those skilled in the art will provide specific parameters for this study, such as the number of tumor cells injected, the site of injection, the number of human T cells transplanted, the amount of bispecific antibody constructs administered, and the timeline. We still know how to achieve meaningful and reproducible results while modifying or adapting. Preferably, the tumor growth inhibitory T / C [%] is 70 or 60, more preferably 50 or 40, even more preferably 30 or 20, most preferably 10 or 5 or even 0. It is 2.5.

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

A polynucleotide is a biopolymer composed of 13 or more nucleotide monomers covalently linked within a chain. DNA (such as cDNA) and RNA (such as mRNA) are examples of polynucleotides with different biological functions. Nucleotides are organic molecules that function as monomers or subunits of nucleic acid molecules such as DNA or RNA. Nucleic acid molecules or polynucleotides can be double-stranded and single-stranded, linear and cyclic. It is preferably contained in the vector contained in the host cell. Host cells can express antibody constructs, for example, after transformation or transfection with the vectors or polynucleotides of the invention. To that end, the polynucleotide or nucleic acid molecule is functionally linked to the control sequence.

A genetic code is a set of rules that translates information encoded in a genetic substance (nucleic acid) into a protein. Biological decoding in living cells is accomplished by ribosomes that carry amino acids using tRNA molecules, read three nucleotides of mRNA at a time, and link the amino acids in the order specified by the mRNA. The code defines how the sequence of these three nucleotides, called codons, specifies the amino acids that will then be added during protein synthesis. With a few exceptions, the 3-nucleotide codon in the nucleic acid sequence specifies a single amino acid. This particular code is often referred to as the canonical or standard genetic code, as most genes are coded in exactly the same code. While the genetic code determines the protein sequence of a given coding region, other genomic regions can influence when and where these proteins are produced.

Furthermore, the present invention provides a vector containing the polynucleotide / nucleic acid molecule of the present invention.

Vectors are nucleic acid molecules used as vehicles to transport (foreign) genetic material into cells. The term "vector" includes, but is not limited to, plasmids, viruses, cosmids and artificial chromosomes. In general, genetically engineered vectors include origins of replication, multicloning sites and selectable markers. The vector itself is generally a nucleotide sequence, usually a DNA sequence, that contains an insert (transgene) and a larger sequence that acts as the "skeleton" of the vector. Recent vectors may include additional features: promoters, genetic markers, antibiotic resistance, reporter genes, targeting sequences, protein purification tags, in addition to transgene inserts and skeletons. A vector called an expression vector (expression construct) is specifically for expressing a transgene in a target cell and generally has a control sequence.

The term "regulatory sequence" refers to the DNA sequence required to express a functionally linked coding sequence in a particular host organism. Suitable control sequences for prokaryotes include, for example, promoters, optionally operator sequences, and ribosome binding sites. Eukaryotic cells are known to utilize promoters, polyadenylation signals, and enhancers.

Nucleic acids are "functionally linked" when placed in a functional relationship with another nucleic acid sequence. For example, the DNA of a pre-sequence or secretory leader is functionally linked to the DNA of a polypeptide when expressed as a pre-protein involved in the secretion of the polypeptide; a promoter or enhancer affects the transcription of that sequence. Is it functionally linked to the coding sequence when exerting; or the ribosome binding site is functionally linked to the coding sequence when arranged to facilitate translation. In general, "functionally linked" means that the linked DNA sequences are contiguous, contiguous in the case of a secretory leader, and in the leading phase. However, the enhancers do not have to be continuous. The connection is made by ligation at a convenient restricted site. In the absence of such sites, synthetic oligonucleotide adapters or linkers are used according to conventional practice.

"Transfection" is a method of intentionally introducing a nucleic acid molecule or polynucleotide (including a vector) into a target cell. This term is most often used for non-viral methods in eukaryotic cells. Transduction is often used to account for virus-mediated transfer of nucleic acid molecules or polynucleotides. Transfection of animal cells typically involves transient pore or "hole" openings in the cell membrane, allowing uptake of material. Transfection uses calcium phosphate, by electroporation, by cell squeezing, or by mixing cationic lipids with substances to make liposomes, which are fused with cell membranes to accumulate their cargo inside. Can be implemented by

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 a genetic alteration of a bacterial or non-animal eukaryotic cell due to direct uptake from its surroundings by the cell membrane (s), followed by integration of an exogenous genetic material (nucleic acid molecule). Transformation can be done by artificial means. In order to carry out transformation, the cells or bacteria must be in a competent state, which can occur as a timed response to environmental conditions such as starvation and cell density.

In addition, the invention provides host cells transformed with the polynucleotide / nucleic acid molecule or vector of the invention or transfected with the polynucleotide / nucleic acid molecule or vector of the invention.

As used herein, the terms "host cell" or "recipient cell" are recipients of vectors, exogenous nucleic acid molecules, and polynucleotides encoding antibody constructs of the invention; and / or antibody constructs. It is intended to include any individual cell or cell culture that can or was a recipient of its own. Each substance is introduced into cells by transformation, transfection, or the like. The term "host cell" is also intended to include single-cell progeny or potential progeny. Such progeny are actually with the parent cell, as in subsequent generations, certain changes may occur due to either spontaneous, accidental, or intentional mutations, or due to environmental effects. It may not be exactly the same (in morphology, or in the genome or whole DNA complement), but is still within the scope of this term as it is 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 mice, rats, etc. Examples include, but are not limited to, macaques or humans.

The antibody constructs of the present invention can be produced by bacteria. After expression, the antibody constructs of the invention can be isolated from E. coli cell paste in soluble fractions and purified, for example, by affinity chromatography and / or size exclusion. Final purification can be performed, for example, in the same manner as the process for purifying antibodies expressed in CHO cells.

In addition to prokaryotes, eukaryotic microorganisms such as filamentous fungi or yeast are suitable cloning or expression hosts for the antibody constructs of the invention. Saccharomyces cerevisiae, or common baker's yeast, is the most commonly used lower eukaryotic host microorganism. However, a number of other genera, species, and strains are generally available and are useful herein, eg, Schizosaccharomyces pombe, Kluyveromyces hosts, eg. K. lactis, K. fragilis (ATCC 12424), K. bulgaricus (ATCC 16045), K. wickeramii (ATCC 24178), K. waltii (ATCC 56500), K. drosophilarum (ATCC) 36906), K. thermotolerans, and K. marxianus, etc .; Yarrowia (EP402226); Pichia pastoris (EP183070); Candida; Trichoderma reesia ) (EP244234); Neurospora crassa; the genus Schwanniomyces, such as Schwanniomyces occidentalis; And Aspergillus hosts, such as A. nidulans and A. niger.

Suitable host cells for the expression of the glycosylated antibody constructs of the present invention are obtained from multicellular organisms. Examples of invertebrate cells include plant cells and insect cells. Corresponding from numerous baculovirus strains and variants as well as hosts such as Aedes aegypti, Aedes aegypti, Aedes albopictus, Drosophila melanogaster, and Bombyx mori. Aedes albopictus has been identified. Various viral strains for transfection, such as the L-1 variant of Spodoptera frugiperda (Autographa californica) NPV and the Bm-5 strain of Spodoptera frugiperda, are publicly available, and such viruses are described in accordance with the present invention. In the specification, it may be used as a virus specifically for transfection of Spodoptera frugiperda cells.

Plant cell cultures of cotton, corn, potato, soybean, petunia, tomato, arabidopsis and tobacco can also be used as hosts. Cloning vectors and expression vectors useful for protein production in plant cell cultures are known to those of skill in the art. For example, Hitatt 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 See Fecker et al. (1996) Plant Mol Biol 32: 979-986.

However, interest in vertebrate cells is highest, and reproduction of vertebrate cells in culture (tissue culture) has become a routine procedure. Examples of useful mammalian host cell lines are monkey kidney CV1 strains transformed by SV40 (COS-7, ATCC CRL1651); human fetal kidney strains (293 cells or subcloned for growth in suspension culture). 293 cells, Graham et al., J.Gen Virol. 36:59 (1977)); Baby hamster kidney cells (BHK, ATCC CCL10); Chinese hamster ovary cells /-DHFR (CHO, Urlaub et al., Proc. Natl.Acad.Sci.USA77: 4216 (1980)); Mouse Sertri cells (TM4, Mather, Biol.Reprod. 23: 243-251 (1980)); Monkey kidney cells (CVI ATCC CCL70); African green monkey kidney cells (TM4, Mather, Biol.Reprod.23: 243-251 (1980)); VERO-76, ATCC CRL1587); Human cervical cancer cells (HELA, ATCC CCL2); Canine kidney cells (MDCK, ATCC) CCL34); Buffalo Lat hepatocytes (BRL 3A, ATCC CRL1442); Human lung cells (W138, ATCC CCL75); Human hepatocytes (Hep G2, 1413 8065); Mouse breast tumors (MMT 060562, ATCC CCL5 1); TRI cells (Mather et al., Annals NY Acad. Sci. (1982) 383: 44-68); MRC5 cells; FS4 cells; and human hepatoma strain (Hep G2).

In a further embodiment, the invention provides a method of producing the antibody construct of the invention, the method comprising culturing the host cell of the invention under conditions that allow expression of the antibody construct of the invention. It involves recovering the antibody construct produced from the culture.

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

When using recombinant techniques, antibody constructs can be produced intracellularly in the periplasmic space or secreted directly into the medium. When the antibody construct is produced intracellularly, the first step is to remove particulate debris from either the host cell or the lysed fragment, for example by centrifugation or ultrafiltration. Carter et al., Bio / Technology 10: 163-167 (1992) describe the procedure for isolating antibodies secreted into the periplasmic space of E. coli. Briefly, the cell paste is thawed in the presence of sodium acetate (pH 3.5), EDTA, and phenylmethylsulfonylfluoride (PMSF) for more than about 30 minutes. Cell debris can be removed by centrifugation. When the antibody is secreted into the medium, the supernatant from such an expression system is generally a commercially available protein concentration filter, such as Amicon or Millipore. First concentrated using a Pellicon ultrafiltration unit. Protease inhibitors such as PMSF may be included in any of the above steps to inhibit proteolysis, and antibiotics may be included to prevent the growth of foreign contaminants.

The antibody constructs of the invention prepared from host cells can be recovered or purified using, for example, 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 , anion or cation exchange resin. Chromatography on (such as polyaspartate columns), chromatographic focusing, SDS-PAGE, and ammonium sulfate precipitation are also available depending on the antibody recovered. When the antibody construct of the present invention contains the CH3 domain, Bakerbond ABX resin (JTBaker, Phillipsburg, NJ) is useful for purification.

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

Furthermore, the present invention provides a pharmaceutical composition comprising the antibody construct of the present invention or the antibody construct produced by the method of the present invention.

As used herein, the term "pharmaceutical composition" refers to a composition suitable for administration to a patient, preferably a human patient. A particularly preferred pharmaceutical composition of the invention comprises one or more antibody constructs of the invention (s), preferably in therapeutically effective amounts. Preferably, the pharmaceutical composition is suitable for one or more (pharmaceutically effective) carriers, stabilizers, excipients, diluents, solubilizers, surfactants, emulsifiers, preservatives and / or adjuvants. Includes additional formulations. The acceptable components of the composition are preferably non-toxic to the recipient at the dosage and concentration used. Examples of the pharmaceutical composition of the present invention include, but are not limited to, liquid compositions, frozen compositions, and freeze-dried compositions.

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

Certain embodiments provide pharmaceutical compositions that include the antibody constructs of the invention and one additional excipient, such as those exemplified herein in this section and elsewhere in the specification. Excipients are effective in this regard for a variety of purposes, such as adjusting the physical, chemical, or biological properties of a formulation, such as adjusting viscosity, and / or adjusting the processes of the invention. Improves sexuality and / or stabilizes such formulations and processes against degradation and damage due to stress that occurs, for example, during manufacture, transport, storage, pre-use preparation, administration, and subsequent stress. It can be used in the present invention for such reasons.

In certain embodiments, the pharmaceutical composition modifies, for example, the pH, osmolality, viscosity, transparency, color, isotonicity, odor, asepticity, stability, dissolution rate or release rate, adsorption or penetration of the composition. , May contain pharmaceutical material for maintenance or retention (REMINGTON'S PHARMACEUTICAL SCIENCES, 18'' Edition, (ARGenrmo, ed.), 1990, Mack Publishing See Company). In such embodiments, suitable formulation materials include: Amino acids, including charged amino acids, preferably lysine, lysine acetate, arginine, glutamate and / or histidine, such as glycine, alanine, glutamine, asparagine, threonine, Proline, 2-phenylalanine, etc.-Antibacterial agents, such as antibacterial and antifungal agents, etc.-Antioxidants, such as ascorbic acid, methionine, sodium sulfite or sodium hydrogen sulfite; Buffers, buffers and buffers used to maintain a slightly lower pH, typically in the pH range of about 5 to about 8 or 9; buffer examples are borates. , Dicarbonate, Tris-HCl, citrate, phosphate or other organic acid, succinate, phosphate, histidine and acetate; for example, Tris buffer, about pH 7.0-8.5, Or about pH 4.0-5.5 acetate buffer; Non-aqueous solvents such as propylene glycol, polyethylene glycol, vegetable oils such as olive oil, and injectable organic esters such as ethyl oleate; And aqueous carriers containing water, alcohol / aqueous solution, emulsions or suspensions, including buffer media; biodegradable polymers such as polyester; bulking agents such as mannitol or glycine; Ethylenediamine tetraacetic acid (EDTA), etc .; Isotonic and absorption retarders; Complexing agents, such as caffeine, polyvinylpyrrolidone, beta-cyclodextrin or hydroxypropyl-beta-cyclodextrin) Sugars; disaccharides; and other carbohydrates (such as glucose, mannose or dextrin); carbohydrates may be non-reducing sugars, preferably trehalose, sucrose, octasulfate, sorbitol or xylitol;

It will be apparent to those skilled in the art that different components of a pharmaceutical composition (eg, those listed above) may have different effects, eg, amino acids as buffers, stabilizers and / or antioxidants. Can function; mannitol can function as a bulking agent and / or tonicity enhancer; sodium chloride can function as a delivery vehicle and / or tonicity enhancer, and so on.

It is assumed that the composition of the present invention may contain a biologically active agent in addition to the polypeptide of the present invention as defined herein, depending on the intended use of the composition. .. Such drugs include drugs that act on the gastrointestinal system, drugs that act as cell growth inhibitors, drugs that prevent hyperurikemia, drugs that inhibit the immune response (eg, corticosteroids), It may be a drug that regulates the inflammatory response, a drug that acts on the circulatory system and / or a drug known in the art, such as a cytokine. It is also envisioned that the antibody constructs of the present invention will be applied in co-therapy, i.e. in combination with another anti-cancer drug.

In certain embodiments, the optimal pharmaceutical composition will be determined by one of ordinary skill in the art, for example, depending on the route of administration of interest, the form of delivery and the desired dosage. See, for example, REMINGTON'S PHARMACEUTICAL SCIENCES above. In certain embodiments, such compositions may affect the physical condition, stability, in vivo release rate and in vivo clearance rate of the antibody constructs of the invention. In certain embodiments, the main vehicle or carrier in the pharmaceutical composition may be essentially either aqueous or non-aqueous. For example, a suitable vehicle or carrier may be water for injection, saline or artificial cerebrospinal fluid, which may supplement other materials common in compositions for parenteral administration. Phosphate buffered saline or saline mixed with serum albumin is a further exemplary vehicle. In certain embodiments, the antibody constructs of the compositions of the present invention are the selected compositions having the desired purity, any pharmaceutical agent (REMINGTON'S PHARMACEUTICAL above). By mixing with SCIENCES), it may be prepared for storage in the form of lyophilized cakes or aqueous solutions. Further, in certain embodiments, the antibody constructs of the invention may be formulated as lyophilized using suitable excipients such as sucrose.

Where parenteral administration is considered, therapeutic compositions for use in the present invention will contain the desired antibody construct of the invention in a pharmaceutically acceptable vehicle, parenterally without pyrogens. It may be provided in the form of an acceptable aqueous solution. A vehicle particularly suitable for parenteral injection is sterile distilled water, in which the antibody construct of the present invention is formulated as a sterile isotonic solution and retained appropriately. In certain embodiments, the preparation may provide a controlled or sustained release of the desired molecule and product that can be delivered by depot injection, such as an injectable microsphere, bioerodible particles, a polymeric compound. It may include formulation with (such as polylactic acid or polyglycolic acid), beads or liposomes. In certain embodiments, hyaluronic acid, which has the effect of promoting the persistence of the period of circulation, may also be used. In certain embodiments, a transplantable drug delivery device may be used to introduce the desired antibody construct.

Additional pharmaceutical compositions will be apparent to those of skill in the art that include a formulation comprising the antibody construct of the invention in a continuous or controlled delivery / release formulation. Techniques for formulating a variety of other sustained or controlled delivery means, such as liposome carriers, bioerodible microparticles or porous beads and depot injections, are also known to those of skill in the art. See, for example, International Patent Application No. PCT / US93 / 00829. It describes the controlled release of porous polymer microparticles for delivery of pharmaceutical compositions. The sustained release preparation may include a semipermeable polymer matrix in the form of an article, eg, a film, or microcapsules. Sustained release matrices include polyesters, hydrogels, polylactides (disclosed in US Patent No. 3,773,919 and European Patent Application Publication No. EP058481), L-glutamic acid and gamma ethyl-L-glutamate copolymers (Sidman et al). ., 1983, Biopolymers 2: 547-556), Poly (2-Hydroxyethyl-methacrylate) (Langer et. al., 1981, J.Biomed.Mater.Res.15: 167-277 and Langer, 1982, Chem.Tech.12: 98-105), ethylene vinyl acetate (Langer et al., 1981 above) or poly- D (-)-3-Hydroxybutyric acid (European Patent Application Publication No. EP133,988) may be mentioned. Sustained release compositions may also include liposomes that can be prepared by any of several methods known in the art. For example, Eppstein et al., 1985, Proc.Natl.Acad.Sci.USA82: 3688-3692; European Patent Application Publication Nos. EP036,676; EP088,046 and EP143,949. checking ...

The antibody constructs were also prepared in colloidal drug delivery systems (eg, liposomes, albumin microspheres, microemulsions, nanoparticles and nanocapsules) or in macroemulsions, eg, by coacervation techniques or by interfacial polymerization. It may be encapsulated in microcapsules (eg, hydroxymethyl cellulose or gelatin microcapsules and poly (methyl methacrylate) microcapsules, respectively). Such techniques are disclosed in Remington's Pharmaceutical Sciences, 16th edition, Oslo, A., Ed., (1980).

Pharmaceutical compositions used for in vivo administration are typically provided as sterile preparations. Sterilization can be performed by filtering through a sterile filtration membrane. If the composition is lyophilized, sterilization using this method may be performed either before or after lyophilization and reconstitution. Compositions for parenteral administration can be stored in lyophilized form or in solution. The parenteral composition is generally placed in a container with a sterile access port, eg, an intravenous infusion bag or vial with a stopper that can be pierced by a hypodermic needle.

Another aspect of the invention comprises a self-buffering antibody construct of the formulation of the invention, which has a pharmaceutical composition as described in International Patent Application WO 06138181 (A2) (PCT / US 2006/022599). Can be used as a thing. Various descriptions of protein stabilization and useful formulation materials and methods in this regard, such as 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), etc. are available. See in particular with respect to excipients and processes thereof for self-buffering protein formulations according to the present invention, particularly with respect to protein pharmaceutical products and processes for veterinary and / or human medical applications.

The salt according to a particular embodiment of the invention, for example, to adjust the ionic strength and / or isotonicity of the formulation, and / or the solubility and / or physics of the protein or other component of the composition according to the invention. It may be used to improve physical stability. As is well known, ions bind to charged residues on the surface of a protein and shield the charged and polar groups in the protein to make their electrostatic, attractive, and repulsive interactions. The natural state of the protein can be stabilized by reducing the strength of the protein. Ions can also stabilize the denatured state of a protein, especially by binding to the denatured peptide bond (--CONH) of the protein. In addition, ionic interactions with charged and polar groups in proteins can also reduce intramolecular electrostatic interactions, thereby preventing or reducing protein aggregation and insolubility.

Ionic species differ significantly in their effect on proteins. Numerous categorical rankings for their effects on ions and proteins have been developed and can be used in formulating pharmaceutical compositions according to the present invention. One example is the Hofmeister series, which ranks ionic and polar nonionic solutes by their effect on the conformational stability of proteins in solution. The stabilized solute is referred to as "cosmotropic". The destabilizing solute is referred to as the "chaotropic". Cosmotropes are commonly used in high concentrations (eg,&gt; 1 mol of ammonium sulphate) to precipitate proteins from solution (salting out). Chaotropes are commonly used to denature and / or solubilize proteins (salt-dissolving). The relative effectiveness of the ions for "salting" and "salting out" defines their position in the Hofmeister series.

Free amino acids can be used in antibody constructs of the formulations of the invention according to various embodiments of the invention as bulking agents, stabilizers, and antioxidants, as well as other standard applications. Lysine, proline, serine, and alanine can be used to stabilize the protein in the formulation. Glycine is useful in lyophilization to ensure accurate cake structure and properties. Arginine can be useful for inhibiting protein aggregation in both liquid and lyophilized formulations. Methionine is useful as an antioxidant.

Polycols include sugars such as mannitol, sucrose, and sorbitol and polyhydric alcohols such as glycerol and propylene glycol, as well as polyethylene glycol (PEG) and related substances for the purposes of consideration herein. The polyol is kosmotropic. They are stabilizers that are useful in both liquid and lyophilized formulations to protect proteins from physical and chemical degradation processes. The polyol is also useful for adjusting the tonicity of the formulation. Among the polyols, one useful in the selected embodiment of the present invention is mannitol, which is commonly used in lyophilized formulations to ensure the structural stability of the cake. Mannitol ensures the structural stability of the cake. Mannitol is commonly used with Rio protectants, such as sucrose. Sorbitol and sucrose are preferred agents, among other preferred agents, as agents for adjusting tonicity and as stabilizers for protection from freeze-thaw stress during transport or bulk preparation during the manufacturing process. Reducing sugars (containing free aldehydes or ketone groups), such as glucose and lactose, can saccharify surface lysine and arginine residues. Therefore, they are generally not among the polyols preferred for use in accordance with the present invention. In addition, sugars that form such reactive species, such as sucrose, are also hydrolyzed to fructose and glucose under acidic conditions, resulting in saccharification, which is among the preferred polyols of the invention in this regard. No. PEG is useful for stabilizing proteins and as a cryoprotectant and can be used in the present invention in this regard.

Embodiments relating to antibody constructs of the formulations of the present invention further comprise a surfactant. Protein molecules may be susceptible to adsorption on the surface and denaturation at the gas-liquid, solid-liquid, and liquid-liquid interfaces and the resulting aggregation. These effects generally increase or decrease in inverse proportion to protein concentration. These harmful interactions generally increase or decrease in inverse proportion to the protein concentration and are typically exacerbated by physical agitation, such as agitation that occurs during product transport and handling. Surfactants are routinely used to prevent, minimize, or reduce surface adsorption. Surfactants useful in the present invention in this regard include polysorbate 20, polysorbate 80, other fatty acid esters of sorbitan polyethoxylate, and poloxamer 188. Surfactants are also commonly used to control the conformational stability of proteins. The use of detergents in this regard is protein-specific because any given surfactant typically stabilizes one protein and destabilizes another. Because it becomes.

Polysorbates are susceptible to oxidative degradation and, when supplied, often contain sufficient peroxide amounts to cause the oxidation of protein residue side chains, especially methionine. As a result, polysorbates should be used with caution and, if used, at their lowest effective concentrations. In this regard, polysorbates exemplify the principle that excipients should be used at their lowest effective concentrations.

Embodiments of antibody constructs of the formulations of the present invention further comprise one or more antioxidants. Harmful oxidation of proteins can be prevented to some extent by maintaining appropriate levels of ambient oxygen and ambient temperature in pharmaceutical formulations and by avoiding exposure to light. Antioxidant excipients can also be used to prevent oxidative degradation of proteins. Among the antioxidants useful in this regard are reducing agents, oxygen / free radical scavengers, and chelating agents. Antioxidants for use in therapeutic protein formulations according to the present invention are preferably water soluble and maintain their activity throughout the expiration date of the product. EDTA is the preferred antioxidant according to the present invention in this regard. Antioxidants can damage proteins. For example, reducing agents, especially glutathione, can disrupt intramolecular disulfide bonds. Therefore, antioxidants for use in the present invention are selected, among other things, to eliminate or sufficiently reduce the possibility of themselves damaging proteins in the formulation.

A formulation according to the present invention may contain a metal ion that is a protein cofactor and is required to form a protein coordination complex, for example zinc, which is required to form a particular insulin suspension. Metal ions can also interfere with several processes that break down proteins. However, metal ions also catalyze the physical and chemical processes that break down proteins. Magnesium ion (10-120 mM) can be used to inhibit the isomerization of aspartic acid to isoaspartic acid. Ca⁺²Ions (up to 100 mM) can increase the stability of human deoxyribonuclease. However, Mg⁺², Mn⁺², And Zn⁺²Can destabilize rhDNase. Similarly, Ca⁺²And Sr⁺²Can stabilize factor VIII, which is Mg⁺², Mn⁺²And Zn⁺², Cu⁺²And Fe⁺²Can be destabilized by Al⁺³Can be increased by ions.

Embodiments of antibody constructs of the formulations of the present invention further comprise one or more preservatives. Preservatives are needed when developing multiple dose parenteral formulations with two or more extractions from the same container. Their main function is to inhibit the growth of microorganisms and ensure the sterility of the product throughout the expiration date or period of use of the drug. Commonly used preservatives include benzyl alcohol, phenol and m-cresol. Although preservatives have a long history of being used with small molecule parenteral agents, the development of protein formulations containing preservatives can be difficult. Preservatives almost always have a destabilizing effect (aggregation) on proteins, which is a major factor limiting their use in multiple dose protein formulations. To date, most protein drugs have been formulated for single use only. However, if multiple dose formulations are possible, they have the additional advantage of allowing increased patient convenience and marketability. A good example is human growth hormone (hGH), and the development of preservatives has led to the commercialization of a more convenient multi-use injection pen. At least four such pen devices containing hGH preservatives are currently available on the market. Norditropin (liquid, Novo Nordisk), Nutropin AQ (liquid, Genentech) and Genotropin (freeze-dried--dual chamber cartridge, Pharmacia &amp; Upjohn) contain phenol, while Somatrope (Eli Lilly) is formulated with m-cresol. Will be transformed. Some aspects need to be considered during the formulation and development of the preservative form. The concentration of effective preservatives in medicines must be optimized. This requires testing given preservatives in dosage forms with a concentration range that confer antibacterial efficacy without compromising protein stability.

As may be expected, the development of liquid formulations containing preservatives is more difficult than lyophilized formulations. The lyophilized product can be lyophilized without a preservative and reconstituted with a diluent containing a preservative in use. This reduces the amount of time the preservative is in contact with the protein and significantly minimizes the associated stability risk. For liquid formulations, the effectiveness and stability of the preservative should be maintained throughout the product's shelf life (approximately 18-24 months). An important point to keep in mind is that the effectiveness of preservatives should be demonstrated in the final formulation containing the active drug and all excipient components.

The antibody constructs disclosed herein may also be formulated as immunoliposomes. A "liposome" is a small vesicle composed of various types of lipids, phospholipids and / or surfactants that are useful for drug delivery to mammals. Liposomal components are usually arranged in a bilayer format similar to the lipid arrangement of biological membranes. Liposomes containing antibody constructs include, for example, Epstein et al., Proc. Natl. Acad. Sci. USA, 82: 3688 (1985); Hwang et al., Proc. Natl. Prepared by methods known in the art, such as those described in Acad.Sci.USA, 77:4030 (1980); US Pat. Nos. 4,485,045 and 4,544,545; and WO 97/38731. Will be done. Liposomes with improved circulation time are disclosed in US Pat. No. 5,013,556. Particularly useful liposomes can be produced by reverse phase evaporation using a lipid composition containing phosphatidylcholine, cholesterol and PEG derivatized phosphatidylethanolamine (PEG-PE). Liposomes are extruded through a filter of defined pore size to obtain liposomes with the desired diameter. Fab'fragments of the antibody constructs of the invention can be complexed with liposomes as described in Martin et al. J. Biol. Chem. 257: 286-288 (1982) by a disulfide exchange reaction. The chemotherapeutic agent is optionally contained within the liposome. See Gabizon et al. J. National Cancer Inst. 81 (19) 1484 (1989).

Once formulated, the pharmaceutical composition may be stored in sterilized vials as a solution, suspension, gel, emulsion, solid, crystal, or as a dehydrated or lyophilized powder. Such formulations may be stored in either ready-to-use or pre-administration reconstituted forms (eg, lyophilized).

The biological activity of the pharmaceutical compositions defined herein is described, for example, in WO99 / 54440, or by Schlereth et al. (Cancer Immunol. Immunol. 20 (2005), 1-12), in the following examples. As such, it can be determined by a cytotoxic assay. "Efficacy" or "in vivo efficacy" as used herein, for example, refers to a response to therapy with a pharmaceutical composition of the invention using standardized NCI response criteria. The success or in vivo efficacy of therapy using the pharmaceutical compositions of the present invention causes the effectiveness of the composition for its intended purpose, i.e. its desired effect, i.e. depletion of diseased cells, eg, tumor cells. Refers to the ability of the composition. In vivo efficacy may be monitored by established standard methods for each disease unit, including, but not limited to, white blood cell counting, differences, fluorescence activated cell sorting, and bone marrow aspiration. In addition, various disease-specific clinical chemistry parameters and other established standard methods may be used. In addition, computer-assisted tomography, X-rays, nuclear magnetic resonance tomography (eg National) For response assessment based on Cancer Institute criteria [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, white blood cell counting, differences, fluorescence activated cell selection, bone marrow aspiration, lymph node biopsy / histology, and various lymphoma-specific clinical chemistry parameters (eg, Lactate dehydrogenase) and other established standard methods may be used.

Another major challenge in the development of drugs, such as the pharmaceutical compositions of the present invention, is the predictable regulation of pharmacokinetic properties. For this purpose, a pharmacokinetic profile of a drug candidate, i.e. a profile of pharmacokinetic parameters that affect the ability of a particular drug to treat a given condition, can be established. Pharmacokinetic parameters of a drug that affect the ability of the drug to treat a particular disease unit include, but are not limited to, half-life, volume of distribution, first-pass metabolism of the liver and serum binding. The efficacy of a given drug can be influenced by each of the above parameters.

"Half-life" means the time at which 50% of the administered drug is eliminated by biological processes such as metabolism, excretion, etc. "First-pass metabolism of the liver" means the nature of a drug that is metabolized upon first contact with the liver, i.e., during the first passage of the drug through the liver. "Volume of distribution" means the retention of a drug across various compartments of the body, such as intracellular and extracellular spaces, tissues and organs, and the distribution of the drug within these compartments. "Serum binding" means the nature of a drug that interacts with and binds to serum proteins, such as albumin, resulting in diminished or lost biological activity of the drug.

Pharmacokinetic parameters also include bioavailability, delay time (Tlag), Tmax, rate of absorption, more expression and / or Cmax for a given dose of drug administered. "Bioavailability" means the amount of drug in a blood compartment. "Delay time" means a time delay between the administration of a drug and its detection and measurable in blood or plasma. "Tmax" is the time to reach the maximum blood concentration of the drug thereafter, and "Cmax" is the maximum blood concentration obtained with a given drug. The time to reach the blood or tissue concentration of the drug required for its biological effect is affected by all parameters. Pharmacokinetic parameters of bispecific antibody constructs exhibiting interspecificity may be determined in preclinical animal studies in non-chimpanzee primates as outlined above, eg Schlereth et al. (Cancer Immunol. Immunoother). It is also described in the publications of .20 (2005), 1-12).

In one embodiment, the invention provides an antibody construct of the invention or an antibody construct produced by the methods of the invention for use in the prevention, treatment or amelioration of cancer.

The formulations described herein are useful as pharmaceutical compositions in the treatment, amelioration and / or prevention of pathological health conditions as described herein in patients in need of treatment and the like. The term "treatment" refers to both therapeutic treatment and prophylactic or prophylactic measures. Treatment is to cure, cure, alleviate, alleviate, change, treat, ameliorate, improve, or influence a disease, a symptom of a disease, or a predisposition to a disease. Includes application or administration of the formulation to the body, isolated tissue, or cells of a patient with a disease / disorder, a sign of the disease / disorder, or a predisposition to the disease / disorder.

The term "amelioration", as used herein, is of the type (metastatic) as specified herein by administration of an antibody construct according to the invention to a subject in need of improvement. ) Refers to any improvement in the disease state of a patient with one of the tumors or cancers. Such improvements may also be seen as slowing or stopping the progression of a patient's (metastatic) tumor or cancer. The term "prevention" as used herein refers to a (metastatic) tumor of the type as specified herein or by administration of an antibody construct according to the invention to a subject in need of prevention. It means avoidance of onset or recurrence in patients with one of the cancers.

The term "disease" refers to any condition that would benefit from treatment with the antibody constructs or pharmaceutical compositions described herein. This includes chronic and acute disorders or diseases, including conditions that make mammals susceptible to the disease in question.

A "neoplasm" is a tissue that grows abnormally and usually forms a mass, but this is not always the case. Even when forming a mass, neoplasms are commonly referred to as "tumors." The neoplasm or tumor can be benign, latently malignant (precancerous), or malignant. Malignant neoplasms are commonly referred to as cancer. Neoplasms can usually invade surrounding tissue and destroy it, forming metastases. That is, the neoplasm spreads to other parts of the body, tissues or organs. Therefore, the term "metastatic cancer" includes cancers that metastasize to other tissues or organs rather than one of the original tumors. Lymphoma and leukemia are lymphoid neoplasms. In the present invention, they are also included by the terms "tumor" or "cancer".

In a preferred embodiment, the invention provides an antibody construct of the invention or an antibody construct produced by the method of the invention for use in the prevention, treatment or amelioration of cancer, in which case the cancer is: Lung cancer, head and neck cancer, primary or secondary CNS tumor, primary or secondary brain tumor, primary CNS lymphoma, spinal axis tumor, brain stem glioma, glioblastoma, pituitary adenoma, corticocarcinoma , Esophageal cancer, colon cancer, breast cancer, ovarian cancer, NSCLC (non-small cell lung cancer), SCLC (small cell lung cancer), endometrial cancer, cervical cancer, uterine cancer, transition epithelial cancer, bone Cancer, pancreatic cancer, skin cancer, skin or intraocular melanoma, liver cancer, bile duct cancer, bile sac cancer, kidney cancer, rectal cancer, anal cancer, gastric cancer, gastrointestinal (stomach, colon) Rectal and duodenal) cancer, small intestine cancer, biliary tract cancer, urinary tract cancer, renal cell carcinoma, endometrial carcinoma, thyroid cancer, testis cancer, cutaneous squamous cell carcinoma, melanoma, Gastric cancer, prostate cancer, bladder cancer, osteosarcoma, mesenteric tumor, Hodgkin's disease, non-Hodgkin's lymphoma, chronic or acute leukemia, chronic myeloid leukemia, lymphocytic lymphoma, multiple myeloma, fibrosarcoma, neuroblasts It is selected from the group consisting of tumors, retinoblastomas, and soft sarcomas, and metastatic cancers derived from any of the above. (Metastatic) cancers are preferably P-cadherin-positive or P-cadherin-expressing cancers.

In a more preferred embodiment, the invention provides an antibody construct of the invention or an antibody construct produced by the method of the invention for use in the prevention, treatment or amelioration of cancer, in which case the cancer. (Metastatic) Squamous epithelial carcinoma.

The present invention also provides a method for treating or ameliorating a (metastatic) tumor or cancer, which method is produced by an antibody construct of the present invention or a method of the present invention in a subject in need of treatment or amelioration. Includes the step of administering the antibody construct.

In a preferred embodiment, the invention provides a method of treating or ameliorating a tumor or cancer, where the cancer is lung cancer, head and neck cancer, primary or secondary CNS tumor, primary or secondary. Sexual brain tumor, primary CNS lymphoma, spinal axis tumor, brain stem glioma, pituitary adenoma, adrenal cortex cancer, esophageal cancer, colon cancer, breast cancer, ovarian cancer, NSCLC (non-small cell lung cancer), SCLC (small) Cellular lung cancer), endometrial cancer, cervical cancer, uterine cancer, transitional epithelial cancer, bone cancer, pancreatic cancer, skin cancer, skin or intraocular melanoma, liver cancer, bile duct cancer, Biliary sac cancer, kidney cancer, rectal cancer, anal cancer, gastric cancer, gastrointestinal (gastric, colonic rectal, and duodenal) cancer, small intestinal cancer, biliary tract cancer, urinary tract cancer, renal cell carcinoma , Endometrial cancer, thyroid cancer, testis cancer, cutaneous squamous cell carcinoma, melanoma, gastric cancer, prostate cancer, bladder cancer, osteosarcoma, mesenteric tumor, Hodgkin's disease, non-Hodgkin's lymphoma, chronic Or from acute leukemia, chronic myeloid leukemia, lymphocytic lymphoma, multiple myeloma, fibrosarcoma, neuroblastoma, retinoblastoma, and soft sarcoma, and metastatic cancer diseases derived from any of the above. The method comprises administering the antibody construct of the invention or the antibody construct produced by the method of the invention to a subject in need of treatment or improvement.

In a more preferred embodiment, the invention provides a tumor or cancer or a metastatic tumor or a method of treating or ameliorating a cancer, in which case the cancer is a (metastatic) squamous epithelial carcinoma.

The term "subject in need" or "subject in need of treatment" includes subjects who already have a disability, as well as subjects whose disability should be prevented. Subjects or "patients" in need include human and other mammalian subjects undergoing either prophylactic or therapeutic treatment.

The antibody constructs of the present invention generally have a range of bioavailability and persistence, for a particular route of administration and method of dosing, for a particular dose and frequency, for a particular treatment in a particular disease, among others. Will be designed in. The material of the composition is preferably formulated at an concentration acceptable to the site of administration.

Therefore, the formulations and compositions may be designed according to the present invention for delivery by any suitable route of administration. In the context of the present invention, the routes of administration include: -local routes (on the skin, inhalation, nose, eyes, ears / ears, vagina, mucous membranes, etc.); , Buccal, rectal, etc.); and Parenteral routes (intravenous, intraarterial, intraosseous, intramuscular, intracerebral, intraventricular, extradural, intraluminal, subcutaneous, intraperitoneal, extralamellar, intrajoint , Intracardiac, intradermal, intralesional, intrauterine, intravitreal, intravitreal, transdermal, intranasal, transmucosal, intrasulcular, intraluminal, etc.), but not limited to these.

The pharmaceutical compositions and antibody constructs of the present invention are particularly useful for parenteral administration, eg, subcutaneous or intravenous delivery, by injection, such as bolus injection, or by injection, such as continuous infusion. The pharmaceutical composition may be administered using a medical device. Examples of medical devices for administering pharmaceutical compositions are U.S. Pat. Nos. 4,475,196; 4,439,196; 4,447,224; 4,447,233; 4,486,194; 4,487,603. Documents; 4,596,556; 4,790,824; 4,941,880; 5,064,413; 5,312,335; 5,312,335; 5,383,851; and 5,399,163. It is described in.

In particular, the present invention provides uninterrupted administration of suitable compositions. As a non-limiting example, uninterrupted or substantially uninterrupted, ie, continuous administration is achieved by a small patient-worn pump system for measuring the influx of therapeutic agent into the patient's body. It's okay. Pharmaceutical compositions containing the antibody constructs of the invention can be administered by using this pump system. Such pump systems are generally known in the art and generally rely on regular replacement of cartridges containing the infused therapeutic agent. Replacing the cartridges in such a pump system would otherwise allow the therapeutic agent to flow uninterrupted into the patient's body, but with temporary interruptions. In such cases, the dosing phase before cartridge replacement and the dosing phase after cartridge replacement are, within the meaning of the pharmaceutical means and methods of the invention, a single "uninterrupted administration" of such therapeutic agent. Is still considered to constitute together.

Continuous or uninterrupted administration of the antibody constructs of the invention is intravenous or subcutaneous administration by a fluid delivery device or small pump system, including a fluid delivery mechanism for delivering fluid from the reservoir and an actuating mechanism for activating the delivery mechanism. It may be. A pump system for subcutaneous administration may include a needle or cannula to penetrate the patient's skin and deliver the suitable composition into the patient's body. The pump system may be fixed or attached directly to the patient's skin, independent of veins, arteries or blood vessels, thereby allowing direct contact between the pump system and the patient's skin. The pump system can be worn on the patient's skin for 24 hours up to several days. The pump system may be of small size with a small amount of reservoir. As a non-limiting example, the reservoir volume of a suitable pharmaceutical composition administered can be 0.1-50 ml.

Continuous administration may also be transdermal administration with patches that are worn on the skin and replaced at intervals. Those skilled in the art are aware of patch systems for drug delivery suitable for this purpose. For transdermal administration, replacement of the first used patch places a new second patch, for example, on the skin surface immediately adjacent to the first used patch, just before removal of the first used patch. It should be noted that it is particularly suitable for uninterrupted administration because it is advantageous to be achieved at the same time. There is no problem of inflow interruption or power battery stop.

If the pharmaceutical composition has been lyophilized, the lyophilized material is first reconstituted with the appropriate liquid prior to administration. The lyophilized material is reconstituted with, for example, bacteriostatic water for injection (BWFI), saline, phosphate buffered saline (PBS), or the same formulation that contained the protein prior to frost drying. good.

The compositions of the invention are determined, for example, by dose escalation studies by administering to non-chimpanzee primates, eg, macaques, escalating doses of the antibody constructs of the invention exhibiting the interspecific specificities described herein. Can be administered to the subject at a suitable dose that can be administered. As mentioned above, the antibody constructs of the invention exhibiting the interspecificity described herein can be advantageously used in preclinical studies in non-chimpanzee primates and in the same form as drugs in humans. The dosing regimen will be determined by the attending physician and clinical factors. As is well known in the medical technology field, the dose for any one patient is the patient's body size, body surface area, age, specific compound to be administered, gender, time of administration and route of administration, general health, It also depends on many factors, including other drugs being administered at the same time.

The term "effective dose" or "effective dose" is defined as an amount sufficient to achieve the desired effect, or at least partially. The term "therapeutically effective dose" is defined as an amount sufficient to cure, or at least partially stop, the disease and its complications in a patient already suffering from the disease. Effective amounts or doses for this use include the condition to be treated (indication), antibody construct to be delivered, treatment content and purpose, severity of disease, previous therapy, patient history and response to therapeutic agent, administration. It will depend on the pathway, size (body surface area or organ size) and / or the patient's condition (age and general health), as well as the patient's own immune system overall condition. Appropriate doses may be adjusted at the discretion of the attending physician so that they can be administered to the patient in a single or series of doses and for optimal therapeutic effect.

Typical doses may range from about 0.1 μg / kg up to about 30 mg / kg or more, depending on the factors mentioned above. In certain embodiments, the dose may range from 1.0 μg / kg up to about 20 mg / kg, and optionally from 10 μg / kg up to about 10 mg / kg or 100 μg / kg up to about 5 mg / kg.

A therapeutically effective amount of an antibody construct of the invention preferably results in reduced severity of disease symptoms, increased frequency or duration of asymptomatic disease periods, or prevention of injury or disability due to distress of the disease. To treat CDH3-expressing tumors, therapeutically effective amounts of antibody constructs of the invention, such as anti-CDH3 / anti-CD3 antibody constructs, preferably have at least about 20% cell growth or tumor growth compared to untreated patients. Inhibits 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%. The ability of compounds to inhibit tumor growth may be evaluated in animal models that predict efficacy in human tumors.

The pharmaceutical composition may be administered as a single therapeutic agent or in combination with additional therapies such as anti-cancer therapy, eg, other proteinaceous and non-proteinaceous drugs, as needed. These drugs are administered simultaneously with the composition comprising the antibody constructs of the invention as defined herein, or separately at time-defined intervals and doses before or after administration of the antibody constructs. good.

The term "effective and non-toxic dose", as used herein, of pathological cell depletion, tumor elimination, tumor shrinkage or disease, without or essentially causing significant toxic effects. Refers to an acceptable dose of the antibody construct of the invention that is high enough to provide stabilization. Such effective and non-toxic doses may be determined, for example, by dose escalation studies described in the art and should be less than the dose that induces serious adverse side events (dose limiting toxicity, DLT).

The term "toxicity" as used herein refers to the toxic effects of a drug that appear in an adverse event or a serious adverse event. These side events may refer to a general lack of tolerability of the drug and / or a lack of local tolerance after administration. Toxicity can also include teratogenic or carcinogenic effects caused by the drug.

The terms "safety," "in vivo safety," or "tolerability," as used herein, are serious adverse effects immediately after administration (local tolerance) and during longer-term drug application. Defined as administration of a drug that does not induce an event. "Safety," "in vivo safety," or "tolerability" can be assessed, for example, at regular intervals during treatment and follow-up. Measurements include clinical evaluation, such as screening for organ signs and laboratory abnormalities. A clinical assessment may be performed and differences from normal findings may be recorded / coded according to NCI-CTC and / or MedDRA standards. Symptoms of organs include, for example, Common Terminology Criteria for adverse events. Criteria such as allergy / immunology, blood / bone marrow, cardiac arrhythmia, and coagulation as described in v3.0 (CTCAE) may be mentioned. Test parameters that may be tested include, for example, hematology, clinical chemistry, coagulation profile and urinalysis and other body fluid tests such as serum, plasma, lymph or spinal fluid, cerebrospinal fluid and the like. Therefore, safety includes, for example, physical examination, imaging techniques (ie, ultrasound, X-rays, CT scans, magnetic resonance imaging (MRI), other measurements using technical devices (ie, electrocardiogram), vital signs. Adverse events of non-chimpanzee primates in the uses and methods according to the invention can be assessed by measuring test parameters and recording adverse events, eg, by histopathology and / or histochemical methods. It's okay.

The above terms are also mentioned, for example, in the Preclinical safety evaluation of biotechnology-derived pharmaceuticals S6; ICH Harmonized Tripartite Guideline; ICH Steering Committee meeting (July 16, 1997).

In a further embodiment, the invention provides a kit comprising the antibody construct of the invention, the antibody construct produced by the method of the invention, the vector of the invention, and / or the host cell of the invention.

In the context of the present invention, the term "kit" refers to two or more components packaged together in a container, container or other object, one of which is the antibody construct, pharmaceutical composition, vector of the present invention. Alternatively, it means a component corresponding to the host cell. Thus, the kit may be described as a set of products and / or instruments sufficient to achieve a particular purpose, which may be sold separately.

The kit contains any suitable shape, size and material (preferably waterproof, eg plastic or glass) containing the antibody constructs or pharmaceutical compositions of the invention in dosages suitable for administration (see above). ) May contain one or more containers (vials, ampoules, containers, syringes, bottles, bags, etc.). The kit reconstructs the antibody constructs of the invention, such as instructions for use (eg, in the form of leaflets or instruction manuals), means for administering the antibody constructs of the invention, such as syringes, pumps, injectors, etc. And / or means for diluting the antibody construct of the present invention may be further included.

The present invention also provides kits for single dose dose units. The kit of the present invention may also contain a first container containing a dried / lyophilized antibody construct and a second container containing an aqueous formulation. In certain embodiments of the invention, kits containing single-chamber and multi-chamber prefilled syringes (eg, liquid syringes and riosyringes) are provided.

<figref num="1">FIG. 6 is a schematic representation of the five extracellular domains D1 to D5 of human CDH3, mouse CDH3, and one exemplary chimeric CDH3 (here "D1B mouse"). Below, the five domains are shown again, and each of them divided into three subdomains. An exemplary FACS signal interpretation (see Epitope Clustering in Example 2) is shown in the lower right.</figref><figref num="2-1">Sequence alignment of human CDH3 and mouse CDH3 and different domains: Signal peptides, propeptides, extracellular domains D1-D5, transmembrane domains and cytoplasmic domains are shown exemplary. Mouse sequence exchangers of 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). ..</figref><figref num="2-2">Sequence alignment of human CDH3 and mouse CDH3 and different domains: Signal peptides, propeptides, extracellular domains D1-D5, transmembrane domains and cytoplasmic domains are shown exemplary. Mouse sequence exchangers of 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). ..</figref><figref num="3">Shows 20 chimeric human-mouse CDH3 constructs (5 extracellular domains (ECDs) and 3 subdomains for each ECD) in addition to human and mouse CDH3s expressed on the surface of CHO cells, as shown by flow cytometry. .. Expression of human wild-type CDH3, mouse wild-type CDH3 and chimeric CDH3 constructs on CHO cells was verified with mouse cross-reactive monoclonal mouse IgG1 anti-human CDH3 antibody. Bound monoclonal antibody was detected with anti-mouse IgG Fcγ-PE (1: 100, 50 μl; Jackson Immunoresearch No. 115-116-071). D1 *): Hu CDH3 D1 mu-CHO.</figref><figref num="4A">The epitope mapping of the CDH3 construct is shown. Examples of binding molecules specific for different epitope clusters / extracellular subdomains, such as those detected by epitope mapping of chimeric CDH3 constructs. See Example 2. FIG. 4A shows the D1B binder.</figref><figref num="4B">The epitope mapping of the CDH3 construct is shown. Examples of binding molecules specific for different epitope clusters / extracellular subdomains, such as those detected by epitope mapping of chimeric CDH3 constructs. See Example 2. FIG. 4B shows a D2C binder.</figref><figref num="4C">The epitope mapping of the CDH3 construct is shown. Examples of binding molecules specific for different epitope clusters / extracellular subdomains, such as those detected by epitope mapping of chimeric CDH3 constructs. See Example 2. Figure 4C shows the D3A binder.</figref><figref num="5A">FACS binding analysis of 5 μg / mL purified bispecific antibody monomers for the cell lines shown is shown. See also Example 5. Detection of CDH3 × CD3 bispecific antibody binding was performed with in-house mouse antibodies specific for the CD3 binding portion of the bispecific antibody, followed by goat anti-mouse Fcγ-PE. The negative control was buffer only, followed by the detection antibody. CDH3 × CD3 bispecific binder (epitogen cluster / extracellular subdomain D2C), human CDH3 transfected CHO cells, human CD3 on human T cell line HPB-all, cynoCDH3 transfected CHO cells, cynoCD3 expressing T cell line Their binding to cynoCD3 on HSC-F, human CDH3-positive cell line A431, and mouse CDH3-transfected CHO cells (negative controls) was analyzed. Binding was detected in all cases except the negative control.</figref><figref num="5B">FACS binding analysis of 5 μg / mL purified bispecific antibody monomers for the cell lines shown is shown. See also Example 5. Detection of CDH3 × CD3 bispecific antibody binding was performed with in-house mouse antibodies specific for the CD3 binding portion of the bispecific antibody, followed by goat anti-mouse Fcγ-PE. The negative control was buffer only, followed by the detection antibody. CDH3 × CD3 bispecific binder (epitogen cluster / extracellular subdomain D3A), human CDH3 transfected CHO cells, human CD3 on human T cell line HPB-all, cynoCDH3 transfected CHO cells, cynoCD3 expressing T cell line Their binding to cynoCD3 on HSC-F, human CDH3-positive cell line A431, and mouse CDH3-transfected CHO cells (negative controls) was analyzed. Binding was detected in all cases except the negative control.</figref><figref num="5C">FACS binding analysis of 5 μg / mL purified bispecific antibody monomers for the cell lines shown is shown. See also Example 5. Detection of CDH3 × CD3 bispecific antibody binding was performed with in-house mouse antibodies specific for the CD3 binding portion of the bispecific antibody, followed by goat anti-mouse Fcγ-PE. The negative control was buffer only, followed by the detection antibody. CDH3 × CD3 bispecific binders (epitope clusters / extracellular subdomain D2C) were analyzed for their binding of human CDH3 paralogs to CDH1, CDH2, CDH4 and CDH5. No binding to paralogs was detected. dhfr^-/-No binding to CHO cells (negative control) was also detected.</figref><figref num="5D">FACS binding analysis of 5 μg / mL purified bispecific antibody monomers for the cell lines shown is shown. See also Example 5. Detection of CDH3 × CD3 bispecific antibody binding was performed with in-house mouse antibodies specific for the CD3 binding portion of the bispecific antibody, followed by goat anti-mouse Fcγ-PE. The negative control was buffer only, followed by the detection antibody. CDH3 × CD3 bispecific binders (epitope cluster / extracellular subdomain D3A) were analyzed for their binding of human CDH3 paralogs to CDH1, CDH2, CDH4 and CDH5. No binding to paralogs was detected. dhfr^-/-No binding to CHO cells (negative control) was also detected.</figref><figref num="6">18 hours⁵¹The cytotoxic activity of stimulated human CD8 + T cells against human CDH3 transfected CHO cells in the presence of CDH3 × CD3 bispecific antibody measured by Chromium release assay is shown. Effector cells: Stimulus-enriched human CD8 + T cells. Target cells: Human CDH3 transfected CHO cells. Effector cell to target cell (E: T) ratio: 10: 1. Antibodies specific for epitope clusters / extracellular subdomains D2C (Fig. 6A) and D3A (Fig. 6B).</figref><figref num="7">18 hours⁵¹The cytotoxic activity of human CD8 + T cells stimulated against human CDH3-positive epidermoid carcinoma cell line A431 in the presence of CDH3 × CD3 bispecific antibody measured by Chromium release assay is shown. Effector cells: Stimulus-enriched human CD8 + T cells. Target cells: Human A431 cells. Effector cell to target cell (E: T) ratio: 10: 1. Antibodies specific for epitope clusters / extracellular subdomains D2C (Fig. 7A) and D3A (Fig. 7B).</figref><figref num="8">The cytotoxic activity of unstimulated human PBMCs against human CDH3-transfected CHO cells in the presence of CDH3 × CD3 bispecific antibodies, as measured by a 48-hour FACS-based cytotoxic assay. Effector cells: Unstimulated human PBMC (CD14- / CD56-). Target cells: Human CDH3 transfected CHO cells. Effector cell to target cell (E: T) ratio: 10: 1. Antibodies specific for epitope clusters / extracellular subdomains D2C (Fig. 8A) and D3A (Fig. 8B).</figref><figref num="9">The cytotoxic activity of unstimulated human PBMCs against human CDH3-positive epidermoid carcinoma cell line A431 in the presence of CDH3 × CD3 bispecific antibodies, as measured by a 48-hour FACS-based cytotoxic assay. Effector cells: Unstimulated human PBMC (CD14- / CD56-). Target cells: Human A431 cells. Effector cell to target cell (E: T) ratio: 10: 1. Antibodies specific for epitope clusters / extracellular subdomains D2C (Fig. 9A) and D3A (Fig. 9B).</figref><figref num="10A">The cytotoxic activity of the Makaku T cell line against Makaku CDH3-transfected CHO cells in the presence of the CDH3 × CD3 bispecific antibody, as measured by a 48-hour FACS-based cytotoxic assay. Effector cells: Macaque CD3-positive T cell line LnPx4119. Target cells: Macaque CDH3 transfected CHO cells. Effector cell to target cell (E: T) ratio: 10: 1. Antibodies specific for epitope cluster / extracellular subdomain D2C.</figref><figref num="10B">The cytotoxic activity of the Makaku T cell line against Makaku CDH3-transfected CHO cells in the presence of the CDH3 × CD3 bispecific antibody, as measured by a 48-hour FACS-based cytotoxic assay. Effector cells: Macaque CD3-positive T cell line LnPx4119. Target cells: Macaque CDH3 transfected CHO cells. Effector cell to target cell (E: T) ratio: 10: 1. Antibodies specific for epitope cluster / extracellular subdomain D3A.</figref><figref num="10C">The cytotoxic activity of the Makaku T cell line against Makaku CDH3-transfected CHO cells in the presence of the CDH3 × CD3 bispecific antibody, as measured by a 48-hour FACS-based cytotoxic assay. Effector cells: Macaque CD3-positive T cell line LnPx4119. Target cells: Macaque CDH3 transfected CHO cells. Effector cell to target cell (E: T) ratio: 10: 1. Antibodies specific for epitope cluster / extracellular subdomain D1B.</figref><figref num="11">The antitumor activity of CDH3 × CD3 bispecific antibodies of epitope cluster / extracellular subdomain D2C (CDH3-13) in a human tumor xenograft model is shown (see Example 14). The antibody dose-dependently prevents the formation of A-431 tumors in the presence of human PBMC. The tumor volume was large at the start of measurement (day 5) because a large amount of cell mixture was injected on day 1. ** p &lt;0.01; *** p &lt;0.0001.</figref><figref num="12A">Shows antitumor activity of CDH3 × CD3 bispecific and extended half-life (HLE) antibody of epitope cluster / extracellular subdomain D2C (CDH3-13) in human tumor xenograft model (see Example 15). ). HLE antibodies dose-dependently prevent the formation of human HCT-116 tumors in the presence of human PBMCs. Figure 12A shows the overall results.</figref><figref num="12B">Shows antitumor activity of CDH3 × CD3 bispecific and prolonged half-life (HLE) antibody of epitope cluster / extracellular subdomain D2C (CDH3-13) in human tumor xenograft model (see Example 15). ). HLE antibodies dose-dependently prevent the formation of human HCT-116 tumors in the presence of human PBMCs. Figure 12B distinguishes the results obtained for higher antibody concentrations (Group 2) into responding animals (7/10) and non-responsive animals (3/10).</figref><figref num="13">T cell activation in the absence of target cells with bispecific antibody constructs in the absence (upper panel) and presence (lower panel) of albumin fusion at the C-terminus of the construct is shown.</figref>

The following examples describe the invention. These examples should not be construed as limiting the scope of the invention. The present invention is limited only by the claims.

Example 1 Generation of CHO cells expressing wild-type and chimeric CDH3 Sequences of each of the five extracellular domains Dom1 to Dom5 (or D1 to D5) of human CDH3 and them for the construction of chimeric molecules used for epitope mapping. Subdomains (A, B and C) of were replaced with the corresponding mouse sequences. The following 20 molecules were generated; see also Figure 1 and Figure 2. Hu CDH3 / Dom1 mu (aa 108-215) SEQ ID NO: 13 Hu CDH3 / Dom1A mu (aa 108-143) SEQ ID NO: 14 Hu CDH3 / Dom1B mu (aa 144-179) SEQ ID NO: 15 Hu CDH3 / Dom1C mu (aa 180-215) SEQ ID NO: 16 Hu CDH3 / Dom2 mu (aa 216-327) SEQ ID NO: 17 Hu CDH3 / Dom2A mu (aa 216-252) SEQ ID NO: 18 Hu CDH3 / Dom2B mu (aa 253-290) SEQ ID NO: 19 Hu CDH3 / Dom2C mu (aa 291-327) SEQ ID NO: 20 Hu CDH3 / Dom3 mu (aa 328-440) SEQ ID NO: 21 Hu CDH3 / Dom3A mu (aa 328-363) SEQ ID NO: 22 Hu CDH3 / Dom3B mu (aa 364-403) SEQ ID NO: 23 Hu CDH3 / Dom3C mu (aa 404-440) SEQ ID NO: 24 Hu CDH3 / Dom4 mu (aa 441-546) SEQ ID NO: 25 Hu CDH3 / Dom4A mu (aa 441-474) SEQ ID NO: 26 Hu CDH3 / Dom4B mu (aa 475-511) SEQ ID NO: 27 Hu CDH3 / Dom4C mu (aa 512-546) SEQ ID NO: 28 Hu CDH3 / Dom5 mu (aa 547-650) SEQ ID NO: 29 Hu CDH3 / Dom5A mu (aa) 547-581) SEQ ID NO: 30 Hu CDH3 / Dom5B mu (aa 582-616) SEQ ID NO: 31 Hu CDH3 / Dom5C mu (aa 617-650) SEQ ID NO: 32

The above list shows the positions of the different domains (Dom1 to Dom5) as well as the positions of the respective subdomains A to C within the amino acid sequence of human CDH3 as shown in SEQ ID NO: 1. For example, the subdomain D1A is located at amino acids 108-143 of SEQ ID NO: 1. The same applies to all other domains listed above.

For expression in CHO cells, the coding sequence for the chimeric extracellular domain is followed by the coding sequence for the artificial Ser / Gly linker in frame, followed by the transmembrane / intracellular domain of human EpCAM (GenBank accession number NM_002354). It is followed by a domain derived from amino acids 266-314) in a sequence as published. All chimeric constructs contained an N-terminal signal sequence (signal peptide) and a propeptide.

Human CDH3 (SEQ ID NO: 2, see also GenBank accession number NM_001793), cynoCDH3 (SEQ ID NO: 2) for the generation of CHO cells expressing human, cynomolgus monkey (cyno), mouse and human / mouse chimeric CDH3. : 6), mouse CDH3 (SEQ ID NO: 10, see also GenBank accession number NM_001037809) and 20 human-mouse CDH3 chimeras (see above), respectively, with pEF-DHFR (pEF-). DHFR was cloned into a plasmid called Raum et al. Cancer Immunol Immunother 50 (2001) 141-150). A human sequence-specific oligonucleotide that hybridizes the coding sequence of macaque CDH3 to the untranslated or conserved sequence region of the cynomolgus monkey spleen cDNA library (BioChain) and human CDH3 mRNA transcript (NM_001793).<img file="JP6907124B2_D0002.tif" />Obtained by standard cloning using. Sequence analysis revealed amino acid sequence similarity in the core extracellular domain compared to the rhesus CDH3GenBank sequence (JU473826, JU473827). All cloning procedures were performed according to standard protocols (Sambrook, Molecular Cloning; A Laboratory Manual, 3rd edition, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York (2001)). For each construct, the corresponding plasmid was transfected into DHFR-deficient CHO cells for eukaryotic expression as described by Kaufman RJ (1990) Methods Enzymol. 185,537-566.

Expression of CDH3 (human, mouse and chimeric constructs) on CHO cells was validated by FACS assay using a mouse cross-reactive monoclonal mouse IgG1 anti-human CDH3 antibody. The bound monoclonal antibody was detected with anti-mouse IgG Fcγ-PE. As a negative control, cells were incubated with PBS / 2% FCS instead of the first antibody. The sample was measured by flow cytometry. The results are shown in Figure 3. Expression of human and cynoCDH3 on CHO cells (see Example 5) was detected with PE-complexed R &amp; D 861-P.

Example 2 Epitope clustering of mouse scFv fragments ScFv that binds human or mouse CDH3-transfected cells or chimeric human / mouse CDH3 molecule-transfected cells (see Example 1) to human / macaque CDH3. The cells were stained with the containing crude undiluted periplasmic extract. The bound scFv molecule was subjected to mouse monoclonal anti-FLAG-M2 antibody (1 μg / ml; 50 μl in PBS / 2% FCS; Sigma F1804), followed by anti-mouse IgG Fcγ-PE (1: 100, 50 μl; Jackson Immunoresearch No. 115). Detected in -116-071). All antibodies were diluted with PBS with 2% FCS. As a negative control, cells were incubated with PBS / 2% FCS instead of periplasmic extract. The sample was measured by flow cytometry. The results are shown in Fig. 4.

Specifically, FIG. 4A shows a binder that recognizes the extracellular domain D1 of human CDH3, or more precisely the subdomain D1B (loss of FACS signal in each chimeric CDH3 construct). Note that the binder called CDH3-6 is the parent binder of CDH3-4. The binder referred to as CDH3-10 is the parent binder of CDH3-1, CDH3-2 and CDH3-3. FIG. 4B shows a binder that recognizes the extracellular domain D2 of human CDH3, or more precisely the subdomain D2C. Note that the binder referred to as CDH3-21 is the parent binder for CDH3-11, CDH3-12 and CDH3-14. The binder called CDH3-23 is the parent binder of CDH3-13. Finally, FIG. 4C shows a binder that recognizes the extracellular domain D3 of human CDH3, or more precisely the subdomain D3A. Note that the binder CDH3-32 further binds to the subdomain D3C. The binder referred to as CDH3-32 is the parent binder for CDH3-25, CDH3-26 and CDH3-27. The binder called CDH3-33 is the parent binder of CDH3-24. The term "parent binder" means, in this context, a binder that has been further developed to produce or obtain an optimized binder.

Similar epitope clustering analyzes on the binders CDH3-11, CDH3-12, CDH3-13 and CDH3-14 showed that they recognized the extracellular domain D2 of human CDH3, or more precisely the subdomain D2C. It was shown to do (no data shown). The same analysis was further performed with the binders CDH3-24, CDH3-25, CDH3-26 and CDH3-27. These binders recognized the extracellular domain D3 of human CDH3, or more precisely the subdomain D3A (data not shown).

Example 3 Biacore-based determination of antibody affinity for human and cynomolgus monkey CDH3 Biacore analysis experiments were performed using recombinant CDH3 (human and cynoCDH3) fusion proteins with human albumin (HALB), respectively, and the antibodies of the invention. CDH3 target binding was determined.

Specifically, CM5 sensor chips (GE Healthcare) were immobilized at approximately 600-800 RU for each recombinant antigen using acetate buffer pH 4.5 according to the manufacturer's instructions. CDH3 × CD3 bispecific antibody samples were diluted in HBS-EP running buffer (GE Healthcare) and loaded at five concentrations: 50 nM, 25 nM, 12.5 nM, 6.25 nM and 3.13 nM. The flow rate was 30 μl / min for 3 minutes, then HBS-EP running buffer was applied again for 8-20 minutes at a flow rate of 30 μl / ml. Chip regeneration was performed using a pH 1.5 solution of 10 mM glycine 10 mM NaCl. The dataset was analyzed using BiaEval software. In general, two independent experiments were performed.

The CDH3 × CD3 bispecific antibody according to the present invention showed high affinity for human CDH3 in the single digit nanomolar concentration range. The binding to macaque CDH3 was balanced, but it also showed a similar range of affinity. Table 2 shows the calculated affinity gap in addition to the affinity value. CDH3-25 × F12q-HALB and CDH3-13- × I2C-HLE (Fc) were measured in separate assays, respectively: · 31.8 ± 1.9 nM KD (hu), 40.9 ± 3.89 nM KD (cyno) and It was shown to have an affinity gap of 1.29 and an affinity gap of 12.95 ± 0.5 nM KD (hu), 12.35 ± 0.35 nM KD (cyno) and 0.95, respectively.

(Table 2) Affinity of CDH3 × CD3 bispecific antibody to human and macaque CDH3 determined by Biacore analysis, and calculated interspecific affinity gap<img file="JP6907124B2_D0003.tif" />

Example 4 Scatchard-based analysis of CDH3 × CD3 bispecific antibody affinity for human and macaque CDH3 on target antigen-positive cells and determination of interspecific affinity gaps for human or macaque CDH3-transfected CHO cells The affinity of the CDH3 × CD3 bispecific antibody was also determined using Scatchard analysis as the most reliable method of measuring the potential affinity gap between human CDH3 and macaque CDH3. In Scatchard analysis, saturated binding experiments are performed using a monovalent detection system to accurately determine monovalent binding between CDH3 × CD3 bispecific antibodies to each cell line.

Each cell line of 2 × 104 cells (recombinant human CDH3-expressing CHO cell line, recombinant Macaku CDH3-expressing CHO cell line) is a 50 μl triplet of each CDH3 × CD3 bispecific antibody initiated at 10-20 nM. Each was incubated with a dilution series (12 dilutions at 1: 2) (until saturation was reached), followed by a 16-hour incubation under stirring at 4 ° C., with one residue washing step. The cells were then incubated with 30 μl of CD3 × ALEXA488 complex solution for an additional hour. After one wash step, cells are resuspended in 150 μl FACS buffer containing 3.5% formaldehyde, incubated for an additional 15 minutes, centrifuged, resuspended in FACS buffer, and the FACS Canto II apparatus. And FACS Analyzed using Diva software. Data were generated from two independent experimental sets, each repeated 3 times. Each Scatchard analysis value was calculated and the maximum binding (Bmax) was extrapolated. The concentration of CDH3 × CD3 bispecific antibody at semimaximal binding was determined by reflecting each KD. The values of the three repeated measurements were plotted as a hyperbola and an S-curve to demonstrate an appropriate concentration range from minimal to optimal coupling.

The values shown in Table 3 were obtained from two independent experiments for each CDH3 × CD3 bispecific antibody. In cell-based Scatchard analysis, the CDH3 × CD3 bispecific antibodies of the invention have nanomolar to sub-nanomolar affinities for human CDH3, with a small cyno / human interspecific CDH3 affinity gap of approximately 1. Confirmed to present. CDH3-25 × F12q-HALB and CDH3-13 × I2C-HLE (Fc) were measured in separate assays, respectively, to a cell-based affinity (hu) of 0.37 ± 0.11 nM, 0.35 ± 0.08 nM cells. Based on affinity (cyno) and 0.95 affinity gap, and 0.32 ± 0.003 nM cell-based affinity (hu), 0.5 ± 0.09 nM cell-based affinity (cyno) and 1.56 affinity gap, respectively. Shown to have.

(Table 3) CDH3 × CD3 bispecific antibody affinity (KD) determined by cell-based Scatchard analysis and calculated affinity gap KD macaque CDH3 / KD human CDH3. Antibodies were measured in two independent experiments, each repeated 3 times.<img file="JP6907124B2_D0004.tif" />

Example 5 Bispecific binding and interspecific cross-reactivity To confirm binding to human CDH3 and CD3 and binding to cynoCDH3 and CD3, bispecific antibodies were transfected, respectively, for human and cynoCDH3. Using CHO cells, human CDH3-positive epidermal cancer cell line A431, CD3-expressing human T-cell leukemia cell line HPB-all (DSMZ, Braunschweig, ACC483), and crab monkey CD3-expressing T cell line HSC-F. Was tested by flow cytometry. In addition, mouse CDH3-transfected CHO cells were used as a negative control.

For flow cytometry, each cell line of 200,000 cells was incubated on ice for 30 minutes at a concentration of 5 μg / ml with 50 μl of purified bispecific antibody. Cells were washed twice in PBS / 2% FCS and construct binding was detected with in-house mouse antibody specific for the CD3 binding moiety. After washing, bound mouse antibody was detected with goat anti-mouse Fcγ-PE. The sample was measured by flow cytometry.

The results are shown in FIGS. 5A and 5B. The CDH3 × CD3 bispecific antibodies of the invention stain human CDH3 and CHO cells transfected with cynoCDH3, which are also human in addition to human CDH3-positive epidermal carcinoma cell line A431 (naturally expressed). And cynoT cell expression CD3 were also stained. In addition, negative control cells (mouse CDH3 transfected CHO) were not stained.

Example 6 Confirmation of absence of binding to human and macaque paralogs Human and macaque CDH3 paralogs (CDH1 = E-cadherin, CDH2 = N-cadherin, CDH4 = R-cadherin, and CDH5 = VE-cadherin), dhfr^-/-Stable transfection into CHO cells. Protein expression was confirmed by FACS analysis using antibodies specific for each paralog. Antibodies were R &amp; D MAB18381 (for CDH1), eBioscience 12-3259-41 (for CDH2), R &amp; D Systems polyclonal AF2217 (for CDH4) and BD Bioscience number 555661 (for CDH5).

The paralog sequence used in this example is identified in the sequence listing (SEQ ID NO: 41-44). They can also be found in the GenBank accession numbers below. NM_004360 [Human CDH1]

NM_001792 [Human CDH2]

NM_001794 [Human CDH4]

NM_001795 [Human CDH5]

A flow cytometry assay was performed as described in Example 5. The results are shown in Figures 5C and 5D. Analysis confirmed that none of the CDH3 x CD3 bispecific antibodies of the invention cross-reactive with any of the human CDH3 paralogs tested.

It was further verified that the antibody of the present invention did not cross-react with CDH1, CDH2, CDH4, and CDH5 of the macaque CDH3 paralog (data not shown). Makaku paralog expression on CHO cells was verified with the same antibody as described above for human paralog. The sequence of the macaque paralog used in this example is identified in the sequence listing (SEQ ID NOs: 45 to 48). They can also be found in the GenBank accession numbers below. XM_002802516 [Macaque CDH1]

JU321883 [Macaque CDH2]

XM_002802511 [Macaque CDH5]

In addition, the macaque CDH4 sequence was obtained from the Ensembl Genome Browser (ENSMMUT00000017252) and fused in-frame with the human CDH4 signal peptide (amino acids 1-19) at the N-terminus.

Example 7: Identity with Human Germline To analyze the identity / similarity of antibody sequences to human germline genes, the CDH3 binders of the invention were aligned as follows: all CDRs. Complete VL containing was aligned to the human antibody germline gene (Vbase); complete VH containing CDR1 and but excluding CDR3 was aligned to the human antibody germline gene (Vbase). Further details can be found in the specification of this application. The results are shown in Table 4 below.

(Table 4) Identity of VH and VL to human germline<img file="JP6907124B2_D0005.tif" />

Example 8 Cytotoxic activity The efficacy of CDH3 × CD3 bispecific antibodies of the invention in redirecting effector T cells against CDH3-expressing target cells was analyzed in five in vitro cytotoxicity assays: Stimulation of human CDH3-transfected CHO cells Human CD8 + effectors The efficacy of CDH3 × CD3 bispecific antibodies on T cell redirection was measured by an 18-hour 51 chromium release assay. See Figure 6 for the results. The efficacy of CDH3 × CD3 bispecific antibodies on stimulation of human CD8 + effector T cells against CDH3-positive human carcinoma cell line A431 was measured by an 18-hour 51 chromium release assay. See Figure 7 for the results. The efficacy of CDH3 × CD3 bispecific antibodies on T cell redirection in unstimulated human PBMCs against human CDH3 transfected CHO cells was measured by a 48-hour FACS-based cytotoxic assay. See Figure 8 for the results. The efficacy of CDH3 × CD3 bispecific antibodies on T cell redirection in unstimulated human PBMCs against CDH3-positive human carcinoma cell line A431 was measured by a 48-hour FACS-based cytotoxic assay. See Figure 9 for the results. A 48-hour FACS-based cytotoxic assay was performed to confirm that cross-reactive CDH3 x CD3 bispecific antibodies were able to redirect Makaku T cells against Makaku CDH3 transfected CHO cells. , The Makaku T cell line was used as the effector T cell. See Figure 10 for the results.

Example 8.1 Chromium release assay using stimulated human T cells CD8⁺Stimulated T cells enriched for T cells were obtained as described below. Petri dishes (145 mm in diameter, Greiner bio-one GmbH, Kremsmünster) were coated with commercially available anti-CD3 specific antibodies (OKT3, Orthoclone) at a final concentration of 1 μg / ml at 37 ° C. for 1 hour. Unbound proteins were removed by a single wash step with PBS. 3 ~ 5 × 10⁷Human PBMCs were added to 120 ml RPMI 1640 precoated Petri dishes containing stabilized glutamine / 10% FCS / IL-2 20 U / ml (Proleukin®, Chiron) and stimulated for 2 days. Cells were collected on day 3 and washed once with RPMI 1640. IL-2 was added to a final concentration of 20 U / ml and the cells were cultured again in the same cell culture medium as above for 1 day. CD8⁺Cytotoxic T lymphocytes (CTL), using Dyna-Beads according to the manufacturer's protocol, CD4⁺T cells and CD56⁺Concentrated by depletion of NK cells.

CynoCDH3 or human CDH3 transfected CHO target cells were washed twice with PBS to 11.1 MBq.⁵¹Labeled with Cr in RPMI in a final volume of 100 μl containing 50% FCS at 37 ° C for 60 minutes. The labeled target cells were then washed 3 times with 5 ml RPMI and then used in the cytotoxicity assay. The assay was performed in a 96-well plate in a total volume of 200 μl of supplemented RPMI with a 10: 1 E: T ratio. Purified bispecific antibodies with an starting concentration of 0.01-1 μg / ml and 3-fold dilutions thereof were used. The assay incubation time was 18 hours. Cytotoxicity was determined as a relative value of chromium released in the supernatant compared to the difference between maximal lysis (addition of Triton-X) and spontaneous lysis (without effector cells). All measurements were repeated 4 times. Chromium activity measurements in the supernatant were performed on a Wizard 3-inch gamma counter (Perkin Elmer Life Sciences GmbH, Cologne, Germany). Analysis of results, Prism 5 for Windows (version 5.0, GraphPad Software) Inc., San Diego, CA, USA). The EC50 value calculated by the analytical program from the sigmoid dose-response curve was used for comparison of cytotoxic activity.

Example 8.2 Stimulation of human CDH3-transfected CHO cells Efficacy for redirection of human effector T cells The cytotoxic activity of the CDH3 × CD3 bispecific antibody according to the present invention was used as a target cell for human CDH3-transfected CHO cells, and Using stimulated human CD8 + T cells as effector cells, 51 chromium (⁵¹Cr) Released by cytotoxicity assay. The experiment was carried out as described in Example 8.1.

The results are shown in Figure 6 and Table 5. CDH3 x CD3 bispecific antibodies showed potent cytotoxic activity against human CDH3 transfected CHO cells even when reduced to a single digit picomolar concentration range. The claimed antibody is specific for the epitope cluster corresponding to positions 291 to 363 of human CDH3 and presents a good epitope activity relationship supporting strong bispecific antibody-mediated cytotoxic activity. CDH3-25 × F12q-HALB and CDH3-13 × I2C-HLE (Fc) were measured in separate assays, respectively, and were shown to have EC50s of 3.6pM and 8.9pM, respectively.

(Table 5) 51 chromium (Table 5) using CHO cells transfected with human CDH3 as target cells and stimulated human CD8T cells as effector cells.⁵¹Cr) EC50 value of CDH3 × CD3 bispecific antibody analyzed by released cytotoxicity assay [pg / ml]<img file="JP6907124B2_D0006.tif" />

Example 8.3 Stimulation against CDH3-positive human cancer tumor line A431 Efficacy for redirection of human effector T cells CDH3-positive human epidermal cancer cell line A431, using the cytotoxic activity of CDH3 × CD3 bispecific antibody as a source of target cells, And using stimulated human CD8 + T cells as effector cells, 51 chromium (⁵¹Cr) Released by cytotoxicity assay. The assay was performed as described in Example 8.1.

Consistent with the results of the 51 chromium release assay using stimulated concentrated human CD8 + T lymphocytes as effector cells and human CDH3 transfected CHO cells as target cells, the CDH3 × CD3 bispecific antibody of the invention is also naturally expressed. Strong cytotoxic activity against target cells (Fig. 7 and Table 6). CDH3-25 × F12q-HALB and CDH3-13 × I2C-HLE (Fc) were measured in separate assays, respectively, and were shown to have EC50s of 1.2pM and 16pM, respectively.

(Table 6) Using CDH3-positive human carcinoma cell line A431 as a source of target cells and stimulating concentrated human CD8T cells as effector cells, 51 chromium for 18 hours (Table 6)⁵¹Cr) EC50 value of CDH3 × CD3 bispecific antibody analyzed by released cytotoxicity assay [pg / ml]<img file="JP6907124B2_D0007.tif" />

Example 8.4 FACS-based cytotoxic assay using unstimulated human PBMC Isolation of effector cells Human peripheral blood mononuclear cells (PBMC) are lysed into a concentrated lymphocyte preparation (buffy coat), ie blood for transfusion. Was prepared by Ficoll density gradient centrifugation from blood bank by-products. Buffy coats were supplied by the local blood bank and PBMCs were prepared on the same day as blood collection. After extensive washing with Ficoll density centrifugation and Dulbecco PBS (Gibco), residual erythrocytes are removed in erythrocyte lysis buffer (155 mM NH).₄Cl, 10mM KHCO₃, 100 μM EDTA) was removed from PBMC. Platelets were removed from the supernatant when the PBMC was centrifuged at 100 xg. Residual lymphocytes mainly include B and T lymphocytes, NK cells and monocytes. PBMC 37 ° C / 5% CO₂Was maintained in culture in RPMI medium (Gibco) containing 10% FCS (Gibco).

CD14⁺And CD56⁺Cell depletion CD14⁺Human CD14 microbeads (Milteny Biotec, MACS, number 130-050-201) are used for cell depletion, and human CD56 microbeads (MACS, number 130-050-401) are used for NK cell depletion. bottom. PBMCs were counted and centrifuged at 300 xg at room temperature for 10 minutes. Discard the supernatant and dispose of the cell pellet in MACS isolation buffer [80 μL / 10].⁷Cells; resuspended in PBS (Invitrogen, number 20012-043), 0.5% (v / v) FBS (Gibco, number 10270-106), 2 mM EDTA (Sigma-Aldrich, number E-6511)]. CD14 microbeads and CD56 microbeads (20 μL / 10)⁷Cells) were added and incubated for 15 minutes at 4-8 ° C. MACS isolation buffer (1-2 mL / 10) for cells⁷Washed with cells). After centrifugation (see above), discard the supernatant and isolate cells in MACS isolation buffer (500 μL / 10).⁸Resuspended in cells). CD14 / CD56 negative cells were then isolated using an LS column (Miltenyi Biotec, number 130-042-401). PBMC without CD14 + / CD56 + cells in RPMI complete medium, ie 10% FBS (Biochrom AG, number S0115), 1 x non-essential amino acids (Biochrom AG, number K0293), 10 mM Hepes buffer (Biochrom AG, number). In the incubator at 37 ° C in RPMI1640 (Biochrom AG, number FG1215) supplemented with L1613), 1 mM sodium pyruvate (Biochrom AG, number L0473) and 100 U / mL penicillin / streptomycin (Biochrom AG, number A2213). Incubated until needed.

Fluorescent Membrane Dye DiOC for Analysis of Cytolysis in Targeted Cell Labeled Flow Cytometry Assay₁₈(DiO) (Molecular Probes, number V22886) was used to label human CDH3 or macaque CDH3 transfected CHO cells as target cells and distinguish them from effector cells. Briefly, cells were harvested and washed once with PBS with 2% (v / v) FBS and membrane dye DiO (5 μL / 10).⁶10 in PBS containing cells)⁶Adjusted to cells / mL. After incubation at 37 ° C for 3 minutes, cells were washed twice in complete RPMI medium to increase the number of cells to 1.25 × 10.⁵Adjusted to cells / mL. Cell vitality was measured using 0.5% (v / v) isotonic Eosin G solution (Roth, number 45380).

Flow Cytometry-Based Analysis This assay was designed to quantify lysis of cyno or human CDH3 transfected CHO cells in the presence of serial dilutions of CDH3 bispecific antibody. Equal volumes of DiO-labeled target cells and effector cells (ie, CD14)⁺Cell-free PBMCs) were mixed to give an E: T cell ratio of 10: 1. 160 μL of this suspension was transferred to each well of a 96-well plate. A serial dilution of 40 μL of CDH3 × CD3 bispecific antibody and negative control bispecific (CD3-based bispecific antibody that recognizes unrelated target antigens) or additional RPMI complete medium as a negative control was added. .. Bispecific antibody-mediated cytotoxic response is 7% CO₂It proceeded for 48 hours in a humidifying incubator. Cells were then transferred to a new 96-well plate and 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 pigment that is normally eliminated from living cells, whereas dead cells take it up and can be identified by fluorescent radiation.

Samples were measured by flow cytometry on a FACSCanto II instrument and analyzed by FACSDiva software (both from Becton Dickinson). Target cells were identified as DiO positive cells. PI-negative target cells were classified as target live cells. The percentage of cytotoxicity was calculated by the following formula.<img file="JP6907124B2_D0008.tif" />

Using GraphPad Prism 5 software (San Diego), the percentage of cytotoxicity was plotted against the corresponding bispecific antibody concentration. The dose-response curve was analyzed using a four-parameter logistic regression model to evaluate the sigmoid dose-response curve with a fixed hill slope, and EC50 values were calculated.

Example 8.5 Efficacy for redirection of unstimulated human PBMC to human CDH3-transfected CHO cells The cytotoxic activity of the CDH3 × CD3 bispecific antibody was not applied to human CDH3-transfected CHO cells as target cells and effector cells. Stimulated human PBMCs were used and analyzed in a FACS-based cytotoxicity assay. The assay was performed as described in Example 8.4 above.

The results of a FACS-based cytotoxicity assay using unstimulated human PBMCs as effector cells and human CDH3 transfected CHO cells as targets are shown in FIGS. 8 and 7. CDH3-25 × F12q-HALB and CDH3-13 × I2C-HLE (Fc) were measured in separate assays, respectively, and were shown to have EC50s of 4.6pM and 5.6pM, respectively.

(Table 7) EC50 of CDH3 × CD3 bispecific antibody measured by a 48-hour FACS-based cytotoxicity assay using unstimulated human PBMC as effector cells and human CDH3 transfected CHO cells as target cells. Value [pg / ml]<img file="JP6907124B2_D0009.tif" />

Example 8.6 Efficacy for redirection of unstimulated human PBMC against CDH3-positive human carcinoma cell line A431 CDH3-positive human epidermal carcinoma cell line A431 and effector using the cytotoxic activity of CDH3 × CD3 bispecific antibody as a source of target cells Unstimulated human PBMCs were used as cells for further analysis in a FACS-based cytotoxicity assay. The assay was performed as described in Example 8.4 above. The results are shown in Figure 9 and Table 8. CDH3-25 × F12q-HALB and CDH3-13 × I2C-HLE (Fc) were measured in separate assays, respectively, and were shown to have EC50s of 2.3pM and 32pM, respectively.

(Table 8) EC50 values of CDH3 × CD3 bispecific antibodies measured by a 48-hour FACS-based cytotoxic assay using unstimulated human PBMCs as effector cells and human A431 cell lines as a source of target cells. [pg / ml]<img file="JP6907124B2_D0010.tif" />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.

Example 8.7 Efficacy for redirection of Macaku T cells against Macaku CDH3-expressing CHO cells Finally, the cytotoxic activity of the CDH3 × CD3 bispecific antibody was used as target cells in CHO cells transfected with Macaku (cyno) CDH3, and Makaku T cell lines were used as a source of effector cells and analyzed in a FACS-based cytotoxicity assay. The macaque T cell line 4119LnPx (Knappe et al. Blood 95: 3256-61 (2000)) was used as a source of effector cells. Macaque CDH3 transfection analysis based on flow cytometry target cell labeling and cytotoxic activity of CHO cells was performed as described above.

The results are shown in Figure 10 and Table 9. Macaque T cells derived from cell line 4119LnPx correspond to the CDH3 × CD3 bispecific antibody of the invention, ie, human CDH3 at positions 291 to 363 and adjacent subdomains D2C (positions 291 to 327) and D3A (328). An antibody that binds to an epitope cluster of human CDH3 containing (~ 363) was induced to efficiently kill macaque CDH3-transfected CHO cells. The antibodies strongly presented EC50 values in the 2-digit to very low 4-digit pg / ml in this assay, confirming that these antibodies are very active in the macaque system.

Another group of anti-CDH3 antibodies has been identified during epitope clustering (see Example 2), which bind to the extracellular domain D1 of human CDH3, more specifically to the subdomain D1B. Unexpectedly, this group of CDH3 × CD3 bispecific antibodies has strong cytotoxic activity against human CDH3-transfected CHO cells, but very weak against Makaku CDH3-transfected CHO cells. Demonstrated to show cytotoxic activity (see Figure 10C and Table 9). Antibodies in this group showed significantly weaker efficacy at very high 4-digit pg / ml ranges and even 5-digit pg / ml range EC50 values. CDH3-25 × F12q-HALB and CDH3-13 × I2C-HLE (Fc) were measured in separate assays, respectively, and were shown to have EC50s of 1.4pM and 4.0pM, respectively.

Therefore, the CDH3 × CD3 antibody of the present invention that binds to the epitope cluster of CDH3 corresponding to positions 291 to 363 is more specific to the extracellular domain D1 than the antibody that binds to the CDH3 subdomain D1B in the macaque system. It is about 5 to 1000 times more powerful.

(Table 9) CDH3 corresponding to positions 291 to 363 measured by a 48-hour FACS-based cytotoxic assay using Macaku T cell line 4119LnPx as an effector cell and Macaku CDH3 transfected CHO cells as target cells. EC50 value [pg / ml] of CDH3 × CD3 bispecific antibody (rows 1-8) that binds to epitope clusters and CDH3 × CD3 bispecific antibody (rows 9 ~) that binds to CDH3 epitope cluster / subdomain D1B 12) EC50 value [pg / ml]<img file="JP6907124B2_D0011.tif" />

Example 9 (i) Monomer to dimer conversion after 3 freeze / thaw cycles and (ii) 7 days incubation at 250 μg / ml Bispecific CDH3 × CD3 antibody monomer After exposure to different stress conditions, a high performance SEC was performed to determine the percentage of initial monomeric antibody, which was converted to antibody dimer.

(i) 15 μg of monomeric antibody was adjusted to a concentration of 250 μg / ml using common pharmaceutical buffers, then frozen at -80 ° C for 30 minutes, followed by thawing at room temperature for 30 minutes. .. After 3 freeze / thaw cycles, the dimer content was determined by HP-SEC.

(ii) 15 μg of monomeric antibody was adjusted to a concentration of 250 μg / ml using common pharmaceutical buffers, followed by incubation at 37 ° C. for 7 days. The dimer content was determined by HP-SEC.

A high resolution SEC column TSK Gel G3000 SWXL (Tosoh, Tokyo, Japan) was connected to an Akta Purifier 10 FPLC (GE Lifesciences) equipped with an A905 autosampler. The column equilibration and running buffer consisted of 100 mM KH2PO4 and 200 mM Na2SO4 adjusted to pH 6.6. An antibody solution (15 μg protein) was applied to the equilibration 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 with absorbances at 280, 254 and 210 nm. The analysis was performed by the peak integral of the 210 nm signal recorded on the Akta Unicorn software execution evaluation sheet. The dimer content was calculated by dividing the dimer peak area by the total peak area of the monomer and the dimer.

The results are shown in Table 10 below. CDH3 × CD3 bispecific antibodies that bind to epitope clusters / extracellular subdomain D2C are 1% dimers after 3 freeze / thaw cycles plus 7 days of incubation at 37 ° C. This is considered to be very good, as we have presented a percentage, or more accurately a dimer percentage of 0.0%. This is considered good as the dimer conversion of the CDH3 × CD3 bispecific antibody of the epitope cluster / extracellular subdomain D3A reached a value of 2%, or more accurately 0.2-1.8%. Be done. CDH3-25 × F12q-HALB and CDH3-13 × I2C-HLE (Fc) were measured in separate assays, respectively, with dimer percentages of 1.1 and 0.84, respectively, after 3 freeze / thaw cycles, and for 7 days. It was shown to have a dimeric percentage of 0.0 (both HLE constructs) after incubation.

(Table 10) Monomer vs. dimer CDH3 x CD3 bispecific antibody percentage measured by high performance size exclusion chromatography (HP-SEC)<img file="JP6907124B2_D0012.tif" />

Example 10 Thermal stability The antibody aggregation temperature was determined as follows: 40 μl of antibody solution at 250 μg / ml was transferred to a single-use cuvette and placed in a Wyatt dynamic light scattering device DynaPro Nanostar (Wyatt). .. The sample was heated to 40 ° C to 70 ° C at a heating rate of 0.5 ° C / min, and the measured radius was obtained constantly. The increase in radius indicating protein melting and aggregation was used with the software package supplied with the DLS device to calculate antibody aggregation temperature.

All CDH3 × CD3 bispecific antibodies of the invention tested showed very good thermal stability with aggregation temperatures above 54 ° C, as shown in Table 11 below. CDH3-25 × F12q-HALB was measured in a separate assay and was shown to have thermal stability of 56.3 ° C.

(Table 11) Thermal stability of bispecific antibodies determined by DLS (Dynamic Light Scattering)<img file="JP6907124B2_D0013.tif" />

Example 11 Stability after 24-hour incubation in human plasma Purified bispecific antibodies in a 1: 5 ratio in a human plasma pool at 37 ° C for 24-96 hours, 2-20 μg / ml. Incubated at final concentration. After plasma incubation, the antibody was combined with stimulated human T cells and CDH3 transfected CHO cells in a 51 chromium release assay at a starting concentration of 0.01-0.1 μg / ml in a 10: 1 effector cell to target cell (E: T) ratio. (The assay described in the chromium release assay of Example 8.1 using stimulated human T cells). Fresh thawed bispecific antibodies that were not incubated were included as controls. The results are shown in Table 12. All antibodies tested had good plasma stability (EC) of 4₅₀Plasma / EC₅₀The group of antibodies that had (control) and bound to D3A further had plasma stability of 3. CDH3-25 × F12q-HALB and CDH3-13 × I2C-HLE (Fc) were measured in separate assays, respectively: EC50 with 4.4pM plasma, EC50 without 3.6pM plasma, and 1.2 plasma. It was shown to have a contrast ratio and a plasma contrast ratio of EC50 with 3.4 pM plasma, EC50 without 8.9 pM plasma, and 0.4 plasma, respectively.

(Table 12) EC50 value of antibody with and without plasma incubation and calculated plasma / control value<img file="JP6907124B2_D0014.tif" />

Example 12 Turbidity at an antibody concentration of 2500 μg / ml A 250 μg / ml purified monomeric antibody solution of 1 ml was concentrated to 2500 μg / ml with a spin concentration unit. After storage at 5 ° C. for 16 hours, the turbidity of the antibody solution was measured by OD340 nm light absorption measurement with respect to a general pharmaceutical buffer. The results are shown in Table 13 below. All antibodies tested have a very good turbidity of 0.05. CDH3-25 × F12q-HALB and CDH3-13 × I2C-HLE (Fc) were measured in separate assays, respectively, and were shown to have turbidity of 0.066 and 0.026 at 2500 μg / ml, respectively.

(Table 13) Antibody turbidity after concentration to 2.5 mg / ml overnight<img file="JP6907124B2_D0015.tif" />

Example 14 Therapeutic efficacy of CDH3 × CD3 bispecific antibody in a human tumor xenograft model On day 1 of the study, cells of human epidermal carcinoma cell line A-431 and a newly isolated human PBMC were subjected to female NOD. / SCID mice were co-injected subcutaneously into the right dorsal abdomen (E: T cell ratio 1: 2). Mice in vehicle control group 1 (n = 5) were not injected with effector cells and were used as non-transplanted controls for comparison with vehicle control group 2 (n = 10, effector cells injected). The effect of PBMC on tumor growth in the absence was monitored.

To mice, 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), A CDH3 × CD3 bispecific antibody that specifically binds to the extracellular CDH3 subdomain D2C was started approximately 2 hours after tumor cell injection on day 1 by daily intravenous bolus injection for 10 consecutive days. It was administered.

Tumors were measured by calipers during the study and progression was assessed by group-to-group comparison of tumor volume (TV). Tumor growth inhibition T / C [%] was measured by calculating TV as T / C% = 100 × (median TV in the analyzed group) / (median TV in control group 2).

The results are shown in Fig. 11. Administration of CDH3 x CD3 bispecific antibody to mice at a dose level of 0.5 mg / kg / day resulted in complete inhibition of tumorigenesis and animals that developed tumors by the end of the study (day 40). There wasn't.

Example 15 Antitumor activity assay of CDH3 × CD3 bispecific HLE antibody in HCT-116 tumor model was performed in female NOD / SCID mice subcutaneously injected with human HCT-116 colon carcinoma cells. Human CD3 by growing effector cells in vitro⁺Activated T cells (day 12). Treatment is about 200 mm tumor³It started when the volume of was reached (day 17). The control group was a q5d vehicle treatment group using T cells. Antibodies with SEQ ID NO: 425 were administered by intravenous bolus injection at concentrations of 5 mg / kg / dose (group 2) and 0.5 mg / kg / dose (group 3) every 5 days (q5d). The results are shown in FIGS. 12A and 12B. In particular, FIG. 12B distinguishes the results in terms of responding animals (7/10) and non-responsive animals (3/10). Although the reason for non-response in 30% of animals is unclear, Figure 12B shows tumors in which administration of a half-life-prolonging bispecific construct at a concentration of 5 mg / kg to responding animals began at the time of antibody administration. Shows that it results in growth arrest.

Example 16 T Cell Activation Activation Assay Isolated PBMCs from healthy human donors, increasing the concentration of CDH3-13 × I2C or CDH3-13 × I2C-HALB bispecific antibody constructs (0.001 pM-20 μM). ), Incubated for 48 hours. Expression of the activation marker CD69 on CD4 + and CD8 + T cells was confirmed by immunostaining and flow cytometry and the antigen-specific complex mAb. The results are shown in FIG. 13 and are discussed above herein.

Example 17 Cyno pharmacokinetic study of extended half-life CDH3 × CD3 constructs Female cynomolgus monkeys were given a buffer containing 0.015 mg / kg extended half-life CDH3 × CD3 bispecific antibody construct (administration volume 1 ml / kg) for 60 minutes. Using, iv injection was performed. The three HLE formats tested were P156, HALB, and HALB variant 1 (see SEQ ID NOs: 437, 443, and 444), each fused to the C-terminus of each construct. The half-lives of these HLE constructs in the cyno model were analyzed at 4 (P156) and 6 (HALB, HALB variant 1) plasma concentrations from 96 hours post-dose to the end of the study. Calculated based on. As a result, the HLE construct at P156 each showed a half-life of 57 hours, the HLE construct at HALB showed a half-life of 63-85 hours, and the HLE construct at HALB variant 1 showed a half-life of 68 hours. .. These PK properties suggest weekly iv dosing in humans.

BAYER II array<img file="JP6907124B2_D0016.tif" /><img file="JP6907124B2_D0017.tif" /><img file="JP6907124B2_D0018.tif" /><img file="JP6907124B2_D0019.tif" /><img file="JP6907124B2_D0020.tif" /><img file="JP6907124B2_D0021.tif" /><img file="JP6907124B2_D0022.tif" /><img file="JP6907124B2_D0023.tif" /><img file="JP6907124B2_D0024.tif" /><img file="JP6907124B2_D0025.tif" /><img file="JP6907124B2_D0026.tif" /><img file="JP6907124B2_D0027.tif" /><img file="JP6907124B2_D0028.tif" /><img file="JP6907124B2_D0029.tif" /><img file="JP6907124B2_D0030.tif" /><img file="JP6907124B2_D0031.tif" /><img file="JP6907124B2_D0032.tif" /><img file="JP6907124B2_D0033.tif" /><img file="JP6907124B2_D0034.tif" /><img file="JP6907124B2_D0035.tif" /><img file="JP6907124B2_D0036.tif" /><img file="JP6907124B2_D0037.tif" /><img file="JP6907124B2_D0038.tif" /><img file="JP6907124B2_D0039.tif" /><img file="JP6907124B2_D0040.tif" /><img file="JP6907124B2_D0041.tif" /><img file="JP6907124B2_D0042.tif" /><img file="JP6907124B2_D0043.tif" /><img file="JP6907124B2_D0044.tif" /><img file="JP6907124B2_D0045.tif" /><img file="JP6907124B2_D0046.tif" /><img file="JP6907124B2_D0047.tif" /><img file="JP6907124B2_D0048.tif" /><img file="JP6907124B2_D0049.tif" /><img file="JP6907124B2_D0050.tif" /><img file="JP6907124B2_D0051.tif" /><img file="JP6907124B2_D0052.tif" /><img file="JP6907124B2_D0053.tif" /><img file="JP6907124B2_D0054.tif" /><img file="JP6907124B2_D0055.tif" /><img file="JP6907124B2_D0056.tif" /><img file="JP6907124B2_D0057.tif" /><img file="JP6907124B2_D0058.tif" /><img file="JP6907124B2_D0059.tif" /><img file="JP6907124B2_D0060.tif" /><img file="JP6907124B2_D0061.tif" /><img file="JP6907124B2_D0062.tif" /><img file="JP6907124B2_D0063.tif" /><img file="JP6907124B2_D0064.tif" /><img file="JP6907124B2_D0065.tif" /><img file="JP6907124B2_D0066.tif" /><img file="JP6907124B2_D0067.tif" /><img file="JP6907124B2_D0068.tif" /><img file="JP6907124B2_D0069.tif" /><img file="JP6907124B2_D0070.tif" /><img file="JP6907124B2_D0071.tif" /><img file="JP6907124B2_D0072.tif" /><img file="JP6907124B2_D0073.tif" /><img file="JP6907124B2_D0074.tif" /><img file="JP6907124B2_D0075.tif" /><img file="JP6907124B2_D0076.tif" /><img file="JP6907124B2_D0077.tif" /><img file="JP6907124B2_D0078.tif" /><img file="JP6907124B2_D0079.tif" /><img file="JP6907124B2_D0080.tif" /><img file="JP6907124B2_D0081.tif" /><img file="JP6907124B2_D0082.tif" /><img file="JP6907124B2_D0083.tif" /><img file="JP6907124B2_D0084.tif" /><img file="JP6907124B2_D0085.tif" /><img file="JP6907124B2_D0086.tif" /><img file="JP6907124B2_D0087.tif" /><img file="JP6907124B2_D0088.tif" /><img file="JP6907124B2_D0089.tif" /><img file="JP6907124B2_D0090.tif" /><img file="JP6907124B2_D0091.tif" /><img file="JP6907124B2_D0092.tif" /><img file="JP6907124B2_D0093.tif" /><img file="JP6907124B2_D0094.tif" /><img file="JP6907124B2_D0095.tif" /><img file="JP6907124B2_D0096.tif" /><img file="JP6907124B2_D0097.tif" /><img file="JP6907124B2_D0098.tif" /><img file="JP6907124B2_D0099.tif" /><img file="JP6907124B2_D0100.tif" /><img file="JP6907124B2_D0101.tif" /><img file="JP6907124B2_D0102.tif" /><img file="JP6907124B2_D0103.tif" /><img file="JP6907124B2_D0104.tif" /><img file="JP6907124B2_D0105.tif" /><img file="JP6907124B2_D0106.tif" /><img file="JP6907124B2_D0107.tif" /><img file="JP6907124B2_D0108.tif" /><img file="JP6907124B2_D0109.tif" /><img file="JP6907124B2_D0110.tif" /><img file="JP6907124B2_D0111.tif" /><img file="JP6907124B2_D0112.tif" /><img file="JP6907124B2_D0113.tif" /><img file="JP6907124B2_D0114.tif" /><img file="JP6907124B2_D0115.tif" /><img file="JP6907124B2_D0116.tif" /><img file="JP6907124B2_D0117.tif" /><img file="JP6907124B2_D0118.tif" /><img file="JP6907124B2_D0119.tif" /><img file="JP6907124B2_D0120.tif" /><img file="JP6907124B2_D0121.tif" /><img file="JP6907124B2_D0122.tif" /><img file="JP6907124B2_D0123.tif" /><img file="JP6907124B2_D0124.tif" /><img file="JP6907124B2_D0125.tif" /><img file="JP6907124B2_D0126.tif" /><img file="JP6907124B2_D0127.tif" /><img file="JP6907124B2_D0128.tif" /><img file="JP6907124B2_D0129.tif" /><img file="JP6907124B2_D0130.tif" /><img file="JP6907124B2_D0131.tif" /><img file="JP6907124B2_D0132.tif" /><img file="JP6907124B2_D0133.tif" /><img file="JP6907124B2_D0134.tif" /><img file="JP6907124B2_D0135.tif" /><img file="JP6907124B2_D0136.tif" /><img file="JP6907124B2_D0137.tif" /><img file="JP6907124B2_D0138.tif" />